diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,9328 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "global_step": 1502,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.999999972657395e-05,
+      "loss": 0.6994,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.999999890629579e-05,
+      "loss": 0.6608,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.999999753916553e-05,
+      "loss": 0.6388,
+      "step": 3
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.999999562518318e-05,
+      "loss": 0.6756,
+      "step": 4
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.999999316434876e-05,
+      "loss": 0.6275,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.999999015666233e-05,
+      "loss": 0.6326,
+      "step": 6
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.999998660212388e-05,
+      "loss": 0.6227,
+      "step": 7
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.999998250073347e-05,
+      "loss": 0.5957,
+      "step": 8
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.999997785249113e-05,
+      "loss": 0.5689,
+      "step": 9
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.999997265739693e-05,
+      "loss": 0.5414,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.999996691545092e-05,
+      "loss": 0.5057,
+      "step": 11
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.999996062665318e-05,
+      "loss": 0.5144,
+      "step": 12
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.999995379100373e-05,
+      "loss": 0.494,
+      "step": 13
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.999994640850269e-05,
+      "loss": 0.4604,
+      "step": 14
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.999993847915012e-05,
+      "loss": 0.5193,
+      "step": 15
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.999993000294612e-05,
+      "loss": 0.4891,
+      "step": 16
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.999992097989076e-05,
+      "loss": 0.4434,
+      "step": 17
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.999991140998416e-05,
+      "loss": 0.4517,
+      "step": 18
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999990129322643e-05,
+      "loss": 0.4627,
+      "step": 19
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999989062961765e-05,
+      "loss": 0.4189,
+      "step": 20
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999987941915795e-05,
+      "loss": 0.4412,
+      "step": 21
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999986766184748e-05,
+      "loss": 0.4427,
+      "step": 22
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999985535768635e-05,
+      "loss": 0.4288,
+      "step": 23
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999984250667468e-05,
+      "loss": 0.4352,
+      "step": 24
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999982910881263e-05,
+      "loss": 0.4292,
+      "step": 25
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.999981516410031e-05,
+      "loss": 0.3909,
+      "step": 26
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.999980067253793e-05,
+      "loss": 0.3963,
+      "step": 27
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.999978563412562e-05,
+      "loss": 0.4599,
+      "step": 28
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.999977004886353e-05,
+      "loss": 0.4081,
+      "step": 29
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.999975391675185e-05,
+      "loss": 0.4265,
+      "step": 30
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.999973723779074e-05,
+      "loss": 0.3877,
+      "step": 31
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.999972001198042e-05,
+      "loss": 0.436,
+      "step": 32
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.999970223932103e-05,
+      "loss": 0.3692,
+      "step": 33
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.99996839198128e-05,
+      "loss": 0.3845,
+      "step": 34
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.99996650534559e-05,
+      "loss": 0.394,
+      "step": 35
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.999964564025055e-05,
+      "loss": 0.384,
+      "step": 36
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.999962568019699e-05,
+      "loss": 0.3938,
+      "step": 37
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.999960517329539e-05,
+      "loss": 0.3981,
+      "step": 38
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.9999584119546e-05,
+      "loss": 0.3667,
+      "step": 39
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.999956251894907e-05,
+      "loss": 0.3842,
+      "step": 40
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.999954037150479e-05,
+      "loss": 0.3929,
+      "step": 41
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.999951767721342e-05,
+      "loss": 0.3786,
+      "step": 42
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.999949443607524e-05,
+      "loss": 0.3767,
+      "step": 43
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.999947064809048e-05,
+      "loss": 0.3602,
+      "step": 44
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.999944631325937e-05,
+      "loss": 0.3636,
+      "step": 45
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.999942143158223e-05,
+      "loss": 0.363,
+      "step": 46
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.99993960030593e-05,
+      "loss": 0.3341,
+      "step": 47
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.999937002769086e-05,
+      "loss": 0.361,
+      "step": 48
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.999934350547721e-05,
+      "loss": 0.349,
+      "step": 49
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.999931643641862e-05,
+      "loss": 0.3545,
+      "step": 50
+    },
+    {
+      "epoch": 0.07,
+      "eval_accuracy": 0.838751214487223,
+      "eval_f1": 0.8754171603252268,
+      "eval_loss": 0.361499547958374,
+      "eval_runtime": 110.3932,
+      "eval_samples_per_second": 773.843,
+      "eval_steps_per_second": 0.761,
+      "step": 50
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.999928882051539e-05,
+      "loss": 0.3998,
+      "step": 51
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.999926065776783e-05,
+      "loss": 0.3802,
+      "step": 52
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.999923194817626e-05,
+      "loss": 0.3573,
+      "step": 53
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.999920269174098e-05,
+      "loss": 0.3426,
+      "step": 54
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.99991728884623e-05,
+      "loss": 0.3643,
+      "step": 55
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.999914253834055e-05,
+      "loss": 0.3655,
+      "step": 56
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.999911164137608e-05,
+      "loss": 0.3496,
+      "step": 57
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.999908019756921e-05,
+      "loss": 0.3989,
+      "step": 58
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.999904820692027e-05,
+      "loss": 0.3351,
+      "step": 59
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.999901566942966e-05,
+      "loss": 0.376,
+      "step": 60
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.99989825850977e-05,
+      "loss": 0.3694,
+      "step": 61
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.999894895392474e-05,
+      "loss": 0.3687,
+      "step": 62
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.999891477591118e-05,
+      "loss": 0.346,
+      "step": 63
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.999888005105738e-05,
+      "loss": 0.3349,
+      "step": 64
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.99988447793637e-05,
+      "loss": 0.3686,
+      "step": 65
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.999880896083056e-05,
+      "loss": 0.3433,
+      "step": 66
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.999877259545834e-05,
+      "loss": 0.3706,
+      "step": 67
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.999873568324741e-05,
+      "loss": 0.3658,
+      "step": 68
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.99986982241982e-05,
+      "loss": 0.3708,
+      "step": 69
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.999866021831115e-05,
+      "loss": 0.3419,
+      "step": 70
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.999862166558662e-05,
+      "loss": 0.3414,
+      "step": 71
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.999858256602506e-05,
+      "loss": 0.3262,
+      "step": 72
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.999854291962687e-05,
+      "loss": 0.3674,
+      "step": 73
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.999850272639252e-05,
+      "loss": 0.3449,
+      "step": 74
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.999846198632245e-05,
+      "loss": 0.3624,
+      "step": 75
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.999842069941708e-05,
+      "loss": 0.3334,
+      "step": 76
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.999837886567687e-05,
+      "loss": 0.3668,
+      "step": 77
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.999833648510229e-05,
+      "loss": 0.3829,
+      "step": 78
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.999829355769377e-05,
+      "loss": 0.3576,
+      "step": 79
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.999825008345182e-05,
+      "loss": 0.3691,
+      "step": 80
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.999820606237689e-05,
+      "loss": 0.3692,
+      "step": 81
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.999816149446948e-05,
+      "loss": 0.3636,
+      "step": 82
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.999811637973007e-05,
+      "loss": 0.3509,
+      "step": 83
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.999807071815913e-05,
+      "loss": 0.3857,
+      "step": 84
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.99980245097572e-05,
+      "loss": 0.3423,
+      "step": 85
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.999797775452475e-05,
+      "loss": 0.355,
+      "step": 86
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.999793045246232e-05,
+      "loss": 0.3347,
+      "step": 87
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.99978826035704e-05,
+      "loss": 0.3521,
+      "step": 88
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.999783420784954e-05,
+      "loss": 0.3379,
+      "step": 89
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.999778526530025e-05,
+      "loss": 0.3319,
+      "step": 90
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.999773577592308e-05,
+      "loss": 0.363,
+      "step": 91
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.999768573971855e-05,
+      "loss": 0.335,
+      "step": 92
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.999763515668723e-05,
+      "loss": 0.3615,
+      "step": 93
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.999758402682966e-05,
+      "loss": 0.342,
+      "step": 94
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.999753235014641e-05,
+      "loss": 0.3715,
+      "step": 95
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.999748012663803e-05,
+      "loss": 0.3446,
+      "step": 96
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.99974273563051e-05,
+      "loss": 0.3389,
+      "step": 97
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.999737403914822e-05,
+      "loss": 0.3619,
+      "step": 98
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.999732017516792e-05,
+      "loss": 0.3333,
+      "step": 99
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.999726576436483e-05,
+      "loss": 0.3682,
+      "step": 100
+    },
+    {
+      "epoch": 0.13,
+      "eval_accuracy": 0.8537698853992297,
+      "eval_f1": 0.884870603848706,
+      "eval_loss": 0.3303517997264862,
+      "eval_runtime": 110.3842,
+      "eval_samples_per_second": 773.906,
+      "eval_steps_per_second": 0.761,
+      "step": 100
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.999721080673952e-05,
+      "loss": 0.3209,
+      "step": 101
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.999715530229262e-05,
+      "loss": 0.3175,
+      "step": 102
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.999709925102471e-05,
+      "loss": 0.3436,
+      "step": 103
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.999704265293644e-05,
+      "loss": 0.3371,
+      "step": 104
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.999698550802838e-05,
+      "loss": 0.3351,
+      "step": 105
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.999692781630119e-05,
+      "loss": 0.3356,
+      "step": 106
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.999686957775549e-05,
+      "loss": 0.3343,
+      "step": 107
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.999681079239192e-05,
+      "loss": 0.3456,
+      "step": 108
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.99967514602111e-05,
+      "loss": 0.3117,
+      "step": 109
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.999669158121372e-05,
+      "loss": 0.3291,
+      "step": 110
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.99966311554004e-05,
+      "loss": 0.314,
+      "step": 111
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.999657018277182e-05,
+      "loss": 0.361,
+      "step": 112
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.999650866332863e-05,
+      "loss": 0.3191,
+      "step": 113
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.999644659707154e-05,
+      "loss": 0.3844,
+      "step": 114
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.999638398400118e-05,
+      "loss": 0.3491,
+      "step": 115
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.999632082411827e-05,
+      "loss": 0.313,
+      "step": 116
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.999625711742349e-05,
+      "loss": 0.3732,
+      "step": 117
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.999619286391753e-05,
+      "loss": 0.3449,
+      "step": 118
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.99961280636011e-05,
+      "loss": 0.3411,
+      "step": 119
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.99960627164749e-05,
+      "loss": 0.3195,
+      "step": 120
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.999599682253967e-05,
+      "loss": 0.3302,
+      "step": 121
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.99959303817961e-05,
+      "loss": 0.3101,
+      "step": 122
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.999586339424491e-05,
+      "loss": 0.3325,
+      "step": 123
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.999579585988687e-05,
+      "loss": 0.3694,
+      "step": 124
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.99957277787227e-05,
+      "loss": 0.3316,
+      "step": 125
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.999565915075315e-05,
+      "loss": 0.323,
+      "step": 126
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.999558997597895e-05,
+      "loss": 0.3542,
+      "step": 127
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.999552025440089e-05,
+      "loss": 0.3518,
+      "step": 128
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.999544998601971e-05,
+      "loss": 0.3281,
+      "step": 129
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.999537917083619e-05,
+      "loss": 0.3287,
+      "step": 130
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.999530780885107e-05,
+      "loss": 0.3076,
+      "step": 131
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.999523590006518e-05,
+      "loss": 0.3135,
+      "step": 132
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.999516344447927e-05,
+      "loss": 0.3284,
+      "step": 133
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.999509044209417e-05,
+      "loss": 0.3323,
+      "step": 134
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.999501689291065e-05,
+      "loss": 0.3107,
+      "step": 135
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.999494279692952e-05,
+      "loss": 0.322,
+      "step": 136
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.999486815415157e-05,
+      "loss": 0.3545,
+      "step": 137
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.999479296457766e-05,
+      "loss": 0.3075,
+      "step": 138
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.999471722820857e-05,
+      "loss": 0.3171,
+      "step": 139
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.999464094504516e-05,
+      "loss": 0.3073,
+      "step": 140
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.999456411508824e-05,
+      "loss": 0.2914,
+      "step": 141
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.999448673833867e-05,
+      "loss": 0.3406,
+      "step": 142
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.999440881479729e-05,
+      "loss": 0.2846,
+      "step": 143
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.999433034446495e-05,
+      "loss": 0.3163,
+      "step": 144
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.99942513273425e-05,
+      "loss": 0.351,
+      "step": 145
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.999417176343082e-05,
+      "loss": 0.3439,
+      "step": 146
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.999409165273077e-05,
+      "loss": 0.3357,
+      "step": 147
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.999401099524324e-05,
+      "loss": 0.3154,
+      "step": 148
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.999392979096909e-05,
+      "loss": 0.3111,
+      "step": 149
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.999384803990921e-05,
+      "loss": 0.3231,
+      "step": 150
+    },
+    {
+      "epoch": 0.2,
+      "eval_accuracy": 0.8603251899282428,
+      "eval_f1": 0.889286840982055,
+      "eval_loss": 0.3143615126609802,
+      "eval_runtime": 110.4415,
+      "eval_samples_per_second": 773.505,
+      "eval_steps_per_second": 0.761,
+      "step": 150
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.999376574206451e-05,
+      "loss": 0.3204,
+      "step": 151
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.999368289743591e-05,
+      "loss": 0.3318,
+      "step": 152
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.999359950602425e-05,
+      "loss": 0.346,
+      "step": 153
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.999351556783051e-05,
+      "loss": 0.3152,
+      "step": 154
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.999343108285557e-05,
+      "loss": 0.3332,
+      "step": 155
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.999334605110036e-05,
+      "loss": 0.3289,
+      "step": 156
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.999326047256583e-05,
+      "loss": 0.3169,
+      "step": 157
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.999317434725288e-05,
+      "loss": 0.2985,
+      "step": 158
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.999308767516249e-05,
+      "loss": 0.3266,
+      "step": 159
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.999300045629557e-05,
+      "loss": 0.3345,
+      "step": 160
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.99929126906531e-05,
+      "loss": 0.3415,
+      "step": 161
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.999282437823605e-05,
+      "loss": 0.3601,
+      "step": 162
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.999273551904536e-05,
+      "loss": 0.3205,
+      "step": 163
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.9992646113082e-05,
+      "loss": 0.2974,
+      "step": 164
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.999255616034697e-05,
+      "loss": 0.3292,
+      "step": 165
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.999246566084124e-05,
+      "loss": 0.3228,
+      "step": 166
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.99923746145658e-05,
+      "loss": 0.3282,
+      "step": 167
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.999228302152165e-05,
+      "loss": 0.3541,
+      "step": 168
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.999219088170979e-05,
+      "loss": 0.3385,
+      "step": 169
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.999209819513123e-05,
+      "loss": 0.3379,
+      "step": 170
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.999200496178698e-05,
+      "loss": 0.3317,
+      "step": 171
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.999191118167807e-05,
+      "loss": 0.3441,
+      "step": 172
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.99918168548055e-05,
+      "loss": 0.3426,
+      "step": 173
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.999172198117032e-05,
+      "loss": 0.3518,
+      "step": 174
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.999162656077359e-05,
+      "loss": 0.3172,
+      "step": 175
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.999153059361631e-05,
+      "loss": 0.3309,
+      "step": 176
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 9.999143407969955e-05,
+      "loss": 0.2979,
+      "step": 177
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 9.999133701902438e-05,
+      "loss": 0.3192,
+      "step": 178
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 9.999123941159183e-05,
+      "loss": 0.3422,
+      "step": 179
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 9.999114125740299e-05,
+      "loss": 0.3607,
+      "step": 180
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 9.999104255645893e-05,
+      "loss": 0.3121,
+      "step": 181
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 9.999094330876073e-05,
+      "loss": 0.3034,
+      "step": 182
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 9.999084351430948e-05,
+      "loss": 0.3048,
+      "step": 183
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 9.999074317310624e-05,
+      "loss": 0.3242,
+      "step": 184
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 9.999064228515214e-05,
+      "loss": 0.3049,
+      "step": 185
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 9.999054085044827e-05,
+      "loss": 0.2907,
+      "step": 186
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 9.999043886899577e-05,
+      "loss": 0.2926,
+      "step": 187
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 9.99903363407957e-05,
+      "loss": 0.3155,
+      "step": 188
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 9.999023326584925e-05,
+      "loss": 0.326,
+      "step": 189
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 9.999012964415747e-05,
+      "loss": 0.3052,
+      "step": 190
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 9.999002547572155e-05,
+      "loss": 0.281,
+      "step": 191
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 9.998992076054261e-05,
+      "loss": 0.3194,
+      "step": 192
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 9.998981549862179e-05,
+      "loss": 0.3583,
+      "step": 193
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 9.998970968996025e-05,
+      "loss": 0.3103,
+      "step": 194
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 9.998960333455915e-05,
+      "loss": 0.3117,
+      "step": 195
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 9.998949643241963e-05,
+      "loss": 0.3454,
+      "step": 196
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 9.99893889835429e-05,
+      "loss": 0.3094,
+      "step": 197
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 9.99892809879301e-05,
+      "loss": 0.2976,
+      "step": 198
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 9.998917244558244e-05,
+      "loss": 0.3437,
+      "step": 199
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 9.998906335650107e-05,
+      "loss": 0.3095,
+      "step": 200
+    },
+    {
+      "epoch": 0.27,
+      "eval_accuracy": 0.8668570826553665,
+      "eval_f1": 0.8957718600516834,
+      "eval_loss": 0.30482426285743713,
+      "eval_runtime": 110.2629,
+      "eval_samples_per_second": 774.758,
+      "eval_steps_per_second": 0.762,
+      "step": 200
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 9.998895372068721e-05,
+      "loss": 0.2742,
+      "step": 201
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 9.998884353814207e-05,
+      "loss": 0.3123,
+      "step": 202
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 9.99887328088668e-05,
+      "loss": 0.2748,
+      "step": 203
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 9.998862153286269e-05,
+      "loss": 0.3261,
+      "step": 204
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 9.998850971013089e-05,
+      "loss": 0.3109,
+      "step": 205
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 9.998839734067266e-05,
+      "loss": 0.3038,
+      "step": 206
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 9.998828442448922e-05,
+      "loss": 0.3203,
+      "step": 207
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 9.99881709615818e-05,
+      "loss": 0.308,
+      "step": 208
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 9.998805695195164e-05,
+      "loss": 0.2998,
+      "step": 209
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 9.998794239559998e-05,
+      "loss": 0.3142,
+      "step": 210
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 9.998782729252812e-05,
+      "loss": 0.3085,
+      "step": 211
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 9.998771164273724e-05,
+      "loss": 0.3135,
+      "step": 212
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 9.998759544622866e-05,
+      "loss": 0.307,
+      "step": 213
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 9.998747870300367e-05,
+      "loss": 0.3197,
+      "step": 214
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 9.998736141306349e-05,
+      "loss": 0.3614,
+      "step": 215
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 9.998724357640943e-05,
+      "loss": 0.314,
+      "step": 216
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 9.998712519304276e-05,
+      "loss": 0.3145,
+      "step": 217
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 9.998700626296483e-05,
+      "loss": 0.3143,
+      "step": 218
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 9.998688678617687e-05,
+      "loss": 0.3355,
+      "step": 219
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 9.998676676268023e-05,
+      "loss": 0.2987,
+      "step": 220
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 9.998664619247622e-05,
+      "loss": 0.2727,
+      "step": 221
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 9.998652507556614e-05,
+      "loss": 0.3341,
+      "step": 222
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 9.99864034119513e-05,
+      "loss": 0.2953,
+      "step": 223
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 9.99862812016331e-05,
+      "loss": 0.3044,
+      "step": 224
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 9.998615844461282e-05,
+      "loss": 0.3085,
+      "step": 225
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 9.998603514089181e-05,
+      "loss": 0.3145,
+      "step": 226
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 9.998591129047143e-05,
+      "loss": 0.303,
+      "step": 227
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 9.998578689335302e-05,
+      "loss": 0.3108,
+      "step": 228
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 9.998566194953796e-05,
+      "loss": 0.3065,
+      "step": 229
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 9.998553645902759e-05,
+      "loss": 0.3177,
+      "step": 230
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 9.99854104218233e-05,
+      "loss": 0.3153,
+      "step": 231
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 9.998528383792647e-05,
+      "loss": 0.3124,
+      "step": 232
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 9.998515670733848e-05,
+      "loss": 0.3519,
+      "step": 233
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 9.998502903006073e-05,
+      "loss": 0.3227,
+      "step": 234
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 9.998490080609459e-05,
+      "loss": 0.2731,
+      "step": 235
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 9.99847720354415e-05,
+      "loss": 0.3231,
+      "step": 236
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 9.998464271810284e-05,
+      "loss": 0.3196,
+      "step": 237
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 9.998451285408001e-05,
+      "loss": 0.3286,
+      "step": 238
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 9.998438244337449e-05,
+      "loss": 0.3067,
+      "step": 239
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 9.998425148598764e-05,
+      "loss": 0.3149,
+      "step": 240
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 9.998411998192095e-05,
+      "loss": 0.3235,
+      "step": 241
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 9.99839879311758e-05,
+      "loss": 0.3225,
+      "step": 242
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 9.998385533375369e-05,
+      "loss": 0.2936,
+      "step": 243
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 9.998372218965603e-05,
+      "loss": 0.3247,
+      "step": 244
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 9.998358849888429e-05,
+      "loss": 0.2941,
+      "step": 245
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 9.998345426143994e-05,
+      "loss": 0.3162,
+      "step": 246
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 9.998331947732443e-05,
+      "loss": 0.2808,
+      "step": 247
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 9.998318414653925e-05,
+      "loss": 0.3113,
+      "step": 248
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 9.998304826908588e-05,
+      "loss": 0.2924,
+      "step": 249
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 9.99829118449658e-05,
+      "loss": 0.3137,
+      "step": 250
+    },
+    {
+      "epoch": 0.33,
+      "eval_accuracy": 0.872241797089913,
+      "eval_f1": 0.9005775502395832,
+      "eval_loss": 0.29461291432380676,
+      "eval_runtime": 110.2326,
+      "eval_samples_per_second": 774.97,
+      "eval_steps_per_second": 0.762,
+      "step": 250
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 9.998277487418049e-05,
+      "loss": 0.284,
+      "step": 251
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 9.998263735673149e-05,
+      "loss": 0.294,
+      "step": 252
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 9.998249929262025e-05,
+      "loss": 0.2827,
+      "step": 253
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 9.998236068184831e-05,
+      "loss": 0.2885,
+      "step": 254
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 9.998222152441718e-05,
+      "loss": 0.2995,
+      "step": 255
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 9.998208182032839e-05,
+      "loss": 0.3083,
+      "step": 256
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 9.998194156958345e-05,
+      "loss": 0.315,
+      "step": 257
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 9.998180077218392e-05,
+      "loss": 0.3069,
+      "step": 258
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 9.998165942813131e-05,
+      "loss": 0.2975,
+      "step": 259
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 9.99815175374272e-05,
+      "loss": 0.305,
+      "step": 260
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 9.99813751000731e-05,
+      "loss": 0.3152,
+      "step": 261
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 9.998123211607062e-05,
+      "loss": 0.3487,
+      "step": 262
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 9.998108858542128e-05,
+      "loss": 0.3331,
+      "step": 263
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 9.998094450812665e-05,
+      "loss": 0.318,
+      "step": 264
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 9.998079988418833e-05,
+      "loss": 0.3215,
+      "step": 265
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 9.99806547136079e-05,
+      "loss": 0.3053,
+      "step": 266
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.998050899638694e-05,
+      "loss": 0.3137,
+      "step": 267
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.998036273252703e-05,
+      "loss": 0.2992,
+      "step": 268
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.99802159220298e-05,
+      "loss": 0.3058,
+      "step": 269
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.998006856489682e-05,
+      "loss": 0.3084,
+      "step": 270
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.997992066112973e-05,
+      "loss": 0.3199,
+      "step": 271
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.997977221073013e-05,
+      "loss": 0.3086,
+      "step": 272
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.997962321369967e-05,
+      "loss": 0.3154,
+      "step": 273
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.997947367003994e-05,
+      "loss": 0.3003,
+      "step": 274
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 9.997932357975262e-05,
+      "loss": 0.3173,
+      "step": 275
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 9.997917294283929e-05,
+      "loss": 0.286,
+      "step": 276
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 9.997902175930168e-05,
+      "loss": 0.3094,
+      "step": 277
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 9.997887002914137e-05,
+      "loss": 0.2959,
+      "step": 278
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 9.997871775236005e-05,
+      "loss": 0.2829,
+      "step": 279
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 9.99785649289594e-05,
+      "loss": 0.2925,
+      "step": 280
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 9.997841155894107e-05,
+      "loss": 0.3205,
+      "step": 281
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 9.997825764230674e-05,
+      "loss": 0.2901,
+      "step": 282
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 9.99781031790581e-05,
+      "loss": 0.2741,
+      "step": 283
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 9.997794816919684e-05,
+      "loss": 0.2822,
+      "step": 284
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 9.997779261272464e-05,
+      "loss": 0.2883,
+      "step": 285
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 9.997763650964322e-05,
+      "loss": 0.3418,
+      "step": 286
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 9.997747985995427e-05,
+      "loss": 0.3004,
+      "step": 287
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 9.997732266365951e-05,
+      "loss": 0.296,
+      "step": 288
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 9.997716492076068e-05,
+      "loss": 0.3136,
+      "step": 289
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 9.997700663125949e-05,
+      "loss": 0.3291,
+      "step": 290
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 9.997684779515765e-05,
+      "loss": 0.2975,
+      "step": 291
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 9.997668841245692e-05,
+      "loss": 0.2965,
+      "step": 292
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 9.997652848315904e-05,
+      "loss": 0.3046,
+      "step": 293
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 9.997636800726578e-05,
+      "loss": 0.3121,
+      "step": 294
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 9.997620698477884e-05,
+      "loss": 0.3002,
+      "step": 295
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 9.997604541570003e-05,
+      "loss": 0.3103,
+      "step": 296
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 9.99758833000311e-05,
+      "loss": 0.331,
+      "step": 297
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 9.997572063777382e-05,
+      "loss": 0.328,
+      "step": 298
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 9.997555742892999e-05,
+      "loss": 0.2955,
+      "step": 299
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 9.997539367350137e-05,
+      "loss": 0.3296,
+      "step": 300
+    },
+    {
+      "epoch": 0.4,
+      "eval_accuracy": 0.8743254474580636,
+      "eval_f1": 0.9011418047882137,
+      "eval_loss": 0.2919522821903229,
+      "eval_runtime": 110.3351,
+      "eval_samples_per_second": 774.25,
+      "eval_steps_per_second": 0.761,
+      "step": 300
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 9.997522937148976e-05,
+      "loss": 0.284,
+      "step": 301
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 9.997506452289696e-05,
+      "loss": 0.289,
+      "step": 302
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 9.997489912772475e-05,
+      "loss": 0.315,
+      "step": 303
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 9.997473318597499e-05,
+      "loss": 0.2855,
+      "step": 304
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 9.997456669764945e-05,
+      "loss": 0.3001,
+      "step": 305
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 9.997439966274995e-05,
+      "loss": 0.2993,
+      "step": 306
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 9.997423208127834e-05,
+      "loss": 0.326,
+      "step": 307
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 9.997406395323645e-05,
+      "loss": 0.317,
+      "step": 308
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 9.997389527862611e-05,
+      "loss": 0.3075,
+      "step": 309
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 9.997372605744918e-05,
+      "loss": 0.3032,
+      "step": 310
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 9.997355628970748e-05,
+      "loss": 0.3137,
+      "step": 311
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 9.997338597540288e-05,
+      "loss": 0.29,
+      "step": 312
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 9.997321511453728e-05,
+      "loss": 0.319,
+      "step": 313
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 9.997304370711249e-05,
+      "loss": 0.286,
+      "step": 314
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 9.997287175313041e-05,
+      "loss": 0.3025,
+      "step": 315
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 9.997269925259293e-05,
+      "loss": 0.282,
+      "step": 316
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 9.997252620550193e-05,
+      "loss": 0.2629,
+      "step": 317
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 9.99723526118593e-05,
+      "loss": 0.2908,
+      "step": 318
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 9.997217847166694e-05,
+      "loss": 0.3139,
+      "step": 319
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 9.997200378492676e-05,
+      "loss": 0.296,
+      "step": 320
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 9.997182855164066e-05,
+      "loss": 0.3338,
+      "step": 321
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 9.997165277181056e-05,
+      "loss": 0.305,
+      "step": 322
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 9.997147644543838e-05,
+      "loss": 0.3046,
+      "step": 323
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 9.997129957252607e-05,
+      "loss": 0.3266,
+      "step": 324
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 9.997112215307553e-05,
+      "loss": 0.3143,
+      "step": 325
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 9.997094418708872e-05,
+      "loss": 0.3052,
+      "step": 326
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 9.99707656745676e-05,
+      "loss": 0.3021,
+      "step": 327
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 9.997058661551408e-05,
+      "loss": 0.2708,
+      "step": 328
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 9.997040700993017e-05,
+      "loss": 0.2659,
+      "step": 329
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 9.99702268578178e-05,
+      "loss": 0.2714,
+      "step": 330
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 9.997004615917896e-05,
+      "loss": 0.3382,
+      "step": 331
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 9.996986491401559e-05,
+      "loss": 0.3285,
+      "step": 332
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 9.99696831223297e-05,
+      "loss": 0.2955,
+      "step": 333
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 9.996950078412331e-05,
+      "loss": 0.3214,
+      "step": 334
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 9.996931789939834e-05,
+      "loss": 0.298,
+      "step": 335
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 9.996913446815686e-05,
+      "loss": 0.305,
+      "step": 336
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 9.996895049040083e-05,
+      "loss": 0.3045,
+      "step": 337
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 9.99687659661323e-05,
+      "loss": 0.2887,
+      "step": 338
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 9.996858089535324e-05,
+      "loss": 0.2876,
+      "step": 339
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 9.996839527806572e-05,
+      "loss": 0.2953,
+      "step": 340
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 9.996820911427173e-05,
+      "loss": 0.2936,
+      "step": 341
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 9.996802240397336e-05,
+      "loss": 0.2865,
+      "step": 342
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 9.99678351471726e-05,
+      "loss": 0.3168,
+      "step": 343
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 9.996764734387151e-05,
+      "loss": 0.3185,
+      "step": 344
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 9.996745899407216e-05,
+      "loss": 0.2876,
+      "step": 345
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 9.996727009777661e-05,
+      "loss": 0.3169,
+      "step": 346
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 9.996708065498692e-05,
+      "loss": 0.3007,
+      "step": 347
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 9.996689066570515e-05,
+      "loss": 0.2881,
+      "step": 348
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 9.996670012993338e-05,
+      "loss": 0.2982,
+      "step": 349
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 9.996650904767372e-05,
+      "loss": 0.3162,
+      "step": 350
+    },
+    {
+      "epoch": 0.47,
+      "eval_accuracy": 0.8758940381846488,
+      "eval_f1": 0.9015690279454089,
+      "eval_loss": 0.2856006622314453,
+      "eval_runtime": 110.3808,
+      "eval_samples_per_second": 773.93,
+      "eval_steps_per_second": 0.761,
+      "step": 350
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 9.996631741892823e-05,
+      "loss": 0.3084,
+      "step": 351
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 9.996612524369902e-05,
+      "loss": 0.2985,
+      "step": 352
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 9.996593252198819e-05,
+      "loss": 0.3124,
+      "step": 353
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 9.996573925379783e-05,
+      "loss": 0.296,
+      "step": 354
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 9.99655454391301e-05,
+      "loss": 0.2763,
+      "step": 355
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 9.996535107798707e-05,
+      "loss": 0.3208,
+      "step": 356
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 9.996515617037089e-05,
+      "loss": 0.316,
+      "step": 357
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 9.99649607162837e-05,
+      "loss": 0.3152,
+      "step": 358
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 9.99647647157276e-05,
+      "loss": 0.2862,
+      "step": 359
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 9.996456816870479e-05,
+      "loss": 0.3084,
+      "step": 360
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 9.996437107521737e-05,
+      "loss": 0.3298,
+      "step": 361
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 9.996417343526751e-05,
+      "loss": 0.3073,
+      "step": 362
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 9.996397524885738e-05,
+      "loss": 0.2956,
+      "step": 363
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 9.996377651598916e-05,
+      "loss": 0.3033,
+      "step": 364
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 9.996357723666498e-05,
+      "loss": 0.3091,
+      "step": 365
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 9.996337741088707e-05,
+      "loss": 0.2838,
+      "step": 366
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 9.996317703865758e-05,
+      "loss": 0.2959,
+      "step": 367
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 9.996297611997871e-05,
+      "loss": 0.3001,
+      "step": 368
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 9.996277465485266e-05,
+      "loss": 0.2905,
+      "step": 369
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 9.996257264328166e-05,
+      "loss": 0.2876,
+      "step": 370
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 9.996237008526788e-05,
+      "loss": 0.2964,
+      "step": 371
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.996216698081354e-05,
+      "loss": 0.3077,
+      "step": 372
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.996196332992087e-05,
+      "loss": 0.2955,
+      "step": 373
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.99617591325921e-05,
+      "loss": 0.3191,
+      "step": 374
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.996155438882947e-05,
+      "loss": 0.306,
+      "step": 375
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.99613490986352e-05,
+      "loss": 0.2869,
+      "step": 376
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.996114326201156e-05,
+      "loss": 0.2997,
+      "step": 377
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.996093687896078e-05,
+      "loss": 0.3031,
+      "step": 378
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 9.996072994948512e-05,
+      "loss": 0.2983,
+      "step": 379
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 9.996052247358685e-05,
+      "loss": 0.3062,
+      "step": 380
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 9.996031445126823e-05,
+      "loss": 0.3064,
+      "step": 381
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 9.996010588253155e-05,
+      "loss": 0.2906,
+      "step": 382
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 9.995989676737909e-05,
+      "loss": 0.2888,
+      "step": 383
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 9.995968710581313e-05,
+      "loss": 0.3178,
+      "step": 384
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 9.995947689783596e-05,
+      "loss": 0.3119,
+      "step": 385
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 9.995926614344987e-05,
+      "loss": 0.2935,
+      "step": 386
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 9.995905484265719e-05,
+      "loss": 0.286,
+      "step": 387
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 9.995884299546022e-05,
+      "loss": 0.2978,
+      "step": 388
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 9.995863060186126e-05,
+      "loss": 0.3109,
+      "step": 389
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 9.995841766186265e-05,
+      "loss": 0.3073,
+      "step": 390
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 9.995820417546673e-05,
+      "loss": 0.2701,
+      "step": 391
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 9.995799014267582e-05,
+      "loss": 0.3074,
+      "step": 392
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 9.995777556349226e-05,
+      "loss": 0.2851,
+      "step": 393
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 9.99575604379184e-05,
+      "loss": 0.3282,
+      "step": 394
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 9.995734476595658e-05,
+      "loss": 0.3084,
+      "step": 395
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 9.99571285476092e-05,
+      "loss": 0.2579,
+      "step": 396
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 9.995691178287856e-05,
+      "loss": 0.3171,
+      "step": 397
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 9.99566944717671e-05,
+      "loss": 0.2844,
+      "step": 398
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 9.995647661427714e-05,
+      "loss": 0.294,
+      "step": 399
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 9.99562582104111e-05,
+      "loss": 0.2879,
+      "step": 400
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.8786449249066455,
+      "eval_f1": 0.9048244204728023,
+      "eval_loss": 0.28084805607795715,
+      "eval_runtime": 110.3341,
+      "eval_samples_per_second": 774.258,
+      "eval_steps_per_second": 0.761,
+      "step": 400
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 9.995603926017136e-05,
+      "loss": 0.2661,
+      "step": 401
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 9.99558197635603e-05,
+      "loss": 0.2736,
+      "step": 402
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 9.995559972058033e-05,
+      "loss": 0.2832,
+      "step": 403
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 9.995537913123385e-05,
+      "loss": 0.2726,
+      "step": 404
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 9.99551579955233e-05,
+      "loss": 0.2746,
+      "step": 405
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 9.995493631345106e-05,
+      "loss": 0.297,
+      "step": 406
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 9.995471408501959e-05,
+      "loss": 0.3015,
+      "step": 407
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 9.99544913102313e-05,
+      "loss": 0.2676,
+      "step": 408
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 9.99542679890886e-05,
+      "loss": 0.2732,
+      "step": 409
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 9.995404412159399e-05,
+      "loss": 0.2633,
+      "step": 410
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 9.995381970774989e-05,
+      "loss": 0.2725,
+      "step": 411
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 9.995359474755875e-05,
+      "loss": 0.2876,
+      "step": 412
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 9.995336924102303e-05,
+      "loss": 0.3068,
+      "step": 413
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 9.995314318814522e-05,
+      "loss": 0.318,
+      "step": 414
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 9.995291658892776e-05,
+      "loss": 0.2756,
+      "step": 415
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 9.995268944337315e-05,
+      "loss": 0.2896,
+      "step": 416
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 9.995246175148386e-05,
+      "loss": 0.2954,
+      "step": 417
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 9.995223351326239e-05,
+      "loss": 0.2889,
+      "step": 418
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 9.995200472871123e-05,
+      "loss": 0.284,
+      "step": 419
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 9.995177539783289e-05,
+      "loss": 0.3147,
+      "step": 420
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 9.995154552062987e-05,
+      "loss": 0.3026,
+      "step": 421
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 9.995131509710469e-05,
+      "loss": 0.292,
+      "step": 422
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 9.995108412725988e-05,
+      "loss": 0.3082,
+      "step": 423
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 9.995085261109794e-05,
+      "loss": 0.2913,
+      "step": 424
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 9.99506205486214e-05,
+      "loss": 0.2891,
+      "step": 425
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 9.995038793983285e-05,
+      "loss": 0.2959,
+      "step": 426
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 9.995015478473478e-05,
+      "loss": 0.3299,
+      "step": 427
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 9.994992108332977e-05,
+      "loss": 0.2801,
+      "step": 428
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 9.994968683562036e-05,
+      "loss": 0.289,
+      "step": 429
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 9.99494520416091e-05,
+      "loss": 0.3006,
+      "step": 430
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 9.99492167012986e-05,
+      "loss": 0.3089,
+      "step": 431
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 9.99489808146914e-05,
+      "loss": 0.3047,
+      "step": 432
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 9.994874438179007e-05,
+      "loss": 0.2773,
+      "step": 433
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 9.994850740259722e-05,
+      "loss": 0.2948,
+      "step": 434
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 9.994826987711545e-05,
+      "loss": 0.297,
+      "step": 435
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 9.994803180534734e-05,
+      "loss": 0.2842,
+      "step": 436
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 9.99477931872955e-05,
+      "loss": 0.3203,
+      "step": 437
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 9.994755402296253e-05,
+      "loss": 0.2769,
+      "step": 438
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 9.994731431235106e-05,
+      "loss": 0.2874,
+      "step": 439
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 9.99470740554637e-05,
+      "loss": 0.3025,
+      "step": 440
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 9.994683325230309e-05,
+      "loss": 0.2883,
+      "step": 441
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 9.994659190287184e-05,
+      "loss": 0.3271,
+      "step": 442
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 9.994635000717263e-05,
+      "loss": 0.2828,
+      "step": 443
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 9.994610756520806e-05,
+      "loss": 0.305,
+      "step": 444
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 9.99458645769808e-05,
+      "loss": 0.2962,
+      "step": 445
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 9.994562104249353e-05,
+      "loss": 0.294,
+      "step": 446
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 9.994537696174888e-05,
+      "loss": 0.3001,
+      "step": 447
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 9.994513233474954e-05,
+      "loss": 0.2951,
+      "step": 448
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 9.994488716149819e-05,
+      "loss": 0.2714,
+      "step": 449
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 9.994464144199749e-05,
+      "loss": 0.2994,
+      "step": 450
+    },
+    {
+      "epoch": 0.6,
+      "eval_accuracy": 0.8745595654769569,
+      "eval_f1": 0.9053473951984737,
+      "eval_loss": 0.2921605110168457,
+      "eval_runtime": 110.1799,
+      "eval_samples_per_second": 775.341,
+      "eval_steps_per_second": 0.762,
+      "step": 450
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 9.994439517625014e-05,
+      "loss": 0.3174,
+      "step": 451
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 9.994414836425883e-05,
+      "loss": 0.2803,
+      "step": 452
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 9.994390100602626e-05,
+      "loss": 0.2929,
+      "step": 453
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 9.994365310155514e-05,
+      "loss": 0.2952,
+      "step": 454
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 9.994340465084818e-05,
+      "loss": 0.314,
+      "step": 455
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 9.994315565390809e-05,
+      "loss": 0.2821,
+      "step": 456
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 9.994290611073759e-05,
+      "loss": 0.2955,
+      "step": 457
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 9.994265602133944e-05,
+      "loss": 0.3225,
+      "step": 458
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 9.994240538571633e-05,
+      "loss": 0.3088,
+      "step": 459
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 9.994215420387103e-05,
+      "loss": 0.3079,
+      "step": 460
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 9.994190247580628e-05,
+      "loss": 0.305,
+      "step": 461
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.994165020152483e-05,
+      "loss": 0.2941,
+      "step": 462
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.994139738102946e-05,
+      "loss": 0.2675,
+      "step": 463
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.994114401432291e-05,
+      "loss": 0.3072,
+      "step": 464
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.994089010140797e-05,
+      "loss": 0.2785,
+      "step": 465
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.994063564228739e-05,
+      "loss": 0.3189,
+      "step": 466
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.994038063696399e-05,
+      "loss": 0.2882,
+      "step": 467
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.994012508544052e-05,
+      "loss": 0.3256,
+      "step": 468
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 9.993986898771981e-05,
+      "loss": 0.3049,
+      "step": 469
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.993961234380463e-05,
+      "loss": 0.303,
+      "step": 470
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.993935515369781e-05,
+      "loss": 0.2906,
+      "step": 471
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.993909741740215e-05,
+      "loss": 0.284,
+      "step": 472
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.993883913492048e-05,
+      "loss": 0.2862,
+      "step": 473
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.993858030625561e-05,
+      "loss": 0.2927,
+      "step": 474
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.993832093141038e-05,
+      "loss": 0.2725,
+      "step": 475
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 9.993806101038762e-05,
+      "loss": 0.2619,
+      "step": 476
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.99378005431902e-05,
+      "loss": 0.2909,
+      "step": 477
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.993753952982093e-05,
+      "loss": 0.2652,
+      "step": 478
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.993727797028269e-05,
+      "loss": 0.2832,
+      "step": 479
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.993701586457833e-05,
+      "loss": 0.2789,
+      "step": 480
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.993675321271072e-05,
+      "loss": 0.2833,
+      "step": 481
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.993649001468272e-05,
+      "loss": 0.2718,
+      "step": 482
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.993622627049724e-05,
+      "loss": 0.3131,
+      "step": 483
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 9.993596198015714e-05,
+      "loss": 0.2781,
+      "step": 484
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 9.99356971436653e-05,
+      "loss": 0.2866,
+      "step": 485
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 9.993543176102465e-05,
+      "loss": 0.2681,
+      "step": 486
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 9.993516583223808e-05,
+      "loss": 0.2828,
+      "step": 487
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 9.993489935730848e-05,
+      "loss": 0.3017,
+      "step": 488
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 9.993463233623879e-05,
+      "loss": 0.2836,
+      "step": 489
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 9.993436476903191e-05,
+      "loss": 0.3034,
+      "step": 490
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 9.993409665569077e-05,
+      "loss": 0.2847,
+      "step": 491
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 9.993382799621831e-05,
+      "loss": 0.2902,
+      "step": 492
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 9.993355879061746e-05,
+      "loss": 0.2857,
+      "step": 493
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 9.993328903889119e-05,
+      "loss": 0.2937,
+      "step": 494
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 9.99330187410424e-05,
+      "loss": 0.2915,
+      "step": 495
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 9.99327478970741e-05,
+      "loss": 0.2899,
+      "step": 496
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 9.993247650698923e-05,
+      "loss": 0.3007,
+      "step": 497
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 9.993220457079073e-05,
+      "loss": 0.2854,
+      "step": 498
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 9.993193208848162e-05,
+      "loss": 0.301,
+      "step": 499
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 9.993165906006488e-05,
+      "loss": 0.276,
+      "step": 500
+    },
+    {
+      "epoch": 0.67,
+      "eval_accuracy": 0.8828824610486146,
+      "eval_f1": 0.907681660899654,
+      "eval_loss": 0.27344685792922974,
+      "eval_runtime": 110.318,
+      "eval_samples_per_second": 774.37,
+      "eval_steps_per_second": 0.761,
+      "step": 500
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 9.993138548554345e-05,
+      "loss": 0.294,
+      "step": 501
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 9.993111136492035e-05,
+      "loss": 0.2633,
+      "step": 502
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 9.993083669819859e-05,
+      "loss": 0.2872,
+      "step": 503
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 9.993056148538115e-05,
+      "loss": 0.2847,
+      "step": 504
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 9.993028572647105e-05,
+      "loss": 0.2878,
+      "step": 505
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 9.993000942147132e-05,
+      "loss": 0.2439,
+      "step": 506
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 9.992973257038495e-05,
+      "loss": 0.3225,
+      "step": 507
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 9.992945517321501e-05,
+      "loss": 0.276,
+      "step": 508
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 9.992917722996451e-05,
+      "loss": 0.2724,
+      "step": 509
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 9.992889874063647e-05,
+      "loss": 0.2864,
+      "step": 510
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 9.992861970523397e-05,
+      "loss": 0.2867,
+      "step": 511
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 9.992834012376004e-05,
+      "loss": 0.2797,
+      "step": 512
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 9.992805999621777e-05,
+      "loss": 0.2791,
+      "step": 513
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 9.992777932261018e-05,
+      "loss": 0.3239,
+      "step": 514
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 9.992749810294037e-05,
+      "loss": 0.3073,
+      "step": 515
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 9.992721633721139e-05,
+      "loss": 0.2578,
+      "step": 516
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 9.992693402542634e-05,
+      "loss": 0.2922,
+      "step": 517
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 9.992665116758832e-05,
+      "loss": 0.3011,
+      "step": 518
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 9.992636776370041e-05,
+      "loss": 0.3103,
+      "step": 519
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 9.992608381376569e-05,
+      "loss": 0.3228,
+      "step": 520
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 9.99257993177873e-05,
+      "loss": 0.2914,
+      "step": 521
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 9.992551427576832e-05,
+      "loss": 0.3035,
+      "step": 522
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 9.99252286877119e-05,
+      "loss": 0.264,
+      "step": 523
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 9.992494255362113e-05,
+      "loss": 0.2701,
+      "step": 524
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 9.992465587349918e-05,
+      "loss": 0.2783,
+      "step": 525
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 9.992436864734913e-05,
+      "loss": 0.3069,
+      "step": 526
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 9.992408087517418e-05,
+      "loss": 0.287,
+      "step": 527
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 9.992379255697743e-05,
+      "loss": 0.2916,
+      "step": 528
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 9.992350369276206e-05,
+      "loss": 0.2808,
+      "step": 529
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 9.992321428253122e-05,
+      "loss": 0.3041,
+      "step": 530
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 9.992292432628809e-05,
+      "loss": 0.3139,
+      "step": 531
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 9.992263382403581e-05,
+      "loss": 0.2983,
+      "step": 532
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 9.99223427757776e-05,
+      "loss": 0.278,
+      "step": 533
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 9.99220511815166e-05,
+      "loss": 0.2806,
+      "step": 534
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 9.992175904125603e-05,
+      "loss": 0.3131,
+      "step": 535
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 9.992146635499906e-05,
+      "loss": 0.2897,
+      "step": 536
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 9.992117312274892e-05,
+      "loss": 0.3143,
+      "step": 537
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 9.99208793445088e-05,
+      "loss": 0.2941,
+      "step": 538
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 9.99205850202819e-05,
+      "loss": 0.2791,
+      "step": 539
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 9.992029015007147e-05,
+      "loss": 0.2764,
+      "step": 540
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 9.991999473388072e-05,
+      "loss": 0.2444,
+      "step": 541
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 9.991969877171287e-05,
+      "loss": 0.2488,
+      "step": 542
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 9.991940226357117e-05,
+      "loss": 0.2993,
+      "step": 543
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 9.991910520945886e-05,
+      "loss": 0.3032,
+      "step": 544
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 9.99188076093792e-05,
+      "loss": 0.2791,
+      "step": 545
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 9.991850946333542e-05,
+      "loss": 0.2598,
+      "step": 546
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 9.991821077133081e-05,
+      "loss": 0.2506,
+      "step": 547
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 9.991791153336861e-05,
+      "loss": 0.2822,
+      "step": 548
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 9.99176117494521e-05,
+      "loss": 0.287,
+      "step": 549
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 9.991731141958458e-05,
+      "loss": 0.2924,
+      "step": 550
+    },
+    {
+      "epoch": 0.73,
+      "eval_accuracy": 0.881536282439978,
+      "eval_f1": 0.9056392660002983,
+      "eval_loss": 0.2759096026420593,
+      "eval_runtime": 110.2362,
+      "eval_samples_per_second": 774.945,
+      "eval_steps_per_second": 0.762,
+      "step": 550
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 9.991701054376932e-05,
+      "loss": 0.3105,
+      "step": 551
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 9.991670912200959e-05,
+      "loss": 0.2634,
+      "step": 552
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 9.99164071543087e-05,
+      "loss": 0.2963,
+      "step": 553
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 9.991610464066999e-05,
+      "loss": 0.2913,
+      "step": 554
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 9.99158015810967e-05,
+      "loss": 0.2748,
+      "step": 555
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 9.991549797559221e-05,
+      "loss": 0.2952,
+      "step": 556
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 9.991519382415979e-05,
+      "loss": 0.2835,
+      "step": 557
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 9.99148891268028e-05,
+      "loss": 0.2684,
+      "step": 558
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 9.991458388352456e-05,
+      "loss": 0.2823,
+      "step": 559
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 9.99142780943284e-05,
+      "loss": 0.3091,
+      "step": 560
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 9.991397175921768e-05,
+      "loss": 0.2692,
+      "step": 561
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 9.991366487819576e-05,
+      "loss": 0.2525,
+      "step": 562
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 9.991335745126595e-05,
+      "loss": 0.2848,
+      "step": 563
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 9.991304947843166e-05,
+      "loss": 0.263,
+      "step": 564
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 9.991274095969624e-05,
+      "loss": 0.297,
+      "step": 565
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 9.991243189506306e-05,
+      "loss": 0.3176,
+      "step": 566
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 9.991212228453551e-05,
+      "loss": 0.2795,
+      "step": 567
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 9.991181212811698e-05,
+      "loss": 0.2737,
+      "step": 568
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 9.991150142581084e-05,
+      "loss": 0.2848,
+      "step": 569
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 9.99111901776205e-05,
+      "loss": 0.2871,
+      "step": 570
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 9.991087838354938e-05,
+      "loss": 0.2873,
+      "step": 571
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 9.991056604360088e-05,
+      "loss": 0.3092,
+      "step": 572
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 9.991025315777839e-05,
+      "loss": 0.3036,
+      "step": 573
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 9.990993972608535e-05,
+      "loss": 0.2829,
+      "step": 574
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 9.990962574852523e-05,
+      "loss": 0.2842,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 9.990931122510139e-05,
+      "loss": 0.262,
+      "step": 576
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 9.990899615581732e-05,
+      "loss": 0.2806,
+      "step": 577
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 9.990868054067646e-05,
+      "loss": 0.3061,
+      "step": 578
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 9.990836437968224e-05,
+      "loss": 0.2729,
+      "step": 579
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 9.990804767283815e-05,
+      "loss": 0.2609,
+      "step": 580
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 9.990773042014761e-05,
+      "loss": 0.2718,
+      "step": 581
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 9.990741262161414e-05,
+      "loss": 0.3026,
+      "step": 582
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 9.990709427724117e-05,
+      "loss": 0.2669,
+      "step": 583
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 9.990677538703221e-05,
+      "loss": 0.2979,
+      "step": 584
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 9.990645595099075e-05,
+      "loss": 0.2552,
+      "step": 585
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 9.990613596912028e-05,
+      "loss": 0.2825,
+      "step": 586
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 9.990581544142428e-05,
+      "loss": 0.3041,
+      "step": 587
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 9.990549436790629e-05,
+      "loss": 0.2752,
+      "step": 588
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 9.990517274856976e-05,
+      "loss": 0.3109,
+      "step": 589
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 9.990485058341829e-05,
+      "loss": 0.2888,
+      "step": 590
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 9.990452787245534e-05,
+      "loss": 0.2788,
+      "step": 591
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 9.990420461568447e-05,
+      "loss": 0.3068,
+      "step": 592
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 9.990388081310921e-05,
+      "loss": 0.2605,
+      "step": 593
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 9.990355646473309e-05,
+      "loss": 0.293,
+      "step": 594
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 9.990323157055968e-05,
+      "loss": 0.2699,
+      "step": 595
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 9.990290613059251e-05,
+      "loss": 0.2991,
+      "step": 596
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 9.990258014483516e-05,
+      "loss": 0.2428,
+      "step": 597
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.990225361329117e-05,
+      "loss": 0.2778,
+      "step": 598
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.990192653596413e-05,
+      "loss": 0.2538,
+      "step": 599
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.990159891285761e-05,
+      "loss": 0.2583,
+      "step": 600
+    },
+    {
+      "epoch": 0.8,
+      "eval_accuracy": 0.8856801713743898,
+      "eval_f1": 0.9111001875216197,
+      "eval_loss": 0.271943598985672,
+      "eval_runtime": 110.8142,
+      "eval_samples_per_second": 770.903,
+      "eval_steps_per_second": 0.758,
+      "step": 600
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.990127074397519e-05,
+      "loss": 0.2746,
+      "step": 601
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.990094202932049e-05,
+      "loss": 0.3115,
+      "step": 602
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.990061276889707e-05,
+      "loss": 0.2867,
+      "step": 603
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 9.990028296270852e-05,
+      "loss": 0.2999,
+      "step": 604
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.989995261075849e-05,
+      "loss": 0.2873,
+      "step": 605
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.989962171305059e-05,
+      "loss": 0.2606,
+      "step": 606
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.989929026958838e-05,
+      "loss": 0.3095,
+      "step": 607
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.989895828037555e-05,
+      "loss": 0.2954,
+      "step": 608
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.989862574541571e-05,
+      "loss": 0.2783,
+      "step": 609
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.989829266471249e-05,
+      "loss": 0.2963,
+      "step": 610
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.989795903826953e-05,
+      "loss": 0.2621,
+      "step": 611
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 9.98976248660905e-05,
+      "loss": 0.2729,
+      "step": 612
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 9.989729014817901e-05,
+      "loss": 0.2557,
+      "step": 613
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 9.989695488453877e-05,
+      "loss": 0.3073,
+      "step": 614
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 9.989661907517343e-05,
+      "loss": 0.2593,
+      "step": 615
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 9.989628272008666e-05,
+      "loss": 0.2437,
+      "step": 616
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 9.989594581928214e-05,
+      "loss": 0.2907,
+      "step": 617
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 9.989560837276355e-05,
+      "loss": 0.2803,
+      "step": 618
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 9.989527038053459e-05,
+      "loss": 0.2884,
+      "step": 619
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.989493184259895e-05,
+      "loss": 0.2854,
+      "step": 620
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.989459275896034e-05,
+      "loss": 0.2724,
+      "step": 621
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.989425312962246e-05,
+      "loss": 0.3106,
+      "step": 622
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.989391295458902e-05,
+      "loss": 0.2618,
+      "step": 623
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.989357223386376e-05,
+      "loss": 0.2897,
+      "step": 624
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.989323096745038e-05,
+      "loss": 0.2712,
+      "step": 625
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.989288915535263e-05,
+      "loss": 0.2483,
+      "step": 626
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.989254679757424e-05,
+      "loss": 0.2851,
+      "step": 627
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 9.989220389411898e-05,
+      "loss": 0.3061,
+      "step": 628
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 9.989186044499058e-05,
+      "loss": 0.2937,
+      "step": 629
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 9.989151645019277e-05,
+      "loss": 0.2966,
+      "step": 630
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 9.989117190972937e-05,
+      "loss": 0.2893,
+      "step": 631
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 9.989082682360408e-05,
+      "loss": 0.2542,
+      "step": 632
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 9.989048119182074e-05,
+      "loss": 0.2831,
+      "step": 633
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 9.989013501438309e-05,
+      "loss": 0.2887,
+      "step": 634
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 9.988978829129493e-05,
+      "loss": 0.2684,
+      "step": 635
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 9.988944102256004e-05,
+      "loss": 0.2603,
+      "step": 636
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 9.988909320818225e-05,
+      "loss": 0.2623,
+      "step": 637
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 9.988874484816533e-05,
+      "loss": 0.307,
+      "step": 638
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 9.988839594251309e-05,
+      "loss": 0.2919,
+      "step": 639
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 9.988804649122937e-05,
+      "loss": 0.3082,
+      "step": 640
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 9.988769649431798e-05,
+      "loss": 0.2911,
+      "step": 641
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 9.988734595178274e-05,
+      "loss": 0.2801,
+      "step": 642
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 9.988699486362749e-05,
+      "loss": 0.2903,
+      "step": 643
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 9.988664322985608e-05,
+      "loss": 0.2686,
+      "step": 644
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 9.988629105047235e-05,
+      "loss": 0.2465,
+      "step": 645
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 9.988593832548014e-05,
+      "loss": 0.275,
+      "step": 646
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 9.988558505488332e-05,
+      "loss": 0.2818,
+      "step": 647
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 9.988523123868575e-05,
+      "loss": 0.2735,
+      "step": 648
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 9.98848768768913e-05,
+      "loss": 0.2849,
+      "step": 649
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 9.988452196950384e-05,
+      "loss": 0.2855,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "eval_accuracy": 0.8854343474545519,
+      "eval_f1": 0.9092948034736188,
+      "eval_loss": 0.2715582549571991,
+      "eval_runtime": 110.3229,
+      "eval_samples_per_second": 774.336,
+      "eval_steps_per_second": 0.761,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 9.988416651652727e-05,
+      "loss": 0.2922,
+      "step": 651
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 9.988381051796545e-05,
+      "loss": 0.2826,
+      "step": 652
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 9.98834539738223e-05,
+      "loss": 0.2481,
+      "step": 653
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 9.988309688410171e-05,
+      "loss": 0.2655,
+      "step": 654
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 9.988273924880757e-05,
+      "loss": 0.2939,
+      "step": 655
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 9.988238106794383e-05,
+      "loss": 0.2949,
+      "step": 656
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 9.988202234151436e-05,
+      "loss": 0.2505,
+      "step": 657
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 9.988166306952312e-05,
+      "loss": 0.2769,
+      "step": 658
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 9.988130325197403e-05,
+      "loss": 0.2866,
+      "step": 659
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 9.9880942888871e-05,
+      "loss": 0.267,
+      "step": 660
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 9.988058198021799e-05,
+      "loss": 0.2562,
+      "step": 661
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 9.988022052601898e-05,
+      "loss": 0.2696,
+      "step": 662
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 9.987985852627786e-05,
+      "loss": 0.2879,
+      "step": 663
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 9.987949598099863e-05,
+      "loss": 0.2808,
+      "step": 664
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 9.987913289018526e-05,
+      "loss": 0.2731,
+      "step": 665
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 9.987876925384169e-05,
+      "loss": 0.2854,
+      "step": 666
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 9.987840507197193e-05,
+      "loss": 0.2517,
+      "step": 667
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 9.987804034457995e-05,
+      "loss": 0.3352,
+      "step": 668
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 9.987767507166971e-05,
+      "loss": 0.278,
+      "step": 669
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 9.987730925324526e-05,
+      "loss": 0.3167,
+      "step": 670
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 9.987694288931058e-05,
+      "loss": 0.2854,
+      "step": 671
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 9.987657597986965e-05,
+      "loss": 0.2542,
+      "step": 672
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 9.987620852492651e-05,
+      "loss": 0.2945,
+      "step": 673
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 9.987584052448518e-05,
+      "loss": 0.3062,
+      "step": 674
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 9.987547197854966e-05,
+      "loss": 0.2935,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 9.9875102887124e-05,
+      "loss": 0.2799,
+      "step": 676
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 9.987473325021224e-05,
+      "loss": 0.3217,
+      "step": 677
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 9.987436306781842e-05,
+      "loss": 0.2793,
+      "step": 678
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 9.987399233994658e-05,
+      "loss": 0.2619,
+      "step": 679
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 9.987362106660079e-05,
+      "loss": 0.2635,
+      "step": 680
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 9.98732492477851e-05,
+      "loss": 0.2952,
+      "step": 681
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 9.987287688350356e-05,
+      "loss": 0.276,
+      "step": 682
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 9.987250397376026e-05,
+      "loss": 0.2825,
+      "step": 683
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 9.987213051855931e-05,
+      "loss": 0.2732,
+      "step": 684
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 9.987175651790475e-05,
+      "loss": 0.2788,
+      "step": 685
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 9.987138197180066e-05,
+      "loss": 0.2588,
+      "step": 686
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 9.987100688025118e-05,
+      "loss": 0.2626,
+      "step": 687
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 9.98706312432604e-05,
+      "loss": 0.296,
+      "step": 688
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 9.98702550608324e-05,
+      "loss": 0.2842,
+      "step": 689
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 9.986987833297134e-05,
+      "loss": 0.2604,
+      "step": 690
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 9.98695010596813e-05,
+      "loss": 0.2862,
+      "step": 691
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 9.986912324096643e-05,
+      "loss": 0.286,
+      "step": 692
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 9.986874487683084e-05,
+      "loss": 0.3072,
+      "step": 693
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 9.986836596727868e-05,
+      "loss": 0.2692,
+      "step": 694
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 9.986798651231411e-05,
+      "loss": 0.2993,
+      "step": 695
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 9.986760651194127e-05,
+      "loss": 0.2942,
+      "step": 696
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 9.98672259661643e-05,
+      "loss": 0.2947,
+      "step": 697
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 9.986684487498739e-05,
+      "loss": 0.2766,
+      "step": 698
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 9.986646323841468e-05,
+      "loss": 0.3258,
+      "step": 699
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 9.986608105645035e-05,
+      "loss": 0.2896,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "eval_accuracy": 0.8867688201622438,
+      "eval_f1": 0.9105337637233049,
+      "eval_loss": 0.26790928840637207,
+      "eval_runtime": 110.6557,
+      "eval_samples_per_second": 772.007,
+      "eval_steps_per_second": 0.759,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 9.98656983290986e-05,
+      "loss": 0.2857,
+      "step": 701
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 9.98653150563636e-05,
+      "loss": 0.2726,
+      "step": 702
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 9.986493123824954e-05,
+      "loss": 0.2523,
+      "step": 703
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 9.986454687476064e-05,
+      "loss": 0.2864,
+      "step": 704
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 9.986416196590106e-05,
+      "loss": 0.2803,
+      "step": 705
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 9.986377651167504e-05,
+      "loss": 0.2856,
+      "step": 706
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 9.98633905120868e-05,
+      "loss": 0.3033,
+      "step": 707
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 9.986300396714054e-05,
+      "loss": 0.2924,
+      "step": 708
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 9.986261687684052e-05,
+      "loss": 0.2709,
+      "step": 709
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 9.986222924119096e-05,
+      "loss": 0.2849,
+      "step": 710
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 9.986184106019607e-05,
+      "loss": 0.2794,
+      "step": 711
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 9.986145233386013e-05,
+      "loss": 0.2846,
+      "step": 712
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 9.986106306218737e-05,
+      "loss": 0.2754,
+      "step": 713
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 9.986067324518208e-05,
+      "loss": 0.2495,
+      "step": 714
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 9.986028288284849e-05,
+      "loss": 0.2639,
+      "step": 715
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 9.985989197519087e-05,
+      "loss": 0.2479,
+      "step": 716
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 9.985950052221354e-05,
+      "loss": 0.3032,
+      "step": 717
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 9.98591085239207e-05,
+      "loss": 0.3031,
+      "step": 718
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 9.985871598031673e-05,
+      "loss": 0.2662,
+      "step": 719
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 9.985832289140587e-05,
+      "loss": 0.2829,
+      "step": 720
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 9.985792925719241e-05,
+      "loss": 0.2587,
+      "step": 721
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 9.985753507768068e-05,
+      "loss": 0.2726,
+      "step": 722
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 9.9857140352875e-05,
+      "loss": 0.2805,
+      "step": 723
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 9.985674508277964e-05,
+      "loss": 0.2835,
+      "step": 724
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 9.985634926739898e-05,
+      "loss": 0.29,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 9.985595290673731e-05,
+      "loss": 0.26,
+      "step": 726
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 9.985555600079899e-05,
+      "loss": 0.2768,
+      "step": 727
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 9.985515854958833e-05,
+      "loss": 0.2852,
+      "step": 728
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 9.985476055310972e-05,
+      "loss": 0.2911,
+      "step": 729
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 9.985436201136748e-05,
+      "loss": 0.2584,
+      "step": 730
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 9.985396292436597e-05,
+      "loss": 0.2896,
+      "step": 731
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 9.985356329210958e-05,
+      "loss": 0.3335,
+      "step": 732
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 9.985316311460264e-05,
+      "loss": 0.2817,
+      "step": 733
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 9.985276239184958e-05,
+      "loss": 0.2886,
+      "step": 734
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 9.985236112385473e-05,
+      "loss": 0.2875,
+      "step": 735
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 9.98519593106225e-05,
+      "loss": 0.2591,
+      "step": 736
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 9.985155695215732e-05,
+      "loss": 0.2433,
+      "step": 737
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 9.985115404846352e-05,
+      "loss": 0.2649,
+      "step": 738
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 9.985075059954556e-05,
+      "loss": 0.2653,
+      "step": 739
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 9.985034660540784e-05,
+      "loss": 0.2944,
+      "step": 740
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 9.984994206605476e-05,
+      "loss": 0.2868,
+      "step": 741
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 9.984953698149078e-05,
+      "loss": 0.295,
+      "step": 742
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 9.98491313517203e-05,
+      "loss": 0.2657,
+      "step": 743
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 9.984872517674775e-05,
+      "loss": 0.258,
+      "step": 744
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 9.98483184565776e-05,
+      "loss": 0.2814,
+      "step": 745
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 9.98479111912143e-05,
+      "loss": 0.2694,
+      "step": 746
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 9.984750338066228e-05,
+      "loss": 0.286,
+      "step": 747
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 9.9847095024926e-05,
+      "loss": 0.2812,
+      "step": 748
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 9.984668612400996e-05,
+      "loss": 0.2724,
+      "step": 749
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 9.984627667791859e-05,
+      "loss": 0.2531,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8866985847565758,
+      "eval_f1": 0.9098982527019354,
+      "eval_loss": 0.26476675271987915,
+      "eval_runtime": 110.3373,
+      "eval_samples_per_second": 774.235,
+      "eval_steps_per_second": 0.761,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 9.98458666866564e-05,
+      "loss": 0.2651,
+      "step": 751
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 9.984545615022786e-05,
+      "loss": 0.2413,
+      "step": 752
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 9.984504506863745e-05,
+      "loss": 0.2266,
+      "step": 753
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 9.98446334418897e-05,
+      "loss": 0.2403,
+      "step": 754
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.984422126998908e-05,
+      "loss": 0.2197,
+      "step": 755
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.984380855294011e-05,
+      "loss": 0.2342,
+      "step": 756
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.984339529074729e-05,
+      "loss": 0.2381,
+      "step": 757
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.984298148341517e-05,
+      "loss": 0.2686,
+      "step": 758
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.984256713094824e-05,
+      "loss": 0.2599,
+      "step": 759
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.984215223335106e-05,
+      "loss": 0.2232,
+      "step": 760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.984173679062816e-05,
+      "loss": 0.2249,
+      "step": 761
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.984132080278407e-05,
+      "loss": 0.2267,
+      "step": 762
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 9.984090426982336e-05,
+      "loss": 0.2379,
+      "step": 763
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 9.984048719175056e-05,
+      "loss": 0.2555,
+      "step": 764
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 9.984006956857026e-05,
+      "loss": 0.219,
+      "step": 765
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 9.983965140028701e-05,
+      "loss": 0.274,
+      "step": 766
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 9.98392326869054e-05,
+      "loss": 0.2323,
+      "step": 767
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 9.983881342843e-05,
+      "loss": 0.2636,
+      "step": 768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 9.983839362486538e-05,
+      "loss": 0.2562,
+      "step": 769
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 9.983797327621614e-05,
+      "loss": 0.2598,
+      "step": 770
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 9.983755238248689e-05,
+      "loss": 0.2519,
+      "step": 771
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 9.983713094368223e-05,
+      "loss": 0.2576,
+      "step": 772
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 9.983670895980677e-05,
+      "loss": 0.2256,
+      "step": 773
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 9.983628643086511e-05,
+      "loss": 0.26,
+      "step": 774
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 9.98358633568619e-05,
+      "loss": 0.2568,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 9.983543973780172e-05,
+      "loss": 0.2735,
+      "step": 776
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 9.983501557368925e-05,
+      "loss": 0.2481,
+      "step": 777
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 9.98345908645291e-05,
+      "loss": 0.2909,
+      "step": 778
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 9.983416561032594e-05,
+      "loss": 0.229,
+      "step": 779
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 9.983373981108439e-05,
+      "loss": 0.2708,
+      "step": 780
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 9.983331346680914e-05,
+      "loss": 0.2679,
+      "step": 781
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 9.983288657750484e-05,
+      "loss": 0.2205,
+      "step": 782
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 9.983245914317613e-05,
+      "loss": 0.2349,
+      "step": 783
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 9.983203116382773e-05,
+      "loss": 0.2449,
+      "step": 784
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.983160263946432e-05,
+      "loss": 0.2656,
+      "step": 785
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.983117357009052e-05,
+      "loss": 0.2555,
+      "step": 786
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.983074395571112e-05,
+      "loss": 0.222,
+      "step": 787
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.983031379633074e-05,
+      "loss": 0.2076,
+      "step": 788
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.982988309195413e-05,
+      "loss": 0.2579,
+      "step": 789
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.982945184258598e-05,
+      "loss": 0.2237,
+      "step": 790
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.9829020048231e-05,
+      "loss": 0.2388,
+      "step": 791
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 9.982858770889394e-05,
+      "loss": 0.2541,
+      "step": 792
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 9.982815482457951e-05,
+      "loss": 0.2608,
+      "step": 793
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 9.982772139529244e-05,
+      "loss": 0.24,
+      "step": 794
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 9.982728742103748e-05,
+      "loss": 0.2309,
+      "step": 795
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 9.982685290181939e-05,
+      "loss": 0.2291,
+      "step": 796
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 9.982641783764288e-05,
+      "loss": 0.2336,
+      "step": 797
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 9.982598222851276e-05,
+      "loss": 0.2528,
+      "step": 798
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 9.982554607443375e-05,
+      "loss": 0.2334,
+      "step": 799
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.982510937541065e-05,
+      "loss": 0.2413,
+      "step": 800
+    },
+    {
+      "epoch": 1.07,
+      "eval_accuracy": 0.8888407646294497,
+      "eval_f1": 0.9135202083674844,
+      "eval_loss": 0.26743802428245544,
+      "eval_runtime": 110.5438,
+      "eval_samples_per_second": 772.789,
+      "eval_steps_per_second": 0.76,
+      "step": 800
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.982467213144822e-05,
+      "loss": 0.2302,
+      "step": 801
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.982423434255125e-05,
+      "loss": 0.2507,
+      "step": 802
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.982379600872452e-05,
+      "loss": 0.2334,
+      "step": 803
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.982335712997283e-05,
+      "loss": 0.2199,
+      "step": 804
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.982291770630099e-05,
+      "loss": 0.2237,
+      "step": 805
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.982247773771379e-05,
+      "loss": 0.2606,
+      "step": 806
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 9.982203722421605e-05,
+      "loss": 0.2743,
+      "step": 807
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.982159616581259e-05,
+      "loss": 0.26,
+      "step": 808
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.982115456250822e-05,
+      "loss": 0.2561,
+      "step": 809
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.982071241430779e-05,
+      "loss": 0.2604,
+      "step": 810
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.982026972121612e-05,
+      "loss": 0.2823,
+      "step": 811
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.981982648323805e-05,
+      "loss": 0.2539,
+      "step": 812
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.981938270037845e-05,
+      "loss": 0.2552,
+      "step": 813
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 9.981893837264213e-05,
+      "loss": 0.2393,
+      "step": 814
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.981849350003401e-05,
+      "loss": 0.2568,
+      "step": 815
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.98180480825589e-05,
+      "loss": 0.2616,
+      "step": 816
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.981760212022171e-05,
+      "loss": 0.2564,
+      "step": 817
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.98171556130273e-05,
+      "loss": 0.2543,
+      "step": 818
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.981670856098055e-05,
+      "loss": 0.2736,
+      "step": 819
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.981626096408637e-05,
+      "loss": 0.2495,
+      "step": 820
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.981581282234962e-05,
+      "loss": 0.2486,
+      "step": 821
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 9.981536413577524e-05,
+      "loss": 0.2738,
+      "step": 822
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.981491490436812e-05,
+      "loss": 0.246,
+      "step": 823
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.981446512813316e-05,
+      "loss": 0.231,
+      "step": 824
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.98140148070753e-05,
+      "loss": 0.2264,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.981356394119944e-05,
+      "loss": 0.2406,
+      "step": 826
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.981311253051056e-05,
+      "loss": 0.2281,
+      "step": 827
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.981266057501355e-05,
+      "loss": 0.2515,
+      "step": 828
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 9.981220807471337e-05,
+      "loss": 0.2371,
+      "step": 829
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.981175502961494e-05,
+      "loss": 0.2705,
+      "step": 830
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.981130143972328e-05,
+      "loss": 0.2564,
+      "step": 831
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.981084730504332e-05,
+      "loss": 0.2253,
+      "step": 832
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.981039262557998e-05,
+      "loss": 0.2376,
+      "step": 833
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.980993740133831e-05,
+      "loss": 0.2668,
+      "step": 834
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.980948163232324e-05,
+      "loss": 0.2167,
+      "step": 835
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.980902531853975e-05,
+      "loss": 0.2257,
+      "step": 836
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 9.980856845999287e-05,
+      "loss": 0.2549,
+      "step": 837
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.980811105668756e-05,
+      "loss": 0.2156,
+      "step": 838
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.980765310862884e-05,
+      "loss": 0.2474,
+      "step": 839
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.980719461582172e-05,
+      "loss": 0.2344,
+      "step": 840
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.98067355782712e-05,
+      "loss": 0.2627,
+      "step": 841
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.980627599598232e-05,
+      "loss": 0.2692,
+      "step": 842
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.980581586896009e-05,
+      "loss": 0.2524,
+      "step": 843
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 9.980535519720957e-05,
+      "loss": 0.2442,
+      "step": 844
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 9.980489398073576e-05,
+      "loss": 0.236,
+      "step": 845
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 9.980443221954372e-05,
+      "loss": 0.2406,
+      "step": 846
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 9.98039699136385e-05,
+      "loss": 0.2702,
+      "step": 847
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 9.980350706302517e-05,
+      "loss": 0.2419,
+      "step": 848
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 9.980304366770878e-05,
+      "loss": 0.2583,
+      "step": 849
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 9.98025797276944e-05,
+      "loss": 0.2316,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "eval_accuracy": 0.890198649139031,
+      "eval_f1": 0.9136502559192842,
+      "eval_loss": 0.2640092968940735,
+      "eval_runtime": 110.6084,
+      "eval_samples_per_second": 772.337,
+      "eval_steps_per_second": 0.759,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 9.980211524298709e-05,
+      "loss": 0.238,
+      "step": 851
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 9.980165021359195e-05,
+      "loss": 0.2375,
+      "step": 852
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 9.980118463951406e-05,
+      "loss": 0.2164,
+      "step": 853
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 9.980071852075852e-05,
+      "loss": 0.2415,
+      "step": 854
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 9.980025185733041e-05,
+      "loss": 0.2251,
+      "step": 855
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 9.979978464923485e-05,
+      "loss": 0.2413,
+      "step": 856
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 9.979931689647694e-05,
+      "loss": 0.233,
+      "step": 857
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 9.979884859906178e-05,
+      "loss": 0.2067,
+      "step": 858
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 9.979837975699454e-05,
+      "loss": 0.2048,
+      "step": 859
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 9.97979103702803e-05,
+      "loss": 0.2355,
+      "step": 860
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 9.979744043892423e-05,
+      "loss": 0.2544,
+      "step": 861
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 9.979696996293144e-05,
+      "loss": 0.2622,
+      "step": 862
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 9.979649894230708e-05,
+      "loss": 0.235,
+      "step": 863
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 9.979602737705633e-05,
+      "loss": 0.2401,
+      "step": 864
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 9.97955552671843e-05,
+      "loss": 0.2459,
+      "step": 865
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 9.979508261269619e-05,
+      "loss": 0.254,
+      "step": 866
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 9.979460941359717e-05,
+      "loss": 0.2642,
+      "step": 867
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 9.97941356698924e-05,
+      "loss": 0.2446,
+      "step": 868
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 9.979366138158705e-05,
+      "loss": 0.231,
+      "step": 869
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 9.979318654868634e-05,
+      "loss": 0.2836,
+      "step": 870
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 9.979271117119546e-05,
+      "loss": 0.2308,
+      "step": 871
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 9.979223524911958e-05,
+      "loss": 0.2449,
+      "step": 872
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 9.979175878246391e-05,
+      "loss": 0.2596,
+      "step": 873
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 9.979128177123369e-05,
+      "loss": 0.24,
+      "step": 874
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 9.979080421543412e-05,
+      "loss": 0.2493,
+      "step": 875
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 9.97903261150704e-05,
+      "loss": 0.2413,
+      "step": 876
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 9.978984747014781e-05,
+      "loss": 0.2352,
+      "step": 877
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 9.978936828067153e-05,
+      "loss": 0.2544,
+      "step": 878
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 9.978888854664685e-05,
+      "loss": 0.2461,
+      "step": 879
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 9.978840826807899e-05,
+      "loss": 0.2413,
+      "step": 880
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 9.978792744497321e-05,
+      "loss": 0.2247,
+      "step": 881
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 9.978744607733475e-05,
+      "loss": 0.2515,
+      "step": 882
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 9.97869641651689e-05,
+      "loss": 0.276,
+      "step": 883
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 9.978648170848092e-05,
+      "loss": 0.2332,
+      "step": 884
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 9.978599870727608e-05,
+      "loss": 0.2566,
+      "step": 885
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 9.978551516155969e-05,
+      "loss": 0.2516,
+      "step": 886
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 9.978503107133701e-05,
+      "loss": 0.2861,
+      "step": 887
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 9.978454643661333e-05,
+      "loss": 0.2459,
+      "step": 888
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 9.978406125739398e-05,
+      "loss": 0.2313,
+      "step": 889
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 9.978357553368424e-05,
+      "loss": 0.2168,
+      "step": 890
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 9.978308926548945e-05,
+      "loss": 0.2207,
+      "step": 891
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 9.97826024528149e-05,
+      "loss": 0.243,
+      "step": 892
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 9.978211509566592e-05,
+      "loss": 0.2207,
+      "step": 893
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 9.978162719404785e-05,
+      "loss": 0.2329,
+      "step": 894
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 9.978113874796602e-05,
+      "loss": 0.2243,
+      "step": 895
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 9.978064975742578e-05,
+      "loss": 0.2976,
+      "step": 896
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 9.978016022243247e-05,
+      "loss": 0.2428,
+      "step": 897
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 9.977967014299145e-05,
+      "loss": 0.2362,
+      "step": 898
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 9.977917951910807e-05,
+      "loss": 0.2589,
+      "step": 899
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 9.97786883507877e-05,
+      "loss": 0.2534,
+      "step": 900
+    },
+    {
+      "epoch": 1.2,
+      "eval_accuracy": 0.8880447633652124,
+      "eval_f1": 0.9139153915391539,
+      "eval_loss": 0.2635560631752014,
+      "eval_runtime": 110.4436,
+      "eval_samples_per_second": 773.489,
+      "eval_steps_per_second": 0.761,
+      "step": 900
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 9.977819663803572e-05,
+      "loss": 0.2512,
+      "step": 901
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 9.977770438085749e-05,
+      "loss": 0.2634,
+      "step": 902
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 9.977721157925843e-05,
+      "loss": 0.2464,
+      "step": 903
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 9.977671823324389e-05,
+      "loss": 0.2661,
+      "step": 904
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 9.977622434281928e-05,
+      "loss": 0.2555,
+      "step": 905
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 9.977572990799e-05,
+      "loss": 0.2448,
+      "step": 906
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 9.977523492876149e-05,
+      "loss": 0.2744,
+      "step": 907
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 9.977473940513911e-05,
+      "loss": 0.2505,
+      "step": 908
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 9.97742433371283e-05,
+      "loss": 0.2188,
+      "step": 909
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 9.977374672473449e-05,
+      "loss": 0.2206,
+      "step": 910
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 9.977324956796312e-05,
+      "loss": 0.2558,
+      "step": 911
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 9.977275186681962e-05,
+      "loss": 0.2288,
+      "step": 912
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 9.977225362130943e-05,
+      "loss": 0.2385,
+      "step": 913
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 9.977175483143799e-05,
+      "loss": 0.2752,
+      "step": 914
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 9.977125549721078e-05,
+      "loss": 0.255,
+      "step": 915
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 9.977075561863324e-05,
+      "loss": 0.2163,
+      "step": 916
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 9.977025519571085e-05,
+      "loss": 0.2386,
+      "step": 917
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 9.976975422844908e-05,
+      "loss": 0.2399,
+      "step": 918
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 9.97692527168534e-05,
+      "loss": 0.2377,
+      "step": 919
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 9.97687506609293e-05,
+      "loss": 0.2318,
+      "step": 920
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 9.976824806068228e-05,
+      "loss": 0.2494,
+      "step": 921
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 9.976774491611783e-05,
+      "loss": 0.2432,
+      "step": 922
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 9.976724122724144e-05,
+      "loss": 0.2435,
+      "step": 923
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 9.976673699405866e-05,
+      "loss": 0.2134,
+      "step": 924
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 9.976623221657496e-05,
+      "loss": 0.2599,
+      "step": 925
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 9.976572689479587e-05,
+      "loss": 0.2391,
+      "step": 926
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 9.976522102872691e-05,
+      "loss": 0.2564,
+      "step": 927
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 9.976471461837366e-05,
+      "loss": 0.2468,
+      "step": 928
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 9.976420766374162e-05,
+      "loss": 0.246,
+      "step": 929
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 9.976370016483633e-05,
+      "loss": 0.2804,
+      "step": 930
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 9.976319212166335e-05,
+      "loss": 0.2913,
+      "step": 931
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 9.976268353422824e-05,
+      "loss": 0.2456,
+      "step": 932
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 9.976217440253657e-05,
+      "loss": 0.2362,
+      "step": 933
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 9.976166472659387e-05,
+      "loss": 0.2458,
+      "step": 934
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 9.976115450640578e-05,
+      "loss": 0.2562,
+      "step": 935
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 9.976064374197782e-05,
+      "loss": 0.2562,
+      "step": 936
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 9.97601324333156e-05,
+      "loss": 0.2438,
+      "step": 937
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 9.975962058042471e-05,
+      "loss": 0.2181,
+      "step": 938
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 9.975910818331076e-05,
+      "loss": 0.2266,
+      "step": 939
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 9.975859524197933e-05,
+      "loss": 0.24,
+      "step": 940
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 9.975808175643606e-05,
+      "loss": 0.2415,
+      "step": 941
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 9.975756772668653e-05,
+      "loss": 0.2401,
+      "step": 942
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 9.975705315273638e-05,
+      "loss": 0.2599,
+      "step": 943
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 9.975653803459126e-05,
+      "loss": 0.2407,
+      "step": 944
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 9.975602237225677e-05,
+      "loss": 0.2631,
+      "step": 945
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 9.975550616573856e-05,
+      "loss": 0.261,
+      "step": 946
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 9.97549894150423e-05,
+      "loss": 0.2505,
+      "step": 947
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 9.97544721201736e-05,
+      "loss": 0.2476,
+      "step": 948
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 9.975395428113814e-05,
+      "loss": 0.252,
+      "step": 949
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 9.975343589794158e-05,
+      "loss": 0.2446,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "eval_accuracy": 0.8872487621009751,
+      "eval_f1": 0.9124840995820461,
+      "eval_loss": 0.2636834979057312,
+      "eval_runtime": 110.5736,
+      "eval_samples_per_second": 772.58,
+      "eval_steps_per_second": 0.76,
+      "step": 950
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 9.975291697058959e-05,
+      "loss": 0.2389,
+      "step": 951
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 9.975239749908786e-05,
+      "loss": 0.2534,
+      "step": 952
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 9.975187748344206e-05,
+      "loss": 0.2591,
+      "step": 953
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 9.975135692365785e-05,
+      "loss": 0.2338,
+      "step": 954
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 9.975083581974097e-05,
+      "loss": 0.2119,
+      "step": 955
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 9.97503141716971e-05,
+      "loss": 0.2416,
+      "step": 956
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 9.974979197953193e-05,
+      "loss": 0.2308,
+      "step": 957
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 9.974926924325121e-05,
+      "loss": 0.2477,
+      "step": 958
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 9.974874596286061e-05,
+      "loss": 0.2289,
+      "step": 959
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 9.974822213836588e-05,
+      "loss": 0.2367,
+      "step": 960
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 9.974769776977274e-05,
+      "loss": 0.2624,
+      "step": 961
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 9.974717285708694e-05,
+      "loss": 0.2641,
+      "step": 962
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 9.97466474003142e-05,
+      "loss": 0.2285,
+      "step": 963
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 9.974612139946029e-05,
+      "loss": 0.224,
+      "step": 964
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 9.974559485453095e-05,
+      "loss": 0.2338,
+      "step": 965
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 9.974506776553192e-05,
+      "loss": 0.2581,
+      "step": 966
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 9.9744540132469e-05,
+      "loss": 0.2451,
+      "step": 967
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 9.974401195534795e-05,
+      "loss": 0.2207,
+      "step": 968
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 9.974348323417453e-05,
+      "loss": 0.2231,
+      "step": 969
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 9.974295396895453e-05,
+      "loss": 0.2508,
+      "step": 970
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 9.974242415969375e-05,
+      "loss": 0.2724,
+      "step": 971
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 9.974189380639797e-05,
+      "loss": 0.2587,
+      "step": 972
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 9.974136290907301e-05,
+      "loss": 0.2411,
+      "step": 973
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 9.974083146772466e-05,
+      "loss": 0.2349,
+      "step": 974
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 9.974029948235873e-05,
+      "loss": 0.2685,
+      "step": 975
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 9.973976695298107e-05,
+      "loss": 0.2439,
+      "step": 976
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 9.973923387959745e-05,
+      "loss": 0.2431,
+      "step": 977
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 9.973870026221372e-05,
+      "loss": 0.2378,
+      "step": 978
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 9.973816610083575e-05,
+      "loss": 0.239,
+      "step": 979
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 9.973763139546935e-05,
+      "loss": 0.2442,
+      "step": 980
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 9.973709614612039e-05,
+      "loss": 0.2276,
+      "step": 981
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 9.973656035279469e-05,
+      "loss": 0.2446,
+      "step": 982
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 9.973602401549814e-05,
+      "loss": 0.2325,
+      "step": 983
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 9.973548713423659e-05,
+      "loss": 0.2483,
+      "step": 984
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 9.973494970901593e-05,
+      "loss": 0.2316,
+      "step": 985
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 9.973441173984201e-05,
+      "loss": 0.2353,
+      "step": 986
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 9.973387322672074e-05,
+      "loss": 0.2525,
+      "step": 987
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 9.9733334169658e-05,
+      "loss": 0.2406,
+      "step": 988
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 9.973279456865967e-05,
+      "loss": 0.2308,
+      "step": 989
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 9.973225442373169e-05,
+      "loss": 0.2547,
+      "step": 990
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 9.973171373487991e-05,
+      "loss": 0.2306,
+      "step": 991
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 9.97311725021103e-05,
+      "loss": 0.2457,
+      "step": 992
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 9.973063072542876e-05,
+      "loss": 0.2358,
+      "step": 993
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 9.973008840484121e-05,
+      "loss": 0.2756,
+      "step": 994
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 9.972954554035358e-05,
+      "loss": 0.2412,
+      "step": 995
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 9.972900213197181e-05,
+      "loss": 0.2586,
+      "step": 996
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 9.972845817970184e-05,
+      "loss": 0.2294,
+      "step": 997
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 9.972791368354964e-05,
+      "loss": 0.2303,
+      "step": 998
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 9.972736864352115e-05,
+      "loss": 0.2123,
+      "step": 999
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 9.972682305962232e-05,
+      "loss": 0.2294,
+      "step": 1000
+    },
+    {
+      "epoch": 1.33,
+      "eval_accuracy": 0.889917707516359,
+      "eval_f1": 0.9145246318851118,
+      "eval_loss": 0.2669387459754944,
+      "eval_runtime": 111.8933,
+      "eval_samples_per_second": 763.468,
+      "eval_steps_per_second": 0.751,
+      "step": 1000
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 9.972627693185912e-05,
+      "loss": 0.2616,
+      "step": 1001
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 9.972573026023753e-05,
+      "loss": 0.2616,
+      "step": 1002
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 9.972518304476355e-05,
+      "loss": 0.2166,
+      "step": 1003
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 9.972463528544312e-05,
+      "loss": 0.2215,
+      "step": 1004
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 9.972408698228227e-05,
+      "loss": 0.2289,
+      "step": 1005
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 9.972353813528696e-05,
+      "loss": 0.2448,
+      "step": 1006
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 9.972298874446324e-05,
+      "loss": 0.2178,
+      "step": 1007
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 9.972243880981709e-05,
+      "loss": 0.2288,
+      "step": 1008
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 9.972188833135452e-05,
+      "loss": 0.2919,
+      "step": 1009
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 9.972133730908155e-05,
+      "loss": 0.2251,
+      "step": 1010
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.972078574300423e-05,
+      "loss": 0.281,
+      "step": 1011
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.972023363312858e-05,
+      "loss": 0.2404,
+      "step": 1012
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.971968097946063e-05,
+      "loss": 0.2527,
+      "step": 1013
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.971912778200642e-05,
+      "loss": 0.2103,
+      "step": 1014
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.971857404077203e-05,
+      "loss": 0.2495,
+      "step": 1015
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.971801975576348e-05,
+      "loss": 0.2745,
+      "step": 1016
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.971746492698686e-05,
+      "loss": 0.2607,
+      "step": 1017
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 9.971690955444823e-05,
+      "loss": 0.2245,
+      "step": 1018
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 9.971635363815365e-05,
+      "loss": 0.2726,
+      "step": 1019
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 9.971579717810921e-05,
+      "loss": 0.2547,
+      "step": 1020
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 9.971524017432101e-05,
+      "loss": 0.2508,
+      "step": 1021
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 9.971468262679512e-05,
+      "loss": 0.2666,
+      "step": 1022
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 9.971412453553766e-05,
+      "loss": 0.2657,
+      "step": 1023
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 9.97135659005547e-05,
+      "loss": 0.2513,
+      "step": 1024
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 9.971300672185239e-05,
+      "loss": 0.2697,
+      "step": 1025
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 9.971244699943682e-05,
+      "loss": 0.2536,
+      "step": 1026
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 9.971188673331412e-05,
+      "loss": 0.2341,
+      "step": 1027
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 9.971132592349042e-05,
+      "loss": 0.2398,
+      "step": 1028
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 9.971076456997183e-05,
+      "loss": 0.2279,
+      "step": 1029
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 9.971020267276454e-05,
+      "loss": 0.2761,
+      "step": 1030
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 9.970964023187464e-05,
+      "loss": 0.2387,
+      "step": 1031
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 9.970907724730831e-05,
+      "loss": 0.2134,
+      "step": 1032
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 9.970851371907173e-05,
+      "loss": 0.2572,
+      "step": 1033
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 9.970794964717101e-05,
+      "loss": 0.244,
+      "step": 1034
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 9.970738503161235e-05,
+      "loss": 0.2739,
+      "step": 1035
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 9.970681987240193e-05,
+      "loss": 0.2525,
+      "step": 1036
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 9.970625416954591e-05,
+      "loss": 0.2337,
+      "step": 1037
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 9.970568792305051e-05,
+      "loss": 0.2317,
+      "step": 1038
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 9.970512113292187e-05,
+      "loss": 0.266,
+      "step": 1039
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 9.970455379916625e-05,
+      "loss": 0.2392,
+      "step": 1040
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 9.970398592178982e-05,
+      "loss": 0.2536,
+      "step": 1041
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 9.97034175007988e-05,
+      "loss": 0.2735,
+      "step": 1042
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 9.97028485361994e-05,
+      "loss": 0.2699,
+      "step": 1043
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 9.970227902799785e-05,
+      "loss": 0.2358,
+      "step": 1044
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 9.970170897620038e-05,
+      "loss": 0.2448,
+      "step": 1045
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 9.970113838081322e-05,
+      "loss": 0.2494,
+      "step": 1046
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 9.97005672418426e-05,
+      "loss": 0.2821,
+      "step": 1047
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9.969999555929479e-05,
+      "loss": 0.2335,
+      "step": 1048
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9.969942333317603e-05,
+      "loss": 0.2563,
+      "step": 1049
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9.969885056349257e-05,
+      "loss": 0.2424,
+      "step": 1050
+    },
+    {
+      "epoch": 1.4,
+      "eval_accuracy": 0.8908541795919322,
+      "eval_f1": 0.9154669084315504,
+      "eval_loss": 0.2650482654571533,
+      "eval_runtime": 110.6683,
+      "eval_samples_per_second": 771.92,
+      "eval_steps_per_second": 0.759,
+      "step": 1050
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9.969827725025069e-05,
+      "loss": 0.2702,
+      "step": 1051
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9.969770339345666e-05,
+      "loss": 0.2409,
+      "step": 1052
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9.969712899311674e-05,
+      "loss": 0.2606,
+      "step": 1053
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9.969655404923724e-05,
+      "loss": 0.2527,
+      "step": 1054
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9.96959785618244e-05,
+      "loss": 0.2213,
+      "step": 1055
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 9.969540253088456e-05,
+      "loss": 0.2777,
+      "step": 1056
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 9.969482595642401e-05,
+      "loss": 0.2701,
+      "step": 1057
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 9.969424883844904e-05,
+      "loss": 0.249,
+      "step": 1058
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 9.969367117696598e-05,
+      "loss": 0.2348,
+      "step": 1059
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 9.969309297198116e-05,
+      "loss": 0.2414,
+      "step": 1060
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 9.969251422350085e-05,
+      "loss": 0.2434,
+      "step": 1061
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 9.969193493153144e-05,
+      "loss": 0.2315,
+      "step": 1062
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.969135509607921e-05,
+      "loss": 0.2706,
+      "step": 1063
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.969077471715055e-05,
+      "loss": 0.2291,
+      "step": 1064
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.96901937947518e-05,
+      "loss": 0.2525,
+      "step": 1065
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.96896123288893e-05,
+      "loss": 0.246,
+      "step": 1066
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.968903031956939e-05,
+      "loss": 0.25,
+      "step": 1067
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.968844776679849e-05,
+      "loss": 0.2281,
+      "step": 1068
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.96878646705829e-05,
+      "loss": 0.225,
+      "step": 1069
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.968728103092907e-05,
+      "loss": 0.2448,
+      "step": 1070
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 9.968669684784334e-05,
+      "loss": 0.2259,
+      "step": 1071
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 9.968611212133211e-05,
+      "loss": 0.2468,
+      "step": 1072
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 9.968552685140178e-05,
+      "loss": 0.2485,
+      "step": 1073
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 9.968494103805872e-05,
+      "loss": 0.2665,
+      "step": 1074
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 9.968435468130939e-05,
+      "loss": 0.2635,
+      "step": 1075
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 9.968376778116017e-05,
+      "loss": 0.227,
+      "step": 1076
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 9.968318033761748e-05,
+      "loss": 0.2495,
+      "step": 1077
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.968259235068775e-05,
+      "loss": 0.2475,
+      "step": 1078
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.96820038203774e-05,
+      "loss": 0.2625,
+      "step": 1079
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.968141474669289e-05,
+      "loss": 0.2423,
+      "step": 1080
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.968082512964066e-05,
+      "loss": 0.2354,
+      "step": 1081
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.968023496922712e-05,
+      "loss": 0.2409,
+      "step": 1082
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.967964426545878e-05,
+      "loss": 0.231,
+      "step": 1083
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.967905301834206e-05,
+      "loss": 0.244,
+      "step": 1084
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.967846122788343e-05,
+      "loss": 0.2595,
+      "step": 1085
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 9.967786889408938e-05,
+      "loss": 0.2251,
+      "step": 1086
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 9.967727601696639e-05,
+      "loss": 0.2512,
+      "step": 1087
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 9.967668259652093e-05,
+      "loss": 0.271,
+      "step": 1088
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 9.96760886327595e-05,
+      "loss": 0.227,
+      "step": 1089
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 9.967549412568858e-05,
+      "loss": 0.2488,
+      "step": 1090
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 9.96748990753147e-05,
+      "loss": 0.2662,
+      "step": 1091
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 9.967430348164434e-05,
+      "loss": 0.2102,
+      "step": 1092
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 9.967370734468403e-05,
+      "loss": 0.2522,
+      "step": 1093
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 9.967311066444028e-05,
+      "loss": 0.2721,
+      "step": 1094
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 9.967251344091964e-05,
+      "loss": 0.2714,
+      "step": 1095
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 9.96719156741286e-05,
+      "loss": 0.2512,
+      "step": 1096
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 9.967131736407373e-05,
+      "loss": 0.2278,
+      "step": 1097
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 9.967071851076158e-05,
+      "loss": 0.2355,
+      "step": 1098
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 9.967011911419866e-05,
+      "loss": 0.2386,
+      "step": 1099
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 9.966951917439156e-05,
+      "loss": 0.2671,
+      "step": 1100
+    },
+    {
+      "epoch": 1.46,
+      "eval_accuracy": 0.890514708464537,
+      "eval_f1": 0.9135046655507569,
+      "eval_loss": 0.26096296310424805,
+      "eval_runtime": 110.3878,
+      "eval_samples_per_second": 773.881,
+      "eval_steps_per_second": 0.761,
+      "step": 1100
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 9.966891869134684e-05,
+      "loss": 0.2563,
+      "step": 1101
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 9.966831766507106e-05,
+      "loss": 0.2501,
+      "step": 1102
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 9.966771609557077e-05,
+      "loss": 0.2416,
+      "step": 1103
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 9.96671139828526e-05,
+      "loss": 0.2494,
+      "step": 1104
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 9.96665113269231e-05,
+      "loss": 0.2368,
+      "step": 1105
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 9.966590812778887e-05,
+      "loss": 0.26,
+      "step": 1106
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 9.96653043854565e-05,
+      "loss": 0.2432,
+      "step": 1107
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 9.966470009993259e-05,
+      "loss": 0.2315,
+      "step": 1108
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 9.966409527122378e-05,
+      "loss": 0.2451,
+      "step": 1109
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 9.966348989933664e-05,
+      "loss": 0.2716,
+      "step": 1110
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 9.966288398427785e-05,
+      "loss": 0.2326,
+      "step": 1111
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 9.966227752605397e-05,
+      "loss": 0.2489,
+      "step": 1112
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 9.966167052467168e-05,
+      "loss": 0.2356,
+      "step": 1113
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 9.96610629801376e-05,
+      "loss": 0.2494,
+      "step": 1114
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 9.966045489245837e-05,
+      "loss": 0.2477,
+      "step": 1115
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 9.965984626164065e-05,
+      "loss": 0.242,
+      "step": 1116
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 9.96592370876911e-05,
+      "loss": 0.2734,
+      "step": 1117
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 9.965862737061638e-05,
+      "loss": 0.2544,
+      "step": 1118
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 9.965801711042315e-05,
+      "loss": 0.2283,
+      "step": 1119
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 9.965740630711809e-05,
+      "loss": 0.2621,
+      "step": 1120
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 9.96567949607079e-05,
+      "loss": 0.2194,
+      "step": 1121
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 9.965618307119921e-05,
+      "loss": 0.2523,
+      "step": 1122
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.965557063859879e-05,
+      "loss": 0.2517,
+      "step": 1123
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.965495766291328e-05,
+      "loss": 0.2346,
+      "step": 1124
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.965434414414941e-05,
+      "loss": 0.2761,
+      "step": 1125
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.965373008231386e-05,
+      "loss": 0.2339,
+      "step": 1126
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.965311547741339e-05,
+      "loss": 0.2521,
+      "step": 1127
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.965250032945468e-05,
+      "loss": 0.2101,
+      "step": 1128
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.965188463844449e-05,
+      "loss": 0.2474,
+      "step": 1129
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.965126840438955e-05,
+      "loss": 0.2464,
+      "step": 1130
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.965065162729658e-05,
+      "loss": 0.2351,
+      "step": 1131
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.965003430717235e-05,
+      "loss": 0.2459,
+      "step": 1132
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.964941644402357e-05,
+      "loss": 0.2558,
+      "step": 1133
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.964879803785705e-05,
+      "loss": 0.2234,
+      "step": 1134
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.964817908867953e-05,
+      "loss": 0.2385,
+      "step": 1135
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.964755959649778e-05,
+      "loss": 0.2435,
+      "step": 1136
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.964693956131856e-05,
+      "loss": 0.2482,
+      "step": 1137
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.964631898314868e-05,
+      "loss": 0.2438,
+      "step": 1138
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.964569786199492e-05,
+      "loss": 0.2384,
+      "step": 1139
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.964507619786406e-05,
+      "loss": 0.2317,
+      "step": 1140
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.96444539907629e-05,
+      "loss": 0.2313,
+      "step": 1141
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.964383124069824e-05,
+      "loss": 0.2304,
+      "step": 1142
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.964320794767692e-05,
+      "loss": 0.2472,
+      "step": 1143
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.964258411170573e-05,
+      "loss": 0.2503,
+      "step": 1144
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.96419597327915e-05,
+      "loss": 0.2448,
+      "step": 1145
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.964133481094109e-05,
+      "loss": 0.2588,
+      "step": 1146
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.964070934616128e-05,
+      "loss": 0.2372,
+      "step": 1147
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.964008333845894e-05,
+      "loss": 0.2453,
+      "step": 1148
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.96394567878409e-05,
+      "loss": 0.2479,
+      "step": 1149
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.963882969431404e-05,
+      "loss": 0.2262,
+      "step": 1150
+    },
+    {
+      "epoch": 1.53,
+      "eval_accuracy": 0.8908775913938216,
+      "eval_f1": 0.915916511825086,
+      "eval_loss": 0.2618601322174072,
+      "eval_runtime": 110.7143,
+      "eval_samples_per_second": 771.599,
+      "eval_steps_per_second": 0.759,
+      "step": 1150
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.96382020578852e-05,
+      "loss": 0.2001,
+      "step": 1151
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.963757387856126e-05,
+      "loss": 0.2165,
+      "step": 1152
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.963694515634907e-05,
+      "loss": 0.2142,
+      "step": 1153
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.963631589125553e-05,
+      "loss": 0.2625,
+      "step": 1154
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.963568608328748e-05,
+      "loss": 0.247,
+      "step": 1155
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.963505573245186e-05,
+      "loss": 0.2464,
+      "step": 1156
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.963442483875553e-05,
+      "loss": 0.2572,
+      "step": 1157
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.96337934022054e-05,
+      "loss": 0.2442,
+      "step": 1158
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.963316142280837e-05,
+      "loss": 0.2475,
+      "step": 1159
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 9.963252890057138e-05,
+      "loss": 0.2389,
+      "step": 1160
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.96318958355013e-05,
+      "loss": 0.2165,
+      "step": 1161
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.963126222760509e-05,
+      "loss": 0.262,
+      "step": 1162
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.963062807688967e-05,
+      "loss": 0.2507,
+      "step": 1163
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.962999338336198e-05,
+      "loss": 0.262,
+      "step": 1164
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.962935814702897e-05,
+      "loss": 0.2243,
+      "step": 1165
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.962872236789756e-05,
+      "loss": 0.2451,
+      "step": 1166
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.962808604597471e-05,
+      "loss": 0.2521,
+      "step": 1167
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.962744918126739e-05,
+      "loss": 0.2773,
+      "step": 1168
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.962681177378256e-05,
+      "loss": 0.2422,
+      "step": 1169
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.962617382352721e-05,
+      "loss": 0.2487,
+      "step": 1170
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.96255353305083e-05,
+      "loss": 0.2625,
+      "step": 1171
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.962489629473282e-05,
+      "loss": 0.2536,
+      "step": 1172
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.962425671620775e-05,
+      "loss": 0.223,
+      "step": 1173
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.962361659494008e-05,
+      "loss": 0.2354,
+      "step": 1174
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.962297593093683e-05,
+      "loss": 0.2313,
+      "step": 1175
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.9622334724205e-05,
+      "loss": 0.2545,
+      "step": 1176
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.962169297475158e-05,
+      "loss": 0.248,
+      "step": 1177
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.962105068258363e-05,
+      "loss": 0.2292,
+      "step": 1178
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.962040784770815e-05,
+      "loss": 0.2439,
+      "step": 1179
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.961976447013217e-05,
+      "loss": 0.2493,
+      "step": 1180
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.961912054986273e-05,
+      "loss": 0.2447,
+      "step": 1181
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.961847608690687e-05,
+      "loss": 0.2341,
+      "step": 1182
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.961783108127165e-05,
+      "loss": 0.2319,
+      "step": 1183
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.961718553296413e-05,
+      "loss": 0.2285,
+      "step": 1184
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.961653944199134e-05,
+      "loss": 0.2386,
+      "step": 1185
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.961589280836036e-05,
+      "loss": 0.2279,
+      "step": 1186
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.961524563207828e-05,
+      "loss": 0.2436,
+      "step": 1187
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.961459791315215e-05,
+      "loss": 0.2542,
+      "step": 1188
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.961394965158908e-05,
+      "loss": 0.2681,
+      "step": 1189
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.961330084739615e-05,
+      "loss": 0.2261,
+      "step": 1190
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.961265150058046e-05,
+      "loss": 0.2294,
+      "step": 1191
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.961200161114909e-05,
+      "loss": 0.2382,
+      "step": 1192
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.961135117910915e-05,
+      "loss": 0.2603,
+      "step": 1193
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.96107002044678e-05,
+      "loss": 0.2694,
+      "step": 1194
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.961004868723212e-05,
+      "loss": 0.2344,
+      "step": 1195
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.960939662740923e-05,
+      "loss": 0.2638,
+      "step": 1196
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.960874402500629e-05,
+      "loss": 0.2646,
+      "step": 1197
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.960809088003042e-05,
+      "loss": 0.2305,
+      "step": 1198
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.960743719248875e-05,
+      "loss": 0.2537,
+      "step": 1199
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.960678296238845e-05,
+      "loss": 0.2601,
+      "step": 1200
+    },
+    {
+      "epoch": 1.6,
+      "eval_accuracy": 0.890163531436197,
+      "eval_f1": 0.9122058479532164,
+      "eval_loss": 0.26000529527664185,
+      "eval_runtime": 110.4089,
+      "eval_samples_per_second": 773.733,
+      "eval_steps_per_second": 0.761,
+      "step": 1200
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.960612818973669e-05,
+      "loss": 0.226,
+      "step": 1201
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.960547287454061e-05,
+      "loss": 0.2423,
+      "step": 1202
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.960481701680736e-05,
+      "loss": 0.2616,
+      "step": 1203
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.960416061654414e-05,
+      "loss": 0.2613,
+      "step": 1204
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.960350367375813e-05,
+      "loss": 0.2361,
+      "step": 1205
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.96028461884565e-05,
+      "loss": 0.2731,
+      "step": 1206
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.960218816064643e-05,
+      "loss": 0.2502,
+      "step": 1207
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.960152959033515e-05,
+      "loss": 0.2515,
+      "step": 1208
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.960087047752986e-05,
+      "loss": 0.2054,
+      "step": 1209
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.960021082223774e-05,
+      "loss": 0.2374,
+      "step": 1210
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.959955062446602e-05,
+      "loss": 0.2534,
+      "step": 1211
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.959888988422192e-05,
+      "loss": 0.2522,
+      "step": 1212
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.959822860151268e-05,
+      "loss": 0.2526,
+      "step": 1213
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.959756677634552e-05,
+      "loss": 0.2334,
+      "step": 1214
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.959690440872766e-05,
+      "loss": 0.2526,
+      "step": 1215
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.959624149866637e-05,
+      "loss": 0.229,
+      "step": 1216
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.95955780461689e-05,
+      "loss": 0.2363,
+      "step": 1217
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.95949140512425e-05,
+      "loss": 0.2507,
+      "step": 1218
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.959424951389443e-05,
+      "loss": 0.2646,
+      "step": 1219
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.959358443413196e-05,
+      "loss": 0.2315,
+      "step": 1220
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.959291881196236e-05,
+      "loss": 0.2527,
+      "step": 1221
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.959225264739292e-05,
+      "loss": 0.2453,
+      "step": 1222
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.959158594043092e-05,
+      "loss": 0.2272,
+      "step": 1223
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.959091869108363e-05,
+      "loss": 0.2288,
+      "step": 1224
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.959025089935839e-05,
+      "loss": 0.2488,
+      "step": 1225
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.958958256526247e-05,
+      "loss": 0.2264,
+      "step": 1226
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.958891368880322e-05,
+      "loss": 0.2233,
+      "step": 1227
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.95882442699879e-05,
+      "loss": 0.2105,
+      "step": 1228
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.958757430882387e-05,
+      "loss": 0.2613,
+      "step": 1229
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.958690380531843e-05,
+      "loss": 0.2548,
+      "step": 1230
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.958623275947895e-05,
+      "loss": 0.2401,
+      "step": 1231
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.958556117131274e-05,
+      "loss": 0.2611,
+      "step": 1232
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.958488904082715e-05,
+      "loss": 0.2293,
+      "step": 1233
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.958421636802955e-05,
+      "loss": 0.2553,
+      "step": 1234
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.958354315292726e-05,
+      "loss": 0.2392,
+      "step": 1235
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.958286939552769e-05,
+      "loss": 0.2369,
+      "step": 1236
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.958219509583818e-05,
+      "loss": 0.2543,
+      "step": 1237
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.958152025386609e-05,
+      "loss": 0.2521,
+      "step": 1238
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.958084486961882e-05,
+      "loss": 0.2535,
+      "step": 1239
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.958016894310377e-05,
+      "loss": 0.2666,
+      "step": 1240
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.957949247432834e-05,
+      "loss": 0.238,
+      "step": 1241
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 9.957881546329987e-05,
+      "loss": 0.2479,
+      "step": 1242
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 9.957813791002583e-05,
+      "loss": 0.2487,
+      "step": 1243
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 9.957745981451358e-05,
+      "loss": 0.2412,
+      "step": 1244
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 9.957678117677058e-05,
+      "loss": 0.2484,
+      "step": 1245
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 9.957610199680423e-05,
+      "loss": 0.2582,
+      "step": 1246
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 9.957542227462195e-05,
+      "loss": 0.2423,
+      "step": 1247
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 9.957474201023118e-05,
+      "loss": 0.2328,
+      "step": 1248
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 9.957406120363937e-05,
+      "loss": 0.2439,
+      "step": 1249
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 9.957337985485397e-05,
+      "loss": 0.2461,
+      "step": 1250
+    },
+    {
+      "epoch": 1.66,
+      "eval_accuracy": 0.8925515352289088,
+      "eval_f1": 0.9158530660139527,
+      "eval_loss": 0.259185254573822,
+      "eval_runtime": 110.3049,
+      "eval_samples_per_second": 774.463,
+      "eval_steps_per_second": 0.762,
+      "step": 1250
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 9.95726979638824e-05,
+      "loss": 0.2612,
+      "step": 1251
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 9.957201553073217e-05,
+      "loss": 0.2644,
+      "step": 1252
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 9.957133255541069e-05,
+      "loss": 0.2361,
+      "step": 1253
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 9.957064903792545e-05,
+      "loss": 0.2541,
+      "step": 1254
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 9.956996497828395e-05,
+      "loss": 0.2378,
+      "step": 1255
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 9.956928037649362e-05,
+      "loss": 0.2105,
+      "step": 1256
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 9.956859523256202e-05,
+      "loss": 0.2587,
+      "step": 1257
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 9.956790954649657e-05,
+      "loss": 0.2782,
+      "step": 1258
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 9.95672233183048e-05,
+      "loss": 0.2477,
+      "step": 1259
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 9.956653654799423e-05,
+      "loss": 0.2594,
+      "step": 1260
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 9.956584923557235e-05,
+      "loss": 0.2348,
+      "step": 1261
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 9.95651613810467e-05,
+      "loss": 0.2069,
+      "step": 1262
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 9.956447298442476e-05,
+      "loss": 0.2234,
+      "step": 1263
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 9.95637840457141e-05,
+      "loss": 0.2503,
+      "step": 1264
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 9.956309456492225e-05,
+      "loss": 0.2486,
+      "step": 1265
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 9.956240454205673e-05,
+      "loss": 0.2363,
+      "step": 1266
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 9.956171397712511e-05,
+      "loss": 0.2396,
+      "step": 1267
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 9.956102287013492e-05,
+      "loss": 0.2401,
+      "step": 1268
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 9.956033122109374e-05,
+      "loss": 0.2566,
+      "step": 1269
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 9.955963903000914e-05,
+      "loss": 0.2829,
+      "step": 1270
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 9.955894629688864e-05,
+      "loss": 0.246,
+      "step": 1271
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 9.955825302173988e-05,
+      "loss": 0.2503,
+      "step": 1272
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.95575592045704e-05,
+      "loss": 0.2357,
+      "step": 1273
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.95568648453878e-05,
+      "loss": 0.2436,
+      "step": 1274
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.95561699441997e-05,
+      "loss": 0.2331,
+      "step": 1275
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.955547450101365e-05,
+      "loss": 0.2157,
+      "step": 1276
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.955477851583732e-05,
+      "loss": 0.2393,
+      "step": 1277
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.955408198867825e-05,
+      "loss": 0.2369,
+      "step": 1278
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.955338491954411e-05,
+      "loss": 0.2529,
+      "step": 1279
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 9.95526873084425e-05,
+      "loss": 0.272,
+      "step": 1280
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.955198915538107e-05,
+      "loss": 0.2464,
+      "step": 1281
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.955129046036742e-05,
+      "loss": 0.2574,
+      "step": 1282
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.955059122340924e-05,
+      "loss": 0.2645,
+      "step": 1283
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.954989144451413e-05,
+      "loss": 0.2451,
+      "step": 1284
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.954919112368978e-05,
+      "loss": 0.2315,
+      "step": 1285
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.954849026094384e-05,
+      "loss": 0.2448,
+      "step": 1286
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.954778885628395e-05,
+      "loss": 0.2353,
+      "step": 1287
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.954708690971781e-05,
+      "loss": 0.2447,
+      "step": 1288
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.95463844212531e-05,
+      "loss": 0.2125,
+      "step": 1289
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.954568139089748e-05,
+      "loss": 0.2297,
+      "step": 1290
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.954497781865865e-05,
+      "loss": 0.2782,
+      "step": 1291
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.954427370454431e-05,
+      "loss": 0.2059,
+      "step": 1292
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.954356904856217e-05,
+      "loss": 0.2714,
+      "step": 1293
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.954286385071991e-05,
+      "loss": 0.236,
+      "step": 1294
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 9.954215811102525e-05,
+      "loss": 0.2403,
+      "step": 1295
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 9.954145182948594e-05,
+      "loss": 0.2861,
+      "step": 1296
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 9.954074500610966e-05,
+      "loss": 0.2423,
+      "step": 1297
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 9.954003764090416e-05,
+      "loss": 0.2457,
+      "step": 1298
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 9.953932973387719e-05,
+      "loss": 0.2358,
+      "step": 1299
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 9.953862128503646e-05,
+      "loss": 0.2485,
+      "step": 1300
+    },
+    {
+      "epoch": 1.73,
+      "eval_accuracy": 0.8932304774836995,
+      "eval_f1": 0.9162404150787455,
+      "eval_loss": 0.25554707646369934,
+      "eval_runtime": 110.4111,
+      "eval_samples_per_second": 773.717,
+      "eval_steps_per_second": 0.761,
+      "step": 1300
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 9.953791229438975e-05,
+      "loss": 0.2338,
+      "step": 1301
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 9.95372027619448e-05,
+      "loss": 0.2662,
+      "step": 1302
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 9.953649268770939e-05,
+      "loss": 0.2397,
+      "step": 1303
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 9.953578207169125e-05,
+      "loss": 0.2366,
+      "step": 1304
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 9.953507091389818e-05,
+      "loss": 0.2094,
+      "step": 1305
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 9.953435921433793e-05,
+      "loss": 0.2316,
+      "step": 1306
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 9.953364697301831e-05,
+      "loss": 0.2818,
+      "step": 1307
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 9.953293418994711e-05,
+      "loss": 0.2305,
+      "step": 1308
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 9.953222086513211e-05,
+      "loss": 0.2745,
+      "step": 1309
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 9.953150699858113e-05,
+      "loss": 0.2513,
+      "step": 1310
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 9.953079259030196e-05,
+      "loss": 0.236,
+      "step": 1311
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 9.953007764030242e-05,
+      "loss": 0.254,
+      "step": 1312
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 9.952936214859032e-05,
+      "loss": 0.2158,
+      "step": 1313
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 9.952864611517352e-05,
+      "loss": 0.2434,
+      "step": 1314
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 9.95279295400598e-05,
+      "loss": 0.2348,
+      "step": 1315
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 9.952721242325704e-05,
+      "loss": 0.2276,
+      "step": 1316
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 9.952649476477306e-05,
+      "loss": 0.2132,
+      "step": 1317
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 9.952577656461572e-05,
+      "loss": 0.244,
+      "step": 1318
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 9.952505782279289e-05,
+      "loss": 0.2533,
+      "step": 1319
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 9.952433853931238e-05,
+      "loss": 0.2428,
+      "step": 1320
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 9.95236187141821e-05,
+      "loss": 0.2326,
+      "step": 1321
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 9.952289834740992e-05,
+      "loss": 0.2051,
+      "step": 1322
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 9.95221774390037e-05,
+      "loss": 0.2221,
+      "step": 1323
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 9.952145598897134e-05,
+      "loss": 0.2769,
+      "step": 1324
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 9.952073399732072e-05,
+      "loss": 0.2393,
+      "step": 1325
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 9.952001146405976e-05,
+      "loss": 0.2211,
+      "step": 1326
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 9.951928838919633e-05,
+      "loss": 0.2259,
+      "step": 1327
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 9.951856477273836e-05,
+      "loss": 0.2143,
+      "step": 1328
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 9.951784061469374e-05,
+      "loss": 0.2391,
+      "step": 1329
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 9.951711591507042e-05,
+      "loss": 0.2853,
+      "step": 1330
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 9.95163906738763e-05,
+      "loss": 0.2275,
+      "step": 1331
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 9.951566489111935e-05,
+      "loss": 0.2352,
+      "step": 1332
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 9.951493856680745e-05,
+      "loss": 0.2243,
+      "step": 1333
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 9.95142117009486e-05,
+      "loss": 0.2395,
+      "step": 1334
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 9.951348429355072e-05,
+      "loss": 0.2355,
+      "step": 1335
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 9.951275634462177e-05,
+      "loss": 0.2323,
+      "step": 1336
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 9.951202785416971e-05,
+      "loss": 0.2681,
+      "step": 1337
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 9.951129882220253e-05,
+      "loss": 0.2439,
+      "step": 1338
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 9.951056924872817e-05,
+      "loss": 0.2183,
+      "step": 1339
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 9.950983913375463e-05,
+      "loss": 0.263,
+      "step": 1340
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 9.950910847728988e-05,
+      "loss": 0.2295,
+      "step": 1341
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 9.950837727934192e-05,
+      "loss": 0.2348,
+      "step": 1342
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 9.950764553991876e-05,
+      "loss": 0.263,
+      "step": 1343
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 9.950691325902838e-05,
+      "loss": 0.2325,
+      "step": 1344
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 9.950618043667881e-05,
+      "loss": 0.2445,
+      "step": 1345
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 9.950544707287806e-05,
+      "loss": 0.206,
+      "step": 1346
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 9.950471316763412e-05,
+      "loss": 0.2391,
+      "step": 1347
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 9.950397872095507e-05,
+      "loss": 0.2485,
+      "step": 1348
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 9.95032437328489e-05,
+      "loss": 0.2365,
+      "step": 1349
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 9.950250820332367e-05,
+      "loss": 0.2245,
+      "step": 1350
+    },
+    {
+      "epoch": 1.8,
+      "eval_accuracy": 0.8936635958186522,
+      "eval_f1": 0.9175201568969275,
+      "eval_loss": 0.2583000957965851,
+      "eval_runtime": 110.1261,
+      "eval_samples_per_second": 775.72,
+      "eval_steps_per_second": 0.763,
+      "step": 1350
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 9.95017721323874e-05,
+      "loss": 0.2236,
+      "step": 1351
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 9.950103552004819e-05,
+      "loss": 0.2485,
+      "step": 1352
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 9.950029836631404e-05,
+      "loss": 0.2398,
+      "step": 1353
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 9.949956067119304e-05,
+      "loss": 0.2705,
+      "step": 1354
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 9.949882243469327e-05,
+      "loss": 0.2573,
+      "step": 1355
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 9.949808365682278e-05,
+      "loss": 0.275,
+      "step": 1356
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 9.949734433758966e-05,
+      "loss": 0.2539,
+      "step": 1357
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 9.949660447700199e-05,
+      "loss": 0.2659,
+      "step": 1358
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 9.949586407506788e-05,
+      "loss": 0.2315,
+      "step": 1359
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 9.94951231317954e-05,
+      "loss": 0.2505,
+      "step": 1360
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 9.94943816471927e-05,
+      "loss": 0.2179,
+      "step": 1361
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 9.949363962126783e-05,
+      "loss": 0.2447,
+      "step": 1362
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 9.949289705402893e-05,
+      "loss": 0.2355,
+      "step": 1363
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 9.949215394548414e-05,
+      "loss": 0.2556,
+      "step": 1364
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 9.949141029564158e-05,
+      "loss": 0.2169,
+      "step": 1365
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 9.949066610450936e-05,
+      "loss": 0.2345,
+      "step": 1366
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 9.948992137209563e-05,
+      "loss": 0.2138,
+      "step": 1367
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 9.948917609840856e-05,
+      "loss": 0.2217,
+      "step": 1368
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 9.948843028345627e-05,
+      "loss": 0.2641,
+      "step": 1369
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 9.948768392724693e-05,
+      "loss": 0.2457,
+      "step": 1370
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.948693702978871e-05,
+      "loss": 0.2843,
+      "step": 1371
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.948618959108978e-05,
+      "loss": 0.2195,
+      "step": 1372
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.948544161115827e-05,
+      "loss": 0.2539,
+      "step": 1373
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.948469309000242e-05,
+      "loss": 0.2307,
+      "step": 1374
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.94839440276304e-05,
+      "loss": 0.2761,
+      "step": 1375
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.948319442405038e-05,
+      "loss": 0.2389,
+      "step": 1376
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.948244427927058e-05,
+      "loss": 0.2331,
+      "step": 1377
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.94816935932992e-05,
+      "loss": 0.2374,
+      "step": 1378
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 9.948094236614444e-05,
+      "loss": 0.2599,
+      "step": 1379
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 9.948019059781453e-05,
+      "loss": 0.2466,
+      "step": 1380
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 9.94794382883177e-05,
+      "loss": 0.2358,
+      "step": 1381
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 9.947868543766214e-05,
+      "loss": 0.2461,
+      "step": 1382
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 9.947793204585611e-05,
+      "loss": 0.2365,
+      "step": 1383
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 9.947717811290787e-05,
+      "loss": 0.243,
+      "step": 1384
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 9.947642363882563e-05,
+      "loss": 0.2404,
+      "step": 1385
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 9.947566862361766e-05,
+      "loss": 0.2496,
+      "step": 1386
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 9.947491306729221e-05,
+      "loss": 0.2432,
+      "step": 1387
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 9.947415696985755e-05,
+      "loss": 0.2079,
+      "step": 1388
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 9.947340033132195e-05,
+      "loss": 0.2127,
+      "step": 1389
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 9.947264315169369e-05,
+      "loss": 0.2513,
+      "step": 1390
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 9.947188543098103e-05,
+      "loss": 0.2517,
+      "step": 1391
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 9.947112716919227e-05,
+      "loss": 0.2522,
+      "step": 1392
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 9.94703683663357e-05,
+      "loss": 0.245,
+      "step": 1393
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 9.946960902241965e-05,
+      "loss": 0.2281,
+      "step": 1394
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 9.946884913745237e-05,
+      "loss": 0.2436,
+      "step": 1395
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 9.94680887114422e-05,
+      "loss": 0.2521,
+      "step": 1396
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 9.946732774439746e-05,
+      "loss": 0.262,
+      "step": 1397
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 9.946656623632649e-05,
+      "loss": 0.2498,
+      "step": 1398
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 9.946580418723756e-05,
+      "loss": 0.251,
+      "step": 1399
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 9.946504159713906e-05,
+      "loss": 0.229,
+      "step": 1400
+    },
+    {
+      "epoch": 1.86,
+      "eval_accuracy": 0.8938860079366009,
+      "eval_f1": 0.9165815457950289,
+      "eval_loss": 0.2521640956401825,
+      "eval_runtime": 110.1489,
+      "eval_samples_per_second": 775.559,
+      "eval_steps_per_second": 0.763,
+      "step": 1400
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 9.946427846603931e-05,
+      "loss": 0.2551,
+      "step": 1401
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 9.946351479394667e-05,
+      "loss": 0.208,
+      "step": 1402
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 9.946275058086946e-05,
+      "loss": 0.2572,
+      "step": 1403
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 9.946198582681607e-05,
+      "loss": 0.2169,
+      "step": 1404
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 9.946122053179485e-05,
+      "loss": 0.2435,
+      "step": 1405
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 9.946045469581419e-05,
+      "loss": 0.2423,
+      "step": 1406
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 9.945968831888244e-05,
+      "loss": 0.2224,
+      "step": 1407
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 9.945892140100798e-05,
+      "loss": 0.2443,
+      "step": 1408
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 9.945815394219922e-05,
+      "loss": 0.2039,
+      "step": 1409
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 9.945738594246452e-05,
+      "loss": 0.2211,
+      "step": 1410
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 9.945661740181234e-05,
+      "loss": 0.2508,
+      "step": 1411
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 9.945584832025102e-05,
+      "loss": 0.2353,
+      "step": 1412
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 9.945507869778901e-05,
+      "loss": 0.2431,
+      "step": 1413
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 9.945430853443471e-05,
+      "loss": 0.2442,
+      "step": 1414
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 9.945353783019656e-05,
+      "loss": 0.2259,
+      "step": 1415
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 9.945276658508295e-05,
+      "loss": 0.2381,
+      "step": 1416
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 9.945199479910238e-05,
+      "loss": 0.2398,
+      "step": 1417
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 9.945122247226323e-05,
+      "loss": 0.2381,
+      "step": 1418
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 9.9450449604574e-05,
+      "loss": 0.2641,
+      "step": 1419
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 9.944967619604309e-05,
+      "loss": 0.273,
+      "step": 1420
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 9.9448902246679e-05,
+      "loss": 0.2431,
+      "step": 1421
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 9.944812775649017e-05,
+      "loss": 0.2523,
+      "step": 1422
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 9.944735272548509e-05,
+      "loss": 0.2631,
+      "step": 1423
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 9.944657715367221e-05,
+      "loss": 0.2377,
+      "step": 1424
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 9.944580104106003e-05,
+      "loss": 0.2501,
+      "step": 1425
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 9.944502438765705e-05,
+      "loss": 0.2491,
+      "step": 1426
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 9.944424719347174e-05,
+      "loss": 0.2675,
+      "step": 1427
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 9.944346945851262e-05,
+      "loss": 0.2562,
+      "step": 1428
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 9.944269118278817e-05,
+      "loss": 0.257,
+      "step": 1429
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 9.944191236630693e-05,
+      "loss": 0.2494,
+      "step": 1430
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 9.944113300907742e-05,
+      "loss": 0.2355,
+      "step": 1431
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 9.944035311110814e-05,
+      "loss": 0.2485,
+      "step": 1432
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 9.943957267240763e-05,
+      "loss": 0.2292,
+      "step": 1433
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 9.943879169298442e-05,
+      "loss": 0.2665,
+      "step": 1434
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 9.943801017284708e-05,
+      "loss": 0.2614,
+      "step": 1435
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 9.943722811200412e-05,
+      "loss": 0.2248,
+      "step": 1436
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 9.943644551046412e-05,
+      "loss": 0.2385,
+      "step": 1437
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 9.94356623682356e-05,
+      "loss": 0.2087,
+      "step": 1438
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 9.943487868532719e-05,
+      "loss": 0.2333,
+      "step": 1439
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 9.943409446174741e-05,
+      "loss": 0.2433,
+      "step": 1440
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 9.943330969750485e-05,
+      "loss": 0.224,
+      "step": 1441
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 9.943252439260809e-05,
+      "loss": 0.2508,
+      "step": 1442
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 9.943173854706574e-05,
+      "loss": 0.2383,
+      "step": 1443
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 9.943095216088637e-05,
+      "loss": 0.2258,
+      "step": 1444
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 9.943016523407858e-05,
+      "loss": 0.2182,
+      "step": 1445
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 9.9429377766651e-05,
+      "loss": 0.2437,
+      "step": 1446
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 9.942858975861222e-05,
+      "loss": 0.2689,
+      "step": 1447
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 9.942780120997087e-05,
+      "loss": 0.2419,
+      "step": 1448
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 9.942701212073558e-05,
+      "loss": 0.2222,
+      "step": 1449
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 9.942622249091495e-05,
+      "loss": 0.2327,
+      "step": 1450
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8937806548280989,
+      "eval_f1": 0.9156550352289416,
+      "eval_loss": 0.2551378011703491,
+      "eval_runtime": 110.4306,
+      "eval_samples_per_second": 773.581,
+      "eval_steps_per_second": 0.761,
+      "step": 1450
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 9.942543232051766e-05,
+      "loss": 0.2284,
+      "step": 1451
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 9.942464160955233e-05,
+      "loss": 0.2415,
+      "step": 1452
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 9.942385035802759e-05,
+      "loss": 0.2431,
+      "step": 1453
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 9.942305856595214e-05,
+      "loss": 0.2225,
+      "step": 1454
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 9.94222662333346e-05,
+      "loss": 0.2393,
+      "step": 1455
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 9.942147336018363e-05,
+      "loss": 0.2346,
+      "step": 1456
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 9.942067994650795e-05,
+      "loss": 0.2537,
+      "step": 1457
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 9.941988599231618e-05,
+      "loss": 0.2372,
+      "step": 1458
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 9.941909149761704e-05,
+      "loss": 0.2266,
+      "step": 1459
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 9.941829646241922e-05,
+      "loss": 0.2481,
+      "step": 1460
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 9.941750088673138e-05,
+      "loss": 0.2387,
+      "step": 1461
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 9.941670477056227e-05,
+      "loss": 0.2015,
+      "step": 1462
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 9.941590811392056e-05,
+      "loss": 0.2329,
+      "step": 1463
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 9.9415110916815e-05,
+      "loss": 0.2324,
+      "step": 1464
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 9.941431317925427e-05,
+      "loss": 0.2543,
+      "step": 1465
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 9.94135149012471e-05,
+      "loss": 0.2279,
+      "step": 1466
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 9.941271608280224e-05,
+      "loss": 0.2467,
+      "step": 1467
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 9.941191672392843e-05,
+      "loss": 0.2822,
+      "step": 1468
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 9.94111168246344e-05,
+      "loss": 0.2637,
+      "step": 1469
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 9.941031638492889e-05,
+      "loss": 0.2552,
+      "step": 1470
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 9.940951540482067e-05,
+      "loss": 0.2455,
+      "step": 1471
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 9.940871388431849e-05,
+      "loss": 0.2401,
+      "step": 1472
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 9.940791182343112e-05,
+      "loss": 0.2548,
+      "step": 1473
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 9.940710922216734e-05,
+      "loss": 0.2618,
+      "step": 1474
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 9.940630608053592e-05,
+      "loss": 0.2569,
+      "step": 1475
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 9.940550239854562e-05,
+      "loss": 0.2306,
+      "step": 1476
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 9.940469817620528e-05,
+      "loss": 0.2375,
+      "step": 1477
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 9.940389341352367e-05,
+      "loss": 0.2847,
+      "step": 1478
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 9.940308811050958e-05,
+      "loss": 0.2363,
+      "step": 1479
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 9.940228226717184e-05,
+      "loss": 0.255,
+      "step": 1480
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 9.940147588351924e-05,
+      "loss": 0.2424,
+      "step": 1481
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 9.940066895956064e-05,
+      "loss": 0.2717,
+      "step": 1482
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 9.939986149530482e-05,
+      "loss": 0.2447,
+      "step": 1483
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 9.939905349076062e-05,
+      "loss": 0.2939,
+      "step": 1484
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 9.93982449459369e-05,
+      "loss": 0.24,
+      "step": 1485
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 9.939743586084249e-05,
+      "loss": 0.2224,
+      "step": 1486
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 9.939662623548625e-05,
+      "loss": 0.2501,
+      "step": 1487
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 9.9395816069877e-05,
+      "loss": 0.2472,
+      "step": 1488
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 9.939500536402364e-05,
+      "loss": 0.2394,
+      "step": 1489
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 9.939419411793501e-05,
+      "loss": 0.2254,
+      "step": 1490
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 9.939338233162e-05,
+      "loss": 0.2352,
+      "step": 1491
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 9.939257000508748e-05,
+      "loss": 0.2269,
+      "step": 1492
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 9.939175713834635e-05,
+      "loss": 0.2279,
+      "step": 1493
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 9.939094373140548e-05,
+      "loss": 0.2266,
+      "step": 1494
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 9.939012978427375e-05,
+      "loss": 0.2609,
+      "step": 1495
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 9.938931529696011e-05,
+      "loss": 0.2558,
+      "step": 1496
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 9.938850026947343e-05,
+      "loss": 0.2502,
+      "step": 1497
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 9.938768470182265e-05,
+      "loss": 0.2242,
+      "step": 1498
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 9.938686859401669e-05,
+      "loss": 0.2491,
+      "step": 1499
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 9.938605194606443e-05,
+      "loss": 0.2507,
+      "step": 1500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8938040666299881,
+      "eval_f1": 0.9176201372997712,
+      "eval_loss": 0.2536262273788452,
+      "eval_runtime": 110.3623,
+      "eval_samples_per_second": 774.06,
+      "eval_steps_per_second": 0.761,
+      "step": 1500
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 9.938523475797487e-05,
+      "loss": 0.2562,
+      "step": 1501
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 9.938441702975689e-05,
+      "loss": 0.2332,
+      "step": 1502
+    }
+  ],
+  "max_steps": 30040,
+  "num_train_epochs": 40,
+  "total_flos": 7.9566853473192e+16,
+  "trial_name": null,
+  "trial_params": null
+}