diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,12955 @@
+{
+  "best_metric": 0.19643232226371765,
+  "best_model_checkpoint": "./qlora-Math3-data-07/checkpoint-4224",
+  "epoch": 1.9496884375721208,
+  "eval_steps": 128,
+  "global_step": 4224,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.0000000000000003e-07,
+      "loss": 0.6401,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.000000000000001e-07,
+      "loss": 0.6396,
+      "step": 4
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2000000000000002e-06,
+      "loss": 0.6553,
+      "step": 6
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6000000000000001e-06,
+      "loss": 0.632,
+      "step": 8
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 0.619,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4000000000000003e-06,
+      "loss": 0.5579,
+      "step": 12
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8000000000000003e-06,
+      "loss": 0.5585,
+      "step": 14
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2000000000000003e-06,
+      "loss": 0.515,
+      "step": 16
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6000000000000003e-06,
+      "loss": 0.4867,
+      "step": 18
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.4558,
+      "step": 20
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4e-06,
+      "loss": 0.4728,
+      "step": 22
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.800000000000001e-06,
+      "loss": 0.4643,
+      "step": 24
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.2e-06,
+      "loss": 0.4609,
+      "step": 26
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.600000000000001e-06,
+      "loss": 0.4274,
+      "step": 28
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6e-06,
+      "loss": 0.4226,
+      "step": 30
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.4000000000000006e-06,
+      "loss": 0.3975,
+      "step": 32
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.800000000000001e-06,
+      "loss": 0.4339,
+      "step": 34
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.2000000000000005e-06,
+      "loss": 0.4114,
+      "step": 36
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.600000000000001e-06,
+      "loss": 0.409,
+      "step": 38
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.3817,
+      "step": 40
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.400000000000001e-06,
+      "loss": 0.4023,
+      "step": 42
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.8e-06,
+      "loss": 0.408,
+      "step": 44
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.200000000000002e-06,
+      "loss": 0.3758,
+      "step": 46
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.600000000000001e-06,
+      "loss": 0.3662,
+      "step": 48
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 0.3922,
+      "step": 50
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.04e-05,
+      "loss": 0.361,
+      "step": 52
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.0800000000000002e-05,
+      "loss": 0.3468,
+      "step": 54
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.1200000000000001e-05,
+      "loss": 0.3672,
+      "step": 56
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.16e-05,
+      "loss": 0.3338,
+      "step": 58
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.2e-05,
+      "loss": 0.352,
+      "step": 60
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.2400000000000002e-05,
+      "loss": 0.3668,
+      "step": 62
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.2800000000000001e-05,
+      "loss": 0.3527,
+      "step": 64
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.3200000000000002e-05,
+      "loss": 0.3551,
+      "step": 66
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.3600000000000002e-05,
+      "loss": 0.3561,
+      "step": 68
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.4e-05,
+      "loss": 0.3636,
+      "step": 70
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.4400000000000001e-05,
+      "loss": 0.3673,
+      "step": 72
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.48e-05,
+      "loss": 0.3571,
+      "step": 74
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.5200000000000002e-05,
+      "loss": 0.3354,
+      "step": 76
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.5600000000000003e-05,
+      "loss": 0.3432,
+      "step": 78
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.3754,
+      "step": 80
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.64e-05,
+      "loss": 0.3321,
+      "step": 82
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.6800000000000002e-05,
+      "loss": 0.3538,
+      "step": 84
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.72e-05,
+      "loss": 0.3131,
+      "step": 86
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.76e-05,
+      "loss": 0.3569,
+      "step": 88
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.8e-05,
+      "loss": 0.3453,
+      "step": 90
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.8400000000000003e-05,
+      "loss": 0.3422,
+      "step": 92
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.88e-05,
+      "loss": 0.3454,
+      "step": 94
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9200000000000003e-05,
+      "loss": 0.33,
+      "step": 96
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9600000000000002e-05,
+      "loss": 0.3312,
+      "step": 98
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 2e-05,
+      "loss": 0.3435,
+      "step": 100
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9999998285528446e-05,
+      "loss": 0.3134,
+      "step": 102
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.999999314211436e-05,
+      "loss": 0.345,
+      "step": 104
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9999984569759512e-05,
+      "loss": 0.3274,
+      "step": 106
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9999972568466842e-05,
+      "loss": 0.3563,
+      "step": 108
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.999995713824046e-05,
+      "loss": 0.322,
+      "step": 110
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.999993827908566e-05,
+      "loss": 0.3221,
+      "step": 112
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.999991599100891e-05,
+      "loss": 0.3325,
+      "step": 114
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.999989027401785e-05,
+      "loss": 0.3403,
+      "step": 116
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.99998611281213e-05,
+      "loss": 0.3286,
+      "step": 118
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9999828553329254e-05,
+      "loss": 0.3368,
+      "step": 120
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9999792549652882e-05,
+      "loss": 0.3413,
+      "step": 122
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9999753117104527e-05,
+      "loss": 0.3423,
+      "step": 124
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9999710255697713e-05,
+      "loss": 0.3073,
+      "step": 126
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9999663965447133e-05,
+      "loss": 0.3315,
+      "step": 128
+    },
+    {
+      "epoch": 0.06,
+      "eval_loss": 0.3296878933906555,
+      "eval_runtime": 72.3883,
+      "eval_samples_per_second": 27.629,
+      "eval_steps_per_second": 13.814,
+      "step": 128
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9999614246368666e-05,
+      "loss": 0.3214,
+      "step": 130
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9999561098479355e-05,
+      "loss": 0.3477,
+      "step": 132
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9999504521797427e-05,
+      "loss": 0.3152,
+      "step": 134
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.999944451634228e-05,
+      "loss": 0.343,
+      "step": 136
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.999938108213449e-05,
+      "loss": 0.3361,
+      "step": 138
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9999314219195808e-05,
+      "loss": 0.3074,
+      "step": 140
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9999243927549162e-05,
+      "loss": 0.3381,
+      "step": 142
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9999170207218653e-05,
+      "loss": 0.2995,
+      "step": 144
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9999093058229562e-05,
+      "loss": 0.3361,
+      "step": 146
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.999901248060834e-05,
+      "loss": 0.3319,
+      "step": 148
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9998928474382617e-05,
+      "loss": 0.3143,
+      "step": 150
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.99988410395812e-05,
+      "loss": 0.3111,
+      "step": 152
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9998750176234067e-05,
+      "loss": 0.31,
+      "step": 154
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.999865588437238e-05,
+      "loss": 0.3205,
+      "step": 156
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9998558164028466e-05,
+      "loss": 0.3,
+      "step": 158
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9998457015235836e-05,
+      "loss": 0.3318,
+      "step": 160
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.999835243802917e-05,
+      "loss": 0.3114,
+      "step": 162
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9998244432444328e-05,
+      "loss": 0.3199,
+      "step": 164
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9998132998518347e-05,
+      "loss": 0.334,
+      "step": 166
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.999801813628943e-05,
+      "loss": 0.3301,
+      "step": 168
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9997899845796975e-05,
+      "loss": 0.3245,
+      "step": 170
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9997778127081532e-05,
+      "loss": 0.3161,
+      "step": 172
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.999765298018484e-05,
+      "loss": 0.3391,
+      "step": 174
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.999752440514982e-05,
+      "loss": 0.3167,
+      "step": 176
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9997392402020547e-05,
+      "loss": 0.3182,
+      "step": 178
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9997256970842288e-05,
+      "loss": 0.3057,
+      "step": 180
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9997118111661485e-05,
+      "loss": 0.3285,
+      "step": 182
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.999697582452575e-05,
+      "loss": 0.3034,
+      "step": 184
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9996830109483874e-05,
+      "loss": 0.3069,
+      "step": 186
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.999668096658582e-05,
+      "loss": 0.3197,
+      "step": 188
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.999652839588273e-05,
+      "loss": 0.3081,
+      "step": 190
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9996372397426914e-05,
+      "loss": 0.3127,
+      "step": 192
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.999621297127187e-05,
+      "loss": 0.3301,
+      "step": 194
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.999605011747226e-05,
+      "loss": 0.3174,
+      "step": 196
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9995883836083928e-05,
+      "loss": 0.3236,
+      "step": 198
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.999571412716389e-05,
+      "loss": 0.3176,
+      "step": 200
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.999554099077034e-05,
+      "loss": 0.2994,
+      "step": 202
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9995364426962647e-05,
+      "loss": 0.3079,
+      "step": 204
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9995184435801346e-05,
+      "loss": 0.3163,
+      "step": 206
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999500101734816e-05,
+      "loss": 0.3038,
+      "step": 208
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999481417166598e-05,
+      "loss": 0.3112,
+      "step": 210
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999462389881888e-05,
+      "loss": 0.3009,
+      "step": 212
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.99944301988721e-05,
+      "loss": 0.3154,
+      "step": 214
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9994233071892056e-05,
+      "loss": 0.3238,
+      "step": 216
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9994032517946342e-05,
+      "loss": 0.299,
+      "step": 218
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999382853710373e-05,
+      "loss": 0.3184,
+      "step": 220
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999362112943416e-05,
+      "loss": 0.3048,
+      "step": 222
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999341029500876e-05,
+      "loss": 0.2982,
+      "step": 224
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9993196033899816e-05,
+      "loss": 0.3003,
+      "step": 226
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9992978346180793e-05,
+      "loss": 0.3242,
+      "step": 228
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9992757231926344e-05,
+      "loss": 0.2987,
+      "step": 230
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9992532691212285e-05,
+      "loss": 0.2973,
+      "step": 232
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9992304724115605e-05,
+      "loss": 0.2934,
+      "step": 234
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9992073330714477e-05,
+      "loss": 0.2805,
+      "step": 236
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999183851108825e-05,
+      "loss": 0.3004,
+      "step": 238
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9991600265317428e-05,
+      "loss": 0.2968,
+      "step": 240
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9991358593483716e-05,
+      "loss": 0.2878,
+      "step": 242
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9991113495669977e-05,
+      "loss": 0.2906,
+      "step": 244
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9990864971960255e-05,
+      "loss": 0.3078,
+      "step": 246
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999061302243977e-05,
+      "loss": 0.2951,
+      "step": 248
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9990357647194907e-05,
+      "loss": 0.2944,
+      "step": 250
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.999009884631324e-05,
+      "loss": 0.2919,
+      "step": 252
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9989836619883503e-05,
+      "loss": 0.3139,
+      "step": 254
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9989570967995622e-05,
+      "loss": 0.2956,
+      "step": 256
+    },
+    {
+      "epoch": 0.12,
+      "eval_loss": 0.304767370223999,
+      "eval_runtime": 72.4187,
+      "eval_samples_per_second": 27.617,
+      "eval_steps_per_second": 13.809,
+      "step": 256
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.998930189074068e-05,
+      "loss": 0.2967,
+      "step": 258
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.998902938821094e-05,
+      "loss": 0.3226,
+      "step": 260
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9988753460499852e-05,
+      "loss": 0.3067,
+      "step": 262
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9988474107702017e-05,
+      "loss": 0.3257,
+      "step": 264
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9988191329913236e-05,
+      "loss": 0.3052,
+      "step": 266
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.998790512723046e-05,
+      "loss": 0.3138,
+      "step": 268
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9987615499751837e-05,
+      "loss": 0.3053,
+      "step": 270
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9987322447576673e-05,
+      "loss": 0.3183,
+      "step": 272
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.998702597080545e-05,
+      "loss": 0.2944,
+      "step": 274
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9986726069539837e-05,
+      "loss": 0.2955,
+      "step": 276
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9986422743882663e-05,
+      "loss": 0.3107,
+      "step": 278
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9986115993937936e-05,
+      "loss": 0.3026,
+      "step": 280
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9985805819810844e-05,
+      "loss": 0.2965,
+      "step": 282
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.998549222160774e-05,
+      "loss": 0.2781,
+      "step": 284
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9985175199436155e-05,
+      "loss": 0.3054,
+      "step": 286
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9984854753404795e-05,
+      "loss": 0.301,
+      "step": 288
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9984530883623538e-05,
+      "loss": 0.2834,
+      "step": 290
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.998420359020344e-05,
+      "loss": 0.2776,
+      "step": 292
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9983872873256725e-05,
+      "loss": 0.2862,
+      "step": 294
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9983538732896792e-05,
+      "loss": 0.3304,
+      "step": 296
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9983201169238224e-05,
+      "loss": 0.3152,
+      "step": 298
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9982860182396763e-05,
+      "loss": 0.2881,
+      "step": 300
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.998251577248933e-05,
+      "loss": 0.2823,
+      "step": 302
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9982167939634028e-05,
+      "loss": 0.3101,
+      "step": 304
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9981816683950122e-05,
+      "loss": 0.283,
+      "step": 306
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9981462005558055e-05,
+      "loss": 0.2777,
+      "step": 308
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.998110390457945e-05,
+      "loss": 0.2964,
+      "step": 310
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.998074238113709e-05,
+      "loss": 0.2984,
+      "step": 312
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.998037743535494e-05,
+      "loss": 0.2841,
+      "step": 314
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9980009067358147e-05,
+      "loss": 0.2946,
+      "step": 316
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9979637277273012e-05,
+      "loss": 0.2758,
+      "step": 318
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.997926206522703e-05,
+      "loss": 0.2959,
+      "step": 320
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9978883431348845e-05,
+      "loss": 0.2928,
+      "step": 322
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.99785013757683e-05,
+      "loss": 0.2931,
+      "step": 324
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.99781158986164e-05,
+      "loss": 0.3,
+      "step": 326
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.997772700002531e-05,
+      "loss": 0.2934,
+      "step": 328
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9977334680128394e-05,
+      "loss": 0.2874,
+      "step": 330
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9976938939060172e-05,
+      "loss": 0.3075,
+      "step": 332
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9976539776956344e-05,
+      "loss": 0.2855,
+      "step": 334
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9976137193953775e-05,
+      "loss": 0.2887,
+      "step": 336
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9975731190190512e-05,
+      "loss": 0.2874,
+      "step": 338
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9975321765805777e-05,
+      "loss": 0.2838,
+      "step": 340
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9974908920939942e-05,
+      "loss": 0.2839,
+      "step": 342
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.997449265573459e-05,
+      "loss": 0.3064,
+      "step": 344
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9974072970332443e-05,
+      "loss": 0.3035,
+      "step": 346
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9973649864877413e-05,
+      "loss": 0.2927,
+      "step": 348
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9973223339514573e-05,
+      "loss": 0.3183,
+      "step": 350
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9972793394390192e-05,
+      "loss": 0.2807,
+      "step": 352
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.997236002965168e-05,
+      "loss": 0.3008,
+      "step": 354
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.997192324544764e-05,
+      "loss": 0.2672,
+      "step": 356
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.997148304192785e-05,
+      "loss": 0.2945,
+      "step": 358
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9971039419243246e-05,
+      "loss": 0.2828,
+      "step": 360
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9970592377545945e-05,
+      "loss": 0.2919,
+      "step": 362
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9970141916989235e-05,
+      "loss": 0.2804,
+      "step": 364
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9969688037727576e-05,
+      "loss": 0.2848,
+      "step": 366
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9969230739916606e-05,
+      "loss": 0.3199,
+      "step": 368
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9968770023713123e-05,
+      "loss": 0.2829,
+      "step": 370
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9968305889275104e-05,
+      "loss": 0.2746,
+      "step": 372
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9967838336761702e-05,
+      "loss": 0.2773,
+      "step": 374
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9967367366333233e-05,
+      "loss": 0.2817,
+      "step": 376
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9966892978151196e-05,
+      "loss": 0.2762,
+      "step": 378
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9966415172378256e-05,
+      "loss": 0.2856,
+      "step": 380
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9965933949178248e-05,
+      "loss": 0.2964,
+      "step": 382
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9965449308716175e-05,
+      "loss": 0.2742,
+      "step": 384
+    },
+    {
+      "epoch": 0.18,
+      "eval_loss": 0.2901277542114258,
+      "eval_runtime": 72.4126,
+      "eval_samples_per_second": 27.62,
+      "eval_steps_per_second": 13.81,
+      "step": 384
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9964961251158224e-05,
+      "loss": 0.3035,
+      "step": 386
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9964469776671747e-05,
+      "loss": 0.2868,
+      "step": 388
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9963974885425267e-05,
+      "loss": 0.2881,
+      "step": 390
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9963476577588477e-05,
+      "loss": 0.2873,
+      "step": 392
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9962974853332247e-05,
+      "loss": 0.2733,
+      "step": 394
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9962469712828613e-05,
+      "loss": 0.2625,
+      "step": 396
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.996196115625079e-05,
+      "loss": 0.2909,
+      "step": 398
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9961449183773148e-05,
+      "loss": 0.2841,
+      "step": 400
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9960933795571255e-05,
+      "loss": 0.2884,
+      "step": 402
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.996041499182182e-05,
+      "loss": 0.287,
+      "step": 404
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9959892772702747e-05,
+      "loss": 0.2989,
+      "step": 406
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.99593671383931e-05,
+      "loss": 0.266,
+      "step": 408
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.995883808907311e-05,
+      "loss": 0.2665,
+      "step": 410
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9958305624924193e-05,
+      "loss": 0.2758,
+      "step": 412
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9957769746128927e-05,
+      "loss": 0.2693,
+      "step": 414
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9957230452871056e-05,
+      "loss": 0.2805,
+      "step": 416
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.995668774533551e-05,
+      "loss": 0.3119,
+      "step": 418
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9956141623708373e-05,
+      "loss": 0.2823,
+      "step": 420
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9955592088176906e-05,
+      "loss": 0.2907,
+      "step": 422
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9955039138929546e-05,
+      "loss": 0.2921,
+      "step": 424
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.995448277615589e-05,
+      "loss": 0.2833,
+      "step": 426
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.995392300004672e-05,
+      "loss": 0.2831,
+      "step": 428
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.995335981079398e-05,
+      "loss": 0.2875,
+      "step": 430
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9952793208590778e-05,
+      "loss": 0.276,
+      "step": 432
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9952223193631398e-05,
+      "loss": 0.3026,
+      "step": 434
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9951649766111304e-05,
+      "loss": 0.2667,
+      "step": 436
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9951072926227112e-05,
+      "loss": 0.2794,
+      "step": 438
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.995049267417662e-05,
+      "loss": 0.2763,
+      "step": 440
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.994990901015879e-05,
+      "loss": 0.2681,
+      "step": 442
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9949321934373763e-05,
+      "loss": 0.2618,
+      "step": 444
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9948731447022843e-05,
+      "loss": 0.2787,
+      "step": 446
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9948137548308504e-05,
+      "loss": 0.3006,
+      "step": 448
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.994754023843439e-05,
+      "loss": 0.281,
+      "step": 450
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9946939517605314e-05,
+      "loss": 0.2874,
+      "step": 452
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9946335386027254e-05,
+      "loss": 0.3117,
+      "step": 454
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9945727843907377e-05,
+      "loss": 0.2872,
+      "step": 456
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9945116891453998e-05,
+      "loss": 0.2987,
+      "step": 458
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.994450252887661e-05,
+      "loss": 0.2967,
+      "step": 460
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9943884756385873e-05,
+      "loss": 0.2702,
+      "step": 462
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9943263574193618e-05,
+      "loss": 0.2884,
+      "step": 464
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9942638982512847e-05,
+      "loss": 0.2899,
+      "step": 466
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9942010981557728e-05,
+      "loss": 0.268,
+      "step": 468
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9941379571543597e-05,
+      "loss": 0.2701,
+      "step": 470
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9940744752686966e-05,
+      "loss": 0.2804,
+      "step": 472
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9940106525205502e-05,
+      "loss": 0.2619,
+      "step": 474
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9939464889318058e-05,
+      "loss": 0.278,
+      "step": 476
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9938819845244642e-05,
+      "loss": 0.2737,
+      "step": 478
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.993817139320644e-05,
+      "loss": 0.2713,
+      "step": 480
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.99375195334258e-05,
+      "loss": 0.2746,
+      "step": 482
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9936864266126243e-05,
+      "loss": 0.2782,
+      "step": 484
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9936205591532453e-05,
+      "loss": 0.2657,
+      "step": 486
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.993554350987029e-05,
+      "loss": 0.2733,
+      "step": 488
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9934878021366773e-05,
+      "loss": 0.2682,
+      "step": 490
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.99342091262501e-05,
+      "loss": 0.2712,
+      "step": 492
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9933536824749625e-05,
+      "loss": 0.2772,
+      "step": 494
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9932861117095885e-05,
+      "loss": 0.2635,
+      "step": 496
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9932182003520567e-05,
+      "loss": 0.2713,
+      "step": 498
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9931499484256537e-05,
+      "loss": 0.2892,
+      "step": 500
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.993081355953783e-05,
+      "loss": 0.2939,
+      "step": 502
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9930124229599647e-05,
+      "loss": 0.2814,
+      "step": 504
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9929431494678355e-05,
+      "loss": 0.2767,
+      "step": 506
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9928735355011485e-05,
+      "loss": 0.2742,
+      "step": 508
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.992803581083774e-05,
+      "loss": 0.293,
+      "step": 510
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9927332862396996e-05,
+      "loss": 0.2829,
+      "step": 512
+    },
+    {
+      "epoch": 0.24,
+      "eval_loss": 0.28032785654067993,
+      "eval_runtime": 72.3833,
+      "eval_samples_per_second": 27.631,
+      "eval_steps_per_second": 13.815,
+      "step": 512
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.992662650993028e-05,
+      "loss": 0.2724,
+      "step": 514
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9925916753679805e-05,
+      "loss": 0.2943,
+      "step": 516
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9925203593888937e-05,
+      "loss": 0.2862,
+      "step": 518
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9924487030802214e-05,
+      "loss": 0.2707,
+      "step": 520
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9923767064665345e-05,
+      "loss": 0.2782,
+      "step": 522
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9923043695725202e-05,
+      "loss": 0.2969,
+      "step": 524
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9922316924229824e-05,
+      "loss": 0.2619,
+      "step": 526
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9921586750428414e-05,
+      "loss": 0.2736,
+      "step": 528
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9920853174571345e-05,
+      "loss": 0.2704,
+      "step": 530
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9920116196910162e-05,
+      "loss": 0.2853,
+      "step": 532
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9919375817697563e-05,
+      "loss": 0.2615,
+      "step": 534
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.991863203718742e-05,
+      "loss": 0.2851,
+      "step": 536
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9917884855634778e-05,
+      "loss": 0.265,
+      "step": 538
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9917134273295833e-05,
+      "loss": 0.2703,
+      "step": 540
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.991638029042796e-05,
+      "loss": 0.2852,
+      "step": 542
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9915622907289695e-05,
+      "loss": 0.2592,
+      "step": 544
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9914862124140743e-05,
+      "loss": 0.2883,
+      "step": 546
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9914097941241966e-05,
+      "loss": 0.2792,
+      "step": 548
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.99133303588554e-05,
+      "loss": 0.2611,
+      "step": 550
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.991255937724425e-05,
+      "loss": 0.2881,
+      "step": 552
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9911784996672875e-05,
+      "loss": 0.2734,
+      "step": 554
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.991100721740681e-05,
+      "loss": 0.2643,
+      "step": 556
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9910226039712746e-05,
+      "loss": 0.2707,
+      "step": 558
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.990944146385855e-05,
+      "loss": 0.2775,
+      "step": 560
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9908653490113243e-05,
+      "loss": 0.2763,
+      "step": 562
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.990786211874702e-05,
+      "loss": 0.2733,
+      "step": 564
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.990706735003124e-05,
+      "loss": 0.2701,
+      "step": 566
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9906269184238422e-05,
+      "loss": 0.253,
+      "step": 568
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9905467621642248e-05,
+      "loss": 0.2657,
+      "step": 570
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.990466266251758e-05,
+      "loss": 0.2675,
+      "step": 572
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9903854307140423e-05,
+      "loss": 0.2683,
+      "step": 574
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9903042555787967e-05,
+      "loss": 0.2692,
+      "step": 576
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9902227408738545e-05,
+      "loss": 0.2654,
+      "step": 578
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.990140886627168e-05,
+      "loss": 0.275,
+      "step": 580
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9900586928668037e-05,
+      "loss": 0.2899,
+      "step": 582
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9899761596209455e-05,
+      "loss": 0.2475,
+      "step": 584
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9898932869178936e-05,
+      "loss": 0.2646,
+      "step": 586
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9898100747860644e-05,
+      "loss": 0.2665,
+      "step": 588
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9897265232539917e-05,
+      "loss": 0.2733,
+      "step": 590
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9896426323503237e-05,
+      "loss": 0.2593,
+      "step": 592
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.989558402103827e-05,
+      "loss": 0.2771,
+      "step": 594
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.989473832543383e-05,
+      "loss": 0.2728,
+      "step": 596
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9893889236979906e-05,
+      "loss": 0.2672,
+      "step": 598
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9893036755967645e-05,
+      "loss": 0.2677,
+      "step": 600
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9892180882689356e-05,
+      "loss": 0.2763,
+      "step": 602
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9891321617438515e-05,
+      "loss": 0.2754,
+      "step": 604
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.989045896050976e-05,
+      "loss": 0.262,
+      "step": 606
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9889592912198883e-05,
+      "loss": 0.2689,
+      "step": 608
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.988872347280286e-05,
+      "loss": 0.2763,
+      "step": 610
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9887850642619804e-05,
+      "loss": 0.2786,
+      "step": 612
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9886974421949015e-05,
+      "loss": 0.263,
+      "step": 614
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9886094811090933e-05,
+      "loss": 0.2671,
+      "step": 616
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9885211810347185e-05,
+      "loss": 0.2537,
+      "step": 618
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9884325420020535e-05,
+      "loss": 0.2651,
+      "step": 620
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9883435640414922e-05,
+      "loss": 0.2825,
+      "step": 622
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9882542471835458e-05,
+      "loss": 0.2686,
+      "step": 624
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.988164591458839e-05,
+      "loss": 0.2818,
+      "step": 626
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9880745968981152e-05,
+      "loss": 0.267,
+      "step": 628
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.987984263532233e-05,
+      "loss": 0.2578,
+      "step": 630
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9878935913921668e-05,
+      "loss": 0.2756,
+      "step": 632
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.987802580509008e-05,
+      "loss": 0.2456,
+      "step": 634
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9877112309139632e-05,
+      "loss": 0.2643,
+      "step": 636
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.987619542638356e-05,
+      "loss": 0.263,
+      "step": 638
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9875275157136254e-05,
+      "loss": 0.2551,
+      "step": 640
+    },
+    {
+      "epoch": 0.3,
+      "eval_loss": 0.27295249700546265,
+      "eval_runtime": 72.3829,
+      "eval_samples_per_second": 27.631,
+      "eval_steps_per_second": 13.815,
+      "step": 640
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9874351501713275e-05,
+      "loss": 0.2652,
+      "step": 642
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9873424460431336e-05,
+      "loss": 0.2408,
+      "step": 644
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9872494033608313e-05,
+      "loss": 0.2731,
+      "step": 646
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.987156022156325e-05,
+      "loss": 0.2595,
+      "step": 648
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9870623024616338e-05,
+      "loss": 0.2715,
+      "step": 650
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9869682443088943e-05,
+      "loss": 0.2572,
+      "step": 652
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9868738477303576e-05,
+      "loss": 0.2588,
+      "step": 654
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.986779112758393e-05,
+      "loss": 0.2652,
+      "step": 656
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9866840394254833e-05,
+      "loss": 0.2795,
+      "step": 658
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9865886277642297e-05,
+      "loss": 0.2544,
+      "step": 660
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9864928778073474e-05,
+      "loss": 0.2982,
+      "step": 662
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9863967895876697e-05,
+      "loss": 0.2668,
+      "step": 664
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9863003631381433e-05,
+      "loss": 0.2734,
+      "step": 666
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9862035984918333e-05,
+      "loss": 0.2832,
+      "step": 668
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9861064956819195e-05,
+      "loss": 0.2662,
+      "step": 670
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9860090547416976e-05,
+      "loss": 0.313,
+      "step": 672
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9859112757045794e-05,
+      "loss": 0.2771,
+      "step": 674
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9858131586040938e-05,
+      "loss": 0.26,
+      "step": 676
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9857147034738836e-05,
+      "loss": 0.2572,
+      "step": 678
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9856159103477085e-05,
+      "loss": 0.2767,
+      "step": 680
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9855167792594447e-05,
+      "loss": 0.2564,
+      "step": 682
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9854173102430837e-05,
+      "loss": 0.2649,
+      "step": 684
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.985317503332732e-05,
+      "loss": 0.2696,
+      "step": 686
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9852173585626137e-05,
+      "loss": 0.2733,
+      "step": 688
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9851168759670677e-05,
+      "loss": 0.261,
+      "step": 690
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9850160555805485e-05,
+      "loss": 0.2593,
+      "step": 692
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9849148974376274e-05,
+      "loss": 0.2654,
+      "step": 694
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9848134015729904e-05,
+      "loss": 0.2762,
+      "step": 696
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9847115680214402e-05,
+      "loss": 0.2482,
+      "step": 698
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9846093968178947e-05,
+      "loss": 0.2506,
+      "step": 700
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9845068879973882e-05,
+      "loss": 0.2957,
+      "step": 702
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9844040415950702e-05,
+      "loss": 0.2824,
+      "step": 704
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.984300857646206e-05,
+      "loss": 0.2643,
+      "step": 706
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.984197336186177e-05,
+      "loss": 0.2544,
+      "step": 708
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.98409347725048e-05,
+      "loss": 0.266,
+      "step": 710
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9839892808747273e-05,
+      "loss": 0.2521,
+      "step": 712
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.983884747094648e-05,
+      "loss": 0.2745,
+      "step": 714
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9837798759460857e-05,
+      "loss": 0.2671,
+      "step": 716
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.983674667465e-05,
+      "loss": 0.2862,
+      "step": 718
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9835691216874667e-05,
+      "loss": 0.3079,
+      "step": 720
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9834632386496766e-05,
+      "loss": 0.2765,
+      "step": 722
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9833570183879364e-05,
+      "loss": 0.2603,
+      "step": 724
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.983250460938668e-05,
+      "loss": 0.2617,
+      "step": 726
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9831435663384106e-05,
+      "loss": 0.2461,
+      "step": 728
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9830363346238163e-05,
+      "loss": 0.2537,
+      "step": 730
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.982928765831655e-05,
+      "loss": 0.2804,
+      "step": 732
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9828208599988114e-05,
+      "loss": 0.2809,
+      "step": 734
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9827126171622858e-05,
+      "loss": 0.2736,
+      "step": 736
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9826040373591934e-05,
+      "loss": 0.2684,
+      "step": 738
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9824951206267668e-05,
+      "loss": 0.2531,
+      "step": 740
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9823858670023516e-05,
+      "loss": 0.2801,
+      "step": 742
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.982276276523411e-05,
+      "loss": 0.2633,
+      "step": 744
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9821663492275233e-05,
+      "loss": 0.2706,
+      "step": 746
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9820560851523813e-05,
+      "loss": 0.2638,
+      "step": 748
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9819454843357942e-05,
+      "loss": 0.2614,
+      "step": 750
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.981834546815686e-05,
+      "loss": 0.2806,
+      "step": 752
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9817232726300972e-05,
+      "loss": 0.2653,
+      "step": 754
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9816116618171826e-05,
+      "loss": 0.2567,
+      "step": 756
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9814997144152134e-05,
+      "loss": 0.2595,
+      "step": 758
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.981387430462575e-05,
+      "loss": 0.2637,
+      "step": 760
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9812748099977693e-05,
+      "loss": 0.2694,
+      "step": 762
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.981161853059413e-05,
+      "loss": 0.2824,
+      "step": 764
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.981048559686239e-05,
+      "loss": 0.2448,
+      "step": 766
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.980934929917095e-05,
+      "loss": 0.2579,
+      "step": 768
+    },
+    {
+      "epoch": 0.35,
+      "eval_loss": 0.2658372223377228,
+      "eval_runtime": 72.3849,
+      "eval_samples_per_second": 27.63,
+      "eval_steps_per_second": 13.815,
+      "step": 768
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.980820963790943e-05,
+      "loss": 0.2631,
+      "step": 770
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9807066613468622e-05,
+      "loss": 0.2719,
+      "step": 772
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9805920226240458e-05,
+      "loss": 0.2549,
+      "step": 774
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9804770476618028e-05,
+      "loss": 0.2354,
+      "step": 776
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9803617364995578e-05,
+      "loss": 0.2757,
+      "step": 778
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9802460891768504e-05,
+      "loss": 0.2751,
+      "step": 780
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9801301057333345e-05,
+      "loss": 0.2306,
+      "step": 782
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9800137862087814e-05,
+      "loss": 0.2435,
+      "step": 784
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9798971306430756e-05,
+      "loss": 0.2568,
+      "step": 786
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.979780139076218e-05,
+      "loss": 0.2528,
+      "step": 788
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.979662811548324e-05,
+      "loss": 0.2807,
+      "step": 790
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9795451480996248e-05,
+      "loss": 0.2442,
+      "step": 792
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9794271487704663e-05,
+      "loss": 0.2627,
+      "step": 794
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.97930881360131e-05,
+      "loss": 0.261,
+      "step": 796
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9791901426327327e-05,
+      "loss": 0.2448,
+      "step": 798
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9790711359054254e-05,
+      "loss": 0.2436,
+      "step": 800
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9789517934601948e-05,
+      "loss": 0.2589,
+      "step": 802
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9788321153379634e-05,
+      "loss": 0.2706,
+      "step": 804
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9787121015797672e-05,
+      "loss": 0.2975,
+      "step": 806
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9785917522267593e-05,
+      "loss": 0.2606,
+      "step": 808
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.978471067320206e-05,
+      "loss": 0.2758,
+      "step": 810
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9783500469014897e-05,
+      "loss": 0.2482,
+      "step": 812
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.978228691012108e-05,
+      "loss": 0.2318,
+      "step": 814
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9781069996936726e-05,
+      "loss": 0.2636,
+      "step": 816
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.977984972987911e-05,
+      "loss": 0.246,
+      "step": 818
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9778626109366653e-05,
+      "loss": 0.2637,
+      "step": 820
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.977739913581893e-05,
+      "loss": 0.2591,
+      "step": 822
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.977616880965666e-05,
+      "loss": 0.2591,
+      "step": 824
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9774935131301723e-05,
+      "loss": 0.2807,
+      "step": 826
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9773698101177128e-05,
+      "loss": 0.2585,
+      "step": 828
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9772457719707053e-05,
+      "loss": 0.2431,
+      "step": 830
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.977121398731682e-05,
+      "loss": 0.2624,
+      "step": 832
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.976996690443289e-05,
+      "loss": 0.2611,
+      "step": 834
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9768716471482885e-05,
+      "loss": 0.2599,
+      "step": 836
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9767462688895575e-05,
+      "loss": 0.2528,
+      "step": 838
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.976620555710087e-05,
+      "loss": 0.2623,
+      "step": 840
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.976494507652983e-05,
+      "loss": 0.2689,
+      "step": 842
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9763681247614676e-05,
+      "loss": 0.2535,
+      "step": 844
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9762414070788763e-05,
+      "loss": 0.2625,
+      "step": 846
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.97611435464866e-05,
+      "loss": 0.264,
+      "step": 848
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9759869675143835e-05,
+      "loss": 0.2597,
+      "step": 850
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.975859245719728e-05,
+      "loss": 0.2652,
+      "step": 852
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9757311893084887e-05,
+      "loss": 0.2567,
+      "step": 854
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9756027983245748e-05,
+      "loss": 0.2656,
+      "step": 856
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.975474072812011e-05,
+      "loss": 0.2495,
+      "step": 858
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9753450128149366e-05,
+      "loss": 0.2511,
+      "step": 860
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9752156183776056e-05,
+      "loss": 0.2591,
+      "step": 862
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9750858895443864e-05,
+      "loss": 0.2498,
+      "step": 864
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.974955826359763e-05,
+      "loss": 0.2596,
+      "step": 866
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9748254288683322e-05,
+      "loss": 0.2755,
+      "step": 868
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9746946971148073e-05,
+      "loss": 0.2638,
+      "step": 870
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9745636311440152e-05,
+      "loss": 0.2468,
+      "step": 872
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9744322310008983e-05,
+      "loss": 0.2735,
+      "step": 874
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.974300496730512e-05,
+      "loss": 0.2749,
+      "step": 876
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9741684283780277e-05,
+      "loss": 0.2554,
+      "step": 878
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9740360259887312e-05,
+      "loss": 0.2541,
+      "step": 880
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9739032896080217e-05,
+      "loss": 0.2583,
+      "step": 882
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9737702192814146e-05,
+      "loss": 0.2548,
+      "step": 884
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9736368150545384e-05,
+      "loss": 0.2368,
+      "step": 886
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.973503076973137e-05,
+      "loss": 0.2658,
+      "step": 888
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9733690050830684e-05,
+      "loss": 0.2506,
+      "step": 890
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9732345994303046e-05,
+      "loss": 0.2484,
+      "step": 892
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9730998600609334e-05,
+      "loss": 0.2625,
+      "step": 894
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.972964787021155e-05,
+      "loss": 0.2609,
+      "step": 896
+    },
+    {
+      "epoch": 0.41,
+      "eval_loss": 0.2601650357246399,
+      "eval_runtime": 72.3835,
+      "eval_samples_per_second": 27.631,
+      "eval_steps_per_second": 13.815,
+      "step": 896
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9728293803572862e-05,
+      "loss": 0.2638,
+      "step": 898
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.972693640115757e-05,
+      "loss": 0.2561,
+      "step": 900
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9725575663431114e-05,
+      "loss": 0.25,
+      "step": 902
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.972421159086009e-05,
+      "loss": 0.2631,
+      "step": 904
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9722844183912226e-05,
+      "loss": 0.2552,
+      "step": 906
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.97214734430564e-05,
+      "loss": 0.2427,
+      "step": 908
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9720099368762633e-05,
+      "loss": 0.2524,
+      "step": 910
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.971872196150208e-05,
+      "loss": 0.2364,
+      "step": 912
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9717341221747056e-05,
+      "loss": 0.2608,
+      "step": 914
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9715957149971006e-05,
+      "loss": 0.2473,
+      "step": 916
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9714569746648516e-05,
+      "loss": 0.2498,
+      "step": 918
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.971317901225532e-05,
+      "loss": 0.2733,
+      "step": 920
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9711784947268295e-05,
+      "loss": 0.288,
+      "step": 922
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9710387552165455e-05,
+      "loss": 0.2446,
+      "step": 924
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9708986827425963e-05,
+      "loss": 0.2588,
+      "step": 926
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.970758277353012e-05,
+      "loss": 0.2711,
+      "step": 928
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9706175390959362e-05,
+      "loss": 0.2407,
+      "step": 930
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.970476468019628e-05,
+      "loss": 0.2633,
+      "step": 932
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.970335064172459e-05,
+      "loss": 0.2498,
+      "step": 934
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.970193327602916e-05,
+      "loss": 0.2591,
+      "step": 936
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.970051258359601e-05,
+      "loss": 0.274,
+      "step": 938
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9699088564912266e-05,
+      "loss": 0.2454,
+      "step": 940
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.969766122046623e-05,
+      "loss": 0.2358,
+      "step": 942
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.969623055074733e-05,
+      "loss": 0.241,
+      "step": 944
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9694796556246125e-05,
+      "loss": 0.2615,
+      "step": 946
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.969335923745433e-05,
+      "loss": 0.2606,
+      "step": 948
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9691918594864798e-05,
+      "loss": 0.2595,
+      "step": 950
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.969047462897151e-05,
+      "loss": 0.2598,
+      "step": 952
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.96890273402696e-05,
+      "loss": 0.2775,
+      "step": 954
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9687576729255325e-05,
+      "loss": 0.2607,
+      "step": 956
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9686122796426098e-05,
+      "loss": 0.24,
+      "step": 958
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9684665542280466e-05,
+      "loss": 0.2563,
+      "step": 960
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9683204967318113e-05,
+      "loss": 0.2695,
+      "step": 962
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9681741072039858e-05,
+      "loss": 0.2557,
+      "step": 964
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.968027385694766e-05,
+      "loss": 0.2566,
+      "step": 966
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9678803322544627e-05,
+      "loss": 0.2564,
+      "step": 968
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.967732946933499e-05,
+      "loss": 0.2739,
+      "step": 970
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.967585229782413e-05,
+      "loss": 0.2804,
+      "step": 972
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9674371808518556e-05,
+      "loss": 0.2648,
+      "step": 974
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9672888001925922e-05,
+      "loss": 0.2448,
+      "step": 976
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9671400878555017e-05,
+      "loss": 0.2675,
+      "step": 978
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9669910438915763e-05,
+      "loss": 0.2558,
+      "step": 980
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9668416683519234e-05,
+      "loss": 0.2574,
+      "step": 982
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9666919612877616e-05,
+      "loss": 0.2438,
+      "step": 984
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.966541922750426e-05,
+      "loss": 0.2373,
+      "step": 986
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9663915527913628e-05,
+      "loss": 0.2555,
+      "step": 988
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9662408514621334e-05,
+      "loss": 0.2672,
+      "step": 990
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9660898188144125e-05,
+      "loss": 0.2454,
+      "step": 992
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9659384548999885e-05,
+      "loss": 0.2468,
+      "step": 994
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.965786759770763e-05,
+      "loss": 0.252,
+      "step": 996
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9656347334787515e-05,
+      "loss": 0.2619,
+      "step": 998
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9654823760760832e-05,
+      "loss": 0.2464,
+      "step": 1000
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.965329687615e-05,
+      "loss": 0.2522,
+      "step": 1002
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.965176668147858e-05,
+      "loss": 0.2478,
+      "step": 1004
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.965023317727127e-05,
+      "loss": 0.2641,
+      "step": 1006
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.96486963640539e-05,
+      "loss": 0.2489,
+      "step": 1008
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.964715624235343e-05,
+      "loss": 0.2808,
+      "step": 1010
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.964561281269797e-05,
+      "loss": 0.2694,
+      "step": 1012
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.964406607561674e-05,
+      "loss": 0.2581,
+      "step": 1014
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9642516031640114e-05,
+      "loss": 0.2449,
+      "step": 1016
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.964096268129959e-05,
+      "loss": 0.244,
+      "step": 1018
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.963940602512781e-05,
+      "loss": 0.2475,
+      "step": 1020
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9637846063658534e-05,
+      "loss": 0.2344,
+      "step": 1022
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.963628279742667e-05,
+      "loss": 0.2639,
+      "step": 1024
+    },
+    {
+      "epoch": 0.47,
+      "eval_loss": 0.25564315915107727,
+      "eval_runtime": 72.4544,
+      "eval_samples_per_second": 27.604,
+      "eval_steps_per_second": 13.802,
+      "step": 1024
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9634716226968245e-05,
+      "loss": 0.2586,
+      "step": 1026
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.963314635282044e-05,
+      "loss": 0.2533,
+      "step": 1028
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9631573175521547e-05,
+      "loss": 0.228,
+      "step": 1030
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9629996695611e-05,
+      "loss": 0.2494,
+      "step": 1032
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9628416913629368e-05,
+      "loss": 0.2493,
+      "step": 1034
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9626833830118347e-05,
+      "loss": 0.2634,
+      "step": 1036
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.962524744562077e-05,
+      "loss": 0.2609,
+      "step": 1038
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.962365776068059e-05,
+      "loss": 0.2501,
+      "step": 1040
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9622064775842916e-05,
+      "loss": 0.2661,
+      "step": 1042
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9620468491653957e-05,
+      "loss": 0.2533,
+      "step": 1044
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9618868908661088e-05,
+      "loss": 0.2502,
+      "step": 1046
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.961726602741278e-05,
+      "loss": 0.2417,
+      "step": 1048
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.961565984845866e-05,
+      "loss": 0.2497,
+      "step": 1050
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.961405037234948e-05,
+      "loss": 0.2574,
+      "step": 1052
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9612437599637114e-05,
+      "loss": 0.2437,
+      "step": 1054
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.961082153087458e-05,
+      "loss": 0.2506,
+      "step": 1056
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.960920216661601e-05,
+      "loss": 0.263,
+      "step": 1058
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9607579507416677e-05,
+      "loss": 0.2536,
+      "step": 1060
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9605953553832987e-05,
+      "loss": 0.2749,
+      "step": 1062
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.960432430642247e-05,
+      "loss": 0.2421,
+      "step": 1064
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9602691765743778e-05,
+      "loss": 0.2638,
+      "step": 1066
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9601055932356705e-05,
+      "loss": 0.2588,
+      "step": 1068
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9599416806822167e-05,
+      "loss": 0.2392,
+      "step": 1070
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9597774389702217e-05,
+      "loss": 0.2531,
+      "step": 1072
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.959612868156002e-05,
+      "loss": 0.2414,
+      "step": 1074
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.959447968295989e-05,
+      "loss": 0.2454,
+      "step": 1076
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9592827394467252e-05,
+      "loss": 0.2467,
+      "step": 1078
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.959117181664867e-05,
+      "loss": 0.2543,
+      "step": 1080
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.958951295007183e-05,
+      "loss": 0.2503,
+      "step": 1082
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9587850795305556e-05,
+      "loss": 0.2339,
+      "step": 1084
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.958618535291978e-05,
+      "loss": 0.2522,
+      "step": 1086
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9584516623485576e-05,
+      "loss": 0.2332,
+      "step": 1088
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9582844607575146e-05,
+      "loss": 0.2374,
+      "step": 1090
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.958116930576181e-05,
+      "loss": 0.2747,
+      "step": 1092
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9579490718620025e-05,
+      "loss": 0.2447,
+      "step": 1094
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9577808846725362e-05,
+      "loss": 0.2383,
+      "step": 1096
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.957612369065453e-05,
+      "loss": 0.2504,
+      "step": 1098
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.957443525098536e-05,
+      "loss": 0.241,
+      "step": 1100
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.95727435282968e-05,
+      "loss": 0.2435,
+      "step": 1102
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.957104852316895e-05,
+      "loss": 0.257,
+      "step": 1104
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9569350236182997e-05,
+      "loss": 0.2503,
+      "step": 1106
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9567648667921285e-05,
+      "loss": 0.2486,
+      "step": 1108
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9565943818967276e-05,
+      "loss": 0.2521,
+      "step": 1110
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.956423568990554e-05,
+      "loss": 0.2658,
+      "step": 1112
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9562524281321797e-05,
+      "loss": 0.2296,
+      "step": 1114
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.956080959380287e-05,
+      "loss": 0.2372,
+      "step": 1116
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.9559091627936728e-05,
+      "loss": 0.2466,
+      "step": 1118
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.9557370384312436e-05,
+      "loss": 0.243,
+      "step": 1120
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.9555645863520206e-05,
+      "loss": 0.2482,
+      "step": 1122
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.955391806615137e-05,
+      "loss": 0.2641,
+      "step": 1124
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.9552186992798377e-05,
+      "loss": 0.2276,
+      "step": 1126
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.95504526440548e-05,
+      "loss": 0.2511,
+      "step": 1128
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.954871502051534e-05,
+      "loss": 0.2468,
+      "step": 1130
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.954697412277582e-05,
+      "loss": 0.238,
+      "step": 1132
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.9545229951433176e-05,
+      "loss": 0.2449,
+      "step": 1134
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.9543482507085484e-05,
+      "loss": 0.2672,
+      "step": 1136
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.9541731790331923e-05,
+      "loss": 0.2252,
+      "step": 1138
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.9539977801772812e-05,
+      "loss": 0.2648,
+      "step": 1140
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.9538220542009586e-05,
+      "loss": 0.2548,
+      "step": 1142
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.9536460011644788e-05,
+      "loss": 0.2494,
+      "step": 1144
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.9534696211282106e-05,
+      "loss": 0.2547,
+      "step": 1146
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.9532929141526324e-05,
+      "loss": 0.2522,
+      "step": 1148
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.9531158802983374e-05,
+      "loss": 0.2443,
+      "step": 1150
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.9529385196260284e-05,
+      "loss": 0.2492,
+      "step": 1152
+    },
+    {
+      "epoch": 0.53,
+      "eval_loss": 0.25091081857681274,
+      "eval_runtime": 72.3803,
+      "eval_samples_per_second": 27.632,
+      "eval_steps_per_second": 13.816,
+      "step": 1152
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.952760832196522e-05,
+      "loss": 0.2552,
+      "step": 1154
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.952582818070746e-05,
+      "loss": 0.2584,
+      "step": 1156
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.9524044773097405e-05,
+      "loss": 0.2343,
+      "step": 1158
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.9522258099746572e-05,
+      "loss": 0.2343,
+      "step": 1160
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.952046816126761e-05,
+      "loss": 0.2291,
+      "step": 1162
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.9518674958274268e-05,
+      "loss": 0.2447,
+      "step": 1164
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.9516878491381427e-05,
+      "loss": 0.2545,
+      "step": 1166
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.9515078761205096e-05,
+      "loss": 0.2534,
+      "step": 1168
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.9513275768362377e-05,
+      "loss": 0.2465,
+      "step": 1170
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.9511469513471518e-05,
+      "loss": 0.2558,
+      "step": 1172
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.9509659997151866e-05,
+      "loss": 0.229,
+      "step": 1174
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.9507847220023898e-05,
+      "loss": 0.2398,
+      "step": 1176
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.9506031182709205e-05,
+      "loss": 0.2575,
+      "step": 1178
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.9504211885830494e-05,
+      "loss": 0.2373,
+      "step": 1180
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.950238933001159e-05,
+      "loss": 0.2696,
+      "step": 1182
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.950056351587744e-05,
+      "loss": 0.2532,
+      "step": 1184
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.9498734444054105e-05,
+      "loss": 0.2415,
+      "step": 1186
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.949690211516876e-05,
+      "loss": 0.2586,
+      "step": 1188
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.9495066529849706e-05,
+      "loss": 0.2538,
+      "step": 1190
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.9493227688726354e-05,
+      "loss": 0.2426,
+      "step": 1192
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.9491385592429223e-05,
+      "loss": 0.2238,
+      "step": 1194
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.948954024158997e-05,
+      "loss": 0.2462,
+      "step": 1196
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.9487691636841347e-05,
+      "loss": 0.2502,
+      "step": 1198
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.9485839778817236e-05,
+      "loss": 0.2393,
+      "step": 1200
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.9483984668152618e-05,
+      "loss": 0.2423,
+      "step": 1202
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.948212630548361e-05,
+      "loss": 0.2404,
+      "step": 1204
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.9480264691447434e-05,
+      "loss": 0.2564,
+      "step": 1206
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.9478399826682422e-05,
+      "loss": 0.2499,
+      "step": 1208
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.9476531711828027e-05,
+      "loss": 0.2533,
+      "step": 1210
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.9474660347524813e-05,
+      "loss": 0.2475,
+      "step": 1212
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.9472785734414465e-05,
+      "loss": 0.2572,
+      "step": 1214
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.9470907873139776e-05,
+      "loss": 0.229,
+      "step": 1216
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.946902676434465e-05,
+      "loss": 0.2612,
+      "step": 1218
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.9467142408674112e-05,
+      "loss": 0.2289,
+      "step": 1220
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.9465254806774294e-05,
+      "loss": 0.2468,
+      "step": 1222
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.946336395929245e-05,
+      "loss": 0.2657,
+      "step": 1224
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.9461469866876935e-05,
+      "loss": 0.2509,
+      "step": 1226
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.9459572530177223e-05,
+      "loss": 0.2474,
+      "step": 1228
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.94576719498439e-05,
+      "loss": 0.2605,
+      "step": 1230
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.945576812652867e-05,
+      "loss": 0.252,
+      "step": 1232
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.9453861060884332e-05,
+      "loss": 0.2506,
+      "step": 1234
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.9451950753564823e-05,
+      "loss": 0.2251,
+      "step": 1236
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.945003720522516e-05,
+      "loss": 0.2406,
+      "step": 1238
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.9448120416521502e-05,
+      "loss": 0.2568,
+      "step": 1240
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.9446200388111093e-05,
+      "loss": 0.2572,
+      "step": 1242
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.944427712065231e-05,
+      "loss": 0.2424,
+      "step": 1244
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.9442350614804626e-05,
+      "loss": 0.2267,
+      "step": 1246
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.9440420871228625e-05,
+      "loss": 0.2624,
+      "step": 1248
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.9438487890586014e-05,
+      "loss": 0.2508,
+      "step": 1250
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.9436551673539595e-05,
+      "loss": 0.264,
+      "step": 1252
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.9434612220753288e-05,
+      "loss": 0.2488,
+      "step": 1254
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.9432669532892113e-05,
+      "loss": 0.2582,
+      "step": 1256
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.943072361062222e-05,
+      "loss": 0.23,
+      "step": 1258
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.9428774454610845e-05,
+      "loss": 0.2338,
+      "step": 1260
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.9426822065526342e-05,
+      "loss": 0.2617,
+      "step": 1262
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.9424866444038183e-05,
+      "loss": 0.256,
+      "step": 1264
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.9422907590816932e-05,
+      "loss": 0.2413,
+      "step": 1266
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9420945506534266e-05,
+      "loss": 0.2348,
+      "step": 1268
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.941898019186298e-05,
+      "loss": 0.2504,
+      "step": 1270
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9417011647476966e-05,
+      "loss": 0.2458,
+      "step": 1272
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9415039874051228e-05,
+      "loss": 0.2574,
+      "step": 1274
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9413064872261875e-05,
+      "loss": 0.2523,
+      "step": 1276
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.941108664278612e-05,
+      "loss": 0.2374,
+      "step": 1278
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9409105186302292e-05,
+      "loss": 0.2323,
+      "step": 1280
+    },
+    {
+      "epoch": 0.59,
+      "eval_loss": 0.24687859416007996,
+      "eval_runtime": 72.3804,
+      "eval_samples_per_second": 27.632,
+      "eval_steps_per_second": 13.816,
+      "step": 1280
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9407120503489822e-05,
+      "loss": 0.2518,
+      "step": 1282
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.940513259502924e-05,
+      "loss": 0.2287,
+      "step": 1284
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9403141461602195e-05,
+      "loss": 0.2444,
+      "step": 1286
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9401147103891432e-05,
+      "loss": 0.2511,
+      "step": 1288
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.9399149522580805e-05,
+      "loss": 0.2462,
+      "step": 1290
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.9397148718355275e-05,
+      "loss": 0.2456,
+      "step": 1292
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.9395144691900903e-05,
+      "loss": 0.2521,
+      "step": 1294
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.9393137443904862e-05,
+      "loss": 0.2409,
+      "step": 1296
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.939112697505542e-05,
+      "loss": 0.2379,
+      "step": 1298
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.9389113286041965e-05,
+      "loss": 0.2286,
+      "step": 1300
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.9387096377554974e-05,
+      "loss": 0.2461,
+      "step": 1302
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.938507625028603e-05,
+      "loss": 0.243,
+      "step": 1304
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.938305290492783e-05,
+      "loss": 0.2404,
+      "step": 1306
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.9381026342174166e-05,
+      "loss": 0.2564,
+      "step": 1308
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.937899656271993e-05,
+      "loss": 0.2443,
+      "step": 1310
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.9376963567261126e-05,
+      "loss": 0.2426,
+      "step": 1312
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.9374927356494856e-05,
+      "loss": 0.2466,
+      "step": 1314
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.9372887931119324e-05,
+      "loss": 0.2374,
+      "step": 1316
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.937084529183384e-05,
+      "loss": 0.237,
+      "step": 1318
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.936879943933881e-05,
+      "loss": 0.262,
+      "step": 1320
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.9366750374335744e-05,
+      "loss": 0.2303,
+      "step": 1322
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.936469809752726e-05,
+      "loss": 0.2538,
+      "step": 1324
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.936264260961707e-05,
+      "loss": 0.2223,
+      "step": 1326
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.9360583911309987e-05,
+      "loss": 0.2375,
+      "step": 1328
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.9358522003311927e-05,
+      "loss": 0.2316,
+      "step": 1330
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.9356456886329912e-05,
+      "loss": 0.2304,
+      "step": 1332
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.9354388561072055e-05,
+      "loss": 0.2346,
+      "step": 1334
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.935231702824757e-05,
+      "loss": 0.2235,
+      "step": 1336
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.935024228856678e-05,
+      "loss": 0.251,
+      "step": 1338
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.9348164342741095e-05,
+      "loss": 0.2684,
+      "step": 1340
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.9346083191483034e-05,
+      "loss": 0.2389,
+      "step": 1342
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.9343998835506216e-05,
+      "loss": 0.2309,
+      "step": 1344
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.9341911275525346e-05,
+      "loss": 0.2598,
+      "step": 1346
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.933982051225624e-05,
+      "loss": 0.2386,
+      "step": 1348
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.9337726546415815e-05,
+      "loss": 0.2338,
+      "step": 1350
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.9335629378722072e-05,
+      "loss": 0.2506,
+      "step": 1352
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.9333529009894124e-05,
+      "loss": 0.2303,
+      "step": 1354
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.9331425440652168e-05,
+      "loss": 0.2377,
+      "step": 1356
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.932931867171751e-05,
+      "loss": 0.2544,
+      "step": 1358
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.9327208703812553e-05,
+      "loss": 0.2491,
+      "step": 1360
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.9325095537660786e-05,
+      "loss": 0.2784,
+      "step": 1362
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.9322979173986805e-05,
+      "loss": 0.2349,
+      "step": 1364
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.9320859613516302e-05,
+      "loss": 0.2275,
+      "step": 1366
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.931873685697606e-05,
+      "loss": 0.2353,
+      "step": 1368
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.9316610905093957e-05,
+      "loss": 0.2353,
+      "step": 1370
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.931448175859897e-05,
+      "loss": 0.2595,
+      "step": 1372
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.9312349418221178e-05,
+      "loss": 0.2404,
+      "step": 1374
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.931021388469174e-05,
+      "loss": 0.2353,
+      "step": 1376
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.9308075158742926e-05,
+      "loss": 0.2686,
+      "step": 1378
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.9305933241108086e-05,
+      "loss": 0.233,
+      "step": 1380
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.9303788132521673e-05,
+      "loss": 0.2556,
+      "step": 1382
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.9301639833719238e-05,
+      "loss": 0.2349,
+      "step": 1384
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.9299488345437415e-05,
+      "loss": 0.2277,
+      "step": 1386
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.9297333668413935e-05,
+      "loss": 0.2305,
+      "step": 1388
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.9295175803387633e-05,
+      "loss": 0.2523,
+      "step": 1390
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.929301475109842e-05,
+      "loss": 0.2194,
+      "step": 1392
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.9290850512287314e-05,
+      "loss": 0.2457,
+      "step": 1394
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.928868308769642e-05,
+      "loss": 0.2444,
+      "step": 1396
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.928651247806893e-05,
+      "loss": 0.22,
+      "step": 1398
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.928433868414914e-05,
+      "loss": 0.2311,
+      "step": 1400
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.928216170668243e-05,
+      "loss": 0.2478,
+      "step": 1402
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.927998154641527e-05,
+      "loss": 0.2358,
+      "step": 1404
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.927779820409523e-05,
+      "loss": 0.2435,
+      "step": 1406
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.9275611680470958e-05,
+      "loss": 0.2449,
+      "step": 1408
+    },
+    {
+      "epoch": 0.65,
+      "eval_loss": 0.24367311596870422,
+      "eval_runtime": 72.3787,
+      "eval_samples_per_second": 27.632,
+      "eval_steps_per_second": 13.816,
+      "step": 1408
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.927342197629221e-05,
+      "loss": 0.2655,
+      "step": 1410
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.9271229092309815e-05,
+      "loss": 0.2374,
+      "step": 1412
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.9269033029275707e-05,
+      "loss": 0.2326,
+      "step": 1414
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.9266833787942896e-05,
+      "loss": 0.2482,
+      "step": 1416
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.9264631369065496e-05,
+      "loss": 0.242,
+      "step": 1418
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.92624257733987e-05,
+      "loss": 0.2483,
+      "step": 1420
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.9260217001698798e-05,
+      "loss": 0.2361,
+      "step": 1422
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.925800505472316e-05,
+      "loss": 0.2291,
+      "step": 1424
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.9255789933230255e-05,
+      "loss": 0.2274,
+      "step": 1426
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.9253571637979634e-05,
+      "loss": 0.2466,
+      "step": 1428
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.9251350169731935e-05,
+      "loss": 0.2349,
+      "step": 1430
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.924912552924889e-05,
+      "loss": 0.2388,
+      "step": 1432
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.9246897717293316e-05,
+      "loss": 0.2263,
+      "step": 1434
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.9244666734629113e-05,
+      "loss": 0.2463,
+      "step": 1436
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.9242432582021276e-05,
+      "loss": 0.2345,
+      "step": 1438
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.9240195260235883e-05,
+      "loss": 0.2423,
+      "step": 1440
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.92379547700401e-05,
+      "loss": 0.263,
+      "step": 1442
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.9235711112202175e-05,
+      "loss": 0.2587,
+      "step": 1444
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.9233464287491442e-05,
+      "loss": 0.2464,
+      "step": 1446
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.9231214296678336e-05,
+      "loss": 0.2209,
+      "step": 1448
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.922896114053436e-05,
+      "loss": 0.2197,
+      "step": 1450
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.9226704819832105e-05,
+      "loss": 0.2297,
+      "step": 1452
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.922444533534525e-05,
+      "loss": 0.2422,
+      "step": 1454
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.9222182687848566e-05,
+      "loss": 0.2313,
+      "step": 1456
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.92199168781179e-05,
+      "loss": 0.2272,
+      "step": 1458
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.9217647906930183e-05,
+      "loss": 0.2228,
+      "step": 1460
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.9215375775063435e-05,
+      "loss": 0.2249,
+      "step": 1462
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.9213100483296753e-05,
+      "loss": 0.2395,
+      "step": 1464
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.9210822032410326e-05,
+      "loss": 0.2388,
+      "step": 1466
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.920854042318542e-05,
+      "loss": 0.2533,
+      "step": 1468
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.9206255656404384e-05,
+      "loss": 0.2501,
+      "step": 1470
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.9203967732850654e-05,
+      "loss": 0.2375,
+      "step": 1472
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.9201676653308744e-05,
+      "loss": 0.2354,
+      "step": 1474
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.9199382418564253e-05,
+      "loss": 0.2363,
+      "step": 1476
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.9197085029403863e-05,
+      "loss": 0.219,
+      "step": 1478
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.9194784486615333e-05,
+      "loss": 0.2297,
+      "step": 1480
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.919248079098751e-05,
+      "loss": 0.2356,
+      "step": 1482
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.919017394331031e-05,
+      "loss": 0.2484,
+      "step": 1484
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.9187863944374748e-05,
+      "loss": 0.2332,
+      "step": 1486
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.91855507949729e-05,
+      "loss": 0.2445,
+      "step": 1488
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.918323449589794e-05,
+      "loss": 0.2471,
+      "step": 1490
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.918091504794411e-05,
+      "loss": 0.2387,
+      "step": 1492
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.9178592451906736e-05,
+      "loss": 0.2299,
+      "step": 1494
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.9176266708582222e-05,
+      "loss": 0.2534,
+      "step": 1496
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.917393781876805e-05,
+      "loss": 0.2445,
+      "step": 1498
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.917160578326279e-05,
+      "loss": 0.2147,
+      "step": 1500
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.9169270602866075e-05,
+      "loss": 0.2378,
+      "step": 1502
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.9166932278378637e-05,
+      "loss": 0.2469,
+      "step": 1504
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.9164590810602262e-05,
+      "loss": 0.2211,
+      "step": 1506
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.9162246200339833e-05,
+      "loss": 0.2548,
+      "step": 1508
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.91598984483953e-05,
+      "loss": 0.2364,
+      "step": 1510
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.9157547555573693e-05,
+      "loss": 0.2592,
+      "step": 1512
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.9155193522681125e-05,
+      "loss": 0.24,
+      "step": 1514
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.9152836350524774e-05,
+      "loss": 0.2107,
+      "step": 1516
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.9150476039912907e-05,
+      "loss": 0.2043,
+      "step": 1518
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.9148112591654858e-05,
+      "loss": 0.2333,
+      "step": 1520
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.914574600656104e-05,
+      "loss": 0.2306,
+      "step": 1522
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.9143376285442942e-05,
+      "loss": 0.239,
+      "step": 1524
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.914100342911313e-05,
+      "loss": 0.2288,
+      "step": 1526
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.9138627438385236e-05,
+      "loss": 0.2177,
+      "step": 1528
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.9136248314073985e-05,
+      "loss": 0.2385,
+      "step": 1530
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.9133866056995153e-05,
+      "loss": 0.2373,
+      "step": 1532
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.913148066796561e-05,
+      "loss": 0.2296,
+      "step": 1534
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.9129092147803293e-05,
+      "loss": 0.2227,
+      "step": 1536
+    },
+    {
+      "epoch": 0.71,
+      "eval_loss": 0.24045754969120026,
+      "eval_runtime": 72.382,
+      "eval_samples_per_second": 27.631,
+      "eval_steps_per_second": 13.816,
+      "step": 1536
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.9126700497327207e-05,
+      "loss": 0.2561,
+      "step": 1538
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.9124305717357437e-05,
+      "loss": 0.2363,
+      "step": 1540
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.912190780871514e-05,
+      "loss": 0.2487,
+      "step": 1542
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.9119506772222545e-05,
+      "loss": 0.238,
+      "step": 1544
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.9117102608702957e-05,
+      "loss": 0.2264,
+      "step": 1546
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.9114695318980744e-05,
+      "loss": 0.2312,
+      "step": 1548
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.911228490388136e-05,
+      "loss": 0.2319,
+      "step": 1550
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.910987136423131e-05,
+      "loss": 0.2352,
+      "step": 1552
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.9107454700858195e-05,
+      "loss": 0.2349,
+      "step": 1554
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.9105034914590673e-05,
+      "loss": 0.2323,
+      "step": 1556
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.910261200625847e-05,
+      "loss": 0.235,
+      "step": 1558
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.910018597669239e-05,
+      "loss": 0.2397,
+      "step": 1560
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.909775682672431e-05,
+      "loss": 0.2248,
+      "step": 1562
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.9095324557187162e-05,
+      "loss": 0.227,
+      "step": 1564
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.9092889168914963e-05,
+      "loss": 0.2371,
+      "step": 1566
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.909045066274279e-05,
+      "loss": 0.2354,
+      "step": 1568
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.90880090395068e-05,
+      "loss": 0.2303,
+      "step": 1570
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.9085564300044207e-05,
+      "loss": 0.2489,
+      "step": 1572
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.9083116445193298e-05,
+      "loss": 0.2434,
+      "step": 1574
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.908066547579343e-05,
+      "loss": 0.2321,
+      "step": 1576
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.9078211392685023e-05,
+      "loss": 0.2345,
+      "step": 1578
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.9075754196709574e-05,
+      "loss": 0.2289,
+      "step": 1580
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.9073293888709635e-05,
+      "loss": 0.2493,
+      "step": 1582
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.9070830469528837e-05,
+      "loss": 0.2418,
+      "step": 1584
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.906836394001187e-05,
+      "loss": 0.2396,
+      "step": 1586
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.9065894301004495e-05,
+      "loss": 0.2209,
+      "step": 1588
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.9063421553353535e-05,
+      "loss": 0.2346,
+      "step": 1590
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.906094569790688e-05,
+      "loss": 0.242,
+      "step": 1592
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.905846673551349e-05,
+      "loss": 0.2474,
+      "step": 1594
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.9055984667023384e-05,
+      "loss": 0.2285,
+      "step": 1596
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.9053499493287652e-05,
+      "loss": 0.2343,
+      "step": 1598
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.9051011215158445e-05,
+      "loss": 0.2216,
+      "step": 1600
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.904851983348898e-05,
+      "loss": 0.2411,
+      "step": 1602
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.9046025349133532e-05,
+      "loss": 0.2345,
+      "step": 1604
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.9043527762947453e-05,
+      "loss": 0.2258,
+      "step": 1606
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.9041027075787148e-05,
+      "loss": 0.2401,
+      "step": 1608
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.9038523288510088e-05,
+      "loss": 0.2172,
+      "step": 1610
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.9036016401974812e-05,
+      "loss": 0.2324,
+      "step": 1612
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.9033506417040906e-05,
+      "loss": 0.2354,
+      "step": 1614
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.903099333456904e-05,
+      "loss": 0.2397,
+      "step": 1616
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.9028477155420937e-05,
+      "loss": 0.2489,
+      "step": 1618
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.902595788045937e-05,
+      "loss": 0.225,
+      "step": 1620
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.902343551054819e-05,
+      "loss": 0.2421,
+      "step": 1622
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.9020910046552308e-05,
+      "loss": 0.2372,
+      "step": 1624
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.901838148933768e-05,
+      "loss": 0.2307,
+      "step": 1626
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.9015849839771346e-05,
+      "loss": 0.2423,
+      "step": 1628
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.901331509872139e-05,
+      "loss": 0.2413,
+      "step": 1630
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.9010777267056954e-05,
+      "loss": 0.2295,
+      "step": 1632
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.900823634564825e-05,
+      "loss": 0.235,
+      "step": 1634
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.900569233536655e-05,
+      "loss": 0.2314,
+      "step": 1636
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.9003145237084177e-05,
+      "loss": 0.2272,
+      "step": 1638
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.9000595051674518e-05,
+      "loss": 0.2439,
+      "step": 1640
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.899804178001201e-05,
+      "loss": 0.2363,
+      "step": 1642
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.8995485422972167e-05,
+      "loss": 0.2119,
+      "step": 1644
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.8992925981431537e-05,
+      "loss": 0.2381,
+      "step": 1646
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.899036345626775e-05,
+      "loss": 0.2339,
+      "step": 1648
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.8987797848359472e-05,
+      "loss": 0.2388,
+      "step": 1650
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.8985229158586443e-05,
+      "loss": 0.259,
+      "step": 1652
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.8982657387829447e-05,
+      "loss": 0.2376,
+      "step": 1654
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.898008253697033e-05,
+      "loss": 0.2205,
+      "step": 1656
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.897750460689199e-05,
+      "loss": 0.2321,
+      "step": 1658
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8974923598478393e-05,
+      "loss": 0.223,
+      "step": 1660
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8972339512614546e-05,
+      "loss": 0.2296,
+      "step": 1662
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8969752350186522e-05,
+      "loss": 0.2341,
+      "step": 1664
+    },
+    {
+      "epoch": 0.77,
+      "eval_loss": 0.236300989985466,
+      "eval_runtime": 72.4186,
+      "eval_samples_per_second": 27.617,
+      "eval_steps_per_second": 13.809,
+      "step": 1664
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.896716211208144e-05,
+      "loss": 0.2187,
+      "step": 1666
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.896456879918748e-05,
+      "loss": 0.2261,
+      "step": 1668
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8961972412393873e-05,
+      "loss": 0.22,
+      "step": 1670
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8959372952590905e-05,
+      "loss": 0.2644,
+      "step": 1672
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.895677042066992e-05,
+      "loss": 0.2189,
+      "step": 1674
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8954164817523302e-05,
+      "loss": 0.2389,
+      "step": 1676
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8951556144044508e-05,
+      "loss": 0.2635,
+      "step": 1678
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8948944401128035e-05,
+      "loss": 0.2316,
+      "step": 1680
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8946329589669435e-05,
+      "loss": 0.2255,
+      "step": 1682
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8943711710565304e-05,
+      "loss": 0.235,
+      "step": 1684
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8941090764713305e-05,
+      "loss": 0.2373,
+      "step": 1686
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8938466753012146e-05,
+      "loss": 0.2495,
+      "step": 1688
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8935839676361584e-05,
+      "loss": 0.2171,
+      "step": 1690
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8933209535662428e-05,
+      "loss": 0.2353,
+      "step": 1692
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8930576331816537e-05,
+      "loss": 0.2287,
+      "step": 1694
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8927940065726824e-05,
+      "loss": 0.2233,
+      "step": 1696
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.892530073829725e-05,
+      "loss": 0.2333,
+      "step": 1698
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.8922658350432827e-05,
+      "loss": 0.2294,
+      "step": 1700
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.8920012903039607e-05,
+      "loss": 0.2312,
+      "step": 1702
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.891736439702471e-05,
+      "loss": 0.2141,
+      "step": 1704
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.8914712833296282e-05,
+      "loss": 0.2151,
+      "step": 1706
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.8912058212763536e-05,
+      "loss": 0.2394,
+      "step": 1708
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.8909400536336728e-05,
+      "loss": 0.2282,
+      "step": 1710
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.8906739804927156e-05,
+      "loss": 0.2307,
+      "step": 1712
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.890407601944717e-05,
+      "loss": 0.2259,
+      "step": 1714
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.890140918081017e-05,
+      "loss": 0.2399,
+      "step": 1716
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.8898739289930595e-05,
+      "loss": 0.2305,
+      "step": 1718
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.889606634772394e-05,
+      "loss": 0.2286,
+      "step": 1720
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.889339035510674e-05,
+      "loss": 0.2235,
+      "step": 1722
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.889071131299657e-05,
+      "loss": 0.2396,
+      "step": 1724
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8888029222312073e-05,
+      "loss": 0.2413,
+      "step": 1726
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8885344083972912e-05,
+      "loss": 0.2108,
+      "step": 1728
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8882655898899812e-05,
+      "loss": 0.2417,
+      "step": 1730
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8879964668014533e-05,
+      "loss": 0.2345,
+      "step": 1732
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.887727039223988e-05,
+      "loss": 0.2557,
+      "step": 1734
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.887457307249971e-05,
+      "loss": 0.2487,
+      "step": 1736
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8871872709718916e-05,
+      "loss": 0.2311,
+      "step": 1738
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8869169304823438e-05,
+      "loss": 0.2278,
+      "step": 1740
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.8866462858740256e-05,
+      "loss": 0.2295,
+      "step": 1742
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.88637533723974e-05,
+      "loss": 0.226,
+      "step": 1744
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8861040846723933e-05,
+      "loss": 0.2377,
+      "step": 1746
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8858325282649962e-05,
+      "loss": 0.2391,
+      "step": 1748
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8855606681106646e-05,
+      "loss": 0.2423,
+      "step": 1750
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8852885043026173e-05,
+      "loss": 0.2233,
+      "step": 1752
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.885016036934178e-05,
+      "loss": 0.2109,
+      "step": 1754
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8847432660987737e-05,
+      "loss": 0.2403,
+      "step": 1756
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8844701918899365e-05,
+      "loss": 0.2277,
+      "step": 1758
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.884196814401302e-05,
+      "loss": 0.226,
+      "step": 1760
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8839231337266092e-05,
+      "loss": 0.2285,
+      "step": 1762
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8836491499597024e-05,
+      "loss": 0.2305,
+      "step": 1764
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.8833748631945285e-05,
+      "loss": 0.2196,
+      "step": 1766
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.8831002735251394e-05,
+      "loss": 0.2329,
+      "step": 1768
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.8828253810456898e-05,
+      "loss": 0.2288,
+      "step": 1770
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.882550185850439e-05,
+      "loss": 0.2164,
+      "step": 1772
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.8822746880337498e-05,
+      "loss": 0.2352,
+      "step": 1774
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.881998887690089e-05,
+      "loss": 0.2467,
+      "step": 1776
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.8817227849140268e-05,
+      "loss": 0.228,
+      "step": 1778
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.881446379800237e-05,
+      "loss": 0.2221,
+      "step": 1780
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.881169672443498e-05,
+      "loss": 0.2264,
+      "step": 1782
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.8808926629386913e-05,
+      "loss": 0.2297,
+      "step": 1784
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.880615351380801e-05,
+      "loss": 0.2466,
+      "step": 1786
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.880337737864916e-05,
+      "loss": 0.2308,
+      "step": 1788
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8800598224862286e-05,
+      "loss": 0.2293,
+      "step": 1790
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.879781605340034e-05,
+      "loss": 0.215,
+      "step": 1792
+    },
+    {
+      "epoch": 0.83,
+      "eval_loss": 0.23247940838336945,
+      "eval_runtime": 72.3922,
+      "eval_samples_per_second": 27.627,
+      "eval_steps_per_second": 13.814,
+      "step": 1792
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.879503086521732e-05,
+      "loss": 0.236,
+      "step": 1794
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8792242661268245e-05,
+      "loss": 0.2149,
+      "step": 1796
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8789451442509178e-05,
+      "loss": 0.2379,
+      "step": 1798
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8786657209897207e-05,
+      "loss": 0.2313,
+      "step": 1800
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8783859964390466e-05,
+      "loss": 0.2401,
+      "step": 1802
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8781059706948107e-05,
+      "loss": 0.226,
+      "step": 1804
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8778256438530325e-05,
+      "loss": 0.2095,
+      "step": 1806
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.8775450160098348e-05,
+      "loss": 0.2364,
+      "step": 1808
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.877264087261443e-05,
+      "loss": 0.2335,
+      "step": 1810
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8769828577041857e-05,
+      "loss": 0.2393,
+      "step": 1812
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8767013274344953e-05,
+      "loss": 0.2195,
+      "step": 1814
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.876419496548907e-05,
+      "loss": 0.2439,
+      "step": 1816
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8761373651440587e-05,
+      "loss": 0.2258,
+      "step": 1818
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.875854933316692e-05,
+      "loss": 0.2439,
+      "step": 1820
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.875572201163651e-05,
+      "loss": 0.2392,
+      "step": 1822
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8752891687818827e-05,
+      "loss": 0.2211,
+      "step": 1824
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8750058362684377e-05,
+      "loss": 0.2181,
+      "step": 1826
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8747222037204685e-05,
+      "loss": 0.2282,
+      "step": 1828
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8744382712352317e-05,
+      "loss": 0.2194,
+      "step": 1830
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.8741540389100864e-05,
+      "loss": 0.2285,
+      "step": 1832
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.873869506842493e-05,
+      "loss": 0.228,
+      "step": 1834
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.8735846751300173e-05,
+      "loss": 0.2294,
+      "step": 1836
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.8732995438703257e-05,
+      "loss": 0.2078,
+      "step": 1838
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.8730141131611882e-05,
+      "loss": 0.2213,
+      "step": 1840
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.8727283831004776e-05,
+      "loss": 0.2228,
+      "step": 1842
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.872442353786169e-05,
+      "loss": 0.2226,
+      "step": 1844
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.8721560253163397e-05,
+      "loss": 0.2256,
+      "step": 1846
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.8718693977891713e-05,
+      "loss": 0.2284,
+      "step": 1848
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.8715824713029455e-05,
+      "loss": 0.2129,
+      "step": 1850
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.8712952459560484e-05,
+      "loss": 0.2169,
+      "step": 1852
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.8710077218469678e-05,
+      "loss": 0.2283,
+      "step": 1854
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.8707198990742946e-05,
+      "loss": 0.2238,
+      "step": 1856
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.8704317777367204e-05,
+      "loss": 0.2272,
+      "step": 1858
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.8701433579330414e-05,
+      "loss": 0.2496,
+      "step": 1860
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.8698546397621548e-05,
+      "loss": 0.2411,
+      "step": 1862
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.86956562332306e-05,
+      "loss": 0.2187,
+      "step": 1864
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.86927630871486e-05,
+      "loss": 0.2472,
+      "step": 1866
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.8689866960367588e-05,
+      "loss": 0.2238,
+      "step": 1868
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.868696785388062e-05,
+      "loss": 0.2051,
+      "step": 1870
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.8684065768681794e-05,
+      "loss": 0.2147,
+      "step": 1872
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.8681160705766216e-05,
+      "loss": 0.2218,
+      "step": 1874
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.8678252666130016e-05,
+      "loss": 0.2313,
+      "step": 1876
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.867534165077034e-05,
+      "loss": 0.225,
+      "step": 1878
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.8672427660685365e-05,
+      "loss": 0.2289,
+      "step": 1880
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.866951069687428e-05,
+      "loss": 0.2436,
+      "step": 1882
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.8666590760337286e-05,
+      "loss": 0.225,
+      "step": 1884
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.8663667852075624e-05,
+      "loss": 0.2141,
+      "step": 1886
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.866074197309154e-05,
+      "loss": 0.2254,
+      "step": 1888
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.86578131243883e-05,
+      "loss": 0.2211,
+      "step": 1890
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.8654881306970186e-05,
+      "loss": 0.2213,
+      "step": 1892
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.865194652184251e-05,
+      "loss": 0.2291,
+      "step": 1894
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.8649008770011586e-05,
+      "loss": 0.2229,
+      "step": 1896
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.8646068052484757e-05,
+      "loss": 0.2428,
+      "step": 1898
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.8643124370270373e-05,
+      "loss": 0.2211,
+      "step": 1900
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.864017772437781e-05,
+      "loss": 0.222,
+      "step": 1902
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.8637228115817456e-05,
+      "loss": 0.2247,
+      "step": 1904
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.8634275545600712e-05,
+      "loss": 0.2186,
+      "step": 1906
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.863132001474e-05,
+      "loss": 0.238,
+      "step": 1908
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.8628361524248755e-05,
+      "loss": 0.2403,
+      "step": 1910
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.8625400075141424e-05,
+      "loss": 0.2207,
+      "step": 1912
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.8622435668433474e-05,
+      "loss": 0.2229,
+      "step": 1914
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.8619468305141382e-05,
+      "loss": 0.2401,
+      "step": 1916
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.8616497986282642e-05,
+      "loss": 0.2203,
+      "step": 1918
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.8613524712875755e-05,
+      "loss": 0.2221,
+      "step": 1920
+    },
+    {
+      "epoch": 0.89,
+      "eval_loss": 0.22889545559883118,
+      "eval_runtime": 72.3869,
+      "eval_samples_per_second": 27.629,
+      "eval_steps_per_second": 13.815,
+      "step": 1920
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.8610548485940243e-05,
+      "loss": 0.2083,
+      "step": 1922
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.8607569306496636e-05,
+      "loss": 0.2379,
+      "step": 1924
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.860458717556648e-05,
+      "loss": 0.2306,
+      "step": 1926
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.8601602094172327e-05,
+      "loss": 0.2388,
+      "step": 1928
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.8598614063337743e-05,
+      "loss": 0.2163,
+      "step": 1930
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.8595623084087317e-05,
+      "loss": 0.2259,
+      "step": 1932
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.8592629157446625e-05,
+      "loss": 0.2364,
+      "step": 1934
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.8589632284442273e-05,
+      "loss": 0.2242,
+      "step": 1936
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.858663246610188e-05,
+      "loss": 0.2438,
+      "step": 1938
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8583629703454055e-05,
+      "loss": 0.2384,
+      "step": 1940
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8580623997528434e-05,
+      "loss": 0.2157,
+      "step": 1942
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8577615349355653e-05,
+      "loss": 0.2074,
+      "step": 1944
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8574603759967364e-05,
+      "loss": 0.2244,
+      "step": 1946
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8571589230396225e-05,
+      "loss": 0.2175,
+      "step": 1948
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8568571761675893e-05,
+      "loss": 0.2174,
+      "step": 1950
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.856555135484105e-05,
+      "loss": 0.2256,
+      "step": 1952
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8562528010927373e-05,
+      "loss": 0.2206,
+      "step": 1954
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8559501730971544e-05,
+      "loss": 0.2201,
+      "step": 1956
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.855647251601127e-05,
+      "loss": 0.2254,
+      "step": 1958
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8553440367085238e-05,
+      "loss": 0.2283,
+      "step": 1960
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.8550405285233163e-05,
+      "loss": 0.2241,
+      "step": 1962
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.854736727149575e-05,
+      "loss": 0.2291,
+      "step": 1964
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.8544326326914727e-05,
+      "loss": 0.2115,
+      "step": 1966
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.8541282452532813e-05,
+      "loss": 0.237,
+      "step": 1968
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.8538235649393727e-05,
+      "loss": 0.22,
+      "step": 1970
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.8535185918542207e-05,
+      "loss": 0.2197,
+      "step": 1972
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.853213326102399e-05,
+      "loss": 0.2335,
+      "step": 1974
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.8529077677885813e-05,
+      "loss": 0.2326,
+      "step": 1976
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.8526019170175418e-05,
+      "loss": 0.2211,
+      "step": 1978
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.852295773894155e-05,
+      "loss": 0.2189,
+      "step": 1980
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.8519893385233955e-05,
+      "loss": 0.2301,
+      "step": 1982
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.8516826110103383e-05,
+      "loss": 0.2165,
+      "step": 1984
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.8513755914601588e-05,
+      "loss": 0.2139,
+      "step": 1986
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.8510682799781316e-05,
+      "loss": 0.2299,
+      "step": 1988
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.850760676669633e-05,
+      "loss": 0.2289,
+      "step": 1990
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.850452781640138e-05,
+      "loss": 0.2201,
+      "step": 1992
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.8501445949952216e-05,
+      "loss": 0.2236,
+      "step": 1994
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.84983611684056e-05,
+      "loss": 0.2391,
+      "step": 1996
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.849527347281928e-05,
+      "loss": 0.2301,
+      "step": 1998
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.849218286425201e-05,
+      "loss": 0.2144,
+      "step": 2000
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.8489089343763543e-05,
+      "loss": 0.2209,
+      "step": 2002
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.848599291241463e-05,
+      "loss": 0.2307,
+      "step": 2004
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.8482893571267026e-05,
+      "loss": 0.244,
+      "step": 2006
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.8479791321383468e-05,
+      "loss": 0.2026,
+      "step": 2008
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.84766861638277e-05,
+      "loss": 0.237,
+      "step": 2010
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.847357809966447e-05,
+      "loss": 0.2394,
+      "step": 2012
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.847046712995951e-05,
+      "loss": 0.2221,
+      "step": 2014
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.8467353255779556e-05,
+      "loss": 0.2243,
+      "step": 2016
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.8464236478192336e-05,
+      "loss": 0.2218,
+      "step": 2018
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.846111679826658e-05,
+      "loss": 0.2222,
+      "step": 2020
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.8457994217072002e-05,
+      "loss": 0.2132,
+      "step": 2022
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.845486873567932e-05,
+      "loss": 0.2116,
+      "step": 2024
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.8451740355160245e-05,
+      "loss": 0.2339,
+      "step": 2026
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.8448609076587482e-05,
+      "loss": 0.2219,
+      "step": 2028
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.844547490103472e-05,
+      "loss": 0.2114,
+      "step": 2030
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.844233782957666e-05,
+      "loss": 0.221,
+      "step": 2032
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.8439197863288986e-05,
+      "loss": 0.2165,
+      "step": 2034
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.8436055003248366e-05,
+      "loss": 0.2168,
+      "step": 2036
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.8432909250532477e-05,
+      "loss": 0.2396,
+      "step": 2038
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.8429760606219974e-05,
+      "loss": 0.234,
+      "step": 2040
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.8426609071390514e-05,
+      "loss": 0.2187,
+      "step": 2042
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.842345464712474e-05,
+      "loss": 0.2238,
+      "step": 2044
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.8420297334504284e-05,
+      "loss": 0.2184,
+      "step": 2046
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.841713713461177e-05,
+      "loss": 0.2345,
+      "step": 2048
+    },
+    {
+      "epoch": 0.95,
+      "eval_loss": 0.2246369868516922,
+      "eval_runtime": 72.3804,
+      "eval_samples_per_second": 27.632,
+      "eval_steps_per_second": 13.816,
+      "step": 2048
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.8413974048530813e-05,
+      "loss": 0.2353,
+      "step": 2050
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.8410808077346012e-05,
+      "loss": 0.2038,
+      "step": 2052
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.8407639222142976e-05,
+      "loss": 0.2132,
+      "step": 2054
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.8404467484008273e-05,
+      "loss": 0.2213,
+      "step": 2056
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.8401292864029477e-05,
+      "loss": 0.2345,
+      "step": 2058
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.8398115363295152e-05,
+      "loss": 0.2291,
+      "step": 2060
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.8394934982894835e-05,
+      "loss": 0.2152,
+      "step": 2062
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.8391751723919075e-05,
+      "loss": 0.2108,
+      "step": 2064
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.8388565587459382e-05,
+      "loss": 0.239,
+      "step": 2066
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.8385376574608265e-05,
+      "loss": 0.2157,
+      "step": 2068
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.8382184686459225e-05,
+      "loss": 0.2311,
+      "step": 2070
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.837898992410674e-05,
+      "loss": 0.2176,
+      "step": 2072
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.837579228864627e-05,
+      "loss": 0.2277,
+      "step": 2074
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.837259178117427e-05,
+      "loss": 0.2203,
+      "step": 2076
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.8369388402788173e-05,
+      "loss": 0.2324,
+      "step": 2078
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.8366182154586408e-05,
+      "loss": 0.2254,
+      "step": 2080
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.8362973037668368e-05,
+      "loss": 0.2362,
+      "step": 2082
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.8359761053134448e-05,
+      "loss": 0.2243,
+      "step": 2084
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.8356546202086017e-05,
+      "loss": 0.1984,
+      "step": 2086
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.835332848562543e-05,
+      "loss": 0.2404,
+      "step": 2088
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.8350107904856026e-05,
+      "loss": 0.2033,
+      "step": 2090
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8346884460882117e-05,
+      "loss": 0.2273,
+      "step": 2092
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.834365815480901e-05,
+      "loss": 0.2136,
+      "step": 2094
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8340428987742983e-05,
+      "loss": 0.2481,
+      "step": 2096
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8337196960791305e-05,
+      "loss": 0.2216,
+      "step": 2098
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.833396207506221e-05,
+      "loss": 0.2195,
+      "step": 2100
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8330724331664932e-05,
+      "loss": 0.211,
+      "step": 2102
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8327483731709663e-05,
+      "loss": 0.2013,
+      "step": 2104
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.83242402763076e-05,
+      "loss": 0.2203,
+      "step": 2106
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8320993966570896e-05,
+      "loss": 0.2025,
+      "step": 2108
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8317744803612693e-05,
+      "loss": 0.2186,
+      "step": 2110
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8314492788547115e-05,
+      "loss": 0.2342,
+      "step": 2112
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8311237922489254e-05,
+      "loss": 0.2147,
+      "step": 2114
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8307980206555195e-05,
+      "loss": 0.2202,
+      "step": 2116
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8304719641861977e-05,
+      "loss": 0.2212,
+      "step": 2118
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8301456229527637e-05,
+      "loss": 0.231,
+      "step": 2120
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8298189970671176e-05,
+      "loss": 0.2107,
+      "step": 2122
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.829492086641258e-05,
+      "loss": 0.2146,
+      "step": 2124
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8291648917872807e-05,
+      "loss": 0.214,
+      "step": 2126
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8288374126173785e-05,
+      "loss": 0.2288,
+      "step": 2128
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8285096492438424e-05,
+      "loss": 0.2001,
+      "step": 2130
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8281816017790604e-05,
+      "loss": 0.2144,
+      "step": 2132
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.8278532703355182e-05,
+      "loss": 0.2159,
+      "step": 2134
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.8275246550257993e-05,
+      "loss": 0.2349,
+      "step": 2136
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.827195755962583e-05,
+      "loss": 0.2076,
+      "step": 2138
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.826866573258648e-05,
+      "loss": 0.2424,
+      "step": 2140
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.8265371070268685e-05,
+      "loss": 0.2124,
+      "step": 2142
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.8262073573802167e-05,
+      "loss": 0.2154,
+      "step": 2144
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.825877324431762e-05,
+      "loss": 0.1979,
+      "step": 2146
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.8255470082946707e-05,
+      "loss": 0.2334,
+      "step": 2148
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.8252164090822064e-05,
+      "loss": 0.2352,
+      "step": 2150
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.8248855269077296e-05,
+      "loss": 0.2299,
+      "step": 2152
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.824554361884698e-05,
+      "loss": 0.2253,
+      "step": 2154
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8242229141266663e-05,
+      "loss": 0.2175,
+      "step": 2156
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8238911837472858e-05,
+      "loss": 0.2085,
+      "step": 2158
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.823559170860305e-05,
+      "loss": 0.2007,
+      "step": 2160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8232268755795695e-05,
+      "loss": 0.2292,
+      "step": 2162
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8228942980190215e-05,
+      "loss": 0.2241,
+      "step": 2164
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8225614382926996e-05,
+      "loss": 0.2165,
+      "step": 2166
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8222282965147396e-05,
+      "loss": 0.1876,
+      "step": 2168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8218948727993736e-05,
+      "loss": 0.1677,
+      "step": 2170
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8215611672609316e-05,
+      "loss": 0.2014,
+      "step": 2172
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8212271800138387e-05,
+      "loss": 0.1759,
+      "step": 2174
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8208929111726174e-05,
+      "loss": 0.1882,
+      "step": 2176
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.22453388571739197,
+      "eval_runtime": 72.3818,
+      "eval_samples_per_second": 27.631,
+      "eval_steps_per_second": 13.816,
+      "step": 2176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.820558360851886e-05,
+      "loss": 0.1752,
+      "step": 2178
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.820223529166361e-05,
+      "loss": 0.1926,
+      "step": 2180
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8198884162308534e-05,
+      "loss": 0.2013,
+      "step": 2182
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.819553022160272e-05,
+      "loss": 0.1823,
+      "step": 2184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.819217347069621e-05,
+      "loss": 0.1967,
+      "step": 2186
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.818881391074002e-05,
+      "loss": 0.1986,
+      "step": 2188
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8185451542886123e-05,
+      "loss": 0.1787,
+      "step": 2190
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8182086368287454e-05,
+      "loss": 0.186,
+      "step": 2192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8178718388097915e-05,
+      "loss": 0.1701,
+      "step": 2194
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.817534760347236e-05,
+      "loss": 0.1831,
+      "step": 2196
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8171974015566623e-05,
+      "loss": 0.188,
+      "step": 2198
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.816859762553748e-05,
+      "loss": 0.1984,
+      "step": 2200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8165218434542677e-05,
+      "loss": 0.1713,
+      "step": 2202
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8161836443740925e-05,
+      "loss": 0.1811,
+      "step": 2204
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8158451654291883e-05,
+      "loss": 0.1829,
+      "step": 2206
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8155064067356175e-05,
+      "loss": 0.1968,
+      "step": 2208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8151673684095393e-05,
+      "loss": 0.1842,
+      "step": 2210
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8148280505672072e-05,
+      "loss": 0.1783,
+      "step": 2212
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8144884533249717e-05,
+      "loss": 0.1904,
+      "step": 2214
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.814148576799279e-05,
+      "loss": 0.1813,
+      "step": 2216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8138084211066703e-05,
+      "loss": 0.1893,
+      "step": 2218
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.813467986363784e-05,
+      "loss": 0.1802,
+      "step": 2220
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.8131272726873518e-05,
+      "loss": 0.2053,
+      "step": 2222
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.812786280194204e-05,
+      "loss": 0.1879,
+      "step": 2224
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.812445009001264e-05,
+      "loss": 0.1773,
+      "step": 2226
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.8121034592255514e-05,
+      "loss": 0.1692,
+      "step": 2228
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.811761630984183e-05,
+      "loss": 0.1804,
+      "step": 2230
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.8114195243943686e-05,
+      "loss": 0.19,
+      "step": 2232
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.8110771395734156e-05,
+      "loss": 0.2059,
+      "step": 2234
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.8107344766387248e-05,
+      "loss": 0.1913,
+      "step": 2236
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.810391535707794e-05,
+      "loss": 0.18,
+      "step": 2238
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.8100483168982152e-05,
+      "loss": 0.1846,
+      "step": 2240
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.809704820327677e-05,
+      "loss": 0.1883,
+      "step": 2242
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.8093610461139617e-05,
+      "loss": 0.1973,
+      "step": 2244
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.8090169943749477e-05,
+      "loss": 0.1816,
+      "step": 2246
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.8086726652286085e-05,
+      "loss": 0.172,
+      "step": 2248
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.8083280587930124e-05,
+      "loss": 0.1779,
+      "step": 2250
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.807983175186323e-05,
+      "loss": 0.1905,
+      "step": 2252
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.8076380145267994e-05,
+      "loss": 0.1891,
+      "step": 2254
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.8072925769327943e-05,
+      "loss": 0.1862,
+      "step": 2256
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.8069468625227574e-05,
+      "loss": 0.1727,
+      "step": 2258
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.8066008714152317e-05,
+      "loss": 0.1833,
+      "step": 2260
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.8062546037288556e-05,
+      "loss": 0.1863,
+      "step": 2262
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.8059080595823615e-05,
+      "loss": 0.1746,
+      "step": 2264
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.805561239094579e-05,
+      "loss": 0.1891,
+      "step": 2266
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.8052141423844297e-05,
+      "loss": 0.1756,
+      "step": 2268
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.8048667695709317e-05,
+      "loss": 0.1775,
+      "step": 2270
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.8045191207731964e-05,
+      "loss": 0.1759,
+      "step": 2272
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.8041711961104315e-05,
+      "loss": 0.1865,
+      "step": 2274
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.8038229957019378e-05,
+      "loss": 0.1825,
+      "step": 2276
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.8034745196671115e-05,
+      "loss": 0.2005,
+      "step": 2278
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.803125768125443e-05,
+      "loss": 0.1747,
+      "step": 2280
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.802776741196517e-05,
+      "loss": 0.1731,
+      "step": 2282
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.8024274390000134e-05,
+      "loss": 0.1864,
+      "step": 2284
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.8020778616557053e-05,
+      "loss": 0.1891,
+      "step": 2286
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.801728009283461e-05,
+      "loss": 0.1819,
+      "step": 2288
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.801377882003243e-05,
+      "loss": 0.1864,
+      "step": 2290
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.8010274799351082e-05,
+      "loss": 0.179,
+      "step": 2292
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.800676803199207e-05,
+      "loss": 0.1984,
+      "step": 2294
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.8003258519157846e-05,
+      "loss": 0.1769,
+      "step": 2296
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.79997462620518e-05,
+      "loss": 0.1684,
+      "step": 2298
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.799623126187827e-05,
+      "loss": 0.1768,
+      "step": 2300
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.7992713519842528e-05,
+      "loss": 0.1885,
+      "step": 2302
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.7989193037150785e-05,
+      "loss": 0.1908,
+      "step": 2304
+    },
+    {
+      "epoch": 1.06,
+      "eval_loss": 0.22056737542152405,
+      "eval_runtime": 72.4524,
+      "eval_samples_per_second": 27.604,
+      "eval_steps_per_second": 13.802,
+      "step": 2304
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.7985669815010197e-05,
+      "loss": 0.1806,
+      "step": 2306
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.7982143854628854e-05,
+      "loss": 0.1985,
+      "step": 2308
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.7978615157215792e-05,
+      "loss": 0.2027,
+      "step": 2310
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.7975083723980975e-05,
+      "loss": 0.1962,
+      "step": 2312
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.797154955613532e-05,
+      "loss": 0.2063,
+      "step": 2314
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.7968012654890666e-05,
+      "loss": 0.2052,
+      "step": 2316
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.79644730214598e-05,
+      "loss": 0.1888,
+      "step": 2318
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.796093065705644e-05,
+      "loss": 0.1765,
+      "step": 2320
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.795738556289524e-05,
+      "loss": 0.1832,
+      "step": 2322
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.79538377401918e-05,
+      "loss": 0.192,
+      "step": 2324
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.795028719016265e-05,
+      "loss": 0.1779,
+      "step": 2326
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.794673391402524e-05,
+      "loss": 0.1901,
+      "step": 2328
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.7943177912997974e-05,
+      "loss": 0.1869,
+      "step": 2330
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.7939619188300193e-05,
+      "loss": 0.1811,
+      "step": 2332
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.793605774115215e-05,
+      "loss": 0.1921,
+      "step": 2334
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.793249357277506e-05,
+      "loss": 0.1936,
+      "step": 2336
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.792892668439104e-05,
+      "loss": 0.1851,
+      "step": 2338
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.792535707722317e-05,
+      "loss": 0.192,
+      "step": 2340
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.7921784752495435e-05,
+      "loss": 0.1747,
+      "step": 2342
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.7918209711432776e-05,
+      "loss": 0.1747,
+      "step": 2344
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.791463195526105e-05,
+      "loss": 0.1919,
+      "step": 2346
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.7911051485207047e-05,
+      "loss": 0.2045,
+      "step": 2348
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.7907468302498495e-05,
+      "loss": 0.1811,
+      "step": 2350
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.790388240836404e-05,
+      "loss": 0.1895,
+      "step": 2352
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.790029380403327e-05,
+      "loss": 0.1892,
+      "step": 2354
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.7896702490736696e-05,
+      "loss": 0.168,
+      "step": 2356
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.7893108469705755e-05,
+      "loss": 0.1905,
+      "step": 2358
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.7889511742172822e-05,
+      "loss": 0.1855,
+      "step": 2360
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.788591230937119e-05,
+      "loss": 0.1994,
+      "step": 2362
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.7882310172535093e-05,
+      "loss": 0.1955,
+      "step": 2364
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.7878705332899672e-05,
+      "loss": 0.1826,
+      "step": 2366
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.7875097791701007e-05,
+      "loss": 0.1743,
+      "step": 2368
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.787148755017611e-05,
+      "loss": 0.1852,
+      "step": 2370
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.786787460956291e-05,
+      "loss": 0.1805,
+      "step": 2372
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.7864258971100265e-05,
+      "loss": 0.1865,
+      "step": 2374
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.786064063602795e-05,
+      "loss": 0.1761,
+      "step": 2376
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.7857019605586677e-05,
+      "loss": 0.1848,
+      "step": 2378
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.7853395881018075e-05,
+      "loss": 0.1883,
+      "step": 2380
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.78497694635647e-05,
+      "loss": 0.1741,
+      "step": 2382
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.7846140354470032e-05,
+      "loss": 0.2019,
+      "step": 2384
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.784250855497847e-05,
+      "loss": 0.1835,
+      "step": 2386
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.7838874066335333e-05,
+      "loss": 0.1848,
+      "step": 2388
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.7835236889786868e-05,
+      "loss": 0.1911,
+      "step": 2390
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.783159702658025e-05,
+      "loss": 0.1905,
+      "step": 2392
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.782795447796356e-05,
+      "loss": 0.1748,
+      "step": 2394
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.7824309245185805e-05,
+      "loss": 0.1857,
+      "step": 2396
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.782066132949692e-05,
+      "loss": 0.1819,
+      "step": 2398
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.7817010732147758e-05,
+      "loss": 0.1909,
+      "step": 2400
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.781335745439008e-05,
+      "loss": 0.1858,
+      "step": 2402
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.7809701497476574e-05,
+      "loss": 0.1909,
+      "step": 2404
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.7806042862660854e-05,
+      "loss": 0.1886,
+      "step": 2406
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.780238155119744e-05,
+      "loss": 0.1869,
+      "step": 2408
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.7798717564341775e-05,
+      "loss": 0.1868,
+      "step": 2410
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.779505090335022e-05,
+      "loss": 0.1972,
+      "step": 2412
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.7791381569480056e-05,
+      "loss": 0.1812,
+      "step": 2414
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.7787709563989472e-05,
+      "loss": 0.1777,
+      "step": 2416
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.7784034888137577e-05,
+      "loss": 0.1891,
+      "step": 2418
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.7780357543184396e-05,
+      "loss": 0.1916,
+      "step": 2420
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.7776677530390875e-05,
+      "loss": 0.1878,
+      "step": 2422
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.7772994851018865e-05,
+      "loss": 0.1758,
+      "step": 2424
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.776930950633114e-05,
+      "loss": 0.1856,
+      "step": 2426
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.7765621497591377e-05,
+      "loss": 0.1806,
+      "step": 2428
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.776193082606418e-05,
+      "loss": 0.188,
+      "step": 2430
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.775823749301506e-05,
+      "loss": 0.1728,
+      "step": 2432
+    },
+    {
+      "epoch": 1.12,
+      "eval_loss": 0.21872420608997345,
+      "eval_runtime": 72.3811,
+      "eval_samples_per_second": 27.632,
+      "eval_steps_per_second": 13.816,
+      "step": 2432
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.775454149971043e-05,
+      "loss": 0.1798,
+      "step": 2434
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.7750842847417637e-05,
+      "loss": 0.1704,
+      "step": 2436
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7747141537404916e-05,
+      "loss": 0.185,
+      "step": 2438
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7743437570941435e-05,
+      "loss": 0.1805,
+      "step": 2440
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7739730949297262e-05,
+      "loss": 0.1797,
+      "step": 2442
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.773602167374337e-05,
+      "loss": 0.185,
+      "step": 2444
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7732309745551655e-05,
+      "loss": 0.1935,
+      "step": 2446
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.772859516599491e-05,
+      "loss": 0.1822,
+      "step": 2448
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.772487793634685e-05,
+      "loss": 0.1931,
+      "step": 2450
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7721158057882085e-05,
+      "loss": 0.1859,
+      "step": 2452
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7717435531876148e-05,
+      "loss": 0.1893,
+      "step": 2454
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7713710359605466e-05,
+      "loss": 0.2003,
+      "step": 2456
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7709982542347382e-05,
+      "loss": 0.1852,
+      "step": 2458
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.7706252081380143e-05,
+      "loss": 0.1614,
+      "step": 2460
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.77025189779829e-05,
+      "loss": 0.1783,
+      "step": 2462
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.7698783233435715e-05,
+      "loss": 0.1931,
+      "step": 2464
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.7695044849019558e-05,
+      "loss": 0.1626,
+      "step": 2466
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.769130382601629e-05,
+      "loss": 0.1783,
+      "step": 2468
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.7687560165708695e-05,
+      "loss": 0.1735,
+      "step": 2470
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.7683813869380452e-05,
+      "loss": 0.1938,
+      "step": 2472
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.7680064938316138e-05,
+      "loss": 0.1725,
+      "step": 2474
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.7676313373801245e-05,
+      "loss": 0.1738,
+      "step": 2476
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.7672559177122167e-05,
+      "loss": 0.1647,
+      "step": 2478
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.766880234956619e-05,
+      "loss": 0.1742,
+      "step": 2480
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7665042892421512e-05,
+      "loss": 0.179,
+      "step": 2482
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.766128080697723e-05,
+      "loss": 0.1938,
+      "step": 2484
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7657516094523336e-05,
+      "loss": 0.2011,
+      "step": 2486
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7653748756350736e-05,
+      "loss": 0.184,
+      "step": 2488
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7649978793751226e-05,
+      "loss": 0.1793,
+      "step": 2490
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7646206208017506e-05,
+      "loss": 0.1879,
+      "step": 2492
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7642431000443168e-05,
+      "loss": 0.1776,
+      "step": 2494
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7638653172322716e-05,
+      "loss": 0.1973,
+      "step": 2496
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7634872724951545e-05,
+      "loss": 0.1843,
+      "step": 2498
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.763108965962595e-05,
+      "loss": 0.1747,
+      "step": 2500
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.7627303977643115e-05,
+      "loss": 0.1958,
+      "step": 2502
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.7623515680301135e-05,
+      "loss": 0.1797,
+      "step": 2504
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.7619724768899e-05,
+      "loss": 0.1844,
+      "step": 2506
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.761593124473658e-05,
+      "loss": 0.1823,
+      "step": 2508
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.761213510911466e-05,
+      "loss": 0.1886,
+      "step": 2510
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.7608336363334915e-05,
+      "loss": 0.1895,
+      "step": 2512
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.7604535008699916e-05,
+      "loss": 0.182,
+      "step": 2514
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.7600731046513114e-05,
+      "loss": 0.1688,
+      "step": 2516
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.7596924478078873e-05,
+      "loss": 0.1987,
+      "step": 2518
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.7593115304702448e-05,
+      "loss": 0.1976,
+      "step": 2520
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.7589303527689972e-05,
+      "loss": 0.1782,
+      "step": 2522
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7585489148348493e-05,
+      "loss": 0.1815,
+      "step": 2524
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7581672167985934e-05,
+      "loss": 0.196,
+      "step": 2526
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7577852587911113e-05,
+      "loss": 0.1873,
+      "step": 2528
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7574030409433755e-05,
+      "loss": 0.1817,
+      "step": 2530
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7570205633864448e-05,
+      "loss": 0.1811,
+      "step": 2532
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7566378262514696e-05,
+      "loss": 0.1806,
+      "step": 2534
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7562548296696876e-05,
+      "loss": 0.18,
+      "step": 2536
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7558715737724266e-05,
+      "loss": 0.1713,
+      "step": 2538
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7554880586911025e-05,
+      "loss": 0.1839,
+      "step": 2540
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.755104284557221e-05,
+      "loss": 0.1855,
+      "step": 2542
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.7547202515023753e-05,
+      "loss": 0.1815,
+      "step": 2544
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.7543359596582493e-05,
+      "loss": 0.1936,
+      "step": 2546
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.753951409156613e-05,
+      "loss": 0.2066,
+      "step": 2548
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.753566600129328e-05,
+      "loss": 0.1967,
+      "step": 2550
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.753181532708342e-05,
+      "loss": 0.162,
+      "step": 2552
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.7527962070256933e-05,
+      "loss": 0.1948,
+      "step": 2554
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.752410623213507e-05,
+      "loss": 0.1753,
+      "step": 2556
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.7520247814039985e-05,
+      "loss": 0.1832,
+      "step": 2558
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.75163868172947e-05,
+      "loss": 0.1796,
+      "step": 2560
+    },
+    {
+      "epoch": 1.18,
+      "eval_loss": 0.21623384952545166,
+      "eval_runtime": 72.3784,
+      "eval_samples_per_second": 27.633,
+      "eval_steps_per_second": 13.816,
+      "step": 2560
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.7512523243223137e-05,
+      "loss": 0.1823,
+      "step": 2562
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.7508657093150086e-05,
+      "loss": 0.1801,
+      "step": 2564
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.7504788368401232e-05,
+      "loss": 0.1795,
+      "step": 2566
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7500917070303135e-05,
+      "loss": 0.1697,
+      "step": 2568
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7497043200183247e-05,
+      "loss": 0.1752,
+      "step": 2570
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.749316675936989e-05,
+      "loss": 0.1826,
+      "step": 2572
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.748928774919228e-05,
+      "loss": 0.1721,
+      "step": 2574
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.74854061709805e-05,
+      "loss": 0.1842,
+      "step": 2576
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7481522026065523e-05,
+      "loss": 0.2052,
+      "step": 2578
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7477635315779205e-05,
+      "loss": 0.1768,
+      "step": 2580
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7473746041454274e-05,
+      "loss": 0.1849,
+      "step": 2582
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.746985420442434e-05,
+      "loss": 0.1804,
+      "step": 2584
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7465959806023887e-05,
+      "loss": 0.1789,
+      "step": 2586
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7462062847588292e-05,
+      "loss": 0.176,
+      "step": 2588
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7458163330453794e-05,
+      "loss": 0.1714,
+      "step": 2590
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7454261255957514e-05,
+      "loss": 0.1737,
+      "step": 2592
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7450356625437454e-05,
+      "loss": 0.1658,
+      "step": 2594
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7446449440232486e-05,
+      "loss": 0.1862,
+      "step": 2596
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7442539701682366e-05,
+      "loss": 0.1763,
+      "step": 2598
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.743862741112772e-05,
+      "loss": 0.1828,
+      "step": 2600
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.743471256991005e-05,
+      "loss": 0.1835,
+      "step": 2602
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7430795179371728e-05,
+      "loss": 0.1757,
+      "step": 2604
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7426875240856013e-05,
+      "loss": 0.1763,
+      "step": 2606
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7422952755707025e-05,
+      "loss": 0.1661,
+      "step": 2608
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7419027725269763e-05,
+      "loss": 0.1848,
+      "step": 2610
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7415100150890096e-05,
+      "loss": 0.174,
+      "step": 2612
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7411170033914767e-05,
+      "loss": 0.1802,
+      "step": 2614
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7407237375691394e-05,
+      "loss": 0.1938,
+      "step": 2616
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.740330217756846e-05,
+      "loss": 0.1928,
+      "step": 2618
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7399364440895323e-05,
+      "loss": 0.1848,
+      "step": 2620
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.739542416702221e-05,
+      "loss": 0.1782,
+      "step": 2622
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.739148135730022e-05,
+      "loss": 0.1641,
+      "step": 2624
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7387536013081317e-05,
+      "loss": 0.1805,
+      "step": 2626
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7383588135718342e-05,
+      "loss": 0.1736,
+      "step": 2628
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7379637726564994e-05,
+      "loss": 0.1738,
+      "step": 2630
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.737568478697585e-05,
+      "loss": 0.1941,
+      "step": 2632
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.7371729318306353e-05,
+      "loss": 0.1939,
+      "step": 2634
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.7367771321912802e-05,
+      "loss": 0.1789,
+      "step": 2636
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.7363810799152376e-05,
+      "loss": 0.1721,
+      "step": 2638
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.7359847751383115e-05,
+      "loss": 0.1772,
+      "step": 2640
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.735588217996393e-05,
+      "loss": 0.1793,
+      "step": 2642
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.735191408625459e-05,
+      "loss": 0.1688,
+      "step": 2644
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.734794347161573e-05,
+      "loss": 0.177,
+      "step": 2646
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.7343970337408852e-05,
+      "loss": 0.1848,
+      "step": 2648
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.733999468499632e-05,
+      "loss": 0.169,
+      "step": 2650
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.733601651574137e-05,
+      "loss": 0.1703,
+      "step": 2652
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7332035831008082e-05,
+      "loss": 0.1897,
+      "step": 2654
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7328052632161413e-05,
+      "loss": 0.1749,
+      "step": 2656
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7324066920567187e-05,
+      "loss": 0.19,
+      "step": 2658
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7320078697592077e-05,
+      "loss": 0.1747,
+      "step": 2660
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7316087964603624e-05,
+      "loss": 0.1681,
+      "step": 2662
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.731209472297022e-05,
+      "loss": 0.1888,
+      "step": 2664
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7308098974061132e-05,
+      "loss": 0.189,
+      "step": 2666
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7304100719246483e-05,
+      "loss": 0.1794,
+      "step": 2668
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.730009995989724e-05,
+      "loss": 0.1823,
+      "step": 2670
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7296096697385252e-05,
+      "loss": 0.1853,
+      "step": 2672
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.7292090933083207e-05,
+      "loss": 0.187,
+      "step": 2674
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.7288082668364665e-05,
+      "loss": 0.1922,
+      "step": 2676
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.728407190460403e-05,
+      "loss": 0.1817,
+      "step": 2678
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.728005864317658e-05,
+      "loss": 0.1834,
+      "step": 2680
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.7276042885458428e-05,
+      "loss": 0.1789,
+      "step": 2682
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.7272024632826564e-05,
+      "loss": 0.1757,
+      "step": 2684
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.726800388665882e-05,
+      "loss": 0.1874,
+      "step": 2686
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.7263980648333886e-05,
+      "loss": 0.1769,
+      "step": 2688
+    },
+    {
+      "epoch": 1.24,
+      "eval_loss": 0.21456414461135864,
+      "eval_runtime": 72.4572,
+      "eval_samples_per_second": 27.602,
+      "eval_steps_per_second": 13.801,
+      "step": 2688
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.725995491923131e-05,
+      "loss": 0.1747,
+      "step": 2690
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.725592670073149e-05,
+      "loss": 0.174,
+      "step": 2692
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.725189599421568e-05,
+      "loss": 0.1656,
+      "step": 2694
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.7247862801065988e-05,
+      "loss": 0.1788,
+      "step": 2696
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.724382712266537e-05,
+      "loss": 0.1834,
+      "step": 2698
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.7239788960397636e-05,
+      "loss": 0.1727,
+      "step": 2700
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.723574831564745e-05,
+      "loss": 0.1861,
+      "step": 2702
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.723170518980033e-05,
+      "loss": 0.1824,
+      "step": 2704
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.7227659584242638e-05,
+      "loss": 0.1758,
+      "step": 2706
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.722361150036159e-05,
+      "loss": 0.1903,
+      "step": 2708
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.7219560939545246e-05,
+      "loss": 0.1765,
+      "step": 2710
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.7215507903182528e-05,
+      "loss": 0.1909,
+      "step": 2712
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.721145239266319e-05,
+      "loss": 0.1711,
+      "step": 2714
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.7207394409377855e-05,
+      "loss": 0.1948,
+      "step": 2716
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.7203333954717976e-05,
+      "loss": 0.1791,
+      "step": 2718
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.719927103007586e-05,
+      "loss": 0.1802,
+      "step": 2720
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.7195205636844657e-05,
+      "loss": 0.1879,
+      "step": 2722
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.7191137776418375e-05,
+      "loss": 0.1887,
+      "step": 2724
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.7187067450191852e-05,
+      "loss": 0.185,
+      "step": 2726
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.7182994659560786e-05,
+      "loss": 0.1788,
+      "step": 2728
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.7178919405921717e-05,
+      "loss": 0.1801,
+      "step": 2730
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.7174841690672016e-05,
+      "loss": 0.1981,
+      "step": 2732
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.717076151520991e-05,
+      "loss": 0.1854,
+      "step": 2734
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.7166678880934478e-05,
+      "loss": 0.182,
+      "step": 2736
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.716259378924562e-05,
+      "loss": 0.1862,
+      "step": 2738
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.71585062415441e-05,
+      "loss": 0.1753,
+      "step": 2740
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.715441623923151e-05,
+      "loss": 0.1927,
+      "step": 2742
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.715032378371029e-05,
+      "loss": 0.2002,
+      "step": 2744
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.7146228876383718e-05,
+      "loss": 0.1815,
+      "step": 2746
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.7142131518655917e-05,
+      "loss": 0.1749,
+      "step": 2748
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.7138031711931842e-05,
+      "loss": 0.1891,
+      "step": 2750
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.7133929457617303e-05,
+      "loss": 0.1775,
+      "step": 2752
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.712982475711893e-05,
+      "loss": 0.173,
+      "step": 2754
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.712571761184421e-05,
+      "loss": 0.1755,
+      "step": 2756
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.7121608023201453e-05,
+      "loss": 0.1741,
+      "step": 2758
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.7117495992599814e-05,
+      "loss": 0.1737,
+      "step": 2760
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.7113381521449294e-05,
+      "loss": 0.1906,
+      "step": 2762
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.710926461116071e-05,
+      "loss": 0.1674,
+      "step": 2764
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.710514526314573e-05,
+      "loss": 0.1736,
+      "step": 2766
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.7101023478816857e-05,
+      "loss": 0.1797,
+      "step": 2768
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.7096899259587432e-05,
+      "loss": 0.1839,
+      "step": 2770
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.709277260687162e-05,
+      "loss": 0.1844,
+      "step": 2772
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.7088643522084426e-05,
+      "loss": 0.1807,
+      "step": 2774
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.7084512006641695e-05,
+      "loss": 0.1852,
+      "step": 2776
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.7080378061960097e-05,
+      "loss": 0.1802,
+      "step": 2778
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.7076241689457134e-05,
+      "loss": 0.1643,
+      "step": 2780
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.7072102890551153e-05,
+      "loss": 0.1826,
+      "step": 2782
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.706796166666132e-05,
+      "loss": 0.1898,
+      "step": 2784
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7063818019207638e-05,
+      "loss": 0.1884,
+      "step": 2786
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7059671949610937e-05,
+      "loss": 0.179,
+      "step": 2788
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7055523459292888e-05,
+      "loss": 0.1715,
+      "step": 2790
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7051372549675974e-05,
+      "loss": 0.1814,
+      "step": 2792
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7047219222183532e-05,
+      "loss": 0.1751,
+      "step": 2794
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.70430634782397e-05,
+      "loss": 0.195,
+      "step": 2796
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.703890531926947e-05,
+      "loss": 0.1699,
+      "step": 2798
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7034744746698644e-05,
+      "loss": 0.1762,
+      "step": 2800
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7030581761953865e-05,
+      "loss": 0.1766,
+      "step": 2802
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7026416366462588e-05,
+      "loss": 0.1598,
+      "step": 2804
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.7022248561653113e-05,
+      "loss": 0.1891,
+      "step": 2806
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.701807834895455e-05,
+      "loss": 0.176,
+      "step": 2808
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.7013905729796845e-05,
+      "loss": 0.1848,
+      "step": 2810
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.7009730705610764e-05,
+      "loss": 0.1759,
+      "step": 2812
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.70055532778279e-05,
+      "loss": 0.1798,
+      "step": 2814
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.7001373447880664e-05,
+      "loss": 0.1751,
+      "step": 2816
+    },
+    {
+      "epoch": 1.3,
+      "eval_loss": 0.2121935337781906,
+      "eval_runtime": 72.3805,
+      "eval_samples_per_second": 27.632,
+      "eval_steps_per_second": 13.816,
+      "step": 2816
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.6997191217202303e-05,
+      "loss": 0.1756,
+      "step": 2818
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.6993006587226876e-05,
+      "loss": 0.1971,
+      "step": 2820
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.6988819559389274e-05,
+      "loss": 0.1794,
+      "step": 2822
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.6984630135125196e-05,
+      "loss": 0.1758,
+      "step": 2824
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.6980438315871178e-05,
+      "loss": 0.1901,
+      "step": 2826
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.6976244103064572e-05,
+      "loss": 0.1697,
+      "step": 2828
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.6972047498143544e-05,
+      "loss": 0.1846,
+      "step": 2830
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.6967848502547092e-05,
+      "loss": 0.171,
+      "step": 2832
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.696364711771502e-05,
+      "loss": 0.1655,
+      "step": 2834
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.6959443345087975e-05,
+      "loss": 0.1832,
+      "step": 2836
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.6955237186107387e-05,
+      "loss": 0.1684,
+      "step": 2838
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.6951028642215533e-05,
+      "loss": 0.1708,
+      "step": 2840
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.69468177148555e-05,
+      "loss": 0.1885,
+      "step": 2842
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.6942604405471193e-05,
+      "loss": 0.191,
+      "step": 2844
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.6938388715507322e-05,
+      "loss": 0.1816,
+      "step": 2846
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.6934170646409432e-05,
+      "loss": 0.1692,
+      "step": 2848
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.6929950199623875e-05,
+      "loss": 0.1804,
+      "step": 2850
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.6925727376597814e-05,
+      "loss": 0.1797,
+      "step": 2852
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.6921502178779232e-05,
+      "loss": 0.1843,
+      "step": 2854
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.6917274607616927e-05,
+      "loss": 0.1712,
+      "step": 2856
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.6913044664560508e-05,
+      "loss": 0.1763,
+      "step": 2858
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.6908812351060397e-05,
+      "loss": 0.1676,
+      "step": 2860
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.6904577668567833e-05,
+      "loss": 0.1818,
+      "step": 2862
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.690034061853486e-05,
+      "loss": 0.1647,
+      "step": 2864
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.6896101202414344e-05,
+      "loss": 0.1935,
+      "step": 2866
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.6891859421659955e-05,
+      "loss": 0.1791,
+      "step": 2868
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.6887615277726174e-05,
+      "loss": 0.1713,
+      "step": 2870
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6883368772068293e-05,
+      "loss": 0.1875,
+      "step": 2872
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6879119906142417e-05,
+      "loss": 0.1713,
+      "step": 2874
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.687486868140546e-05,
+      "loss": 0.1673,
+      "step": 2876
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6870615099315137e-05,
+      "loss": 0.1804,
+      "step": 2878
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.686635916132998e-05,
+      "loss": 0.1627,
+      "step": 2880
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6862100868909324e-05,
+      "loss": 0.1815,
+      "step": 2882
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6857840223513315e-05,
+      "loss": 0.1776,
+      "step": 2884
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6853577226602907e-05,
+      "loss": 0.1653,
+      "step": 2886
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6849311879639855e-05,
+      "loss": 0.1999,
+      "step": 2888
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.684504418408672e-05,
+      "loss": 0.1855,
+      "step": 2890
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.6840774141406874e-05,
+      "loss": 0.1708,
+      "step": 2892
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.6836501753064488e-05,
+      "loss": 0.1825,
+      "step": 2894
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.683222702052454e-05,
+      "loss": 0.1795,
+      "step": 2896
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.6827949945252814e-05,
+      "loss": 0.1788,
+      "step": 2898
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.6823670528715886e-05,
+      "loss": 0.1887,
+      "step": 2900
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.6819388772381156e-05,
+      "loss": 0.1742,
+      "step": 2902
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.6815104677716806e-05,
+      "loss": 0.1668,
+      "step": 2904
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.681081824619183e-05,
+      "loss": 0.1848,
+      "step": 2906
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.680652947927602e-05,
+      "loss": 0.1782,
+      "step": 2908
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.6802238378439976e-05,
+      "loss": 0.1798,
+      "step": 2910
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.679794494515508e-05,
+      "loss": 0.1659,
+      "step": 2912
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.6793649180893534e-05,
+      "loss": 0.1685,
+      "step": 2914
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.678935108712833e-05,
+      "loss": 0.1706,
+      "step": 2916
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.678505066533326e-05,
+      "loss": 0.1822,
+      "step": 2918
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.6780747916982912e-05,
+      "loss": 0.1869,
+      "step": 2920
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.6776442843552678e-05,
+      "loss": 0.168,
+      "step": 2922
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.677213544651874e-05,
+      "loss": 0.1634,
+      "step": 2924
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.676782572735808e-05,
+      "loss": 0.1852,
+      "step": 2926
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.6763513687548477e-05,
+      "loss": 0.1672,
+      "step": 2928
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.6759199328568506e-05,
+      "loss": 0.1765,
+      "step": 2930
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.6754882651897534e-05,
+      "loss": 0.1772,
+      "step": 2932
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.6750563659015727e-05,
+      "loss": 0.1973,
+      "step": 2934
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.674624235140404e-05,
+      "loss": 0.1836,
+      "step": 2936
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6741918730544225e-05,
+      "loss": 0.1695,
+      "step": 2938
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.673759279791883e-05,
+      "loss": 0.1998,
+      "step": 2940
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6733264555011196e-05,
+      "loss": 0.1688,
+      "step": 2942
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6728934003305442e-05,
+      "loss": 0.174,
+      "step": 2944
+    },
+    {
+      "epoch": 1.36,
+      "eval_loss": 0.20981821417808533,
+      "eval_runtime": 72.3799,
+      "eval_samples_per_second": 27.632,
+      "eval_steps_per_second": 13.816,
+      "step": 2944
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6724601144286498e-05,
+      "loss": 0.1829,
+      "step": 2946
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6720265979440074e-05,
+      "loss": 0.1898,
+      "step": 2948
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6715928510252678e-05,
+      "loss": 0.1757,
+      "step": 2950
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6711588738211597e-05,
+      "loss": 0.189,
+      "step": 2952
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6707246664804916e-05,
+      "loss": 0.1879,
+      "step": 2954
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.6702902291521508e-05,
+      "loss": 0.1691,
+      "step": 2956
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.6698555619851033e-05,
+      "loss": 0.1699,
+      "step": 2958
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.6694206651283938e-05,
+      "loss": 0.1819,
+      "step": 2960
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.6689855387311468e-05,
+      "loss": 0.1775,
+      "step": 2962
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.6685501829425637e-05,
+      "loss": 0.1703,
+      "step": 2964
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.6681145979119263e-05,
+      "loss": 0.1857,
+      "step": 2966
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.6676787837885937e-05,
+      "loss": 0.1747,
+      "step": 2968
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.6672427407220038e-05,
+      "loss": 0.1736,
+      "step": 2970
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.6668064688616745e-05,
+      "loss": 0.1863,
+      "step": 2972
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.6663699683571995e-05,
+      "loss": 0.1707,
+      "step": 2974
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.665933239358253e-05,
+      "loss": 0.1739,
+      "step": 2976
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.665496282014587e-05,
+      "loss": 0.1761,
+      "step": 2978
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.665059096476032e-05,
+      "loss": 0.1807,
+      "step": 2980
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.6646216828924955e-05,
+      "loss": 0.1609,
+      "step": 2982
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.664184041413965e-05,
+      "loss": 0.2013,
+      "step": 2984
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.6637461721905047e-05,
+      "loss": 0.1694,
+      "step": 2986
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.6633080753722575e-05,
+      "loss": 0.1666,
+      "step": 2988
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.6628697511094446e-05,
+      "loss": 0.1801,
+      "step": 2990
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.6624311995523653e-05,
+      "loss": 0.1897,
+      "step": 2992
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.6619924208513954e-05,
+      "loss": 0.1778,
+      "step": 2994
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.6615534151569902e-05,
+      "loss": 0.1746,
+      "step": 2996
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.661114182619682e-05,
+      "loss": 0.1779,
+      "step": 2998
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.6606747233900816e-05,
+      "loss": 0.182,
+      "step": 3000
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.6602350376188767e-05,
+      "loss": 0.1757,
+      "step": 3002
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.6597951254568334e-05,
+      "loss": 0.1906,
+      "step": 3004
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.6593549870547944e-05,
+      "loss": 0.182,
+      "step": 3006
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.6589146225636813e-05,
+      "loss": 0.1716,
+      "step": 3008
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.6584740321344927e-05,
+      "loss": 0.181,
+      "step": 3010
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.6580332159183036e-05,
+      "loss": 0.1609,
+      "step": 3012
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.6575921740662684e-05,
+      "loss": 0.1678,
+      "step": 3014
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.657150906729617e-05,
+      "loss": 0.1744,
+      "step": 3016
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.6567094140596583e-05,
+      "loss": 0.1762,
+      "step": 3018
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.656267696207777e-05,
+      "loss": 0.1753,
+      "step": 3020
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.655825753325436e-05,
+      "loss": 0.163,
+      "step": 3022
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6553835855641746e-05,
+      "loss": 0.1885,
+      "step": 3024
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6549411930756104e-05,
+      "loss": 0.1738,
+      "step": 3026
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6544985760114364e-05,
+      "loss": 0.181,
+      "step": 3028
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6540557345234235e-05,
+      "loss": 0.1802,
+      "step": 3030
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.65361266876342e-05,
+      "loss": 0.1676,
+      "step": 3032
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6531693788833507e-05,
+      "loss": 0.173,
+      "step": 3034
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6527258650352167e-05,
+      "loss": 0.1702,
+      "step": 3036
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6522821273710966e-05,
+      "loss": 0.1704,
+      "step": 3038
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6518381660431456e-05,
+      "loss": 0.1826,
+      "step": 3040
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.6513939812035956e-05,
+      "loss": 0.1732,
+      "step": 3042
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.6509495730047548e-05,
+      "loss": 0.1809,
+      "step": 3044
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.6505049415990084e-05,
+      "loss": 0.1785,
+      "step": 3046
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.650060087138818e-05,
+      "loss": 0.1641,
+      "step": 3048
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.6496150097767212e-05,
+      "loss": 0.1714,
+      "step": 3050
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.6491697096653332e-05,
+      "loss": 0.1799,
+      "step": 3052
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.6487241869573445e-05,
+      "loss": 0.1853,
+      "step": 3054
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.6482784418055226e-05,
+      "loss": 0.1721,
+      "step": 3056
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.6478324743627104e-05,
+      "loss": 0.1606,
+      "step": 3058
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.647386284781828e-05,
+      "loss": 0.1922,
+      "step": 3060
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.646939873215871e-05,
+      "loss": 0.181,
+      "step": 3062
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.646493239817912e-05,
+      "loss": 0.2019,
+      "step": 3064
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.6460463847410983e-05,
+      "loss": 0.1713,
+      "step": 3066
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.6455993081386545e-05,
+      "loss": 0.1762,
+      "step": 3068
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.6451520101638805e-05,
+      "loss": 0.1689,
+      "step": 3070
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.644704490970152e-05,
+      "loss": 0.1763,
+      "step": 3072
+    },
+    {
+      "epoch": 1.42,
+      "eval_loss": 0.20855751633644104,
+      "eval_runtime": 72.4551,
+      "eval_samples_per_second": 27.603,
+      "eval_steps_per_second": 13.802,
+      "step": 3072
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.644256750710921e-05,
+      "loss": 0.1778,
+      "step": 3074
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.6438087895397152e-05,
+      "loss": 0.1695,
+      "step": 3076
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.6433606076101375e-05,
+      "loss": 0.168,
+      "step": 3078
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.642912205075867e-05,
+      "loss": 0.1809,
+      "step": 3080
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.6424635820906593e-05,
+      "loss": 0.1727,
+      "step": 3082
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.6420147388083436e-05,
+      "loss": 0.1799,
+      "step": 3084
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.6415656753828256e-05,
+      "loss": 0.1629,
+      "step": 3086
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.6411163919680878e-05,
+      "loss": 0.184,
+      "step": 3088
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.640666888718186e-05,
+      "loss": 0.1827,
+      "step": 3090
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.6402171657872524e-05,
+      "loss": 0.1777,
+      "step": 3092
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.6397672233294945e-05,
+      "loss": 0.1749,
+      "step": 3094
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.639317061499195e-05,
+      "loss": 0.1882,
+      "step": 3096
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.6388666804507118e-05,
+      "loss": 0.1741,
+      "step": 3098
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.6384160803384782e-05,
+      "loss": 0.1824,
+      "step": 3100
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.6379652613170024e-05,
+      "loss": 0.1813,
+      "step": 3102
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.637514223540867e-05,
+      "loss": 0.1928,
+      "step": 3104
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.6370629671647316e-05,
+      "loss": 0.1812,
+      "step": 3106
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.6366114923433284e-05,
+      "loss": 0.1783,
+      "step": 3108
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.6361597992314657e-05,
+      "loss": 0.1787,
+      "step": 3110
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.6357078879840267e-05,
+      "loss": 0.1751,
+      "step": 3112
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.6352557587559693e-05,
+      "loss": 0.1807,
+      "step": 3114
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.634803411702326e-05,
+      "loss": 0.1662,
+      "step": 3116
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.6343508469782035e-05,
+      "loss": 0.1762,
+      "step": 3118
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.6338980647387843e-05,
+      "loss": 0.1751,
+      "step": 3120
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.6334450651393243e-05,
+      "loss": 0.1744,
+      "step": 3122
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.6329918483351552e-05,
+      "loss": 0.1702,
+      "step": 3124
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.632538414481682e-05,
+      "loss": 0.1725,
+      "step": 3126
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.6320847637343846e-05,
+      "loss": 0.1739,
+      "step": 3128
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.6316308962488173e-05,
+      "loss": 0.1772,
+      "step": 3130
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.631176812180609e-05,
+      "loss": 0.1648,
+      "step": 3132
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.6307225116854623e-05,
+      "loss": 0.1904,
+      "step": 3134
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.630267994919154e-05,
+      "loss": 0.1664,
+      "step": 3136
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.629813262037535e-05,
+      "loss": 0.1815,
+      "step": 3138
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.6293583131965317e-05,
+      "loss": 0.1818,
+      "step": 3140
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.628903148552143e-05,
+      "loss": 0.1802,
+      "step": 3142
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.6284477682604418e-05,
+      "loss": 0.1824,
+      "step": 3144
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.627992172477576e-05,
+      "loss": 0.1661,
+      "step": 3146
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.627536361359767e-05,
+      "loss": 0.19,
+      "step": 3148
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.6270803350633085e-05,
+      "loss": 0.1731,
+      "step": 3150
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.626624093744571e-05,
+      "loss": 0.174,
+      "step": 3152
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.626167637559996e-05,
+      "loss": 0.1831,
+      "step": 3154
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.6257109666661e-05,
+      "loss": 0.1771,
+      "step": 3156
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.625254081219473e-05,
+      "loss": 0.1762,
+      "step": 3158
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.6247969813767784e-05,
+      "loss": 0.1828,
+      "step": 3160
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.6243396672947527e-05,
+      "loss": 0.1926,
+      "step": 3162
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.623882139130207e-05,
+      "loss": 0.1699,
+      "step": 3164
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.623424397040025e-05,
+      "loss": 0.1845,
+      "step": 3166
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.622966441181163e-05,
+      "loss": 0.188,
+      "step": 3168
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.6225082717106525e-05,
+      "loss": 0.1702,
+      "step": 3170
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.6220498887855964e-05,
+      "loss": 0.1848,
+      "step": 3172
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.6215912925631724e-05,
+      "loss": 0.1649,
+      "step": 3174
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.6211324832006296e-05,
+      "loss": 0.1805,
+      "step": 3176
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.620673460855292e-05,
+      "loss": 0.1757,
+      "step": 3178
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.6202142256845553e-05,
+      "loss": 0.1821,
+      "step": 3180
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.6197547778458885e-05,
+      "loss": 0.1782,
+      "step": 3182
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.619295117496834e-05,
+      "loss": 0.1744,
+      "step": 3184
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.6188352447950065e-05,
+      "loss": 0.1641,
+      "step": 3186
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.6183751598980943e-05,
+      "loss": 0.1783,
+      "step": 3188
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.6179148629638567e-05,
+      "loss": 0.1757,
+      "step": 3190
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.617454354150128e-05,
+      "loss": 0.183,
+      "step": 3192
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.6169936336148136e-05,
+      "loss": 0.1786,
+      "step": 3194
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.6165327015158918e-05,
+      "loss": 0.1661,
+      "step": 3196
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.616071558011414e-05,
+      "loss": 0.1915,
+      "step": 3198
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.6156102032595032e-05,
+      "loss": 0.1657,
+      "step": 3200
+    },
+    {
+      "epoch": 1.48,
+      "eval_loss": 0.2072741985321045,
+      "eval_runtime": 72.4312,
+      "eval_samples_per_second": 27.612,
+      "eval_steps_per_second": 13.806,
+      "step": 3200
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.615148637418356e-05,
+      "loss": 0.1678,
+      "step": 3202
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.61468686064624e-05,
+      "loss": 0.1802,
+      "step": 3204
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.6142248731014964e-05,
+      "loss": 0.1872,
+      "step": 3206
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.6137626749425377e-05,
+      "loss": 0.1844,
+      "step": 3208
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.613300266327849e-05,
+      "loss": 0.1824,
+      "step": 3210
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.6128376474159883e-05,
+      "loss": 0.1802,
+      "step": 3212
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.612374818365584e-05,
+      "loss": 0.1796,
+      "step": 3214
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.611911779335338e-05,
+      "loss": 0.1697,
+      "step": 3216
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6114485304840234e-05,
+      "loss": 0.1686,
+      "step": 3218
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6109850719704864e-05,
+      "loss": 0.179,
+      "step": 3220
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6105214039536433e-05,
+      "loss": 0.1692,
+      "step": 3222
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6100575265924843e-05,
+      "loss": 0.1554,
+      "step": 3224
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6095934400460693e-05,
+      "loss": 0.1817,
+      "step": 3226
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6091291444735318e-05,
+      "loss": 0.1933,
+      "step": 3228
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6086646400340756e-05,
+      "loss": 0.1647,
+      "step": 3230
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6081999268869765e-05,
+      "loss": 0.1643,
+      "step": 3232
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.6077350051915824e-05,
+      "loss": 0.176,
+      "step": 3234
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.607269875107312e-05,
+      "loss": 0.1822,
+      "step": 3236
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.606804536793656e-05,
+      "loss": 0.176,
+      "step": 3238
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6063389904101764e-05,
+      "loss": 0.1686,
+      "step": 3240
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6058732361165054e-05,
+      "loss": 0.1702,
+      "step": 3242
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6054072740723488e-05,
+      "loss": 0.1783,
+      "step": 3244
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6049411044374815e-05,
+      "loss": 0.1739,
+      "step": 3246
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6044747273717512e-05,
+      "loss": 0.1745,
+      "step": 3248
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.604008143035075e-05,
+      "loss": 0.1723,
+      "step": 3250
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6035413515874427e-05,
+      "loss": 0.1757,
+      "step": 3252
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.603074353188914e-05,
+      "loss": 0.176,
+      "step": 3254
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.60260714799962e-05,
+      "loss": 0.1894,
+      "step": 3256
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.6021397361797632e-05,
+      "loss": 0.1707,
+      "step": 3258
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.601672117889616e-05,
+      "loss": 0.199,
+      "step": 3260
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.6012042932895224e-05,
+      "loss": 0.1734,
+      "step": 3262
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.6007362625398965e-05,
+      "loss": 0.1683,
+      "step": 3264
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.6002680258012233e-05,
+      "loss": 0.18,
+      "step": 3266
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.5997995832340586e-05,
+      "loss": 0.1703,
+      "step": 3268
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.5993309349990293e-05,
+      "loss": 0.1721,
+      "step": 3270
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.598862081256831e-05,
+      "loss": 0.175,
+      "step": 3272
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.598393022168232e-05,
+      "loss": 0.1766,
+      "step": 3274
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.5979237578940696e-05,
+      "loss": 0.1808,
+      "step": 3276
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.5974542885952518e-05,
+      "loss": 0.1743,
+      "step": 3278
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.5969846144327574e-05,
+      "loss": 0.1711,
+      "step": 3280
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.5965147355676344e-05,
+      "loss": 0.1701,
+      "step": 3282
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.5960446521610017e-05,
+      "loss": 0.1711,
+      "step": 3284
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.5955743643740488e-05,
+      "loss": 0.1765,
+      "step": 3286
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.5951038723680343e-05,
+      "loss": 0.1813,
+      "step": 3288
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.594633176304287e-05,
+      "loss": 0.1804,
+      "step": 3290
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.5941622763442058e-05,
+      "loss": 0.1691,
+      "step": 3292
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.5936911726492607e-05,
+      "loss": 0.1666,
+      "step": 3294
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.5932198653809886e-05,
+      "loss": 0.1807,
+      "step": 3296
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.592748354701e-05,
+      "loss": 0.1658,
+      "step": 3298
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.5922766407709724e-05,
+      "loss": 0.1872,
+      "step": 3300
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.591804723752654e-05,
+      "loss": 0.1832,
+      "step": 3302
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.5913326038078617e-05,
+      "loss": 0.1718,
+      "step": 3304
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.5908602810984835e-05,
+      "loss": 0.178,
+      "step": 3306
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.5903877557864765e-05,
+      "loss": 0.1793,
+      "step": 3308
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.589915028033866e-05,
+      "loss": 0.1754,
+      "step": 3310
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.5894420980027484e-05,
+      "loss": 0.1764,
+      "step": 3312
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.588968965855288e-05,
+      "loss": 0.1688,
+      "step": 3314
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.5884956317537203e-05,
+      "loss": 0.1763,
+      "step": 3316
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.5880220958603475e-05,
+      "loss": 0.1826,
+      "step": 3318
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.5875483583375434e-05,
+      "loss": 0.1736,
+      "step": 3320
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.587074419347749e-05,
+      "loss": 0.182,
+      "step": 3322
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.586600279053476e-05,
+      "loss": 0.1798,
+      "step": 3324
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5861259376173042e-05,
+      "loss": 0.1765,
+      "step": 3326
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5856513952018827e-05,
+      "loss": 0.1717,
+      "step": 3328
+    },
+    {
+      "epoch": 1.54,
+      "eval_loss": 0.2057955116033554,
+      "eval_runtime": 72.4415,
+      "eval_samples_per_second": 27.608,
+      "eval_steps_per_second": 13.804,
+      "step": 3328
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5851766519699294e-05,
+      "loss": 0.1739,
+      "step": 3330
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5847017080842308e-05,
+      "loss": 0.1697,
+      "step": 3332
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5842265637076424e-05,
+      "loss": 0.1934,
+      "step": 3334
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5837512190030884e-05,
+      "loss": 0.164,
+      "step": 3336
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5832756741335627e-05,
+      "loss": 0.1656,
+      "step": 3338
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5827999292621263e-05,
+      "loss": 0.1773,
+      "step": 3340
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5823239845519087e-05,
+      "loss": 0.1749,
+      "step": 3342
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.58184784016611e-05,
+      "loss": 0.1923,
+      "step": 3344
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.5813714962679964e-05,
+      "loss": 0.1669,
+      "step": 3346
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.580894953020904e-05,
+      "loss": 0.1723,
+      "step": 3348
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.580418210588237e-05,
+      "loss": 0.1798,
+      "step": 3350
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.579941269133467e-05,
+      "loss": 0.1731,
+      "step": 3352
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.5794641288201346e-05,
+      "loss": 0.1771,
+      "step": 3354
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.578986789811849e-05,
+      "loss": 0.1656,
+      "step": 3356
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.5785092522722863e-05,
+      "loss": 0.1675,
+      "step": 3358
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.5780315163651922e-05,
+      "loss": 0.1629,
+      "step": 3360
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.5775535822543796e-05,
+      "loss": 0.1708,
+      "step": 3362
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.5770754501037284e-05,
+      "loss": 0.1921,
+      "step": 3364
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.5765971200771883e-05,
+      "loss": 0.1829,
+      "step": 3366
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.576118592338776e-05,
+      "loss": 0.1668,
+      "step": 3368
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.5756398670525753e-05,
+      "loss": 0.1737,
+      "step": 3370
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.5751609443827387e-05,
+      "loss": 0.1727,
+      "step": 3372
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.574681824493486e-05,
+      "loss": 0.1752,
+      "step": 3374
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.574202507549105e-05,
+      "loss": 0.1796,
+      "step": 3376
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.5737229937139504e-05,
+      "loss": 0.1845,
+      "step": 3378
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.5732432831524448e-05,
+      "loss": 0.1914,
+      "step": 3380
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.5727633760290785e-05,
+      "loss": 0.1719,
+      "step": 3382
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.5722832725084088e-05,
+      "loss": 0.1699,
+      "step": 3384
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.5718029727550602e-05,
+      "loss": 0.1761,
+      "step": 3386
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.571322476933725e-05,
+      "loss": 0.1677,
+      "step": 3388
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.5708417852091625e-05,
+      "loss": 0.1664,
+      "step": 3390
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.570360897746199e-05,
+      "loss": 0.1877,
+      "step": 3392
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.569879814709728e-05,
+      "loss": 0.1785,
+      "step": 3394
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.5693985362647105e-05,
+      "loss": 0.1708,
+      "step": 3396
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.568917062576174e-05,
+      "loss": 0.1804,
+      "step": 3398
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.568435393809213e-05,
+      "loss": 0.1715,
+      "step": 3400
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.567953530128989e-05,
+      "loss": 0.1571,
+      "step": 3402
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.56747147170073e-05,
+      "loss": 0.1747,
+      "step": 3404
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.5669892186897317e-05,
+      "loss": 0.1839,
+      "step": 3406
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.566506771261356e-05,
+      "loss": 0.1781,
+      "step": 3408
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.5660241295810307e-05,
+      "loss": 0.189,
+      "step": 3410
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 1.565541293814251e-05,
+      "loss": 0.172,
+      "step": 3412
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.5650582641265784e-05,
+      "loss": 0.193,
+      "step": 3414
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.564575040683642e-05,
+      "loss": 0.1849,
+      "step": 3416
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.5640916236511357e-05,
+      "loss": 0.1806,
+      "step": 3418
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.5636080131948204e-05,
+      "loss": 0.1712,
+      "step": 3420
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.5631242094805235e-05,
+      "loss": 0.1691,
+      "step": 3422
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.562640212674138e-05,
+      "loss": 0.1831,
+      "step": 3424
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.5621560229416248e-05,
+      "loss": 0.1702,
+      "step": 3426
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.561671640449009e-05,
+      "loss": 0.1728,
+      "step": 3428
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.5611870653623826e-05,
+      "loss": 0.1805,
+      "step": 3430
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.560702297847904e-05,
+      "loss": 0.1719,
+      "step": 3432
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.5602173380717966e-05,
+      "loss": 0.1746,
+      "step": 3434
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.559732186200351e-05,
+      "loss": 0.1687,
+      "step": 3436
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.559246842399923e-05,
+      "loss": 0.1827,
+      "step": 3438
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.5587613068369337e-05,
+      "loss": 0.173,
+      "step": 3440
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.5582755796778712e-05,
+      "loss": 0.1766,
+      "step": 3442
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.557789661089288e-05,
+      "loss": 0.1601,
+      "step": 3444
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.5573035512378026e-05,
+      "loss": 0.172,
+      "step": 3446
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.5568172502901006e-05,
+      "loss": 0.1737,
+      "step": 3448
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.5563307584129302e-05,
+      "loss": 0.1637,
+      "step": 3450
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.5558440757731076e-05,
+      "loss": 0.1659,
+      "step": 3452
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 1.5553572025375135e-05,
+      "loss": 0.1646,
+      "step": 3454
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.5548701388730936e-05,
+      "loss": 0.1754,
+      "step": 3456
+    },
+    {
+      "epoch": 1.6,
+      "eval_loss": 0.20447595417499542,
+      "eval_runtime": 72.4542,
+      "eval_samples_per_second": 27.604,
+      "eval_steps_per_second": 13.802,
+      "step": 3456
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.5543828849468596e-05,
+      "loss": 0.1781,
+      "step": 3458
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.553895440925888e-05,
+      "loss": 0.1663,
+      "step": 3460
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.55340780697732e-05,
+      "loss": 0.1762,
+      "step": 3462
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.5529199832683634e-05,
+      "loss": 0.1791,
+      "step": 3464
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.55243196996629e-05,
+      "loss": 0.1801,
+      "step": 3466
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.5519437672384366e-05,
+      "loss": 0.1771,
+      "step": 3468
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.551455375252205e-05,
+      "loss": 0.1662,
+      "step": 3470
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.5509667941750622e-05,
+      "loss": 0.1664,
+      "step": 3472
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.55047802417454e-05,
+      "loss": 0.1682,
+      "step": 3474
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.5499890654182346e-05,
+      "loss": 0.1818,
+      "step": 3476
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.549499918073807e-05,
+      "loss": 0.167,
+      "step": 3478
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.549010582308984e-05,
+      "loss": 0.17,
+      "step": 3480
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.5485210582915548e-05,
+      "loss": 0.1849,
+      "step": 3482
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.548031346189375e-05,
+      "loss": 0.1808,
+      "step": 3484
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.5475414461703645e-05,
+      "loss": 0.1711,
+      "step": 3486
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.5470513584025063e-05,
+      "loss": 0.1797,
+      "step": 3488
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.5465610830538496e-05,
+      "loss": 0.1724,
+      "step": 3490
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.546070620292506e-05,
+      "loss": 0.1845,
+      "step": 3492
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.545579970286654e-05,
+      "loss": 0.1701,
+      "step": 3494
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.545089133204533e-05,
+      "loss": 0.1941,
+      "step": 3496
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.544598109214449e-05,
+      "loss": 0.1616,
+      "step": 3498
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.5441068984847718e-05,
+      "loss": 0.1813,
+      "step": 3500
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.543615501183934e-05,
+      "loss": 0.1709,
+      "step": 3502
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.543123917480433e-05,
+      "loss": 0.1731,
+      "step": 3504
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.5426321475428303e-05,
+      "loss": 0.1823,
+      "step": 3506
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.5421401915397512e-05,
+      "loss": 0.1748,
+      "step": 3508
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.5416480496398843e-05,
+      "loss": 0.172,
+      "step": 3510
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.5411557220119823e-05,
+      "loss": 0.1785,
+      "step": 3512
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.5406632088248616e-05,
+      "loss": 0.1798,
+      "step": 3514
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.540170510247402e-05,
+      "loss": 0.174,
+      "step": 3516
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.539677626448547e-05,
+      "loss": 0.1766,
+      "step": 3518
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.5391845575973046e-05,
+      "loss": 0.172,
+      "step": 3520
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.538691303862744e-05,
+      "loss": 0.1642,
+      "step": 3522
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.5381978654139997e-05,
+      "loss": 0.1874,
+      "step": 3524
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.537704242420269e-05,
+      "loss": 0.1746,
+      "step": 3526
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.5372104350508125e-05,
+      "loss": 0.1748,
+      "step": 3528
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.5367164434749536e-05,
+      "loss": 0.1659,
+      "step": 3530
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.536222267862079e-05,
+      "loss": 0.1806,
+      "step": 3532
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.5357279083816392e-05,
+      "loss": 0.1692,
+      "step": 3534
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.5352333652031474e-05,
+      "loss": 0.161,
+      "step": 3536
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.534738638496179e-05,
+      "loss": 0.2041,
+      "step": 3538
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.5342437284303738e-05,
+      "loss": 0.1796,
+      "step": 3540
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.5337486351754326e-05,
+      "loss": 0.1741,
+      "step": 3542
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.533253358901121e-05,
+      "loss": 0.1636,
+      "step": 3544
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.5327578997772657e-05,
+      "loss": 0.1667,
+      "step": 3546
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.5322622579737577e-05,
+      "loss": 0.1695,
+      "step": 3548
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.5317664336605488e-05,
+      "loss": 0.175,
+      "step": 3550
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.5312704270076548e-05,
+      "loss": 0.1661,
+      "step": 3552
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.5307742381851537e-05,
+      "loss": 0.1656,
+      "step": 3554
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.5302778673631854e-05,
+      "loss": 0.1695,
+      "step": 3556
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.529781314711953e-05,
+      "loss": 0.1827,
+      "step": 3558
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.5292845804017216e-05,
+      "loss": 0.1717,
+      "step": 3560
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.528787664602818e-05,
+      "loss": 0.1791,
+      "step": 3562
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.5282905674856326e-05,
+      "loss": 0.1699,
+      "step": 3564
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.5277932892206165e-05,
+      "loss": 0.1708,
+      "step": 3566
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.5272958299782842e-05,
+      "loss": 0.163,
+      "step": 3568
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.526798189929211e-05,
+      "loss": 0.1801,
+      "step": 3570
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.5263003692440355e-05,
+      "loss": 0.1784,
+      "step": 3572
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.5258023680934569e-05,
+      "loss": 0.1696,
+      "step": 3574
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.5253041866482377e-05,
+      "loss": 0.1726,
+      "step": 3576
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.5248058250792008e-05,
+      "loss": 0.1825,
+      "step": 3578
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.5243072835572319e-05,
+      "loss": 0.1678,
+      "step": 3580
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.5238085622532778e-05,
+      "loss": 0.1676,
+      "step": 3582
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 1.5233096613383478e-05,
+      "loss": 0.1761,
+      "step": 3584
+    },
+    {
+      "epoch": 1.65,
+      "eval_loss": 0.2031940519809723,
+      "eval_runtime": 72.4441,
+      "eval_samples_per_second": 27.608,
+      "eval_steps_per_second": 13.804,
+      "step": 3584
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.5228105809835114e-05,
+      "loss": 0.18,
+      "step": 3586
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.5223113213599013e-05,
+      "loss": 0.1765,
+      "step": 3588
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.5218118826387099e-05,
+      "loss": 0.169,
+      "step": 3590
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.521312264991192e-05,
+      "loss": 0.1714,
+      "step": 3592
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.5208124685886643e-05,
+      "loss": 0.1791,
+      "step": 3594
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.5203124936025038e-05,
+      "loss": 0.1587,
+      "step": 3596
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.5198123402041489e-05,
+      "loss": 0.1709,
+      "step": 3598
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.5193120085650996e-05,
+      "loss": 0.1772,
+      "step": 3600
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.5188114988569166e-05,
+      "loss": 0.1699,
+      "step": 3602
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.5183108112512217e-05,
+      "loss": 0.1755,
+      "step": 3604
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.5178099459196985e-05,
+      "loss": 0.1892,
+      "step": 3606
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5173089030340897e-05,
+      "loss": 0.1621,
+      "step": 3608
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5168076827662014e-05,
+      "loss": 0.1774,
+      "step": 3610
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5163062852878984e-05,
+      "loss": 0.1712,
+      "step": 3612
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5158047107711071e-05,
+      "loss": 0.1818,
+      "step": 3614
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5153029593878145e-05,
+      "loss": 0.1777,
+      "step": 3616
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5148010313100688e-05,
+      "loss": 0.178,
+      "step": 3618
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5142989267099775e-05,
+      "loss": 0.17,
+      "step": 3620
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.51379664575971e-05,
+      "loss": 0.1792,
+      "step": 3622
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5132941886314957e-05,
+      "loss": 0.173,
+      "step": 3624
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.5127915554976235e-05,
+      "loss": 0.1815,
+      "step": 3626
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 1.512288746530444e-05,
+      "loss": 0.1668,
+      "step": 3628
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5117857619023677e-05,
+      "loss": 0.1613,
+      "step": 3630
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5112826017858648e-05,
+      "loss": 0.1682,
+      "step": 3632
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5107792663534663e-05,
+      "loss": 0.1724,
+      "step": 3634
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5102757557777633e-05,
+      "loss": 0.1681,
+      "step": 3636
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5097720702314055e-05,
+      "loss": 0.1718,
+      "step": 3638
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5092682098871052e-05,
+      "loss": 0.1747,
+      "step": 3640
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5087641749176327e-05,
+      "loss": 0.1646,
+      "step": 3642
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5082599654958183e-05,
+      "loss": 0.1716,
+      "step": 3644
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5077555817945536e-05,
+      "loss": 0.1765,
+      "step": 3646
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.507251023986788e-05,
+      "loss": 0.1675,
+      "step": 3648
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.5067462922455314e-05,
+      "loss": 0.1751,
+      "step": 3650
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.506241386743854e-05,
+      "loss": 0.1804,
+      "step": 3652
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.5057363076548852e-05,
+      "loss": 0.1743,
+      "step": 3654
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.505231055151813e-05,
+      "loss": 0.1713,
+      "step": 3656
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.5047256294078866e-05,
+      "loss": 0.1808,
+      "step": 3658
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.5042200305964123e-05,
+      "loss": 0.1746,
+      "step": 3660
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.503714258890758e-05,
+      "loss": 0.1851,
+      "step": 3662
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.5032083144643496e-05,
+      "loss": 0.1822,
+      "step": 3664
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.502702197490673e-05,
+      "loss": 0.1625,
+      "step": 3666
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.5021959081432722e-05,
+      "loss": 0.1655,
+      "step": 3668
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.5016894465957514e-05,
+      "loss": 0.1528,
+      "step": 3670
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.5011828130217732e-05,
+      "loss": 0.1736,
+      "step": 3672
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.5006760075950592e-05,
+      "loss": 0.1772,
+      "step": 3674
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.5001690304893907e-05,
+      "loss": 0.1628,
+      "step": 3676
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.4996618818786066e-05,
+      "loss": 0.1675,
+      "step": 3678
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.4991545619366055e-05,
+      "loss": 0.1693,
+      "step": 3680
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.4986470708373446e-05,
+      "loss": 0.1738,
+      "step": 3682
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.4981394087548396e-05,
+      "loss": 0.1635,
+      "step": 3684
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.497631575863165e-05,
+      "loss": 0.1765,
+      "step": 3686
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.4971235723364542e-05,
+      "loss": 0.1594,
+      "step": 3688
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.4966153983488978e-05,
+      "loss": 0.1741,
+      "step": 3690
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 1.4961070540747464e-05,
+      "loss": 0.1725,
+      "step": 3692
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4955985396883086e-05,
+      "loss": 0.1713,
+      "step": 3694
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4950898553639505e-05,
+      "loss": 0.183,
+      "step": 3696
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4945810012760972e-05,
+      "loss": 0.1726,
+      "step": 3698
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4940719775992326e-05,
+      "loss": 0.1673,
+      "step": 3700
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4935627845078966e-05,
+      "loss": 0.1677,
+      "step": 3702
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4930534221766899e-05,
+      "loss": 0.1633,
+      "step": 3704
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4925438907802694e-05,
+      "loss": 0.1711,
+      "step": 3706
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4920341904933504e-05,
+      "loss": 0.1837,
+      "step": 3708
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4915243214907067e-05,
+      "loss": 0.1807,
+      "step": 3710
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.491014283947169e-05,
+      "loss": 0.1542,
+      "step": 3712
+    },
+    {
+      "epoch": 1.71,
+      "eval_loss": 0.2018110603094101,
+      "eval_runtime": 72.3821,
+      "eval_samples_per_second": 27.631,
+      "eval_steps_per_second": 13.816,
+      "step": 3712
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4905040780376264e-05,
+      "loss": 0.1665,
+      "step": 3714
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.489993703937026e-05,
+      "loss": 0.1689,
+      "step": 3716
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.4894831618203718e-05,
+      "loss": 0.1753,
+      "step": 3718
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.4889724518627255e-05,
+      "loss": 0.1668,
+      "step": 3720
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.4884615742392072e-05,
+      "loss": 0.1742,
+      "step": 3722
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.4879505291249937e-05,
+      "loss": 0.1819,
+      "step": 3724
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.4874393166953194e-05,
+      "loss": 0.1779,
+      "step": 3726
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.4869279371254764e-05,
+      "loss": 0.1663,
+      "step": 3728
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.4864163905908133e-05,
+      "loss": 0.1642,
+      "step": 3730
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.4859046772667372e-05,
+      "loss": 0.1801,
+      "step": 3732
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.485392797328711e-05,
+      "loss": 0.1792,
+      "step": 3734
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.4848807509522561e-05,
+      "loss": 0.1763,
+      "step": 3736
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.4843685383129498e-05,
+      "loss": 0.1724,
+      "step": 3738
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.4838561595864269e-05,
+      "loss": 0.1606,
+      "step": 3740
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.4833436149483795e-05,
+      "loss": 0.1739,
+      "step": 3742
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.4828309045745559e-05,
+      "loss": 0.1666,
+      "step": 3744
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.4823180286407617e-05,
+      "loss": 0.1744,
+      "step": 3746
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.4818049873228593e-05,
+      "loss": 0.1736,
+      "step": 3748
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.4812917807967672e-05,
+      "loss": 0.1652,
+      "step": 3750
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.4807784092384618e-05,
+      "loss": 0.1821,
+      "step": 3752
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.4802648728239743e-05,
+      "loss": 0.162,
+      "step": 3754
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.479751171729394e-05,
+      "loss": 0.175,
+      "step": 3756
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 1.4792373061308662e-05,
+      "loss": 0.1726,
+      "step": 3758
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.478723276204592e-05,
+      "loss": 0.1639,
+      "step": 3760
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.4782090821268298e-05,
+      "loss": 0.1698,
+      "step": 3762
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.4776947240738936e-05,
+      "loss": 0.1714,
+      "step": 3764
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.4771802022221537e-05,
+      "loss": 0.1581,
+      "step": 3766
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.4766655167480373e-05,
+      "loss": 0.1773,
+      "step": 3768
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.4761506678280264e-05,
+      "loss": 0.1723,
+      "step": 3770
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.47563565563866e-05,
+      "loss": 0.1596,
+      "step": 3772
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.4751204803565334e-05,
+      "loss": 0.1772,
+      "step": 3774
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.4746051421582964e-05,
+      "loss": 0.1662,
+      "step": 3776
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.474089641220656e-05,
+      "loss": 0.1668,
+      "step": 3778
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.4735739777203746e-05,
+      "loss": 0.1829,
+      "step": 3780
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.4730581518342699e-05,
+      "loss": 0.1739,
+      "step": 3782
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.4725421637392161e-05,
+      "loss": 0.1609,
+      "step": 3784
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.4720260136121424e-05,
+      "loss": 0.1811,
+      "step": 3786
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.4715097016300337e-05,
+      "loss": 0.158,
+      "step": 3788
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.4709932279699309e-05,
+      "loss": 0.1675,
+      "step": 3790
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.470476592808929e-05,
+      "loss": 0.1704,
+      "step": 3792
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.46995979632418e-05,
+      "loss": 0.1575,
+      "step": 3794
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.4694428386928898e-05,
+      "loss": 0.177,
+      "step": 3796
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.468925720092321e-05,
+      "loss": 0.1807,
+      "step": 3798
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.4684084406997903e-05,
+      "loss": 0.1598,
+      "step": 3800
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.4678910006926697e-05,
+      "loss": 0.1723,
+      "step": 3802
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.4673734002483866e-05,
+      "loss": 0.1853,
+      "step": 3804
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.4668556395444231e-05,
+      "loss": 0.173,
+      "step": 3806
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.4663377187583167e-05,
+      "loss": 0.1746,
+      "step": 3808
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.465819638067659e-05,
+      "loss": 0.1628,
+      "step": 3810
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.4653013976500977e-05,
+      "loss": 0.1678,
+      "step": 3812
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.4647829976833336e-05,
+      "loss": 0.1665,
+      "step": 3814
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.4642644383451239e-05,
+      "loss": 0.1689,
+      "step": 3816
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.4637457198132788e-05,
+      "loss": 0.1699,
+      "step": 3818
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.4632268422656645e-05,
+      "loss": 0.1655,
+      "step": 3820
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.4627078058802014e-05,
+      "loss": 0.1793,
+      "step": 3822
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4621886108348633e-05,
+      "loss": 0.1705,
+      "step": 3824
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4616692573076803e-05,
+      "loss": 0.1658,
+      "step": 3826
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4611497454767347e-05,
+      "loss": 0.1634,
+      "step": 3828
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4606300755201646e-05,
+      "loss": 0.1753,
+      "step": 3830
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4601102476161621e-05,
+      "loss": 0.1705,
+      "step": 3832
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4595902619429732e-05,
+      "loss": 0.1727,
+      "step": 3834
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4590701186788976e-05,
+      "loss": 0.1688,
+      "step": 3836
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4585498180022896e-05,
+      "loss": 0.1606,
+      "step": 3838
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4580293600915579e-05,
+      "loss": 0.1695,
+      "step": 3840
+    },
+    {
+      "epoch": 1.77,
+      "eval_loss": 0.19979771971702576,
+      "eval_runtime": 72.4526,
+      "eval_samples_per_second": 27.604,
+      "eval_steps_per_second": 13.802,
+      "step": 3840
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4575087451251637e-05,
+      "loss": 0.163,
+      "step": 3842
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4569879732816235e-05,
+      "loss": 0.1754,
+      "step": 3844
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.4564670447395067e-05,
+      "loss": 0.1772,
+      "step": 3846
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.455945959677437e-05,
+      "loss": 0.1611,
+      "step": 3848
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.4554247182740915e-05,
+      "loss": 0.1649,
+      "step": 3850
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.4549033207082006e-05,
+      "loss": 0.1682,
+      "step": 3852
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.454381767158549e-05,
+      "loss": 0.171,
+      "step": 3854
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.4538600578039739e-05,
+      "loss": 0.179,
+      "step": 3856
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.4533381928233668e-05,
+      "loss": 0.1837,
+      "step": 3858
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.4528161723956725e-05,
+      "loss": 0.173,
+      "step": 3860
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.4522939966998882e-05,
+      "loss": 0.1703,
+      "step": 3862
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.4517716659150654e-05,
+      "loss": 0.1584,
+      "step": 3864
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.4512491802203082e-05,
+      "loss": 0.1605,
+      "step": 3866
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.450726539794774e-05,
+      "loss": 0.1736,
+      "step": 3868
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.4502037448176734e-05,
+      "loss": 0.1696,
+      "step": 3870
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.4496807954682697e-05,
+      "loss": 0.1794,
+      "step": 3872
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.4491576919258794e-05,
+      "loss": 0.1552,
+      "step": 3874
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.448634434369871e-05,
+      "loss": 0.1524,
+      "step": 3876
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.4481110229796676e-05,
+      "loss": 0.1717,
+      "step": 3878
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.4475874579347436e-05,
+      "loss": 0.19,
+      "step": 3880
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.4470637394146258e-05,
+      "loss": 0.1719,
+      "step": 3882
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.4465398675988955e-05,
+      "loss": 0.1845,
+      "step": 3884
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.4460158426671847e-05,
+      "loss": 0.1835,
+      "step": 3886
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.4454916647991782e-05,
+      "loss": 0.1671,
+      "step": 3888
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.4449673341746142e-05,
+      "loss": 0.1722,
+      "step": 3890
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.4444428509732827e-05,
+      "loss": 0.1723,
+      "step": 3892
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.4439182153750252e-05,
+      "loss": 0.1694,
+      "step": 3894
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.4433934275597374e-05,
+      "loss": 0.1772,
+      "step": 3896
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.4428684877073656e-05,
+      "loss": 0.161,
+      "step": 3898
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.4423433959979085e-05,
+      "loss": 0.1528,
+      "step": 3900
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.441818152611417e-05,
+      "loss": 0.1736,
+      "step": 3902
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.4412927577279943e-05,
+      "loss": 0.1865,
+      "step": 3904
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.4407672115277955e-05,
+      "loss": 0.1753,
+      "step": 3906
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.4402415141910271e-05,
+      "loss": 0.1661,
+      "step": 3908
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.439715665897948e-05,
+      "loss": 0.1774,
+      "step": 3910
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4391896668288678e-05,
+      "loss": 0.1725,
+      "step": 3912
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4386635171641497e-05,
+      "loss": 0.1706,
+      "step": 3914
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4381372170842067e-05,
+      "loss": 0.1875,
+      "step": 3916
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4376107667695042e-05,
+      "loss": 0.1775,
+      "step": 3918
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4370841664005592e-05,
+      "loss": 0.1733,
+      "step": 3920
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4365574161579401e-05,
+      "loss": 0.1514,
+      "step": 3922
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4360305162222662e-05,
+      "loss": 0.1645,
+      "step": 3924
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4355034667742085e-05,
+      "loss": 0.169,
+      "step": 3926
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4349762679944896e-05,
+      "loss": 0.1657,
+      "step": 3928
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4344489200638827e-05,
+      "loss": 0.1702,
+      "step": 3930
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4339214231632128e-05,
+      "loss": 0.1667,
+      "step": 3932
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.4333937774733547e-05,
+      "loss": 0.1608,
+      "step": 3934
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.4328659831752358e-05,
+      "loss": 0.1548,
+      "step": 3936
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.4323380404498339e-05,
+      "loss": 0.1634,
+      "step": 3938
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.431809949478177e-05,
+      "loss": 0.1733,
+      "step": 3940
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.431281710441345e-05,
+      "loss": 0.1701,
+      "step": 3942
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.4307533235204678e-05,
+      "loss": 0.1726,
+      "step": 3944
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.4302247888967262e-05,
+      "loss": 0.1741,
+      "step": 3946
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.4296961067513519e-05,
+      "loss": 0.1734,
+      "step": 3948
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.429167277265627e-05,
+      "loss": 0.1683,
+      "step": 3950
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.4286383006208837e-05,
+      "loss": 0.1649,
+      "step": 3952
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.4281091769985058e-05,
+      "loss": 0.1708,
+      "step": 3954
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.4275799065799263e-05,
+      "loss": 0.172,
+      "step": 3956
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.427050489546629e-05,
+      "loss": 0.1686,
+      "step": 3958
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.4265209260801483e-05,
+      "loss": 0.1856,
+      "step": 3960
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.4259912163620682e-05,
+      "loss": 0.1593,
+      "step": 3962
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.4254613605740233e-05,
+      "loss": 0.164,
+      "step": 3964
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.424931358897698e-05,
+      "loss": 0.1844,
+      "step": 3966
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.4244012115148272e-05,
+      "loss": 0.1679,
+      "step": 3968
+    },
+    {
+      "epoch": 1.83,
+      "eval_loss": 0.19876763224601746,
+      "eval_runtime": 72.3753,
+      "eval_samples_per_second": 27.634,
+      "eval_steps_per_second": 13.817,
+      "step": 3968
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.4238709186071948e-05,
+      "loss": 0.1658,
+      "step": 3970
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.423340480356636e-05,
+      "loss": 0.1717,
+      "step": 3972
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.4228098969450343e-05,
+      "loss": 0.1719,
+      "step": 3974
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.4222791685543245e-05,
+      "loss": 0.1666,
+      "step": 3976
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.4217482953664896e-05,
+      "loss": 0.1694,
+      "step": 3978
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.4212172775635633e-05,
+      "loss": 0.1672,
+      "step": 3980
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.4206861153276288e-05,
+      "loss": 0.1732,
+      "step": 3982
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.4201548088408184e-05,
+      "loss": 0.1584,
+      "step": 3984
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.419623358285314e-05,
+      "loss": 0.1693,
+      "step": 3986
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.4190917638433472e-05,
+      "loss": 0.181,
+      "step": 3988
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.4185600256971987e-05,
+      "loss": 0.1714,
+      "step": 3990
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.4180281440291982e-05,
+      "loss": 0.1733,
+      "step": 3992
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.417496119021725e-05,
+      "loss": 0.1799,
+      "step": 3994
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.4169639508572077e-05,
+      "loss": 0.1627,
+      "step": 3996
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.4164316397181233e-05,
+      "loss": 0.1722,
+      "step": 3998
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.4158991857869987e-05,
+      "loss": 0.1731,
+      "step": 4000
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.4153665892464092e-05,
+      "loss": 0.1677,
+      "step": 4002
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.4148338502789786e-05,
+      "loss": 0.1679,
+      "step": 4004
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.4143009690673812e-05,
+      "loss": 0.1642,
+      "step": 4006
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.413767945794338e-05,
+      "loss": 0.1727,
+      "step": 4008
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.4132347806426197e-05,
+      "loss": 0.1768,
+      "step": 4010
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.4127014737950459e-05,
+      "loss": 0.1716,
+      "step": 4012
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.4121680254344845e-05,
+      "loss": 0.164,
+      "step": 4014
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.4116344357438518e-05,
+      "loss": 0.1743,
+      "step": 4016
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.4111007049061126e-05,
+      "loss": 0.1822,
+      "step": 4018
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.41056683310428e-05,
+      "loss": 0.1574,
+      "step": 4020
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4100328205214161e-05,
+      "loss": 0.1591,
+      "step": 4022
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4094986673406306e-05,
+      "loss": 0.1695,
+      "step": 4024
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4089643737450813e-05,
+      "loss": 0.176,
+      "step": 4026
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4084299399179749e-05,
+      "loss": 0.1608,
+      "step": 4028
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4078953660425651e-05,
+      "loss": 0.1745,
+      "step": 4030
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4073606523021547e-05,
+      "loss": 0.1664,
+      "step": 4032
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4068257988800943e-05,
+      "loss": 0.1645,
+      "step": 4034
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4062908059597813e-05,
+      "loss": 0.1734,
+      "step": 4036
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.405755673724662e-05,
+      "loss": 0.168,
+      "step": 4038
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 1.4052204023582305e-05,
+      "loss": 0.1737,
+      "step": 4040
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.4046849920440282e-05,
+      "loss": 0.1735,
+      "step": 4042
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.4041494429656442e-05,
+      "loss": 0.1701,
+      "step": 4044
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.4036137553067151e-05,
+      "loss": 0.1833,
+      "step": 4046
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.4030779292509254e-05,
+      "loss": 0.1636,
+      "step": 4048
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.4025419649820065e-05,
+      "loss": 0.1715,
+      "step": 4050
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.4020058626837376e-05,
+      "loss": 0.1583,
+      "step": 4052
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.4014696225399456e-05,
+      "loss": 0.1767,
+      "step": 4054
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.4009332447345032e-05,
+      "loss": 0.1772,
+      "step": 4056
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.4003967294513323e-05,
+      "loss": 0.1696,
+      "step": 4058
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.3998600768744006e-05,
+      "loss": 0.1656,
+      "step": 4060
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.399323287187723e-05,
+      "loss": 0.1747,
+      "step": 4062
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.3987863605753617e-05,
+      "loss": 0.1706,
+      "step": 4064
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.3982492972214261e-05,
+      "loss": 0.1825,
+      "step": 4066
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.3977120973100715e-05,
+      "loss": 0.1722,
+      "step": 4068
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.3971747610255015e-05,
+      "loss": 0.1673,
+      "step": 4070
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.3966372885519652e-05,
+      "loss": 0.1621,
+      "step": 4072
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.3960996800737587e-05,
+      "loss": 0.1669,
+      "step": 4074
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.3955619357752252e-05,
+      "loss": 0.1717,
+      "step": 4076
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.3950240558407541e-05,
+      "loss": 0.1623,
+      "step": 4078
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.3944860404547816e-05,
+      "loss": 0.1756,
+      "step": 4080
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 1.3939478898017896e-05,
+      "loss": 0.1661,
+      "step": 4082
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.393409604066307e-05,
+      "loss": 0.1729,
+      "step": 4084
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.3928711834329096e-05,
+      "loss": 0.1604,
+      "step": 4086
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.392332628086218e-05,
+      "loss": 0.1656,
+      "step": 4088
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.3917939382108998e-05,
+      "loss": 0.1752,
+      "step": 4090
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.3912551139916692e-05,
+      "loss": 0.1602,
+      "step": 4092
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.3907161556132856e-05,
+      "loss": 0.1758,
+      "step": 4094
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.3901770632605546e-05,
+      "loss": 0.1735,
+      "step": 4096
+    },
+    {
+      "epoch": 1.89,
+      "eval_loss": 0.19798384606838226,
+      "eval_runtime": 72.4529,
+      "eval_samples_per_second": 27.604,
+      "eval_steps_per_second": 13.802,
+      "step": 4096
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.3896378371183283e-05,
+      "loss": 0.1823,
+      "step": 4098
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.3890984773715042e-05,
+      "loss": 0.1717,
+      "step": 4100
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.3885589842050253e-05,
+      "loss": 0.178,
+      "step": 4102
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.3880193578038814e-05,
+      "loss": 0.158,
+      "step": 4104
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.387479598353107e-05,
+      "loss": 0.1657,
+      "step": 4106
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.3869397060377821e-05,
+      "loss": 0.1762,
+      "step": 4108
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.3863996810430333e-05,
+      "loss": 0.1746,
+      "step": 4110
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.3858595235540323e-05,
+      "loss": 0.1619,
+      "step": 4112
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.385319233755995e-05,
+      "loss": 0.1812,
+      "step": 4114
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.3847788118341843e-05,
+      "loss": 0.1762,
+      "step": 4116
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.3842382579739084e-05,
+      "loss": 0.1749,
+      "step": 4118
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.383697572360519e-05,
+      "loss": 0.1709,
+      "step": 4120
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.3831567551794153e-05,
+      "loss": 0.155,
+      "step": 4122
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.3826158066160395e-05,
+      "loss": 0.1704,
+      "step": 4124
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.38207472685588e-05,
+      "loss": 0.172,
+      "step": 4126
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.3815335160844705e-05,
+      "loss": 0.187,
+      "step": 4128
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.3809921744873885e-05,
+      "loss": 0.1744,
+      "step": 4130
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.380450702250257e-05,
+      "loss": 0.1719,
+      "step": 4132
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.3799090995587441e-05,
+      "loss": 0.1719,
+      "step": 4134
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.3793673665985619e-05,
+      "loss": 0.17,
+      "step": 4136
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.3788255035554676e-05,
+      "loss": 0.1633,
+      "step": 4138
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.3782835106152634e-05,
+      "loss": 0.1616,
+      "step": 4140
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.377741387963795e-05,
+      "loss": 0.163,
+      "step": 4142
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.3771991357869536e-05,
+      "loss": 0.1716,
+      "step": 4144
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.3766567542706743e-05,
+      "loss": 0.1699,
+      "step": 4146
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.3761142436009365e-05,
+      "loss": 0.172,
+      "step": 4148
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.375571603963764e-05,
+      "loss": 0.1655,
+      "step": 4150
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.375028835545225e-05,
+      "loss": 0.1703,
+      "step": 4152
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.3744859385314318e-05,
+      "loss": 0.1761,
+      "step": 4154
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.3739429131085405e-05,
+      "loss": 0.1737,
+      "step": 4156
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.3733997594627513e-05,
+      "loss": 0.1769,
+      "step": 4158
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.3728564777803089e-05,
+      "loss": 0.179,
+      "step": 4160
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.372313068247501e-05,
+      "loss": 0.189,
+      "step": 4162
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.37176953105066e-05,
+      "loss": 0.1813,
+      "step": 4164
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.3712258663761615e-05,
+      "loss": 0.1668,
+      "step": 4166
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.3706820744104255e-05,
+      "loss": 0.1519,
+      "step": 4168
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.3701381553399147e-05,
+      "loss": 0.1639,
+      "step": 4170
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.3695941093511356e-05,
+      "loss": 0.1732,
+      "step": 4172
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.369049936630639e-05,
+      "loss": 0.1685,
+      "step": 4174
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.3685056373650189e-05,
+      "loss": 0.1631,
+      "step": 4176
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.3679612117409114e-05,
+      "loss": 0.1728,
+      "step": 4178
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.3674166599449978e-05,
+      "loss": 0.1645,
+      "step": 4180
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.3668719821640018e-05,
+      "loss": 0.1738,
+      "step": 4182
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.3663271785846895e-05,
+      "loss": 0.1852,
+      "step": 4184
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.3657822493938722e-05,
+      "loss": 0.1658,
+      "step": 4186
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.3652371947784017e-05,
+      "loss": 0.1753,
+      "step": 4188
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.364692014925175e-05,
+      "loss": 0.1624,
+      "step": 4190
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.3641467100211312e-05,
+      "loss": 0.169,
+      "step": 4192
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.3636012802532517e-05,
+      "loss": 0.1717,
+      "step": 4194
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.3630557258085615e-05,
+      "loss": 0.1709,
+      "step": 4196
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.3625100468741283e-05,
+      "loss": 0.1616,
+      "step": 4198
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.361964243637062e-05,
+      "loss": 0.1715,
+      "step": 4200
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.3614183162845158e-05,
+      "loss": 0.1755,
+      "step": 4202
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.3608722650036849e-05,
+      "loss": 0.1726,
+      "step": 4204
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.360326089981807e-05,
+      "loss": 0.1719,
+      "step": 4206
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.3597797914061626e-05,
+      "loss": 0.1771,
+      "step": 4208
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.3592333694640743e-05,
+      "loss": 0.1653,
+      "step": 4210
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.3586868243429073e-05,
+      "loss": 0.1572,
+      "step": 4212
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.3581401562300682e-05,
+      "loss": 0.1794,
+      "step": 4214
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.3575933653130072e-05,
+      "loss": 0.1643,
+      "step": 4216
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.3570464517792155e-05,
+      "loss": 0.1747,
+      "step": 4218
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.3564994158162261e-05,
+      "loss": 0.1763,
+      "step": 4220
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.3559522576116154e-05,
+      "loss": 0.1633,
+      "step": 4222
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.355404977353e-05,
+      "loss": 0.1717,
+      "step": 4224
+    },
+    {
+      "epoch": 1.95,
+      "eval_loss": 0.19643232226371765,
+      "eval_runtime": 72.3894,
+      "eval_samples_per_second": 27.628,
+      "eval_steps_per_second": 13.814,
+      "step": 4224
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 10830,
+  "num_train_epochs": 5,
+  "save_steps": 128,
+  "total_flos": 1.23677314068712e+19,
+  "trial_name": null,
+  "trial_params": null
+}