diff --git "a/trainer_state.json" "b/trainer_state.json"
--- "a/trainer_state.json"
+++ "b/trainer_state.json"
@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.486668109893799,
-  "best_model_checkpoint": "checkpoints-mistral-300M/checkpoint-56000",
-  "epoch": 7.5630581204949445,
+  "best_metric": 2.401254177093506,
+  "best_model_checkpoint": "checkpoints-mistral-300M/checkpoint-96000",
+  "epoch": 13.64072654838873,
   "eval_steps": 1000,
-  "global_step": 56000,
+  "global_step": 101000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -34055,13 +34055,27373 @@
       "eval_samples_per_second": 59.99,
       "eval_steps_per_second": 5.0,
       "step": 56000
+    },
+    {
+      "epoch": 7.56,
+      "learning_rate": 4.1870747339754796e-05,
+      "loss": 2.4059,
+      "step": 56010
+    },
+    {
+      "epoch": 7.57,
+      "learning_rate": 4.182664262331522e-05,
+      "loss": 2.3906,
+      "step": 56020
+    },
+    {
+      "epoch": 7.57,
+      "learning_rate": 4.1782557384417925e-05,
+      "loss": 2.4007,
+      "step": 56030
+    },
+    {
+      "epoch": 7.57,
+      "learning_rate": 4.173849163100073e-05,
+      "loss": 2.3591,
+      "step": 56040
+    },
+    {
+      "epoch": 7.57,
+      "learning_rate": 4.169444537099814e-05,
+      "loss": 2.3478,
+      "step": 56050
+    },
+    {
+      "epoch": 7.57,
+      "learning_rate": 4.165041861234106e-05,
+      "loss": 2.3565,
+      "step": 56060
+    },
+    {
+      "epoch": 7.57,
+      "learning_rate": 4.160641136295678e-05,
+      "loss": 2.3565,
+      "step": 56070
+    },
+    {
+      "epoch": 7.57,
+      "learning_rate": 4.156242363076929e-05,
+      "loss": 2.3755,
+      "step": 56080
+    },
+    {
+      "epoch": 7.58,
+      "learning_rate": 4.151845542369888e-05,
+      "loss": 2.3951,
+      "step": 56090
+    },
+    {
+      "epoch": 7.58,
+      "learning_rate": 4.147450674966253e-05,
+      "loss": 2.3777,
+      "step": 56100
+    },
+    {
+      "epoch": 7.58,
+      "learning_rate": 4.1430577616573415e-05,
+      "loss": 2.3661,
+      "step": 56110
+    },
+    {
+      "epoch": 7.58,
+      "learning_rate": 4.138666803234147e-05,
+      "loss": 2.3462,
+      "step": 56120
+    },
+    {
+      "epoch": 7.58,
+      "learning_rate": 4.134277800487289e-05,
+      "loss": 2.3647,
+      "step": 56130
+    },
+    {
+      "epoch": 7.58,
+      "learning_rate": 4.129890754207058e-05,
+      "loss": 2.3753,
+      "step": 56140
+    },
+    {
+      "epoch": 7.58,
+      "learning_rate": 4.1255056651833775e-05,
+      "loss": 2.3671,
+      "step": 56150
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 4.121122534205806e-05,
+      "loss": 2.361,
+      "step": 56160
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 4.116741362063576e-05,
+      "loss": 2.3947,
+      "step": 56170
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 4.1123621495455515e-05,
+      "loss": 2.3554,
+      "step": 56180
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 4.107984897440252e-05,
+      "loss": 2.369,
+      "step": 56190
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 4.1036096065358364e-05,
+      "loss": 2.339,
+      "step": 56200
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 4.099236277620112e-05,
+      "loss": 2.3562,
+      "step": 56210
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 4.094864911480531e-05,
+      "loss": 2.358,
+      "step": 56220
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 4.0904955089042054e-05,
+      "loss": 2.3706,
+      "step": 56230
+    },
+    {
+      "epoch": 7.6,
+      "learning_rate": 4.086128070677877e-05,
+      "loss": 2.364,
+      "step": 56240
+    },
+    {
+      "epoch": 7.6,
+      "learning_rate": 4.081762597587941e-05,
+      "loss": 2.3612,
+      "step": 56250
+    },
+    {
+      "epoch": 7.6,
+      "learning_rate": 4.07739909042044e-05,
+      "loss": 2.3381,
+      "step": 56260
+    },
+    {
+      "epoch": 7.6,
+      "learning_rate": 4.073037549961054e-05,
+      "loss": 2.3497,
+      "step": 56270
+    },
+    {
+      "epoch": 7.6,
+      "learning_rate": 4.068677976995126e-05,
+      "loss": 2.3506,
+      "step": 56280
+    },
+    {
+      "epoch": 7.6,
+      "learning_rate": 4.0643203723076315e-05,
+      "loss": 2.3855,
+      "step": 56290
+    },
+    {
+      "epoch": 7.6,
+      "learning_rate": 4.059964736683192e-05,
+      "loss": 2.3654,
+      "step": 56300
+    },
+    {
+      "epoch": 7.61,
+      "learning_rate": 4.055611070906075e-05,
+      "loss": 2.379,
+      "step": 56310
+    },
+    {
+      "epoch": 7.61,
+      "learning_rate": 4.051259375760203e-05,
+      "loss": 2.3614,
+      "step": 56320
+    },
+    {
+      "epoch": 7.61,
+      "learning_rate": 4.046909652029132e-05,
+      "loss": 2.3609,
+      "step": 56330
+    },
+    {
+      "epoch": 7.61,
+      "learning_rate": 4.042561900496068e-05,
+      "loss": 2.3697,
+      "step": 56340
+    },
+    {
+      "epoch": 7.61,
+      "learning_rate": 4.038216121943859e-05,
+      "loss": 2.3966,
+      "step": 56350
+    },
+    {
+      "epoch": 7.61,
+      "learning_rate": 4.0338723171549966e-05,
+      "loss": 2.3521,
+      "step": 56360
+    },
+    {
+      "epoch": 7.61,
+      "learning_rate": 4.0295304869116293e-05,
+      "loss": 2.368,
+      "step": 56370
+    },
+    {
+      "epoch": 7.61,
+      "learning_rate": 4.025190631995535e-05,
+      "loss": 2.3571,
+      "step": 56380
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 4.020852753188143e-05,
+      "loss": 2.3475,
+      "step": 56390
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 4.0165168512705206e-05,
+      "loss": 2.3615,
+      "step": 56400
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 4.012182927023395e-05,
+      "loss": 2.3585,
+      "step": 56410
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 4.00785098122712e-05,
+      "loss": 2.318,
+      "step": 56420
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 4.0035210146617005e-05,
+      "loss": 2.3888,
+      "step": 56430
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 3.999193028106784e-05,
+      "loss": 2.3632,
+      "step": 56440
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 3.9948670223416565e-05,
+      "loss": 2.3704,
+      "step": 56450
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 3.990542998145262e-05,
+      "loss": 2.3451,
+      "step": 56460
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 3.986220956296175e-05,
+      "loss": 2.3573,
+      "step": 56470
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 3.981900897572615e-05,
+      "loss": 2.3571,
+      "step": 56480
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 3.977582822752439e-05,
+      "loss": 2.3243,
+      "step": 56490
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 3.973266732613167e-05,
+      "loss": 2.3619,
+      "step": 56500
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 3.9689526279319405e-05,
+      "loss": 2.3474,
+      "step": 56510
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 3.964640509485553e-05,
+      "loss": 2.3313,
+      "step": 56520
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 3.960330378050435e-05,
+      "loss": 2.3591,
+      "step": 56530
+    },
+    {
+      "epoch": 7.64,
+      "learning_rate": 3.956022234402663e-05,
+      "loss": 2.3622,
+      "step": 56540
+    },
+    {
+      "epoch": 7.64,
+      "learning_rate": 3.951716079317961e-05,
+      "loss": 2.3607,
+      "step": 56550
+    },
+    {
+      "epoch": 7.64,
+      "learning_rate": 3.9474119135716844e-05,
+      "loss": 2.3115,
+      "step": 56560
+    },
+    {
+      "epoch": 7.64,
+      "learning_rate": 3.9431097379388376e-05,
+      "loss": 2.362,
+      "step": 56570
+    },
+    {
+      "epoch": 7.64,
+      "learning_rate": 3.938809553194063e-05,
+      "loss": 2.3156,
+      "step": 56580
+    },
+    {
+      "epoch": 7.64,
+      "learning_rate": 3.9345113601116405e-05,
+      "loss": 2.3482,
+      "step": 56590
+    },
+    {
+      "epoch": 7.64,
+      "learning_rate": 3.930215159465507e-05,
+      "loss": 2.3345,
+      "step": 56600
+    },
+    {
+      "epoch": 7.65,
+      "learning_rate": 3.925920952029226e-05,
+      "loss": 2.3703,
+      "step": 56610
+    },
+    {
+      "epoch": 7.65,
+      "learning_rate": 3.921628738576005e-05,
+      "loss": 2.3394,
+      "step": 56620
+    },
+    {
+      "epoch": 7.65,
+      "learning_rate": 3.9173385198786916e-05,
+      "loss": 2.3617,
+      "step": 56630
+    },
+    {
+      "epoch": 7.65,
+      "learning_rate": 3.9130502967097834e-05,
+      "loss": 2.3196,
+      "step": 56640
+    },
+    {
+      "epoch": 7.65,
+      "learning_rate": 3.9087640698414075e-05,
+      "loss": 2.349,
+      "step": 56650
+    },
+    {
+      "epoch": 7.65,
+      "learning_rate": 3.904479840045337e-05,
+      "loss": 2.3665,
+      "step": 56660
+    },
+    {
+      "epoch": 7.65,
+      "learning_rate": 3.9001976080929835e-05,
+      "loss": 2.3308,
+      "step": 56670
+    },
+    {
+      "epoch": 7.66,
+      "learning_rate": 3.895917374755396e-05,
+      "loss": 2.3443,
+      "step": 56680
+    },
+    {
+      "epoch": 7.66,
+      "learning_rate": 3.8916391408032746e-05,
+      "loss": 2.3335,
+      "step": 56690
+    },
+    {
+      "epoch": 7.66,
+      "learning_rate": 3.88736290700695e-05,
+      "loss": 2.3683,
+      "step": 56700
+    },
+    {
+      "epoch": 7.66,
+      "learning_rate": 3.883088674136392e-05,
+      "loss": 2.3684,
+      "step": 56710
+    },
+    {
+      "epoch": 7.66,
+      "learning_rate": 3.87881644296121e-05,
+      "loss": 2.3324,
+      "step": 56720
+    },
+    {
+      "epoch": 7.66,
+      "learning_rate": 3.874546214250666e-05,
+      "loss": 2.3486,
+      "step": 56730
+    },
+    {
+      "epoch": 7.66,
+      "learning_rate": 3.8702779887736405e-05,
+      "loss": 2.339,
+      "step": 56740
+    },
+    {
+      "epoch": 7.66,
+      "learning_rate": 3.866011767298677e-05,
+      "loss": 2.3221,
+      "step": 56750
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 3.8617475505939345e-05,
+      "loss": 2.3272,
+      "step": 56760
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 3.85748533942722e-05,
+      "loss": 2.3575,
+      "step": 56770
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 3.853225134565991e-05,
+      "loss": 2.3334,
+      "step": 56780
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 3.848966936777331e-05,
+      "loss": 2.3462,
+      "step": 56790
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 3.8447107468279616e-05,
+      "loss": 2.3183,
+      "step": 56800
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 3.8404565654842424e-05,
+      "loss": 2.3276,
+      "step": 56810
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 3.8362043935121874e-05,
+      "loss": 2.3532,
+      "step": 56820
+    },
+    {
+      "epoch": 7.68,
+      "learning_rate": 3.8319542316774255e-05,
+      "loss": 2.3531,
+      "step": 56830
+    },
+    {
+      "epoch": 7.68,
+      "learning_rate": 3.8277060807452484e-05,
+      "loss": 2.3539,
+      "step": 56840
+    },
+    {
+      "epoch": 7.68,
+      "learning_rate": 3.8234599414805586e-05,
+      "loss": 2.338,
+      "step": 56850
+    },
+    {
+      "epoch": 7.68,
+      "learning_rate": 3.81921581464791e-05,
+      "loss": 2.3432,
+      "step": 56860
+    },
+    {
+      "epoch": 7.68,
+      "learning_rate": 3.8149737010115037e-05,
+      "loss": 2.326,
+      "step": 56870
+    },
+    {
+      "epoch": 7.68,
+      "learning_rate": 3.810733601335159e-05,
+      "loss": 2.3237,
+      "step": 56880
+    },
+    {
+      "epoch": 7.68,
+      "learning_rate": 3.8064955163823564e-05,
+      "loss": 2.3555,
+      "step": 56890
+    },
+    {
+      "epoch": 7.68,
+      "learning_rate": 3.80225944691618e-05,
+      "loss": 2.3421,
+      "step": 56900
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 3.7980253936993875e-05,
+      "loss": 2.3284,
+      "step": 56910
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 3.793793357494343e-05,
+      "loss": 2.3377,
+      "step": 56920
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 3.789563339063078e-05,
+      "loss": 2.3624,
+      "step": 56930
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 3.785335339167229e-05,
+      "loss": 2.3399,
+      "step": 56940
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 3.7811093585680825e-05,
+      "loss": 2.3223,
+      "step": 56950
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 3.776885398026574e-05,
+      "loss": 2.3471,
+      "step": 56960
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 3.7726634583032536e-05,
+      "loss": 2.3401,
+      "step": 56970
+    },
+    {
+      "epoch": 7.7,
+      "learning_rate": 3.768443540158329e-05,
+      "loss": 2.35,
+      "step": 56980
+    },
+    {
+      "epoch": 7.7,
+      "learning_rate": 3.7642256443516185e-05,
+      "loss": 2.3585,
+      "step": 56990
+    },
+    {
+      "epoch": 7.7,
+      "learning_rate": 3.7600097716426035e-05,
+      "loss": 2.3568,
+      "step": 57000
+    },
+    {
+      "epoch": 7.7,
+      "eval_loss": 2.4452409744262695,
+      "eval_runtime": 1269.6433,
+      "eval_samples_per_second": 59.966,
+      "eval_steps_per_second": 4.997,
+      "step": 57000
+    },
+    {
+      "epoch": 7.7,
+      "learning_rate": 3.755795922790376e-05,
+      "loss": 2.3457,
+      "step": 57010
+    },
+    {
+      "epoch": 7.7,
+      "learning_rate": 3.751584098553689e-05,
+      "loss": 2.3495,
+      "step": 57020
+    },
+    {
+      "epoch": 7.7,
+      "learning_rate": 3.7473742996909134e-05,
+      "loss": 2.3283,
+      "step": 57030
+    },
+    {
+      "epoch": 7.7,
+      "learning_rate": 3.7431665269600484e-05,
+      "loss": 2.3556,
+      "step": 57040
+    },
+    {
+      "epoch": 7.71,
+      "learning_rate": 3.7389607811187534e-05,
+      "loss": 2.3424,
+      "step": 57050
+    },
+    {
+      "epoch": 7.71,
+      "learning_rate": 3.7347570629243e-05,
+      "loss": 2.3273,
+      "step": 57060
+    },
+    {
+      "epoch": 7.71,
+      "learning_rate": 3.7305553731336116e-05,
+      "loss": 2.3028,
+      "step": 57070
+    },
+    {
+      "epoch": 7.71,
+      "learning_rate": 3.726355712503235e-05,
+      "loss": 2.3354,
+      "step": 57080
+    },
+    {
+      "epoch": 7.71,
+      "learning_rate": 3.7221580817893545e-05,
+      "loss": 2.3538,
+      "step": 57090
+    },
+    {
+      "epoch": 7.71,
+      "learning_rate": 3.71796248174779e-05,
+      "loss": 2.3368,
+      "step": 57100
+    },
+    {
+      "epoch": 7.71,
+      "learning_rate": 3.713768913133991e-05,
+      "loss": 2.3206,
+      "step": 57110
+    },
+    {
+      "epoch": 7.71,
+      "learning_rate": 3.709577376703059e-05,
+      "loss": 2.3328,
+      "step": 57120
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 3.705387873209699e-05,
+      "loss": 2.3531,
+      "step": 57130
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 3.7012004034082764e-05,
+      "loss": 2.3424,
+      "step": 57140
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 3.697014968052778e-05,
+      "loss": 2.3309,
+      "step": 57150
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 3.692831567896833e-05,
+      "loss": 2.3564,
+      "step": 57160
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 3.6886502036936944e-05,
+      "loss": 2.3151,
+      "step": 57170
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 3.684470876196253e-05,
+      "loss": 2.3369,
+      "step": 57180
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 3.6802935861570345e-05,
+      "loss": 2.3401,
+      "step": 57190
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 3.676118334328188e-05,
+      "loss": 2.3318,
+      "step": 57200
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 3.671945121461513e-05,
+      "loss": 2.3529,
+      "step": 57210
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 3.66777394830843e-05,
+      "loss": 2.3475,
+      "step": 57220
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 3.663604815619993e-05,
+      "loss": 2.3294,
+      "step": 57230
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 3.6594377241468866e-05,
+      "loss": 2.3572,
+      "step": 57240
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 3.655272674639439e-05,
+      "loss": 2.3487,
+      "step": 57250
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 3.6511096678476006e-05,
+      "loss": 2.3178,
+      "step": 57260
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 3.6469487045209535e-05,
+      "loss": 2.3433,
+      "step": 57270
+    },
+    {
+      "epoch": 7.74,
+      "learning_rate": 3.642789785408719e-05,
+      "loss": 2.3478,
+      "step": 57280
+    },
+    {
+      "epoch": 7.74,
+      "learning_rate": 3.6386329112597384e-05,
+      "loss": 2.3432,
+      "step": 57290
+    },
+    {
+      "epoch": 7.74,
+      "learning_rate": 3.634478082822503e-05,
+      "loss": 2.3185,
+      "step": 57300
+    },
+    {
+      "epoch": 7.74,
+      "learning_rate": 3.630325300845121e-05,
+      "loss": 2.3277,
+      "step": 57310
+    },
+    {
+      "epoch": 7.74,
+      "learning_rate": 3.6261745660753335e-05,
+      "loss": 2.3458,
+      "step": 57320
+    },
+    {
+      "epoch": 7.74,
+      "learning_rate": 3.6220258792605166e-05,
+      "loss": 2.3412,
+      "step": 57330
+    },
+    {
+      "epoch": 7.74,
+      "learning_rate": 3.6178792411476833e-05,
+      "loss": 2.3556,
+      "step": 57340
+    },
+    {
+      "epoch": 7.75,
+      "learning_rate": 3.6137346524834655e-05,
+      "loss": 2.3503,
+      "step": 57350
+    },
+    {
+      "epoch": 7.75,
+      "learning_rate": 3.609592114014135e-05,
+      "loss": 2.3389,
+      "step": 57360
+    },
+    {
+      "epoch": 7.75,
+      "learning_rate": 3.605451626485588e-05,
+      "loss": 2.3499,
+      "step": 57370
+    },
+    {
+      "epoch": 7.75,
+      "learning_rate": 3.601313190643354e-05,
+      "loss": 2.3472,
+      "step": 57380
+    },
+    {
+      "epoch": 7.75,
+      "learning_rate": 3.5971768072326015e-05,
+      "loss": 2.3494,
+      "step": 57390
+    },
+    {
+      "epoch": 7.75,
+      "learning_rate": 3.593042476998116e-05,
+      "loss": 2.3429,
+      "step": 57400
+    },
+    {
+      "epoch": 7.75,
+      "learning_rate": 3.588910200684321e-05,
+      "loss": 2.3304,
+      "step": 57410
+    },
+    {
+      "epoch": 7.76,
+      "learning_rate": 3.584779979035262e-05,
+      "loss": 2.3484,
+      "step": 57420
+    },
+    {
+      "epoch": 7.76,
+      "learning_rate": 3.580651812794632e-05,
+      "loss": 2.3575,
+      "step": 57430
+    },
+    {
+      "epoch": 7.76,
+      "learning_rate": 3.576525702705736e-05,
+      "loss": 2.3632,
+      "step": 57440
+    },
+    {
+      "epoch": 7.76,
+      "learning_rate": 3.5724016495115175e-05,
+      "loss": 2.3281,
+      "step": 57450
+    },
+    {
+      "epoch": 7.76,
+      "learning_rate": 3.5682796539545485e-05,
+      "loss": 2.3477,
+      "step": 57460
+    },
+    {
+      "epoch": 7.76,
+      "learning_rate": 3.564159716777021e-05,
+      "loss": 2.3106,
+      "step": 57470
+    },
+    {
+      "epoch": 7.76,
+      "learning_rate": 3.5600418387207784e-05,
+      "loss": 2.3171,
+      "step": 57480
+    },
+    {
+      "epoch": 7.76,
+      "learning_rate": 3.555926020527273e-05,
+      "loss": 2.3303,
+      "step": 57490
+    },
+    {
+      "epoch": 7.77,
+      "learning_rate": 3.5518122629375954e-05,
+      "loss": 2.3352,
+      "step": 57500
+    },
+    {
+      "epoch": 7.77,
+      "learning_rate": 3.547700566692458e-05,
+      "loss": 2.3431,
+      "step": 57510
+    },
+    {
+      "epoch": 7.77,
+      "learning_rate": 3.5435909325322135e-05,
+      "loss": 2.3256,
+      "step": 57520
+    },
+    {
+      "epoch": 7.77,
+      "learning_rate": 3.539483361196834e-05,
+      "loss": 2.3364,
+      "step": 57530
+    },
+    {
+      "epoch": 7.77,
+      "learning_rate": 3.535377853425925e-05,
+      "loss": 2.3369,
+      "step": 57540
+    },
+    {
+      "epoch": 7.77,
+      "learning_rate": 3.5312744099587146e-05,
+      "loss": 2.3373,
+      "step": 57550
+    },
+    {
+      "epoch": 7.77,
+      "learning_rate": 3.527173031534062e-05,
+      "loss": 2.3123,
+      "step": 57560
+    },
+    {
+      "epoch": 7.78,
+      "learning_rate": 3.523073718890462e-05,
+      "loss": 2.3213,
+      "step": 57570
+    },
+    {
+      "epoch": 7.78,
+      "learning_rate": 3.518976472766028e-05,
+      "loss": 2.3136,
+      "step": 57580
+    },
+    {
+      "epoch": 7.78,
+      "learning_rate": 3.514881293898504e-05,
+      "loss": 2.3088,
+      "step": 57590
+    },
+    {
+      "epoch": 7.78,
+      "learning_rate": 3.510788183025258e-05,
+      "loss": 2.347,
+      "step": 57600
+    },
+    {
+      "epoch": 7.78,
+      "learning_rate": 3.506697140883298e-05,
+      "loss": 2.334,
+      "step": 57610
+    },
+    {
+      "epoch": 7.78,
+      "learning_rate": 3.5026081682092466e-05,
+      "loss": 2.3216,
+      "step": 57620
+    },
+    {
+      "epoch": 7.78,
+      "learning_rate": 3.4985212657393605e-05,
+      "loss": 2.3527,
+      "step": 57630
+    },
+    {
+      "epoch": 7.78,
+      "learning_rate": 3.494436434209518e-05,
+      "loss": 2.3176,
+      "step": 57640
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 3.4903536743552274e-05,
+      "loss": 2.3175,
+      "step": 57650
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 3.486272986911633e-05,
+      "loss": 2.3292,
+      "step": 57660
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 3.48219437261349e-05,
+      "loss": 2.3441,
+      "step": 57670
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 3.478117832195192e-05,
+      "loss": 2.313,
+      "step": 57680
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 3.474043366390749e-05,
+      "loss": 2.323,
+      "step": 57690
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 3.46997097593381e-05,
+      "loss": 2.334,
+      "step": 57700
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 3.465900661557642e-05,
+      "loss": 2.332,
+      "step": 57710
+    },
+    {
+      "epoch": 7.8,
+      "learning_rate": 3.461832423995142e-05,
+      "loss": 2.3329,
+      "step": 57720
+    },
+    {
+      "epoch": 7.8,
+      "learning_rate": 3.457766263978827e-05,
+      "loss": 2.3681,
+      "step": 57730
+    },
+    {
+      "epoch": 7.8,
+      "learning_rate": 3.4537021822408435e-05,
+      "loss": 2.356,
+      "step": 57740
+    },
+    {
+      "epoch": 7.8,
+      "learning_rate": 3.449640179512969e-05,
+      "loss": 2.3066,
+      "step": 57750
+    },
+    {
+      "epoch": 7.8,
+      "learning_rate": 3.445580256526601e-05,
+      "loss": 2.3342,
+      "step": 57760
+    },
+    {
+      "epoch": 7.8,
+      "learning_rate": 3.441522414012761e-05,
+      "loss": 2.347,
+      "step": 57770
+    },
+    {
+      "epoch": 7.8,
+      "learning_rate": 3.437466652702096e-05,
+      "loss": 2.339,
+      "step": 57780
+    },
+    {
+      "epoch": 7.81,
+      "learning_rate": 3.43341297332489e-05,
+      "loss": 2.343,
+      "step": 57790
+    },
+    {
+      "epoch": 7.81,
+      "learning_rate": 3.4293613766110356e-05,
+      "loss": 2.3327,
+      "step": 57800
+    },
+    {
+      "epoch": 7.81,
+      "learning_rate": 3.4253118632900594e-05,
+      "loss": 2.309,
+      "step": 57810
+    },
+    {
+      "epoch": 7.81,
+      "learning_rate": 3.42126443409111e-05,
+      "loss": 2.3335,
+      "step": 57820
+    },
+    {
+      "epoch": 7.81,
+      "learning_rate": 3.417219089742959e-05,
+      "loss": 2.3286,
+      "step": 57830
+    },
+    {
+      "epoch": 7.81,
+      "learning_rate": 3.4131758309740144e-05,
+      "loss": 2.3525,
+      "step": 57840
+    },
+    {
+      "epoch": 7.81,
+      "learning_rate": 3.409134658512288e-05,
+      "loss": 2.3236,
+      "step": 57850
+    },
+    {
+      "epoch": 7.81,
+      "learning_rate": 3.405095573085444e-05,
+      "loss": 2.3538,
+      "step": 57860
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 3.401058575420734e-05,
+      "loss": 2.3131,
+      "step": 57870
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 3.397023666245069e-05,
+      "loss": 2.3593,
+      "step": 57880
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 3.3929908462849665e-05,
+      "loss": 2.3258,
+      "step": 57890
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 3.3889601162665655e-05,
+      "loss": 2.35,
+      "step": 57900
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 3.3849314769156385e-05,
+      "loss": 2.3291,
+      "step": 57910
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 3.3809049289575684e-05,
+      "loss": 2.3268,
+      "step": 57920
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 3.376880473117383e-05,
+      "loss": 2.3231,
+      "step": 57930
+    },
+    {
+      "epoch": 7.83,
+      "learning_rate": 3.372858110119706e-05,
+      "loss": 2.3359,
+      "step": 57940
+    },
+    {
+      "epoch": 7.83,
+      "learning_rate": 3.368837840688815e-05,
+      "loss": 2.3608,
+      "step": 57950
+    },
+    {
+      "epoch": 7.83,
+      "learning_rate": 3.364819665548575e-05,
+      "loss": 2.3473,
+      "step": 57960
+    },
+    {
+      "epoch": 7.83,
+      "learning_rate": 3.360803585422509e-05,
+      "loss": 2.3503,
+      "step": 57970
+    },
+    {
+      "epoch": 7.83,
+      "learning_rate": 3.356789601033737e-05,
+      "loss": 2.3186,
+      "step": 57980
+    },
+    {
+      "epoch": 7.83,
+      "learning_rate": 3.3527777131050124e-05,
+      "loss": 2.362,
+      "step": 57990
+    },
+    {
+      "epoch": 7.83,
+      "learning_rate": 3.348767922358719e-05,
+      "loss": 2.3553,
+      "step": 58000
+    },
+    {
+      "epoch": 7.83,
+      "eval_loss": 2.4380388259887695,
+      "eval_runtime": 1269.0276,
+      "eval_samples_per_second": 59.995,
+      "eval_steps_per_second": 5.0,
+      "step": 58000
+    },
+    {
+      "epoch": 7.83,
+      "learning_rate": 3.344760229516841e-05,
+      "loss": 2.3205,
+      "step": 58010
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 3.3407546353010075e-05,
+      "loss": 2.366,
+      "step": 58020
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 3.336751140432452e-05,
+      "loss": 2.3337,
+      "step": 58030
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 3.3327497456320475e-05,
+      "loss": 2.3345,
+      "step": 58040
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 3.3287504516202756e-05,
+      "loss": 2.3426,
+      "step": 58050
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 3.3247532591172426e-05,
+      "loss": 2.3443,
+      "step": 58060
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 3.320758168842678e-05,
+      "loss": 2.3106,
+      "step": 58070
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 3.316765181515925e-05,
+      "loss": 2.3283,
+      "step": 58080
+    },
+    {
+      "epoch": 7.85,
+      "learning_rate": 3.3127742978559735e-05,
+      "loss": 2.3228,
+      "step": 58090
+    },
+    {
+      "epoch": 7.85,
+      "learning_rate": 3.308785518581396e-05,
+      "loss": 2.3414,
+      "step": 58100
+    },
+    {
+      "epoch": 7.85,
+      "learning_rate": 3.3047988444104185e-05,
+      "loss": 2.3316,
+      "step": 58110
+    },
+    {
+      "epoch": 7.85,
+      "learning_rate": 3.30081427606087e-05,
+      "loss": 2.3269,
+      "step": 58120
+    },
+    {
+      "epoch": 7.85,
+      "learning_rate": 3.296831814250211e-05,
+      "loss": 2.3251,
+      "step": 58130
+    },
+    {
+      "epoch": 7.85,
+      "learning_rate": 3.292851459695522e-05,
+      "loss": 2.3408,
+      "step": 58140
+    },
+    {
+      "epoch": 7.85,
+      "learning_rate": 3.288873213113486e-05,
+      "loss": 2.3318,
+      "step": 58150
+    },
+    {
+      "epoch": 7.86,
+      "learning_rate": 3.284897075220432e-05,
+      "loss": 2.3246,
+      "step": 58160
+    },
+    {
+      "epoch": 7.86,
+      "learning_rate": 3.28092304673229e-05,
+      "loss": 2.352,
+      "step": 58170
+    },
+    {
+      "epoch": 7.86,
+      "learning_rate": 3.2769511283646284e-05,
+      "loss": 2.3389,
+      "step": 58180
+    },
+    {
+      "epoch": 7.86,
+      "learning_rate": 3.272981320832616e-05,
+      "loss": 2.335,
+      "step": 58190
+    },
+    {
+      "epoch": 7.86,
+      "learning_rate": 3.269013624851057e-05,
+      "loss": 2.3162,
+      "step": 58200
+    },
+    {
+      "epoch": 7.86,
+      "learning_rate": 3.265048041134359e-05,
+      "loss": 2.3378,
+      "step": 58210
+    },
+    {
+      "epoch": 7.86,
+      "learning_rate": 3.261084570396573e-05,
+      "loss": 2.3373,
+      "step": 58220
+    },
+    {
+      "epoch": 7.86,
+      "learning_rate": 3.2571232133513526e-05,
+      "loss": 2.3507,
+      "step": 58230
+    },
+    {
+      "epoch": 7.87,
+      "learning_rate": 3.253163970711961e-05,
+      "loss": 2.3384,
+      "step": 58240
+    },
+    {
+      "epoch": 7.87,
+      "learning_rate": 3.24920684319131e-05,
+      "loss": 2.3159,
+      "step": 58250
+    },
+    {
+      "epoch": 7.87,
+      "learning_rate": 3.245251831501904e-05,
+      "loss": 2.3311,
+      "step": 58260
+    },
+    {
+      "epoch": 7.87,
+      "learning_rate": 3.241298936355885e-05,
+      "loss": 2.3213,
+      "step": 58270
+    },
+    {
+      "epoch": 7.87,
+      "learning_rate": 3.237348158465001e-05,
+      "loss": 2.3428,
+      "step": 58280
+    },
+    {
+      "epoch": 7.87,
+      "learning_rate": 3.233399498540625e-05,
+      "loss": 2.3247,
+      "step": 58290
+    },
+    {
+      "epoch": 7.87,
+      "learning_rate": 3.2294529572937415e-05,
+      "loss": 2.3331,
+      "step": 58300
+    },
+    {
+      "epoch": 7.88,
+      "learning_rate": 3.2255085354349696e-05,
+      "loss": 2.3201,
+      "step": 58310
+    },
+    {
+      "epoch": 7.88,
+      "learning_rate": 3.2215662336745294e-05,
+      "loss": 2.2999,
+      "step": 58320
+    },
+    {
+      "epoch": 7.88,
+      "learning_rate": 3.2176260527222684e-05,
+      "loss": 2.3316,
+      "step": 58330
+    },
+    {
+      "epoch": 7.88,
+      "learning_rate": 3.2136879932876494e-05,
+      "loss": 2.3285,
+      "step": 58340
+    },
+    {
+      "epoch": 7.88,
+      "learning_rate": 3.209752056079747e-05,
+      "loss": 2.3398,
+      "step": 58350
+    },
+    {
+      "epoch": 7.88,
+      "learning_rate": 3.205818241807275e-05,
+      "loss": 2.3185,
+      "step": 58360
+    },
+    {
+      "epoch": 7.88,
+      "learning_rate": 3.2018865511785394e-05,
+      "loss": 2.3533,
+      "step": 58370
+    },
+    {
+      "epoch": 7.88,
+      "learning_rate": 3.197956984901481e-05,
+      "loss": 2.333,
+      "step": 58380
+    },
+    {
+      "epoch": 7.89,
+      "learning_rate": 3.194029543683642e-05,
+      "loss": 2.3383,
+      "step": 58390
+    },
+    {
+      "epoch": 7.89,
+      "learning_rate": 3.190104228232205e-05,
+      "loss": 2.3555,
+      "step": 58400
+    },
+    {
+      "epoch": 7.89,
+      "learning_rate": 3.18618103925395e-05,
+      "loss": 2.3419,
+      "step": 58410
+    },
+    {
+      "epoch": 7.89,
+      "learning_rate": 3.1822599774552826e-05,
+      "loss": 2.3144,
+      "step": 58420
+    },
+    {
+      "epoch": 7.89,
+      "learning_rate": 3.178341043542221e-05,
+      "loss": 2.3317,
+      "step": 58430
+    },
+    {
+      "epoch": 7.89,
+      "learning_rate": 3.1744242382204015e-05,
+      "loss": 2.3331,
+      "step": 58440
+    },
+    {
+      "epoch": 7.89,
+      "learning_rate": 3.170509562195086e-05,
+      "loss": 2.3555,
+      "step": 58450
+    },
+    {
+      "epoch": 7.9,
+      "learning_rate": 3.1665970161711406e-05,
+      "loss": 2.3398,
+      "step": 58460
+    },
+    {
+      "epoch": 7.9,
+      "learning_rate": 3.162686600853054e-05,
+      "loss": 2.3318,
+      "step": 58470
+    },
+    {
+      "epoch": 7.9,
+      "learning_rate": 3.1587783169449256e-05,
+      "loss": 2.3225,
+      "step": 58480
+    },
+    {
+      "epoch": 7.9,
+      "learning_rate": 3.154872165150481e-05,
+      "loss": 2.3441,
+      "step": 58490
+    },
+    {
+      "epoch": 7.9,
+      "learning_rate": 3.150968146173058e-05,
+      "loss": 2.3314,
+      "step": 58500
+    },
+    {
+      "epoch": 7.9,
+      "learning_rate": 3.147066260715604e-05,
+      "loss": 2.3313,
+      "step": 58510
+    },
+    {
+      "epoch": 7.9,
+      "learning_rate": 3.1431665094806883e-05,
+      "loss": 2.3191,
+      "step": 58520
+    },
+    {
+      "epoch": 7.91,
+      "learning_rate": 3.1392688931704904e-05,
+      "loss": 2.315,
+      "step": 58530
+    },
+    {
+      "epoch": 7.91,
+      "learning_rate": 3.135373412486819e-05,
+      "loss": 2.3557,
+      "step": 58540
+    },
+    {
+      "epoch": 7.91,
+      "learning_rate": 3.131480068131082e-05,
+      "loss": 2.3326,
+      "step": 58550
+    },
+    {
+      "epoch": 7.91,
+      "learning_rate": 3.1275888608043133e-05,
+      "loss": 2.3353,
+      "step": 58560
+    },
+    {
+      "epoch": 7.91,
+      "learning_rate": 3.123699791207156e-05,
+      "loss": 2.342,
+      "step": 58570
+    },
+    {
+      "epoch": 7.91,
+      "learning_rate": 3.1198128600398656e-05,
+      "loss": 2.3548,
+      "step": 58580
+    },
+    {
+      "epoch": 7.91,
+      "learning_rate": 3.115928068002328e-05,
+      "loss": 2.3248,
+      "step": 58590
+    },
+    {
+      "epoch": 7.91,
+      "learning_rate": 3.112045415794027e-05,
+      "loss": 2.3239,
+      "step": 58600
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 3.108164904114069e-05,
+      "loss": 2.336,
+      "step": 58610
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 3.104286533661169e-05,
+      "loss": 2.3429,
+      "step": 58620
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 3.100410305133669e-05,
+      "loss": 2.3351,
+      "step": 58630
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 3.096536219229512e-05,
+      "loss": 2.3383,
+      "step": 58640
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 3.0926642766462636e-05,
+      "loss": 2.321,
+      "step": 58650
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 3.0887944780810956e-05,
+      "loss": 2.3209,
+      "step": 58660
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 3.0849268242307986e-05,
+      "loss": 2.3739,
+      "step": 58670
+    },
+    {
+      "epoch": 7.93,
+      "learning_rate": 3.081061315791782e-05,
+      "loss": 2.3601,
+      "step": 58680
+    },
+    {
+      "epoch": 7.93,
+      "learning_rate": 3.077197953460063e-05,
+      "loss": 2.3201,
+      "step": 58690
+    },
+    {
+      "epoch": 7.93,
+      "learning_rate": 3.073336737931272e-05,
+      "loss": 2.3414,
+      "step": 58700
+    },
+    {
+      "epoch": 7.93,
+      "learning_rate": 3.069477669900647e-05,
+      "loss": 2.3366,
+      "step": 58710
+    },
+    {
+      "epoch": 7.93,
+      "learning_rate": 3.065620750063059e-05,
+      "loss": 2.3282,
+      "step": 58720
+    },
+    {
+      "epoch": 7.93,
+      "learning_rate": 3.061765979112972e-05,
+      "loss": 2.3203,
+      "step": 58730
+    },
+    {
+      "epoch": 7.93,
+      "learning_rate": 3.057913357744471e-05,
+      "loss": 2.3464,
+      "step": 58740
+    },
+    {
+      "epoch": 7.93,
+      "learning_rate": 3.0540628866512564e-05,
+      "loss": 2.3164,
+      "step": 58750
+    },
+    {
+      "epoch": 7.94,
+      "learning_rate": 3.0502145665266314e-05,
+      "loss": 2.3191,
+      "step": 58760
+    },
+    {
+      "epoch": 7.94,
+      "learning_rate": 3.046368398063529e-05,
+      "loss": 2.3284,
+      "step": 58770
+    },
+    {
+      "epoch": 7.94,
+      "learning_rate": 3.04252438195448e-05,
+      "loss": 2.3408,
+      "step": 58780
+    },
+    {
+      "epoch": 7.94,
+      "learning_rate": 3.0386825188916337e-05,
+      "loss": 2.3226,
+      "step": 58790
+    },
+    {
+      "epoch": 7.94,
+      "learning_rate": 3.0348428095667437e-05,
+      "loss": 2.3367,
+      "step": 58800
+    },
+    {
+      "epoch": 7.94,
+      "learning_rate": 3.031005254671194e-05,
+      "loss": 2.3602,
+      "step": 58810
+    },
+    {
+      "epoch": 7.94,
+      "learning_rate": 3.0271698548959612e-05,
+      "loss": 2.3462,
+      "step": 58820
+    },
+    {
+      "epoch": 7.95,
+      "learning_rate": 3.0233366109316498e-05,
+      "loss": 2.3453,
+      "step": 58830
+    },
+    {
+      "epoch": 7.95,
+      "learning_rate": 3.0195055234684612e-05,
+      "loss": 2.3334,
+      "step": 58840
+    },
+    {
+      "epoch": 7.95,
+      "learning_rate": 3.0156765931962135e-05,
+      "loss": 2.333,
+      "step": 58850
+    },
+    {
+      "epoch": 7.95,
+      "learning_rate": 3.0118498208043453e-05,
+      "loss": 2.3251,
+      "step": 58860
+    },
+    {
+      "epoch": 7.95,
+      "learning_rate": 3.0080252069818933e-05,
+      "loss": 2.3161,
+      "step": 58870
+    },
+    {
+      "epoch": 7.95,
+      "learning_rate": 3.0042027524175255e-05,
+      "loss": 2.3088,
+      "step": 58880
+    },
+    {
+      "epoch": 7.95,
+      "learning_rate": 3.0003824577994878e-05,
+      "loss": 2.3215,
+      "step": 58890
+    },
+    {
+      "epoch": 7.96,
+      "learning_rate": 2.9965643238156734e-05,
+      "loss": 2.3246,
+      "step": 58900
+    },
+    {
+      "epoch": 7.96,
+      "learning_rate": 2.992748351153559e-05,
+      "loss": 2.3314,
+      "step": 58910
+    },
+    {
+      "epoch": 7.96,
+      "learning_rate": 2.988934540500256e-05,
+      "loss": 2.3285,
+      "step": 58920
+    },
+    {
+      "epoch": 7.96,
+      "learning_rate": 2.985122892542463e-05,
+      "loss": 2.3271,
+      "step": 58930
+    },
+    {
+      "epoch": 7.96,
+      "learning_rate": 2.9813134079664984e-05,
+      "loss": 2.3191,
+      "step": 58940
+    },
+    {
+      "epoch": 7.96,
+      "learning_rate": 2.9775060874583022e-05,
+      "loss": 2.3279,
+      "step": 58950
+    },
+    {
+      "epoch": 7.96,
+      "learning_rate": 2.973700931703406e-05,
+      "loss": 2.3639,
+      "step": 58960
+    },
+    {
+      "epoch": 7.96,
+      "learning_rate": 2.9698979413869718e-05,
+      "loss": 2.3162,
+      "step": 58970
+    },
+    {
+      "epoch": 7.97,
+      "learning_rate": 2.9660971171937463e-05,
+      "loss": 2.3423,
+      "step": 58980
+    },
+    {
+      "epoch": 7.97,
+      "learning_rate": 2.9622984598081128e-05,
+      "loss": 2.3376,
+      "step": 58990
+    },
+    {
+      "epoch": 7.97,
+      "learning_rate": 2.958501969914044e-05,
+      "loss": 2.3136,
+      "step": 59000
+    },
+    {
+      "epoch": 7.97,
+      "eval_loss": 2.4326837062835693,
+      "eval_runtime": 1269.0324,
+      "eval_samples_per_second": 59.995,
+      "eval_steps_per_second": 5.0,
+      "step": 59000
+    },
+    {
+      "epoch": 7.97,
+      "learning_rate": 2.954707648195136e-05,
+      "loss": 2.3448,
+      "step": 59010
+    },
+    {
+      "epoch": 7.97,
+      "learning_rate": 2.9509154953345926e-05,
+      "loss": 2.3091,
+      "step": 59020
+    },
+    {
+      "epoch": 7.97,
+      "learning_rate": 2.947125512015209e-05,
+      "loss": 2.3125,
+      "step": 59030
+    },
+    {
+      "epoch": 7.97,
+      "learning_rate": 2.9433376989194186e-05,
+      "loss": 2.3406,
+      "step": 59040
+    },
+    {
+      "epoch": 7.98,
+      "learning_rate": 2.9395520567292396e-05,
+      "loss": 2.3109,
+      "step": 59050
+    },
+    {
+      "epoch": 7.98,
+      "learning_rate": 2.9357685861263224e-05,
+      "loss": 2.3572,
+      "step": 59060
+    },
+    {
+      "epoch": 7.98,
+      "learning_rate": 2.931987287791897e-05,
+      "loss": 2.3314,
+      "step": 59070
+    },
+    {
+      "epoch": 7.98,
+      "learning_rate": 2.9282081624068305e-05,
+      "loss": 2.3333,
+      "step": 59080
+    },
+    {
+      "epoch": 7.98,
+      "learning_rate": 2.9244312106515833e-05,
+      "loss": 2.368,
+      "step": 59090
+    },
+    {
+      "epoch": 7.98,
+      "learning_rate": 2.9206564332062232e-05,
+      "loss": 2.3189,
+      "step": 59100
+    },
+    {
+      "epoch": 7.98,
+      "learning_rate": 2.916883830750445e-05,
+      "loss": 2.3473,
+      "step": 59110
+    },
+    {
+      "epoch": 7.98,
+      "learning_rate": 2.9131134039635202e-05,
+      "loss": 2.2917,
+      "step": 59120
+    },
+    {
+      "epoch": 7.99,
+      "learning_rate": 2.909345153524359e-05,
+      "loss": 2.3035,
+      "step": 59130
+    },
+    {
+      "epoch": 7.99,
+      "learning_rate": 2.9055790801114603e-05,
+      "loss": 2.3223,
+      "step": 59140
+    },
+    {
+      "epoch": 7.99,
+      "learning_rate": 2.9018151844029452e-05,
+      "loss": 2.3257,
+      "step": 59150
+    },
+    {
+      "epoch": 7.99,
+      "learning_rate": 2.898053467076533e-05,
+      "loss": 2.3013,
+      "step": 59160
+    },
+    {
+      "epoch": 7.99,
+      "learning_rate": 2.8942939288095506e-05,
+      "loss": 2.3185,
+      "step": 59170
+    },
+    {
+      "epoch": 7.99,
+      "learning_rate": 2.8905365702789362e-05,
+      "loss": 2.3406,
+      "step": 59180
+    },
+    {
+      "epoch": 7.99,
+      "learning_rate": 2.8867813921612305e-05,
+      "loss": 2.3565,
+      "step": 59190
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 2.8830283951325987e-05,
+      "loss": 2.3323,
+      "step": 59200
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 2.879277579868783e-05,
+      "loss": 2.3337,
+      "step": 59210
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 2.8755289470451622e-05,
+      "loss": 2.3194,
+      "step": 59220
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 2.8717824973367014e-05,
+      "loss": 2.32,
+      "step": 59230
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 2.8680382314179883e-05,
+      "loss": 2.305,
+      "step": 59240
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 2.8642961499632084e-05,
+      "loss": 2.2792,
+      "step": 59250
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 2.8605562536461557e-05,
+      "loss": 2.2701,
+      "step": 59260
+    },
+    {
+      "epoch": 8.01,
+      "learning_rate": 2.856818543140227e-05,
+      "loss": 2.2728,
+      "step": 59270
+    },
+    {
+      "epoch": 8.01,
+      "learning_rate": 2.85308301911843e-05,
+      "loss": 2.261,
+      "step": 59280
+    },
+    {
+      "epoch": 8.01,
+      "learning_rate": 2.8493496822533824e-05,
+      "loss": 2.2888,
+      "step": 59290
+    },
+    {
+      "epoch": 8.01,
+      "learning_rate": 2.8456185332172992e-05,
+      "loss": 2.2691,
+      "step": 59300
+    },
+    {
+      "epoch": 8.01,
+      "learning_rate": 2.84188957268201e-05,
+      "loss": 2.2804,
+      "step": 59310
+    },
+    {
+      "epoch": 8.01,
+      "learning_rate": 2.8381628013189376e-05,
+      "loss": 2.261,
+      "step": 59320
+    },
+    {
+      "epoch": 8.01,
+      "learning_rate": 2.834438219799129e-05,
+      "loss": 2.2752,
+      "step": 59330
+    },
+    {
+      "epoch": 8.01,
+      "learning_rate": 2.8307158287932253e-05,
+      "loss": 2.2448,
+      "step": 59340
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 2.8269956289714706e-05,
+      "loss": 2.2735,
+      "step": 59350
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 2.823277621003721e-05,
+      "loss": 2.2932,
+      "step": 59360
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 2.8195618055594344e-05,
+      "loss": 2.3017,
+      "step": 59370
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 2.8158481833076785e-05,
+      "loss": 2.2872,
+      "step": 59380
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 2.812136754917122e-05,
+      "loss": 2.2832,
+      "step": 59390
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 2.8084275210560386e-05,
+      "loss": 2.3165,
+      "step": 59400
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 2.8047204823923024e-05,
+      "loss": 2.2778,
+      "step": 59410
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 2.801015639593408e-05,
+      "loss": 2.292,
+      "step": 59420
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 2.79731299332644e-05,
+      "loss": 2.2853,
+      "step": 59430
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 2.79361254425809e-05,
+      "loss": 2.2774,
+      "step": 59440
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 2.7899142930546547e-05,
+      "loss": 2.2974,
+      "step": 59450
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 2.7862182403820355e-05,
+      "loss": 2.2687,
+      "step": 59460
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 2.7825243869057444e-05,
+      "loss": 2.2674,
+      "step": 59470
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 2.7788327332908865e-05,
+      "loss": 2.2996,
+      "step": 59480
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 2.7751432802021783e-05,
+      "loss": 2.2919,
+      "step": 59490
+    },
+    {
+      "epoch": 8.04,
+      "learning_rate": 2.7714560283039316e-05,
+      "loss": 2.2939,
+      "step": 59500
+    },
+    {
+      "epoch": 8.04,
+      "learning_rate": 2.7677709782600756e-05,
+      "loss": 2.2781,
+      "step": 59510
+    },
+    {
+      "epoch": 8.04,
+      "learning_rate": 2.764088130734134e-05,
+      "loss": 2.2742,
+      "step": 59520
+    },
+    {
+      "epoch": 8.04,
+      "learning_rate": 2.76040748638923e-05,
+      "loss": 2.2869,
+      "step": 59530
+    },
+    {
+      "epoch": 8.04,
+      "learning_rate": 2.7567290458881003e-05,
+      "loss": 2.2744,
+      "step": 59540
+    },
+    {
+      "epoch": 8.04,
+      "learning_rate": 2.753052809893073e-05,
+      "loss": 2.2876,
+      "step": 59550
+    },
+    {
+      "epoch": 8.04,
+      "learning_rate": 2.749378779066096e-05,
+      "loss": 2.2808,
+      "step": 59560
+    },
+    {
+      "epoch": 8.05,
+      "learning_rate": 2.7457069540687026e-05,
+      "loss": 2.2729,
+      "step": 59570
+    },
+    {
+      "epoch": 8.05,
+      "learning_rate": 2.7420373355620395e-05,
+      "loss": 2.2865,
+      "step": 59580
+    },
+    {
+      "epoch": 8.05,
+      "learning_rate": 2.738369924206848e-05,
+      "loss": 2.279,
+      "step": 59590
+    },
+    {
+      "epoch": 8.05,
+      "learning_rate": 2.7347047206634866e-05,
+      "loss": 2.2872,
+      "step": 59600
+    },
+    {
+      "epoch": 8.05,
+      "learning_rate": 2.7310417255918986e-05,
+      "loss": 2.2802,
+      "step": 59610
+    },
+    {
+      "epoch": 8.05,
+      "learning_rate": 2.7273809396516422e-05,
+      "loss": 2.2981,
+      "step": 59620
+    },
+    {
+      "epoch": 8.05,
+      "learning_rate": 2.7237223635018714e-05,
+      "loss": 2.262,
+      "step": 59630
+    },
+    {
+      "epoch": 8.05,
+      "learning_rate": 2.7200659978013406e-05,
+      "loss": 2.2753,
+      "step": 59640
+    },
+    {
+      "epoch": 8.06,
+      "learning_rate": 2.716411843208417e-05,
+      "loss": 2.3008,
+      "step": 59650
+    },
+    {
+      "epoch": 8.06,
+      "learning_rate": 2.712759900381059e-05,
+      "loss": 2.2897,
+      "step": 59660
+    },
+    {
+      "epoch": 8.06,
+      "learning_rate": 2.709110169976831e-05,
+      "loss": 2.2793,
+      "step": 59670
+    },
+    {
+      "epoch": 8.06,
+      "learning_rate": 2.7054626526528932e-05,
+      "loss": 2.2619,
+      "step": 59680
+    },
+    {
+      "epoch": 8.06,
+      "learning_rate": 2.701817349066021e-05,
+      "loss": 2.2667,
+      "step": 59690
+    },
+    {
+      "epoch": 8.06,
+      "learning_rate": 2.6981742598725792e-05,
+      "loss": 2.2816,
+      "step": 59700
+    },
+    {
+      "epoch": 8.06,
+      "learning_rate": 2.694533385728536e-05,
+      "loss": 2.269,
+      "step": 59710
+    },
+    {
+      "epoch": 8.07,
+      "learning_rate": 2.6908947272894626e-05,
+      "loss": 2.2879,
+      "step": 59720
+    },
+    {
+      "epoch": 8.07,
+      "learning_rate": 2.687258285210528e-05,
+      "loss": 2.2922,
+      "step": 59730
+    },
+    {
+      "epoch": 8.07,
+      "learning_rate": 2.6836240601465097e-05,
+      "loss": 2.2951,
+      "step": 59740
+    },
+    {
+      "epoch": 8.07,
+      "learning_rate": 2.6799920527517817e-05,
+      "loss": 2.2841,
+      "step": 59750
+    },
+    {
+      "epoch": 8.07,
+      "learning_rate": 2.6763622636803133e-05,
+      "loss": 2.2619,
+      "step": 59760
+    },
+    {
+      "epoch": 8.07,
+      "learning_rate": 2.6727346935856793e-05,
+      "loss": 2.2761,
+      "step": 59770
+    },
+    {
+      "epoch": 8.07,
+      "learning_rate": 2.6691093431210596e-05,
+      "loss": 2.2632,
+      "step": 59780
+    },
+    {
+      "epoch": 8.08,
+      "learning_rate": 2.665486212939228e-05,
+      "loss": 2.2778,
+      "step": 59790
+    },
+    {
+      "epoch": 8.08,
+      "learning_rate": 2.6618653036925586e-05,
+      "loss": 2.2617,
+      "step": 59800
+    },
+    {
+      "epoch": 8.08,
+      "learning_rate": 2.6582466160330284e-05,
+      "loss": 2.2832,
+      "step": 59810
+    },
+    {
+      "epoch": 8.08,
+      "learning_rate": 2.6546301506122098e-05,
+      "loss": 2.3091,
+      "step": 59820
+    },
+    {
+      "epoch": 8.08,
+      "learning_rate": 2.6510159080812832e-05,
+      "loss": 2.2926,
+      "step": 59830
+    },
+    {
+      "epoch": 8.08,
+      "learning_rate": 2.6474038890910197e-05,
+      "loss": 2.3009,
+      "step": 59840
+    },
+    {
+      "epoch": 8.08,
+      "learning_rate": 2.6437940942918045e-05,
+      "loss": 2.2843,
+      "step": 59850
+    },
+    {
+      "epoch": 8.08,
+      "learning_rate": 2.6401865243335974e-05,
+      "loss": 2.2865,
+      "step": 59860
+    },
+    {
+      "epoch": 8.09,
+      "learning_rate": 2.6365811798659818e-05,
+      "loss": 2.2808,
+      "step": 59870
+    },
+    {
+      "epoch": 8.09,
+      "learning_rate": 2.6329780615381307e-05,
+      "loss": 2.283,
+      "step": 59880
+    },
+    {
+      "epoch": 8.09,
+      "learning_rate": 2.6293771699988132e-05,
+      "loss": 2.3176,
+      "step": 59890
+    },
+    {
+      "epoch": 8.09,
+      "learning_rate": 2.625778505896402e-05,
+      "loss": 2.3082,
+      "step": 59900
+    },
+    {
+      "epoch": 8.09,
+      "learning_rate": 2.6221820698788644e-05,
+      "loss": 2.2947,
+      "step": 59910
+    },
+    {
+      "epoch": 8.09,
+      "learning_rate": 2.618587862593775e-05,
+      "loss": 2.28,
+      "step": 59920
+    },
+    {
+      "epoch": 8.09,
+      "learning_rate": 2.614995884688297e-05,
+      "loss": 2.2873,
+      "step": 59930
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 2.6114061368092053e-05,
+      "loss": 2.2876,
+      "step": 59940
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 2.6078186196028522e-05,
+      "loss": 2.2325,
+      "step": 59950
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 2.604233333715212e-05,
+      "loss": 2.2868,
+      "step": 59960
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 2.6006502797918404e-05,
+      "loss": 2.2874,
+      "step": 59970
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 2.5970694584778962e-05,
+      "loss": 2.2909,
+      "step": 59980
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 2.5934908704181467e-05,
+      "loss": 2.2785,
+      "step": 59990
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 2.589914516256932e-05,
+      "loss": 2.2752,
+      "step": 60000
+    },
+    {
+      "epoch": 8.1,
+      "eval_loss": 2.434035539627075,
+      "eval_runtime": 1269.0314,
+      "eval_samples_per_second": 59.995,
+      "eval_steps_per_second": 5.0,
+      "step": 60000
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 2.586340396638216e-05,
+      "loss": 2.2876,
+      "step": 60010
+    },
+    {
+      "epoch": 8.11,
+      "learning_rate": 2.582768512205545e-05,
+      "loss": 2.2852,
+      "step": 60020
+    },
+    {
+      "epoch": 8.11,
+      "learning_rate": 2.579198863602076e-05,
+      "loss": 2.2998,
+      "step": 60030
+    },
+    {
+      "epoch": 8.11,
+      "learning_rate": 2.5756314514705413e-05,
+      "loss": 2.2887,
+      "step": 60040
+    },
+    {
+      "epoch": 8.11,
+      "learning_rate": 2.572066276453295e-05,
+      "loss": 2.2831,
+      "step": 60050
+    },
+    {
+      "epoch": 8.11,
+      "learning_rate": 2.5685033391922744e-05,
+      "loss": 2.2774,
+      "step": 60060
+    },
+    {
+      "epoch": 8.11,
+      "learning_rate": 2.5649426403290123e-05,
+      "loss": 2.2884,
+      "step": 60070
+    },
+    {
+      "epoch": 8.11,
+      "learning_rate": 2.561384180504653e-05,
+      "loss": 2.2725,
+      "step": 60080
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 2.5578279603599156e-05,
+      "loss": 2.3079,
+      "step": 60090
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 2.5542739805351376e-05,
+      "loss": 2.2875,
+      "step": 60100
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 2.5507222416702374e-05,
+      "loss": 2.3166,
+      "step": 60110
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 2.5471727444047413e-05,
+      "loss": 2.2641,
+      "step": 60120
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 2.5436254893777704e-05,
+      "loss": 2.2877,
+      "step": 60130
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 2.540080477228023e-05,
+      "loss": 2.3026,
+      "step": 60140
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 2.5365377085938228e-05,
+      "loss": 2.2566,
+      "step": 60150
+    },
+    {
+      "epoch": 8.13,
+      "learning_rate": 2.5329971841130687e-05,
+      "loss": 2.2719,
+      "step": 60160
+    },
+    {
+      "epoch": 8.13,
+      "learning_rate": 2.529458904423271e-05,
+      "loss": 2.2718,
+      "step": 60170
+    },
+    {
+      "epoch": 8.13,
+      "learning_rate": 2.5259228701615154e-05,
+      "loss": 2.2822,
+      "step": 60180
+    },
+    {
+      "epoch": 8.13,
+      "learning_rate": 2.5223890819645055e-05,
+      "loss": 2.2728,
+      "step": 60190
+    },
+    {
+      "epoch": 8.13,
+      "learning_rate": 2.5188575404685223e-05,
+      "loss": 2.2512,
+      "step": 60200
+    },
+    {
+      "epoch": 8.13,
+      "learning_rate": 2.5153282463094575e-05,
+      "loss": 2.2953,
+      "step": 60210
+    },
+    {
+      "epoch": 8.13,
+      "learning_rate": 2.5118012001227912e-05,
+      "loss": 2.2905,
+      "step": 60220
+    },
+    {
+      "epoch": 8.13,
+      "learning_rate": 2.5082764025435865e-05,
+      "loss": 2.2818,
+      "step": 60230
+    },
+    {
+      "epoch": 8.14,
+      "learning_rate": 2.504753854206526e-05,
+      "loss": 2.28,
+      "step": 60240
+    },
+    {
+      "epoch": 8.14,
+      "learning_rate": 2.5012335557458658e-05,
+      "loss": 2.2842,
+      "step": 60250
+    },
+    {
+      "epoch": 8.14,
+      "learning_rate": 2.4977155077954752e-05,
+      "loss": 2.2634,
+      "step": 60260
+    },
+    {
+      "epoch": 8.14,
+      "learning_rate": 2.4941997109888028e-05,
+      "loss": 2.2988,
+      "step": 60270
+    },
+    {
+      "epoch": 8.14,
+      "learning_rate": 2.4906861659589005e-05,
+      "loss": 2.2756,
+      "step": 60280
+    },
+    {
+      "epoch": 8.14,
+      "learning_rate": 2.4871748733384073e-05,
+      "loss": 2.2797,
+      "step": 60290
+    },
+    {
+      "epoch": 8.14,
+      "learning_rate": 2.4836658337595705e-05,
+      "loss": 2.2994,
+      "step": 60300
+    },
+    {
+      "epoch": 8.15,
+      "learning_rate": 2.480159047854215e-05,
+      "loss": 2.3133,
+      "step": 60310
+    },
+    {
+      "epoch": 8.15,
+      "learning_rate": 2.4766545162537722e-05,
+      "loss": 2.314,
+      "step": 60320
+    },
+    {
+      "epoch": 8.15,
+      "learning_rate": 2.4731522395892624e-05,
+      "loss": 2.2846,
+      "step": 60330
+    },
+    {
+      "epoch": 8.15,
+      "learning_rate": 2.4696522184912944e-05,
+      "loss": 2.2929,
+      "step": 60340
+    },
+    {
+      "epoch": 8.15,
+      "learning_rate": 2.4661544535900857e-05,
+      "loss": 2.2917,
+      "step": 60350
+    },
+    {
+      "epoch": 8.15,
+      "learning_rate": 2.462658945515436e-05,
+      "loss": 2.293,
+      "step": 60360
+    },
+    {
+      "epoch": 8.15,
+      "learning_rate": 2.459165694896742e-05,
+      "loss": 2.2715,
+      "step": 60370
+    },
+    {
+      "epoch": 8.15,
+      "learning_rate": 2.4556747023629868e-05,
+      "loss": 2.3106,
+      "step": 60380
+    },
+    {
+      "epoch": 8.16,
+      "learning_rate": 2.4521859685427647e-05,
+      "loss": 2.2864,
+      "step": 60390
+    },
+    {
+      "epoch": 8.16,
+      "learning_rate": 2.4486994940642463e-05,
+      "loss": 2.2673,
+      "step": 60400
+    },
+    {
+      "epoch": 8.16,
+      "learning_rate": 2.4452152795552028e-05,
+      "loss": 2.275,
+      "step": 60410
+    },
+    {
+      "epoch": 8.16,
+      "learning_rate": 2.441733325642994e-05,
+      "loss": 2.2626,
+      "step": 60420
+    },
+    {
+      "epoch": 8.16,
+      "learning_rate": 2.4382536329545748e-05,
+      "loss": 2.2847,
+      "step": 60430
+    },
+    {
+      "epoch": 8.16,
+      "learning_rate": 2.4347762021165012e-05,
+      "loss": 2.276,
+      "step": 60440
+    },
+    {
+      "epoch": 8.16,
+      "learning_rate": 2.4313010337549087e-05,
+      "loss": 2.2585,
+      "step": 60450
+    },
+    {
+      "epoch": 8.17,
+      "learning_rate": 2.427828128495532e-05,
+      "loss": 2.3073,
+      "step": 60460
+    },
+    {
+      "epoch": 8.17,
+      "learning_rate": 2.4243574869636955e-05,
+      "loss": 2.303,
+      "step": 60470
+    },
+    {
+      "epoch": 8.17,
+      "learning_rate": 2.4208891097843232e-05,
+      "loss": 2.2554,
+      "step": 60480
+    },
+    {
+      "epoch": 8.17,
+      "learning_rate": 2.417422997581923e-05,
+      "loss": 2.2796,
+      "step": 60490
+    },
+    {
+      "epoch": 8.17,
+      "learning_rate": 2.4139591509806e-05,
+      "loss": 2.269,
+      "step": 60500
+    },
+    {
+      "epoch": 8.17,
+      "learning_rate": 2.4104975706040463e-05,
+      "loss": 2.2763,
+      "step": 60510
+    },
+    {
+      "epoch": 8.17,
+      "learning_rate": 2.4070382570755486e-05,
+      "loss": 2.2793,
+      "step": 60520
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 2.403581211017992e-05,
+      "loss": 2.2965,
+      "step": 60530
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 2.4001264330538434e-05,
+      "loss": 2.2731,
+      "step": 60540
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 2.3966739238051653e-05,
+      "loss": 2.2783,
+      "step": 60550
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 2.3932236838936088e-05,
+      "loss": 2.2589,
+      "step": 60560
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 2.3897757139404268e-05,
+      "loss": 2.2824,
+      "step": 60570
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 2.3863300145664492e-05,
+      "loss": 2.2762,
+      "step": 60580
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 2.3828865863921086e-05,
+      "loss": 2.2991,
+      "step": 60590
+    },
+    {
+      "epoch": 8.18,
+      "learning_rate": 2.3794454300374188e-05,
+      "loss": 2.2801,
+      "step": 60600
+    },
+    {
+      "epoch": 8.19,
+      "learning_rate": 2.3760065461219906e-05,
+      "loss": 2.2592,
+      "step": 60610
+    },
+    {
+      "epoch": 8.19,
+      "learning_rate": 2.372569935265029e-05,
+      "loss": 2.2746,
+      "step": 60620
+    },
+    {
+      "epoch": 8.19,
+      "learning_rate": 2.369135598085322e-05,
+      "loss": 2.2901,
+      "step": 60630
+    },
+    {
+      "epoch": 8.19,
+      "learning_rate": 2.365703535201252e-05,
+      "loss": 2.2802,
+      "step": 60640
+    },
+    {
+      "epoch": 8.19,
+      "learning_rate": 2.3622737472307914e-05,
+      "loss": 2.3105,
+      "step": 60650
+    },
+    {
+      "epoch": 8.19,
+      "learning_rate": 2.3588462347914987e-05,
+      "loss": 2.2866,
+      "step": 60660
+    },
+    {
+      "epoch": 8.19,
+      "learning_rate": 2.3554209985005345e-05,
+      "loss": 2.2836,
+      "step": 60670
+    },
+    {
+      "epoch": 8.2,
+      "learning_rate": 2.3519980389746402e-05,
+      "loss": 2.307,
+      "step": 60680
+    },
+    {
+      "epoch": 8.2,
+      "learning_rate": 2.348577356830146e-05,
+      "loss": 2.2732,
+      "step": 60690
+    },
+    {
+      "epoch": 8.2,
+      "learning_rate": 2.3451589526829723e-05,
+      "loss": 2.3188,
+      "step": 60700
+    },
+    {
+      "epoch": 8.2,
+      "learning_rate": 2.341742827148642e-05,
+      "loss": 2.2739,
+      "step": 60710
+    },
+    {
+      "epoch": 8.2,
+      "learning_rate": 2.3383289808422516e-05,
+      "loss": 2.2923,
+      "step": 60720
+    },
+    {
+      "epoch": 8.2,
+      "learning_rate": 2.3349174143784943e-05,
+      "loss": 2.3051,
+      "step": 60730
+    },
+    {
+      "epoch": 8.2,
+      "learning_rate": 2.3315081283716502e-05,
+      "loss": 2.3147,
+      "step": 60740
+    },
+    {
+      "epoch": 8.2,
+      "learning_rate": 2.32810112343559e-05,
+      "loss": 2.3078,
+      "step": 60750
+    },
+    {
+      "epoch": 8.21,
+      "learning_rate": 2.3246964001837798e-05,
+      "loss": 2.271,
+      "step": 60760
+    },
+    {
+      "epoch": 8.21,
+      "learning_rate": 2.3212939592292656e-05,
+      "loss": 2.2644,
+      "step": 60770
+    },
+    {
+      "epoch": 8.21,
+      "learning_rate": 2.317893801184686e-05,
+      "loss": 2.2431,
+      "step": 60780
+    },
+    {
+      "epoch": 8.21,
+      "learning_rate": 2.314495926662266e-05,
+      "loss": 2.2762,
+      "step": 60790
+    },
+    {
+      "epoch": 8.21,
+      "learning_rate": 2.311100336273828e-05,
+      "loss": 2.2673,
+      "step": 60800
+    },
+    {
+      "epoch": 8.21,
+      "learning_rate": 2.307707030630771e-05,
+      "loss": 2.2979,
+      "step": 60810
+    },
+    {
+      "epoch": 8.21,
+      "learning_rate": 2.3043160103440982e-05,
+      "loss": 2.2687,
+      "step": 60820
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 2.3009272760243848e-05,
+      "loss": 2.2788,
+      "step": 60830
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 2.297540828281797e-05,
+      "loss": 2.3006,
+      "step": 60840
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 2.2941566677261027e-05,
+      "loss": 2.261,
+      "step": 60850
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 2.2907747949666472e-05,
+      "loss": 2.2649,
+      "step": 60860
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 2.2873952106123623e-05,
+      "loss": 2.2747,
+      "step": 60870
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 2.2840179152717707e-05,
+      "loss": 2.2773,
+      "step": 60880
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 2.28064290955299e-05,
+      "loss": 2.2716,
+      "step": 60890
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 2.2772701940637117e-05,
+      "loss": 2.2998,
+      "step": 60900
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 2.2738997694112304e-05,
+      "loss": 2.2623,
+      "step": 60910
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 2.270531636202414e-05,
+      "loss": 2.2775,
+      "step": 60920
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 2.267165795043719e-05,
+      "loss": 2.2618,
+      "step": 60930
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 2.2638022465412054e-05,
+      "loss": 2.2746,
+      "step": 60940
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 2.260440991300499e-05,
+      "loss": 2.2979,
+      "step": 60950
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 2.257082029926834e-05,
+      "loss": 2.2893,
+      "step": 60960
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 2.2537253630250064e-05,
+      "loss": 2.2937,
+      "step": 60970
+    },
+    {
+      "epoch": 8.24,
+      "learning_rate": 2.250370991199423e-05,
+      "loss": 2.2874,
+      "step": 60980
+    },
+    {
+      "epoch": 8.24,
+      "learning_rate": 2.24701891505406e-05,
+      "loss": 2.3037,
+      "step": 60990
+    },
+    {
+      "epoch": 8.24,
+      "learning_rate": 2.243669135192499e-05,
+      "loss": 2.2609,
+      "step": 61000
+    },
+    {
+      "epoch": 8.24,
+      "eval_loss": 2.4321961402893066,
+      "eval_runtime": 1269.1578,
+      "eval_samples_per_second": 59.989,
+      "eval_steps_per_second": 4.999,
+      "step": 61000
+    },
+    {
+      "epoch": 8.24,
+      "learning_rate": 2.2403216522178868e-05,
+      "loss": 2.2852,
+      "step": 61010
+    },
+    {
+      "epoch": 8.24,
+      "learning_rate": 2.2369764667329648e-05,
+      "loss": 2.2954,
+      "step": 61020
+    },
+    {
+      "epoch": 8.24,
+      "learning_rate": 2.2336335793400712e-05,
+      "loss": 2.2837,
+      "step": 61030
+    },
+    {
+      "epoch": 8.24,
+      "learning_rate": 2.230292990641111e-05,
+      "loss": 2.2713,
+      "step": 61040
+    },
+    {
+      "epoch": 8.25,
+      "learning_rate": 2.2269547012376015e-05,
+      "loss": 2.2678,
+      "step": 61050
+    },
+    {
+      "epoch": 8.25,
+      "learning_rate": 2.223618711730612e-05,
+      "loss": 2.2934,
+      "step": 61060
+    },
+    {
+      "epoch": 8.25,
+      "learning_rate": 2.2202850227208284e-05,
+      "loss": 2.2755,
+      "step": 61070
+    },
+    {
+      "epoch": 8.25,
+      "learning_rate": 2.2169536348085025e-05,
+      "loss": 2.2793,
+      "step": 61080
+    },
+    {
+      "epoch": 8.25,
+      "learning_rate": 2.2136245485934877e-05,
+      "loss": 2.2855,
+      "step": 61090
+    },
+    {
+      "epoch": 8.25,
+      "learning_rate": 2.2102977646752128e-05,
+      "loss": 2.2748,
+      "step": 61100
+    },
+    {
+      "epoch": 8.25,
+      "learning_rate": 2.2069732836526856e-05,
+      "loss": 2.2693,
+      "step": 61110
+    },
+    {
+      "epoch": 8.25,
+      "learning_rate": 2.203651106124515e-05,
+      "loss": 2.2964,
+      "step": 61120
+    },
+    {
+      "epoch": 8.26,
+      "learning_rate": 2.2003312326888828e-05,
+      "loss": 2.2533,
+      "step": 61130
+    },
+    {
+      "epoch": 8.26,
+      "learning_rate": 2.1970136639435658e-05,
+      "loss": 2.2673,
+      "step": 61140
+    },
+    {
+      "epoch": 8.26,
+      "learning_rate": 2.1936984004859197e-05,
+      "loss": 2.2711,
+      "step": 61150
+    },
+    {
+      "epoch": 8.26,
+      "learning_rate": 2.1903854429128836e-05,
+      "loss": 2.27,
+      "step": 61160
+    },
+    {
+      "epoch": 8.26,
+      "learning_rate": 2.187074791820987e-05,
+      "loss": 2.2796,
+      "step": 61170
+    },
+    {
+      "epoch": 8.26,
+      "learning_rate": 2.1837664478063343e-05,
+      "loss": 2.3061,
+      "step": 61180
+    },
+    {
+      "epoch": 8.26,
+      "learning_rate": 2.1804604114646345e-05,
+      "loss": 2.26,
+      "step": 61190
+    },
+    {
+      "epoch": 8.27,
+      "learning_rate": 2.1771566833911513e-05,
+      "loss": 2.3028,
+      "step": 61200
+    },
+    {
+      "epoch": 8.27,
+      "learning_rate": 2.1738552641807622e-05,
+      "loss": 2.2921,
+      "step": 61210
+    },
+    {
+      "epoch": 8.27,
+      "learning_rate": 2.1705561544279083e-05,
+      "loss": 2.2792,
+      "step": 61220
+    },
+    {
+      "epoch": 8.27,
+      "learning_rate": 2.1672593547266294e-05,
+      "loss": 2.2622,
+      "step": 61230
+    },
+    {
+      "epoch": 8.27,
+      "learning_rate": 2.163964865670539e-05,
+      "loss": 2.2936,
+      "step": 61240
+    },
+    {
+      "epoch": 8.27,
+      "learning_rate": 2.1606726878528375e-05,
+      "loss": 2.2644,
+      "step": 61250
+    },
+    {
+      "epoch": 8.27,
+      "learning_rate": 2.1573828218663107e-05,
+      "loss": 2.2586,
+      "step": 61260
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 2.1540952683033235e-05,
+      "loss": 2.2662,
+      "step": 61270
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 2.1508100277558353e-05,
+      "loss": 2.2812,
+      "step": 61280
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 2.1475271008153798e-05,
+      "loss": 2.2783,
+      "step": 61290
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 2.1442464880730725e-05,
+      "loss": 2.2923,
+      "step": 61300
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 2.1409681901196175e-05,
+      "loss": 2.2988,
+      "step": 61310
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 2.1376922075453023e-05,
+      "loss": 2.2611,
+      "step": 61320
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 2.1344185409399955e-05,
+      "loss": 2.2842,
+      "step": 61330
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 2.131147190893147e-05,
+      "loss": 2.28,
+      "step": 61340
+    },
+    {
+      "epoch": 8.29,
+      "learning_rate": 2.1278781579937924e-05,
+      "loss": 2.2854,
+      "step": 61350
+    },
+    {
+      "epoch": 8.29,
+      "learning_rate": 2.1246114428305477e-05,
+      "loss": 2.2854,
+      "step": 61360
+    },
+    {
+      "epoch": 8.29,
+      "learning_rate": 2.1213470459916153e-05,
+      "loss": 2.289,
+      "step": 61370
+    },
+    {
+      "epoch": 8.29,
+      "learning_rate": 2.1180849680647788e-05,
+      "loss": 2.2661,
+      "step": 61380
+    },
+    {
+      "epoch": 8.29,
+      "learning_rate": 2.1148252096374014e-05,
+      "loss": 2.2584,
+      "step": 61390
+    },
+    {
+      "epoch": 8.29,
+      "learning_rate": 2.111567771296427e-05,
+      "loss": 2.294,
+      "step": 61400
+    },
+    {
+      "epoch": 8.29,
+      "learning_rate": 2.1083126536283927e-05,
+      "loss": 2.2849,
+      "step": 61410
+    },
+    {
+      "epoch": 8.3,
+      "learning_rate": 2.105059857219409e-05,
+      "loss": 2.2901,
+      "step": 61420
+    },
+    {
+      "epoch": 8.3,
+      "learning_rate": 2.1018093826551645e-05,
+      "loss": 2.2806,
+      "step": 61430
+    },
+    {
+      "epoch": 8.3,
+      "learning_rate": 2.0985612305209404e-05,
+      "loss": 2.2917,
+      "step": 61440
+    },
+    {
+      "epoch": 8.3,
+      "learning_rate": 2.0953154014015883e-05,
+      "loss": 2.2958,
+      "step": 61450
+    },
+    {
+      "epoch": 8.3,
+      "learning_rate": 2.092071895881554e-05,
+      "loss": 2.2824,
+      "step": 61460
+    },
+    {
+      "epoch": 8.3,
+      "learning_rate": 2.0888307145448575e-05,
+      "loss": 2.2924,
+      "step": 61470
+    },
+    {
+      "epoch": 8.3,
+      "learning_rate": 2.0855918579750997e-05,
+      "loss": 2.2495,
+      "step": 61480
+    },
+    {
+      "epoch": 8.3,
+      "learning_rate": 2.082355326755461e-05,
+      "loss": 2.2817,
+      "step": 61490
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 2.0791211214687153e-05,
+      "loss": 2.2901,
+      "step": 61500
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 2.075889242697202e-05,
+      "loss": 2.2789,
+      "step": 61510
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 2.0726596910228532e-05,
+      "loss": 2.2678,
+      "step": 61520
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 2.069432467027173e-05,
+      "loss": 2.2762,
+      "step": 61530
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 2.066207571291251e-05,
+      "loss": 2.2976,
+      "step": 61540
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 2.0629850043957626e-05,
+      "loss": 2.2842,
+      "step": 61550
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 2.059764766920954e-05,
+      "loss": 2.2797,
+      "step": 61560
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 2.0565468594466595e-05,
+      "loss": 2.2791,
+      "step": 61570
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 2.0533312825522842e-05,
+      "loss": 2.2667,
+      "step": 61580
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 2.0501180368168313e-05,
+      "loss": 2.2845,
+      "step": 61590
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 2.046907122818865e-05,
+      "loss": 2.278,
+      "step": 61600
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 2.0436985411365427e-05,
+      "loss": 2.2807,
+      "step": 61610
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 2.0404922923475935e-05,
+      "loss": 2.2776,
+      "step": 61620
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 2.0372883770293297e-05,
+      "loss": 2.299,
+      "step": 61630
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 2.0340867957586478e-05,
+      "loss": 2.2773,
+      "step": 61640
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 2.0308875491120165e-05,
+      "loss": 2.3009,
+      "step": 61650
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 2.02769063766549e-05,
+      "loss": 2.2633,
+      "step": 61660
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 2.0244960619946966e-05,
+      "loss": 2.2805,
+      "step": 61670
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 2.0213038226748506e-05,
+      "loss": 2.298,
+      "step": 61680
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 2.018113920280741e-05,
+      "loss": 2.2873,
+      "step": 61690
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 2.014926355386737e-05,
+      "loss": 2.2981,
+      "step": 61700
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 2.0117411285667867e-05,
+      "loss": 2.2564,
+      "step": 61710
+    },
+    {
+      "epoch": 8.34,
+      "learning_rate": 2.0085582403944138e-05,
+      "loss": 2.2906,
+      "step": 61720
+    },
+    {
+      "epoch": 8.34,
+      "learning_rate": 2.0053776914427315e-05,
+      "loss": 2.265,
+      "step": 61730
+    },
+    {
+      "epoch": 8.34,
+      "learning_rate": 2.0021994822844217e-05,
+      "loss": 2.2789,
+      "step": 61740
+    },
+    {
+      "epoch": 8.34,
+      "learning_rate": 1.9990236134917497e-05,
+      "loss": 2.2742,
+      "step": 61750
+    },
+    {
+      "epoch": 8.34,
+      "learning_rate": 1.9958500856365535e-05,
+      "loss": 2.2842,
+      "step": 61760
+    },
+    {
+      "epoch": 8.34,
+      "learning_rate": 1.9926788992902586e-05,
+      "loss": 2.2905,
+      "step": 61770
+    },
+    {
+      "epoch": 8.34,
+      "learning_rate": 1.9895100550238642e-05,
+      "loss": 2.2777,
+      "step": 61780
+    },
+    {
+      "epoch": 8.35,
+      "learning_rate": 1.9863435534079463e-05,
+      "loss": 2.287,
+      "step": 61790
+    },
+    {
+      "epoch": 8.35,
+      "learning_rate": 1.9831793950126628e-05,
+      "loss": 2.2573,
+      "step": 61800
+    },
+    {
+      "epoch": 8.35,
+      "learning_rate": 1.980017580407741e-05,
+      "loss": 2.2612,
+      "step": 61810
+    },
+    {
+      "epoch": 8.35,
+      "learning_rate": 1.9768581101625008e-05,
+      "loss": 2.3016,
+      "step": 61820
+    },
+    {
+      "epoch": 8.35,
+      "learning_rate": 1.9737009848458303e-05,
+      "loss": 2.3136,
+      "step": 61830
+    },
+    {
+      "epoch": 8.35,
+      "learning_rate": 1.9705462050261945e-05,
+      "loss": 2.2676,
+      "step": 61840
+    },
+    {
+      "epoch": 8.35,
+      "learning_rate": 1.9673937712716374e-05,
+      "loss": 2.2641,
+      "step": 61850
+    },
+    {
+      "epoch": 8.35,
+      "learning_rate": 1.9642436841497857e-05,
+      "loss": 2.2677,
+      "step": 61860
+    },
+    {
+      "epoch": 8.36,
+      "learning_rate": 1.9610959442278396e-05,
+      "loss": 2.2945,
+      "step": 61870
+    },
+    {
+      "epoch": 8.36,
+      "learning_rate": 1.9579505520725747e-05,
+      "loss": 2.2853,
+      "step": 61880
+    },
+    {
+      "epoch": 8.36,
+      "learning_rate": 1.954807508250346e-05,
+      "loss": 2.2961,
+      "step": 61890
+    },
+    {
+      "epoch": 8.36,
+      "learning_rate": 1.951666813327083e-05,
+      "loss": 2.2755,
+      "step": 61900
+    },
+    {
+      "epoch": 8.36,
+      "learning_rate": 1.9485284678683012e-05,
+      "loss": 2.2876,
+      "step": 61910
+    },
+    {
+      "epoch": 8.36,
+      "learning_rate": 1.9453924724390785e-05,
+      "loss": 2.2632,
+      "step": 61920
+    },
+    {
+      "epoch": 8.36,
+      "learning_rate": 1.9422588276040902e-05,
+      "loss": 2.2581,
+      "step": 61930
+    },
+    {
+      "epoch": 8.37,
+      "learning_rate": 1.9391275339275615e-05,
+      "loss": 2.2938,
+      "step": 61940
+    },
+    {
+      "epoch": 8.37,
+      "learning_rate": 1.935998591973318e-05,
+      "loss": 2.2856,
+      "step": 61950
+    },
+    {
+      "epoch": 8.37,
+      "learning_rate": 1.9328720023047507e-05,
+      "loss": 2.2981,
+      "step": 61960
+    },
+    {
+      "epoch": 8.37,
+      "learning_rate": 1.9297477654848293e-05,
+      "loss": 2.2867,
+      "step": 61970
+    },
+    {
+      "epoch": 8.37,
+      "learning_rate": 1.926625882076097e-05,
+      "loss": 2.2768,
+      "step": 61980
+    },
+    {
+      "epoch": 8.37,
+      "learning_rate": 1.9235063526406745e-05,
+      "loss": 2.2806,
+      "step": 61990
+    },
+    {
+      "epoch": 8.37,
+      "learning_rate": 1.920389177740264e-05,
+      "loss": 2.2746,
+      "step": 62000
+    },
+    {
+      "epoch": 8.37,
+      "eval_loss": 2.4299051761627197,
+      "eval_runtime": 1269.0207,
+      "eval_samples_per_second": 59.995,
+      "eval_steps_per_second": 5.0,
+      "step": 62000
+    },
+    {
+      "epoch": 8.38,
+      "learning_rate": 1.9172743579361365e-05,
+      "loss": 2.2811,
+      "step": 62010
+    },
+    {
+      "epoch": 8.38,
+      "learning_rate": 1.91416189378915e-05,
+      "loss": 2.2749,
+      "step": 62020
+    },
+    {
+      "epoch": 8.38,
+      "learning_rate": 1.911051785859718e-05,
+      "loss": 2.2429,
+      "step": 62030
+    },
+    {
+      "epoch": 8.38,
+      "learning_rate": 1.907944034707852e-05,
+      "loss": 2.2934,
+      "step": 62040
+    },
+    {
+      "epoch": 8.38,
+      "learning_rate": 1.9048386408931238e-05,
+      "loss": 2.2662,
+      "step": 62050
+    },
+    {
+      "epoch": 8.38,
+      "learning_rate": 1.9017356049746867e-05,
+      "loss": 2.2893,
+      "step": 62060
+    },
+    {
+      "epoch": 8.38,
+      "learning_rate": 1.8986349275112746e-05,
+      "loss": 2.3018,
+      "step": 62070
+    },
+    {
+      "epoch": 8.38,
+      "learning_rate": 1.895536609061182e-05,
+      "loss": 2.2645,
+      "step": 62080
+    },
+    {
+      "epoch": 8.39,
+      "learning_rate": 1.892440650182296e-05,
+      "loss": 2.2715,
+      "step": 62090
+    },
+    {
+      "epoch": 8.39,
+      "learning_rate": 1.8893470514320624e-05,
+      "loss": 2.2491,
+      "step": 62100
+    },
+    {
+      "epoch": 8.39,
+      "learning_rate": 1.8862558133675175e-05,
+      "loss": 2.2795,
+      "step": 62110
+    },
+    {
+      "epoch": 8.39,
+      "learning_rate": 1.8831669365452646e-05,
+      "loss": 2.2328,
+      "step": 62120
+    },
+    {
+      "epoch": 8.39,
+      "learning_rate": 1.8800804215214798e-05,
+      "loss": 2.2655,
+      "step": 62130
+    },
+    {
+      "epoch": 8.39,
+      "learning_rate": 1.8769962688519175e-05,
+      "loss": 2.2631,
+      "step": 62140
+    },
+    {
+      "epoch": 8.39,
+      "learning_rate": 1.873914479091903e-05,
+      "loss": 2.3048,
+      "step": 62150
+    },
+    {
+      "epoch": 8.4,
+      "learning_rate": 1.8708350527963478e-05,
+      "loss": 2.2834,
+      "step": 62160
+    },
+    {
+      "epoch": 8.4,
+      "learning_rate": 1.8677579905197183e-05,
+      "loss": 2.2881,
+      "step": 62170
+    },
+    {
+      "epoch": 8.4,
+      "learning_rate": 1.8646832928160726e-05,
+      "loss": 2.2884,
+      "step": 62180
+    },
+    {
+      "epoch": 8.4,
+      "learning_rate": 1.8616109602390317e-05,
+      "loss": 2.2844,
+      "step": 62190
+    },
+    {
+      "epoch": 8.4,
+      "learning_rate": 1.858540993341801e-05,
+      "loss": 2.2877,
+      "step": 62200
+    },
+    {
+      "epoch": 8.4,
+      "learning_rate": 1.8554733926771552e-05,
+      "loss": 2.2808,
+      "step": 62210
+    },
+    {
+      "epoch": 8.4,
+      "learning_rate": 1.8524081587974326e-05,
+      "loss": 2.2845,
+      "step": 62220
+    },
+    {
+      "epoch": 8.4,
+      "learning_rate": 1.8493452922545647e-05,
+      "loss": 2.2897,
+      "step": 62230
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 1.846284793600039e-05,
+      "loss": 2.2766,
+      "step": 62240
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 1.843226663384933e-05,
+      "loss": 2.289,
+      "step": 62250
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 1.8401709021598844e-05,
+      "loss": 2.2796,
+      "step": 62260
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 1.8371175104751114e-05,
+      "loss": 2.2715,
+      "step": 62270
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 1.8340664888803974e-05,
+      "loss": 2.2957,
+      "step": 62280
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 1.8310178379251134e-05,
+      "loss": 2.2833,
+      "step": 62290
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 1.827971558158195e-05,
+      "loss": 2.2884,
+      "step": 62300
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 1.824927650128142e-05,
+      "loss": 2.2699,
+      "step": 62310
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 1.821886114383044e-05,
+      "loss": 2.3088,
+      "step": 62320
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 1.8188469514705495e-05,
+      "loss": 2.281,
+      "step": 62330
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 1.815810161937893e-05,
+      "loss": 2.2912,
+      "step": 62340
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 1.8127757463318716e-05,
+      "loss": 2.2623,
+      "step": 62350
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 1.8097437051988584e-05,
+      "loss": 2.2628,
+      "step": 62360
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 1.8067140390847935e-05,
+      "loss": 2.2404,
+      "step": 62370
+    },
+    {
+      "epoch": 8.43,
+      "learning_rate": 1.803686748535202e-05,
+      "loss": 2.3156,
+      "step": 62380
+    },
+    {
+      "epoch": 8.43,
+      "learning_rate": 1.8006618340951694e-05,
+      "loss": 2.2653,
+      "step": 62390
+    },
+    {
+      "epoch": 8.43,
+      "learning_rate": 1.797639296309359e-05,
+      "loss": 2.2976,
+      "step": 62400
+    },
+    {
+      "epoch": 8.43,
+      "learning_rate": 1.794619135722003e-05,
+      "loss": 2.2754,
+      "step": 62410
+    },
+    {
+      "epoch": 8.43,
+      "learning_rate": 1.7916013528769053e-05,
+      "loss": 2.2638,
+      "step": 62420
+    },
+    {
+      "epoch": 8.43,
+      "learning_rate": 1.7885859483174507e-05,
+      "loss": 2.2594,
+      "step": 62430
+    },
+    {
+      "epoch": 8.43,
+      "learning_rate": 1.7855729225865823e-05,
+      "loss": 2.2798,
+      "step": 62440
+    },
+    {
+      "epoch": 8.43,
+      "learning_rate": 1.7825622762268254e-05,
+      "loss": 2.2801,
+      "step": 62450
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 1.7795540097802668e-05,
+      "loss": 2.2868,
+      "step": 62460
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 1.7765481237885764e-05,
+      "loss": 2.2868,
+      "step": 62470
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 1.7735446187929898e-05,
+      "loss": 2.2753,
+      "step": 62480
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 1.7705434953343122e-05,
+      "loss": 2.259,
+      "step": 62490
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 1.76754475395292e-05,
+      "loss": 2.267,
+      "step": 62500
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 1.764548395188762e-05,
+      "loss": 2.2693,
+      "step": 62510
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 1.7615544195813636e-05,
+      "loss": 2.2592,
+      "step": 62520
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 1.758562827669811e-05,
+      "loss": 2.2892,
+      "step": 62530
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 1.7555736199927684e-05,
+      "loss": 2.2727,
+      "step": 62540
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 1.752586797088465e-05,
+      "loss": 2.2869,
+      "step": 62550
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 1.749602359494709e-05,
+      "loss": 2.2947,
+      "step": 62560
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 1.7466203077488706e-05,
+      "loss": 2.2713,
+      "step": 62570
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 1.743640642387895e-05,
+      "loss": 2.2738,
+      "step": 62580
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 1.7406633639482954e-05,
+      "loss": 2.3112,
+      "step": 62590
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 1.7376884729661556e-05,
+      "loss": 2.2771,
+      "step": 62600
+    },
+    {
+      "epoch": 8.46,
+      "learning_rate": 1.734715969977134e-05,
+      "loss": 2.2967,
+      "step": 62610
+    },
+    {
+      "epoch": 8.46,
+      "learning_rate": 1.7317458555164526e-05,
+      "loss": 2.2596,
+      "step": 62620
+    },
+    {
+      "epoch": 8.46,
+      "learning_rate": 1.728778130118907e-05,
+      "loss": 2.3161,
+      "step": 62630
+    },
+    {
+      "epoch": 8.46,
+      "learning_rate": 1.7258127943188575e-05,
+      "loss": 2.2729,
+      "step": 62640
+    },
+    {
+      "epoch": 8.46,
+      "learning_rate": 1.722849848650246e-05,
+      "loss": 2.2861,
+      "step": 62650
+    },
+    {
+      "epoch": 8.46,
+      "learning_rate": 1.7198892936465725e-05,
+      "loss": 2.2826,
+      "step": 62660
+    },
+    {
+      "epoch": 8.46,
+      "learning_rate": 1.7169311298409082e-05,
+      "loss": 2.2829,
+      "step": 62670
+    },
+    {
+      "epoch": 8.47,
+      "learning_rate": 1.7139753577658994e-05,
+      "loss": 2.2992,
+      "step": 62680
+    },
+    {
+      "epoch": 8.47,
+      "learning_rate": 1.7110219779537533e-05,
+      "loss": 2.2886,
+      "step": 62690
+    },
+    {
+      "epoch": 8.47,
+      "learning_rate": 1.7080709909362577e-05,
+      "loss": 2.2754,
+      "step": 62700
+    },
+    {
+      "epoch": 8.47,
+      "learning_rate": 1.7051223972447596e-05,
+      "loss": 2.2973,
+      "step": 62710
+    },
+    {
+      "epoch": 8.47,
+      "learning_rate": 1.7021761974101794e-05,
+      "loss": 2.2816,
+      "step": 62720
+    },
+    {
+      "epoch": 8.47,
+      "learning_rate": 1.6992323919630057e-05,
+      "loss": 2.2888,
+      "step": 62730
+    },
+    {
+      "epoch": 8.47,
+      "learning_rate": 1.696290981433293e-05,
+      "loss": 2.2714,
+      "step": 62740
+    },
+    {
+      "epoch": 8.47,
+      "learning_rate": 1.6933519663506727e-05,
+      "loss": 2.3208,
+      "step": 62750
+    },
+    {
+      "epoch": 8.48,
+      "learning_rate": 1.6904153472443365e-05,
+      "loss": 2.2687,
+      "step": 62760
+    },
+    {
+      "epoch": 8.48,
+      "learning_rate": 1.6874811246430497e-05,
+      "loss": 2.2654,
+      "step": 62770
+    },
+    {
+      "epoch": 8.48,
+      "learning_rate": 1.6845492990751397e-05,
+      "loss": 2.2864,
+      "step": 62780
+    },
+    {
+      "epoch": 8.48,
+      "learning_rate": 1.681619871068514e-05,
+      "loss": 2.2779,
+      "step": 62790
+    },
+    {
+      "epoch": 8.48,
+      "learning_rate": 1.678692841150638e-05,
+      "loss": 2.2802,
+      "step": 62800
+    },
+    {
+      "epoch": 8.48,
+      "learning_rate": 1.6757682098485474e-05,
+      "loss": 2.3053,
+      "step": 62810
+    },
+    {
+      "epoch": 8.48,
+      "learning_rate": 1.672845977688847e-05,
+      "loss": 2.2806,
+      "step": 62820
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 1.6699261451977104e-05,
+      "loss": 2.2745,
+      "step": 62830
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 1.6670087129008802e-05,
+      "loss": 2.3039,
+      "step": 62840
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 1.6640936813236648e-05,
+      "loss": 2.2992,
+      "step": 62850
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 1.6611810509909396e-05,
+      "loss": 2.2998,
+      "step": 62860
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 1.6582708224271446e-05,
+      "loss": 2.2618,
+      "step": 62870
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 1.6553629961562992e-05,
+      "loss": 2.2724,
+      "step": 62880
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 1.652457572701978e-05,
+      "loss": 2.2947,
+      "step": 62890
+    },
+    {
+      "epoch": 8.5,
+      "learning_rate": 1.6495545525873322e-05,
+      "loss": 2.2803,
+      "step": 62900
+    },
+    {
+      "epoch": 8.5,
+      "learning_rate": 1.6466539363350706e-05,
+      "loss": 2.3,
+      "step": 62910
+    },
+    {
+      "epoch": 8.5,
+      "learning_rate": 1.6437557244674705e-05,
+      "loss": 2.2713,
+      "step": 62920
+    },
+    {
+      "epoch": 8.5,
+      "learning_rate": 1.640859917506389e-05,
+      "loss": 2.2815,
+      "step": 62930
+    },
+    {
+      "epoch": 8.5,
+      "learning_rate": 1.637966515973233e-05,
+      "loss": 2.2895,
+      "step": 62940
+    },
+    {
+      "epoch": 8.5,
+      "learning_rate": 1.6350755203889954e-05,
+      "loss": 2.2591,
+      "step": 62950
+    },
+    {
+      "epoch": 8.5,
+      "learning_rate": 1.6321869312742097e-05,
+      "loss": 2.2809,
+      "step": 62960
+    },
+    {
+      "epoch": 8.5,
+      "learning_rate": 1.6293007491490018e-05,
+      "loss": 2.302,
+      "step": 62970
+    },
+    {
+      "epoch": 8.51,
+      "learning_rate": 1.6264169745330475e-05,
+      "loss": 2.3023,
+      "step": 62980
+    },
+    {
+      "epoch": 8.51,
+      "learning_rate": 1.6235356079456048e-05,
+      "loss": 2.2526,
+      "step": 62990
+    },
+    {
+      "epoch": 8.51,
+      "learning_rate": 1.6206566499054803e-05,
+      "loss": 2.2666,
+      "step": 63000
+    },
+    {
+      "epoch": 8.51,
+      "eval_loss": 2.427412986755371,
+      "eval_runtime": 1269.0254,
+      "eval_samples_per_second": 59.995,
+      "eval_steps_per_second": 5.0,
+      "step": 63000
+    },
+    {
+      "epoch": 8.51,
+      "learning_rate": 1.6177801009310527e-05,
+      "loss": 2.2902,
+      "step": 63010
+    },
+    {
+      "epoch": 8.51,
+      "learning_rate": 1.614905961540276e-05,
+      "loss": 2.2971,
+      "step": 63020
+    },
+    {
+      "epoch": 8.51,
+      "learning_rate": 1.612034232250658e-05,
+      "loss": 2.277,
+      "step": 63030
+    },
+    {
+      "epoch": 8.51,
+      "learning_rate": 1.6091649135792884e-05,
+      "loss": 2.2956,
+      "step": 63040
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 1.6062980060427988e-05,
+      "loss": 2.2661,
+      "step": 63050
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 1.6034335101574113e-05,
+      "loss": 2.2718,
+      "step": 63060
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 1.600571426438895e-05,
+      "loss": 2.2972,
+      "step": 63070
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 1.5977117554026e-05,
+      "loss": 2.2645,
+      "step": 63080
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 1.5948544975634346e-05,
+      "loss": 2.277,
+      "step": 63090
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 1.5919996534358635e-05,
+      "loss": 2.3066,
+      "step": 63100
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 1.589147223533935e-05,
+      "loss": 2.2673,
+      "step": 63110
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 1.5862972083712473e-05,
+      "loss": 2.2743,
+      "step": 63120
+    },
+    {
+      "epoch": 8.53,
+      "learning_rate": 1.5834496084609787e-05,
+      "loss": 2.2895,
+      "step": 63130
+    },
+    {
+      "epoch": 8.53,
+      "learning_rate": 1.5806044243158532e-05,
+      "loss": 2.2617,
+      "step": 63140
+    },
+    {
+      "epoch": 8.53,
+      "learning_rate": 1.5777616564481794e-05,
+      "loss": 2.2837,
+      "step": 63150
+    },
+    {
+      "epoch": 8.53,
+      "learning_rate": 1.5749213053698156e-05,
+      "loss": 2.267,
+      "step": 63160
+    },
+    {
+      "epoch": 8.53,
+      "learning_rate": 1.5720833715921965e-05,
+      "loss": 2.2818,
+      "step": 63170
+    },
+    {
+      "epoch": 8.53,
+      "learning_rate": 1.569247855626318e-05,
+      "loss": 2.2688,
+      "step": 63180
+    },
+    {
+      "epoch": 8.53,
+      "learning_rate": 1.566414757982729e-05,
+      "loss": 2.2786,
+      "step": 63190
+    },
+    {
+      "epoch": 8.54,
+      "learning_rate": 1.563584079171564e-05,
+      "loss": 2.2873,
+      "step": 63200
+    },
+    {
+      "epoch": 8.54,
+      "learning_rate": 1.5607558197025028e-05,
+      "loss": 2.2677,
+      "step": 63210
+    },
+    {
+      "epoch": 8.54,
+      "learning_rate": 1.5579299800848038e-05,
+      "loss": 2.269,
+      "step": 63220
+    },
+    {
+      "epoch": 8.54,
+      "learning_rate": 1.5551065608272823e-05,
+      "loss": 2.2796,
+      "step": 63230
+    },
+    {
+      "epoch": 8.54,
+      "learning_rate": 1.5522855624383195e-05,
+      "loss": 2.2842,
+      "step": 63240
+    },
+    {
+      "epoch": 8.54,
+      "learning_rate": 1.549466985425858e-05,
+      "loss": 2.2856,
+      "step": 63250
+    },
+    {
+      "epoch": 8.54,
+      "learning_rate": 1.5466508302974063e-05,
+      "loss": 2.2495,
+      "step": 63260
+    },
+    {
+      "epoch": 8.55,
+      "learning_rate": 1.543837097560044e-05,
+      "loss": 2.2996,
+      "step": 63270
+    },
+    {
+      "epoch": 8.55,
+      "learning_rate": 1.5410257877203976e-05,
+      "loss": 2.2769,
+      "step": 63280
+    },
+    {
+      "epoch": 8.55,
+      "learning_rate": 1.538216901284675e-05,
+      "loss": 2.2873,
+      "step": 63290
+    },
+    {
+      "epoch": 8.55,
+      "learning_rate": 1.5354104387586373e-05,
+      "loss": 2.3165,
+      "step": 63300
+    },
+    {
+      "epoch": 8.55,
+      "learning_rate": 1.5326064006476153e-05,
+      "loss": 2.2599,
+      "step": 63310
+    },
+    {
+      "epoch": 8.55,
+      "learning_rate": 1.5298047874564977e-05,
+      "loss": 2.3048,
+      "step": 63320
+    },
+    {
+      "epoch": 8.55,
+      "learning_rate": 1.5270055996897384e-05,
+      "loss": 2.2763,
+      "step": 63330
+    },
+    {
+      "epoch": 8.55,
+      "learning_rate": 1.5242088378513572e-05,
+      "loss": 2.2756,
+      "step": 63340
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 1.5214145024449315e-05,
+      "loss": 2.2992,
+      "step": 63350
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 1.5186225939736118e-05,
+      "loss": 2.2985,
+      "step": 63360
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 1.5158331129400996e-05,
+      "loss": 2.3101,
+      "step": 63370
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 1.5130460598466681e-05,
+      "loss": 2.2582,
+      "step": 63380
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 1.5102614351951443e-05,
+      "loss": 2.2992,
+      "step": 63390
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 1.5074792394869335e-05,
+      "loss": 2.2854,
+      "step": 63400
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 1.5046994732229866e-05,
+      "loss": 2.2806,
+      "step": 63410
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 1.501922136903827e-05,
+      "loss": 2.2873,
+      "step": 63420
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 1.4991472310295377e-05,
+      "loss": 2.2736,
+      "step": 63430
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 1.4963747560997608e-05,
+      "loss": 2.3065,
+      "step": 63440
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 1.4936047126137101e-05,
+      "loss": 2.27,
+      "step": 63450
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 1.4908371010701536e-05,
+      "loss": 2.2951,
+      "step": 63460
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 1.4880719219674243e-05,
+      "loss": 2.2653,
+      "step": 63470
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 1.4853091758034102e-05,
+      "loss": 2.2834,
+      "step": 63480
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 1.4825488630755772e-05,
+      "loss": 2.2953,
+      "step": 63490
+    },
+    {
+      "epoch": 8.58,
+      "learning_rate": 1.4797909842809374e-05,
+      "loss": 2.3167,
+      "step": 63500
+    },
+    {
+      "epoch": 8.58,
+      "learning_rate": 1.4770355399160721e-05,
+      "loss": 2.2873,
+      "step": 63510
+    },
+    {
+      "epoch": 8.58,
+      "learning_rate": 1.4742825304771227e-05,
+      "loss": 2.2828,
+      "step": 63520
+    },
+    {
+      "epoch": 8.58,
+      "learning_rate": 1.4715319564597893e-05,
+      "loss": 2.286,
+      "step": 63530
+    },
+    {
+      "epoch": 8.58,
+      "learning_rate": 1.468783818359341e-05,
+      "loss": 2.268,
+      "step": 63540
+    },
+    {
+      "epoch": 8.58,
+      "learning_rate": 1.466038116670603e-05,
+      "loss": 2.2843,
+      "step": 63550
+    },
+    {
+      "epoch": 8.58,
+      "learning_rate": 1.463294851887959e-05,
+      "loss": 2.2516,
+      "step": 63560
+    },
+    {
+      "epoch": 8.59,
+      "learning_rate": 1.4605540245053581e-05,
+      "loss": 2.285,
+      "step": 63570
+    },
+    {
+      "epoch": 8.59,
+      "learning_rate": 1.4578156350163144e-05,
+      "loss": 2.2638,
+      "step": 63580
+    },
+    {
+      "epoch": 8.59,
+      "learning_rate": 1.4550796839138951e-05,
+      "loss": 2.2527,
+      "step": 63590
+    },
+    {
+      "epoch": 8.59,
+      "learning_rate": 1.4523461716907308e-05,
+      "loss": 2.2711,
+      "step": 63600
+    },
+    {
+      "epoch": 8.59,
+      "learning_rate": 1.4496150988390164e-05,
+      "loss": 2.2791,
+      "step": 63610
+    },
+    {
+      "epoch": 8.59,
+      "learning_rate": 1.4468864658505004e-05,
+      "loss": 2.3034,
+      "step": 63620
+    },
+    {
+      "epoch": 8.59,
+      "learning_rate": 1.4441602732165008e-05,
+      "loss": 2.2935,
+      "step": 63630
+    },
+    {
+      "epoch": 8.6,
+      "learning_rate": 1.4414365214278922e-05,
+      "loss": 2.3195,
+      "step": 63640
+    },
+    {
+      "epoch": 8.6,
+      "learning_rate": 1.4387152109751088e-05,
+      "loss": 2.2611,
+      "step": 63650
+    },
+    {
+      "epoch": 8.6,
+      "learning_rate": 1.4359963423481435e-05,
+      "loss": 2.2949,
+      "step": 63660
+    },
+    {
+      "epoch": 8.6,
+      "learning_rate": 1.433279916036555e-05,
+      "loss": 2.2934,
+      "step": 63670
+    },
+    {
+      "epoch": 8.6,
+      "learning_rate": 1.4305659325294583e-05,
+      "loss": 2.2653,
+      "step": 63680
+    },
+    {
+      "epoch": 8.6,
+      "learning_rate": 1.4278543923155295e-05,
+      "loss": 2.2755,
+      "step": 63690
+    },
+    {
+      "epoch": 8.6,
+      "learning_rate": 1.4251452958830045e-05,
+      "loss": 2.2872,
+      "step": 63700
+    },
+    {
+      "epoch": 8.6,
+      "learning_rate": 1.4224386437196766e-05,
+      "loss": 2.2571,
+      "step": 63710
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 1.4197344363129076e-05,
+      "loss": 2.2757,
+      "step": 63720
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 1.4170326741496096e-05,
+      "loss": 2.2808,
+      "step": 63730
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 1.4143333577162586e-05,
+      "loss": 2.2571,
+      "step": 63740
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 1.4116364874988855e-05,
+      "loss": 2.2572,
+      "step": 63750
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 1.408942063983094e-05,
+      "loss": 2.2562,
+      "step": 63760
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 1.4062500876540339e-05,
+      "loss": 2.2908,
+      "step": 63770
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 1.403560558996416e-05,
+      "loss": 2.2855,
+      "step": 63780
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 1.4008734784945164e-05,
+      "loss": 2.2775,
+      "step": 63790
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 1.3981888466321629e-05,
+      "loss": 2.2717,
+      "step": 63800
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 1.395506663892752e-05,
+      "loss": 2.2655,
+      "step": 63810
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 1.3928269307592327e-05,
+      "loss": 2.317,
+      "step": 63820
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 1.3901496477141106e-05,
+      "loss": 2.2426,
+      "step": 63830
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 1.3874748152394533e-05,
+      "loss": 2.2666,
+      "step": 63840
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 1.3848024338168923e-05,
+      "loss": 2.2785,
+      "step": 63850
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 1.3821325039276094e-05,
+      "loss": 2.2828,
+      "step": 63860
+    },
+    {
+      "epoch": 8.63,
+      "learning_rate": 1.37946502605235e-05,
+      "loss": 2.2498,
+      "step": 63870
+    },
+    {
+      "epoch": 8.63,
+      "learning_rate": 1.3768000006714147e-05,
+      "loss": 2.2693,
+      "step": 63880
+    },
+    {
+      "epoch": 8.63,
+      "learning_rate": 1.3741374282646616e-05,
+      "loss": 2.3123,
+      "step": 63890
+    },
+    {
+      "epoch": 8.63,
+      "learning_rate": 1.3714773093115156e-05,
+      "loss": 2.2944,
+      "step": 63900
+    },
+    {
+      "epoch": 8.63,
+      "learning_rate": 1.36881964429095e-05,
+      "loss": 2.2677,
+      "step": 63910
+    },
+    {
+      "epoch": 8.63,
+      "learning_rate": 1.3661644336815069e-05,
+      "loss": 2.2872,
+      "step": 63920
+    },
+    {
+      "epoch": 8.63,
+      "learning_rate": 1.3635116779612675e-05,
+      "loss": 2.2669,
+      "step": 63930
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 1.3608613776078946e-05,
+      "loss": 2.3057,
+      "step": 63940
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 1.3582135330985933e-05,
+      "loss": 2.2931,
+      "step": 63950
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 1.3555681449101301e-05,
+      "loss": 2.2718,
+      "step": 63960
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 1.3529252135188296e-05,
+      "loss": 2.262,
+      "step": 63970
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 1.350284739400574e-05,
+      "loss": 2.2919,
+      "step": 63980
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 1.3476467230308047e-05,
+      "loss": 2.269,
+      "step": 63990
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 1.3450111648845185e-05,
+      "loss": 2.2876,
+      "step": 64000
+    },
+    {
+      "epoch": 8.64,
+      "eval_loss": 2.4255166053771973,
+      "eval_runtime": 1269.0542,
+      "eval_samples_per_second": 59.993,
+      "eval_steps_per_second": 5.0,
+      "step": 64000
+    },
+    {
+      "epoch": 8.65,
+      "learning_rate": 1.342378065436276e-05,
+      "loss": 2.292,
+      "step": 64010
+    },
+    {
+      "epoch": 8.65,
+      "learning_rate": 1.3397474251601774e-05,
+      "loss": 2.2803,
+      "step": 64020
+    },
+    {
+      "epoch": 8.65,
+      "learning_rate": 1.3371192445299028e-05,
+      "loss": 2.2966,
+      "step": 64030
+    },
+    {
+      "epoch": 8.65,
+      "learning_rate": 1.3344935240186733e-05,
+      "loss": 2.2961,
+      "step": 64040
+    },
+    {
+      "epoch": 8.65,
+      "learning_rate": 1.3318702640992724e-05,
+      "loss": 2.2714,
+      "step": 64050
+    },
+    {
+      "epoch": 8.65,
+      "learning_rate": 1.3292494652440472e-05,
+      "loss": 2.2835,
+      "step": 64060
+    },
+    {
+      "epoch": 8.65,
+      "learning_rate": 1.3266311279248853e-05,
+      "loss": 2.286,
+      "step": 64070
+    },
+    {
+      "epoch": 8.65,
+      "learning_rate": 1.3240152526132463e-05,
+      "loss": 2.256,
+      "step": 64080
+    },
+    {
+      "epoch": 8.66,
+      "learning_rate": 1.3214018397801385e-05,
+      "loss": 2.2812,
+      "step": 64090
+    },
+    {
+      "epoch": 8.66,
+      "learning_rate": 1.3187908898961352e-05,
+      "loss": 2.2926,
+      "step": 64100
+    },
+    {
+      "epoch": 8.66,
+      "learning_rate": 1.3161824034313522e-05,
+      "loss": 2.2627,
+      "step": 64110
+    },
+    {
+      "epoch": 8.66,
+      "learning_rate": 1.3135763808554755e-05,
+      "loss": 2.2695,
+      "step": 64120
+    },
+    {
+      "epoch": 8.66,
+      "learning_rate": 1.3109728226377415e-05,
+      "loss": 2.2766,
+      "step": 64130
+    },
+    {
+      "epoch": 8.66,
+      "learning_rate": 1.3083717292469382e-05,
+      "loss": 2.2909,
+      "step": 64140
+    },
+    {
+      "epoch": 8.66,
+      "learning_rate": 1.3057731011514227e-05,
+      "loss": 2.2592,
+      "step": 64150
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 1.3031769388190927e-05,
+      "loss": 2.254,
+      "step": 64160
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 1.300583242717414e-05,
+      "loss": 2.2959,
+      "step": 64170
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 1.2979920133134013e-05,
+      "loss": 2.2831,
+      "step": 64180
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 1.2954032510736318e-05,
+      "loss": 2.2567,
+      "step": 64190
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 1.2928169564642338e-05,
+      "loss": 2.285,
+      "step": 64200
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 1.290233129950885e-05,
+      "loss": 2.2854,
+      "step": 64210
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 1.2876517719988333e-05,
+      "loss": 2.3168,
+      "step": 64220
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 1.2850728830728668e-05,
+      "loss": 2.2701,
+      "step": 64230
+    },
+    {
+      "epoch": 8.68,
+      "learning_rate": 1.2824964636373492e-05,
+      "loss": 2.2567,
+      "step": 64240
+    },
+    {
+      "epoch": 8.68,
+      "learning_rate": 1.2799225141561726e-05,
+      "loss": 2.2829,
+      "step": 64250
+    },
+    {
+      "epoch": 8.68,
+      "learning_rate": 1.2773510350928079e-05,
+      "loss": 2.2806,
+      "step": 64260
+    },
+    {
+      "epoch": 8.68,
+      "learning_rate": 1.2747820269102665e-05,
+      "loss": 2.2743,
+      "step": 64270
+    },
+    {
+      "epoch": 8.68,
+      "learning_rate": 1.272215490071125e-05,
+      "loss": 2.3117,
+      "step": 64280
+    },
+    {
+      "epoch": 8.68,
+      "learning_rate": 1.2696514250375122e-05,
+      "loss": 2.2668,
+      "step": 64290
+    },
+    {
+      "epoch": 8.68,
+      "learning_rate": 1.2670898322711003e-05,
+      "loss": 2.2694,
+      "step": 64300
+    },
+    {
+      "epoch": 8.69,
+      "learning_rate": 1.2645307122331354e-05,
+      "loss": 2.2511,
+      "step": 64310
+    },
+    {
+      "epoch": 8.69,
+      "learning_rate": 1.261974065384404e-05,
+      "loss": 2.2613,
+      "step": 64320
+    },
+    {
+      "epoch": 8.69,
+      "learning_rate": 1.259419892185256e-05,
+      "loss": 2.276,
+      "step": 64330
+    },
+    {
+      "epoch": 8.69,
+      "learning_rate": 1.2568681930955888e-05,
+      "loss": 2.2722,
+      "step": 64340
+    },
+    {
+      "epoch": 8.69,
+      "learning_rate": 1.2543189685748611e-05,
+      "loss": 2.3039,
+      "step": 64350
+    },
+    {
+      "epoch": 8.69,
+      "learning_rate": 1.2517722190820777e-05,
+      "loss": 2.293,
+      "step": 64360
+    },
+    {
+      "epoch": 8.69,
+      "learning_rate": 1.2492279450758086e-05,
+      "loss": 2.2538,
+      "step": 64370
+    },
+    {
+      "epoch": 8.7,
+      "learning_rate": 1.2466861470141703e-05,
+      "loss": 2.293,
+      "step": 64380
+    },
+    {
+      "epoch": 8.7,
+      "learning_rate": 1.2441468253548337e-05,
+      "loss": 2.2807,
+      "step": 64390
+    },
+    {
+      "epoch": 8.7,
+      "learning_rate": 1.2416099805550261e-05,
+      "loss": 2.2693,
+      "step": 64400
+    },
+    {
+      "epoch": 8.7,
+      "learning_rate": 1.2390756130715257e-05,
+      "loss": 2.2907,
+      "step": 64410
+    },
+    {
+      "epoch": 8.7,
+      "learning_rate": 1.2365437233606723e-05,
+      "loss": 2.28,
+      "step": 64420
+    },
+    {
+      "epoch": 8.7,
+      "learning_rate": 1.234014311878353e-05,
+      "loss": 2.2629,
+      "step": 64430
+    },
+    {
+      "epoch": 8.7,
+      "learning_rate": 1.2314873790800084e-05,
+      "loss": 2.2919,
+      "step": 64440
+    },
+    {
+      "epoch": 8.7,
+      "learning_rate": 1.2289629254206313e-05,
+      "loss": 2.249,
+      "step": 64450
+    },
+    {
+      "epoch": 8.71,
+      "learning_rate": 1.2264409513547796e-05,
+      "loss": 2.2752,
+      "step": 64460
+    },
+    {
+      "epoch": 8.71,
+      "learning_rate": 1.223921457336552e-05,
+      "loss": 2.2788,
+      "step": 64470
+    },
+    {
+      "epoch": 8.71,
+      "learning_rate": 1.2214044438196036e-05,
+      "loss": 2.2908,
+      "step": 64480
+    },
+    {
+      "epoch": 8.71,
+      "learning_rate": 1.2188899112571455e-05,
+      "loss": 2.2839,
+      "step": 64490
+    },
+    {
+      "epoch": 8.71,
+      "learning_rate": 1.2163778601019385e-05,
+      "loss": 2.2606,
+      "step": 64500
+    },
+    {
+      "epoch": 8.71,
+      "learning_rate": 1.2138682908063047e-05,
+      "loss": 2.2914,
+      "step": 64510
+    },
+    {
+      "epoch": 8.71,
+      "learning_rate": 1.2113612038221105e-05,
+      "loss": 2.2536,
+      "step": 64520
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 1.2088565996007764e-05,
+      "loss": 2.2732,
+      "step": 64530
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 1.2063544785932789e-05,
+      "loss": 2.2746,
+      "step": 64540
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 1.2038548412501503e-05,
+      "loss": 2.2831,
+      "step": 64550
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 1.2013576880214676e-05,
+      "loss": 2.3174,
+      "step": 64560
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 1.198863019356866e-05,
+      "loss": 2.2805,
+      "step": 64570
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 1.196370835705533e-05,
+      "loss": 2.3062,
+      "step": 64580
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 1.1938811375162027e-05,
+      "loss": 2.2877,
+      "step": 64590
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 1.1913939252371735e-05,
+      "loss": 2.255,
+      "step": 64600
+    },
+    {
+      "epoch": 8.73,
+      "learning_rate": 1.188909199316287e-05,
+      "loss": 2.2826,
+      "step": 64610
+    },
+    {
+      "epoch": 8.73,
+      "learning_rate": 1.1864269602009385e-05,
+      "loss": 2.2846,
+      "step": 64620
+    },
+    {
+      "epoch": 8.73,
+      "learning_rate": 1.1839472083380742e-05,
+      "loss": 2.2832,
+      "step": 64630
+    },
+    {
+      "epoch": 8.73,
+      "learning_rate": 1.1814699441741998e-05,
+      "loss": 2.2613,
+      "step": 64640
+    },
+    {
+      "epoch": 8.73,
+      "learning_rate": 1.1789951681553655e-05,
+      "loss": 2.2787,
+      "step": 64650
+    },
+    {
+      "epoch": 8.73,
+      "learning_rate": 1.1765228807271749e-05,
+      "loss": 2.2943,
+      "step": 64660
+    },
+    {
+      "epoch": 8.73,
+      "learning_rate": 1.1740530823347866e-05,
+      "loss": 2.2558,
+      "step": 64670
+    },
+    {
+      "epoch": 8.74,
+      "learning_rate": 1.171585773422905e-05,
+      "loss": 2.3028,
+      "step": 64680
+    },
+    {
+      "epoch": 8.74,
+      "learning_rate": 1.1691209544357965e-05,
+      "loss": 2.2798,
+      "step": 64690
+    },
+    {
+      "epoch": 8.74,
+      "learning_rate": 1.1666586258172694e-05,
+      "loss": 2.2811,
+      "step": 64700
+    },
+    {
+      "epoch": 8.74,
+      "learning_rate": 1.164198788010689e-05,
+      "loss": 2.2955,
+      "step": 64710
+    },
+    {
+      "epoch": 8.74,
+      "learning_rate": 1.1617414414589676e-05,
+      "loss": 2.2789,
+      "step": 64720
+    },
+    {
+      "epoch": 8.74,
+      "learning_rate": 1.1592865866045697e-05,
+      "loss": 2.2768,
+      "step": 64730
+    },
+    {
+      "epoch": 8.74,
+      "learning_rate": 1.1568342238895184e-05,
+      "loss": 2.2719,
+      "step": 64740
+    },
+    {
+      "epoch": 8.75,
+      "learning_rate": 1.1543843537553804e-05,
+      "loss": 2.2734,
+      "step": 64750
+    },
+    {
+      "epoch": 8.75,
+      "learning_rate": 1.1519369766432762e-05,
+      "loss": 2.297,
+      "step": 64760
+    },
+    {
+      "epoch": 8.75,
+      "learning_rate": 1.1494920929938733e-05,
+      "loss": 2.2954,
+      "step": 64770
+    },
+    {
+      "epoch": 8.75,
+      "learning_rate": 1.1470497032474013e-05,
+      "loss": 2.279,
+      "step": 64780
+    },
+    {
+      "epoch": 8.75,
+      "learning_rate": 1.1446098078436283e-05,
+      "loss": 2.2633,
+      "step": 64790
+    },
+    {
+      "epoch": 8.75,
+      "learning_rate": 1.1421724072218797e-05,
+      "loss": 2.2575,
+      "step": 64800
+    },
+    {
+      "epoch": 8.75,
+      "learning_rate": 1.1397375018210292e-05,
+      "loss": 2.2838,
+      "step": 64810
+    },
+    {
+      "epoch": 8.75,
+      "learning_rate": 1.1373050920795029e-05,
+      "loss": 2.2937,
+      "step": 64820
+    },
+    {
+      "epoch": 8.76,
+      "learning_rate": 1.1348751784352788e-05,
+      "loss": 2.2422,
+      "step": 64830
+    },
+    {
+      "epoch": 8.76,
+      "learning_rate": 1.1324477613258837e-05,
+      "loss": 2.2794,
+      "step": 64840
+    },
+    {
+      "epoch": 8.76,
+      "learning_rate": 1.1300228411883927e-05,
+      "loss": 2.3019,
+      "step": 64850
+    },
+    {
+      "epoch": 8.76,
+      "learning_rate": 1.1276004184594333e-05,
+      "loss": 2.2891,
+      "step": 64860
+    },
+    {
+      "epoch": 8.76,
+      "learning_rate": 1.1251804935751868e-05,
+      "loss": 2.2586,
+      "step": 64870
+    },
+    {
+      "epoch": 8.76,
+      "learning_rate": 1.122763066971376e-05,
+      "loss": 2.2816,
+      "step": 64880
+    },
+    {
+      "epoch": 8.76,
+      "learning_rate": 1.1203481390832898e-05,
+      "loss": 2.2883,
+      "step": 64890
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 1.1179357103457466e-05,
+      "loss": 2.2593,
+      "step": 64900
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 1.115525781193124e-05,
+      "loss": 2.3009,
+      "step": 64910
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 1.1131183520593584e-05,
+      "loss": 2.3058,
+      "step": 64920
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 1.1107134233779247e-05,
+      "loss": 2.2622,
+      "step": 64930
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 1.1083109955818497e-05,
+      "loss": 2.2731,
+      "step": 64940
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 1.1059110691037093e-05,
+      "loss": 2.2951,
+      "step": 64950
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 1.1035136443756359e-05,
+      "loss": 2.254,
+      "step": 64960
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 1.1011187218293011e-05,
+      "loss": 2.2429,
+      "step": 64970
+    },
+    {
+      "epoch": 8.78,
+      "learning_rate": 1.0987263018959397e-05,
+      "loss": 2.2607,
+      "step": 64980
+    },
+    {
+      "epoch": 8.78,
+      "learning_rate": 1.0963363850063205e-05,
+      "loss": 2.2862,
+      "step": 64990
+    },
+    {
+      "epoch": 8.78,
+      "learning_rate": 1.0939489715907662e-05,
+      "loss": 2.2789,
+      "step": 65000
+    },
+    {
+      "epoch": 8.78,
+      "eval_loss": 2.423799991607666,
+      "eval_runtime": 1269.0127,
+      "eval_samples_per_second": 59.995,
+      "eval_steps_per_second": 5.0,
+      "step": 65000
+    },
+    {
+      "epoch": 8.78,
+      "learning_rate": 0.00011032802507499742,
+      "loss": 2.298,
+      "step": 65010
+    },
+    {
+      "epoch": 8.78,
+      "learning_rate": 0.00011028710508691679,
+      "loss": 2.2818,
+      "step": 65020
+    },
+    {
+      "epoch": 8.78,
+      "learning_rate": 0.00011024618827678742,
+      "loss": 2.2924,
+      "step": 65030
+    },
+    {
+      "epoch": 8.78,
+      "learning_rate": 0.00011020527464788362,
+      "loss": 2.3126,
+      "step": 65040
+    },
+    {
+      "epoch": 8.79,
+      "learning_rate": 0.00011016436420347948,
+      "loss": 2.2822,
+      "step": 65050
+    },
+    {
+      "epoch": 8.79,
+      "learning_rate": 0.0001101234569468487,
+      "loss": 2.3013,
+      "step": 65060
+    },
+    {
+      "epoch": 8.79,
+      "learning_rate": 0.00011008255288126484,
+      "loss": 2.3096,
+      "step": 65070
+    },
+    {
+      "epoch": 8.79,
+      "learning_rate": 0.00011004165201000116,
+      "loss": 2.3275,
+      "step": 65080
+    },
+    {
+      "epoch": 8.79,
+      "learning_rate": 0.00011000075433633074,
+      "loss": 2.2697,
+      "step": 65090
+    },
+    {
+      "epoch": 8.79,
+      "learning_rate": 0.00010995985986352624,
+      "loss": 2.3019,
+      "step": 65100
+    },
+    {
+      "epoch": 8.79,
+      "learning_rate": 0.00010991896859486023,
+      "loss": 2.3125,
+      "step": 65110
+    },
+    {
+      "epoch": 8.8,
+      "learning_rate": 0.00010987808053360495,
+      "loss": 2.3284,
+      "step": 65120
+    },
+    {
+      "epoch": 8.8,
+      "learning_rate": 0.00010983719568303241,
+      "loss": 2.3277,
+      "step": 65130
+    },
+    {
+      "epoch": 8.8,
+      "learning_rate": 0.00010979631404641428,
+      "loss": 2.3207,
+      "step": 65140
+    },
+    {
+      "epoch": 8.8,
+      "learning_rate": 0.0001097554356270221,
+      "loss": 2.3009,
+      "step": 65150
+    },
+    {
+      "epoch": 8.8,
+      "learning_rate": 0.00010971456042812707,
+      "loss": 2.3218,
+      "step": 65160
+    },
+    {
+      "epoch": 8.8,
+      "learning_rate": 0.0001096736884530002,
+      "loss": 2.3457,
+      "step": 65170
+    },
+    {
+      "epoch": 8.8,
+      "learning_rate": 0.00010963281970491208,
+      "loss": 2.3026,
+      "step": 65180
+    },
+    {
+      "epoch": 8.8,
+      "learning_rate": 0.00010959195418713318,
+      "loss": 2.3229,
+      "step": 65190
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 0.00010955109190293381,
+      "loss": 2.3555,
+      "step": 65200
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 0.00010951023285558385,
+      "loss": 2.3307,
+      "step": 65210
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 0.00010946937704835289,
+      "loss": 2.3357,
+      "step": 65220
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 0.00010942852448451037,
+      "loss": 2.3314,
+      "step": 65230
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 0.00010938767516732549,
+      "loss": 2.2866,
+      "step": 65240
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 0.00010934682910006712,
+      "loss": 2.3265,
+      "step": 65250
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 0.00010930598628600387,
+      "loss": 2.3542,
+      "step": 65260
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 0.0001092651467284041,
+      "loss": 2.3234,
+      "step": 65270
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 0.00010922431043053598,
+      "loss": 2.3185,
+      "step": 65280
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 0.00010918347739566734,
+      "loss": 2.3011,
+      "step": 65290
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 0.00010914264762706571,
+      "loss": 2.3168,
+      "step": 65300
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 0.00010910182112799848,
+      "loss": 2.3058,
+      "step": 65310
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 0.0001090609979017327,
+      "loss": 2.3258,
+      "step": 65320
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 0.00010902017795153518,
+      "loss": 2.3216,
+      "step": 65330
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 0.00010897936128067245,
+      "loss": 2.3203,
+      "step": 65340
+    },
+    {
+      "epoch": 8.83,
+      "learning_rate": 0.00010893854789241075,
+      "loss": 2.3575,
+      "step": 65350
+    },
+    {
+      "epoch": 8.83,
+      "learning_rate": 0.00010889773779001617,
+      "loss": 2.336,
+      "step": 65360
+    },
+    {
+      "epoch": 8.83,
+      "learning_rate": 0.00010885693097675448,
+      "loss": 2.3056,
+      "step": 65370
+    },
+    {
+      "epoch": 8.83,
+      "learning_rate": 0.00010881612745589104,
+      "loss": 2.32,
+      "step": 65380
+    },
+    {
+      "epoch": 8.83,
+      "learning_rate": 0.00010877532723069114,
+      "loss": 2.3128,
+      "step": 65390
+    },
+    {
+      "epoch": 8.83,
+      "learning_rate": 0.00010873453030441981,
+      "loss": 2.3343,
+      "step": 65400
+    },
+    {
+      "epoch": 8.83,
+      "learning_rate": 0.00010869373668034174,
+      "loss": 2.3212,
+      "step": 65410
+    },
+    {
+      "epoch": 8.84,
+      "learning_rate": 0.00010865294636172126,
+      "loss": 2.3265,
+      "step": 65420
+    },
+    {
+      "epoch": 8.84,
+      "learning_rate": 0.0001086121593518226,
+      "loss": 2.2803,
+      "step": 65430
+    },
+    {
+      "epoch": 8.84,
+      "learning_rate": 0.00010857137565390967,
+      "loss": 2.3106,
+      "step": 65440
+    },
+    {
+      "epoch": 8.84,
+      "learning_rate": 0.00010853059527124616,
+      "loss": 2.3361,
+      "step": 65450
+    },
+    {
+      "epoch": 8.84,
+      "learning_rate": 0.00010848981820709534,
+      "loss": 2.3107,
+      "step": 65460
+    },
+    {
+      "epoch": 8.84,
+      "learning_rate": 0.00010844904446472038,
+      "loss": 2.3249,
+      "step": 65470
+    },
+    {
+      "epoch": 8.84,
+      "learning_rate": 0.00010840827404738412,
+      "loss": 2.3392,
+      "step": 65480
+    },
+    {
+      "epoch": 8.85,
+      "learning_rate": 0.00010836750695834916,
+      "loss": 2.3613,
+      "step": 65490
+    },
+    {
+      "epoch": 8.85,
+      "learning_rate": 0.00010832674320087774,
+      "loss": 2.352,
+      "step": 65500
+    },
+    {
+      "epoch": 8.85,
+      "learning_rate": 0.00010828598277823196,
+      "loss": 2.3089,
+      "step": 65510
+    },
+    {
+      "epoch": 8.85,
+      "learning_rate": 0.00010824522569367361,
+      "loss": 2.3179,
+      "step": 65520
+    },
+    {
+      "epoch": 8.85,
+      "learning_rate": 0.00010820447195046418,
+      "loss": 2.3291,
+      "step": 65530
+    },
+    {
+      "epoch": 8.85,
+      "learning_rate": 0.00010816372155186489,
+      "loss": 2.333,
+      "step": 65540
+    },
+    {
+      "epoch": 8.85,
+      "learning_rate": 0.00010812297450113673,
+      "loss": 2.3415,
+      "step": 65550
+    },
+    {
+      "epoch": 8.85,
+      "learning_rate": 0.00010808223080154044,
+      "loss": 2.3432,
+      "step": 65560
+    },
+    {
+      "epoch": 8.86,
+      "learning_rate": 0.00010804149045633646,
+      "loss": 2.3165,
+      "step": 65570
+    },
+    {
+      "epoch": 8.86,
+      "learning_rate": 0.00010800075346878492,
+      "loss": 2.3421,
+      "step": 65580
+    },
+    {
+      "epoch": 8.86,
+      "learning_rate": 0.00010796001984214567,
+      "loss": 2.3369,
+      "step": 65590
+    },
+    {
+      "epoch": 8.86,
+      "learning_rate": 0.00010791928957967845,
+      "loss": 2.3183,
+      "step": 65600
+    },
+    {
+      "epoch": 8.86,
+      "learning_rate": 0.00010787856268464267,
+      "loss": 2.3005,
+      "step": 65610
+    },
+    {
+      "epoch": 8.86,
+      "learning_rate": 0.00010783783916029723,
+      "loss": 2.3284,
+      "step": 65620
+    },
+    {
+      "epoch": 8.86,
+      "learning_rate": 0.00010779711900990111,
+      "loss": 2.3353,
+      "step": 65630
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 0.00010775640223671279,
+      "loss": 2.3338,
+      "step": 65640
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 0.00010771568884399063,
+      "loss": 2.3238,
+      "step": 65650
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 0.00010767497883499258,
+      "loss": 2.323,
+      "step": 65660
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 0.0001076342722129764,
+      "loss": 2.3202,
+      "step": 65670
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 0.00010759356898119956,
+      "loss": 2.3298,
+      "step": 65680
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 0.00010755286914291929,
+      "loss": 2.3072,
+      "step": 65690
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 0.00010751217270139248,
+      "loss": 2.3452,
+      "step": 65700
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 0.00010747147965987578,
+      "loss": 2.3392,
+      "step": 65710
+    },
+    {
+      "epoch": 8.88,
+      "learning_rate": 0.00010743079002162561,
+      "loss": 2.3526,
+      "step": 65720
+    },
+    {
+      "epoch": 8.88,
+      "learning_rate": 0.00010739010378989813,
+      "loss": 2.2986,
+      "step": 65730
+    },
+    {
+      "epoch": 8.88,
+      "learning_rate": 0.00010734942096794905,
+      "loss": 2.3602,
+      "step": 65740
+    },
+    {
+      "epoch": 8.88,
+      "learning_rate": 0.00010730874155903405,
+      "loss": 2.332,
+      "step": 65750
+    },
+    {
+      "epoch": 8.88,
+      "learning_rate": 0.00010726806556640836,
+      "loss": 2.3365,
+      "step": 65760
+    },
+    {
+      "epoch": 8.88,
+      "learning_rate": 0.0001072273929933271,
+      "loss": 2.3384,
+      "step": 65770
+    },
+    {
+      "epoch": 8.88,
+      "learning_rate": 0.00010718672384304488,
+      "loss": 2.302,
+      "step": 65780
+    },
+    {
+      "epoch": 8.89,
+      "learning_rate": 0.00010714605811881622,
+      "loss": 2.2957,
+      "step": 65790
+    },
+    {
+      "epoch": 8.89,
+      "learning_rate": 0.00010710539582389537,
+      "loss": 2.3329,
+      "step": 65800
+    },
+    {
+      "epoch": 8.89,
+      "learning_rate": 0.00010706473696153627,
+      "loss": 2.3449,
+      "step": 65810
+    },
+    {
+      "epoch": 8.89,
+      "learning_rate": 0.00010702408153499247,
+      "loss": 2.3305,
+      "step": 65820
+    },
+    {
+      "epoch": 8.89,
+      "learning_rate": 0.0001069834295475174,
+      "loss": 2.3192,
+      "step": 65830
+    },
+    {
+      "epoch": 8.89,
+      "learning_rate": 0.00010694278100236417,
+      "loss": 2.3124,
+      "step": 65840
+    },
+    {
+      "epoch": 8.89,
+      "learning_rate": 0.00010690213590278565,
+      "loss": 2.3043,
+      "step": 65850
+    },
+    {
+      "epoch": 8.9,
+      "learning_rate": 0.00010686149425203426,
+      "loss": 2.3322,
+      "step": 65860
+    },
+    {
+      "epoch": 8.9,
+      "learning_rate": 0.00010682085605336236,
+      "loss": 2.346,
+      "step": 65870
+    },
+    {
+      "epoch": 8.9,
+      "learning_rate": 0.00010678022131002195,
+      "loss": 2.3445,
+      "step": 65880
+    },
+    {
+      "epoch": 8.9,
+      "learning_rate": 0.00010673959002526477,
+      "loss": 2.3325,
+      "step": 65890
+    },
+    {
+      "epoch": 8.9,
+      "learning_rate": 0.00010669896220234216,
+      "loss": 2.331,
+      "step": 65900
+    },
+    {
+      "epoch": 8.9,
+      "learning_rate": 0.00010665833784450539,
+      "loss": 2.3241,
+      "step": 65910
+    },
+    {
+      "epoch": 8.9,
+      "learning_rate": 0.00010661771695500532,
+      "loss": 2.3273,
+      "step": 65920
+    },
+    {
+      "epoch": 8.9,
+      "learning_rate": 0.00010657709953709258,
+      "loss": 2.3786,
+      "step": 65930
+    },
+    {
+      "epoch": 8.91,
+      "learning_rate": 0.00010653648559401745,
+      "loss": 2.3494,
+      "step": 65940
+    },
+    {
+      "epoch": 8.91,
+      "learning_rate": 0.00010649587512903001,
+      "loss": 2.3227,
+      "step": 65950
+    },
+    {
+      "epoch": 8.91,
+      "learning_rate": 0.00010645526814538006,
+      "loss": 2.3135,
+      "step": 65960
+    },
+    {
+      "epoch": 8.91,
+      "learning_rate": 0.00010641466464631714,
+      "loss": 2.3401,
+      "step": 65970
+    },
+    {
+      "epoch": 8.91,
+      "learning_rate": 0.00010637406463509034,
+      "loss": 2.3354,
+      "step": 65980
+    },
+    {
+      "epoch": 8.91,
+      "learning_rate": 0.00010633346811494866,
+      "loss": 2.3334,
+      "step": 65990
+    },
+    {
+      "epoch": 8.91,
+      "learning_rate": 0.0001062928750891408,
+      "loss": 2.3245,
+      "step": 66000
+    },
+    {
+      "epoch": 8.91,
+      "eval_loss": 2.463181734085083,
+      "eval_runtime": 1269.4771,
+      "eval_samples_per_second": 59.974,
+      "eval_steps_per_second": 4.998,
+      "step": 66000
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 0.0001062522855609152,
+      "loss": 2.3069,
+      "step": 66010
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 0.00010621169953351982,
+      "loss": 2.3352,
+      "step": 66020
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 0.00010617111701020252,
+      "loss": 2.3111,
+      "step": 66030
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 0.0001061305379942109,
+      "loss": 2.3607,
+      "step": 66040
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 0.00010608996248879219,
+      "loss": 2.308,
+      "step": 66050
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 0.00010604939049719334,
+      "loss": 2.3331,
+      "step": 66060
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 0.00010600882202266109,
+      "loss": 2.335,
+      "step": 66070
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 0.00010596825706844184,
+      "loss": 2.3553,
+      "step": 66080
+    },
+    {
+      "epoch": 8.93,
+      "learning_rate": 0.00010592769563778174,
+      "loss": 2.3339,
+      "step": 66090
+    },
+    {
+      "epoch": 8.93,
+      "learning_rate": 0.00010588713773392661,
+      "loss": 2.3259,
+      "step": 66100
+    },
+    {
+      "epoch": 8.93,
+      "learning_rate": 0.00010584658336012205,
+      "loss": 2.3591,
+      "step": 66110
+    },
+    {
+      "epoch": 8.93,
+      "learning_rate": 0.00010580603251961335,
+      "loss": 2.3095,
+      "step": 66120
+    },
+    {
+      "epoch": 8.93,
+      "learning_rate": 0.00010576548521564552,
+      "loss": 2.3324,
+      "step": 66130
+    },
+    {
+      "epoch": 8.93,
+      "learning_rate": 0.00010572494145146329,
+      "loss": 2.3383,
+      "step": 66140
+    },
+    {
+      "epoch": 8.93,
+      "learning_rate": 0.00010568440123031105,
+      "loss": 2.3536,
+      "step": 66150
+    },
+    {
+      "epoch": 8.94,
+      "learning_rate": 0.00010564386455543297,
+      "loss": 2.3267,
+      "step": 66160
+    },
+    {
+      "epoch": 8.94,
+      "learning_rate": 0.00010560333143007304,
+      "loss": 2.3479,
+      "step": 66170
+    },
+    {
+      "epoch": 8.94,
+      "learning_rate": 0.00010556280185747467,
+      "loss": 2.3463,
+      "step": 66180
+    },
+    {
+      "epoch": 8.94,
+      "learning_rate": 0.00010552227584088125,
+      "loss": 2.3545,
+      "step": 66190
+    },
+    {
+      "epoch": 8.94,
+      "learning_rate": 0.00010548175338353576,
+      "loss": 2.3268,
+      "step": 66200
+    },
+    {
+      "epoch": 8.94,
+      "learning_rate": 0.00010544123448868109,
+      "loss": 2.3448,
+      "step": 66210
+    },
+    {
+      "epoch": 8.94,
+      "learning_rate": 0.00010540071915955947,
+      "loss": 2.3611,
+      "step": 66220
+    },
+    {
+      "epoch": 8.95,
+      "learning_rate": 0.00010536020739941319,
+      "loss": 2.3535,
+      "step": 66230
+    },
+    {
+      "epoch": 8.95,
+      "learning_rate": 0.00010531969921148411,
+      "loss": 2.3222,
+      "step": 66240
+    },
+    {
+      "epoch": 8.95,
+      "learning_rate": 0.00010527919459901383,
+      "loss": 2.3093,
+      "step": 66250
+    },
+    {
+      "epoch": 8.95,
+      "learning_rate": 0.00010523869356524362,
+      "loss": 2.3169,
+      "step": 66260
+    },
+    {
+      "epoch": 8.95,
+      "learning_rate": 0.00010519819611341452,
+      "loss": 2.3183,
+      "step": 66270
+    },
+    {
+      "epoch": 8.95,
+      "learning_rate": 0.00010515770224676728,
+      "loss": 2.3573,
+      "step": 66280
+    },
+    {
+      "epoch": 8.95,
+      "learning_rate": 0.00010511721196854237,
+      "loss": 2.3544,
+      "step": 66290
+    },
+    {
+      "epoch": 8.95,
+      "learning_rate": 0.00010507672528197987,
+      "loss": 2.3199,
+      "step": 66300
+    },
+    {
+      "epoch": 8.96,
+      "learning_rate": 0.00010503624219031972,
+      "loss": 2.3108,
+      "step": 66310
+    },
+    {
+      "epoch": 8.96,
+      "learning_rate": 0.00010499576269680146,
+      "loss": 2.3443,
+      "step": 66320
+    },
+    {
+      "epoch": 8.96,
+      "learning_rate": 0.00010495528680466448,
+      "loss": 2.3269,
+      "step": 66330
+    },
+    {
+      "epoch": 8.96,
+      "learning_rate": 0.00010491481451714764,
+      "loss": 2.3324,
+      "step": 66340
+    },
+    {
+      "epoch": 8.96,
+      "learning_rate": 0.00010487434583748973,
+      "loss": 2.3503,
+      "step": 66350
+    },
+    {
+      "epoch": 8.96,
+      "learning_rate": 0.00010483388076892924,
+      "loss": 2.3197,
+      "step": 66360
+    },
+    {
+      "epoch": 8.96,
+      "learning_rate": 0.00010479341931470431,
+      "loss": 2.324,
+      "step": 66370
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 0.00010475296147805267,
+      "loss": 2.3471,
+      "step": 66380
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 0.00010471250726221197,
+      "loss": 2.3493,
+      "step": 66390
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 0.00010467205667041946,
+      "loss": 2.3294,
+      "step": 66400
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 0.0001046316097059122,
+      "loss": 2.3345,
+      "step": 66410
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 0.00010459116637192679,
+      "loss": 2.3262,
+      "step": 66420
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 0.00010455072667169964,
+      "loss": 2.3483,
+      "step": 66430
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 0.00010451029060846691,
+      "loss": 2.2973,
+      "step": 66440
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 0.00010446985818546443,
+      "loss": 2.3341,
+      "step": 66450
+    },
+    {
+      "epoch": 8.98,
+      "learning_rate": 0.00010442942940592766,
+      "loss": 2.3696,
+      "step": 66460
+    },
+    {
+      "epoch": 8.98,
+      "learning_rate": 0.0001043890042730919,
+      "loss": 2.3518,
+      "step": 66470
+    },
+    {
+      "epoch": 8.98,
+      "learning_rate": 0.00010434858279019205,
+      "loss": 2.3238,
+      "step": 66480
+    },
+    {
+      "epoch": 8.98,
+      "learning_rate": 0.00010430816496046284,
+      "loss": 2.3411,
+      "step": 66490
+    },
+    {
+      "epoch": 8.98,
+      "learning_rate": 0.00010426775078713854,
+      "loss": 2.3636,
+      "step": 66500
+    },
+    {
+      "epoch": 8.98,
+      "learning_rate": 0.00010422734027345326,
+      "loss": 2.3255,
+      "step": 66510
+    },
+    {
+      "epoch": 8.98,
+      "learning_rate": 0.00010418693342264078,
+      "loss": 2.2983,
+      "step": 66520
+    },
+    {
+      "epoch": 8.99,
+      "learning_rate": 0.00010414653023793464,
+      "loss": 2.3351,
+      "step": 66530
+    },
+    {
+      "epoch": 8.99,
+      "learning_rate": 0.0001041061307225679,
+      "loss": 2.3587,
+      "step": 66540
+    },
+    {
+      "epoch": 8.99,
+      "learning_rate": 0.00010406573487977356,
+      "loss": 2.3191,
+      "step": 66550
+    },
+    {
+      "epoch": 8.99,
+      "learning_rate": 0.00010402534271278417,
+      "loss": 2.3563,
+      "step": 66560
+    },
+    {
+      "epoch": 8.99,
+      "learning_rate": 0.00010398495422483215,
+      "loss": 2.358,
+      "step": 66570
+    },
+    {
+      "epoch": 8.99,
+      "learning_rate": 0.00010394456941914936,
+      "loss": 2.3265,
+      "step": 66580
+    },
+    {
+      "epoch": 8.99,
+      "learning_rate": 0.00010390418829896758,
+      "loss": 2.3412,
+      "step": 66590
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.00010386381086751822,
+      "loss": 2.3406,
+      "step": 66600
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.00010382343712803253,
+      "loss": 2.3635,
+      "step": 66610
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.00010378306708374117,
+      "loss": 2.3523,
+      "step": 66620
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.00010374270073787475,
+      "loss": 2.3207,
+      "step": 66630
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.0001037023380936635,
+      "loss": 2.3281,
+      "step": 66640
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.00010366197915433745,
+      "loss": 2.3065,
+      "step": 66650
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.00010362162392312614,
+      "loss": 2.3141,
+      "step": 66660
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.00010358127240325896,
+      "loss": 2.303,
+      "step": 66670
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 0.00010354092459796496,
+      "loss": 2.3302,
+      "step": 66680
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 0.00010350058051047298,
+      "loss": 2.3243,
+      "step": 66690
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 0.00010346024014401133,
+      "loss": 2.3088,
+      "step": 66700
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 0.00010341990350180829,
+      "loss": 2.2969,
+      "step": 66710
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 0.00010337957058709167,
+      "loss": 2.2866,
+      "step": 66720
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 0.00010333924140308915,
+      "loss": 2.2995,
+      "step": 66730
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 0.00010329891595302778,
+      "loss": 2.3179,
+      "step": 66740
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 0.0001032585942401347,
+      "loss": 2.304,
+      "step": 66750
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 0.00010321827626763653,
+      "loss": 2.3148,
+      "step": 66760
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 0.00010317796203875972,
+      "loss": 2.3425,
+      "step": 66770
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 0.00010313765155673021,
+      "loss": 2.3126,
+      "step": 66780
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 0.00010309734482477383,
+      "loss": 2.3193,
+      "step": 66790
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 0.00010305704184611602,
+      "loss": 2.3163,
+      "step": 66800
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 0.00010301674262398209,
+      "loss": 2.3048,
+      "step": 66810
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 0.00010297644716159673,
+      "loss": 2.323,
+      "step": 66820
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 0.00010293615546218458,
+      "loss": 2.3172,
+      "step": 66830
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 0.00010289586752896995,
+      "loss": 2.2832,
+      "step": 66840
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 0.00010285558336517677,
+      "loss": 2.3142,
+      "step": 66850
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 0.00010281530297402869,
+      "loss": 2.3155,
+      "step": 66860
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 0.00010277502635874912,
+      "loss": 2.2911,
+      "step": 66870
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 0.00010273475352256106,
+      "loss": 2.3134,
+      "step": 66880
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 0.00010269448446868735,
+      "loss": 2.3272,
+      "step": 66890
+    },
+    {
+      "epoch": 9.04,
+      "learning_rate": 0.00010265421920035037,
+      "loss": 2.279,
+      "step": 66900
+    },
+    {
+      "epoch": 9.04,
+      "learning_rate": 0.00010261395772077232,
+      "loss": 2.3218,
+      "step": 66910
+    },
+    {
+      "epoch": 9.04,
+      "learning_rate": 0.00010257370003317502,
+      "loss": 2.3008,
+      "step": 66920
+    },
+    {
+      "epoch": 9.04,
+      "learning_rate": 0.0001025334461407801,
+      "loss": 2.3192,
+      "step": 66930
+    },
+    {
+      "epoch": 9.04,
+      "learning_rate": 0.00010249319604680865,
+      "loss": 2.3154,
+      "step": 66940
+    },
+    {
+      "epoch": 9.04,
+      "learning_rate": 0.00010245294975448173,
+      "loss": 2.2971,
+      "step": 66950
+    },
+    {
+      "epoch": 9.04,
+      "learning_rate": 0.00010241270726701993,
+      "loss": 2.3211,
+      "step": 66960
+    },
+    {
+      "epoch": 9.04,
+      "learning_rate": 0.00010237246858764368,
+      "loss": 2.325,
+      "step": 66970
+    },
+    {
+      "epoch": 9.05,
+      "learning_rate": 0.00010233223371957284,
+      "loss": 2.3014,
+      "step": 66980
+    },
+    {
+      "epoch": 9.05,
+      "learning_rate": 0.00010229200266602723,
+      "loss": 2.3403,
+      "step": 66990
+    },
+    {
+      "epoch": 9.05,
+      "learning_rate": 0.0001022517754302262,
+      "loss": 2.337,
+      "step": 67000
+    },
+    {
+      "epoch": 9.05,
+      "eval_loss": 2.464355707168579,
+      "eval_runtime": 1269.0426,
+      "eval_samples_per_second": 59.994,
+      "eval_steps_per_second": 5.0,
+      "step": 67000
+    },
+    {
+      "epoch": 9.05,
+      "learning_rate": 0.000102211552015389,
+      "loss": 2.3159,
+      "step": 67010
+    },
+    {
+      "epoch": 9.05,
+      "learning_rate": 0.00010217133242473427,
+      "loss": 2.3263,
+      "step": 67020
+    },
+    {
+      "epoch": 9.05,
+      "learning_rate": 0.00010213111666148062,
+      "loss": 2.3228,
+      "step": 67030
+    },
+    {
+      "epoch": 9.05,
+      "learning_rate": 0.00010209090472884618,
+      "loss": 2.3465,
+      "step": 67040
+    },
+    {
+      "epoch": 9.06,
+      "learning_rate": 0.0001020506966300489,
+      "loss": 2.3447,
+      "step": 67050
+    },
+    {
+      "epoch": 9.06,
+      "learning_rate": 0.00010201049236830628,
+      "loss": 2.323,
+      "step": 67060
+    },
+    {
+      "epoch": 9.06,
+      "learning_rate": 0.00010197029194683563,
+      "loss": 2.3342,
+      "step": 67070
+    },
+    {
+      "epoch": 9.06,
+      "learning_rate": 0.00010193009536885391,
+      "loss": 2.3161,
+      "step": 67080
+    },
+    {
+      "epoch": 9.06,
+      "learning_rate": 0.0001018899026375778,
+      "loss": 2.3052,
+      "step": 67090
+    },
+    {
+      "epoch": 9.06,
+      "learning_rate": 0.00010184971375622363,
+      "loss": 2.3071,
+      "step": 67100
+    },
+    {
+      "epoch": 9.06,
+      "learning_rate": 0.00010180952872800744,
+      "loss": 2.3062,
+      "step": 67110
+    },
+    {
+      "epoch": 9.07,
+      "learning_rate": 0.00010176934755614496,
+      "loss": 2.3128,
+      "step": 67120
+    },
+    {
+      "epoch": 9.07,
+      "learning_rate": 0.00010172917024385169,
+      "loss": 2.3296,
+      "step": 67130
+    },
+    {
+      "epoch": 9.07,
+      "learning_rate": 0.00010168899679434258,
+      "loss": 2.3431,
+      "step": 67140
+    },
+    {
+      "epoch": 9.07,
+      "learning_rate": 0.00010164882721083253,
+      "loss": 2.3048,
+      "step": 67150
+    },
+    {
+      "epoch": 9.07,
+      "learning_rate": 0.00010160866149653609,
+      "loss": 2.3151,
+      "step": 67160
+    },
+    {
+      "epoch": 9.07,
+      "learning_rate": 0.00010156849965466743,
+      "loss": 2.3131,
+      "step": 67170
+    },
+    {
+      "epoch": 9.07,
+      "learning_rate": 0.00010152834168844033,
+      "loss": 2.3155,
+      "step": 67180
+    },
+    {
+      "epoch": 9.07,
+      "learning_rate": 0.00010148818760106844,
+      "loss": 2.3135,
+      "step": 67190
+    },
+    {
+      "epoch": 9.08,
+      "learning_rate": 0.00010144803739576499,
+      "loss": 2.3157,
+      "step": 67200
+    },
+    {
+      "epoch": 9.08,
+      "learning_rate": 0.00010140789107574295,
+      "loss": 2.3193,
+      "step": 67210
+    },
+    {
+      "epoch": 9.08,
+      "learning_rate": 0.00010136774864421492,
+      "loss": 2.3274,
+      "step": 67220
+    },
+    {
+      "epoch": 9.08,
+      "learning_rate": 0.00010132761010439324,
+      "loss": 2.2954,
+      "step": 67230
+    },
+    {
+      "epoch": 9.08,
+      "learning_rate": 0.00010128747545948996,
+      "loss": 2.3066,
+      "step": 67240
+    },
+    {
+      "epoch": 9.08,
+      "learning_rate": 0.00010124734471271673,
+      "loss": 2.293,
+      "step": 67250
+    },
+    {
+      "epoch": 9.08,
+      "learning_rate": 0.00010120721786728494,
+      "loss": 2.3072,
+      "step": 67260
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 0.00010116709492640569,
+      "loss": 2.3308,
+      "step": 67270
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 0.00010112697589328973,
+      "loss": 2.324,
+      "step": 67280
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 0.00010108686077114757,
+      "loss": 2.3202,
+      "step": 67290
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 0.00010104674956318925,
+      "loss": 2.3002,
+      "step": 67300
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 0.00010100664227262465,
+      "loss": 2.3189,
+      "step": 67310
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 0.00010096653890266327,
+      "loss": 2.3028,
+      "step": 67320
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 0.00010092643945651438,
+      "loss": 2.3149,
+      "step": 67330
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 0.00010088634393738675,
+      "loss": 2.3036,
+      "step": 67340
+    },
+    {
+      "epoch": 9.1,
+      "learning_rate": 0.00010084625234848895,
+      "loss": 2.34,
+      "step": 67350
+    },
+    {
+      "epoch": 9.1,
+      "learning_rate": 0.00010080616469302931,
+      "loss": 2.3238,
+      "step": 67360
+    },
+    {
+      "epoch": 9.1,
+      "learning_rate": 0.00010076608097421582,
+      "loss": 2.2953,
+      "step": 67370
+    },
+    {
+      "epoch": 9.1,
+      "learning_rate": 0.00010072600119525597,
+      "loss": 2.3213,
+      "step": 67380
+    },
+    {
+      "epoch": 9.1,
+      "learning_rate": 0.00010068592535935714,
+      "loss": 2.327,
+      "step": 67390
+    },
+    {
+      "epoch": 9.1,
+      "learning_rate": 0.00010064585346972628,
+      "loss": 2.3276,
+      "step": 67400
+    },
+    {
+      "epoch": 9.1,
+      "learning_rate": 0.00010060578552957017,
+      "loss": 2.3176,
+      "step": 67410
+    },
+    {
+      "epoch": 9.11,
+      "learning_rate": 0.00010056572154209509,
+      "loss": 2.3327,
+      "step": 67420
+    },
+    {
+      "epoch": 9.11,
+      "learning_rate": 0.00010052566151050709,
+      "loss": 2.3009,
+      "step": 67430
+    },
+    {
+      "epoch": 9.11,
+      "learning_rate": 0.0001004856054380119,
+      "loss": 2.3336,
+      "step": 67440
+    },
+    {
+      "epoch": 9.11,
+      "learning_rate": 0.00010044555332781498,
+      "loss": 2.3316,
+      "step": 67450
+    },
+    {
+      "epoch": 9.11,
+      "learning_rate": 0.00010040550518312138,
+      "loss": 2.3167,
+      "step": 67460
+    },
+    {
+      "epoch": 9.11,
+      "learning_rate": 0.0001003654610071359,
+      "loss": 2.3088,
+      "step": 67470
+    },
+    {
+      "epoch": 9.11,
+      "learning_rate": 0.00010032542080306297,
+      "loss": 2.3311,
+      "step": 67480
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 0.00010028538457410679,
+      "loss": 2.3262,
+      "step": 67490
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 0.00010024535232347115,
+      "loss": 2.3034,
+      "step": 67500
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 0.00010020532405435957,
+      "loss": 2.3428,
+      "step": 67510
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 0.00010016529976997519,
+      "loss": 2.3214,
+      "step": 67520
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 0.00010012527947352102,
+      "loss": 2.3312,
+      "step": 67530
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 0.00010008526316819942,
+      "loss": 2.3563,
+      "step": 67540
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 0.0001000452508572127,
+      "loss": 2.3126,
+      "step": 67550
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 0.00010000524254376281,
+      "loss": 2.3431,
+      "step": 67560
+    },
+    {
+      "epoch": 9.13,
+      "learning_rate": 9.996523823105139e-05,
+      "loss": 2.3198,
+      "step": 67570
+    },
+    {
+      "epoch": 9.13,
+      "learning_rate": 9.992523792227955e-05,
+      "loss": 2.3234,
+      "step": 67580
+    },
+    {
+      "epoch": 9.13,
+      "learning_rate": 9.988524162064833e-05,
+      "loss": 2.3005,
+      "step": 67590
+    },
+    {
+      "epoch": 9.13,
+      "learning_rate": 9.984524932935838e-05,
+      "loss": 2.3091,
+      "step": 67600
+    },
+    {
+      "epoch": 9.13,
+      "learning_rate": 9.980526105161001e-05,
+      "loss": 2.3043,
+      "step": 67610
+    },
+    {
+      "epoch": 9.13,
+      "learning_rate": 9.976527679060316e-05,
+      "loss": 2.3296,
+      "step": 67620
+    },
+    {
+      "epoch": 9.13,
+      "learning_rate": 9.972529654953751e-05,
+      "loss": 2.3063,
+      "step": 67630
+    },
+    {
+      "epoch": 9.14,
+      "learning_rate": 9.968532033161242e-05,
+      "loss": 2.3231,
+      "step": 67640
+    },
+    {
+      "epoch": 9.14,
+      "learning_rate": 9.964534814002695e-05,
+      "loss": 2.3274,
+      "step": 67650
+    },
+    {
+      "epoch": 9.14,
+      "learning_rate": 9.960537997797972e-05,
+      "loss": 2.3275,
+      "step": 67660
+    },
+    {
+      "epoch": 9.14,
+      "learning_rate": 9.956541584866915e-05,
+      "loss": 2.3183,
+      "step": 67670
+    },
+    {
+      "epoch": 9.14,
+      "learning_rate": 9.95254557552933e-05,
+      "loss": 2.3342,
+      "step": 67680
+    },
+    {
+      "epoch": 9.14,
+      "learning_rate": 9.948549970104993e-05,
+      "loss": 2.3329,
+      "step": 67690
+    },
+    {
+      "epoch": 9.14,
+      "learning_rate": 9.94455476891364e-05,
+      "loss": 2.3375,
+      "step": 67700
+    },
+    {
+      "epoch": 9.14,
+      "learning_rate": 9.94055997227498e-05,
+      "loss": 2.3058,
+      "step": 67710
+    },
+    {
+      "epoch": 9.15,
+      "learning_rate": 9.93656558050869e-05,
+      "loss": 2.3194,
+      "step": 67720
+    },
+    {
+      "epoch": 9.15,
+      "learning_rate": 9.932571593934424e-05,
+      "loss": 2.2998,
+      "step": 67730
+    },
+    {
+      "epoch": 9.15,
+      "learning_rate": 9.928578012871775e-05,
+      "loss": 2.3435,
+      "step": 67740
+    },
+    {
+      "epoch": 9.15,
+      "learning_rate": 9.92458483764033e-05,
+      "loss": 2.324,
+      "step": 67750
+    },
+    {
+      "epoch": 9.15,
+      "learning_rate": 9.920592068559639e-05,
+      "loss": 2.3025,
+      "step": 67760
+    },
+    {
+      "epoch": 9.15,
+      "learning_rate": 9.916599705949218e-05,
+      "loss": 2.3239,
+      "step": 67770
+    },
+    {
+      "epoch": 9.15,
+      "learning_rate": 9.912607750128539e-05,
+      "loss": 2.3278,
+      "step": 67780
+    },
+    {
+      "epoch": 9.16,
+      "learning_rate": 9.908616201417056e-05,
+      "loss": 2.32,
+      "step": 67790
+    },
+    {
+      "epoch": 9.16,
+      "learning_rate": 9.904625060134184e-05,
+      "loss": 2.3236,
+      "step": 67800
+    },
+    {
+      "epoch": 9.16,
+      "learning_rate": 9.900634326599311e-05,
+      "loss": 2.3224,
+      "step": 67810
+    },
+    {
+      "epoch": 9.16,
+      "learning_rate": 9.896644001131784e-05,
+      "loss": 2.3148,
+      "step": 67820
+    },
+    {
+      "epoch": 9.16,
+      "learning_rate": 9.892654084050923e-05,
+      "loss": 2.3264,
+      "step": 67830
+    },
+    {
+      "epoch": 9.16,
+      "learning_rate": 9.888664575676009e-05,
+      "loss": 2.3043,
+      "step": 67840
+    },
+    {
+      "epoch": 9.16,
+      "learning_rate": 9.884675476326306e-05,
+      "loss": 2.3107,
+      "step": 67850
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 9.880686786321024e-05,
+      "loss": 2.3066,
+      "step": 67860
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 9.876698505979355e-05,
+      "loss": 2.3422,
+      "step": 67870
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 9.87271063562045e-05,
+      "loss": 2.2833,
+      "step": 67880
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 9.868723175563439e-05,
+      "loss": 2.3166,
+      "step": 67890
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 9.864736126127402e-05,
+      "loss": 2.3381,
+      "step": 67900
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 9.860749487631401e-05,
+      "loss": 2.3354,
+      "step": 67910
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 9.856763260394456e-05,
+      "loss": 2.2924,
+      "step": 67920
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 9.852777444735569e-05,
+      "loss": 2.3272,
+      "step": 67930
+    },
+    {
+      "epoch": 9.18,
+      "learning_rate": 9.848792040973677e-05,
+      "loss": 2.3109,
+      "step": 67940
+    },
+    {
+      "epoch": 9.18,
+      "learning_rate": 9.844807049427715e-05,
+      "loss": 2.3272,
+      "step": 67950
+    },
+    {
+      "epoch": 9.18,
+      "learning_rate": 9.84082247041658e-05,
+      "loss": 2.3328,
+      "step": 67960
+    },
+    {
+      "epoch": 9.18,
+      "learning_rate": 9.836838304259127e-05,
+      "loss": 2.3022,
+      "step": 67970
+    },
+    {
+      "epoch": 9.18,
+      "learning_rate": 9.832854551274179e-05,
+      "loss": 2.321,
+      "step": 67980
+    },
+    {
+      "epoch": 9.18,
+      "learning_rate": 9.828871211780527e-05,
+      "loss": 2.3118,
+      "step": 67990
+    },
+    {
+      "epoch": 9.18,
+      "learning_rate": 9.824888286096935e-05,
+      "loss": 2.3403,
+      "step": 68000
+    },
+    {
+      "epoch": 9.18,
+      "eval_loss": 2.463167190551758,
+      "eval_runtime": 1269.0658,
+      "eval_samples_per_second": 59.993,
+      "eval_steps_per_second": 5.0,
+      "step": 68000
+    },
+    {
+      "epoch": 9.19,
+      "learning_rate": 9.820905774542127e-05,
+      "loss": 2.3087,
+      "step": 68010
+    },
+    {
+      "epoch": 9.19,
+      "learning_rate": 9.816923677434796e-05,
+      "loss": 2.3569,
+      "step": 68020
+    },
+    {
+      "epoch": 9.19,
+      "learning_rate": 9.812941995093603e-05,
+      "loss": 2.307,
+      "step": 68030
+    },
+    {
+      "epoch": 9.19,
+      "learning_rate": 9.808960727837173e-05,
+      "loss": 2.3398,
+      "step": 68040
+    },
+    {
+      "epoch": 9.19,
+      "learning_rate": 9.804979875984105e-05,
+      "loss": 2.3211,
+      "step": 68050
+    },
+    {
+      "epoch": 9.19,
+      "learning_rate": 9.80099943985295e-05,
+      "loss": 2.3359,
+      "step": 68060
+    },
+    {
+      "epoch": 9.19,
+      "learning_rate": 9.797019419762245e-05,
+      "loss": 2.3151,
+      "step": 68070
+    },
+    {
+      "epoch": 9.19,
+      "learning_rate": 9.793039816030475e-05,
+      "loss": 2.3348,
+      "step": 68080
+    },
+    {
+      "epoch": 9.2,
+      "learning_rate": 9.78906062897611e-05,
+      "loss": 2.3121,
+      "step": 68090
+    },
+    {
+      "epoch": 9.2,
+      "learning_rate": 9.785081858917564e-05,
+      "loss": 2.2957,
+      "step": 68100
+    },
+    {
+      "epoch": 9.2,
+      "learning_rate": 9.781103506173244e-05,
+      "loss": 2.3039,
+      "step": 68110
+    },
+    {
+      "epoch": 9.2,
+      "learning_rate": 9.777125571061503e-05,
+      "loss": 2.3177,
+      "step": 68120
+    },
+    {
+      "epoch": 9.2,
+      "learning_rate": 9.773148053900676e-05,
+      "loss": 2.3103,
+      "step": 68130
+    },
+    {
+      "epoch": 9.2,
+      "learning_rate": 9.769170955009045e-05,
+      "loss": 2.3279,
+      "step": 68140
+    },
+    {
+      "epoch": 9.2,
+      "learning_rate": 9.76519427470487e-05,
+      "loss": 2.3679,
+      "step": 68150
+    },
+    {
+      "epoch": 9.21,
+      "learning_rate": 9.761218013306386e-05,
+      "loss": 2.3181,
+      "step": 68160
+    },
+    {
+      "epoch": 9.21,
+      "learning_rate": 9.75724217113179e-05,
+      "loss": 2.3381,
+      "step": 68170
+    },
+    {
+      "epoch": 9.21,
+      "learning_rate": 9.753266748499226e-05,
+      "loss": 2.3194,
+      "step": 68180
+    },
+    {
+      "epoch": 9.21,
+      "learning_rate": 9.749291745726831e-05,
+      "loss": 2.3146,
+      "step": 68190
+    },
+    {
+      "epoch": 9.21,
+      "learning_rate": 9.745317163132694e-05,
+      "loss": 2.315,
+      "step": 68200
+    },
+    {
+      "epoch": 9.21,
+      "learning_rate": 9.741343001034876e-05,
+      "loss": 2.3128,
+      "step": 68210
+    },
+    {
+      "epoch": 9.21,
+      "learning_rate": 9.737369259751397e-05,
+      "loss": 2.323,
+      "step": 68220
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 9.733395939600253e-05,
+      "loss": 2.3243,
+      "step": 68230
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 9.729423040899397e-05,
+      "loss": 2.3692,
+      "step": 68240
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 9.72545056396676e-05,
+      "loss": 2.3279,
+      "step": 68250
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 9.721478509120225e-05,
+      "loss": 2.3441,
+      "step": 68260
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 9.717506876677652e-05,
+      "loss": 2.3289,
+      "step": 68270
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 9.713535666956863e-05,
+      "loss": 2.3018,
+      "step": 68280
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 9.70956488027565e-05,
+      "loss": 2.3238,
+      "step": 68290
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 9.705594516951762e-05,
+      "loss": 2.3407,
+      "step": 68300
+    },
+    {
+      "epoch": 9.23,
+      "learning_rate": 9.70162457730292e-05,
+      "loss": 2.3301,
+      "step": 68310
+    },
+    {
+      "epoch": 9.23,
+      "learning_rate": 9.697655061646817e-05,
+      "loss": 2.3171,
+      "step": 68320
+    },
+    {
+      "epoch": 9.23,
+      "learning_rate": 9.693685970301106e-05,
+      "loss": 2.3432,
+      "step": 68330
+    },
+    {
+      "epoch": 9.23,
+      "learning_rate": 9.689717303583398e-05,
+      "loss": 2.3146,
+      "step": 68340
+    },
+    {
+      "epoch": 9.23,
+      "learning_rate": 9.685749061811285e-05,
+      "loss": 2.3328,
+      "step": 68350
+    },
+    {
+      "epoch": 9.23,
+      "learning_rate": 9.681781245302313e-05,
+      "loss": 2.3317,
+      "step": 68360
+    },
+    {
+      "epoch": 9.23,
+      "learning_rate": 9.67781385437401e-05,
+      "loss": 2.3272,
+      "step": 68370
+    },
+    {
+      "epoch": 9.24,
+      "learning_rate": 9.673846889343849e-05,
+      "loss": 2.3403,
+      "step": 68380
+    },
+    {
+      "epoch": 9.24,
+      "learning_rate": 9.66988035052928e-05,
+      "loss": 2.3277,
+      "step": 68390
+    },
+    {
+      "epoch": 9.24,
+      "learning_rate": 9.665914238247721e-05,
+      "loss": 2.3301,
+      "step": 68400
+    },
+    {
+      "epoch": 9.24,
+      "learning_rate": 9.661948552816553e-05,
+      "loss": 2.3414,
+      "step": 68410
+    },
+    {
+      "epoch": 9.24,
+      "learning_rate": 9.657983294553119e-05,
+      "loss": 2.3295,
+      "step": 68420
+    },
+    {
+      "epoch": 9.24,
+      "learning_rate": 9.654018463774735e-05,
+      "loss": 2.3345,
+      "step": 68430
+    },
+    {
+      "epoch": 9.24,
+      "learning_rate": 9.650054060798678e-05,
+      "loss": 2.3263,
+      "step": 68440
+    },
+    {
+      "epoch": 9.24,
+      "learning_rate": 9.646090085942195e-05,
+      "loss": 2.3041,
+      "step": 68450
+    },
+    {
+      "epoch": 9.25,
+      "learning_rate": 9.64212653952249e-05,
+      "loss": 2.3022,
+      "step": 68460
+    },
+    {
+      "epoch": 9.25,
+      "learning_rate": 9.638163421856745e-05,
+      "loss": 2.3272,
+      "step": 68470
+    },
+    {
+      "epoch": 9.25,
+      "learning_rate": 9.634200733262093e-05,
+      "loss": 2.3312,
+      "step": 68480
+    },
+    {
+      "epoch": 9.25,
+      "learning_rate": 9.630238474055654e-05,
+      "loss": 2.3396,
+      "step": 68490
+    },
+    {
+      "epoch": 9.25,
+      "learning_rate": 9.626276644554483e-05,
+      "loss": 2.3296,
+      "step": 68500
+    },
+    {
+      "epoch": 9.25,
+      "learning_rate": 9.622315245075629e-05,
+      "loss": 2.3463,
+      "step": 68510
+    },
+    {
+      "epoch": 9.25,
+      "learning_rate": 9.618354275936094e-05,
+      "loss": 2.3321,
+      "step": 68520
+    },
+    {
+      "epoch": 9.26,
+      "learning_rate": 9.614393737452855e-05,
+      "loss": 2.3488,
+      "step": 68530
+    },
+    {
+      "epoch": 9.26,
+      "learning_rate": 9.610433629942833e-05,
+      "loss": 2.3242,
+      "step": 68540
+    },
+    {
+      "epoch": 9.26,
+      "learning_rate": 9.606473953722931e-05,
+      "loss": 2.3349,
+      "step": 68550
+    },
+    {
+      "epoch": 9.26,
+      "learning_rate": 9.60251470911002e-05,
+      "loss": 2.3217,
+      "step": 68560
+    },
+    {
+      "epoch": 9.26,
+      "learning_rate": 9.598555896420936e-05,
+      "loss": 2.3729,
+      "step": 68570
+    },
+    {
+      "epoch": 9.26,
+      "learning_rate": 9.594597515972462e-05,
+      "loss": 2.3303,
+      "step": 68580
+    },
+    {
+      "epoch": 9.26,
+      "learning_rate": 9.590639568081369e-05,
+      "loss": 2.3319,
+      "step": 68590
+    },
+    {
+      "epoch": 9.27,
+      "learning_rate": 9.586682053064383e-05,
+      "loss": 2.3337,
+      "step": 68600
+    },
+    {
+      "epoch": 9.27,
+      "learning_rate": 9.582724971238199e-05,
+      "loss": 2.3069,
+      "step": 68610
+    },
+    {
+      "epoch": 9.27,
+      "learning_rate": 9.578768322919468e-05,
+      "loss": 2.3141,
+      "step": 68620
+    },
+    {
+      "epoch": 9.27,
+      "learning_rate": 9.57481210842482e-05,
+      "loss": 2.3114,
+      "step": 68630
+    },
+    {
+      "epoch": 9.27,
+      "learning_rate": 9.570856328070841e-05,
+      "loss": 2.3228,
+      "step": 68640
+    },
+    {
+      "epoch": 9.27,
+      "learning_rate": 9.56690098217409e-05,
+      "loss": 2.3154,
+      "step": 68650
+    },
+    {
+      "epoch": 9.27,
+      "learning_rate": 9.562946071051079e-05,
+      "loss": 2.3135,
+      "step": 68660
+    },
+    {
+      "epoch": 9.27,
+      "learning_rate": 9.558991595018295e-05,
+      "loss": 2.3163,
+      "step": 68670
+    },
+    {
+      "epoch": 9.28,
+      "learning_rate": 9.55503755439219e-05,
+      "loss": 2.3264,
+      "step": 68680
+    },
+    {
+      "epoch": 9.28,
+      "learning_rate": 9.551083949489183e-05,
+      "loss": 2.335,
+      "step": 68690
+    },
+    {
+      "epoch": 9.28,
+      "learning_rate": 9.547130780625639e-05,
+      "loss": 2.3277,
+      "step": 68700
+    },
+    {
+      "epoch": 9.28,
+      "learning_rate": 9.543178048117914e-05,
+      "loss": 2.2883,
+      "step": 68710
+    },
+    {
+      "epoch": 9.28,
+      "learning_rate": 9.539225752282319e-05,
+      "loss": 2.3468,
+      "step": 68720
+    },
+    {
+      "epoch": 9.28,
+      "learning_rate": 9.535273893435131e-05,
+      "loss": 2.3042,
+      "step": 68730
+    },
+    {
+      "epoch": 9.28,
+      "learning_rate": 9.531322471892579e-05,
+      "loss": 2.308,
+      "step": 68740
+    },
+    {
+      "epoch": 9.29,
+      "learning_rate": 9.527371487970877e-05,
+      "loss": 2.3368,
+      "step": 68750
+    },
+    {
+      "epoch": 9.29,
+      "learning_rate": 9.523420941986187e-05,
+      "loss": 2.2988,
+      "step": 68760
+    },
+    {
+      "epoch": 9.29,
+      "learning_rate": 9.519470834254664e-05,
+      "loss": 2.33,
+      "step": 68770
+    },
+    {
+      "epoch": 9.29,
+      "learning_rate": 9.515521165092386e-05,
+      "loss": 2.3532,
+      "step": 68780
+    },
+    {
+      "epoch": 9.29,
+      "learning_rate": 9.511571934815427e-05,
+      "loss": 2.3027,
+      "step": 68790
+    },
+    {
+      "epoch": 9.29,
+      "learning_rate": 9.507623143739816e-05,
+      "loss": 2.315,
+      "step": 68800
+    },
+    {
+      "epoch": 9.29,
+      "learning_rate": 9.503674792181551e-05,
+      "loss": 2.315,
+      "step": 68810
+    },
+    {
+      "epoch": 9.29,
+      "learning_rate": 9.499726880456587e-05,
+      "loss": 2.3514,
+      "step": 68820
+    },
+    {
+      "epoch": 9.3,
+      "learning_rate": 9.495779408880848e-05,
+      "loss": 2.308,
+      "step": 68830
+    },
+    {
+      "epoch": 9.3,
+      "learning_rate": 9.491832377770227e-05,
+      "loss": 2.3033,
+      "step": 68840
+    },
+    {
+      "epoch": 9.3,
+      "learning_rate": 9.48788578744058e-05,
+      "loss": 2.3131,
+      "step": 68850
+    },
+    {
+      "epoch": 9.3,
+      "learning_rate": 9.483939638207716e-05,
+      "loss": 2.3533,
+      "step": 68860
+    },
+    {
+      "epoch": 9.3,
+      "learning_rate": 9.479993930387425e-05,
+      "loss": 2.3156,
+      "step": 68870
+    },
+    {
+      "epoch": 9.3,
+      "learning_rate": 9.476048664295455e-05,
+      "loss": 2.3362,
+      "step": 68880
+    },
+    {
+      "epoch": 9.3,
+      "learning_rate": 9.472103840247524e-05,
+      "loss": 2.3109,
+      "step": 68890
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 9.468159458559294e-05,
+      "loss": 2.3415,
+      "step": 68900
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 9.464215519546419e-05,
+      "loss": 2.31,
+      "step": 68910
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 9.4602720235245e-05,
+      "loss": 2.3433,
+      "step": 68920
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 9.456328970809118e-05,
+      "loss": 2.3171,
+      "step": 68930
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 9.452386361715795e-05,
+      "loss": 2.3198,
+      "step": 68940
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 9.448444196560037e-05,
+      "loss": 2.3121,
+      "step": 68950
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 9.444502475657306e-05,
+      "loss": 2.3249,
+      "step": 68960
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 9.440561199323041e-05,
+      "loss": 2.3408,
+      "step": 68970
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 9.436620367872623e-05,
+      "loss": 2.3187,
+      "step": 68980
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 9.432679981621413e-05,
+      "loss": 2.3065,
+      "step": 68990
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 9.428740040884736e-05,
+      "loss": 2.324,
+      "step": 69000
+    },
+    {
+      "epoch": 9.32,
+      "eval_loss": 2.458608388900757,
+      "eval_runtime": 1269.1836,
+      "eval_samples_per_second": 59.987,
+      "eval_steps_per_second": 4.999,
+      "step": 69000
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 9.42480054597788e-05,
+      "loss": 2.3242,
+      "step": 69010
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 9.420861497216091e-05,
+      "loss": 2.3136,
+      "step": 69020
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 9.416922894914584e-05,
+      "loss": 2.3549,
+      "step": 69030
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 9.412984739388543e-05,
+      "loss": 2.3234,
+      "step": 69040
+    },
+    {
+      "epoch": 9.33,
+      "learning_rate": 9.409047030953111e-05,
+      "loss": 2.3454,
+      "step": 69050
+    },
+    {
+      "epoch": 9.33,
+      "learning_rate": 9.40510976992339e-05,
+      "loss": 2.3244,
+      "step": 69060
+    },
+    {
+      "epoch": 9.33,
+      "learning_rate": 9.401172956614456e-05,
+      "loss": 2.3477,
+      "step": 69070
+    },
+    {
+      "epoch": 9.33,
+      "learning_rate": 9.397236591341349e-05,
+      "loss": 2.3063,
+      "step": 69080
+    },
+    {
+      "epoch": 9.33,
+      "learning_rate": 9.393300674419067e-05,
+      "loss": 2.3015,
+      "step": 69090
+    },
+    {
+      "epoch": 9.33,
+      "learning_rate": 9.389365206162566e-05,
+      "loss": 2.3391,
+      "step": 69100
+    },
+    {
+      "epoch": 9.33,
+      "learning_rate": 9.385430186886786e-05,
+      "loss": 2.3336,
+      "step": 69110
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 9.381495616906616e-05,
+      "loss": 2.3343,
+      "step": 69120
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 9.377561496536914e-05,
+      "loss": 2.3251,
+      "step": 69130
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 9.373627826092497e-05,
+      "loss": 2.3175,
+      "step": 69140
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 9.369694605888151e-05,
+      "loss": 2.3035,
+      "step": 69150
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 9.365761836238621e-05,
+      "loss": 2.3307,
+      "step": 69160
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 9.361829517458636e-05,
+      "loss": 2.3145,
+      "step": 69170
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 9.357897649862852e-05,
+      "loss": 2.3139,
+      "step": 69180
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 9.353966233765919e-05,
+      "loss": 2.346,
+      "step": 69190
+    },
+    {
+      "epoch": 9.35,
+      "learning_rate": 9.35003526948244e-05,
+      "loss": 2.3071,
+      "step": 69200
+    },
+    {
+      "epoch": 9.35,
+      "learning_rate": 9.346104757326987e-05,
+      "loss": 2.3352,
+      "step": 69210
+    },
+    {
+      "epoch": 9.35,
+      "learning_rate": 9.342174697614084e-05,
+      "loss": 2.3201,
+      "step": 69220
+    },
+    {
+      "epoch": 9.35,
+      "learning_rate": 9.338245090658236e-05,
+      "loss": 2.3216,
+      "step": 69230
+    },
+    {
+      "epoch": 9.35,
+      "learning_rate": 9.334315936773893e-05,
+      "loss": 2.3065,
+      "step": 69240
+    },
+    {
+      "epoch": 9.35,
+      "learning_rate": 9.330387236275488e-05,
+      "loss": 2.307,
+      "step": 69250
+    },
+    {
+      "epoch": 9.35,
+      "learning_rate": 9.3264589894774e-05,
+      "loss": 2.3361,
+      "step": 69260
+    },
+    {
+      "epoch": 9.36,
+      "learning_rate": 9.322531196693983e-05,
+      "loss": 2.3054,
+      "step": 69270
+    },
+    {
+      "epoch": 9.36,
+      "learning_rate": 9.318603858239551e-05,
+      "loss": 2.3097,
+      "step": 69280
+    },
+    {
+      "epoch": 9.36,
+      "learning_rate": 9.314676974428392e-05,
+      "loss": 2.3278,
+      "step": 69290
+    },
+    {
+      "epoch": 9.36,
+      "learning_rate": 9.310750545574723e-05,
+      "loss": 2.3351,
+      "step": 69300
+    },
+    {
+      "epoch": 9.36,
+      "learning_rate": 9.30682457199277e-05,
+      "loss": 2.344,
+      "step": 69310
+    },
+    {
+      "epoch": 9.36,
+      "learning_rate": 9.302899053996698e-05,
+      "loss": 2.3426,
+      "step": 69320
+    },
+    {
+      "epoch": 9.36,
+      "learning_rate": 9.298973991900642e-05,
+      "loss": 2.3465,
+      "step": 69330
+    },
+    {
+      "epoch": 9.37,
+      "learning_rate": 9.295049386018686e-05,
+      "loss": 2.313,
+      "step": 69340
+    },
+    {
+      "epoch": 9.37,
+      "learning_rate": 9.291125236664899e-05,
+      "loss": 2.2986,
+      "step": 69350
+    },
+    {
+      "epoch": 9.37,
+      "learning_rate": 9.287201544153298e-05,
+      "loss": 2.3205,
+      "step": 69360
+    },
+    {
+      "epoch": 9.37,
+      "learning_rate": 9.28327830879788e-05,
+      "loss": 2.3039,
+      "step": 69370
+    },
+    {
+      "epoch": 9.37,
+      "learning_rate": 9.279355530912582e-05,
+      "loss": 2.3362,
+      "step": 69380
+    },
+    {
+      "epoch": 9.37,
+      "learning_rate": 9.275433210811324e-05,
+      "loss": 2.3084,
+      "step": 69390
+    },
+    {
+      "epoch": 9.37,
+      "learning_rate": 9.271511348807978e-05,
+      "loss": 2.3303,
+      "step": 69400
+    },
+    {
+      "epoch": 9.37,
+      "learning_rate": 9.267589945216393e-05,
+      "loss": 2.3196,
+      "step": 69410
+    },
+    {
+      "epoch": 9.38,
+      "learning_rate": 9.263669000350362e-05,
+      "loss": 2.3229,
+      "step": 69420
+    },
+    {
+      "epoch": 9.38,
+      "learning_rate": 9.259748514523653e-05,
+      "loss": 2.2878,
+      "step": 69430
+    },
+    {
+      "epoch": 9.38,
+      "learning_rate": 9.25582848805e-05,
+      "loss": 2.316,
+      "step": 69440
+    },
+    {
+      "epoch": 9.38,
+      "learning_rate": 9.251908921243093e-05,
+      "loss": 2.3384,
+      "step": 69450
+    },
+    {
+      "epoch": 9.38,
+      "learning_rate": 9.247989814416593e-05,
+      "loss": 2.3094,
+      "step": 69460
+    },
+    {
+      "epoch": 9.38,
+      "learning_rate": 9.24407116788411e-05,
+      "loss": 2.3308,
+      "step": 69470
+    },
+    {
+      "epoch": 9.38,
+      "learning_rate": 9.240152981959234e-05,
+      "loss": 2.3529,
+      "step": 69480
+    },
+    {
+      "epoch": 9.39,
+      "learning_rate": 9.236235256955507e-05,
+      "loss": 2.3226,
+      "step": 69490
+    },
+    {
+      "epoch": 9.39,
+      "learning_rate": 9.232317993186444e-05,
+      "loss": 2.335,
+      "step": 69500
+    },
+    {
+      "epoch": 9.39,
+      "learning_rate": 9.228401190965508e-05,
+      "loss": 2.3262,
+      "step": 69510
+    },
+    {
+      "epoch": 9.39,
+      "learning_rate": 9.224484850606136e-05,
+      "loss": 2.3487,
+      "step": 69520
+    },
+    {
+      "epoch": 9.39,
+      "learning_rate": 9.220568972421729e-05,
+      "loss": 2.2983,
+      "step": 69530
+    },
+    {
+      "epoch": 9.39,
+      "learning_rate": 9.216653556725652e-05,
+      "loss": 2.3295,
+      "step": 69540
+    },
+    {
+      "epoch": 9.39,
+      "learning_rate": 9.212738603831216e-05,
+      "loss": 2.3028,
+      "step": 69550
+    },
+    {
+      "epoch": 9.39,
+      "learning_rate": 9.208824114051718e-05,
+      "loss": 2.3422,
+      "step": 69560
+    },
+    {
+      "epoch": 9.4,
+      "learning_rate": 9.204910087700403e-05,
+      "loss": 2.3227,
+      "step": 69570
+    },
+    {
+      "epoch": 9.4,
+      "learning_rate": 9.200996525090493e-05,
+      "loss": 2.3214,
+      "step": 69580
+    },
+    {
+      "epoch": 9.4,
+      "learning_rate": 9.197083426535152e-05,
+      "loss": 2.3138,
+      "step": 69590
+    },
+    {
+      "epoch": 9.4,
+      "learning_rate": 9.193170792347521e-05,
+      "loss": 2.3419,
+      "step": 69600
+    },
+    {
+      "epoch": 9.4,
+      "learning_rate": 9.189258622840707e-05,
+      "loss": 2.2931,
+      "step": 69610
+    },
+    {
+      "epoch": 9.4,
+      "learning_rate": 9.18534691832777e-05,
+      "loss": 2.3191,
+      "step": 69620
+    },
+    {
+      "epoch": 9.4,
+      "learning_rate": 9.181435679121736e-05,
+      "loss": 2.3183,
+      "step": 69630
+    },
+    {
+      "epoch": 9.41,
+      "learning_rate": 9.177524905535596e-05,
+      "loss": 2.3209,
+      "step": 69640
+    },
+    {
+      "epoch": 9.41,
+      "learning_rate": 9.173614597882303e-05,
+      "loss": 2.305,
+      "step": 69650
+    },
+    {
+      "epoch": 9.41,
+      "learning_rate": 9.169704756474775e-05,
+      "loss": 2.3379,
+      "step": 69660
+    },
+    {
+      "epoch": 9.41,
+      "learning_rate": 9.165795381625884e-05,
+      "loss": 2.3386,
+      "step": 69670
+    },
+    {
+      "epoch": 9.41,
+      "learning_rate": 9.161886473648473e-05,
+      "loss": 2.338,
+      "step": 69680
+    },
+    {
+      "epoch": 9.41,
+      "learning_rate": 9.157978032855347e-05,
+      "loss": 2.3081,
+      "step": 69690
+    },
+    {
+      "epoch": 9.41,
+      "learning_rate": 9.154070059559274e-05,
+      "loss": 2.3197,
+      "step": 69700
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 9.15016255407297e-05,
+      "loss": 2.3385,
+      "step": 69710
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 9.146255516709139e-05,
+      "loss": 2.3233,
+      "step": 69720
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 9.142348947780429e-05,
+      "loss": 2.3355,
+      "step": 69730
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 9.138442847599463e-05,
+      "loss": 2.339,
+      "step": 69740
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 9.134537216478808e-05,
+      "loss": 2.3234,
+      "step": 69750
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 9.13063205473101e-05,
+      "loss": 2.3408,
+      "step": 69760
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 9.12672736266857e-05,
+      "loss": 2.3325,
+      "step": 69770
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 9.122823140603965e-05,
+      "loss": 2.3183,
+      "step": 69780
+    },
+    {
+      "epoch": 9.43,
+      "learning_rate": 9.11891938884961e-05,
+      "loss": 2.3217,
+      "step": 69790
+    },
+    {
+      "epoch": 9.43,
+      "learning_rate": 9.115016107717901e-05,
+      "loss": 2.3456,
+      "step": 69800
+    },
+    {
+      "epoch": 9.43,
+      "learning_rate": 9.111113297521189e-05,
+      "loss": 2.3273,
+      "step": 69810
+    },
+    {
+      "epoch": 9.43,
+      "learning_rate": 9.107210958571794e-05,
+      "loss": 2.2984,
+      "step": 69820
+    },
+    {
+      "epoch": 9.43,
+      "learning_rate": 9.103309091181985e-05,
+      "loss": 2.3115,
+      "step": 69830
+    },
+    {
+      "epoch": 9.43,
+      "learning_rate": 9.099407695664008e-05,
+      "loss": 2.332,
+      "step": 69840
+    },
+    {
+      "epoch": 9.43,
+      "learning_rate": 9.095506772330064e-05,
+      "loss": 2.3563,
+      "step": 69850
+    },
+    {
+      "epoch": 9.44,
+      "learning_rate": 9.091606321492317e-05,
+      "loss": 2.317,
+      "step": 69860
+    },
+    {
+      "epoch": 9.44,
+      "learning_rate": 9.087706343462893e-05,
+      "loss": 2.3531,
+      "step": 69870
+    },
+    {
+      "epoch": 9.44,
+      "learning_rate": 9.08380683855388e-05,
+      "loss": 2.3175,
+      "step": 69880
+    },
+    {
+      "epoch": 9.44,
+      "learning_rate": 9.079907807077331e-05,
+      "loss": 2.3307,
+      "step": 69890
+    },
+    {
+      "epoch": 9.44,
+      "learning_rate": 9.076009249345261e-05,
+      "loss": 2.3144,
+      "step": 69900
+    },
+    {
+      "epoch": 9.44,
+      "learning_rate": 9.072111165669634e-05,
+      "loss": 2.2892,
+      "step": 69910
+    },
+    {
+      "epoch": 9.44,
+      "learning_rate": 9.068213556362398e-05,
+      "loss": 2.3341,
+      "step": 69920
+    },
+    {
+      "epoch": 9.44,
+      "learning_rate": 9.064316421735446e-05,
+      "loss": 2.3626,
+      "step": 69930
+    },
+    {
+      "epoch": 9.45,
+      "learning_rate": 9.060419762100648e-05,
+      "loss": 2.3031,
+      "step": 69940
+    },
+    {
+      "epoch": 9.45,
+      "learning_rate": 9.056523577769814e-05,
+      "loss": 2.3293,
+      "step": 69950
+    },
+    {
+      "epoch": 9.45,
+      "learning_rate": 9.052627869054734e-05,
+      "loss": 2.3267,
+      "step": 69960
+    },
+    {
+      "epoch": 9.45,
+      "learning_rate": 9.048732636267155e-05,
+      "loss": 2.31,
+      "step": 69970
+    },
+    {
+      "epoch": 9.45,
+      "learning_rate": 9.044837879718796e-05,
+      "loss": 2.3295,
+      "step": 69980
+    },
+    {
+      "epoch": 9.45,
+      "learning_rate": 9.040943599721313e-05,
+      "loss": 2.3445,
+      "step": 69990
+    },
+    {
+      "epoch": 9.45,
+      "learning_rate": 9.037049796586341e-05,
+      "loss": 2.3394,
+      "step": 70000
+    },
+    {
+      "epoch": 9.45,
+      "eval_loss": 2.4543914794921875,
+      "eval_runtime": 1269.1098,
+      "eval_samples_per_second": 59.991,
+      "eval_steps_per_second": 5.0,
+      "step": 70000
+    },
+    {
+      "epoch": 9.46,
+      "learning_rate": 9.033156470625479e-05,
+      "loss": 2.3182,
+      "step": 70010
+    },
+    {
+      "epoch": 9.46,
+      "learning_rate": 9.029263622150287e-05,
+      "loss": 2.3422,
+      "step": 70020
+    },
+    {
+      "epoch": 9.46,
+      "learning_rate": 9.025371251472274e-05,
+      "loss": 2.3575,
+      "step": 70030
+    },
+    {
+      "epoch": 9.46,
+      "learning_rate": 9.021479358902924e-05,
+      "loss": 2.3077,
+      "step": 70040
+    },
+    {
+      "epoch": 9.46,
+      "learning_rate": 9.017587944753678e-05,
+      "loss": 2.3227,
+      "step": 70050
+    },
+    {
+      "epoch": 9.46,
+      "learning_rate": 9.013697009335941e-05,
+      "loss": 2.3383,
+      "step": 70060
+    },
+    {
+      "epoch": 9.46,
+      "learning_rate": 9.009806552961075e-05,
+      "loss": 2.333,
+      "step": 70070
+    },
+    {
+      "epoch": 9.46,
+      "learning_rate": 9.005916575940407e-05,
+      "loss": 2.3024,
+      "step": 70080
+    },
+    {
+      "epoch": 9.47,
+      "learning_rate": 9.002027078585227e-05,
+      "loss": 2.3309,
+      "step": 70090
+    },
+    {
+      "epoch": 9.47,
+      "learning_rate": 8.998138061206791e-05,
+      "loss": 2.3202,
+      "step": 70100
+    },
+    {
+      "epoch": 9.47,
+      "learning_rate": 8.994249524116294e-05,
+      "loss": 2.3323,
+      "step": 70110
+    },
+    {
+      "epoch": 9.47,
+      "learning_rate": 8.990361467624921e-05,
+      "loss": 2.3041,
+      "step": 70120
+    },
+    {
+      "epoch": 9.47,
+      "learning_rate": 8.986473892043805e-05,
+      "loss": 2.3313,
+      "step": 70130
+    },
+    {
+      "epoch": 9.47,
+      "learning_rate": 8.982586797684048e-05,
+      "loss": 2.3133,
+      "step": 70140
+    },
+    {
+      "epoch": 9.47,
+      "learning_rate": 8.978700184856694e-05,
+      "loss": 2.3407,
+      "step": 70150
+    },
+    {
+      "epoch": 9.48,
+      "learning_rate": 8.974814053872769e-05,
+      "loss": 2.3172,
+      "step": 70160
+    },
+    {
+      "epoch": 9.48,
+      "learning_rate": 8.970928405043251e-05,
+      "loss": 2.3258,
+      "step": 70170
+    },
+    {
+      "epoch": 9.48,
+      "learning_rate": 8.967043238679091e-05,
+      "loss": 2.3141,
+      "step": 70180
+    },
+    {
+      "epoch": 9.48,
+      "learning_rate": 8.96315855509118e-05,
+      "loss": 2.3215,
+      "step": 70190
+    },
+    {
+      "epoch": 9.48,
+      "learning_rate": 8.959274354590392e-05,
+      "loss": 2.3508,
+      "step": 70200
+    },
+    {
+      "epoch": 9.48,
+      "learning_rate": 8.955390637487546e-05,
+      "loss": 2.3328,
+      "step": 70210
+    },
+    {
+      "epoch": 9.48,
+      "learning_rate": 8.951507404093434e-05,
+      "loss": 2.3137,
+      "step": 70220
+    },
+    {
+      "epoch": 9.49,
+      "learning_rate": 8.947624654718803e-05,
+      "loss": 2.2952,
+      "step": 70230
+    },
+    {
+      "epoch": 9.49,
+      "learning_rate": 8.943742389674363e-05,
+      "loss": 2.3256,
+      "step": 70240
+    },
+    {
+      "epoch": 9.49,
+      "learning_rate": 8.939860609270783e-05,
+      "loss": 2.3407,
+      "step": 70250
+    },
+    {
+      "epoch": 9.49,
+      "learning_rate": 8.935979313818701e-05,
+      "loss": 2.2986,
+      "step": 70260
+    },
+    {
+      "epoch": 9.49,
+      "learning_rate": 8.932098503628705e-05,
+      "loss": 2.3394,
+      "step": 70270
+    },
+    {
+      "epoch": 9.49,
+      "learning_rate": 8.928218179011351e-05,
+      "loss": 2.3699,
+      "step": 70280
+    },
+    {
+      "epoch": 9.49,
+      "learning_rate": 8.924338340277155e-05,
+      "loss": 2.3234,
+      "step": 70290
+    },
+    {
+      "epoch": 9.49,
+      "learning_rate": 8.920458987736601e-05,
+      "loss": 2.329,
+      "step": 70300
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 8.916580121700112e-05,
+      "loss": 2.3334,
+      "step": 70310
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 8.912701742478097e-05,
+      "loss": 2.3382,
+      "step": 70320
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 8.908823850380916e-05,
+      "loss": 2.3237,
+      "step": 70330
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 8.904946445718894e-05,
+      "loss": 2.3514,
+      "step": 70340
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 8.901069528802301e-05,
+      "loss": 2.3058,
+      "step": 70350
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 8.89719309994139e-05,
+      "loss": 2.3269,
+      "step": 70360
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 8.893317159446358e-05,
+      "loss": 2.3193,
+      "step": 70370
+    },
+    {
+      "epoch": 9.51,
+      "learning_rate": 8.889441707627383e-05,
+      "loss": 2.3449,
+      "step": 70380
+    },
+    {
+      "epoch": 9.51,
+      "learning_rate": 8.885566744794576e-05,
+      "loss": 2.3272,
+      "step": 70390
+    },
+    {
+      "epoch": 9.51,
+      "learning_rate": 8.88169227125803e-05,
+      "loss": 2.3205,
+      "step": 70400
+    },
+    {
+      "epoch": 9.51,
+      "learning_rate": 8.877818287327793e-05,
+      "loss": 2.3173,
+      "step": 70410
+    },
+    {
+      "epoch": 9.51,
+      "learning_rate": 8.873944793313875e-05,
+      "loss": 2.3537,
+      "step": 70420
+    },
+    {
+      "epoch": 9.51,
+      "learning_rate": 8.870071789526242e-05,
+      "loss": 2.3131,
+      "step": 70430
+    },
+    {
+      "epoch": 9.51,
+      "learning_rate": 8.866199276274824e-05,
+      "loss": 2.3306,
+      "step": 70440
+    },
+    {
+      "epoch": 9.51,
+      "learning_rate": 8.862327253869513e-05,
+      "loss": 2.3514,
+      "step": 70450
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 8.858455722620164e-05,
+      "loss": 2.3121,
+      "step": 70460
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 8.85458468283658e-05,
+      "loss": 2.3258,
+      "step": 70470
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 8.850714134828542e-05,
+      "loss": 2.3298,
+      "step": 70480
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 8.84684407890578e-05,
+      "loss": 2.3286,
+      "step": 70490
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 8.842974515377995e-05,
+      "loss": 2.3207,
+      "step": 70500
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 8.839105444554826e-05,
+      "loss": 2.3344,
+      "step": 70510
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 8.835236866745903e-05,
+      "loss": 2.3464,
+      "step": 70520
+    },
+    {
+      "epoch": 9.53,
+      "learning_rate": 8.831368782260794e-05,
+      "loss": 2.3397,
+      "step": 70530
+    },
+    {
+      "epoch": 9.53,
+      "learning_rate": 8.827501191409046e-05,
+      "loss": 2.3446,
+      "step": 70540
+    },
+    {
+      "epoch": 9.53,
+      "learning_rate": 8.82363409450014e-05,
+      "loss": 2.3302,
+      "step": 70550
+    },
+    {
+      "epoch": 9.53,
+      "learning_rate": 8.819767491843544e-05,
+      "loss": 2.3161,
+      "step": 70560
+    },
+    {
+      "epoch": 9.53,
+      "learning_rate": 8.81590138374867e-05,
+      "loss": 2.3395,
+      "step": 70570
+    },
+    {
+      "epoch": 9.53,
+      "learning_rate": 8.812035770524905e-05,
+      "loss": 2.3591,
+      "step": 70580
+    },
+    {
+      "epoch": 9.53,
+      "learning_rate": 8.808170652481577e-05,
+      "loss": 2.3224,
+      "step": 70590
+    },
+    {
+      "epoch": 9.54,
+      "learning_rate": 8.804306029927992e-05,
+      "loss": 2.3132,
+      "step": 70600
+    },
+    {
+      "epoch": 9.54,
+      "learning_rate": 8.800441903173403e-05,
+      "loss": 2.3382,
+      "step": 70610
+    },
+    {
+      "epoch": 9.54,
+      "learning_rate": 8.79657827252704e-05,
+      "loss": 2.3273,
+      "step": 70620
+    },
+    {
+      "epoch": 9.54,
+      "learning_rate": 8.79271513829807e-05,
+      "loss": 2.3078,
+      "step": 70630
+    },
+    {
+      "epoch": 9.54,
+      "learning_rate": 8.788852500795643e-05,
+      "loss": 2.2964,
+      "step": 70640
+    },
+    {
+      "epoch": 9.54,
+      "learning_rate": 8.784990360328854e-05,
+      "loss": 2.3204,
+      "step": 70650
+    },
+    {
+      "epoch": 9.54,
+      "learning_rate": 8.781128717206772e-05,
+      "loss": 2.352,
+      "step": 70660
+    },
+    {
+      "epoch": 9.54,
+      "learning_rate": 8.777267571738405e-05,
+      "loss": 2.3287,
+      "step": 70670
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 8.773406924232742e-05,
+      "loss": 2.3417,
+      "step": 70680
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 8.769546774998724e-05,
+      "loss": 2.3419,
+      "step": 70690
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 8.765687124345255e-05,
+      "loss": 2.3211,
+      "step": 70700
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 8.76182797258119e-05,
+      "loss": 2.3009,
+      "step": 70710
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 8.757969320015348e-05,
+      "loss": 2.3271,
+      "step": 70720
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 8.754111166956522e-05,
+      "loss": 2.3154,
+      "step": 70730
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 8.750253513713447e-05,
+      "loss": 2.3081,
+      "step": 70740
+    },
+    {
+      "epoch": 9.56,
+      "learning_rate": 8.746396360594826e-05,
+      "loss": 2.3252,
+      "step": 70750
+    },
+    {
+      "epoch": 9.56,
+      "learning_rate": 8.742539707909313e-05,
+      "loss": 2.3703,
+      "step": 70760
+    },
+    {
+      "epoch": 9.56,
+      "learning_rate": 8.738683555965543e-05,
+      "loss": 2.2995,
+      "step": 70770
+    },
+    {
+      "epoch": 9.56,
+      "learning_rate": 8.734827905072092e-05,
+      "loss": 2.3247,
+      "step": 70780
+    },
+    {
+      "epoch": 9.56,
+      "learning_rate": 8.730972755537491e-05,
+      "loss": 2.3403,
+      "step": 70790
+    },
+    {
+      "epoch": 9.56,
+      "learning_rate": 8.727118107670258e-05,
+      "loss": 2.3188,
+      "step": 70800
+    },
+    {
+      "epoch": 9.56,
+      "learning_rate": 8.723263961778839e-05,
+      "loss": 2.3202,
+      "step": 70810
+    },
+    {
+      "epoch": 9.56,
+      "learning_rate": 8.719410318171676e-05,
+      "loss": 2.3129,
+      "step": 70820
+    },
+    {
+      "epoch": 9.57,
+      "learning_rate": 8.715557177157124e-05,
+      "loss": 2.3438,
+      "step": 70830
+    },
+    {
+      "epoch": 9.57,
+      "learning_rate": 8.711704539043541e-05,
+      "loss": 2.346,
+      "step": 70840
+    },
+    {
+      "epoch": 9.57,
+      "learning_rate": 8.707852404139214e-05,
+      "loss": 2.3373,
+      "step": 70850
+    },
+    {
+      "epoch": 9.57,
+      "learning_rate": 8.704000772752418e-05,
+      "loss": 2.3394,
+      "step": 70860
+    },
+    {
+      "epoch": 9.57,
+      "learning_rate": 8.700149645191365e-05,
+      "loss": 2.3205,
+      "step": 70870
+    },
+    {
+      "epoch": 9.57,
+      "learning_rate": 8.69629902176423e-05,
+      "loss": 2.3391,
+      "step": 70880
+    },
+    {
+      "epoch": 9.57,
+      "learning_rate": 8.692448902779163e-05,
+      "loss": 2.3416,
+      "step": 70890
+    },
+    {
+      "epoch": 9.58,
+      "learning_rate": 8.688599288544256e-05,
+      "loss": 2.3124,
+      "step": 70900
+    },
+    {
+      "epoch": 9.58,
+      "learning_rate": 8.684750179367561e-05,
+      "loss": 2.3494,
+      "step": 70910
+    },
+    {
+      "epoch": 9.58,
+      "learning_rate": 8.68090157555711e-05,
+      "loss": 2.3271,
+      "step": 70920
+    },
+    {
+      "epoch": 9.58,
+      "learning_rate": 8.677053477420868e-05,
+      "loss": 2.3397,
+      "step": 70930
+    },
+    {
+      "epoch": 9.58,
+      "learning_rate": 8.673205885266782e-05,
+      "loss": 2.3117,
+      "step": 70940
+    },
+    {
+      "epoch": 9.58,
+      "learning_rate": 8.669358799402743e-05,
+      "loss": 2.3373,
+      "step": 70950
+    },
+    {
+      "epoch": 9.58,
+      "learning_rate": 8.665512220136601e-05,
+      "loss": 2.3492,
+      "step": 70960
+    },
+    {
+      "epoch": 9.59,
+      "learning_rate": 8.661666147776185e-05,
+      "loss": 2.3136,
+      "step": 70970
+    },
+    {
+      "epoch": 9.59,
+      "learning_rate": 8.657820582629265e-05,
+      "loss": 2.3141,
+      "step": 70980
+    },
+    {
+      "epoch": 9.59,
+      "learning_rate": 8.653975525003564e-05,
+      "loss": 2.3282,
+      "step": 70990
+    },
+    {
+      "epoch": 9.59,
+      "learning_rate": 8.650130975206793e-05,
+      "loss": 2.3457,
+      "step": 71000
+    },
+    {
+      "epoch": 9.59,
+      "eval_loss": 2.4499146938323975,
+      "eval_runtime": 1269.4328,
+      "eval_samples_per_second": 59.976,
+      "eval_steps_per_second": 4.998,
+      "step": 71000
+    },
+    {
+      "epoch": 9.59,
+      "learning_rate": 8.646286933546589e-05,
+      "loss": 2.2982,
+      "step": 71010
+    },
+    {
+      "epoch": 9.59,
+      "learning_rate": 8.642443400330585e-05,
+      "loss": 2.3313,
+      "step": 71020
+    },
+    {
+      "epoch": 9.59,
+      "learning_rate": 8.638600375866327e-05,
+      "loss": 2.3166,
+      "step": 71030
+    },
+    {
+      "epoch": 9.59,
+      "learning_rate": 8.634757860461366e-05,
+      "loss": 2.3371,
+      "step": 71040
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 8.630915854423178e-05,
+      "loss": 2.3362,
+      "step": 71050
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 8.627074358059225e-05,
+      "loss": 2.3279,
+      "step": 71060
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 8.623233371676911e-05,
+      "loss": 2.3305,
+      "step": 71070
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 8.619392895583598e-05,
+      "loss": 2.3236,
+      "step": 71080
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 8.615552930086623e-05,
+      "loss": 2.3074,
+      "step": 71090
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 8.611713475493268e-05,
+      "loss": 2.3289,
+      "step": 71100
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 8.607874532110769e-05,
+      "loss": 2.3121,
+      "step": 71110
+    },
+    {
+      "epoch": 9.61,
+      "learning_rate": 8.604036100246346e-05,
+      "loss": 2.3154,
+      "step": 71120
+    },
+    {
+      "epoch": 9.61,
+      "learning_rate": 8.60019818020715e-05,
+      "loss": 2.3401,
+      "step": 71130
+    },
+    {
+      "epoch": 9.61,
+      "learning_rate": 8.596360772300316e-05,
+      "loss": 2.3392,
+      "step": 71140
+    },
+    {
+      "epoch": 9.61,
+      "learning_rate": 8.592523876832917e-05,
+      "loss": 2.3236,
+      "step": 71150
+    },
+    {
+      "epoch": 9.61,
+      "learning_rate": 8.588687494111989e-05,
+      "loss": 2.2912,
+      "step": 71160
+    },
+    {
+      "epoch": 9.61,
+      "learning_rate": 8.584851624444542e-05,
+      "loss": 2.3346,
+      "step": 71170
+    },
+    {
+      "epoch": 9.61,
+      "learning_rate": 8.581016268137531e-05,
+      "loss": 2.3203,
+      "step": 71180
+    },
+    {
+      "epoch": 9.61,
+      "learning_rate": 8.577181425497866e-05,
+      "loss": 2.3119,
+      "step": 71190
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 8.573347096832433e-05,
+      "loss": 2.3134,
+      "step": 71200
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 8.569513282448059e-05,
+      "loss": 2.335,
+      "step": 71210
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 8.565679982651555e-05,
+      "loss": 2.3435,
+      "step": 71220
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 8.561847197749648e-05,
+      "loss": 2.3026,
+      "step": 71230
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 8.558014928049068e-05,
+      "loss": 2.3272,
+      "step": 71240
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 8.554183173856474e-05,
+      "loss": 2.3089,
+      "step": 71250
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 8.550351935478506e-05,
+      "loss": 2.3481,
+      "step": 71260
+    },
+    {
+      "epoch": 9.63,
+      "learning_rate": 8.54652121322175e-05,
+      "loss": 2.3044,
+      "step": 71270
+    },
+    {
+      "epoch": 9.63,
+      "learning_rate": 8.54269100739274e-05,
+      "loss": 2.3281,
+      "step": 71280
+    },
+    {
+      "epoch": 9.63,
+      "learning_rate": 8.538861318298e-05,
+      "loss": 2.3636,
+      "step": 71290
+    },
+    {
+      "epoch": 9.63,
+      "learning_rate": 8.53503214624398e-05,
+      "loss": 2.3192,
+      "step": 71300
+    },
+    {
+      "epoch": 9.63,
+      "learning_rate": 8.531203491537104e-05,
+      "loss": 2.3563,
+      "step": 71310
+    },
+    {
+      "epoch": 9.63,
+      "learning_rate": 8.527375354483758e-05,
+      "loss": 2.315,
+      "step": 71320
+    },
+    {
+      "epoch": 9.63,
+      "learning_rate": 8.523547735390276e-05,
+      "loss": 2.3046,
+      "step": 71330
+    },
+    {
+      "epoch": 9.64,
+      "learning_rate": 8.519720634562965e-05,
+      "loss": 2.3066,
+      "step": 71340
+    },
+    {
+      "epoch": 9.64,
+      "learning_rate": 8.515894052308075e-05,
+      "loss": 2.3259,
+      "step": 71350
+    },
+    {
+      "epoch": 9.64,
+      "learning_rate": 8.512067988931818e-05,
+      "loss": 2.313,
+      "step": 71360
+    },
+    {
+      "epoch": 9.64,
+      "learning_rate": 8.508242444740377e-05,
+      "loss": 2.3035,
+      "step": 71370
+    },
+    {
+      "epoch": 9.64,
+      "learning_rate": 8.504417420039878e-05,
+      "loss": 2.3352,
+      "step": 71380
+    },
+    {
+      "epoch": 9.64,
+      "learning_rate": 8.50059291513641e-05,
+      "loss": 2.3481,
+      "step": 71390
+    },
+    {
+      "epoch": 9.64,
+      "learning_rate": 8.496768930336028e-05,
+      "loss": 2.3325,
+      "step": 71400
+    },
+    {
+      "epoch": 9.64,
+      "learning_rate": 8.492945465944731e-05,
+      "loss": 2.3498,
+      "step": 71410
+    },
+    {
+      "epoch": 9.65,
+      "learning_rate": 8.489122522268501e-05,
+      "loss": 2.3107,
+      "step": 71420
+    },
+    {
+      "epoch": 9.65,
+      "learning_rate": 8.48530009961324e-05,
+      "loss": 2.3329,
+      "step": 71430
+    },
+    {
+      "epoch": 9.65,
+      "learning_rate": 8.481478198284846e-05,
+      "loss": 2.3168,
+      "step": 71440
+    },
+    {
+      "epoch": 9.65,
+      "learning_rate": 8.477656818589149e-05,
+      "loss": 2.3044,
+      "step": 71450
+    },
+    {
+      "epoch": 9.65,
+      "learning_rate": 8.473835960831959e-05,
+      "loss": 2.3306,
+      "step": 71460
+    },
+    {
+      "epoch": 9.65,
+      "learning_rate": 8.470015625319028e-05,
+      "loss": 2.3128,
+      "step": 71470
+    },
+    {
+      "epoch": 9.65,
+      "learning_rate": 8.466195812356066e-05,
+      "loss": 2.3233,
+      "step": 71480
+    },
+    {
+      "epoch": 9.66,
+      "learning_rate": 8.462376522248755e-05,
+      "loss": 2.3202,
+      "step": 71490
+    },
+    {
+      "epoch": 9.66,
+      "learning_rate": 8.458557755302727e-05,
+      "loss": 2.3412,
+      "step": 71500
+    },
+    {
+      "epoch": 9.66,
+      "learning_rate": 8.454739511823559e-05,
+      "loss": 2.3196,
+      "step": 71510
+    },
+    {
+      "epoch": 9.66,
+      "learning_rate": 8.450921792116814e-05,
+      "loss": 2.3198,
+      "step": 71520
+    },
+    {
+      "epoch": 9.66,
+      "learning_rate": 8.447104596487988e-05,
+      "loss": 2.3234,
+      "step": 71530
+    },
+    {
+      "epoch": 9.66,
+      "learning_rate": 8.443287925242553e-05,
+      "loss": 2.3257,
+      "step": 71540
+    },
+    {
+      "epoch": 9.66,
+      "learning_rate": 8.439471778685927e-05,
+      "loss": 2.3132,
+      "step": 71550
+    },
+    {
+      "epoch": 9.66,
+      "learning_rate": 8.435656157123484e-05,
+      "loss": 2.3372,
+      "step": 71560
+    },
+    {
+      "epoch": 9.67,
+      "learning_rate": 8.431841060860574e-05,
+      "loss": 2.3342,
+      "step": 71570
+    },
+    {
+      "epoch": 9.67,
+      "learning_rate": 8.428026490202489e-05,
+      "loss": 2.3362,
+      "step": 71580
+    },
+    {
+      "epoch": 9.67,
+      "learning_rate": 8.424212445454473e-05,
+      "loss": 2.3153,
+      "step": 71590
+    },
+    {
+      "epoch": 9.67,
+      "learning_rate": 8.420398926921752e-05,
+      "loss": 2.314,
+      "step": 71600
+    },
+    {
+      "epoch": 9.67,
+      "learning_rate": 8.416585934909485e-05,
+      "loss": 2.3185,
+      "step": 71610
+    },
+    {
+      "epoch": 9.67,
+      "learning_rate": 8.412773469722816e-05,
+      "loss": 2.318,
+      "step": 71620
+    },
+    {
+      "epoch": 9.67,
+      "learning_rate": 8.408961531666807e-05,
+      "loss": 2.358,
+      "step": 71630
+    },
+    {
+      "epoch": 9.68,
+      "learning_rate": 8.40515012104652e-05,
+      "loss": 2.3115,
+      "step": 71640
+    },
+    {
+      "epoch": 9.68,
+      "learning_rate": 8.401339238166944e-05,
+      "loss": 2.335,
+      "step": 71650
+    },
+    {
+      "epoch": 9.68,
+      "learning_rate": 8.397528883333048e-05,
+      "loss": 2.3397,
+      "step": 71660
+    },
+    {
+      "epoch": 9.68,
+      "learning_rate": 8.393719056849746e-05,
+      "loss": 2.2946,
+      "step": 71670
+    },
+    {
+      "epoch": 9.68,
+      "learning_rate": 8.3899097590219e-05,
+      "loss": 2.3319,
+      "step": 71680
+    },
+    {
+      "epoch": 9.68,
+      "learning_rate": 8.386100990154364e-05,
+      "loss": 2.3021,
+      "step": 71690
+    },
+    {
+      "epoch": 9.68,
+      "learning_rate": 8.382292750551912e-05,
+      "loss": 2.3035,
+      "step": 71700
+    },
+    {
+      "epoch": 9.69,
+      "learning_rate": 8.378485040519292e-05,
+      "loss": 2.316,
+      "step": 71710
+    },
+    {
+      "epoch": 9.69,
+      "learning_rate": 8.374677860361219e-05,
+      "loss": 2.3185,
+      "step": 71720
+    },
+    {
+      "epoch": 9.69,
+      "learning_rate": 8.37087121038234e-05,
+      "loss": 2.329,
+      "step": 71730
+    },
+    {
+      "epoch": 9.69,
+      "learning_rate": 8.36706509088729e-05,
+      "loss": 2.3385,
+      "step": 71740
+    },
+    {
+      "epoch": 9.69,
+      "learning_rate": 8.363259502180645e-05,
+      "loss": 2.3062,
+      "step": 71750
+    },
+    {
+      "epoch": 9.69,
+      "learning_rate": 8.359454444566927e-05,
+      "loss": 2.3465,
+      "step": 71760
+    },
+    {
+      "epoch": 9.69,
+      "learning_rate": 8.355649918350646e-05,
+      "loss": 2.3339,
+      "step": 71770
+    },
+    {
+      "epoch": 9.69,
+      "learning_rate": 8.351845923836241e-05,
+      "loss": 2.2929,
+      "step": 71780
+    },
+    {
+      "epoch": 9.7,
+      "learning_rate": 8.348042461328124e-05,
+      "loss": 2.3191,
+      "step": 71790
+    },
+    {
+      "epoch": 9.7,
+      "learning_rate": 8.344239531130653e-05,
+      "loss": 2.3026,
+      "step": 71800
+    },
+    {
+      "epoch": 9.7,
+      "learning_rate": 8.340437133548157e-05,
+      "loss": 2.3352,
+      "step": 71810
+    },
+    {
+      "epoch": 9.7,
+      "learning_rate": 8.336635268884926e-05,
+      "loss": 2.3239,
+      "step": 71820
+    },
+    {
+      "epoch": 9.7,
+      "learning_rate": 8.332833937445177e-05,
+      "loss": 2.3088,
+      "step": 71830
+    },
+    {
+      "epoch": 9.7,
+      "learning_rate": 8.329033139533118e-05,
+      "loss": 2.3457,
+      "step": 71840
+    },
+    {
+      "epoch": 9.7,
+      "learning_rate": 8.32523287545289e-05,
+      "loss": 2.3309,
+      "step": 71850
+    },
+    {
+      "epoch": 9.71,
+      "learning_rate": 8.321433145508616e-05,
+      "loss": 2.3408,
+      "step": 71860
+    },
+    {
+      "epoch": 9.71,
+      "learning_rate": 8.317633950004355e-05,
+      "loss": 2.3164,
+      "step": 71870
+    },
+    {
+      "epoch": 9.71,
+      "learning_rate": 8.313835289244124e-05,
+      "loss": 2.3096,
+      "step": 71880
+    },
+    {
+      "epoch": 9.71,
+      "learning_rate": 8.310037163531919e-05,
+      "loss": 2.3449,
+      "step": 71890
+    },
+    {
+      "epoch": 9.71,
+      "learning_rate": 8.306239573171667e-05,
+      "loss": 2.2952,
+      "step": 71900
+    },
+    {
+      "epoch": 9.71,
+      "learning_rate": 8.302442518467261e-05,
+      "loss": 2.328,
+      "step": 71910
+    },
+    {
+      "epoch": 9.71,
+      "learning_rate": 8.298645999722565e-05,
+      "loss": 2.3518,
+      "step": 71920
+    },
+    {
+      "epoch": 9.71,
+      "learning_rate": 8.294850017241376e-05,
+      "loss": 2.3238,
+      "step": 71930
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 8.291054571327469e-05,
+      "loss": 2.34,
+      "step": 71940
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 8.287259662284568e-05,
+      "loss": 2.3127,
+      "step": 71950
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 8.283465290416342e-05,
+      "loss": 2.3194,
+      "step": 71960
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 8.279671456026446e-05,
+      "loss": 2.3169,
+      "step": 71970
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 8.275878159418463e-05,
+      "loss": 2.3029,
+      "step": 71980
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 8.272085400895949e-05,
+      "loss": 2.3107,
+      "step": 71990
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 8.268293180762404e-05,
+      "loss": 2.3384,
+      "step": 72000
+    },
+    {
+      "epoch": 9.72,
+      "eval_loss": 2.446279525756836,
+      "eval_runtime": 1269.5485,
+      "eval_samples_per_second": 59.97,
+      "eval_steps_per_second": 4.998,
+      "step": 72000
+    },
+    {
+      "epoch": 9.73,
+      "learning_rate": 8.264501499321299e-05,
+      "loss": 2.3068,
+      "step": 72010
+    },
+    {
+      "epoch": 9.73,
+      "learning_rate": 8.260710356876071e-05,
+      "loss": 2.3127,
+      "step": 72020
+    },
+    {
+      "epoch": 9.73,
+      "learning_rate": 8.256919753730074e-05,
+      "loss": 2.3312,
+      "step": 72030
+    },
+    {
+      "epoch": 9.73,
+      "learning_rate": 8.253129690186659e-05,
+      "loss": 2.3133,
+      "step": 72040
+    },
+    {
+      "epoch": 9.73,
+      "learning_rate": 8.249340166549112e-05,
+      "loss": 2.3072,
+      "step": 72050
+    },
+    {
+      "epoch": 9.73,
+      "learning_rate": 8.245551183120692e-05,
+      "loss": 2.3097,
+      "step": 72060
+    },
+    {
+      "epoch": 9.73,
+      "learning_rate": 8.241762740204599e-05,
+      "loss": 2.3782,
+      "step": 72070
+    },
+    {
+      "epoch": 9.74,
+      "learning_rate": 8.237974838103991e-05,
+      "loss": 2.3228,
+      "step": 72080
+    },
+    {
+      "epoch": 9.74,
+      "learning_rate": 8.234187477121998e-05,
+      "loss": 2.3365,
+      "step": 72090
+    },
+    {
+      "epoch": 9.74,
+      "learning_rate": 8.230400657561693e-05,
+      "loss": 2.3165,
+      "step": 72100
+    },
+    {
+      "epoch": 9.74,
+      "learning_rate": 8.226614379726103e-05,
+      "loss": 2.34,
+      "step": 72110
+    },
+    {
+      "epoch": 9.74,
+      "learning_rate": 8.222828643918231e-05,
+      "loss": 2.3741,
+      "step": 72120
+    },
+    {
+      "epoch": 9.74,
+      "learning_rate": 8.219043450441008e-05,
+      "loss": 2.3117,
+      "step": 72130
+    },
+    {
+      "epoch": 9.74,
+      "learning_rate": 8.215258799597353e-05,
+      "loss": 2.3264,
+      "step": 72140
+    },
+    {
+      "epoch": 9.74,
+      "learning_rate": 8.211474691690115e-05,
+      "loss": 2.31,
+      "step": 72150
+    },
+    {
+      "epoch": 9.75,
+      "learning_rate": 8.207691127022108e-05,
+      "loss": 2.3216,
+      "step": 72160
+    },
+    {
+      "epoch": 9.75,
+      "learning_rate": 8.203908105896117e-05,
+      "loss": 2.3223,
+      "step": 72170
+    },
+    {
+      "epoch": 9.75,
+      "learning_rate": 8.200125628614864e-05,
+      "loss": 2.3061,
+      "step": 72180
+    },
+    {
+      "epoch": 9.75,
+      "learning_rate": 8.196343695481035e-05,
+      "loss": 2.317,
+      "step": 72190
+    },
+    {
+      "epoch": 9.75,
+      "learning_rate": 8.192562306797265e-05,
+      "loss": 2.3211,
+      "step": 72200
+    },
+    {
+      "epoch": 9.75,
+      "learning_rate": 8.188781462866159e-05,
+      "loss": 2.3296,
+      "step": 72210
+    },
+    {
+      "epoch": 9.75,
+      "learning_rate": 8.185001163990287e-05,
+      "loss": 2.3312,
+      "step": 72220
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 8.181221410472135e-05,
+      "loss": 2.3524,
+      "step": 72230
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 8.177442202614184e-05,
+      "loss": 2.3249,
+      "step": 72240
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 8.173663540718854e-05,
+      "loss": 2.3314,
+      "step": 72250
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 8.169885425088534e-05,
+      "loss": 2.3096,
+      "step": 72260
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 8.166107856025553e-05,
+      "loss": 2.3418,
+      "step": 72270
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 8.162330833832198e-05,
+      "loss": 2.3204,
+      "step": 72280
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 8.158554358810736e-05,
+      "loss": 2.3556,
+      "step": 72290
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 8.15477843126336e-05,
+      "loss": 2.3055,
+      "step": 72300
+    },
+    {
+      "epoch": 9.77,
+      "learning_rate": 8.151003051492226e-05,
+      "loss": 2.3065,
+      "step": 72310
+    },
+    {
+      "epoch": 9.77,
+      "learning_rate": 8.14722821979947e-05,
+      "loss": 2.3045,
+      "step": 72320
+    },
+    {
+      "epoch": 9.77,
+      "learning_rate": 8.143453936487146e-05,
+      "loss": 2.3292,
+      "step": 72330
+    },
+    {
+      "epoch": 9.77,
+      "learning_rate": 8.139680201857302e-05,
+      "loss": 2.3238,
+      "step": 72340
+    },
+    {
+      "epoch": 9.77,
+      "learning_rate": 8.135907016211913e-05,
+      "loss": 2.3392,
+      "step": 72350
+    },
+    {
+      "epoch": 9.77,
+      "learning_rate": 8.132134379852921e-05,
+      "loss": 2.3239,
+      "step": 72360
+    },
+    {
+      "epoch": 9.77,
+      "learning_rate": 8.128362293082231e-05,
+      "loss": 2.3551,
+      "step": 72370
+    },
+    {
+      "epoch": 9.78,
+      "learning_rate": 8.124590756201697e-05,
+      "loss": 2.3554,
+      "step": 72380
+    },
+    {
+      "epoch": 9.78,
+      "learning_rate": 8.120819769513123e-05,
+      "loss": 2.339,
+      "step": 72390
+    },
+    {
+      "epoch": 9.78,
+      "learning_rate": 8.117049333318273e-05,
+      "loss": 2.343,
+      "step": 72400
+    },
+    {
+      "epoch": 9.78,
+      "learning_rate": 8.113279447918877e-05,
+      "loss": 2.34,
+      "step": 72410
+    },
+    {
+      "epoch": 9.78,
+      "learning_rate": 8.10951011361662e-05,
+      "loss": 2.3268,
+      "step": 72420
+    },
+    {
+      "epoch": 9.78,
+      "learning_rate": 8.105741330713115e-05,
+      "loss": 2.3213,
+      "step": 72430
+    },
+    {
+      "epoch": 9.78,
+      "learning_rate": 8.101973099509969e-05,
+      "loss": 2.3621,
+      "step": 72440
+    },
+    {
+      "epoch": 9.79,
+      "learning_rate": 8.098205420308716e-05,
+      "loss": 2.3484,
+      "step": 72450
+    },
+    {
+      "epoch": 9.79,
+      "learning_rate": 8.094438293410871e-05,
+      "loss": 2.3508,
+      "step": 72460
+    },
+    {
+      "epoch": 9.79,
+      "learning_rate": 8.090671719117884e-05,
+      "loss": 2.314,
+      "step": 72470
+    },
+    {
+      "epoch": 9.79,
+      "learning_rate": 8.086905697731164e-05,
+      "loss": 2.3237,
+      "step": 72480
+    },
+    {
+      "epoch": 9.79,
+      "learning_rate": 8.083140229552088e-05,
+      "loss": 2.3521,
+      "step": 72490
+    },
+    {
+      "epoch": 9.79,
+      "learning_rate": 8.079375314881978e-05,
+      "loss": 2.345,
+      "step": 72500
+    },
+    {
+      "epoch": 9.79,
+      "learning_rate": 8.075610954022109e-05,
+      "loss": 2.336,
+      "step": 72510
+    },
+    {
+      "epoch": 9.79,
+      "learning_rate": 8.071847147273726e-05,
+      "loss": 2.3152,
+      "step": 72520
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 8.06808389493801e-05,
+      "loss": 2.3125,
+      "step": 72530
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 8.06432119731612e-05,
+      "loss": 2.3431,
+      "step": 72540
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 8.060559054709154e-05,
+      "loss": 2.3159,
+      "step": 72550
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 8.056797467418164e-05,
+      "loss": 2.3364,
+      "step": 72560
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 8.053036435744176e-05,
+      "loss": 2.3274,
+      "step": 72570
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 8.049275959988153e-05,
+      "loss": 2.3399,
+      "step": 72580
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 8.045516040451023e-05,
+      "loss": 2.3344,
+      "step": 72590
+    },
+    {
+      "epoch": 9.81,
+      "learning_rate": 8.041756677433657e-05,
+      "loss": 2.3327,
+      "step": 72600
+    },
+    {
+      "epoch": 9.81,
+      "learning_rate": 8.037997871236899e-05,
+      "loss": 2.3487,
+      "step": 72610
+    },
+    {
+      "epoch": 9.81,
+      "learning_rate": 8.034239622161552e-05,
+      "loss": 2.3181,
+      "step": 72620
+    },
+    {
+      "epoch": 9.81,
+      "learning_rate": 8.030481930508338e-05,
+      "loss": 2.3124,
+      "step": 72630
+    },
+    {
+      "epoch": 9.81,
+      "learning_rate": 8.02672479657798e-05,
+      "loss": 2.3469,
+      "step": 72640
+    },
+    {
+      "epoch": 9.81,
+      "learning_rate": 8.022968220671123e-05,
+      "loss": 2.3582,
+      "step": 72650
+    },
+    {
+      "epoch": 9.81,
+      "learning_rate": 8.01921220308839e-05,
+      "loss": 2.3381,
+      "step": 72660
+    },
+    {
+      "epoch": 9.81,
+      "learning_rate": 8.015456744130344e-05,
+      "loss": 2.3173,
+      "step": 72670
+    },
+    {
+      "epoch": 9.82,
+      "learning_rate": 8.011701844097504e-05,
+      "loss": 2.3267,
+      "step": 72680
+    },
+    {
+      "epoch": 9.82,
+      "learning_rate": 8.00794750329036e-05,
+      "loss": 2.3173,
+      "step": 72690
+    },
+    {
+      "epoch": 9.82,
+      "learning_rate": 8.00419372200934e-05,
+      "loss": 2.3142,
+      "step": 72700
+    },
+    {
+      "epoch": 9.82,
+      "learning_rate": 8.00044050055483e-05,
+      "loss": 2.3247,
+      "step": 72710
+    },
+    {
+      "epoch": 9.82,
+      "learning_rate": 7.996687839227181e-05,
+      "loss": 2.3183,
+      "step": 72720
+    },
+    {
+      "epoch": 9.82,
+      "learning_rate": 7.992935738326689e-05,
+      "loss": 2.3312,
+      "step": 72730
+    },
+    {
+      "epoch": 9.82,
+      "learning_rate": 7.989184198153613e-05,
+      "loss": 2.3302,
+      "step": 72740
+    },
+    {
+      "epoch": 9.83,
+      "learning_rate": 7.98543321900816e-05,
+      "loss": 2.296,
+      "step": 72750
+    },
+    {
+      "epoch": 9.83,
+      "learning_rate": 7.981682801190493e-05,
+      "loss": 2.3502,
+      "step": 72760
+    },
+    {
+      "epoch": 9.83,
+      "learning_rate": 7.977932945000738e-05,
+      "loss": 2.2959,
+      "step": 72770
+    },
+    {
+      "epoch": 9.83,
+      "learning_rate": 7.974183650738968e-05,
+      "loss": 2.3102,
+      "step": 72780
+    },
+    {
+      "epoch": 9.83,
+      "learning_rate": 7.970434918705214e-05,
+      "loss": 2.3401,
+      "step": 72790
+    },
+    {
+      "epoch": 9.83,
+      "learning_rate": 7.966686749199452e-05,
+      "loss": 2.3505,
+      "step": 72800
+    },
+    {
+      "epoch": 9.83,
+      "learning_rate": 7.962939142521632e-05,
+      "loss": 2.3316,
+      "step": 72810
+    },
+    {
+      "epoch": 9.84,
+      "learning_rate": 7.959192098971658e-05,
+      "loss": 2.3281,
+      "step": 72820
+    },
+    {
+      "epoch": 9.84,
+      "learning_rate": 7.95544561884936e-05,
+      "loss": 2.3243,
+      "step": 72830
+    },
+    {
+      "epoch": 9.84,
+      "learning_rate": 7.951699702454555e-05,
+      "loss": 2.3329,
+      "step": 72840
+    },
+    {
+      "epoch": 9.84,
+      "learning_rate": 7.947954350087e-05,
+      "loss": 2.3442,
+      "step": 72850
+    },
+    {
+      "epoch": 9.84,
+      "learning_rate": 7.944209562046416e-05,
+      "loss": 2.351,
+      "step": 72860
+    },
+    {
+      "epoch": 9.84,
+      "learning_rate": 7.940465338632466e-05,
+      "loss": 2.3243,
+      "step": 72870
+    },
+    {
+      "epoch": 9.84,
+      "learning_rate": 7.936721680144771e-05,
+      "loss": 2.3261,
+      "step": 72880
+    },
+    {
+      "epoch": 9.84,
+      "learning_rate": 7.932978586882923e-05,
+      "loss": 2.3444,
+      "step": 72890
+    },
+    {
+      "epoch": 9.85,
+      "learning_rate": 7.929236059146448e-05,
+      "loss": 2.3373,
+      "step": 72900
+    },
+    {
+      "epoch": 9.85,
+      "learning_rate": 7.925494097234831e-05,
+      "loss": 2.335,
+      "step": 72910
+    },
+    {
+      "epoch": 9.85,
+      "learning_rate": 7.921752701447526e-05,
+      "loss": 2.3288,
+      "step": 72920
+    },
+    {
+      "epoch": 9.85,
+      "learning_rate": 7.918011872083919e-05,
+      "loss": 2.3072,
+      "step": 72930
+    },
+    {
+      "epoch": 9.85,
+      "learning_rate": 7.914271609443381e-05,
+      "loss": 2.3305,
+      "step": 72940
+    },
+    {
+      "epoch": 9.85,
+      "learning_rate": 7.910531913825198e-05,
+      "loss": 2.3355,
+      "step": 72950
+    },
+    {
+      "epoch": 9.85,
+      "learning_rate": 7.906792785528643e-05,
+      "loss": 2.3132,
+      "step": 72960
+    },
+    {
+      "epoch": 9.86,
+      "learning_rate": 7.903054224852938e-05,
+      "loss": 2.3183,
+      "step": 72970
+    },
+    {
+      "epoch": 9.86,
+      "learning_rate": 7.899316232097247e-05,
+      "loss": 2.3356,
+      "step": 72980
+    },
+    {
+      "epoch": 9.86,
+      "learning_rate": 7.8955788075607e-05,
+      "loss": 2.2969,
+      "step": 72990
+    },
+    {
+      "epoch": 9.86,
+      "learning_rate": 7.89184195154237e-05,
+      "loss": 2.3195,
+      "step": 73000
+    },
+    {
+      "epoch": 9.86,
+      "eval_loss": 2.442723274230957,
+      "eval_runtime": 1269.4894,
+      "eval_samples_per_second": 59.973,
+      "eval_steps_per_second": 4.998,
+      "step": 73000
+    },
+    {
+      "epoch": 9.86,
+      "learning_rate": 7.888105664341296e-05,
+      "loss": 2.3405,
+      "step": 73010
+    },
+    {
+      "epoch": 9.86,
+      "learning_rate": 7.884369946256483e-05,
+      "loss": 2.3412,
+      "step": 73020
+    },
+    {
+      "epoch": 9.86,
+      "learning_rate": 7.880634797586846e-05,
+      "loss": 2.3368,
+      "step": 73030
+    },
+    {
+      "epoch": 9.86,
+      "learning_rate": 7.876900218631305e-05,
+      "loss": 2.3505,
+      "step": 73040
+    },
+    {
+      "epoch": 9.87,
+      "learning_rate": 7.8731662096887e-05,
+      "loss": 2.34,
+      "step": 73050
+    },
+    {
+      "epoch": 9.87,
+      "learning_rate": 7.869432771057852e-05,
+      "loss": 2.3367,
+      "step": 73060
+    },
+    {
+      "epoch": 9.87,
+      "learning_rate": 7.865699903037514e-05,
+      "loss": 2.3125,
+      "step": 73070
+    },
+    {
+      "epoch": 9.87,
+      "learning_rate": 7.861967605926396e-05,
+      "loss": 2.3156,
+      "step": 73080
+    },
+    {
+      "epoch": 9.87,
+      "learning_rate": 7.858235880023182e-05,
+      "loss": 2.3428,
+      "step": 73090
+    },
+    {
+      "epoch": 9.87,
+      "learning_rate": 7.85450472562649e-05,
+      "loss": 2.3348,
+      "step": 73100
+    },
+    {
+      "epoch": 9.87,
+      "learning_rate": 7.85077414303489e-05,
+      "loss": 2.3366,
+      "step": 73110
+    },
+    {
+      "epoch": 9.88,
+      "learning_rate": 7.847044132546932e-05,
+      "loss": 2.3302,
+      "step": 73120
+    },
+    {
+      "epoch": 9.88,
+      "learning_rate": 7.843314694461087e-05,
+      "loss": 2.3401,
+      "step": 73130
+    },
+    {
+      "epoch": 9.88,
+      "learning_rate": 7.839585829075815e-05,
+      "loss": 2.3344,
+      "step": 73140
+    },
+    {
+      "epoch": 9.88,
+      "learning_rate": 7.835857536689489e-05,
+      "loss": 2.3426,
+      "step": 73150
+    },
+    {
+      "epoch": 9.88,
+      "learning_rate": 7.83212981760047e-05,
+      "loss": 2.3094,
+      "step": 73160
+    },
+    {
+      "epoch": 9.88,
+      "learning_rate": 7.828402672107068e-05,
+      "loss": 2.3044,
+      "step": 73170
+    },
+    {
+      "epoch": 9.88,
+      "learning_rate": 7.824676100507536e-05,
+      "loss": 2.3035,
+      "step": 73180
+    },
+    {
+      "epoch": 9.88,
+      "learning_rate": 7.820950103100084e-05,
+      "loss": 2.3315,
+      "step": 73190
+    },
+    {
+      "epoch": 9.89,
+      "learning_rate": 7.817224680182873e-05,
+      "loss": 2.3362,
+      "step": 73200
+    },
+    {
+      "epoch": 9.89,
+      "learning_rate": 7.813499832054031e-05,
+      "loss": 2.3506,
+      "step": 73210
+    },
+    {
+      "epoch": 9.89,
+      "learning_rate": 7.80977555901164e-05,
+      "loss": 2.2887,
+      "step": 73220
+    },
+    {
+      "epoch": 9.89,
+      "learning_rate": 7.806051861353707e-05,
+      "loss": 2.3251,
+      "step": 73230
+    },
+    {
+      "epoch": 9.89,
+      "learning_rate": 7.802328739378232e-05,
+      "loss": 2.3021,
+      "step": 73240
+    },
+    {
+      "epoch": 9.89,
+      "learning_rate": 7.798606193383136e-05,
+      "loss": 2.3161,
+      "step": 73250
+    },
+    {
+      "epoch": 9.89,
+      "learning_rate": 7.794884223666327e-05,
+      "loss": 2.3154,
+      "step": 73260
+    },
+    {
+      "epoch": 9.9,
+      "learning_rate": 7.791162830525637e-05,
+      "loss": 2.3101,
+      "step": 73270
+    },
+    {
+      "epoch": 9.9,
+      "learning_rate": 7.78744201425886e-05,
+      "loss": 2.3441,
+      "step": 73280
+    },
+    {
+      "epoch": 9.9,
+      "learning_rate": 7.78372177516376e-05,
+      "loss": 2.3122,
+      "step": 73290
+    },
+    {
+      "epoch": 9.9,
+      "learning_rate": 7.780002113538035e-05,
+      "loss": 2.3155,
+      "step": 73300
+    },
+    {
+      "epoch": 9.9,
+      "learning_rate": 7.776283029679339e-05,
+      "loss": 2.2871,
+      "step": 73310
+    },
+    {
+      "epoch": 9.9,
+      "learning_rate": 7.772564523885294e-05,
+      "loss": 2.3248,
+      "step": 73320
+    },
+    {
+      "epoch": 9.9,
+      "learning_rate": 7.76884659645346e-05,
+      "loss": 2.3508,
+      "step": 73330
+    },
+    {
+      "epoch": 9.91,
+      "learning_rate": 7.765129247681372e-05,
+      "loss": 2.3322,
+      "step": 73340
+    },
+    {
+      "epoch": 9.91,
+      "learning_rate": 7.761412477866477e-05,
+      "loss": 2.3289,
+      "step": 73350
+    },
+    {
+      "epoch": 9.91,
+      "learning_rate": 7.75769628730622e-05,
+      "loss": 2.3293,
+      "step": 73360
+    },
+    {
+      "epoch": 9.91,
+      "learning_rate": 7.753980676297986e-05,
+      "loss": 2.3257,
+      "step": 73370
+    },
+    {
+      "epoch": 9.91,
+      "learning_rate": 7.750265645139104e-05,
+      "loss": 2.3062,
+      "step": 73380
+    },
+    {
+      "epoch": 9.91,
+      "learning_rate": 7.746551194126864e-05,
+      "loss": 2.2948,
+      "step": 73390
+    },
+    {
+      "epoch": 9.91,
+      "learning_rate": 7.742837323558498e-05,
+      "loss": 2.3186,
+      "step": 73400
+    },
+    {
+      "epoch": 9.91,
+      "learning_rate": 7.739124033731214e-05,
+      "loss": 2.3132,
+      "step": 73410
+    },
+    {
+      "epoch": 9.92,
+      "learning_rate": 7.735411324942167e-05,
+      "loss": 2.2895,
+      "step": 73420
+    },
+    {
+      "epoch": 9.92,
+      "learning_rate": 7.732070384065256e-05,
+      "loss": 2.3114,
+      "step": 73430
+    },
+    {
+      "epoch": 9.92,
+      "learning_rate": 7.72835878006731e-05,
+      "loss": 2.3158,
+      "step": 73440
+    },
+    {
+      "epoch": 9.92,
+      "learning_rate": 7.724647757969068e-05,
+      "loss": 2.3355,
+      "step": 73450
+    },
+    {
+      "epoch": 9.92,
+      "learning_rate": 7.720937318067489e-05,
+      "loss": 2.3104,
+      "step": 73460
+    },
+    {
+      "epoch": 9.92,
+      "learning_rate": 7.71722746065949e-05,
+      "loss": 2.3438,
+      "step": 73470
+    },
+    {
+      "epoch": 9.92,
+      "learning_rate": 7.713518186041959e-05,
+      "loss": 2.3307,
+      "step": 73480
+    },
+    {
+      "epoch": 9.93,
+      "learning_rate": 7.709809494511716e-05,
+      "loss": 2.3443,
+      "step": 73490
+    },
+    {
+      "epoch": 9.93,
+      "learning_rate": 7.706101386365537e-05,
+      "loss": 2.331,
+      "step": 73500
+    },
+    {
+      "epoch": 9.93,
+      "learning_rate": 7.702393861900165e-05,
+      "loss": 2.326,
+      "step": 73510
+    },
+    {
+      "epoch": 9.93,
+      "learning_rate": 7.698686921412282e-05,
+      "loss": 2.3392,
+      "step": 73520
+    },
+    {
+      "epoch": 9.93,
+      "learning_rate": 7.694980565198543e-05,
+      "loss": 2.271,
+      "step": 73530
+    },
+    {
+      "epoch": 9.93,
+      "learning_rate": 7.691274793555519e-05,
+      "loss": 2.3062,
+      "step": 73540
+    },
+    {
+      "epoch": 9.93,
+      "learning_rate": 7.687569606779777e-05,
+      "loss": 2.3192,
+      "step": 73550
+    },
+    {
+      "epoch": 9.93,
+      "learning_rate": 7.683865005167803e-05,
+      "loss": 2.3228,
+      "step": 73560
+    },
+    {
+      "epoch": 9.94,
+      "learning_rate": 7.680160989016067e-05,
+      "loss": 2.3511,
+      "step": 73570
+    },
+    {
+      "epoch": 9.94,
+      "learning_rate": 7.676457558620965e-05,
+      "loss": 2.3147,
+      "step": 73580
+    },
+    {
+      "epoch": 9.94,
+      "learning_rate": 7.672754714278855e-05,
+      "loss": 2.3201,
+      "step": 73590
+    },
+    {
+      "epoch": 9.94,
+      "learning_rate": 7.669052456286062e-05,
+      "loss": 2.3203,
+      "step": 73600
+    },
+    {
+      "epoch": 9.94,
+      "learning_rate": 7.665350784938845e-05,
+      "loss": 2.3201,
+      "step": 73610
+    },
+    {
+      "epoch": 9.94,
+      "learning_rate": 7.661649700533417e-05,
+      "loss": 2.3442,
+      "step": 73620
+    },
+    {
+      "epoch": 9.94,
+      "learning_rate": 7.657949203365967e-05,
+      "loss": 2.3143,
+      "step": 73630
+    },
+    {
+      "epoch": 9.95,
+      "learning_rate": 7.654249293732601e-05,
+      "loss": 2.3344,
+      "step": 73640
+    },
+    {
+      "epoch": 9.95,
+      "learning_rate": 7.650549971929414e-05,
+      "loss": 2.3371,
+      "step": 73650
+    },
+    {
+      "epoch": 9.95,
+      "learning_rate": 7.646851238252432e-05,
+      "loss": 2.2976,
+      "step": 73660
+    },
+    {
+      "epoch": 9.95,
+      "learning_rate": 7.643153092997632e-05,
+      "loss": 2.3583,
+      "step": 73670
+    },
+    {
+      "epoch": 9.95,
+      "learning_rate": 7.639455536460963e-05,
+      "loss": 2.3362,
+      "step": 73680
+    },
+    {
+      "epoch": 9.95,
+      "learning_rate": 7.635758568938309e-05,
+      "loss": 2.3178,
+      "step": 73690
+    },
+    {
+      "epoch": 9.95,
+      "learning_rate": 7.632062190725508e-05,
+      "loss": 2.3236,
+      "step": 73700
+    },
+    {
+      "epoch": 9.96,
+      "learning_rate": 7.628366402118367e-05,
+      "loss": 2.335,
+      "step": 73710
+    },
+    {
+      "epoch": 9.96,
+      "learning_rate": 7.624671203412623e-05,
+      "loss": 2.3298,
+      "step": 73720
+    },
+    {
+      "epoch": 9.96,
+      "learning_rate": 7.620976594903997e-05,
+      "loss": 2.3284,
+      "step": 73730
+    },
+    {
+      "epoch": 9.96,
+      "learning_rate": 7.617282576888115e-05,
+      "loss": 2.3405,
+      "step": 73740
+    },
+    {
+      "epoch": 9.96,
+      "learning_rate": 7.613589149660607e-05,
+      "loss": 2.3345,
+      "step": 73750
+    },
+    {
+      "epoch": 9.96,
+      "learning_rate": 7.609896313517017e-05,
+      "loss": 2.342,
+      "step": 73760
+    },
+    {
+      "epoch": 9.96,
+      "learning_rate": 7.606204068752874e-05,
+      "loss": 2.3303,
+      "step": 73770
+    },
+    {
+      "epoch": 9.96,
+      "learning_rate": 7.602512415663632e-05,
+      "loss": 2.3287,
+      "step": 73780
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 7.598821354544702e-05,
+      "loss": 2.3382,
+      "step": 73790
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 7.595130885691474e-05,
+      "loss": 2.3479,
+      "step": 73800
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 7.591441009399258e-05,
+      "loss": 2.339,
+      "step": 73810
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 7.587751725963326e-05,
+      "loss": 2.3257,
+      "step": 73820
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 7.584063035678917e-05,
+      "loss": 2.3039,
+      "step": 73830
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 7.5803749388412e-05,
+      "loss": 2.3207,
+      "step": 73840
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 7.576687435745323e-05,
+      "loss": 2.3073,
+      "step": 73850
+    },
+    {
+      "epoch": 9.98,
+      "learning_rate": 7.573000526686359e-05,
+      "loss": 2.3336,
+      "step": 73860
+    },
+    {
+      "epoch": 9.98,
+      "learning_rate": 7.569314211959346e-05,
+      "loss": 2.2973,
+      "step": 73870
+    },
+    {
+      "epoch": 9.98,
+      "learning_rate": 7.565628491859286e-05,
+      "loss": 2.3278,
+      "step": 73880
+    },
+    {
+      "epoch": 9.98,
+      "learning_rate": 7.561943366681114e-05,
+      "loss": 2.3097,
+      "step": 73890
+    },
+    {
+      "epoch": 9.98,
+      "learning_rate": 7.55825883671972e-05,
+      "loss": 2.3274,
+      "step": 73900
+    },
+    {
+      "epoch": 9.98,
+      "learning_rate": 7.554574902269967e-05,
+      "loss": 2.3112,
+      "step": 73910
+    },
+    {
+      "epoch": 9.98,
+      "learning_rate": 7.550891563626636e-05,
+      "loss": 2.3235,
+      "step": 73920
+    },
+    {
+      "epoch": 9.98,
+      "learning_rate": 7.547208821084501e-05,
+      "loss": 2.3231,
+      "step": 73930
+    },
+    {
+      "epoch": 9.99,
+      "learning_rate": 7.543526674938246e-05,
+      "loss": 2.3159,
+      "step": 73940
+    },
+    {
+      "epoch": 9.99,
+      "learning_rate": 7.539845125482543e-05,
+      "loss": 2.2928,
+      "step": 73950
+    },
+    {
+      "epoch": 9.99,
+      "learning_rate": 7.536164173011988e-05,
+      "loss": 2.3303,
+      "step": 73960
+    },
+    {
+      "epoch": 9.99,
+      "learning_rate": 7.532483817821156e-05,
+      "loss": 2.3305,
+      "step": 73970
+    },
+    {
+      "epoch": 9.99,
+      "learning_rate": 7.528804060204554e-05,
+      "loss": 2.3057,
+      "step": 73980
+    },
+    {
+      "epoch": 9.99,
+      "learning_rate": 7.525124900456641e-05,
+      "loss": 2.3563,
+      "step": 73990
+    },
+    {
+      "epoch": 9.99,
+      "learning_rate": 7.52144633887185e-05,
+      "loss": 2.3183,
+      "step": 74000
+    },
+    {
+      "epoch": 9.99,
+      "eval_loss": 2.4382996559143066,
+      "eval_runtime": 1269.0959,
+      "eval_samples_per_second": 59.992,
+      "eval_steps_per_second": 5.0,
+      "step": 74000
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7.517768375744541e-05,
+      "loss": 2.3382,
+      "step": 74010
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7.514091011369036e-05,
+      "loss": 2.3106,
+      "step": 74020
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7.510414246039616e-05,
+      "loss": 2.3396,
+      "step": 74030
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7.506738080050496e-05,
+      "loss": 2.3195,
+      "step": 74040
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7.503062513695868e-05,
+      "loss": 2.2994,
+      "step": 74050
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7.499387547269857e-05,
+      "loss": 2.2542,
+      "step": 74060
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 7.495713181066538e-05,
+      "loss": 2.285,
+      "step": 74070
+    },
+    {
+      "epoch": 10.01,
+      "learning_rate": 7.492039415379957e-05,
+      "loss": 2.2853,
+      "step": 74080
+    },
+    {
+      "epoch": 10.01,
+      "learning_rate": 7.488366250504095e-05,
+      "loss": 2.274,
+      "step": 74090
+    },
+    {
+      "epoch": 10.01,
+      "learning_rate": 7.484693686732887e-05,
+      "loss": 2.3008,
+      "step": 74100
+    },
+    {
+      "epoch": 10.01,
+      "learning_rate": 7.481021724360231e-05,
+      "loss": 2.2662,
+      "step": 74110
+    },
+    {
+      "epoch": 10.01,
+      "learning_rate": 7.477350363679957e-05,
+      "loss": 2.2901,
+      "step": 74120
+    },
+    {
+      "epoch": 10.01,
+      "learning_rate": 7.47367960498588e-05,
+      "loss": 2.2318,
+      "step": 74130
+    },
+    {
+      "epoch": 10.01,
+      "learning_rate": 7.47000944857172e-05,
+      "loss": 2.2923,
+      "step": 74140
+    },
+    {
+      "epoch": 10.01,
+      "learning_rate": 7.466339894731193e-05,
+      "loss": 2.2659,
+      "step": 74150
+    },
+    {
+      "epoch": 10.02,
+      "learning_rate": 7.462670943757936e-05,
+      "loss": 2.2859,
+      "step": 74160
+    },
+    {
+      "epoch": 10.02,
+      "learning_rate": 7.45900259594556e-05,
+      "loss": 2.2853,
+      "step": 74170
+    },
+    {
+      "epoch": 10.02,
+      "learning_rate": 7.455334851587615e-05,
+      "loss": 2.284,
+      "step": 74180
+    },
+    {
+      "epoch": 10.02,
+      "learning_rate": 7.4516677109776e-05,
+      "loss": 2.2795,
+      "step": 74190
+    },
+    {
+      "epoch": 10.02,
+      "learning_rate": 7.44800117440898e-05,
+      "loss": 2.2537,
+      "step": 74200
+    },
+    {
+      "epoch": 10.02,
+      "learning_rate": 7.444335242175158e-05,
+      "loss": 2.2663,
+      "step": 74210
+    },
+    {
+      "epoch": 10.02,
+      "learning_rate": 7.440669914569488e-05,
+      "loss": 2.2674,
+      "step": 74220
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 7.437005191885293e-05,
+      "loss": 2.2734,
+      "step": 74230
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 7.433341074415822e-05,
+      "loss": 2.2975,
+      "step": 74240
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 7.429677562454305e-05,
+      "loss": 2.2357,
+      "step": 74250
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 7.426014656293899e-05,
+      "loss": 2.2686,
+      "step": 74260
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 7.422352356227715e-05,
+      "loss": 2.2922,
+      "step": 74270
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 7.418690662548836e-05,
+      "loss": 2.3174,
+      "step": 74280
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 7.415029575550274e-05,
+      "loss": 2.2796,
+      "step": 74290
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 7.411369095524997e-05,
+      "loss": 2.26,
+      "step": 74300
+    },
+    {
+      "epoch": 10.04,
+      "learning_rate": 7.407709222765936e-05,
+      "loss": 2.2817,
+      "step": 74310
+    },
+    {
+      "epoch": 10.04,
+      "learning_rate": 7.404049957565961e-05,
+      "loss": 2.2818,
+      "step": 74320
+    },
+    {
+      "epoch": 10.04,
+      "learning_rate": 7.400391300217908e-05,
+      "loss": 2.2876,
+      "step": 74330
+    },
+    {
+      "epoch": 10.04,
+      "learning_rate": 7.396733251014533e-05,
+      "loss": 2.2688,
+      "step": 74340
+    },
+    {
+      "epoch": 10.04,
+      "learning_rate": 7.393075810248586e-05,
+      "loss": 2.2792,
+      "step": 74350
+    },
+    {
+      "epoch": 10.04,
+      "learning_rate": 7.389418978212732e-05,
+      "loss": 2.2799,
+      "step": 74360
+    },
+    {
+      "epoch": 10.04,
+      "learning_rate": 7.385762755199613e-05,
+      "loss": 2.2871,
+      "step": 74370
+    },
+    {
+      "epoch": 10.05,
+      "learning_rate": 7.38210714150181e-05,
+      "loss": 2.2777,
+      "step": 74380
+    },
+    {
+      "epoch": 10.05,
+      "learning_rate": 7.378452137411846e-05,
+      "loss": 2.2812,
+      "step": 74390
+    },
+    {
+      "epoch": 10.05,
+      "learning_rate": 7.374797743222221e-05,
+      "loss": 2.257,
+      "step": 74400
+    },
+    {
+      "epoch": 10.05,
+      "learning_rate": 7.371143959225363e-05,
+      "loss": 2.2838,
+      "step": 74410
+    },
+    {
+      "epoch": 10.05,
+      "learning_rate": 7.367490785713654e-05,
+      "loss": 2.278,
+      "step": 74420
+    },
+    {
+      "epoch": 10.05,
+      "learning_rate": 7.363838222979445e-05,
+      "loss": 2.2827,
+      "step": 74430
+    },
+    {
+      "epoch": 10.05,
+      "learning_rate": 7.360186271315013e-05,
+      "loss": 2.2847,
+      "step": 74440
+    },
+    {
+      "epoch": 10.06,
+      "learning_rate": 7.356534931012613e-05,
+      "loss": 2.2956,
+      "step": 74450
+    },
+    {
+      "epoch": 10.06,
+      "learning_rate": 7.352884202364428e-05,
+      "loss": 2.2798,
+      "step": 74460
+    },
+    {
+      "epoch": 10.06,
+      "learning_rate": 7.349234085662596e-05,
+      "loss": 2.2965,
+      "step": 74470
+    },
+    {
+      "epoch": 10.06,
+      "learning_rate": 7.345584581199223e-05,
+      "loss": 2.2983,
+      "step": 74480
+    },
+    {
+      "epoch": 10.06,
+      "learning_rate": 7.341935689266346e-05,
+      "loss": 2.2888,
+      "step": 74490
+    },
+    {
+      "epoch": 10.06,
+      "learning_rate": 7.338287410155956e-05,
+      "loss": 2.2695,
+      "step": 74500
+    },
+    {
+      "epoch": 10.06,
+      "learning_rate": 7.334639744160015e-05,
+      "loss": 2.2846,
+      "step": 74510
+    },
+    {
+      "epoch": 10.06,
+      "learning_rate": 7.330992691570405e-05,
+      "loss": 2.2776,
+      "step": 74520
+    },
+    {
+      "epoch": 10.07,
+      "learning_rate": 7.327346252678993e-05,
+      "loss": 2.2856,
+      "step": 74530
+    },
+    {
+      "epoch": 10.07,
+      "learning_rate": 7.323700427777555e-05,
+      "loss": 2.2801,
+      "step": 74540
+    },
+    {
+      "epoch": 10.07,
+      "learning_rate": 7.32005521715786e-05,
+      "loss": 2.3014,
+      "step": 74550
+    },
+    {
+      "epoch": 10.07,
+      "learning_rate": 7.316410621111597e-05,
+      "loss": 2.2945,
+      "step": 74560
+    },
+    {
+      "epoch": 10.07,
+      "learning_rate": 7.31276663993043e-05,
+      "loss": 2.2861,
+      "step": 74570
+    },
+    {
+      "epoch": 10.07,
+      "learning_rate": 7.309123273905955e-05,
+      "loss": 2.2648,
+      "step": 74580
+    },
+    {
+      "epoch": 10.07,
+      "learning_rate": 7.30548052332972e-05,
+      "loss": 2.2642,
+      "step": 74590
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 7.301838388493244e-05,
+      "loss": 2.2877,
+      "step": 74600
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 7.298196869687973e-05,
+      "loss": 2.2649,
+      "step": 74610
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 7.294555967205309e-05,
+      "loss": 2.2688,
+      "step": 74620
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 7.290915681336618e-05,
+      "loss": 2.2914,
+      "step": 74630
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 7.287276012373196e-05,
+      "loss": 2.2729,
+      "step": 74640
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 7.283636960606314e-05,
+      "loss": 2.2764,
+      "step": 74650
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 7.279998526327174e-05,
+      "loss": 2.2877,
+      "step": 74660
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 7.27636070982693e-05,
+      "loss": 2.2891,
+      "step": 74670
+    },
+    {
+      "epoch": 10.09,
+      "learning_rate": 7.272723511396702e-05,
+      "loss": 2.2695,
+      "step": 74680
+    },
+    {
+      "epoch": 10.09,
+      "learning_rate": 7.269086931327543e-05,
+      "loss": 2.2822,
+      "step": 74690
+    },
+    {
+      "epoch": 10.09,
+      "learning_rate": 7.265450969910465e-05,
+      "loss": 2.2802,
+      "step": 74700
+    },
+    {
+      "epoch": 10.09,
+      "learning_rate": 7.261815627436425e-05,
+      "loss": 2.3001,
+      "step": 74710
+    },
+    {
+      "epoch": 10.09,
+      "learning_rate": 7.258180904196338e-05,
+      "loss": 2.2765,
+      "step": 74720
+    },
+    {
+      "epoch": 10.09,
+      "learning_rate": 7.254546800481077e-05,
+      "loss": 2.2653,
+      "step": 74730
+    },
+    {
+      "epoch": 10.09,
+      "learning_rate": 7.250913316581436e-05,
+      "loss": 2.2761,
+      "step": 74740
+    },
+    {
+      "epoch": 10.1,
+      "learning_rate": 7.247280452788189e-05,
+      "loss": 2.2761,
+      "step": 74750
+    },
+    {
+      "epoch": 10.1,
+      "learning_rate": 7.243648209392042e-05,
+      "loss": 2.2513,
+      "step": 74760
+    },
+    {
+      "epoch": 10.1,
+      "learning_rate": 7.240016586683668e-05,
+      "loss": 2.2945,
+      "step": 74770
+    },
+    {
+      "epoch": 10.1,
+      "learning_rate": 7.236385584953676e-05,
+      "loss": 2.2818,
+      "step": 74780
+    },
+    {
+      "epoch": 10.1,
+      "learning_rate": 7.232755204492624e-05,
+      "loss": 2.2997,
+      "step": 74790
+    },
+    {
+      "epoch": 10.1,
+      "learning_rate": 7.229125445591039e-05,
+      "loss": 2.2752,
+      "step": 74800
+    },
+    {
+      "epoch": 10.1,
+      "learning_rate": 7.225496308539379e-05,
+      "loss": 2.2933,
+      "step": 74810
+    },
+    {
+      "epoch": 10.11,
+      "learning_rate": 7.221867793628054e-05,
+      "loss": 2.2772,
+      "step": 74820
+    },
+    {
+      "epoch": 10.11,
+      "learning_rate": 7.21823990114744e-05,
+      "loss": 2.2908,
+      "step": 74830
+    },
+    {
+      "epoch": 10.11,
+      "learning_rate": 7.214612631387842e-05,
+      "loss": 2.3003,
+      "step": 74840
+    },
+    {
+      "epoch": 10.11,
+      "learning_rate": 7.210985984639536e-05,
+      "loss": 2.2767,
+      "step": 74850
+    },
+    {
+      "epoch": 10.11,
+      "learning_rate": 7.207359961192732e-05,
+      "loss": 2.2683,
+      "step": 74860
+    },
+    {
+      "epoch": 10.11,
+      "learning_rate": 7.20373456133759e-05,
+      "loss": 2.2635,
+      "step": 74870
+    },
+    {
+      "epoch": 10.11,
+      "learning_rate": 7.200109785364239e-05,
+      "loss": 2.2833,
+      "step": 74880
+    },
+    {
+      "epoch": 10.11,
+      "learning_rate": 7.196485633562739e-05,
+      "loss": 2.2758,
+      "step": 74890
+    },
+    {
+      "epoch": 10.12,
+      "learning_rate": 7.192862106223103e-05,
+      "loss": 2.2966,
+      "step": 74900
+    },
+    {
+      "epoch": 10.12,
+      "learning_rate": 7.189239203635295e-05,
+      "loss": 2.2873,
+      "step": 74910
+    },
+    {
+      "epoch": 10.12,
+      "learning_rate": 7.185616926089235e-05,
+      "loss": 2.2645,
+      "step": 74920
+    },
+    {
+      "epoch": 10.12,
+      "learning_rate": 7.1819952738748e-05,
+      "loss": 2.2807,
+      "step": 74930
+    },
+    {
+      "epoch": 10.12,
+      "learning_rate": 7.178374247281786e-05,
+      "loss": 2.2707,
+      "step": 74940
+    },
+    {
+      "epoch": 10.12,
+      "learning_rate": 7.174753846599974e-05,
+      "loss": 2.284,
+      "step": 74950
+    },
+    {
+      "epoch": 10.12,
+      "learning_rate": 7.171134072119067e-05,
+      "loss": 2.2731,
+      "step": 74960
+    },
+    {
+      "epoch": 10.13,
+      "learning_rate": 7.167514924128745e-05,
+      "loss": 2.2863,
+      "step": 74970
+    },
+    {
+      "epoch": 10.13,
+      "learning_rate": 7.163896402918619e-05,
+      "loss": 2.2899,
+      "step": 74980
+    },
+    {
+      "epoch": 10.13,
+      "learning_rate": 7.160278508778243e-05,
+      "loss": 2.2872,
+      "step": 74990
+    },
+    {
+      "epoch": 10.13,
+      "learning_rate": 7.156661241997149e-05,
+      "loss": 2.3089,
+      "step": 75000
+    },
+    {
+      "epoch": 10.13,
+      "eval_loss": 2.4407835006713867,
+      "eval_runtime": 1269.632,
+      "eval_samples_per_second": 59.966,
+      "eval_steps_per_second": 4.998,
+      "step": 75000
+    },
+    {
+      "epoch": 10.13,
+      "learning_rate": 7.153044602864796e-05,
+      "loss": 2.272,
+      "step": 75010
+    },
+    {
+      "epoch": 10.13,
+      "learning_rate": 7.14942859167059e-05,
+      "loss": 2.2891,
+      "step": 75020
+    },
+    {
+      "epoch": 10.13,
+      "learning_rate": 7.145813208703911e-05,
+      "loss": 2.2711,
+      "step": 75030
+    },
+    {
+      "epoch": 10.13,
+      "learning_rate": 7.142198454254059e-05,
+      "loss": 2.2848,
+      "step": 75040
+    },
+    {
+      "epoch": 10.14,
+      "learning_rate": 7.13858432861031e-05,
+      "loss": 2.2882,
+      "step": 75050
+    },
+    {
+      "epoch": 10.14,
+      "learning_rate": 7.134970832061874e-05,
+      "loss": 2.2699,
+      "step": 75060
+    },
+    {
+      "epoch": 10.14,
+      "learning_rate": 7.131357964897906e-05,
+      "loss": 2.2703,
+      "step": 75070
+    },
+    {
+      "epoch": 10.14,
+      "learning_rate": 7.127745727407532e-05,
+      "loss": 2.2887,
+      "step": 75080
+    },
+    {
+      "epoch": 10.14,
+      "learning_rate": 7.124134119879807e-05,
+      "loss": 2.2635,
+      "step": 75090
+    },
+    {
+      "epoch": 10.14,
+      "learning_rate": 7.120523142603748e-05,
+      "loss": 2.2539,
+      "step": 75100
+    },
+    {
+      "epoch": 10.14,
+      "learning_rate": 7.116912795868307e-05,
+      "loss": 2.309,
+      "step": 75110
+    },
+    {
+      "epoch": 10.15,
+      "learning_rate": 7.113303079962401e-05,
+      "loss": 2.2625,
+      "step": 75120
+    },
+    {
+      "epoch": 10.15,
+      "learning_rate": 7.109693995174904e-05,
+      "loss": 2.276,
+      "step": 75130
+    },
+    {
+      "epoch": 10.15,
+      "learning_rate": 7.106085541794602e-05,
+      "loss": 2.2887,
+      "step": 75140
+    },
+    {
+      "epoch": 10.15,
+      "learning_rate": 7.102477720110275e-05,
+      "loss": 2.3016,
+      "step": 75150
+    },
+    {
+      "epoch": 10.15,
+      "learning_rate": 7.098870530410618e-05,
+      "loss": 2.2968,
+      "step": 75160
+    },
+    {
+      "epoch": 10.15,
+      "learning_rate": 7.0952639729843e-05,
+      "loss": 2.298,
+      "step": 75170
+    },
+    {
+      "epoch": 10.15,
+      "learning_rate": 7.091658048119926e-05,
+      "loss": 2.2978,
+      "step": 75180
+    },
+    {
+      "epoch": 10.16,
+      "learning_rate": 7.088052756106047e-05,
+      "loss": 2.3082,
+      "step": 75190
+    },
+    {
+      "epoch": 10.16,
+      "learning_rate": 7.084448097231182e-05,
+      "loss": 2.2648,
+      "step": 75200
+    },
+    {
+      "epoch": 10.16,
+      "learning_rate": 7.08084407178378e-05,
+      "loss": 2.3002,
+      "step": 75210
+    },
+    {
+      "epoch": 10.16,
+      "learning_rate": 7.077240680052242e-05,
+      "loss": 2.2578,
+      "step": 75220
+    },
+    {
+      "epoch": 10.16,
+      "learning_rate": 7.073637922324934e-05,
+      "loss": 2.2742,
+      "step": 75230
+    },
+    {
+      "epoch": 10.16,
+      "learning_rate": 7.070035798890147e-05,
+      "loss": 2.2786,
+      "step": 75240
+    },
+    {
+      "epoch": 10.16,
+      "learning_rate": 7.066434310036149e-05,
+      "loss": 2.2527,
+      "step": 75250
+    },
+    {
+      "epoch": 10.16,
+      "learning_rate": 7.062833456051134e-05,
+      "loss": 2.2931,
+      "step": 75260
+    },
+    {
+      "epoch": 10.17,
+      "learning_rate": 7.059233237223248e-05,
+      "loss": 2.2707,
+      "step": 75270
+    },
+    {
+      "epoch": 10.17,
+      "learning_rate": 7.055633653840605e-05,
+      "loss": 2.305,
+      "step": 75280
+    },
+    {
+      "epoch": 10.17,
+      "learning_rate": 7.05203470619125e-05,
+      "loss": 2.267,
+      "step": 75290
+    },
+    {
+      "epoch": 10.17,
+      "learning_rate": 7.048436394563178e-05,
+      "loss": 2.3066,
+      "step": 75300
+    },
+    {
+      "epoch": 10.17,
+      "learning_rate": 7.044838719244335e-05,
+      "loss": 2.2763,
+      "step": 75310
+    },
+    {
+      "epoch": 10.17,
+      "learning_rate": 7.041241680522624e-05,
+      "loss": 2.3101,
+      "step": 75320
+    },
+    {
+      "epoch": 10.17,
+      "learning_rate": 7.037645278685901e-05,
+      "loss": 2.2961,
+      "step": 75330
+    },
+    {
+      "epoch": 10.18,
+      "learning_rate": 7.034049514021941e-05,
+      "loss": 2.3077,
+      "step": 75340
+    },
+    {
+      "epoch": 10.18,
+      "learning_rate": 7.030454386818505e-05,
+      "loss": 2.2848,
+      "step": 75350
+    },
+    {
+      "epoch": 10.18,
+      "learning_rate": 7.026859897363272e-05,
+      "loss": 2.286,
+      "step": 75360
+    },
+    {
+      "epoch": 10.18,
+      "learning_rate": 7.0232660459439e-05,
+      "loss": 2.301,
+      "step": 75370
+    },
+    {
+      "epoch": 10.18,
+      "learning_rate": 7.019672832847972e-05,
+      "loss": 2.3134,
+      "step": 75380
+    },
+    {
+      "epoch": 10.18,
+      "learning_rate": 7.016080258363023e-05,
+      "loss": 2.2789,
+      "step": 75390
+    },
+    {
+      "epoch": 10.18,
+      "learning_rate": 7.012488322776555e-05,
+      "loss": 2.2761,
+      "step": 75400
+    },
+    {
+      "epoch": 10.18,
+      "learning_rate": 7.008897026375997e-05,
+      "loss": 2.2731,
+      "step": 75410
+    },
+    {
+      "epoch": 10.19,
+      "learning_rate": 7.005306369448735e-05,
+      "loss": 2.2786,
+      "step": 75420
+    },
+    {
+      "epoch": 10.19,
+      "learning_rate": 7.00171635228211e-05,
+      "loss": 2.2785,
+      "step": 75430
+    },
+    {
+      "epoch": 10.19,
+      "learning_rate": 6.998126975163401e-05,
+      "loss": 2.3212,
+      "step": 75440
+    },
+    {
+      "epoch": 10.19,
+      "learning_rate": 6.99453823837985e-05,
+      "loss": 2.2888,
+      "step": 75450
+    },
+    {
+      "epoch": 10.19,
+      "learning_rate": 6.990950142218635e-05,
+      "loss": 2.2683,
+      "step": 75460
+    },
+    {
+      "epoch": 10.19,
+      "learning_rate": 6.987362686966879e-05,
+      "loss": 2.2721,
+      "step": 75470
+    },
+    {
+      "epoch": 10.19,
+      "learning_rate": 6.983775872911673e-05,
+      "loss": 2.302,
+      "step": 75480
+    },
+    {
+      "epoch": 10.2,
+      "learning_rate": 6.98018970034004e-05,
+      "loss": 2.3044,
+      "step": 75490
+    },
+    {
+      "epoch": 10.2,
+      "learning_rate": 6.976604169538955e-05,
+      "loss": 2.2885,
+      "step": 75500
+    },
+    {
+      "epoch": 10.2,
+      "learning_rate": 6.973019280795343e-05,
+      "loss": 2.2901,
+      "step": 75510
+    },
+    {
+      "epoch": 10.2,
+      "learning_rate": 6.969435034396079e-05,
+      "loss": 2.2962,
+      "step": 75520
+    },
+    {
+      "epoch": 10.2,
+      "learning_rate": 6.965851430628e-05,
+      "loss": 2.2963,
+      "step": 75530
+    },
+    {
+      "epoch": 10.2,
+      "learning_rate": 6.962268469777851e-05,
+      "loss": 2.2809,
+      "step": 75540
+    },
+    {
+      "epoch": 10.2,
+      "learning_rate": 6.958686152132373e-05,
+      "loss": 2.2868,
+      "step": 75550
+    },
+    {
+      "epoch": 10.21,
+      "learning_rate": 6.95510447797822e-05,
+      "loss": 2.2925,
+      "step": 75560
+    },
+    {
+      "epoch": 10.21,
+      "learning_rate": 6.951523447602021e-05,
+      "loss": 2.2987,
+      "step": 75570
+    },
+    {
+      "epoch": 10.21,
+      "learning_rate": 6.947943061290338e-05,
+      "loss": 2.2927,
+      "step": 75580
+    },
+    {
+      "epoch": 10.21,
+      "learning_rate": 6.944363319329675e-05,
+      "loss": 2.257,
+      "step": 75590
+    },
+    {
+      "epoch": 10.21,
+      "learning_rate": 6.94078422200651e-05,
+      "loss": 2.276,
+      "step": 75600
+    },
+    {
+      "epoch": 10.21,
+      "learning_rate": 6.937205769607245e-05,
+      "loss": 2.2825,
+      "step": 75610
+    },
+    {
+      "epoch": 10.21,
+      "learning_rate": 6.933627962418233e-05,
+      "loss": 2.2841,
+      "step": 75620
+    },
+    {
+      "epoch": 10.21,
+      "learning_rate": 6.930050800725795e-05,
+      "loss": 2.2816,
+      "step": 75630
+    },
+    {
+      "epoch": 10.22,
+      "learning_rate": 6.926474284816175e-05,
+      "loss": 2.2938,
+      "step": 75640
+    },
+    {
+      "epoch": 10.22,
+      "learning_rate": 6.922898414975588e-05,
+      "loss": 2.2804,
+      "step": 75650
+    },
+    {
+      "epoch": 10.22,
+      "learning_rate": 6.919323191490179e-05,
+      "loss": 2.2823,
+      "step": 75660
+    },
+    {
+      "epoch": 10.22,
+      "learning_rate": 6.915748614646045e-05,
+      "loss": 2.2804,
+      "step": 75670
+    },
+    {
+      "epoch": 10.22,
+      "learning_rate": 6.912174684729246e-05,
+      "loss": 2.3074,
+      "step": 75680
+    },
+    {
+      "epoch": 10.22,
+      "learning_rate": 6.908601402025774e-05,
+      "loss": 2.2661,
+      "step": 75690
+    },
+    {
+      "epoch": 10.22,
+      "learning_rate": 6.905028766821573e-05,
+      "loss": 2.2593,
+      "step": 75700
+    },
+    {
+      "epoch": 10.23,
+      "learning_rate": 6.901456779402531e-05,
+      "loss": 2.2593,
+      "step": 75710
+    },
+    {
+      "epoch": 10.23,
+      "learning_rate": 6.897885440054497e-05,
+      "loss": 2.2788,
+      "step": 75720
+    },
+    {
+      "epoch": 10.23,
+      "learning_rate": 6.894314749063271e-05,
+      "loss": 2.2988,
+      "step": 75730
+    },
+    {
+      "epoch": 10.23,
+      "learning_rate": 6.890744706714568e-05,
+      "loss": 2.2889,
+      "step": 75740
+    },
+    {
+      "epoch": 10.23,
+      "learning_rate": 6.887175313294092e-05,
+      "loss": 2.2764,
+      "step": 75750
+    },
+    {
+      "epoch": 10.23,
+      "learning_rate": 6.883606569087466e-05,
+      "loss": 2.2834,
+      "step": 75760
+    },
+    {
+      "epoch": 10.23,
+      "learning_rate": 6.880038474380281e-05,
+      "loss": 2.2765,
+      "step": 75770
+    },
+    {
+      "epoch": 10.23,
+      "learning_rate": 6.876471029458065e-05,
+      "loss": 2.3067,
+      "step": 75780
+    },
+    {
+      "epoch": 10.24,
+      "learning_rate": 6.872904234606287e-05,
+      "loss": 2.2822,
+      "step": 75790
+    },
+    {
+      "epoch": 10.24,
+      "learning_rate": 6.869338090110389e-05,
+      "loss": 2.2782,
+      "step": 75800
+    },
+    {
+      "epoch": 10.24,
+      "learning_rate": 6.865772596255734e-05,
+      "loss": 2.2905,
+      "step": 75810
+    },
+    {
+      "epoch": 10.24,
+      "learning_rate": 6.862207753327644e-05,
+      "loss": 2.2804,
+      "step": 75820
+    },
+    {
+      "epoch": 10.24,
+      "learning_rate": 6.858643561611395e-05,
+      "loss": 2.271,
+      "step": 75830
+    },
+    {
+      "epoch": 10.24,
+      "learning_rate": 6.8550800213922e-05,
+      "loss": 2.2572,
+      "step": 75840
+    },
+    {
+      "epoch": 10.24,
+      "learning_rate": 6.851517132955229e-05,
+      "loss": 2.2733,
+      "step": 75850
+    },
+    {
+      "epoch": 10.25,
+      "learning_rate": 6.847954896585595e-05,
+      "loss": 2.3055,
+      "step": 75860
+    },
+    {
+      "epoch": 10.25,
+      "learning_rate": 6.844393312568352e-05,
+      "loss": 2.2974,
+      "step": 75870
+    },
+    {
+      "epoch": 10.25,
+      "learning_rate": 6.840832381188518e-05,
+      "loss": 2.2699,
+      "step": 75880
+    },
+    {
+      "epoch": 10.25,
+      "learning_rate": 6.837272102731049e-05,
+      "loss": 2.2937,
+      "step": 75890
+    },
+    {
+      "epoch": 10.25,
+      "learning_rate": 6.833712477480848e-05,
+      "loss": 2.2778,
+      "step": 75900
+    },
+    {
+      "epoch": 10.25,
+      "learning_rate": 6.830153505722759e-05,
+      "loss": 2.2821,
+      "step": 75910
+    },
+    {
+      "epoch": 10.25,
+      "learning_rate": 6.826595187741593e-05,
+      "loss": 2.3049,
+      "step": 75920
+    },
+    {
+      "epoch": 10.26,
+      "learning_rate": 6.823037523822103e-05,
+      "loss": 2.3026,
+      "step": 75930
+    },
+    {
+      "epoch": 10.26,
+      "learning_rate": 6.819480514248968e-05,
+      "loss": 2.2822,
+      "step": 75940
+    },
+    {
+      "epoch": 10.26,
+      "learning_rate": 6.815924159306844e-05,
+      "loss": 2.3048,
+      "step": 75950
+    },
+    {
+      "epoch": 10.26,
+      "learning_rate": 6.812368459280311e-05,
+      "loss": 2.2935,
+      "step": 75960
+    },
+    {
+      "epoch": 10.26,
+      "learning_rate": 6.808813414453922e-05,
+      "loss": 2.319,
+      "step": 75970
+    },
+    {
+      "epoch": 10.26,
+      "learning_rate": 6.805259025112152e-05,
+      "loss": 2.2787,
+      "step": 75980
+    },
+    {
+      "epoch": 10.26,
+      "learning_rate": 6.801705291539433e-05,
+      "loss": 2.2846,
+      "step": 75990
+    },
+    {
+      "epoch": 10.26,
+      "learning_rate": 6.798152214020154e-05,
+      "loss": 2.279,
+      "step": 76000
+    },
+    {
+      "epoch": 10.26,
+      "eval_loss": 2.4384381771087646,
+      "eval_runtime": 1269.5674,
+      "eval_samples_per_second": 59.969,
+      "eval_steps_per_second": 4.998,
+      "step": 76000
+    },
+    {
+      "epoch": 10.27,
+      "learning_rate": 6.794599792838641e-05,
+      "loss": 2.2895,
+      "step": 76010
+    },
+    {
+      "epoch": 10.27,
+      "learning_rate": 6.791048028279162e-05,
+      "loss": 2.2829,
+      "step": 76020
+    },
+    {
+      "epoch": 10.27,
+      "learning_rate": 6.787496920625954e-05,
+      "loss": 2.3112,
+      "step": 76030
+    },
+    {
+      "epoch": 10.27,
+      "learning_rate": 6.783946470163173e-05,
+      "loss": 2.2966,
+      "step": 76040
+    },
+    {
+      "epoch": 10.27,
+      "learning_rate": 6.78039667717496e-05,
+      "loss": 2.2717,
+      "step": 76050
+    },
+    {
+      "epoch": 10.27,
+      "learning_rate": 6.776847541945351e-05,
+      "loss": 2.2848,
+      "step": 76060
+    },
+    {
+      "epoch": 10.27,
+      "learning_rate": 6.773299064758374e-05,
+      "loss": 2.2937,
+      "step": 76070
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 6.769751245897997e-05,
+      "loss": 2.289,
+      "step": 76080
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 6.766204085648118e-05,
+      "loss": 2.2535,
+      "step": 76090
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 6.762657584292593e-05,
+      "loss": 2.281,
+      "step": 76100
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 6.75911174211522e-05,
+      "loss": 2.2784,
+      "step": 76110
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 6.755566559399755e-05,
+      "loss": 2.2903,
+      "step": 76120
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 6.752022036429902e-05,
+      "loss": 2.2918,
+      "step": 76130
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 6.748478173489286e-05,
+      "loss": 2.2799,
+      "step": 76140
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 6.744934970861513e-05,
+      "loss": 2.2952,
+      "step": 76150
+    },
+    {
+      "epoch": 10.29,
+      "learning_rate": 6.741392428830113e-05,
+      "loss": 2.2851,
+      "step": 76160
+    },
+    {
+      "epoch": 10.29,
+      "learning_rate": 6.737850547678579e-05,
+      "loss": 2.309,
+      "step": 76170
+    },
+    {
+      "epoch": 10.29,
+      "learning_rate": 6.73430932769034e-05,
+      "loss": 2.2763,
+      "step": 76180
+    },
+    {
+      "epoch": 10.29,
+      "learning_rate": 6.730768769148768e-05,
+      "loss": 2.2792,
+      "step": 76190
+    },
+    {
+      "epoch": 10.29,
+      "learning_rate": 6.727228872337202e-05,
+      "loss": 2.3108,
+      "step": 76200
+    },
+    {
+      "epoch": 10.29,
+      "learning_rate": 6.723689637538912e-05,
+      "loss": 2.2822,
+      "step": 76210
+    },
+    {
+      "epoch": 10.29,
+      "learning_rate": 6.720151065037112e-05,
+      "loss": 2.2966,
+      "step": 76220
+    },
+    {
+      "epoch": 10.3,
+      "learning_rate": 6.716613155114979e-05,
+      "loss": 2.3092,
+      "step": 76230
+    },
+    {
+      "epoch": 10.3,
+      "learning_rate": 6.713075908055617e-05,
+      "loss": 2.2931,
+      "step": 76240
+    },
+    {
+      "epoch": 10.3,
+      "learning_rate": 6.709539324142107e-05,
+      "loss": 2.2901,
+      "step": 76250
+    },
+    {
+      "epoch": 10.3,
+      "learning_rate": 6.706003403657434e-05,
+      "loss": 2.2975,
+      "step": 76260
+    },
+    {
+      "epoch": 10.3,
+      "learning_rate": 6.702468146884564e-05,
+      "loss": 2.3185,
+      "step": 76270
+    },
+    {
+      "epoch": 10.3,
+      "learning_rate": 6.698933554106406e-05,
+      "loss": 2.2842,
+      "step": 76280
+    },
+    {
+      "epoch": 10.3,
+      "learning_rate": 6.695399625605802e-05,
+      "loss": 2.2913,
+      "step": 76290
+    },
+    {
+      "epoch": 10.31,
+      "learning_rate": 6.691866361665548e-05,
+      "loss": 2.2785,
+      "step": 76300
+    },
+    {
+      "epoch": 10.31,
+      "learning_rate": 6.688333762568386e-05,
+      "loss": 2.2681,
+      "step": 76310
+    },
+    {
+      "epoch": 10.31,
+      "learning_rate": 6.684801828597005e-05,
+      "loss": 2.2691,
+      "step": 76320
+    },
+    {
+      "epoch": 10.31,
+      "learning_rate": 6.681270560034056e-05,
+      "loss": 2.2916,
+      "step": 76330
+    },
+    {
+      "epoch": 10.31,
+      "learning_rate": 6.677739957162102e-05,
+      "loss": 2.2904,
+      "step": 76340
+    },
+    {
+      "epoch": 10.31,
+      "learning_rate": 6.674210020263684e-05,
+      "loss": 2.2826,
+      "step": 76350
+    },
+    {
+      "epoch": 10.31,
+      "learning_rate": 6.670680749621273e-05,
+      "loss": 2.2954,
+      "step": 76360
+    },
+    {
+      "epoch": 10.31,
+      "learning_rate": 6.667152145517301e-05,
+      "loss": 2.2645,
+      "step": 76370
+    },
+    {
+      "epoch": 10.32,
+      "learning_rate": 6.663624208234132e-05,
+      "loss": 2.3007,
+      "step": 76380
+    },
+    {
+      "epoch": 10.32,
+      "learning_rate": 6.660096938054077e-05,
+      "loss": 2.3112,
+      "step": 76390
+    },
+    {
+      "epoch": 10.32,
+      "learning_rate": 6.656570335259414e-05,
+      "loss": 2.2731,
+      "step": 76400
+    },
+    {
+      "epoch": 10.32,
+      "learning_rate": 6.653044400132343e-05,
+      "loss": 2.2733,
+      "step": 76410
+    },
+    {
+      "epoch": 10.32,
+      "learning_rate": 6.649519132955016e-05,
+      "loss": 2.2939,
+      "step": 76420
+    },
+    {
+      "epoch": 10.32,
+      "learning_rate": 6.64599453400955e-05,
+      "loss": 2.2834,
+      "step": 76430
+    },
+    {
+      "epoch": 10.32,
+      "learning_rate": 6.64247060357798e-05,
+      "loss": 2.3035,
+      "step": 76440
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 6.638947341942319e-05,
+      "loss": 2.2923,
+      "step": 76450
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 6.63542474938449e-05,
+      "loss": 2.2919,
+      "step": 76460
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 6.631902826186389e-05,
+      "loss": 2.2661,
+      "step": 76470
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 6.62838157262986e-05,
+      "loss": 2.3292,
+      "step": 76480
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 6.62486098899668e-05,
+      "loss": 2.2783,
+      "step": 76490
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 6.621341075568573e-05,
+      "loss": 2.2618,
+      "step": 76500
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 6.617821832627213e-05,
+      "loss": 2.2797,
+      "step": 76510
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 6.614303260454223e-05,
+      "loss": 2.3157,
+      "step": 76520
+    },
+    {
+      "epoch": 10.34,
+      "learning_rate": 6.610785359331182e-05,
+      "loss": 2.275,
+      "step": 76530
+    },
+    {
+      "epoch": 10.34,
+      "learning_rate": 6.607268129539585e-05,
+      "loss": 2.3163,
+      "step": 76540
+    },
+    {
+      "epoch": 10.34,
+      "learning_rate": 6.603751571360903e-05,
+      "loss": 2.2885,
+      "step": 76550
+    },
+    {
+      "epoch": 10.34,
+      "learning_rate": 6.600235685076534e-05,
+      "loss": 2.2623,
+      "step": 76560
+    },
+    {
+      "epoch": 10.34,
+      "learning_rate": 6.59672047096784e-05,
+      "loss": 2.3004,
+      "step": 76570
+    },
+    {
+      "epoch": 10.34,
+      "learning_rate": 6.593205929316115e-05,
+      "loss": 2.2942,
+      "step": 76580
+    },
+    {
+      "epoch": 10.34,
+      "learning_rate": 6.589692060402599e-05,
+      "loss": 2.2927,
+      "step": 76590
+    },
+    {
+      "epoch": 10.35,
+      "learning_rate": 6.586178864508493e-05,
+      "loss": 2.307,
+      "step": 76600
+    },
+    {
+      "epoch": 10.35,
+      "learning_rate": 6.582666341914928e-05,
+      "loss": 2.3065,
+      "step": 76610
+    },
+    {
+      "epoch": 10.35,
+      "learning_rate": 6.579154492902982e-05,
+      "loss": 2.3072,
+      "step": 76620
+    },
+    {
+      "epoch": 10.35,
+      "learning_rate": 6.575643317753696e-05,
+      "loss": 2.2873,
+      "step": 76630
+    },
+    {
+      "epoch": 10.35,
+      "learning_rate": 6.572132816748034e-05,
+      "loss": 2.2848,
+      "step": 76640
+    },
+    {
+      "epoch": 10.35,
+      "learning_rate": 6.568622990166931e-05,
+      "loss": 2.287,
+      "step": 76650
+    },
+    {
+      "epoch": 10.35,
+      "learning_rate": 6.565113838291237e-05,
+      "loss": 2.2848,
+      "step": 76660
+    },
+    {
+      "epoch": 10.36,
+      "learning_rate": 6.561605361401774e-05,
+      "loss": 2.3058,
+      "step": 76670
+    },
+    {
+      "epoch": 10.36,
+      "learning_rate": 6.558097559779308e-05,
+      "loss": 2.3029,
+      "step": 76680
+    },
+    {
+      "epoch": 10.36,
+      "learning_rate": 6.554590433704539e-05,
+      "loss": 2.2928,
+      "step": 76690
+    },
+    {
+      "epoch": 10.36,
+      "learning_rate": 6.551083983458115e-05,
+      "loss": 2.2914,
+      "step": 76700
+    },
+    {
+      "epoch": 10.36,
+      "learning_rate": 6.547578209320632e-05,
+      "loss": 2.285,
+      "step": 76710
+    },
+    {
+      "epoch": 10.36,
+      "learning_rate": 6.544073111572643e-05,
+      "loss": 2.2699,
+      "step": 76720
+    },
+    {
+      "epoch": 10.36,
+      "learning_rate": 6.54056869049463e-05,
+      "loss": 2.2716,
+      "step": 76730
+    },
+    {
+      "epoch": 10.36,
+      "learning_rate": 6.537064946367021e-05,
+      "loss": 2.3183,
+      "step": 76740
+    },
+    {
+      "epoch": 10.37,
+      "learning_rate": 6.533561879470212e-05,
+      "loss": 2.2722,
+      "step": 76750
+    },
+    {
+      "epoch": 10.37,
+      "learning_rate": 6.530059490084517e-05,
+      "loss": 2.2961,
+      "step": 76760
+    },
+    {
+      "epoch": 10.37,
+      "learning_rate": 6.526557778490218e-05,
+      "loss": 2.2785,
+      "step": 76770
+    },
+    {
+      "epoch": 10.37,
+      "learning_rate": 6.523056744967527e-05,
+      "loss": 2.3033,
+      "step": 76780
+    },
+    {
+      "epoch": 10.37,
+      "learning_rate": 6.519556389796604e-05,
+      "loss": 2.3019,
+      "step": 76790
+    },
+    {
+      "epoch": 10.37,
+      "learning_rate": 6.51605671325757e-05,
+      "loss": 2.2826,
+      "step": 76800
+    },
+    {
+      "epoch": 10.37,
+      "learning_rate": 6.512557715630472e-05,
+      "loss": 2.2742,
+      "step": 76810
+    },
+    {
+      "epoch": 10.38,
+      "learning_rate": 6.509059397195306e-05,
+      "loss": 2.3159,
+      "step": 76820
+    },
+    {
+      "epoch": 10.38,
+      "learning_rate": 6.505561758232032e-05,
+      "loss": 2.2892,
+      "step": 76830
+    },
+    {
+      "epoch": 10.38,
+      "learning_rate": 6.502064799020529e-05,
+      "loss": 2.2933,
+      "step": 76840
+    },
+    {
+      "epoch": 10.38,
+      "learning_rate": 6.498568519840653e-05,
+      "loss": 2.2943,
+      "step": 76850
+    },
+    {
+      "epoch": 10.38,
+      "learning_rate": 6.495072920972162e-05,
+      "loss": 2.3162,
+      "step": 76860
+    },
+    {
+      "epoch": 10.38,
+      "learning_rate": 6.491578002694799e-05,
+      "loss": 2.3133,
+      "step": 76870
+    },
+    {
+      "epoch": 10.38,
+      "learning_rate": 6.488083765288242e-05,
+      "loss": 2.2815,
+      "step": 76880
+    },
+    {
+      "epoch": 10.38,
+      "learning_rate": 6.484590209032106e-05,
+      "loss": 2.2605,
+      "step": 76890
+    },
+    {
+      "epoch": 10.39,
+      "learning_rate": 6.481097334205956e-05,
+      "loss": 2.3016,
+      "step": 76900
+    },
+    {
+      "epoch": 10.39,
+      "learning_rate": 6.4776051410893e-05,
+      "loss": 2.2971,
+      "step": 76910
+    },
+    {
+      "epoch": 10.39,
+      "learning_rate": 6.474113629961603e-05,
+      "loss": 2.2624,
+      "step": 76920
+    },
+    {
+      "epoch": 10.39,
+      "learning_rate": 6.470622801102259e-05,
+      "loss": 2.2972,
+      "step": 76930
+    },
+    {
+      "epoch": 10.39,
+      "learning_rate": 6.467132654790615e-05,
+      "loss": 2.2702,
+      "step": 76940
+    },
+    {
+      "epoch": 10.39,
+      "learning_rate": 6.463643191305972e-05,
+      "loss": 2.2898,
+      "step": 76950
+    },
+    {
+      "epoch": 10.39,
+      "learning_rate": 6.460154410927554e-05,
+      "loss": 2.298,
+      "step": 76960
+    },
+    {
+      "epoch": 10.4,
+      "learning_rate": 6.456666313934559e-05,
+      "loss": 2.2669,
+      "step": 76970
+    },
+    {
+      "epoch": 10.4,
+      "learning_rate": 6.45317890060611e-05,
+      "loss": 2.2724,
+      "step": 76980
+    },
+    {
+      "epoch": 10.4,
+      "learning_rate": 6.449692171221271e-05,
+      "loss": 2.3033,
+      "step": 76990
+    },
+    {
+      "epoch": 10.4,
+      "learning_rate": 6.446206126059076e-05,
+      "loss": 2.2874,
+      "step": 77000
+    },
+    {
+      "epoch": 10.4,
+      "eval_loss": 2.435051679611206,
+      "eval_runtime": 1269.1365,
+      "eval_samples_per_second": 59.99,
+      "eval_steps_per_second": 4.999,
+      "step": 77000
+    },
+    {
+      "epoch": 10.4,
+      "learning_rate": 6.442720765398487e-05,
+      "loss": 2.3107,
+      "step": 77010
+    },
+    {
+      "epoch": 10.4,
+      "learning_rate": 6.439236089518399e-05,
+      "loss": 2.282,
+      "step": 77020
+    },
+    {
+      "epoch": 10.4,
+      "learning_rate": 6.435752098697687e-05,
+      "loss": 2.2794,
+      "step": 77030
+    },
+    {
+      "epoch": 10.41,
+      "learning_rate": 6.432268793215135e-05,
+      "loss": 2.2946,
+      "step": 77040
+    },
+    {
+      "epoch": 10.41,
+      "learning_rate": 6.428786173349504e-05,
+      "loss": 2.2772,
+      "step": 77050
+    },
+    {
+      "epoch": 10.41,
+      "learning_rate": 6.425304239379463e-05,
+      "loss": 2.2728,
+      "step": 77060
+    },
+    {
+      "epoch": 10.41,
+      "learning_rate": 6.421822991583663e-05,
+      "loss": 2.2928,
+      "step": 77070
+    },
+    {
+      "epoch": 10.41,
+      "learning_rate": 6.418342430240683e-05,
+      "loss": 2.2939,
+      "step": 77080
+    },
+    {
+      "epoch": 10.41,
+      "learning_rate": 6.414862555629048e-05,
+      "loss": 2.2903,
+      "step": 77090
+    },
+    {
+      "epoch": 10.41,
+      "learning_rate": 6.411383368027225e-05,
+      "loss": 2.2955,
+      "step": 77100
+    },
+    {
+      "epoch": 10.41,
+      "learning_rate": 6.407904867713626e-05,
+      "loss": 2.2993,
+      "step": 77110
+    },
+    {
+      "epoch": 10.42,
+      "learning_rate": 6.404427054966622e-05,
+      "loss": 2.271,
+      "step": 77120
+    },
+    {
+      "epoch": 10.42,
+      "learning_rate": 6.400949930064514e-05,
+      "loss": 2.2961,
+      "step": 77130
+    },
+    {
+      "epoch": 10.42,
+      "learning_rate": 6.397473493285545e-05,
+      "loss": 2.3345,
+      "step": 77140
+    },
+    {
+      "epoch": 10.42,
+      "learning_rate": 6.393997744907925e-05,
+      "loss": 2.2854,
+      "step": 77150
+    },
+    {
+      "epoch": 10.42,
+      "learning_rate": 6.390522685209781e-05,
+      "loss": 2.2907,
+      "step": 77160
+    },
+    {
+      "epoch": 10.42,
+      "learning_rate": 6.387048314469207e-05,
+      "loss": 2.2951,
+      "step": 77170
+    },
+    {
+      "epoch": 10.42,
+      "learning_rate": 6.383574632964231e-05,
+      "loss": 2.2926,
+      "step": 77180
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 6.380101640972823e-05,
+      "loss": 2.2937,
+      "step": 77190
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 6.376629338772912e-05,
+      "loss": 2.2831,
+      "step": 77200
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 6.373157726642355e-05,
+      "loss": 2.2804,
+      "step": 77210
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 6.36968680485896e-05,
+      "loss": 2.2992,
+      "step": 77220
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 6.36621657370049e-05,
+      "loss": 2.2946,
+      "step": 77230
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 6.362747033444635e-05,
+      "loss": 2.2904,
+      "step": 77240
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 6.359278184369052e-05,
+      "loss": 2.3105,
+      "step": 77250
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 6.355810026751309e-05,
+      "loss": 2.2957,
+      "step": 77260
+    },
+    {
+      "epoch": 10.44,
+      "learning_rate": 6.352342560868947e-05,
+      "loss": 2.2827,
+      "step": 77270
+    },
+    {
+      "epoch": 10.44,
+      "learning_rate": 6.348875786999454e-05,
+      "loss": 2.2912,
+      "step": 77280
+    },
+    {
+      "epoch": 10.44,
+      "learning_rate": 6.345409705420244e-05,
+      "loss": 2.2772,
+      "step": 77290
+    },
+    {
+      "epoch": 10.44,
+      "learning_rate": 6.341944316408685e-05,
+      "loss": 2.3104,
+      "step": 77300
+    },
+    {
+      "epoch": 10.44,
+      "learning_rate": 6.33847962024208e-05,
+      "loss": 2.2841,
+      "step": 77310
+    },
+    {
+      "epoch": 10.44,
+      "learning_rate": 6.335015617197698e-05,
+      "loss": 2.312,
+      "step": 77320
+    },
+    {
+      "epoch": 10.44,
+      "learning_rate": 6.331552307552734e-05,
+      "loss": 2.3048,
+      "step": 77330
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 6.328089691584328e-05,
+      "loss": 2.3141,
+      "step": 77340
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 6.324627769569578e-05,
+      "loss": 2.3042,
+      "step": 77350
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 6.321166541785512e-05,
+      "loss": 2.2926,
+      "step": 77360
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 6.317706008509113e-05,
+      "loss": 2.2865,
+      "step": 77370
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 6.314246170017302e-05,
+      "loss": 2.2755,
+      "step": 77380
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 6.310787026586942e-05,
+      "loss": 2.2796,
+      "step": 77390
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 6.307328578494852e-05,
+      "loss": 2.2872,
+      "step": 77400
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 6.303870826017784e-05,
+      "loss": 2.2898,
+      "step": 77410
+    },
+    {
+      "epoch": 10.46,
+      "learning_rate": 6.300413769432433e-05,
+      "loss": 2.2934,
+      "step": 77420
+    },
+    {
+      "epoch": 10.46,
+      "learning_rate": 6.296957409015455e-05,
+      "loss": 2.2928,
+      "step": 77430
+    },
+    {
+      "epoch": 10.46,
+      "learning_rate": 6.293501745043427e-05,
+      "loss": 2.2736,
+      "step": 77440
+    },
+    {
+      "epoch": 10.46,
+      "learning_rate": 6.290046777792899e-05,
+      "loss": 2.2939,
+      "step": 77450
+    },
+    {
+      "epoch": 10.46,
+      "learning_rate": 6.286592507540326e-05,
+      "loss": 2.2945,
+      "step": 77460
+    },
+    {
+      "epoch": 10.46,
+      "learning_rate": 6.283138934562141e-05,
+      "loss": 2.2923,
+      "step": 77470
+    },
+    {
+      "epoch": 10.46,
+      "learning_rate": 6.279686059134717e-05,
+      "loss": 2.3005,
+      "step": 77480
+    },
+    {
+      "epoch": 10.47,
+      "learning_rate": 6.276233881534356e-05,
+      "loss": 2.2809,
+      "step": 77490
+    },
+    {
+      "epoch": 10.47,
+      "learning_rate": 6.272782402037312e-05,
+      "loss": 2.2752,
+      "step": 77500
+    },
+    {
+      "epoch": 10.47,
+      "learning_rate": 6.269331620919779e-05,
+      "loss": 2.2721,
+      "step": 77510
+    },
+    {
+      "epoch": 10.47,
+      "learning_rate": 6.265881538457909e-05,
+      "loss": 2.2643,
+      "step": 77520
+    },
+    {
+      "epoch": 10.47,
+      "learning_rate": 6.262432154927786e-05,
+      "loss": 2.2974,
+      "step": 77530
+    },
+    {
+      "epoch": 10.47,
+      "learning_rate": 6.258983470605434e-05,
+      "loss": 2.2949,
+      "step": 77540
+    },
+    {
+      "epoch": 10.47,
+      "learning_rate": 6.255535485766836e-05,
+      "loss": 2.2999,
+      "step": 77550
+    },
+    {
+      "epoch": 10.48,
+      "learning_rate": 6.252088200687902e-05,
+      "loss": 2.35,
+      "step": 77560
+    },
+    {
+      "epoch": 10.48,
+      "learning_rate": 6.248641615644505e-05,
+      "loss": 2.3174,
+      "step": 77570
+    },
+    {
+      "epoch": 10.48,
+      "learning_rate": 6.245195730912446e-05,
+      "loss": 2.3191,
+      "step": 77580
+    },
+    {
+      "epoch": 10.48,
+      "learning_rate": 6.241750546767471e-05,
+      "loss": 2.3094,
+      "step": 77590
+    },
+    {
+      "epoch": 10.48,
+      "learning_rate": 6.238306063485284e-05,
+      "loss": 2.2964,
+      "step": 77600
+    },
+    {
+      "epoch": 10.48,
+      "learning_rate": 6.234862281341519e-05,
+      "loss": 2.3059,
+      "step": 77610
+    },
+    {
+      "epoch": 10.48,
+      "learning_rate": 6.231419200611752e-05,
+      "loss": 2.3062,
+      "step": 77620
+    },
+    {
+      "epoch": 10.48,
+      "learning_rate": 6.227976821571522e-05,
+      "loss": 2.264,
+      "step": 77630
+    },
+    {
+      "epoch": 10.49,
+      "learning_rate": 6.224535144496286e-05,
+      "loss": 2.3142,
+      "step": 77640
+    },
+    {
+      "epoch": 10.49,
+      "learning_rate": 6.221094169661476e-05,
+      "loss": 2.2902,
+      "step": 77650
+    },
+    {
+      "epoch": 10.49,
+      "learning_rate": 6.217653897342425e-05,
+      "loss": 2.2781,
+      "step": 77660
+    },
+    {
+      "epoch": 10.49,
+      "learning_rate": 6.21421432781445e-05,
+      "loss": 2.2672,
+      "step": 77670
+    },
+    {
+      "epoch": 10.49,
+      "learning_rate": 6.210775461352798e-05,
+      "loss": 2.2872,
+      "step": 77680
+    },
+    {
+      "epoch": 10.49,
+      "learning_rate": 6.207337298232653e-05,
+      "loss": 2.2814,
+      "step": 77690
+    },
+    {
+      "epoch": 10.49,
+      "learning_rate": 6.203899838729148e-05,
+      "loss": 2.302,
+      "step": 77700
+    },
+    {
+      "epoch": 10.5,
+      "learning_rate": 6.200463083117355e-05,
+      "loss": 2.3124,
+      "step": 77710
+    },
+    {
+      "epoch": 10.5,
+      "learning_rate": 6.197027031672303e-05,
+      "loss": 2.2752,
+      "step": 77720
+    },
+    {
+      "epoch": 10.5,
+      "learning_rate": 6.193591684668952e-05,
+      "loss": 2.2752,
+      "step": 77730
+    },
+    {
+      "epoch": 10.5,
+      "learning_rate": 6.190157042382202e-05,
+      "loss": 2.2728,
+      "step": 77740
+    },
+    {
+      "epoch": 10.5,
+      "learning_rate": 6.186723105086917e-05,
+      "loss": 2.2881,
+      "step": 77750
+    },
+    {
+      "epoch": 10.5,
+      "learning_rate": 6.18328987305788e-05,
+      "loss": 2.2903,
+      "step": 77760
+    },
+    {
+      "epoch": 10.5,
+      "learning_rate": 6.179857346569838e-05,
+      "loss": 2.2977,
+      "step": 77770
+    },
+    {
+      "epoch": 10.5,
+      "learning_rate": 6.176425525897471e-05,
+      "loss": 2.2901,
+      "step": 77780
+    },
+    {
+      "epoch": 10.51,
+      "learning_rate": 6.172994411315394e-05,
+      "loss": 2.2786,
+      "step": 77790
+    },
+    {
+      "epoch": 10.51,
+      "learning_rate": 6.169564003098189e-05,
+      "loss": 2.2717,
+      "step": 77800
+    },
+    {
+      "epoch": 10.51,
+      "learning_rate": 6.166134301520363e-05,
+      "loss": 2.3026,
+      "step": 77810
+    },
+    {
+      "epoch": 10.51,
+      "learning_rate": 6.162705306856366e-05,
+      "loss": 2.3037,
+      "step": 77820
+    },
+    {
+      "epoch": 10.51,
+      "learning_rate": 6.159277019380609e-05,
+      "loss": 2.3024,
+      "step": 77830
+    },
+    {
+      "epoch": 10.51,
+      "learning_rate": 6.15584943936742e-05,
+      "loss": 2.2744,
+      "step": 77840
+    },
+    {
+      "epoch": 10.51,
+      "learning_rate": 6.152422567091104e-05,
+      "loss": 2.2887,
+      "step": 77850
+    },
+    {
+      "epoch": 10.52,
+      "learning_rate": 6.148996402825866e-05,
+      "loss": 2.2872,
+      "step": 77860
+    },
+    {
+      "epoch": 10.52,
+      "learning_rate": 6.145570946845892e-05,
+      "loss": 2.2682,
+      "step": 77870
+    },
+    {
+      "epoch": 10.52,
+      "learning_rate": 6.142146199425301e-05,
+      "loss": 2.2777,
+      "step": 77880
+    },
+    {
+      "epoch": 10.52,
+      "learning_rate": 6.138722160838151e-05,
+      "loss": 2.2728,
+      "step": 77890
+    },
+    {
+      "epoch": 10.52,
+      "learning_rate": 6.135298831358438e-05,
+      "loss": 2.2684,
+      "step": 77900
+    },
+    {
+      "epoch": 10.52,
+      "learning_rate": 6.131876211260107e-05,
+      "loss": 2.2897,
+      "step": 77910
+    },
+    {
+      "epoch": 10.52,
+      "learning_rate": 6.128454300817055e-05,
+      "loss": 2.272,
+      "step": 77920
+    },
+    {
+      "epoch": 10.53,
+      "learning_rate": 6.125033100303104e-05,
+      "loss": 2.3143,
+      "step": 77930
+    },
+    {
+      "epoch": 10.53,
+      "learning_rate": 6.121612609992043e-05,
+      "loss": 2.2875,
+      "step": 77940
+    },
+    {
+      "epoch": 10.53,
+      "learning_rate": 6.11819283015758e-05,
+      "loss": 2.3092,
+      "step": 77950
+    },
+    {
+      "epoch": 10.53,
+      "learning_rate": 6.114773761073374e-05,
+      "loss": 2.3021,
+      "step": 77960
+    },
+    {
+      "epoch": 10.53,
+      "learning_rate": 6.11135540301304e-05,
+      "loss": 2.2813,
+      "step": 77970
+    },
+    {
+      "epoch": 10.53,
+      "learning_rate": 6.107937756250122e-05,
+      "loss": 2.2903,
+      "step": 77980
+    },
+    {
+      "epoch": 10.53,
+      "learning_rate": 6.104520821058103e-05,
+      "loss": 2.2945,
+      "step": 77990
+    },
+    {
+      "epoch": 10.53,
+      "learning_rate": 6.101104597710428e-05,
+      "loss": 2.274,
+      "step": 78000
+    },
+    {
+      "epoch": 10.53,
+      "eval_loss": 2.43227481842041,
+      "eval_runtime": 1269.5204,
+      "eval_samples_per_second": 59.971,
+      "eval_steps_per_second": 4.998,
+      "step": 78000
+    },
+    {
+      "epoch": 10.54,
+      "learning_rate": 6.0976890864804656e-05,
+      "loss": 2.295,
+      "step": 78010
+    },
+    {
+      "epoch": 10.54,
+      "learning_rate": 6.094274287641544e-05,
+      "loss": 2.308,
+      "step": 78020
+    },
+    {
+      "epoch": 10.54,
+      "learning_rate": 6.0908602014669235e-05,
+      "loss": 2.2972,
+      "step": 78030
+    },
+    {
+      "epoch": 10.54,
+      "learning_rate": 6.0874468282298016e-05,
+      "loss": 2.2975,
+      "step": 78040
+    },
+    {
+      "epoch": 10.54,
+      "learning_rate": 6.084034168203339e-05,
+      "loss": 2.2966,
+      "step": 78050
+    },
+    {
+      "epoch": 10.54,
+      "learning_rate": 6.0806222216606214e-05,
+      "loss": 2.3113,
+      "step": 78060
+    },
+    {
+      "epoch": 10.54,
+      "learning_rate": 6.077210988874686e-05,
+      "loss": 2.3093,
+      "step": 78070
+    },
+    {
+      "epoch": 10.55,
+      "learning_rate": 6.0738004701185046e-05,
+      "loss": 2.2944,
+      "step": 78080
+    },
+    {
+      "epoch": 10.55,
+      "learning_rate": 6.0703906656649997e-05,
+      "loss": 2.2833,
+      "step": 78090
+    },
+    {
+      "epoch": 10.55,
+      "learning_rate": 6.0669815757870506e-05,
+      "loss": 2.2676,
+      "step": 78100
+    },
+    {
+      "epoch": 10.55,
+      "learning_rate": 6.0635732007574366e-05,
+      "loss": 2.268,
+      "step": 78110
+    },
+    {
+      "epoch": 10.55,
+      "learning_rate": 6.060165540848927e-05,
+      "loss": 2.2755,
+      "step": 78120
+    },
+    {
+      "epoch": 10.55,
+      "learning_rate": 6.0567585963341994e-05,
+      "loss": 2.2766,
+      "step": 78130
+    },
+    {
+      "epoch": 10.55,
+      "learning_rate": 6.053352367485902e-05,
+      "loss": 2.2857,
+      "step": 78140
+    },
+    {
+      "epoch": 10.55,
+      "learning_rate": 6.0499468545766054e-05,
+      "loss": 2.2834,
+      "step": 78150
+    },
+    {
+      "epoch": 10.56,
+      "learning_rate": 6.0465420578788225e-05,
+      "loss": 2.3107,
+      "step": 78160
+    },
+    {
+      "epoch": 10.56,
+      "learning_rate": 6.0431379776650304e-05,
+      "loss": 2.2658,
+      "step": 78170
+    },
+    {
+      "epoch": 10.56,
+      "learning_rate": 6.0397346142076256e-05,
+      "loss": 2.2906,
+      "step": 78180
+    },
+    {
+      "epoch": 10.56,
+      "learning_rate": 6.03633196777895e-05,
+      "loss": 2.2933,
+      "step": 78190
+    },
+    {
+      "epoch": 10.56,
+      "learning_rate": 6.0329300386513086e-05,
+      "loss": 2.2867,
+      "step": 78200
+    },
+    {
+      "epoch": 10.56,
+      "learning_rate": 6.0295288270969213e-05,
+      "loss": 2.3138,
+      "step": 78210
+    },
+    {
+      "epoch": 10.56,
+      "learning_rate": 6.026128333387975e-05,
+      "loss": 2.2977,
+      "step": 78220
+    },
+    {
+      "epoch": 10.57,
+      "learning_rate": 6.022728557796583e-05,
+      "loss": 2.265,
+      "step": 78230
+    },
+    {
+      "epoch": 10.57,
+      "learning_rate": 6.0193295005947985e-05,
+      "loss": 2.2741,
+      "step": 78240
+    },
+    {
+      "epoch": 10.57,
+      "learning_rate": 6.015931162054639e-05,
+      "loss": 2.2933,
+      "step": 78250
+    },
+    {
+      "epoch": 10.57,
+      "learning_rate": 6.0125335424480425e-05,
+      "loss": 2.2991,
+      "step": 78260
+    },
+    {
+      "epoch": 10.57,
+      "learning_rate": 6.009136642046898e-05,
+      "loss": 2.2717,
+      "step": 78270
+    },
+    {
+      "epoch": 10.57,
+      "learning_rate": 6.005740461123031e-05,
+      "loss": 2.3055,
+      "step": 78280
+    },
+    {
+      "epoch": 10.57,
+      "learning_rate": 6.0023449999482194e-05,
+      "loss": 2.2672,
+      "step": 78290
+    },
+    {
+      "epoch": 10.58,
+      "learning_rate": 5.9989502587941904e-05,
+      "loss": 2.2768,
+      "step": 78300
+    },
+    {
+      "epoch": 10.58,
+      "learning_rate": 5.99555623793258e-05,
+      "loss": 2.31,
+      "step": 78310
+    },
+    {
+      "epoch": 10.58,
+      "learning_rate": 5.992162937635005e-05,
+      "loss": 2.2949,
+      "step": 78320
+    },
+    {
+      "epoch": 10.58,
+      "learning_rate": 5.988770358172995e-05,
+      "loss": 2.2786,
+      "step": 78330
+    },
+    {
+      "epoch": 10.58,
+      "learning_rate": 5.985378499818048e-05,
+      "loss": 2.2778,
+      "step": 78340
+    },
+    {
+      "epoch": 10.58,
+      "learning_rate": 5.981987362841586e-05,
+      "loss": 2.2803,
+      "step": 78350
+    },
+    {
+      "epoch": 10.58,
+      "learning_rate": 5.978596947514971e-05,
+      "loss": 2.2771,
+      "step": 78360
+    },
+    {
+      "epoch": 10.58,
+      "learning_rate": 5.975207254109529e-05,
+      "loss": 2.2877,
+      "step": 78370
+    },
+    {
+      "epoch": 10.59,
+      "learning_rate": 5.971818282896505e-05,
+      "loss": 2.3114,
+      "step": 78380
+    },
+    {
+      "epoch": 10.59,
+      "learning_rate": 5.968430034147092e-05,
+      "loss": 2.2926,
+      "step": 78390
+    },
+    {
+      "epoch": 10.59,
+      "learning_rate": 5.9650425081324374e-05,
+      "loss": 2.2887,
+      "step": 78400
+    },
+    {
+      "epoch": 10.59,
+      "learning_rate": 5.961655705123612e-05,
+      "loss": 2.2973,
+      "step": 78410
+    },
+    {
+      "epoch": 10.59,
+      "learning_rate": 5.958269625391649e-05,
+      "loss": 2.2772,
+      "step": 78420
+    },
+    {
+      "epoch": 10.59,
+      "learning_rate": 5.954884269207509e-05,
+      "loss": 2.2997,
+      "step": 78430
+    },
+    {
+      "epoch": 10.59,
+      "learning_rate": 5.9514996368420915e-05,
+      "loss": 2.2547,
+      "step": 78440
+    },
+    {
+      "epoch": 10.6,
+      "learning_rate": 5.948115728566259e-05,
+      "loss": 2.2791,
+      "step": 78450
+    },
+    {
+      "epoch": 10.6,
+      "learning_rate": 5.9447325446507934e-05,
+      "loss": 2.2986,
+      "step": 78460
+    },
+    {
+      "epoch": 10.6,
+      "learning_rate": 5.941350085366431e-05,
+      "loss": 2.2761,
+      "step": 78470
+    },
+    {
+      "epoch": 10.6,
+      "learning_rate": 5.937968350983839e-05,
+      "loss": 2.2805,
+      "step": 78480
+    },
+    {
+      "epoch": 10.6,
+      "learning_rate": 5.9345873417736426e-05,
+      "loss": 2.2661,
+      "step": 78490
+    },
+    {
+      "epoch": 10.6,
+      "learning_rate": 5.9312070580064095e-05,
+      "loss": 2.2993,
+      "step": 78500
+    },
+    {
+      "epoch": 10.6,
+      "learning_rate": 5.92782749995262e-05,
+      "loss": 2.2781,
+      "step": 78510
+    },
+    {
+      "epoch": 10.6,
+      "learning_rate": 5.9244486678827315e-05,
+      "loss": 2.2731,
+      "step": 78520
+    },
+    {
+      "epoch": 10.61,
+      "learning_rate": 5.921070562067121e-05,
+      "loss": 2.2779,
+      "step": 78530
+    },
+    {
+      "epoch": 10.61,
+      "learning_rate": 5.917693182776123e-05,
+      "loss": 2.2961,
+      "step": 78540
+    },
+    {
+      "epoch": 10.61,
+      "learning_rate": 5.9143165302800026e-05,
+      "loss": 2.3046,
+      "step": 78550
+    },
+    {
+      "epoch": 10.61,
+      "learning_rate": 5.910940604848966e-05,
+      "loss": 2.2645,
+      "step": 78560
+    },
+    {
+      "epoch": 10.61,
+      "learning_rate": 5.9075654067531716e-05,
+      "loss": 2.2972,
+      "step": 78570
+    },
+    {
+      "epoch": 10.61,
+      "learning_rate": 5.904190936262713e-05,
+      "loss": 2.2676,
+      "step": 78580
+    },
+    {
+      "epoch": 10.61,
+      "learning_rate": 5.900817193647619e-05,
+      "loss": 2.3157,
+      "step": 78590
+    },
+    {
+      "epoch": 10.62,
+      "learning_rate": 5.8974441791778746e-05,
+      "loss": 2.2877,
+      "step": 78600
+    },
+    {
+      "epoch": 10.62,
+      "learning_rate": 5.8940718931233944e-05,
+      "loss": 2.2691,
+      "step": 78610
+    },
+    {
+      "epoch": 10.62,
+      "learning_rate": 5.8907003357540446e-05,
+      "loss": 2.2873,
+      "step": 78620
+    },
+    {
+      "epoch": 10.62,
+      "learning_rate": 5.887329507339626e-05,
+      "loss": 2.2925,
+      "step": 78630
+    },
+    {
+      "epoch": 10.62,
+      "learning_rate": 5.883959408149877e-05,
+      "loss": 2.278,
+      "step": 78640
+    },
+    {
+      "epoch": 10.62,
+      "learning_rate": 5.880590038454492e-05,
+      "loss": 2.303,
+      "step": 78650
+    },
+    {
+      "epoch": 10.62,
+      "learning_rate": 5.877221398523095e-05,
+      "loss": 2.3034,
+      "step": 78660
+    },
+    {
+      "epoch": 10.63,
+      "learning_rate": 5.8738534886252544e-05,
+      "loss": 2.3074,
+      "step": 78670
+    },
+    {
+      "epoch": 10.63,
+      "learning_rate": 5.870486309030478e-05,
+      "loss": 2.3157,
+      "step": 78680
+    },
+    {
+      "epoch": 10.63,
+      "learning_rate": 5.8671198600082216e-05,
+      "loss": 2.2929,
+      "step": 78690
+    },
+    {
+      "epoch": 10.63,
+      "learning_rate": 5.863754141827889e-05,
+      "loss": 2.3178,
+      "step": 78700
+    },
+    {
+      "epoch": 10.63,
+      "learning_rate": 5.860389154758796e-05,
+      "loss": 2.296,
+      "step": 78710
+    },
+    {
+      "epoch": 10.63,
+      "learning_rate": 5.857024899070235e-05,
+      "loss": 2.2602,
+      "step": 78720
+    },
+    {
+      "epoch": 10.63,
+      "learning_rate": 5.853661375031414e-05,
+      "loss": 2.2957,
+      "step": 78730
+    },
+    {
+      "epoch": 10.63,
+      "learning_rate": 5.850298582911502e-05,
+      "loss": 2.3065,
+      "step": 78740
+    },
+    {
+      "epoch": 10.64,
+      "learning_rate": 5.846936522979595e-05,
+      "loss": 2.2874,
+      "step": 78750
+    },
+    {
+      "epoch": 10.64,
+      "learning_rate": 5.8435751955047314e-05,
+      "loss": 2.2855,
+      "step": 78760
+    },
+    {
+      "epoch": 10.64,
+      "learning_rate": 5.8402146007559065e-05,
+      "loss": 2.2957,
+      "step": 78770
+    },
+    {
+      "epoch": 10.64,
+      "learning_rate": 5.836854739002039e-05,
+      "loss": 2.2741,
+      "step": 78780
+    },
+    {
+      "epoch": 10.64,
+      "learning_rate": 5.8334956105119915e-05,
+      "loss": 2.295,
+      "step": 78790
+    },
+    {
+      "epoch": 10.64,
+      "learning_rate": 5.8301372155545824e-05,
+      "loss": 2.2828,
+      "step": 78800
+    },
+    {
+      "epoch": 10.64,
+      "learning_rate": 5.826779554398549e-05,
+      "loss": 2.316,
+      "step": 78810
+    },
+    {
+      "epoch": 10.65,
+      "learning_rate": 5.823422627312594e-05,
+      "loss": 2.2687,
+      "step": 78820
+    },
+    {
+      "epoch": 10.65,
+      "learning_rate": 5.820066434565344e-05,
+      "loss": 2.291,
+      "step": 78830
+    },
+    {
+      "epoch": 10.65,
+      "learning_rate": 5.816710976425365e-05,
+      "loss": 2.2914,
+      "step": 78840
+    },
+    {
+      "epoch": 10.65,
+      "learning_rate": 5.813356253161185e-05,
+      "loss": 2.2706,
+      "step": 78850
+    },
+    {
+      "epoch": 10.65,
+      "learning_rate": 5.810002265041252e-05,
+      "loss": 2.2933,
+      "step": 78860
+    },
+    {
+      "epoch": 10.65,
+      "learning_rate": 5.806649012333965e-05,
+      "loss": 2.2978,
+      "step": 78870
+    },
+    {
+      "epoch": 10.65,
+      "learning_rate": 5.803296495307654e-05,
+      "loss": 2.3164,
+      "step": 78880
+    },
+    {
+      "epoch": 10.65,
+      "learning_rate": 5.799944714230604e-05,
+      "loss": 2.2892,
+      "step": 78890
+    },
+    {
+      "epoch": 10.66,
+      "learning_rate": 5.796593669371047e-05,
+      "loss": 2.296,
+      "step": 78900
+    },
+    {
+      "epoch": 10.66,
+      "learning_rate": 5.79324336099712e-05,
+      "loss": 2.2453,
+      "step": 78910
+    },
+    {
+      "epoch": 10.66,
+      "learning_rate": 5.789893789376945e-05,
+      "loss": 2.2795,
+      "step": 78920
+    },
+    {
+      "epoch": 10.66,
+      "learning_rate": 5.786544954778553e-05,
+      "loss": 2.3039,
+      "step": 78930
+    },
+    {
+      "epoch": 10.66,
+      "learning_rate": 5.783196857469937e-05,
+      "loss": 2.3022,
+      "step": 78940
+    },
+    {
+      "epoch": 10.66,
+      "learning_rate": 5.7798494977190214e-05,
+      "loss": 2.2918,
+      "step": 78950
+    },
+    {
+      "epoch": 10.66,
+      "learning_rate": 5.7765028757936625e-05,
+      "loss": 2.2881,
+      "step": 78960
+    },
+    {
+      "epoch": 10.67,
+      "learning_rate": 5.773156991961682e-05,
+      "loss": 2.2795,
+      "step": 78970
+    },
+    {
+      "epoch": 10.67,
+      "learning_rate": 5.769811846490819e-05,
+      "loss": 2.2911,
+      "step": 78980
+    },
+    {
+      "epoch": 10.67,
+      "learning_rate": 5.766467439648762e-05,
+      "loss": 2.2898,
+      "step": 78990
+    },
+    {
+      "epoch": 10.67,
+      "learning_rate": 5.7631237717031475e-05,
+      "loss": 2.2707,
+      "step": 79000
+    },
+    {
+      "epoch": 10.67,
+      "eval_loss": 2.4284002780914307,
+      "eval_runtime": 1269.6161,
+      "eval_samples_per_second": 59.967,
+      "eval_steps_per_second": 4.998,
+      "step": 79000
+    },
+    {
+      "epoch": 10.67,
+      "learning_rate": 5.759780842921537e-05,
+      "loss": 2.2848,
+      "step": 79010
+    },
+    {
+      "epoch": 10.67,
+      "learning_rate": 5.756438653571453e-05,
+      "loss": 2.29,
+      "step": 79020
+    },
+    {
+      "epoch": 10.67,
+      "learning_rate": 5.7530972039203435e-05,
+      "loss": 2.3164,
+      "step": 79030
+    },
+    {
+      "epoch": 10.68,
+      "learning_rate": 5.749756494235595e-05,
+      "loss": 2.2912,
+      "step": 79040
+    },
+    {
+      "epoch": 10.68,
+      "learning_rate": 5.746416524784554e-05,
+      "loss": 2.2782,
+      "step": 79050
+    },
+    {
+      "epoch": 10.68,
+      "learning_rate": 5.74307729583449e-05,
+      "loss": 2.3148,
+      "step": 79060
+    },
+    {
+      "epoch": 10.68,
+      "learning_rate": 5.739738807652616e-05,
+      "loss": 2.2975,
+      "step": 79070
+    },
+    {
+      "epoch": 10.68,
+      "learning_rate": 5.7364010605060865e-05,
+      "loss": 2.2501,
+      "step": 79080
+    },
+    {
+      "epoch": 10.68,
+      "learning_rate": 5.7330640546620006e-05,
+      "loss": 2.2705,
+      "step": 79090
+    },
+    {
+      "epoch": 10.68,
+      "learning_rate": 5.729727790387409e-05,
+      "loss": 2.3016,
+      "step": 79100
+    },
+    {
+      "epoch": 10.68,
+      "learning_rate": 5.726392267949268e-05,
+      "loss": 2.304,
+      "step": 79110
+    },
+    {
+      "epoch": 10.69,
+      "learning_rate": 5.723057487614514e-05,
+      "loss": 2.2975,
+      "step": 79120
+    },
+    {
+      "epoch": 10.69,
+      "learning_rate": 5.719723449649994e-05,
+      "loss": 2.2867,
+      "step": 79130
+    },
+    {
+      "epoch": 10.69,
+      "learning_rate": 5.716390154322521e-05,
+      "loss": 2.2845,
+      "step": 79140
+    },
+    {
+      "epoch": 10.69,
+      "learning_rate": 5.7130576018988276e-05,
+      "loss": 2.2987,
+      "step": 79150
+    },
+    {
+      "epoch": 10.69,
+      "learning_rate": 5.709725792645592e-05,
+      "loss": 2.282,
+      "step": 79160
+    },
+    {
+      "epoch": 10.69,
+      "learning_rate": 5.7063947268294463e-05,
+      "loss": 2.2686,
+      "step": 79170
+    },
+    {
+      "epoch": 10.69,
+      "learning_rate": 5.703064404716947e-05,
+      "loss": 2.2966,
+      "step": 79180
+    },
+    {
+      "epoch": 10.7,
+      "learning_rate": 5.699734826574594e-05,
+      "loss": 2.2845,
+      "step": 79190
+    },
+    {
+      "epoch": 10.7,
+      "learning_rate": 5.696405992668837e-05,
+      "loss": 2.2971,
+      "step": 79200
+    },
+    {
+      "epoch": 10.7,
+      "learning_rate": 5.6930779032660536e-05,
+      "loss": 2.2675,
+      "step": 79210
+    },
+    {
+      "epoch": 10.7,
+      "learning_rate": 5.689750558632579e-05,
+      "loss": 2.2866,
+      "step": 79220
+    },
+    {
+      "epoch": 10.7,
+      "learning_rate": 5.6864239590346636e-05,
+      "loss": 2.2823,
+      "step": 79230
+    },
+    {
+      "epoch": 10.7,
+      "learning_rate": 5.683098104738516e-05,
+      "loss": 2.2909,
+      "step": 79240
+    },
+    {
+      "epoch": 10.7,
+      "learning_rate": 5.6797729960102914e-05,
+      "loss": 2.2884,
+      "step": 79250
+    },
+    {
+      "epoch": 10.7,
+      "learning_rate": 5.676448633116067e-05,
+      "loss": 2.2737,
+      "step": 79260
+    },
+    {
+      "epoch": 10.71,
+      "learning_rate": 5.673125016321872e-05,
+      "loss": 2.3109,
+      "step": 79270
+    },
+    {
+      "epoch": 10.71,
+      "learning_rate": 5.669802145893666e-05,
+      "loss": 2.2788,
+      "step": 79280
+    },
+    {
+      "epoch": 10.71,
+      "learning_rate": 5.666480022097361e-05,
+      "loss": 2.2852,
+      "step": 79290
+    },
+    {
+      "epoch": 10.71,
+      "learning_rate": 5.6631586451988134e-05,
+      "loss": 2.2868,
+      "step": 79300
+    },
+    {
+      "epoch": 10.71,
+      "learning_rate": 5.6598380154637915e-05,
+      "loss": 2.2887,
+      "step": 79310
+    },
+    {
+      "epoch": 10.71,
+      "learning_rate": 5.6565181331580344e-05,
+      "loss": 2.2737,
+      "step": 79320
+    },
+    {
+      "epoch": 10.71,
+      "learning_rate": 5.6531989985472036e-05,
+      "loss": 2.2874,
+      "step": 79330
+    },
+    {
+      "epoch": 10.72,
+      "learning_rate": 5.649880611896913e-05,
+      "loss": 2.2738,
+      "step": 79340
+    },
+    {
+      "epoch": 10.72,
+      "learning_rate": 5.646562973472709e-05,
+      "loss": 2.2959,
+      "step": 79350
+    },
+    {
+      "epoch": 10.72,
+      "learning_rate": 5.643246083540073e-05,
+      "loss": 2.2878,
+      "step": 79360
+    },
+    {
+      "epoch": 10.72,
+      "learning_rate": 5.6399299423644414e-05,
+      "loss": 2.2994,
+      "step": 79370
+    },
+    {
+      "epoch": 10.72,
+      "learning_rate": 5.63661455021118e-05,
+      "loss": 2.283,
+      "step": 79380
+    },
+    {
+      "epoch": 10.72,
+      "learning_rate": 5.63329990734559e-05,
+      "loss": 2.2741,
+      "step": 79390
+    },
+    {
+      "epoch": 10.72,
+      "learning_rate": 5.6299860140329294e-05,
+      "loss": 2.2766,
+      "step": 79400
+    },
+    {
+      "epoch": 10.73,
+      "learning_rate": 5.626672870538379e-05,
+      "loss": 2.2966,
+      "step": 79410
+    },
+    {
+      "epoch": 10.73,
+      "learning_rate": 5.623360477127078e-05,
+      "loss": 2.2676,
+      "step": 79420
+    },
+    {
+      "epoch": 10.73,
+      "learning_rate": 5.620048834064078e-05,
+      "loss": 2.2815,
+      "step": 79430
+    },
+    {
+      "epoch": 10.73,
+      "learning_rate": 5.6167379416143956e-05,
+      "loss": 2.2872,
+      "step": 79440
+    },
+    {
+      "epoch": 10.73,
+      "learning_rate": 5.613427800042985e-05,
+      "loss": 2.3198,
+      "step": 79450
+    },
+    {
+      "epoch": 10.73,
+      "learning_rate": 5.610118409614728e-05,
+      "loss": 2.3054,
+      "step": 79460
+    },
+    {
+      "epoch": 10.73,
+      "learning_rate": 5.606809770594454e-05,
+      "loss": 2.2876,
+      "step": 79470
+    },
+    {
+      "epoch": 10.73,
+      "learning_rate": 5.603501883246922e-05,
+      "loss": 2.2764,
+      "step": 79480
+    },
+    {
+      "epoch": 10.74,
+      "learning_rate": 5.600194747836849e-05,
+      "loss": 2.2756,
+      "step": 79490
+    },
+    {
+      "epoch": 10.74,
+      "learning_rate": 5.59688836462889e-05,
+      "loss": 2.2961,
+      "step": 79500
+    },
+    {
+      "epoch": 10.74,
+      "learning_rate": 5.593582733887614e-05,
+      "loss": 2.3134,
+      "step": 79510
+    },
+    {
+      "epoch": 10.74,
+      "learning_rate": 5.590277855877561e-05,
+      "loss": 2.2926,
+      "step": 79520
+    },
+    {
+      "epoch": 10.74,
+      "learning_rate": 5.5869737308631876e-05,
+      "loss": 2.254,
+      "step": 79530
+    },
+    {
+      "epoch": 10.74,
+      "learning_rate": 5.583670359108914e-05,
+      "loss": 2.2954,
+      "step": 79540
+    },
+    {
+      "epoch": 10.74,
+      "learning_rate": 5.580367740879078e-05,
+      "loss": 2.2802,
+      "step": 79550
+    },
+    {
+      "epoch": 10.75,
+      "learning_rate": 5.57706587643796e-05,
+      "loss": 2.2873,
+      "step": 79560
+    },
+    {
+      "epoch": 10.75,
+      "learning_rate": 5.5737647660497964e-05,
+      "loss": 2.2875,
+      "step": 79570
+    },
+    {
+      "epoch": 10.75,
+      "learning_rate": 5.5704644099787504e-05,
+      "loss": 2.2896,
+      "step": 79580
+    },
+    {
+      "epoch": 10.75,
+      "learning_rate": 5.567164808488918e-05,
+      "loss": 2.301,
+      "step": 79590
+    },
+    {
+      "epoch": 10.75,
+      "learning_rate": 5.5638659618443564e-05,
+      "loss": 2.2731,
+      "step": 79600
+    },
+    {
+      "epoch": 10.75,
+      "learning_rate": 5.560567870309038e-05,
+      "loss": 2.2719,
+      "step": 79610
+    },
+    {
+      "epoch": 10.75,
+      "learning_rate": 5.557270534146902e-05,
+      "loss": 2.2989,
+      "step": 79620
+    },
+    {
+      "epoch": 10.75,
+      "learning_rate": 5.5539739536217913e-05,
+      "loss": 2.274,
+      "step": 79630
+    },
+    {
+      "epoch": 10.76,
+      "learning_rate": 5.55067812899752e-05,
+      "loss": 2.2845,
+      "step": 79640
+    },
+    {
+      "epoch": 10.76,
+      "learning_rate": 5.547383060537834e-05,
+      "loss": 2.3124,
+      "step": 79650
+    },
+    {
+      "epoch": 10.76,
+      "learning_rate": 5.544088748506412e-05,
+      "loss": 2.2739,
+      "step": 79660
+    },
+    {
+      "epoch": 10.76,
+      "learning_rate": 5.5407951931668725e-05,
+      "loss": 2.2712,
+      "step": 79670
+    },
+    {
+      "epoch": 10.76,
+      "learning_rate": 5.537502394782773e-05,
+      "loss": 2.2952,
+      "step": 79680
+    },
+    {
+      "epoch": 10.76,
+      "learning_rate": 5.5342103536176194e-05,
+      "loss": 2.2665,
+      "step": 79690
+    },
+    {
+      "epoch": 10.76,
+      "learning_rate": 5.530919069934859e-05,
+      "loss": 2.2824,
+      "step": 79700
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 5.5276285439978515e-05,
+      "loss": 2.2958,
+      "step": 79710
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 5.524338776069931e-05,
+      "loss": 2.2555,
+      "step": 79720
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 5.521049766414344e-05,
+      "loss": 2.2854,
+      "step": 79730
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 5.517761515294299e-05,
+      "loss": 2.2866,
+      "step": 79740
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 5.514474022972926e-05,
+      "loss": 2.2921,
+      "step": 79750
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 5.5111872897132954e-05,
+      "loss": 2.2582,
+      "step": 79760
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 5.5079013157784316e-05,
+      "loss": 2.2779,
+      "step": 79770
+    },
+    {
+      "epoch": 10.78,
+      "learning_rate": 5.504616101431285e-05,
+      "loss": 2.2926,
+      "step": 79780
+    },
+    {
+      "epoch": 10.78,
+      "learning_rate": 5.501331646934743e-05,
+      "loss": 2.2771,
+      "step": 79790
+    },
+    {
+      "epoch": 10.78,
+      "learning_rate": 5.4980479525516495e-05,
+      "loss": 2.2586,
+      "step": 79800
+    },
+    {
+      "epoch": 10.78,
+      "learning_rate": 5.494765018544763e-05,
+      "loss": 2.2869,
+      "step": 79810
+    },
+    {
+      "epoch": 10.78,
+      "learning_rate": 5.491482845176812e-05,
+      "loss": 2.2987,
+      "step": 79820
+    },
+    {
+      "epoch": 10.78,
+      "learning_rate": 5.488201432710426e-05,
+      "loss": 2.2827,
+      "step": 79830
+    },
+    {
+      "epoch": 10.78,
+      "learning_rate": 5.484920781408203e-05,
+      "loss": 2.2845,
+      "step": 79840
+    },
+    {
+      "epoch": 10.78,
+      "learning_rate": 5.481640891532676e-05,
+      "loss": 2.2711,
+      "step": 79850
+    },
+    {
+      "epoch": 10.79,
+      "learning_rate": 5.4783617633463085e-05,
+      "loss": 2.2954,
+      "step": 79860
+    },
+    {
+      "epoch": 10.79,
+      "learning_rate": 5.475083397111506e-05,
+      "loss": 2.2819,
+      "step": 79870
+    },
+    {
+      "epoch": 10.79,
+      "learning_rate": 5.471805793090607e-05,
+      "loss": 2.3067,
+      "step": 79880
+    },
+    {
+      "epoch": 10.79,
+      "learning_rate": 5.468528951545904e-05,
+      "loss": 2.269,
+      "step": 79890
+    },
+    {
+      "epoch": 10.79,
+      "learning_rate": 5.465252872739628e-05,
+      "loss": 2.2707,
+      "step": 79900
+    },
+    {
+      "epoch": 10.79,
+      "learning_rate": 5.461977556933922e-05,
+      "loss": 2.301,
+      "step": 79910
+    },
+    {
+      "epoch": 10.79,
+      "learning_rate": 5.4587030043909e-05,
+      "loss": 2.2915,
+      "step": 79920
+    },
+    {
+      "epoch": 10.8,
+      "learning_rate": 5.4554292153725955e-05,
+      "loss": 2.3048,
+      "step": 79930
+    },
+    {
+      "epoch": 10.8,
+      "learning_rate": 5.452156190140996e-05,
+      "loss": 2.2852,
+      "step": 79940
+    },
+    {
+      "epoch": 10.8,
+      "learning_rate": 5.4488839289580127e-05,
+      "loss": 2.2639,
+      "step": 79950
+    },
+    {
+      "epoch": 10.8,
+      "learning_rate": 5.445612432085499e-05,
+      "loss": 2.3,
+      "step": 79960
+    },
+    {
+      "epoch": 10.8,
+      "learning_rate": 5.4423416997852613e-05,
+      "loss": 2.2824,
+      "step": 79970
+    },
+    {
+      "epoch": 10.8,
+      "learning_rate": 5.439071732319027e-05,
+      "loss": 2.2732,
+      "step": 79980
+    },
+    {
+      "epoch": 10.8,
+      "learning_rate": 5.4358025299484645e-05,
+      "loss": 2.308,
+      "step": 79990
+    },
+    {
+      "epoch": 10.8,
+      "learning_rate": 5.4325340929351976e-05,
+      "loss": 2.273,
+      "step": 80000
+    },
+    {
+      "epoch": 10.8,
+      "eval_loss": 2.4251482486724854,
+      "eval_runtime": 1267.5227,
+      "eval_samples_per_second": 60.066,
+      "eval_steps_per_second": 5.006,
+      "step": 80000
+    },
+    {
+      "epoch": 10.81,
+      "learning_rate": 5.429266421540763e-05,
+      "loss": 2.2751,
+      "step": 80010
+    },
+    {
+      "epoch": 10.81,
+      "learning_rate": 5.42599951602667e-05,
+      "loss": 2.2898,
+      "step": 80020
+    },
+    {
+      "epoch": 10.81,
+      "learning_rate": 5.4227333766543233e-05,
+      "loss": 2.2951,
+      "step": 80030
+    },
+    {
+      "epoch": 10.81,
+      "learning_rate": 5.419468003685101e-05,
+      "loss": 2.302,
+      "step": 80040
+    },
+    {
+      "epoch": 10.81,
+      "learning_rate": 5.416203397380313e-05,
+      "loss": 2.2958,
+      "step": 80050
+    },
+    {
+      "epoch": 10.81,
+      "learning_rate": 5.412939558001199e-05,
+      "loss": 2.3045,
+      "step": 80060
+    },
+    {
+      "epoch": 10.81,
+      "learning_rate": 5.4096764858089406e-05,
+      "loss": 2.281,
+      "step": 80070
+    },
+    {
+      "epoch": 10.82,
+      "learning_rate": 5.406414181064654e-05,
+      "loss": 2.2879,
+      "step": 80080
+    },
+    {
+      "epoch": 10.82,
+      "learning_rate": 5.403152644029412e-05,
+      "loss": 2.2757,
+      "step": 80090
+    },
+    {
+      "epoch": 10.82,
+      "learning_rate": 5.3998918749642046e-05,
+      "loss": 2.29,
+      "step": 80100
+    },
+    {
+      "epoch": 10.82,
+      "learning_rate": 5.3966318741299655e-05,
+      "loss": 2.2615,
+      "step": 80110
+    },
+    {
+      "epoch": 10.82,
+      "learning_rate": 5.39337264178758e-05,
+      "loss": 2.2581,
+      "step": 80120
+    },
+    {
+      "epoch": 10.82,
+      "learning_rate": 5.390114178197854e-05,
+      "loss": 2.3091,
+      "step": 80130
+    },
+    {
+      "epoch": 10.82,
+      "learning_rate": 5.386856483621548e-05,
+      "loss": 2.2911,
+      "step": 80140
+    },
+    {
+      "epoch": 10.83,
+      "learning_rate": 5.3835995583193464e-05,
+      "loss": 2.2865,
+      "step": 80150
+    },
+    {
+      "epoch": 10.83,
+      "learning_rate": 5.380343402551878e-05,
+      "loss": 2.2772,
+      "step": 80160
+    },
+    {
+      "epoch": 10.83,
+      "learning_rate": 5.377088016579717e-05,
+      "loss": 2.3031,
+      "step": 80170
+    },
+    {
+      "epoch": 10.83,
+      "learning_rate": 5.373833400663366e-05,
+      "loss": 2.306,
+      "step": 80180
+    },
+    {
+      "epoch": 10.83,
+      "learning_rate": 5.370579555063265e-05,
+      "loss": 2.2853,
+      "step": 80190
+    },
+    {
+      "epoch": 10.83,
+      "learning_rate": 5.367326480039806e-05,
+      "loss": 2.2855,
+      "step": 80200
+    },
+    {
+      "epoch": 10.83,
+      "learning_rate": 5.3640741758533015e-05,
+      "loss": 2.3047,
+      "step": 80210
+    },
+    {
+      "epoch": 10.83,
+      "learning_rate": 5.360822642764025e-05,
+      "loss": 2.3235,
+      "step": 80220
+    },
+    {
+      "epoch": 10.84,
+      "learning_rate": 5.3575718810321555e-05,
+      "loss": 2.3011,
+      "step": 80230
+    },
+    {
+      "epoch": 10.84,
+      "learning_rate": 5.354321890917838e-05,
+      "loss": 2.265,
+      "step": 80240
+    },
+    {
+      "epoch": 10.84,
+      "learning_rate": 5.351072672681153e-05,
+      "loss": 2.275,
+      "step": 80250
+    },
+    {
+      "epoch": 10.84,
+      "learning_rate": 5.347824226582108e-05,
+      "loss": 2.2868,
+      "step": 80260
+    },
+    {
+      "epoch": 10.84,
+      "learning_rate": 5.344576552880653e-05,
+      "loss": 2.2789,
+      "step": 80270
+    },
+    {
+      "epoch": 10.84,
+      "learning_rate": 5.341329651836676e-05,
+      "loss": 2.3153,
+      "step": 80280
+    },
+    {
+      "epoch": 10.84,
+      "learning_rate": 5.338083523710008e-05,
+      "loss": 2.2779,
+      "step": 80290
+    },
+    {
+      "epoch": 10.85,
+      "learning_rate": 5.334838168760414e-05,
+      "loss": 2.3076,
+      "step": 80300
+    },
+    {
+      "epoch": 10.85,
+      "learning_rate": 5.3315935872475904e-05,
+      "loss": 2.3193,
+      "step": 80310
+    },
+    {
+      "epoch": 10.85,
+      "learning_rate": 5.328349779431191e-05,
+      "loss": 2.3109,
+      "step": 80320
+    },
+    {
+      "epoch": 10.85,
+      "learning_rate": 5.325106745570784e-05,
+      "loss": 2.3218,
+      "step": 80330
+    },
+    {
+      "epoch": 10.85,
+      "learning_rate": 5.321864485925896e-05,
+      "loss": 2.3136,
+      "step": 80340
+    },
+    {
+      "epoch": 10.85,
+      "learning_rate": 5.318623000755981e-05,
+      "loss": 2.2929,
+      "step": 80350
+    },
+    {
+      "epoch": 10.85,
+      "learning_rate": 5.3153822903204255e-05,
+      "loss": 2.2945,
+      "step": 80360
+    },
+    {
+      "epoch": 10.85,
+      "learning_rate": 5.3121423548785707e-05,
+      "loss": 2.2844,
+      "step": 80370
+    },
+    {
+      "epoch": 10.86,
+      "learning_rate": 5.308903194689685e-05,
+      "loss": 2.2494,
+      "step": 80380
+    },
+    {
+      "epoch": 10.86,
+      "learning_rate": 5.305664810012967e-05,
+      "loss": 2.2886,
+      "step": 80390
+    },
+    {
+      "epoch": 10.86,
+      "learning_rate": 5.302427201107576e-05,
+      "loss": 2.2692,
+      "step": 80400
+    },
+    {
+      "epoch": 10.86,
+      "learning_rate": 5.299190368232583e-05,
+      "loss": 2.2851,
+      "step": 80410
+    },
+    {
+      "epoch": 10.86,
+      "learning_rate": 5.295954311647027e-05,
+      "loss": 2.2946,
+      "step": 80420
+    },
+    {
+      "epoch": 10.86,
+      "learning_rate": 5.2927190316098455e-05,
+      "loss": 2.2967,
+      "step": 80430
+    },
+    {
+      "epoch": 10.86,
+      "learning_rate": 5.289484528379946e-05,
+      "loss": 2.28,
+      "step": 80440
+    },
+    {
+      "epoch": 10.87,
+      "learning_rate": 5.28625080221617e-05,
+      "loss": 2.2691,
+      "step": 80450
+    },
+    {
+      "epoch": 10.87,
+      "learning_rate": 5.2830178533772856e-05,
+      "loss": 2.274,
+      "step": 80460
+    },
+    {
+      "epoch": 10.87,
+      "learning_rate": 5.2797856821220016e-05,
+      "loss": 2.305,
+      "step": 80470
+    },
+    {
+      "epoch": 10.87,
+      "learning_rate": 5.2765542887089626e-05,
+      "loss": 2.2657,
+      "step": 80480
+    },
+    {
+      "epoch": 10.87,
+      "learning_rate": 5.2733236733967666e-05,
+      "loss": 2.3138,
+      "step": 80490
+    },
+    {
+      "epoch": 10.87,
+      "learning_rate": 5.2700938364439286e-05,
+      "loss": 2.2829,
+      "step": 80500
+    },
+    {
+      "epoch": 10.87,
+      "learning_rate": 5.2668647781089096e-05,
+      "loss": 2.2951,
+      "step": 80510
+    },
+    {
+      "epoch": 10.87,
+      "learning_rate": 5.2636364986501175e-05,
+      "loss": 2.2903,
+      "step": 80520
+    },
+    {
+      "epoch": 10.88,
+      "learning_rate": 5.2604089983258766e-05,
+      "loss": 2.2999,
+      "step": 80530
+    },
+    {
+      "epoch": 10.88,
+      "learning_rate": 5.257182277394477e-05,
+      "loss": 2.2873,
+      "step": 80540
+    },
+    {
+      "epoch": 10.88,
+      "learning_rate": 5.253956336114122e-05,
+      "loss": 2.306,
+      "step": 80550
+    },
+    {
+      "epoch": 10.88,
+      "learning_rate": 5.250731174742958e-05,
+      "loss": 2.287,
+      "step": 80560
+    },
+    {
+      "epoch": 10.88,
+      "learning_rate": 5.247506793539082e-05,
+      "loss": 2.3244,
+      "step": 80570
+    },
+    {
+      "epoch": 10.88,
+      "learning_rate": 5.244283192760515e-05,
+      "loss": 2.2891,
+      "step": 80580
+    },
+    {
+      "epoch": 10.88,
+      "learning_rate": 5.241060372665214e-05,
+      "loss": 2.3038,
+      "step": 80590
+    },
+    {
+      "epoch": 10.89,
+      "learning_rate": 5.237838333511091e-05,
+      "loss": 2.2691,
+      "step": 80600
+    },
+    {
+      "epoch": 10.89,
+      "learning_rate": 5.2346170755559706e-05,
+      "loss": 2.3065,
+      "step": 80610
+    },
+    {
+      "epoch": 10.89,
+      "learning_rate": 5.231396599057646e-05,
+      "loss": 2.2722,
+      "step": 80620
+    },
+    {
+      "epoch": 10.89,
+      "learning_rate": 5.2281769042738084e-05,
+      "loss": 2.2847,
+      "step": 80630
+    },
+    {
+      "epoch": 10.89,
+      "learning_rate": 5.224957991462119e-05,
+      "loss": 2.2837,
+      "step": 80640
+    },
+    {
+      "epoch": 10.89,
+      "learning_rate": 5.2217398608801685e-05,
+      "loss": 2.2845,
+      "step": 80650
+    },
+    {
+      "epoch": 10.89,
+      "learning_rate": 5.218522512785478e-05,
+      "loss": 2.2731,
+      "step": 80660
+    },
+    {
+      "epoch": 10.9,
+      "learning_rate": 5.215305947435511e-05,
+      "loss": 2.2979,
+      "step": 80670
+    },
+    {
+      "epoch": 10.9,
+      "learning_rate": 5.212090165087661e-05,
+      "loss": 2.2906,
+      "step": 80680
+    },
+    {
+      "epoch": 10.9,
+      "learning_rate": 5.208875165999274e-05,
+      "loss": 2.285,
+      "step": 80690
+    },
+    {
+      "epoch": 10.9,
+      "learning_rate": 5.205660950427622e-05,
+      "loss": 2.3084,
+      "step": 80700
+    },
+    {
+      "epoch": 10.9,
+      "learning_rate": 5.202447518629908e-05,
+      "loss": 2.2918,
+      "step": 80710
+    },
+    {
+      "epoch": 10.9,
+      "learning_rate": 5.1992348708632946e-05,
+      "loss": 2.2896,
+      "step": 80720
+    },
+    {
+      "epoch": 10.9,
+      "learning_rate": 5.1960230073848574e-05,
+      "loss": 2.32,
+      "step": 80730
+    },
+    {
+      "epoch": 10.9,
+      "learning_rate": 5.1928119284516274e-05,
+      "loss": 2.2792,
+      "step": 80740
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 5.1896016343205624e-05,
+      "loss": 2.2994,
+      "step": 80750
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 5.186392125248555e-05,
+      "loss": 2.2998,
+      "step": 80760
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 5.183183401492451e-05,
+      "loss": 2.3022,
+      "step": 80770
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 5.1799754633090144e-05,
+      "loss": 2.2932,
+      "step": 80780
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 5.1767683109549545e-05,
+      "loss": 2.2625,
+      "step": 80790
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 5.1735619446869225e-05,
+      "loss": 2.2682,
+      "step": 80800
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 5.170356364761497e-05,
+      "loss": 2.2892,
+      "step": 80810
+    },
+    {
+      "epoch": 10.92,
+      "learning_rate": 5.1671515714352096e-05,
+      "loss": 2.2909,
+      "step": 80820
+    },
+    {
+      "epoch": 10.92,
+      "learning_rate": 5.1639475649645e-05,
+      "loss": 2.2913,
+      "step": 80830
+    },
+    {
+      "epoch": 10.92,
+      "learning_rate": 5.160744345605771e-05,
+      "loss": 2.2669,
+      "step": 80840
+    },
+    {
+      "epoch": 10.92,
+      "learning_rate": 5.157541913615363e-05,
+      "loss": 2.29,
+      "step": 80850
+    },
+    {
+      "epoch": 10.92,
+      "learning_rate": 5.154340269249537e-05,
+      "loss": 2.2832,
+      "step": 80860
+    },
+    {
+      "epoch": 10.92,
+      "learning_rate": 5.151139412764498e-05,
+      "loss": 2.288,
+      "step": 80870
+    },
+    {
+      "epoch": 10.92,
+      "learning_rate": 5.1479393444163865e-05,
+      "loss": 2.3023,
+      "step": 80880
+    },
+    {
+      "epoch": 10.92,
+      "learning_rate": 5.144740064461289e-05,
+      "loss": 2.3043,
+      "step": 80890
+    },
+    {
+      "epoch": 10.93,
+      "learning_rate": 5.141541573155219e-05,
+      "loss": 2.2672,
+      "step": 80900
+    },
+    {
+      "epoch": 10.93,
+      "learning_rate": 5.138343870754125e-05,
+      "loss": 2.3026,
+      "step": 80910
+    },
+    {
+      "epoch": 10.93,
+      "learning_rate": 5.135146957513905e-05,
+      "loss": 2.2905,
+      "step": 80920
+    },
+    {
+      "epoch": 10.93,
+      "learning_rate": 5.131950833690379e-05,
+      "loss": 2.2972,
+      "step": 80930
+    },
+    {
+      "epoch": 10.93,
+      "learning_rate": 5.128755499539318e-05,
+      "loss": 2.2918,
+      "step": 80940
+    },
+    {
+      "epoch": 10.93,
+      "learning_rate": 5.1255609553164206e-05,
+      "loss": 2.2765,
+      "step": 80950
+    },
+    {
+      "epoch": 10.93,
+      "learning_rate": 5.122367201277318e-05,
+      "loss": 2.2796,
+      "step": 80960
+    },
+    {
+      "epoch": 10.94,
+      "learning_rate": 5.119174237677594e-05,
+      "loss": 2.2653,
+      "step": 80970
+    },
+    {
+      "epoch": 10.94,
+      "learning_rate": 5.115982064772756e-05,
+      "loss": 2.2742,
+      "step": 80980
+    },
+    {
+      "epoch": 10.94,
+      "learning_rate": 5.112790682818245e-05,
+      "loss": 2.2945,
+      "step": 80990
+    },
+    {
+      "epoch": 10.94,
+      "learning_rate": 5.109600092069456e-05,
+      "loss": 2.2764,
+      "step": 81000
+    },
+    {
+      "epoch": 10.94,
+      "eval_loss": 2.421452760696411,
+      "eval_runtime": 1269.5215,
+      "eval_samples_per_second": 59.971,
+      "eval_steps_per_second": 4.998,
+      "step": 81000
+    },
+    {
+      "epoch": 10.94,
+      "learning_rate": 5.106410292781701e-05,
+      "loss": 2.3018,
+      "step": 81010
+    },
+    {
+      "epoch": 10.94,
+      "learning_rate": 5.103221285210253e-05,
+      "loss": 2.2967,
+      "step": 81020
+    },
+    {
+      "epoch": 10.94,
+      "learning_rate": 5.1000330696102846e-05,
+      "loss": 2.2983,
+      "step": 81030
+    },
+    {
+      "epoch": 10.95,
+      "learning_rate": 5.0968456462369385e-05,
+      "loss": 2.3023,
+      "step": 81040
+    },
+    {
+      "epoch": 10.95,
+      "learning_rate": 5.0936590153452875e-05,
+      "loss": 2.2806,
+      "step": 81050
+    },
+    {
+      "epoch": 10.95,
+      "learning_rate": 5.09047317719033e-05,
+      "loss": 2.313,
+      "step": 81060
+    },
+    {
+      "epoch": 10.95,
+      "learning_rate": 5.087288132027006e-05,
+      "loss": 2.2668,
+      "step": 81070
+    },
+    {
+      "epoch": 10.95,
+      "learning_rate": 5.0841038801101895e-05,
+      "loss": 2.2847,
+      "step": 81080
+    },
+    {
+      "epoch": 10.95,
+      "learning_rate": 5.080920421694706e-05,
+      "loss": 2.326,
+      "step": 81090
+    },
+    {
+      "epoch": 10.95,
+      "learning_rate": 5.077737757035296e-05,
+      "loss": 2.2891,
+      "step": 81100
+    },
+    {
+      "epoch": 10.95,
+      "learning_rate": 5.0745558863866456e-05,
+      "loss": 2.2859,
+      "step": 81110
+    },
+    {
+      "epoch": 10.96,
+      "learning_rate": 5.0713748100033866e-05,
+      "loss": 2.3108,
+      "step": 81120
+    },
+    {
+      "epoch": 10.96,
+      "learning_rate": 5.0681945281400686e-05,
+      "loss": 2.3087,
+      "step": 81130
+    },
+    {
+      "epoch": 10.96,
+      "learning_rate": 5.065015041051197e-05,
+      "loss": 2.273,
+      "step": 81140
+    },
+    {
+      "epoch": 10.96,
+      "learning_rate": 5.061836348991202e-05,
+      "loss": 2.2738,
+      "step": 81150
+    },
+    {
+      "epoch": 10.96,
+      "learning_rate": 5.0586584522144466e-05,
+      "loss": 2.2836,
+      "step": 81160
+    },
+    {
+      "epoch": 10.96,
+      "learning_rate": 5.0554813509752445e-05,
+      "loss": 2.2825,
+      "step": 81170
+    },
+    {
+      "epoch": 10.96,
+      "learning_rate": 5.052305045527833e-05,
+      "loss": 2.2935,
+      "step": 81180
+    },
+    {
+      "epoch": 10.97,
+      "learning_rate": 5.049129536126387e-05,
+      "loss": 2.2856,
+      "step": 81190
+    },
+    {
+      "epoch": 10.97,
+      "learning_rate": 5.045954823025029e-05,
+      "loss": 2.2946,
+      "step": 81200
+    },
+    {
+      "epoch": 10.97,
+      "learning_rate": 5.042780906477799e-05,
+      "loss": 2.2605,
+      "step": 81210
+    },
+    {
+      "epoch": 10.97,
+      "learning_rate": 5.0396077867387e-05,
+      "loss": 2.2499,
+      "step": 81220
+    },
+    {
+      "epoch": 10.97,
+      "learning_rate": 5.036435464061636e-05,
+      "loss": 2.2831,
+      "step": 81230
+    },
+    {
+      "epoch": 10.97,
+      "learning_rate": 5.0332639387004786e-05,
+      "loss": 2.2789,
+      "step": 81240
+    },
+    {
+      "epoch": 10.97,
+      "learning_rate": 5.0300932109090165e-05,
+      "loss": 2.2675,
+      "step": 81250
+    },
+    {
+      "epoch": 10.97,
+      "learning_rate": 5.026923280940988e-05,
+      "loss": 2.3216,
+      "step": 81260
+    },
+    {
+      "epoch": 10.98,
+      "learning_rate": 5.023754149050058e-05,
+      "loss": 2.3191,
+      "step": 81270
+    },
+    {
+      "epoch": 10.98,
+      "learning_rate": 5.0205858154898255e-05,
+      "loss": 2.2762,
+      "step": 81280
+    },
+    {
+      "epoch": 10.98,
+      "learning_rate": 5.017418280513839e-05,
+      "loss": 2.2838,
+      "step": 81290
+    },
+    {
+      "epoch": 10.98,
+      "learning_rate": 5.014251544375571e-05,
+      "loss": 2.2784,
+      "step": 81300
+    },
+    {
+      "epoch": 10.98,
+      "learning_rate": 5.01108560732843e-05,
+      "loss": 2.2607,
+      "step": 81310
+    },
+    {
+      "epoch": 10.98,
+      "learning_rate": 5.007920469625771e-05,
+      "loss": 2.2754,
+      "step": 81320
+    },
+    {
+      "epoch": 10.98,
+      "learning_rate": 5.004756131520871e-05,
+      "loss": 2.2746,
+      "step": 81330
+    },
+    {
+      "epoch": 10.99,
+      "learning_rate": 5.001592593266959e-05,
+      "loss": 2.2856,
+      "step": 81340
+    },
+    {
+      "epoch": 10.99,
+      "learning_rate": 4.9984298551171863e-05,
+      "loss": 2.2638,
+      "step": 81350
+    },
+    {
+      "epoch": 10.99,
+      "learning_rate": 4.9952679173246396e-05,
+      "loss": 2.2619,
+      "step": 81360
+    },
+    {
+      "epoch": 10.99,
+      "learning_rate": 4.992106780142359e-05,
+      "loss": 2.2789,
+      "step": 81370
+    },
+    {
+      "epoch": 10.99,
+      "learning_rate": 4.9889464438233046e-05,
+      "loss": 2.2674,
+      "step": 81380
+    },
+    {
+      "epoch": 10.99,
+      "learning_rate": 4.9857869086203686e-05,
+      "loss": 2.3019,
+      "step": 81390
+    },
+    {
+      "epoch": 10.99,
+      "learning_rate": 4.9826281747863965e-05,
+      "loss": 2.3092,
+      "step": 81400
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 4.979470242574154e-05,
+      "loss": 2.282,
+      "step": 81410
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 4.9763131122363595e-05,
+      "loss": 2.2997,
+      "step": 81420
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 4.9731567840256406e-05,
+      "loss": 2.3026,
+      "step": 81430
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 4.970001258194588e-05,
+      "loss": 2.2893,
+      "step": 81440
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 4.9668465349957095e-05,
+      "loss": 2.2785,
+      "step": 81450
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 4.963692614681464e-05,
+      "loss": 2.243,
+      "step": 81460
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 4.960539497504235e-05,
+      "loss": 2.2427,
+      "step": 81470
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 4.9573871837163384e-05,
+      "loss": 2.2414,
+      "step": 81480
+    },
+    {
+      "epoch": 11.01,
+      "learning_rate": 4.954235673570044e-05,
+      "loss": 2.245,
+      "step": 81490
+    },
+    {
+      "epoch": 11.01,
+      "learning_rate": 4.951084967317539e-05,
+      "loss": 2.2313,
+      "step": 81500
+    },
+    {
+      "epoch": 11.01,
+      "learning_rate": 4.9479350652109486e-05,
+      "loss": 2.219,
+      "step": 81510
+    },
+    {
+      "epoch": 11.01,
+      "learning_rate": 4.944785967502348e-05,
+      "loss": 2.2516,
+      "step": 81520
+    },
+    {
+      "epoch": 11.01,
+      "learning_rate": 4.941637674443727e-05,
+      "loss": 2.2551,
+      "step": 81530
+    },
+    {
+      "epoch": 11.01,
+      "learning_rate": 4.938490186287033e-05,
+      "loss": 2.241,
+      "step": 81540
+    },
+    {
+      "epoch": 11.01,
+      "learning_rate": 4.935343503284131e-05,
+      "loss": 2.2383,
+      "step": 81550
+    },
+    {
+      "epoch": 11.02,
+      "learning_rate": 4.932197625686828e-05,
+      "loss": 2.213,
+      "step": 81560
+    },
+    {
+      "epoch": 11.02,
+      "learning_rate": 4.9290525537468724e-05,
+      "loss": 2.2449,
+      "step": 81570
+    },
+    {
+      "epoch": 11.02,
+      "learning_rate": 4.92590828771594e-05,
+      "loss": 2.2708,
+      "step": 81580
+    },
+    {
+      "epoch": 11.02,
+      "learning_rate": 4.922764827845638e-05,
+      "loss": 2.2274,
+      "step": 81590
+    },
+    {
+      "epoch": 11.02,
+      "learning_rate": 4.919622174387529e-05,
+      "loss": 2.2443,
+      "step": 81600
+    },
+    {
+      "epoch": 11.02,
+      "learning_rate": 4.916480327593088e-05,
+      "loss": 2.2273,
+      "step": 81610
+    },
+    {
+      "epoch": 11.02,
+      "learning_rate": 4.913339287713746e-05,
+      "loss": 2.2218,
+      "step": 81620
+    },
+    {
+      "epoch": 11.02,
+      "learning_rate": 4.910199055000842e-05,
+      "loss": 2.2225,
+      "step": 81630
+    },
+    {
+      "epoch": 11.03,
+      "learning_rate": 4.907059629705685e-05,
+      "loss": 2.2662,
+      "step": 81640
+    },
+    {
+      "epoch": 11.03,
+      "learning_rate": 4.9039210120794864e-05,
+      "loss": 2.2634,
+      "step": 81650
+    },
+    {
+      "epoch": 11.03,
+      "learning_rate": 4.900783202373422e-05,
+      "loss": 2.2342,
+      "step": 81660
+    },
+    {
+      "epoch": 11.03,
+      "learning_rate": 4.897646200838583e-05,
+      "loss": 2.2585,
+      "step": 81670
+    },
+    {
+      "epoch": 11.03,
+      "learning_rate": 4.894510007725998e-05,
+      "loss": 2.2083,
+      "step": 81680
+    },
+    {
+      "epoch": 11.03,
+      "learning_rate": 4.891374623286644e-05,
+      "loss": 2.2787,
+      "step": 81690
+    },
+    {
+      "epoch": 11.03,
+      "learning_rate": 4.888240047771419e-05,
+      "loss": 2.2284,
+      "step": 81700
+    },
+    {
+      "epoch": 11.04,
+      "learning_rate": 4.885106281431157e-05,
+      "loss": 2.2471,
+      "step": 81710
+    },
+    {
+      "epoch": 11.04,
+      "learning_rate": 4.881973324516644e-05,
+      "loss": 2.2164,
+      "step": 81720
+    },
+    {
+      "epoch": 11.04,
+      "learning_rate": 4.878841177278576e-05,
+      "loss": 2.2687,
+      "step": 81730
+    },
+    {
+      "epoch": 11.04,
+      "learning_rate": 4.875709839967609e-05,
+      "loss": 2.2488,
+      "step": 81740
+    },
+    {
+      "epoch": 11.04,
+      "learning_rate": 4.872579312834318e-05,
+      "loss": 2.2631,
+      "step": 81750
+    },
+    {
+      "epoch": 11.04,
+      "learning_rate": 4.869449596129211e-05,
+      "loss": 2.2441,
+      "step": 81760
+    },
+    {
+      "epoch": 11.04,
+      "learning_rate": 4.86632069010275e-05,
+      "loss": 2.2561,
+      "step": 81770
+    },
+    {
+      "epoch": 11.05,
+      "learning_rate": 4.8631925950053105e-05,
+      "loss": 2.2234,
+      "step": 81780
+    },
+    {
+      "epoch": 11.05,
+      "learning_rate": 4.860065311087214e-05,
+      "loss": 2.2471,
+      "step": 81790
+    },
+    {
+      "epoch": 11.05,
+      "learning_rate": 4.8569388385987205e-05,
+      "loss": 2.2537,
+      "step": 81800
+    },
+    {
+      "epoch": 11.05,
+      "learning_rate": 4.8538131777900124e-05,
+      "loss": 2.2409,
+      "step": 81810
+    },
+    {
+      "epoch": 11.05,
+      "learning_rate": 4.850688328911229e-05,
+      "loss": 2.2457,
+      "step": 81820
+    },
+    {
+      "epoch": 11.05,
+      "learning_rate": 4.847564292212413e-05,
+      "loss": 2.2707,
+      "step": 81830
+    },
+    {
+      "epoch": 11.05,
+      "learning_rate": 4.844441067943571e-05,
+      "loss": 2.2723,
+      "step": 81840
+    },
+    {
+      "epoch": 11.05,
+      "learning_rate": 4.841318656354626e-05,
+      "loss": 2.2319,
+      "step": 81850
+    },
+    {
+      "epoch": 11.06,
+      "learning_rate": 4.838197057695451e-05,
+      "loss": 2.2326,
+      "step": 81860
+    },
+    {
+      "epoch": 11.06,
+      "learning_rate": 4.835076272215843e-05,
+      "loss": 2.2585,
+      "step": 81870
+    },
+    {
+      "epoch": 11.06,
+      "learning_rate": 4.831956300165533e-05,
+      "loss": 2.2301,
+      "step": 81880
+    },
+    {
+      "epoch": 11.06,
+      "learning_rate": 4.828837141794198e-05,
+      "loss": 2.2566,
+      "step": 81890
+    },
+    {
+      "epoch": 11.06,
+      "learning_rate": 4.8257187973514394e-05,
+      "loss": 2.2452,
+      "step": 81900
+    },
+    {
+      "epoch": 11.06,
+      "learning_rate": 4.8226012670867944e-05,
+      "loss": 2.2716,
+      "step": 81910
+    },
+    {
+      "epoch": 11.06,
+      "learning_rate": 4.8194845512497425e-05,
+      "loss": 2.2636,
+      "step": 81920
+    },
+    {
+      "epoch": 11.07,
+      "learning_rate": 4.816368650089689e-05,
+      "loss": 2.2385,
+      "step": 81930
+    },
+    {
+      "epoch": 11.07,
+      "learning_rate": 4.8132535638559826e-05,
+      "loss": 2.2484,
+      "step": 81940
+    },
+    {
+      "epoch": 11.07,
+      "learning_rate": 4.8101392927979005e-05,
+      "loss": 2.2401,
+      "step": 81950
+    },
+    {
+      "epoch": 11.07,
+      "learning_rate": 4.807025837164651e-05,
+      "loss": 2.2419,
+      "step": 81960
+    },
+    {
+      "epoch": 11.07,
+      "learning_rate": 4.803913197205393e-05,
+      "loss": 2.2662,
+      "step": 81970
+    },
+    {
+      "epoch": 11.07,
+      "learning_rate": 4.8008013731692024e-05,
+      "loss": 2.2746,
+      "step": 81980
+    },
+    {
+      "epoch": 11.07,
+      "learning_rate": 4.797690365305097e-05,
+      "loss": 2.2405,
+      "step": 81990
+    },
+    {
+      "epoch": 11.07,
+      "learning_rate": 4.794580173862034e-05,
+      "loss": 2.2423,
+      "step": 82000
+    },
+    {
+      "epoch": 11.07,
+      "eval_loss": 2.424318552017212,
+      "eval_runtime": 1269.6246,
+      "eval_samples_per_second": 59.967,
+      "eval_steps_per_second": 4.998,
+      "step": 82000
+    },
+    {
+      "epoch": 11.08,
+      "learning_rate": 4.791470799088894e-05,
+      "loss": 2.245,
+      "step": 82010
+    },
+    {
+      "epoch": 11.08,
+      "learning_rate": 4.7883622412345136e-05,
+      "loss": 2.26,
+      "step": 82020
+    },
+    {
+      "epoch": 11.08,
+      "learning_rate": 4.78525450054763e-05,
+      "loss": 2.2459,
+      "step": 82030
+    },
+    {
+      "epoch": 11.08,
+      "learning_rate": 4.782147577276948e-05,
+      "loss": 2.2524,
+      "step": 82040
+    },
+    {
+      "epoch": 11.08,
+      "learning_rate": 4.779041471671084e-05,
+      "loss": 2.2293,
+      "step": 82050
+    },
+    {
+      "epoch": 11.08,
+      "learning_rate": 4.7759361839786074e-05,
+      "loss": 2.2178,
+      "step": 82060
+    },
+    {
+      "epoch": 11.08,
+      "learning_rate": 4.77283171444801e-05,
+      "loss": 2.2805,
+      "step": 82070
+    },
+    {
+      "epoch": 11.09,
+      "learning_rate": 4.769728063327715e-05,
+      "loss": 2.2572,
+      "step": 82080
+    },
+    {
+      "epoch": 11.09,
+      "learning_rate": 4.7666252308660964e-05,
+      "loss": 2.2411,
+      "step": 82090
+    },
+    {
+      "epoch": 11.09,
+      "learning_rate": 4.763523217311448e-05,
+      "loss": 2.2728,
+      "step": 82100
+    },
+    {
+      "epoch": 11.09,
+      "learning_rate": 4.760422022911998e-05,
+      "loss": 2.2542,
+      "step": 82110
+    },
+    {
+      "epoch": 11.09,
+      "learning_rate": 4.757321647915924e-05,
+      "loss": 2.265,
+      "step": 82120
+    },
+    {
+      "epoch": 11.09,
+      "learning_rate": 4.754222092571316e-05,
+      "loss": 2.2536,
+      "step": 82130
+    },
+    {
+      "epoch": 11.09,
+      "learning_rate": 4.751123357126221e-05,
+      "loss": 2.2627,
+      "step": 82140
+    },
+    {
+      "epoch": 11.1,
+      "learning_rate": 4.7480254418286066e-05,
+      "loss": 2.2304,
+      "step": 82150
+    },
+    {
+      "epoch": 11.1,
+      "learning_rate": 4.744928346926371e-05,
+      "loss": 2.2527,
+      "step": 82160
+    },
+    {
+      "epoch": 11.1,
+      "learning_rate": 4.7418320726673625e-05,
+      "loss": 2.2282,
+      "step": 82170
+    },
+    {
+      "epoch": 11.1,
+      "learning_rate": 4.7387366192993516e-05,
+      "loss": 2.2566,
+      "step": 82180
+    },
+    {
+      "epoch": 11.1,
+      "learning_rate": 4.735641987070041e-05,
+      "loss": 2.264,
+      "step": 82190
+    },
+    {
+      "epoch": 11.1,
+      "learning_rate": 4.732548176227082e-05,
+      "loss": 2.2314,
+      "step": 82200
+    },
+    {
+      "epoch": 11.1,
+      "learning_rate": 4.729455187018043e-05,
+      "loss": 2.2534,
+      "step": 82210
+    },
+    {
+      "epoch": 11.1,
+      "learning_rate": 4.7263630196904465e-05,
+      "loss": 2.2341,
+      "step": 82220
+    },
+    {
+      "epoch": 11.11,
+      "learning_rate": 4.723580772008744e-05,
+      "loss": 2.2528,
+      "step": 82230
+    },
+    {
+      "epoch": 11.11,
+      "learning_rate": 4.72049016693752e-05,
+      "loss": 2.2426,
+      "step": 82240
+    },
+    {
+      "epoch": 11.11,
+      "learning_rate": 4.717400384465145e-05,
+      "loss": 2.2633,
+      "step": 82250
+    },
+    {
+      "epoch": 11.11,
+      "learning_rate": 4.714311424838869e-05,
+      "loss": 2.2218,
+      "step": 82260
+    },
+    {
+      "epoch": 11.11,
+      "learning_rate": 4.7112232883058736e-05,
+      "loss": 2.2453,
+      "step": 82270
+    },
+    {
+      "epoch": 11.11,
+      "learning_rate": 4.7081359751132914e-05,
+      "loss": 2.2553,
+      "step": 82280
+    },
+    {
+      "epoch": 11.11,
+      "learning_rate": 4.7050494855081735e-05,
+      "loss": 2.259,
+      "step": 82290
+    },
+    {
+      "epoch": 11.12,
+      "learning_rate": 4.701963819737507e-05,
+      "loss": 2.2393,
+      "step": 82300
+    },
+    {
+      "epoch": 11.12,
+      "learning_rate": 4.698878978048227e-05,
+      "loss": 2.2465,
+      "step": 82310
+    },
+    {
+      "epoch": 11.12,
+      "learning_rate": 4.6957949606871785e-05,
+      "loss": 2.2521,
+      "step": 82320
+    },
+    {
+      "epoch": 11.12,
+      "learning_rate": 4.692711767901174e-05,
+      "loss": 2.2667,
+      "step": 82330
+    },
+    {
+      "epoch": 11.12,
+      "learning_rate": 4.689629399936918e-05,
+      "loss": 2.247,
+      "step": 82340
+    },
+    {
+      "epoch": 11.12,
+      "learning_rate": 4.686547857041083e-05,
+      "loss": 2.2632,
+      "step": 82350
+    },
+    {
+      "epoch": 11.12,
+      "learning_rate": 4.683467139460266e-05,
+      "loss": 2.2376,
+      "step": 82360
+    },
+    {
+      "epoch": 11.12,
+      "learning_rate": 4.6803872474409924e-05,
+      "loss": 2.2602,
+      "step": 82370
+    },
+    {
+      "epoch": 11.13,
+      "learning_rate": 4.677308181229725e-05,
+      "loss": 2.2368,
+      "step": 82380
+    },
+    {
+      "epoch": 11.13,
+      "learning_rate": 4.674229941072855e-05,
+      "loss": 2.2497,
+      "step": 82390
+    },
+    {
+      "epoch": 11.13,
+      "learning_rate": 4.671152527216718e-05,
+      "loss": 2.2623,
+      "step": 82400
+    },
+    {
+      "epoch": 11.13,
+      "learning_rate": 4.668075939907589e-05,
+      "loss": 2.2419,
+      "step": 82410
+    },
+    {
+      "epoch": 11.13,
+      "learning_rate": 4.665000179391645e-05,
+      "loss": 2.2412,
+      "step": 82420
+    },
+    {
+      "epoch": 11.13,
+      "learning_rate": 4.6619252459150355e-05,
+      "loss": 2.2682,
+      "step": 82430
+    },
+    {
+      "epoch": 11.13,
+      "learning_rate": 4.658851139723812e-05,
+      "loss": 2.2694,
+      "step": 82440
+    },
+    {
+      "epoch": 11.14,
+      "learning_rate": 4.655777861063989e-05,
+      "loss": 2.2511,
+      "step": 82450
+    },
+    {
+      "epoch": 11.14,
+      "learning_rate": 4.652705410181491e-05,
+      "loss": 2.2464,
+      "step": 82460
+    },
+    {
+      "epoch": 11.14,
+      "learning_rate": 4.649633787322183e-05,
+      "loss": 2.2587,
+      "step": 82470
+    },
+    {
+      "epoch": 11.14,
+      "learning_rate": 4.6465629927318745e-05,
+      "loss": 2.2305,
+      "step": 82480
+    },
+    {
+      "epoch": 11.14,
+      "learning_rate": 4.6434930266562956e-05,
+      "loss": 2.2508,
+      "step": 82490
+    },
+    {
+      "epoch": 11.14,
+      "learning_rate": 4.6404238893411114e-05,
+      "loss": 2.2462,
+      "step": 82500
+    },
+    {
+      "epoch": 11.14,
+      "learning_rate": 4.637355581031929e-05,
+      "loss": 2.2292,
+      "step": 82510
+    },
+    {
+      "epoch": 11.15,
+      "learning_rate": 4.6342881019742794e-05,
+      "loss": 2.2241,
+      "step": 82520
+    },
+    {
+      "epoch": 11.15,
+      "learning_rate": 4.631221452413646e-05,
+      "loss": 2.2549,
+      "step": 82530
+    },
+    {
+      "epoch": 11.15,
+      "learning_rate": 4.628155632595409e-05,
+      "loss": 2.2514,
+      "step": 82540
+    },
+    {
+      "epoch": 11.15,
+      "learning_rate": 4.625090642764917e-05,
+      "loss": 2.2664,
+      "step": 82550
+    },
+    {
+      "epoch": 11.15,
+      "learning_rate": 4.622026483167446e-05,
+      "loss": 2.265,
+      "step": 82560
+    },
+    {
+      "epoch": 11.15,
+      "learning_rate": 4.6189631540481926e-05,
+      "loss": 2.2529,
+      "step": 82570
+    },
+    {
+      "epoch": 11.15,
+      "learning_rate": 4.615900655652294e-05,
+      "loss": 2.2489,
+      "step": 82580
+    },
+    {
+      "epoch": 11.15,
+      "learning_rate": 4.6128389882248186e-05,
+      "loss": 2.2335,
+      "step": 82590
+    },
+    {
+      "epoch": 11.16,
+      "learning_rate": 4.609778152010774e-05,
+      "loss": 2.2803,
+      "step": 82600
+    },
+    {
+      "epoch": 11.16,
+      "learning_rate": 4.606718147255108e-05,
+      "loss": 2.2643,
+      "step": 82610
+    },
+    {
+      "epoch": 11.16,
+      "learning_rate": 4.603658974202672e-05,
+      "loss": 2.2049,
+      "step": 82620
+    },
+    {
+      "epoch": 11.16,
+      "learning_rate": 4.600600633098287e-05,
+      "loss": 2.2738,
+      "step": 82630
+    },
+    {
+      "epoch": 11.16,
+      "learning_rate": 4.597543124186678e-05,
+      "loss": 2.2619,
+      "step": 82640
+    },
+    {
+      "epoch": 11.16,
+      "learning_rate": 4.594486447712529e-05,
+      "loss": 2.2404,
+      "step": 82650
+    },
+    {
+      "epoch": 11.16,
+      "learning_rate": 4.591430603920439e-05,
+      "loss": 2.2526,
+      "step": 82660
+    },
+    {
+      "epoch": 11.17,
+      "learning_rate": 4.588375593054943e-05,
+      "loss": 2.2237,
+      "step": 82670
+    },
+    {
+      "epoch": 11.17,
+      "learning_rate": 4.585321415360519e-05,
+      "loss": 2.2154,
+      "step": 82680
+    },
+    {
+      "epoch": 11.17,
+      "learning_rate": 4.582268071081571e-05,
+      "loss": 2.2483,
+      "step": 82690
+    },
+    {
+      "epoch": 11.17,
+      "learning_rate": 4.579215560462428e-05,
+      "loss": 2.2691,
+      "step": 82700
+    },
+    {
+      "epoch": 11.17,
+      "learning_rate": 4.5761638837473763e-05,
+      "loss": 2.2285,
+      "step": 82710
+    },
+    {
+      "epoch": 11.17,
+      "learning_rate": 4.573113041180607e-05,
+      "loss": 2.2747,
+      "step": 82720
+    },
+    {
+      "epoch": 11.17,
+      "learning_rate": 4.5700630330062735e-05,
+      "loss": 2.2642,
+      "step": 82730
+    },
+    {
+      "epoch": 11.17,
+      "learning_rate": 4.567013859468431e-05,
+      "loss": 2.2658,
+      "step": 82740
+    },
+    {
+      "epoch": 11.18,
+      "learning_rate": 4.563965520811089e-05,
+      "loss": 2.2462,
+      "step": 82750
+    },
+    {
+      "epoch": 11.18,
+      "learning_rate": 4.560918017278192e-05,
+      "loss": 2.2521,
+      "step": 82760
+    },
+    {
+      "epoch": 11.18,
+      "learning_rate": 4.557871349113606e-05,
+      "loss": 2.2667,
+      "step": 82770
+    },
+    {
+      "epoch": 11.18,
+      "learning_rate": 4.5548255165611356e-05,
+      "loss": 2.2677,
+      "step": 82780
+    },
+    {
+      "epoch": 11.18,
+      "learning_rate": 4.551780519864514e-05,
+      "loss": 2.2459,
+      "step": 82790
+    },
+    {
+      "epoch": 11.18,
+      "learning_rate": 4.5487363592674116e-05,
+      "loss": 2.2578,
+      "step": 82800
+    },
+    {
+      "epoch": 11.18,
+      "learning_rate": 4.5456930350134476e-05,
+      "loss": 2.2453,
+      "step": 82810
+    },
+    {
+      "epoch": 11.19,
+      "learning_rate": 4.542650547346135e-05,
+      "loss": 2.2658,
+      "step": 82820
+    },
+    {
+      "epoch": 11.19,
+      "learning_rate": 4.539608896508958e-05,
+      "loss": 2.2381,
+      "step": 82830
+    },
+    {
+      "epoch": 11.19,
+      "learning_rate": 4.536568082745313e-05,
+      "loss": 2.2516,
+      "step": 82840
+    },
+    {
+      "epoch": 11.19,
+      "learning_rate": 4.5335281062985405e-05,
+      "loss": 2.242,
+      "step": 82850
+    },
+    {
+      "epoch": 11.19,
+      "learning_rate": 4.5304889674119056e-05,
+      "loss": 2.2449,
+      "step": 82860
+    },
+    {
+      "epoch": 11.19,
+      "learning_rate": 4.5274506663286074e-05,
+      "loss": 2.2474,
+      "step": 82870
+    },
+    {
+      "epoch": 11.19,
+      "learning_rate": 4.5244132032917866e-05,
+      "loss": 2.2831,
+      "step": 82880
+    },
+    {
+      "epoch": 11.2,
+      "learning_rate": 4.521376578544509e-05,
+      "loss": 2.2647,
+      "step": 82890
+    },
+    {
+      "epoch": 11.2,
+      "learning_rate": 4.518340792329768e-05,
+      "loss": 2.2316,
+      "step": 82900
+    },
+    {
+      "epoch": 11.2,
+      "learning_rate": 4.515305844890507e-05,
+      "loss": 2.2553,
+      "step": 82910
+    },
+    {
+      "epoch": 11.2,
+      "learning_rate": 4.5122717364695816e-05,
+      "loss": 2.2428,
+      "step": 82920
+    },
+    {
+      "epoch": 11.2,
+      "learning_rate": 4.5092384673098064e-05,
+      "loss": 2.2477,
+      "step": 82930
+    },
+    {
+      "epoch": 11.2,
+      "learning_rate": 4.506206037653894e-05,
+      "loss": 2.2453,
+      "step": 82940
+    },
+    {
+      "epoch": 11.2,
+      "learning_rate": 4.503174447744518e-05,
+      "loss": 2.2453,
+      "step": 82950
+    },
+    {
+      "epoch": 11.2,
+      "learning_rate": 4.50014369782428e-05,
+      "loss": 2.2698,
+      "step": 82960
+    },
+    {
+      "epoch": 11.21,
+      "learning_rate": 4.497113788135706e-05,
+      "loss": 2.2357,
+      "step": 82970
+    },
+    {
+      "epoch": 11.21,
+      "learning_rate": 4.4940847189212595e-05,
+      "loss": 2.2562,
+      "step": 82980
+    },
+    {
+      "epoch": 11.21,
+      "learning_rate": 4.4910564904233304e-05,
+      "loss": 2.2335,
+      "step": 82990
+    },
+    {
+      "epoch": 11.21,
+      "learning_rate": 4.4880291028842576e-05,
+      "loss": 2.2444,
+      "step": 83000
+    },
+    {
+      "epoch": 11.21,
+      "eval_loss": 2.4225544929504395,
+      "eval_runtime": 1269.7097,
+      "eval_samples_per_second": 59.963,
+      "eval_steps_per_second": 4.997,
+      "step": 83000
+    },
+    {
+      "epoch": 11.21,
+      "learning_rate": 4.485002556546297e-05,
+      "loss": 2.2795,
+      "step": 83010
+    },
+    {
+      "epoch": 11.21,
+      "learning_rate": 4.4819768516516375e-05,
+      "loss": 2.2435,
+      "step": 83020
+    },
+    {
+      "epoch": 11.21,
+      "learning_rate": 4.478951988442415e-05,
+      "loss": 2.2505,
+      "step": 83030
+    },
+    {
+      "epoch": 11.22,
+      "learning_rate": 4.47592796716068e-05,
+      "loss": 2.2499,
+      "step": 83040
+    },
+    {
+      "epoch": 11.22,
+      "learning_rate": 4.472904788048432e-05,
+      "loss": 2.2537,
+      "step": 83050
+    },
+    {
+      "epoch": 11.22,
+      "learning_rate": 4.469882451347593e-05,
+      "loss": 2.2569,
+      "step": 83060
+    },
+    {
+      "epoch": 11.22,
+      "learning_rate": 4.4668609573000136e-05,
+      "loss": 2.2565,
+      "step": 83070
+    },
+    {
+      "epoch": 11.22,
+      "learning_rate": 4.4638403061474925e-05,
+      "loss": 2.2531,
+      "step": 83080
+    },
+    {
+      "epoch": 11.22,
+      "learning_rate": 4.4608204981317467e-05,
+      "loss": 2.2469,
+      "step": 83090
+    },
+    {
+      "epoch": 11.22,
+      "learning_rate": 4.4578015334944285e-05,
+      "loss": 2.2534,
+      "step": 83100
+    },
+    {
+      "epoch": 11.22,
+      "learning_rate": 4.454783412477132e-05,
+      "loss": 2.2593,
+      "step": 83110
+    },
+    {
+      "epoch": 11.23,
+      "learning_rate": 4.4517661353213676e-05,
+      "loss": 2.2562,
+      "step": 83120
+    },
+    {
+      "epoch": 11.23,
+      "learning_rate": 4.4487497022686045e-05,
+      "loss": 2.2485,
+      "step": 83130
+    },
+    {
+      "epoch": 11.23,
+      "learning_rate": 4.4457341135602044e-05,
+      "loss": 2.247,
+      "step": 83140
+    },
+    {
+      "epoch": 11.23,
+      "learning_rate": 4.442719369437494e-05,
+      "loss": 2.2531,
+      "step": 83150
+    },
+    {
+      "epoch": 11.23,
+      "learning_rate": 4.4397054701417294e-05,
+      "loss": 2.2328,
+      "step": 83160
+    },
+    {
+      "epoch": 11.23,
+      "learning_rate": 4.436692415914086e-05,
+      "loss": 2.2112,
+      "step": 83170
+    },
+    {
+      "epoch": 11.23,
+      "learning_rate": 4.43368020699568e-05,
+      "loss": 2.2432,
+      "step": 83180
+    },
+    {
+      "epoch": 11.24,
+      "learning_rate": 4.430668843627551e-05,
+      "loss": 2.2441,
+      "step": 83190
+    },
+    {
+      "epoch": 11.24,
+      "learning_rate": 4.427658326050688e-05,
+      "loss": 2.2463,
+      "step": 83200
+    },
+    {
+      "epoch": 11.24,
+      "learning_rate": 4.424648654505998e-05,
+      "loss": 2.2686,
+      "step": 83210
+    },
+    {
+      "epoch": 11.24,
+      "learning_rate": 4.421639829234318e-05,
+      "loss": 2.2617,
+      "step": 83220
+    },
+    {
+      "epoch": 11.24,
+      "learning_rate": 4.4186318504764365e-05,
+      "loss": 2.2578,
+      "step": 83230
+    },
+    {
+      "epoch": 11.24,
+      "learning_rate": 4.4156247184730484e-05,
+      "loss": 2.229,
+      "step": 83240
+    },
+    {
+      "epoch": 11.24,
+      "learning_rate": 4.4126184334648054e-05,
+      "loss": 2.2447,
+      "step": 83250
+    },
+    {
+      "epoch": 11.25,
+      "learning_rate": 4.409612995692273e-05,
+      "loss": 2.2324,
+      "step": 83260
+    },
+    {
+      "epoch": 11.25,
+      "learning_rate": 4.406608405395954e-05,
+      "loss": 2.2686,
+      "step": 83270
+    },
+    {
+      "epoch": 11.25,
+      "learning_rate": 4.403604662816293e-05,
+      "loss": 2.2553,
+      "step": 83280
+    },
+    {
+      "epoch": 11.25,
+      "learning_rate": 4.4006017681936534e-05,
+      "loss": 2.2566,
+      "step": 83290
+    },
+    {
+      "epoch": 11.25,
+      "learning_rate": 4.3975997217683346e-05,
+      "loss": 2.2256,
+      "step": 83300
+    },
+    {
+      "epoch": 11.25,
+      "learning_rate": 4.394598523780576e-05,
+      "loss": 2.2668,
+      "step": 83310
+    },
+    {
+      "epoch": 11.25,
+      "learning_rate": 4.391598174470537e-05,
+      "loss": 2.2709,
+      "step": 83320
+    },
+    {
+      "epoch": 11.25,
+      "learning_rate": 4.388598674078324e-05,
+      "loss": 2.25,
+      "step": 83330
+    },
+    {
+      "epoch": 11.26,
+      "learning_rate": 4.3856000228439536e-05,
+      "loss": 2.2677,
+      "step": 83340
+    },
+    {
+      "epoch": 11.26,
+      "learning_rate": 4.382602221007392e-05,
+      "loss": 2.2464,
+      "step": 83350
+    },
+    {
+      "epoch": 11.26,
+      "learning_rate": 4.3796052688085406e-05,
+      "loss": 2.2578,
+      "step": 83360
+    },
+    {
+      "epoch": 11.26,
+      "learning_rate": 4.3766091664872186e-05,
+      "loss": 2.2509,
+      "step": 83370
+    },
+    {
+      "epoch": 11.26,
+      "learning_rate": 4.373613914283184e-05,
+      "loss": 2.2421,
+      "step": 83380
+    },
+    {
+      "epoch": 11.26,
+      "learning_rate": 4.370619512436122e-05,
+      "loss": 2.2545,
+      "step": 83390
+    },
+    {
+      "epoch": 11.26,
+      "learning_rate": 4.3676259611856636e-05,
+      "loss": 2.2455,
+      "step": 83400
+    },
+    {
+      "epoch": 11.27,
+      "learning_rate": 4.364633260771355e-05,
+      "loss": 2.2507,
+      "step": 83410
+    },
+    {
+      "epoch": 11.27,
+      "learning_rate": 4.361641411432681e-05,
+      "loss": 2.2511,
+      "step": 83420
+    },
+    {
+      "epoch": 11.27,
+      "learning_rate": 4.358650413409065e-05,
+      "loss": 2.244,
+      "step": 83430
+    },
+    {
+      "epoch": 11.27,
+      "learning_rate": 4.355660266939849e-05,
+      "loss": 2.2235,
+      "step": 83440
+    },
+    {
+      "epoch": 11.27,
+      "learning_rate": 4.3526709722643214e-05,
+      "loss": 2.2432,
+      "step": 83450
+    },
+    {
+      "epoch": 11.27,
+      "learning_rate": 4.349682529621691e-05,
+      "loss": 2.2656,
+      "step": 83460
+    },
+    {
+      "epoch": 11.27,
+      "learning_rate": 4.3466949392510985e-05,
+      "loss": 2.2714,
+      "step": 83470
+    },
+    {
+      "epoch": 11.27,
+      "learning_rate": 4.343708201391628e-05,
+      "loss": 2.2752,
+      "step": 83480
+    },
+    {
+      "epoch": 11.28,
+      "learning_rate": 4.340722316282286e-05,
+      "loss": 2.2648,
+      "step": 83490
+    },
+    {
+      "epoch": 11.28,
+      "learning_rate": 4.337737284162004e-05,
+      "loss": 2.2249,
+      "step": 83500
+    },
+    {
+      "epoch": 11.28,
+      "learning_rate": 4.334753105269667e-05,
+      "loss": 2.2159,
+      "step": 83510
+    },
+    {
+      "epoch": 11.28,
+      "learning_rate": 4.331769779844065e-05,
+      "loss": 2.2438,
+      "step": 83520
+    },
+    {
+      "epoch": 11.28,
+      "learning_rate": 4.3287873081239495e-05,
+      "loss": 2.2394,
+      "step": 83530
+    },
+    {
+      "epoch": 11.28,
+      "learning_rate": 4.3258056903479675e-05,
+      "loss": 2.2439,
+      "step": 83540
+    },
+    {
+      "epoch": 11.28,
+      "learning_rate": 4.322824926754729e-05,
+      "loss": 2.2597,
+      "step": 83550
+    },
+    {
+      "epoch": 11.29,
+      "learning_rate": 4.319845017582767e-05,
+      "loss": 2.2512,
+      "step": 83560
+    },
+    {
+      "epoch": 11.29,
+      "learning_rate": 4.316865963070538e-05,
+      "loss": 2.2526,
+      "step": 83570
+    },
+    {
+      "epoch": 11.29,
+      "learning_rate": 4.313887763456439e-05,
+      "loss": 2.2582,
+      "step": 83580
+    },
+    {
+      "epoch": 11.29,
+      "learning_rate": 4.310910418978785e-05,
+      "loss": 2.2597,
+      "step": 83590
+    },
+    {
+      "epoch": 11.29,
+      "learning_rate": 4.307933929875845e-05,
+      "loss": 2.2518,
+      "step": 83600
+    },
+    {
+      "epoch": 11.29,
+      "learning_rate": 4.3049582963858014e-05,
+      "loss": 2.2475,
+      "step": 83610
+    },
+    {
+      "epoch": 11.29,
+      "learning_rate": 4.30198351874677e-05,
+      "loss": 2.2622,
+      "step": 83620
+    },
+    {
+      "epoch": 11.29,
+      "learning_rate": 4.29900959719681e-05,
+      "loss": 2.2321,
+      "step": 83630
+    },
+    {
+      "epoch": 11.3,
+      "learning_rate": 4.296036531973895e-05,
+      "loss": 2.2449,
+      "step": 83640
+    },
+    {
+      "epoch": 11.3,
+      "learning_rate": 4.2930643233159516e-05,
+      "loss": 2.2612,
+      "step": 83650
+    },
+    {
+      "epoch": 11.3,
+      "learning_rate": 4.290092971460815e-05,
+      "loss": 2.2513,
+      "step": 83660
+    },
+    {
+      "epoch": 11.3,
+      "learning_rate": 4.287122476646262e-05,
+      "loss": 2.2369,
+      "step": 83670
+    },
+    {
+      "epoch": 11.3,
+      "learning_rate": 4.284152839110007e-05,
+      "loss": 2.2449,
+      "step": 83680
+    },
+    {
+      "epoch": 11.3,
+      "learning_rate": 4.281184059089689e-05,
+      "loss": 2.2765,
+      "step": 83690
+    },
+    {
+      "epoch": 11.3,
+      "learning_rate": 4.278216136822871e-05,
+      "loss": 2.2751,
+      "step": 83700
+    },
+    {
+      "epoch": 11.31,
+      "learning_rate": 4.2752490725470655e-05,
+      "loss": 2.2421,
+      "step": 83710
+    },
+    {
+      "epoch": 11.31,
+      "learning_rate": 4.272282866499699e-05,
+      "loss": 2.2443,
+      "step": 83720
+    },
+    {
+      "epoch": 11.31,
+      "learning_rate": 4.269317518918147e-05,
+      "loss": 2.248,
+      "step": 83730
+    },
+    {
+      "epoch": 11.31,
+      "learning_rate": 4.2663530300396926e-05,
+      "loss": 2.2586,
+      "step": 83740
+    },
+    {
+      "epoch": 11.31,
+      "learning_rate": 4.263389400101568e-05,
+      "loss": 2.2591,
+      "step": 83750
+    },
+    {
+      "epoch": 11.31,
+      "learning_rate": 4.26042662934094e-05,
+      "loss": 2.2521,
+      "step": 83760
+    },
+    {
+      "epoch": 11.31,
+      "learning_rate": 4.257464717994891e-05,
+      "loss": 2.2299,
+      "step": 83770
+    },
+    {
+      "epoch": 11.32,
+      "learning_rate": 4.2545036663004453e-05,
+      "loss": 2.2469,
+      "step": 83780
+    },
+    {
+      "epoch": 11.32,
+      "learning_rate": 4.251543474494552e-05,
+      "loss": 2.253,
+      "step": 83790
+    },
+    {
+      "epoch": 11.32,
+      "learning_rate": 4.248584142814101e-05,
+      "loss": 2.2619,
+      "step": 83800
+    },
+    {
+      "epoch": 11.32,
+      "learning_rate": 4.2456256714959045e-05,
+      "loss": 2.2554,
+      "step": 83810
+    },
+    {
+      "epoch": 11.32,
+      "learning_rate": 4.242668060776704e-05,
+      "loss": 2.2346,
+      "step": 83820
+    },
+    {
+      "epoch": 11.32,
+      "learning_rate": 4.239711310893184e-05,
+      "loss": 2.2427,
+      "step": 83830
+    },
+    {
+      "epoch": 11.32,
+      "learning_rate": 4.2367554220819475e-05,
+      "loss": 2.2736,
+      "step": 83840
+    },
+    {
+      "epoch": 11.32,
+      "learning_rate": 4.2338003945795405e-05,
+      "loss": 2.2432,
+      "step": 83850
+    },
+    {
+      "epoch": 11.33,
+      "learning_rate": 4.2308462286224304e-05,
+      "loss": 2.2663,
+      "step": 83860
+    },
+    {
+      "epoch": 11.33,
+      "learning_rate": 4.2278929244470146e-05,
+      "loss": 2.2277,
+      "step": 83870
+    },
+    {
+      "epoch": 11.33,
+      "learning_rate": 4.224940482289634e-05,
+      "loss": 2.2424,
+      "step": 83880
+    },
+    {
+      "epoch": 11.33,
+      "learning_rate": 4.221988902386546e-05,
+      "loss": 2.2595,
+      "step": 83890
+    },
+    {
+      "epoch": 11.33,
+      "learning_rate": 4.219038184973944e-05,
+      "loss": 2.2832,
+      "step": 83900
+    },
+    {
+      "epoch": 11.33,
+      "learning_rate": 4.216088330287962e-05,
+      "loss": 2.2704,
+      "step": 83910
+    },
+    {
+      "epoch": 11.33,
+      "learning_rate": 4.2131393385646466e-05,
+      "loss": 2.2289,
+      "step": 83920
+    },
+    {
+      "epoch": 11.34,
+      "learning_rate": 4.2101912100400005e-05,
+      "loss": 2.2214,
+      "step": 83930
+    },
+    {
+      "epoch": 11.34,
+      "learning_rate": 4.207243944949923e-05,
+      "loss": 2.2472,
+      "step": 83940
+    },
+    {
+      "epoch": 11.34,
+      "learning_rate": 4.204297543530272e-05,
+      "loss": 2.2427,
+      "step": 83950
+    },
+    {
+      "epoch": 11.34,
+      "learning_rate": 4.201352006016833e-05,
+      "loss": 2.2418,
+      "step": 83960
+    },
+    {
+      "epoch": 11.34,
+      "learning_rate": 4.1984073326453155e-05,
+      "loss": 2.2555,
+      "step": 83970
+    },
+    {
+      "epoch": 11.34,
+      "learning_rate": 4.195463523651357e-05,
+      "loss": 2.2512,
+      "step": 83980
+    },
+    {
+      "epoch": 11.34,
+      "learning_rate": 4.192520579270529e-05,
+      "loss": 2.2549,
+      "step": 83990
+    },
+    {
+      "epoch": 11.34,
+      "learning_rate": 4.1895784997383405e-05,
+      "loss": 2.2831,
+      "step": 84000
+    },
+    {
+      "epoch": 11.34,
+      "eval_loss": 2.420787811279297,
+      "eval_runtime": 1269.4908,
+      "eval_samples_per_second": 59.973,
+      "eval_steps_per_second": 4.998,
+      "step": 84000
+    },
+    {
+      "epoch": 11.35,
+      "learning_rate": 4.1866372852902255e-05,
+      "loss": 2.2513,
+      "step": 84010
+    },
+    {
+      "epoch": 11.35,
+      "learning_rate": 4.183696936161545e-05,
+      "loss": 2.2547,
+      "step": 84020
+    },
+    {
+      "epoch": 11.35,
+      "learning_rate": 4.1807574525876006e-05,
+      "loss": 2.2535,
+      "step": 84030
+    },
+    {
+      "epoch": 11.35,
+      "learning_rate": 4.177818834803613e-05,
+      "loss": 2.2693,
+      "step": 84040
+    },
+    {
+      "epoch": 11.35,
+      "learning_rate": 4.174881083044746e-05,
+      "loss": 2.2306,
+      "step": 84050
+    },
+    {
+      "epoch": 11.35,
+      "learning_rate": 4.171944197546087e-05,
+      "loss": 2.2402,
+      "step": 84060
+    },
+    {
+      "epoch": 11.35,
+      "learning_rate": 4.169008178542646e-05,
+      "loss": 2.2731,
+      "step": 84070
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 4.166073026269385e-05,
+      "loss": 2.2621,
+      "step": 84080
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 4.163138740961179e-05,
+      "loss": 2.2426,
+      "step": 84090
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 4.160205322852831e-05,
+      "loss": 2.2714,
+      "step": 84100
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 4.157272772179097e-05,
+      "loss": 2.2529,
+      "step": 84110
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 4.1543410891746346e-05,
+      "loss": 2.2735,
+      "step": 84120
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 4.151410274074063e-05,
+      "loss": 2.2529,
+      "step": 84130
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 4.148480327111897e-05,
+      "loss": 2.2617,
+      "step": 84140
+    },
+    {
+      "epoch": 11.37,
+      "learning_rate": 4.1455512485226086e-05,
+      "loss": 2.2798,
+      "step": 84150
+    },
+    {
+      "epoch": 11.37,
+      "learning_rate": 4.1426230385405956e-05,
+      "loss": 2.2402,
+      "step": 84160
+    },
+    {
+      "epoch": 11.37,
+      "learning_rate": 4.13969569740018e-05,
+      "loss": 2.2565,
+      "step": 84170
+    },
+    {
+      "epoch": 11.37,
+      "learning_rate": 4.136769225335615e-05,
+      "loss": 2.2643,
+      "step": 84180
+    },
+    {
+      "epoch": 11.37,
+      "learning_rate": 4.1338436225810834e-05,
+      "loss": 2.2671,
+      "step": 84190
+    },
+    {
+      "epoch": 11.37,
+      "learning_rate": 4.130918889370709e-05,
+      "loss": 2.2604,
+      "step": 84200
+    },
+    {
+      "epoch": 11.37,
+      "learning_rate": 4.1279950259385335e-05,
+      "loss": 2.2724,
+      "step": 84210
+    },
+    {
+      "epoch": 11.37,
+      "learning_rate": 4.12507203251853e-05,
+      "loss": 2.273,
+      "step": 84220
+    },
+    {
+      "epoch": 11.38,
+      "learning_rate": 4.122149909344615e-05,
+      "loss": 2.2556,
+      "step": 84230
+    },
+    {
+      "epoch": 11.38,
+      "learning_rate": 4.119228656650616e-05,
+      "loss": 2.2618,
+      "step": 84240
+    },
+    {
+      "epoch": 11.38,
+      "learning_rate": 4.1163082746703104e-05,
+      "loss": 2.2444,
+      "step": 84250
+    },
+    {
+      "epoch": 11.38,
+      "learning_rate": 4.1133887636373916e-05,
+      "loss": 2.2378,
+      "step": 84260
+    },
+    {
+      "epoch": 11.38,
+      "learning_rate": 4.110470123785482e-05,
+      "loss": 2.2552,
+      "step": 84270
+    },
+    {
+      "epoch": 11.38,
+      "learning_rate": 4.1075523553481536e-05,
+      "loss": 2.2419,
+      "step": 84280
+    },
+    {
+      "epoch": 11.38,
+      "learning_rate": 4.104635458558887e-05,
+      "loss": 2.2753,
+      "step": 84290
+    },
+    {
+      "epoch": 11.39,
+      "learning_rate": 4.101719433651099e-05,
+      "loss": 2.2632,
+      "step": 84300
+    },
+    {
+      "epoch": 11.39,
+      "learning_rate": 4.0988042808581476e-05,
+      "loss": 2.2506,
+      "step": 84310
+    },
+    {
+      "epoch": 11.39,
+      "learning_rate": 4.095890000413304e-05,
+      "loss": 2.2524,
+      "step": 84320
+    },
+    {
+      "epoch": 11.39,
+      "learning_rate": 4.0929765925497905e-05,
+      "loss": 2.2707,
+      "step": 84330
+    },
+    {
+      "epoch": 11.39,
+      "learning_rate": 4.0900640575007295e-05,
+      "loss": 2.2508,
+      "step": 84340
+    },
+    {
+      "epoch": 11.39,
+      "learning_rate": 4.087152395499206e-05,
+      "loss": 2.2335,
+      "step": 84350
+    },
+    {
+      "epoch": 11.39,
+      "learning_rate": 4.08424160677821e-05,
+      "loss": 2.2286,
+      "step": 84360
+    },
+    {
+      "epoch": 11.39,
+      "learning_rate": 4.081331691570681e-05,
+      "loss": 2.2349,
+      "step": 84370
+    },
+    {
+      "epoch": 11.4,
+      "learning_rate": 4.078713514930374e-05,
+      "loss": 2.2661,
+      "step": 84380
+    },
+    {
+      "epoch": 11.4,
+      "learning_rate": 4.0758052600398977e-05,
+      "loss": 2.2446,
+      "step": 84390
+    },
+    {
+      "epoch": 11.4,
+      "learning_rate": 4.072897879337989e-05,
+      "loss": 2.274,
+      "step": 84400
+    },
+    {
+      "epoch": 11.4,
+      "learning_rate": 4.0699913730572993e-05,
+      "loss": 2.2652,
+      "step": 84410
+    },
+    {
+      "epoch": 11.4,
+      "learning_rate": 4.067085741430424e-05,
+      "loss": 2.2663,
+      "step": 84420
+    },
+    {
+      "epoch": 11.4,
+      "learning_rate": 4.064180984689875e-05,
+      "loss": 2.2684,
+      "step": 84430
+    },
+    {
+      "epoch": 11.4,
+      "learning_rate": 4.0612771030681084e-05,
+      "loss": 2.2479,
+      "step": 84440
+    },
+    {
+      "epoch": 11.41,
+      "learning_rate": 4.058374096797495e-05,
+      "loss": 2.2714,
+      "step": 84450
+    },
+    {
+      "epoch": 11.41,
+      "learning_rate": 4.055471966110341e-05,
+      "loss": 2.2633,
+      "step": 84460
+    },
+    {
+      "epoch": 11.41,
+      "learning_rate": 4.0525707112388936e-05,
+      "loss": 2.2269,
+      "step": 84470
+    },
+    {
+      "epoch": 11.41,
+      "learning_rate": 4.049670332415312e-05,
+      "loss": 2.2486,
+      "step": 84480
+    },
+    {
+      "epoch": 11.41,
+      "learning_rate": 4.046770829871698e-05,
+      "loss": 2.2664,
+      "step": 84490
+    },
+    {
+      "epoch": 11.41,
+      "learning_rate": 4.043872203840071e-05,
+      "loss": 2.2472,
+      "step": 84500
+    },
+    {
+      "epoch": 11.41,
+      "learning_rate": 4.0409744545523934e-05,
+      "loss": 2.2534,
+      "step": 84510
+    },
+    {
+      "epoch": 11.42,
+      "learning_rate": 4.038077582240563e-05,
+      "loss": 2.2595,
+      "step": 84520
+    },
+    {
+      "epoch": 11.42,
+      "learning_rate": 4.0351815871363775e-05,
+      "loss": 2.2801,
+      "step": 84530
+    },
+    {
+      "epoch": 11.42,
+      "learning_rate": 4.032286469471598e-05,
+      "loss": 2.2843,
+      "step": 84540
+    },
+    {
+      "epoch": 11.42,
+      "learning_rate": 4.029392229477889e-05,
+      "loss": 2.2524,
+      "step": 84550
+    },
+    {
+      "epoch": 11.42,
+      "learning_rate": 4.026498867386869e-05,
+      "loss": 2.25,
+      "step": 84560
+    },
+    {
+      "epoch": 11.42,
+      "learning_rate": 4.02360638343007e-05,
+      "loss": 2.2444,
+      "step": 84570
+    },
+    {
+      "epoch": 11.42,
+      "learning_rate": 4.0207147778389484e-05,
+      "loss": 2.2624,
+      "step": 84580
+    },
+    {
+      "epoch": 11.42,
+      "learning_rate": 4.017824050844914e-05,
+      "loss": 2.2451,
+      "step": 84590
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 4.014934202679285e-05,
+      "loss": 2.2231,
+      "step": 84600
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 4.012045233573311e-05,
+      "loss": 2.2637,
+      "step": 84610
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 4.009157143758187e-05,
+      "loss": 2.2492,
+      "step": 84620
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 4.0062699334650163e-05,
+      "loss": 2.2424,
+      "step": 84630
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 4.003383602924855e-05,
+      "loss": 2.254,
+      "step": 84640
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 4.000498152368667e-05,
+      "loss": 2.2343,
+      "step": 84650
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 3.9976135820273544e-05,
+      "loss": 2.2545,
+      "step": 84660
+    },
+    {
+      "epoch": 11.44,
+      "learning_rate": 3.994729892131757e-05,
+      "loss": 2.2581,
+      "step": 84670
+    },
+    {
+      "epoch": 11.44,
+      "learning_rate": 3.991847082912634e-05,
+      "loss": 2.2587,
+      "step": 84680
+    },
+    {
+      "epoch": 11.44,
+      "learning_rate": 3.988965154600674e-05,
+      "loss": 2.242,
+      "step": 84690
+    },
+    {
+      "epoch": 11.44,
+      "learning_rate": 3.986084107426496e-05,
+      "loss": 2.2592,
+      "step": 84700
+    },
+    {
+      "epoch": 11.44,
+      "learning_rate": 3.983203941620653e-05,
+      "loss": 2.2715,
+      "step": 84710
+    },
+    {
+      "epoch": 11.44,
+      "learning_rate": 3.980324657413637e-05,
+      "loss": 2.2668,
+      "step": 84720
+    },
+    {
+      "epoch": 11.44,
+      "learning_rate": 3.9774462550358356e-05,
+      "loss": 2.2558,
+      "step": 84730
+    },
+    {
+      "epoch": 11.44,
+      "learning_rate": 3.974568734717604e-05,
+      "loss": 2.2395,
+      "step": 84740
+    },
+    {
+      "epoch": 11.45,
+      "learning_rate": 3.971692096689199e-05,
+      "loss": 2.2669,
+      "step": 84750
+    },
+    {
+      "epoch": 11.45,
+      "learning_rate": 3.968816341180831e-05,
+      "loss": 2.2434,
+      "step": 84760
+    },
+    {
+      "epoch": 11.45,
+      "learning_rate": 3.9659414684226195e-05,
+      "loss": 2.2447,
+      "step": 84770
+    },
+    {
+      "epoch": 11.45,
+      "learning_rate": 3.963067478644617e-05,
+      "loss": 2.2484,
+      "step": 84780
+    },
+    {
+      "epoch": 11.45,
+      "learning_rate": 3.960194372076817e-05,
+      "loss": 2.2804,
+      "step": 84790
+    },
+    {
+      "epoch": 11.45,
+      "learning_rate": 3.9573221489491335e-05,
+      "loss": 2.2325,
+      "step": 84800
+    },
+    {
+      "epoch": 11.45,
+      "learning_rate": 3.954450809491404e-05,
+      "loss": 2.263,
+      "step": 84810
+    },
+    {
+      "epoch": 11.46,
+      "learning_rate": 3.9515803539334114e-05,
+      "loss": 2.2624,
+      "step": 84820
+    },
+    {
+      "epoch": 11.46,
+      "learning_rate": 3.9487107825048494e-05,
+      "loss": 2.2328,
+      "step": 84830
+    },
+    {
+      "epoch": 11.46,
+      "learning_rate": 3.945842095435359e-05,
+      "loss": 2.2699,
+      "step": 84840
+    },
+    {
+      "epoch": 11.46,
+      "learning_rate": 3.9429742929544976e-05,
+      "loss": 2.2685,
+      "step": 84850
+    },
+    {
+      "epoch": 11.46,
+      "learning_rate": 3.9401073752917535e-05,
+      "loss": 2.2674,
+      "step": 84860
+    },
+    {
+      "epoch": 11.46,
+      "learning_rate": 3.937241342676551e-05,
+      "loss": 2.2804,
+      "step": 84870
+    },
+    {
+      "epoch": 11.46,
+      "learning_rate": 3.9343761953382366e-05,
+      "loss": 2.2374,
+      "step": 84880
+    },
+    {
+      "epoch": 11.47,
+      "learning_rate": 3.9315119335060915e-05,
+      "loss": 2.2469,
+      "step": 84890
+    },
+    {
+      "epoch": 11.47,
+      "learning_rate": 3.928648557409314e-05,
+      "loss": 2.2643,
+      "step": 84900
+    },
+    {
+      "epoch": 11.47,
+      "learning_rate": 3.925786067277045e-05,
+      "loss": 2.2521,
+      "step": 84910
+    },
+    {
+      "epoch": 11.47,
+      "learning_rate": 3.922924463338363e-05,
+      "loss": 2.2354,
+      "step": 84920
+    },
+    {
+      "epoch": 11.47,
+      "learning_rate": 3.9200637458222415e-05,
+      "loss": 2.2813,
+      "step": 84930
+    },
+    {
+      "epoch": 11.47,
+      "learning_rate": 3.917203914957619e-05,
+      "loss": 2.2346,
+      "step": 84940
+    },
+    {
+      "epoch": 11.47,
+      "learning_rate": 3.914344970973339e-05,
+      "loss": 2.249,
+      "step": 84950
+    },
+    {
+      "epoch": 11.47,
+      "learning_rate": 3.9114869140981914e-05,
+      "loss": 2.2502,
+      "step": 84960
+    },
+    {
+      "epoch": 11.48,
+      "learning_rate": 3.9086297445608825e-05,
+      "loss": 2.2471,
+      "step": 84970
+    },
+    {
+      "epoch": 11.48,
+      "learning_rate": 3.9057734625900485e-05,
+      "loss": 2.2483,
+      "step": 84980
+    },
+    {
+      "epoch": 11.48,
+      "learning_rate": 3.902918068414265e-05,
+      "loss": 2.275,
+      "step": 84990
+    },
+    {
+      "epoch": 11.48,
+      "learning_rate": 3.9000635622620285e-05,
+      "loss": 2.2273,
+      "step": 85000
+    },
+    {
+      "epoch": 11.48,
+      "eval_loss": 2.41872239112854,
+      "eval_runtime": 1269.1084,
+      "eval_samples_per_second": 59.991,
+      "eval_steps_per_second": 5.0,
+      "step": 85000
+    },
+    {
+      "epoch": 11.48,
+      "learning_rate": 3.8972099443617585e-05,
+      "loss": 2.2679,
+      "step": 85010
+    },
+    {
+      "epoch": 11.48,
+      "learning_rate": 3.89435721494182e-05,
+      "loss": 2.2703,
+      "step": 85020
+    },
+    {
+      "epoch": 11.48,
+      "learning_rate": 3.89150537423049e-05,
+      "loss": 2.2657,
+      "step": 85030
+    },
+    {
+      "epoch": 11.49,
+      "learning_rate": 3.888654422455989e-05,
+      "loss": 2.2604,
+      "step": 85040
+    },
+    {
+      "epoch": 11.49,
+      "learning_rate": 3.885804359846455e-05,
+      "loss": 2.2762,
+      "step": 85050
+    },
+    {
+      "epoch": 11.49,
+      "learning_rate": 3.882955186629955e-05,
+      "loss": 2.2538,
+      "step": 85060
+    },
+    {
+      "epoch": 11.49,
+      "learning_rate": 3.880106903034497e-05,
+      "loss": 2.2577,
+      "step": 85070
+    },
+    {
+      "epoch": 11.49,
+      "learning_rate": 3.877259509288006e-05,
+      "loss": 2.2308,
+      "step": 85080
+    },
+    {
+      "epoch": 11.49,
+      "learning_rate": 3.874413005618339e-05,
+      "loss": 2.2528,
+      "step": 85090
+    },
+    {
+      "epoch": 11.49,
+      "learning_rate": 3.8715673922532774e-05,
+      "loss": 2.2535,
+      "step": 85100
+    },
+    {
+      "epoch": 11.49,
+      "learning_rate": 3.868722669420542e-05,
+      "loss": 2.2477,
+      "step": 85110
+    },
+    {
+      "epoch": 11.5,
+      "learning_rate": 3.865878837347786e-05,
+      "loss": 2.2688,
+      "step": 85120
+    },
+    {
+      "epoch": 11.5,
+      "learning_rate": 3.8630358962625596e-05,
+      "loss": 2.2746,
+      "step": 85130
+    },
+    {
+      "epoch": 11.5,
+      "learning_rate": 3.860193846392382e-05,
+      "loss": 2.2542,
+      "step": 85140
+    },
+    {
+      "epoch": 11.5,
+      "learning_rate": 3.857352687964671e-05,
+      "loss": 2.2324,
+      "step": 85150
+    },
+    {
+      "epoch": 11.5,
+      "learning_rate": 3.854512421206796e-05,
+      "loss": 2.2465,
+      "step": 85160
+    },
+    {
+      "epoch": 11.5,
+      "learning_rate": 3.851673046346039e-05,
+      "loss": 2.2471,
+      "step": 85170
+    },
+    {
+      "epoch": 11.5,
+      "learning_rate": 3.848834563609612e-05,
+      "loss": 2.2383,
+      "step": 85180
+    },
+    {
+      "epoch": 11.51,
+      "learning_rate": 3.845996973224667e-05,
+      "loss": 2.2257,
+      "step": 85190
+    },
+    {
+      "epoch": 11.51,
+      "learning_rate": 3.843160275418273e-05,
+      "loss": 2.2336,
+      "step": 85200
+    },
+    {
+      "epoch": 11.51,
+      "learning_rate": 3.840324470417427e-05,
+      "loss": 2.2658,
+      "step": 85210
+    },
+    {
+      "epoch": 11.51,
+      "learning_rate": 3.837489558449069e-05,
+      "loss": 2.2721,
+      "step": 85220
+    },
+    {
+      "epoch": 11.51,
+      "learning_rate": 3.834655539740046e-05,
+      "loss": 2.2607,
+      "step": 85230
+    },
+    {
+      "epoch": 11.51,
+      "learning_rate": 3.8318224145171616e-05,
+      "loss": 2.2669,
+      "step": 85240
+    },
+    {
+      "epoch": 11.51,
+      "learning_rate": 3.828990183007112e-05,
+      "loss": 2.2332,
+      "step": 85250
+    },
+    {
+      "epoch": 11.52,
+      "learning_rate": 3.82615884543655e-05,
+      "loss": 2.2397,
+      "step": 85260
+    },
+    {
+      "epoch": 11.52,
+      "learning_rate": 3.8233284020320546e-05,
+      "loss": 2.2456,
+      "step": 85270
+    },
+    {
+      "epoch": 11.52,
+      "learning_rate": 3.820498853020119e-05,
+      "loss": 2.2724,
+      "step": 85280
+    },
+    {
+      "epoch": 11.52,
+      "learning_rate": 3.817670198627176e-05,
+      "loss": 2.2354,
+      "step": 85290
+    },
+    {
+      "epoch": 11.52,
+      "learning_rate": 3.814842439079577e-05,
+      "loss": 2.2894,
+      "step": 85300
+    },
+    {
+      "epoch": 11.52,
+      "learning_rate": 3.812015574603613e-05,
+      "loss": 2.2673,
+      "step": 85310
+    },
+    {
+      "epoch": 11.52,
+      "learning_rate": 3.8091896054255085e-05,
+      "loss": 2.2554,
+      "step": 85320
+    },
+    {
+      "epoch": 11.52,
+      "learning_rate": 3.806364531771387e-05,
+      "loss": 2.243,
+      "step": 85330
+    },
+    {
+      "epoch": 11.53,
+      "learning_rate": 3.803540353867335e-05,
+      "loss": 2.2548,
+      "step": 85340
+    },
+    {
+      "epoch": 11.53,
+      "learning_rate": 3.800717071939341e-05,
+      "loss": 2.2415,
+      "step": 85350
+    },
+    {
+      "epoch": 11.53,
+      "learning_rate": 3.7978946862133446e-05,
+      "loss": 2.2653,
+      "step": 85360
+    },
+    {
+      "epoch": 11.53,
+      "learning_rate": 3.795073196915195e-05,
+      "loss": 2.2381,
+      "step": 85370
+    },
+    {
+      "epoch": 11.53,
+      "learning_rate": 3.792252604270674e-05,
+      "loss": 2.2625,
+      "step": 85380
+    },
+    {
+      "epoch": 11.53,
+      "learning_rate": 3.789432908505502e-05,
+      "loss": 2.2628,
+      "step": 85390
+    },
+    {
+      "epoch": 11.53,
+      "learning_rate": 3.7866141098453166e-05,
+      "loss": 2.2617,
+      "step": 85400
+    },
+    {
+      "epoch": 11.54,
+      "learning_rate": 3.783796208515683e-05,
+      "loss": 2.252,
+      "step": 85410
+    },
+    {
+      "epoch": 11.54,
+      "learning_rate": 3.7809792047421055e-05,
+      "loss": 2.2558,
+      "step": 85420
+    },
+    {
+      "epoch": 11.54,
+      "learning_rate": 3.7781630987500026e-05,
+      "loss": 2.2576,
+      "step": 85430
+    },
+    {
+      "epoch": 11.54,
+      "learning_rate": 3.7753478907647394e-05,
+      "loss": 2.2186,
+      "step": 85440
+    },
+    {
+      "epoch": 11.54,
+      "learning_rate": 3.772533581011582e-05,
+      "loss": 2.2627,
+      "step": 85450
+    },
+    {
+      "epoch": 11.54,
+      "learning_rate": 3.769720169715748e-05,
+      "loss": 2.253,
+      "step": 85460
+    },
+    {
+      "epoch": 11.54,
+      "learning_rate": 3.7669076571023795e-05,
+      "loss": 2.268,
+      "step": 85470
+    },
+    {
+      "epoch": 11.54,
+      "learning_rate": 3.764096043396539e-05,
+      "loss": 2.2438,
+      "step": 85480
+    },
+    {
+      "epoch": 11.55,
+      "learning_rate": 3.761285328823222e-05,
+      "loss": 2.2613,
+      "step": 85490
+    },
+    {
+      "epoch": 11.55,
+      "learning_rate": 3.7584755136073434e-05,
+      "loss": 2.2621,
+      "step": 85500
+    },
+    {
+      "epoch": 11.55,
+      "learning_rate": 3.75566659797376e-05,
+      "loss": 2.255,
+      "step": 85510
+    },
+    {
+      "epoch": 11.55,
+      "learning_rate": 3.752858582147258e-05,
+      "loss": 2.2866,
+      "step": 85520
+    },
+    {
+      "epoch": 11.55,
+      "learning_rate": 3.7500514663525275e-05,
+      "loss": 2.2806,
+      "step": 85530
+    },
+    {
+      "epoch": 11.55,
+      "learning_rate": 3.747245250814214e-05,
+      "loss": 2.2515,
+      "step": 85540
+    },
+    {
+      "epoch": 11.55,
+      "learning_rate": 3.7444399357568725e-05,
+      "loss": 2.2399,
+      "step": 85550
+    },
+    {
+      "epoch": 11.56,
+      "learning_rate": 3.741635521405e-05,
+      "loss": 2.2913,
+      "step": 85560
+    },
+    {
+      "epoch": 11.56,
+      "learning_rate": 3.7388320079830135e-05,
+      "loss": 2.2761,
+      "step": 85570
+    },
+    {
+      "epoch": 11.56,
+      "learning_rate": 3.736029395715251e-05,
+      "loss": 2.2556,
+      "step": 85580
+    },
+    {
+      "epoch": 11.56,
+      "learning_rate": 3.7332276848259975e-05,
+      "loss": 2.241,
+      "step": 85590
+    },
+    {
+      "epoch": 11.56,
+      "learning_rate": 3.73042687553945e-05,
+      "loss": 2.2683,
+      "step": 85600
+    },
+    {
+      "epoch": 11.56,
+      "learning_rate": 3.727626968079732e-05,
+      "loss": 2.2806,
+      "step": 85610
+    },
+    {
+      "epoch": 11.56,
+      "learning_rate": 3.724827962670913e-05,
+      "loss": 2.2419,
+      "step": 85620
+    },
+    {
+      "epoch": 11.57,
+      "learning_rate": 3.7220298595369655e-05,
+      "loss": 2.2304,
+      "step": 85630
+    },
+    {
+      "epoch": 11.57,
+      "learning_rate": 3.719232658901819e-05,
+      "loss": 2.2557,
+      "step": 85640
+    },
+    {
+      "epoch": 11.57,
+      "learning_rate": 3.716436360989295e-05,
+      "loss": 2.2472,
+      "step": 85650
+    },
+    {
+      "epoch": 11.57,
+      "learning_rate": 3.7136409660231734e-05,
+      "loss": 2.2533,
+      "step": 85660
+    },
+    {
+      "epoch": 11.57,
+      "learning_rate": 3.710846474227151e-05,
+      "loss": 2.2458,
+      "step": 85670
+    },
+    {
+      "epoch": 11.57,
+      "learning_rate": 3.7080528858248485e-05,
+      "loss": 2.2836,
+      "step": 85680
+    },
+    {
+      "epoch": 11.57,
+      "learning_rate": 3.705260201039821e-05,
+      "loss": 2.2657,
+      "step": 85690
+    },
+    {
+      "epoch": 11.57,
+      "learning_rate": 3.70246842009554e-05,
+      "loss": 2.243,
+      "step": 85700
+    },
+    {
+      "epoch": 11.58,
+      "learning_rate": 3.699677543215418e-05,
+      "loss": 2.2652,
+      "step": 85710
+    },
+    {
+      "epoch": 11.58,
+      "learning_rate": 3.696887570622799e-05,
+      "loss": 2.266,
+      "step": 85720
+    },
+    {
+      "epoch": 11.58,
+      "learning_rate": 3.6940985025409284e-05,
+      "loss": 2.2536,
+      "step": 85730
+    },
+    {
+      "epoch": 11.58,
+      "learning_rate": 3.6913103391930065e-05,
+      "loss": 2.2461,
+      "step": 85740
+    },
+    {
+      "epoch": 11.58,
+      "learning_rate": 3.688523080802146e-05,
+      "loss": 2.2437,
+      "step": 85750
+    },
+    {
+      "epoch": 11.58,
+      "learning_rate": 3.6857367275913976e-05,
+      "loss": 2.2458,
+      "step": 85760
+    },
+    {
+      "epoch": 11.58,
+      "learning_rate": 3.6829512797837324e-05,
+      "loss": 2.2815,
+      "step": 85770
+    },
+    {
+      "epoch": 11.59,
+      "learning_rate": 3.680166737602046e-05,
+      "loss": 2.2649,
+      "step": 85780
+    },
+    {
+      "epoch": 11.59,
+      "learning_rate": 3.6773831012691727e-05,
+      "loss": 2.2384,
+      "step": 85790
+    },
+    {
+      "epoch": 11.59,
+      "learning_rate": 3.674600371007866e-05,
+      "loss": 2.2623,
+      "step": 85800
+    },
+    {
+      "epoch": 11.59,
+      "learning_rate": 3.671818547040804e-05,
+      "loss": 2.2425,
+      "step": 85810
+    },
+    {
+      "epoch": 11.59,
+      "learning_rate": 3.669037629590602e-05,
+      "loss": 2.2642,
+      "step": 85820
+    },
+    {
+      "epoch": 11.59,
+      "learning_rate": 3.6662576188797957e-05,
+      "loss": 2.2341,
+      "step": 85830
+    },
+    {
+      "epoch": 11.59,
+      "learning_rate": 3.663478515130859e-05,
+      "loss": 2.2547,
+      "step": 85840
+    },
+    {
+      "epoch": 11.59,
+      "learning_rate": 3.660700318566168e-05,
+      "loss": 2.2511,
+      "step": 85850
+    },
+    {
+      "epoch": 11.6,
+      "learning_rate": 3.6579230294080494e-05,
+      "loss": 2.237,
+      "step": 85860
+    },
+    {
+      "epoch": 11.6,
+      "learning_rate": 3.655146647878761e-05,
+      "loss": 2.2527,
+      "step": 85870
+    },
+    {
+      "epoch": 11.6,
+      "learning_rate": 3.652371174200468e-05,
+      "loss": 2.2485,
+      "step": 85880
+    },
+    {
+      "epoch": 11.6,
+      "learning_rate": 3.649596608595273e-05,
+      "loss": 2.2424,
+      "step": 85890
+    },
+    {
+      "epoch": 11.6,
+      "learning_rate": 3.646822951285202e-05,
+      "loss": 2.2348,
+      "step": 85900
+    },
+    {
+      "epoch": 11.6,
+      "learning_rate": 3.6440502024922166e-05,
+      "loss": 2.2715,
+      "step": 85910
+    },
+    {
+      "epoch": 11.6,
+      "learning_rate": 3.6412783624382095e-05,
+      "loss": 2.2591,
+      "step": 85920
+    },
+    {
+      "epoch": 11.61,
+      "learning_rate": 3.638507431344976e-05,
+      "loss": 2.2422,
+      "step": 85930
+    },
+    {
+      "epoch": 11.61,
+      "learning_rate": 3.635737409434266e-05,
+      "loss": 2.2611,
+      "step": 85940
+    },
+    {
+      "epoch": 11.61,
+      "learning_rate": 3.6329682969277355e-05,
+      "loss": 2.269,
+      "step": 85950
+    },
+    {
+      "epoch": 11.61,
+      "learning_rate": 3.6302000940469876e-05,
+      "loss": 2.2405,
+      "step": 85960
+    },
+    {
+      "epoch": 11.61,
+      "learning_rate": 3.627432801013541e-05,
+      "loss": 2.2216,
+      "step": 85970
+    },
+    {
+      "epoch": 11.61,
+      "learning_rate": 3.624666418048835e-05,
+      "loss": 2.2494,
+      "step": 85980
+    },
+    {
+      "epoch": 11.61,
+      "learning_rate": 3.621900945374254e-05,
+      "loss": 2.2437,
+      "step": 85990
+    },
+    {
+      "epoch": 11.62,
+      "learning_rate": 3.619136383211097e-05,
+      "loss": 2.24,
+      "step": 86000
+    },
+    {
+      "epoch": 11.62,
+      "eval_loss": 2.4159932136535645,
+      "eval_runtime": 1266.6001,
+      "eval_samples_per_second": 60.11,
+      "eval_steps_per_second": 5.009,
+      "step": 86000
+    },
+    {
+      "epoch": 11.62,
+      "learning_rate": 3.616372731780586e-05,
+      "loss": 2.2454,
+      "step": 86010
+    },
+    {
+      "epoch": 11.62,
+      "learning_rate": 3.6136099913038896e-05,
+      "loss": 2.2753,
+      "step": 86020
+    },
+    {
+      "epoch": 11.62,
+      "learning_rate": 3.610848162002078e-05,
+      "loss": 2.2292,
+      "step": 86030
+    },
+    {
+      "epoch": 11.62,
+      "learning_rate": 3.608087244096178e-05,
+      "loss": 2.2566,
+      "step": 86040
+    },
+    {
+      "epoch": 11.62,
+      "learning_rate": 3.605327237807109e-05,
+      "loss": 2.2602,
+      "step": 86050
+    },
+    {
+      "epoch": 11.62,
+      "learning_rate": 3.6025681433557423e-05,
+      "loss": 2.2812,
+      "step": 86060
+    },
+    {
+      "epoch": 11.62,
+      "learning_rate": 3.599809960962874e-05,
+      "loss": 2.2561,
+      "step": 86070
+    },
+    {
+      "epoch": 11.63,
+      "learning_rate": 3.59705269084922e-05,
+      "loss": 2.2744,
+      "step": 86080
+    },
+    {
+      "epoch": 11.63,
+      "learning_rate": 3.594296333235425e-05,
+      "loss": 2.2212,
+      "step": 86090
+    },
+    {
+      "epoch": 11.63,
+      "learning_rate": 3.591540888342056e-05,
+      "loss": 2.2672,
+      "step": 86100
+    },
+    {
+      "epoch": 11.63,
+      "learning_rate": 3.588786356389623e-05,
+      "loss": 2.2501,
+      "step": 86110
+    },
+    {
+      "epoch": 11.63,
+      "learning_rate": 3.5860327375985464e-05,
+      "loss": 2.2923,
+      "step": 86120
+    },
+    {
+      "epoch": 11.63,
+      "learning_rate": 3.583280032189175e-05,
+      "loss": 2.2696,
+      "step": 86130
+    },
+    {
+      "epoch": 11.63,
+      "learning_rate": 3.5805282403817996e-05,
+      "loss": 2.2665,
+      "step": 86140
+    },
+    {
+      "epoch": 11.64,
+      "learning_rate": 3.577777362396617e-05,
+      "loss": 2.2346,
+      "step": 86150
+    },
+    {
+      "epoch": 11.64,
+      "learning_rate": 3.5750273984537696e-05,
+      "loss": 2.2338,
+      "step": 86160
+    },
+    {
+      "epoch": 11.64,
+      "learning_rate": 3.572278348773316e-05,
+      "loss": 2.2661,
+      "step": 86170
+    },
+    {
+      "epoch": 11.64,
+      "learning_rate": 3.569530213575236e-05,
+      "loss": 2.24,
+      "step": 86180
+    },
+    {
+      "epoch": 11.64,
+      "learning_rate": 3.566782993079455e-05,
+      "loss": 2.2366,
+      "step": 86190
+    },
+    {
+      "epoch": 11.64,
+      "learning_rate": 3.56403668750581e-05,
+      "loss": 2.2734,
+      "step": 86200
+    },
+    {
+      "epoch": 11.64,
+      "learning_rate": 3.561291297074066e-05,
+      "loss": 2.2457,
+      "step": 86210
+    },
+    {
+      "epoch": 11.64,
+      "learning_rate": 3.558546822003923e-05,
+      "loss": 2.2572,
+      "step": 86220
+    },
+    {
+      "epoch": 11.65,
+      "learning_rate": 3.5558032625149944e-05,
+      "loss": 2.2711,
+      "step": 86230
+    },
+    {
+      "epoch": 11.65,
+      "learning_rate": 3.5530606188268446e-05,
+      "loss": 2.2589,
+      "step": 86240
+    },
+    {
+      "epoch": 11.65,
+      "learning_rate": 3.5503188911589295e-05,
+      "loss": 2.2484,
+      "step": 86250
+    },
+    {
+      "epoch": 11.65,
+      "learning_rate": 3.547578079730658e-05,
+      "loss": 2.2606,
+      "step": 86260
+    },
+    {
+      "epoch": 11.65,
+      "learning_rate": 3.544838184761365e-05,
+      "loss": 2.2673,
+      "step": 86270
+    },
+    {
+      "epoch": 11.65,
+      "learning_rate": 3.5420992064703014e-05,
+      "loss": 2.2539,
+      "step": 86280
+    },
+    {
+      "epoch": 11.65,
+      "learning_rate": 3.539361145076647e-05,
+      "loss": 2.2982,
+      "step": 86290
+    },
+    {
+      "epoch": 11.66,
+      "learning_rate": 3.5366240007995066e-05,
+      "loss": 2.2708,
+      "step": 86300
+    },
+    {
+      "epoch": 11.66,
+      "learning_rate": 3.533887773857926e-05,
+      "loss": 2.2685,
+      "step": 86310
+    },
+    {
+      "epoch": 11.66,
+      "learning_rate": 3.531152464470857e-05,
+      "loss": 2.2644,
+      "step": 86320
+    },
+    {
+      "epoch": 11.66,
+      "learning_rate": 3.5284180728571895e-05,
+      "loss": 2.2739,
+      "step": 86330
+    },
+    {
+      "epoch": 11.66,
+      "learning_rate": 3.525684599235742e-05,
+      "loss": 2.2678,
+      "step": 86340
+    },
+    {
+      "epoch": 11.66,
+      "learning_rate": 3.52295204382525e-05,
+      "loss": 2.2461,
+      "step": 86350
+    },
+    {
+      "epoch": 11.66,
+      "learning_rate": 3.520220406844389e-05,
+      "loss": 2.2525,
+      "step": 86360
+    },
+    {
+      "epoch": 11.67,
+      "learning_rate": 3.517489688511749e-05,
+      "loss": 2.2821,
+      "step": 86370
+    },
+    {
+      "epoch": 11.67,
+      "learning_rate": 3.5147598890458446e-05,
+      "loss": 2.2609,
+      "step": 86380
+    },
+    {
+      "epoch": 11.67,
+      "learning_rate": 3.5120310086651346e-05,
+      "loss": 2.2762,
+      "step": 86390
+    },
+    {
+      "epoch": 11.67,
+      "learning_rate": 3.509303047587987e-05,
+      "loss": 2.2621,
+      "step": 86400
+    },
+    {
+      "epoch": 11.67,
+      "learning_rate": 3.506576006032698e-05,
+      "loss": 2.2454,
+      "step": 86410
+    },
+    {
+      "epoch": 11.67,
+      "learning_rate": 3.5038498842175004e-05,
+      "loss": 2.2658,
+      "step": 86420
+    },
+    {
+      "epoch": 11.67,
+      "learning_rate": 3.501124682360542e-05,
+      "loss": 2.262,
+      "step": 86430
+    },
+    {
+      "epoch": 11.67,
+      "learning_rate": 3.498400400679911e-05,
+      "loss": 2.2762,
+      "step": 86440
+    },
+    {
+      "epoch": 11.68,
+      "learning_rate": 3.4956770393936e-05,
+      "loss": 2.2713,
+      "step": 86450
+    },
+    {
+      "epoch": 11.68,
+      "learning_rate": 3.492954598719546e-05,
+      "loss": 2.2415,
+      "step": 86460
+    },
+    {
+      "epoch": 11.68,
+      "learning_rate": 3.490233078875615e-05,
+      "loss": 2.2533,
+      "step": 86470
+    },
+    {
+      "epoch": 11.68,
+      "learning_rate": 3.487512480079583e-05,
+      "loss": 2.2633,
+      "step": 86480
+    },
+    {
+      "epoch": 11.68,
+      "learning_rate": 3.484792802549166e-05,
+      "loss": 2.243,
+      "step": 86490
+    },
+    {
+      "epoch": 11.68,
+      "learning_rate": 3.482074046501992e-05,
+      "loss": 2.2435,
+      "step": 86500
+    },
+    {
+      "epoch": 11.68,
+      "learning_rate": 3.4793562121556385e-05,
+      "loss": 2.2606,
+      "step": 86510
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 3.476639299727584e-05,
+      "loss": 2.2324,
+      "step": 86520
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 3.473923309435246e-05,
+      "loss": 2.261,
+      "step": 86530
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 3.471208241495974e-05,
+      "loss": 2.2802,
+      "step": 86540
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 3.4684940961270255e-05,
+      "loss": 2.2521,
+      "step": 86550
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 3.465780873545604e-05,
+      "loss": 2.2448,
+      "step": 86560
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 3.463068573968828e-05,
+      "loss": 2.2636,
+      "step": 86570
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 3.460357197613737e-05,
+      "loss": 2.2837,
+      "step": 86580
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 3.4576467446973174e-05,
+      "loss": 2.2476,
+      "step": 86590
+    },
+    {
+      "epoch": 11.7,
+      "learning_rate": 3.4549372154364573e-05,
+      "loss": 2.248,
+      "step": 86600
+    },
+    {
+      "epoch": 11.7,
+      "learning_rate": 3.452228610047983e-05,
+      "loss": 2.2325,
+      "step": 86610
+    },
+    {
+      "epoch": 11.7,
+      "learning_rate": 3.4495209287486505e-05,
+      "loss": 2.2653,
+      "step": 86620
+    },
+    {
+      "epoch": 11.7,
+      "learning_rate": 3.446814171755131e-05,
+      "loss": 2.2671,
+      "step": 86630
+    },
+    {
+      "epoch": 11.7,
+      "learning_rate": 3.444108339284041e-05,
+      "loss": 2.2438,
+      "step": 86640
+    },
+    {
+      "epoch": 11.7,
+      "learning_rate": 3.441403431551889e-05,
+      "loss": 2.243,
+      "step": 86650
+    },
+    {
+      "epoch": 11.7,
+      "learning_rate": 3.438699448775141e-05,
+      "loss": 2.2426,
+      "step": 86660
+    },
+    {
+      "epoch": 11.71,
+      "learning_rate": 3.435996391170184e-05,
+      "loss": 2.2596,
+      "step": 86670
+    },
+    {
+      "epoch": 11.71,
+      "learning_rate": 3.433564430526381e-05,
+      "loss": 2.2849,
+      "step": 86680
+    },
+    {
+      "epoch": 11.71,
+      "learning_rate": 3.430863131343676e-05,
+      "loss": 2.2715,
+      "step": 86690
+    },
+    {
+      "epoch": 11.71,
+      "learning_rate": 3.4281627579598484e-05,
+      "loss": 2.2612,
+      "step": 86700
+    },
+    {
+      "epoch": 11.71,
+      "learning_rate": 3.425463310590985e-05,
+      "loss": 2.2472,
+      "step": 86710
+    },
+    {
+      "epoch": 11.71,
+      "learning_rate": 3.4227647894531e-05,
+      "loss": 2.2421,
+      "step": 86720
+    },
+    {
+      "epoch": 11.71,
+      "learning_rate": 3.420067194762147e-05,
+      "loss": 2.2762,
+      "step": 86730
+    },
+    {
+      "epoch": 11.71,
+      "learning_rate": 3.417370526733986e-05,
+      "loss": 2.275,
+      "step": 86740
+    },
+    {
+      "epoch": 11.72,
+      "learning_rate": 3.414674785584422e-05,
+      "loss": 2.25,
+      "step": 86750
+    },
+    {
+      "epoch": 11.72,
+      "learning_rate": 3.4119799715291753e-05,
+      "loss": 2.2738,
+      "step": 86760
+    },
+    {
+      "epoch": 11.72,
+      "learning_rate": 3.4092860847838846e-05,
+      "loss": 2.2784,
+      "step": 86770
+    },
+    {
+      "epoch": 11.72,
+      "learning_rate": 3.406593125564134e-05,
+      "loss": 2.2555,
+      "step": 86780
+    },
+    {
+      "epoch": 11.72,
+      "learning_rate": 3.403901094085418e-05,
+      "loss": 2.2682,
+      "step": 86790
+    },
+    {
+      "epoch": 11.72,
+      "learning_rate": 3.401209990563158e-05,
+      "loss": 2.2311,
+      "step": 86800
+    },
+    {
+      "epoch": 11.72,
+      "learning_rate": 3.3985198152127126e-05,
+      "loss": 2.2455,
+      "step": 86810
+    },
+    {
+      "epoch": 11.73,
+      "learning_rate": 3.3958305682493494e-05,
+      "loss": 2.2813,
+      "step": 86820
+    },
+    {
+      "epoch": 11.73,
+      "learning_rate": 3.393142249888283e-05,
+      "loss": 2.2639,
+      "step": 86830
+    },
+    {
+      "epoch": 11.73,
+      "learning_rate": 3.390454860344624e-05,
+      "loss": 2.2514,
+      "step": 86840
+    },
+    {
+      "epoch": 11.73,
+      "learning_rate": 3.387768399833438e-05,
+      "loss": 2.2602,
+      "step": 86850
+    },
+    {
+      "epoch": 11.73,
+      "learning_rate": 3.385082868569698e-05,
+      "loss": 2.242,
+      "step": 86860
+    },
+    {
+      "epoch": 11.73,
+      "learning_rate": 3.3823982667683164e-05,
+      "loss": 2.2474,
+      "step": 86870
+    },
+    {
+      "epoch": 11.73,
+      "learning_rate": 3.3797145946441166e-05,
+      "loss": 2.2528,
+      "step": 86880
+    },
+    {
+      "epoch": 11.74,
+      "learning_rate": 3.3770318524118536e-05,
+      "loss": 2.2699,
+      "step": 86890
+    },
+    {
+      "epoch": 11.74,
+      "learning_rate": 3.3743500402862144e-05,
+      "loss": 2.2558,
+      "step": 86900
+    },
+    {
+      "epoch": 11.74,
+      "learning_rate": 3.371669158481805e-05,
+      "loss": 2.2727,
+      "step": 86910
+    },
+    {
+      "epoch": 11.74,
+      "learning_rate": 3.3689892072131526e-05,
+      "loss": 2.2383,
+      "step": 86920
+    },
+    {
+      "epoch": 11.74,
+      "learning_rate": 3.3663101866947216e-05,
+      "loss": 2.253,
+      "step": 86930
+    },
+    {
+      "epoch": 11.74,
+      "learning_rate": 3.363632097140892e-05,
+      "loss": 2.2679,
+      "step": 86940
+    },
+    {
+      "epoch": 11.74,
+      "learning_rate": 3.360954938765975e-05,
+      "loss": 2.2306,
+      "step": 86950
+    },
+    {
+      "epoch": 11.74,
+      "learning_rate": 3.358278711784207e-05,
+      "loss": 2.2673,
+      "step": 86960
+    },
+    {
+      "epoch": 11.75,
+      "learning_rate": 3.355603416409741e-05,
+      "loss": 2.2441,
+      "step": 86970
+    },
+    {
+      "epoch": 11.75,
+      "learning_rate": 3.3529290528566716e-05,
+      "loss": 2.2482,
+      "step": 86980
+    },
+    {
+      "epoch": 11.75,
+      "learning_rate": 3.350255621339005e-05,
+      "loss": 2.228,
+      "step": 86990
+    },
+    {
+      "epoch": 11.75,
+      "learning_rate": 3.347583122070679e-05,
+      "loss": 2.2574,
+      "step": 87000
+    },
+    {
+      "epoch": 11.75,
+      "eval_loss": 2.4128317832946777,
+      "eval_runtime": 1268.9991,
+      "eval_samples_per_second": 59.996,
+      "eval_steps_per_second": 5.0,
+      "step": 87000
+    },
+    {
+      "epoch": 11.75,
+      "learning_rate": 3.3449115552655495e-05,
+      "loss": 2.2396,
+      "step": 87010
+    },
+    {
+      "epoch": 11.75,
+      "learning_rate": 3.3422409211374096e-05,
+      "loss": 2.2359,
+      "step": 87020
+    },
+    {
+      "epoch": 11.75,
+      "learning_rate": 3.339571219899979e-05,
+      "loss": 2.2363,
+      "step": 87030
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 3.336902451766879e-05,
+      "loss": 2.2553,
+      "step": 87040
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 3.334234616951685e-05,
+      "loss": 2.284,
+      "step": 87050
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 3.33156771566788e-05,
+      "loss": 2.2833,
+      "step": 87060
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 3.328901748128882e-05,
+      "loss": 2.2608,
+      "step": 87070
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 3.326236714548029e-05,
+      "loss": 2.2577,
+      "step": 87080
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 3.323572615138582e-05,
+      "loss": 2.2444,
+      "step": 87090
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 3.3209094501137364e-05,
+      "loss": 2.2534,
+      "step": 87100
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 3.318247219686603e-05,
+      "loss": 2.2393,
+      "step": 87110
+    },
+    {
+      "epoch": 11.77,
+      "learning_rate": 3.31558592407022e-05,
+      "loss": 2.2499,
+      "step": 87120
+    },
+    {
+      "epoch": 11.77,
+      "learning_rate": 3.312925563477557e-05,
+      "loss": 2.259,
+      "step": 87130
+    },
+    {
+      "epoch": 11.77,
+      "learning_rate": 3.310266138121501e-05,
+      "loss": 2.2143,
+      "step": 87140
+    },
+    {
+      "epoch": 11.77,
+      "learning_rate": 3.307607648214874e-05,
+      "loss": 2.2542,
+      "step": 87150
+    },
+    {
+      "epoch": 11.77,
+      "learning_rate": 3.30495009397041e-05,
+      "loss": 2.255,
+      "step": 87160
+    },
+    {
+      "epoch": 11.77,
+      "learning_rate": 3.302293475600775e-05,
+      "loss": 2.2248,
+      "step": 87170
+    },
+    {
+      "epoch": 11.77,
+      "learning_rate": 3.299637793318565e-05,
+      "loss": 2.2314,
+      "step": 87180
+    },
+    {
+      "epoch": 11.78,
+      "learning_rate": 3.296983047336295e-05,
+      "loss": 2.2742,
+      "step": 87190
+    },
+    {
+      "epoch": 11.78,
+      "learning_rate": 3.294329237866403e-05,
+      "loss": 2.2717,
+      "step": 87200
+    },
+    {
+      "epoch": 11.78,
+      "learning_rate": 3.2916763651212525e-05,
+      "loss": 2.2772,
+      "step": 87210
+    },
+    {
+      "epoch": 11.78,
+      "learning_rate": 3.28902442931314e-05,
+      "loss": 2.2465,
+      "step": 87220
+    },
+    {
+      "epoch": 11.78,
+      "learning_rate": 3.2863734306542865e-05,
+      "loss": 2.2283,
+      "step": 87230
+    },
+    {
+      "epoch": 11.78,
+      "learning_rate": 3.2837233693568204e-05,
+      "loss": 2.2491,
+      "step": 87240
+    },
+    {
+      "epoch": 11.78,
+      "learning_rate": 3.28107424563282e-05,
+      "loss": 2.2533,
+      "step": 87250
+    },
+    {
+      "epoch": 11.79,
+      "learning_rate": 3.278426059694267e-05,
+      "loss": 2.2307,
+      "step": 87260
+    },
+    {
+      "epoch": 11.79,
+      "learning_rate": 3.275778811753085e-05,
+      "loss": 2.243,
+      "step": 87270
+    },
+    {
+      "epoch": 11.79,
+      "learning_rate": 3.273132502021114e-05,
+      "loss": 2.274,
+      "step": 87280
+    },
+    {
+      "epoch": 11.79,
+      "learning_rate": 3.270487130710114e-05,
+      "loss": 2.26,
+      "step": 87290
+    },
+    {
+      "epoch": 11.79,
+      "learning_rate": 3.267842698031784e-05,
+      "loss": 2.2644,
+      "step": 87300
+    },
+    {
+      "epoch": 11.79,
+      "learning_rate": 3.265199204197736e-05,
+      "loss": 2.2322,
+      "step": 87310
+    },
+    {
+      "epoch": 11.79,
+      "learning_rate": 3.262556649419505e-05,
+      "loss": 2.2847,
+      "step": 87320
+    },
+    {
+      "epoch": 11.79,
+      "learning_rate": 3.2599150339085694e-05,
+      "loss": 2.232,
+      "step": 87330
+    },
+    {
+      "epoch": 11.8,
+      "learning_rate": 3.257274357876308e-05,
+      "loss": 2.2685,
+      "step": 87340
+    },
+    {
+      "epoch": 11.8,
+      "learning_rate": 3.2546346215340444e-05,
+      "loss": 2.2581,
+      "step": 87350
+    },
+    {
+      "epoch": 11.8,
+      "learning_rate": 3.251995825093014e-05,
+      "loss": 2.2493,
+      "step": 87360
+    },
+    {
+      "epoch": 11.8,
+      "learning_rate": 3.249357968764381e-05,
+      "loss": 2.2833,
+      "step": 87370
+    },
+    {
+      "epoch": 11.8,
+      "learning_rate": 3.246721052759239e-05,
+      "loss": 2.2727,
+      "step": 87380
+    },
+    {
+      "epoch": 11.8,
+      "learning_rate": 3.244085077288599e-05,
+      "loss": 2.2455,
+      "step": 87390
+    },
+    {
+      "epoch": 11.8,
+      "learning_rate": 3.2414500425634035e-05,
+      "loss": 2.2481,
+      "step": 87400
+    },
+    {
+      "epoch": 11.81,
+      "learning_rate": 3.238815948794508e-05,
+      "loss": 2.2802,
+      "step": 87410
+    },
+    {
+      "epoch": 11.81,
+      "learning_rate": 3.236182796192707e-05,
+      "loss": 2.2547,
+      "step": 87420
+    },
+    {
+      "epoch": 11.81,
+      "learning_rate": 3.233550584968722e-05,
+      "loss": 2.2344,
+      "step": 87430
+    },
+    {
+      "epoch": 11.81,
+      "learning_rate": 3.230919315333174e-05,
+      "loss": 2.275,
+      "step": 87440
+    },
+    {
+      "epoch": 11.81,
+      "learning_rate": 3.2282889874966385e-05,
+      "loss": 2.2507,
+      "step": 87450
+    },
+    {
+      "epoch": 11.81,
+      "learning_rate": 3.225659601669595e-05,
+      "loss": 2.2678,
+      "step": 87460
+    },
+    {
+      "epoch": 11.81,
+      "learning_rate": 3.22303115806246e-05,
+      "loss": 2.2468,
+      "step": 87470
+    },
+    {
+      "epoch": 11.81,
+      "learning_rate": 3.220403656885568e-05,
+      "loss": 2.2318,
+      "step": 87480
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 3.217777098349178e-05,
+      "loss": 2.2436,
+      "step": 87490
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 3.21515148266348e-05,
+      "loss": 2.2264,
+      "step": 87500
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 3.212526810038583e-05,
+      "loss": 2.259,
+      "step": 87510
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 3.209903080684516e-05,
+      "loss": 2.2481,
+      "step": 87520
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 3.207280294811246e-05,
+      "loss": 2.2694,
+      "step": 87530
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 3.204658452628648e-05,
+      "loss": 2.2597,
+      "step": 87540
+    },
+    {
+      "epoch": 11.82,
+      "learning_rate": 3.202037554346539e-05,
+      "loss": 2.2742,
+      "step": 87550
+    },
+    {
+      "epoch": 11.83,
+      "learning_rate": 3.1994176001746494e-05,
+      "loss": 2.2713,
+      "step": 87560
+    },
+    {
+      "epoch": 11.83,
+      "learning_rate": 3.1967985903226294e-05,
+      "loss": 2.2684,
+      "step": 87570
+    },
+    {
+      "epoch": 11.83,
+      "learning_rate": 3.19418052500007e-05,
+      "loss": 2.2764,
+      "step": 87580
+    },
+    {
+      "epoch": 11.83,
+      "learning_rate": 3.1915634044164714e-05,
+      "loss": 2.2284,
+      "step": 87590
+    },
+    {
+      "epoch": 11.83,
+      "learning_rate": 3.1889472287812665e-05,
+      "loss": 2.2504,
+      "step": 87600
+    },
+    {
+      "epoch": 11.83,
+      "learning_rate": 3.1863319983038035e-05,
+      "loss": 2.2722,
+      "step": 87610
+    },
+    {
+      "epoch": 11.83,
+      "learning_rate": 3.183717713193365e-05,
+      "loss": 2.2573,
+      "step": 87620
+    },
+    {
+      "epoch": 11.84,
+      "learning_rate": 3.1811043736591664e-05,
+      "loss": 2.2577,
+      "step": 87630
+    },
+    {
+      "epoch": 11.84,
+      "learning_rate": 3.1784919799103154e-05,
+      "loss": 2.2515,
+      "step": 87640
+    },
+    {
+      "epoch": 11.84,
+      "learning_rate": 3.175880532155877e-05,
+      "loss": 2.2441,
+      "step": 87650
+    },
+    {
+      "epoch": 11.84,
+      "learning_rate": 3.17327003060482e-05,
+      "loss": 2.2554,
+      "step": 87660
+    },
+    {
+      "epoch": 11.84,
+      "learning_rate": 3.1706604754660525e-05,
+      "loss": 2.2694,
+      "step": 87670
+    },
+    {
+      "epoch": 11.84,
+      "learning_rate": 3.168051866948395e-05,
+      "loss": 2.275,
+      "step": 87680
+    },
+    {
+      "epoch": 11.84,
+      "learning_rate": 3.165444205260594e-05,
+      "loss": 2.2448,
+      "step": 87690
+    },
+    {
+      "epoch": 11.84,
+      "learning_rate": 3.1628374906113305e-05,
+      "loss": 2.2709,
+      "step": 87700
+    },
+    {
+      "epoch": 11.85,
+      "learning_rate": 3.1602317232091954e-05,
+      "loss": 2.2691,
+      "step": 87710
+    },
+    {
+      "epoch": 11.85,
+      "learning_rate": 3.15762690326271e-05,
+      "loss": 2.2457,
+      "step": 87720
+    },
+    {
+      "epoch": 11.85,
+      "learning_rate": 3.1550230309803266e-05,
+      "loss": 2.2549,
+      "step": 87730
+    },
+    {
+      "epoch": 11.85,
+      "learning_rate": 3.152420106570408e-05,
+      "loss": 2.265,
+      "step": 87740
+    },
+    {
+      "epoch": 11.85,
+      "learning_rate": 3.149818130241256e-05,
+      "loss": 2.2417,
+      "step": 87750
+    },
+    {
+      "epoch": 11.85,
+      "learning_rate": 3.147217102201084e-05,
+      "loss": 2.26,
+      "step": 87760
+    },
+    {
+      "epoch": 11.85,
+      "learning_rate": 3.144617022658032e-05,
+      "loss": 2.277,
+      "step": 87770
+    },
+    {
+      "epoch": 11.86,
+      "learning_rate": 3.142017891820172e-05,
+      "loss": 2.2828,
+      "step": 87780
+    },
+    {
+      "epoch": 11.86,
+      "learning_rate": 3.139419709895495e-05,
+      "loss": 2.2539,
+      "step": 87790
+    },
+    {
+      "epoch": 11.86,
+      "learning_rate": 3.1368224770919124e-05,
+      "loss": 2.2667,
+      "step": 87800
+    },
+    {
+      "epoch": 11.86,
+      "learning_rate": 3.1342261936172594e-05,
+      "loss": 2.2719,
+      "step": 87810
+    },
+    {
+      "epoch": 11.86,
+      "learning_rate": 3.131630859679304e-05,
+      "loss": 2.2575,
+      "step": 87820
+    },
+    {
+      "epoch": 11.86,
+      "learning_rate": 3.12903647548574e-05,
+      "loss": 2.2725,
+      "step": 87830
+    },
+    {
+      "epoch": 11.86,
+      "learning_rate": 3.1264430412441633e-05,
+      "loss": 2.2761,
+      "step": 87840
+    },
+    {
+      "epoch": 11.86,
+      "learning_rate": 3.123850557162119e-05,
+      "loss": 2.2616,
+      "step": 87850
+    },
+    {
+      "epoch": 11.87,
+      "learning_rate": 3.1212590234470584e-05,
+      "loss": 2.2614,
+      "step": 87860
+    },
+    {
+      "epoch": 11.87,
+      "learning_rate": 3.118668440306374e-05,
+      "loss": 2.2485,
+      "step": 87870
+    },
+    {
+      "epoch": 11.87,
+      "learning_rate": 3.116078807947366e-05,
+      "loss": 2.2588,
+      "step": 87880
+    },
+    {
+      "epoch": 11.87,
+      "learning_rate": 3.113490126577262e-05,
+      "loss": 2.2815,
+      "step": 87890
+    },
+    {
+      "epoch": 11.87,
+      "learning_rate": 3.110902396403224e-05,
+      "loss": 2.2565,
+      "step": 87900
+    },
+    {
+      "epoch": 11.87,
+      "learning_rate": 3.1083156176323264e-05,
+      "loss": 2.2716,
+      "step": 87910
+    },
+    {
+      "epoch": 11.87,
+      "learning_rate": 3.10572979047157e-05,
+      "loss": 2.2572,
+      "step": 87920
+    },
+    {
+      "epoch": 11.88,
+      "learning_rate": 3.1031449151278854e-05,
+      "loss": 2.2618,
+      "step": 87930
+    },
+    {
+      "epoch": 11.88,
+      "learning_rate": 3.1005609918081154e-05,
+      "loss": 2.2723,
+      "step": 87940
+    },
+    {
+      "epoch": 11.88,
+      "learning_rate": 3.097978020719043e-05,
+      "loss": 2.2762,
+      "step": 87950
+    },
+    {
+      "epoch": 11.88,
+      "learning_rate": 3.0953960020673593e-05,
+      "loss": 2.2567,
+      "step": 87960
+    },
+    {
+      "epoch": 11.88,
+      "learning_rate": 3.0928149360596834e-05,
+      "loss": 2.2481,
+      "step": 87970
+    },
+    {
+      "epoch": 11.88,
+      "learning_rate": 3.090234822902568e-05,
+      "loss": 2.2415,
+      "step": 87980
+    },
+    {
+      "epoch": 11.88,
+      "learning_rate": 3.0876556628024754e-05,
+      "loss": 2.2489,
+      "step": 87990
+    },
+    {
+      "epoch": 11.89,
+      "learning_rate": 3.0850774559658075e-05,
+      "loss": 2.269,
+      "step": 88000
+    },
+    {
+      "epoch": 11.89,
+      "eval_loss": 2.410313606262207,
+      "eval_runtime": 1269.0076,
+      "eval_samples_per_second": 59.996,
+      "eval_steps_per_second": 5.0,
+      "step": 88000
+    },
+    {
+      "epoch": 11.89,
+      "learning_rate": 3.082500202598866e-05,
+      "loss": 2.2366,
+      "step": 88010
+    },
+    {
+      "epoch": 11.89,
+      "learning_rate": 3.0799239029078975e-05,
+      "loss": 2.2552,
+      "step": 88020
+    },
+    {
+      "epoch": 11.89,
+      "learning_rate": 3.077348557099072e-05,
+      "loss": 2.2643,
+      "step": 88030
+    },
+    {
+      "epoch": 11.89,
+      "learning_rate": 3.074774165378471e-05,
+      "loss": 2.2255,
+      "step": 88040
+    },
+    {
+      "epoch": 11.89,
+      "learning_rate": 3.072200727952106e-05,
+      "loss": 2.26,
+      "step": 88050
+    },
+    {
+      "epoch": 11.89,
+      "learning_rate": 3.069628245025907e-05,
+      "loss": 2.252,
+      "step": 88060
+    },
+    {
+      "epoch": 11.89,
+      "learning_rate": 3.067056716805742e-05,
+      "loss": 2.2495,
+      "step": 88070
+    },
+    {
+      "epoch": 11.9,
+      "learning_rate": 3.064486143497386e-05,
+      "loss": 2.2467,
+      "step": 88080
+    },
+    {
+      "epoch": 11.9,
+      "learning_rate": 3.061916525306543e-05,
+      "loss": 2.241,
+      "step": 88090
+    },
+    {
+      "epoch": 11.9,
+      "learning_rate": 3.0593478624388485e-05,
+      "loss": 2.2292,
+      "step": 88100
+    },
+    {
+      "epoch": 11.9,
+      "learning_rate": 3.056780155099848e-05,
+      "loss": 2.2353,
+      "step": 88110
+    },
+    {
+      "epoch": 11.9,
+      "learning_rate": 3.054213403495023e-05,
+      "loss": 2.2547,
+      "step": 88120
+    },
+    {
+      "epoch": 11.9,
+      "learning_rate": 3.0516476078297743e-05,
+      "loss": 2.257,
+      "step": 88130
+    },
+    {
+      "epoch": 11.9,
+      "learning_rate": 3.049082768309415e-05,
+      "loss": 2.2603,
+      "step": 88140
+    },
+    {
+      "epoch": 11.91,
+      "learning_rate": 3.046518885139203e-05,
+      "loss": 2.2199,
+      "step": 88150
+    },
+    {
+      "epoch": 11.91,
+      "learning_rate": 3.0439559585243046e-05,
+      "loss": 2.2634,
+      "step": 88160
+    },
+    {
+      "epoch": 11.91,
+      "learning_rate": 3.0413939886698084e-05,
+      "loss": 2.2445,
+      "step": 88170
+    },
+    {
+      "epoch": 11.91,
+      "learning_rate": 3.038832975780738e-05,
+      "loss": 2.271,
+      "step": 88180
+    },
+    {
+      "epoch": 11.91,
+      "learning_rate": 3.0362729200620285e-05,
+      "loss": 2.2581,
+      "step": 88190
+    },
+    {
+      "epoch": 11.91,
+      "learning_rate": 3.033713821718553e-05,
+      "loss": 2.2765,
+      "step": 88200
+    },
+    {
+      "epoch": 11.91,
+      "learning_rate": 3.0311556809550847e-05,
+      "loss": 2.2611,
+      "step": 88210
+    },
+    {
+      "epoch": 11.91,
+      "learning_rate": 3.028598497976341e-05,
+      "loss": 2.2454,
+      "step": 88220
+    },
+    {
+      "epoch": 11.92,
+      "learning_rate": 3.0260422729869587e-05,
+      "loss": 2.2627,
+      "step": 88230
+    },
+    {
+      "epoch": 11.92,
+      "learning_rate": 3.023487006191494e-05,
+      "loss": 2.2672,
+      "step": 88240
+    },
+    {
+      "epoch": 11.92,
+      "learning_rate": 3.0209326977944247e-05,
+      "loss": 2.272,
+      "step": 88250
+    },
+    {
+      "epoch": 11.92,
+      "learning_rate": 3.0183793480001528e-05,
+      "loss": 2.2612,
+      "step": 88260
+    },
+    {
+      "epoch": 11.92,
+      "learning_rate": 3.0158269570130112e-05,
+      "loss": 2.2085,
+      "step": 88270
+    },
+    {
+      "epoch": 11.92,
+      "learning_rate": 3.013275525037248e-05,
+      "loss": 2.2667,
+      "step": 88280
+    },
+    {
+      "epoch": 11.92,
+      "learning_rate": 3.010725052277032e-05,
+      "loss": 2.2384,
+      "step": 88290
+    },
+    {
+      "epoch": 11.93,
+      "learning_rate": 3.0081755389364682e-05,
+      "loss": 2.2616,
+      "step": 88300
+    },
+    {
+      "epoch": 11.93,
+      "learning_rate": 3.005626985219568e-05,
+      "loss": 2.2669,
+      "step": 88310
+    },
+    {
+      "epoch": 11.93,
+      "learning_rate": 3.0030793913302838e-05,
+      "loss": 2.2448,
+      "step": 88320
+    },
+    {
+      "epoch": 11.93,
+      "learning_rate": 3.0005327574724787e-05,
+      "loss": 2.2678,
+      "step": 88330
+    },
+    {
+      "epoch": 11.93,
+      "learning_rate": 2.9979870838499363e-05,
+      "loss": 2.2565,
+      "step": 88340
+    },
+    {
+      "epoch": 11.93,
+      "learning_rate": 2.9954423706663794e-05,
+      "loss": 2.256,
+      "step": 88350
+    },
+    {
+      "epoch": 11.93,
+      "learning_rate": 2.9928986181254394e-05,
+      "loss": 2.2335,
+      "step": 88360
+    },
+    {
+      "epoch": 11.94,
+      "learning_rate": 2.9903558264306705e-05,
+      "loss": 2.2402,
+      "step": 88370
+    },
+    {
+      "epoch": 11.94,
+      "learning_rate": 2.9878139957855646e-05,
+      "loss": 2.2611,
+      "step": 88380
+    },
+    {
+      "epoch": 11.94,
+      "learning_rate": 2.9852731263935176e-05,
+      "loss": 2.2709,
+      "step": 88390
+    },
+    {
+      "epoch": 11.94,
+      "learning_rate": 2.98273321845787e-05,
+      "loss": 2.2896,
+      "step": 88400
+    },
+    {
+      "epoch": 11.94,
+      "learning_rate": 2.980194272181858e-05,
+      "loss": 2.2541,
+      "step": 88410
+    },
+    {
+      "epoch": 11.94,
+      "learning_rate": 2.9776562877686634e-05,
+      "loss": 2.2442,
+      "step": 88420
+    },
+    {
+      "epoch": 11.94,
+      "learning_rate": 2.975119265421389e-05,
+      "loss": 2.2512,
+      "step": 88430
+    },
+    {
+      "epoch": 11.94,
+      "learning_rate": 2.9725832053430493e-05,
+      "loss": 2.2721,
+      "step": 88440
+    },
+    {
+      "epoch": 11.95,
+      "learning_rate": 2.97004810773659e-05,
+      "loss": 2.2637,
+      "step": 88450
+    },
+    {
+      "epoch": 11.95,
+      "learning_rate": 2.9675139728048708e-05,
+      "loss": 2.2517,
+      "step": 88460
+    },
+    {
+      "epoch": 11.95,
+      "learning_rate": 2.9649808007506925e-05,
+      "loss": 2.2501,
+      "step": 88470
+    },
+    {
+      "epoch": 11.95,
+      "learning_rate": 2.9624485917767615e-05,
+      "loss": 2.2345,
+      "step": 88480
+    },
+    {
+      "epoch": 11.95,
+      "learning_rate": 2.959917346085709e-05,
+      "loss": 2.2742,
+      "step": 88490
+    },
+    {
+      "epoch": 11.95,
+      "learning_rate": 2.9573870638801027e-05,
+      "loss": 2.2481,
+      "step": 88500
+    },
+    {
+      "epoch": 11.95,
+      "learning_rate": 2.9548577453624145e-05,
+      "loss": 2.2561,
+      "step": 88510
+    },
+    {
+      "epoch": 11.96,
+      "learning_rate": 2.952329390735057e-05,
+      "loss": 2.2489,
+      "step": 88520
+    },
+    {
+      "epoch": 11.96,
+      "learning_rate": 2.9498020002003537e-05,
+      "loss": 2.2667,
+      "step": 88530
+    },
+    {
+      "epoch": 11.96,
+      "learning_rate": 2.947275573960549e-05,
+      "loss": 2.2675,
+      "step": 88540
+    },
+    {
+      "epoch": 11.96,
+      "learning_rate": 2.9447501122178253e-05,
+      "loss": 2.2633,
+      "step": 88550
+    },
+    {
+      "epoch": 11.96,
+      "learning_rate": 2.9422256151742718e-05,
+      "loss": 2.2637,
+      "step": 88560
+    },
+    {
+      "epoch": 11.96,
+      "learning_rate": 2.939702083031904e-05,
+      "loss": 2.2385,
+      "step": 88570
+    },
+    {
+      "epoch": 11.96,
+      "learning_rate": 2.9371795159926704e-05,
+      "loss": 2.2439,
+      "step": 88580
+    },
+    {
+      "epoch": 11.96,
+      "learning_rate": 2.9346579142584275e-05,
+      "loss": 2.2778,
+      "step": 88590
+    },
+    {
+      "epoch": 11.97,
+      "learning_rate": 2.9321372780309743e-05,
+      "loss": 2.2562,
+      "step": 88600
+    },
+    {
+      "epoch": 11.97,
+      "learning_rate": 2.9296176075120025e-05,
+      "loss": 2.2638,
+      "step": 88610
+    },
+    {
+      "epoch": 11.97,
+      "learning_rate": 2.9270989029031527e-05,
+      "loss": 2.2845,
+      "step": 88620
+    },
+    {
+      "epoch": 11.97,
+      "learning_rate": 2.9245811644059835e-05,
+      "loss": 2.2529,
+      "step": 88630
+    },
+    {
+      "epoch": 11.97,
+      "learning_rate": 2.9220643922219688e-05,
+      "loss": 2.2809,
+      "step": 88640
+    },
+    {
+      "epoch": 11.97,
+      "learning_rate": 2.9195485865525094e-05,
+      "loss": 2.2461,
+      "step": 88650
+    },
+    {
+      "epoch": 11.97,
+      "learning_rate": 2.9170337475989218e-05,
+      "loss": 2.2421,
+      "step": 88660
+    },
+    {
+      "epoch": 11.98,
+      "learning_rate": 2.914519875562461e-05,
+      "loss": 2.281,
+      "step": 88670
+    },
+    {
+      "epoch": 11.98,
+      "learning_rate": 2.9120069706442904e-05,
+      "loss": 2.2486,
+      "step": 88680
+    },
+    {
+      "epoch": 11.98,
+      "learning_rate": 2.909495033045498e-05,
+      "loss": 2.2641,
+      "step": 88690
+    },
+    {
+      "epoch": 11.98,
+      "learning_rate": 2.9069840629671025e-05,
+      "loss": 2.2388,
+      "step": 88700
+    },
+    {
+      "epoch": 11.98,
+      "learning_rate": 2.904474060610032e-05,
+      "loss": 2.2881,
+      "step": 88710
+    },
+    {
+      "epoch": 11.98,
+      "learning_rate": 2.9019650261751554e-05,
+      "loss": 2.2415,
+      "step": 88720
+    },
+    {
+      "epoch": 11.98,
+      "learning_rate": 2.8994569598632472e-05,
+      "loss": 2.2381,
+      "step": 88730
+    },
+    {
+      "epoch": 11.99,
+      "learning_rate": 2.896949861875007e-05,
+      "loss": 2.2533,
+      "step": 88740
+    },
+    {
+      "epoch": 11.99,
+      "learning_rate": 2.8944437324110708e-05,
+      "loss": 2.2655,
+      "step": 88750
+    },
+    {
+      "epoch": 11.99,
+      "learning_rate": 2.89193857167198e-05,
+      "loss": 2.2376,
+      "step": 88760
+    },
+    {
+      "epoch": 11.99,
+      "learning_rate": 2.8894343798582025e-05,
+      "loss": 2.2569,
+      "step": 88770
+    },
+    {
+      "epoch": 11.99,
+      "learning_rate": 2.8869311571701413e-05,
+      "loss": 2.2375,
+      "step": 88780
+    },
+    {
+      "epoch": 11.99,
+      "learning_rate": 2.8844289038081013e-05,
+      "loss": 2.254,
+      "step": 88790
+    },
+    {
+      "epoch": 11.99,
+      "learning_rate": 2.8819276199723357e-05,
+      "loss": 2.2432,
+      "step": 88800
+    },
+    {
+      "epoch": 11.99,
+      "learning_rate": 2.8794273058629858e-05,
+      "loss": 2.2417,
+      "step": 88810
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 2.8769279616801466e-05,
+      "loss": 2.2788,
+      "step": 88820
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 2.874429587623825e-05,
+      "loss": 2.2439,
+      "step": 88830
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 2.871932183893944e-05,
+      "loss": 2.2454,
+      "step": 88840
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 2.869435750690356e-05,
+      "loss": 2.2812,
+      "step": 88850
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 2.8669402882128273e-05,
+      "loss": 2.2306,
+      "step": 88860
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 2.8644457966610634e-05,
+      "loss": 2.2076,
+      "step": 88870
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 2.861952276234675e-05,
+      "loss": 2.2041,
+      "step": 88880
+    },
+    {
+      "epoch": 12.01,
+      "learning_rate": 2.8594597271331982e-05,
+      "loss": 2.2114,
+      "step": 88890
+    },
+    {
+      "epoch": 12.01,
+      "learning_rate": 2.8569681495561027e-05,
+      "loss": 2.2496,
+      "step": 88900
+    },
+    {
+      "epoch": 12.01,
+      "learning_rate": 2.8544775437027652e-05,
+      "loss": 2.2097,
+      "step": 88910
+    },
+    {
+      "epoch": 12.01,
+      "learning_rate": 2.851987909772498e-05,
+      "loss": 2.2543,
+      "step": 88920
+    },
+    {
+      "epoch": 12.01,
+      "learning_rate": 2.8494992479645267e-05,
+      "loss": 2.2301,
+      "step": 88930
+    },
+    {
+      "epoch": 12.01,
+      "learning_rate": 2.847011558477999e-05,
+      "loss": 2.2203,
+      "step": 88940
+    },
+    {
+      "epoch": 12.01,
+      "learning_rate": 2.8445248415119936e-05,
+      "loss": 2.2437,
+      "step": 88950
+    },
+    {
+      "epoch": 12.01,
+      "learning_rate": 2.8420390972655017e-05,
+      "loss": 2.2308,
+      "step": 88960
+    },
+    {
+      "epoch": 12.02,
+      "learning_rate": 2.8395543259374387e-05,
+      "loss": 2.2088,
+      "step": 88970
+    },
+    {
+      "epoch": 12.02,
+      "learning_rate": 2.8370705277266477e-05,
+      "loss": 2.2255,
+      "step": 88980
+    },
+    {
+      "epoch": 12.02,
+      "learning_rate": 2.834587702831886e-05,
+      "loss": 2.2249,
+      "step": 88990
+    },
+    {
+      "epoch": 12.02,
+      "learning_rate": 2.832105851451848e-05,
+      "loss": 2.2249,
+      "step": 89000
+    },
+    {
+      "epoch": 12.02,
+      "eval_loss": 2.4120471477508545,
+      "eval_runtime": 1269.0105,
+      "eval_samples_per_second": 59.996,
+      "eval_steps_per_second": 5.0,
+      "step": 89000
+    },
+    {
+      "epoch": 12.02,
+      "learning_rate": 2.8296249737851216e-05,
+      "loss": 2.1843,
+      "step": 89010
+    },
+    {
+      "epoch": 12.02,
+      "learning_rate": 2.8271450700302485e-05,
+      "loss": 2.1899,
+      "step": 89020
+    },
+    {
+      "epoch": 12.02,
+      "learning_rate": 2.824666140385669e-05,
+      "loss": 2.1925,
+      "step": 89030
+    },
+    {
+      "epoch": 12.03,
+      "learning_rate": 2.8221881850497634e-05,
+      "loss": 2.1798,
+      "step": 89040
+    },
+    {
+      "epoch": 12.03,
+      "learning_rate": 2.8197112042208232e-05,
+      "loss": 2.2077,
+      "step": 89050
+    },
+    {
+      "epoch": 12.03,
+      "learning_rate": 2.8172351980970575e-05,
+      "loss": 2.236,
+      "step": 89060
+    },
+    {
+      "epoch": 12.03,
+      "learning_rate": 2.8147601668766133e-05,
+      "loss": 2.1932,
+      "step": 89070
+    },
+    {
+      "epoch": 12.03,
+      "learning_rate": 2.8122861107575473e-05,
+      "loss": 2.2448,
+      "step": 89080
+    },
+    {
+      "epoch": 12.03,
+      "learning_rate": 2.8098130299378357e-05,
+      "loss": 2.1892,
+      "step": 89090
+    },
+    {
+      "epoch": 12.03,
+      "learning_rate": 2.8075880912446173e-05,
+      "loss": 2.2049,
+      "step": 89100
+    },
+    {
+      "epoch": 12.04,
+      "learning_rate": 2.8051168640388515e-05,
+      "loss": 2.2447,
+      "step": 89110
+    },
+    {
+      "epoch": 12.04,
+      "learning_rate": 2.8026466127061515e-05,
+      "loss": 2.2093,
+      "step": 89120
+    },
+    {
+      "epoch": 12.04,
+      "learning_rate": 2.8001773374441905e-05,
+      "loss": 2.2221,
+      "step": 89130
+    },
+    {
+      "epoch": 12.04,
+      "learning_rate": 2.797709038450574e-05,
+      "loss": 2.246,
+      "step": 89140
+    },
+    {
+      "epoch": 12.04,
+      "learning_rate": 2.795241715922819e-05,
+      "loss": 2.2311,
+      "step": 89150
+    },
+    {
+      "epoch": 12.04,
+      "learning_rate": 2.792775370058371e-05,
+      "loss": 2.2173,
+      "step": 89160
+    },
+    {
+      "epoch": 12.04,
+      "learning_rate": 2.7903100010545886e-05,
+      "loss": 2.2397,
+      "step": 89170
+    },
+    {
+      "epoch": 12.04,
+      "learning_rate": 2.7878456091087626e-05,
+      "loss": 2.2228,
+      "step": 89180
+    },
+    {
+      "epoch": 12.05,
+      "learning_rate": 2.7853821944181116e-05,
+      "loss": 2.2414,
+      "step": 89190
+    },
+    {
+      "epoch": 12.05,
+      "learning_rate": 2.782919757179746e-05,
+      "loss": 2.2332,
+      "step": 89200
+    },
+    {
+      "epoch": 12.05,
+      "learning_rate": 2.7804582975907325e-05,
+      "loss": 2.2225,
+      "step": 89210
+    },
+    {
+      "epoch": 12.05,
+      "learning_rate": 2.7779978158480378e-05,
+      "loss": 2.2515,
+      "step": 89220
+    },
+    {
+      "epoch": 12.05,
+      "learning_rate": 2.7755383121485608e-05,
+      "loss": 2.2106,
+      "step": 89230
+    },
+    {
+      "epoch": 12.05,
+      "learning_rate": 2.7730797866891198e-05,
+      "loss": 2.2259,
+      "step": 89240
+    },
+    {
+      "epoch": 12.05,
+      "learning_rate": 2.7706222396664485e-05,
+      "loss": 2.2184,
+      "step": 89250
+    },
+    {
+      "epoch": 12.06,
+      "learning_rate": 2.7681656712772132e-05,
+      "loss": 2.2025,
+      "step": 89260
+    },
+    {
+      "epoch": 12.06,
+      "learning_rate": 2.765710081717995e-05,
+      "loss": 2.2369,
+      "step": 89270
+    },
+    {
+      "epoch": 12.06,
+      "learning_rate": 2.7632554711852917e-05,
+      "loss": 2.2263,
+      "step": 89280
+    },
+    {
+      "epoch": 12.06,
+      "learning_rate": 2.7608018398755393e-05,
+      "loss": 2.2289,
+      "step": 89290
+    },
+    {
+      "epoch": 12.06,
+      "learning_rate": 2.758349187985076e-05,
+      "loss": 2.222,
+      "step": 89300
+    },
+    {
+      "epoch": 12.06,
+      "learning_rate": 2.7558975157101782e-05,
+      "loss": 2.2158,
+      "step": 89310
+    },
+    {
+      "epoch": 12.06,
+      "learning_rate": 2.7534468232470335e-05,
+      "loss": 2.1979,
+      "step": 89320
+    },
+    {
+      "epoch": 12.06,
+      "learning_rate": 2.750997110791749e-05,
+      "loss": 2.2182,
+      "step": 89330
+    },
+    {
+      "epoch": 12.07,
+      "learning_rate": 2.7485483785403674e-05,
+      "loss": 2.2072,
+      "step": 89340
+    },
+    {
+      "epoch": 12.07,
+      "learning_rate": 2.7461006266888386e-05,
+      "loss": 2.2172,
+      "step": 89350
+    },
+    {
+      "epoch": 12.07,
+      "learning_rate": 2.743653855433041e-05,
+      "loss": 2.2009,
+      "step": 89360
+    },
+    {
+      "epoch": 12.07,
+      "learning_rate": 2.7412080649687684e-05,
+      "loss": 2.2324,
+      "step": 89370
+    },
+    {
+      "epoch": 12.07,
+      "learning_rate": 2.7387632554917453e-05,
+      "loss": 2.2004,
+      "step": 89380
+    },
+    {
+      "epoch": 12.07,
+      "learning_rate": 2.736319427197619e-05,
+      "loss": 2.2142,
+      "step": 89390
+    },
+    {
+      "epoch": 12.07,
+      "learning_rate": 2.7338765802819397e-05,
+      "loss": 2.2462,
+      "step": 89400
+    },
+    {
+      "epoch": 12.08,
+      "learning_rate": 2.7314347149402005e-05,
+      "loss": 2.2158,
+      "step": 89410
+    },
+    {
+      "epoch": 12.08,
+      "learning_rate": 2.7289938313678006e-05,
+      "loss": 2.2372,
+      "step": 89420
+    },
+    {
+      "epoch": 12.08,
+      "learning_rate": 2.7265539297600752e-05,
+      "loss": 2.2128,
+      "step": 89430
+    },
+    {
+      "epoch": 12.08,
+      "learning_rate": 2.7241150103122683e-05,
+      "loss": 2.2097,
+      "step": 89440
+    },
+    {
+      "epoch": 12.08,
+      "learning_rate": 2.721677073219547e-05,
+      "loss": 2.2011,
+      "step": 89450
+    },
+    {
+      "epoch": 12.08,
+      "learning_rate": 2.71924011867701e-05,
+      "loss": 2.2142,
+      "step": 89460
+    },
+    {
+      "epoch": 12.08,
+      "learning_rate": 2.716804146879667e-05,
+      "loss": 2.2278,
+      "step": 89470
+    },
+    {
+      "epoch": 12.08,
+      "learning_rate": 2.7143691580224465e-05,
+      "loss": 2.2481,
+      "step": 89480
+    },
+    {
+      "epoch": 12.09,
+      "learning_rate": 2.7119351523002126e-05,
+      "loss": 2.2027,
+      "step": 89490
+    },
+    {
+      "epoch": 12.09,
+      "learning_rate": 2.7095021299077364e-05,
+      "loss": 2.2063,
+      "step": 89500
+    },
+    {
+      "epoch": 12.09,
+      "learning_rate": 2.7070700910397203e-05,
+      "loss": 2.2193,
+      "step": 89510
+    },
+    {
+      "epoch": 12.09,
+      "learning_rate": 2.704639035890782e-05,
+      "loss": 2.2277,
+      "step": 89520
+    },
+    {
+      "epoch": 12.09,
+      "learning_rate": 2.702208964655458e-05,
+      "loss": 2.2209,
+      "step": 89530
+    },
+    {
+      "epoch": 12.09,
+      "learning_rate": 2.699779877528217e-05,
+      "loss": 2.2195,
+      "step": 89540
+    },
+    {
+      "epoch": 12.09,
+      "learning_rate": 2.6973517747034408e-05,
+      "loss": 2.2129,
+      "step": 89550
+    },
+    {
+      "epoch": 12.1,
+      "learning_rate": 2.69492465637543e-05,
+      "loss": 2.1983,
+      "step": 89560
+    },
+    {
+      "epoch": 12.1,
+      "learning_rate": 2.6924985227384117e-05,
+      "loss": 2.2166,
+      "step": 89570
+    },
+    {
+      "epoch": 12.1,
+      "learning_rate": 2.6900733739865314e-05,
+      "loss": 2.2486,
+      "step": 89580
+    },
+    {
+      "epoch": 12.1,
+      "learning_rate": 2.687649210313868e-05,
+      "loss": 2.2188,
+      "step": 89590
+    },
+    {
+      "epoch": 12.1,
+      "learning_rate": 2.6852260319143943e-05,
+      "loss": 2.2352,
+      "step": 89600
+    },
+    {
+      "epoch": 12.1,
+      "learning_rate": 2.6828038389820316e-05,
+      "loss": 2.1857,
+      "step": 89610
+    },
+    {
+      "epoch": 12.1,
+      "learning_rate": 2.6803826317106052e-05,
+      "loss": 2.221,
+      "step": 89620
+    },
+    {
+      "epoch": 12.11,
+      "learning_rate": 2.677962410293875e-05,
+      "loss": 2.2093,
+      "step": 89630
+    },
+    {
+      "epoch": 12.11,
+      "learning_rate": 2.6755431749255112e-05,
+      "loss": 2.1951,
+      "step": 89640
+    },
+    {
+      "epoch": 12.11,
+      "learning_rate": 2.673124925799104e-05,
+      "loss": 2.241,
+      "step": 89650
+    },
+    {
+      "epoch": 12.11,
+      "learning_rate": 2.6707076631081763e-05,
+      "loss": 2.2325,
+      "step": 89660
+    },
+    {
+      "epoch": 12.11,
+      "learning_rate": 2.6682913870461625e-05,
+      "loss": 2.2569,
+      "step": 89670
+    },
+    {
+      "epoch": 12.11,
+      "learning_rate": 2.6658760978064174e-05,
+      "loss": 2.2314,
+      "step": 89680
+    },
+    {
+      "epoch": 12.11,
+      "learning_rate": 2.663461795582225e-05,
+      "loss": 2.2247,
+      "step": 89690
+    },
+    {
+      "epoch": 12.11,
+      "learning_rate": 2.6610484805667824e-05,
+      "loss": 2.2419,
+      "step": 89700
+    },
+    {
+      "epoch": 12.12,
+      "learning_rate": 2.6586361529532134e-05,
+      "loss": 2.238,
+      "step": 89710
+    },
+    {
+      "epoch": 12.12,
+      "learning_rate": 2.6562248129345586e-05,
+      "loss": 2.2297,
+      "step": 89720
+    },
+    {
+      "epoch": 12.12,
+      "learning_rate": 2.6538144607037786e-05,
+      "loss": 2.2392,
+      "step": 89730
+    },
+    {
+      "epoch": 12.12,
+      "learning_rate": 2.6514050964537624e-05,
+      "loss": 2.2517,
+      "step": 89740
+    },
+    {
+      "epoch": 12.12,
+      "learning_rate": 2.6489967203773127e-05,
+      "loss": 2.2141,
+      "step": 89750
+    },
+    {
+      "epoch": 12.12,
+      "learning_rate": 2.646589332667155e-05,
+      "loss": 2.2244,
+      "step": 89760
+    },
+    {
+      "epoch": 12.12,
+      "learning_rate": 2.6441829335159303e-05,
+      "loss": 2.2319,
+      "step": 89770
+    },
+    {
+      "epoch": 12.13,
+      "learning_rate": 2.6417775231162147e-05,
+      "loss": 2.201,
+      "step": 89780
+    },
+    {
+      "epoch": 12.13,
+      "learning_rate": 2.6393731016604993e-05,
+      "loss": 2.2452,
+      "step": 89790
+    },
+    {
+      "epoch": 12.13,
+      "learning_rate": 2.636969669341181e-05,
+      "loss": 2.2169,
+      "step": 89800
+    },
+    {
+      "epoch": 12.13,
+      "learning_rate": 2.6345672263506025e-05,
+      "loss": 2.2075,
+      "step": 89810
+    },
+    {
+      "epoch": 12.13,
+      "learning_rate": 2.6321657728810036e-05,
+      "loss": 2.2206,
+      "step": 89820
+    },
+    {
+      "epoch": 12.13,
+      "learning_rate": 2.629765309124568e-05,
+      "loss": 2.2091,
+      "step": 89830
+    },
+    {
+      "epoch": 12.13,
+      "learning_rate": 2.62736583527338e-05,
+      "loss": 2.2429,
+      "step": 89840
+    },
+    {
+      "epoch": 12.13,
+      "learning_rate": 2.6249673515194546e-05,
+      "loss": 2.2293,
+      "step": 89850
+    },
+    {
+      "epoch": 12.14,
+      "learning_rate": 2.622569858054728e-05,
+      "loss": 2.233,
+      "step": 89860
+    },
+    {
+      "epoch": 12.14,
+      "learning_rate": 2.6201733550710574e-05,
+      "loss": 2.1981,
+      "step": 89870
+    },
+    {
+      "epoch": 12.14,
+      "learning_rate": 2.6177778427602085e-05,
+      "loss": 2.2054,
+      "step": 89880
+    },
+    {
+      "epoch": 12.14,
+      "learning_rate": 2.6153833213138904e-05,
+      "loss": 2.2526,
+      "step": 89890
+    },
+    {
+      "epoch": 12.14,
+      "learning_rate": 2.6129897909237106e-05,
+      "loss": 2.2377,
+      "step": 89900
+    },
+    {
+      "epoch": 12.14,
+      "learning_rate": 2.6105972517812136e-05,
+      "loss": 2.1945,
+      "step": 89910
+    },
+    {
+      "epoch": 12.14,
+      "learning_rate": 2.608205704077857e-05,
+      "loss": 2.2187,
+      "step": 89920
+    },
+    {
+      "epoch": 12.15,
+      "learning_rate": 2.6058151480050132e-05,
+      "loss": 2.2255,
+      "step": 89930
+    },
+    {
+      "epoch": 12.15,
+      "learning_rate": 2.603425583753991e-05,
+      "loss": 2.2325,
+      "step": 89940
+    },
+    {
+      "epoch": 12.15,
+      "learning_rate": 2.6010370115160067e-05,
+      "loss": 2.2379,
+      "step": 89950
+    },
+    {
+      "epoch": 12.15,
+      "learning_rate": 2.5986494314822014e-05,
+      "loss": 2.2331,
+      "step": 89960
+    },
+    {
+      "epoch": 12.15,
+      "learning_rate": 2.596262843843634e-05,
+      "loss": 2.2421,
+      "step": 89970
+    },
+    {
+      "epoch": 12.15,
+      "learning_rate": 2.5938772487912895e-05,
+      "loss": 2.2329,
+      "step": 89980
+    },
+    {
+      "epoch": 12.15,
+      "learning_rate": 2.591492646516079e-05,
+      "loss": 2.2268,
+      "step": 89990
+    },
+    {
+      "epoch": 12.16,
+      "learning_rate": 2.5891090372088112e-05,
+      "loss": 2.2236,
+      "step": 90000
+    },
+    {
+      "epoch": 12.16,
+      "eval_loss": 2.4117822647094727,
+      "eval_runtime": 1269.0092,
+      "eval_samples_per_second": 59.996,
+      "eval_steps_per_second": 5.0,
+      "step": 90000
+    },
+    {
+      "epoch": 12.16,
+      "learning_rate": 2.586726421060239e-05,
+      "loss": 2.2347,
+      "step": 90010
+    },
+    {
+      "epoch": 12.16,
+      "learning_rate": 2.584344798261021e-05,
+      "loss": 2.191,
+      "step": 90020
+    },
+    {
+      "epoch": 12.16,
+      "learning_rate": 2.58196416900175e-05,
+      "loss": 2.2093,
+      "step": 90030
+    },
+    {
+      "epoch": 12.16,
+      "learning_rate": 2.579584533472929e-05,
+      "loss": 2.2308,
+      "step": 90040
+    },
+    {
+      "epoch": 12.16,
+      "learning_rate": 2.5772058918649764e-05,
+      "loss": 2.236,
+      "step": 90050
+    },
+    {
+      "epoch": 12.16,
+      "learning_rate": 2.5748282443682488e-05,
+      "loss": 2.2309,
+      "step": 90060
+    },
+    {
+      "epoch": 12.16,
+      "learning_rate": 2.572451591173009e-05,
+      "loss": 2.2295,
+      "step": 90070
+    },
+    {
+      "epoch": 12.17,
+      "learning_rate": 2.5700759324694404e-05,
+      "loss": 2.2123,
+      "step": 90080
+    },
+    {
+      "epoch": 12.17,
+      "learning_rate": 2.5677012684476582e-05,
+      "loss": 2.2227,
+      "step": 90090
+    },
+    {
+      "epoch": 12.17,
+      "learning_rate": 2.5653275992976828e-05,
+      "loss": 2.2371,
+      "step": 90100
+    },
+    {
+      "epoch": 12.17,
+      "learning_rate": 2.562954925209474e-05,
+      "loss": 2.2377,
+      "step": 90110
+    },
+    {
+      "epoch": 12.17,
+      "learning_rate": 2.5605832463728843e-05,
+      "loss": 2.2337,
+      "step": 90120
+    },
+    {
+      "epoch": 12.17,
+      "learning_rate": 2.558212562977714e-05,
+      "loss": 2.1929,
+      "step": 90130
+    },
+    {
+      "epoch": 12.17,
+      "learning_rate": 2.5558428752136745e-05,
+      "loss": 2.223,
+      "step": 90140
+    },
+    {
+      "epoch": 12.18,
+      "learning_rate": 2.5534741832703908e-05,
+      "loss": 2.2286,
+      "step": 90150
+    },
+    {
+      "epoch": 12.18,
+      "learning_rate": 2.5511064873374132e-05,
+      "loss": 2.2231,
+      "step": 90160
+    },
+    {
+      "epoch": 12.18,
+      "learning_rate": 2.5487397876042116e-05,
+      "loss": 2.2101,
+      "step": 90170
+    },
+    {
+      "epoch": 12.18,
+      "learning_rate": 2.546374084260176e-05,
+      "loss": 2.212,
+      "step": 90180
+    },
+    {
+      "epoch": 12.18,
+      "learning_rate": 2.5440093774946284e-05,
+      "loss": 2.2127,
+      "step": 90190
+    },
+    {
+      "epoch": 12.18,
+      "learning_rate": 2.5416456674967846e-05,
+      "loss": 2.2208,
+      "step": 90200
+    },
+    {
+      "epoch": 12.18,
+      "learning_rate": 2.539282954455806e-05,
+      "loss": 2.2211,
+      "step": 90210
+    },
+    {
+      "epoch": 12.18,
+      "learning_rate": 2.5369212385607562e-05,
+      "loss": 2.2418,
+      "step": 90220
+    },
+    {
+      "epoch": 12.19,
+      "learning_rate": 2.534560520000638e-05,
+      "loss": 2.2295,
+      "step": 90230
+    },
+    {
+      "epoch": 12.19,
+      "learning_rate": 2.5322007989643566e-05,
+      "loss": 2.2137,
+      "step": 90240
+    },
+    {
+      "epoch": 12.19,
+      "learning_rate": 2.5298420756407428e-05,
+      "loss": 2.2211,
+      "step": 90250
+    },
+    {
+      "epoch": 12.19,
+      "learning_rate": 2.5274843502185555e-05,
+      "loss": 2.2179,
+      "step": 90260
+    },
+    {
+      "epoch": 12.19,
+      "learning_rate": 2.5251276228864637e-05,
+      "loss": 2.2197,
+      "step": 90270
+    },
+    {
+      "epoch": 12.19,
+      "learning_rate": 2.5227718938330565e-05,
+      "loss": 2.2251,
+      "step": 90280
+    },
+    {
+      "epoch": 12.19,
+      "learning_rate": 2.5204171632468533e-05,
+      "loss": 2.2385,
+      "step": 90290
+    },
+    {
+      "epoch": 12.2,
+      "learning_rate": 2.5180634313162816e-05,
+      "loss": 2.2138,
+      "step": 90300
+    },
+    {
+      "epoch": 12.2,
+      "learning_rate": 2.515710698229704e-05,
+      "loss": 2.2252,
+      "step": 90310
+    },
+    {
+      "epoch": 12.2,
+      "learning_rate": 2.5133589641753787e-05,
+      "loss": 2.2274,
+      "step": 90320
+    },
+    {
+      "epoch": 12.2,
+      "learning_rate": 2.511008229341508e-05,
+      "loss": 2.2187,
+      "step": 90330
+    },
+    {
+      "epoch": 12.2,
+      "learning_rate": 2.508658493916208e-05,
+      "loss": 2.2317,
+      "step": 90340
+    },
+    {
+      "epoch": 12.2,
+      "learning_rate": 2.506309758087507e-05,
+      "loss": 2.1969,
+      "step": 90350
+    },
+    {
+      "epoch": 12.2,
+      "learning_rate": 2.5039620220433594e-05,
+      "loss": 2.1977,
+      "step": 90360
+    },
+    {
+      "epoch": 12.21,
+      "learning_rate": 2.501615285971635e-05,
+      "loss": 2.2503,
+      "step": 90370
+    },
+    {
+      "epoch": 12.21,
+      "learning_rate": 2.4992695500601302e-05,
+      "loss": 2.2278,
+      "step": 90380
+    },
+    {
+      "epoch": 12.21,
+      "learning_rate": 2.4969248144965646e-05,
+      "loss": 2.2408,
+      "step": 90390
+    },
+    {
+      "epoch": 12.21,
+      "learning_rate": 2.4945810794685588e-05,
+      "loss": 2.2304,
+      "step": 90400
+    },
+    {
+      "epoch": 12.21,
+      "learning_rate": 2.492238345163676e-05,
+      "loss": 2.1981,
+      "step": 90410
+    },
+    {
+      "epoch": 12.21,
+      "learning_rate": 2.489896611769382e-05,
+      "loss": 2.1961,
+      "step": 90420
+    },
+    {
+      "epoch": 12.21,
+      "learning_rate": 2.487555879473076e-05,
+      "loss": 2.2441,
+      "step": 90430
+    },
+    {
+      "epoch": 12.21,
+      "learning_rate": 2.485216148462068e-05,
+      "loss": 2.2077,
+      "step": 90440
+    },
+    {
+      "epoch": 12.22,
+      "learning_rate": 2.482877418923587e-05,
+      "loss": 2.2143,
+      "step": 90450
+    },
+    {
+      "epoch": 12.22,
+      "learning_rate": 2.4805396910447923e-05,
+      "loss": 2.262,
+      "step": 90460
+    },
+    {
+      "epoch": 12.22,
+      "learning_rate": 2.478202965012753e-05,
+      "loss": 2.264,
+      "step": 90470
+    },
+    {
+      "epoch": 12.22,
+      "learning_rate": 2.4758672410144585e-05,
+      "loss": 2.2188,
+      "step": 90480
+    },
+    {
+      "epoch": 12.22,
+      "learning_rate": 2.4735325192368272e-05,
+      "loss": 2.2468,
+      "step": 90490
+    },
+    {
+      "epoch": 12.22,
+      "learning_rate": 2.4711987998666828e-05,
+      "loss": 2.2417,
+      "step": 90500
+    },
+    {
+      "epoch": 12.22,
+      "learning_rate": 2.4688660830907893e-05,
+      "loss": 2.2169,
+      "step": 90510
+    },
+    {
+      "epoch": 12.23,
+      "learning_rate": 2.4665343690958023e-05,
+      "loss": 2.246,
+      "step": 90520
+    },
+    {
+      "epoch": 12.23,
+      "learning_rate": 2.464203658068321e-05,
+      "loss": 2.2194,
+      "step": 90530
+    },
+    {
+      "epoch": 12.23,
+      "learning_rate": 2.4618739501948613e-05,
+      "loss": 2.2148,
+      "step": 90540
+    },
+    {
+      "epoch": 12.23,
+      "learning_rate": 2.4595452456618474e-05,
+      "loss": 2.2389,
+      "step": 90550
+    },
+    {
+      "epoch": 12.23,
+      "learning_rate": 2.45721754465563e-05,
+      "loss": 2.2065,
+      "step": 90560
+    },
+    {
+      "epoch": 12.23,
+      "learning_rate": 2.4548908473624774e-05,
+      "loss": 2.2102,
+      "step": 90570
+    },
+    {
+      "epoch": 12.23,
+      "learning_rate": 2.4525651539685796e-05,
+      "loss": 2.2317,
+      "step": 90580
+    },
+    {
+      "epoch": 12.23,
+      "learning_rate": 2.450240464660057e-05,
+      "loss": 2.2173,
+      "step": 90590
+    },
+    {
+      "epoch": 12.24,
+      "learning_rate": 2.447916779622922e-05,
+      "loss": 2.2194,
+      "step": 90600
+    },
+    {
+      "epoch": 12.24,
+      "learning_rate": 2.4455940990431343e-05,
+      "loss": 2.2093,
+      "step": 90610
+    },
+    {
+      "epoch": 12.24,
+      "learning_rate": 2.4432724231065553e-05,
+      "loss": 2.2255,
+      "step": 90620
+    },
+    {
+      "epoch": 12.24,
+      "learning_rate": 2.4409517519989796e-05,
+      "loss": 2.2059,
+      "step": 90630
+    },
+    {
+      "epoch": 12.24,
+      "learning_rate": 2.438632085906112e-05,
+      "loss": 2.234,
+      "step": 90640
+    },
+    {
+      "epoch": 12.24,
+      "learning_rate": 2.436313425013574e-05,
+      "loss": 2.2152,
+      "step": 90650
+    },
+    {
+      "epoch": 12.24,
+      "learning_rate": 2.4339957695069205e-05,
+      "loss": 2.2235,
+      "step": 90660
+    },
+    {
+      "epoch": 12.25,
+      "learning_rate": 2.431679119571615e-05,
+      "loss": 2.2403,
+      "step": 90670
+    },
+    {
+      "epoch": 12.25,
+      "learning_rate": 2.429363475393039e-05,
+      "loss": 2.2133,
+      "step": 90680
+    },
+    {
+      "epoch": 12.25,
+      "learning_rate": 2.4270488371565027e-05,
+      "loss": 2.2104,
+      "step": 90690
+    },
+    {
+      "epoch": 12.25,
+      "learning_rate": 2.4247352050472264e-05,
+      "loss": 2.2267,
+      "step": 90700
+    },
+    {
+      "epoch": 12.25,
+      "learning_rate": 2.4224225792503632e-05,
+      "loss": 2.2354,
+      "step": 90710
+    },
+    {
+      "epoch": 12.25,
+      "learning_rate": 2.4201109599509638e-05,
+      "loss": 2.2462,
+      "step": 90720
+    },
+    {
+      "epoch": 12.25,
+      "learning_rate": 2.4178003473340167e-05,
+      "loss": 2.2512,
+      "step": 90730
+    },
+    {
+      "epoch": 12.26,
+      "learning_rate": 2.415490741584431e-05,
+      "loss": 2.2069,
+      "step": 90740
+    },
+    {
+      "epoch": 12.26,
+      "learning_rate": 2.413182142887022e-05,
+      "loss": 2.2433,
+      "step": 90750
+    },
+    {
+      "epoch": 12.26,
+      "learning_rate": 2.4108745514265315e-05,
+      "loss": 2.2401,
+      "step": 90760
+    },
+    {
+      "epoch": 12.26,
+      "learning_rate": 2.4085679673876175e-05,
+      "loss": 2.2531,
+      "step": 90770
+    },
+    {
+      "epoch": 12.26,
+      "learning_rate": 2.4062623909548683e-05,
+      "loss": 2.2216,
+      "step": 90780
+    },
+    {
+      "epoch": 12.26,
+      "learning_rate": 2.403957822312777e-05,
+      "loss": 2.2111,
+      "step": 90790
+    },
+    {
+      "epoch": 12.26,
+      "learning_rate": 2.4016542616457607e-05,
+      "loss": 2.2432,
+      "step": 90800
+    },
+    {
+      "epoch": 12.26,
+      "learning_rate": 2.3993517091381655e-05,
+      "loss": 2.22,
+      "step": 90810
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 2.397050164974242e-05,
+      "loss": 2.254,
+      "step": 90820
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 2.3947496293381713e-05,
+      "loss": 2.2162,
+      "step": 90830
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 2.392450102414048e-05,
+      "loss": 2.2313,
+      "step": 90840
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 2.3901515843858833e-05,
+      "loss": 2.2249,
+      "step": 90850
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 2.3878540754376197e-05,
+      "loss": 2.2189,
+      "step": 90860
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 2.385557575753107e-05,
+      "loss": 2.2189,
+      "step": 90870
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 2.383262085516115e-05,
+      "loss": 2.2333,
+      "step": 90880
+    },
+    {
+      "epoch": 12.28,
+      "learning_rate": 2.380967604910345e-05,
+      "loss": 2.2353,
+      "step": 90890
+    },
+    {
+      "epoch": 12.28,
+      "learning_rate": 2.3786741341193964e-05,
+      "loss": 2.2305,
+      "step": 90900
+    },
+    {
+      "epoch": 12.28,
+      "learning_rate": 2.3763816733268164e-05,
+      "loss": 2.2415,
+      "step": 90910
+    },
+    {
+      "epoch": 12.28,
+      "learning_rate": 2.374090222716038e-05,
+      "loss": 2.231,
+      "step": 90920
+    },
+    {
+      "epoch": 12.28,
+      "learning_rate": 2.3717997824704377e-05,
+      "loss": 2.2359,
+      "step": 90930
+    },
+    {
+      "epoch": 12.28,
+      "learning_rate": 2.3695103527733057e-05,
+      "loss": 2.2225,
+      "step": 90940
+    },
+    {
+      "epoch": 12.28,
+      "learning_rate": 2.3672219338078485e-05,
+      "loss": 2.2064,
+      "step": 90950
+    },
+    {
+      "epoch": 12.28,
+      "learning_rate": 2.3649345257571915e-05,
+      "loss": 2.2369,
+      "step": 90960
+    },
+    {
+      "epoch": 12.29,
+      "learning_rate": 2.362648128804377e-05,
+      "loss": 2.2174,
+      "step": 90970
+    },
+    {
+      "epoch": 12.29,
+      "learning_rate": 2.3603627431323762e-05,
+      "loss": 2.2254,
+      "step": 90980
+    },
+    {
+      "epoch": 12.29,
+      "learning_rate": 2.35807836892407e-05,
+      "loss": 2.2378,
+      "step": 90990
+    },
+    {
+      "epoch": 12.29,
+      "learning_rate": 2.355795006362259e-05,
+      "loss": 2.2325,
+      "step": 91000
+    },
+    {
+      "epoch": 12.29,
+      "eval_loss": 2.4098644256591797,
+      "eval_runtime": 1268.9861,
+      "eval_samples_per_second": 59.997,
+      "eval_steps_per_second": 5.0,
+      "step": 91000
+    },
+    {
+      "epoch": 12.29,
+      "learning_rate": 2.353512655629668e-05,
+      "loss": 2.2031,
+      "step": 91010
+    },
+    {
+      "epoch": 12.29,
+      "learning_rate": 2.3512313169089346e-05,
+      "loss": 2.2083,
+      "step": 91020
+    },
+    {
+      "epoch": 12.29,
+      "learning_rate": 2.348950990382625e-05,
+      "loss": 2.2523,
+      "step": 91030
+    },
+    {
+      "epoch": 12.3,
+      "learning_rate": 2.346671676233214e-05,
+      "loss": 2.2391,
+      "step": 91040
+    },
+    {
+      "epoch": 12.3,
+      "learning_rate": 2.344393374643096e-05,
+      "loss": 2.2306,
+      "step": 91050
+    },
+    {
+      "epoch": 12.3,
+      "learning_rate": 2.3421160857945953e-05,
+      "loss": 2.2282,
+      "step": 91060
+    },
+    {
+      "epoch": 12.3,
+      "learning_rate": 2.3398398098699444e-05,
+      "loss": 2.2384,
+      "step": 91070
+    },
+    {
+      "epoch": 12.3,
+      "learning_rate": 2.3375645470512928e-05,
+      "loss": 2.2188,
+      "step": 91080
+    },
+    {
+      "epoch": 12.3,
+      "learning_rate": 2.3352902975207243e-05,
+      "loss": 2.2409,
+      "step": 91090
+    },
+    {
+      "epoch": 12.3,
+      "learning_rate": 2.333017061460222e-05,
+      "loss": 2.2024,
+      "step": 91100
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 2.330972015673037e-05,
+      "loss": 2.2323,
+      "step": 91110
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 2.3287007057067725e-05,
+      "loss": 2.2336,
+      "step": 91120
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 2.3264304097378937e-05,
+      "loss": 2.2283,
+      "step": 91130
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 2.3241611279480877e-05,
+      "loss": 2.22,
+      "step": 91140
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 2.3218928605189417e-05,
+      "loss": 2.2265,
+      "step": 91150
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 2.319625607631968e-05,
+      "loss": 2.2311,
+      "step": 91160
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 2.3173593694686078e-05,
+      "loss": 2.2058,
+      "step": 91170
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 2.3150941462102058e-05,
+      "loss": 2.204,
+      "step": 91180
+    },
+    {
+      "epoch": 12.32,
+      "learning_rate": 2.3128299380380322e-05,
+      "loss": 2.2183,
+      "step": 91190
+    },
+    {
+      "epoch": 12.32,
+      "learning_rate": 2.310566745133282e-05,
+      "loss": 2.2477,
+      "step": 91200
+    },
+    {
+      "epoch": 12.32,
+      "learning_rate": 2.3083045676770538e-05,
+      "loss": 2.2467,
+      "step": 91210
+    },
+    {
+      "epoch": 12.32,
+      "learning_rate": 2.3060434058503847e-05,
+      "loss": 2.2145,
+      "step": 91220
+    },
+    {
+      "epoch": 12.32,
+      "learning_rate": 2.3037832598342142e-05,
+      "loss": 2.2072,
+      "step": 91230
+    },
+    {
+      "epoch": 12.32,
+      "learning_rate": 2.3015241298094033e-05,
+      "loss": 2.2248,
+      "step": 91240
+    },
+    {
+      "epoch": 12.32,
+      "learning_rate": 2.299266015956744e-05,
+      "loss": 2.237,
+      "step": 91250
+    },
+    {
+      "epoch": 12.33,
+      "learning_rate": 2.2970089184569294e-05,
+      "loss": 2.2178,
+      "step": 91260
+    },
+    {
+      "epoch": 12.33,
+      "learning_rate": 2.2947528374905795e-05,
+      "loss": 2.2073,
+      "step": 91270
+    },
+    {
+      "epoch": 12.33,
+      "learning_rate": 2.2924977732382405e-05,
+      "loss": 2.2328,
+      "step": 91280
+    },
+    {
+      "epoch": 12.33,
+      "learning_rate": 2.2902437258803603e-05,
+      "loss": 2.2266,
+      "step": 91290
+    },
+    {
+      "epoch": 12.33,
+      "learning_rate": 2.2879906955973277e-05,
+      "loss": 2.2358,
+      "step": 91300
+    },
+    {
+      "epoch": 12.33,
+      "learning_rate": 2.285738682569422e-05,
+      "loss": 2.2397,
+      "step": 91310
+    },
+    {
+      "epoch": 12.33,
+      "learning_rate": 2.2834876869768655e-05,
+      "loss": 2.2108,
+      "step": 91320
+    },
+    {
+      "epoch": 12.33,
+      "learning_rate": 2.2812377089997873e-05,
+      "loss": 2.1901,
+      "step": 91330
+    },
+    {
+      "epoch": 12.34,
+      "learning_rate": 2.2789887488182406e-05,
+      "loss": 2.2082,
+      "step": 91340
+    },
+    {
+      "epoch": 12.34,
+      "learning_rate": 2.2767408066121913e-05,
+      "loss": 2.2253,
+      "step": 91350
+    },
+    {
+      "epoch": 12.34,
+      "learning_rate": 2.2744938825615265e-05,
+      "loss": 2.2077,
+      "step": 91360
+    },
+    {
+      "epoch": 12.34,
+      "learning_rate": 2.272247976846055e-05,
+      "loss": 2.2416,
+      "step": 91370
+    },
+    {
+      "epoch": 12.34,
+      "learning_rate": 2.2700030896455007e-05,
+      "loss": 2.2264,
+      "step": 91380
+    },
+    {
+      "epoch": 12.34,
+      "learning_rate": 2.2677592211395035e-05,
+      "loss": 2.2437,
+      "step": 91390
+    },
+    {
+      "epoch": 12.34,
+      "learning_rate": 2.265516371507628e-05,
+      "loss": 2.2331,
+      "step": 91400
+    },
+    {
+      "epoch": 12.35,
+      "learning_rate": 2.2632745409293507e-05,
+      "loss": 2.2451,
+      "step": 91410
+    },
+    {
+      "epoch": 12.35,
+      "learning_rate": 2.2610337295840762e-05,
+      "loss": 2.2276,
+      "step": 91420
+    },
+    {
+      "epoch": 12.35,
+      "learning_rate": 2.2587939376511167e-05,
+      "loss": 2.2378,
+      "step": 91430
+    },
+    {
+      "epoch": 12.35,
+      "learning_rate": 2.2565551653097048e-05,
+      "loss": 2.2306,
+      "step": 91440
+    },
+    {
+      "epoch": 12.35,
+      "learning_rate": 2.2543174127390025e-05,
+      "loss": 2.2094,
+      "step": 91450
+    },
+    {
+      "epoch": 12.35,
+      "learning_rate": 2.2520806801180746e-05,
+      "loss": 2.2243,
+      "step": 91460
+    },
+    {
+      "epoch": 12.35,
+      "learning_rate": 2.2498449676259117e-05,
+      "loss": 2.237,
+      "step": 91470
+    },
+    {
+      "epoch": 12.36,
+      "learning_rate": 2.247610275441427e-05,
+      "loss": 2.2329,
+      "step": 91480
+    },
+    {
+      "epoch": 12.36,
+      "learning_rate": 2.245376603743443e-05,
+      "loss": 2.2165,
+      "step": 91490
+    },
+    {
+      "epoch": 12.36,
+      "learning_rate": 2.2431439527107126e-05,
+      "loss": 2.2316,
+      "step": 91500
+    },
+    {
+      "epoch": 12.36,
+      "learning_rate": 2.2409123225218883e-05,
+      "loss": 2.2224,
+      "step": 91510
+    },
+    {
+      "epoch": 12.36,
+      "learning_rate": 2.2386817133555622e-05,
+      "loss": 2.2212,
+      "step": 91520
+    },
+    {
+      "epoch": 12.36,
+      "learning_rate": 2.2364521253902266e-05,
+      "loss": 2.2403,
+      "step": 91530
+    },
+    {
+      "epoch": 12.36,
+      "learning_rate": 2.234223558804309e-05,
+      "loss": 2.219,
+      "step": 91540
+    },
+    {
+      "epoch": 12.36,
+      "learning_rate": 2.2319960137761412e-05,
+      "loss": 2.226,
+      "step": 91550
+    },
+    {
+      "epoch": 12.37,
+      "learning_rate": 2.2297694904839743e-05,
+      "loss": 2.2057,
+      "step": 91560
+    },
+    {
+      "epoch": 12.37,
+      "learning_rate": 2.2275439891059904e-05,
+      "loss": 2.2426,
+      "step": 91570
+    },
+    {
+      "epoch": 12.37,
+      "learning_rate": 2.2253195098202776e-05,
+      "loss": 2.2635,
+      "step": 91580
+    },
+    {
+      "epoch": 12.37,
+      "learning_rate": 2.2230960528048415e-05,
+      "loss": 2.2232,
+      "step": 91590
+    },
+    {
+      "epoch": 12.37,
+      "learning_rate": 2.2208736182376163e-05,
+      "loss": 2.2171,
+      "step": 91600
+    },
+    {
+      "epoch": 12.37,
+      "learning_rate": 2.2186522062964433e-05,
+      "loss": 2.2224,
+      "step": 91610
+    },
+    {
+      "epoch": 12.37,
+      "learning_rate": 2.2164318171590938e-05,
+      "loss": 2.2499,
+      "step": 91620
+    },
+    {
+      "epoch": 12.38,
+      "learning_rate": 2.2142124510032437e-05,
+      "loss": 2.2365,
+      "step": 91630
+    },
+    {
+      "epoch": 12.38,
+      "learning_rate": 2.2119941080064946e-05,
+      "loss": 2.2284,
+      "step": 91640
+    },
+    {
+      "epoch": 12.38,
+      "learning_rate": 2.2097767883463692e-05,
+      "loss": 2.2085,
+      "step": 91650
+    },
+    {
+      "epoch": 12.38,
+      "learning_rate": 2.207560492200302e-05,
+      "loss": 2.2084,
+      "step": 91660
+    },
+    {
+      "epoch": 12.38,
+      "learning_rate": 2.2053452197456435e-05,
+      "loss": 2.2434,
+      "step": 91670
+    },
+    {
+      "epoch": 12.38,
+      "learning_rate": 2.2031309711596745e-05,
+      "loss": 2.223,
+      "step": 91680
+    },
+    {
+      "epoch": 12.38,
+      "learning_rate": 2.2009177466195804e-05,
+      "loss": 2.2459,
+      "step": 91690
+    },
+    {
+      "epoch": 12.38,
+      "learning_rate": 2.1987055463024793e-05,
+      "loss": 2.252,
+      "step": 91700
+    },
+    {
+      "epoch": 12.39,
+      "learning_rate": 2.1964943703853867e-05,
+      "loss": 2.2,
+      "step": 91710
+    },
+    {
+      "epoch": 12.39,
+      "learning_rate": 2.1942842190452554e-05,
+      "loss": 2.2486,
+      "step": 91720
+    },
+    {
+      "epoch": 12.39,
+      "learning_rate": 2.1920750924589424e-05,
+      "loss": 2.2214,
+      "step": 91730
+    },
+    {
+      "epoch": 12.39,
+      "learning_rate": 2.1898669908032367e-05,
+      "loss": 2.2453,
+      "step": 91740
+    },
+    {
+      "epoch": 12.39,
+      "learning_rate": 2.187659914254835e-05,
+      "loss": 2.2261,
+      "step": 91750
+    },
+    {
+      "epoch": 12.39,
+      "learning_rate": 2.1854538629903494e-05,
+      "loss": 2.2244,
+      "step": 91760
+    },
+    {
+      "epoch": 12.39,
+      "learning_rate": 2.183248837186322e-05,
+      "loss": 2.2138,
+      "step": 91770
+    },
+    {
+      "epoch": 12.4,
+      "learning_rate": 2.1810448370192035e-05,
+      "loss": 2.223,
+      "step": 91780
+    },
+    {
+      "epoch": 12.4,
+      "learning_rate": 2.1788418626653625e-05,
+      "loss": 2.2262,
+      "step": 91790
+    },
+    {
+      "epoch": 12.4,
+      "learning_rate": 2.1766399143010914e-05,
+      "loss": 2.205,
+      "step": 91800
+    },
+    {
+      "epoch": 12.4,
+      "learning_rate": 2.174438992102593e-05,
+      "loss": 2.1996,
+      "step": 91810
+    },
+    {
+      "epoch": 12.4,
+      "learning_rate": 2.172239096245999e-05,
+      "loss": 2.2337,
+      "step": 91820
+    },
+    {
+      "epoch": 12.4,
+      "learning_rate": 2.1700402269073475e-05,
+      "loss": 2.2384,
+      "step": 91830
+    },
+    {
+      "epoch": 12.4,
+      "learning_rate": 2.1678423842625942e-05,
+      "loss": 2.2182,
+      "step": 91840
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 2.1656455684876267e-05,
+      "loss": 2.2115,
+      "step": 91850
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 2.1634497797582377e-05,
+      "loss": 2.2451,
+      "step": 91860
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 2.1612550182501404e-05,
+      "loss": 2.2411,
+      "step": 91870
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 2.159061284138962e-05,
+      "loss": 2.2488,
+      "step": 91880
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 2.1568685776002558e-05,
+      "loss": 2.2493,
+      "step": 91890
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 2.1546768988094983e-05,
+      "loss": 2.2714,
+      "step": 91900
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 2.1524862479420584e-05,
+      "loss": 2.1985,
+      "step": 91910
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 2.15029662517325e-05,
+      "loss": 2.2081,
+      "step": 91920
+    },
+    {
+      "epoch": 12.42,
+      "learning_rate": 2.1481080306782877e-05,
+      "loss": 2.2127,
+      "step": 91930
+    },
+    {
+      "epoch": 12.42,
+      "learning_rate": 2.1459204646323148e-05,
+      "loss": 2.231,
+      "step": 91940
+    },
+    {
+      "epoch": 12.42,
+      "learning_rate": 2.1437339272103844e-05,
+      "loss": 2.2594,
+      "step": 91950
+    },
+    {
+      "epoch": 12.42,
+      "learning_rate": 2.1415484185874666e-05,
+      "loss": 2.216,
+      "step": 91960
+    },
+    {
+      "epoch": 12.42,
+      "learning_rate": 2.1393639389384615e-05,
+      "loss": 2.2125,
+      "step": 91970
+    },
+    {
+      "epoch": 12.42,
+      "learning_rate": 2.137180488438173e-05,
+      "loss": 2.2177,
+      "step": 91980
+    },
+    {
+      "epoch": 12.42,
+      "learning_rate": 2.134998067261322e-05,
+      "loss": 2.2235,
+      "step": 91990
+    },
+    {
+      "epoch": 12.43,
+      "learning_rate": 2.1328166755825633e-05,
+      "loss": 2.2176,
+      "step": 92000
+    },
+    {
+      "epoch": 12.43,
+      "eval_loss": 2.4086060523986816,
+      "eval_runtime": 1269.3719,
+      "eval_samples_per_second": 59.978,
+      "eval_steps_per_second": 4.999,
+      "step": 92000
+    },
+    {
+      "epoch": 12.43,
+      "learning_rate": 2.1306363135764516e-05,
+      "loss": 2.2402,
+      "step": 92010
+    },
+    {
+      "epoch": 12.43,
+      "learning_rate": 2.1284569814174708e-05,
+      "loss": 2.2256,
+      "step": 92020
+    },
+    {
+      "epoch": 12.43,
+      "learning_rate": 2.1262786792800163e-05,
+      "loss": 2.2356,
+      "step": 92030
+    },
+    {
+      "epoch": 12.43,
+      "learning_rate": 2.1241014073383987e-05,
+      "loss": 2.2135,
+      "step": 92040
+    },
+    {
+      "epoch": 12.43,
+      "learning_rate": 2.1219251657668572e-05,
+      "loss": 2.2379,
+      "step": 92050
+    },
+    {
+      "epoch": 12.43,
+      "learning_rate": 2.119749954739539e-05,
+      "loss": 2.2327,
+      "step": 92060
+    },
+    {
+      "epoch": 12.43,
+      "learning_rate": 2.1175757744305083e-05,
+      "loss": 2.222,
+      "step": 92070
+    },
+    {
+      "epoch": 12.44,
+      "learning_rate": 2.1154026250137505e-05,
+      "loss": 2.2457,
+      "step": 92080
+    },
+    {
+      "epoch": 12.44,
+      "learning_rate": 2.113230506663169e-05,
+      "loss": 2.2255,
+      "step": 92090
+    },
+    {
+      "epoch": 12.44,
+      "learning_rate": 2.1110594195525905e-05,
+      "loss": 2.2326,
+      "step": 92100
+    },
+    {
+      "epoch": 12.44,
+      "learning_rate": 2.1088893638557398e-05,
+      "loss": 2.2207,
+      "step": 92110
+    },
+    {
+      "epoch": 12.44,
+      "learning_rate": 2.1067203397462796e-05,
+      "loss": 2.2028,
+      "step": 92120
+    },
+    {
+      "epoch": 12.44,
+      "learning_rate": 2.104552347397776e-05,
+      "loss": 2.2093,
+      "step": 92130
+    },
+    {
+      "epoch": 12.44,
+      "learning_rate": 2.102385386983727e-05,
+      "loss": 2.2001,
+      "step": 92140
+    },
+    {
+      "epoch": 12.45,
+      "learning_rate": 2.1002194586775356e-05,
+      "loss": 2.2245,
+      "step": 92150
+    },
+    {
+      "epoch": 12.45,
+      "learning_rate": 2.0980545626525212e-05,
+      "loss": 2.2229,
+      "step": 92160
+    },
+    {
+      "epoch": 12.45,
+      "learning_rate": 2.0958906990819358e-05,
+      "loss": 2.2274,
+      "step": 92170
+    },
+    {
+      "epoch": 12.45,
+      "learning_rate": 2.0937278681389302e-05,
+      "loss": 2.181,
+      "step": 92180
+    },
+    {
+      "epoch": 12.45,
+      "learning_rate": 2.091566069996583e-05,
+      "loss": 2.1901,
+      "step": 92190
+    },
+    {
+      "epoch": 12.45,
+      "learning_rate": 2.0894053048278908e-05,
+      "loss": 2.2071,
+      "step": 92200
+    },
+    {
+      "epoch": 12.45,
+      "learning_rate": 2.0872455728057602e-05,
+      "loss": 2.2281,
+      "step": 92210
+    },
+    {
+      "epoch": 12.46,
+      "learning_rate": 2.0850868741030246e-05,
+      "loss": 2.2247,
+      "step": 92220
+    },
+    {
+      "epoch": 12.46,
+      "learning_rate": 2.0829292088924292e-05,
+      "loss": 2.2183,
+      "step": 92230
+    },
+    {
+      "epoch": 12.46,
+      "learning_rate": 2.0807725773466326e-05,
+      "loss": 2.232,
+      "step": 92240
+    },
+    {
+      "epoch": 12.46,
+      "learning_rate": 2.0786169796382203e-05,
+      "loss": 2.2621,
+      "step": 92250
+    },
+    {
+      "epoch": 12.46,
+      "learning_rate": 2.0764624159396903e-05,
+      "loss": 2.1754,
+      "step": 92260
+    },
+    {
+      "epoch": 12.46,
+      "learning_rate": 2.0743088864234535e-05,
+      "loss": 2.2388,
+      "step": 92270
+    },
+    {
+      "epoch": 12.46,
+      "learning_rate": 2.072156391261842e-05,
+      "loss": 2.2055,
+      "step": 92280
+    },
+    {
+      "epoch": 12.46,
+      "learning_rate": 2.0700049306271067e-05,
+      "loss": 2.2291,
+      "step": 92290
+    },
+    {
+      "epoch": 12.47,
+      "learning_rate": 2.0678545046914226e-05,
+      "loss": 2.237,
+      "step": 92300
+    },
+    {
+      "epoch": 12.47,
+      "learning_rate": 2.065705113626859e-05,
+      "loss": 2.2516,
+      "step": 92310
+    },
+    {
+      "epoch": 12.47,
+      "learning_rate": 2.0635567576054262e-05,
+      "loss": 2.2014,
+      "step": 92320
+    },
+    {
+      "epoch": 12.47,
+      "learning_rate": 2.0614094367990374e-05,
+      "loss": 2.2167,
+      "step": 92330
+    },
+    {
+      "epoch": 12.47,
+      "learning_rate": 2.059263151379535e-05,
+      "loss": 2.2223,
+      "step": 92340
+    },
+    {
+      "epoch": 12.47,
+      "learning_rate": 2.057117901518666e-05,
+      "loss": 2.2324,
+      "step": 92350
+    },
+    {
+      "epoch": 12.47,
+      "learning_rate": 2.054973687388099e-05,
+      "loss": 2.2325,
+      "step": 92360
+    },
+    {
+      "epoch": 12.48,
+      "learning_rate": 2.0528305091594262e-05,
+      "loss": 2.2096,
+      "step": 92370
+    },
+    {
+      "epoch": 12.48,
+      "learning_rate": 2.0506883670041468e-05,
+      "loss": 2.1884,
+      "step": 92380
+    },
+    {
+      "epoch": 12.48,
+      "learning_rate": 2.0485472610936827e-05,
+      "loss": 2.2401,
+      "step": 92390
+    },
+    {
+      "epoch": 12.48,
+      "learning_rate": 2.0464071915993745e-05,
+      "loss": 2.2122,
+      "step": 92400
+    },
+    {
+      "epoch": 12.48,
+      "learning_rate": 2.0442681586924726e-05,
+      "loss": 2.2141,
+      "step": 92410
+    },
+    {
+      "epoch": 12.48,
+      "learning_rate": 2.0421301625441545e-05,
+      "loss": 2.2109,
+      "step": 92420
+    },
+    {
+      "epoch": 12.48,
+      "learning_rate": 2.0399932033255085e-05,
+      "loss": 2.2521,
+      "step": 92430
+    },
+    {
+      "epoch": 12.48,
+      "learning_rate": 2.0378572812075344e-05,
+      "loss": 2.238,
+      "step": 92440
+    },
+    {
+      "epoch": 12.49,
+      "learning_rate": 2.0357223963611642e-05,
+      "loss": 2.2535,
+      "step": 92450
+    },
+    {
+      "epoch": 12.49,
+      "learning_rate": 2.033588548957234e-05,
+      "loss": 2.2248,
+      "step": 92460
+    },
+    {
+      "epoch": 12.49,
+      "learning_rate": 2.0314557391665025e-05,
+      "loss": 2.2302,
+      "step": 92470
+    },
+    {
+      "epoch": 12.49,
+      "learning_rate": 2.0293239671596385e-05,
+      "loss": 2.219,
+      "step": 92480
+    },
+    {
+      "epoch": 12.49,
+      "learning_rate": 2.0271932331072367e-05,
+      "loss": 2.2474,
+      "step": 92490
+    },
+    {
+      "epoch": 12.49,
+      "learning_rate": 2.0250635371798124e-05,
+      "loss": 2.2179,
+      "step": 92500
+    },
+    {
+      "epoch": 12.49,
+      "learning_rate": 2.022934879547778e-05,
+      "loss": 2.2209,
+      "step": 92510
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 2.020807260381485e-05,
+      "loss": 2.1919,
+      "step": 92520
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 2.0186806798511845e-05,
+      "loss": 2.2113,
+      "step": 92530
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 2.0165551381270585e-05,
+      "loss": 2.2179,
+      "step": 92540
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 2.014430635379199e-05,
+      "loss": 2.2343,
+      "step": 92550
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 2.01230717177761e-05,
+      "loss": 2.2336,
+      "step": 92560
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 2.0101847474922246e-05,
+      "loss": 2.2326,
+      "step": 92570
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 2.0080633626928826e-05,
+      "loss": 2.2414,
+      "step": 92580
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 2.0059430175493407e-05,
+      "loss": 2.2063,
+      "step": 92590
+    },
+    {
+      "epoch": 12.51,
+      "learning_rate": 2.0038237122312827e-05,
+      "loss": 2.2023,
+      "step": 92600
+    },
+    {
+      "epoch": 12.51,
+      "learning_rate": 2.001705446908295e-05,
+      "loss": 2.2186,
+      "step": 92610
+    },
+    {
+      "epoch": 12.51,
+      "learning_rate": 1.999588221749895e-05,
+      "loss": 2.2591,
+      "step": 92620
+    },
+    {
+      "epoch": 12.51,
+      "learning_rate": 1.9974720369255054e-05,
+      "loss": 2.2078,
+      "step": 92630
+    },
+    {
+      "epoch": 12.51,
+      "learning_rate": 1.9953568926044662e-05,
+      "loss": 2.248,
+      "step": 92640
+    },
+    {
+      "epoch": 12.51,
+      "learning_rate": 1.9932427889560465e-05,
+      "loss": 2.2084,
+      "step": 92650
+    },
+    {
+      "epoch": 12.51,
+      "learning_rate": 1.9911297261494214e-05,
+      "loss": 2.2405,
+      "step": 92660
+    },
+    {
+      "epoch": 12.52,
+      "learning_rate": 1.9890177043536804e-05,
+      "loss": 2.2453,
+      "step": 92670
+    },
+    {
+      "epoch": 12.52,
+      "learning_rate": 1.9869067237378333e-05,
+      "loss": 2.2229,
+      "step": 92680
+    },
+    {
+      "epoch": 12.52,
+      "learning_rate": 1.984796784470812e-05,
+      "loss": 2.2225,
+      "step": 92690
+    },
+    {
+      "epoch": 12.52,
+      "learning_rate": 1.9826878867214657e-05,
+      "loss": 2.2131,
+      "step": 92700
+    },
+    {
+      "epoch": 12.52,
+      "learning_rate": 1.9805800306585418e-05,
+      "loss": 2.2373,
+      "step": 92710
+    },
+    {
+      "epoch": 12.52,
+      "learning_rate": 1.9784732164507268e-05,
+      "loss": 2.2226,
+      "step": 92720
+    },
+    {
+      "epoch": 12.52,
+      "learning_rate": 1.9763674442666104e-05,
+      "loss": 2.2329,
+      "step": 92730
+    },
+    {
+      "epoch": 12.53,
+      "learning_rate": 1.9742627142747068e-05,
+      "loss": 2.1925,
+      "step": 92740
+    },
+    {
+      "epoch": 12.53,
+      "learning_rate": 1.9721590266434427e-05,
+      "loss": 2.2206,
+      "step": 92750
+    },
+    {
+      "epoch": 12.53,
+      "learning_rate": 1.970056381541157e-05,
+      "loss": 2.2343,
+      "step": 92760
+    },
+    {
+      "epoch": 12.53,
+      "learning_rate": 1.967954779136116e-05,
+      "loss": 2.2207,
+      "step": 92770
+    },
+    {
+      "epoch": 12.53,
+      "learning_rate": 1.9658542195964945e-05,
+      "loss": 2.248,
+      "step": 92780
+    },
+    {
+      "epoch": 12.53,
+      "learning_rate": 1.9637547030903822e-05,
+      "loss": 2.2269,
+      "step": 92790
+    },
+    {
+      "epoch": 12.53,
+      "learning_rate": 1.961656229785795e-05,
+      "loss": 2.1977,
+      "step": 92800
+    },
+    {
+      "epoch": 12.53,
+      "learning_rate": 1.9595587998506542e-05,
+      "loss": 2.2045,
+      "step": 92810
+    },
+    {
+      "epoch": 12.54,
+      "learning_rate": 1.9574624134528076e-05,
+      "loss": 2.2152,
+      "step": 92820
+    },
+    {
+      "epoch": 12.54,
+      "learning_rate": 1.9553670707600123e-05,
+      "loss": 2.2466,
+      "step": 92830
+    },
+    {
+      "epoch": 12.54,
+      "learning_rate": 1.9532727719399415e-05,
+      "loss": 2.2343,
+      "step": 92840
+    },
+    {
+      "epoch": 12.54,
+      "learning_rate": 1.9511795171601934e-05,
+      "loss": 2.235,
+      "step": 92850
+    },
+    {
+      "epoch": 12.54,
+      "learning_rate": 1.9490873065882724e-05,
+      "loss": 2.2112,
+      "step": 92860
+    },
+    {
+      "epoch": 12.54,
+      "learning_rate": 1.9469961403916072e-05,
+      "loss": 2.2192,
+      "step": 92870
+    },
+    {
+      "epoch": 12.54,
+      "learning_rate": 1.944906018737533e-05,
+      "loss": 2.2214,
+      "step": 92880
+    },
+    {
+      "epoch": 12.55,
+      "learning_rate": 1.9428169417933127e-05,
+      "loss": 2.2339,
+      "step": 92890
+    },
+    {
+      "epoch": 12.55,
+      "learning_rate": 1.940728909726127e-05,
+      "loss": 2.2471,
+      "step": 92900
+    },
+    {
+      "epoch": 12.55,
+      "learning_rate": 1.9386419227030542e-05,
+      "loss": 2.2155,
+      "step": 92910
+    },
+    {
+      "epoch": 12.55,
+      "learning_rate": 1.936555980891111e-05,
+      "loss": 2.2397,
+      "step": 92920
+    },
+    {
+      "epoch": 12.55,
+      "learning_rate": 1.934471084457215e-05,
+      "loss": 2.231,
+      "step": 92930
+    },
+    {
+      "epoch": 12.55,
+      "learning_rate": 1.932387233568211e-05,
+      "loss": 2.2353,
+      "step": 92940
+    },
+    {
+      "epoch": 12.55,
+      "learning_rate": 1.930304428390855e-05,
+      "loss": 2.2124,
+      "step": 92950
+    },
+    {
+      "epoch": 12.55,
+      "learning_rate": 1.9282226690918122e-05,
+      "loss": 2.2021,
+      "step": 92960
+    },
+    {
+      "epoch": 12.56,
+      "learning_rate": 1.926141955837682e-05,
+      "loss": 2.2739,
+      "step": 92970
+    },
+    {
+      "epoch": 12.56,
+      "learning_rate": 1.9240622887949663e-05,
+      "loss": 2.219,
+      "step": 92980
+    },
+    {
+      "epoch": 12.56,
+      "learning_rate": 1.9219836681300797e-05,
+      "loss": 2.2419,
+      "step": 92990
+    },
+    {
+      "epoch": 12.56,
+      "learning_rate": 1.9199060940093714e-05,
+      "loss": 2.237,
+      "step": 93000
+    },
+    {
+      "epoch": 12.56,
+      "eval_loss": 2.406857490539551,
+      "eval_runtime": 1269.1794,
+      "eval_samples_per_second": 59.988,
+      "eval_steps_per_second": 4.999,
+      "step": 93000
+    },
+    {
+      "epoch": 12.56,
+      "learning_rate": 1.9178295665990844e-05,
+      "loss": 2.2253,
+      "step": 93010
+    },
+    {
+      "epoch": 12.56,
+      "learning_rate": 1.9157540860654013e-05,
+      "loss": 2.2572,
+      "step": 93020
+    },
+    {
+      "epoch": 12.56,
+      "learning_rate": 1.9136796525743957e-05,
+      "loss": 2.21,
+      "step": 93030
+    },
+    {
+      "epoch": 12.57,
+      "learning_rate": 1.911606266292075e-05,
+      "loss": 2.2103,
+      "step": 93040
+    },
+    {
+      "epoch": 12.57,
+      "learning_rate": 1.909533927384363e-05,
+      "loss": 2.2428,
+      "step": 93050
+    },
+    {
+      "epoch": 12.57,
+      "learning_rate": 1.9074626360170918e-05,
+      "loss": 2.2502,
+      "step": 93060
+    },
+    {
+      "epoch": 12.57,
+      "learning_rate": 1.905392392356011e-05,
+      "loss": 2.208,
+      "step": 93070
+    },
+    {
+      "epoch": 12.57,
+      "learning_rate": 1.903323196566786e-05,
+      "loss": 2.2661,
+      "step": 93080
+    },
+    {
+      "epoch": 12.57,
+      "learning_rate": 1.901255048815003e-05,
+      "loss": 2.2563,
+      "step": 93090
+    },
+    {
+      "epoch": 12.57,
+      "learning_rate": 1.8991879492661694e-05,
+      "loss": 2.2292,
+      "step": 93100
+    },
+    {
+      "epoch": 12.58,
+      "learning_rate": 1.8971218980856863e-05,
+      "loss": 2.1974,
+      "step": 93110
+    },
+    {
+      "epoch": 12.58,
+      "learning_rate": 1.895056895438898e-05,
+      "loss": 2.2258,
+      "step": 93120
+    },
+    {
+      "epoch": 12.58,
+      "learning_rate": 1.892992941491042e-05,
+      "loss": 2.2105,
+      "step": 93130
+    },
+    {
+      "epoch": 12.58,
+      "learning_rate": 1.8909300364072937e-05,
+      "loss": 2.2175,
+      "step": 93140
+    },
+    {
+      "epoch": 12.58,
+      "learning_rate": 1.8888681803527257e-05,
+      "loss": 2.2368,
+      "step": 93150
+    },
+    {
+      "epoch": 12.58,
+      "learning_rate": 1.886807373492334e-05,
+      "loss": 2.2172,
+      "step": 93160
+    },
+    {
+      "epoch": 12.58,
+      "learning_rate": 1.884747615991035e-05,
+      "loss": 2.2145,
+      "step": 93170
+    },
+    {
+      "epoch": 12.58,
+      "learning_rate": 1.8826889080136558e-05,
+      "loss": 2.2076,
+      "step": 93180
+    },
+    {
+      "epoch": 12.59,
+      "learning_rate": 1.8806312497249376e-05,
+      "loss": 2.2249,
+      "step": 93190
+    },
+    {
+      "epoch": 12.59,
+      "learning_rate": 1.8785746412895454e-05,
+      "loss": 2.2347,
+      "step": 93200
+    },
+    {
+      "epoch": 12.59,
+      "learning_rate": 1.8765190828720505e-05,
+      "loss": 2.23,
+      "step": 93210
+    },
+    {
+      "epoch": 12.59,
+      "learning_rate": 1.8744645746369536e-05,
+      "loss": 2.2404,
+      "step": 93220
+    },
+    {
+      "epoch": 12.59,
+      "learning_rate": 1.872411116748651e-05,
+      "loss": 2.2183,
+      "step": 93230
+    },
+    {
+      "epoch": 12.59,
+      "learning_rate": 1.870358709371472e-05,
+      "loss": 2.2372,
+      "step": 93240
+    },
+    {
+      "epoch": 12.59,
+      "learning_rate": 1.8683073526696623e-05,
+      "loss": 2.2558,
+      "step": 93250
+    },
+    {
+      "epoch": 12.6,
+      "learning_rate": 1.8662570468073735e-05,
+      "loss": 2.2441,
+      "step": 93260
+    },
+    {
+      "epoch": 12.6,
+      "learning_rate": 1.8642077919486752e-05,
+      "loss": 2.23,
+      "step": 93270
+    },
+    {
+      "epoch": 12.6,
+      "learning_rate": 1.862159588257557e-05,
+      "loss": 2.2571,
+      "step": 93280
+    },
+    {
+      "epoch": 12.6,
+      "learning_rate": 1.860112435897922e-05,
+      "loss": 2.2288,
+      "step": 93290
+    },
+    {
+      "epoch": 12.6,
+      "learning_rate": 1.8580663350335977e-05,
+      "loss": 2.2151,
+      "step": 93300
+    },
+    {
+      "epoch": 12.6,
+      "learning_rate": 1.856021285828305e-05,
+      "loss": 2.2203,
+      "step": 93310
+    },
+    {
+      "epoch": 12.6,
+      "learning_rate": 1.8539772884457084e-05,
+      "loss": 2.2385,
+      "step": 93320
+    },
+    {
+      "epoch": 12.6,
+      "learning_rate": 1.8519343430493645e-05,
+      "loss": 2.2213,
+      "step": 93330
+    },
+    {
+      "epoch": 12.61,
+      "learning_rate": 1.849892449802766e-05,
+      "loss": 2.2235,
+      "step": 93340
+    },
+    {
+      "epoch": 12.61,
+      "learning_rate": 1.8478516088693064e-05,
+      "loss": 2.2109,
+      "step": 93350
+    },
+    {
+      "epoch": 12.61,
+      "learning_rate": 1.8458118204122975e-05,
+      "loss": 2.2397,
+      "step": 93360
+    },
+    {
+      "epoch": 12.61,
+      "learning_rate": 1.8437730845949755e-05,
+      "loss": 2.2284,
+      "step": 93370
+    },
+    {
+      "epoch": 12.61,
+      "learning_rate": 1.8417354015804854e-05,
+      "loss": 2.2067,
+      "step": 93380
+    },
+    {
+      "epoch": 12.61,
+      "learning_rate": 1.8396987715318827e-05,
+      "loss": 2.2162,
+      "step": 93390
+    },
+    {
+      "epoch": 12.61,
+      "learning_rate": 1.8376631946121546e-05,
+      "loss": 2.2135,
+      "step": 93400
+    },
+    {
+      "epoch": 12.62,
+      "learning_rate": 1.835628670984185e-05,
+      "loss": 2.2441,
+      "step": 93410
+    },
+    {
+      "epoch": 12.62,
+      "learning_rate": 1.8335952008107956e-05,
+      "loss": 2.2199,
+      "step": 93420
+    },
+    {
+      "epoch": 12.62,
+      "learning_rate": 1.8315627842546947e-05,
+      "loss": 2.1905,
+      "step": 93430
+    },
+    {
+      "epoch": 12.62,
+      "learning_rate": 1.829531421478532e-05,
+      "loss": 2.2261,
+      "step": 93440
+    },
+    {
+      "epoch": 12.62,
+      "learning_rate": 1.82770409609619e-05,
+      "loss": 2.2382,
+      "step": 93450
+    },
+    {
+      "epoch": 12.62,
+      "learning_rate": 1.8256747359496832e-05,
+      "loss": 2.2406,
+      "step": 93460
+    },
+    {
+      "epoch": 12.62,
+      "learning_rate": 1.8236464300542925e-05,
+      "loss": 2.2395,
+      "step": 93470
+    },
+    {
+      "epoch": 12.63,
+      "learning_rate": 1.8216191785723316e-05,
+      "loss": 2.2318,
+      "step": 93480
+    },
+    {
+      "epoch": 12.63,
+      "learning_rate": 1.8195929816660276e-05,
+      "loss": 2.2226,
+      "step": 93490
+    },
+    {
+      "epoch": 12.63,
+      "learning_rate": 1.817567839497519e-05,
+      "loss": 2.2375,
+      "step": 93500
+    },
+    {
+      "epoch": 12.63,
+      "learning_rate": 1.8155437522288694e-05,
+      "loss": 2.1982,
+      "step": 93510
+    },
+    {
+      "epoch": 12.63,
+      "learning_rate": 1.813520720022048e-05,
+      "loss": 2.2262,
+      "step": 93520
+    },
+    {
+      "epoch": 12.63,
+      "learning_rate": 1.81149874303895e-05,
+      "loss": 2.2067,
+      "step": 93530
+    },
+    {
+      "epoch": 12.63,
+      "learning_rate": 1.8094778214413775e-05,
+      "loss": 2.2605,
+      "step": 93540
+    },
+    {
+      "epoch": 12.63,
+      "learning_rate": 1.8074579553910478e-05,
+      "loss": 2.2324,
+      "step": 93550
+    },
+    {
+      "epoch": 12.64,
+      "learning_rate": 1.805439145049602e-05,
+      "loss": 2.2287,
+      "step": 93560
+    },
+    {
+      "epoch": 12.64,
+      "learning_rate": 1.8034213905785903e-05,
+      "loss": 2.2306,
+      "step": 93570
+    },
+    {
+      "epoch": 12.64,
+      "learning_rate": 1.801404692139476e-05,
+      "loss": 2.235,
+      "step": 93580
+    },
+    {
+      "epoch": 12.64,
+      "learning_rate": 1.7993890498936474e-05,
+      "loss": 2.2293,
+      "step": 93590
+    },
+    {
+      "epoch": 12.64,
+      "learning_rate": 1.797374464002398e-05,
+      "loss": 2.2242,
+      "step": 93600
+    },
+    {
+      "epoch": 12.64,
+      "learning_rate": 1.7953609346269483e-05,
+      "loss": 2.2394,
+      "step": 93610
+    },
+    {
+      "epoch": 12.64,
+      "learning_rate": 1.793348461928416e-05,
+      "loss": 2.2355,
+      "step": 93620
+    },
+    {
+      "epoch": 12.65,
+      "learning_rate": 1.7913370460678545e-05,
+      "loss": 2.2109,
+      "step": 93630
+    },
+    {
+      "epoch": 12.65,
+      "learning_rate": 1.789326687206218e-05,
+      "loss": 2.2391,
+      "step": 93640
+    },
+    {
+      "epoch": 12.65,
+      "learning_rate": 1.7873173855043876e-05,
+      "loss": 2.2023,
+      "step": 93650
+    },
+    {
+      "epoch": 12.65,
+      "learning_rate": 1.7853091411231503e-05,
+      "loss": 2.2184,
+      "step": 93660
+    },
+    {
+      "epoch": 12.65,
+      "learning_rate": 1.78330195422321e-05,
+      "loss": 2.2166,
+      "step": 93670
+    },
+    {
+      "epoch": 12.65,
+      "learning_rate": 1.781295824965196e-05,
+      "loss": 2.2084,
+      "step": 93680
+    },
+    {
+      "epoch": 12.65,
+      "learning_rate": 1.779290753509639e-05,
+      "loss": 2.2446,
+      "step": 93690
+    },
+    {
+      "epoch": 12.65,
+      "learning_rate": 1.7772867400169878e-05,
+      "loss": 2.2098,
+      "step": 93700
+    },
+    {
+      "epoch": 12.66,
+      "learning_rate": 1.7752837846476187e-05,
+      "loss": 2.2003,
+      "step": 93710
+    },
+    {
+      "epoch": 12.66,
+      "learning_rate": 1.773281887561807e-05,
+      "loss": 2.2262,
+      "step": 93720
+    },
+    {
+      "epoch": 12.66,
+      "learning_rate": 1.7712810489197565e-05,
+      "loss": 2.2422,
+      "step": 93730
+    },
+    {
+      "epoch": 12.66,
+      "learning_rate": 1.769281268881579e-05,
+      "loss": 2.2122,
+      "step": 93740
+    },
+    {
+      "epoch": 12.66,
+      "learning_rate": 1.7672825476072977e-05,
+      "loss": 2.2549,
+      "step": 93750
+    },
+    {
+      "epoch": 12.66,
+      "learning_rate": 1.7652848852568653e-05,
+      "loss": 2.221,
+      "step": 93760
+    },
+    {
+      "epoch": 12.66,
+      "learning_rate": 1.7632882819901367e-05,
+      "loss": 2.1962,
+      "step": 93770
+    },
+    {
+      "epoch": 12.67,
+      "learning_rate": 1.761292737966883e-05,
+      "loss": 2.2582,
+      "step": 93780
+    },
+    {
+      "epoch": 12.67,
+      "learning_rate": 1.7592982533468008e-05,
+      "loss": 2.2077,
+      "step": 93790
+    },
+    {
+      "epoch": 12.67,
+      "learning_rate": 1.7573048282894897e-05,
+      "loss": 2.2301,
+      "step": 93800
+    },
+    {
+      "epoch": 12.67,
+      "learning_rate": 1.7553124629544785e-05,
+      "loss": 2.201,
+      "step": 93810
+    },
+    {
+      "epoch": 12.67,
+      "learning_rate": 1.753321157501188e-05,
+      "loss": 2.2377,
+      "step": 93820
+    },
+    {
+      "epoch": 12.67,
+      "learning_rate": 1.7513309120889808e-05,
+      "loss": 2.226,
+      "step": 93830
+    },
+    {
+      "epoch": 12.67,
+      "learning_rate": 1.7493417268771158e-05,
+      "loss": 2.2366,
+      "step": 93840
+    },
+    {
+      "epoch": 12.68,
+      "learning_rate": 1.7473536020247797e-05,
+      "loss": 2.2078,
+      "step": 93850
+    },
+    {
+      "epoch": 12.68,
+      "learning_rate": 1.7453665376910664e-05,
+      "loss": 2.2469,
+      "step": 93860
+    },
+    {
+      "epoch": 12.68,
+      "learning_rate": 1.743380534034982e-05,
+      "loss": 2.2306,
+      "step": 93870
+    },
+    {
+      "epoch": 12.68,
+      "learning_rate": 1.74139559121546e-05,
+      "loss": 2.2248,
+      "step": 93880
+    },
+    {
+      "epoch": 12.68,
+      "learning_rate": 1.739411709391339e-05,
+      "loss": 2.2633,
+      "step": 93890
+    },
+    {
+      "epoch": 12.68,
+      "learning_rate": 1.7374288887213728e-05,
+      "loss": 2.2426,
+      "step": 93900
+    },
+    {
+      "epoch": 12.68,
+      "learning_rate": 1.7354471293642374e-05,
+      "loss": 2.2229,
+      "step": 93910
+    },
+    {
+      "epoch": 12.68,
+      "learning_rate": 1.7334664314785145e-05,
+      "loss": 2.2352,
+      "step": 93920
+    },
+    {
+      "epoch": 12.69,
+      "learning_rate": 1.7314867952227123e-05,
+      "loss": 2.2329,
+      "step": 93930
+    },
+    {
+      "epoch": 12.69,
+      "learning_rate": 1.729508220755244e-05,
+      "loss": 2.2049,
+      "step": 93940
+    },
+    {
+      "epoch": 12.69,
+      "learning_rate": 1.7275307082344396e-05,
+      "loss": 2.2222,
+      "step": 93950
+    },
+    {
+      "epoch": 12.69,
+      "learning_rate": 1.7255542578185495e-05,
+      "loss": 2.2275,
+      "step": 93960
+    },
+    {
+      "epoch": 12.69,
+      "learning_rate": 1.7235788696657334e-05,
+      "loss": 2.2433,
+      "step": 93970
+    },
+    {
+      "epoch": 12.69,
+      "learning_rate": 1.7216045439340664e-05,
+      "loss": 2.229,
+      "step": 93980
+    },
+    {
+      "epoch": 12.69,
+      "learning_rate": 1.7196312807815443e-05,
+      "loss": 2.2099,
+      "step": 93990
+    },
+    {
+      "epoch": 12.7,
+      "learning_rate": 1.7176590803660707e-05,
+      "loss": 2.2432,
+      "step": 94000
+    },
+    {
+      "epoch": 12.7,
+      "eval_loss": 2.4050042629241943,
+      "eval_runtime": 1269.1883,
+      "eval_samples_per_second": 59.987,
+      "eval_steps_per_second": 4.999,
+      "step": 94000
+    },
+    {
+      "epoch": 12.7,
+      "learning_rate": 1.7156879428454724e-05,
+      "loss": 2.2023,
+      "step": 94010
+    },
+    {
+      "epoch": 12.7,
+      "learning_rate": 1.7137178683774766e-05,
+      "loss": 2.2124,
+      "step": 94020
+    },
+    {
+      "epoch": 12.7,
+      "learning_rate": 1.711748857119744e-05,
+      "loss": 2.2149,
+      "step": 94030
+    },
+    {
+      "epoch": 12.7,
+      "learning_rate": 1.7097809092298336e-05,
+      "loss": 2.2453,
+      "step": 94040
+    },
+    {
+      "epoch": 12.7,
+      "learning_rate": 1.7078140248652344e-05,
+      "loss": 2.2021,
+      "step": 94050
+    },
+    {
+      "epoch": 12.7,
+      "learning_rate": 1.705848204183339e-05,
+      "loss": 2.2236,
+      "step": 94060
+    },
+    {
+      "epoch": 12.7,
+      "learning_rate": 1.7038834473414558e-05,
+      "loss": 2.21,
+      "step": 94070
+    },
+    {
+      "epoch": 12.71,
+      "learning_rate": 1.7019197544968167e-05,
+      "loss": 2.2574,
+      "step": 94080
+    },
+    {
+      "epoch": 12.71,
+      "learning_rate": 1.6999571258065585e-05,
+      "loss": 2.2274,
+      "step": 94090
+    },
+    {
+      "epoch": 12.71,
+      "learning_rate": 1.6979955614277347e-05,
+      "loss": 2.2381,
+      "step": 94100
+    },
+    {
+      "epoch": 12.71,
+      "learning_rate": 1.696035061517324e-05,
+      "loss": 2.2442,
+      "step": 94110
+    },
+    {
+      "epoch": 12.71,
+      "learning_rate": 1.694075626232203e-05,
+      "loss": 2.2079,
+      "step": 94120
+    },
+    {
+      "epoch": 12.71,
+      "learning_rate": 1.692117255729178e-05,
+      "loss": 2.2392,
+      "step": 94130
+    },
+    {
+      "epoch": 12.71,
+      "learning_rate": 1.6901599501649633e-05,
+      "loss": 2.2378,
+      "step": 94140
+    },
+    {
+      "epoch": 12.72,
+      "learning_rate": 1.6882037096961837e-05,
+      "loss": 2.2142,
+      "step": 94150
+    },
+    {
+      "epoch": 12.72,
+      "learning_rate": 1.686248534479389e-05,
+      "loss": 2.2211,
+      "step": 94160
+    },
+    {
+      "epoch": 12.72,
+      "learning_rate": 1.684294424671038e-05,
+      "loss": 2.2312,
+      "step": 94170
+    },
+    {
+      "epoch": 12.72,
+      "learning_rate": 1.682341380427501e-05,
+      "loss": 2.1923,
+      "step": 94180
+    },
+    {
+      "epoch": 12.72,
+      "learning_rate": 1.6803894019050707e-05,
+      "loss": 2.2344,
+      "step": 94190
+    },
+    {
+      "epoch": 12.72,
+      "learning_rate": 1.6784384892599485e-05,
+      "loss": 2.2057,
+      "step": 94200
+    },
+    {
+      "epoch": 12.72,
+      "learning_rate": 1.6764886426482577e-05,
+      "loss": 2.2196,
+      "step": 94210
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 1.6745398622260204e-05,
+      "loss": 2.2263,
+      "step": 94220
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 1.6725921481491934e-05,
+      "loss": 2.2255,
+      "step": 94230
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 1.6706455005736334e-05,
+      "loss": 2.2155,
+      "step": 94240
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 1.6686999196551233e-05,
+      "loss": 2.2351,
+      "step": 94250
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 1.666755405549351e-05,
+      "loss": 2.2147,
+      "step": 94260
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 1.6648119584119192e-05,
+      "loss": 2.2225,
+      "step": 94270
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 1.6628695783983564e-05,
+      "loss": 2.2286,
+      "step": 94280
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 1.6609282656640938e-05,
+      "loss": 2.1909,
+      "step": 94290
+    },
+    {
+      "epoch": 12.74,
+      "learning_rate": 1.6589880203644783e-05,
+      "loss": 2.2199,
+      "step": 94300
+    },
+    {
+      "epoch": 12.74,
+      "learning_rate": 1.6570488426547795e-05,
+      "loss": 2.2403,
+      "step": 94310
+    },
+    {
+      "epoch": 12.74,
+      "learning_rate": 1.655110732690173e-05,
+      "loss": 2.2352,
+      "step": 94320
+    },
+    {
+      "epoch": 12.74,
+      "learning_rate": 1.653173690625758e-05,
+      "loss": 2.2309,
+      "step": 94330
+    },
+    {
+      "epoch": 12.74,
+      "learning_rate": 1.6512377166165376e-05,
+      "loss": 2.2153,
+      "step": 94340
+    },
+    {
+      "epoch": 12.74,
+      "learning_rate": 1.6493028108174323e-05,
+      "loss": 2.2505,
+      "step": 94350
+    },
+    {
+      "epoch": 12.74,
+      "learning_rate": 1.6473689733832867e-05,
+      "loss": 2.2075,
+      "step": 94360
+    },
+    {
+      "epoch": 12.75,
+      "learning_rate": 1.645436204468849e-05,
+      "loss": 2.2326,
+      "step": 94370
+    },
+    {
+      "epoch": 12.75,
+      "learning_rate": 1.6435045042287836e-05,
+      "loss": 2.235,
+      "step": 94380
+    },
+    {
+      "epoch": 12.75,
+      "learning_rate": 1.6415738728176735e-05,
+      "loss": 2.2348,
+      "step": 94390
+    },
+    {
+      "epoch": 12.75,
+      "learning_rate": 1.639644310390012e-05,
+      "loss": 2.236,
+      "step": 94400
+    },
+    {
+      "epoch": 12.75,
+      "learning_rate": 1.6377158171002154e-05,
+      "loss": 2.2238,
+      "step": 94410
+    },
+    {
+      "epoch": 12.75,
+      "learning_rate": 1.6357883931025966e-05,
+      "loss": 2.2461,
+      "step": 94420
+    },
+    {
+      "epoch": 12.75,
+      "learning_rate": 1.6338620385514047e-05,
+      "loss": 2.2448,
+      "step": 94430
+    },
+    {
+      "epoch": 12.75,
+      "learning_rate": 1.6319367536007838e-05,
+      "loss": 2.2146,
+      "step": 94440
+    },
+    {
+      "epoch": 12.76,
+      "learning_rate": 1.630012538404808e-05,
+      "loss": 2.2232,
+      "step": 94450
+    },
+    {
+      "epoch": 12.76,
+      "learning_rate": 1.628089393117457e-05,
+      "loss": 2.1952,
+      "step": 94460
+    },
+    {
+      "epoch": 12.76,
+      "learning_rate": 1.626167317892625e-05,
+      "loss": 2.2556,
+      "step": 94470
+    },
+    {
+      "epoch": 12.76,
+      "learning_rate": 1.6242463128841248e-05,
+      "loss": 2.2131,
+      "step": 94480
+    },
+    {
+      "epoch": 12.76,
+      "learning_rate": 1.6223263782456825e-05,
+      "loss": 2.2171,
+      "step": 94490
+    },
+    {
+      "epoch": 12.76,
+      "learning_rate": 1.6204075141309315e-05,
+      "loss": 2.2195,
+      "step": 94500
+    },
+    {
+      "epoch": 12.76,
+      "learning_rate": 1.6184897206934323e-05,
+      "loss": 2.2304,
+      "step": 94510
+    },
+    {
+      "epoch": 12.77,
+      "learning_rate": 1.616572998086647e-05,
+      "loss": 2.2207,
+      "step": 94520
+    },
+    {
+      "epoch": 12.77,
+      "learning_rate": 1.614657346463965e-05,
+      "loss": 2.2163,
+      "step": 94530
+    },
+    {
+      "epoch": 12.77,
+      "learning_rate": 1.6127427659786768e-05,
+      "loss": 2.2206,
+      "step": 94540
+    },
+    {
+      "epoch": 12.77,
+      "learning_rate": 1.610829256783993e-05,
+      "loss": 2.2491,
+      "step": 94550
+    },
+    {
+      "epoch": 12.77,
+      "learning_rate": 1.6089168190330432e-05,
+      "loss": 2.2124,
+      "step": 94560
+    },
+    {
+      "epoch": 12.77,
+      "learning_rate": 1.607005452878863e-05,
+      "loss": 2.2511,
+      "step": 94570
+    },
+    {
+      "epoch": 12.77,
+      "learning_rate": 1.6050951584744047e-05,
+      "loss": 2.2353,
+      "step": 94580
+    },
+    {
+      "epoch": 12.78,
+      "learning_rate": 1.6031859359725402e-05,
+      "loss": 2.2038,
+      "step": 94590
+    },
+    {
+      "epoch": 12.78,
+      "learning_rate": 1.6012777855260484e-05,
+      "loss": 2.2213,
+      "step": 94600
+    },
+    {
+      "epoch": 12.78,
+      "learning_rate": 1.599370707287632e-05,
+      "loss": 2.2117,
+      "step": 94610
+    },
+    {
+      "epoch": 12.78,
+      "learning_rate": 1.5974647014098897e-05,
+      "loss": 2.2269,
+      "step": 94620
+    },
+    {
+      "epoch": 12.78,
+      "learning_rate": 1.5955597680453552e-05,
+      "loss": 2.216,
+      "step": 94630
+    },
+    {
+      "epoch": 12.78,
+      "learning_rate": 1.593655907346462e-05,
+      "loss": 2.2351,
+      "step": 94640
+    },
+    {
+      "epoch": 12.78,
+      "learning_rate": 1.5917531194655703e-05,
+      "loss": 2.2221,
+      "step": 94650
+    },
+    {
+      "epoch": 12.78,
+      "learning_rate": 1.589851404554941e-05,
+      "loss": 2.1902,
+      "step": 94660
+    },
+    {
+      "epoch": 12.79,
+      "learning_rate": 1.5879507627667543e-05,
+      "loss": 2.2454,
+      "step": 94670
+    },
+    {
+      "epoch": 12.79,
+      "learning_rate": 1.5860511942531116e-05,
+      "loss": 2.2018,
+      "step": 94680
+    },
+    {
+      "epoch": 12.79,
+      "learning_rate": 1.5841526991660186e-05,
+      "loss": 2.2391,
+      "step": 94690
+    },
+    {
+      "epoch": 12.79,
+      "learning_rate": 1.5822552776573967e-05,
+      "loss": 2.1972,
+      "step": 94700
+    },
+    {
+      "epoch": 12.79,
+      "learning_rate": 1.5803589298790874e-05,
+      "loss": 2.2223,
+      "step": 94710
+    },
+    {
+      "epoch": 12.79,
+      "learning_rate": 1.57846365598284e-05,
+      "loss": 2.2209,
+      "step": 94720
+    },
+    {
+      "epoch": 12.79,
+      "learning_rate": 1.5765694561203245e-05,
+      "loss": 2.2384,
+      "step": 94730
+    },
+    {
+      "epoch": 12.8,
+      "learning_rate": 1.5746763304431165e-05,
+      "loss": 2.1929,
+      "step": 94740
+    },
+    {
+      "epoch": 12.8,
+      "learning_rate": 1.5727842791027095e-05,
+      "loss": 2.2589,
+      "step": 94750
+    },
+    {
+      "epoch": 12.8,
+      "learning_rate": 1.5708933022505142e-05,
+      "loss": 2.2285,
+      "step": 94760
+    },
+    {
+      "epoch": 12.8,
+      "learning_rate": 1.5690034000378532e-05,
+      "loss": 2.238,
+      "step": 94770
+    },
+    {
+      "epoch": 12.8,
+      "learning_rate": 1.56711457261596e-05,
+      "loss": 2.2629,
+      "step": 94780
+    },
+    {
+      "epoch": 12.8,
+      "learning_rate": 1.5652268201359825e-05,
+      "loss": 2.2511,
+      "step": 94790
+    },
+    {
+      "epoch": 12.8,
+      "learning_rate": 1.5633401427489862e-05,
+      "loss": 2.2315,
+      "step": 94800
+    },
+    {
+      "epoch": 12.8,
+      "learning_rate": 1.561454540605957e-05,
+      "loss": 2.2393,
+      "step": 94810
+    },
+    {
+      "epoch": 12.81,
+      "learning_rate": 1.5595700138577728e-05,
+      "loss": 2.2301,
+      "step": 94820
+    },
+    {
+      "epoch": 12.81,
+      "learning_rate": 1.557686562655249e-05,
+      "loss": 2.2271,
+      "step": 94830
+    },
+    {
+      "epoch": 12.81,
+      "learning_rate": 1.555804187149101e-05,
+      "loss": 2.2408,
+      "step": 94840
+    },
+    {
+      "epoch": 12.81,
+      "learning_rate": 1.553922887489968e-05,
+      "loss": 2.2371,
+      "step": 94850
+    },
+    {
+      "epoch": 12.81,
+      "learning_rate": 1.5520426638283923e-05,
+      "loss": 2.2448,
+      "step": 94860
+    },
+    {
+      "epoch": 12.81,
+      "learning_rate": 1.550163516314834e-05,
+      "loss": 2.219,
+      "step": 94870
+    },
+    {
+      "epoch": 12.81,
+      "learning_rate": 1.5482854450996728e-05,
+      "loss": 2.2602,
+      "step": 94880
+    },
+    {
+      "epoch": 12.82,
+      "learning_rate": 1.546408450333197e-05,
+      "loss": 2.2157,
+      "step": 94890
+    },
+    {
+      "epoch": 12.82,
+      "learning_rate": 1.5445325321656077e-05,
+      "loss": 2.2229,
+      "step": 94900
+    },
+    {
+      "epoch": 12.82,
+      "learning_rate": 1.5426576907470235e-05,
+      "loss": 2.2313,
+      "step": 94910
+    },
+    {
+      "epoch": 12.82,
+      "learning_rate": 1.5407839262274725e-05,
+      "loss": 2.2283,
+      "step": 94920
+    },
+    {
+      "epoch": 12.82,
+      "learning_rate": 1.5389112387569047e-05,
+      "loss": 2.2115,
+      "step": 94930
+    },
+    {
+      "epoch": 12.82,
+      "learning_rate": 1.537039628485174e-05,
+      "loss": 2.2187,
+      "step": 94940
+    },
+    {
+      "epoch": 12.82,
+      "learning_rate": 1.5351690955620515e-05,
+      "loss": 2.2225,
+      "step": 94950
+    },
+    {
+      "epoch": 12.83,
+      "learning_rate": 1.5332996401372284e-05,
+      "loss": 2.2266,
+      "step": 94960
+    },
+    {
+      "epoch": 12.83,
+      "learning_rate": 1.531431262360301e-05,
+      "loss": 2.2102,
+      "step": 94970
+    },
+    {
+      "epoch": 12.83,
+      "learning_rate": 1.5295639623807832e-05,
+      "loss": 2.2361,
+      "step": 94980
+    },
+    {
+      "epoch": 12.83,
+      "learning_rate": 1.5276977403481007e-05,
+      "loss": 2.2307,
+      "step": 94990
+    },
+    {
+      "epoch": 12.83,
+      "learning_rate": 1.5258325964115936e-05,
+      "loss": 2.2214,
+      "step": 95000
+    },
+    {
+      "epoch": 12.83,
+      "eval_loss": 2.4032440185546875,
+      "eval_runtime": 1275.175,
+      "eval_samples_per_second": 59.706,
+      "eval_steps_per_second": 4.976,
+      "step": 95000
+    },
+    {
+      "epoch": 12.83,
+      "learning_rate": 1.5239685307205263e-05,
+      "loss": 2.2442,
+      "step": 95010
+    },
+    {
+      "epoch": 12.83,
+      "learning_rate": 1.5221055434240525e-05,
+      "loss": 2.2017,
+      "step": 95020
+    },
+    {
+      "epoch": 12.83,
+      "learning_rate": 1.5202436346712665e-05,
+      "loss": 2.2418,
+      "step": 95030
+    },
+    {
+      "epoch": 12.84,
+      "learning_rate": 1.5183828046111557e-05,
+      "loss": 2.2177,
+      "step": 95040
+    },
+    {
+      "epoch": 12.84,
+      "learning_rate": 1.516523053392636e-05,
+      "loss": 2.2339,
+      "step": 95050
+    },
+    {
+      "epoch": 12.84,
+      "learning_rate": 1.5146643811645282e-05,
+      "loss": 2.2128,
+      "step": 95060
+    },
+    {
+      "epoch": 12.84,
+      "learning_rate": 1.5128067880755651e-05,
+      "loss": 2.2292,
+      "step": 95070
+    },
+    {
+      "epoch": 12.84,
+      "learning_rate": 1.5109502742744046e-05,
+      "loss": 2.2461,
+      "step": 95080
+    },
+    {
+      "epoch": 12.84,
+      "learning_rate": 1.509094839909606e-05,
+      "loss": 2.2163,
+      "step": 95090
+    },
+    {
+      "epoch": 12.84,
+      "learning_rate": 1.5072404851296455e-05,
+      "loss": 2.2202,
+      "step": 95100
+    },
+    {
+      "epoch": 12.85,
+      "learning_rate": 1.5053872100829211e-05,
+      "loss": 2.2311,
+      "step": 95110
+    },
+    {
+      "epoch": 12.85,
+      "learning_rate": 1.5035350149177295e-05,
+      "loss": 2.2363,
+      "step": 95120
+    },
+    {
+      "epoch": 12.85,
+      "learning_rate": 1.5016838997822949e-05,
+      "loss": 2.234,
+      "step": 95130
+    },
+    {
+      "epoch": 12.85,
+      "learning_rate": 1.4998338648247493e-05,
+      "loss": 2.2202,
+      "step": 95140
+    },
+    {
+      "epoch": 12.85,
+      "learning_rate": 1.4979849101931341e-05,
+      "loss": 2.2105,
+      "step": 95150
+    },
+    {
+      "epoch": 12.85,
+      "learning_rate": 1.4961370360354142e-05,
+      "loss": 2.2215,
+      "step": 95160
+    },
+    {
+      "epoch": 12.85,
+      "learning_rate": 1.494290242499458e-05,
+      "loss": 2.2078,
+      "step": 95170
+    },
+    {
+      "epoch": 12.85,
+      "learning_rate": 1.492444529733054e-05,
+      "loss": 2.2321,
+      "step": 95180
+    },
+    {
+      "epoch": 12.86,
+      "learning_rate": 1.4905998978838974e-05,
+      "loss": 2.2373,
+      "step": 95190
+    },
+    {
+      "epoch": 12.86,
+      "learning_rate": 1.4887563470996067e-05,
+      "loss": 2.2266,
+      "step": 95200
+    },
+    {
+      "epoch": 12.86,
+      "learning_rate": 1.4869138775277107e-05,
+      "loss": 2.2079,
+      "step": 95210
+    },
+    {
+      "epoch": 12.86,
+      "learning_rate": 1.4850724893156413e-05,
+      "loss": 2.2098,
+      "step": 95220
+    },
+    {
+      "epoch": 12.86,
+      "learning_rate": 1.4832321826107608e-05,
+      "loss": 2.2089,
+      "step": 95230
+    },
+    {
+      "epoch": 12.86,
+      "learning_rate": 1.4813929575603279e-05,
+      "loss": 2.208,
+      "step": 95240
+    },
+    {
+      "epoch": 12.86,
+      "learning_rate": 1.4795548143115299e-05,
+      "loss": 2.2312,
+      "step": 95250
+    },
+    {
+      "epoch": 12.87,
+      "learning_rate": 1.477717753011461e-05,
+      "loss": 2.2004,
+      "step": 95260
+    },
+    {
+      "epoch": 12.87,
+      "learning_rate": 1.4758817738071217e-05,
+      "loss": 2.2263,
+      "step": 95270
+    },
+    {
+      "epoch": 12.87,
+      "learning_rate": 1.474046876845441e-05,
+      "loss": 2.2272,
+      "step": 95280
+    },
+    {
+      "epoch": 12.87,
+      "learning_rate": 1.4722130622732503e-05,
+      "loss": 2.2327,
+      "step": 95290
+    },
+    {
+      "epoch": 12.87,
+      "learning_rate": 1.4703803302372918e-05,
+      "loss": 2.2065,
+      "step": 95300
+    },
+    {
+      "epoch": 12.87,
+      "learning_rate": 1.4685486808842366e-05,
+      "loss": 2.2016,
+      "step": 95310
+    },
+    {
+      "epoch": 12.87,
+      "learning_rate": 1.4667181143606493e-05,
+      "loss": 2.2217,
+      "step": 95320
+    },
+    {
+      "epoch": 12.88,
+      "learning_rate": 1.4648886308130242e-05,
+      "loss": 2.2177,
+      "step": 95330
+    },
+    {
+      "epoch": 12.88,
+      "learning_rate": 1.4630602303877609e-05,
+      "loss": 2.1962,
+      "step": 95340
+    },
+    {
+      "epoch": 12.88,
+      "learning_rate": 1.4612329132311706e-05,
+      "loss": 2.2205,
+      "step": 95350
+    },
+    {
+      "epoch": 12.88,
+      "learning_rate": 1.4594066794894866e-05,
+      "loss": 2.2505,
+      "step": 95360
+    },
+    {
+      "epoch": 12.88,
+      "learning_rate": 1.4575815293088466e-05,
+      "loss": 2.2108,
+      "step": 95370
+    },
+    {
+      "epoch": 12.88,
+      "learning_rate": 1.4557574628353042e-05,
+      "loss": 2.2425,
+      "step": 95380
+    },
+    {
+      "epoch": 12.88,
+      "learning_rate": 1.4539344802148257e-05,
+      "loss": 2.2357,
+      "step": 95390
+    },
+    {
+      "epoch": 12.88,
+      "learning_rate": 1.4521125815932927e-05,
+      "loss": 2.2045,
+      "step": 95400
+    },
+    {
+      "epoch": 12.89,
+      "learning_rate": 1.4502917671165071e-05,
+      "loss": 2.2355,
+      "step": 95410
+    },
+    {
+      "epoch": 12.89,
+      "learning_rate": 1.4484720369301656e-05,
+      "loss": 2.2541,
+      "step": 95420
+    },
+    {
+      "epoch": 12.89,
+      "learning_rate": 1.4466533911798933e-05,
+      "loss": 2.2417,
+      "step": 95430
+    },
+    {
+      "epoch": 12.89,
+      "learning_rate": 1.4448358300112222e-05,
+      "loss": 2.2523,
+      "step": 95440
+    },
+    {
+      "epoch": 12.89,
+      "learning_rate": 1.4430193535696027e-05,
+      "loss": 2.2342,
+      "step": 95450
+    },
+    {
+      "epoch": 12.89,
+      "learning_rate": 1.4412039620003935e-05,
+      "loss": 2.2277,
+      "step": 95460
+    },
+    {
+      "epoch": 12.89,
+      "learning_rate": 1.4393896554488649e-05,
+      "loss": 2.2605,
+      "step": 95470
+    },
+    {
+      "epoch": 12.9,
+      "learning_rate": 1.4375764340602076e-05,
+      "loss": 2.2173,
+      "step": 95480
+    },
+    {
+      "epoch": 12.9,
+      "learning_rate": 1.4357642979795204e-05,
+      "loss": 2.2135,
+      "step": 95490
+    },
+    {
+      "epoch": 12.9,
+      "learning_rate": 1.4339532473518122e-05,
+      "loss": 2.2177,
+      "step": 95500
+    },
+    {
+      "epoch": 12.9,
+      "learning_rate": 1.4321432823220158e-05,
+      "loss": 2.2349,
+      "step": 95510
+    },
+    {
+      "epoch": 12.9,
+      "learning_rate": 1.4303344030349617e-05,
+      "loss": 2.2337,
+      "step": 95520
+    },
+    {
+      "epoch": 12.9,
+      "learning_rate": 1.4285266096354125e-05,
+      "loss": 2.2493,
+      "step": 95530
+    },
+    {
+      "epoch": 12.9,
+      "learning_rate": 1.4267199022680276e-05,
+      "loss": 2.2083,
+      "step": 95540
+    },
+    {
+      "epoch": 12.9,
+      "learning_rate": 1.4249142810773829e-05,
+      "loss": 2.2264,
+      "step": 95550
+    },
+    {
+      "epoch": 12.91,
+      "learning_rate": 1.4231097462079744e-05,
+      "loss": 2.2402,
+      "step": 95560
+    },
+    {
+      "epoch": 12.91,
+      "learning_rate": 1.4214865937495167e-05,
+      "loss": 2.2057,
+      "step": 95570
+    },
+    {
+      "epoch": 12.91,
+      "learning_rate": 1.4196841232882161e-05,
+      "loss": 2.2152,
+      "step": 95580
+    },
+    {
+      "epoch": 12.91,
+      "learning_rate": 1.4178827395666853e-05,
+      "loss": 2.2399,
+      "step": 95590
+    },
+    {
+      "epoch": 12.91,
+      "learning_rate": 1.4160824427290773e-05,
+      "loss": 2.2006,
+      "step": 95600
+    },
+    {
+      "epoch": 12.91,
+      "learning_rate": 1.414283232919452e-05,
+      "loss": 2.2183,
+      "step": 95610
+    },
+    {
+      "epoch": 12.91,
+      "learning_rate": 1.4124851102817942e-05,
+      "loss": 2.2225,
+      "step": 95620
+    },
+    {
+      "epoch": 12.92,
+      "learning_rate": 1.4106880749599903e-05,
+      "loss": 2.2176,
+      "step": 95630
+    },
+    {
+      "epoch": 12.92,
+      "learning_rate": 1.4088921270978487e-05,
+      "loss": 2.2053,
+      "step": 95640
+    },
+    {
+      "epoch": 12.92,
+      "learning_rate": 1.4070972668390878e-05,
+      "loss": 2.2646,
+      "step": 95650
+    },
+    {
+      "epoch": 12.92,
+      "learning_rate": 1.405303494327331e-05,
+      "loss": 2.2232,
+      "step": 95660
+    },
+    {
+      "epoch": 12.92,
+      "learning_rate": 1.4035108097061288e-05,
+      "loss": 2.2142,
+      "step": 95670
+    },
+    {
+      "epoch": 12.92,
+      "learning_rate": 1.4017192131189342e-05,
+      "loss": 2.2419,
+      "step": 95680
+    },
+    {
+      "epoch": 12.92,
+      "learning_rate": 1.3999287047091162e-05,
+      "loss": 2.231,
+      "step": 95690
+    },
+    {
+      "epoch": 12.92,
+      "learning_rate": 1.3981392846199585e-05,
+      "loss": 2.2128,
+      "step": 95700
+    },
+    {
+      "epoch": 12.93,
+      "learning_rate": 1.3963509529946532e-05,
+      "loss": 2.2251,
+      "step": 95710
+    },
+    {
+      "epoch": 12.93,
+      "learning_rate": 1.3945637099763173e-05,
+      "loss": 2.1967,
+      "step": 95720
+    },
+    {
+      "epoch": 12.93,
+      "learning_rate": 1.3927775557079584e-05,
+      "loss": 2.2515,
+      "step": 95730
+    },
+    {
+      "epoch": 12.93,
+      "learning_rate": 1.3909924903325154e-05,
+      "loss": 2.2066,
+      "step": 95740
+    },
+    {
+      "epoch": 12.93,
+      "learning_rate": 1.3892085139928406e-05,
+      "loss": 2.2197,
+      "step": 95750
+    },
+    {
+      "epoch": 12.93,
+      "learning_rate": 1.387425626831688e-05,
+      "loss": 2.2296,
+      "step": 95760
+    },
+    {
+      "epoch": 12.93,
+      "learning_rate": 1.3856438289917305e-05,
+      "loss": 2.2124,
+      "step": 95770
+    },
+    {
+      "epoch": 12.94,
+      "learning_rate": 1.3838631206155503e-05,
+      "loss": 2.2324,
+      "step": 95780
+    },
+    {
+      "epoch": 12.94,
+      "learning_rate": 1.3820835018456522e-05,
+      "loss": 2.2035,
+      "step": 95790
+    },
+    {
+      "epoch": 12.94,
+      "learning_rate": 1.380304972824442e-05,
+      "loss": 2.2161,
+      "step": 95800
+    },
+    {
+      "epoch": 12.94,
+      "learning_rate": 1.3785275336942425e-05,
+      "loss": 2.2176,
+      "step": 95810
+    },
+    {
+      "epoch": 12.94,
+      "learning_rate": 1.376751184597295e-05,
+      "loss": 2.2354,
+      "step": 95820
+    },
+    {
+      "epoch": 12.94,
+      "learning_rate": 1.3749759256757426e-05,
+      "loss": 2.2225,
+      "step": 95830
+    },
+    {
+      "epoch": 12.94,
+      "learning_rate": 1.3732017570716514e-05,
+      "loss": 2.2259,
+      "step": 95840
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 1.3714286789269946e-05,
+      "loss": 2.2125,
+      "step": 95850
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 1.3696566913836571e-05,
+      "loss": 2.2176,
+      "step": 95860
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 1.3678857945834454e-05,
+      "loss": 2.2234,
+      "step": 95870
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 1.3661159886680662e-05,
+      "loss": 2.235,
+      "step": 95880
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 1.3643472737791444e-05,
+      "loss": 2.2236,
+      "step": 95890
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 1.3625796500582237e-05,
+      "loss": 2.2285,
+      "step": 95900
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 1.3608131176467508e-05,
+      "loss": 2.2357,
+      "step": 95910
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 1.359047676686094e-05,
+      "loss": 2.2251,
+      "step": 95920
+    },
+    {
+      "epoch": 12.96,
+      "learning_rate": 1.3572833273175193e-05,
+      "loss": 2.2444,
+      "step": 95930
+    },
+    {
+      "epoch": 12.96,
+      "learning_rate": 1.3555200696822232e-05,
+      "loss": 2.2235,
+      "step": 95940
+    },
+    {
+      "epoch": 12.96,
+      "learning_rate": 1.3537579039213082e-05,
+      "loss": 2.2119,
+      "step": 95950
+    },
+    {
+      "epoch": 12.96,
+      "learning_rate": 1.3519968301757861e-05,
+      "loss": 2.2608,
+      "step": 95960
+    },
+    {
+      "epoch": 12.96,
+      "learning_rate": 1.3502368485865844e-05,
+      "loss": 2.2363,
+      "step": 95970
+    },
+    {
+      "epoch": 12.96,
+      "learning_rate": 1.3484779592945388e-05,
+      "loss": 2.256,
+      "step": 95980
+    },
+    {
+      "epoch": 12.96,
+      "learning_rate": 1.346720162440405e-05,
+      "loss": 2.2607,
+      "step": 95990
+    },
+    {
+      "epoch": 12.97,
+      "learning_rate": 1.3449634581648472e-05,
+      "loss": 2.2032,
+      "step": 96000
+    },
+    {
+      "epoch": 12.97,
+      "eval_loss": 2.401254177093506,
+      "eval_runtime": 1270.5885,
+      "eval_samples_per_second": 59.921,
+      "eval_steps_per_second": 4.994,
+      "step": 96000
+    },
+    {
+      "epoch": 12.97,
+      "learning_rate": 1.3432078466084395e-05,
+      "loss": 2.2359,
+      "step": 96010
+    },
+    {
+      "epoch": 12.97,
+      "learning_rate": 1.3414533279116747e-05,
+      "loss": 2.2226,
+      "step": 96020
+    },
+    {
+      "epoch": 12.97,
+      "learning_rate": 1.3396999022149535e-05,
+      "loss": 2.2261,
+      "step": 96030
+    },
+    {
+      "epoch": 12.97,
+      "learning_rate": 1.3379475696585906e-05,
+      "loss": 2.2086,
+      "step": 96040
+    },
+    {
+      "epoch": 12.97,
+      "learning_rate": 1.3361963303828156e-05,
+      "loss": 2.24,
+      "step": 96050
+    },
+    {
+      "epoch": 12.97,
+      "learning_rate": 1.3344461845277626e-05,
+      "loss": 2.2254,
+      "step": 96060
+    },
+    {
+      "epoch": 12.97,
+      "learning_rate": 1.33269713223349e-05,
+      "loss": 2.1811,
+      "step": 96070
+    },
+    {
+      "epoch": 12.98,
+      "learning_rate": 1.3309491736399591e-05,
+      "loss": 2.2479,
+      "step": 96080
+    },
+    {
+      "epoch": 12.98,
+      "learning_rate": 1.3292023088870463e-05,
+      "loss": 2.2206,
+      "step": 96090
+    },
+    {
+      "epoch": 12.98,
+      "learning_rate": 1.3274565381145447e-05,
+      "loss": 2.2302,
+      "step": 96100
+    },
+    {
+      "epoch": 12.98,
+      "learning_rate": 1.325711861462151e-05,
+      "loss": 2.2533,
+      "step": 96110
+    },
+    {
+      "epoch": 12.98,
+      "learning_rate": 1.3239682790694899e-05,
+      "loss": 2.1956,
+      "step": 96120
+    },
+    {
+      "epoch": 12.98,
+      "learning_rate": 1.3222257910760753e-05,
+      "loss": 2.2296,
+      "step": 96130
+    },
+    {
+      "epoch": 12.98,
+      "learning_rate": 1.3204843976213519e-05,
+      "loss": 2.2355,
+      "step": 96140
+    },
+    {
+      "epoch": 12.99,
+      "learning_rate": 1.318744098844675e-05,
+      "loss": 2.2399,
+      "step": 96150
+    },
+    {
+      "epoch": 12.99,
+      "learning_rate": 1.3170048948853084e-05,
+      "loss": 2.21,
+      "step": 96160
+    },
+    {
+      "epoch": 12.99,
+      "learning_rate": 1.3152667858824239e-05,
+      "loss": 2.2089,
+      "step": 96170
+    },
+    {
+      "epoch": 12.99,
+      "learning_rate": 1.313529771975112e-05,
+      "loss": 2.2111,
+      "step": 96180
+    },
+    {
+      "epoch": 12.99,
+      "learning_rate": 1.311793853302378e-05,
+      "loss": 2.2256,
+      "step": 96190
+    },
+    {
+      "epoch": 12.99,
+      "learning_rate": 1.3100590300031327e-05,
+      "loss": 2.2305,
+      "step": 96200
+    },
+    {
+      "epoch": 12.99,
+      "learning_rate": 1.3083253022161982e-05,
+      "loss": 2.2326,
+      "step": 96210
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 1.3065926700803203e-05,
+      "loss": 2.2231,
+      "step": 96220
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 1.304861133734143e-05,
+      "loss": 2.2583,
+      "step": 96230
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 1.3031306933162373e-05,
+      "loss": 2.2178,
+      "step": 96240
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 1.301401348965072e-05,
+      "loss": 2.2307,
+      "step": 96250
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 1.2996731008190336e-05,
+      "loss": 2.1926,
+      "step": 96260
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 1.297945949016428e-05,
+      "loss": 2.1905,
+      "step": 96270
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 1.2962198936954626e-05,
+      "loss": 2.238,
+      "step": 96280
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 1.2944949349942624e-05,
+      "loss": 2.2113,
+      "step": 96290
+    },
+    {
+      "epoch": 13.01,
+      "learning_rate": 1.2927710730508684e-05,
+      "loss": 2.1695,
+      "step": 96300
+    },
+    {
+      "epoch": 13.01,
+      "learning_rate": 1.291048308003222e-05,
+      "loss": 2.2097,
+      "step": 96310
+    },
+    {
+      "epoch": 13.01,
+      "learning_rate": 1.2893266399891943e-05,
+      "loss": 2.2106,
+      "step": 96320
+    },
+    {
+      "epoch": 13.01,
+      "learning_rate": 1.2876060691465473e-05,
+      "loss": 2.2055,
+      "step": 96330
+    },
+    {
+      "epoch": 13.01,
+      "learning_rate": 1.2858865956129704e-05,
+      "loss": 2.1814,
+      "step": 96340
+    },
+    {
+      "epoch": 13.01,
+      "learning_rate": 1.2841682195260671e-05,
+      "loss": 2.1985,
+      "step": 96350
+    },
+    {
+      "epoch": 13.01,
+      "learning_rate": 1.2824509410233425e-05,
+      "loss": 2.2073,
+      "step": 96360
+    },
+    {
+      "epoch": 13.02,
+      "learning_rate": 1.2807347602422196e-05,
+      "loss": 2.223,
+      "step": 96370
+    },
+    {
+      "epoch": 13.02,
+      "learning_rate": 1.279019677320029e-05,
+      "loss": 2.1933,
+      "step": 96380
+    },
+    {
+      "epoch": 13.02,
+      "learning_rate": 1.2773056923940223e-05,
+      "loss": 2.1855,
+      "step": 96390
+    },
+    {
+      "epoch": 13.02,
+      "learning_rate": 1.2755928056013565e-05,
+      "loss": 2.2056,
+      "step": 96400
+    },
+    {
+      "epoch": 13.02,
+      "learning_rate": 1.2738810170791004e-05,
+      "loss": 2.1966,
+      "step": 96410
+    },
+    {
+      "epoch": 13.02,
+      "learning_rate": 1.2721703269642392e-05,
+      "loss": 2.2055,
+      "step": 96420
+    },
+    {
+      "epoch": 13.02,
+      "learning_rate": 1.2704607353936635e-05,
+      "loss": 2.209,
+      "step": 96430
+    },
+    {
+      "epoch": 13.02,
+      "learning_rate": 1.2687522425041857e-05,
+      "loss": 2.1966,
+      "step": 96440
+    },
+    {
+      "epoch": 13.03,
+      "learning_rate": 1.267044848432523e-05,
+      "loss": 2.2037,
+      "step": 96450
+    },
+    {
+      "epoch": 13.03,
+      "learning_rate": 1.2653385533153027e-05,
+      "loss": 2.1976,
+      "step": 96460
+    },
+    {
+      "epoch": 13.03,
+      "learning_rate": 1.2636333572890722e-05,
+      "loss": 2.2281,
+      "step": 96470
+    },
+    {
+      "epoch": 13.03,
+      "learning_rate": 1.2619292604902875e-05,
+      "loss": 2.1796,
+      "step": 96480
+    },
+    {
+      "epoch": 13.03,
+      "learning_rate": 1.2602262630553112e-05,
+      "loss": 2.2005,
+      "step": 96490
+    },
+    {
+      "epoch": 13.03,
+      "learning_rate": 1.2585243651204258e-05,
+      "loss": 2.2076,
+      "step": 96500
+    },
+    {
+      "epoch": 13.03,
+      "learning_rate": 1.2568235668218207e-05,
+      "loss": 2.2003,
+      "step": 96510
+    },
+    {
+      "epoch": 13.04,
+      "learning_rate": 1.2551238682956055e-05,
+      "loss": 2.2062,
+      "step": 96520
+    },
+    {
+      "epoch": 13.04,
+      "learning_rate": 1.2534252696777847e-05,
+      "loss": 2.2026,
+      "step": 96530
+    },
+    {
+      "epoch": 13.04,
+      "learning_rate": 1.2517277711042927e-05,
+      "loss": 2.1971,
+      "step": 96540
+    },
+    {
+      "epoch": 13.04,
+      "learning_rate": 1.2500313727109645e-05,
+      "loss": 2.1913,
+      "step": 96550
+    },
+    {
+      "epoch": 13.04,
+      "learning_rate": 1.248336074633558e-05,
+      "loss": 2.1996,
+      "step": 96560
+    },
+    {
+      "epoch": 13.04,
+      "learning_rate": 1.2466418770077313e-05,
+      "loss": 2.1909,
+      "step": 96570
+    },
+    {
+      "epoch": 13.04,
+      "learning_rate": 1.2449487799690578e-05,
+      "loss": 2.1935,
+      "step": 96580
+    },
+    {
+      "epoch": 13.05,
+      "learning_rate": 1.2432567836530288e-05,
+      "loss": 2.183,
+      "step": 96590
+    },
+    {
+      "epoch": 13.05,
+      "learning_rate": 1.2415658881950414e-05,
+      "loss": 2.2118,
+      "step": 96600
+    },
+    {
+      "epoch": 13.05,
+      "learning_rate": 1.2398760937304036e-05,
+      "loss": 2.21,
+      "step": 96610
+    },
+    {
+      "epoch": 13.05,
+      "learning_rate": 1.2381874003943443e-05,
+      "loss": 2.2043,
+      "step": 96620
+    },
+    {
+      "epoch": 13.05,
+      "learning_rate": 1.2364998083219902e-05,
+      "loss": 2.2018,
+      "step": 96630
+    },
+    {
+      "epoch": 13.05,
+      "learning_rate": 1.2348133176483965e-05,
+      "loss": 2.1986,
+      "step": 96640
+    },
+    {
+      "epoch": 13.05,
+      "learning_rate": 1.2331279285085172e-05,
+      "loss": 2.2115,
+      "step": 96650
+    },
+    {
+      "epoch": 13.05,
+      "learning_rate": 1.2314436410372191e-05,
+      "loss": 2.2182,
+      "step": 96660
+    },
+    {
+      "epoch": 13.06,
+      "learning_rate": 1.2297604553692896e-05,
+      "loss": 2.2184,
+      "step": 96670
+    },
+    {
+      "epoch": 13.06,
+      "learning_rate": 1.228078371639421e-05,
+      "loss": 2.2278,
+      "step": 96680
+    },
+    {
+      "epoch": 13.06,
+      "learning_rate": 1.2263973899822154e-05,
+      "loss": 2.2251,
+      "step": 96690
+    },
+    {
+      "epoch": 13.06,
+      "learning_rate": 1.2247175105321954e-05,
+      "loss": 2.1951,
+      "step": 96700
+    },
+    {
+      "epoch": 13.06,
+      "learning_rate": 1.2230387334237868e-05,
+      "loss": 2.1906,
+      "step": 96710
+    },
+    {
+      "epoch": 13.06,
+      "learning_rate": 1.221361058791337e-05,
+      "loss": 2.2135,
+      "step": 96720
+    },
+    {
+      "epoch": 13.06,
+      "learning_rate": 1.219684486769087e-05,
+      "loss": 2.2049,
+      "step": 96730
+    },
+    {
+      "epoch": 13.07,
+      "learning_rate": 1.2180090174912127e-05,
+      "loss": 2.1704,
+      "step": 96740
+    },
+    {
+      "epoch": 13.07,
+      "learning_rate": 1.216334651091782e-05,
+      "loss": 2.1809,
+      "step": 96750
+    },
+    {
+      "epoch": 13.07,
+      "learning_rate": 1.2146613877047896e-05,
+      "loss": 2.2163,
+      "step": 96760
+    },
+    {
+      "epoch": 13.07,
+      "learning_rate": 1.2129892274641328e-05,
+      "loss": 2.1892,
+      "step": 96770
+    },
+    {
+      "epoch": 13.07,
+      "learning_rate": 1.2113181705036201e-05,
+      "loss": 2.2093,
+      "step": 96780
+    },
+    {
+      "epoch": 13.07,
+      "learning_rate": 1.209648216956981e-05,
+      "loss": 2.1915,
+      "step": 96790
+    },
+    {
+      "epoch": 13.07,
+      "learning_rate": 1.2079793669578469e-05,
+      "loss": 2.189,
+      "step": 96800
+    },
+    {
+      "epoch": 13.07,
+      "learning_rate": 1.2063116206397611e-05,
+      "loss": 2.2267,
+      "step": 96810
+    },
+    {
+      "epoch": 13.08,
+      "learning_rate": 1.2046449781361883e-05,
+      "loss": 2.2041,
+      "step": 96820
+    },
+    {
+      "epoch": 13.08,
+      "learning_rate": 1.2029794395804937e-05,
+      "loss": 2.2201,
+      "step": 96830
+    },
+    {
+      "epoch": 13.08,
+      "learning_rate": 1.201315005105964e-05,
+      "loss": 2.1665,
+      "step": 96840
+    },
+    {
+      "epoch": 13.08,
+      "learning_rate": 1.1996516748457873e-05,
+      "loss": 2.1785,
+      "step": 96850
+    },
+    {
+      "epoch": 13.08,
+      "learning_rate": 1.1979894489330694e-05,
+      "loss": 2.2242,
+      "step": 96860
+    },
+    {
+      "epoch": 13.08,
+      "learning_rate": 1.19632832750083e-05,
+      "loss": 2.2418,
+      "step": 96870
+    },
+    {
+      "epoch": 13.08,
+      "learning_rate": 1.1946683106819966e-05,
+      "loss": 2.2096,
+      "step": 96880
+    },
+    {
+      "epoch": 13.09,
+      "learning_rate": 1.1930093986094041e-05,
+      "loss": 2.2091,
+      "step": 96890
+    },
+    {
+      "epoch": 13.09,
+      "learning_rate": 1.19135159141581e-05,
+      "loss": 2.1805,
+      "step": 96900
+    },
+    {
+      "epoch": 13.09,
+      "learning_rate": 1.1896948892338731e-05,
+      "loss": 2.1938,
+      "step": 96910
+    },
+    {
+      "epoch": 13.09,
+      "learning_rate": 1.188039292196174e-05,
+      "loss": 2.2056,
+      "step": 96920
+    },
+    {
+      "epoch": 13.09,
+      "learning_rate": 1.18638480043519e-05,
+      "loss": 2.2127,
+      "step": 96930
+    },
+    {
+      "epoch": 13.09,
+      "learning_rate": 1.1847314140833254e-05,
+      "loss": 2.1924,
+      "step": 96940
+    },
+    {
+      "epoch": 13.09,
+      "learning_rate": 1.1830791332728856e-05,
+      "loss": 2.2031,
+      "step": 96950
+    },
+    {
+      "epoch": 13.1,
+      "learning_rate": 1.1814279581360935e-05,
+      "loss": 2.1959,
+      "step": 96960
+    },
+    {
+      "epoch": 13.1,
+      "learning_rate": 1.1797778888050818e-05,
+      "loss": 2.217,
+      "step": 96970
+    },
+    {
+      "epoch": 13.1,
+      "learning_rate": 1.1781289254118914e-05,
+      "loss": 2.2072,
+      "step": 96980
+    },
+    {
+      "epoch": 13.1,
+      "learning_rate": 1.1764810680884818e-05,
+      "loss": 2.1911,
+      "step": 96990
+    },
+    {
+      "epoch": 13.1,
+      "learning_rate": 1.1748343169667174e-05,
+      "loss": 2.2097,
+      "step": 97000
+    },
+    {
+      "epoch": 13.1,
+      "eval_loss": 2.4041683673858643,
+      "eval_runtime": 1269.5496,
+      "eval_samples_per_second": 59.97,
+      "eval_steps_per_second": 4.998,
+      "step": 97000
+    },
+    {
+      "epoch": 13.1,
+      "learning_rate": 1.1731886721783729e-05,
+      "loss": 2.1956,
+      "step": 97010
+    },
+    {
+      "epoch": 13.1,
+      "learning_rate": 1.1715441338551446e-05,
+      "loss": 2.1909,
+      "step": 97020
+    },
+    {
+      "epoch": 13.1,
+      "learning_rate": 1.1699007021286272e-05,
+      "loss": 2.1952,
+      "step": 97030
+    },
+    {
+      "epoch": 13.11,
+      "learning_rate": 1.1682583771303405e-05,
+      "loss": 2.209,
+      "step": 97040
+    },
+    {
+      "epoch": 13.11,
+      "learning_rate": 1.1666171589917046e-05,
+      "loss": 2.1874,
+      "step": 97050
+    },
+    {
+      "epoch": 13.11,
+      "learning_rate": 1.1649770478440507e-05,
+      "loss": 2.2329,
+      "step": 97060
+    },
+    {
+      "epoch": 13.11,
+      "learning_rate": 1.163338043818634e-05,
+      "loss": 2.2008,
+      "step": 97070
+    },
+    {
+      "epoch": 13.11,
+      "learning_rate": 1.161700147046608e-05,
+      "loss": 2.2056,
+      "step": 97080
+    },
+    {
+      "epoch": 13.11,
+      "learning_rate": 1.1600633576590412e-05,
+      "loss": 2.1786,
+      "step": 97090
+    },
+    {
+      "epoch": 13.11,
+      "learning_rate": 1.1584276757869204e-05,
+      "loss": 2.1981,
+      "step": 97100
+    },
+    {
+      "epoch": 13.12,
+      "learning_rate": 1.1567931015611293e-05,
+      "loss": 2.2216,
+      "step": 97110
+    },
+    {
+      "epoch": 13.12,
+      "learning_rate": 1.155159635112483e-05,
+      "loss": 2.2227,
+      "step": 97120
+    },
+    {
+      "epoch": 13.12,
+      "learning_rate": 1.1535272765716857e-05,
+      "loss": 2.1962,
+      "step": 97130
+    },
+    {
+      "epoch": 13.12,
+      "learning_rate": 1.1518960260693693e-05,
+      "loss": 2.2304,
+      "step": 97140
+    },
+    {
+      "epoch": 13.12,
+      "learning_rate": 1.1502658837360696e-05,
+      "loss": 2.2086,
+      "step": 97150
+    },
+    {
+      "epoch": 13.12,
+      "learning_rate": 1.1486368497022386e-05,
+      "loss": 2.1812,
+      "step": 97160
+    },
+    {
+      "epoch": 13.12,
+      "learning_rate": 1.1470089240982355e-05,
+      "loss": 2.1969,
+      "step": 97170
+    },
+    {
+      "epoch": 13.12,
+      "learning_rate": 1.1453821070543295e-05,
+      "loss": 2.2183,
+      "step": 97180
+    },
+    {
+      "epoch": 13.13,
+      "learning_rate": 1.143756398700708e-05,
+      "loss": 2.2113,
+      "step": 97190
+    },
+    {
+      "epoch": 13.13,
+      "learning_rate": 1.1421317991674623e-05,
+      "loss": 2.197,
+      "step": 97200
+    },
+    {
+      "epoch": 13.13,
+      "learning_rate": 1.1405083085845961e-05,
+      "loss": 2.1736,
+      "step": 97210
+    },
+    {
+      "epoch": 13.13,
+      "learning_rate": 1.1388859270820328e-05,
+      "loss": 2.1797,
+      "step": 97220
+    },
+    {
+      "epoch": 13.13,
+      "learning_rate": 1.1372646547895931e-05,
+      "loss": 2.2025,
+      "step": 97230
+    },
+    {
+      "epoch": 13.13,
+      "learning_rate": 1.1356444918370234e-05,
+      "loss": 2.2154,
+      "step": 97240
+    },
+    {
+      "epoch": 13.13,
+      "learning_rate": 1.1340254383539698e-05,
+      "loss": 2.1828,
+      "step": 97250
+    },
+    {
+      "epoch": 13.14,
+      "learning_rate": 1.1324074944699923e-05,
+      "loss": 2.2228,
+      "step": 97260
+    },
+    {
+      "epoch": 13.14,
+      "learning_rate": 1.1307906603145689e-05,
+      "loss": 2.1822,
+      "step": 97270
+    },
+    {
+      "epoch": 13.14,
+      "learning_rate": 1.1291749360170826e-05,
+      "loss": 2.1942,
+      "step": 97280
+    },
+    {
+      "epoch": 13.14,
+      "learning_rate": 1.1275603217068235e-05,
+      "loss": 2.2061,
+      "step": 97290
+    },
+    {
+      "epoch": 13.14,
+      "learning_rate": 1.1259468175130049e-05,
+      "loss": 2.1987,
+      "step": 97300
+    },
+    {
+      "epoch": 13.14,
+      "learning_rate": 1.12433442356474e-05,
+      "loss": 2.1863,
+      "step": 97310
+    },
+    {
+      "epoch": 13.14,
+      "learning_rate": 1.122723139991064e-05,
+      "loss": 2.2207,
+      "step": 97320
+    },
+    {
+      "epoch": 13.15,
+      "learning_rate": 1.1211129669209091e-05,
+      "loss": 2.2236,
+      "step": 97330
+    },
+    {
+      "epoch": 13.15,
+      "learning_rate": 1.11950390448313e-05,
+      "loss": 2.1885,
+      "step": 97340
+    },
+    {
+      "epoch": 13.15,
+      "learning_rate": 1.1178959528064873e-05,
+      "loss": 2.2132,
+      "step": 97350
+    },
+    {
+      "epoch": 13.15,
+      "learning_rate": 1.1162891120196582e-05,
+      "loss": 2.2034,
+      "step": 97360
+    },
+    {
+      "epoch": 13.15,
+      "learning_rate": 1.114683382251223e-05,
+      "loss": 2.1783,
+      "step": 97370
+    },
+    {
+      "epoch": 13.15,
+      "learning_rate": 1.113078763629679e-05,
+      "loss": 2.2299,
+      "step": 97380
+    },
+    {
+      "epoch": 13.15,
+      "learning_rate": 1.1114752562834333e-05,
+      "loss": 2.2016,
+      "step": 97390
+    },
+    {
+      "epoch": 13.15,
+      "learning_rate": 1.1098728603408036e-05,
+      "loss": 2.2128,
+      "step": 97400
+    },
+    {
+      "epoch": 13.16,
+      "learning_rate": 1.1082715759300154e-05,
+      "loss": 2.2036,
+      "step": 97410
+    },
+    {
+      "epoch": 13.16,
+      "learning_rate": 1.1066714031792145e-05,
+      "loss": 2.2044,
+      "step": 97420
+    },
+    {
+      "epoch": 13.16,
+      "learning_rate": 1.105072342216447e-05,
+      "loss": 2.1874,
+      "step": 97430
+    },
+    {
+      "epoch": 13.16,
+      "learning_rate": 1.103474393169677e-05,
+      "loss": 2.1992,
+      "step": 97440
+    },
+    {
+      "epoch": 13.16,
+      "learning_rate": 1.1018775561667791e-05,
+      "loss": 2.1804,
+      "step": 97450
+    },
+    {
+      "epoch": 13.16,
+      "learning_rate": 1.100281831335531e-05,
+      "loss": 2.2247,
+      "step": 97460
+    },
+    {
+      "epoch": 13.16,
+      "learning_rate": 1.0986872188036339e-05,
+      "loss": 2.2081,
+      "step": 97470
+    },
+    {
+      "epoch": 13.17,
+      "learning_rate": 1.0970937186986922e-05,
+      "loss": 2.2026,
+      "step": 97480
+    },
+    {
+      "epoch": 13.17,
+      "learning_rate": 1.0955013311482192e-05,
+      "loss": 2.212,
+      "step": 97490
+    },
+    {
+      "epoch": 13.17,
+      "learning_rate": 1.0939100562796493e-05,
+      "loss": 2.2071,
+      "step": 97500
+    },
+    {
+      "epoch": 13.17,
+      "learning_rate": 1.0923198942203144e-05,
+      "loss": 2.1814,
+      "step": 97510
+    },
+    {
+      "epoch": 13.17,
+      "learning_rate": 1.0907308450974722e-05,
+      "loss": 2.2042,
+      "step": 97520
+    },
+    {
+      "epoch": 13.17,
+      "learning_rate": 1.0891429090382747e-05,
+      "loss": 2.214,
+      "step": 97530
+    },
+    {
+      "epoch": 13.17,
+      "learning_rate": 1.0875560861697985e-05,
+      "loss": 2.2327,
+      "step": 97540
+    },
+    {
+      "epoch": 13.17,
+      "learning_rate": 1.0859703766190253e-05,
+      "loss": 2.2025,
+      "step": 97550
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 1.0843857805128486e-05,
+      "loss": 2.2154,
+      "step": 97560
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 1.0828022979780754e-05,
+      "loss": 2.1975,
+      "step": 97570
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 1.0812199291414142e-05,
+      "loss": 2.2298,
+      "step": 97580
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 1.0796386741294987e-05,
+      "loss": 2.2021,
+      "step": 97590
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 1.0780585330688612e-05,
+      "loss": 2.2064,
+      "step": 97600
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 1.0764795060859504e-05,
+      "loss": 2.2031,
+      "step": 97610
+    },
+    {
+      "epoch": 13.18,
+      "learning_rate": 1.074901593307127e-05,
+      "loss": 2.1919,
+      "step": 97620
+    },
+    {
+      "epoch": 13.19,
+      "learning_rate": 1.0733247948586566e-05,
+      "loss": 2.2038,
+      "step": 97630
+    },
+    {
+      "epoch": 13.19,
+      "learning_rate": 1.0717491108667248e-05,
+      "loss": 2.2185,
+      "step": 97640
+    },
+    {
+      "epoch": 13.19,
+      "learning_rate": 1.0701745414574209e-05,
+      "loss": 2.2328,
+      "step": 97650
+    },
+    {
+      "epoch": 13.19,
+      "learning_rate": 1.0686010867567424e-05,
+      "loss": 2.2361,
+      "step": 97660
+    },
+    {
+      "epoch": 13.19,
+      "learning_rate": 1.0670287468906086e-05,
+      "loss": 2.1944,
+      "step": 97670
+    },
+    {
+      "epoch": 13.19,
+      "learning_rate": 1.0656145942986177e-05,
+      "loss": 2.2046,
+      "step": 97680
+    },
+    {
+      "epoch": 13.19,
+      "learning_rate": 1.0640443729646818e-05,
+      "loss": 2.1981,
+      "step": 97690
+    },
+    {
+      "epoch": 13.2,
+      "learning_rate": 1.0624752668299308e-05,
+      "loss": 2.2069,
+      "step": 97700
+    },
+    {
+      "epoch": 13.2,
+      "learning_rate": 1.0609072760199356e-05,
+      "loss": 2.194,
+      "step": 97710
+    },
+    {
+      "epoch": 13.2,
+      "learning_rate": 1.0593404006601575e-05,
+      "loss": 2.2102,
+      "step": 97720
+    },
+    {
+      "epoch": 13.2,
+      "learning_rate": 1.0577746408759946e-05,
+      "loss": 2.2007,
+      "step": 97730
+    },
+    {
+      "epoch": 13.2,
+      "learning_rate": 1.0562099967927384e-05,
+      "loss": 2.1784,
+      "step": 97740
+    },
+    {
+      "epoch": 13.2,
+      "learning_rate": 1.0546464685356004e-05,
+      "loss": 2.2163,
+      "step": 97750
+    },
+    {
+      "epoch": 13.2,
+      "learning_rate": 1.0530840562296971e-05,
+      "loss": 2.2025,
+      "step": 97760
+    },
+    {
+      "epoch": 13.2,
+      "learning_rate": 1.0515227600000536e-05,
+      "loss": 2.2133,
+      "step": 97770
+    },
+    {
+      "epoch": 13.21,
+      "learning_rate": 1.0499625799716183e-05,
+      "loss": 2.2313,
+      "step": 97780
+    },
+    {
+      "epoch": 13.21,
+      "learning_rate": 1.0484035162692362e-05,
+      "loss": 2.1915,
+      "step": 97790
+    },
+    {
+      "epoch": 13.21,
+      "learning_rate": 1.0468455690176659e-05,
+      "loss": 2.2054,
+      "step": 97800
+    },
+    {
+      "epoch": 13.21,
+      "learning_rate": 1.0452887383415876e-05,
+      "loss": 2.1988,
+      "step": 97810
+    },
+    {
+      "epoch": 13.21,
+      "learning_rate": 1.043733024365575e-05,
+      "loss": 2.2011,
+      "step": 97820
+    },
+    {
+      "epoch": 13.21,
+      "learning_rate": 1.0421784272141303e-05,
+      "loss": 2.2051,
+      "step": 97830
+    },
+    {
+      "epoch": 13.21,
+      "learning_rate": 1.0406249470116473e-05,
+      "loss": 2.2106,
+      "step": 97840
+    },
+    {
+      "epoch": 13.22,
+      "learning_rate": 1.0390725838824465e-05,
+      "loss": 2.2154,
+      "step": 97850
+    },
+    {
+      "epoch": 13.22,
+      "learning_rate": 1.0375213379507536e-05,
+      "loss": 2.237,
+      "step": 97860
+    },
+    {
+      "epoch": 13.22,
+      "learning_rate": 1.0359712093407024e-05,
+      "loss": 2.208,
+      "step": 97870
+    },
+    {
+      "epoch": 13.22,
+      "learning_rate": 1.034422198176339e-05,
+      "loss": 2.1729,
+      "step": 97880
+    },
+    {
+      "epoch": 13.22,
+      "learning_rate": 1.0328743045816174e-05,
+      "loss": 2.2057,
+      "step": 97890
+    },
+    {
+      "epoch": 13.22,
+      "learning_rate": 1.0313275286804068e-05,
+      "loss": 2.2039,
+      "step": 97900
+    },
+    {
+      "epoch": 13.22,
+      "learning_rate": 1.0297818705964933e-05,
+      "loss": 2.1946,
+      "step": 97910
+    },
+    {
+      "epoch": 13.22,
+      "learning_rate": 1.0282373304535513e-05,
+      "loss": 2.2165,
+      "step": 97920
+    },
+    {
+      "epoch": 13.23,
+      "learning_rate": 1.0266939083751902e-05,
+      "loss": 2.2333,
+      "step": 97930
+    },
+    {
+      "epoch": 13.23,
+      "learning_rate": 1.0251516044849128e-05,
+      "loss": 2.1843,
+      "step": 97940
+    },
+    {
+      "epoch": 13.23,
+      "learning_rate": 1.023610418906144e-05,
+      "loss": 2.2095,
+      "step": 97950
+    },
+    {
+      "epoch": 13.23,
+      "learning_rate": 1.022070351762213e-05,
+      "loss": 2.2169,
+      "step": 97960
+    },
+    {
+      "epoch": 13.23,
+      "learning_rate": 1.0205314031763583e-05,
+      "loss": 2.1808,
+      "step": 97970
+    },
+    {
+      "epoch": 13.23,
+      "learning_rate": 1.0189935732717363e-05,
+      "loss": 2.2231,
+      "step": 97980
+    },
+    {
+      "epoch": 13.23,
+      "learning_rate": 1.0174568621714052e-05,
+      "loss": 2.2224,
+      "step": 97990
+    },
+    {
+      "epoch": 13.24,
+      "learning_rate": 1.0159212699983366e-05,
+      "loss": 2.2055,
+      "step": 98000
+    },
+    {
+      "epoch": 13.24,
+      "eval_loss": 2.4040699005126953,
+      "eval_runtime": 1269.6142,
+      "eval_samples_per_second": 59.967,
+      "eval_steps_per_second": 4.998,
+      "step": 98000
+    },
+    {
+      "epoch": 13.24,
+      "learning_rate": 1.0143867968754188e-05,
+      "loss": 2.2126,
+      "step": 98010
+    },
+    {
+      "epoch": 13.24,
+      "learning_rate": 1.012853442925437e-05,
+      "loss": 2.1931,
+      "step": 98020
+    },
+    {
+      "epoch": 13.24,
+      "learning_rate": 1.0113212082711064e-05,
+      "loss": 2.2014,
+      "step": 98030
+    },
+    {
+      "epoch": 13.24,
+      "learning_rate": 1.009790093035029e-05,
+      "loss": 2.1955,
+      "step": 98040
+    },
+    {
+      "epoch": 13.24,
+      "learning_rate": 1.0082600973397348e-05,
+      "loss": 2.225,
+      "step": 98050
+    },
+    {
+      "epoch": 13.24,
+      "learning_rate": 1.0067312213076629e-05,
+      "loss": 2.2284,
+      "step": 98060
+    },
+    {
+      "epoch": 13.25,
+      "learning_rate": 1.0052034650611534e-05,
+      "loss": 2.1989,
+      "step": 98070
+    },
+    {
+      "epoch": 13.25,
+      "learning_rate": 1.003676828722464e-05,
+      "loss": 2.1932,
+      "step": 98080
+    },
+    {
+      "epoch": 13.25,
+      "learning_rate": 1.0021513124137597e-05,
+      "loss": 2.1943,
+      "step": 98090
+    },
+    {
+      "epoch": 13.25,
+      "learning_rate": 1.0006269162571167e-05,
+      "loss": 2.2094,
+      "step": 98100
+    },
+    {
+      "epoch": 13.25,
+      "learning_rate": 9.99103640374529e-06,
+      "loss": 2.1887,
+      "step": 98110
+    },
+    {
+      "epoch": 13.25,
+      "learning_rate": 9.975814848878837e-06,
+      "loss": 2.2046,
+      "step": 98120
+    },
+    {
+      "epoch": 13.25,
+      "learning_rate": 9.960604499189955e-06,
+      "loss": 2.2195,
+      "step": 98130
+    },
+    {
+      "epoch": 13.25,
+      "learning_rate": 9.945405355895785e-06,
+      "loss": 2.1998,
+      "step": 98140
+    },
+    {
+      "epoch": 13.26,
+      "learning_rate": 9.930217420212639e-06,
+      "loss": 2.186,
+      "step": 98150
+    },
+    {
+      "epoch": 13.26,
+      "learning_rate": 9.915040693355892e-06,
+      "loss": 2.1855,
+      "step": 98160
+    },
+    {
+      "epoch": 13.26,
+      "learning_rate": 9.899875176540023e-06,
+      "loss": 2.2076,
+      "step": 98170
+    },
+    {
+      "epoch": 13.26,
+      "learning_rate": 9.884720870978647e-06,
+      "loss": 2.19,
+      "step": 98180
+    },
+    {
+      "epoch": 13.26,
+      "learning_rate": 9.869577777884457e-06,
+      "loss": 2.2035,
+      "step": 98190
+    },
+    {
+      "epoch": 13.26,
+      "learning_rate": 9.854445898469204e-06,
+      "loss": 2.2067,
+      "step": 98200
+    },
+    {
+      "epoch": 13.26,
+      "learning_rate": 9.839325233943852e-06,
+      "loss": 2.1957,
+      "step": 98210
+    },
+    {
+      "epoch": 13.27,
+      "learning_rate": 9.824215785518363e-06,
+      "loss": 2.1803,
+      "step": 98220
+    },
+    {
+      "epoch": 13.27,
+      "learning_rate": 9.809117554401908e-06,
+      "loss": 2.1956,
+      "step": 98230
+    },
+    {
+      "epoch": 13.27,
+      "learning_rate": 9.794030541802584e-06,
+      "loss": 2.1995,
+      "step": 98240
+    },
+    {
+      "epoch": 13.27,
+      "learning_rate": 9.778954748927775e-06,
+      "loss": 2.1806,
+      "step": 98250
+    },
+    {
+      "epoch": 13.27,
+      "learning_rate": 9.763890176983901e-06,
+      "loss": 2.1953,
+      "step": 98260
+    },
+    {
+      "epoch": 13.27,
+      "learning_rate": 9.748836827176465e-06,
+      "loss": 2.2072,
+      "step": 98270
+    },
+    {
+      "epoch": 13.27,
+      "learning_rate": 9.733794700710067e-06,
+      "loss": 2.2062,
+      "step": 98280
+    },
+    {
+      "epoch": 13.27,
+      "learning_rate": 9.718763798788415e-06,
+      "loss": 2.193,
+      "step": 98290
+    },
+    {
+      "epoch": 13.28,
+      "learning_rate": 9.703744122614343e-06,
+      "loss": 2.2153,
+      "step": 98300
+    },
+    {
+      "epoch": 13.28,
+      "learning_rate": 9.688735673389824e-06,
+      "loss": 2.2049,
+      "step": 98310
+    },
+    {
+      "epoch": 13.28,
+      "learning_rate": 9.6737384523158e-06,
+      "loss": 2.2114,
+      "step": 98320
+    },
+    {
+      "epoch": 13.28,
+      "learning_rate": 9.658752460592439e-06,
+      "loss": 2.2023,
+      "step": 98330
+    },
+    {
+      "epoch": 13.28,
+      "learning_rate": 9.643777699418953e-06,
+      "loss": 2.2146,
+      "step": 98340
+    },
+    {
+      "epoch": 13.28,
+      "learning_rate": 9.628814169993682e-06,
+      "loss": 2.1882,
+      "step": 98350
+    },
+    {
+      "epoch": 13.28,
+      "learning_rate": 9.613861873514066e-06,
+      "loss": 2.1742,
+      "step": 98360
+    },
+    {
+      "epoch": 13.29,
+      "learning_rate": 9.598920811176586e-06,
+      "loss": 2.203,
+      "step": 98370
+    },
+    {
+      "epoch": 13.29,
+      "learning_rate": 9.583990984176932e-06,
+      "loss": 2.229,
+      "step": 98380
+    },
+    {
+      "epoch": 13.29,
+      "learning_rate": 9.569072393709814e-06,
+      "loss": 2.224,
+      "step": 98390
+    },
+    {
+      "epoch": 13.29,
+      "learning_rate": 9.55416504096903e-06,
+      "loss": 2.2064,
+      "step": 98400
+    },
+    {
+      "epoch": 13.29,
+      "learning_rate": 9.539268927147576e-06,
+      "loss": 2.1951,
+      "step": 98410
+    },
+    {
+      "epoch": 13.29,
+      "learning_rate": 9.524384053437428e-06,
+      "loss": 2.2224,
+      "step": 98420
+    },
+    {
+      "epoch": 13.29,
+      "learning_rate": 9.509510421029787e-06,
+      "loss": 2.199,
+      "step": 98430
+    },
+    {
+      "epoch": 13.3,
+      "learning_rate": 9.494648031114798e-06,
+      "loss": 2.1838,
+      "step": 98440
+    },
+    {
+      "epoch": 13.3,
+      "learning_rate": 9.479796884881846e-06,
+      "loss": 2.1865,
+      "step": 98450
+    },
+    {
+      "epoch": 13.3,
+      "learning_rate": 9.464956983519411e-06,
+      "loss": 2.1849,
+      "step": 98460
+    },
+    {
+      "epoch": 13.3,
+      "learning_rate": 9.450128328214962e-06,
+      "loss": 2.2035,
+      "step": 98470
+    },
+    {
+      "epoch": 13.3,
+      "learning_rate": 9.435310920155165e-06,
+      "loss": 2.2059,
+      "step": 98480
+    },
+    {
+      "epoch": 13.3,
+      "learning_rate": 9.420504760525721e-06,
+      "loss": 2.1655,
+      "step": 98490
+    },
+    {
+      "epoch": 13.3,
+      "learning_rate": 9.405709850511517e-06,
+      "loss": 2.1981,
+      "step": 98500
+    },
+    {
+      "epoch": 13.3,
+      "learning_rate": 9.390926191296472e-06,
+      "loss": 2.2255,
+      "step": 98510
+    },
+    {
+      "epoch": 13.31,
+      "learning_rate": 9.376153784063573e-06,
+      "loss": 2.1945,
+      "step": 98520
+    },
+    {
+      "epoch": 13.31,
+      "learning_rate": 9.36139262999504e-06,
+      "loss": 2.2264,
+      "step": 98530
+    },
+    {
+      "epoch": 13.31,
+      "learning_rate": 9.34664273027203e-06,
+      "loss": 2.1924,
+      "step": 98540
+    },
+    {
+      "epoch": 13.31,
+      "learning_rate": 9.331904086074914e-06,
+      "loss": 2.205,
+      "step": 98550
+    },
+    {
+      "epoch": 13.31,
+      "learning_rate": 9.317176698583151e-06,
+      "loss": 2.2151,
+      "step": 98560
+    },
+    {
+      "epoch": 13.31,
+      "learning_rate": 9.302460568975195e-06,
+      "loss": 2.1788,
+      "step": 98570
+    },
+    {
+      "epoch": 13.31,
+      "learning_rate": 9.287755698428756e-06,
+      "loss": 2.2102,
+      "step": 98580
+    },
+    {
+      "epoch": 13.32,
+      "learning_rate": 9.273062088120526e-06,
+      "loss": 2.2078,
+      "step": 98590
+    },
+    {
+      "epoch": 13.32,
+      "learning_rate": 9.25837973922633e-06,
+      "loss": 2.2053,
+      "step": 98600
+    },
+    {
+      "epoch": 13.32,
+      "learning_rate": 9.243708652921112e-06,
+      "loss": 2.1796,
+      "step": 98610
+    },
+    {
+      "epoch": 13.32,
+      "learning_rate": 9.229048830378882e-06,
+      "loss": 2.2332,
+      "step": 98620
+    },
+    {
+      "epoch": 13.32,
+      "learning_rate": 9.214400272772804e-06,
+      "loss": 2.1989,
+      "step": 98630
+    },
+    {
+      "epoch": 13.32,
+      "learning_rate": 9.199762981275054e-06,
+      "loss": 2.1648,
+      "step": 98640
+    },
+    {
+      "epoch": 13.32,
+      "learning_rate": 9.185136957056948e-06,
+      "loss": 2.1907,
+      "step": 98650
+    },
+    {
+      "epoch": 13.32,
+      "learning_rate": 9.170522201288965e-06,
+      "loss": 2.1855,
+      "step": 98660
+    },
+    {
+      "epoch": 13.33,
+      "learning_rate": 9.155918715140602e-06,
+      "loss": 2.2157,
+      "step": 98670
+    },
+    {
+      "epoch": 13.33,
+      "learning_rate": 9.14132649978046e-06,
+      "loss": 2.1988,
+      "step": 98680
+    },
+    {
+      "epoch": 13.33,
+      "learning_rate": 9.126745556376236e-06,
+      "loss": 2.2296,
+      "step": 98690
+    },
+    {
+      "epoch": 13.33,
+      "learning_rate": 9.11217588609478e-06,
+      "loss": 2.2232,
+      "step": 98700
+    },
+    {
+      "epoch": 13.33,
+      "learning_rate": 9.09761749010201e-06,
+      "loss": 2.2242,
+      "step": 98710
+    },
+    {
+      "epoch": 13.33,
+      "learning_rate": 9.083070369562878e-06,
+      "loss": 2.2118,
+      "step": 98720
+    },
+    {
+      "epoch": 13.33,
+      "learning_rate": 9.068534525641552e-06,
+      "loss": 2.199,
+      "step": 98730
+    },
+    {
+      "epoch": 13.34,
+      "learning_rate": 9.054009959501185e-06,
+      "loss": 2.2029,
+      "step": 98740
+    },
+    {
+      "epoch": 13.34,
+      "learning_rate": 9.03949667230413e-06,
+      "loss": 2.2034,
+      "step": 98750
+    },
+    {
+      "epoch": 13.34,
+      "learning_rate": 9.024994665211776e-06,
+      "loss": 2.1922,
+      "step": 98760
+    },
+    {
+      "epoch": 13.34,
+      "learning_rate": 9.010503939384561e-06,
+      "loss": 2.195,
+      "step": 98770
+    },
+    {
+      "epoch": 13.34,
+      "learning_rate": 8.996024495982157e-06,
+      "loss": 2.1968,
+      "step": 98780
+    },
+    {
+      "epoch": 13.34,
+      "learning_rate": 8.98155633616322e-06,
+      "loss": 2.2254,
+      "step": 98790
+    },
+    {
+      "epoch": 13.34,
+      "learning_rate": 8.967099461085509e-06,
+      "loss": 2.1825,
+      "step": 98800
+    },
+    {
+      "epoch": 13.34,
+      "learning_rate": 8.952653871905963e-06,
+      "loss": 2.2418,
+      "step": 98810
+    },
+    {
+      "epoch": 13.35,
+      "learning_rate": 8.938219569780507e-06,
+      "loss": 2.1962,
+      "step": 98820
+    },
+    {
+      "epoch": 13.35,
+      "learning_rate": 8.923796555864304e-06,
+      "loss": 2.1682,
+      "step": 98830
+    },
+    {
+      "epoch": 13.35,
+      "learning_rate": 8.909384831311444e-06,
+      "loss": 2.1915,
+      "step": 98840
+    },
+    {
+      "epoch": 13.35,
+      "learning_rate": 8.894984397275207e-06,
+      "loss": 2.2083,
+      "step": 98850
+    },
+    {
+      "epoch": 13.35,
+      "learning_rate": 8.880595254908019e-06,
+      "loss": 2.2114,
+      "step": 98860
+    },
+    {
+      "epoch": 13.35,
+      "learning_rate": 8.866217405361308e-06,
+      "loss": 2.2084,
+      "step": 98870
+    },
+    {
+      "epoch": 13.35,
+      "learning_rate": 8.85185084978564e-06,
+      "loss": 2.197,
+      "step": 98880
+    },
+    {
+      "epoch": 13.36,
+      "learning_rate": 8.837495589330662e-06,
+      "loss": 2.2128,
+      "step": 98890
+    },
+    {
+      "epoch": 13.36,
+      "learning_rate": 8.823151625145136e-06,
+      "loss": 2.205,
+      "step": 98900
+    },
+    {
+      "epoch": 13.36,
+      "learning_rate": 8.808818958376913e-06,
+      "loss": 2.2021,
+      "step": 98910
+    },
+    {
+      "epoch": 13.36,
+      "learning_rate": 8.794497590172922e-06,
+      "loss": 2.2189,
+      "step": 98920
+    },
+    {
+      "epoch": 13.36,
+      "learning_rate": 8.780187521679233e-06,
+      "loss": 2.2026,
+      "step": 98930
+    },
+    {
+      "epoch": 13.36,
+      "learning_rate": 8.765888754040945e-06,
+      "loss": 2.2042,
+      "step": 98940
+    },
+    {
+      "epoch": 13.36,
+      "learning_rate": 8.751601288402327e-06,
+      "loss": 2.2082,
+      "step": 98950
+    },
+    {
+      "epoch": 13.37,
+      "learning_rate": 8.737325125906713e-06,
+      "loss": 2.2109,
+      "step": 98960
+    },
+    {
+      "epoch": 13.37,
+      "learning_rate": 8.723060267696474e-06,
+      "loss": 2.2003,
+      "step": 98970
+    },
+    {
+      "epoch": 13.37,
+      "learning_rate": 8.70880671491318e-06,
+      "loss": 2.2204,
+      "step": 98980
+    },
+    {
+      "epoch": 13.37,
+      "learning_rate": 8.694564468697434e-06,
+      "loss": 2.2051,
+      "step": 98990
+    },
+    {
+      "epoch": 13.37,
+      "learning_rate": 8.68033353018891e-06,
+      "loss": 2.2026,
+      "step": 99000
+    },
+    {
+      "epoch": 13.37,
+      "eval_loss": 2.403158664703369,
+      "eval_runtime": 1269.4248,
+      "eval_samples_per_second": 59.976,
+      "eval_steps_per_second": 4.998,
+      "step": 99000
+    },
+    {
+      "epoch": 13.37,
+      "learning_rate": 8.666113900526461e-06,
+      "loss": 2.2028,
+      "step": 99010
+    },
+    {
+      "epoch": 13.37,
+      "learning_rate": 8.651905580847963e-06,
+      "loss": 2.1781,
+      "step": 99020
+    },
+    {
+      "epoch": 13.37,
+      "learning_rate": 8.637708572290435e-06,
+      "loss": 2.1987,
+      "step": 99030
+    },
+    {
+      "epoch": 13.38,
+      "learning_rate": 8.623522875989924e-06,
+      "loss": 2.2305,
+      "step": 99040
+    },
+    {
+      "epoch": 13.38,
+      "learning_rate": 8.60934849308162e-06,
+      "loss": 2.1856,
+      "step": 99050
+    },
+    {
+      "epoch": 13.38,
+      "learning_rate": 8.595185424699846e-06,
+      "loss": 2.1764,
+      "step": 99060
+    },
+    {
+      "epoch": 13.38,
+      "learning_rate": 8.581033671977949e-06,
+      "loss": 2.2065,
+      "step": 99070
+    },
+    {
+      "epoch": 13.38,
+      "learning_rate": 8.566893236048389e-06,
+      "loss": 2.2348,
+      "step": 99080
+    },
+    {
+      "epoch": 13.38,
+      "learning_rate": 8.552764118042727e-06,
+      "loss": 2.2146,
+      "step": 99090
+    },
+    {
+      "epoch": 13.38,
+      "learning_rate": 8.53864631909164e-06,
+      "loss": 2.1937,
+      "step": 99100
+    },
+    {
+      "epoch": 13.39,
+      "learning_rate": 8.524539840324863e-06,
+      "loss": 2.1876,
+      "step": 99110
+    },
+    {
+      "epoch": 13.39,
+      "learning_rate": 8.51044468287122e-06,
+      "loss": 2.1938,
+      "step": 99120
+    },
+    {
+      "epoch": 13.39,
+      "learning_rate": 8.496360847858712e-06,
+      "loss": 2.2026,
+      "step": 99130
+    },
+    {
+      "epoch": 13.39,
+      "learning_rate": 8.482288336414289e-06,
+      "loss": 2.229,
+      "step": 99140
+    },
+    {
+      "epoch": 13.39,
+      "learning_rate": 8.468227149664164e-06,
+      "loss": 2.2138,
+      "step": 99150
+    },
+    {
+      "epoch": 13.39,
+      "learning_rate": 8.454177288733504e-06,
+      "loss": 2.2373,
+      "step": 99160
+    },
+    {
+      "epoch": 13.39,
+      "learning_rate": 8.440138754746629e-06,
+      "loss": 2.1873,
+      "step": 99170
+    },
+    {
+      "epoch": 13.39,
+      "learning_rate": 8.426111548826969e-06,
+      "loss": 2.2202,
+      "step": 99180
+    },
+    {
+      "epoch": 13.4,
+      "learning_rate": 8.412095672097013e-06,
+      "loss": 2.223,
+      "step": 99190
+    },
+    {
+      "epoch": 13.4,
+      "learning_rate": 8.398091125678314e-06,
+      "loss": 2.1832,
+      "step": 99200
+    },
+    {
+      "epoch": 13.4,
+      "learning_rate": 8.384097910691639e-06,
+      "loss": 2.1799,
+      "step": 99210
+    },
+    {
+      "epoch": 13.4,
+      "learning_rate": 8.370116028256696e-06,
+      "loss": 2.2099,
+      "step": 99220
+    },
+    {
+      "epoch": 13.4,
+      "learning_rate": 8.356145479492454e-06,
+      "loss": 2.2105,
+      "step": 99230
+    },
+    {
+      "epoch": 13.4,
+      "learning_rate": 8.342186265516754e-06,
+      "loss": 2.1805,
+      "step": 99240
+    },
+    {
+      "epoch": 13.4,
+      "learning_rate": 8.328238387446734e-06,
+      "loss": 2.1914,
+      "step": 99250
+    },
+    {
+      "epoch": 13.41,
+      "learning_rate": 8.314301846398552e-06,
+      "loss": 2.2172,
+      "step": 99260
+    },
+    {
+      "epoch": 13.41,
+      "learning_rate": 8.300376643487434e-06,
+      "loss": 2.1886,
+      "step": 99270
+    },
+    {
+      "epoch": 13.41,
+      "learning_rate": 8.286462779827736e-06,
+      "loss": 2.2083,
+      "step": 99280
+    },
+    {
+      "epoch": 13.41,
+      "learning_rate": 8.272560256532834e-06,
+      "loss": 2.213,
+      "step": 99290
+    },
+    {
+      "epoch": 13.41,
+      "learning_rate": 8.258669074715324e-06,
+      "loss": 2.2437,
+      "step": 99300
+    },
+    {
+      "epoch": 13.41,
+      "learning_rate": 8.244789235486782e-06,
+      "loss": 2.2031,
+      "step": 99310
+    },
+    {
+      "epoch": 13.41,
+      "learning_rate": 8.230920739957919e-06,
+      "loss": 2.1893,
+      "step": 99320
+    },
+    {
+      "epoch": 13.42,
+      "learning_rate": 8.217063589238548e-06,
+      "loss": 2.1995,
+      "step": 99330
+    },
+    {
+      "epoch": 13.42,
+      "learning_rate": 8.203217784437549e-06,
+      "loss": 2.1842,
+      "step": 99340
+    },
+    {
+      "epoch": 13.42,
+      "learning_rate": 8.189383326662918e-06,
+      "loss": 2.2006,
+      "step": 99350
+    },
+    {
+      "epoch": 13.42,
+      "learning_rate": 8.175560217021754e-06,
+      "loss": 2.2078,
+      "step": 99360
+    },
+    {
+      "epoch": 13.42,
+      "learning_rate": 8.161748456620154e-06,
+      "loss": 2.2194,
+      "step": 99370
+    },
+    {
+      "epoch": 13.42,
+      "learning_rate": 8.147948046563451e-06,
+      "loss": 2.2261,
+      "step": 99380
+    },
+    {
+      "epoch": 13.42,
+      "learning_rate": 8.13415898795598e-06,
+      "loss": 2.22,
+      "step": 99390
+    },
+    {
+      "epoch": 13.42,
+      "learning_rate": 8.120381281901156e-06,
+      "loss": 2.2186,
+      "step": 99400
+    },
+    {
+      "epoch": 13.43,
+      "learning_rate": 8.106614929501548e-06,
+      "loss": 2.2172,
+      "step": 99410
+    },
+    {
+      "epoch": 13.43,
+      "learning_rate": 8.092859931858741e-06,
+      "loss": 2.2189,
+      "step": 99420
+    },
+    {
+      "epoch": 13.43,
+      "learning_rate": 8.07911629007354e-06,
+      "loss": 2.2094,
+      "step": 99430
+    },
+    {
+      "epoch": 13.43,
+      "learning_rate": 8.065384005245645e-06,
+      "loss": 2.2385,
+      "step": 99440
+    },
+    {
+      "epoch": 13.43,
+      "learning_rate": 8.051663078474013e-06,
+      "loss": 2.2044,
+      "step": 99450
+    },
+    {
+      "epoch": 13.43,
+      "learning_rate": 8.037953510856648e-06,
+      "loss": 2.2138,
+      "step": 99460
+    },
+    {
+      "epoch": 13.43,
+      "learning_rate": 8.024255303490623e-06,
+      "loss": 2.203,
+      "step": 99470
+    },
+    {
+      "epoch": 13.44,
+      "learning_rate": 8.01056845747211e-06,
+      "loss": 2.1965,
+      "step": 99480
+    },
+    {
+      "epoch": 13.44,
+      "learning_rate": 7.996892973896336e-06,
+      "loss": 2.2139,
+      "step": 99490
+    },
+    {
+      "epoch": 13.44,
+      "learning_rate": 7.983228853857703e-06,
+      "loss": 2.2075,
+      "step": 99500
+    },
+    {
+      "epoch": 13.44,
+      "learning_rate": 7.969576098449659e-06,
+      "loss": 2.2002,
+      "step": 99510
+    },
+    {
+      "epoch": 13.44,
+      "learning_rate": 7.955934708764694e-06,
+      "loss": 2.1902,
+      "step": 99520
+    },
+    {
+      "epoch": 13.44,
+      "learning_rate": 7.942304685894502e-06,
+      "loss": 2.2314,
+      "step": 99530
+    },
+    {
+      "epoch": 13.44,
+      "learning_rate": 7.928686030929726e-06,
+      "loss": 2.1924,
+      "step": 99540
+    },
+    {
+      "epoch": 13.44,
+      "learning_rate": 7.915078744960245e-06,
+      "loss": 2.1927,
+      "step": 99550
+    },
+    {
+      "epoch": 13.45,
+      "learning_rate": 7.901482829074906e-06,
+      "loss": 2.2137,
+      "step": 99560
+    },
+    {
+      "epoch": 13.45,
+      "learning_rate": 7.887898284361704e-06,
+      "loss": 2.1887,
+      "step": 99570
+    },
+    {
+      "epoch": 13.45,
+      "learning_rate": 7.874325111907754e-06,
+      "loss": 2.1827,
+      "step": 99580
+    },
+    {
+      "epoch": 13.45,
+      "learning_rate": 7.860763312799184e-06,
+      "loss": 2.1719,
+      "step": 99590
+    },
+    {
+      "epoch": 13.45,
+      "learning_rate": 7.84721288812128e-06,
+      "loss": 2.215,
+      "step": 99600
+    },
+    {
+      "epoch": 13.45,
+      "learning_rate": 7.83367383895837e-06,
+      "loss": 2.2269,
+      "step": 99610
+    },
+    {
+      "epoch": 13.45,
+      "learning_rate": 7.820146166393892e-06,
+      "loss": 2.1944,
+      "step": 99620
+    },
+    {
+      "epoch": 13.46,
+      "learning_rate": 7.806629871510378e-06,
+      "loss": 2.1946,
+      "step": 99630
+    },
+    {
+      "epoch": 13.46,
+      "learning_rate": 7.793124955389463e-06,
+      "loss": 2.2048,
+      "step": 99640
+    },
+    {
+      "epoch": 13.46,
+      "learning_rate": 7.779631419111832e-06,
+      "loss": 2.1786,
+      "step": 99650
+    },
+    {
+      "epoch": 13.46,
+      "learning_rate": 7.766149263757255e-06,
+      "loss": 2.2084,
+      "step": 99660
+    },
+    {
+      "epoch": 13.46,
+      "learning_rate": 7.75267849040465e-06,
+      "loss": 2.1908,
+      "step": 99670
+    },
+    {
+      "epoch": 13.46,
+      "learning_rate": 7.739219100132043e-06,
+      "loss": 2.2126,
+      "step": 99680
+    },
+    {
+      "epoch": 13.46,
+      "learning_rate": 7.725771094016387e-06,
+      "loss": 2.2214,
+      "step": 99690
+    },
+    {
+      "epoch": 13.47,
+      "learning_rate": 7.712334473133902e-06,
+      "loss": 2.2415,
+      "step": 99700
+    },
+    {
+      "epoch": 13.47,
+      "learning_rate": 7.698909238559797e-06,
+      "loss": 2.2259,
+      "step": 99710
+    },
+    {
+      "epoch": 13.47,
+      "learning_rate": 7.685495391368445e-06,
+      "loss": 2.2036,
+      "step": 99720
+    },
+    {
+      "epoch": 13.47,
+      "learning_rate": 7.672092932633223e-06,
+      "loss": 2.1904,
+      "step": 99730
+    },
+    {
+      "epoch": 13.47,
+      "learning_rate": 7.658701863426653e-06,
+      "loss": 2.193,
+      "step": 99740
+    },
+    {
+      "epoch": 13.47,
+      "learning_rate": 7.645322184820329e-06,
+      "loss": 2.2005,
+      "step": 99750
+    },
+    {
+      "epoch": 13.47,
+      "learning_rate": 7.631953897884962e-06,
+      "loss": 2.2124,
+      "step": 99760
+    },
+    {
+      "epoch": 13.47,
+      "learning_rate": 7.618597003690263e-06,
+      "loss": 2.21,
+      "step": 99770
+    },
+    {
+      "epoch": 13.48,
+      "learning_rate": 7.6052515033051595e-06,
+      "loss": 2.218,
+      "step": 99780
+    },
+    {
+      "epoch": 13.48,
+      "learning_rate": 7.591917397797531e-06,
+      "loss": 2.2163,
+      "step": 99790
+    },
+    {
+      "epoch": 13.48,
+      "learning_rate": 7.578594688234491e-06,
+      "loss": 2.1915,
+      "step": 99800
+    },
+    {
+      "epoch": 13.48,
+      "learning_rate": 7.565283375682135e-06,
+      "loss": 2.2252,
+      "step": 99810
+    },
+    {
+      "epoch": 13.48,
+      "learning_rate": 7.55198346120563e-06,
+      "loss": 2.1862,
+      "step": 99820
+    },
+    {
+      "epoch": 13.48,
+      "learning_rate": 7.538694945869356e-06,
+      "loss": 2.1918,
+      "step": 99830
+    },
+    {
+      "epoch": 13.48,
+      "learning_rate": 7.526745029210468e-06,
+      "loss": 2.2057,
+      "step": 99840
+    },
+    {
+      "epoch": 13.49,
+      "learning_rate": 7.513478175169419e-06,
+      "loss": 2.1799,
+      "step": 99850
+    },
+    {
+      "epoch": 13.49,
+      "learning_rate": 7.500222723349886e-06,
+      "loss": 2.1973,
+      "step": 99860
+    },
+    {
+      "epoch": 13.49,
+      "learning_rate": 7.4869786748126044e-06,
+      "loss": 2.2012,
+      "step": 99870
+    },
+    {
+      "epoch": 13.49,
+      "learning_rate": 7.473746030617378e-06,
+      "loss": 2.2168,
+      "step": 99880
+    },
+    {
+      "epoch": 13.49,
+      "learning_rate": 7.460524791823174e-06,
+      "loss": 2.2098,
+      "step": 99890
+    },
+    {
+      "epoch": 13.49,
+      "learning_rate": 7.447314959487966e-06,
+      "loss": 2.2106,
+      "step": 99900
+    },
+    {
+      "epoch": 13.49,
+      "learning_rate": 7.434116534668838e-06,
+      "loss": 2.2046,
+      "step": 99910
+    },
+    {
+      "epoch": 13.49,
+      "learning_rate": 7.420929518422014e-06,
+      "loss": 2.1923,
+      "step": 99920
+    },
+    {
+      "epoch": 13.5,
+      "learning_rate": 7.4077539118027e-06,
+      "loss": 2.1912,
+      "step": 99930
+    },
+    {
+      "epoch": 13.5,
+      "learning_rate": 7.394589715865317e-06,
+      "loss": 2.1908,
+      "step": 99940
+    },
+    {
+      "epoch": 13.5,
+      "learning_rate": 7.381436931663259e-06,
+      "loss": 2.2282,
+      "step": 99950
+    },
+    {
+      "epoch": 13.5,
+      "learning_rate": 7.368295560249049e-06,
+      "loss": 2.2188,
+      "step": 99960
+    },
+    {
+      "epoch": 13.5,
+      "learning_rate": 7.355165602674329e-06,
+      "loss": 2.23,
+      "step": 99970
+    },
+    {
+      "epoch": 13.5,
+      "learning_rate": 7.342047059989792e-06,
+      "loss": 2.1958,
+      "step": 99980
+    },
+    {
+      "epoch": 13.5,
+      "learning_rate": 7.3289399332452145e-06,
+      "loss": 2.2049,
+      "step": 99990
+    },
+    {
+      "epoch": 13.51,
+      "learning_rate": 7.315844223489459e-06,
+      "loss": 2.1868,
+      "step": 100000
+    },
+    {
+      "epoch": 13.51,
+      "eval_loss": 2.402578592300415,
+      "eval_runtime": 1269.336,
+      "eval_samples_per_second": 59.98,
+      "eval_steps_per_second": 4.999,
+      "step": 100000
+    },
+    {
+      "epoch": 13.51,
+      "learning_rate": 7.302759931770502e-06,
+      "loss": 2.2089,
+      "step": 100010
+    },
+    {
+      "epoch": 13.51,
+      "learning_rate": 7.289687059135441e-06,
+      "loss": 2.1781,
+      "step": 100020
+    },
+    {
+      "epoch": 13.51,
+      "learning_rate": 7.276625606630304e-06,
+      "loss": 2.2477,
+      "step": 100030
+    },
+    {
+      "epoch": 13.51,
+      "learning_rate": 7.26357557530039e-06,
+      "loss": 2.1791,
+      "step": 100040
+    },
+    {
+      "epoch": 13.51,
+      "learning_rate": 7.250536966189962e-06,
+      "loss": 2.1935,
+      "step": 100050
+    },
+    {
+      "epoch": 13.51,
+      "learning_rate": 7.237509780342438e-06,
+      "loss": 2.1836,
+      "step": 100060
+    },
+    {
+      "epoch": 13.52,
+      "learning_rate": 7.224494018800297e-06,
+      "loss": 2.2214,
+      "step": 100070
+    },
+    {
+      "epoch": 13.52,
+      "learning_rate": 7.211489682605059e-06,
+      "loss": 2.215,
+      "step": 100080
+    },
+    {
+      "epoch": 13.52,
+      "learning_rate": 7.198496772797407e-06,
+      "loss": 2.2051,
+      "step": 100090
+    },
+    {
+      "epoch": 13.52,
+      "learning_rate": 7.185515290417077e-06,
+      "loss": 2.1906,
+      "step": 100100
+    },
+    {
+      "epoch": 13.52,
+      "learning_rate": 7.172545236502869e-06,
+      "loss": 2.2039,
+      "step": 100110
+    },
+    {
+      "epoch": 13.52,
+      "learning_rate": 7.1595866120927045e-06,
+      "loss": 2.2314,
+      "step": 100120
+    },
+    {
+      "epoch": 13.52,
+      "learning_rate": 7.146639418223554e-06,
+      "loss": 2.2213,
+      "step": 100130
+    },
+    {
+      "epoch": 13.52,
+      "learning_rate": 7.133703655931522e-06,
+      "loss": 2.2124,
+      "step": 100140
+    },
+    {
+      "epoch": 13.53,
+      "learning_rate": 7.120779326251747e-06,
+      "loss": 2.2092,
+      "step": 100150
+    },
+    {
+      "epoch": 13.53,
+      "learning_rate": 7.107866430218467e-06,
+      "loss": 2.1987,
+      "step": 100160
+    },
+    {
+      "epoch": 13.53,
+      "learning_rate": 7.094964968865058e-06,
+      "loss": 2.1872,
+      "step": 100170
+    },
+    {
+      "epoch": 13.53,
+      "learning_rate": 7.082074943223876e-06,
+      "loss": 2.2151,
+      "step": 100180
+    },
+    {
+      "epoch": 13.53,
+      "learning_rate": 7.069196354326479e-06,
+      "loss": 2.1733,
+      "step": 100190
+    },
+    {
+      "epoch": 13.53,
+      "learning_rate": 7.056329203203376e-06,
+      "loss": 2.2054,
+      "step": 100200
+    },
+    {
+      "epoch": 13.53,
+      "learning_rate": 7.043473490884294e-06,
+      "loss": 2.195,
+      "step": 100210
+    },
+    {
+      "epoch": 13.54,
+      "learning_rate": 7.030629218398026e-06,
+      "loss": 2.2032,
+      "step": 100220
+    },
+    {
+      "epoch": 13.54,
+      "learning_rate": 7.017796386772317e-06,
+      "loss": 2.2249,
+      "step": 100230
+    },
+    {
+      "epoch": 13.54,
+      "learning_rate": 7.004974997034163e-06,
+      "loss": 2.1902,
+      "step": 100240
+    },
+    {
+      "epoch": 13.54,
+      "learning_rate": 6.992165050209508e-06,
+      "loss": 2.2267,
+      "step": 100250
+    },
+    {
+      "epoch": 13.54,
+      "learning_rate": 6.979366547323517e-06,
+      "loss": 2.1868,
+      "step": 100260
+    },
+    {
+      "epoch": 13.54,
+      "learning_rate": 6.9665794894003204e-06,
+      "loss": 2.1873,
+      "step": 100270
+    },
+    {
+      "epoch": 13.54,
+      "learning_rate": 6.953803877463182e-06,
+      "loss": 2.1794,
+      "step": 100280
+    },
+    {
+      "epoch": 13.54,
+      "learning_rate": 6.941039712534469e-06,
+      "loss": 2.2042,
+      "step": 100290
+    },
+    {
+      "epoch": 13.55,
+      "learning_rate": 6.928286995635579e-06,
+      "loss": 2.2281,
+      "step": 100300
+    },
+    {
+      "epoch": 13.55,
+      "learning_rate": 6.915545727787047e-06,
+      "loss": 2.1981,
+      "step": 100310
+    },
+    {
+      "epoch": 13.55,
+      "learning_rate": 6.902815910008474e-06,
+      "loss": 2.2023,
+      "step": 100320
+    },
+    {
+      "epoch": 13.55,
+      "learning_rate": 6.890097543318513e-06,
+      "loss": 2.2254,
+      "step": 100330
+    },
+    {
+      "epoch": 13.55,
+      "learning_rate": 6.877390628734964e-06,
+      "loss": 2.1974,
+      "step": 100340
+    },
+    {
+      "epoch": 13.55,
+      "learning_rate": 6.8646951672746645e-06,
+      "loss": 2.1984,
+      "step": 100350
+    },
+    {
+      "epoch": 13.55,
+      "learning_rate": 6.852011159953502e-06,
+      "loss": 2.2353,
+      "step": 100360
+    },
+    {
+      "epoch": 13.56,
+      "learning_rate": 6.839338607786549e-06,
+      "loss": 2.2099,
+      "step": 100370
+    },
+    {
+      "epoch": 13.56,
+      "learning_rate": 6.826677511787859e-06,
+      "loss": 2.191,
+      "step": 100380
+    },
+    {
+      "epoch": 13.56,
+      "learning_rate": 6.81402787297069e-06,
+      "loss": 2.215,
+      "step": 100390
+    },
+    {
+      "epoch": 13.56,
+      "learning_rate": 6.801389692347181e-06,
+      "loss": 2.2121,
+      "step": 100400
+    },
+    {
+      "epoch": 13.56,
+      "learning_rate": 6.788762970928774e-06,
+      "loss": 2.2293,
+      "step": 100410
+    },
+    {
+      "epoch": 13.56,
+      "learning_rate": 6.776147709725876e-06,
+      "loss": 2.1984,
+      "step": 100420
+    },
+    {
+      "epoch": 13.56,
+      "learning_rate": 6.763543909747998e-06,
+      "loss": 2.1935,
+      "step": 100430
+    },
+    {
+      "epoch": 13.57,
+      "learning_rate": 6.750951572003732e-06,
+      "loss": 2.1866,
+      "step": 100440
+    },
+    {
+      "epoch": 13.57,
+      "learning_rate": 6.738370697500755e-06,
+      "loss": 2.1829,
+      "step": 100450
+    },
+    {
+      "epoch": 13.57,
+      "learning_rate": 6.725801287245846e-06,
+      "loss": 2.2139,
+      "step": 100460
+    },
+    {
+      "epoch": 13.57,
+      "learning_rate": 6.713243342244834e-06,
+      "loss": 2.185,
+      "step": 100470
+    },
+    {
+      "epoch": 13.57,
+      "learning_rate": 6.7006968635026135e-06,
+      "loss": 2.213,
+      "step": 100480
+    },
+    {
+      "epoch": 13.57,
+      "learning_rate": 6.688161852023266e-06,
+      "loss": 2.2146,
+      "step": 100490
+    },
+    {
+      "epoch": 13.57,
+      "learning_rate": 6.6756383088098056e-06,
+      "loss": 2.198,
+      "step": 100500
+    },
+    {
+      "epoch": 13.57,
+      "learning_rate": 6.66312623486448e-06,
+      "loss": 2.2188,
+      "step": 100510
+    },
+    {
+      "epoch": 13.58,
+      "learning_rate": 6.650625631188505e-06,
+      "loss": 2.1988,
+      "step": 100520
+    },
+    {
+      "epoch": 13.58,
+      "learning_rate": 6.6381364987821986e-06,
+      "loss": 2.1983,
+      "step": 100530
+    },
+    {
+      "epoch": 13.58,
+      "learning_rate": 6.625658838645026e-06,
+      "loss": 2.2031,
+      "step": 100540
+    },
+    {
+      "epoch": 13.58,
+      "learning_rate": 6.613192651775456e-06,
+      "loss": 2.1922,
+      "step": 100550
+    },
+    {
+      "epoch": 13.58,
+      "learning_rate": 6.600737939171074e-06,
+      "loss": 2.2062,
+      "step": 100560
+    },
+    {
+      "epoch": 13.58,
+      "learning_rate": 6.588294701828584e-06,
+      "loss": 2.1853,
+      "step": 100570
+    },
+    {
+      "epoch": 13.58,
+      "learning_rate": 6.57586294074367e-06,
+      "loss": 2.1875,
+      "step": 100580
+    },
+    {
+      "epoch": 13.59,
+      "learning_rate": 6.563442656911238e-06,
+      "loss": 2.2224,
+      "step": 100590
+    },
+    {
+      "epoch": 13.59,
+      "learning_rate": 6.5510338513251104e-06,
+      "loss": 2.1992,
+      "step": 100600
+    },
+    {
+      "epoch": 13.59,
+      "learning_rate": 6.538636524978324e-06,
+      "loss": 2.1961,
+      "step": 100610
+    },
+    {
+      "epoch": 13.59,
+      "learning_rate": 6.526250678862987e-06,
+      "loss": 2.1907,
+      "step": 100620
+    },
+    {
+      "epoch": 13.59,
+      "learning_rate": 6.513876313970206e-06,
+      "loss": 2.189,
+      "step": 100630
+    },
+    {
+      "epoch": 13.59,
+      "learning_rate": 6.501513431290223e-06,
+      "loss": 2.2031,
+      "step": 100640
+    },
+    {
+      "epoch": 13.59,
+      "learning_rate": 6.489162031812345e-06,
+      "loss": 2.2186,
+      "step": 100650
+    },
+    {
+      "epoch": 13.59,
+      "learning_rate": 6.476822116525015e-06,
+      "loss": 2.1978,
+      "step": 100660
+    },
+    {
+      "epoch": 13.6,
+      "learning_rate": 6.464493686415678e-06,
+      "loss": 2.2134,
+      "step": 100670
+    },
+    {
+      "epoch": 13.6,
+      "learning_rate": 6.452176742470877e-06,
+      "loss": 2.1969,
+      "step": 100680
+    },
+    {
+      "epoch": 13.6,
+      "learning_rate": 6.4398712856762905e-06,
+      "loss": 2.1947,
+      "step": 100690
+    },
+    {
+      "epoch": 13.6,
+      "learning_rate": 6.4275773170166136e-06,
+      "loss": 2.1936,
+      "step": 100700
+    },
+    {
+      "epoch": 13.6,
+      "learning_rate": 6.415294837475676e-06,
+      "loss": 2.1839,
+      "step": 100710
+    },
+    {
+      "epoch": 13.6,
+      "learning_rate": 6.403023848036342e-06,
+      "loss": 2.2028,
+      "step": 100720
+    },
+    {
+      "epoch": 13.6,
+      "learning_rate": 6.3907643496805605e-06,
+      "loss": 2.2015,
+      "step": 100730
+    },
+    {
+      "epoch": 13.61,
+      "learning_rate": 6.378516343389411e-06,
+      "loss": 2.2001,
+      "step": 100740
+    },
+    {
+      "epoch": 13.61,
+      "learning_rate": 6.3662798301429955e-06,
+      "loss": 2.2124,
+      "step": 100750
+    },
+    {
+      "epoch": 13.61,
+      "learning_rate": 6.3540548109204945e-06,
+      "loss": 2.1924,
+      "step": 100760
+    },
+    {
+      "epoch": 13.61,
+      "learning_rate": 6.341841286700245e-06,
+      "loss": 2.2121,
+      "step": 100770
+    },
+    {
+      "epoch": 13.61,
+      "learning_rate": 6.329639258459579e-06,
+      "loss": 2.2218,
+      "step": 100780
+    },
+    {
+      "epoch": 13.61,
+      "learning_rate": 6.3174487271749675e-06,
+      "loss": 2.1911,
+      "step": 100790
+    },
+    {
+      "epoch": 13.61,
+      "learning_rate": 6.305269693821879e-06,
+      "loss": 2.2245,
+      "step": 100800
+    },
+    {
+      "epoch": 13.62,
+      "learning_rate": 6.293102159374969e-06,
+      "loss": 2.2129,
+      "step": 100810
+    },
+    {
+      "epoch": 13.62,
+      "learning_rate": 6.280946124807923e-06,
+      "loss": 2.173,
+      "step": 100820
+    },
+    {
+      "epoch": 13.62,
+      "learning_rate": 6.268801591093481e-06,
+      "loss": 2.1948,
+      "step": 100830
+    },
+    {
+      "epoch": 13.62,
+      "learning_rate": 6.2566685592034985e-06,
+      "loss": 2.2074,
+      "step": 100840
+    },
+    {
+      "epoch": 13.62,
+      "learning_rate": 6.244547030108899e-06,
+      "loss": 2.1877,
+      "step": 100850
+    },
+    {
+      "epoch": 13.62,
+      "learning_rate": 6.2324370047796725e-06,
+      "loss": 2.1952,
+      "step": 100860
+    },
+    {
+      "epoch": 13.62,
+      "learning_rate": 6.220338484184928e-06,
+      "loss": 2.2106,
+      "step": 100870
+    },
+    {
+      "epoch": 13.62,
+      "learning_rate": 6.208251469292791e-06,
+      "loss": 2.2037,
+      "step": 100880
+    },
+    {
+      "epoch": 13.63,
+      "learning_rate": 6.196175961070554e-06,
+      "loss": 2.2145,
+      "step": 100890
+    },
+    {
+      "epoch": 13.63,
+      "learning_rate": 6.184111960484478e-06,
+      "loss": 2.2098,
+      "step": 100900
+    },
+    {
+      "epoch": 13.63,
+      "learning_rate": 6.172059468500007e-06,
+      "loss": 2.1998,
+      "step": 100910
+    },
+    {
+      "epoch": 13.63,
+      "learning_rate": 6.160018486081603e-06,
+      "loss": 2.2052,
+      "step": 100920
+    },
+    {
+      "epoch": 13.63,
+      "learning_rate": 6.147989014192811e-06,
+      "loss": 2.2069,
+      "step": 100930
+    },
+    {
+      "epoch": 13.63,
+      "learning_rate": 6.135971053796296e-06,
+      "loss": 2.2099,
+      "step": 100940
+    },
+    {
+      "epoch": 13.63,
+      "learning_rate": 6.123964605853754e-06,
+      "loss": 2.2023,
+      "step": 100950
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 6.111969671325967e-06,
+      "loss": 2.1889,
+      "step": 100960
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 6.09998625117285e-06,
+      "loss": 2.1995,
+      "step": 100970
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 6.0880143463532874e-06,
+      "loss": 2.218,
+      "step": 100980
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 6.076053957825411e-06,
+      "loss": 2.2252,
+      "step": 100990
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 6.064105086546223e-06,
+      "loss": 2.1972,
+      "step": 101000
+    },
+    {
+      "epoch": 13.64,
+      "eval_loss": 2.4018149375915527,
+      "eval_runtime": 1270.4303,
+      "eval_samples_per_second": 59.929,
+      "eval_steps_per_second": 4.994,
+      "step": 101000
     }
   ],
   "logging_steps": 10,
   "max_steps": 111060,
   "num_train_epochs": 15,
   "save_steps": 1000,
-  "total_flos": 1.5657271333748736e+19,
+  "total_flos": 3.1792627288375296e+19,
   "trial_name": null,
   "trial_params": null
 }