diff --git "a/checkpoint-3052726/trainer_state.json" "b/checkpoint-3052726/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-3052726/trainer_state.json"
@@ -0,0 +1,36117 @@
+{
+  "best_metric": 3.9892959594726562,
+  "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/outputs/superlative-quantifier/lstm/3/checkpoints/checkpoint-3052726",
+  "epoch": 1.0249763653862154,
+  "eval_steps": 10,
+  "global_step": 3052726,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999998362119627e-05,
+      "loss": 10.8181,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999161405248948e-05,
+      "loss": 7.554,
+      "step": 512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.998322810497896e-05,
+      "loss": 7.0541,
+      "step": 1024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.997484215746844e-05,
+      "loss": 6.9869,
+      "step": 1536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.996645620995792e-05,
+      "loss": 6.9413,
+      "step": 2048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99580702624474e-05,
+      "loss": 6.8806,
+      "step": 2560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.994968431493688e-05,
+      "loss": 6.7151,
+      "step": 3072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.994129836742636e-05,
+      "loss": 6.609,
+      "step": 3584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.993292879871958e-05,
+      "loss": 6.521,
+      "step": 4096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.992454285120906e-05,
+      "loss": 6.4483,
+      "step": 4608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.991617328250227e-05,
+      "loss": 6.386,
+      "step": 5120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.990778733499175e-05,
+      "loss": 6.3152,
+      "step": 5632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.989940138748123e-05,
+      "loss": 6.2486,
+      "step": 6144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.989101543997071e-05,
+      "loss": 6.189,
+      "step": 6656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.988262949246018e-05,
+      "loss": 6.1376,
+      "step": 7168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9874259923753396e-05,
+      "loss": 6.0919,
+      "step": 7680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9865873976242876e-05,
+      "loss": 6.047,
+      "step": 8192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9857488028732356e-05,
+      "loss": 6.0107,
+      "step": 8704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.984910208122183e-05,
+      "loss": 5.9639,
+      "step": 9216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9840716133711316e-05,
+      "loss": 5.9378,
+      "step": 9728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9832330186200796e-05,
+      "loss": 5.9037,
+      "step": 10240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9823944238690276e-05,
+      "loss": 5.8591,
+      "step": 10752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9815558291179756e-05,
+      "loss": 5.8305,
+      "step": 11264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9807172343669236e-05,
+      "loss": 5.7959,
+      "step": 11776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9798786396158716e-05,
+      "loss": 5.7762,
+      "step": 12288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9790400448648195e-05,
+      "loss": 5.7448,
+      "step": 12800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9782014501137675e-05,
+      "loss": 5.7162,
+      "step": 13312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9773628553627155e-05,
+      "loss": 5.6881,
+      "step": 13824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9765242606116635e-05,
+      "loss": 5.6819,
+      "step": 14336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9756856658606115e-05,
+      "loss": 5.643,
+      "step": 14848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9748470711095595e-05,
+      "loss": 5.6307,
+      "step": 15360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9740101142388804e-05,
+      "loss": 5.6222,
+      "step": 15872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9731715194878284e-05,
+      "loss": 5.5913,
+      "step": 16384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9723329247367764e-05,
+      "loss": 5.5736,
+      "step": 16896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9714943299857244e-05,
+      "loss": 5.5687,
+      "step": 17408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9706557352346724e-05,
+      "loss": 5.5461,
+      "step": 17920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9698171404836204e-05,
+      "loss": 5.5371,
+      "step": 18432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.968980183612942e-05,
+      "loss": 5.4986,
+      "step": 18944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.96814158886189e-05,
+      "loss": 5.4921,
+      "step": 19456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.967302994110837e-05,
+      "loss": 5.4639,
+      "step": 19968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.966466037240159e-05,
+      "loss": 5.4617,
+      "step": 20480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.965627442489107e-05,
+      "loss": 5.4486,
+      "step": 20992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.964788847738054e-05,
+      "loss": 5.443,
+      "step": 21504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.963950252987002e-05,
+      "loss": 5.4127,
+      "step": 22016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.96311165823595e-05,
+      "loss": 5.411,
+      "step": 22528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.962273063484898e-05,
+      "loss": 5.3966,
+      "step": 23040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.961434468733847e-05,
+      "loss": 5.3982,
+      "step": 23552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.960595873982795e-05,
+      "loss": 5.3902,
+      "step": 24064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.959757279231743e-05,
+      "loss": 5.362,
+      "step": 24576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.958920322361064e-05,
+      "loss": 5.346,
+      "step": 25088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.958081727610012e-05,
+      "loss": 5.3559,
+      "step": 25600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.95724313285896e-05,
+      "loss": 5.3494,
+      "step": 26112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.956404538107908e-05,
+      "loss": 5.3248,
+      "step": 26624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.955567581237229e-05,
+      "loss": 5.3156,
+      "step": 27136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9547306243665496e-05,
+      "loss": 5.3127,
+      "step": 27648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9538920296154976e-05,
+      "loss": 5.2933,
+      "step": 28160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9530534348644456e-05,
+      "loss": 5.3016,
+      "step": 28672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9522148401133936e-05,
+      "loss": 5.268,
+      "step": 29184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.951376245362342e-05,
+      "loss": 5.2718,
+      "step": 29696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.95053765061129e-05,
+      "loss": 5.2606,
+      "step": 30208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.949700693740611e-05,
+      "loss": 5.2606,
+      "step": 30720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.948862098989559e-05,
+      "loss": 5.2366,
+      "step": 31232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.948023504238507e-05,
+      "loss": 5.2443,
+      "step": 31744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.947184909487455e-05,
+      "loss": 5.2165,
+      "step": 32256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.946346314736403e-05,
+      "loss": 5.2226,
+      "step": 32768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.945509357865724e-05,
+      "loss": 5.2228,
+      "step": 33280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.944670763114672e-05,
+      "loss": 5.215,
+      "step": 33792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.94383216836362e-05,
+      "loss": 5.194,
+      "step": 34304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.942993573612568e-05,
+      "loss": 5.178,
+      "step": 34816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.942154978861516e-05,
+      "loss": 5.1707,
+      "step": 35328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.941316384110464e-05,
+      "loss": 5.1857,
+      "step": 35840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.940477789359412e-05,
+      "loss": 5.175,
+      "step": 36352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.93963919460836e-05,
+      "loss": 5.1593,
+      "step": 36864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.938800599857309e-05,
+      "loss": 5.1699,
+      "step": 37376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9379636429866296e-05,
+      "loss": 5.1678,
+      "step": 37888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9371250482355776e-05,
+      "loss": 5.153,
+      "step": 38400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9362864534845256e-05,
+      "loss": 5.141,
+      "step": 38912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.935447858733473e-05,
+      "loss": 5.1305,
+      "step": 39424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9346109018627945e-05,
+      "loss": 5.128,
+      "step": 39936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9337723071117425e-05,
+      "loss": 5.1128,
+      "step": 40448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9329337123606905e-05,
+      "loss": 5.1165,
+      "step": 40960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9320967554900114e-05,
+      "loss": 5.1039,
+      "step": 41472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9312581607389594e-05,
+      "loss": 5.1159,
+      "step": 41984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9304195659879074e-05,
+      "loss": 5.1033,
+      "step": 42496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9295809712368554e-05,
+      "loss": 5.0802,
+      "step": 43008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9287423764858034e-05,
+      "loss": 5.0832,
+      "step": 43520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9279037817347514e-05,
+      "loss": 5.0818,
+      "step": 44032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9270651869836994e-05,
+      "loss": 5.0803,
+      "step": 44544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9262265922326474e-05,
+      "loss": 5.0756,
+      "step": 45056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9253879974815954e-05,
+      "loss": 5.057,
+      "step": 45568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.924551040610916e-05,
+      "loss": 5.0545,
+      "step": 46080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.923712445859864e-05,
+      "loss": 5.052,
+      "step": 46592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.922873851108812e-05,
+      "loss": 5.0378,
+      "step": 47104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.92203525635776e-05,
+      "loss": 5.0278,
+      "step": 47616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.921196661606708e-05,
+      "loss": 5.0278,
+      "step": 48128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.920358066855656e-05,
+      "loss": 5.0352,
+      "step": 48640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.919519472104604e-05,
+      "loss": 5.0228,
+      "step": 49152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.918680877353552e-05,
+      "loss": 5.0122,
+      "step": 49664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.917843920482874e-05,
+      "loss": 5.003,
+      "step": 50176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.917005325731822e-05,
+      "loss": 5.0043,
+      "step": 50688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.916168368861143e-05,
+      "loss": 5.0084,
+      "step": 51200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.915329774110091e-05,
+      "loss": 4.9993,
+      "step": 51712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.914491179359039e-05,
+      "loss": 4.9931,
+      "step": 52224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.913652584607987e-05,
+      "loss": 4.9805,
+      "step": 52736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.912813989856935e-05,
+      "loss": 4.9902,
+      "step": 53248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.911975395105883e-05,
+      "loss": 4.9681,
+      "step": 53760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9111384382352036e-05,
+      "loss": 4.9724,
+      "step": 54272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9103014813645245e-05,
+      "loss": 4.965,
+      "step": 54784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.909462886613473e-05,
+      "loss": 4.9575,
+      "step": 55296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.908624291862421e-05,
+      "loss": 4.9525,
+      "step": 55808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.907785697111369e-05,
+      "loss": 4.9576,
+      "step": 56320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.906947102360317e-05,
+      "loss": 4.9438,
+      "step": 56832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.906108507609265e-05,
+      "loss": 4.9418,
+      "step": 57344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.905269912858213e-05,
+      "loss": 4.9395,
+      "step": 57856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.904431318107161e-05,
+      "loss": 4.9347,
+      "step": 58368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.903592723356109e-05,
+      "loss": 4.938,
+      "step": 58880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9027541286050565e-05,
+      "loss": 4.9293,
+      "step": 59392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.901917171734378e-05,
+      "loss": 4.9274,
+      "step": 59904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.901078576983326e-05,
+      "loss": 4.9239,
+      "step": 60416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.900239982232274e-05,
+      "loss": 4.9175,
+      "step": 60928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8994013874812214e-05,
+      "loss": 4.9161,
+      "step": 61440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.89856279273017e-05,
+      "loss": 4.9034,
+      "step": 61952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.897724197979118e-05,
+      "loss": 4.9007,
+      "step": 62464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.896885603228066e-05,
+      "loss": 4.8986,
+      "step": 62976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.896047008477014e-05,
+      "loss": 4.8971,
+      "step": 63488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.895210051606335e-05,
+      "loss": 4.8903,
+      "step": 64000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.894371456855283e-05,
+      "loss": 4.8877,
+      "step": 64512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.893532862104231e-05,
+      "loss": 4.8865,
+      "step": 65024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.892695905233552e-05,
+      "loss": 4.8742,
+      "step": 65536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8918573104825e-05,
+      "loss": 4.883,
+      "step": 66048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.891018715731448e-05,
+      "loss": 4.8813,
+      "step": 66560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.890180120980396e-05,
+      "loss": 4.8616,
+      "step": 67072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.889341526229344e-05,
+      "loss": 4.8506,
+      "step": 67584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.888502931478292e-05,
+      "loss": 4.8572,
+      "step": 68096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.88766433672724e-05,
+      "loss": 4.8643,
+      "step": 68608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.886825741976188e-05,
+      "loss": 4.8714,
+      "step": 69120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8859887851055094e-05,
+      "loss": 4.8537,
+      "step": 69632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8851501903544574e-05,
+      "loss": 4.8524,
+      "step": 70144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8843115956034054e-05,
+      "loss": 4.8537,
+      "step": 70656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8834730008523534e-05,
+      "loss": 4.8516,
+      "step": 71168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.882636043981674e-05,
+      "loss": 4.8381,
+      "step": 71680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.881797449230622e-05,
+      "loss": 4.8321,
+      "step": 72192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.88095885447957e-05,
+      "loss": 4.8387,
+      "step": 72704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.880120259728518e-05,
+      "loss": 4.829,
+      "step": 73216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.879283302857839e-05,
+      "loss": 4.8262,
+      "step": 73728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.878444708106787e-05,
+      "loss": 4.8299,
+      "step": 74240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.877606113355735e-05,
+      "loss": 4.8144,
+      "step": 74752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.876769156485057e-05,
+      "loss": 4.8243,
+      "step": 75264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.875930561734005e-05,
+      "loss": 4.8211,
+      "step": 75776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.875091966982953e-05,
+      "loss": 4.7985,
+      "step": 76288
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.786604404449463,
+      "eval_runtime": 295.0564,
+      "eval_samples_per_second": 1293.281,
+      "eval_steps_per_second": 40.416,
+      "step": 76320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.874253372231901e-05,
+      "loss": 4.8134,
+      "step": 76800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.873414777480849e-05,
+      "loss": 4.8015,
+      "step": 77312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.872576182729797e-05,
+      "loss": 4.8128,
+      "step": 77824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.871737587978745e-05,
+      "loss": 4.7994,
+      "step": 78336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.870898993227693e-05,
+      "loss": 4.8006,
+      "step": 78848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8700636742373866e-05,
+      "loss": 4.7861,
+      "step": 79360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8692250794863346e-05,
+      "loss": 4.7865,
+      "step": 79872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8683864847352826e-05,
+      "loss": 4.7683,
+      "step": 80384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8675478899842306e-05,
+      "loss": 4.7937,
+      "step": 80896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8667092952331785e-05,
+      "loss": 4.7873,
+      "step": 81408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.865870700482127e-05,
+      "loss": 4.7827,
+      "step": 81920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.865033743611448e-05,
+      "loss": 4.7861,
+      "step": 82432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.864196786740769e-05,
+      "loss": 4.7686,
+      "step": 82944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.863358191989717e-05,
+      "loss": 4.7666,
+      "step": 83456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.862519597238665e-05,
+      "loss": 4.7624,
+      "step": 83968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.861681002487613e-05,
+      "loss": 4.7606,
+      "step": 84480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.860842407736561e-05,
+      "loss": 4.76,
+      "step": 84992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.860003812985509e-05,
+      "loss": 4.7537,
+      "step": 85504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.859165218234457e-05,
+      "loss": 4.7632,
+      "step": 86016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.858326623483405e-05,
+      "loss": 4.7654,
+      "step": 86528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.857488028732352e-05,
+      "loss": 4.7522,
+      "step": 87040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.856651071861674e-05,
+      "loss": 4.7431,
+      "step": 87552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8558124771106226e-05,
+      "loss": 4.7412,
+      "step": 88064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.85497388235957e-05,
+      "loss": 4.7478,
+      "step": 88576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.854135287608518e-05,
+      "loss": 4.7399,
+      "step": 89088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8532983307378395e-05,
+      "loss": 4.7264,
+      "step": 89600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8524597359867875e-05,
+      "loss": 4.7271,
+      "step": 90112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.851621141235735e-05,
+      "loss": 4.7422,
+      "step": 90624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.850782546484683e-05,
+      "loss": 4.7146,
+      "step": 91136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.849943951733631e-05,
+      "loss": 4.7201,
+      "step": 91648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.849105356982579e-05,
+      "loss": 4.7295,
+      "step": 92160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.848266762231527e-05,
+      "loss": 4.7252,
+      "step": 92672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.847428167480475e-05,
+      "loss": 4.7236,
+      "step": 93184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8465912106097964e-05,
+      "loss": 4.7233,
+      "step": 93696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.845754253739117e-05,
+      "loss": 4.7193,
+      "step": 94208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.844915658988065e-05,
+      "loss": 4.7198,
+      "step": 94720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.844077064237013e-05,
+      "loss": 4.6977,
+      "step": 95232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.843238469485961e-05,
+      "loss": 4.7067,
+      "step": 95744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.842399874734909e-05,
+      "loss": 4.6894,
+      "step": 96256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.841561279983857e-05,
+      "loss": 4.6974,
+      "step": 96768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.840722685232805e-05,
+      "loss": 4.6985,
+      "step": 97280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.839884090481753e-05,
+      "loss": 4.6972,
+      "step": 97792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.839045495730701e-05,
+      "loss": 4.6903,
+      "step": 98304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.838208538860022e-05,
+      "loss": 4.6906,
+      "step": 98816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.83736994410897e-05,
+      "loss": 4.6898,
+      "step": 99328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.836531349357918e-05,
+      "loss": 4.6925,
+      "step": 99840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.835692754606866e-05,
+      "loss": 4.6957,
+      "step": 100352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.834855797736188e-05,
+      "loss": 4.6779,
+      "step": 100864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8340188408655086e-05,
+      "loss": 4.6695,
+      "step": 101376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8331802461144566e-05,
+      "loss": 4.683,
+      "step": 101888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8323416513634046e-05,
+      "loss": 4.69,
+      "step": 102400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8315030566123526e-05,
+      "loss": 4.6743,
+      "step": 102912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8306644618613006e-05,
+      "loss": 4.6693,
+      "step": 103424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8298258671102486e-05,
+      "loss": 4.6677,
+      "step": 103936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8289889102395695e-05,
+      "loss": 4.6634,
+      "step": 104448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8281503154885175e-05,
+      "loss": 4.6726,
+      "step": 104960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8273117207374655e-05,
+      "loss": 4.6521,
+      "step": 105472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8264731259864135e-05,
+      "loss": 4.6644,
+      "step": 105984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8256345312353615e-05,
+      "loss": 4.6592,
+      "step": 106496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.824797574364683e-05,
+      "loss": 4.6611,
+      "step": 107008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.823958979613631e-05,
+      "loss": 4.6424,
+      "step": 107520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.823120384862579e-05,
+      "loss": 4.6629,
+      "step": 108032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.822281790111527e-05,
+      "loss": 4.6368,
+      "step": 108544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.821443195360475e-05,
+      "loss": 4.6521,
+      "step": 109056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.820604600609423e-05,
+      "loss": 4.6516,
+      "step": 109568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.819766005858371e-05,
+      "loss": 4.6515,
+      "step": 110080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8189274111073184e-05,
+      "loss": 4.637,
+      "step": 110592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.81809045423664e-05,
+      "loss": 4.6318,
+      "step": 111104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.817253497365961e-05,
+      "loss": 4.6295,
+      "step": 111616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.816414902614909e-05,
+      "loss": 4.6408,
+      "step": 112128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.815576307863857e-05,
+      "loss": 4.6419,
+      "step": 112640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.814737713112805e-05,
+      "loss": 4.6326,
+      "step": 113152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8138991183617535e-05,
+      "loss": 4.6403,
+      "step": 113664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.813060523610701e-05,
+      "loss": 4.6443,
+      "step": 114176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.812221928859649e-05,
+      "loss": 4.6411,
+      "step": 114688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8113849719889704e-05,
+      "loss": 4.6233,
+      "step": 115200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8105463772379184e-05,
+      "loss": 4.6292,
+      "step": 115712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.809707782486866e-05,
+      "loss": 4.6306,
+      "step": 116224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.808869187735814e-05,
+      "loss": 4.6084,
+      "step": 116736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.808030592984762e-05,
+      "loss": 4.6297,
+      "step": 117248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8071936361140827e-05,
+      "loss": 4.6125,
+      "step": 117760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8063550413630307e-05,
+      "loss": 4.6328,
+      "step": 118272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8055164466119786e-05,
+      "loss": 4.6229,
+      "step": 118784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.804677851860927e-05,
+      "loss": 4.6001,
+      "step": 119296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.803839257109875e-05,
+      "loss": 4.6169,
+      "step": 119808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.803000662358823e-05,
+      "loss": 4.6057,
+      "step": 120320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.802162067607771e-05,
+      "loss": 4.6198,
+      "step": 120832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.801323472856719e-05,
+      "loss": 4.6119,
+      "step": 121344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.80048651598604e-05,
+      "loss": 4.6087,
+      "step": 121856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.799647921234988e-05,
+      "loss": 4.6041,
+      "step": 122368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.798809326483936e-05,
+      "loss": 4.6021,
+      "step": 122880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.797970731732884e-05,
+      "loss": 4.5941,
+      "step": 123392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.797133774862205e-05,
+      "loss": 4.5879,
+      "step": 123904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.796295180111153e-05,
+      "loss": 4.5919,
+      "step": 124416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.795456585360101e-05,
+      "loss": 4.6018,
+      "step": 124928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.794617990609049e-05,
+      "loss": 4.5938,
+      "step": 125440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.793781033738371e-05,
+      "loss": 4.5877,
+      "step": 125952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.792942438987319e-05,
+      "loss": 4.5818,
+      "step": 126464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.792103844236267e-05,
+      "loss": 4.59,
+      "step": 126976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.791265249485215e-05,
+      "loss": 4.5948,
+      "step": 127488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.790426654734163e-05,
+      "loss": 4.5879,
+      "step": 128000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7895880599831107e-05,
+      "loss": 4.5828,
+      "step": 128512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7887494652320587e-05,
+      "loss": 4.582,
+      "step": 129024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7879125083613796e-05,
+      "loss": 4.5877,
+      "step": 129536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7870739136103276e-05,
+      "loss": 4.5679,
+      "step": 130048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7862353188592756e-05,
+      "loss": 4.5772,
+      "step": 130560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7853967241082236e-05,
+      "loss": 4.5755,
+      "step": 131072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7845581293571715e-05,
+      "loss": 4.5686,
+      "step": 131584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.783722810366866e-05,
+      "loss": 4.5678,
+      "step": 132096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.782884215615814e-05,
+      "loss": 4.5715,
+      "step": 132608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.782045620864762e-05,
+      "loss": 4.5681,
+      "step": 133120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.78120702611371e-05,
+      "loss": 4.5654,
+      "step": 133632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.780368431362658e-05,
+      "loss": 4.5663,
+      "step": 134144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.779529836611606e-05,
+      "loss": 4.5629,
+      "step": 134656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.778691241860554e-05,
+      "loss": 4.5746,
+      "step": 135168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.777852647109502e-05,
+      "loss": 4.5689,
+      "step": 135680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.777014052358449e-05,
+      "loss": 4.5643,
+      "step": 136192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.776175457607397e-05,
+      "loss": 4.5629,
+      "step": 136704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.775336862856345e-05,
+      "loss": 4.5641,
+      "step": 137216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.774498268105293e-05,
+      "loss": 4.5608,
+      "step": 137728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.773661311234615e-05,
+      "loss": 4.5529,
+      "step": 138240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.772822716483563e-05,
+      "loss": 4.5508,
+      "step": 138752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.771984121732511e-05,
+      "loss": 4.5562,
+      "step": 139264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.771145526981459e-05,
+      "loss": 4.556,
+      "step": 139776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.770306932230407e-05,
+      "loss": 4.5522,
+      "step": 140288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.769468337479355e-05,
+      "loss": 4.5518,
+      "step": 140800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.768629742728303e-05,
+      "loss": 4.5458,
+      "step": 141312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.767791147977251e-05,
+      "loss": 4.5493,
+      "step": 141824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.766954191106572e-05,
+      "loss": 4.5472,
+      "step": 142336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.76611559635552e-05,
+      "loss": 4.5552,
+      "step": 142848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.765278639484841e-05,
+      "loss": 4.5362,
+      "step": 143360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.764440044733789e-05,
+      "loss": 4.5268,
+      "step": 143872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.763601449982737e-05,
+      "loss": 4.5338,
+      "step": 144384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.762762855231685e-05,
+      "loss": 4.5464,
+      "step": 144896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.761924260480633e-05,
+      "loss": 4.5498,
+      "step": 145408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7610856657295813e-05,
+      "loss": 4.5408,
+      "step": 145920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.760248708858902e-05,
+      "loss": 4.5408,
+      "step": 146432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.75941011410785e-05,
+      "loss": 4.5421,
+      "step": 146944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.758571519356798e-05,
+      "loss": 4.5419,
+      "step": 147456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.757734562486119e-05,
+      "loss": 4.5334,
+      "step": 147968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.756895967735067e-05,
+      "loss": 4.5298,
+      "step": 148480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.756057372984015e-05,
+      "loss": 4.5333,
+      "step": 148992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.755218778232963e-05,
+      "loss": 4.5304,
+      "step": 149504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.754380183481911e-05,
+      "loss": 4.5236,
+      "step": 150016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.753541588730859e-05,
+      "loss": 4.5353,
+      "step": 150528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.752702993979807e-05,
+      "loss": 4.5203,
+      "step": 151040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.751864399228755e-05,
+      "loss": 4.5303,
+      "step": 151552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.751025804477703e-05,
+      "loss": 4.5276,
+      "step": 152064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.750187209726651e-05,
+      "loss": 4.5144,
+      "step": 152576
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.5020432472229,
+      "eval_runtime": 295.1253,
+      "eval_samples_per_second": 1292.98,
+      "eval_steps_per_second": 40.407,
+      "step": 152640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.749348614975599e-05,
+      "loss": 4.526,
+      "step": 153088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.748510020224547e-05,
+      "loss": 4.5155,
+      "step": 153600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.747671425473495e-05,
+      "loss": 4.5255,
+      "step": 154112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.746832830722443e-05,
+      "loss": 4.519,
+      "step": 154624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.745994235971391e-05,
+      "loss": 4.5225,
+      "step": 155136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.745155641220339e-05,
+      "loss": 4.5087,
+      "step": 155648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.744317046469287e-05,
+      "loss": 4.5075,
+      "step": 156160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.743478451718235e-05,
+      "loss": 4.4953,
+      "step": 156672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.742639856967183e-05,
+      "loss": 4.524,
+      "step": 157184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.741801262216131e-05,
+      "loss": 4.5157,
+      "step": 157696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.740962667465079e-05,
+      "loss": 4.5078,
+      "step": 158208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7401240727140264e-05,
+      "loss": 4.5211,
+      "step": 158720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.739287115843348e-05,
+      "loss": 4.4986,
+      "step": 159232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7384485210922966e-05,
+      "loss": 4.5064,
+      "step": 159744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7376099263412446e-05,
+      "loss": 4.4974,
+      "step": 160256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.736771331590192e-05,
+      "loss": 4.4954,
+      "step": 160768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.73593273683914e-05,
+      "loss": 4.5049,
+      "step": 161280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.735094142088088e-05,
+      "loss": 4.4916,
+      "step": 161792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.734255547337036e-05,
+      "loss": 4.5056,
+      "step": 162304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.733416952585984e-05,
+      "loss": 4.5073,
+      "step": 162816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.732578357834932e-05,
+      "loss": 4.4995,
+      "step": 163328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.73173976308388e-05,
+      "loss": 4.4933,
+      "step": 163840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.730901168332828e-05,
+      "loss": 4.4933,
+      "step": 164352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.730064211462149e-05,
+      "loss": 4.4976,
+      "step": 164864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7292272545914704e-05,
+      "loss": 4.4858,
+      "step": 165376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7283886598404184e-05,
+      "loss": 4.4864,
+      "step": 165888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7275500650893664e-05,
+      "loss": 4.4749,
+      "step": 166400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7267114703383144e-05,
+      "loss": 4.4991,
+      "step": 166912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7258728755872624e-05,
+      "loss": 4.4695,
+      "step": 167424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7250342808362104e-05,
+      "loss": 4.4818,
+      "step": 167936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7241956860851584e-05,
+      "loss": 4.4873,
+      "step": 168448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.723358729214479e-05,
+      "loss": 4.4863,
+      "step": 168960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.722520134463427e-05,
+      "loss": 4.4843,
+      "step": 169472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.721681539712375e-05,
+      "loss": 4.4869,
+      "step": 169984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.720844582841696e-05,
+      "loss": 4.4832,
+      "step": 170496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.720005988090644e-05,
+      "loss": 4.4864,
+      "step": 171008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.719167393339592e-05,
+      "loss": 4.4695,
+      "step": 171520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.71832879858854e-05,
+      "loss": 4.4745,
+      "step": 172032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.717490203837488e-05,
+      "loss": 4.4609,
+      "step": 172544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.716651609086437e-05,
+      "loss": 4.469,
+      "step": 173056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.715813014335385e-05,
+      "loss": 4.4668,
+      "step": 173568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.714974419584333e-05,
+      "loss": 4.4742,
+      "step": 174080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.714135824833281e-05,
+      "loss": 4.464,
+      "step": 174592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.713297230082228e-05,
+      "loss": 4.4676,
+      "step": 175104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.712458635331176e-05,
+      "loss": 4.4689,
+      "step": 175616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.711620040580124e-05,
+      "loss": 4.4701,
+      "step": 176128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.710783083709446e-05,
+      "loss": 4.4734,
+      "step": 176640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.709944488958393e-05,
+      "loss": 4.4584,
+      "step": 177152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7091075320877147e-05,
+      "loss": 4.4481,
+      "step": 177664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7082689373366627e-05,
+      "loss": 4.4696,
+      "step": 178176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7074303425856106e-05,
+      "loss": 4.469,
+      "step": 178688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.706593385714932e-05,
+      "loss": 4.4604,
+      "step": 179200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.70575479096388e-05,
+      "loss": 4.4592,
+      "step": 179712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.704916196212828e-05,
+      "loss": 4.4499,
+      "step": 180224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7040776014617755e-05,
+      "loss": 4.4548,
+      "step": 180736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7032390067107235e-05,
+      "loss": 4.4607,
+      "step": 181248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7024004119596715e-05,
+      "loss": 4.4421,
+      "step": 181760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7015618172086195e-05,
+      "loss": 4.4542,
+      "step": 182272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7007232224575675e-05,
+      "loss": 4.4518,
+      "step": 182784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6998862655868884e-05,
+      "loss": 4.4487,
+      "step": 183296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6990476708358364e-05,
+      "loss": 4.4375,
+      "step": 183808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6982090760847844e-05,
+      "loss": 4.4602,
+      "step": 184320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6973704813337324e-05,
+      "loss": 4.4365,
+      "step": 184832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6965318865826804e-05,
+      "loss": 4.448,
+      "step": 185344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.695693291831629e-05,
+      "loss": 4.4454,
+      "step": 185856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.694854697080577e-05,
+      "loss": 4.4522,
+      "step": 186368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.694016102329525e-05,
+      "loss": 4.4354,
+      "step": 186880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.693179145458846e-05,
+      "loss": 4.4359,
+      "step": 187392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.692342188588167e-05,
+      "loss": 4.4259,
+      "step": 187904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.691503593837115e-05,
+      "loss": 4.4451,
+      "step": 188416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.690666636966436e-05,
+      "loss": 4.4462,
+      "step": 188928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.689828042215384e-05,
+      "loss": 4.4389,
+      "step": 189440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.688989447464332e-05,
+      "loss": 4.4434,
+      "step": 189952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.68815085271328e-05,
+      "loss": 4.4514,
+      "step": 190464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.687312257962228e-05,
+      "loss": 4.4512,
+      "step": 190976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.686473663211176e-05,
+      "loss": 4.4289,
+      "step": 191488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6856350684601245e-05,
+      "loss": 4.4373,
+      "step": 192000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6847964737090725e-05,
+      "loss": 4.4454,
+      "step": 192512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6839578789580204e-05,
+      "loss": 4.4215,
+      "step": 193024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6831192842069684e-05,
+      "loss": 4.4382,
+      "step": 193536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6822806894559164e-05,
+      "loss": 4.4228,
+      "step": 194048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6814437325852373e-05,
+      "loss": 4.4475,
+      "step": 194560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6806051378341853e-05,
+      "loss": 4.4381,
+      "step": 195072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6797665430831333e-05,
+      "loss": 4.4172,
+      "step": 195584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.678927948332081e-05,
+      "loss": 4.4277,
+      "step": 196096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6780893535810287e-05,
+      "loss": 4.4276,
+      "step": 196608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6772507588299766e-05,
+      "loss": 4.436,
+      "step": 197120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6764121640789246e-05,
+      "loss": 4.4293,
+      "step": 197632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.675575207208246e-05,
+      "loss": 4.427,
+      "step": 198144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.674736612457194e-05,
+      "loss": 4.4236,
+      "step": 198656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.673898017706142e-05,
+      "loss": 4.4204,
+      "step": 199168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.67305942295509e-05,
+      "loss": 4.4175,
+      "step": 199680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.672220828204038e-05,
+      "loss": 4.4062,
+      "step": 200192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.671382233452986e-05,
+      "loss": 4.418,
+      "step": 200704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.670543638701934e-05,
+      "loss": 4.4256,
+      "step": 201216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.669705043950882e-05,
+      "loss": 4.4187,
+      "step": 201728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.668868087080203e-05,
+      "loss": 4.4116,
+      "step": 202240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.668029492329151e-05,
+      "loss": 4.4095,
+      "step": 202752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.667190897578099e-05,
+      "loss": 4.4139,
+      "step": 203264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.666352302827047e-05,
+      "loss": 4.4248,
+      "step": 203776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6655169838367416e-05,
+      "loss": 4.415,
+      "step": 204288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6646783890856896e-05,
+      "loss": 4.4126,
+      "step": 204800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6638397943346376e-05,
+      "loss": 4.4057,
+      "step": 205312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6630011995835856e-05,
+      "loss": 4.4203,
+      "step": 205824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6621626048325336e-05,
+      "loss": 4.3974,
+      "step": 206336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6613240100814816e-05,
+      "loss": 4.4107,
+      "step": 206848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6604870532108025e-05,
+      "loss": 4.404,
+      "step": 207360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6596484584597505e-05,
+      "loss": 4.4061,
+      "step": 207872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6588098637086985e-05,
+      "loss": 4.4021,
+      "step": 208384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6579712689576465e-05,
+      "loss": 4.4053,
+      "step": 208896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6571343120869674e-05,
+      "loss": 4.4046,
+      "step": 209408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6562957173359154e-05,
+      "loss": 4.3995,
+      "step": 209920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6554571225848634e-05,
+      "loss": 4.4055,
+      "step": 210432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6546185278338114e-05,
+      "loss": 4.4003,
+      "step": 210944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.65377993308276e-05,
+      "loss": 4.4072,
+      "step": 211456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.652941338331708e-05,
+      "loss": 4.4074,
+      "step": 211968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.652104381461029e-05,
+      "loss": 4.4045,
+      "step": 212480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.651265786709977e-05,
+      "loss": 4.4039,
+      "step": 212992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.650428829839298e-05,
+      "loss": 4.401,
+      "step": 213504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.649590235088246e-05,
+      "loss": 4.3966,
+      "step": 214016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.648751640337194e-05,
+      "loss": 4.3975,
+      "step": 214528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.647913045586142e-05,
+      "loss": 4.3951,
+      "step": 215040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.64707445083509e-05,
+      "loss": 4.3928,
+      "step": 215552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.646235856084038e-05,
+      "loss": 4.4013,
+      "step": 216064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.645397261332986e-05,
+      "loss": 4.3925,
+      "step": 216576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.644558666581934e-05,
+      "loss": 4.3971,
+      "step": 217088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.643720071830882e-05,
+      "loss": 4.3907,
+      "step": 217600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.64288147707983e-05,
+      "loss": 4.3927,
+      "step": 218112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.642042882328778e-05,
+      "loss": 4.3882,
+      "step": 218624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.641204287577726e-05,
+      "loss": 4.4022,
+      "step": 219136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6403673307070474e-05,
+      "loss": 4.3834,
+      "step": 219648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.639528735955995e-05,
+      "loss": 4.38,
+      "step": 220160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.638691779085316e-05,
+      "loss": 4.3776,
+      "step": 220672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.637853184334264e-05,
+      "loss": 4.3922,
+      "step": 221184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.637014589583212e-05,
+      "loss": 4.3965,
+      "step": 221696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6361759948321596e-05,
+      "loss": 4.3946,
+      "step": 222208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6353374000811076e-05,
+      "loss": 4.3891,
+      "step": 222720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.634500443210429e-05,
+      "loss": 4.3946,
+      "step": 223232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.633661848459377e-05,
+      "loss": 4.3882,
+      "step": 223744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.632823253708325e-05,
+      "loss": 4.3889,
+      "step": 224256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.631984658957273e-05,
+      "loss": 4.3784,
+      "step": 224768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.631146064206221e-05,
+      "loss": 4.3883,
+      "step": 225280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.630307469455169e-05,
+      "loss": 4.384,
+      "step": 225792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.62947051258449e-05,
+      "loss": 4.3742,
+      "step": 226304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.628631917833438e-05,
+      "loss": 4.3892,
+      "step": 226816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.627793323082386e-05,
+      "loss": 4.3811,
+      "step": 227328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.626954728331334e-05,
+      "loss": 4.3818,
+      "step": 227840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.626117771460655e-05,
+      "loss": 4.3832,
+      "step": 228352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.625279176709603e-05,
+      "loss": 4.3706,
+      "step": 228864
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.365301609039307,
+      "eval_runtime": 290.8234,
+      "eval_samples_per_second": 1312.106,
+      "eval_steps_per_second": 41.004,
+      "step": 228960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.624440581958551e-05,
+      "loss": 4.3854,
+      "step": 229376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.623601987207499e-05,
+      "loss": 4.3747,
+      "step": 229888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6227633924564476e-05,
+      "loss": 4.3818,
+      "step": 230400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6219264355857685e-05,
+      "loss": 4.3751,
+      "step": 230912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6210878408347165e-05,
+      "loss": 4.3856,
+      "step": 231424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6202492460836645e-05,
+      "loss": 4.3676,
+      "step": 231936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6194106513326125e-05,
+      "loss": 4.3645,
+      "step": 232448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6185720565815605e-05,
+      "loss": 4.3583,
+      "step": 232960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6177334618305085e-05,
+      "loss": 4.3817,
+      "step": 233472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6168948670794565e-05,
+      "loss": 4.381,
+      "step": 233984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6160579102087774e-05,
+      "loss": 4.3635,
+      "step": 234496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6152193154577254e-05,
+      "loss": 4.3812,
+      "step": 235008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.614382358587046e-05,
+      "loss": 4.3641,
+      "step": 235520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.613543763835994e-05,
+      "loss": 4.3679,
+      "step": 236032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.612705169084942e-05,
+      "loss": 4.3643,
+      "step": 236544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.611866574333891e-05,
+      "loss": 4.3582,
+      "step": 237056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.611027979582839e-05,
+      "loss": 4.3668,
+      "step": 237568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.610189384831787e-05,
+      "loss": 4.3577,
+      "step": 238080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.609350790080735e-05,
+      "loss": 4.3707,
+      "step": 238592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.608512195329683e-05,
+      "loss": 4.3709,
+      "step": 239104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.607673600578631e-05,
+      "loss": 4.3707,
+      "step": 239616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.606835005827578e-05,
+      "loss": 4.3614,
+      "step": 240128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6059980489569e-05,
+      "loss": 4.3564,
+      "step": 240640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.605159454205848e-05,
+      "loss": 4.3708,
+      "step": 241152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.604322497335169e-05,
+      "loss": 4.3522,
+      "step": 241664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.603483902584117e-05,
+      "loss": 4.3567,
+      "step": 242176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.602645307833065e-05,
+      "loss": 4.3447,
+      "step": 242688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.601806713082013e-05,
+      "loss": 4.3667,
+      "step": 243200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.600968118330961e-05,
+      "loss": 4.3418,
+      "step": 243712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.600129523579909e-05,
+      "loss": 4.3509,
+      "step": 244224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.599290928828857e-05,
+      "loss": 4.3611,
+      "step": 244736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.598452334077805e-05,
+      "loss": 4.355,
+      "step": 245248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.597613739326753e-05,
+      "loss": 4.3558,
+      "step": 245760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5967767824560737e-05,
+      "loss": 4.3592,
+      "step": 246272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5959381877050217e-05,
+      "loss": 4.3564,
+      "step": 246784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5950995929539696e-05,
+      "loss": 4.3579,
+      "step": 247296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5942609982029176e-05,
+      "loss": 4.3481,
+      "step": 247808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5934224034518656e-05,
+      "loss": 4.3422,
+      "step": 248320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5925838087008136e-05,
+      "loss": 4.3392,
+      "step": 248832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5917452139497616e-05,
+      "loss": 4.3449,
+      "step": 249344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5909066191987096e-05,
+      "loss": 4.3405,
+      "step": 249856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.590069662328031e-05,
+      "loss": 4.3534,
+      "step": 250368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.589231067576979e-05,
+      "loss": 4.3391,
+      "step": 250880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.588392472825927e-05,
+      "loss": 4.3439,
+      "step": 251392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.587553878074875e-05,
+      "loss": 4.3482,
+      "step": 251904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.586716921204196e-05,
+      "loss": 4.342,
+      "step": 252416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.585879964333517e-05,
+      "loss": 4.3556,
+      "step": 252928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.585041369582465e-05,
+      "loss": 4.3374,
+      "step": 253440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.584202774831413e-05,
+      "loss": 4.3209,
+      "step": 253952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.583364180080361e-05,
+      "loss": 4.3534,
+      "step": 254464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.582525585329309e-05,
+      "loss": 4.3466,
+      "step": 254976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.581686990578257e-05,
+      "loss": 4.3453,
+      "step": 255488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.580848395827205e-05,
+      "loss": 4.3372,
+      "step": 256000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.580009801076153e-05,
+      "loss": 4.3312,
+      "step": 256512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5791728442054746e-05,
+      "loss": 4.3336,
+      "step": 257024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5783358873347955e-05,
+      "loss": 4.3395,
+      "step": 257536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5774972925837435e-05,
+      "loss": 4.3259,
+      "step": 258048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5766586978326915e-05,
+      "loss": 4.3316,
+      "step": 258560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5758201030816395e-05,
+      "loss": 4.3394,
+      "step": 259072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5749831462109604e-05,
+      "loss": 4.3305,
+      "step": 259584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5741445514599084e-05,
+      "loss": 4.3201,
+      "step": 260096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5733059567088564e-05,
+      "loss": 4.3441,
+      "step": 260608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5724673619578044e-05,
+      "loss": 4.3191,
+      "step": 261120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.571630405087125e-05,
+      "loss": 4.3267,
+      "step": 261632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.570791810336074e-05,
+      "loss": 4.3344,
+      "step": 262144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.569953215585022e-05,
+      "loss": 4.3367,
+      "step": 262656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.56911462083397e-05,
+      "loss": 4.3214,
+      "step": 263168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.568276026082918e-05,
+      "loss": 4.324,
+      "step": 263680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.567437431331866e-05,
+      "loss": 4.3082,
+      "step": 264192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.566600474461187e-05,
+      "loss": 4.336,
+      "step": 264704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.565763517590508e-05,
+      "loss": 4.3309,
+      "step": 265216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.564924922839456e-05,
+      "loss": 4.3284,
+      "step": 265728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.564086328088404e-05,
+      "loss": 4.3265,
+      "step": 266240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.563247733337352e-05,
+      "loss": 4.3387,
+      "step": 266752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5624091385863e-05,
+      "loss": 4.341,
+      "step": 267264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.561570543835248e-05,
+      "loss": 4.3162,
+      "step": 267776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.560731949084196e-05,
+      "loss": 4.3254,
+      "step": 268288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.559893354333144e-05,
+      "loss": 4.3339,
+      "step": 268800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.559054759582092e-05,
+      "loss": 4.3092,
+      "step": 269312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.558217802711413e-05,
+      "loss": 4.3285,
+      "step": 269824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.557379207960361e-05,
+      "loss": 4.3151,
+      "step": 270336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.556540613209309e-05,
+      "loss": 4.3324,
+      "step": 270848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5557020184582566e-05,
+      "loss": 4.3313,
+      "step": 271360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.554865061587578e-05,
+      "loss": 4.3086,
+      "step": 271872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.554026466836526e-05,
+      "loss": 4.3201,
+      "step": 272384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.553187872085474e-05,
+      "loss": 4.3158,
+      "step": 272896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5523492773344215e-05,
+      "loss": 4.3293,
+      "step": 273408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5515106825833695e-05,
+      "loss": 4.3189,
+      "step": 273920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.550673725712691e-05,
+      "loss": 4.3173,
+      "step": 274432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.549835130961639e-05,
+      "loss": 4.3151,
+      "step": 274944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.548996536210587e-05,
+      "loss": 4.3149,
+      "step": 275456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.548157941459535e-05,
+      "loss": 4.3118,
+      "step": 275968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.547319346708483e-05,
+      "loss": 4.2993,
+      "step": 276480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.546480751957431e-05,
+      "loss": 4.3111,
+      "step": 276992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.545642157206379e-05,
+      "loss": 4.3184,
+      "step": 277504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.544803562455327e-05,
+      "loss": 4.3183,
+      "step": 278016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.543966605584648e-05,
+      "loss": 4.2998,
+      "step": 278528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.543129648713969e-05,
+      "loss": 4.3053,
+      "step": 279040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.542291053962917e-05,
+      "loss": 4.3088,
+      "step": 279552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.541452459211865e-05,
+      "loss": 4.319,
+      "step": 280064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.540613864460813e-05,
+      "loss": 4.3087,
+      "step": 280576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.539775269709761e-05,
+      "loss": 4.3109,
+      "step": 281088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5389366749587095e-05,
+      "loss": 4.302,
+      "step": 281600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5380980802076575e-05,
+      "loss": 4.3174,
+      "step": 282112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5372594854566055e-05,
+      "loss": 4.2922,
+      "step": 282624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5364225285859264e-05,
+      "loss": 4.3079,
+      "step": 283136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5355855717152473e-05,
+      "loss": 4.3,
+      "step": 283648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5347469769641953e-05,
+      "loss": 4.3046,
+      "step": 284160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.533908382213143e-05,
+      "loss": 4.2983,
+      "step": 284672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.533071425342464e-05,
+      "loss": 4.3091,
+      "step": 285184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.532232830591412e-05,
+      "loss": 4.2972,
+      "step": 285696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.53139423584036e-05,
+      "loss": 4.2978,
+      "step": 286208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.530555641089308e-05,
+      "loss": 4.3032,
+      "step": 286720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.529717046338256e-05,
+      "loss": 4.3005,
+      "step": 287232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.528878451587205e-05,
+      "loss": 4.3089,
+      "step": 287744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.528039856836153e-05,
+      "loss": 4.3095,
+      "step": 288256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.527202899965474e-05,
+      "loss": 4.3028,
+      "step": 288768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.526364305214422e-05,
+      "loss": 4.3068,
+      "step": 289280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.52552571046337e-05,
+      "loss": 4.2999,
+      "step": 289792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.524687115712318e-05,
+      "loss": 4.2956,
+      "step": 290304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.523848520961266e-05,
+      "loss": 4.2969,
+      "step": 290816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.523009926210214e-05,
+      "loss": 4.2971,
+      "step": 291328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.522171331459162e-05,
+      "loss": 4.2986,
+      "step": 291840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.52133273670811e-05,
+      "loss": 4.3025,
+      "step": 292352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.520494141957057e-05,
+      "loss": 4.2953,
+      "step": 292864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.519655547206005e-05,
+      "loss": 4.3007,
+      "step": 293376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.518816952454953e-05,
+      "loss": 4.2907,
+      "step": 293888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.517978357703902e-05,
+      "loss": 4.2955,
+      "step": 294400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.517141400833223e-05,
+      "loss": 4.2933,
+      "step": 294912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.516302806082171e-05,
+      "loss": 4.307,
+      "step": 295424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5154642113311187e-05,
+      "loss": 4.2896,
+      "step": 295936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.51462725446044e-05,
+      "loss": 4.2833,
+      "step": 296448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5137886597093876e-05,
+      "loss": 4.2781,
+      "step": 296960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5129500649583356e-05,
+      "loss": 4.2958,
+      "step": 297472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5121114702072836e-05,
+      "loss": 4.3003,
+      "step": 297984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5112728754562316e-05,
+      "loss": 4.3006,
+      "step": 298496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5104342807051795e-05,
+      "loss": 4.2966,
+      "step": 299008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5095956859541275e-05,
+      "loss": 4.2997,
+      "step": 299520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5087587290834485e-05,
+      "loss": 4.2932,
+      "step": 300032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.507920134332397e-05,
+      "loss": 4.2934,
+      "step": 300544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.507081539581345e-05,
+      "loss": 4.286,
+      "step": 301056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.506242944830293e-05,
+      "loss": 4.2935,
+      "step": 301568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.505404350079241e-05,
+      "loss": 4.2929,
+      "step": 302080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.504565755328189e-05,
+      "loss": 4.2795,
+      "step": 302592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.50372879845751e-05,
+      "loss": 4.2973,
+      "step": 303104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.502890203706458e-05,
+      "loss": 4.2894,
+      "step": 303616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.502053246835779e-05,
+      "loss": 4.2855,
+      "step": 304128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.501214652084727e-05,
+      "loss": 4.2879,
+      "step": 304640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.500376057333675e-05,
+      "loss": 4.2853,
+      "step": 305152
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.280682563781738,
+      "eval_runtime": 301.0163,
+      "eval_samples_per_second": 1267.675,
+      "eval_steps_per_second": 39.616,
+      "step": 305280
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.499537462582623e-05,
+      "loss": 4.2872,
+      "step": 305664
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.498698867831571e-05,
+      "loss": 4.285,
+      "step": 306176
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.497860273080519e-05,
+      "loss": 4.2922,
+      "step": 306688
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.497021678329467e-05,
+      "loss": 4.2819,
+      "step": 307200
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.496183083578415e-05,
+      "loss": 4.2943,
+      "step": 307712
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4953477645881094e-05,
+      "loss": 4.2808,
+      "step": 308224
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4945091698370574e-05,
+      "loss": 4.2711,
+      "step": 308736
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4936705750860054e-05,
+      "loss": 4.2731,
+      "step": 309248
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4928319803349534e-05,
+      "loss": 4.2872,
+      "step": 309760
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4919933855839014e-05,
+      "loss": 4.2903,
+      "step": 310272
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.491156428713222e-05,
+      "loss": 4.2777,
+      "step": 310784
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.49031783396217e-05,
+      "loss": 4.2892,
+      "step": 311296
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.489480877091491e-05,
+      "loss": 4.278,
+      "step": 311808
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.488642282340439e-05,
+      "loss": 4.279,
+      "step": 312320
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.487803687589387e-05,
+      "loss": 4.2736,
+      "step": 312832
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.486965092838336e-05,
+      "loss": 4.2697,
+      "step": 313344
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.486126498087284e-05,
+      "loss": 4.2754,
+      "step": 313856
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.485287903336232e-05,
+      "loss": 4.2706,
+      "step": 314368
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.48444930858518e-05,
+      "loss": 4.2851,
+      "step": 314880
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.483610713834128e-05,
+      "loss": 4.2837,
+      "step": 315392
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.482773756963449e-05,
+      "loss": 4.2843,
+      "step": 315904
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.481935162212397e-05,
+      "loss": 4.2772,
+      "step": 316416
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.481096567461345e-05,
+      "loss": 4.2696,
+      "step": 316928
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.480257972710293e-05,
+      "loss": 4.2811,
+      "step": 317440
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.479419377959241e-05,
+      "loss": 4.2687,
+      "step": 317952
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.478580783208188e-05,
+      "loss": 4.2686,
+      "step": 318464
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.477742188457136e-05,
+      "loss": 4.2589,
+      "step": 318976
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.476903593706084e-05,
+      "loss": 4.2756,
+      "step": 319488
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4760666368354056e-05,
+      "loss": 4.2624,
+      "step": 320000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.4752280420843536e-05,
+      "loss": 4.26,
+      "step": 320512
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4743894473333016e-05,
+      "loss": 4.2765,
+      "step": 321024
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.473552490462623e-05,
+      "loss": 4.267,
+      "step": 321536
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.472713895711571e-05,
+      "loss": 4.2759,
+      "step": 322048
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4718753009605185e-05,
+      "loss": 4.2755,
+      "step": 322560
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4710367062094665e-05,
+      "loss": 4.266,
+      "step": 323072
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4701981114584145e-05,
+      "loss": 4.276,
+      "step": 323584
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4693611545877354e-05,
+      "loss": 4.2679,
+      "step": 324096
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4685225598366834e-05,
+      "loss": 4.2566,
+      "step": 324608
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4676839650856314e-05,
+      "loss": 4.2534,
+      "step": 325120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4668453703345794e-05,
+      "loss": 4.2592,
+      "step": 325632
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.466008413463901e-05,
+      "loss": 4.2588,
+      "step": 326144
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4651714565932226e-05,
+      "loss": 4.2702,
+      "step": 326656
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4643328618421706e-05,
+      "loss": 4.2584,
+      "step": 327168
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.463494267091118e-05,
+      "loss": 4.2597,
+      "step": 327680
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.462655672340066e-05,
+      "loss": 4.2683,
+      "step": 328192
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.461817077589014e-05,
+      "loss": 4.2645,
+      "step": 328704
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.460978482837962e-05,
+      "loss": 4.2696,
+      "step": 329216
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.46013988808691e-05,
+      "loss": 4.2535,
+      "step": 329728
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.459301293335858e-05,
+      "loss": 4.238,
+      "step": 330240
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.458462698584806e-05,
+      "loss": 4.2715,
+      "step": 330752
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.457624103833754e-05,
+      "loss": 4.2691,
+      "step": 331264
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.456785509082702e-05,
+      "loss": 4.263,
+      "step": 331776
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.45594691433165e-05,
+      "loss": 4.2568,
+      "step": 332288
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4551099574609714e-05,
+      "loss": 4.2493,
+      "step": 332800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4542713627099194e-05,
+      "loss": 4.2524,
+      "step": 333312
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4534327679588674e-05,
+      "loss": 4.2564,
+      "step": 333824
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4525941732078154e-05,
+      "loss": 4.2487,
+      "step": 334336
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.451757216337136e-05,
+      "loss": 4.2516,
+      "step": 334848
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.450918621586084e-05,
+      "loss": 4.2595,
+      "step": 335360
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.450080026835032e-05,
+      "loss": 4.2542,
+      "step": 335872
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.44924143208398e-05,
+      "loss": 4.2416,
+      "step": 336384
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.448402837332928e-05,
+      "loss": 4.2666,
+      "step": 336896
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.447565880462249e-05,
+      "loss": 4.2409,
+      "step": 337408
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.446727285711197e-05,
+      "loss": 4.2441,
+      "step": 337920
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.445888690960145e-05,
+      "loss": 4.2586,
+      "step": 338432
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.445051734089467e-05,
+      "loss": 4.2619,
+      "step": 338944
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.444213139338415e-05,
+      "loss": 4.239,
+      "step": 339456
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.443374544587363e-05,
+      "loss": 4.2473,
+      "step": 339968
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.442535949836311e-05,
+      "loss": 4.2307,
+      "step": 340480
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.441697355085259e-05,
+      "loss": 4.2568,
+      "step": 340992
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.440858760334207e-05,
+      "loss": 4.2567,
+      "step": 341504
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.440020165583154e-05,
+      "loss": 4.2471,
+      "step": 342016
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.439181570832102e-05,
+      "loss": 4.2525,
+      "step": 342528
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.438344613961424e-05,
+      "loss": 4.2579,
+      "step": 343040
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4375076570907446e-05,
+      "loss": 4.2675,
+      "step": 343552
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4366690623396926e-05,
+      "loss": 4.2394,
+      "step": 344064
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4358304675886406e-05,
+      "loss": 4.2504,
+      "step": 344576
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4349918728375886e-05,
+      "loss": 4.2573,
+      "step": 345088
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4341532780865366e-05,
+      "loss": 4.2324,
+      "step": 345600
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4333146833354846e-05,
+      "loss": 4.2547,
+      "step": 346112
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4324760885844326e-05,
+      "loss": 4.241,
+      "step": 346624
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4316374938333806e-05,
+      "loss": 4.2552,
+      "step": 347136
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4308005369627015e-05,
+      "loss": 4.2556,
+      "step": 347648
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4299619422116495e-05,
+      "loss": 4.2355,
+      "step": 348160
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.429124985340971e-05,
+      "loss": 4.2422,
+      "step": 348672
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.428286390589919e-05,
+      "loss": 4.2405,
+      "step": 349184
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4274477958388664e-05,
+      "loss": 4.2581,
+      "step": 349696
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4266092010878144e-05,
+      "loss": 4.2447,
+      "step": 350208
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.4257706063367624e-05,
+      "loss": 4.2428,
+      "step": 350720
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.424933649466084e-05,
+      "loss": 4.2392,
+      "step": 351232
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.424095054715032e-05,
+      "loss": 4.2446,
+      "step": 351744
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.42325645996398e-05,
+      "loss": 4.2385,
+      "step": 352256
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.422417865212928e-05,
+      "loss": 4.2249,
+      "step": 352768
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.421579270461876e-05,
+      "loss": 4.2393,
+      "step": 353280
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.420740675710824e-05,
+      "loss": 4.2441,
+      "step": 353792
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.419902080959772e-05,
+      "loss": 4.2448,
+      "step": 354304
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.41906348620872e-05,
+      "loss": 4.2268,
+      "step": 354816
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.418226529338041e-05,
+      "loss": 4.2318,
+      "step": 355328
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.417387934586989e-05,
+      "loss": 4.2382,
+      "step": 355840
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.416549339835937e-05,
+      "loss": 4.2428,
+      "step": 356352
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.415710745084885e-05,
+      "loss": 4.2411,
+      "step": 356864
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.414873788214206e-05,
+      "loss": 4.2377,
+      "step": 357376
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4140351934631544e-05,
+      "loss": 4.231,
+      "step": 357888
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4131965987121024e-05,
+      "loss": 4.2451,
+      "step": 358400
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4123580039610504e-05,
+      "loss": 4.2202,
+      "step": 358912
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.411521047090371e-05,
+      "loss": 4.2363,
+      "step": 359424
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.410684090219692e-05,
+      "loss": 4.2265,
+      "step": 359936
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.40984549546864e-05,
+      "loss": 4.2369,
+      "step": 360448
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.409008538597961e-05,
+      "loss": 4.2229,
+      "step": 360960
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.408169943846909e-05,
+      "loss": 4.2384,
+      "step": 361472
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.407331349095857e-05,
+      "loss": 4.2283,
+      "step": 361984
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.406492754344805e-05,
+      "loss": 4.2287,
+      "step": 362496
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.405654159593753e-05,
+      "loss": 4.2308,
+      "step": 363008
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.404815564842701e-05,
+      "loss": 4.2316,
+      "step": 363520
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.40397697009165e-05,
+      "loss": 4.2361,
+      "step": 364032
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.403138375340598e-05,
+      "loss": 4.2426,
+      "step": 364544
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.402301418469919e-05,
+      "loss": 4.2354,
+      "step": 365056
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.401462823718867e-05,
+      "loss": 4.2365,
+      "step": 365568
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.4006242289678147e-05,
+      "loss": 4.2286,
+      "step": 366080
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3997856342167627e-05,
+      "loss": 4.2254,
+      "step": 366592
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3989486773460836e-05,
+      "loss": 4.2242,
+      "step": 367104
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3981100825950316e-05,
+      "loss": 4.2343,
+      "step": 367616
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3972714878439796e-05,
+      "loss": 4.2259,
+      "step": 368128
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3964328930929276e-05,
+      "loss": 4.2312,
+      "step": 368640
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3955942983418755e-05,
+      "loss": 4.2255,
+      "step": 369152
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3947557035908235e-05,
+      "loss": 4.2331,
+      "step": 369664
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3939171088397715e-05,
+      "loss": 4.2256,
+      "step": 370176
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3930785140887195e-05,
+      "loss": 4.2269,
+      "step": 370688
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3922399193376675e-05,
+      "loss": 4.2239,
+      "step": 371200
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3914013245866155e-05,
+      "loss": 4.2352,
+      "step": 371712
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3905627298355635e-05,
+      "loss": 4.2262,
+      "step": 372224
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3897241350845115e-05,
+      "loss": 4.2171,
+      "step": 372736
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.388888816094206e-05,
+      "loss": 4.2106,
+      "step": 373248
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.388050221343154e-05,
+      "loss": 4.2296,
+      "step": 373760
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.387211626592102e-05,
+      "loss": 4.2332,
+      "step": 374272
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.38637303184105e-05,
+      "loss": 4.2313,
+      "step": 374784
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.385534437089997e-05,
+      "loss": 4.2314,
+      "step": 375296
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.384697480219319e-05,
+      "loss": 4.2308,
+      "step": 375808
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.383858885468267e-05,
+      "loss": 4.2251,
+      "step": 376320
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.383020290717215e-05,
+      "loss": 4.2312,
+      "step": 376832
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.382181695966163e-05,
+      "loss": 4.2168,
+      "step": 377344
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.381343101215111e-05,
+      "loss": 4.2257,
+      "step": 377856
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.380504506464059e-05,
+      "loss": 4.2238,
+      "step": 378368
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.379665911713007e-05,
+      "loss": 4.218,
+      "step": 378880
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.378827316961955e-05,
+      "loss": 4.2297,
+      "step": 379392
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.377988722210903e-05,
+      "loss": 4.2231,
+      "step": 379904
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.377150127459851e-05,
+      "loss": 4.22,
+      "step": 380416
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.376314808469545e-05,
+      "loss": 4.2207,
+      "step": 380928
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.375476213718493e-05,
+      "loss": 4.2215,
+      "step": 381440
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.223392963409424,
+      "eval_runtime": 304.2674,
+      "eval_samples_per_second": 1254.13,
+      "eval_steps_per_second": 39.193,
+      "step": 381600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.374637618967441e-05,
+      "loss": 4.2224,
+      "step": 381952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.373799024216389e-05,
+      "loss": 4.2202,
+      "step": 382464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.3729604294653374e-05,
+      "loss": 4.2256,
+      "step": 382976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.3721218347142853e-05,
+      "loss": 4.2159,
+      "step": 383488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.3712832399632333e-05,
+      "loss": 4.2295,
+      "step": 384000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.3704446452121813e-05,
+      "loss": 4.2167,
+      "step": 384512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.369606050461129e-05,
+      "loss": 4.2083,
+      "step": 385024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.368767455710077e-05,
+      "loss": 4.2064,
+      "step": 385536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.367928860959025e-05,
+      "loss": 4.2232,
+      "step": 386048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.367090266207973e-05,
+      "loss": 4.2242,
+      "step": 386560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.366251671456921e-05,
+      "loss": 4.2164,
+      "step": 387072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.3654130767058686e-05,
+      "loss": 4.222,
+      "step": 387584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.36457611983519e-05,
+      "loss": 4.2164,
+      "step": 388096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.363737525084138e-05,
+      "loss": 4.2169,
+      "step": 388608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.362898930333086e-05,
+      "loss": 4.2103,
+      "step": 389120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.3620603355820335e-05,
+      "loss": 4.2041,
+      "step": 389632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.361223378711355e-05,
+      "loss": 4.2146,
+      "step": 390144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.360384783960304e-05,
+      "loss": 4.2061,
+      "step": 390656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.359546189209251e-05,
+      "loss": 4.2206,
+      "step": 391168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.358707594458199e-05,
+      "loss": 4.222,
+      "step": 391680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.357868999707147e-05,
+      "loss": 4.2256,
+      "step": 392192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.357030404956095e-05,
+      "loss": 4.2133,
+      "step": 392704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.356191810205043e-05,
+      "loss": 4.2104,
+      "step": 393216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.355353215453991e-05,
+      "loss": 4.2154,
+      "step": 393728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.354514620702939e-05,
+      "loss": 4.2061,
+      "step": 394240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.353676025951887e-05,
+      "loss": 4.2092,
+      "step": 394752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.352837431200835e-05,
+      "loss": 4.193,
+      "step": 395264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.351998836449783e-05,
+      "loss": 4.2171,
+      "step": 395776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.351160241698731e-05,
+      "loss": 4.2037,
+      "step": 396288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.350323284828052e-05,
+      "loss": 4.1982,
+      "step": 396800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3494846900770006e-05,
+      "loss": 4.209,
+      "step": 397312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3486460953259486e-05,
+      "loss": 4.2122,
+      "step": 397824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3478091384552696e-05,
+      "loss": 4.2165,
+      "step": 398336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3469721815845905e-05,
+      "loss": 4.2107,
+      "step": 398848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3461335868335385e-05,
+      "loss": 4.2059,
+      "step": 399360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3452949920824865e-05,
+      "loss": 4.2144,
+      "step": 399872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3444563973314344e-05,
+      "loss": 4.2101,
+      "step": 400384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3436178025803824e-05,
+      "loss": 4.1945,
+      "step": 400896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3427792078293304e-05,
+      "loss": 4.1962,
+      "step": 401408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3419406130782784e-05,
+      "loss": 4.1968,
+      "step": 401920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3411020183272264e-05,
+      "loss": 4.2032,
+      "step": 402432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3402634235761744e-05,
+      "loss": 4.2099,
+      "step": 402944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3394248288251224e-05,
+      "loss": 4.2005,
+      "step": 403456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3385862340740704e-05,
+      "loss": 4.1964,
+      "step": 403968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3377476393230184e-05,
+      "loss": 4.2082,
+      "step": 404480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.33691068245234e-05,
+      "loss": 4.2011,
+      "step": 404992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.336072087701287e-05,
+      "loss": 4.2154,
+      "step": 405504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.335233492950235e-05,
+      "loss": 4.1961,
+      "step": 406016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.334396536079557e-05,
+      "loss": 4.1767,
+      "step": 406528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.333557941328505e-05,
+      "loss": 4.2148,
+      "step": 407040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.332719346577452e-05,
+      "loss": 4.2072,
+      "step": 407552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3318807518264e-05,
+      "loss": 4.208,
+      "step": 408064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.331042157075348e-05,
+      "loss": 4.2,
+      "step": 408576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.330203562324296e-05,
+      "loss": 4.1929,
+      "step": 409088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.329364967573244e-05,
+      "loss": 4.1946,
+      "step": 409600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.328526372822193e-05,
+      "loss": 4.1959,
+      "step": 410112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.327687778071141e-05,
+      "loss": 4.1911,
+      "step": 410624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.326850821200462e-05,
+      "loss": 4.1926,
+      "step": 411136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.32601222644941e-05,
+      "loss": 4.2002,
+      "step": 411648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.325173631698358e-05,
+      "loss": 4.1967,
+      "step": 412160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.324335036947306e-05,
+      "loss": 4.1864,
+      "step": 412672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.323498080076627e-05,
+      "loss": 4.2075,
+      "step": 413184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.322659485325575e-05,
+      "loss": 4.1855,
+      "step": 413696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.321820890574523e-05,
+      "loss": 4.1871,
+      "step": 414208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3209822958234707e-05,
+      "loss": 4.2012,
+      "step": 414720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3201437010724187e-05,
+      "loss": 4.2018,
+      "step": 415232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3193067442017396e-05,
+      "loss": 4.1824,
+      "step": 415744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3184681494506876e-05,
+      "loss": 4.1927,
+      "step": 416256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.317629554699636e-05,
+      "loss": 4.1749,
+      "step": 416768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.316790959948584e-05,
+      "loss": 4.1964,
+      "step": 417280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.315952365197532e-05,
+      "loss": 4.1976,
+      "step": 417792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.315115408326853e-05,
+      "loss": 4.1944,
+      "step": 418304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.314276813575801e-05,
+      "loss": 4.193,
+      "step": 418816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.313438218824749e-05,
+      "loss": 4.2027,
+      "step": 419328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.312599624073697e-05,
+      "loss": 4.2079,
+      "step": 419840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.311762667203018e-05,
+      "loss": 4.183,
+      "step": 420352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.310924072451966e-05,
+      "loss": 4.1966,
+      "step": 420864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.310085477700914e-05,
+      "loss": 4.2015,
+      "step": 421376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.309246882949862e-05,
+      "loss": 4.177,
+      "step": 421888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.308409926079183e-05,
+      "loss": 4.1987,
+      "step": 422400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3075713313281316e-05,
+      "loss": 4.1847,
+      "step": 422912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3067327365770796e-05,
+      "loss": 4.2008,
+      "step": 423424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3058941418260276e-05,
+      "loss": 4.1988,
+      "step": 423936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3050571849553485e-05,
+      "loss": 4.1832,
+      "step": 424448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3042185902042965e-05,
+      "loss": 4.1817,
+      "step": 424960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3033799954532445e-05,
+      "loss": 4.1874,
+      "step": 425472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3025414007021925e-05,
+      "loss": 4.2016,
+      "step": 425984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3017028059511405e-05,
+      "loss": 4.1871,
+      "step": 426496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3008642112000885e-05,
+      "loss": 4.1886,
+      "step": 427008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.300025616449036e-05,
+      "loss": 4.1848,
+      "step": 427520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.299187021697984e-05,
+      "loss": 4.1902,
+      "step": 428032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2983500648273054e-05,
+      "loss": 4.1846,
+      "step": 428544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.297513107956627e-05,
+      "loss": 4.1677,
+      "step": 429056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.296674513205575e-05,
+      "loss": 4.1847,
+      "step": 429568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.295835918454523e-05,
+      "loss": 4.1914,
+      "step": 430080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.294997323703471e-05,
+      "loss": 4.1895,
+      "step": 430592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.294158728952418e-05,
+      "loss": 4.1759,
+      "step": 431104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.293320134201366e-05,
+      "loss": 4.1766,
+      "step": 431616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.292483177330688e-05,
+      "loss": 4.1844,
+      "step": 432128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.291644582579636e-05,
+      "loss": 4.1873,
+      "step": 432640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.290805987828583e-05,
+      "loss": 4.1922,
+      "step": 433152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.289967393077531e-05,
+      "loss": 4.178,
+      "step": 433664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.289128798326479e-05,
+      "loss": 4.1813,
+      "step": 434176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.288290203575427e-05,
+      "loss": 4.1925,
+      "step": 434688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.287451608824375e-05,
+      "loss": 4.1698,
+      "step": 435200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.286613014073324e-05,
+      "loss": 4.1794,
+      "step": 435712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.285777695083018e-05,
+      "loss": 4.173,
+      "step": 436224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2849391003319656e-05,
+      "loss": 4.1856,
+      "step": 436736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2841005055809136e-05,
+      "loss": 4.1704,
+      "step": 437248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2832619108298616e-05,
+      "loss": 4.1868,
+      "step": 437760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2824233160788096e-05,
+      "loss": 4.1751,
+      "step": 438272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2815847213277576e-05,
+      "loss": 4.1788,
+      "step": 438784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2807461265767056e-05,
+      "loss": 4.1755,
+      "step": 439296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2799075318256536e-05,
+      "loss": 4.1814,
+      "step": 439808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2790705749549745e-05,
+      "loss": 4.1797,
+      "step": 440320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2782319802039225e-05,
+      "loss": 4.196,
+      "step": 440832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.277395023333244e-05,
+      "loss": 4.1832,
+      "step": 441344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.276556428582192e-05,
+      "loss": 4.1843,
+      "step": 441856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.27571783383114e-05,
+      "loss": 4.175,
+      "step": 442368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.274879239080088e-05,
+      "loss": 4.1805,
+      "step": 442880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.274040644329036e-05,
+      "loss": 4.1744,
+      "step": 443392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.273203687458357e-05,
+      "loss": 4.178,
+      "step": 443904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.272365092707305e-05,
+      "loss": 4.1769,
+      "step": 444416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.271526497956253e-05,
+      "loss": 4.1793,
+      "step": 444928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.270687903205201e-05,
+      "loss": 4.1782,
+      "step": 445440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.269849308454149e-05,
+      "loss": 4.1763,
+      "step": 445952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.269010713703097e-05,
+      "loss": 4.175,
+      "step": 446464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.268172118952045e-05,
+      "loss": 4.181,
+      "step": 446976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.267333524200993e-05,
+      "loss": 4.1732,
+      "step": 447488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2664965673303146e-05,
+      "loss": 4.18,
+      "step": 448000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2656579725792626e-05,
+      "loss": 4.1752,
+      "step": 448512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2648193778282105e-05,
+      "loss": 4.1695,
+      "step": 449024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2639824209575315e-05,
+      "loss": 4.1556,
+      "step": 449536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2631438262064795e-05,
+      "loss": 4.1811,
+      "step": 450048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2623052314554274e-05,
+      "loss": 4.1807,
+      "step": 450560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2614666367043754e-05,
+      "loss": 4.182,
+      "step": 451072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2606280419533234e-05,
+      "loss": 4.1802,
+      "step": 451584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2597894472022714e-05,
+      "loss": 4.1797,
+      "step": 452096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.258950852451219e-05,
+      "loss": 4.1736,
+      "step": 452608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.258112257700167e-05,
+      "loss": 4.186,
+      "step": 453120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.257275300829488e-05,
+      "loss": 4.1649,
+      "step": 453632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.256438343958809e-05,
+      "loss": 4.1749,
+      "step": 454144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.255599749207758e-05,
+      "loss": 4.1723,
+      "step": 454656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.254761154456706e-05,
+      "loss": 4.172,
+      "step": 455168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.253922559705654e-05,
+      "loss": 4.183,
+      "step": 455680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.253083964954602e-05,
+      "loss": 4.1689,
+      "step": 456192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.252245370203549e-05,
+      "loss": 4.1768,
+      "step": 456704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.251406775452497e-05,
+      "loss": 4.1684,
+      "step": 457216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.250569818581819e-05,
+      "loss": 4.1723,
+      "step": 457728
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.182225704193115,
+      "eval_runtime": 296.6807,
+      "eval_samples_per_second": 1286.201,
+      "eval_steps_per_second": 40.195,
+      "step": 457920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.249731223830766e-05,
+      "loss": 4.1707,
+      "step": 458240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.248892629079714e-05,
+      "loss": 4.1722,
+      "step": 458752
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.248054034328662e-05,
+      "loss": 4.1764,
+      "step": 459264
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.247217077457984e-05,
+      "loss": 4.167,
+      "step": 459776
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.246378482706932e-05,
+      "loss": 4.1818,
+      "step": 460288
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.24553988795588e-05,
+      "loss": 4.1674,
+      "step": 460800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.244701293204828e-05,
+      "loss": 4.1629,
+      "step": 461312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.243862698453776e-05,
+      "loss": 4.1573,
+      "step": 461824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.243024103702724e-05,
+      "loss": 4.1752,
+      "step": 462336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.2421871468320446e-05,
+      "loss": 4.1761,
+      "step": 462848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.2413485520809926e-05,
+      "loss": 4.167,
+      "step": 463360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.2405099573299406e-05,
+      "loss": 4.1747,
+      "step": 463872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.2396730004592615e-05,
+      "loss": 4.1676,
+      "step": 464384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.2388344057082095e-05,
+      "loss": 4.1686,
+      "step": 464896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.2379958109571575e-05,
+      "loss": 4.1595,
+      "step": 465408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.2371572162061055e-05,
+      "loss": 4.1576,
+      "step": 465920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.2363186214550535e-05,
+      "loss": 4.1659,
+      "step": 466432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.2354800267040015e-05,
+      "loss": 4.1633,
+      "step": 466944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.23464143195295e-05,
+      "loss": 4.1697,
+      "step": 467456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.233802837201898e-05,
+      "loss": 4.1734,
+      "step": 467968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.232964242450846e-05,
+      "loss": 4.1795,
+      "step": 468480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.232125647699794e-05,
+      "loss": 4.1668,
+      "step": 468992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.231288690829115e-05,
+      "loss": 4.1633,
+      "step": 469504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.230450096078063e-05,
+      "loss": 4.1645,
+      "step": 470016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.229611501327011e-05,
+      "loss": 4.1635,
+      "step": 470528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.228774544456332e-05,
+      "loss": 4.1623,
+      "step": 471040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.22793594970528e-05,
+      "loss": 4.1428,
+      "step": 471552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.227097354954228e-05,
+      "loss": 4.1719,
+      "step": 472064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.226258760203176e-05,
+      "loss": 4.1613,
+      "step": 472576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.225420165452124e-05,
+      "loss": 4.152,
+      "step": 473088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.224581570701072e-05,
+      "loss": 4.1614,
+      "step": 473600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.22374297595002e-05,
+      "loss": 4.167,
+      "step": 474112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.222904381198968e-05,
+      "loss": 4.1649,
+      "step": 474624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.222065786447916e-05,
+      "loss": 4.1665,
+      "step": 475136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.221227191696864e-05,
+      "loss": 4.1595,
+      "step": 475648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.220388596945812e-05,
+      "loss": 4.1674,
+      "step": 476160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.219551640075133e-05,
+      "loss": 4.1639,
+      "step": 476672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.218713045324081e-05,
+      "loss": 4.1487,
+      "step": 477184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.217874450573029e-05,
+      "loss": 4.1518,
+      "step": 477696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.217035855821977e-05,
+      "loss": 4.1503,
+      "step": 478208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.216197261070925e-05,
+      "loss": 4.1572,
+      "step": 478720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.215360304200246e-05,
+      "loss": 4.1661,
+      "step": 479232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.214521709449194e-05,
+      "loss": 4.1536,
+      "step": 479744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.2136831146981424e-05,
+      "loss": 4.1533,
+      "step": 480256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.2128445199470904e-05,
+      "loss": 4.1609,
+      "step": 480768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.2120059251960384e-05,
+      "loss": 4.1552,
+      "step": 481280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.2111673304449864e-05,
+      "loss": 4.1705,
+      "step": 481792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.2103287356939343e-05,
+      "loss": 4.1533,
+      "step": 482304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.2094901409428823e-05,
+      "loss": 4.1284,
+      "step": 482816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.20865154619183e-05,
+      "loss": 4.1694,
+      "step": 483328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.207814589321151e-05,
+      "loss": 4.1631,
+      "step": 483840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.206975994570099e-05,
+      "loss": 4.1627,
+      "step": 484352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.206137399819047e-05,
+      "loss": 4.1536,
+      "step": 484864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.205298805067995e-05,
+      "loss": 4.1489,
+      "step": 485376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.204461848197316e-05,
+      "loss": 4.1484,
+      "step": 485888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.203623253446264e-05,
+      "loss": 4.1537,
+      "step": 486400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.202784658695212e-05,
+      "loss": 4.1475,
+      "step": 486912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.20194606394416e-05,
+      "loss": 4.1459,
+      "step": 487424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.201107469193109e-05,
+      "loss": 4.1569,
+      "step": 487936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.20027051232243e-05,
+      "loss": 4.1532,
+      "step": 488448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.199431917571378e-05,
+      "loss": 4.1408,
+      "step": 488960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.198593322820326e-05,
+      "loss": 4.1626,
+      "step": 489472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.197754728069274e-05,
+      "loss": 4.1393,
+      "step": 489984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1969177711985946e-05,
+      "loss": 4.1467,
+      "step": 490496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1960791764475426e-05,
+      "loss": 4.1564,
+      "step": 491008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1952405816964906e-05,
+      "loss": 4.157,
+      "step": 491520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1944019869454386e-05,
+      "loss": 4.1387,
+      "step": 492032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.193563392194386e-05,
+      "loss": 4.1462,
+      "step": 492544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.192724797443334e-05,
+      "loss": 4.1302,
+      "step": 493056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1918862026922826e-05,
+      "loss": 4.1531,
+      "step": 493568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.191049245821604e-05,
+      "loss": 4.1529,
+      "step": 494080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1902106510705515e-05,
+      "loss": 4.1538,
+      "step": 494592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1893720563194995e-05,
+      "loss": 4.1479,
+      "step": 495104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1885334615684475e-05,
+      "loss": 4.1574,
+      "step": 495616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1876948668173955e-05,
+      "loss": 4.1657,
+      "step": 496128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1868562720663435e-05,
+      "loss": 4.1418,
+      "step": 496640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1860176773152915e-05,
+      "loss": 4.1523,
+      "step": 497152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1851790825642395e-05,
+      "loss": 4.1553,
+      "step": 497664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1843404878131875e-05,
+      "loss": 4.1372,
+      "step": 498176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.1835035309425084e-05,
+      "loss": 4.1527,
+      "step": 498688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.182666574071829e-05,
+      "loss": 4.1471,
+      "step": 499200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.181829617201151e-05,
+      "loss": 4.1542,
+      "step": 499712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.180991022450099e-05,
+      "loss": 4.157,
+      "step": 500224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.180152427699047e-05,
+      "loss": 4.1413,
+      "step": 500736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.179313832947995e-05,
+      "loss": 4.14,
+      "step": 501248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.178475238196943e-05,
+      "loss": 4.146,
+      "step": 501760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.177636643445891e-05,
+      "loss": 4.1525,
+      "step": 502272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.176798048694839e-05,
+      "loss": 4.1477,
+      "step": 502784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.175959453943787e-05,
+      "loss": 4.1449,
+      "step": 503296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.175120859192735e-05,
+      "loss": 4.1447,
+      "step": 503808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.174282264441683e-05,
+      "loss": 4.149,
+      "step": 504320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.173443669690631e-05,
+      "loss": 4.1399,
+      "step": 504832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.172605074939579e-05,
+      "loss": 4.1266,
+      "step": 505344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1717681180689e-05,
+      "loss": 4.1437,
+      "step": 505856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.170931161198221e-05,
+      "loss": 4.146,
+      "step": 506368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.170092566447169e-05,
+      "loss": 4.1469,
+      "step": 506880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.169253971696117e-05,
+      "loss": 4.1366,
+      "step": 507392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.168415376945065e-05,
+      "loss": 4.1347,
+      "step": 507904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.167576782194013e-05,
+      "loss": 4.1437,
+      "step": 508416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.166738187442961e-05,
+      "loss": 4.1443,
+      "step": 508928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.165899592691909e-05,
+      "loss": 4.1516,
+      "step": 509440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.16506263582123e-05,
+      "loss": 4.1375,
+      "step": 509952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.164224041070178e-05,
+      "loss": 4.1415,
+      "step": 510464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.163385446319126e-05,
+      "loss": 4.1487,
+      "step": 510976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.162546851568074e-05,
+      "loss": 4.1272,
+      "step": 511488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.161708256817022e-05,
+      "loss": 4.141,
+      "step": 512000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.160871299946343e-05,
+      "loss": 4.1257,
+      "step": 512512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.160032705195292e-05,
+      "loss": 4.1479,
+      "step": 513024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.15919411044424e-05,
+      "loss": 4.1285,
+      "step": 513536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.158355515693187e-05,
+      "loss": 4.1445,
+      "step": 514048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.157516920942135e-05,
+      "loss": 4.1347,
+      "step": 514560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.156678326191083e-05,
+      "loss": 4.1351,
+      "step": 515072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.155839731440031e-05,
+      "loss": 4.136,
+      "step": 515584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.155001136688979e-05,
+      "loss": 4.1418,
+      "step": 516096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.154162541937927e-05,
+      "loss": 4.1389,
+      "step": 516608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.153325585067248e-05,
+      "loss": 4.1531,
+      "step": 517120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1524886281965695e-05,
+      "loss": 4.1362,
+      "step": 517632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.151650033445517e-05,
+      "loss": 4.151,
+      "step": 518144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1508114386944655e-05,
+      "loss": 4.1347,
+      "step": 518656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1499728439434135e-05,
+      "loss": 4.1363,
+      "step": 519168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1491342491923615e-05,
+      "loss": 4.1364,
+      "step": 519680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1482972923216824e-05,
+      "loss": 4.133,
+      "step": 520192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1474586975706304e-05,
+      "loss": 4.1335,
+      "step": 520704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1466201028195784e-05,
+      "loss": 4.1396,
+      "step": 521216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1457815080685264e-05,
+      "loss": 4.1355,
+      "step": 521728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1449429133174744e-05,
+      "loss": 4.1391,
+      "step": 522240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1441043185664224e-05,
+      "loss": 4.1349,
+      "step": 522752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1432657238153704e-05,
+      "loss": 4.1398,
+      "step": 523264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1424271290643184e-05,
+      "loss": 4.1355,
+      "step": 523776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1415885343132664e-05,
+      "loss": 4.1359,
+      "step": 524288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.140751577442587e-05,
+      "loss": 4.1326,
+      "step": 524800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.139912982691535e-05,
+      "loss": 4.1322,
+      "step": 525312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.139074387940483e-05,
+      "loss": 4.1165,
+      "step": 525824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.138237431069805e-05,
+      "loss": 4.1411,
+      "step": 526336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.137398836318753e-05,
+      "loss": 4.1417,
+      "step": 526848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.136560241567701e-05,
+      "loss": 4.143,
+      "step": 527360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.135721646816649e-05,
+      "loss": 4.1358,
+      "step": 527872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.13488468994597e-05,
+      "loss": 4.1425,
+      "step": 528384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.134046095194918e-05,
+      "loss": 4.1368,
+      "step": 528896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.133207500443866e-05,
+      "loss": 4.1461,
+      "step": 529408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.132368905692814e-05,
+      "loss": 4.1243,
+      "step": 529920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.131530310941762e-05,
+      "loss": 4.133,
+      "step": 530432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.13069171619071e-05,
+      "loss": 4.1352,
+      "step": 530944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.129853121439658e-05,
+      "loss": 4.1335,
+      "step": 531456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.129016164568979e-05,
+      "loss": 4.1386,
+      "step": 531968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1281775698179273e-05,
+      "loss": 4.1323,
+      "step": 532480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.1273389750668753e-05,
+      "loss": 4.1365,
+      "step": 532992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.126502018196196e-05,
+      "loss": 4.1288,
+      "step": 533504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.125663423445144e-05,
+      "loss": 4.1396,
+      "step": 534016
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.151095867156982,
+      "eval_runtime": 302.9133,
+      "eval_samples_per_second": 1259.737,
+      "eval_steps_per_second": 39.368,
+      "step": 534240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.124824828694092e-05,
+      "loss": 4.1378,
+      "step": 534528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.12398623394304e-05,
+      "loss": 4.1374,
+      "step": 535040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.123147639191988e-05,
+      "loss": 4.1316,
+      "step": 535552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1223090444409356e-05,
+      "loss": 4.1271,
+      "step": 536064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1214704496898835e-05,
+      "loss": 4.1405,
+      "step": 536576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1206318549388315e-05,
+      "loss": 4.1303,
+      "step": 537088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1197932601877795e-05,
+      "loss": 4.1248,
+      "step": 537600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1189546654367275e-05,
+      "loss": 4.1221,
+      "step": 538112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1181160706856755e-05,
+      "loss": 4.1309,
+      "step": 538624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.117277475934624e-05,
+      "loss": 4.1411,
+      "step": 539136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.116438881183572e-05,
+      "loss": 4.1277,
+      "step": 539648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.11560028643252e-05,
+      "loss": 4.134,
+      "step": 540160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.114763329561841e-05,
+      "loss": 4.1363,
+      "step": 540672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.113924734810789e-05,
+      "loss": 4.1248,
+      "step": 541184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.113086140059737e-05,
+      "loss": 4.123,
+      "step": 541696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.112247545308685e-05,
+      "loss": 4.1187,
+      "step": 542208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.111408950557633e-05,
+      "loss": 4.1272,
+      "step": 542720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.110570355806581e-05,
+      "loss": 4.1233,
+      "step": 543232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.109733398935902e-05,
+      "loss": 4.1279,
+      "step": 543744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.10889480418485e-05,
+      "loss": 4.1382,
+      "step": 544256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.108056209433798e-05,
+      "loss": 4.1413,
+      "step": 544768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.107217614682746e-05,
+      "loss": 4.1292,
+      "step": 545280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1063806578120676e-05,
+      "loss": 4.1244,
+      "step": 545792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1055420630610156e-05,
+      "loss": 4.1246,
+      "step": 546304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1047034683099636e-05,
+      "loss": 4.1267,
+      "step": 546816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1038648735589115e-05,
+      "loss": 4.125,
+      "step": 547328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1030262788078595e-05,
+      "loss": 4.1083,
+      "step": 547840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.102187684056807e-05,
+      "loss": 4.132,
+      "step": 548352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.101349089305755e-05,
+      "loss": 4.1243,
+      "step": 548864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.100510494554703e-05,
+      "loss": 4.1151,
+      "step": 549376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.099671899803651e-05,
+      "loss": 4.1179,
+      "step": 549888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.098833305052599e-05,
+      "loss": 4.1306,
+      "step": 550400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.097994710301547e-05,
+      "loss": 4.1305,
+      "step": 550912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.097156115550495e-05,
+      "loss": 4.1268,
+      "step": 551424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.096317520799443e-05,
+      "loss": 4.1222,
+      "step": 551936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0954805639287644e-05,
+      "loss": 4.1281,
+      "step": 552448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0946419691777124e-05,
+      "loss": 4.1301,
+      "step": 552960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0938033744266604e-05,
+      "loss": 4.1124,
+      "step": 553472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0929647796756084e-05,
+      "loss": 4.1139,
+      "step": 553984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.092127822804929e-05,
+      "loss": 4.1137,
+      "step": 554496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.091289228053877e-05,
+      "loss": 4.1198,
+      "step": 555008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.090450633302825e-05,
+      "loss": 4.1244,
+      "step": 555520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.089612038551773e-05,
+      "loss": 4.1224,
+      "step": 556032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.088775081681094e-05,
+      "loss": 4.1147,
+      "step": 556544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.087936486930042e-05,
+      "loss": 4.1236,
+      "step": 557056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.08709789217899e-05,
+      "loss": 4.1172,
+      "step": 557568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.086260935308311e-05,
+      "loss": 4.1295,
+      "step": 558080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.08542234055726e-05,
+      "loss": 4.1209,
+      "step": 558592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.084583745806208e-05,
+      "loss": 4.0924,
+      "step": 559104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.083745151055156e-05,
+      "loss": 4.1304,
+      "step": 559616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.082908194184477e-05,
+      "loss": 4.1247,
+      "step": 560128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.082069599433425e-05,
+      "loss": 4.1281,
+      "step": 560640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.081231004682373e-05,
+      "loss": 4.1139,
+      "step": 561152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.080392409931321e-05,
+      "loss": 4.1109,
+      "step": 561664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0795554530606416e-05,
+      "loss": 4.115,
+      "step": 562176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0787168583095896e-05,
+      "loss": 4.1154,
+      "step": 562688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0778782635585376e-05,
+      "loss": 4.1159,
+      "step": 563200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0770396688074856e-05,
+      "loss": 4.1049,
+      "step": 563712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0762010740564336e-05,
+      "loss": 4.1199,
+      "step": 564224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.075364117185755e-05,
+      "loss": 4.1197,
+      "step": 564736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.074525522434703e-05,
+      "loss": 4.1039,
+      "step": 565248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.073686927683651e-05,
+      "loss": 4.1273,
+      "step": 565760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.072848332932599e-05,
+      "loss": 4.1044,
+      "step": 566272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.072009738181547e-05,
+      "loss": 4.1079,
+      "step": 566784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.071171143430495e-05,
+      "loss": 4.12,
+      "step": 567296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.070332548679443e-05,
+      "loss": 4.1217,
+      "step": 567808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0694939539283904e-05,
+      "loss": 4.1043,
+      "step": 568320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0686553591773384e-05,
+      "loss": 4.112,
+      "step": 568832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.06781840230666e-05,
+      "loss": 4.0948,
+      "step": 569344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.066981445435981e-05,
+      "loss": 4.1175,
+      "step": 569856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.066142850684929e-05,
+      "loss": 4.1169,
+      "step": 570368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.065304255933877e-05,
+      "loss": 4.1188,
+      "step": 570880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.064465661182825e-05,
+      "loss": 4.1108,
+      "step": 571392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.063627066431773e-05,
+      "loss": 4.1206,
+      "step": 571904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.062788471680721e-05,
+      "loss": 4.1337,
+      "step": 572416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.061949876929669e-05,
+      "loss": 4.1033,
+      "step": 572928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.061111282178617e-05,
+      "loss": 4.1167,
+      "step": 573440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.060274325307938e-05,
+      "loss": 4.1224,
+      "step": 573952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.059435730556886e-05,
+      "loss": 4.1028,
+      "step": 574464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0585987736862074e-05,
+      "loss": 4.1174,
+      "step": 574976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0577601789351554e-05,
+      "loss": 4.1145,
+      "step": 575488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.056921584184103e-05,
+      "loss": 4.1169,
+      "step": 576000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.056082989433051e-05,
+      "loss": 4.1245,
+      "step": 576512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.055244394681999e-05,
+      "loss": 4.1018,
+      "step": 577024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.05440743781132e-05,
+      "loss": 4.1063,
+      "step": 577536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.053568843060268e-05,
+      "loss": 4.1101,
+      "step": 578048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.052730248309216e-05,
+      "loss": 4.1189,
+      "step": 578560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.051891653558164e-05,
+      "loss": 4.1094,
+      "step": 579072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.051053058807112e-05,
+      "loss": 4.1153,
+      "step": 579584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.05021446405606e-05,
+      "loss": 4.1125,
+      "step": 580096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.049375869305008e-05,
+      "loss": 4.108,
+      "step": 580608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.048537274553956e-05,
+      "loss": 4.1082,
+      "step": 581120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.047698679802904e-05,
+      "loss": 4.0917,
+      "step": 581632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.046861722932225e-05,
+      "loss": 4.1069,
+      "step": 582144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.046023128181173e-05,
+      "loss": 4.1113,
+      "step": 582656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.045184533430121e-05,
+      "loss": 4.1144,
+      "step": 583168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.044345938679069e-05,
+      "loss": 4.1025,
+      "step": 583680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.043508981808391e-05,
+      "loss": 4.0981,
+      "step": 584192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.042670387057339e-05,
+      "loss": 4.1099,
+      "step": 584704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.041831792306287e-05,
+      "loss": 4.1094,
+      "step": 585216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.040993197555235e-05,
+      "loss": 4.1136,
+      "step": 585728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0401562406845556e-05,
+      "loss": 4.1067,
+      "step": 586240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0393176459335036e-05,
+      "loss": 4.1086,
+      "step": 586752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0384790511824516e-05,
+      "loss": 4.1078,
+      "step": 587264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0376404564313996e-05,
+      "loss": 4.0989,
+      "step": 587776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0368018616803476e-05,
+      "loss": 4.1026,
+      "step": 588288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0359649048096685e-05,
+      "loss": 4.0943,
+      "step": 588800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0351263100586165e-05,
+      "loss": 4.1144,
+      "step": 589312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0342877153075645e-05,
+      "loss": 4.0942,
+      "step": 589824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0334491205565125e-05,
+      "loss": 4.1091,
+      "step": 590336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0326105258054605e-05,
+      "loss": 4.0993,
+      "step": 590848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.031771931054409e-05,
+      "loss": 4.1058,
+      "step": 591360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0309333363033565e-05,
+      "loss": 4.1004,
+      "step": 591872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0300947415523045e-05,
+      "loss": 4.1054,
+      "step": 592384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.029257784681626e-05,
+      "loss": 4.1041,
+      "step": 592896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.028420827810947e-05,
+      "loss": 4.1212,
+      "step": 593408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.027583870940268e-05,
+      "loss": 4.1014,
+      "step": 593920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.026745276189216e-05,
+      "loss": 4.1165,
+      "step": 594432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.025906681438164e-05,
+      "loss": 4.1,
+      "step": 594944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.025068086687112e-05,
+      "loss": 4.1029,
+      "step": 595456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.02422949193606e-05,
+      "loss": 4.1034,
+      "step": 595968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.023390897185008e-05,
+      "loss": 4.1052,
+      "step": 596480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.022552302433956e-05,
+      "loss": 4.0976,
+      "step": 596992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.021713707682904e-05,
+      "loss": 4.1059,
+      "step": 597504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0208767508122255e-05,
+      "loss": 4.0991,
+      "step": 598016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0200381560611735e-05,
+      "loss": 4.1056,
+      "step": 598528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0191995613101214e-05,
+      "loss": 4.1064,
+      "step": 599040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.018360966559069e-05,
+      "loss": 4.1011,
+      "step": 599552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.017522371808017e-05,
+      "loss": 4.1039,
+      "step": 600064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.016683777056965e-05,
+      "loss": 4.1089,
+      "step": 600576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.015845182305913e-05,
+      "loss": 4.0954,
+      "step": 601088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.015006587554861e-05,
+      "loss": 4.1003,
+      "step": 601600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.014167992803809e-05,
+      "loss": 4.0807,
+      "step": 602112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0133310359331297e-05,
+      "loss": 4.1079,
+      "step": 602624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.012492441182078e-05,
+      "loss": 4.1094,
+      "step": 603136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.011653846431026e-05,
+      "loss": 4.1099,
+      "step": 603648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.010815251679974e-05,
+      "loss": 4.102,
+      "step": 604160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.009978294809295e-05,
+      "loss": 4.1076,
+      "step": 604672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.009139700058243e-05,
+      "loss": 4.1042,
+      "step": 605184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.008301105307191e-05,
+      "loss": 4.1109,
+      "step": 605696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.007462510556139e-05,
+      "loss": 4.0916,
+      "step": 606208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.006623915805087e-05,
+      "loss": 4.1016,
+      "step": 606720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.005786958934408e-05,
+      "loss": 4.0999,
+      "step": 607232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.004950002063729e-05,
+      "loss": 4.1024,
+      "step": 607744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.004111407312677e-05,
+      "loss": 4.104,
+      "step": 608256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.003272812561625e-05,
+      "loss": 4.0965,
+      "step": 608768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.002434217810574e-05,
+      "loss": 4.1087,
+      "step": 609280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0015972609398946e-05,
+      "loss": 4.0947,
+      "step": 609792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0007586661888426e-05,
+      "loss": 4.1084,
+      "step": 610304
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.126674652099609,
+      "eval_runtime": 293.4932,
+      "eval_samples_per_second": 1300.17,
+      "eval_steps_per_second": 40.631,
+      "step": 610560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9999200714377906e-05,
+      "loss": 4.0936,
+      "step": 610816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9990814766867386e-05,
+      "loss": 4.1029,
+      "step": 611328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9982428819356866e-05,
+      "loss": 4.1028,
+      "step": 611840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9974059250650075e-05,
+      "loss": 4.0966,
+      "step": 612352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9965673303139555e-05,
+      "loss": 4.1093,
+      "step": 612864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9957287355629035e-05,
+      "loss": 4.0983,
+      "step": 613376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9948901408118515e-05,
+      "loss": 4.0914,
+      "step": 613888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9940515460607995e-05,
+      "loss": 4.0891,
+      "step": 614400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9932129513097475e-05,
+      "loss": 4.098,
+      "step": 614912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9923743565586955e-05,
+      "loss": 4.11,
+      "step": 615424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9915357618076435e-05,
+      "loss": 4.101,
+      "step": 615936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.990698804936965e-05,
+      "loss": 4.101,
+      "step": 616448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.989861848066286e-05,
+      "loss": 4.1029,
+      "step": 616960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.989023253315234e-05,
+      "loss": 4.0918,
+      "step": 617472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.988184658564182e-05,
+      "loss": 4.0902,
+      "step": 617984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.98734606381313e-05,
+      "loss": 4.0873,
+      "step": 618496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.986507469062078e-05,
+      "loss": 4.097,
+      "step": 619008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.985668874311026e-05,
+      "loss": 4.0915,
+      "step": 619520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.984831917440347e-05,
+      "loss": 4.0966,
+      "step": 620032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.983993322689295e-05,
+      "loss": 4.1054,
+      "step": 620544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.983154727938243e-05,
+      "loss": 4.1071,
+      "step": 621056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.982316133187191e-05,
+      "loss": 4.0949,
+      "step": 621568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.981477538436139e-05,
+      "loss": 4.096,
+      "step": 622080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.980638943685087e-05,
+      "loss": 4.0927,
+      "step": 622592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.979800348934035e-05,
+      "loss": 4.0948,
+      "step": 623104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.978961754182983e-05,
+      "loss": 4.0934,
+      "step": 623616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.978123159431931e-05,
+      "loss": 4.0799,
+      "step": 624128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.977284564680879e-05,
+      "loss": 4.102,
+      "step": 624640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9764476078102e-05,
+      "loss": 4.0947,
+      "step": 625152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.975609013059148e-05,
+      "loss": 4.0806,
+      "step": 625664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.974770418308096e-05,
+      "loss": 4.0877,
+      "step": 626176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.973931823557044e-05,
+      "loss": 4.1008,
+      "step": 626688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.973096504566738e-05,
+      "loss": 4.1025,
+      "step": 627200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.972257909815686e-05,
+      "loss": 4.0958,
+      "step": 627712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.971419315064634e-05,
+      "loss": 4.0917,
+      "step": 628224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.970580720313582e-05,
+      "loss": 4.0915,
+      "step": 628736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.96974212556253e-05,
+      "loss": 4.1035,
+      "step": 629248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.968903530811478e-05,
+      "loss": 4.0811,
+      "step": 629760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.968064936060426e-05,
+      "loss": 4.0799,
+      "step": 630272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.967227979189747e-05,
+      "loss": 4.0861,
+      "step": 630784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.966389384438695e-05,
+      "loss": 4.0863,
+      "step": 631296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.965552427568017e-05,
+      "loss": 4.0944,
+      "step": 631808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.964713832816964e-05,
+      "loss": 4.0854,
+      "step": 632320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.963875238065912e-05,
+      "loss": 4.092,
+      "step": 632832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.96303664331486e-05,
+      "loss": 4.087,
+      "step": 633344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.962198048563808e-05,
+      "loss": 4.0841,
+      "step": 633856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.961359453812756e-05,
+      "loss": 4.1034,
+      "step": 634368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9605208590617046e-05,
+      "loss": 4.088,
+      "step": 634880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9596822643106526e-05,
+      "loss": 4.0668,
+      "step": 635392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9588436695596006e-05,
+      "loss": 4.0985,
+      "step": 635904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9580067126889215e-05,
+      "loss": 4.0908,
+      "step": 636416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9571681179378695e-05,
+      "loss": 4.0995,
+      "step": 636928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9563295231868175e-05,
+      "loss": 4.0851,
+      "step": 637440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9554909284357655e-05,
+      "loss": 4.0802,
+      "step": 637952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9546523336847135e-05,
+      "loss": 4.0838,
+      "step": 638464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9538137389336615e-05,
+      "loss": 4.0815,
+      "step": 638976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9529751441826095e-05,
+      "loss": 4.0901,
+      "step": 639488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9521365494315575e-05,
+      "loss": 4.0714,
+      "step": 640000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9512979546805055e-05,
+      "loss": 4.0936,
+      "step": 640512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9504626356902e-05,
+      "loss": 4.0866,
+      "step": 641024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.949625678819521e-05,
+      "loss": 4.0756,
+      "step": 641536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.948787084068469e-05,
+      "loss": 4.0943,
+      "step": 642048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.947948489317417e-05,
+      "loss": 4.0733,
+      "step": 642560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.947109894566365e-05,
+      "loss": 4.0764,
+      "step": 643072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.946271299815313e-05,
+      "loss": 4.0887,
+      "step": 643584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.945432705064261e-05,
+      "loss": 4.0953,
+      "step": 644096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.944594110313209e-05,
+      "loss": 4.0695,
+      "step": 644608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.943755515562157e-05,
+      "loss": 4.0872,
+      "step": 645120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.942916920811105e-05,
+      "loss": 4.0659,
+      "step": 645632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.942078326060053e-05,
+      "loss": 4.0847,
+      "step": 646144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.941239731309e-05,
+      "loss": 4.0904,
+      "step": 646656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.940402774438322e-05,
+      "loss": 4.0812,
+      "step": 647168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.93956417968727e-05,
+      "loss": 4.0863,
+      "step": 647680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9387255849362185e-05,
+      "loss": 4.0876,
+      "step": 648192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.937886990185166e-05,
+      "loss": 4.1052,
+      "step": 648704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.937048395434114e-05,
+      "loss": 4.0718,
+      "step": 649216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.936209800683062e-05,
+      "loss": 4.0876,
+      "step": 649728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.93537120593201e-05,
+      "loss": 4.0865,
+      "step": 650240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.934532611180958e-05,
+      "loss": 4.0792,
+      "step": 650752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.933695654310279e-05,
+      "loss": 4.0866,
+      "step": 651264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9328586974396e-05,
+      "loss": 4.0823,
+      "step": 651776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9320201026885476e-05,
+      "loss": 4.0877,
+      "step": 652288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9311815079374956e-05,
+      "loss": 4.0952,
+      "step": 652800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9303429131864436e-05,
+      "loss": 4.07,
+      "step": 653312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.929505956315765e-05,
+      "loss": 4.0784,
+      "step": 653824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.928667361564713e-05,
+      "loss": 4.0796,
+      "step": 654336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.927830404694035e-05,
+      "loss": 4.0932,
+      "step": 654848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.926991809942983e-05,
+      "loss": 4.0761,
+      "step": 655360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.92615321519193e-05,
+      "loss": 4.0847,
+      "step": 655872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.925314620440878e-05,
+      "loss": 4.0839,
+      "step": 656384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.924476025689826e-05,
+      "loss": 4.0779,
+      "step": 656896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.923637430938774e-05,
+      "loss": 4.0781,
+      "step": 657408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.922798836187722e-05,
+      "loss": 4.0621,
+      "step": 657920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.92196024143667e-05,
+      "loss": 4.0772,
+      "step": 658432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.921121646685618e-05,
+      "loss": 4.081,
+      "step": 658944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.920283051934566e-05,
+      "loss": 4.0867,
+      "step": 659456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.919444457183514e-05,
+      "loss": 4.0726,
+      "step": 659968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9186075003128356e-05,
+      "loss": 4.0684,
+      "step": 660480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9177689055617836e-05,
+      "loss": 4.0772,
+      "step": 660992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9169303108107316e-05,
+      "loss": 4.0831,
+      "step": 661504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9160933539400525e-05,
+      "loss": 4.0847,
+      "step": 662016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9152547591890005e-05,
+      "loss": 4.0784,
+      "step": 662528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9144161644379485e-05,
+      "loss": 4.0815,
+      "step": 663040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9135775696868965e-05,
+      "loss": 4.0748,
+      "step": 663552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9127389749358445e-05,
+      "loss": 4.0743,
+      "step": 664064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9119003801847925e-05,
+      "loss": 4.0694,
+      "step": 664576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9110617854337405e-05,
+      "loss": 4.0712,
+      "step": 665088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9102231906826885e-05,
+      "loss": 4.0842,
+      "step": 665600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9093845959316365e-05,
+      "loss": 4.0665,
+      "step": 666112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.908546001180584e-05,
+      "loss": 4.0768,
+      "step": 666624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9077074064295325e-05,
+      "loss": 4.0678,
+      "step": 667136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9068688116784804e-05,
+      "loss": 4.0797,
+      "step": 667648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9060318548078014e-05,
+      "loss": 4.0723,
+      "step": 668160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9051932600567494e-05,
+      "loss": 4.0758,
+      "step": 668672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9043546653056974e-05,
+      "loss": 4.0777,
+      "step": 669184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9035160705546453e-05,
+      "loss": 4.0903,
+      "step": 669696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.902679113683966e-05,
+      "loss": 4.0761,
+      "step": 670208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.901840518932914e-05,
+      "loss": 4.0853,
+      "step": 670720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.901001924181862e-05,
+      "loss": 4.0743,
+      "step": 671232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.900164967311184e-05,
+      "loss": 4.0742,
+      "step": 671744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.899326372560131e-05,
+      "loss": 4.0771,
+      "step": 672256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.898487777809079e-05,
+      "loss": 4.077,
+      "step": 672768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.897649183058028e-05,
+      "loss": 4.069,
+      "step": 673280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.896810588306976e-05,
+      "loss": 4.0773,
+      "step": 673792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.895971993555924e-05,
+      "loss": 4.0718,
+      "step": 674304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.895133398804872e-05,
+      "loss": 4.0775,
+      "step": 674816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.894296441934193e-05,
+      "loss": 4.0734,
+      "step": 675328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.893457847183141e-05,
+      "loss": 4.0779,
+      "step": 675840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.892619252432089e-05,
+      "loss": 4.0745,
+      "step": 676352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.891780657681037e-05,
+      "loss": 4.0829,
+      "step": 676864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.890942062929985e-05,
+      "loss": 4.0674,
+      "step": 677376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8901051060593056e-05,
+      "loss": 4.0723,
+      "step": 677888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8892665113082536e-05,
+      "loss": 4.0502,
+      "step": 678400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8884279165572016e-05,
+      "loss": 4.0826,
+      "step": 678912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8875893218061496e-05,
+      "loss": 4.0789,
+      "step": 679424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8867507270550976e-05,
+      "loss": 4.0862,
+      "step": 679936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.885912132304046e-05,
+      "loss": 4.0705,
+      "step": 680448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.885073537552994e-05,
+      "loss": 4.0822,
+      "step": 680960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.884234942801942e-05,
+      "loss": 4.0757,
+      "step": 681472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.883397985931263e-05,
+      "loss": 4.0855,
+      "step": 681984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.882561029060584e-05,
+      "loss": 4.0631,
+      "step": 682496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.881722434309532e-05,
+      "loss": 4.0717,
+      "step": 683008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.88088383955848e-05,
+      "loss": 4.0707,
+      "step": 683520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.880045244807428e-05,
+      "loss": 4.0815,
+      "step": 684032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.879206650056376e-05,
+      "loss": 4.0712,
+      "step": 684544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.878368055305324e-05,
+      "loss": 4.0698,
+      "step": 685056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.877531098434645e-05,
+      "loss": 4.0778,
+      "step": 685568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.876692503683593e-05,
+      "loss": 4.0731,
+      "step": 686080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8758539089325416e-05,
+      "loss": 4.0794,
+      "step": 686592
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.106430530548096,
+      "eval_runtime": 294.2988,
+      "eval_samples_per_second": 1296.611,
+      "eval_steps_per_second": 40.52,
+      "step": 686880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8750153141814896e-05,
+      "loss": 4.0702,
+      "step": 687104
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8741767194304376e-05,
+      "loss": 4.0791,
+      "step": 687616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.873338124679385e-05,
+      "loss": 4.0694,
+      "step": 688128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.872499529928333e-05,
+      "loss": 4.0728,
+      "step": 688640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.871660935177281e-05,
+      "loss": 4.086,
+      "step": 689152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.870822340426229e-05,
+      "loss": 4.0691,
+      "step": 689664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.869983745675177e-05,
+      "loss": 4.0649,
+      "step": 690176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.869145150924125e-05,
+      "loss": 4.0624,
+      "step": 690688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.868306556173073e-05,
+      "loss": 4.0718,
+      "step": 691200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.867467961422021e-05,
+      "loss": 4.0812,
+      "step": 691712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.866629366670969e-05,
+      "loss": 4.0697,
+      "step": 692224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.865790771919917e-05,
+      "loss": 4.0748,
+      "step": 692736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.864953815049238e-05,
+      "loss": 4.077,
+      "step": 693248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8641152202981865e-05,
+      "loss": 4.0674,
+      "step": 693760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8632766255471345e-05,
+      "loss": 4.0622,
+      "step": 694272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8624380307960825e-05,
+      "loss": 4.0591,
+      "step": 694784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8615994360450305e-05,
+      "loss": 4.0721,
+      "step": 695296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8607608412939785e-05,
+      "loss": 4.0668,
+      "step": 695808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8599222465429265e-05,
+      "loss": 4.0639,
+      "step": 696320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.859083651791874e-05,
+      "loss": 4.0778,
+      "step": 696832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.858245057040822e-05,
+      "loss": 4.0814,
+      "step": 697344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.85740646228977e-05,
+      "loss": 4.0696,
+      "step": 697856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.856567867538718e-05,
+      "loss": 4.0678,
+      "step": 698368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.855730910668039e-05,
+      "loss": 4.0685,
+      "step": 698880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.854892315916987e-05,
+      "loss": 4.0693,
+      "step": 699392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.854053721165935e-05,
+      "loss": 4.0661,
+      "step": 699904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8532151264148833e-05,
+      "loss": 4.052,
+      "step": 700416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.852376531663831e-05,
+      "loss": 4.0713,
+      "step": 700928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.851537936912779e-05,
+      "loss": 4.0723,
+      "step": 701440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.850699342161727e-05,
+      "loss": 4.054,
+      "step": 701952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.849860747410675e-05,
+      "loss": 4.0617,
+      "step": 702464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.849023790539996e-05,
+      "loss": 4.0699,
+      "step": 702976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.848186833669317e-05,
+      "loss": 4.0735,
+      "step": 703488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.847348238918265e-05,
+      "loss": 4.0728,
+      "step": 704000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.846509644167213e-05,
+      "loss": 4.0693,
+      "step": 704512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.845671049416161e-05,
+      "loss": 4.0602,
+      "step": 705024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.844834092545482e-05,
+      "loss": 4.0806,
+      "step": 705536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.84399549779443e-05,
+      "loss": 4.0524,
+      "step": 706048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.843156903043379e-05,
+      "loss": 4.0542,
+      "step": 706560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.842318308292327e-05,
+      "loss": 4.0572,
+      "step": 707072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.841479713541275e-05,
+      "loss": 4.0616,
+      "step": 707584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.840641118790223e-05,
+      "loss": 4.0666,
+      "step": 708096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.839802524039171e-05,
+      "loss": 4.061,
+      "step": 708608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8389655671684916e-05,
+      "loss": 4.0662,
+      "step": 709120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8381269724174396e-05,
+      "loss": 4.0632,
+      "step": 709632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8372883776663876e-05,
+      "loss": 4.0607,
+      "step": 710144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8364497829153356e-05,
+      "loss": 4.0719,
+      "step": 710656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8356128260446565e-05,
+      "loss": 4.0687,
+      "step": 711168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8347742312936045e-05,
+      "loss": 4.04,
+      "step": 711680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8339356365425525e-05,
+      "loss": 4.0695,
+      "step": 712192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8330970417915005e-05,
+      "loss": 4.0673,
+      "step": 712704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8322584470404485e-05,
+      "loss": 4.0745,
+      "step": 713216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.831419852289397e-05,
+      "loss": 4.058,
+      "step": 713728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.830581257538345e-05,
+      "loss": 4.0569,
+      "step": 714240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8297426627872925e-05,
+      "loss": 4.0537,
+      "step": 714752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.828905705916614e-05,
+      "loss": 4.0577,
+      "step": 715264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.828067111165562e-05,
+      "loss": 4.0629,
+      "step": 715776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.82722851641451e-05,
+      "loss": 4.0486,
+      "step": 716288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8263899216634574e-05,
+      "loss": 4.0645,
+      "step": 716800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8255513269124054e-05,
+      "loss": 4.0621,
+      "step": 717312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.824714370041727e-05,
+      "loss": 4.0532,
+      "step": 717824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.823875775290675e-05,
+      "loss": 4.0639,
+      "step": 718336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.823037180539622e-05,
+      "loss": 4.0494,
+      "step": 718848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.822198585788571e-05,
+      "loss": 4.0546,
+      "step": 719360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8213616289178925e-05,
+      "loss": 4.0574,
+      "step": 719872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.82052303416684e-05,
+      "loss": 4.0696,
+      "step": 720384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.819684439415788e-05,
+      "loss": 4.0493,
+      "step": 720896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.818845844664736e-05,
+      "loss": 4.0637,
+      "step": 721408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.818007249913684e-05,
+      "loss": 4.0366,
+      "step": 721920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.817168655162632e-05,
+      "loss": 4.0609,
+      "step": 722432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.816331698291953e-05,
+      "loss": 4.0629,
+      "step": 722944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.815493103540901e-05,
+      "loss": 4.0615,
+      "step": 723456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.814654508789849e-05,
+      "loss": 4.0573,
+      "step": 723968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.813815914038797e-05,
+      "loss": 4.0646,
+      "step": 724480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8129789571681176e-05,
+      "loss": 4.0782,
+      "step": 724992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.812142000297439e-05,
+      "loss": 4.0501,
+      "step": 725504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.811303405546387e-05,
+      "loss": 4.0635,
+      "step": 726016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.810464810795335e-05,
+      "loss": 4.0564,
+      "step": 726528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.809626216044283e-05,
+      "loss": 4.0577,
+      "step": 727040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.808787621293231e-05,
+      "loss": 4.062,
+      "step": 727552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.807949026542179e-05,
+      "loss": 4.0548,
+      "step": 728064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8071120696715e-05,
+      "loss": 4.0659,
+      "step": 728576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.806273474920448e-05,
+      "loss": 4.0725,
+      "step": 729088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.805434880169396e-05,
+      "loss": 4.0459,
+      "step": 729600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.804596285418344e-05,
+      "loss": 4.054,
+      "step": 730112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.803757690667292e-05,
+      "loss": 4.0542,
+      "step": 730624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.80291909591624e-05,
+      "loss": 4.0676,
+      "step": 731136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.802080501165188e-05,
+      "loss": 4.0501,
+      "step": 731648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8012435442945097e-05,
+      "loss": 4.0651,
+      "step": 732160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8004049495434577e-05,
+      "loss": 4.0577,
+      "step": 732672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7995663547924056e-05,
+      "loss": 4.0528,
+      "step": 733184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7987277600413536e-05,
+      "loss": 4.0521,
+      "step": 733696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7978891652903016e-05,
+      "loss": 4.0405,
+      "step": 734208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7970505705392496e-05,
+      "loss": 4.0517,
+      "step": 734720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7962119757881976e-05,
+      "loss": 4.0593,
+      "step": 735232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7953733810371456e-05,
+      "loss": 4.0656,
+      "step": 735744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7945347862860936e-05,
+      "loss": 4.0455,
+      "step": 736256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.793696191535041e-05,
+      "loss": 4.0478,
+      "step": 736768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.792857596783989e-05,
+      "loss": 4.0536,
+      "step": 737280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7920222777936834e-05,
+      "loss": 4.0566,
+      "step": 737792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7911836830426314e-05,
+      "loss": 4.0601,
+      "step": 738304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7903450882915794e-05,
+      "loss": 4.0529,
+      "step": 738816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.789506493540528e-05,
+      "loss": 4.0553,
+      "step": 739328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.788667898789476e-05,
+      "loss": 4.0532,
+      "step": 739840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7878293040384234e-05,
+      "loss": 4.0542,
+      "step": 740352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7869907092873714e-05,
+      "loss": 4.0426,
+      "step": 740864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.786153752416693e-05,
+      "loss": 4.0479,
+      "step": 741376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.785315157665641e-05,
+      "loss": 4.0579,
+      "step": 741888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.784476562914588e-05,
+      "loss": 4.0458,
+      "step": 742400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.783637968163536e-05,
+      "loss": 4.051,
+      "step": 742912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.782801011292858e-05,
+      "loss": 4.0436,
+      "step": 743424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.781962416541805e-05,
+      "loss": 4.0556,
+      "step": 743936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.781123821790753e-05,
+      "loss": 4.0489,
+      "step": 744448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.780285227039702e-05,
+      "loss": 4.0515,
+      "step": 744960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.77944663228865e-05,
+      "loss": 4.0567,
+      "step": 745472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.778608037537598e-05,
+      "loss": 4.0683,
+      "step": 745984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.777769442786546e-05,
+      "loss": 4.0513,
+      "step": 746496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.776932485915867e-05,
+      "loss": 4.0571,
+      "step": 747008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.776093891164815e-05,
+      "loss": 4.0547,
+      "step": 747520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.775255296413763e-05,
+      "loss": 4.0463,
+      "step": 748032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.774416701662711e-05,
+      "loss": 4.0591,
+      "step": 748544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.773578106911659e-05,
+      "loss": 4.0467,
+      "step": 749056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.772739512160607e-05,
+      "loss": 4.0465,
+      "step": 749568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.771900917409555e-05,
+      "loss": 4.0536,
+      "step": 750080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.771062322658503e-05,
+      "loss": 4.0506,
+      "step": 750592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7702253657878237e-05,
+      "loss": 4.0549,
+      "step": 751104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7693867710367717e-05,
+      "loss": 4.0478,
+      "step": 751616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.76854817628572e-05,
+      "loss": 4.0585,
+      "step": 752128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.767711219415041e-05,
+      "loss": 4.0459,
+      "step": 752640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.766872624663989e-05,
+      "loss": 4.0609,
+      "step": 753152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.766034029912937e-05,
+      "loss": 4.0445,
+      "step": 753664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.765197073042258e-05,
+      "loss": 4.0491,
+      "step": 754176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.764358478291206e-05,
+      "loss": 4.0257,
+      "step": 754688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.763519883540154e-05,
+      "loss": 4.0593,
+      "step": 755200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.762681288789102e-05,
+      "loss": 4.0537,
+      "step": 755712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.76184269403805e-05,
+      "loss": 4.0627,
+      "step": 756224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.761004099286998e-05,
+      "loss": 4.0478,
+      "step": 756736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.760165504535946e-05,
+      "loss": 4.0572,
+      "step": 757248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.759326909784894e-05,
+      "loss": 4.0504,
+      "step": 757760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.758488315033842e-05,
+      "loss": 4.0666,
+      "step": 758272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.75764972028279e-05,
+      "loss": 4.0408,
+      "step": 758784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.756811125531738e-05,
+      "loss": 4.047,
+      "step": 759296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.755972530780686e-05,
+      "loss": 4.0503,
+      "step": 759808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.755133936029634e-05,
+      "loss": 4.0578,
+      "step": 760320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.754296979158955e-05,
+      "loss": 4.0472,
+      "step": 760832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.753458384407903e-05,
+      "loss": 4.0466,
+      "step": 761344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.752619789656851e-05,
+      "loss": 4.0517,
+      "step": 761856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.751781194905799e-05,
+      "loss": 4.0532,
+      "step": 762368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.75094423803512e-05,
+      "loss": 4.0535,
+      "step": 762880
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.090473175048828,
+      "eval_runtime": 288.0306,
+      "eval_samples_per_second": 1324.828,
+      "eval_steps_per_second": 41.402,
+      "step": 763200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.750105643284068e-05,
+      "loss": 4.0382,
+      "step": 763392
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.749268686413389e-05,
+      "loss": 4.0559,
+      "step": 763904
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7484300916623375e-05,
+      "loss": 4.0496,
+      "step": 764416
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7475914969112855e-05,
+      "loss": 4.0484,
+      "step": 764928
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7467529021602335e-05,
+      "loss": 4.0597,
+      "step": 765440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7459143074091815e-05,
+      "loss": 4.0461,
+      "step": 765952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7450757126581294e-05,
+      "loss": 4.0478,
+      "step": 766464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7442387557874504e-05,
+      "loss": 4.0387,
+      "step": 766976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7434001610363984e-05,
+      "loss": 4.0461,
+      "step": 767488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7425615662853463e-05,
+      "loss": 4.0552,
+      "step": 768000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7417229715342943e-05,
+      "loss": 4.0525,
+      "step": 768512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7408843767832423e-05,
+      "loss": 4.0503,
+      "step": 769024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.740047419912563e-05,
+      "loss": 4.0556,
+      "step": 769536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.739208825161511e-05,
+      "loss": 4.0434,
+      "step": 770048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.738370230410459e-05,
+      "loss": 4.0412,
+      "step": 770560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.737531635659407e-05,
+      "loss": 4.0335,
+      "step": 771072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.736693040908356e-05,
+      "loss": 4.0502,
+      "step": 771584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.735854446157304e-05,
+      "loss": 4.0441,
+      "step": 772096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.735015851406252e-05,
+      "loss": 4.0401,
+      "step": 772608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7341772566552e-05,
+      "loss": 4.0526,
+      "step": 773120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.733340299784521e-05,
+      "loss": 4.061,
+      "step": 773632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.732501705033469e-05,
+      "loss": 4.0476,
+      "step": 774144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.731663110282417e-05,
+      "loss": 4.0448,
+      "step": 774656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.730824515531365e-05,
+      "loss": 4.047,
+      "step": 775168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.729985920780313e-05,
+      "loss": 4.0453,
+      "step": 775680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7291506017900066e-05,
+      "loss": 4.0452,
+      "step": 776192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7283120070389546e-05,
+      "loss": 4.0274,
+      "step": 776704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7274734122879026e-05,
+      "loss": 4.0469,
+      "step": 777216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.726636455417224e-05,
+      "loss": 4.0502,
+      "step": 777728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.725797860666172e-05,
+      "loss": 4.0315,
+      "step": 778240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.72495926591512e-05,
+      "loss": 4.0393,
+      "step": 778752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.724120671164068e-05,
+      "loss": 4.0494,
+      "step": 779264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.723282076413016e-05,
+      "loss": 4.0453,
+      "step": 779776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.722443481661964e-05,
+      "loss": 4.0529,
+      "step": 780288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.721604886910912e-05,
+      "loss": 4.0448,
+      "step": 780800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.72076629215986e-05,
+      "loss": 4.0386,
+      "step": 781312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.719929335289181e-05,
+      "loss": 4.0615,
+      "step": 781824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.719090740538129e-05,
+      "loss": 4.027,
+      "step": 782336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.718252145787077e-05,
+      "loss": 4.0326,
+      "step": 782848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.717413551036025e-05,
+      "loss": 4.0398,
+      "step": 783360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.716574956284973e-05,
+      "loss": 4.0363,
+      "step": 783872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.715736361533921e-05,
+      "loss": 4.0407,
+      "step": 784384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.714897766782869e-05,
+      "loss": 4.042,
+      "step": 784896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.714059172031817e-05,
+      "loss": 4.0432,
+      "step": 785408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.713222215161138e-05,
+      "loss": 4.0388,
+      "step": 785920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.712383620410086e-05,
+      "loss": 4.0361,
+      "step": 786432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.711545025659034e-05,
+      "loss": 4.0474,
+      "step": 786944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.710706430907982e-05,
+      "loss": 4.0449,
+      "step": 787456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.70986783615693e-05,
+      "loss": 4.0214,
+      "step": 787968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.709030879286251e-05,
+      "loss": 4.0461,
+      "step": 788480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7081939224155724e-05,
+      "loss": 4.0451,
+      "step": 788992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7073553276645204e-05,
+      "loss": 4.053,
+      "step": 789504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7065167329134684e-05,
+      "loss": 4.035,
+      "step": 790016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7056781381624164e-05,
+      "loss": 4.0371,
+      "step": 790528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7048395434113644e-05,
+      "loss": 4.0284,
+      "step": 791040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.704002586540685e-05,
+      "loss": 4.0316,
+      "step": 791552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.703163991789633e-05,
+      "loss": 4.0464,
+      "step": 792064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.702325397038581e-05,
+      "loss": 4.026,
+      "step": 792576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.701486802287529e-05,
+      "loss": 4.0414,
+      "step": 793088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.700648207536477e-05,
+      "loss": 4.0387,
+      "step": 793600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.699809612785425e-05,
+      "loss": 4.0338,
+      "step": 794112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.698971018034373e-05,
+      "loss": 4.0414,
+      "step": 794624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.698132423283321e-05,
+      "loss": 4.0266,
+      "step": 795136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.697295466412642e-05,
+      "loss": 4.0333,
+      "step": 795648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.69645687166159e-05,
+      "loss": 4.0341,
+      "step": 796160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.695619914790912e-05,
+      "loss": 4.0477,
+      "step": 796672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.69478132003986e-05,
+      "loss": 4.0273,
+      "step": 797184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.693942725288808e-05,
+      "loss": 4.0415,
+      "step": 797696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.693104130537756e-05,
+      "loss": 4.0162,
+      "step": 798208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.692265535786704e-05,
+      "loss": 4.0356,
+      "step": 798720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.691426941035652e-05,
+      "loss": 4.0394,
+      "step": 799232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6905883462846e-05,
+      "loss": 4.0425,
+      "step": 799744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.689749751533548e-05,
+      "loss": 4.0346,
+      "step": 800256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6889127946628687e-05,
+      "loss": 4.0386,
+      "step": 800768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6880741999118167e-05,
+      "loss": 4.0557,
+      "step": 801280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6872356051607646e-05,
+      "loss": 4.0299,
+      "step": 801792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6863986482900856e-05,
+      "loss": 4.046,
+      "step": 802304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6855600535390336e-05,
+      "loss": 4.0319,
+      "step": 802816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.684721458787982e-05,
+      "loss": 4.0355,
+      "step": 803328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.68388286403693e-05,
+      "loss": 4.0389,
+      "step": 803840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.683044269285878e-05,
+      "loss": 4.0322,
+      "step": 804352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.682205674534826e-05,
+      "loss": 4.0457,
+      "step": 804864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6813670797837735e-05,
+      "loss": 4.0485,
+      "step": 805376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6805284850327215e-05,
+      "loss": 4.0314,
+      "step": 805888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.679691528162043e-05,
+      "loss": 4.0265,
+      "step": 806400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.678852933410991e-05,
+      "loss": 4.0322,
+      "step": 806912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6780143386599384e-05,
+      "loss": 4.0433,
+      "step": 807424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6771757439088864e-05,
+      "loss": 4.0295,
+      "step": 807936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.676338787038208e-05,
+      "loss": 4.0421,
+      "step": 808448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.675500192287156e-05,
+      "loss": 4.0346,
+      "step": 808960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.674661597536104e-05,
+      "loss": 4.0344,
+      "step": 809472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6738246406654256e-05,
+      "loss": 4.0293,
+      "step": 809984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6729860459143736e-05,
+      "loss": 4.0202,
+      "step": 810496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.672147451163321e-05,
+      "loss": 4.0276,
+      "step": 811008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.671308856412269e-05,
+      "loss": 4.039,
+      "step": 811520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.670470261661217e-05,
+      "loss": 4.0411,
+      "step": 812032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.669631666910165e-05,
+      "loss": 4.0238,
+      "step": 812544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.668793072159113e-05,
+      "loss": 4.0235,
+      "step": 813056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.667954477408061e-05,
+      "loss": 4.0305,
+      "step": 813568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.667117520537382e-05,
+      "loss": 4.0372,
+      "step": 814080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.66627892578633e-05,
+      "loss": 4.0423,
+      "step": 814592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6654419689156514e-05,
+      "loss": 4.0279,
+      "step": 815104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6646033741645994e-05,
+      "loss": 4.0363,
+      "step": 815616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6637647794135474e-05,
+      "loss": 4.0324,
+      "step": 816128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6629261846624954e-05,
+      "loss": 4.0299,
+      "step": 816640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6620875899114434e-05,
+      "loss": 4.0248,
+      "step": 817152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.661250633040764e-05,
+      "loss": 4.0293,
+      "step": 817664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.660412038289712e-05,
+      "loss": 4.0328,
+      "step": 818176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.65957344353866e-05,
+      "loss": 4.0238,
+      "step": 818688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.658734848787608e-05,
+      "loss": 4.0297,
+      "step": 819200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.657896254036556e-05,
+      "loss": 4.0245,
+      "step": 819712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.657057659285504e-05,
+      "loss": 4.03,
+      "step": 820224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.656219064534452e-05,
+      "loss": 4.0287,
+      "step": 820736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6553804697834e-05,
+      "loss": 4.0276,
+      "step": 821248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.654541875032348e-05,
+      "loss": 4.0333,
+      "step": 821760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.653703280281296e-05,
+      "loss": 4.0456,
+      "step": 822272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.652866323410618e-05,
+      "loss": 4.0318,
+      "step": 822784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.652029366539939e-05,
+      "loss": 4.0359,
+      "step": 823296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.651190771788887e-05,
+      "loss": 4.0327,
+      "step": 823808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6503538149182076e-05,
+      "loss": 4.0265,
+      "step": 824320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6495152201671556e-05,
+      "loss": 4.0416,
+      "step": 824832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6486766254161036e-05,
+      "loss": 4.0218,
+      "step": 825344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6478380306650516e-05,
+      "loss": 4.0259,
+      "step": 825856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6469994359139996e-05,
+      "loss": 4.0338,
+      "step": 826368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6461608411629476e-05,
+      "loss": 4.0321,
+      "step": 826880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6453222464118956e-05,
+      "loss": 4.0303,
+      "step": 827392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6444836516608436e-05,
+      "loss": 4.0281,
+      "step": 827904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.643646694790165e-05,
+      "loss": 4.0352,
+      "step": 828416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.642808100039113e-05,
+      "loss": 4.0262,
+      "step": 828928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.641969505288061e-05,
+      "loss": 4.037,
+      "step": 829440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.641130910537009e-05,
+      "loss": 4.027,
+      "step": 829952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.640292315785957e-05,
+      "loss": 4.0231,
+      "step": 830464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6394537210349045e-05,
+      "loss": 4.0047,
+      "step": 830976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6386151262838525e-05,
+      "loss": 4.0398,
+      "step": 831488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6377765315328005e-05,
+      "loss": 4.03,
+      "step": 832000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.636939574662122e-05,
+      "loss": 4.0442,
+      "step": 832512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6361009799110694e-05,
+      "loss": 4.0277,
+      "step": 833024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6352623851600174e-05,
+      "loss": 4.0333,
+      "step": 833536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6344237904089654e-05,
+      "loss": 4.0316,
+      "step": 834048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6335851956579134e-05,
+      "loss": 4.047,
+      "step": 834560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.632748238787235e-05,
+      "loss": 4.0166,
+      "step": 835072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6319112819165565e-05,
+      "loss": 4.0276,
+      "step": 835584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6310726871655045e-05,
+      "loss": 4.0297,
+      "step": 836096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.630234092414452e-05,
+      "loss": 4.0348,
+      "step": 836608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6293954976634e-05,
+      "loss": 4.0288,
+      "step": 837120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.628556902912348e-05,
+      "loss": 4.0279,
+      "step": 837632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.627718308161296e-05,
+      "loss": 4.0275,
+      "step": 838144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.626879713410244e-05,
+      "loss": 4.0372,
+      "step": 838656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.626041118659192e-05,
+      "loss": 4.0315,
+      "step": 839168
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.07681941986084,
+      "eval_runtime": 288.6201,
+      "eval_samples_per_second": 1322.122,
+      "eval_steps_per_second": 41.317,
+      "step": 839520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.625204161788513e-05,
+      "loss": 4.0324,
+      "step": 839680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.624365567037461e-05,
+      "loss": 4.0318,
+      "step": 840192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.623526972286409e-05,
+      "loss": 4.0283,
+      "step": 840704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.622688377535357e-05,
+      "loss": 4.0312,
+      "step": 841216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6218497827843054e-05,
+      "loss": 4.0364,
+      "step": 841728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6210111880332534e-05,
+      "loss": 4.026,
+      "step": 842240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6201725932822014e-05,
+      "loss": 4.0292,
+      "step": 842752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6193339985311494e-05,
+      "loss": 4.0205,
+      "step": 843264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6184954037800974e-05,
+      "loss": 4.0226,
+      "step": 843776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6176568090290454e-05,
+      "loss": 4.0331,
+      "step": 844288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6168182142779934e-05,
+      "loss": 4.0353,
+      "step": 844800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.615979619526941e-05,
+      "loss": 4.0293,
+      "step": 845312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.615142662656262e-05,
+      "loss": 4.0417,
+      "step": 845824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.61430406790521e-05,
+      "loss": 4.0179,
+      "step": 846336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.613465473154158e-05,
+      "loss": 4.0214,
+      "step": 846848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6126268784031056e-05,
+      "loss": 4.0152,
+      "step": 847360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6117882836520536e-05,
+      "loss": 4.0322,
+      "step": 847872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.610949688901002e-05,
+      "loss": 4.0215,
+      "step": 848384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.61011109414995e-05,
+      "loss": 4.0233,
+      "step": 848896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.609274137279271e-05,
+      "loss": 4.0302,
+      "step": 849408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.608435542528219e-05,
+      "loss": 4.04,
+      "step": 849920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.607596947777167e-05,
+      "loss": 4.0296,
+      "step": 850432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.606758353026115e-05,
+      "loss": 4.0231,
+      "step": 850944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.605919758275063e-05,
+      "loss": 4.0281,
+      "step": 851456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.605081163524011e-05,
+      "loss": 4.0272,
+      "step": 851968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.604242568772959e-05,
+      "loss": 4.0243,
+      "step": 852480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.603403974021907e-05,
+      "loss": 4.0091,
+      "step": 852992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.602565379270855e-05,
+      "loss": 4.0243,
+      "step": 853504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.601726784519803e-05,
+      "loss": 4.0292,
+      "step": 854016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.600888189768751e-05,
+      "loss": 4.0096,
+      "step": 854528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.600049595017699e-05,
+      "loss": 4.0218,
+      "step": 855040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5992142760273936e-05,
+      "loss": 4.0285,
+      "step": 855552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5983756812763416e-05,
+      "loss": 4.027,
+      "step": 856064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5975370865252896e-05,
+      "loss": 4.0328,
+      "step": 856576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5966984917742376e-05,
+      "loss": 4.0251,
+      "step": 857088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5958598970231856e-05,
+      "loss": 4.0213,
+      "step": 857600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5950213022721336e-05,
+      "loss": 4.0426,
+      "step": 858112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5941827075210816e-05,
+      "loss": 4.0042,
+      "step": 858624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5933441127700296e-05,
+      "loss": 4.0129,
+      "step": 859136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.592505518018977e-05,
+      "loss": 4.0228,
+      "step": 859648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.591666923267925e-05,
+      "loss": 4.0151,
+      "step": 860160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5908299663972465e-05,
+      "loss": 4.0204,
+      "step": 860672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5899913716461945e-05,
+      "loss": 4.0279,
+      "step": 861184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5891527768951425e-05,
+      "loss": 4.0211,
+      "step": 861696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.588315820024464e-05,
+      "loss": 4.0218,
+      "step": 862208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.587477225273412e-05,
+      "loss": 4.0139,
+      "step": 862720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5866386305223594e-05,
+      "loss": 4.0255,
+      "step": 863232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5858000357713074e-05,
+      "loss": 4.028,
+      "step": 863744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5849614410202554e-05,
+      "loss": 4.0046,
+      "step": 864256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5841228462692034e-05,
+      "loss": 4.0247,
+      "step": 864768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5832842515181514e-05,
+      "loss": 4.0269,
+      "step": 865280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5824456567670994e-05,
+      "loss": 4.0312,
+      "step": 865792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5816070620160474e-05,
+      "loss": 4.0166,
+      "step": 866304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.580770105145368e-05,
+      "loss": 4.0144,
+      "step": 866816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.579931510394316e-05,
+      "loss": 4.0161,
+      "step": 867328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.579094553523638e-05,
+      "loss": 4.0104,
+      "step": 867840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.578255958772586e-05,
+      "loss": 4.0275,
+      "step": 868352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.577419001901907e-05,
+      "loss": 4.0053,
+      "step": 868864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.576580407150855e-05,
+      "loss": 4.0243,
+      "step": 869376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.575741812399803e-05,
+      "loss": 4.0182,
+      "step": 869888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.574903217648751e-05,
+      "loss": 4.0198,
+      "step": 870400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.574064622897699e-05,
+      "loss": 4.0172,
+      "step": 870912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.573226028146647e-05,
+      "loss": 4.0133,
+      "step": 871424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.572387433395595e-05,
+      "loss": 4.0165,
+      "step": 871936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.571548838644543e-05,
+      "loss": 4.0076,
+      "step": 872448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.570710243893491e-05,
+      "loss": 4.0326,
+      "step": 872960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.569871649142439e-05,
+      "loss": 4.0139,
+      "step": 873472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.569033054391387e-05,
+      "loss": 4.0128,
+      "step": 873984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.568194459640335e-05,
+      "loss": 4.0019,
+      "step": 874496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.567357502769656e-05,
+      "loss": 4.0144,
+      "step": 875008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.566518908018604e-05,
+      "loss": 4.0216,
+      "step": 875520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.565680313267552e-05,
+      "loss": 4.0247,
+      "step": 876032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5648417185165e-05,
+      "loss": 4.0168,
+      "step": 876544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.564003123765448e-05,
+      "loss": 4.0216,
+      "step": 877056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.563166166894769e-05,
+      "loss": 4.0326,
+      "step": 877568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.562327572143717e-05,
+      "loss": 4.0135,
+      "step": 878080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.561490615273038e-05,
+      "loss": 4.0273,
+      "step": 878592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.560652020521986e-05,
+      "loss": 4.0133,
+      "step": 879104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.559813425770934e-05,
+      "loss": 4.0166,
+      "step": 879616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.558974831019882e-05,
+      "loss": 4.0196,
+      "step": 880128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.55813623626883e-05,
+      "loss": 4.0116,
+      "step": 880640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.557297641517778e-05,
+      "loss": 4.0258,
+      "step": 881152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.556459046766726e-05,
+      "loss": 4.0316,
+      "step": 881664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.555620452015674e-05,
+      "loss": 4.0184,
+      "step": 882176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.554781857264622e-05,
+      "loss": 4.0041,
+      "step": 882688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.553944900393943e-05,
+      "loss": 4.014,
+      "step": 883200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5531079435232645e-05,
+      "loss": 4.0234,
+      "step": 883712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5522693487722125e-05,
+      "loss": 4.0127,
+      "step": 884224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5514307540211605e-05,
+      "loss": 4.0219,
+      "step": 884736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.550592159270108e-05,
+      "loss": 4.0188,
+      "step": 885248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.549753564519056e-05,
+      "loss": 4.0164,
+      "step": 885760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.548914969768004e-05,
+      "loss": 4.0107,
+      "step": 886272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5480780128973254e-05,
+      "loss": 4.0005,
+      "step": 886784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5472394181462734e-05,
+      "loss": 4.0092,
+      "step": 887296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5464008233952214e-05,
+      "loss": 4.0142,
+      "step": 887808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5455622286441694e-05,
+      "loss": 4.0303,
+      "step": 888320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5447236338931174e-05,
+      "loss": 4.0027,
+      "step": 888832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.543886677022438e-05,
+      "loss": 4.0099,
+      "step": 889344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.543048082271386e-05,
+      "loss": 4.0162,
+      "step": 889856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.542209487520334e-05,
+      "loss": 4.0146,
+      "step": 890368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.541370892769282e-05,
+      "loss": 4.0222,
+      "step": 890880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.54053229801823e-05,
+      "loss": 4.0092,
+      "step": 891392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.539693703267178e-05,
+      "loss": 4.0204,
+      "step": 891904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.538855108516126e-05,
+      "loss": 4.0113,
+      "step": 892416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.538016513765074e-05,
+      "loss": 4.0102,
+      "step": 892928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.537179556894395e-05,
+      "loss": 4.0072,
+      "step": 893440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.536340962143343e-05,
+      "loss": 4.0116,
+      "step": 893952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.535502367392292e-05,
+      "loss": 4.0161,
+      "step": 894464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.53466377264124e-05,
+      "loss": 4.008,
+      "step": 894976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.533825177890188e-05,
+      "loss": 4.0084,
+      "step": 895488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.532988221019509e-05,
+      "loss": 4.0033,
+      "step": 896000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.532149626268457e-05,
+      "loss": 4.0104,
+      "step": 896512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.531312669397778e-05,
+      "loss": 4.0127,
+      "step": 897024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.530474074646726e-05,
+      "loss": 4.0133,
+      "step": 897536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.529635479895674e-05,
+      "loss": 4.0142,
+      "step": 898048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.528796885144622e-05,
+      "loss": 4.0207,
+      "step": 898560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.52795829039357e-05,
+      "loss": 4.0184,
+      "step": 899072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5271213335228906e-05,
+      "loss": 4.0205,
+      "step": 899584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5262827387718386e-05,
+      "loss": 4.0171,
+      "step": 900096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.525444144020787e-05,
+      "loss": 4.0117,
+      "step": 900608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.524605549269735e-05,
+      "loss": 4.0182,
+      "step": 901120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.523768592399056e-05,
+      "loss": 4.0064,
+      "step": 901632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.522929997648004e-05,
+      "loss": 4.0083,
+      "step": 902144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.522091402896952e-05,
+      "loss": 4.015,
+      "step": 902656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5212528081459e-05,
+      "loss": 4.0127,
+      "step": 903168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.520414213394848e-05,
+      "loss": 4.0124,
+      "step": 903680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.519575618643796e-05,
+      "loss": 4.0109,
+      "step": 904192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5187370238927434e-05,
+      "loss": 4.0161,
+      "step": 904704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5178984291416914e-05,
+      "loss": 4.0102,
+      "step": 905216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.517061472271013e-05,
+      "loss": 4.0186,
+      "step": 905728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.516222877519961e-05,
+      "loss": 4.0109,
+      "step": 906240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.515384282768909e-05,
+      "loss": 4.0084,
+      "step": 906752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.514545688017857e-05,
+      "loss": 3.9851,
+      "step": 907264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5137087311471786e-05,
+      "loss": 4.0212,
+      "step": 907776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5128701363961266e-05,
+      "loss": 4.0113,
+      "step": 908288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.512031541645074e-05,
+      "loss": 4.0267,
+      "step": 908800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.511192946894022e-05,
+      "loss": 4.0118,
+      "step": 909312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.51035435214297e-05,
+      "loss": 4.0129,
+      "step": 909824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.509517395272291e-05,
+      "loss": 4.0156,
+      "step": 910336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.508678800521239e-05,
+      "loss": 4.0273,
+      "step": 910848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.507840205770187e-05,
+      "loss": 3.9995,
+      "step": 911360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.507001611019135e-05,
+      "loss": 4.0118,
+      "step": 911872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.506163016268083e-05,
+      "loss": 4.0102,
+      "step": 912384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.5053260593974044e-05,
+      "loss": 4.0163,
+      "step": 912896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.504489102526726e-05,
+      "loss": 4.0119,
+      "step": 913408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.503650507775674e-05,
+      "loss": 4.0113,
+      "step": 913920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.502813550904995e-05,
+      "loss": 4.0088,
+      "step": 914432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.501974956153943e-05,
+      "loss": 4.019,
+      "step": 914944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.501136361402891e-05,
+      "loss": 4.0098,
+      "step": 915456
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.0655999183654785,
+      "eval_runtime": 294.1983,
+      "eval_samples_per_second": 1297.054,
+      "eval_steps_per_second": 40.534,
+      "step": 915840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.500297766651838e-05,
+      "loss": 4.0074,
+      "step": 915968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.499459171900786e-05,
+      "loss": 4.0155,
+      "step": 916480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.498620577149734e-05,
+      "loss": 4.0071,
+      "step": 916992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.497781982398682e-05,
+      "loss": 4.0152,
+      "step": 917504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.49694338764763e-05,
+      "loss": 4.0141,
+      "step": 918016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.496104792896578e-05,
+      "loss": 4.0146,
+      "step": 918528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.495266198145526e-05,
+      "loss": 4.0099,
+      "step": 919040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.494427603394475e-05,
+      "loss": 4.0017,
+      "step": 919552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.493589008643423e-05,
+      "loss": 4.0046,
+      "step": 920064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.492752051772744e-05,
+      "loss": 4.021,
+      "step": 920576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.491913457021692e-05,
+      "loss": 4.0138,
+      "step": 921088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.49107486227064e-05,
+      "loss": 4.0059,
+      "step": 921600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.4902379053999606e-05,
+      "loss": 4.0256,
+      "step": 922112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.4893993106489086e-05,
+      "loss": 4.0002,
+      "step": 922624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.4885607158978566e-05,
+      "loss": 4.0028,
+      "step": 923136
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.4877221211468046e-05,
+      "loss": 3.9998,
+      "step": 923648
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.4868851642761255e-05,
+      "loss": 4.012,
+      "step": 924160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.4860465695250735e-05,
+      "loss": 4.0069,
+      "step": 924672
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.4852079747740215e-05,
+      "loss": 4.0077,
+      "step": 925184
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.48436938002297e-05,
+      "loss": 4.0143,
+      "step": 925696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.483530785271918e-05,
+      "loss": 4.0189,
+      "step": 926208
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.482692190520866e-05,
+      "loss": 4.0137,
+      "step": 926720
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.481853595769814e-05,
+      "loss": 4.0026,
+      "step": 927232
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.481015001018762e-05,
+      "loss": 4.0097,
+      "step": 927744
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.48017640626771e-05,
+      "loss": 4.0123,
+      "step": 928256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.479339449397031e-05,
+      "loss": 4.0059,
+      "step": 928768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.478500854645979e-05,
+      "loss": 3.9941,
+      "step": 929280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.477662259894927e-05,
+      "loss": 4.0024,
+      "step": 929792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.476825303024248e-05,
+      "loss": 4.0127,
+      "step": 930304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.475986708273196e-05,
+      "loss": 3.993,
+      "step": 930816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.475148113522144e-05,
+      "loss": 4.0062,
+      "step": 931328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.474309518771092e-05,
+      "loss": 4.0068,
+      "step": 931840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.47347092402004e-05,
+      "loss": 4.0115,
+      "step": 932352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.472632329268988e-05,
+      "loss": 4.0144,
+      "step": 932864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.471793734517936e-05,
+      "loss": 4.01,
+      "step": 933376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4709567776472575e-05,
+      "loss": 4.0092,
+      "step": 933888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.470118182896205e-05,
+      "loss": 4.0212,
+      "step": 934400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.469279588145153e-05,
+      "loss": 3.9863,
+      "step": 934912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4684426312744744e-05,
+      "loss": 3.9987,
+      "step": 935424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.467604036523422e-05,
+      "loss": 4.0078,
+      "step": 935936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.46676544177237e-05,
+      "loss": 3.9959,
+      "step": 936448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.465926847021318e-05,
+      "loss": 4.0059,
+      "step": 936960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.465088252270266e-05,
+      "loss": 4.0076,
+      "step": 937472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.464249657519214e-05,
+      "loss": 4.0031,
+      "step": 937984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.463411062768162e-05,
+      "loss": 4.0044,
+      "step": 938496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4625724680171104e-05,
+      "loss": 4.0019,
+      "step": 939008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4617338732660584e-05,
+      "loss": 4.005,
+      "step": 939520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.460896916395379e-05,
+      "loss": 4.016,
+      "step": 940032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.460058321644327e-05,
+      "loss": 3.9825,
+      "step": 940544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.459219726893275e-05,
+      "loss": 4.0053,
+      "step": 941056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.458381132142223e-05,
+      "loss": 4.0129,
+      "step": 941568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.457544175271544e-05,
+      "loss": 4.0125,
+      "step": 942080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.456705580520492e-05,
+      "loss": 4.0053,
+      "step": 942592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.45586698576944e-05,
+      "loss": 3.9973,
+      "step": 943104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.455028391018388e-05,
+      "loss": 3.9989,
+      "step": 943616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.454189796267336e-05,
+      "loss": 3.9928,
+      "step": 944128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.453351201516284e-05,
+      "loss": 4.0091,
+      "step": 944640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.452512606765232e-05,
+      "loss": 3.9895,
+      "step": 945152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.45167401201418e-05,
+      "loss": 4.0103,
+      "step": 945664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.450837055143502e-05,
+      "loss": 3.998,
+      "step": 946176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.450000098272823e-05,
+      "loss": 4.0013,
+      "step": 946688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.449161503521771e-05,
+      "loss": 3.9963,
+      "step": 947200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.448322908770719e-05,
+      "loss": 4.0001,
+      "step": 947712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.447484314019667e-05,
+      "loss": 3.9969,
+      "step": 948224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.446645719268615e-05,
+      "loss": 3.9937,
+      "step": 948736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4458071245175627e-05,
+      "loss": 4.0114,
+      "step": 949248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4449685297665107e-05,
+      "loss": 4.0035,
+      "step": 949760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.444129935015458e-05,
+      "loss": 3.9948,
+      "step": 950272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.443291340264406e-05,
+      "loss": 3.9844,
+      "step": 950784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4424543833937276e-05,
+      "loss": 4.0018,
+      "step": 951296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4416157886426756e-05,
+      "loss": 4.0001,
+      "step": 951808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4407771938916235e-05,
+      "loss": 4.0116,
+      "step": 952320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4399385991405715e-05,
+      "loss": 3.9961,
+      "step": 952832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.439103280150266e-05,
+      "loss": 4.01,
+      "step": 953344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.438264685399214e-05,
+      "loss": 4.014,
+      "step": 953856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.437426090648162e-05,
+      "loss": 4.0006,
+      "step": 954368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.43658749589711e-05,
+      "loss": 4.0082,
+      "step": 954880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.435748901146058e-05,
+      "loss": 3.9987,
+      "step": 955392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4349103063950053e-05,
+      "loss": 3.9984,
+      "step": 955904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.4340717116439533e-05,
+      "loss": 4.0054,
+      "step": 956416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.433234754773275e-05,
+      "loss": 3.9966,
+      "step": 956928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.432396160022223e-05,
+      "loss": 4.002,
+      "step": 957440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.431557565271171e-05,
+      "loss": 4.018,
+      "step": 957952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.430718970520119e-05,
+      "loss": 4.0038,
+      "step": 958464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.429880375769067e-05,
+      "loss": 3.9855,
+      "step": 958976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.429041781018015e-05,
+      "loss": 3.9974,
+      "step": 959488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.428203186266963e-05,
+      "loss": 4.0034,
+      "step": 960000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.427364591515911e-05,
+      "loss": 4.0023,
+      "step": 960512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.426525996764859e-05,
+      "loss": 4.001,
+      "step": 961024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.42568903989418e-05,
+      "loss": 4.0032,
+      "step": 961536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.424850445143128e-05,
+      "loss": 4.0001,
+      "step": 962048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.424013488272449e-05,
+      "loss": 3.9929,
+      "step": 962560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.423174893521397e-05,
+      "loss": 3.9825,
+      "step": 963072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.422336298770345e-05,
+      "loss": 3.9932,
+      "step": 963584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4214977040192934e-05,
+      "loss": 3.9962,
+      "step": 964096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4206591092682414e-05,
+      "loss": 4.0153,
+      "step": 964608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4198205145171894e-05,
+      "loss": 3.9864,
+      "step": 965120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4189819197661374e-05,
+      "loss": 3.9918,
+      "step": 965632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4181433250150854e-05,
+      "loss": 3.9989,
+      "step": 966144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.417306368144406e-05,
+      "loss": 4.0,
+      "step": 966656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.416467773393354e-05,
+      "loss": 4.0062,
+      "step": 967168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.415629178642302e-05,
+      "loss": 3.994,
+      "step": 967680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.41479058389125e-05,
+      "loss": 3.9978,
+      "step": 968192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.413953627020571e-05,
+      "loss": 3.9978,
+      "step": 968704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.413116670149892e-05,
+      "loss": 3.9953,
+      "step": 969216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.41227807539884e-05,
+      "loss": 3.9927,
+      "step": 969728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.411439480647789e-05,
+      "loss": 3.9979,
+      "step": 970240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.410600885896737e-05,
+      "loss": 4.0003,
+      "step": 970752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.409762291145685e-05,
+      "loss": 3.9927,
+      "step": 971264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.408923696394633e-05,
+      "loss": 3.9922,
+      "step": 971776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4080867395239536e-05,
+      "loss": 3.9887,
+      "step": 972288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4072481447729016e-05,
+      "loss": 3.9964,
+      "step": 972800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4064095500218496e-05,
+      "loss": 3.9957,
+      "step": 973312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4055709552707976e-05,
+      "loss": 3.9962,
+      "step": 973824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4047339984001185e-05,
+      "loss": 3.9984,
+      "step": 974336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4038954036490665e-05,
+      "loss": 4.0016,
+      "step": 974848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4030568088980145e-05,
+      "loss": 4.0067,
+      "step": 975360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4022182141469625e-05,
+      "loss": 4.0014,
+      "step": 975872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4013796193959105e-05,
+      "loss": 4.0017,
+      "step": 976384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.4005410246448585e-05,
+      "loss": 3.9975,
+      "step": 976896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.3997024298938065e-05,
+      "loss": 3.9973,
+      "step": 977408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.3988638351427545e-05,
+      "loss": 3.9904,
+      "step": 977920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.3980252403917025e-05,
+      "loss": 3.9945,
+      "step": 978432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.397188283521024e-05,
+      "loss": 3.9976,
+      "step": 978944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.3963496887699714e-05,
+      "loss": 3.9977,
+      "step": 979456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.395512731899293e-05,
+      "loss": 3.9998,
+      "step": 979968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.394674137148241e-05,
+      "loss": 3.9939,
+      "step": 980480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.393835542397189e-05,
+      "loss": 3.9981,
+      "step": 980992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.392996947646136e-05,
+      "loss": 3.9919,
+      "step": 981504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.392158352895084e-05,
+      "loss": 4.0047,
+      "step": 982016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.391321396024406e-05,
+      "loss": 3.9928,
+      "step": 982528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.390482801273354e-05,
+      "loss": 3.9942,
+      "step": 983040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.389644206522302e-05,
+      "loss": 3.9712,
+      "step": 983552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.38880561177125e-05,
+      "loss": 3.9986,
+      "step": 984064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.387967017020198e-05,
+      "loss": 4.0021,
+      "step": 984576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.387130060149519e-05,
+      "loss": 4.0101,
+      "step": 985088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.386291465398467e-05,
+      "loss": 3.9978,
+      "step": 985600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.385452870647415e-05,
+      "loss": 3.9934,
+      "step": 986112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.384614275896363e-05,
+      "loss": 4.0004,
+      "step": 986624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.383775681145311e-05,
+      "loss": 4.0101,
+      "step": 987136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.382938724274632e-05,
+      "loss": 3.9844,
+      "step": 987648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.3821001295235797e-05,
+      "loss": 3.9998,
+      "step": 988160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.3812615347725277e-05,
+      "loss": 3.9932,
+      "step": 988672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.3804229400214757e-05,
+      "loss": 4.0006,
+      "step": 989184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.379584345270424e-05,
+      "loss": 3.9968,
+      "step": 989696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.378745750519372e-05,
+      "loss": 3.9936,
+      "step": 990208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.377908793648693e-05,
+      "loss": 3.9933,
+      "step": 990720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.377070198897641e-05,
+      "loss": 4.0026,
+      "step": 991232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.376231604146589e-05,
+      "loss": 4.0005,
+      "step": 991744
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.055794715881348,
+      "eval_runtime": 294.3624,
+      "eval_samples_per_second": 1296.331,
+      "eval_steps_per_second": 40.511,
+      "step": 992160
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.375393009395537e-05,
+      "loss": 3.9869,
+      "step": 992256
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.374554414644485e-05,
+      "loss": 3.9998,
+      "step": 992768
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.373715819893433e-05,
+      "loss": 3.9927,
+      "step": 993280
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.372877225142381e-05,
+      "loss": 4.0018,
+      "step": 993792
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.372038630391329e-05,
+      "loss": 4.0018,
+      "step": 994304
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.371200035640277e-05,
+      "loss": 3.9954,
+      "step": 994816
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.370361440889225e-05,
+      "loss": 3.9946,
+      "step": 995328
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.369524484018546e-05,
+      "loss": 3.9839,
+      "step": 995840
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.368687527147868e-05,
+      "loss": 3.9917,
+      "step": 996352
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.367848932396816e-05,
+      "loss": 4.003,
+      "step": 996864
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.367010337645764e-05,
+      "loss": 3.9982,
+      "step": 997376
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.366171742894712e-05,
+      "loss": 3.9971,
+      "step": 997888
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.3653347860240326e-05,
+      "loss": 4.0071,
+      "step": 998400
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.3644961912729806e-05,
+      "loss": 3.988,
+      "step": 998912
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.3636575965219286e-05,
+      "loss": 3.9871,
+      "step": 999424
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.3628190017708766e-05,
+      "loss": 3.9846,
+      "step": 999936
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.3619804070198246e-05,
+      "loss": 3.9977,
+      "step": 1000448
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.3611418122687726e-05,
+      "loss": 3.9904,
+      "step": 1000960
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.36030321751772e-05,
+      "loss": 3.9903,
+      "step": 1001472
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.359464622766668e-05,
+      "loss": 3.9989,
+      "step": 1001984
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.3586260280156165e-05,
+      "loss": 4.0053,
+      "step": 1002496
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.3577890711449375e-05,
+      "loss": 3.9982,
+      "step": 1003008
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.356952114274259e-05,
+      "loss": 3.9919,
+      "step": 1003520
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.356113519523207e-05,
+      "loss": 3.992,
+      "step": 1004032
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.355274924772155e-05,
+      "loss": 3.9996,
+      "step": 1004544
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.3544363300211024e-05,
+      "loss": 3.9914,
+      "step": 1005056
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.3535977352700503e-05,
+      "loss": 3.9796,
+      "step": 1005568
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.352760778399372e-05,
+      "loss": 3.9891,
+      "step": 1006080
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.35192218364832e-05,
+      "loss": 3.9909,
+      "step": 1006592
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.351083588897267e-05,
+      "loss": 3.9813,
+      "step": 1007104
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.350244994146215e-05,
+      "loss": 3.9918,
+      "step": 1007616
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.349406399395163e-05,
+      "loss": 3.9908,
+      "step": 1008128
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.348567804644111e-05,
+      "loss": 3.9976,
+      "step": 1008640
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.34772920989306e-05,
+      "loss": 3.9979,
+      "step": 1009152
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.346890615142008e-05,
+      "loss": 3.9977,
+      "step": 1009664
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3460552961517024e-05,
+      "loss": 3.9933,
+      "step": 1010176
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.34521670140065e-05,
+      "loss": 4.0069,
+      "step": 1010688
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.344378106649598e-05,
+      "loss": 3.9745,
+      "step": 1011200
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.343539511898546e-05,
+      "loss": 3.9861,
+      "step": 1011712
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.342700917147494e-05,
+      "loss": 3.9871,
+      "step": 1012224
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.341862322396442e-05,
+      "loss": 3.9749,
+      "step": 1012736
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.34102372764539e-05,
+      "loss": 3.9935,
+      "step": 1013248
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.340185132894338e-05,
+      "loss": 3.9947,
+      "step": 1013760
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.339346538143286e-05,
+      "loss": 3.9903,
+      "step": 1014272
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.338507943392234e-05,
+      "loss": 3.986,
+      "step": 1014784
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.337670986521555e-05,
+      "loss": 3.9897,
+      "step": 1015296
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.336832391770503e-05,
+      "loss": 3.9891,
+      "step": 1015808
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.335993797019451e-05,
+      "loss": 3.9999,
+      "step": 1016320
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.335155202268399e-05,
+      "loss": 3.9702,
+      "step": 1016832
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.334316607517347e-05,
+      "loss": 3.9839,
+      "step": 1017344
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.333478012766295e-05,
+      "loss": 4.0034,
+      "step": 1017856
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.332639418015243e-05,
+      "loss": 3.9984,
+      "step": 1018368
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.331800823264191e-05,
+      "loss": 3.9883,
+      "step": 1018880
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.330963866393512e-05,
+      "loss": 3.9797,
+      "step": 1019392
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.33012527164246e-05,
+      "loss": 3.9855,
+      "step": 1019904
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.329286676891408e-05,
+      "loss": 3.9762,
+      "step": 1020416
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3284480821403555e-05,
+      "loss": 3.9959,
+      "step": 1020928
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3276094873893035e-05,
+      "loss": 3.9756,
+      "step": 1021440
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.326772530518625e-05,
+      "loss": 3.9915,
+      "step": 1021952
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.325933935767574e-05,
+      "loss": 3.986,
+      "step": 1022464
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3250969788968946e-05,
+      "loss": 3.9881,
+      "step": 1022976
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3242583841458426e-05,
+      "loss": 3.9826,
+      "step": 1023488
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3234197893947906e-05,
+      "loss": 3.9887,
+      "step": 1024000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3225811946437386e-05,
+      "loss": 3.979,
+      "step": 1024512
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.321742599892686e-05,
+      "loss": 3.9766,
+      "step": 1025024
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.320904005141634e-05,
+      "loss": 3.9971,
+      "step": 1025536
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.320065410390582e-05,
+      "loss": 3.9904,
+      "step": 1026048
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.31922681563953e-05,
+      "loss": 3.9801,
+      "step": 1026560
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.318388220888478e-05,
+      "loss": 3.9711,
+      "step": 1027072
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.317551264017799e-05,
+      "loss": 3.9837,
+      "step": 1027584
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3167126692667475e-05,
+      "loss": 3.9855,
+      "step": 1028096
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3158740745156955e-05,
+      "loss": 4.0008,
+      "step": 1028608
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3150354797646435e-05,
+      "loss": 3.9813,
+      "step": 1029120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3141968850135915e-05,
+      "loss": 3.9936,
+      "step": 1029632
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3133599281429124e-05,
+      "loss": 4.0012,
+      "step": 1030144
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.3125213333918604e-05,
+      "loss": 3.9819,
+      "step": 1030656
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.311684376521181e-05,
+      "loss": 3.99,
+      "step": 1031168
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.310845781770129e-05,
+      "loss": 3.99,
+      "step": 1031680
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.310007187019077e-05,
+      "loss": 3.9873,
+      "step": 1032192
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.309168592268025e-05,
+      "loss": 3.9867,
+      "step": 1032704
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.308329997516973e-05,
+      "loss": 3.9814,
+      "step": 1033216
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.307491402765921e-05,
+      "loss": 3.992,
+      "step": 1033728
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.306652808014869e-05,
+      "loss": 4.0036,
+      "step": 1034240
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.305814213263817e-05,
+      "loss": 3.99,
+      "step": 1034752
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.304975618512766e-05,
+      "loss": 3.9706,
+      "step": 1035264
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.304138661642087e-05,
+      "loss": 3.9848,
+      "step": 1035776
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.303301704771408e-05,
+      "loss": 3.991,
+      "step": 1036288
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.302464747900729e-05,
+      "loss": 3.9869,
+      "step": 1036800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.301626153149677e-05,
+      "loss": 3.9904,
+      "step": 1037312
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.300787558398625e-05,
+      "loss": 3.9872,
+      "step": 1037824
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2999489636475727e-05,
+      "loss": 3.9902,
+      "step": 1038336
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2991103688965207e-05,
+      "loss": 3.9766,
+      "step": 1038848
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2982717741454686e-05,
+      "loss": 3.972,
+      "step": 1039360
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2974348172747896e-05,
+      "loss": 3.9805,
+      "step": 1039872
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.296596222523738e-05,
+      "loss": 3.9796,
+      "step": 1040384
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.295757627772686e-05,
+      "loss": 3.9963,
+      "step": 1040896
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.294919033021634e-05,
+      "loss": 3.9803,
+      "step": 1041408
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.294080438270582e-05,
+      "loss": 3.9757,
+      "step": 1041920
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.29324184351953e-05,
+      "loss": 3.9875,
+      "step": 1042432
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.292403248768478e-05,
+      "loss": 3.9823,
+      "step": 1042944
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.291564654017426e-05,
+      "loss": 3.9919,
+      "step": 1043456
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.290726059266374e-05,
+      "loss": 3.9834,
+      "step": 1043968
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.289887464515322e-05,
+      "loss": 3.9842,
+      "step": 1044480
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.289050507644643e-05,
+      "loss": 3.9811,
+      "step": 1044992
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.288211912893591e-05,
+      "loss": 3.9852,
+      "step": 1045504
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.287373318142539e-05,
+      "loss": 3.9769,
+      "step": 1046016
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2865347233914864e-05,
+      "loss": 3.9804,
+      "step": 1046528
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2856961286404344e-05,
+      "loss": 3.9897,
+      "step": 1047040
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.284857533889383e-05,
+      "loss": 3.9798,
+      "step": 1047552
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.284018939138331e-05,
+      "loss": 3.9771,
+      "step": 1048064
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.283180344387279e-05,
+      "loss": 3.9769,
+      "step": 1048576
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.282341749636227e-05,
+      "loss": 3.9819,
+      "step": 1049088
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.281504792765548e-05,
+      "loss": 3.9831,
+      "step": 1049600
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.280666198014496e-05,
+      "loss": 3.9808,
+      "step": 1050112
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.279827603263444e-05,
+      "loss": 3.9839,
+      "step": 1050624
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.278989008512392e-05,
+      "loss": 3.9911,
+      "step": 1051136
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.27815041376134e-05,
+      "loss": 3.9904,
+      "step": 1051648
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.277315094771034e-05,
+      "loss": 3.9868,
+      "step": 1052160
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.276476500019982e-05,
+      "loss": 3.9882,
+      "step": 1052672
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.27563790526893e-05,
+      "loss": 3.9818,
+      "step": 1053184
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2747993105178784e-05,
+      "loss": 3.9836,
+      "step": 1053696
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2739623536471994e-05,
+      "loss": 3.9775,
+      "step": 1054208
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2731237588961474e-05,
+      "loss": 3.9833,
+      "step": 1054720
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2722851641450954e-05,
+      "loss": 3.9761,
+      "step": 1055232
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2714465693940433e-05,
+      "loss": 3.9851,
+      "step": 1055744
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.270609612523364e-05,
+      "loss": 3.9877,
+      "step": 1056256
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.269771017772312e-05,
+      "loss": 3.983,
+      "step": 1056768
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.26893242302126e-05,
+      "loss": 3.9811,
+      "step": 1057280
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.268093828270208e-05,
+      "loss": 3.9763,
+      "step": 1057792
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.267255233519156e-05,
+      "loss": 3.9939,
+      "step": 1058304
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.266416638768104e-05,
+      "loss": 3.9814,
+      "step": 1058816
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.265579681897425e-05,
+      "loss": 3.9797,
+      "step": 1059328
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.264741087146374e-05,
+      "loss": 3.9586,
+      "step": 1059840
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.263902492395322e-05,
+      "loss": 3.9831,
+      "step": 1060352
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.26306389764427e-05,
+      "loss": 3.9888,
+      "step": 1060864
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.262225302893218e-05,
+      "loss": 3.9956,
+      "step": 1061376
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.261386708142166e-05,
+      "loss": 3.9837,
+      "step": 1061888
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.260548113391114e-05,
+      "loss": 3.9784,
+      "step": 1062400
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.259709518640062e-05,
+      "loss": 3.9837,
+      "step": 1062912
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.25887092388901e-05,
+      "loss": 3.9992,
+      "step": 1063424
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.258033967018331e-05,
+      "loss": 3.9731,
+      "step": 1063936
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.257195372267279e-05,
+      "loss": 3.9848,
+      "step": 1064448
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2563584153965996e-05,
+      "loss": 3.9818,
+      "step": 1064960
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2555198206455476e-05,
+      "loss": 3.9852,
+      "step": 1065472
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2546812258944956e-05,
+      "loss": 3.9839,
+      "step": 1065984
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.2538426311434436e-05,
+      "loss": 3.9833,
+      "step": 1066496
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.253004036392392e-05,
+      "loss": 3.9759,
+      "step": 1067008
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.25216544164134e-05,
+      "loss": 3.9906,
+      "step": 1067520
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.251326846890288e-05,
+      "loss": 3.9838,
+      "step": 1068032
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.048372268676758,
+      "eval_runtime": 288.8396,
+      "eval_samples_per_second": 1321.117,
+      "eval_steps_per_second": 41.286,
+      "step": 1068480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2504882521392356e-05,
+      "loss": 3.9771,
+      "step": 1068544
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2496496573881836e-05,
+      "loss": 3.9882,
+      "step": 1069056
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2488110626371316e-05,
+      "loss": 3.9795,
+      "step": 1069568
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2479724678860796e-05,
+      "loss": 3.9887,
+      "step": 1070080
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2471338731350276e-05,
+      "loss": 3.9877,
+      "step": 1070592
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2462952783839755e-05,
+      "loss": 3.9873,
+      "step": 1071104
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2454566836329235e-05,
+      "loss": 3.9787,
+      "step": 1071616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2446180888818715e-05,
+      "loss": 3.9702,
+      "step": 1072128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2437794941308195e-05,
+      "loss": 3.9772,
+      "step": 1072640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2429408993797675e-05,
+      "loss": 3.9928,
+      "step": 1073152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2421023046287155e-05,
+      "loss": 3.9824,
+      "step": 1073664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2412637098776635e-05,
+      "loss": 3.9823,
+      "step": 1074176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.240426753006985e-05,
+      "loss": 3.9949,
+      "step": 1074688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.239588158255933e-05,
+      "loss": 3.976,
+      "step": 1075200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.238749563504881e-05,
+      "loss": 3.9747,
+      "step": 1075712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.237910968753829e-05,
+      "loss": 3.9692,
+      "step": 1076224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.237072374002777e-05,
+      "loss": 3.9805,
+      "step": 1076736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2362337792517244e-05,
+      "loss": 3.978,
+      "step": 1077248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2353951845006724e-05,
+      "loss": 3.9773,
+      "step": 1077760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2345565897496204e-05,
+      "loss": 3.9814,
+      "step": 1078272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.233719632878941e-05,
+      "loss": 3.9971,
+      "step": 1078784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.232881038127889e-05,
+      "loss": 3.9823,
+      "step": 1079296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.232042443376837e-05,
+      "loss": 3.9778,
+      "step": 1079808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.231203848625785e-05,
+      "loss": 3.982,
+      "step": 1080320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.230365253874734e-05,
+      "loss": 3.9795,
+      "step": 1080832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.229526659123682e-05,
+      "loss": 3.9821,
+      "step": 1081344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.22868806437263e-05,
+      "loss": 3.9667,
+      "step": 1081856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.227849469621578e-05,
+      "loss": 3.9731,
+      "step": 1082368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.227012512750899e-05,
+      "loss": 3.9789,
+      "step": 1082880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.226173917999847e-05,
+      "loss": 3.97,
+      "step": 1083392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.225335323248795e-05,
+      "loss": 3.973,
+      "step": 1083904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.224496728497743e-05,
+      "loss": 3.9835,
+      "step": 1084416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.223658133746691e-05,
+      "loss": 3.9788,
+      "step": 1084928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.222819538995639e-05,
+      "loss": 3.9872,
+      "step": 1085440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.221980944244587e-05,
+      "loss": 3.9825,
+      "step": 1085952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.221143987373908e-05,
+      "loss": 3.9798,
+      "step": 1086464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.220305392622856e-05,
+      "loss": 3.9893,
+      "step": 1086976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.219466797871804e-05,
+      "loss": 3.9664,
+      "step": 1087488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2186282031207524e-05,
+      "loss": 3.9739,
+      "step": 1088000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2177896083697004e-05,
+      "loss": 3.9711,
+      "step": 1088512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2169510136186484e-05,
+      "loss": 3.9671,
+      "step": 1089024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.216112418867596e-05,
+      "loss": 3.9801,
+      "step": 1089536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.215273824116544e-05,
+      "loss": 3.9799,
+      "step": 1090048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.214436867245865e-05,
+      "loss": 3.9757,
+      "step": 1090560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2135982724948126e-05,
+      "loss": 3.9762,
+      "step": 1091072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2127596777437606e-05,
+      "loss": 3.9767,
+      "step": 1091584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2119210829927086e-05,
+      "loss": 3.9792,
+      "step": 1092096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.21108412612203e-05,
+      "loss": 3.9811,
+      "step": 1092608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2102455313709775e-05,
+      "loss": 3.9601,
+      "step": 1093120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.209406936619926e-05,
+      "loss": 3.9711,
+      "step": 1093632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.208568341868874e-05,
+      "loss": 3.9873,
+      "step": 1094144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.207729747117822e-05,
+      "loss": 3.9836,
+      "step": 1094656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.20689115236677e-05,
+      "loss": 3.9756,
+      "step": 1095168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.206052557615718e-05,
+      "loss": 3.9676,
+      "step": 1095680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.205215600745039e-05,
+      "loss": 3.9743,
+      "step": 1096192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.204377005993987e-05,
+      "loss": 3.9675,
+      "step": 1096704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.203538411242935e-05,
+      "loss": 3.9757,
+      "step": 1097216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.202699816491883e-05,
+      "loss": 3.9642,
+      "step": 1097728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.201861221740831e-05,
+      "loss": 3.9802,
+      "step": 1098240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.201022626989779e-05,
+      "loss": 3.9757,
+      "step": 1098752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.200184032238727e-05,
+      "loss": 3.9728,
+      "step": 1099264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.199347075368048e-05,
+      "loss": 3.9678,
+      "step": 1099776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.198508480616996e-05,
+      "loss": 3.9762,
+      "step": 1100288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.197669885865944e-05,
+      "loss": 3.9643,
+      "step": 1100800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1968312911148926e-05,
+      "loss": 3.9666,
+      "step": 1101312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1959926963638406e-05,
+      "loss": 3.9823,
+      "step": 1101824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1951541016127886e-05,
+      "loss": 3.9776,
+      "step": 1102336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1943155068617366e-05,
+      "loss": 3.968,
+      "step": 1102848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1934769121106846e-05,
+      "loss": 3.9563,
+      "step": 1103360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1926399552400055e-05,
+      "loss": 3.9701,
+      "step": 1103872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1918013604889535e-05,
+      "loss": 3.9744,
+      "step": 1104384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1909627657379015e-05,
+      "loss": 3.9843,
+      "step": 1104896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1901258088672224e-05,
+      "loss": 3.9661,
+      "step": 1105408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1892872141161704e-05,
+      "loss": 3.9826,
+      "step": 1105920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1884486193651184e-05,
+      "loss": 3.98,
+      "step": 1106432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1876100246140664e-05,
+      "loss": 3.9744,
+      "step": 1106944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1867714298630144e-05,
+      "loss": 3.9757,
+      "step": 1107456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1859328351119624e-05,
+      "loss": 3.9734,
+      "step": 1107968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1850942403609104e-05,
+      "loss": 3.9777,
+      "step": 1108480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1842556456098584e-05,
+      "loss": 3.9718,
+      "step": 1108992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1834170508588064e-05,
+      "loss": 3.9703,
+      "step": 1109504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.182580093988127e-05,
+      "loss": 3.9773,
+      "step": 1110016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.181741499237075e-05,
+      "loss": 3.9881,
+      "step": 1110528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.180902904486023e-05,
+      "loss": 3.9753,
+      "step": 1111040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.180064309734971e-05,
+      "loss": 3.9598,
+      "step": 1111552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.179228990744666e-05,
+      "loss": 3.9722,
+      "step": 1112064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.178390395993614e-05,
+      "loss": 3.9755,
+      "step": 1112576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.177551801242562e-05,
+      "loss": 3.978,
+      "step": 1113088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.17671320649151e-05,
+      "loss": 3.9775,
+      "step": 1113600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1758762496208314e-05,
+      "loss": 3.9735,
+      "step": 1114112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1750376548697794e-05,
+      "loss": 3.9733,
+      "step": 1114624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.174199060118727e-05,
+      "loss": 3.9637,
+      "step": 1115136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.173360465367675e-05,
+      "loss": 3.9592,
+      "step": 1115648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.172521870616623e-05,
+      "loss": 3.9647,
+      "step": 1116160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1716849137459436e-05,
+      "loss": 3.9661,
+      "step": 1116672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1708463189948916e-05,
+      "loss": 3.9862,
+      "step": 1117184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1700077242438396e-05,
+      "loss": 3.9655,
+      "step": 1117696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1691691294927876e-05,
+      "loss": 3.9603,
+      "step": 1118208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1683305347417356e-05,
+      "loss": 3.9772,
+      "step": 1118720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1674919399906836e-05,
+      "loss": 3.971,
+      "step": 1119232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.166654983120005e-05,
+      "loss": 3.981,
+      "step": 1119744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.165816388368953e-05,
+      "loss": 3.9702,
+      "step": 1120256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.164977793617901e-05,
+      "loss": 3.9682,
+      "step": 1120768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.164139198866849e-05,
+      "loss": 3.9706,
+      "step": 1121280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.163300604115797e-05,
+      "loss": 3.9728,
+      "step": 1121792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.162462009364745e-05,
+      "loss": 3.9605,
+      "step": 1122304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.161623414613693e-05,
+      "loss": 3.9701,
+      "step": 1122816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.160784819862641e-05,
+      "loss": 3.9706,
+      "step": 1123328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.159947862991962e-05,
+      "loss": 3.966,
+      "step": 1123840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.15910926824091e-05,
+      "loss": 3.9636,
+      "step": 1124352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.158272311370231e-05,
+      "loss": 3.9659,
+      "step": 1124864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.157433716619179e-05,
+      "loss": 3.9631,
+      "step": 1125376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.156595121868127e-05,
+      "loss": 3.9696,
+      "step": 1125888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1557565271170756e-05,
+      "loss": 3.9684,
+      "step": 1126400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1549179323660236e-05,
+      "loss": 3.9685,
+      "step": 1126912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1540809754953445e-05,
+      "loss": 3.9766,
+      "step": 1127424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1532423807442925e-05,
+      "loss": 3.9803,
+      "step": 1127936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1524037859932405e-05,
+      "loss": 3.9735,
+      "step": 1128448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1515651912421885e-05,
+      "loss": 3.9721,
+      "step": 1128960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1507265964911365e-05,
+      "loss": 3.9709,
+      "step": 1129472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1498880017400845e-05,
+      "loss": 3.9716,
+      "step": 1129984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1490494069890325e-05,
+      "loss": 3.9601,
+      "step": 1130496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.14821081223798e-05,
+      "loss": 3.9703,
+      "step": 1131008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1473738553673014e-05,
+      "loss": 3.9665,
+      "step": 1131520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1465352606162494e-05,
+      "loss": 3.9708,
+      "step": 1132032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1456966658651974e-05,
+      "loss": 3.9708,
+      "step": 1132544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1448580711141454e-05,
+      "loss": 3.9747,
+      "step": 1133056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.144021114243467e-05,
+      "loss": 3.9667,
+      "step": 1133568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.143184157372788e-05,
+      "loss": 3.9644,
+      "step": 1134080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.142345562621736e-05,
+      "loss": 3.9797,
+      "step": 1134592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.141506967870684e-05,
+      "loss": 3.9714,
+      "step": 1135104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.140668373119632e-05,
+      "loss": 3.9645,
+      "step": 1135616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.13982977836858e-05,
+      "loss": 3.9474,
+      "step": 1136128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.138991183617527e-05,
+      "loss": 3.9663,
+      "step": 1136640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.138152588866475e-05,
+      "loss": 3.9751,
+      "step": 1137152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.137313994115423e-05,
+      "loss": 3.9835,
+      "step": 1137664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.136477037244745e-05,
+      "loss": 3.9713,
+      "step": 1138176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.135638442493693e-05,
+      "loss": 3.9649,
+      "step": 1138688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.134799847742641e-05,
+      "loss": 3.9718,
+      "step": 1139200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.133961252991589e-05,
+      "loss": 3.9847,
+      "step": 1139712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.133122658240537e-05,
+      "loss": 3.9637,
+      "step": 1140224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1322857013698576e-05,
+      "loss": 3.9687,
+      "step": 1140736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1314471066188056e-05,
+      "loss": 3.9703,
+      "step": 1141248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1306085118677536e-05,
+      "loss": 3.9724,
+      "step": 1141760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1297715549970745e-05,
+      "loss": 3.972,
+      "step": 1142272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1289329602460225e-05,
+      "loss": 3.9687,
+      "step": 1142784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1280943654949705e-05,
+      "loss": 3.9605,
+      "step": 1143296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1272557707439185e-05,
+      "loss": 3.9807,
+      "step": 1143808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1264188138732394e-05,
+      "loss": 3.9698,
+      "step": 1144320
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.040899753570557,
+      "eval_runtime": 294.7617,
+      "eval_samples_per_second": 1294.574,
+      "eval_steps_per_second": 40.456,
+      "step": 1144800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.125580219122188e-05,
+      "loss": 3.9629,
+      "step": 1144832
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.124741624371136e-05,
+      "loss": 3.9745,
+      "step": 1145344
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.123903029620084e-05,
+      "loss": 3.9629,
+      "step": 1145856
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.123064434869032e-05,
+      "loss": 3.9786,
+      "step": 1146368
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.122227477998353e-05,
+      "loss": 3.9713,
+      "step": 1146880
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.121388883247301e-05,
+      "loss": 3.9775,
+      "step": 1147392
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.120550288496249e-05,
+      "loss": 3.9633,
+      "step": 1147904
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.119711693745197e-05,
+      "loss": 3.9573,
+      "step": 1148416
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.118874736874518e-05,
+      "loss": 3.9589,
+      "step": 1148928
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.118036142123466e-05,
+      "loss": 3.9867,
+      "step": 1149440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.117197547372414e-05,
+      "loss": 3.9661,
+      "step": 1149952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.116358952621362e-05,
+      "loss": 3.97,
+      "step": 1150464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.11552035787031e-05,
+      "loss": 3.9772,
+      "step": 1150976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1146834009996315e-05,
+      "loss": 3.9708,
+      "step": 1151488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1138448062485795e-05,
+      "loss": 3.9577,
+      "step": 1152000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1130062114975274e-05,
+      "loss": 3.9602,
+      "step": 1152512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1121676167464754e-05,
+      "loss": 3.9619,
+      "step": 1153024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1113290219954234e-05,
+      "loss": 3.9648,
+      "step": 1153536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1104904272443714e-05,
+      "loss": 3.9652,
+      "step": 1154048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1096518324933194e-05,
+      "loss": 3.9714,
+      "step": 1154560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1088132377422674e-05,
+      "loss": 3.9853,
+      "step": 1155072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.107977918751961e-05,
+      "loss": 3.968,
+      "step": 1155584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.107139324000909e-05,
+      "loss": 3.9671,
+      "step": 1156096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.106300729249857e-05,
+      "loss": 3.9651,
+      "step": 1156608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.105462134498805e-05,
+      "loss": 3.9708,
+      "step": 1157120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.104623539747753e-05,
+      "loss": 3.9649,
+      "step": 1157632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.103784944996702e-05,
+      "loss": 3.9549,
+      "step": 1158144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.102947988126023e-05,
+      "loss": 3.9591,
+      "step": 1158656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.102109393374971e-05,
+      "loss": 3.9665,
+      "step": 1159168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.101270798623919e-05,
+      "loss": 3.9562,
+      "step": 1159680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.100432203872867e-05,
+      "loss": 3.9561,
+      "step": 1160192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.099593609121815e-05,
+      "loss": 3.9713,
+      "step": 1160704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.098755014370763e-05,
+      "loss": 3.9666,
+      "step": 1161216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.097918057500084e-05,
+      "loss": 3.9678,
+      "step": 1161728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.097079462749032e-05,
+      "loss": 3.9741,
+      "step": 1162240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0962425058783526e-05,
+      "loss": 3.9685,
+      "step": 1162752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0954039111273006e-05,
+      "loss": 3.9712,
+      "step": 1163264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0945653163762486e-05,
+      "loss": 3.9568,
+      "step": 1163776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.093726721625197e-05,
+      "loss": 3.9609,
+      "step": 1164288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.092889764754518e-05,
+      "loss": 3.9574,
+      "step": 1164800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.092051170003466e-05,
+      "loss": 3.9576,
+      "step": 1165312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.091212575252414e-05,
+      "loss": 3.9607,
+      "step": 1165824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.090373980501362e-05,
+      "loss": 3.9717,
+      "step": 1166336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.08953538575031e-05,
+      "loss": 3.9573,
+      "step": 1166848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.088696790999258e-05,
+      "loss": 3.9657,
+      "step": 1167360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0878581962482055e-05,
+      "loss": 3.9627,
+      "step": 1167872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0870196014971535e-05,
+      "loss": 3.9668,
+      "step": 1168384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.086182644626475e-05,
+      "loss": 3.9688,
+      "step": 1168896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0853440498754224e-05,
+      "loss": 3.9484,
+      "step": 1169408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.084505455124371e-05,
+      "loss": 3.9586,
+      "step": 1169920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.083666860373319e-05,
+      "loss": 3.9717,
+      "step": 1170432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0828299035026406e-05,
+      "loss": 3.9733,
+      "step": 1170944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.081991308751588e-05,
+      "loss": 3.9576,
+      "step": 1171456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.081152714000536e-05,
+      "loss": 3.9585,
+      "step": 1171968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.080314119249484e-05,
+      "loss": 3.962,
+      "step": 1172480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.079475524498432e-05,
+      "loss": 3.9526,
+      "step": 1172992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.07863692974738e-05,
+      "loss": 3.9643,
+      "step": 1173504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.077798334996328e-05,
+      "loss": 3.9499,
+      "step": 1174016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.076959740245276e-05,
+      "loss": 3.9676,
+      "step": 1174528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.076121145494224e-05,
+      "loss": 3.9601,
+      "step": 1175040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.075282550743172e-05,
+      "loss": 3.9633,
+      "step": 1175552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.07444395599212e-05,
+      "loss": 3.9518,
+      "step": 1176064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.073605361241068e-05,
+      "loss": 3.9648,
+      "step": 1176576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0727700422507624e-05,
+      "loss": 3.9473,
+      "step": 1177088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0719314474997104e-05,
+      "loss": 3.9576,
+      "step": 1177600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0710928527486584e-05,
+      "loss": 3.9672,
+      "step": 1178112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0702542579976064e-05,
+      "loss": 3.9663,
+      "step": 1178624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0694156632465544e-05,
+      "loss": 3.9525,
+      "step": 1179136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0685770684955024e-05,
+      "loss": 3.9492,
+      "step": 1179648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0677384737444504e-05,
+      "loss": 3.9574,
+      "step": 1180160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0668998789933984e-05,
+      "loss": 3.9603,
+      "step": 1180672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.066062922122719e-05,
+      "loss": 3.9725,
+      "step": 1181184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.065224327371667e-05,
+      "loss": 3.9533,
+      "step": 1181696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.064385732620615e-05,
+      "loss": 3.9718,
+      "step": 1182208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.063547137869563e-05,
+      "loss": 3.9682,
+      "step": 1182720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.062710180998884e-05,
+      "loss": 3.9644,
+      "step": 1183232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.061871586247833e-05,
+      "loss": 3.9642,
+      "step": 1183744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.061032991496781e-05,
+      "loss": 3.962,
+      "step": 1184256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.060194396745729e-05,
+      "loss": 3.9653,
+      "step": 1184768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.059355801994677e-05,
+      "loss": 3.9584,
+      "step": 1185280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.058520483004371e-05,
+      "loss": 3.9583,
+      "step": 1185792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.057681888253319e-05,
+      "loss": 3.9633,
+      "step": 1186304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0568432935022667e-05,
+      "loss": 3.9774,
+      "step": 1186816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0560046987512147e-05,
+      "loss": 3.9621,
+      "step": 1187328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0551661040001626e-05,
+      "loss": 3.9538,
+      "step": 1187840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0543275092491106e-05,
+      "loss": 3.9517,
+      "step": 1188352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0534889144980586e-05,
+      "loss": 3.9659,
+      "step": 1188864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0526519576273796e-05,
+      "loss": 3.9612,
+      "step": 1189376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.051813362876328e-05,
+      "loss": 3.9679,
+      "step": 1189888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.050974768125276e-05,
+      "loss": 3.9584,
+      "step": 1190400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.050136173374224e-05,
+      "loss": 3.9622,
+      "step": 1190912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0492975786231715e-05,
+      "loss": 3.9515,
+      "step": 1191424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0484589838721195e-05,
+      "loss": 3.9487,
+      "step": 1191936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0476203891210675e-05,
+      "loss": 3.9493,
+      "step": 1192448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0467817943700155e-05,
+      "loss": 3.9528,
+      "step": 1192960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0459448374993364e-05,
+      "loss": 3.9706,
+      "step": 1193472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.045107880628658e-05,
+      "loss": 3.9605,
+      "step": 1193984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.044269285877606e-05,
+      "loss": 3.9478,
+      "step": 1194496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0434306911265537e-05,
+      "loss": 3.9605,
+      "step": 1195008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0425920963755017e-05,
+      "loss": 3.9616,
+      "step": 1195520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0417551395048233e-05,
+      "loss": 3.9658,
+      "step": 1196032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0409165447537712e-05,
+      "loss": 3.9579,
+      "step": 1196544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.040077950002719e-05,
+      "loss": 3.9566,
+      "step": 1197056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.039239355251667e-05,
+      "loss": 3.9566,
+      "step": 1197568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.038400760500615e-05,
+      "loss": 3.9654,
+      "step": 1198080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.037562165749563e-05,
+      "loss": 3.9456,
+      "step": 1198592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.036723570998511e-05,
+      "loss": 3.9559,
+      "step": 1199104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.035884976247459e-05,
+      "loss": 3.9619,
+      "step": 1199616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.035046381496407e-05,
+      "loss": 3.9528,
+      "step": 1200128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.034209424625728e-05,
+      "loss": 3.9525,
+      "step": 1200640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.033370829874676e-05,
+      "loss": 3.9586,
+      "step": 1201152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.032532235123624e-05,
+      "loss": 3.9513,
+      "step": 1201664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.031693640372572e-05,
+      "loss": 3.9588,
+      "step": 1202176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.03085504562152e-05,
+      "loss": 3.9569,
+      "step": 1202688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.030018088750841e-05,
+      "loss": 3.9534,
+      "step": 1203200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.029179493999789e-05,
+      "loss": 3.9669,
+      "step": 1203712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0283408992487373e-05,
+      "loss": 3.9663,
+      "step": 1204224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0275023044976853e-05,
+      "loss": 3.9604,
+      "step": 1204736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0266637097466333e-05,
+      "loss": 3.963,
+      "step": 1205248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0258251149955813e-05,
+      "loss": 3.9575,
+      "step": 1205760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0249865202445293e-05,
+      "loss": 3.9584,
+      "step": 1206272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0241479254934773e-05,
+      "loss": 3.9525,
+      "step": 1206784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.023309330742425e-05,
+      "loss": 3.9523,
+      "step": 1207296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0224723738717466e-05,
+      "loss": 3.9574,
+      "step": 1207808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0216337791206946e-05,
+      "loss": 3.9581,
+      "step": 1208320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0207951843696426e-05,
+      "loss": 3.9546,
+      "step": 1208832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.01995658961859e-05,
+      "loss": 3.9637,
+      "step": 1209344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0191196327479115e-05,
+      "loss": 3.9566,
+      "step": 1209856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0182826758772327e-05,
+      "loss": 3.9525,
+      "step": 1210368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0174440811261807e-05,
+      "loss": 3.9649,
+      "step": 1210880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0166054863751287e-05,
+      "loss": 3.9604,
+      "step": 1211392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0157668916240767e-05,
+      "loss": 3.9512,
+      "step": 1211904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0149282968730247e-05,
+      "loss": 3.9395,
+      "step": 1212416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0140913400023456e-05,
+      "loss": 3.9543,
+      "step": 1212928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0132527452512936e-05,
+      "loss": 3.9677,
+      "step": 1213440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.012414150500242e-05,
+      "loss": 3.9686,
+      "step": 1213952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.01157555574919e-05,
+      "loss": 3.9578,
+      "step": 1214464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0107369609981372e-05,
+      "loss": 3.9539,
+      "step": 1214976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0098983662470852e-05,
+      "loss": 3.9604,
+      "step": 1215488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0090597714960332e-05,
+      "loss": 3.9732,
+      "step": 1216000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0082228146253548e-05,
+      "loss": 3.9532,
+      "step": 1216512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0073842198743025e-05,
+      "loss": 3.9583,
+      "step": 1217024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0065456251232505e-05,
+      "loss": 3.9564,
+      "step": 1217536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0057070303721985e-05,
+      "loss": 3.9586,
+      "step": 1218048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0048700735015194e-05,
+      "loss": 3.9581,
+      "step": 1218560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0040314787504674e-05,
+      "loss": 3.9647,
+      "step": 1219072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0031928839994157e-05,
+      "loss": 3.949,
+      "step": 1219584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0023542892483637e-05,
+      "loss": 3.9674,
+      "step": 1220096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0015156944973117e-05,
+      "loss": 3.9585,
+      "step": 1220608
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.0006770997462597e-05,
+      "loss": 3.9498,
+      "step": 1221120
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.034625053405762,
+      "eval_runtime": 292.0933,
+      "eval_samples_per_second": 1306.401,
+      "eval_steps_per_second": 40.826,
+      "step": 1221120
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9998385049952077e-05,
+      "loss": 3.9623,
+      "step": 1221632
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9989999102441557e-05,
+      "loss": 3.9531,
+      "step": 1222144
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9981629533734766e-05,
+      "loss": 3.9651,
+      "step": 1222656
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.997324358622425e-05,
+      "loss": 3.9595,
+      "step": 1223168
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.996485763871373e-05,
+      "loss": 3.9668,
+      "step": 1223680
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.995647169120321e-05,
+      "loss": 3.9522,
+      "step": 1224192
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.994808574369269e-05,
+      "loss": 3.9466,
+      "step": 1224704
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.99397161749859e-05,
+      "loss": 3.948,
+      "step": 1225216
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9931330227475378e-05,
+      "loss": 3.9732,
+      "step": 1225728
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9922944279964858e-05,
+      "loss": 3.9579,
+      "step": 1226240
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.991455833245434e-05,
+      "loss": 3.9536,
+      "step": 1226752
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.990618876374755e-05,
+      "loss": 3.9653,
+      "step": 1227264
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.989781919504076e-05,
+      "loss": 3.9604,
+      "step": 1227776
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.988943324753024e-05,
+      "loss": 3.9504,
+      "step": 1228288
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.988104730001972e-05,
+      "loss": 3.9528,
+      "step": 1228800
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9872661352509203e-05,
+      "loss": 3.9481,
+      "step": 1229312
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9864275404998683e-05,
+      "loss": 3.9556,
+      "step": 1229824
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9855905836291892e-05,
+      "loss": 3.9516,
+      "step": 1230336
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9847519888781372e-05,
+      "loss": 3.9627,
+      "step": 1230848
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9839133941270852e-05,
+      "loss": 3.9756,
+      "step": 1231360
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9830747993760332e-05,
+      "loss": 3.957,
+      "step": 1231872
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9822362046249812e-05,
+      "loss": 3.9509,
+      "step": 1232384
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9813976098739292e-05,
+      "loss": 3.9564,
+      "step": 1232896
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9805590151228775e-05,
+      "loss": 3.9607,
+      "step": 1233408
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9797204203718255e-05,
+      "loss": 3.9509,
+      "step": 1233920
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9788834635011464e-05,
+      "loss": 3.9478,
+      "step": 1234432
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9780448687500944e-05,
+      "loss": 3.9453,
+      "step": 1234944
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9772079118794153e-05,
+      "loss": 3.9617,
+      "step": 1235456
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.9763693171283637e-05,
+      "loss": 3.9419,
+      "step": 1235968
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9755307223773117e-05,
+      "loss": 3.9462,
+      "step": 1236480
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9746921276262597e-05,
+      "loss": 3.96,
+      "step": 1236992
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9738551707555806e-05,
+      "loss": 3.9602,
+      "step": 1237504
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9730165760045286e-05,
+      "loss": 3.9571,
+      "step": 1238016
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9721779812534766e-05,
+      "loss": 3.9628,
+      "step": 1238528
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9713410243827978e-05,
+      "loss": 3.955,
+      "step": 1239040
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9705024296317458e-05,
+      "loss": 3.9614,
+      "step": 1239552
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9696638348806938e-05,
+      "loss": 3.9448,
+      "step": 1240064
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9688252401296418e-05,
+      "loss": 3.9527,
+      "step": 1240576
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9679866453785898e-05,
+      "loss": 3.9458,
+      "step": 1241088
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9671480506275378e-05,
+      "loss": 3.947,
+      "step": 1241600
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9663094558764854e-05,
+      "loss": 3.9513,
+      "step": 1242112
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9654708611254334e-05,
+      "loss": 3.9606,
+      "step": 1242624
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9646322663743814e-05,
+      "loss": 3.9484,
+      "step": 1243136
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9637936716233294e-05,
+      "loss": 3.9524,
+      "step": 1243648
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9629550768722774e-05,
+      "loss": 3.9513,
+      "step": 1244160
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9621164821212254e-05,
+      "loss": 3.9561,
+      "step": 1244672
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9612795252505467e-05,
+      "loss": 3.9603,
+      "step": 1245184
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9604409304994947e-05,
+      "loss": 3.9429,
+      "step": 1245696
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9596023357484427e-05,
+      "loss": 3.9423,
+      "step": 1246208
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9587637409973907e-05,
+      "loss": 3.9607,
+      "step": 1246720
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9579251462463386e-05,
+      "loss": 3.9636,
+      "step": 1247232
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9570881893756596e-05,
+      "loss": 3.9462,
+      "step": 1247744
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9562495946246076e-05,
+      "loss": 3.9506,
+      "step": 1248256
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9554126377539288e-05,
+      "loss": 3.9522,
+      "step": 1248768
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9545740430028768e-05,
+      "loss": 3.9402,
+      "step": 1249280
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9537354482518248e-05,
+      "loss": 3.9514,
+      "step": 1249792
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9528984913811457e-05,
+      "loss": 3.9404,
+      "step": 1250304
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9520598966300937e-05,
+      "loss": 3.9569,
+      "step": 1250816
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.951221301879042e-05,
+      "loss": 3.9519,
+      "step": 1251328
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.95038270712799e-05,
+      "loss": 3.9513,
+      "step": 1251840
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.949544112376938e-05,
+      "loss": 3.9387,
+      "step": 1252352
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.948705517625886e-05,
+      "loss": 3.9578,
+      "step": 1252864
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.947866922874834e-05,
+      "loss": 3.9381,
+      "step": 1253376
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.947028328123782e-05,
+      "loss": 3.9491,
+      "step": 1253888
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.946191371253103e-05,
+      "loss": 3.9556,
+      "step": 1254400
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9453527765020513e-05,
+      "loss": 3.9578,
+      "step": 1254912
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9445141817509993e-05,
+      "loss": 3.9412,
+      "step": 1255424
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9436755869999472e-05,
+      "loss": 3.9407,
+      "step": 1255936
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.942838630129268e-05,
+      "loss": 3.9413,
+      "step": 1256448
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.942000035378216e-05,
+      "loss": 3.9532,
+      "step": 1256960
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.941161440627164e-05,
+      "loss": 3.9538,
+      "step": 1257472
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.940322845876112e-05,
+      "loss": 3.9467,
+      "step": 1257984
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9394842511250605e-05,
+      "loss": 3.9587,
+      "step": 1258496
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9386456563740085e-05,
+      "loss": 3.9621,
+      "step": 1259008
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9378070616229565e-05,
+      "loss": 3.953,
+      "step": 1259520
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9369684668719045e-05,
+      "loss": 3.9538,
+      "step": 1260032
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9361315100012254e-05,
+      "loss": 3.9518,
+      "step": 1260544
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9352945531305466e-05,
+      "loss": 3.958,
+      "step": 1261056
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9344559583794946e-05,
+      "loss": 3.9459,
+      "step": 1261568
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9336173636284426e-05,
+      "loss": 3.9497,
+      "step": 1262080
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9327787688773906e-05,
+      "loss": 3.9513,
+      "step": 1262592
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9319401741263386e-05,
+      "loss": 3.9654,
+      "step": 1263104
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9311015793752866e-05,
+      "loss": 3.9542,
+      "step": 1263616
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9302629846242343e-05,
+      "loss": 3.9391,
+      "step": 1264128
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9294243898731823e-05,
+      "loss": 3.9468,
+      "step": 1264640
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9285857951221302e-05,
+      "loss": 3.9494,
+      "step": 1265152
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.927748838251452e-05,
+      "loss": 3.9558,
+      "step": 1265664
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.926910243500399e-05,
+      "loss": 3.9554,
+      "step": 1266176
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.926071648749347e-05,
+      "loss": 3.9478,
+      "step": 1266688
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.925233053998295e-05,
+      "loss": 3.9502,
+      "step": 1267200
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9243944592472435e-05,
+      "loss": 3.9445,
+      "step": 1267712
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9235575023765644e-05,
+      "loss": 3.9378,
+      "step": 1268224
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9227189076255124e-05,
+      "loss": 3.9361,
+      "step": 1268736
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9218803128744604e-05,
+      "loss": 3.9486,
+      "step": 1269248
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9210417181234084e-05,
+      "loss": 3.9583,
+      "step": 1269760
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9202047612527296e-05,
+      "loss": 3.9496,
+      "step": 1270272
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9193661665016776e-05,
+      "loss": 3.9399,
+      "step": 1270784
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9185275717506256e-05,
+      "loss": 3.9466,
+      "step": 1271296
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9176889769995736e-05,
+      "loss": 3.9507,
+      "step": 1271808
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9168520201288945e-05,
+      "loss": 3.9554,
+      "step": 1272320
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9160134253778425e-05,
+      "loss": 3.9472,
+      "step": 1272832
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9151748306267905e-05,
+      "loss": 3.9478,
+      "step": 1273344
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9143362358757385e-05,
+      "loss": 3.9442,
+      "step": 1273856
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.913497641124687e-05,
+      "loss": 3.9566,
+      "step": 1274368
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9126606842540078e-05,
+      "loss": 3.9379,
+      "step": 1274880
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9118237273833287e-05,
+      "loss": 3.9428,
+      "step": 1275392
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9109851326322767e-05,
+      "loss": 3.9518,
+      "step": 1275904
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.910146537881225e-05,
+      "loss": 3.9467,
+      "step": 1276416
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.909307943130173e-05,
+      "loss": 3.9402,
+      "step": 1276928
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.908469348379121e-05,
+      "loss": 3.9461,
+      "step": 1277440
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.907630753628069e-05,
+      "loss": 3.9432,
+      "step": 1277952
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.906792158877017e-05,
+      "loss": 3.9468,
+      "step": 1278464
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.905953564125965e-05,
+      "loss": 3.9451,
+      "step": 1278976
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.905114969374913e-05,
+      "loss": 3.946,
+      "step": 1279488
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.904276374623861e-05,
+      "loss": 3.9547,
+      "step": 1280000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9034394177531822e-05,
+      "loss": 3.9571,
+      "step": 1280512
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9026008230021302e-05,
+      "loss": 3.9483,
+      "step": 1281024
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9017622282510782e-05,
+      "loss": 3.9566,
+      "step": 1281536
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9009236335000262e-05,
+      "loss": 3.9467,
+      "step": 1282048
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.900086676629347e-05,
+      "loss": 3.9506,
+      "step": 1282560
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.899248081878295e-05,
+      "loss": 3.9414,
+      "step": 1283072
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.898409487127243e-05,
+      "loss": 3.9452,
+      "step": 1283584
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8975725302565643e-05,
+      "loss": 3.9462,
+      "step": 1284096
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8967339355055123e-05,
+      "loss": 3.9475,
+      "step": 1284608
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8958953407544603e-05,
+      "loss": 3.9481,
+      "step": 1285120
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8950567460034083e-05,
+      "loss": 3.9501,
+      "step": 1285632
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8942181512523563e-05,
+      "loss": 3.95,
+      "step": 1286144
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8933811943816776e-05,
+      "loss": 3.9399,
+      "step": 1286656
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8925425996306256e-05,
+      "loss": 3.9534,
+      "step": 1287168
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8917040048795736e-05,
+      "loss": 3.9526,
+      "step": 1287680
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8908670480088945e-05,
+      "loss": 3.9389,
+      "step": 1288192
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8900284532578425e-05,
+      "loss": 3.9324,
+      "step": 1288704
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8891898585067905e-05,
+      "loss": 3.9426,
+      "step": 1289216
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8883512637557385e-05,
+      "loss": 3.956,
+      "step": 1289728
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8875126690046868e-05,
+      "loss": 3.9598,
+      "step": 1290240
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8866740742536348e-05,
+      "loss": 3.9418,
+      "step": 1290752
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.885835479502582e-05,
+      "loss": 3.9454,
+      "step": 1291264
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.88499688475153e-05,
+      "loss": 3.9558,
+      "step": 1291776
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.884158290000478e-05,
+      "loss": 3.9567,
+      "step": 1292288
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.883319695249426e-05,
+      "loss": 3.9488,
+      "step": 1292800
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8824811004983744e-05,
+      "loss": 3.9442,
+      "step": 1293312
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8816425057473224e-05,
+      "loss": 3.9494,
+      "step": 1293824
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8808055488766433e-05,
+      "loss": 3.948,
+      "step": 1294336
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8799669541255913e-05,
+      "loss": 3.9438,
+      "step": 1294848
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8791283593745393e-05,
+      "loss": 3.9556,
+      "step": 1295360
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8782897646234873e-05,
+      "loss": 3.9419,
+      "step": 1295872
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8774528077528086e-05,
+      "loss": 3.9553,
+      "step": 1296384
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8766158508821295e-05,
+      "loss": 3.9513,
+      "step": 1296896
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.8757772561310775e-05,
+      "loss": 3.9379,
+      "step": 1297408
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.0304365158081055,
+      "eval_runtime": 295.743,
+      "eval_samples_per_second": 1290.279,
+      "eval_steps_per_second": 40.322,
+      "step": 1297440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8749386613800255e-05,
+      "loss": 3.9523,
+      "step": 1297920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8741000666289735e-05,
+      "loss": 3.946,
+      "step": 1298432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8732614718779215e-05,
+      "loss": 3.956,
+      "step": 1298944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8724228771268698e-05,
+      "loss": 3.9496,
+      "step": 1299456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8715842823758178e-05,
+      "loss": 3.9583,
+      "step": 1299968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8707456876247658e-05,
+      "loss": 3.9429,
+      "step": 1300480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8699070928737138e-05,
+      "loss": 3.939,
+      "step": 1300992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8690684981226618e-05,
+      "loss": 3.9307,
+      "step": 1301504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8682299033716098e-05,
+      "loss": 3.9626,
+      "step": 1302016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8673913086205578e-05,
+      "loss": 3.9501,
+      "step": 1302528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8665527138695058e-05,
+      "loss": 3.9452,
+      "step": 1303040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8657141191184538e-05,
+      "loss": 3.9555,
+      "step": 1303552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.864877162247775e-05,
+      "loss": 3.9464,
+      "step": 1304064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.864040205377096e-05,
+      "loss": 3.9435,
+      "step": 1304576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.863201610626044e-05,
+      "loss": 3.9415,
+      "step": 1305088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.862363015874992e-05,
+      "loss": 3.9423,
+      "step": 1305600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.86152442112394e-05,
+      "loss": 3.9421,
+      "step": 1306112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.860685826372888e-05,
+      "loss": 3.9414,
+      "step": 1306624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8598472316218362e-05,
+      "loss": 3.9526,
+      "step": 1307136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8590086368707836e-05,
+      "loss": 3.9646,
+      "step": 1307648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8581700421197316e-05,
+      "loss": 3.9478,
+      "step": 1308160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8573314473686795e-05,
+      "loss": 3.941,
+      "step": 1308672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8564928526176275e-05,
+      "loss": 3.9464,
+      "step": 1309184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8556542578665755e-05,
+      "loss": 3.9488,
+      "step": 1309696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.854815663115524e-05,
+      "loss": 3.9444,
+      "step": 1310208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.853977068364472e-05,
+      "loss": 3.9365,
+      "step": 1310720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.85313847361342e-05,
+      "loss": 3.9369,
+      "step": 1311232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8523015167427408e-05,
+      "loss": 3.9541,
+      "step": 1311744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.8514629219916888e-05,
+      "loss": 3.9296,
+      "step": 1312256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8506243272406368e-05,
+      "loss": 3.938,
+      "step": 1312768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8497857324895848e-05,
+      "loss": 3.947,
+      "step": 1313280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.848947137738533e-05,
+      "loss": 3.9514,
+      "step": 1313792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.848108542987481e-05,
+      "loss": 3.9533,
+      "step": 1314304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.847269948236429e-05,
+      "loss": 3.9462,
+      "step": 1314816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.84643299136575e-05,
+      "loss": 3.9462,
+      "step": 1315328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.845594396614698e-05,
+      "loss": 3.9529,
+      "step": 1315840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.844755801863646e-05,
+      "loss": 3.9339,
+      "step": 1316352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.843917207112594e-05,
+      "loss": 3.9445,
+      "step": 1316864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8430786123615423e-05,
+      "loss": 3.9343,
+      "step": 1317376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8422400176104896e-05,
+      "loss": 3.9392,
+      "step": 1317888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8414014228594376e-05,
+      "loss": 3.942,
+      "step": 1318400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8405644659887592e-05,
+      "loss": 3.9461,
+      "step": 1318912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8397258712377072e-05,
+      "loss": 3.945,
+      "step": 1319424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8388889143670285e-05,
+      "loss": 3.9386,
+      "step": 1319936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8380503196159765e-05,
+      "loss": 3.9453,
+      "step": 1320448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8372117248649245e-05,
+      "loss": 3.9442,
+      "step": 1320960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8363731301138724e-05,
+      "loss": 3.9476,
+      "step": 1321472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8355345353628198e-05,
+      "loss": 3.9366,
+      "step": 1321984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8346959406117678e-05,
+      "loss": 3.9316,
+      "step": 1322496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.833857345860716e-05,
+      "loss": 3.9481,
+      "step": 1323008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.833018751109664e-05,
+      "loss": 3.953,
+      "step": 1323520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.832180156358612e-05,
+      "loss": 3.9388,
+      "step": 1324032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.83134156160756e-05,
+      "loss": 3.9413,
+      "step": 1324544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.830502966856508e-05,
+      "loss": 3.9382,
+      "step": 1325056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.829664372105456e-05,
+      "loss": 3.9351,
+      "step": 1325568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.828827415234777e-05,
+      "loss": 3.9403,
+      "step": 1326080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8279888204837253e-05,
+      "loss": 3.9316,
+      "step": 1326592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8271502257326733e-05,
+      "loss": 3.9466,
+      "step": 1327104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8263116309816213e-05,
+      "loss": 3.9436,
+      "step": 1327616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8254746741109422e-05,
+      "loss": 3.9411,
+      "step": 1328128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8246360793598902e-05,
+      "loss": 3.9283,
+      "step": 1328640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8237991224892115e-05,
+      "loss": 3.9501,
+      "step": 1329152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8229605277381595e-05,
+      "loss": 3.9258,
+      "step": 1329664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8221219329871075e-05,
+      "loss": 3.9403,
+      "step": 1330176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8212833382360554e-05,
+      "loss": 3.9469,
+      "step": 1330688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8204463813653764e-05,
+      "loss": 3.9476,
+      "step": 1331200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8196094244946976e-05,
+      "loss": 3.9313,
+      "step": 1331712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8187708297436456e-05,
+      "loss": 3.9329,
+      "step": 1332224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8179322349925936e-05,
+      "loss": 3.9294,
+      "step": 1332736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8170936402415416e-05,
+      "loss": 3.9405,
+      "step": 1333248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8162550454904896e-05,
+      "loss": 3.9439,
+      "step": 1333760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8154164507394376e-05,
+      "loss": 3.9412,
+      "step": 1334272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8145778559883856e-05,
+      "loss": 3.9485,
+      "step": 1334784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8137392612373336e-05,
+      "loss": 3.9498,
+      "step": 1335296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8129023043666548e-05,
+      "loss": 3.9483,
+      "step": 1335808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8120637096156028e-05,
+      "loss": 3.935,
+      "step": 1336320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8112251148645508e-05,
+      "loss": 3.946,
+      "step": 1336832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8103865201134988e-05,
+      "loss": 3.9486,
+      "step": 1337344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8095479253624468e-05,
+      "loss": 3.931,
+      "step": 1337856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8087093306113948e-05,
+      "loss": 3.9487,
+      "step": 1338368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8078723737407157e-05,
+      "loss": 3.9355,
+      "step": 1338880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.807033778989664e-05,
+      "loss": 3.9581,
+      "step": 1339392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.806195184238612e-05,
+      "loss": 3.9457,
+      "step": 1339904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.80535658948756e-05,
+      "loss": 3.9232,
+      "step": 1340416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.804519632616881e-05,
+      "loss": 3.9421,
+      "step": 1340928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.803681037865829e-05,
+      "loss": 3.9347,
+      "step": 1341440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.802842443114777e-05,
+      "loss": 3.9483,
+      "step": 1341952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.802003848363725e-05,
+      "loss": 3.9429,
+      "step": 1342464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8011652536126733e-05,
+      "loss": 3.9401,
+      "step": 1342976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8003266588616206e-05,
+      "loss": 3.941,
+      "step": 1343488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7994880641105686e-05,
+      "loss": 3.934,
+      "step": 1344000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7986494693595166e-05,
+      "loss": 3.9307,
+      "step": 1344512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.797812512488838e-05,
+      "loss": 3.9279,
+      "step": 1345024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7969739177377858e-05,
+      "loss": 3.9379,
+      "step": 1345536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7961353229867338e-05,
+      "loss": 3.9479,
+      "step": 1346048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7952983661160554e-05,
+      "loss": 3.9403,
+      "step": 1346560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7944597713650027e-05,
+      "loss": 3.932,
+      "step": 1347072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7936211766139507e-05,
+      "loss": 3.9357,
+      "step": 1347584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7927825818628987e-05,
+      "loss": 3.9404,
+      "step": 1348096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.791943987111847e-05,
+      "loss": 3.9473,
+      "step": 1348608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.791105392360795e-05,
+      "loss": 3.9369,
+      "step": 1349120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.790266797609743e-05,
+      "loss": 3.9372,
+      "step": 1349632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.789428202858691e-05,
+      "loss": 3.9365,
+      "step": 1350144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.788589608107639e-05,
+      "loss": 3.9458,
+      "step": 1350656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.78775265123696e-05,
+      "loss": 3.9257,
+      "step": 1351168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.786914056485908e-05,
+      "loss": 3.9363,
+      "step": 1351680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7860754617348563e-05,
+      "loss": 3.9424,
+      "step": 1352192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7852368669838043e-05,
+      "loss": 3.9352,
+      "step": 1352704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7843999101131252e-05,
+      "loss": 3.9308,
+      "step": 1353216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.783561315362073e-05,
+      "loss": 3.9356,
+      "step": 1353728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.782722720611021e-05,
+      "loss": 3.9349,
+      "step": 1354240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.781884125859969e-05,
+      "loss": 3.9335,
+      "step": 1354752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7810471689892904e-05,
+      "loss": 3.9359,
+      "step": 1355264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7802085742382384e-05,
+      "loss": 3.9366,
+      "step": 1355776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7793699794871864e-05,
+      "loss": 3.9471,
+      "step": 1356288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7785313847361344e-05,
+      "loss": 3.947,
+      "step": 1356800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7776927899850824e-05,
+      "loss": 3.9387,
+      "step": 1357312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7768541952340304e-05,
+      "loss": 3.9432,
+      "step": 1357824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7760172383633516e-05,
+      "loss": 3.941,
+      "step": 1358336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7751786436122996e-05,
+      "loss": 3.9388,
+      "step": 1358848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7743400488612476e-05,
+      "loss": 3.932,
+      "step": 1359360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7735030919905685e-05,
+      "loss": 3.9326,
+      "step": 1359872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7726644972395165e-05,
+      "loss": 3.9386,
+      "step": 1360384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7718259024884645e-05,
+      "loss": 3.942,
+      "step": 1360896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7709873077374125e-05,
+      "loss": 3.9376,
+      "step": 1361408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.770148712986361e-05,
+      "loss": 3.9392,
+      "step": 1361920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.769310118235309e-05,
+      "loss": 3.9371,
+      "step": 1362432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.768471523484257e-05,
+      "loss": 3.9379,
+      "step": 1362944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.767632928733204e-05,
+      "loss": 3.9367,
+      "step": 1363456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.766794333982152e-05,
+      "loss": 3.9473,
+      "step": 1363968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7659557392311e-05,
+      "loss": 3.932,
+      "step": 1364480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7651171444800485e-05,
+      "loss": 3.9227,
+      "step": 1364992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7642801876093694e-05,
+      "loss": 3.9303,
+      "step": 1365504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7634415928583174e-05,
+      "loss": 3.9467,
+      "step": 1366016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7626029981072654e-05,
+      "loss": 3.9468,
+      "step": 1366528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7617644033562134e-05,
+      "loss": 3.9375,
+      "step": 1367040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7609258086051614e-05,
+      "loss": 3.9335,
+      "step": 1367552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7600872138541094e-05,
+      "loss": 3.9479,
+      "step": 1368064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7592486191030574e-05,
+      "loss": 3.9458,
+      "step": 1368576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7584100243520057e-05,
+      "loss": 3.9404,
+      "step": 1369088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7575730674813266e-05,
+      "loss": 3.9345,
+      "step": 1369600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7567344727302746e-05,
+      "loss": 3.9385,
+      "step": 1370112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7558958779792226e-05,
+      "loss": 3.9401,
+      "step": 1370624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7550572832281706e-05,
+      "loss": 3.9313,
+      "step": 1371136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.754220326357492e-05,
+      "loss": 3.9482,
+      "step": 1371648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.75338173160644e-05,
+      "loss": 3.9334,
+      "step": 1372160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.752543136855388e-05,
+      "loss": 3.9458,
+      "step": 1372672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.751704542104336e-05,
+      "loss": 3.9404,
+      "step": 1373184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7508675852336567e-05,
+      "loss": 3.9296,
+      "step": 1373696
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.026124000549316,
+      "eval_runtime": 325.5834,
+      "eval_samples_per_second": 1172.022,
+      "eval_steps_per_second": 36.627,
+      "step": 1373760
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7500289904826047e-05,
+      "loss": 3.9429,
+      "step": 1374208
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7491903957315527e-05,
+      "loss": 3.9371,
+      "step": 1374720
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.748351800980501e-05,
+      "loss": 3.9435,
+      "step": 1375232
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.747516481990195e-05,
+      "loss": 3.9402,
+      "step": 1375744
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.746677887239143e-05,
+      "loss": 3.9497,
+      "step": 1376256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.745839292488091e-05,
+      "loss": 3.9317,
+      "step": 1376768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.745000697737039e-05,
+      "loss": 3.932,
+      "step": 1377280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7441621029859872e-05,
+      "loss": 3.921,
+      "step": 1377792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.743326783995681e-05,
+      "loss": 3.9535,
+      "step": 1378304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.742488189244629e-05,
+      "loss": 3.9425,
+      "step": 1378816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.741649594493577e-05,
+      "loss": 3.9333,
+      "step": 1379328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.740810999742525e-05,
+      "loss": 3.9485,
+      "step": 1379840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7399740428718463e-05,
+      "loss": 3.9332,
+      "step": 1380352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7391354481207943e-05,
+      "loss": 3.9375,
+      "step": 1380864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7382968533697423e-05,
+      "loss": 3.9295,
+      "step": 1381376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7374582586186903e-05,
+      "loss": 3.9314,
+      "step": 1381888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7366196638676383e-05,
+      "loss": 3.9376,
+      "step": 1382400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7357810691165863e-05,
+      "loss": 3.9295,
+      "step": 1382912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7349424743655343e-05,
+      "loss": 3.9435,
+      "step": 1383424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7341038796144826e-05,
+      "loss": 3.9512,
+      "step": 1383936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7332652848634306e-05,
+      "loss": 3.9379,
+      "step": 1384448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7324266901123786e-05,
+      "loss": 3.9342,
+      "step": 1384960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7315880953613266e-05,
+      "loss": 3.9381,
+      "step": 1385472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7307495006102746e-05,
+      "loss": 3.9389,
+      "step": 1385984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7299109058592226e-05,
+      "loss": 3.9312,
+      "step": 1386496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7290723111081702e-05,
+      "loss": 3.9308,
+      "step": 1387008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7282337163571182e-05,
+      "loss": 3.9238,
+      "step": 1387520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7273967594864398e-05,
+      "loss": 3.9457,
+      "step": 1388032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7265581647353878e-05,
+      "loss": 3.9176,
+      "step": 1388544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.725719569984335e-05,
+      "loss": 3.931,
+      "step": 1389056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.724880975233283e-05,
+      "loss": 3.9371,
+      "step": 1389568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.724042380482231e-05,
+      "loss": 3.9411,
+      "step": 1390080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7232054236115524e-05,
+      "loss": 3.9401,
+      "step": 1390592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7223668288605004e-05,
+      "loss": 3.9395,
+      "step": 1391104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.721529871989822e-05,
+      "loss": 3.936,
+      "step": 1391616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.72069127723877e-05,
+      "loss": 3.945,
+      "step": 1392128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7198526824877173e-05,
+      "loss": 3.9288,
+      "step": 1392640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7190140877366656e-05,
+      "loss": 3.9327,
+      "step": 1393152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7181754929856136e-05,
+      "loss": 3.9233,
+      "step": 1393664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7173368982345616e-05,
+      "loss": 3.9308,
+      "step": 1394176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7164983034835096e-05,
+      "loss": 3.9285,
+      "step": 1394688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7156597087324576e-05,
+      "loss": 3.941,
+      "step": 1395200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7148211139814056e-05,
+      "loss": 3.9338,
+      "step": 1395712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7139825192303536e-05,
+      "loss": 3.9298,
+      "step": 1396224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7131439244793016e-05,
+      "loss": 3.9367,
+      "step": 1396736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7123053297282496e-05,
+      "loss": 3.9354,
+      "step": 1397248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7114683728575708e-05,
+      "loss": 3.9366,
+      "step": 1397760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7106297781065188e-05,
+      "loss": 3.9258,
+      "step": 1398272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7097911833554668e-05,
+      "loss": 3.9194,
+      "step": 1398784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7089525886044148e-05,
+      "loss": 3.9433,
+      "step": 1399296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7081139938533628e-05,
+      "loss": 3.9402,
+      "step": 1399808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.707277036982684e-05,
+      "loss": 3.9314,
+      "step": 1400320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.706438442231632e-05,
+      "loss": 3.9322,
+      "step": 1400832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.70559984748058e-05,
+      "loss": 3.9247,
+      "step": 1401344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.704761252729528e-05,
+      "loss": 3.9288,
+      "step": 1401856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.703922657978476e-05,
+      "loss": 3.9318,
+      "step": 1402368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.703085701107797e-05,
+      "loss": 3.9226,
+      "step": 1402880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.702247106356745e-05,
+      "loss": 3.9381,
+      "step": 1403392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.701408511605693e-05,
+      "loss": 3.9332,
+      "step": 1403904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.7005699168546412e-05,
+      "loss": 3.9299,
+      "step": 1404416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6997313221035886e-05,
+      "loss": 3.9203,
+      "step": 1404928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.69889436523291e-05,
+      "loss": 3.9403,
+      "step": 1405440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.698055770481858e-05,
+      "loss": 3.9191,
+      "step": 1405952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.697217175730806e-05,
+      "loss": 3.9311,
+      "step": 1406464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6963785809797538e-05,
+      "loss": 3.9332,
+      "step": 1406976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6955416241090754e-05,
+      "loss": 3.9399,
+      "step": 1407488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6947030293580234e-05,
+      "loss": 3.9227,
+      "step": 1408000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6938644346069714e-05,
+      "loss": 3.928,
+      "step": 1408512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6930258398559187e-05,
+      "loss": 3.915,
+      "step": 1409024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6921888829852403e-05,
+      "loss": 3.9318,
+      "step": 1409536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6913502882341883e-05,
+      "loss": 3.9354,
+      "step": 1410048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6905133313635095e-05,
+      "loss": 3.9324,
+      "step": 1410560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6896747366124575e-05,
+      "loss": 3.9361,
+      "step": 1411072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6888361418614055e-05,
+      "loss": 3.9394,
+      "step": 1411584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6879975471103535e-05,
+      "loss": 3.9438,
+      "step": 1412096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6871622281200477e-05,
+      "loss": 3.9211,
+      "step": 1412608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6863236333689957e-05,
+      "loss": 3.9344,
+      "step": 1413120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6854850386179437e-05,
+      "loss": 3.9422,
+      "step": 1413632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6846464438668917e-05,
+      "loss": 3.9216,
+      "step": 1414144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6838094869962126e-05,
+      "loss": 3.9365,
+      "step": 1414656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6829708922451606e-05,
+      "loss": 3.9247,
+      "step": 1415168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.682132297494109e-05,
+      "loss": 3.9492,
+      "step": 1415680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.681293702743057e-05,
+      "loss": 3.9392,
+      "step": 1416192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.680455107992005e-05,
+      "loss": 3.9174,
+      "step": 1416704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.679616513240953e-05,
+      "loss": 3.9303,
+      "step": 1417216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.678777918489901e-05,
+      "loss": 3.9289,
+      "step": 1417728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6779393237388482e-05,
+      "loss": 3.9383,
+      "step": 1418240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.6771007289877965e-05,
+      "loss": 3.9348,
+      "step": 1418752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.676263772117118e-05,
+      "loss": 3.9303,
+      "step": 1419264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6754251773660654e-05,
+      "loss": 3.9304,
+      "step": 1419776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6745865826150134e-05,
+      "loss": 3.926,
+      "step": 1420288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6737479878639614e-05,
+      "loss": 3.9236,
+      "step": 1420800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.672911030993283e-05,
+      "loss": 3.916,
+      "step": 1421312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6720724362422307e-05,
+      "loss": 3.9308,
+      "step": 1421824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6712338414911787e-05,
+      "loss": 3.9383,
+      "step": 1422336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6703952467401267e-05,
+      "loss": 3.9315,
+      "step": 1422848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6695566519890747e-05,
+      "loss": 3.9225,
+      "step": 1423360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6687180572380227e-05,
+      "loss": 3.9262,
+      "step": 1423872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6678794624869707e-05,
+      "loss": 3.9276,
+      "step": 1424384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.667042505616292e-05,
+      "loss": 3.9404,
+      "step": 1424896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.66620391086524e-05,
+      "loss": 3.9289,
+      "step": 1425408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.665365316114188e-05,
+      "loss": 3.9303,
+      "step": 1425920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.664526721363136e-05,
+      "loss": 3.9255,
+      "step": 1426432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.663688126612084e-05,
+      "loss": 3.9383,
+      "step": 1426944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.662849531861032e-05,
+      "loss": 3.9149,
+      "step": 1427456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6620125749903528e-05,
+      "loss": 3.9301,
+      "step": 1427968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.661173980239301e-05,
+      "loss": 3.9286,
+      "step": 1428480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.660335385488249e-05,
+      "loss": 3.9295,
+      "step": 1428992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.659496790737197e-05,
+      "loss": 3.9203,
+      "step": 1429504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.658658195986145e-05,
+      "loss": 3.9277,
+      "step": 1430016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.657819601235093e-05,
+      "loss": 3.9269,
+      "step": 1430528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.656981006484041e-05,
+      "loss": 3.925,
+      "step": 1431040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.656142411732989e-05,
+      "loss": 3.9287,
+      "step": 1431552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6553054548623103e-05,
+      "loss": 3.9287,
+      "step": 1432064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6544668601112583e-05,
+      "loss": 3.9334,
+      "step": 1432576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6536282653602063e-05,
+      "loss": 3.9381,
+      "step": 1433088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6527913084895273e-05,
+      "loss": 3.9324,
+      "step": 1433600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6519527137384752e-05,
+      "loss": 3.9346,
+      "step": 1434112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6511141189874232e-05,
+      "loss": 3.9289,
+      "step": 1434624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6502755242363712e-05,
+      "loss": 3.9245,
+      "step": 1435136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6494369294853196e-05,
+      "loss": 3.9272,
+      "step": 1435648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6485999726146405e-05,
+      "loss": 3.927,
+      "step": 1436160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6477613778635885e-05,
+      "loss": 3.9235,
+      "step": 1436672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6469227831125365e-05,
+      "loss": 3.9334,
+      "step": 1437184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6460841883614845e-05,
+      "loss": 3.9267,
+      "step": 1437696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.645245593610432e-05,
+      "loss": 3.9315,
+      "step": 1438208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.64440699885938e-05,
+      "loss": 3.9271,
+      "step": 1438720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.643568404108328e-05,
+      "loss": 3.9273,
+      "step": 1439232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.642729809357276e-05,
+      "loss": 3.9239,
+      "step": 1439744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.641892852486597e-05,
+      "loss": 3.9402,
+      "step": 1440256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.641054257735545e-05,
+      "loss": 3.9224,
+      "step": 1440768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6402156629844934e-05,
+      "loss": 3.9202,
+      "step": 1441280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6393787061138143e-05,
+      "loss": 3.9167,
+      "step": 1441792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6385401113627623e-05,
+      "loss": 3.9337,
+      "step": 1442304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6377015166117103e-05,
+      "loss": 3.9375,
+      "step": 1442816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6368629218606582e-05,
+      "loss": 3.9344,
+      "step": 1443328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6360243271096062e-05,
+      "loss": 3.9264,
+      "step": 1443840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6351857323585542e-05,
+      "loss": 3.9385,
+      "step": 1444352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6343471376075022e-05,
+      "loss": 3.9308,
+      "step": 1444864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6335085428564506e-05,
+      "loss": 3.9332,
+      "step": 1445376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6326715859857715e-05,
+      "loss": 3.9235,
+      "step": 1445888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6318329912347195e-05,
+      "loss": 3.9319,
+      "step": 1446400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6309943964836675e-05,
+      "loss": 3.9303,
+      "step": 1446912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6301558017326155e-05,
+      "loss": 3.9204,
+      "step": 1447424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6293172069815635e-05,
+      "loss": 3.9383,
+      "step": 1447936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6284802501108847e-05,
+      "loss": 3.9298,
+      "step": 1448448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6276432932402056e-05,
+      "loss": 3.9308,
+      "step": 1448960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6268046984891536e-05,
+      "loss": 3.9304,
+      "step": 1449472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.6259661037381016e-05,
+      "loss": 3.9193,
+      "step": 1449984
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.021474838256836,
+      "eval_runtime": 348.3451,
+      "eval_samples_per_second": 1095.439,
+      "eval_steps_per_second": 34.233,
+      "step": 1450080
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6251275089870496e-05,
+      "loss": 3.9348,
+      "step": 1450496
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6242889142359976e-05,
+      "loss": 3.9298,
+      "step": 1451008
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.623450319484946e-05,
+      "loss": 3.932,
+      "step": 1451520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.622611724733894e-05,
+      "loss": 3.928,
+      "step": 1452032
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.621773129982842e-05,
+      "loss": 3.9444,
+      "step": 1452544
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.62093453523179e-05,
+      "loss": 3.9256,
+      "step": 1453056
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.620095940480738e-05,
+      "loss": 3.9204,
+      "step": 1453568
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6192573457296852e-05,
+      "loss": 3.9157,
+      "step": 1454080
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6184187509786336e-05,
+      "loss": 3.94,
+      "step": 1454592
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6175801562275816e-05,
+      "loss": 3.9378,
+      "step": 1455104
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6167415614765296e-05,
+      "loss": 3.9198,
+      "step": 1455616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6159029667254776e-05,
+      "loss": 3.94,
+      "step": 1456128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6150660098547985e-05,
+      "loss": 3.925,
+      "step": 1456640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6142274151037465e-05,
+      "loss": 3.9279,
+      "step": 1457152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6133888203526945e-05,
+      "loss": 3.9229,
+      "step": 1457664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6125502256016428e-05,
+      "loss": 3.9213,
+      "step": 1458176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6117116308505908e-05,
+      "loss": 3.9271,
+      "step": 1458688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6108730360995388e-05,
+      "loss": 3.9204,
+      "step": 1459200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6100344413484868e-05,
+      "loss": 3.9347,
+      "step": 1459712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6091974844778077e-05,
+      "loss": 3.9397,
+      "step": 1460224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6083588897267557e-05,
+      "loss": 3.9337,
+      "step": 1460736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6075202949757037e-05,
+      "loss": 3.9253,
+      "step": 1461248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.606681700224652e-05,
+      "loss": 3.924,
+      "step": 1461760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6058431054736e-05,
+      "loss": 3.9339,
+      "step": 1462272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.605004510722548e-05,
+      "loss": 3.9204,
+      "step": 1462784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.604165915971496e-05,
+      "loss": 3.9235,
+      "step": 1463296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.603327321220444e-05,
+      "loss": 3.914,
+      "step": 1463808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.602488726469392e-05,
+      "loss": 3.9352,
+      "step": 1464320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.6016501317183396e-05,
+      "loss": 3.9119,
+      "step": 1464832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6008115369672876e-05,
+      "loss": 3.9199,
+      "step": 1465344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5999729422162356e-05,
+      "loss": 3.9306,
+      "step": 1465856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5991359853455565e-05,
+      "loss": 3.9298,
+      "step": 1466368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5982973905945045e-05,
+      "loss": 3.931,
+      "step": 1466880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5974587958434525e-05,
+      "loss": 3.9325,
+      "step": 1467392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5966234768531474e-05,
+      "loss": 3.9283,
+      "step": 1467904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5957848821020954e-05,
+      "loss": 3.9332,
+      "step": 1468416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5949462873510434e-05,
+      "loss": 3.9229,
+      "step": 1468928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5941076925999914e-05,
+      "loss": 3.918,
+      "step": 1469440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5932690978489394e-05,
+      "loss": 3.9162,
+      "step": 1469952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5924305030978867e-05,
+      "loss": 3.9201,
+      "step": 1470464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.591591908346835e-05,
+      "loss": 3.9178,
+      "step": 1470976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.590753313595783e-05,
+      "loss": 3.9337,
+      "step": 1471488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.589914718844731e-05,
+      "loss": 3.9227,
+      "step": 1472000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.589076124093679e-05,
+      "loss": 3.9223,
+      "step": 1472512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.588239167223e-05,
+      "loss": 3.9307,
+      "step": 1473024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.587400572471948e-05,
+      "loss": 3.9206,
+      "step": 1473536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.586561977720896e-05,
+      "loss": 3.9325,
+      "step": 1474048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.585723382969844e-05,
+      "loss": 3.9188,
+      "step": 1474560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5848847882187922e-05,
+      "loss": 3.9039,
+      "step": 1475072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.584047831348113e-05,
+      "loss": 3.9379,
+      "step": 1475584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.583209236597061e-05,
+      "loss": 3.9293,
+      "step": 1476096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.582370641846009e-05,
+      "loss": 3.9282,
+      "step": 1476608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.581532047094957e-05,
+      "loss": 3.9204,
+      "step": 1477120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.580693452343905e-05,
+      "loss": 3.9175,
+      "step": 1477632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.579854857592853e-05,
+      "loss": 3.9187,
+      "step": 1478144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5790179007221744e-05,
+      "loss": 3.922,
+      "step": 1478656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5781793059711224e-05,
+      "loss": 3.9156,
+      "step": 1479168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5773423491004433e-05,
+      "loss": 3.9213,
+      "step": 1479680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5765037543493913e-05,
+      "loss": 3.9304,
+      "step": 1480192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5756651595983393e-05,
+      "loss": 3.9172,
+      "step": 1480704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5748265648472876e-05,
+      "loss": 3.9101,
+      "step": 1481216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5739879700962356e-05,
+      "loss": 3.9341,
+      "step": 1481728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5731493753451836e-05,
+      "loss": 3.9098,
+      "step": 1482240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5723107805941316e-05,
+      "loss": 3.9157,
+      "step": 1482752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5714721858430796e-05,
+      "loss": 3.9288,
+      "step": 1483264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5706335910920276e-05,
+      "loss": 3.9322,
+      "step": 1483776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5697949963409756e-05,
+      "loss": 3.9133,
+      "step": 1484288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5689564015899232e-05,
+      "loss": 3.9184,
+      "step": 1484800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5681178068388712e-05,
+      "loss": 3.9026,
+      "step": 1485312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5672808499681928e-05,
+      "loss": 3.9288,
+      "step": 1485824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.56644225521714e-05,
+      "loss": 3.9253,
+      "step": 1486336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.565603660466088e-05,
+      "loss": 3.9256,
+      "step": 1486848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.564765065715036e-05,
+      "loss": 3.9247,
+      "step": 1487360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5639264709639845e-05,
+      "loss": 3.933,
+      "step": 1487872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5630895140933054e-05,
+      "loss": 3.9375,
+      "step": 1488384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.562252557222627e-05,
+      "loss": 3.9139,
+      "step": 1488896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.561413962471575e-05,
+      "loss": 3.9271,
+      "step": 1489408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5605753677205223e-05,
+      "loss": 3.9347,
+      "step": 1489920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5597367729694706e-05,
+      "loss": 3.9129,
+      "step": 1490432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5588998160987922e-05,
+      "loss": 3.9274,
+      "step": 1490944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5580612213477402e-05,
+      "loss": 3.9207,
+      "step": 1491456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5572226265966875e-05,
+      "loss": 3.9364,
+      "step": 1491968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5563840318456355e-05,
+      "loss": 3.9317,
+      "step": 1492480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5555454370945835e-05,
+      "loss": 3.9091,
+      "step": 1492992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5547068423435315e-05,
+      "loss": 3.9226,
+      "step": 1493504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5538682475924798e-05,
+      "loss": 3.9191,
+      "step": 1494016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5530296528414278e-05,
+      "loss": 3.9317,
+      "step": 1494528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5521926959707487e-05,
+      "loss": 3.9218,
+      "step": 1495040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.5513541012196967e-05,
+      "loss": 3.9218,
+      "step": 1495552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5505155064686447e-05,
+      "loss": 3.9203,
+      "step": 1496064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5496769117175927e-05,
+      "loss": 3.9202,
+      "step": 1496576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.548839954846914e-05,
+      "loss": 3.9149,
+      "step": 1497088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.548001360095862e-05,
+      "loss": 3.9071,
+      "step": 1497600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.54716276534481e-05,
+      "loss": 3.9203,
+      "step": 1498112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.546325808474131e-05,
+      "loss": 3.9299,
+      "step": 1498624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.545487213723079e-05,
+      "loss": 3.9283,
+      "step": 1499136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.544648618972027e-05,
+      "loss": 3.9082,
+      "step": 1499648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5438100242209752e-05,
+      "loss": 3.9201,
+      "step": 1500160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5429714294699232e-05,
+      "loss": 3.9189,
+      "step": 1500672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.542134472599244e-05,
+      "loss": 3.9294,
+      "step": 1501184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.541295877848192e-05,
+      "loss": 3.9176,
+      "step": 1501696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.54045728309714e-05,
+      "loss": 3.9224,
+      "step": 1502208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.539618688346088e-05,
+      "loss": 3.9162,
+      "step": 1502720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.538780093595036e-05,
+      "loss": 3.9304,
+      "step": 1503232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5379414988439844e-05,
+      "loss": 3.9065,
+      "step": 1503744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5371029040929324e-05,
+      "loss": 3.9228,
+      "step": 1504256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5362643093418804e-05,
+      "loss": 3.9163,
+      "step": 1504768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5354273524712013e-05,
+      "loss": 3.9204,
+      "step": 1505280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5345887577201493e-05,
+      "loss": 3.9124,
+      "step": 1505792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5337501629690973e-05,
+      "loss": 3.9248,
+      "step": 1506304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5329115682180453e-05,
+      "loss": 3.9139,
+      "step": 1506816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5320729734669933e-05,
+      "loss": 3.9171,
+      "step": 1507328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5312343787159416e-05,
+      "loss": 3.9191,
+      "step": 1507840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5303974218452625e-05,
+      "loss": 3.9208,
+      "step": 1508352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5295588270942105e-05,
+      "loss": 3.9266,
+      "step": 1508864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5287202323431585e-05,
+      "loss": 3.9309,
+      "step": 1509376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5278816375921062e-05,
+      "loss": 3.9237,
+      "step": 1509888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5270430428410542e-05,
+      "loss": 3.9266,
+      "step": 1510400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5262044480900022e-05,
+      "loss": 3.9219,
+      "step": 1510912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5253658533389502e-05,
+      "loss": 3.9159,
+      "step": 1511424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.524527258587898e-05,
+      "loss": 3.9173,
+      "step": 1511936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.523690301717219e-05,
+      "loss": 3.9179,
+      "step": 1512448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.522851706966167e-05,
+      "loss": 3.9192,
+      "step": 1512960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5220131122151154e-05,
+      "loss": 3.9247,
+      "step": 1513472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5211745174640634e-05,
+      "loss": 3.9183,
+      "step": 1513984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5203359227130114e-05,
+      "loss": 3.9245,
+      "step": 1514496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5194973279619594e-05,
+      "loss": 3.9173,
+      "step": 1515008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5186603710912803e-05,
+      "loss": 3.9167,
+      "step": 1515520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5178217763402283e-05,
+      "loss": 3.9178,
+      "step": 1516032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5169831815891763e-05,
+      "loss": 3.9303,
+      "step": 1516544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5161445868381246e-05,
+      "loss": 3.9162,
+      "step": 1517056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5153059920870726e-05,
+      "loss": 3.9105,
+      "step": 1517568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5144673973360206e-05,
+      "loss": 3.9079,
+      "step": 1518080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5136288025849686e-05,
+      "loss": 3.9239,
+      "step": 1518592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5127918457142895e-05,
+      "loss": 3.9278,
+      "step": 1519104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5119532509632375e-05,
+      "loss": 3.9268,
+      "step": 1519616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5111146562121855e-05,
+      "loss": 3.92,
+      "step": 1520128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.510276061461134e-05,
+      "loss": 3.931,
+      "step": 1520640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5094391045904548e-05,
+      "loss": 3.9207,
+      "step": 1521152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5086005098394028e-05,
+      "loss": 3.9228,
+      "step": 1521664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5077619150883508e-05,
+      "loss": 3.9175,
+      "step": 1522176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5069233203372987e-05,
+      "loss": 3.9228,
+      "step": 1522688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5060847255862467e-05,
+      "loss": 3.9248,
+      "step": 1523200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5052461308351947e-05,
+      "loss": 3.9117,
+      "step": 1523712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5044075360841424e-05,
+      "loss": 3.9305,
+      "step": 1524224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.503570579213464e-05,
+      "loss": 3.9243,
+      "step": 1524736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.502731984462412e-05,
+      "loss": 3.9187,
+      "step": 1525248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.50189338971136e-05,
+      "loss": 3.9201,
+      "step": 1525760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5010547949603076e-05,
+      "loss": 3.9159,
+      "step": 1526272
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.017867565155029,
+      "eval_runtime": 300.1887,
+      "eval_samples_per_second": 1271.17,
+      "eval_steps_per_second": 39.725,
+      "step": 1526400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.5002162002092556e-05,
+      "loss": 3.9215,
+      "step": 1526784
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.499379243338577e-05,
+      "loss": 3.9236,
+      "step": 1527296
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.498540648587525e-05,
+      "loss": 3.9255,
+      "step": 1527808
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.497702053836473e-05,
+      "loss": 3.9186,
+      "step": 1528320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.496863459085421e-05,
+      "loss": 3.9329,
+      "step": 1528832
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.496024864334369e-05,
+      "loss": 3.9197,
+      "step": 1529344
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.495186269583317e-05,
+      "loss": 3.9114,
+      "step": 1529856
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.494347674832265e-05,
+      "loss": 3.9109,
+      "step": 1530368
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.493509080081213e-05,
+      "loss": 3.9266,
+      "step": 1530880
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4926721232105338e-05,
+      "loss": 3.9278,
+      "step": 1531392
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4918335284594817e-05,
+      "loss": 3.9155,
+      "step": 1531904
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4909949337084297e-05,
+      "loss": 3.9274,
+      "step": 1532416
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.490157976837751e-05,
+      "loss": 3.92,
+      "step": 1532928
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.489319382086699e-05,
+      "loss": 3.9209,
+      "step": 1533440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.488480787335647e-05,
+      "loss": 3.9144,
+      "step": 1533952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.487642192584595e-05,
+      "loss": 3.9129,
+      "step": 1534464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4868052357139162e-05,
+      "loss": 3.9155,
+      "step": 1534976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.485966640962864e-05,
+      "loss": 3.9135,
+      "step": 1535488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4851280462118122e-05,
+      "loss": 3.9285,
+      "step": 1536000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4842894514607602e-05,
+      "loss": 3.9318,
+      "step": 1536512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4834508567097082e-05,
+      "loss": 3.9277,
+      "step": 1537024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4826122619586562e-05,
+      "loss": 3.9185,
+      "step": 1537536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4817736672076042e-05,
+      "loss": 3.9165,
+      "step": 1538048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.480935072456552e-05,
+      "loss": 3.9234,
+      "step": 1538560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4800964777055e-05,
+      "loss": 3.914,
+      "step": 1539072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.479259520834821e-05,
+      "loss": 3.9154,
+      "step": 1539584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4784209260837694e-05,
+      "loss": 3.9074,
+      "step": 1540096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.477582331332717e-05,
+      "loss": 3.921,
+      "step": 1540608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.476743736581665e-05,
+      "loss": 3.9111,
+      "step": 1541120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.475905141830613e-05,
+      "loss": 3.9079,
+      "step": 1541632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4750681849599343e-05,
+      "loss": 3.924,
+      "step": 1542144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4742295902088823e-05,
+      "loss": 3.918,
+      "step": 1542656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4733909954578303e-05,
+      "loss": 3.9277,
+      "step": 1543168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4725524007067783e-05,
+      "loss": 3.9241,
+      "step": 1543680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4717154438360992e-05,
+      "loss": 3.9181,
+      "step": 1544192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4708768490850472e-05,
+      "loss": 3.9259,
+      "step": 1544704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4700382543339952e-05,
+      "loss": 3.9178,
+      "step": 1545216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4691996595829432e-05,
+      "loss": 3.91,
+      "step": 1545728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4683610648318915e-05,
+      "loss": 3.9046,
+      "step": 1546240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4675257458415857e-05,
+      "loss": 3.9111,
+      "step": 1546752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4666871510905337e-05,
+      "loss": 3.9126,
+      "step": 1547264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4658485563394817e-05,
+      "loss": 3.925,
+      "step": 1547776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4650099615884294e-05,
+      "loss": 3.9159,
+      "step": 1548288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4641713668373777e-05,
+      "loss": 3.9136,
+      "step": 1548800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4633327720863257e-05,
+      "loss": 3.9244,
+      "step": 1549312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4624941773352737e-05,
+      "loss": 3.9167,
+      "step": 1549824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4616555825842217e-05,
+      "loss": 3.9213,
+      "step": 1550336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4608169878331697e-05,
+      "loss": 3.908,
+      "step": 1550848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4599783930821173e-05,
+      "loss": 3.8942,
+      "step": 1551360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4591397983310653e-05,
+      "loss": 3.9309,
+      "step": 1551872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.458302841460387e-05,
+      "loss": 3.9243,
+      "step": 1552384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.457464246709335e-05,
+      "loss": 3.9191,
+      "step": 1552896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4566256519582826e-05,
+      "loss": 3.9135,
+      "step": 1553408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4557870572072306e-05,
+      "loss": 3.9105,
+      "step": 1553920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4549501003365518e-05,
+      "loss": 3.9092,
+      "step": 1554432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.454113143465873e-05,
+      "loss": 3.9124,
+      "step": 1554944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.453274548714821e-05,
+      "loss": 3.9091,
+      "step": 1555456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.452435953963769e-05,
+      "loss": 3.915,
+      "step": 1555968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.451597359212717e-05,
+      "loss": 3.9221,
+      "step": 1556480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.450760402342038e-05,
+      "loss": 3.9144,
+      "step": 1556992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.449921807590986e-05,
+      "loss": 3.9043,
+      "step": 1557504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.449083212839934e-05,
+      "loss": 3.9265,
+      "step": 1558016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4482446180888823e-05,
+      "loss": 3.9011,
+      "step": 1558528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.44740602333783e-05,
+      "loss": 3.9043,
+      "step": 1559040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.446567428586778e-05,
+      "loss": 3.9252,
+      "step": 1559552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.445728833835726e-05,
+      "loss": 3.9244,
+      "step": 1560064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.444890239084674e-05,
+      "loss": 3.9019,
+      "step": 1560576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.444051644333622e-05,
+      "loss": 3.9118,
+      "step": 1561088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.44321304958257e-05,
+      "loss": 3.8934,
+      "step": 1561600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.442374454831518e-05,
+      "loss": 3.9201,
+      "step": 1562112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.441535860080466e-05,
+      "loss": 3.9211,
+      "step": 1562624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.440698903209787e-05,
+      "loss": 3.9126,
+      "step": 1563136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.439860308458735e-05,
+      "loss": 3.9187,
+      "step": 1563648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4390217137076828e-05,
+      "loss": 3.9222,
+      "step": 1564160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4381831189566308e-05,
+      "loss": 3.9331,
+      "step": 1564672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4373461620859524e-05,
+      "loss": 3.9058,
+      "step": 1565184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4365092052152733e-05,
+      "loss": 3.9194,
+      "step": 1565696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4356706104642213e-05,
+      "loss": 3.9261,
+      "step": 1566208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4348320157131693e-05,
+      "loss": 3.9032,
+      "step": 1566720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4339934209621173e-05,
+      "loss": 3.9222,
+      "step": 1567232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.433154826211065e-05,
+      "loss": 3.9124,
+      "step": 1567744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4323178693403865e-05,
+      "loss": 3.9258,
+      "step": 1568256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4314792745893345e-05,
+      "loss": 3.9242,
+      "step": 1568768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4306406798382825e-05,
+      "loss": 3.9049,
+      "step": 1569280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4298020850872302e-05,
+      "loss": 3.9094,
+      "step": 1569792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4289634903361782e-05,
+      "loss": 3.9096,
+      "step": 1570304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4281248955851262e-05,
+      "loss": 3.9272,
+      "step": 1570816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.427286300834074e-05,
+      "loss": 3.9134,
+      "step": 1571328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.4264493439633954e-05,
+      "loss": 3.9133,
+      "step": 1571840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4256107492123434e-05,
+      "loss": 3.9121,
+      "step": 1572352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4247721544612914e-05,
+      "loss": 3.9155,
+      "step": 1572864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4239335597102394e-05,
+      "loss": 3.91,
+      "step": 1573376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4230949649591874e-05,
+      "loss": 3.8977,
+      "step": 1573888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4222563702081354e-05,
+      "loss": 3.9148,
+      "step": 1574400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4214194133374566e-05,
+      "loss": 3.92,
+      "step": 1574912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4205808185864046e-05,
+      "loss": 3.9215,
+      "step": 1575424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4197422238353526e-05,
+      "loss": 3.8996,
+      "step": 1575936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4189036290843006e-05,
+      "loss": 3.9123,
+      "step": 1576448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4180650343332483e-05,
+      "loss": 3.9121,
+      "step": 1576960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4172264395821963e-05,
+      "loss": 3.9202,
+      "step": 1577472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4163878448311443e-05,
+      "loss": 3.9163,
+      "step": 1577984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4155492500800926e-05,
+      "loss": 3.9155,
+      "step": 1578496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4147106553290406e-05,
+      "loss": 3.909,
+      "step": 1579008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4138736984583615e-05,
+      "loss": 3.9216,
+      "step": 1579520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4130351037073095e-05,
+      "loss": 3.8984,
+      "step": 1580032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4121965089562575e-05,
+      "loss": 3.9148,
+      "step": 1580544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4113595520855788e-05,
+      "loss": 3.9057,
+      "step": 1581056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4105209573345268e-05,
+      "loss": 3.9174,
+      "step": 1581568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4096823625834747e-05,
+      "loss": 3.8986,
+      "step": 1582080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4088437678324227e-05,
+      "loss": 3.9168,
+      "step": 1582592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4080068109617437e-05,
+      "loss": 3.909,
+      "step": 1583104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4071682162106916e-05,
+      "loss": 3.9097,
+      "step": 1583616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4063296214596396e-05,
+      "loss": 3.9116,
+      "step": 1584128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.405491026708588e-05,
+      "loss": 3.9137,
+      "step": 1584640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.404654069837909e-05,
+      "loss": 3.9176,
+      "step": 1585152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.403815475086857e-05,
+      "loss": 3.9233,
+      "step": 1585664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.402976880335805e-05,
+      "loss": 3.9163,
+      "step": 1586176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.402138285584753e-05,
+      "loss": 3.921,
+      "step": 1586688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.401299690833701e-05,
+      "loss": 3.9148,
+      "step": 1587200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.400461096082649e-05,
+      "loss": 3.9095,
+      "step": 1587712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.399622501331597e-05,
+      "loss": 3.9062,
+      "step": 1588224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.398785544460918e-05,
+      "loss": 3.9169,
+      "step": 1588736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.397946949709866e-05,
+      "loss": 3.9102,
+      "step": 1589248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3971083549588138e-05,
+      "loss": 3.916,
+      "step": 1589760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3962697602077618e-05,
+      "loss": 3.91,
+      "step": 1590272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.39543116545671e-05,
+      "loss": 3.9162,
+      "step": 1590784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.394592570705658e-05,
+      "loss": 3.9115,
+      "step": 1591296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.393755613834979e-05,
+      "loss": 3.9119,
+      "step": 1591808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.392917019083927e-05,
+      "loss": 3.9086,
+      "step": 1592320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.392078424332875e-05,
+      "loss": 3.9229,
+      "step": 1592832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.391239829581823e-05,
+      "loss": 3.911,
+      "step": 1593344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.390401234830771e-05,
+      "loss": 3.9065,
+      "step": 1593856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3895642779600922e-05,
+      "loss": 3.8976,
+      "step": 1594368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3887256832090402e-05,
+      "loss": 3.9155,
+      "step": 1594880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3878870884579882e-05,
+      "loss": 3.9207,
+      "step": 1595392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3870484937069362e-05,
+      "loss": 3.9174,
+      "step": 1595904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3862098989558842e-05,
+      "loss": 3.9151,
+      "step": 1596416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.385371304204832e-05,
+      "loss": 3.9222,
+      "step": 1596928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3845327094537802e-05,
+      "loss": 3.9133,
+      "step": 1597440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3836941147027282e-05,
+      "loss": 3.9198,
+      "step": 1597952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3828555199516762e-05,
+      "loss": 3.9072,
+      "step": 1598464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.382018563080997e-05,
+      "loss": 3.9163,
+      "step": 1598976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3811816062103183e-05,
+      "loss": 3.9116,
+      "step": 1599488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3803430114592663e-05,
+      "loss": 3.9085,
+      "step": 1600000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3795044167082143e-05,
+      "loss": 3.9219,
+      "step": 1600512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3786658219571623e-05,
+      "loss": 3.9183,
+      "step": 1601024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3778272272061103e-05,
+      "loss": 3.9138,
+      "step": 1601536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3769886324550583e-05,
+      "loss": 3.9133,
+      "step": 1602048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3761533134647525e-05,
+      "loss": 3.9123,
+      "step": 1602560
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.014298439025879,
+      "eval_runtime": 300.8331,
+      "eval_samples_per_second": 1268.447,
+      "eval_steps_per_second": 39.64,
+      "step": 1602720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3753147187137005e-05,
+      "loss": 3.9158,
+      "step": 1603072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3744761239626488e-05,
+      "loss": 3.916,
+      "step": 1603584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3736375292115965e-05,
+      "loss": 3.9183,
+      "step": 1604096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3727989344605445e-05,
+      "loss": 3.911,
+      "step": 1604608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3719603397094925e-05,
+      "loss": 3.9256,
+      "step": 1605120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3711217449584405e-05,
+      "loss": 3.9122,
+      "step": 1605632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3702831502073885e-05,
+      "loss": 3.9048,
+      "step": 1606144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3694445554563365e-05,
+      "loss": 3.9031,
+      "step": 1606656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3686059607052848e-05,
+      "loss": 3.9185,
+      "step": 1607168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3677673659542324e-05,
+      "loss": 3.9177,
+      "step": 1607680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3669287712031804e-05,
+      "loss": 3.9113,
+      "step": 1608192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3660901764521284e-05,
+      "loss": 3.9185,
+      "step": 1608704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3652532195814497e-05,
+      "loss": 3.9146,
+      "step": 1609216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3644146248303973e-05,
+      "loss": 3.9128,
+      "step": 1609728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3635760300793457e-05,
+      "loss": 3.908,
+      "step": 1610240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3627374353282937e-05,
+      "loss": 3.9044,
+      "step": 1610752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3618988405772417e-05,
+      "loss": 3.91,
+      "step": 1611264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3610602458261897e-05,
+      "loss": 3.9054,
+      "step": 1611776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3602216510751377e-05,
+      "loss": 3.9193,
+      "step": 1612288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3593830563240853e-05,
+      "loss": 3.9238,
+      "step": 1612800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3585444615730333e-05,
+      "loss": 3.9223,
+      "step": 1613312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.357707504702355e-05,
+      "loss": 3.9122,
+      "step": 1613824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.356868909951303e-05,
+      "loss": 3.9114,
+      "step": 1614336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3560303152002505e-05,
+      "loss": 3.9126,
+      "step": 1614848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3551917204491985e-05,
+      "loss": 3.9071,
+      "step": 1615360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3543531256981465e-05,
+      "loss": 3.9096,
+      "step": 1615872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3535145309470945e-05,
+      "loss": 3.8962,
+      "step": 1616384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3526759361960425e-05,
+      "loss": 3.9166,
+      "step": 1616896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3518373414449905e-05,
+      "loss": 3.9074,
+      "step": 1617408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.350998746693939e-05,
+      "loss": 3.8999,
+      "step": 1617920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3501617898232598e-05,
+      "loss": 3.9114,
+      "step": 1618432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3493231950722078e-05,
+      "loss": 3.9163,
+      "step": 1618944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3484846003211558e-05,
+      "loss": 3.9212,
+      "step": 1619456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3476460055701034e-05,
+      "loss": 3.9138,
+      "step": 1619968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.346809048699425e-05,
+      "loss": 3.9108,
+      "step": 1620480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.345972091828746e-05,
+      "loss": 3.9197,
+      "step": 1620992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.345133497077694e-05,
+      "loss": 3.913,
+      "step": 1621504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.344294902326642e-05,
+      "loss": 3.9009,
+      "step": 1622016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.34345630757559e-05,
+      "loss": 3.9007,
+      "step": 1622528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.342617712824538e-05,
+      "loss": 3.9024,
+      "step": 1623040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.341780755953859e-05,
+      "loss": 3.9073,
+      "step": 1623552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.340942161202807e-05,
+      "loss": 3.9184,
+      "step": 1624064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.340103566451755e-05,
+      "loss": 3.9085,
+      "step": 1624576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.339264971700703e-05,
+      "loss": 3.9023,
+      "step": 1625088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3384263769496508e-05,
+      "loss": 3.9169,
+      "step": 1625600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3375877821985988e-05,
+      "loss": 3.9064,
+      "step": 1626112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3367491874475468e-05,
+      "loss": 3.919,
+      "step": 1626624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.335910592696495e-05,
+      "loss": 3.9022,
+      "step": 1627136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3350752737061893e-05,
+      "loss": 3.8853,
+      "step": 1627648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3342366789551373e-05,
+      "loss": 3.9247,
+      "step": 1628160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3333980842040853e-05,
+      "loss": 3.9137,
+      "step": 1628672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3325594894530333e-05,
+      "loss": 3.9143,
+      "step": 1629184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3317208947019813e-05,
+      "loss": 3.9077,
+      "step": 1629696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3308822999509293e-05,
+      "loss": 3.9035,
+      "step": 1630208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3300453430802505e-05,
+      "loss": 3.9018,
+      "step": 1630720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.329206748329198e-05,
+      "loss": 3.9046,
+      "step": 1631232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.328368153578146e-05,
+      "loss": 3.9024,
+      "step": 1631744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.327529558827094e-05,
+      "loss": 3.9055,
+      "step": 1632256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.326690964076042e-05,
+      "loss": 3.9133,
+      "step": 1632768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3258523693249905e-05,
+      "loss": 3.9066,
+      "step": 1633280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3250137745739385e-05,
+      "loss": 3.8985,
+      "step": 1633792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3241751798228865e-05,
+      "loss": 3.9193,
+      "step": 1634304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3233382229522074e-05,
+      "loss": 3.8953,
+      "step": 1634816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3224996282011554e-05,
+      "loss": 3.8968,
+      "step": 1635328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3216610334501034e-05,
+      "loss": 3.9169,
+      "step": 1635840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3208224386990514e-05,
+      "loss": 3.9163,
+      "step": 1636352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3199838439479994e-05,
+      "loss": 3.8948,
+      "step": 1636864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3191452491969474e-05,
+      "loss": 3.9065,
+      "step": 1637376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3183066544458954e-05,
+      "loss": 3.8896,
+      "step": 1637888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3174680596948434e-05,
+      "loss": 3.9084,
+      "step": 1638400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3166294649437913e-05,
+      "loss": 3.9106,
+      "step": 1638912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.315794145953486e-05,
+      "loss": 3.9093,
+      "step": 1639424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.314955551202434e-05,
+      "loss": 3.9104,
+      "step": 1639936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3141169564513815e-05,
+      "loss": 3.9149,
+      "step": 1640448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3132783617003295e-05,
+      "loss": 3.9234,
+      "step": 1640960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3124414048296507e-05,
+      "loss": 3.8984,
+      "step": 1641472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3116028100785987e-05,
+      "loss": 3.9153,
+      "step": 1641984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3107642153275467e-05,
+      "loss": 3.9164,
+      "step": 1642496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3099256205764947e-05,
+      "loss": 3.8938,
+      "step": 1643008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3090870258254427e-05,
+      "loss": 3.9157,
+      "step": 1643520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3082484310743907e-05,
+      "loss": 3.9044,
+      "step": 1644032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3074098363233387e-05,
+      "loss": 3.9193,
+      "step": 1644544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3065712415722867e-05,
+      "loss": 3.9166,
+      "step": 1645056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.3057326468212344e-05,
+      "loss": 3.9007,
+      "step": 1645568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.304897327830929e-05,
+      "loss": 3.8994,
+      "step": 1646080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.304058733079877e-05,
+      "loss": 3.9042,
+      "step": 1646592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.303220138328825e-05,
+      "loss": 3.9193,
+      "step": 1647104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.302381543577773e-05,
+      "loss": 3.9051,
+      "step": 1647616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.301544586707094e-05,
+      "loss": 3.9059,
+      "step": 1648128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.300705991956042e-05,
+      "loss": 3.9045,
+      "step": 1648640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.29986739720499e-05,
+      "loss": 3.9089,
+      "step": 1649152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.299028802453938e-05,
+      "loss": 3.9034,
+      "step": 1649664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.298190207702886e-05,
+      "loss": 3.8873,
+      "step": 1650176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.297351612951834e-05,
+      "loss": 3.9063,
+      "step": 1650688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2965130182007817e-05,
+      "loss": 3.915,
+      "step": 1651200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.295676061330103e-05,
+      "loss": 3.9138,
+      "step": 1651712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2948374665790513e-05,
+      "loss": 3.8945,
+      "step": 1652224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2939988718279993e-05,
+      "loss": 3.8999,
+      "step": 1652736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2931619149573202e-05,
+      "loss": 3.9052,
+      "step": 1653248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2923233202062682e-05,
+      "loss": 3.9095,
+      "step": 1653760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2914847254552162e-05,
+      "loss": 3.9139,
+      "step": 1654272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2906461307041642e-05,
+      "loss": 3.9022,
+      "step": 1654784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2898075359531122e-05,
+      "loss": 3.9069,
+      "step": 1655296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2889689412020602e-05,
+      "loss": 3.9134,
+      "step": 1655808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2881303464510082e-05,
+      "loss": 3.8949,
+      "step": 1656320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2872917516999562e-05,
+      "loss": 3.904,
+      "step": 1656832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.286454794829277e-05,
+      "loss": 3.8983,
+      "step": 1657344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2856178379585984e-05,
+      "loss": 3.9103,
+      "step": 1657856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2847792432075467e-05,
+      "loss": 3.8947,
+      "step": 1658368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2839406484564943e-05,
+      "loss": 3.9098,
+      "step": 1658880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2831020537054423e-05,
+      "loss": 3.901,
+      "step": 1659392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2822634589543903e-05,
+      "loss": 3.904,
+      "step": 1659904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2814248642033383e-05,
+      "loss": 3.9018,
+      "step": 1660416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2805862694522863e-05,
+      "loss": 3.9084,
+      "step": 1660928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2797476747012343e-05,
+      "loss": 3.9061,
+      "step": 1661440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2789090799501823e-05,
+      "loss": 3.9191,
+      "step": 1661952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2780721230795036e-05,
+      "loss": 3.9091,
+      "step": 1662464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2772335283284516e-05,
+      "loss": 3.9133,
+      "step": 1662976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2763949335773996e-05,
+      "loss": 3.9016,
+      "step": 1663488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2755563388263472e-05,
+      "loss": 3.9086,
+      "step": 1664000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2747177440752952e-05,
+      "loss": 3.8996,
+      "step": 1664512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2738791493242435e-05,
+      "loss": 3.9061,
+      "step": 1665024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2730405545731915e-05,
+      "loss": 3.9045,
+      "step": 1665536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2722019598221395e-05,
+      "loss": 3.9068,
+      "step": 1666048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2713650029514604e-05,
+      "loss": 3.9059,
+      "step": 1666560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2705264082004084e-05,
+      "loss": 3.9036,
+      "step": 1667072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2696878134493564e-05,
+      "loss": 3.9045,
+      "step": 1667584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2688508565786777e-05,
+      "loss": 3.9078,
+      "step": 1668096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2680122618276257e-05,
+      "loss": 3.9027,
+      "step": 1668608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2671736670765737e-05,
+      "loss": 3.91,
+      "step": 1669120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2663350723255217e-05,
+      "loss": 3.9056,
+      "step": 1669632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2654964775744697e-05,
+      "loss": 3.9001,
+      "step": 1670144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2646578828234177e-05,
+      "loss": 3.8887,
+      "step": 1670656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2638192880723653e-05,
+      "loss": 3.9127,
+      "step": 1671168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.262982331201687e-05,
+      "loss": 3.9121,
+      "step": 1671680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.262143736450635e-05,
+      "loss": 3.9127,
+      "step": 1672192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2613051416995826e-05,
+      "loss": 3.9077,
+      "step": 1672704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2604665469485306e-05,
+      "loss": 3.9111,
+      "step": 1673216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2596279521974786e-05,
+      "loss": 3.9055,
+      "step": 1673728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2587893574464265e-05,
+      "loss": 3.9171,
+      "step": 1674240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2579507626953745e-05,
+      "loss": 3.8985,
+      "step": 1674752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.257112167944323e-05,
+      "loss": 3.9079,
+      "step": 1675264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.256273573193271e-05,
+      "loss": 3.9049,
+      "step": 1675776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2554366163225918e-05,
+      "loss": 3.9036,
+      "step": 1676288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2545980215715398e-05,
+      "loss": 3.9154,
+      "step": 1676800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2537594268204878e-05,
+      "loss": 3.9068,
+      "step": 1677312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.2529208320694358e-05,
+      "loss": 3.9118,
+      "step": 1677824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.252083875198757e-05,
+      "loss": 3.9015,
+      "step": 1678336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.251245280447705e-05,
+      "loss": 3.9072,
+      "step": 1678848
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.011335849761963,
+      "eval_runtime": 292.0697,
+      "eval_samples_per_second": 1306.507,
+      "eval_steps_per_second": 40.829,
+      "step": 1679040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.250406685696653e-05,
+      "loss": 3.9067,
+      "step": 1679360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2495680909456007e-05,
+      "loss": 3.9103,
+      "step": 1679872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.248731134074922e-05,
+      "loss": 3.9112,
+      "step": 1680384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.24789253932387e-05,
+      "loss": 3.9019,
+      "step": 1680896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2470539445728182e-05,
+      "loss": 3.9193,
+      "step": 1681408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.246216987702139e-05,
+      "loss": 3.9039,
+      "step": 1681920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.245378392951087e-05,
+      "loss": 3.9007,
+      "step": 1682432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.244539798200035e-05,
+      "loss": 3.8948,
+      "step": 1682944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.243701203448983e-05,
+      "loss": 3.9114,
+      "step": 1683456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2428626086979308e-05,
+      "loss": 3.914,
+      "step": 1683968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.242024013946879e-05,
+      "loss": 3.9043,
+      "step": 1684480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2411870570762004e-05,
+      "loss": 3.9119,
+      "step": 1684992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2403501002055213e-05,
+      "loss": 3.9069,
+      "step": 1685504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2395115054544693e-05,
+      "loss": 3.9057,
+      "step": 1686016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2386729107034173e-05,
+      "loss": 3.8979,
+      "step": 1686528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2378343159523653e-05,
+      "loss": 3.8972,
+      "step": 1687040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2369957212013133e-05,
+      "loss": 3.9036,
+      "step": 1687552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2361571264502613e-05,
+      "loss": 3.8999,
+      "step": 1688064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2353185316992093e-05,
+      "loss": 3.9114,
+      "step": 1688576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2344799369481573e-05,
+      "loss": 3.9145,
+      "step": 1689088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2336413421971053e-05,
+      "loss": 3.9173,
+      "step": 1689600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2328027474460533e-05,
+      "loss": 3.9061,
+      "step": 1690112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2319641526950012e-05,
+      "loss": 3.9052,
+      "step": 1690624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2311255579439492e-05,
+      "loss": 3.9046,
+      "step": 1691136
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2302886010732705e-05,
+      "loss": 3.9032,
+      "step": 1691648
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2294500063222185e-05,
+      "loss": 3.9048,
+      "step": 1692160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2286130494515394e-05,
+      "loss": 3.8868,
+      "step": 1692672
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2277744547004874e-05,
+      "loss": 3.9098,
+      "step": 1693184
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2269358599494354e-05,
+      "loss": 3.9032,
+      "step": 1693696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.2260972651983837e-05,
+      "loss": 3.8922,
+      "step": 1694208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2252603083277046e-05,
+      "loss": 3.9018,
+      "step": 1694720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2244233514570255e-05,
+      "loss": 3.9123,
+      "step": 1695232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2235847567059735e-05,
+      "loss": 3.9085,
+      "step": 1695744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2227461619549215e-05,
+      "loss": 3.9081,
+      "step": 1696256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.22190756720387e-05,
+      "loss": 3.9011,
+      "step": 1696768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.221068972452818e-05,
+      "loss": 3.9139,
+      "step": 1697280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.220230377701766e-05,
+      "loss": 3.9069,
+      "step": 1697792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2193917829507135e-05,
+      "loss": 3.8933,
+      "step": 1698304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2185531881996615e-05,
+      "loss": 3.8951,
+      "step": 1698816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2177145934486095e-05,
+      "loss": 3.8938,
+      "step": 1699328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2168759986975575e-05,
+      "loss": 3.9023,
+      "step": 1699840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2160374039465055e-05,
+      "loss": 3.9106,
+      "step": 1700352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.215198809195454e-05,
+      "loss": 3.9011,
+      "step": 1700864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2143618523247747e-05,
+      "loss": 3.8992,
+      "step": 1701376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2135232575737227e-05,
+      "loss": 3.9054,
+      "step": 1701888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2126846628226707e-05,
+      "loss": 3.9005,
+      "step": 1702400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.211847705951992e-05,
+      "loss": 3.9141,
+      "step": 1702912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.21100911120094e-05,
+      "loss": 3.8985,
+      "step": 1703424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.210170516449888e-05,
+      "loss": 3.8744,
+      "step": 1703936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.209331921698836e-05,
+      "loss": 3.9163,
+      "step": 1704448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.208493326947784e-05,
+      "loss": 3.9097,
+      "step": 1704960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2076547321967316e-05,
+      "loss": 3.9078,
+      "step": 1705472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2068161374456796e-05,
+      "loss": 3.9004,
+      "step": 1705984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2059775426946276e-05,
+      "loss": 3.8977,
+      "step": 1706496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2051405858239492e-05,
+      "loss": 3.8948,
+      "step": 1707008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.204301991072897e-05,
+      "loss": 3.9013,
+      "step": 1707520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.203465034202218e-05,
+      "loss": 3.8975,
+      "step": 1708032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.202626439451166e-05,
+      "loss": 3.8976,
+      "step": 1708544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2017878447001138e-05,
+      "loss": 3.9067,
+      "step": 1709056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.200949249949062e-05,
+      "loss": 3.9022,
+      "step": 1709568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.20011065519801e-05,
+      "loss": 3.8899,
+      "step": 1710080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1992736983273313e-05,
+      "loss": 3.9118,
+      "step": 1710592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.198435103576279e-05,
+      "loss": 3.8891,
+      "step": 1711104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.197596508825227e-05,
+      "loss": 3.8932,
+      "step": 1711616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.196757914074175e-05,
+      "loss": 3.9099,
+      "step": 1712128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1959209572034962e-05,
+      "loss": 3.9092,
+      "step": 1712640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1950823624524442e-05,
+      "loss": 3.8884,
+      "step": 1713152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1942437677013922e-05,
+      "loss": 3.8973,
+      "step": 1713664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1934051729503402e-05,
+      "loss": 3.8833,
+      "step": 1714176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1925665781992882e-05,
+      "loss": 3.9049,
+      "step": 1714688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1917279834482362e-05,
+      "loss": 3.9049,
+      "step": 1715200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1908893886971842e-05,
+      "loss": 3.9049,
+      "step": 1715712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1900507939461322e-05,
+      "loss": 3.9004,
+      "step": 1716224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1892121991950802e-05,
+      "loss": 3.9082,
+      "step": 1716736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1883752423244014e-05,
+      "loss": 3.9182,
+      "step": 1717248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1875366475733494e-05,
+      "loss": 3.8953,
+      "step": 1717760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.186698052822297e-05,
+      "loss": 3.9055,
+      "step": 1718272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.185859458071245e-05,
+      "loss": 3.9101,
+      "step": 1718784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.185020863320193e-05,
+      "loss": 3.8913,
+      "step": 1719296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1841839064495147e-05,
+      "loss": 3.9051,
+      "step": 1719808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1833453116984623e-05,
+      "loss": 3.9045,
+      "step": 1720320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1825067169474103e-05,
+      "loss": 3.9074,
+      "step": 1720832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1816681221963583e-05,
+      "loss": 3.9094,
+      "step": 1721344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1808295274453063e-05,
+      "loss": 3.8952,
+      "step": 1721856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1799925705746276e-05,
+      "loss": 3.8925,
+      "step": 1722368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1791539758235756e-05,
+      "loss": 3.8973,
+      "step": 1722880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1783153810725236e-05,
+      "loss": 3.9078,
+      "step": 1723392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1774767863214716e-05,
+      "loss": 3.901,
+      "step": 1723904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.1766398294507925e-05,
+      "loss": 3.8991,
+      "step": 1724416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1758012346997405e-05,
+      "loss": 3.9006,
+      "step": 1724928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1749626399486885e-05,
+      "loss": 3.9028,
+      "step": 1725440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1741240451976368e-05,
+      "loss": 3.8972,
+      "step": 1725952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1732854504465848e-05,
+      "loss": 3.8814,
+      "step": 1726464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1724468556955328e-05,
+      "loss": 3.9013,
+      "step": 1726976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1716082609444804e-05,
+      "loss": 3.9051,
+      "step": 1727488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1707696661934284e-05,
+      "loss": 3.907,
+      "step": 1728000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1699310714423764e-05,
+      "loss": 3.8894,
+      "step": 1728512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1690941145716977e-05,
+      "loss": 3.8952,
+      "step": 1729024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1682555198206457e-05,
+      "loss": 3.8986,
+      "step": 1729536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1674169250695937e-05,
+      "loss": 3.9023,
+      "step": 1730048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1665783303185417e-05,
+      "loss": 3.9096,
+      "step": 1730560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1657430113282358e-05,
+      "loss": 3.896,
+      "step": 1731072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1649044165771838e-05,
+      "loss": 3.9,
+      "step": 1731584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.164065821826132e-05,
+      "loss": 3.9064,
+      "step": 1732096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.16322722707508e-05,
+      "loss": 3.8865,
+      "step": 1732608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1623886323240278e-05,
+      "loss": 3.8998,
+      "step": 1733120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1615500375729758e-05,
+      "loss": 3.8865,
+      "step": 1733632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1607114428219238e-05,
+      "loss": 3.9094,
+      "step": 1734144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1598744859512447e-05,
+      "loss": 3.8888,
+      "step": 1734656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.159035891200193e-05,
+      "loss": 3.9029,
+      "step": 1735168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.158197296449141e-05,
+      "loss": 3.8953,
+      "step": 1735680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.157358701698089e-05,
+      "loss": 3.8957,
+      "step": 1736192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.156520106947037e-05,
+      "loss": 3.8971,
+      "step": 1736704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.155683150076358e-05,
+      "loss": 3.9038,
+      "step": 1737216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.154844555325306e-05,
+      "loss": 3.8983,
+      "step": 1737728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.154005960574254e-05,
+      "loss": 3.9136,
+      "step": 1738240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1531690037035752e-05,
+      "loss": 3.8986,
+      "step": 1738752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1523304089525232e-05,
+      "loss": 3.913,
+      "step": 1739264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.151491814201471e-05,
+      "loss": 3.8968,
+      "step": 1739776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.150653219450419e-05,
+      "loss": 3.8983,
+      "step": 1740288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.149814624699367e-05,
+      "loss": 3.8968,
+      "step": 1740800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.148976029948315e-05,
+      "loss": 3.8954,
+      "step": 1741312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.148137435197263e-05,
+      "loss": 3.8951,
+      "step": 1741824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.147298840446211e-05,
+      "loss": 3.9043,
+      "step": 1742336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.146460245695159e-05,
+      "loss": 3.8975,
+      "step": 1742848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.145621650944107e-05,
+      "loss": 3.9005,
+      "step": 1743360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.144784694073428e-05,
+      "loss": 3.8998,
+      "step": 1743872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.143946099322376e-05,
+      "loss": 3.9005,
+      "step": 1744384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.143107504571324e-05,
+      "loss": 3.8969,
+      "step": 1744896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1422689098202724e-05,
+      "loss": 3.9,
+      "step": 1745408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1414303150692204e-05,
+      "loss": 3.8948,
+      "step": 1745920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1405917203181684e-05,
+      "loss": 3.8979,
+      "step": 1746432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.139753125567116e-05,
+      "loss": 3.8827,
+      "step": 1746944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.138914530816064e-05,
+      "loss": 3.9053,
+      "step": 1747456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1380775739453853e-05,
+      "loss": 3.9056,
+      "step": 1747968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1372389791943333e-05,
+      "loss": 3.9079,
+      "step": 1748480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1364003844432813e-05,
+      "loss": 3.8973,
+      "step": 1748992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1355617896922293e-05,
+      "loss": 3.9073,
+      "step": 1749504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1347248328215505e-05,
+      "loss": 3.9018,
+      "step": 1750016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1338862380704985e-05,
+      "loss": 3.9095,
+      "step": 1750528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.133047643319446e-05,
+      "loss": 3.8915,
+      "step": 1751040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1322090485683945e-05,
+      "loss": 3.8968,
+      "step": 1751552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1313720916977157e-05,
+      "loss": 3.9021,
+      "step": 1752064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1305334969466634e-05,
+      "loss": 3.9004,
+      "step": 1752576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1296949021956114e-05,
+      "loss": 3.9035,
+      "step": 1753088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1288563074445594e-05,
+      "loss": 3.9015,
+      "step": 1753600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1280177126935074e-05,
+      "loss": 3.9039,
+      "step": 1754112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1271791179424554e-05,
+      "loss": 3.8956,
+      "step": 1754624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.12634379895215e-05,
+      "loss": 3.906,
+      "step": 1755136
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.009059429168701,
+      "eval_runtime": 293.9574,
+      "eval_samples_per_second": 1298.117,
+      "eval_steps_per_second": 40.567,
+      "step": 1755360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.125505204201098e-05,
+      "loss": 3.9032,
+      "step": 1755648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.124666609450046e-05,
+      "loss": 3.9056,
+      "step": 1756160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1238280146989935e-05,
+      "loss": 3.899,
+      "step": 1756672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1229894199479415e-05,
+      "loss": 3.8972,
+      "step": 1757184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1221508251968895e-05,
+      "loss": 3.9098,
+      "step": 1757696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.121312230445838e-05,
+      "loss": 3.8999,
+      "step": 1758208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.120473635694786e-05,
+      "loss": 3.8943,
+      "step": 1758720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.119635040943734e-05,
+      "loss": 3.8918,
+      "step": 1759232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1187964461926815e-05,
+      "loss": 3.899,
+      "step": 1759744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1179578514416295e-05,
+      "loss": 3.9106,
+      "step": 1760256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1171192566905775e-05,
+      "loss": 3.8971,
+      "step": 1760768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1162806619395255e-05,
+      "loss": 3.9063,
+      "step": 1761280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1154437050688467e-05,
+      "loss": 3.9065,
+      "step": 1761792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.114606748198168e-05,
+      "loss": 3.8947,
+      "step": 1762304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.113768153447116e-05,
+      "loss": 3.8929,
+      "step": 1762816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.112929558696064e-05,
+      "loss": 3.8906,
+      "step": 1763328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1120909639450116e-05,
+      "loss": 3.898,
+      "step": 1763840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.11125236919396e-05,
+      "loss": 3.8945,
+      "step": 1764352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.110413774442908e-05,
+      "loss": 3.9017,
+      "step": 1764864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.109575179691856e-05,
+      "loss": 3.9117,
+      "step": 1765376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.108736584940804e-05,
+      "loss": 3.9109,
+      "step": 1765888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.107897990189752e-05,
+      "loss": 3.9001,
+      "step": 1766400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1070593954386996e-05,
+      "loss": 3.8966,
+      "step": 1766912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1062208006876476e-05,
+      "loss": 3.8968,
+      "step": 1767424
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1053822059365956e-05,
+      "loss": 3.8989,
+      "step": 1767936
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.104543611185544e-05,
+      "loss": 3.8989,
+      "step": 1768448
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.103706654314865e-05,
+      "loss": 3.8824,
+      "step": 1768960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.102868059563813e-05,
+      "loss": 3.9047,
+      "step": 1769472
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1020294648127608e-05,
+      "loss": 3.8988,
+      "step": 1769984
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.1011908700617088e-05,
+      "loss": 3.888,
+      "step": 1770496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1003522753106568e-05,
+      "loss": 3.8903,
+      "step": 1771008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0995136805596048e-05,
+      "loss": 3.9077,
+      "step": 1771520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.098675085808553e-05,
+      "loss": 3.9051,
+      "step": 1772032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0978364910575008e-05,
+      "loss": 3.9004,
+      "step": 1772544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.096999534186822e-05,
+      "loss": 3.8961,
+      "step": 1773056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.096162577316143e-05,
+      "loss": 3.9023,
+      "step": 1773568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.095323982565091e-05,
+      "loss": 3.904,
+      "step": 1774080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0944853878140393e-05,
+      "loss": 3.8876,
+      "step": 1774592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0936467930629873e-05,
+      "loss": 3.8871,
+      "step": 1775104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0928098361923082e-05,
+      "loss": 3.8889,
+      "step": 1775616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0919712414412562e-05,
+      "loss": 3.8957,
+      "step": 1776128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0911326466902042e-05,
+      "loss": 3.899,
+      "step": 1776640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0902940519391522e-05,
+      "loss": 3.8985,
+      "step": 1777152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0894570950684734e-05,
+      "loss": 3.8926,
+      "step": 1777664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0886185003174214e-05,
+      "loss": 3.8987,
+      "step": 1778176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0877799055663694e-05,
+      "loss": 3.8927,
+      "step": 1778688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0869413108153174e-05,
+      "loss": 3.9023,
+      "step": 1779200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.086102716064265e-05,
+      "loss": 3.8967,
+      "step": 1779712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.085264121313213e-05,
+      "loss": 3.8701,
+      "step": 1780224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.084425526562161e-05,
+      "loss": 3.9074,
+      "step": 1780736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0835885696914827e-05,
+      "loss": 3.9024,
+      "step": 1781248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0827499749404303e-05,
+      "loss": 3.9052,
+      "step": 1781760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0819113801893783e-05,
+      "loss": 3.8929,
+      "step": 1782272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0810727854383263e-05,
+      "loss": 3.8901,
+      "step": 1782784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0802358285676472e-05,
+      "loss": 3.8912,
+      "step": 1783296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0793972338165955e-05,
+      "loss": 3.8946,
+      "step": 1783808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0785586390655435e-05,
+      "loss": 3.8937,
+      "step": 1784320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0777200443144915e-05,
+      "loss": 3.8856,
+      "step": 1784832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0768814495634395e-05,
+      "loss": 3.9009,
+      "step": 1785344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0760444926927604e-05,
+      "loss": 3.8994,
+      "step": 1785856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0752058979417084e-05,
+      "loss": 3.8835,
+      "step": 1786368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0743673031906564e-05,
+      "loss": 3.9039,
+      "step": 1786880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0735287084396048e-05,
+      "loss": 3.8837,
+      "step": 1787392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0726901136885528e-05,
+      "loss": 3.8848,
+      "step": 1787904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0718531568178737e-05,
+      "loss": 3.9024,
+      "step": 1788416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0710145620668217e-05,
+      "loss": 3.9042,
+      "step": 1788928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0701759673157697e-05,
+      "loss": 3.8841,
+      "step": 1789440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0693373725647177e-05,
+      "loss": 3.8941,
+      "step": 1789952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0684987778136657e-05,
+      "loss": 3.8751,
+      "step": 1790464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0676634588233598e-05,
+      "loss": 3.8975,
+      "step": 1790976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0668248640723078e-05,
+      "loss": 3.899,
+      "step": 1791488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0659862693212558e-05,
+      "loss": 3.8981,
+      "step": 1792000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0651476745702038e-05,
+      "loss": 3.8922,
+      "step": 1792512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.064310717699525e-05,
+      "loss": 3.9016,
+      "step": 1793024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0634737608288463e-05,
+      "loss": 3.9127,
+      "step": 1793536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0626351660777943e-05,
+      "loss": 3.8854,
+      "step": 1794048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.061796571326742e-05,
+      "loss": 3.9005,
+      "step": 1794560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.06095797657569e-05,
+      "loss": 3.9036,
+      "step": 1795072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.060119381824638e-05,
+      "loss": 3.885,
+      "step": 1795584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0592807870735863e-05,
+      "loss": 3.8972,
+      "step": 1796096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0584421923225343e-05,
+      "loss": 3.8985,
+      "step": 1796608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0576035975714823e-05,
+      "loss": 3.9004,
+      "step": 1797120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0567650028204303e-05,
+      "loss": 3.9084,
+      "step": 1797632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.055926408069378e-05,
+      "loss": 3.8848,
+      "step": 1798144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.055087813318326e-05,
+      "loss": 3.8887,
+      "step": 1798656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.054249218567274e-05,
+      "loss": 3.8914,
+      "step": 1799168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.053410623816222e-05,
+      "loss": 3.9019,
+      "step": 1799680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.052573666945543e-05,
+      "loss": 3.8927,
+      "step": 1800192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.051735072194491e-05,
+      "loss": 3.8982,
+      "step": 1800704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.050896477443439e-05,
+      "loss": 3.8962,
+      "step": 1801216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.050057882692387e-05,
+      "loss": 3.8915,
+      "step": 1801728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.049219287941335e-05,
+      "loss": 3.894,
+      "step": 1802240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.048380693190283e-05,
+      "loss": 3.8767,
+      "step": 1802752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.047542098439231e-05,
+      "loss": 3.8932,
+      "step": 1803264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.046703503688179e-05,
+      "loss": 3.8986,
+      "step": 1803776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0458681846978733e-05,
+      "loss": 3.903,
+      "step": 1804288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0450295899468213e-05,
+      "loss": 3.8857,
+      "step": 1804800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0441909951957693e-05,
+      "loss": 3.886,
+      "step": 1805312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0433524004447173e-05,
+      "loss": 3.8937,
+      "step": 1805824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0425138056936656e-05,
+      "loss": 3.8971,
+      "step": 1806336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0416768488229865e-05,
+      "loss": 3.8999,
+      "step": 1806848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0408382540719345e-05,
+      "loss": 3.8934,
+      "step": 1807360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0399996593208825e-05,
+      "loss": 3.8974,
+      "step": 1807872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0391610645698305e-05,
+      "loss": 3.8957,
+      "step": 1808384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0383224698187785e-05,
+      "loss": 3.8855,
+      "step": 1808896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0374838750677265e-05,
+      "loss": 3.8903,
+      "step": 1809408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0366452803166745e-05,
+      "loss": 3.884,
+      "step": 1809920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0358083234459957e-05,
+      "loss": 3.9026,
+      "step": 1810432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0349697286949434e-05,
+      "loss": 3.8833,
+      "step": 1810944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0341311339438914e-05,
+      "loss": 3.8941,
+      "step": 1811456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0332925391928394e-05,
+      "loss": 3.8882,
+      "step": 1811968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.032455582322161e-05,
+      "loss": 3.8945,
+      "step": 1812480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0316169875711086e-05,
+      "loss": 3.8898,
+      "step": 1812992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0307783928200566e-05,
+      "loss": 3.8951,
+      "step": 1813504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0299397980690046e-05,
+      "loss": 3.8943,
+      "step": 1814016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0291012033179526e-05,
+      "loss": 3.9083,
+      "step": 1814528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0282626085669006e-05,
+      "loss": 3.8901,
+      "step": 1815040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.027425651696222e-05,
+      "loss": 3.9049,
+      "step": 1815552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.02658705694517e-05,
+      "loss": 3.8902,
+      "step": 1816064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.025748462194118e-05,
+      "loss": 3.8928,
+      "step": 1816576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0249115053234388e-05,
+      "loss": 3.8917,
+      "step": 1817088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0240729105723868e-05,
+      "loss": 3.8948,
+      "step": 1817600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0232343158213348e-05,
+      "loss": 3.8852,
+      "step": 1818112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0223957210702828e-05,
+      "loss": 3.8963,
+      "step": 1818624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.021557126319231e-05,
+      "loss": 3.8875,
+      "step": 1819136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0207185315681787e-05,
+      "loss": 3.8962,
+      "step": 1819648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0198799368171267e-05,
+      "loss": 3.8956,
+      "step": 1820160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0190413420660747e-05,
+      "loss": 3.8916,
+      "step": 1820672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0182027473150227e-05,
+      "loss": 3.8947,
+      "step": 1821184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.017365790444344e-05,
+      "loss": 3.898,
+      "step": 1821696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.016527195693292e-05,
+      "loss": 3.8874,
+      "step": 1822208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.01568860094224e-05,
+      "loss": 3.8939,
+      "step": 1822720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.014850006191188e-05,
+      "loss": 3.8754,
+      "step": 1823232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.014011411440136e-05,
+      "loss": 3.8994,
+      "step": 1823744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.013172816689084e-05,
+      "loss": 3.902,
+      "step": 1824256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.012334221938032e-05,
+      "loss": 3.9016,
+      "step": 1824768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0114956271869796e-05,
+      "loss": 3.8921,
+      "step": 1825280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0106586703163012e-05,
+      "loss": 3.9007,
+      "step": 1825792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0098200755652492e-05,
+      "loss": 3.8979,
+      "step": 1826304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.008981480814197e-05,
+      "loss": 3.9033,
+      "step": 1826816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.008144523943518e-05,
+      "loss": 3.8848,
+      "step": 1827328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.007305929192466e-05,
+      "loss": 3.8934,
+      "step": 1827840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.006467334441414e-05,
+      "loss": 3.8935,
+      "step": 1828352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.005628739690362e-05,
+      "loss": 3.8968,
+      "step": 1828864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0047917828196833e-05,
+      "loss": 3.8967,
+      "step": 1829376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0039531880686313e-05,
+      "loss": 3.8923,
+      "step": 1829888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.0031145933175793e-05,
+      "loss": 3.9016,
+      "step": 1830400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.002275998566527e-05,
+      "loss": 3.8882,
+      "step": 1830912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.001437403815475e-05,
+      "loss": 3.9027,
+      "step": 1831424
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.00673246383667,
+      "eval_runtime": 504.2103,
+      "eval_samples_per_second": 756.809,
+      "eval_steps_per_second": 23.651,
+      "step": 1831680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.0005988090644233e-05,
+      "loss": 3.888,
+      "step": 1831936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9997618521937442e-05,
+      "loss": 3.898,
+      "step": 1832448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9989232574426922e-05,
+      "loss": 3.8977,
+      "step": 1832960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9980863005720135e-05,
+      "loss": 3.8945,
+      "step": 1833472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9972477058209615e-05,
+      "loss": 3.9058,
+      "step": 1833984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9964091110699095e-05,
+      "loss": 3.8942,
+      "step": 1834496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9955705163188575e-05,
+      "loss": 3.8892,
+      "step": 1835008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9947319215678054e-05,
+      "loss": 3.8859,
+      "step": 1835520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9938933268167534e-05,
+      "loss": 3.8926,
+      "step": 1836032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9930547320657014e-05,
+      "loss": 3.9045,
+      "step": 1836544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9922161373146494e-05,
+      "loss": 3.896,
+      "step": 1837056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9913775425635974e-05,
+      "loss": 3.8977,
+      "step": 1837568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9905405856929187e-05,
+      "loss": 3.9011,
+      "step": 1838080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9897019909418667e-05,
+      "loss": 3.8891,
+      "step": 1838592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9888633961908147e-05,
+      "loss": 3.8879,
+      "step": 1839104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9880248014397623e-05,
+      "loss": 3.8846,
+      "step": 1839616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9871862066887103e-05,
+      "loss": 3.8934,
+      "step": 1840128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9863492498180316e-05,
+      "loss": 3.887,
+      "step": 1840640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9855106550669796e-05,
+      "loss": 3.8946,
+      "step": 1841152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9846720603159276e-05,
+      "loss": 3.9047,
+      "step": 1841664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9838334655648756e-05,
+      "loss": 3.9061,
+      "step": 1842176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9829948708138236e-05,
+      "loss": 3.8926,
+      "step": 1842688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9821579139431448e-05,
+      "loss": 3.8931,
+      "step": 1843200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9813193191920925e-05,
+      "loss": 3.8905,
+      "step": 1843712
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9804807244410405e-05,
+      "loss": 3.8894,
+      "step": 1844224
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9796421296899888e-05,
+      "loss": 3.8916,
+      "step": 1844736
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9788035349389368e-05,
+      "loss": 3.8771,
+      "step": 1845248
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9779649401878848e-05,
+      "loss": 3.8991,
+      "step": 1845760
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9771263454368328e-05,
+      "loss": 3.8939,
+      "step": 1846272
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.9762877506857804e-05,
+      "loss": 3.8785,
+      "step": 1846784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.975452431695475e-05,
+      "loss": 3.8859,
+      "step": 1847296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.974613836944423e-05,
+      "loss": 3.9013,
+      "step": 1847808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.973775242193371e-05,
+      "loss": 3.9015,
+      "step": 1848320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.972936647442319e-05,
+      "loss": 3.8952,
+      "step": 1848832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9720996905716398e-05,
+      "loss": 3.891,
+      "step": 1849344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.971262733700961e-05,
+      "loss": 3.8906,
+      "step": 1849856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.970424138949909e-05,
+      "loss": 3.9043,
+      "step": 1850368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.969585544198857e-05,
+      "loss": 3.8817,
+      "step": 1850880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.968746949447805e-05,
+      "loss": 3.8807,
+      "step": 1851392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.967908354696753e-05,
+      "loss": 3.8859,
+      "step": 1851904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.967069759945701e-05,
+      "loss": 3.887,
+      "step": 1852416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.966231165194649e-05,
+      "loss": 3.8956,
+      "step": 1852928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.965392570443597e-05,
+      "loss": 3.8878,
+      "step": 1853440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.964553975692545e-05,
+      "loss": 3.8937,
+      "step": 1853952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.963715380941493e-05,
+      "loss": 3.8891,
+      "step": 1854464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.962876786190441e-05,
+      "loss": 3.8853,
+      "step": 1854976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.962038191439389e-05,
+      "loss": 3.9026,
+      "step": 1855488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.96120123456871e-05,
+      "loss": 3.8888,
+      "step": 1856000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9603642776980312e-05,
+      "loss": 3.8686,
+      "step": 1856512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9595256829469795e-05,
+      "loss": 3.9006,
+      "step": 1857024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9586870881959275e-05,
+      "loss": 3.8933,
+      "step": 1857536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9578484934448752e-05,
+      "loss": 3.9025,
+      "step": 1858048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.957009898693823e-05,
+      "loss": 3.8891,
+      "step": 1858560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.956171303942771e-05,
+      "loss": 3.8827,
+      "step": 1859072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.955332709191719e-05,
+      "loss": 3.8844,
+      "step": 1859584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.954494114440667e-05,
+      "loss": 3.8856,
+      "step": 1860096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.953655519689615e-05,
+      "loss": 3.8937,
+      "step": 1860608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9528185628189364e-05,
+      "loss": 3.8764,
+      "step": 1861120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9519799680678844e-05,
+      "loss": 3.8989,
+      "step": 1861632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9511430111972053e-05,
+      "loss": 3.89,
+      "step": 1862144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9503044164461533e-05,
+      "loss": 3.8796,
+      "step": 1862656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9494658216951013e-05,
+      "loss": 3.8967,
+      "step": 1863168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9486272269440496e-05,
+      "loss": 3.8793,
+      "step": 1863680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9477886321929976e-05,
+      "loss": 3.8798,
+      "step": 1864192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9469500374419456e-05,
+      "loss": 3.8961,
+      "step": 1864704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9461130805712665e-05,
+      "loss": 3.8993,
+      "step": 1865216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9452744858202145e-05,
+      "loss": 3.8737,
+      "step": 1865728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9444358910691625e-05,
+      "loss": 3.8929,
+      "step": 1866240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9435972963181105e-05,
+      "loss": 3.8701,
+      "step": 1866752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9427587015670585e-05,
+      "loss": 3.8915,
+      "step": 1867264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9419201068160065e-05,
+      "loss": 3.8944,
+      "step": 1867776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9410831499453278e-05,
+      "loss": 3.8871,
+      "step": 1868288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9402445551942754e-05,
+      "loss": 3.8947,
+      "step": 1868800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9394059604432234e-05,
+      "loss": 3.8925,
+      "step": 1869312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9385673656921714e-05,
+      "loss": 3.9101,
+      "step": 1869824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9377287709411197e-05,
+      "loss": 3.8779,
+      "step": 1870336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9368901761900677e-05,
+      "loss": 3.8956,
+      "step": 1870848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9360515814390157e-05,
+      "loss": 3.8944,
+      "step": 1871360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9352129866879637e-05,
+      "loss": 3.8855,
+      "step": 1871872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9343760298172846e-05,
+      "loss": 3.8927,
+      "step": 1872384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.933539072946606e-05,
+      "loss": 3.8905,
+      "step": 1872896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.932700478195554e-05,
+      "loss": 3.8952,
+      "step": 1873408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.931861883444502e-05,
+      "loss": 3.9015,
+      "step": 1873920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.93102328869345e-05,
+      "loss": 3.8775,
+      "step": 1874432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9301863318227708e-05,
+      "loss": 3.885,
+      "step": 1874944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9293477370717188e-05,
+      "loss": 3.8874,
+      "step": 1875456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.9285091423206668e-05,
+      "loss": 3.8976,
+      "step": 1875968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.927670547569615e-05,
+      "loss": 3.8838,
+      "step": 1876480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.926831952818563e-05,
+      "loss": 3.8919,
+      "step": 1876992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.925994995947884e-05,
+      "loss": 3.8924,
+      "step": 1877504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.925156401196832e-05,
+      "loss": 3.8863,
+      "step": 1878016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.92431780644578e-05,
+      "loss": 3.8873,
+      "step": 1878528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.923479211694728e-05,
+      "loss": 3.8718,
+      "step": 1879040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.922640616943676e-05,
+      "loss": 3.8856,
+      "step": 1879552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.921802022192624e-05,
+      "loss": 3.8927,
+      "step": 1880064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.920963427441572e-05,
+      "loss": 3.8974,
+      "step": 1880576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.92012483269052e-05,
+      "loss": 3.8816,
+      "step": 1881088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.919287875819841e-05,
+      "loss": 3.8796,
+      "step": 1881600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.918449281068789e-05,
+      "loss": 3.8864,
+      "step": 1882112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.917610686317737e-05,
+      "loss": 3.8929,
+      "step": 1882624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9167737294470585e-05,
+      "loss": 3.8932,
+      "step": 1883136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.915935134696006e-05,
+      "loss": 3.8876,
+      "step": 1883648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.915096539944954e-05,
+      "loss": 3.8936,
+      "step": 1884160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.914257945193902e-05,
+      "loss": 3.8865,
+      "step": 1884672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.91341935044285e-05,
+      "loss": 3.8843,
+      "step": 1885184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.912580755691798e-05,
+      "loss": 3.8805,
+      "step": 1885696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.911742160940746e-05,
+      "loss": 3.8826,
+      "step": 1886208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9109052040700674e-05,
+      "loss": 3.8972,
+      "step": 1886720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9100666093190153e-05,
+      "loss": 3.8771,
+      "step": 1887232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9092280145679633e-05,
+      "loss": 3.8863,
+      "step": 1887744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9083910576972843e-05,
+      "loss": 3.8801,
+      "step": 1888256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9075524629462322e-05,
+      "loss": 3.8905,
+      "step": 1888768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9067138681951806e-05,
+      "loss": 3.8834,
+      "step": 1889280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9058752734441286e-05,
+      "loss": 3.8897,
+      "step": 1889792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9050366786930766e-05,
+      "loss": 3.8902,
+      "step": 1890304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9041980839420242e-05,
+      "loss": 3.9,
+      "step": 1890816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9033594891909722e-05,
+      "loss": 3.8882,
+      "step": 1891328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9025225323202935e-05,
+      "loss": 3.8973,
+      "step": 1891840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9016839375692415e-05,
+      "loss": 3.8873,
+      "step": 1892352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9008453428181895e-05,
+      "loss": 3.8881,
+      "step": 1892864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.9000067480671375e-05,
+      "loss": 3.8882,
+      "step": 1893376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8991697911964587e-05,
+      "loss": 3.8878,
+      "step": 1893888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8983311964454064e-05,
+      "loss": 3.8816,
+      "step": 1894400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8974926016943544e-05,
+      "loss": 3.8909,
+      "step": 1894912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8966540069433027e-05,
+      "loss": 3.8837,
+      "step": 1895424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8958154121922507e-05,
+      "loss": 3.8915,
+      "step": 1895936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8949784553215716e-05,
+      "loss": 3.8858,
+      "step": 1896448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8941398605705196e-05,
+      "loss": 3.8895,
+      "step": 1896960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8933012658194676e-05,
+      "loss": 3.8864,
+      "step": 1897472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8924626710684156e-05,
+      "loss": 3.8958,
+      "step": 1897984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8916240763173636e-05,
+      "loss": 3.879,
+      "step": 1898496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8907854815663116e-05,
+      "loss": 3.888,
+      "step": 1899008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8899468868152596e-05,
+      "loss": 3.8644,
+      "step": 1899520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8891082920642076e-05,
+      "loss": 3.8957,
+      "step": 1900032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8882713351935288e-05,
+      "loss": 3.8957,
+      "step": 1900544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8874343783228497e-05,
+      "loss": 3.8996,
+      "step": 1901056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.886595783571798e-05,
+      "loss": 3.8835,
+      "step": 1901568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.885757188820746e-05,
+      "loss": 3.8962,
+      "step": 1902080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.884918594069694e-05,
+      "loss": 3.8886,
+      "step": 1902592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.884079999318642e-05,
+      "loss": 3.9017,
+      "step": 1903104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.883243042447963e-05,
+      "loss": 3.8795,
+      "step": 1903616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.882404447696911e-05,
+      "loss": 3.8846,
+      "step": 1904128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.881565852945859e-05,
+      "loss": 3.886,
+      "step": 1904640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.880727258194807e-05,
+      "loss": 3.8981,
+      "step": 1905152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.879888663443755e-05,
+      "loss": 3.8872,
+      "step": 1905664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.879050068692703e-05,
+      "loss": 3.8874,
+      "step": 1906176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.878211473941651e-05,
+      "loss": 3.894,
+      "step": 1906688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.877372879190599e-05,
+      "loss": 3.8886,
+      "step": 1907200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.876534284439547e-05,
+      "loss": 3.8946,
+      "step": 1907712
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.004827499389648,
+      "eval_runtime": 504.5199,
+      "eval_samples_per_second": 756.345,
+      "eval_steps_per_second": 23.636,
+      "step": 1908000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.875695689688495e-05,
+      "loss": 3.8885,
+      "step": 1908224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.874857094937443e-05,
+      "loss": 3.8962,
+      "step": 1908736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.874018500186391e-05,
+      "loss": 3.8865,
+      "step": 1909248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.873179905435339e-05,
+      "loss": 3.8915,
+      "step": 1909760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.872341310684287e-05,
+      "loss": 3.9013,
+      "step": 1910272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.871502715933235e-05,
+      "loss": 3.8889,
+      "step": 1910784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.870664121182183e-05,
+      "loss": 3.8836,
+      "step": 1911296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.869825526431131e-05,
+      "loss": 3.8798,
+      "step": 1911808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8689869316800785e-05,
+      "loss": 3.8889,
+      "step": 1912320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.868148336929027e-05,
+      "loss": 3.8972,
+      "step": 1912832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.867309742177975e-05,
+      "loss": 3.888,
+      "step": 1913344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.866471147426923e-05,
+      "loss": 3.8919,
+      "step": 1913856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8656341905562438e-05,
+      "loss": 3.8962,
+      "step": 1914368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8647955958051918e-05,
+      "loss": 3.8845,
+      "step": 1914880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8639570010541398e-05,
+      "loss": 3.8816,
+      "step": 1915392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8631184063030878e-05,
+      "loss": 3.8781,
+      "step": 1915904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.862279811552036e-05,
+      "loss": 3.8893,
+      "step": 1916416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.861441216800984e-05,
+      "loss": 3.8857,
+      "step": 1916928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.860604259930305e-05,
+      "loss": 3.8844,
+      "step": 1917440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.859765665179253e-05,
+      "loss": 3.8987,
+      "step": 1917952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.858927070428201e-05,
+      "loss": 3.8994,
+      "step": 1918464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.858088475677149e-05,
+      "loss": 3.8892,
+      "step": 1918976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.857249880926097e-05,
+      "loss": 3.889,
+      "step": 1919488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.856411286175045e-05,
+      "loss": 3.8871,
+      "step": 1920000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.855572691423993e-05,
+      "loss": 3.8875,
+      "step": 1920512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.854734096672941e-05,
+      "loss": 3.8859,
+      "step": 1921024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.853895501921889e-05,
+      "loss": 3.8711,
+      "step": 1921536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.853056907170837e-05,
+      "loss": 3.8906,
+      "step": 1922048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8522183124197846e-05,
+      "loss": 3.8926,
+      "step": 1922560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.851379717668733e-05,
+      "loss": 3.8737,
+      "step": 1923072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.850544398678427e-05,
+      "loss": 3.8818,
+      "step": 1923584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.849705803927375e-05,
+      "loss": 3.8897,
+      "step": 1924096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.848867209176323e-05,
+      "loss": 3.8936,
+      "step": 1924608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.848028614425271e-05,
+      "loss": 3.8922,
+      "step": 1925120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.847190019674219e-05,
+      "loss": 3.8895,
+      "step": 1925632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8463530628035403e-05,
+      "loss": 3.8822,
+      "step": 1926144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8455144680524883e-05,
+      "loss": 3.8994,
+      "step": 1926656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8446758733014363e-05,
+      "loss": 3.8763,
+      "step": 1927168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8438389164307573e-05,
+      "loss": 3.8753,
+      "step": 1927680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8430003216797052e-05,
+      "loss": 3.8782,
+      "step": 1928192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8421617269286532e-05,
+      "loss": 3.884,
+      "step": 1928704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8413231321776016e-05,
+      "loss": 3.887,
+      "step": 1929216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8404845374265496e-05,
+      "loss": 3.8827,
+      "step": 1929728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8396459426754972e-05,
+      "loss": 3.8882,
+      "step": 1930240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8388073479244452e-05,
+      "loss": 3.8838,
+      "step": 1930752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8379687531733932e-05,
+      "loss": 3.8814,
+      "step": 1931264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8371301584223412e-05,
+      "loss": 3.8914,
+      "step": 1931776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8362915636712892e-05,
+      "loss": 3.8902,
+      "step": 1932288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8354529689202375e-05,
+      "loss": 3.8633,
+      "step": 1932800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8346143741691852e-05,
+      "loss": 3.8907,
+      "step": 1933312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8337774172985064e-05,
+      "loss": 3.8911,
+      "step": 1933824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8329388225474544e-05,
+      "loss": 3.8965,
+      "step": 1934336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8321002277964024e-05,
+      "loss": 3.8827,
+      "step": 1934848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.83126163304535e-05,
+      "loss": 3.8786,
+      "step": 1935360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8304246761746717e-05,
+      "loss": 3.8762,
+      "step": 1935872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8295877193039926e-05,
+      "loss": 3.8806,
+      "step": 1936384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8287491245529406e-05,
+      "loss": 3.8873,
+      "step": 1936896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8279105298018886e-05,
+      "loss": 3.8726,
+      "step": 1937408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8270719350508366e-05,
+      "loss": 3.8894,
+      "step": 1937920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8262333402997846e-05,
+      "loss": 3.8873,
+      "step": 1938432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8253947455487326e-05,
+      "loss": 3.8766,
+      "step": 1938944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8245561507976806e-05,
+      "loss": 3.8892,
+      "step": 1939456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8237191939270018e-05,
+      "loss": 3.8745,
+      "step": 1939968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8228805991759498e-05,
+      "loss": 3.8781,
+      "step": 1940480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8220420044248975e-05,
+      "loss": 3.8834,
+      "step": 1940992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8212034096738455e-05,
+      "loss": 3.895,
+      "step": 1941504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8203648149227938e-05,
+      "loss": 3.8734,
+      "step": 1942016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8195262201717418e-05,
+      "loss": 3.8873,
+      "step": 1942528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8186876254206898e-05,
+      "loss": 3.8637,
+      "step": 1943040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8178490306696378e-05,
+      "loss": 3.8861,
+      "step": 1943552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8170120737989587e-05,
+      "loss": 3.887,
+      "step": 1944064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.81617511692828e-05,
+      "loss": 3.8866,
+      "step": 1944576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.815336522177228e-05,
+      "loss": 3.8821,
+      "step": 1945088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.814497927426176e-05,
+      "loss": 3.8897,
+      "step": 1945600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.813659332675124e-05,
+      "loss": 3.9024,
+      "step": 1946112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.812822375804445e-05,
+      "loss": 3.8783,
+      "step": 1946624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.811983781053393e-05,
+      "loss": 3.891,
+      "step": 1947136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.811145186302341e-05,
+      "loss": 3.8841,
+      "step": 1947648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.810306591551289e-05,
+      "loss": 3.8837,
+      "step": 1948160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.809467996800237e-05,
+      "loss": 3.887,
+      "step": 1948672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.808629402049185e-05,
+      "loss": 3.8823,
+      "step": 1949184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.807790807298133e-05,
+      "loss": 3.8912,
+      "step": 1949696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8069522125470808e-05,
+      "loss": 3.8973,
+      "step": 1950208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.806115255676402e-05,
+      "loss": 3.8718,
+      "step": 1950720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8052782988057233e-05,
+      "loss": 3.881,
+      "step": 1951232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8044397040546713e-05,
+      "loss": 3.8796,
+      "step": 1951744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8036011093036193e-05,
+      "loss": 3.8933,
+      "step": 1952256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8027625145525673e-05,
+      "loss": 3.8783,
+      "step": 1952768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8019255576818882e-05,
+      "loss": 3.8904,
+      "step": 1953280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8010869629308362e-05,
+      "loss": 3.8847,
+      "step": 1953792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8002483681797845e-05,
+      "loss": 3.8817,
+      "step": 1954304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7994097734287325e-05,
+      "loss": 3.8784,
+      "step": 1954816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7985711786776805e-05,
+      "loss": 3.869,
+      "step": 1955328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7977325839266282e-05,
+      "loss": 3.8803,
+      "step": 1955840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7968939891755762e-05,
+      "loss": 3.8895,
+      "step": 1956352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7960553944245242e-05,
+      "loss": 3.8956,
+      "step": 1956864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7952184375538454e-05,
+      "loss": 3.8728,
+      "step": 1957376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7943798428027934e-05,
+      "loss": 3.8757,
+      "step": 1957888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7935428859321147e-05,
+      "loss": 3.8813,
+      "step": 1958400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7927042911810627e-05,
+      "loss": 3.8863,
+      "step": 1958912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7918656964300103e-05,
+      "loss": 3.888,
+      "step": 1959424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7910271016789583e-05,
+      "loss": 3.8824,
+      "step": 1959936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7901885069279063e-05,
+      "loss": 3.8875,
+      "step": 1960448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7893515500572276e-05,
+      "loss": 3.8834,
+      "step": 1960960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7885129553061756e-05,
+      "loss": 3.8833,
+      "step": 1961472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7876743605551235e-05,
+      "loss": 3.8727,
+      "step": 1961984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7868357658040715e-05,
+      "loss": 3.8769,
+      "step": 1962496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7859971710530195e-05,
+      "loss": 3.888,
+      "step": 1963008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7851585763019675e-05,
+      "loss": 3.874,
+      "step": 1963520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7843199815509155e-05,
+      "loss": 3.8797,
+      "step": 1964032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7834813867998635e-05,
+      "loss": 3.8749,
+      "step": 1964544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7826444299291848e-05,
+      "loss": 3.8853,
+      "step": 1965056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7818058351781328e-05,
+      "loss": 3.8802,
+      "step": 1965568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7809672404270808e-05,
+      "loss": 3.8835,
+      "step": 1966080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7801286456760284e-05,
+      "loss": 3.8846,
+      "step": 1966592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.77929168880535e-05,
+      "loss": 3.8965,
+      "step": 1967104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.778454731934671e-05,
+      "loss": 3.8831,
+      "step": 1967616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.777616137183619e-05,
+      "loss": 3.8867,
+      "step": 1968128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.776777542432567e-05,
+      "loss": 3.8869,
+      "step": 1968640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.775938947681515e-05,
+      "loss": 3.8783,
+      "step": 1969152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.775100352930463e-05,
+      "loss": 3.8879,
+      "step": 1969664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.774263396059784e-05,
+      "loss": 3.877,
+      "step": 1970176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.773424801308732e-05,
+      "loss": 3.8776,
+      "step": 1970688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.77258620655768e-05,
+      "loss": 3.8842,
+      "step": 1971200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.771747611806628e-05,
+      "loss": 3.8804,
+      "step": 1971712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7709090170555758e-05,
+      "loss": 3.8876,
+      "step": 1972224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7700704223045238e-05,
+      "loss": 3.8804,
+      "step": 1972736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7692334654338454e-05,
+      "loss": 3.8892,
+      "step": 1973248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.768394870682793e-05,
+      "loss": 3.8791,
+      "step": 1973760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7675579138121143e-05,
+      "loss": 3.891,
+      "step": 1974272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7667193190610623e-05,
+      "loss": 3.8773,
+      "step": 1974784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7658807243100103e-05,
+      "loss": 3.8799,
+      "step": 1975296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.765042129558958e-05,
+      "loss": 3.859,
+      "step": 1975808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7642035348079063e-05,
+      "loss": 3.8932,
+      "step": 1976320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7633665779372275e-05,
+      "loss": 3.8861,
+      "step": 1976832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7625279831861755e-05,
+      "loss": 3.8961,
+      "step": 1977344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.761689388435123e-05,
+      "loss": 3.8795,
+      "step": 1977856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.760850793684071e-05,
+      "loss": 3.8908,
+      "step": 1978368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.760012198933019e-05,
+      "loss": 3.8835,
+      "step": 1978880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.759173604181967e-05,
+      "loss": 3.9001,
+      "step": 1979392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7583350094309155e-05,
+      "loss": 3.8745,
+      "step": 1979904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7574964146798635e-05,
+      "loss": 3.88,
+      "step": 1980416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.756657819928811e-05,
+      "loss": 3.8828,
+      "step": 1980928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.755819225177759e-05,
+      "loss": 3.8908,
+      "step": 1981440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.754980630426707e-05,
+      "loss": 3.8829,
+      "step": 1981952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.754142035675655e-05,
+      "loss": 3.8816,
+      "step": 1982464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7533067166853496e-05,
+      "loss": 3.8867,
+      "step": 1982976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7524681219342976e-05,
+      "loss": 3.8853,
+      "step": 1983488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7516295271832456e-05,
+      "loss": 3.8892,
+      "step": 1984000
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.002923965454102,
+      "eval_runtime": 293.4422,
+      "eval_samples_per_second": 1300.396,
+      "eval_steps_per_second": 40.638,
+      "step": 1984320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7507909324321936e-05,
+      "loss": 3.8765,
+      "step": 1984512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7499523376811413e-05,
+      "loss": 3.8897,
+      "step": 1985024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7491137429300893e-05,
+      "loss": 3.8852,
+      "step": 1985536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7482751481790373e-05,
+      "loss": 3.8845,
+      "step": 1986048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7474365534279856e-05,
+      "loss": 3.8939,
+      "step": 1986560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7465979586769336e-05,
+      "loss": 3.8815,
+      "step": 1987072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7457593639258816e-05,
+      "loss": 3.8825,
+      "step": 1987584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7449207691748292e-05,
+      "loss": 3.8743,
+      "step": 1988096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7440821744237772e-05,
+      "loss": 3.8808,
+      "step": 1988608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7432435796727252e-05,
+      "loss": 3.8901,
+      "step": 1989120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7424049849216732e-05,
+      "loss": 3.8874,
+      "step": 1989632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7415663901706216e-05,
+      "loss": 3.8863,
+      "step": 1990144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7407294332999425e-05,
+      "loss": 3.8921,
+      "step": 1990656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7398908385488905e-05,
+      "loss": 3.8792,
+      "step": 1991168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7390522437978385e-05,
+      "loss": 3.8776,
+      "step": 1991680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7382136490467865e-05,
+      "loss": 3.871,
+      "step": 1992192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7373750542957345e-05,
+      "loss": 3.8873,
+      "step": 1992704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7365364595446824e-05,
+      "loss": 3.8796,
+      "step": 1993216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7356978647936304e-05,
+      "loss": 3.8801,
+      "step": 1993728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7348592700425784e-05,
+      "loss": 3.8917,
+      "step": 1994240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7340206752915264e-05,
+      "loss": 3.8974,
+      "step": 1994752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7331820805404744e-05,
+      "loss": 3.885,
+      "step": 1995264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7323434857894224e-05,
+      "loss": 3.8818,
+      "step": 1995776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7315048910383704e-05,
+      "loss": 3.8838,
+      "step": 1996288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7306679341676917e-05,
+      "loss": 3.8811,
+      "step": 1996800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7298293394166397e-05,
+      "loss": 3.8819,
+      "step": 1997312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7289907446655877e-05,
+      "loss": 3.8641,
+      "step": 1997824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7281521499145357e-05,
+      "loss": 3.8849,
+      "step": 1998336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7273151930438566e-05,
+      "loss": 3.8873,
+      "step": 1998848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7264765982928046e-05,
+      "loss": 3.8693,
+      "step": 1999360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7256380035417526e-05,
+      "loss": 3.8768,
+      "step": 1999872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7247994087907006e-05,
+      "loss": 3.8868,
+      "step": 2000384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7239608140396485e-05,
+      "loss": 3.8849,
+      "step": 2000896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7231238571689698e-05,
+      "loss": 3.89,
+      "step": 2001408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7222852624179178e-05,
+      "loss": 3.8841,
+      "step": 2001920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7214483055472387e-05,
+      "loss": 3.8786,
+      "step": 2002432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.720609710796187e-05,
+      "loss": 3.8995,
+      "step": 2002944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.719771116045135e-05,
+      "loss": 3.8668,
+      "step": 2003456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.718932521294083e-05,
+      "loss": 3.8699,
+      "step": 2003968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7180939265430307e-05,
+      "loss": 3.8778,
+      "step": 2004480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7172553317919787e-05,
+      "loss": 3.8757,
+      "step": 2004992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7164183749213e-05,
+      "loss": 3.88,
+      "step": 2005504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.715579780170248e-05,
+      "loss": 3.8824,
+      "step": 2006016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.714741185419196e-05,
+      "loss": 3.8825,
+      "step": 2006528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.713902590668144e-05,
+      "loss": 3.8774,
+      "step": 2007040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.713063995917092e-05,
+      "loss": 3.8752,
+      "step": 2007552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.71222540116604e-05,
+      "loss": 3.8867,
+      "step": 2008064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.711386806414988e-05,
+      "loss": 3.8828,
+      "step": 2008576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.710548211663936e-05,
+      "loss": 3.8638,
+      "step": 2009088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.709709616912884e-05,
+      "loss": 3.886,
+      "step": 2009600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.708871022161832e-05,
+      "loss": 3.8853,
+      "step": 2010112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.70803242741078e-05,
+      "loss": 3.8919,
+      "step": 2010624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.707195470540101e-05,
+      "loss": 3.8758,
+      "step": 2011136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7063568757890488e-05,
+      "loss": 3.877,
+      "step": 2011648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7055182810379968e-05,
+      "loss": 3.8695,
+      "step": 2012160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7046796862869448e-05,
+      "loss": 3.8723,
+      "step": 2012672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7038410915358928e-05,
+      "loss": 3.8854,
+      "step": 2013184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.703004134665214e-05,
+      "loss": 3.8677,
+      "step": 2013696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.702165539914162e-05,
+      "loss": 3.8834,
+      "step": 2014208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.70132694516311e-05,
+      "loss": 3.8797,
+      "step": 2014720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.700488350412058e-05,
+      "loss": 3.876,
+      "step": 2015232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.699649755661006e-05,
+      "loss": 3.8829,
+      "step": 2015744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.698811160909954e-05,
+      "loss": 3.866,
+      "step": 2016256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.697972566158902e-05,
+      "loss": 3.8751,
+      "step": 2016768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.69713397140785e-05,
+      "loss": 3.8754,
+      "step": 2017280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6962970145371712e-05,
+      "loss": 3.8887,
+      "step": 2017792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.695460057666492e-05,
+      "loss": 3.8675,
+      "step": 2018304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.69462146291544e-05,
+      "loss": 3.882,
+      "step": 2018816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.693782868164388e-05,
+      "loss": 3.8589,
+      "step": 2019328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6929442734133365e-05,
+      "loss": 3.8793,
+      "step": 2019840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.692105678662284e-05,
+      "loss": 3.8819,
+      "step": 2020352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6912687217916054e-05,
+      "loss": 3.885,
+      "step": 2020864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6904301270405534e-05,
+      "loss": 3.8758,
+      "step": 2021376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6895915322895014e-05,
+      "loss": 3.8813,
+      "step": 2021888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6887529375384494e-05,
+      "loss": 3.8984,
+      "step": 2022400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6879143427873974e-05,
+      "loss": 3.8716,
+      "step": 2022912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6870773859167186e-05,
+      "loss": 3.8899,
+      "step": 2023424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6862387911656666e-05,
+      "loss": 3.8763,
+      "step": 2023936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6854018342949875e-05,
+      "loss": 3.8772,
+      "step": 2024448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6845632395439355e-05,
+      "loss": 3.8811,
+      "step": 2024960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6837246447928835e-05,
+      "loss": 3.8746,
+      "step": 2025472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6828860500418315e-05,
+      "loss": 3.8891,
+      "step": 2025984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6820474552907795e-05,
+      "loss": 3.8915,
+      "step": 2026496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6812088605397275e-05,
+      "loss": 3.8735,
+      "step": 2027008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6803702657886755e-05,
+      "loss": 3.8708,
+      "step": 2027520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6795316710376235e-05,
+      "loss": 3.8761,
+      "step": 2028032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6786947141669447e-05,
+      "loss": 3.8861,
+      "step": 2028544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6778561194158927e-05,
+      "loss": 3.8736,
+      "step": 2029056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6770175246648407e-05,
+      "loss": 3.8862,
+      "step": 2029568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6761789299137887e-05,
+      "loss": 3.8794,
+      "step": 2030080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6753403351627367e-05,
+      "loss": 3.8779,
+      "step": 2030592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6745017404116844e-05,
+      "loss": 3.8736,
+      "step": 2031104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6736631456606324e-05,
+      "loss": 3.8657,
+      "step": 2031616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6728245509095804e-05,
+      "loss": 3.8731,
+      "step": 2032128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.671987594038902e-05,
+      "loss": 3.8839,
+      "step": 2032640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6711489992878496e-05,
+      "loss": 3.8893,
+      "step": 2033152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6703104045367976e-05,
+      "loss": 3.8699,
+      "step": 2033664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6694718097857456e-05,
+      "loss": 3.8702,
+      "step": 2034176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.668634852915067e-05,
+      "loss": 3.8771,
+      "step": 2034688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.667796258164015e-05,
+      "loss": 3.8837,
+      "step": 2035200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.666957663412963e-05,
+      "loss": 3.8864,
+      "step": 2035712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.666119068661911e-05,
+      "loss": 3.8744,
+      "step": 2036224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.665280473910859e-05,
+      "loss": 3.8828,
+      "step": 2036736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6644435170401797e-05,
+      "loss": 3.8764,
+      "step": 2037248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6636049222891277e-05,
+      "loss": 3.8759,
+      "step": 2037760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6627663275380757e-05,
+      "loss": 3.8694,
+      "step": 2038272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.661927732787024e-05,
+      "loss": 3.8756,
+      "step": 2038784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.661089138035972e-05,
+      "loss": 3.8799,
+      "step": 2039296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.660252181165293e-05,
+      "loss": 3.8721,
+      "step": 2039808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.659413586414241e-05,
+      "loss": 3.8727,
+      "step": 2040320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.658574991663189e-05,
+      "loss": 3.8697,
+      "step": 2040832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.657736396912137e-05,
+      "loss": 3.876,
+      "step": 2041344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.656897802161085e-05,
+      "loss": 3.8765,
+      "step": 2041856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.656059207410033e-05,
+      "loss": 3.8759,
+      "step": 2042368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.655220612658981e-05,
+      "loss": 3.8798,
+      "step": 2042880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.654382017907929e-05,
+      "loss": 3.8929,
+      "step": 2043392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.65354506103725e-05,
+      "loss": 3.8799,
+      "step": 2043904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.652708104166571e-05,
+      "loss": 3.881,
+      "step": 2044416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6518695094155194e-05,
+      "loss": 3.881,
+      "step": 2044928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6510309146644674e-05,
+      "loss": 3.8749,
+      "step": 2045440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.650192319913415e-05,
+      "loss": 3.8863,
+      "step": 2045952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6493570009231092e-05,
+      "loss": 3.8689,
+      "step": 2046464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6485184061720572e-05,
+      "loss": 3.8735,
+      "step": 2046976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6476798114210056e-05,
+      "loss": 3.8807,
+      "step": 2047488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6468412166699536e-05,
+      "loss": 3.8782,
+      "step": 2048000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6460026219189016e-05,
+      "loss": 3.877,
+      "step": 2048512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6451640271678496e-05,
+      "loss": 3.8776,
+      "step": 2049024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6443254324167972e-05,
+      "loss": 3.8808,
+      "step": 2049536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6434868376657452e-05,
+      "loss": 3.8769,
+      "step": 2050048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6426498807950665e-05,
+      "loss": 3.8818,
+      "step": 2050560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6418112860440145e-05,
+      "loss": 3.8759,
+      "step": 2051072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6409726912929625e-05,
+      "loss": 3.8721,
+      "step": 2051584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6401340965419105e-05,
+      "loss": 3.8551,
+      "step": 2052096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6392955017908584e-05,
+      "loss": 3.8885,
+      "step": 2052608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6384585449201797e-05,
+      "loss": 3.8787,
+      "step": 2053120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6376199501691274e-05,
+      "loss": 3.8916,
+      "step": 2053632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6367813554180757e-05,
+      "loss": 3.8745,
+      "step": 2054144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6359427606670237e-05,
+      "loss": 3.8819,
+      "step": 2054656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6351041659159717e-05,
+      "loss": 3.8787,
+      "step": 2055168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6342655711649197e-05,
+      "loss": 3.8948,
+      "step": 2055680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6334269764138677e-05,
+      "loss": 3.8672,
+      "step": 2056192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6325883816628153e-05,
+      "loss": 3.8753,
+      "step": 2056704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6317514247921366e-05,
+      "loss": 3.8791,
+      "step": 2057216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6309144679214578e-05,
+      "loss": 3.8842,
+      "step": 2057728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.630077511050779e-05,
+      "loss": 3.8771,
+      "step": 2058240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.629238916299727e-05,
+      "loss": 3.8775,
+      "step": 2058752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6284003215486747e-05,
+      "loss": 3.8779,
+      "step": 2059264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6275617267976227e-05,
+      "loss": 3.8858,
+      "step": 2059776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.626723132046571e-05,
+      "loss": 3.8824,
+      "step": 2060288
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.001179218292236,
+      "eval_runtime": 303.4267,
+      "eval_samples_per_second": 1257.605,
+      "eval_steps_per_second": 39.301,
+      "step": 2060640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.625884537295519e-05,
+      "loss": 3.8832,
+      "step": 2060800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.625045942544467e-05,
+      "loss": 3.884,
+      "step": 2061312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.624207347793415e-05,
+      "loss": 3.8787,
+      "step": 2061824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6233687530423627e-05,
+      "loss": 3.88,
+      "step": 2062336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6225301582913107e-05,
+      "loss": 3.8859,
+      "step": 2062848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6216915635402587e-05,
+      "loss": 3.8784,
+      "step": 2063360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6208529687892067e-05,
+      "loss": 3.8818,
+      "step": 2063872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.620014374038155e-05,
+      "loss": 3.8705,
+      "step": 2064384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.619177417167476e-05,
+      "loss": 3.8745,
+      "step": 2064896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.618338822416424e-05,
+      "loss": 3.885,
+      "step": 2065408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.617500227665372e-05,
+      "loss": 3.8849,
+      "step": 2065920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.61666163291432e-05,
+      "loss": 3.8802,
+      "step": 2066432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.615824676043641e-05,
+      "loss": 3.8931,
+      "step": 2066944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.614986081292589e-05,
+      "loss": 3.8686,
+      "step": 2067456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.614147486541537e-05,
+      "loss": 3.8728,
+      "step": 2067968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.613308891790485e-05,
+      "loss": 3.8676,
+      "step": 2068480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.612470297039433e-05,
+      "loss": 3.881,
+      "step": 2068992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6116317022883808e-05,
+      "loss": 3.8737,
+      "step": 2069504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6107931075373288e-05,
+      "loss": 3.8778,
+      "step": 2070016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6099545127862768e-05,
+      "loss": 3.8841,
+      "step": 2070528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.609115918035225e-05,
+      "loss": 3.8913,
+      "step": 2071040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.608277323284173e-05,
+      "loss": 3.8808,
+      "step": 2071552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.607438728533121e-05,
+      "loss": 3.8758,
+      "step": 2072064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.606600133782069e-05,
+      "loss": 3.8791,
+      "step": 2072576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6057615390310168e-05,
+      "loss": 3.8787,
+      "step": 2073088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6049229442799648e-05,
+      "loss": 3.8746,
+      "step": 2073600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.604085987409286e-05,
+      "loss": 3.8626,
+      "step": 2074112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.603247392658234e-05,
+      "loss": 3.8756,
+      "step": 2074624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.602408797907182e-05,
+      "loss": 3.8813,
+      "step": 2075136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.60157020315613e-05,
+      "loss": 3.863,
+      "step": 2075648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.600731608405078e-05,
+      "loss": 3.8734,
+      "step": 2076160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.599894651534399e-05,
+      "loss": 3.8809,
+      "step": 2076672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5990560567833472e-05,
+      "loss": 3.8775,
+      "step": 2077184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5982174620322952e-05,
+      "loss": 3.8872,
+      "step": 2077696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5973788672812432e-05,
+      "loss": 3.8798,
+      "step": 2078208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.596541910410564e-05,
+      "loss": 3.8746,
+      "step": 2078720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.595703315659512e-05,
+      "loss": 3.8944,
+      "step": 2079232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.59486472090846e-05,
+      "loss": 3.8594,
+      "step": 2079744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.594026126157408e-05,
+      "loss": 3.8665,
+      "step": 2080256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5931891692867294e-05,
+      "loss": 3.8742,
+      "step": 2080768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5923505745356774e-05,
+      "loss": 3.8704,
+      "step": 2081280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5915119797846254e-05,
+      "loss": 3.8735,
+      "step": 2081792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5906733850335734e-05,
+      "loss": 3.8818,
+      "step": 2082304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5898347902825214e-05,
+      "loss": 3.8755,
+      "step": 2082816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5889961955314694e-05,
+      "loss": 3.8754,
+      "step": 2083328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5881576007804174e-05,
+      "loss": 3.8699,
+      "step": 2083840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5873206439097386e-05,
+      "loss": 3.8777,
+      "step": 2084352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5864820491586866e-05,
+      "loss": 3.8807,
+      "step": 2084864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5856434544076346e-05,
+      "loss": 3.8575,
+      "step": 2085376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5848048596565822e-05,
+      "loss": 3.8795,
+      "step": 2085888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5839662649055302e-05,
+      "loss": 3.8805,
+      "step": 2086400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5831276701544782e-05,
+      "loss": 3.8844,
+      "step": 2086912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5822890754034266e-05,
+      "loss": 3.8705,
+      "step": 2087424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5814504806523746e-05,
+      "loss": 3.8704,
+      "step": 2087936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5806135237816955e-05,
+      "loss": 3.8707,
+      "step": 2088448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5797749290306435e-05,
+      "loss": 3.8647,
+      "step": 2088960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5789363342795915e-05,
+      "loss": 3.8822,
+      "step": 2089472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5780977395285395e-05,
+      "loss": 3.8631,
+      "step": 2089984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5772591447774875e-05,
+      "loss": 3.8794,
+      "step": 2090496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5764205500264355e-05,
+      "loss": 3.8745,
+      "step": 2091008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5755819552753835e-05,
+      "loss": 3.8722,
+      "step": 2091520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5747433605243314e-05,
+      "loss": 3.8723,
+      "step": 2092032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5739064036536527e-05,
+      "loss": 3.8673,
+      "step": 2092544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5730678089026004e-05,
+      "loss": 3.871,
+      "step": 2093056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5722292141515483e-05,
+      "loss": 3.8632,
+      "step": 2093568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5713906194004967e-05,
+      "loss": 3.8898,
+      "step": 2094080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5705536625298176e-05,
+      "loss": 3.8697,
+      "step": 2094592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5697150677787656e-05,
+      "loss": 3.8695,
+      "step": 2095104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5688764730277136e-05,
+      "loss": 3.8581,
+      "step": 2095616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5680378782766616e-05,
+      "loss": 3.8727,
+      "step": 2096128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5671992835256096e-05,
+      "loss": 3.8759,
+      "step": 2096640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5663623266549308e-05,
+      "loss": 3.8801,
+      "step": 2097152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.565525369784252e-05,
+      "loss": 3.8727,
+      "step": 2097664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5646867750332e-05,
+      "loss": 3.8799,
+      "step": 2098176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5638481802821477e-05,
+      "loss": 3.8893,
+      "step": 2098688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5630095855310957e-05,
+      "loss": 3.8717,
+      "step": 2099200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5621709907800437e-05,
+      "loss": 3.8855,
+      "step": 2099712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.561332396028992e-05,
+      "loss": 3.8718,
+      "step": 2100224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.56049380127794e-05,
+      "loss": 3.873,
+      "step": 2100736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.559655206526888e-05,
+      "loss": 3.8772,
+      "step": 2101248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.558818249656209e-05,
+      "loss": 3.8691,
+      "step": 2101760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.557979654905157e-05,
+      "loss": 3.8823,
+      "step": 2102272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5571426980344782e-05,
+      "loss": 3.8909,
+      "step": 2102784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5563041032834262e-05,
+      "loss": 3.8743,
+      "step": 2103296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5554655085323742e-05,
+      "loss": 3.8618,
+      "step": 2103808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5546269137813222e-05,
+      "loss": 3.8695,
+      "step": 2104320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5537883190302702e-05,
+      "loss": 3.8796,
+      "step": 2104832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.552949724279218e-05,
+      "loss": 3.8716,
+      "step": 2105344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5521111295281658e-05,
+      "loss": 3.8799,
+      "step": 2105856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.5512725347771138e-05,
+      "loss": 3.8753,
+      "step": 2106368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.550433940026062e-05,
+      "loss": 3.8754,
+      "step": 2106880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.54959534527501e-05,
+      "loss": 3.8692,
+      "step": 2107392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.548756750523958e-05,
+      "loss": 3.8592,
+      "step": 2107904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.547918155772906e-05,
+      "loss": 3.8693,
+      "step": 2108416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.547081198902227e-05,
+      "loss": 3.8736,
+      "step": 2108928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.546242604151175e-05,
+      "loss": 3.891,
+      "step": 2109440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.545404009400123e-05,
+      "loss": 3.8627,
+      "step": 2109952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5445670525294443e-05,
+      "loss": 3.8678,
+      "step": 2110464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5437284577783923e-05,
+      "loss": 3.8736,
+      "step": 2110976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5428898630273403e-05,
+      "loss": 3.8741,
+      "step": 2111488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5420512682762883e-05,
+      "loss": 3.8809,
+      "step": 2112000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.541212673525236e-05,
+      "loss": 3.8689,
+      "step": 2112512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5403740787741843e-05,
+      "loss": 3.8814,
+      "step": 2113024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5395354840231323e-05,
+      "loss": 3.8702,
+      "step": 2113536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5386968892720803e-05,
+      "loss": 3.8699,
+      "step": 2114048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5378599324014012e-05,
+      "loss": 3.8679,
+      "step": 2114560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.537021337650349e-05,
+      "loss": 3.8729,
+      "step": 2115072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.536182742899297e-05,
+      "loss": 3.8755,
+      "step": 2115584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5353457860286184e-05,
+      "loss": 3.8686,
+      "step": 2116096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5345071912775664e-05,
+      "loss": 3.8685,
+      "step": 2116608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5336685965265144e-05,
+      "loss": 3.8639,
+      "step": 2117120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5328300017754624e-05,
+      "loss": 3.8706,
+      "step": 2117632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5319914070244104e-05,
+      "loss": 3.8739,
+      "step": 2118144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5311528122733584e-05,
+      "loss": 3.8737,
+      "step": 2118656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5303142175223064e-05,
+      "loss": 3.8752,
+      "step": 2119168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5294756227712544e-05,
+      "loss": 3.879,
+      "step": 2119680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5286370280202024e-05,
+      "loss": 3.8786,
+      "step": 2120192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5278017090298965e-05,
+      "loss": 3.88,
+      "step": 2120704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5269631142788445e-05,
+      "loss": 3.8776,
+      "step": 2121216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5261245195277925e-05,
+      "loss": 3.87,
+      "step": 2121728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5252859247767407e-05,
+      "loss": 3.8763,
+      "step": 2122240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5244489679060616e-05,
+      "loss": 3.867,
+      "step": 2122752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5236103731550098e-05,
+      "loss": 3.8702,
+      "step": 2123264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5227717784039578e-05,
+      "loss": 3.8762,
+      "step": 2123776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5219348215332788e-05,
+      "loss": 3.8746,
+      "step": 2124288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5210962267822268e-05,
+      "loss": 3.8736,
+      "step": 2124800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5202576320311748e-05,
+      "loss": 3.8715,
+      "step": 2125312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5194190372801228e-05,
+      "loss": 3.8761,
+      "step": 2125824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5185804425290708e-05,
+      "loss": 3.8722,
+      "step": 2126336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.517741847778019e-05,
+      "loss": 3.8784,
+      "step": 2126848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5169032530269666e-05,
+      "loss": 3.8732,
+      "step": 2127360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5160646582759146e-05,
+      "loss": 3.8702,
+      "step": 2127872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5152260635248628e-05,
+      "loss": 3.8464,
+      "step": 2128384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.514389106654184e-05,
+      "loss": 3.8828,
+      "step": 2128896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5135505119031317e-05,
+      "loss": 3.8738,
+      "step": 2129408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5127119171520799e-05,
+      "loss": 3.8877,
+      "step": 2129920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5118733224010279e-05,
+      "loss": 3.8733,
+      "step": 2130432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5110347276499759e-05,
+      "loss": 3.8751,
+      "step": 2130944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.510197770779297e-05,
+      "loss": 3.8779,
+      "step": 2131456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.509359176028245e-05,
+      "loss": 3.8868,
+      "step": 2131968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.508520581277193e-05,
+      "loss": 3.8623,
+      "step": 2132480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.507681986526141e-05,
+      "loss": 3.8753,
+      "step": 2132992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5068433917750891e-05,
+      "loss": 3.8707,
+      "step": 2133504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5060047970240371e-05,
+      "loss": 3.8786,
+      "step": 2134016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5051678401533582e-05,
+      "loss": 3.8762,
+      "step": 2134528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5043292454023062e-05,
+      "loss": 3.8732,
+      "step": 2135040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5034906506512542e-05,
+      "loss": 3.873,
+      "step": 2135552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5026520559002022e-05,
+      "loss": 3.8803,
+      "step": 2136064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.50181346114915e-05,
+      "loss": 3.8758,
+      "step": 2136576
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.999927520751953,
+      "eval_runtime": 294.816,
+      "eval_samples_per_second": 1294.336,
+      "eval_steps_per_second": 40.449,
+      "step": 2136960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.500974866398098e-05,
+      "loss": 3.8842,
+      "step": 2137088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.500136271647046e-05,
+      "loss": 3.8783,
+      "step": 2137600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.499297676895994e-05,
+      "loss": 3.8722,
+      "step": 2138112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4984590821449421e-05,
+      "loss": 3.8776,
+      "step": 2138624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4976204873938901e-05,
+      "loss": 3.8777,
+      "step": 2139136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4967818926428378e-05,
+      "loss": 3.8792,
+      "step": 2139648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.495943297891786e-05,
+      "loss": 3.8745,
+      "step": 2140160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.495104703140734e-05,
+      "loss": 3.8664,
+      "step": 2140672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.494266108389682e-05,
+      "loss": 3.8678,
+      "step": 2141184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.49342751363863e-05,
+      "loss": 3.8864,
+      "step": 2141696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.492588918887578e-05,
+      "loss": 3.8763,
+      "step": 2142208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4917503241365261e-05,
+      "loss": 3.8722,
+      "step": 2142720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.490913367265847e-05,
+      "loss": 3.8916,
+      "step": 2143232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4900747725147952e-05,
+      "loss": 3.8638,
+      "step": 2143744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4892361777637432e-05,
+      "loss": 3.8682,
+      "step": 2144256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4883975830126908e-05,
+      "loss": 3.863,
+      "step": 2144768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.487558988261639e-05,
+      "loss": 3.8768,
+      "step": 2145280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.486720393510587e-05,
+      "loss": 3.87,
+      "step": 2145792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.485881798759535e-05,
+      "loss": 3.8732,
+      "step": 2146304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.485043204008483e-05,
+      "loss": 3.8792,
+      "step": 2146816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.484206247137804e-05,
+      "loss": 3.8823,
+      "step": 2147328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.483367652386752e-05,
+      "loss": 3.8805,
+      "step": 2147840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4825290576357e-05,
+      "loss": 3.8678,
+      "step": 2148352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.481690462884648e-05,
+      "loss": 3.8732,
+      "step": 2148864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4808518681335962e-05,
+      "loss": 3.8753,
+      "step": 2149376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4800132733825442e-05,
+      "loss": 3.87,
+      "step": 2149888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4791746786314919e-05,
+      "loss": 3.8603,
+      "step": 2150400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.47833608388044e-05,
+      "loss": 3.8682,
+      "step": 2150912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.477497489129388e-05,
+      "loss": 3.8797,
+      "step": 2151424
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.476658894378336e-05,
+      "loss": 3.859,
+      "step": 2151936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.475820299627284e-05,
+      "loss": 3.8704,
+      "step": 2152448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4749833427566051e-05,
+      "loss": 3.8724,
+      "step": 2152960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4741447480055531e-05,
+      "loss": 3.8784,
+      "step": 2153472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4733061532545011e-05,
+      "loss": 3.8786,
+      "step": 2153984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4724675585034493e-05,
+      "loss": 3.8742,
+      "step": 2154496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4716306016327702e-05,
+      "loss": 3.876,
+      "step": 2155008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4707920068817183e-05,
+      "loss": 3.8872,
+      "step": 2155520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4699534121306663e-05,
+      "loss": 3.8539,
+      "step": 2156032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4691148173796143e-05,
+      "loss": 3.8639,
+      "step": 2156544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4682762226285623e-05,
+      "loss": 3.8721,
+      "step": 2157056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4674376278775101e-05,
+      "loss": 3.8607,
+      "step": 2157568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4665990331264581e-05,
+      "loss": 3.8729,
+      "step": 2158080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4657604383754061e-05,
+      "loss": 3.877,
+      "step": 2158592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4649234815047272e-05,
+      "loss": 3.8698,
+      "step": 2159104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4640848867536752e-05,
+      "loss": 3.8688,
+      "step": 2159616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4632462920026232e-05,
+      "loss": 3.8672,
+      "step": 2160128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4624076972515714e-05,
+      "loss": 3.8709,
+      "step": 2160640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4615707403808923e-05,
+      "loss": 3.8821,
+      "step": 2161152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4607337835102137e-05,
+      "loss": 3.8497,
+      "step": 2161664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4598951887591617e-05,
+      "loss": 3.8694,
+      "step": 2162176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4590582318884826e-05,
+      "loss": 3.8796,
+      "step": 2162688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4582196371374308e-05,
+      "loss": 3.8794,
+      "step": 2163200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4573810423863788e-05,
+      "loss": 3.8698,
+      "step": 2163712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4565424476353268e-05,
+      "loss": 3.8642,
+      "step": 2164224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4557038528842746e-05,
+      "loss": 3.8674,
+      "step": 2164736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4548652581332226e-05,
+      "loss": 3.8599,
+      "step": 2165248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4540266633821706e-05,
+      "loss": 3.8759,
+      "step": 2165760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4531880686311186e-05,
+      "loss": 3.8581,
+      "step": 2166272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4523494738800666e-05,
+      "loss": 3.879,
+      "step": 2166784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4515108791290147e-05,
+      "loss": 3.8677,
+      "step": 2167296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4506722843779627e-05,
+      "loss": 3.8682,
+      "step": 2167808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4498353275072838e-05,
+      "loss": 3.8653,
+      "step": 2168320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4489967327562318e-05,
+      "loss": 3.8666,
+      "step": 2168832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4481581380051798e-05,
+      "loss": 3.8652,
+      "step": 2169344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4473195432541278e-05,
+      "loss": 3.8622,
+      "step": 2169856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4464809485030756e-05,
+      "loss": 3.8812,
+      "step": 2170368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4456423537520236e-05,
+      "loss": 3.8718,
+      "step": 2170880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4448037590009716e-05,
+      "loss": 3.8618,
+      "step": 2171392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4439651642499196e-05,
+      "loss": 3.8536,
+      "step": 2171904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4431282073792407e-05,
+      "loss": 3.8715,
+      "step": 2172416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.442291250508562e-05,
+      "loss": 3.8684,
+      "step": 2172928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4414526557575101e-05,
+      "loss": 3.8792,
+      "step": 2173440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4406140610064578e-05,
+      "loss": 3.8663,
+      "step": 2173952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4397754662554057e-05,
+      "loss": 3.8763,
+      "step": 2174464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4389385093847272e-05,
+      "loss": 3.8828,
+      "step": 2174976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4380999146336748e-05,
+      "loss": 3.8698,
+      "step": 2175488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.437261319882623e-05,
+      "loss": 3.877,
+      "step": 2176000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.436422725131571e-05,
+      "loss": 3.8693,
+      "step": 2176512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4355857682608922e-05,
+      "loss": 3.8682,
+      "step": 2177024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.43474717350984e-05,
+      "loss": 3.874,
+      "step": 2177536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.433908578758788e-05,
+      "loss": 3.8672,
+      "step": 2178048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.433069984007736e-05,
+      "loss": 3.8723,
+      "step": 2178560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.432231389256684e-05,
+      "loss": 3.8865,
+      "step": 2179072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4313927945056322e-05,
+      "loss": 3.872,
+      "step": 2179584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4305541997545802e-05,
+      "loss": 3.856,
+      "step": 2180096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4297172428839011e-05,
+      "loss": 3.8666,
+      "step": 2180608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4288786481328493e-05,
+      "loss": 3.8735,
+      "step": 2181120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4280400533817973e-05,
+      "loss": 3.8745,
+      "step": 2181632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4272014586307453e-05,
+      "loss": 3.8708,
+      "step": 2182144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4263628638796931e-05,
+      "loss": 3.8736,
+      "step": 2182656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4255259070090143e-05,
+      "loss": 3.8704,
+      "step": 2183168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4246873122579623e-05,
+      "loss": 3.8639,
+      "step": 2183680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4238487175069103e-05,
+      "loss": 3.8529,
+      "step": 2184192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4230101227558582e-05,
+      "loss": 3.8649,
+      "step": 2184704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4221715280048062e-05,
+      "loss": 3.8683,
+      "step": 2185216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4213329332537542e-05,
+      "loss": 3.8885,
+      "step": 2185728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4204943385027023e-05,
+      "loss": 3.8572,
+      "step": 2186240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4196557437516503e-05,
+      "loss": 3.862,
+      "step": 2186752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4188187868809712e-05,
+      "loss": 3.8689,
+      "step": 2187264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4179801921299194e-05,
+      "loss": 3.8706,
+      "step": 2187776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4171415973788674e-05,
+      "loss": 3.8778,
+      "step": 2188288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4163046405081885e-05,
+      "loss": 3.8657,
+      "step": 2188800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4154660457571365e-05,
+      "loss": 3.8694,
+      "step": 2189312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4146274510060845e-05,
+      "loss": 3.868,
+      "step": 2189824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4137888562550325e-05,
+      "loss": 3.8647,
+      "step": 2190336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4129502615039804e-05,
+      "loss": 3.8644,
+      "step": 2190848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4121116667529286e-05,
+      "loss": 3.8698,
+      "step": 2191360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4112730720018763e-05,
+      "loss": 3.871,
+      "step": 2191872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4104344772508243e-05,
+      "loss": 3.8653,
+      "step": 2192384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4095975203801457e-05,
+      "loss": 3.8613,
+      "step": 2192896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4087589256290937e-05,
+      "loss": 3.8609,
+      "step": 2193408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4079203308780415e-05,
+      "loss": 3.8671,
+      "step": 2193920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4070817361269895e-05,
+      "loss": 3.8686,
+      "step": 2194432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4062431413759375e-05,
+      "loss": 3.8668,
+      "step": 2194944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4054078223856318e-05,
+      "loss": 3.8697,
+      "step": 2195456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4045692276345798e-05,
+      "loss": 3.8743,
+      "step": 2195968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4037306328835278e-05,
+      "loss": 3.8794,
+      "step": 2196480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4028920381324758e-05,
+      "loss": 3.8721,
+      "step": 2196992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4020534433814236e-05,
+      "loss": 3.8739,
+      "step": 2197504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4012148486303716e-05,
+      "loss": 3.8702,
+      "step": 2198016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4003762538793196e-05,
+      "loss": 3.8692,
+      "step": 2198528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.399540934889014e-05,
+      "loss": 3.8634,
+      "step": 2199040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.398702340137962e-05,
+      "loss": 3.8669,
+      "step": 2199552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3978637453869101e-05,
+      "loss": 3.87,
+      "step": 2200064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3970251506358581e-05,
+      "loss": 3.8691,
+      "step": 2200576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3961865558848058e-05,
+      "loss": 3.8699,
+      "step": 2201088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.395347961133754e-05,
+      "loss": 3.8674,
+      "step": 2201600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.394509366382702e-05,
+      "loss": 3.8699,
+      "step": 2202112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.39367077163165e-05,
+      "loss": 3.8663,
+      "step": 2202624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.392832176880598e-05,
+      "loss": 3.8759,
+      "step": 2203136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.391995220009919e-05,
+      "loss": 3.8658,
+      "step": 2203648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.391156625258867e-05,
+      "loss": 3.8673,
+      "step": 2204160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3903196683881884e-05,
+      "loss": 3.8443,
+      "step": 2204672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.389481073637136e-05,
+      "loss": 3.8728,
+      "step": 2205184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.388642478886084e-05,
+      "loss": 3.8757,
+      "step": 2205696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.387803884135032e-05,
+      "loss": 3.8819,
+      "step": 2206208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3869652893839802e-05,
+      "loss": 3.87,
+      "step": 2206720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3861266946329282e-05,
+      "loss": 3.8667,
+      "step": 2207232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3852880998818762e-05,
+      "loss": 3.8746,
+      "step": 2207744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.384449505130824e-05,
+      "loss": 3.8839,
+      "step": 2208256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3836125482601453e-05,
+      "loss": 3.8585,
+      "step": 2208768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3827739535090933e-05,
+      "loss": 3.8726,
+      "step": 2209280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3819353587580413e-05,
+      "loss": 3.8679,
+      "step": 2209792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3810967640069891e-05,
+      "loss": 3.8759,
+      "step": 2210304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3802581692559371e-05,
+      "loss": 3.8711,
+      "step": 2210816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3794212123852585e-05,
+      "loss": 3.869,
+      "step": 2211328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3785826176342062e-05,
+      "loss": 3.8678,
+      "step": 2211840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3777440228831542e-05,
+      "loss": 3.8746,
+      "step": 2212352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3769054281321023e-05,
+      "loss": 3.8751,
+      "step": 2212864
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.9982826709747314,
+      "eval_runtime": 300.6474,
+      "eval_samples_per_second": 1269.231,
+      "eval_steps_per_second": 39.664,
+      "step": 2213280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3760684712614236e-05,
+      "loss": 3.873,
+      "step": 2213376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3752298765103713e-05,
+      "loss": 3.8749,
+      "step": 2213888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3743912817593194e-05,
+      "loss": 3.8665,
+      "step": 2214400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3735526870082674e-05,
+      "loss": 3.8759,
+      "step": 2214912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3727140922572154e-05,
+      "loss": 3.8759,
+      "step": 2215424
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3718754975061634e-05,
+      "loss": 3.8709,
+      "step": 2215936
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3710369027551116e-05,
+      "loss": 3.8699,
+      "step": 2216448
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3701983080040596e-05,
+      "loss": 3.8594,
+      "step": 2216960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3693597132530072e-05,
+      "loss": 3.8662,
+      "step": 2217472
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3685211185019554e-05,
+      "loss": 3.8793,
+      "step": 2217984
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3676825237509034e-05,
+      "loss": 3.8696,
+      "step": 2218496
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3668439289998514e-05,
+      "loss": 3.8728,
+      "step": 2219008
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3660069721291725e-05,
+      "loss": 3.8841,
+      "step": 2219520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3651683773781205e-05,
+      "loss": 3.8636,
+      "step": 2220032
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3643314205074417e-05,
+      "loss": 3.8626,
+      "step": 2220544
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3634928257563895e-05,
+      "loss": 3.8604,
+      "step": 2221056
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3626542310053375e-05,
+      "loss": 3.8736,
+      "step": 2221568
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3618156362542855e-05,
+      "loss": 3.8664,
+      "step": 2222080
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3609770415032335e-05,
+      "loss": 3.8675,
+      "step": 2222592
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3601384467521817e-05,
+      "loss": 3.8744,
+      "step": 2223104
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3592998520011297e-05,
+      "loss": 3.8829,
+      "step": 2223616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3584612572500777e-05,
+      "loss": 3.8733,
+      "step": 2224128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3576226624990255e-05,
+      "loss": 3.8669,
+      "step": 2224640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3567840677479735e-05,
+      "loss": 3.8676,
+      "step": 2225152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3559454729969215e-05,
+      "loss": 3.875,
+      "step": 2225664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3551068782458695e-05,
+      "loss": 3.8671,
+      "step": 2226176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3542682834948175e-05,
+      "loss": 3.855,
+      "step": 2226688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3534296887437656e-05,
+      "loss": 3.8658,
+      "step": 2227200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3525910939927133e-05,
+      "loss": 3.8669,
+      "step": 2227712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.3517541371220347e-05,
+      "loss": 3.8592,
+      "step": 2228224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3509155423709827e-05,
+      "loss": 3.8673,
+      "step": 2228736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3500769476199307e-05,
+      "loss": 3.8664,
+      "step": 2229248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3492383528688785e-05,
+      "loss": 3.8735,
+      "step": 2229760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3483997581178265e-05,
+      "loss": 3.8749,
+      "step": 2230272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3475611633667745e-05,
+      "loss": 3.8711,
+      "step": 2230784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3467242064960958e-05,
+      "loss": 3.872,
+      "step": 2231296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3458856117450436e-05,
+      "loss": 3.8831,
+      "step": 2231808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3450470169939916e-05,
+      "loss": 3.8535,
+      "step": 2232320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3442084222429396e-05,
+      "loss": 3.8613,
+      "step": 2232832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3433698274918876e-05,
+      "loss": 3.8656,
+      "step": 2233344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3425328706212087e-05,
+      "loss": 3.8543,
+      "step": 2233856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.34169591375053e-05,
+      "loss": 3.8701,
+      "step": 2234368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.340857318999478e-05,
+      "loss": 3.8723,
+      "step": 2234880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3400187242484257e-05,
+      "loss": 3.8636,
+      "step": 2235392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3391801294973737e-05,
+      "loss": 3.8643,
+      "step": 2235904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3383415347463219e-05,
+      "loss": 3.8684,
+      "step": 2236416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3375029399952699e-05,
+      "loss": 3.8662,
+      "step": 2236928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3366643452442179e-05,
+      "loss": 3.8761,
+      "step": 2237440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3358257504931659e-05,
+      "loss": 3.8486,
+      "step": 2237952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.334988793622487e-05,
+      "loss": 3.8623,
+      "step": 2238464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.334150198871435e-05,
+      "loss": 3.8798,
+      "step": 2238976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.333311604120383e-05,
+      "loss": 3.8752,
+      "step": 2239488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3324730093693311e-05,
+      "loss": 3.8653,
+      "step": 2240000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3316344146182788e-05,
+      "loss": 3.8584,
+      "step": 2240512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3307974577476002e-05,
+      "loss": 3.8647,
+      "step": 2241024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3299588629965482e-05,
+      "loss": 3.8555,
+      "step": 2241536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3291202682454962e-05,
+      "loss": 3.8738,
+      "step": 2242048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.328281673494444e-05,
+      "loss": 3.8552,
+      "step": 2242560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.327443078743392e-05,
+      "loss": 3.8712,
+      "step": 2243072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3266061218727133e-05,
+      "loss": 3.8652,
+      "step": 2243584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.325767527121661e-05,
+      "loss": 3.8665,
+      "step": 2244096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.324928932370609e-05,
+      "loss": 3.8611,
+      "step": 2244608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.324090337619557e-05,
+      "loss": 3.8665,
+      "step": 2245120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.323251742868505e-05,
+      "loss": 3.8576,
+      "step": 2245632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3224131481174532e-05,
+      "loss": 3.8559,
+      "step": 2246144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3215745533664012e-05,
+      "loss": 3.8762,
+      "step": 2246656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3207375964957221e-05,
+      "loss": 3.8702,
+      "step": 2247168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3199006396250436e-05,
+      "loss": 3.8595,
+      "step": 2247680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3190620448739912e-05,
+      "loss": 3.8514,
+      "step": 2248192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3182234501229394e-05,
+      "loss": 3.8647,
+      "step": 2248704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3173848553718874e-05,
+      "loss": 3.8659,
+      "step": 2249216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3165462606208354e-05,
+      "loss": 3.8799,
+      "step": 2249728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3157076658697834e-05,
+      "loss": 3.8589,
+      "step": 2250240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3148690711187314e-05,
+      "loss": 3.8722,
+      "step": 2250752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3140321142480524e-05,
+      "loss": 3.8783,
+      "step": 2251264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3131935194970004e-05,
+      "loss": 3.863,
+      "step": 2251776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3123549247459484e-05,
+      "loss": 3.8693,
+      "step": 2252288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3115163299948966e-05,
+      "loss": 3.8697,
+      "step": 2252800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3106777352438442e-05,
+      "loss": 3.8661,
+      "step": 2253312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3098391404927922e-05,
+      "loss": 3.8672,
+      "step": 2253824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3090005457417404e-05,
+      "loss": 3.8625,
+      "step": 2254336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3081619509906884e-05,
+      "loss": 3.8731,
+      "step": 2254848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3073249941200095e-05,
+      "loss": 3.882,
+      "step": 2255360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3064863993689575e-05,
+      "loss": 3.8677,
+      "step": 2255872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3056478046179055e-05,
+      "loss": 3.8494,
+      "step": 2256384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3048092098668535e-05,
+      "loss": 3.8647,
+      "step": 2256896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3039706151158015e-05,
+      "loss": 3.8697,
+      "step": 2257408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3031320203647496e-05,
+      "loss": 3.8674,
+      "step": 2257920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3022934256136973e-05,
+      "loss": 3.8705,
+      "step": 2258432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.3014564687430187e-05,
+      "loss": 3.8649,
+      "step": 2258944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3006178739919667e-05,
+      "loss": 3.8687,
+      "step": 2259456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2997792792409147e-05,
+      "loss": 3.8562,
+      "step": 2259968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2989406844898625e-05,
+      "loss": 3.8517,
+      "step": 2260480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2981037276191838e-05,
+      "loss": 3.8597,
+      "step": 2260992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2972651328681318e-05,
+      "loss": 3.8603,
+      "step": 2261504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2964281759974528e-05,
+      "loss": 3.878,
+      "step": 2262016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2955895812464008e-05,
+      "loss": 3.8605,
+      "step": 2262528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2947509864953488e-05,
+      "loss": 3.8557,
+      "step": 2263040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2939123917442968e-05,
+      "loss": 3.8677,
+      "step": 2263552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2930737969932447e-05,
+      "loss": 3.8633,
+      "step": 2264064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2922352022421927e-05,
+      "loss": 3.8713,
+      "step": 2264576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2913966074911407e-05,
+      "loss": 3.8638,
+      "step": 2265088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2905580127400888e-05,
+      "loss": 3.8661,
+      "step": 2265600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2897210558694097e-05,
+      "loss": 3.8641,
+      "step": 2266112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2888824611183579e-05,
+      "loss": 3.8669,
+      "step": 2266624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2880438663673059e-05,
+      "loss": 3.8582,
+      "step": 2267136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2872052716162539e-05,
+      "loss": 3.8606,
+      "step": 2267648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2863666768652019e-05,
+      "loss": 3.8697,
+      "step": 2268160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2855280821141499e-05,
+      "loss": 3.8609,
+      "step": 2268672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.284689487363098e-05,
+      "loss": 3.8589,
+      "step": 2269184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2838508926120457e-05,
+      "loss": 3.8577,
+      "step": 2269696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.283013935741367e-05,
+      "loss": 3.8627,
+      "step": 2270208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2821786167510613e-05,
+      "loss": 3.8662,
+      "step": 2270720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2813400220000093e-05,
+      "loss": 3.8626,
+      "step": 2271232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2805014272489571e-05,
+      "loss": 3.8665,
+      "step": 2271744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2796628324979051e-05,
+      "loss": 3.8728,
+      "step": 2272256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2788258756272265e-05,
+      "loss": 3.8728,
+      "step": 2272768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2779872808761745e-05,
+      "loss": 3.8683,
+      "step": 2273280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2771486861251222e-05,
+      "loss": 3.8706,
+      "step": 2273792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2763100913740703e-05,
+      "loss": 3.8641,
+      "step": 2274304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2754714966230183e-05,
+      "loss": 3.8657,
+      "step": 2274816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2746329018719663e-05,
+      "loss": 3.8602,
+      "step": 2275328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2737943071209143e-05,
+      "loss": 3.8671,
+      "step": 2275840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2729557123698623e-05,
+      "loss": 3.8583,
+      "step": 2276352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2721171176188101e-05,
+      "loss": 3.8687,
+      "step": 2276864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2712785228677581e-05,
+      "loss": 3.8705,
+      "step": 2277376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2704415659970795e-05,
+      "loss": 3.8676,
+      "step": 2277888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2696029712460275e-05,
+      "loss": 3.8621,
+      "step": 2278400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2687643764949752e-05,
+      "loss": 3.8583,
+      "step": 2278912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2679257817439234e-05,
+      "loss": 3.8739,
+      "step": 2279424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2670871869928714e-05,
+      "loss": 3.8644,
+      "step": 2279936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2662485922418194e-05,
+      "loss": 3.862,
+      "step": 2280448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2654099974907674e-05,
+      "loss": 3.8409,
+      "step": 2280960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2645730406200884e-05,
+      "loss": 3.8676,
+      "step": 2281472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2637344458690364e-05,
+      "loss": 3.8698,
+      "step": 2281984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2628974889983575e-05,
+      "loss": 3.8784,
+      "step": 2282496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2620588942473055e-05,
+      "loss": 3.866,
+      "step": 2283008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2612202994962535e-05,
+      "loss": 3.8606,
+      "step": 2283520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2603817047452015e-05,
+      "loss": 3.8667,
+      "step": 2284032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2595431099941497e-05,
+      "loss": 3.8834,
+      "step": 2284544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2587045152430977e-05,
+      "loss": 3.8578,
+      "step": 2285056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2578659204920456e-05,
+      "loss": 3.8684,
+      "step": 2285568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2570273257409935e-05,
+      "loss": 3.8641,
+      "step": 2286080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2561887309899415e-05,
+      "loss": 3.8687,
+      "step": 2286592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2553501362388895e-05,
+      "loss": 3.8686,
+      "step": 2287104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2545115414878375e-05,
+      "loss": 3.8658,
+      "step": 2287616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2536729467367855e-05,
+      "loss": 3.8599,
+      "step": 2288128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2528376277464798e-05,
+      "loss": 3.8747,
+      "step": 2288640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2519990329954278e-05,
+      "loss": 3.869,
+      "step": 2289152
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.99674916267395,
+      "eval_runtime": 287.29,
+      "eval_samples_per_second": 1328.243,
+      "eval_steps_per_second": 41.509,
+      "step": 2289600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2511604382443756e-05,
+      "loss": 3.8581,
+      "step": 2289664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2503218434933236e-05,
+      "loss": 3.873,
+      "step": 2290176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2494832487422716e-05,
+      "loss": 3.8634,
+      "step": 2290688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2486462918715929e-05,
+      "loss": 3.8732,
+      "step": 2291200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2478076971205408e-05,
+      "loss": 3.8694,
+      "step": 2291712
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2469691023694888e-05,
+      "loss": 3.8725,
+      "step": 2292224
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2461305076184368e-05,
+      "loss": 3.8626,
+      "step": 2292736
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2452919128673848e-05,
+      "loss": 3.8559,
+      "step": 2293248
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2444533181163328e-05,
+      "loss": 3.8617,
+      "step": 2293760
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2436147233652808e-05,
+      "loss": 3.8776,
+      "step": 2294272
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2427777664946019e-05,
+      "loss": 3.8652,
+      "step": 2294784
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2419391717435499e-05,
+      "loss": 3.8668,
+      "step": 2295296
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2411022148728711e-05,
+      "loss": 3.8796,
+      "step": 2295808
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.240263620121819e-05,
+      "loss": 3.8611,
+      "step": 2296320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.239425025370767e-05,
+      "loss": 3.8594,
+      "step": 2296832
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2385864306197151e-05,
+      "loss": 3.8542,
+      "step": 2297344
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.237749473749036e-05,
+      "loss": 3.8646,
+      "step": 2297856
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2369108789979842e-05,
+      "loss": 3.8643,
+      "step": 2298368
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2360722842469322e-05,
+      "loss": 3.8625,
+      "step": 2298880
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2352336894958802e-05,
+      "loss": 3.8677,
+      "step": 2299392
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.234395094744828e-05,
+      "loss": 3.8821,
+      "step": 2299904
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2335564999937762e-05,
+      "loss": 3.8695,
+      "step": 2300416
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2327179052427242e-05,
+      "loss": 3.8638,
+      "step": 2300928
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.231879310491672e-05,
+      "loss": 3.8659,
+      "step": 2301440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.23104071574062e-05,
+      "loss": 3.8654,
+      "step": 2301952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2302021209895682e-05,
+      "loss": 3.867,
+      "step": 2302464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.229363526238516e-05,
+      "loss": 3.8515,
+      "step": 2302976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.228524931487464e-05,
+      "loss": 3.8588,
+      "step": 2303488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.227686336736412e-05,
+      "loss": 3.8642,
+      "step": 2304000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2268493798657332e-05,
+      "loss": 3.8549,
+      "step": 2304512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.226010785114681e-05,
+      "loss": 3.8578,
+      "step": 2305024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2251738282440023e-05,
+      "loss": 3.8696,
+      "step": 2305536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2243352334929503e-05,
+      "loss": 3.8658,
+      "step": 2306048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2234966387418983e-05,
+      "loss": 3.8722,
+      "step": 2306560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2226580439908463e-05,
+      "loss": 3.8685,
+      "step": 2307072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2218210871201674e-05,
+      "loss": 3.8662,
+      "step": 2307584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2209824923691154e-05,
+      "loss": 3.8753,
+      "step": 2308096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2201438976180634e-05,
+      "loss": 3.8534,
+      "step": 2308608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2193053028670114e-05,
+      "loss": 3.859,
+      "step": 2309120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2184667081159594e-05,
+      "loss": 3.8576,
+      "step": 2309632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2176281133649074e-05,
+      "loss": 3.851,
+      "step": 2310144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2167895186138554e-05,
+      "loss": 3.866,
+      "step": 2310656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2159509238628033e-05,
+      "loss": 3.8677,
+      "step": 2311168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2151123291117513e-05,
+      "loss": 3.8601,
+      "step": 2311680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2142737343606993e-05,
+      "loss": 3.8613,
+      "step": 2312192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2134351396096473e-05,
+      "loss": 3.8622,
+      "step": 2312704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2125965448585953e-05,
+      "loss": 3.8663,
+      "step": 2313216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2117595879879166e-05,
+      "loss": 3.8664,
+      "step": 2313728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2109209932368644e-05,
+      "loss": 3.8455,
+      "step": 2314240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2100823984858124e-05,
+      "loss": 3.8587,
+      "step": 2314752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2092438037347604e-05,
+      "loss": 3.8737,
+      "step": 2315264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2084084847444547e-05,
+      "loss": 3.8688,
+      "step": 2315776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2075698899934027e-05,
+      "loss": 3.8628,
+      "step": 2316288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2067312952423507e-05,
+      "loss": 3.8541,
+      "step": 2316800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2058927004912987e-05,
+      "loss": 3.8634,
+      "step": 2317312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2050541057402465e-05,
+      "loss": 3.8538,
+      "step": 2317824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2042155109891947e-05,
+      "loss": 3.8633,
+      "step": 2318336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2033769162381427e-05,
+      "loss": 3.852,
+      "step": 2318848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2025383214870905e-05,
+      "loss": 3.8678,
+      "step": 2319360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2016997267360385e-05,
+      "loss": 3.8647,
+      "step": 2319872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2008627698653598e-05,
+      "loss": 3.858,
+      "step": 2320384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.2000241751143078e-05,
+      "loss": 3.8547,
+      "step": 2320896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1991855803632558e-05,
+      "loss": 3.8643,
+      "step": 2321408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1983469856122038e-05,
+      "loss": 3.8526,
+      "step": 2321920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1975083908611518e-05,
+      "loss": 3.8543,
+      "step": 2322432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1966697961100996e-05,
+      "loss": 3.8704,
+      "step": 2322944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1958312013590477e-05,
+      "loss": 3.8668,
+      "step": 2323456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1949926066079957e-05,
+      "loss": 3.856,
+      "step": 2323968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1941556497373168e-05,
+      "loss": 3.8459,
+      "step": 2324480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1933170549862648e-05,
+      "loss": 3.86,
+      "step": 2324992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1924784602352128e-05,
+      "loss": 3.8631,
+      "step": 2325504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1916398654841608e-05,
+      "loss": 3.8724,
+      "step": 2326016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1908029086134819e-05,
+      "loss": 3.8565,
+      "step": 2326528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.189965951742803e-05,
+      "loss": 3.8719,
+      "step": 2327040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.189127356991751e-05,
+      "loss": 3.8704,
+      "step": 2327552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1882887622406991e-05,
+      "loss": 3.8653,
+      "step": 2328064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.187450167489647e-05,
+      "loss": 3.8664,
+      "step": 2328576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1866132106189682e-05,
+      "loss": 3.8621,
+      "step": 2329088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1857746158679162e-05,
+      "loss": 3.8676,
+      "step": 2329600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1849360211168642e-05,
+      "loss": 3.8589,
+      "step": 2330112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.184097426365812e-05,
+      "loss": 3.8602,
+      "step": 2330624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1832588316147602e-05,
+      "loss": 3.8655,
+      "step": 2331136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1824202368637082e-05,
+      "loss": 3.8767,
+      "step": 2331648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.181581642112656e-05,
+      "loss": 3.863,
+      "step": 2332160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.180743047361604e-05,
+      "loss": 3.8475,
+      "step": 2332672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1799060904909252e-05,
+      "loss": 3.8612,
+      "step": 2333184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1790691336202463e-05,
+      "loss": 3.8638,
+      "step": 2333696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1782305388691943e-05,
+      "loss": 3.8664,
+      "step": 2334208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1773919441181423e-05,
+      "loss": 3.868,
+      "step": 2334720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1765533493670903e-05,
+      "loss": 3.8633,
+      "step": 2335232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1757147546160383e-05,
+      "loss": 3.8622,
+      "step": 2335744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1748761598649863e-05,
+      "loss": 3.8535,
+      "step": 2336256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1740375651139343e-05,
+      "loss": 3.8505,
+      "step": 2336768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1731989703628823e-05,
+      "loss": 3.8527,
+      "step": 2337280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1723603756118303e-05,
+      "loss": 3.856,
+      "step": 2337792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1715234187411514e-05,
+      "loss": 3.8774,
+      "step": 2338304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1706848239900994e-05,
+      "loss": 3.856,
+      "step": 2338816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1698462292390474e-05,
+      "loss": 3.8486,
+      "step": 2339328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1690076344879954e-05,
+      "loss": 3.8666,
+      "step": 2339840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1681690397369434e-05,
+      "loss": 3.8585,
+      "step": 2340352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1673304449858913e-05,
+      "loss": 3.871,
+      "step": 2340864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1664934881152124e-05,
+      "loss": 3.8595,
+      "step": 2341376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1656548933641604e-05,
+      "loss": 3.859,
+      "step": 2341888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1648162986131086e-05,
+      "loss": 3.8605,
+      "step": 2342400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1639777038620564e-05,
+      "loss": 3.8627,
+      "step": 2342912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1631391091110044e-05,
+      "loss": 3.8517,
+      "step": 2343424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1623021522403257e-05,
+      "loss": 3.8617,
+      "step": 2343936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1614635574892737e-05,
+      "loss": 3.8612,
+      "step": 2344448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1606249627382215e-05,
+      "loss": 3.8581,
+      "step": 2344960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1597863679871695e-05,
+      "loss": 3.8537,
+      "step": 2345472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1589494111164907e-05,
+      "loss": 3.8564,
+      "step": 2345984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1581108163654387e-05,
+      "loss": 3.8557,
+      "step": 2346496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1572722216143867e-05,
+      "loss": 3.8602,
+      "step": 2347008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1564336268633347e-05,
+      "loss": 3.863,
+      "step": 2347520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1555950321122827e-05,
+      "loss": 3.8609,
+      "step": 2348032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1547580752416038e-05,
+      "loss": 3.8693,
+      "step": 2348544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1539194804905518e-05,
+      "loss": 3.871,
+      "step": 2349056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1530808857394998e-05,
+      "loss": 3.8623,
+      "step": 2349568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1522422909884478e-05,
+      "loss": 3.8635,
+      "step": 2350080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1514036962373958e-05,
+      "loss": 3.8636,
+      "step": 2350592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1505651014863438e-05,
+      "loss": 3.8615,
+      "step": 2351104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1497265067352918e-05,
+      "loss": 3.8521,
+      "step": 2351616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1488895498646128e-05,
+      "loss": 3.8644,
+      "step": 2352128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1480509551135608e-05,
+      "loss": 3.8574,
+      "step": 2352640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1472123603625088e-05,
+      "loss": 3.8634,
+      "step": 2353152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1463737656114568e-05,
+      "loss": 3.8627,
+      "step": 2353664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1455368087407779e-05,
+      "loss": 3.8664,
+      "step": 2354176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1446982139897259e-05,
+      "loss": 3.8576,
+      "step": 2354688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.143859619238674e-05,
+      "loss": 3.8546,
+      "step": 2355200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1430210244876219e-05,
+      "loss": 3.8707,
+      "step": 2355712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.142184067616943e-05,
+      "loss": 3.8637,
+      "step": 2356224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1413454728658911e-05,
+      "loss": 3.8569,
+      "step": 2356736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1405068781148391e-05,
+      "loss": 3.8386,
+      "step": 2357248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.139668283363787e-05,
+      "loss": 3.8592,
+      "step": 2357760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.138829688612735e-05,
+      "loss": 3.8692,
+      "step": 2358272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1379910938616831e-05,
+      "loss": 3.8743,
+      "step": 2358784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.137152499110631e-05,
+      "loss": 3.8631,
+      "step": 2359296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1363155422399522e-05,
+      "loss": 3.8565,
+      "step": 2359808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1354769474889002e-05,
+      "loss": 3.8631,
+      "step": 2360320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1346383527378482e-05,
+      "loss": 3.8761,
+      "step": 2360832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.133799757986796e-05,
+      "loss": 3.8576,
+      "step": 2361344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1329611632357442e-05,
+      "loss": 3.8616,
+      "step": 2361856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1321225684846922e-05,
+      "loss": 3.864,
+      "step": 2362368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.13128397373364e-05,
+      "loss": 3.8647,
+      "step": 2362880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.130445378982588e-05,
+      "loss": 3.8641,
+      "step": 2363392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1296084221119092e-05,
+      "loss": 3.8613,
+      "step": 2363904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1287698273608572e-05,
+      "loss": 3.8562,
+      "step": 2364416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1279312326098052e-05,
+      "loss": 3.8722,
+      "step": 2364928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1270942757391263e-05,
+      "loss": 3.8615,
+      "step": 2365440
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.9955921173095703,
+      "eval_runtime": 293.4747,
+      "eval_samples_per_second": 1300.252,
+      "eval_steps_per_second": 40.634,
+      "step": 2365920
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1262556809880743e-05,
+      "loss": 3.8561,
+      "step": 2365952
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1254170862370223e-05,
+      "loss": 3.8692,
+      "step": 2366464
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1245784914859703e-05,
+      "loss": 3.8577,
+      "step": 2366976
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1237398967349183e-05,
+      "loss": 3.8699,
+      "step": 2367488
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1229029398642395e-05,
+      "loss": 3.8661,
+      "step": 2368000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1220643451131874e-05,
+      "loss": 3.8725,
+      "step": 2368512
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1212257503621354e-05,
+      "loss": 3.8568,
+      "step": 2369024
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1203871556110834e-05,
+      "loss": 3.853,
+      "step": 2369536
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1195485608600314e-05,
+      "loss": 3.8536,
+      "step": 2370048
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1187116039893524e-05,
+      "loss": 3.8797,
+      "step": 2370560
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1178730092383006e-05,
+      "loss": 3.8619,
+      "step": 2371072
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1170344144872486e-05,
+      "loss": 3.8646,
+      "step": 2371584
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1161958197361964e-05,
+      "loss": 3.8719,
+      "step": 2372096
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1153588628655177e-05,
+      "loss": 3.8641,
+      "step": 2372608
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1145202681144657e-05,
+      "loss": 3.8527,
+      "step": 2373120
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1136816733634137e-05,
+      "loss": 3.8558,
+      "step": 2373632
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1128430786123615e-05,
+      "loss": 3.8568,
+      "step": 2374144
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1120061217416827e-05,
+      "loss": 3.8595,
+      "step": 2374656
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1111675269906307e-05,
+      "loss": 3.8599,
+      "step": 2375168
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1103289322395787e-05,
+      "loss": 3.8659,
+      "step": 2375680
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1094903374885267e-05,
+      "loss": 3.8804,
+      "step": 2376192
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1086517427374747e-05,
+      "loss": 3.8624,
+      "step": 2376704
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1078131479864227e-05,
+      "loss": 3.8629,
+      "step": 2377216
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1069745532353707e-05,
+      "loss": 3.8602,
+      "step": 2377728
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1061359584843187e-05,
+      "loss": 3.8644,
+      "step": 2378240
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1052973637332667e-05,
+      "loss": 3.8586,
+      "step": 2378752
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1044587689822145e-05,
+      "loss": 3.8495,
+      "step": 2379264
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1036218121115358e-05,
+      "loss": 3.8554,
+      "step": 2379776
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1027832173604838e-05,
+      "loss": 3.8627,
+      "step": 2380288
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1019446226094318e-05,
+      "loss": 3.8518,
+      "step": 2380800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.1011060278583798e-05,
+      "loss": 3.8522,
+      "step": 2381312
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.1002690709877008e-05,
+      "loss": 3.8684,
+      "step": 2381824
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0994304762366488e-05,
+      "loss": 3.862,
+      "step": 2382336
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0985918814855968e-05,
+      "loss": 3.8645,
+      "step": 2382848
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0977532867345448e-05,
+      "loss": 3.8713,
+      "step": 2383360
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.096916329863866e-05,
+      "loss": 3.8635,
+      "step": 2383872
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.096077735112814e-05,
+      "loss": 3.8666,
+      "step": 2384384
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0952391403617619e-05,
+      "loss": 3.8514,
+      "step": 2384896
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0944005456107099e-05,
+      "loss": 3.8545,
+      "step": 2385408
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.093561950859658e-05,
+      "loss": 3.8529,
+      "step": 2385920
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0927233561086059e-05,
+      "loss": 3.8525,
+      "step": 2386432
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.091886399237927e-05,
+      "loss": 3.8576,
+      "step": 2386944
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0910478044868751e-05,
+      "loss": 3.868,
+      "step": 2387456
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0902092097358231e-05,
+      "loss": 3.8519,
+      "step": 2387968
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.089370614984771e-05,
+      "loss": 3.8614,
+      "step": 2388480
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0885320202337191e-05,
+      "loss": 3.8561,
+      "step": 2388992
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0876934254826671e-05,
+      "loss": 3.8623,
+      "step": 2389504
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.086854830731615e-05,
+      "loss": 3.8653,
+      "step": 2390016
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.086016235980563e-05,
+      "loss": 3.8438,
+      "step": 2390528
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0851776412295111e-05,
+      "loss": 3.8533,
+      "step": 2391040
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0843406843588322e-05,
+      "loss": 3.8679,
+      "step": 2391552
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.08350208960778e-05,
+      "loss": 3.8679,
+      "step": 2392064
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0826634948567282e-05,
+      "loss": 3.8538,
+      "step": 2392576
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0818249001056762e-05,
+      "loss": 3.8553,
+      "step": 2393088
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0809879432349972e-05,
+      "loss": 3.8603,
+      "step": 2393600
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0801493484839452e-05,
+      "loss": 3.847,
+      "step": 2394112
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0793107537328932e-05,
+      "loss": 3.8594,
+      "step": 2394624
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0784721589818412e-05,
+      "loss": 3.8481,
+      "step": 2395136
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0776335642307892e-05,
+      "loss": 3.8646,
+      "step": 2395648
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0767949694797372e-05,
+      "loss": 3.8581,
+      "step": 2396160
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0759563747286852e-05,
+      "loss": 3.8589,
+      "step": 2396672
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.075117779977633e-05,
+      "loss": 3.8477,
+      "step": 2397184
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0742791852265812e-05,
+      "loss": 3.8603,
+      "step": 2397696
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0734438662362754e-05,
+      "loss": 3.846,
+      "step": 2398208
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0726052714852235e-05,
+      "loss": 3.8567,
+      "step": 2398720
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0717666767341714e-05,
+      "loss": 3.8646,
+      "step": 2399232
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0709280819831194e-05,
+      "loss": 3.867,
+      "step": 2399744
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0700894872320673e-05,
+      "loss": 3.8473,
+      "step": 2400256
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0692525303613886e-05,
+      "loss": 3.8455,
+      "step": 2400768
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0684139356103364e-05,
+      "loss": 3.8545,
+      "step": 2401280
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0675753408592846e-05,
+      "loss": 3.8567,
+      "step": 2401792
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0667367461082326e-05,
+      "loss": 3.8691,
+      "step": 2402304
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0658981513571804e-05,
+      "loss": 3.8521,
+      "step": 2402816
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0650595566061284e-05,
+      "loss": 3.8691,
+      "step": 2403328
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0642209618550766e-05,
+      "loss": 3.8646,
+      "step": 2403840
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0633840049843976e-05,
+      "loss": 3.8637,
+      "step": 2404352
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0625454102333455e-05,
+      "loss": 3.8616,
+      "step": 2404864
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0617068154822936e-05,
+      "loss": 3.8596,
+      "step": 2405376
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0608682207312416e-05,
+      "loss": 3.8621,
+      "step": 2405888
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0600296259801895e-05,
+      "loss": 3.8572,
+      "step": 2406400
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0591910312291376e-05,
+      "loss": 3.856,
+      "step": 2406912
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0583524364780856e-05,
+      "loss": 3.8606,
+      "step": 2407424
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0575138417270334e-05,
+      "loss": 3.875,
+      "step": 2407936
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0566785227367278e-05,
+      "loss": 3.8591,
+      "step": 2408448
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0558399279856758e-05,
+      "loss": 3.8519,
+      "step": 2408960
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0550013332346238e-05,
+      "loss": 3.8512,
+      "step": 2409472
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0541627384835718e-05,
+      "loss": 3.8624,
+      "step": 2409984
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0533241437325198e-05,
+      "loss": 3.8601,
+      "step": 2410496
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.0524855489814678e-05,
+      "loss": 3.866,
+      "step": 2411008
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 1.051648592110789e-05,
+      "loss": 3.8569,
+      "step": 2411520
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0508099973597368e-05,
+      "loss": 3.8596,
+      "step": 2412032
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0499714026086848e-05,
+      "loss": 3.8501,
+      "step": 2412544
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0491328078576328e-05,
+      "loss": 3.8485,
+      "step": 2413056
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0482974888673272e-05,
+      "loss": 3.849,
+      "step": 2413568
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0474588941162752e-05,
+      "loss": 3.8512,
+      "step": 2414080
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0466202993652231e-05,
+      "loss": 3.8715,
+      "step": 2414592
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0457817046141711e-05,
+      "loss": 3.8608,
+      "step": 2415104
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.044943109863119e-05,
+      "loss": 3.8441,
+      "step": 2415616
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0441045151120671e-05,
+      "loss": 3.8598,
+      "step": 2416128
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0432659203610151e-05,
+      "loss": 3.8601,
+      "step": 2416640
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0424273256099631e-05,
+      "loss": 3.8647,
+      "step": 2417152
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0415887308589111e-05,
+      "loss": 3.8571,
+      "step": 2417664
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0407501361078591e-05,
+      "loss": 3.8573,
+      "step": 2418176
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0399115413568071e-05,
+      "loss": 3.856,
+      "step": 2418688
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.039072946605755e-05,
+      "loss": 3.8643,
+      "step": 2419200
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0382359897350762e-05,
+      "loss": 3.8466,
+      "step": 2419712
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0373973949840242e-05,
+      "loss": 3.8559,
+      "step": 2420224
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0365588002329722e-05,
+      "loss": 3.8613,
+      "step": 2420736
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0357202054819202e-05,
+      "loss": 3.8523,
+      "step": 2421248
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0348816107308682e-05,
+      "loss": 3.8518,
+      "step": 2421760
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0340446538601892e-05,
+      "loss": 3.8558,
+      "step": 2422272
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0332060591091372e-05,
+      "loss": 3.8493,
+      "step": 2422784
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0323674643580852e-05,
+      "loss": 3.8613,
+      "step": 2423296
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0315288696070332e-05,
+      "loss": 3.8573,
+      "step": 2423808
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0306919127363545e-05,
+      "loss": 3.8541,
+      "step": 2424320
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0298533179853023e-05,
+      "loss": 3.8665,
+      "step": 2424832
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0290147232342503e-05,
+      "loss": 3.8665,
+      "step": 2425344
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0281761284831985e-05,
+      "loss": 3.8609,
+      "step": 2425856
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0273375337321463e-05,
+      "loss": 3.8633,
+      "step": 2426368
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0264989389810943e-05,
+      "loss": 3.8594,
+      "step": 2426880
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0256603442300423e-05,
+      "loss": 3.8586,
+      "step": 2427392
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0248217494789903e-05,
+      "loss": 3.8527,
+      "step": 2427904
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0239831547279383e-05,
+      "loss": 3.8541,
+      "step": 2428416
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0231461978572594e-05,
+      "loss": 3.8588,
+      "step": 2428928
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0223076031062075e-05,
+      "loss": 3.8594,
+      "step": 2429440
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0214690083551553e-05,
+      "loss": 3.8562,
+      "step": 2429952
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0206304136041033e-05,
+      "loss": 3.8635,
+      "step": 2430464
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0197918188530513e-05,
+      "loss": 3.8565,
+      "step": 2430976
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0189548619823726e-05,
+      "loss": 3.8523,
+      "step": 2431488
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0181162672313204e-05,
+      "loss": 3.8666,
+      "step": 2432000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0172776724802686e-05,
+      "loss": 3.8609,
+      "step": 2432512
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0164390777292166e-05,
+      "loss": 3.8527,
+      "step": 2433024
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0156021208585375e-05,
+      "loss": 3.84,
+      "step": 2433536
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0147635261074856e-05,
+      "loss": 3.855,
+      "step": 2434048
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0139265692368067e-05,
+      "loss": 3.8677,
+      "step": 2434560
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0130879744857547e-05,
+      "loss": 3.8687,
+      "step": 2435072
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0122493797347027e-05,
+      "loss": 3.8568,
+      "step": 2435584
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0114107849836507e-05,
+      "loss": 3.8531,
+      "step": 2436096
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0105721902325987e-05,
+      "loss": 3.8626,
+      "step": 2436608
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.00973523336192e-05,
+      "loss": 3.8722,
+      "step": 2437120
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0088966386108678e-05,
+      "loss": 3.8547,
+      "step": 2437632
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0080580438598158e-05,
+      "loss": 3.8597,
+      "step": 2438144
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.007219449108764e-05,
+      "loss": 3.8574,
+      "step": 2438656
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0063808543577118e-05,
+      "loss": 3.8602,
+      "step": 2439168
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0055422596066598e-05,
+      "loss": 3.859,
+      "step": 2439680
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0047036648556078e-05,
+      "loss": 3.8654,
+      "step": 2440192
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0038650701045558e-05,
+      "loss": 3.8506,
+      "step": 2440704
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0030281132338768e-05,
+      "loss": 3.8674,
+      "step": 2441216
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.0021895184828248e-05,
+      "loss": 3.8609,
+      "step": 2441728
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.001350923731773e-05,
+      "loss": 3.8512,
+      "step": 2442240
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 3.9942140579223633,
+      "eval_runtime": 292.7501,
+      "eval_samples_per_second": 1303.47,
+      "eval_steps_per_second": 40.734,
+      "step": 2442240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.0005123289807208e-05,
+      "loss": 3.8643,
+      "step": 2442752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.996737342296688e-06,
+      "loss": 3.8564,
+      "step": 2443264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.988351394786168e-06,
+      "loss": 3.8663,
+      "step": 2443776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.979965447275648e-06,
+      "loss": 3.8612,
+      "step": 2444288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.971579499765128e-06,
+      "loss": 3.8698,
+      "step": 2444800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.963193552254608e-06,
+      "loss": 3.8545,
+      "step": 2445312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.95480760474409e-06,
+      "loss": 3.8483,
+      "step": 2445824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.946421657233568e-06,
+      "loss": 3.8497,
+      "step": 2446336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.938035709723048e-06,
+      "loss": 3.8752,
+      "step": 2446848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.929649762212528e-06,
+      "loss": 3.8606,
+      "step": 2447360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.921263814702008e-06,
+      "loss": 3.8559,
+      "step": 2447872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.912877867191488e-06,
+      "loss": 3.8677,
+      "step": 2448384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.904508298484699e-06,
+      "loss": 3.8625,
+      "step": 2448896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.89612235097418e-06,
+      "loss": 3.852,
+      "step": 2449408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.887736403463658e-06,
+      "loss": 3.8527,
+      "step": 2449920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.879350455953138e-06,
+      "loss": 3.8503,
+      "step": 2450432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.870964508442618e-06,
+      "loss": 3.858,
+      "step": 2450944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.862578560932098e-06,
+      "loss": 3.8554,
+      "step": 2451456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.854192613421578e-06,
+      "loss": 3.8659,
+      "step": 2451968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.845806665911058e-06,
+      "loss": 3.8793,
+      "step": 2452480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.83743709720427e-06,
+      "loss": 3.861,
+      "step": 2452992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.829051149693749e-06,
+      "loss": 3.8543,
+      "step": 2453504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.820665202183229e-06,
+      "loss": 3.8589,
+      "step": 2454016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.81227925467271e-06,
+      "loss": 3.8626,
+      "step": 2454528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.803893307162189e-06,
+      "loss": 3.8522,
+      "step": 2455040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.795523738455401e-06,
+      "loss": 3.8498,
+      "step": 2455552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.787137790944881e-06,
+      "loss": 3.8491,
+      "step": 2456064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.778751843434361e-06,
+      "loss": 3.8657,
+      "step": 2456576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.77036589592384e-06,
+      "loss": 3.8446,
+      "step": 2457088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.761979948413321e-06,
+      "loss": 3.85,
+      "step": 2457600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.753610379706532e-06,
+      "loss": 3.8629,
+      "step": 2458112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.74522443219601e-06,
+      "loss": 3.8616,
+      "step": 2458624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.736838484685492e-06,
+      "loss": 3.8603,
+      "step": 2459136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.728452537174972e-06,
+      "loss": 3.867,
+      "step": 2459648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.720082968468183e-06,
+      "loss": 3.8596,
+      "step": 2460160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.711697020957663e-06,
+      "loss": 3.8651,
+      "step": 2460672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.703311073447143e-06,
+      "loss": 3.8478,
+      "step": 2461184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.694925125936622e-06,
+      "loss": 3.8548,
+      "step": 2461696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.686539178426102e-06,
+      "loss": 3.849,
+      "step": 2462208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.678153230915582e-06,
+      "loss": 3.8496,
+      "step": 2462720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.669767283405062e-06,
+      "loss": 3.8555,
+      "step": 2463232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.66138133589454e-06,
+      "loss": 3.8633,
+      "step": 2463744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.653011767187753e-06,
+      "loss": 3.8519,
+      "step": 2464256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.644625819677233e-06,
+      "loss": 3.8548,
+      "step": 2464768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.636239872166713e-06,
+      "loss": 3.8548,
+      "step": 2465280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.627853924656193e-06,
+      "loss": 3.8582,
+      "step": 2465792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.619484355949404e-06,
+      "loss": 3.8632,
+      "step": 2466304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.611114787242616e-06,
+      "loss": 3.8461,
+      "step": 2466816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.602728839732096e-06,
+      "loss": 3.8466,
+      "step": 2467328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.594342892221574e-06,
+      "loss": 3.8635,
+      "step": 2467840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.585956944711056e-06,
+      "loss": 3.8664,
+      "step": 2468352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.577570997200536e-06,
+      "loss": 3.8515,
+      "step": 2468864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.569185049690014e-06,
+      "loss": 3.8545,
+      "step": 2469376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.560799102179494e-06,
+      "loss": 3.8575,
+      "step": 2469888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.552413154668976e-06,
+      "loss": 3.846,
+      "step": 2470400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.544043585962187e-06,
+      "loss": 3.8551,
+      "step": 2470912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.535657638451665e-06,
+      "loss": 3.846,
+      "step": 2471424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.527271690941147e-06,
+      "loss": 3.8606,
+      "step": 2471936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.518885743430627e-06,
+      "loss": 3.8573,
+      "step": 2472448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.51053255352757e-06,
+      "loss": 3.8558,
+      "step": 2472960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.502146606017048e-06,
+      "loss": 3.8432,
+      "step": 2473472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.493760658506528e-06,
+      "loss": 3.8623,
+      "step": 2473984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.48537471099601e-06,
+      "loss": 3.8415,
+      "step": 2474496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.476988763485488e-06,
+      "loss": 3.8543,
+      "step": 2475008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.468619194778699e-06,
+      "loss": 3.8608,
+      "step": 2475520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.46023324726818e-06,
+      "loss": 3.8621,
+      "step": 2476032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.45184729975766e-06,
+      "loss": 3.846,
+      "step": 2476544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.443461352247139e-06,
+      "loss": 3.845,
+      "step": 2477056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.435091783540351e-06,
+      "loss": 3.8461,
+      "step": 2477568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.426705836029831e-06,
+      "loss": 3.8579,
+      "step": 2478080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.418319888519311e-06,
+      "loss": 3.8603,
+      "step": 2478592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.409933941008791e-06,
+      "loss": 3.8524,
+      "step": 2479104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.401547993498271e-06,
+      "loss": 3.8626,
+      "step": 2479616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.393178424791482e-06,
+      "loss": 3.8673,
+      "step": 2480128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.384808856084694e-06,
+      "loss": 3.8581,
+      "step": 2480640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.376422908574173e-06,
+      "loss": 3.8567,
+      "step": 2481152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.368036961063652e-06,
+      "loss": 3.8568,
+      "step": 2481664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.359651013553134e-06,
+      "loss": 3.8623,
+      "step": 2482176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.351265066042612e-06,
+      "loss": 3.85,
+      "step": 2482688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.342879118532092e-06,
+      "loss": 3.8549,
+      "step": 2483200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.334493171021572e-06,
+      "loss": 3.8559,
+      "step": 2483712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.326107223511052e-06,
+      "loss": 3.87,
+      "step": 2484224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.317721276000532e-06,
+      "loss": 3.8603,
+      "step": 2484736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.309335328490012e-06,
+      "loss": 3.8446,
+      "step": 2485248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.300965759783225e-06,
+      "loss": 3.8509,
+      "step": 2485760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.292579812272703e-06,
+      "loss": 3.8539,
+      "step": 2486272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.284193864762183e-06,
+      "loss": 3.8608,
+      "step": 2486784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.275840674859126e-06,
+      "loss": 3.8621,
+      "step": 2487296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.267454727348606e-06,
+      "loss": 3.855,
+      "step": 2487808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.259068779838086e-06,
+      "loss": 3.8549,
+      "step": 2488320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.250682832327566e-06,
+      "loss": 3.8497,
+      "step": 2488832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.242296884817046e-06,
+      "loss": 3.8438,
+      "step": 2489344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.233910937306526e-06,
+      "loss": 3.8434,
+      "step": 2489856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.225524989796006e-06,
+      "loss": 3.8553,
+      "step": 2490368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.217139042285486e-06,
+      "loss": 3.8675,
+      "step": 2490880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.208753094774966e-06,
+      "loss": 3.8579,
+      "step": 2491392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.200367147264446e-06,
+      "loss": 3.8457,
+      "step": 2491904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.191981199753926e-06,
+      "loss": 3.8531,
+      "step": 2492416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.183611631047137e-06,
+      "loss": 3.857,
+      "step": 2492928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.175225683536616e-06,
+      "loss": 3.863,
+      "step": 2493440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.166839736026096e-06,
+      "loss": 3.8545,
+      "step": 2493952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.158453788515576e-06,
+      "loss": 3.8539,
+      "step": 2494464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.150067841005056e-06,
+      "loss": 3.8513,
+      "step": 2494976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.141681893494536e-06,
+      "loss": 3.8632,
+      "step": 2495488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.133295945984016e-06,
+      "loss": 3.843,
+      "step": 2496000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.124909998473496e-06,
+      "loss": 3.8493,
+      "step": 2496512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.116524050962976e-06,
+      "loss": 3.8594,
+      "step": 2497024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.108138103452456e-06,
+      "loss": 3.8529,
+      "step": 2497536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.099768534745667e-06,
+      "loss": 3.8468,
+      "step": 2498048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.091382587235147e-06,
+      "loss": 3.8531,
+      "step": 2498560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.082996639724627e-06,
+      "loss": 3.8502,
+      "step": 2499072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.074610692214107e-06,
+      "loss": 3.8545,
+      "step": 2499584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.066224744703587e-06,
+      "loss": 3.8522,
+      "step": 2500096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.057855175996798e-06,
+      "loss": 3.8528,
+      "step": 2500608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.049469228486277e-06,
+      "loss": 3.8613,
+      "step": 2501120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.041083280975757e-06,
+      "loss": 3.866,
+      "step": 2501632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.032697333465237e-06,
+      "loss": 3.8547,
+      "step": 2502144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.024311385954717e-06,
+      "loss": 3.8651,
+      "step": 2502656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.015925438444197e-06,
+      "loss": 3.8542,
+      "step": 2503168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.007539490933677e-06,
+      "loss": 3.8547,
+      "step": 2503680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.999153543423157e-06,
+      "loss": 3.8485,
+      "step": 2504192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.990783974716368e-06,
+      "loss": 3.8515,
+      "step": 2504704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.98239802720585e-06,
+      "loss": 3.8542,
+      "step": 2505216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.974012079695328e-06,
+      "loss": 3.8555,
+      "step": 2505728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.965626132184808e-06,
+      "loss": 3.8563,
+      "step": 2506240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.957240184674288e-06,
+      "loss": 3.8583,
+      "step": 2506752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.9488706159675e-06,
+      "loss": 3.8561,
+      "step": 2507264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.940484668456979e-06,
+      "loss": 3.8465,
+      "step": 2507776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.932098720946459e-06,
+      "loss": 3.8608,
+      "step": 2508288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.92371277343594e-06,
+      "loss": 3.8596,
+      "step": 2508800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.915326825925418e-06,
+      "loss": 3.8467,
+      "step": 2509312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.906957257218631e-06,
+      "loss": 3.8412,
+      "step": 2509824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.898571309708111e-06,
+      "loss": 3.8521,
+      "step": 2510336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.89018536219759e-06,
+      "loss": 3.864,
+      "step": 2510848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.881799414687069e-06,
+      "loss": 3.8669,
+      "step": 2511360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.87341346717655e-06,
+      "loss": 3.8503,
+      "step": 2511872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.86502751966603e-06,
+      "loss": 3.8523,
+      "step": 2512384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.856657950959241e-06,
+      "loss": 3.8643,
+      "step": 2512896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.848272003448721e-06,
+      "loss": 3.8661,
+      "step": 2513408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.839886055938201e-06,
+      "loss": 3.8569,
+      "step": 2513920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.831500108427681e-06,
+      "loss": 3.853,
+      "step": 2514432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.823114160917161e-06,
+      "loss": 3.8563,
+      "step": 2514944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.814744592210372e-06,
+      "loss": 3.8554,
+      "step": 2515456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.806358644699852e-06,
+      "loss": 3.8532,
+      "step": 2515968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.797972697189332e-06,
+      "loss": 3.8636,
+      "step": 2516480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.789586749678812e-06,
+      "loss": 3.851,
+      "step": 2516992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.781217180972023e-06,
+      "loss": 3.8645,
+      "step": 2517504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.772831233461504e-06,
+      "loss": 3.8587,
+      "step": 2518016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.764445285950983e-06,
+      "loss": 3.8463,
+      "step": 2518528
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.993093729019165,
+      "eval_runtime": 298.7933,
+      "eval_samples_per_second": 1277.107,
+      "eval_steps_per_second": 39.911,
+      "step": 2518560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.756059338440463e-06,
+      "loss": 3.8593,
+      "step": 2519040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.747673390929943e-06,
+      "loss": 3.8548,
+      "step": 2519552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.739287443419424e-06,
+      "loss": 3.8645,
+      "step": 2520064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.730901495908903e-06,
+      "loss": 3.8592,
+      "step": 2520576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.722515548398382e-06,
+      "loss": 3.8667,
+      "step": 2521088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.714145979691595e-06,
+      "loss": 3.8517,
+      "step": 2521600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.705760032181073e-06,
+      "loss": 3.8496,
+      "step": 2522112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.697390463474286e-06,
+      "loss": 3.8397,
+      "step": 2522624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.689004515963766e-06,
+      "loss": 3.8716,
+      "step": 2523136
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.680618568453246e-06,
+      "loss": 3.8592,
+      "step": 2523648
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.672232620942724e-06,
+      "loss": 3.8528,
+      "step": 2524160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.663846673432206e-06,
+      "loss": 3.8666,
+      "step": 2524672
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.655493483529147e-06,
+      "loss": 3.8554,
+      "step": 2525184
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.647107536018629e-06,
+      "loss": 3.851,
+      "step": 2525696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.638721588508107e-06,
+      "loss": 3.8491,
+      "step": 2526208
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.630335640997587e-06,
+      "loss": 3.8527,
+      "step": 2526720
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.621949693487067e-06,
+      "loss": 3.8529,
+      "step": 2527232
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.613563745976547e-06,
+      "loss": 3.8492,
+      "step": 2527744
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.605177798466027e-06,
+      "loss": 3.8632,
+      "step": 2528256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.596791850955507e-06,
+      "loss": 3.8746,
+      "step": 2528768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.58842228224872e-06,
+      "loss": 3.8567,
+      "step": 2529280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.580036334738198e-06,
+      "loss": 3.8505,
+      "step": 2529792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.571650387227678e-06,
+      "loss": 3.8561,
+      "step": 2530304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.56326443971716e-06,
+      "loss": 3.8578,
+      "step": 2530816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.554878492206637e-06,
+      "loss": 3.8536,
+      "step": 2531328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.546492544696117e-06,
+      "loss": 3.8475,
+      "step": 2531840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.538106597185597e-06,
+      "loss": 3.8487,
+      "step": 2532352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.52973702847881e-06,
+      "loss": 3.8628,
+      "step": 2532864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.521351080968288e-06,
+      "loss": 3.8399,
+      "step": 2533376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.51296513345777e-06,
+      "loss": 3.8461,
+      "step": 2533888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.50457918594725e-06,
+      "loss": 3.8561,
+      "step": 2534400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.496193238436728e-06,
+      "loss": 3.8582,
+      "step": 2534912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.487807290926208e-06,
+      "loss": 3.8623,
+      "step": 2535424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.47942134341569e-06,
+      "loss": 3.8579,
+      "step": 2535936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.4710517747089e-06,
+      "loss": 3.8563,
+      "step": 2536448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.462665827198379e-06,
+      "loss": 3.8631,
+      "step": 2536960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.45427987968786e-06,
+      "loss": 3.8445,
+      "step": 2537472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.44589393217734e-06,
+      "loss": 3.8549,
+      "step": 2537984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.437507984666818e-06,
+      "loss": 3.8438,
+      "step": 2538496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.429122037156298e-06,
+      "loss": 3.8477,
+      "step": 2539008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.42073608964578e-06,
+      "loss": 3.8537,
+      "step": 2539520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.412366520938991e-06,
+      "loss": 3.8581,
+      "step": 2540032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.40398057342847e-06,
+      "loss": 3.8549,
+      "step": 2540544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.39559462591795e-06,
+      "loss": 3.8487,
+      "step": 2541056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.38720867840743e-06,
+      "loss": 3.8543,
+      "step": 2541568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.378822730896909e-06,
+      "loss": 3.8556,
+      "step": 2542080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.370453162190121e-06,
+      "loss": 3.8573,
+      "step": 2542592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.362067214679601e-06,
+      "loss": 3.8461,
+      "step": 2543104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.353681267169081e-06,
+      "loss": 3.8434,
+      "step": 2543616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.345295319658561e-06,
+      "loss": 3.857,
+      "step": 2544128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.336909372148041e-06,
+      "loss": 3.864,
+      "step": 2544640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.328539803441252e-06,
+      "loss": 3.8501,
+      "step": 2545152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.320153855930732e-06,
+      "loss": 3.8508,
+      "step": 2545664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.311767908420212e-06,
+      "loss": 3.8486,
+      "step": 2546176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.303381960909692e-06,
+      "loss": 3.8464,
+      "step": 2546688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.295012392202904e-06,
+      "loss": 3.8508,
+      "step": 2547200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.286626444692383e-06,
+      "loss": 3.8415,
+      "step": 2547712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.278240497181863e-06,
+      "loss": 3.8584,
+      "step": 2548224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.269854549671344e-06,
+      "loss": 3.8543,
+      "step": 2548736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.261468602160823e-06,
+      "loss": 3.8536,
+      "step": 2549248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.253099033454035e-06,
+      "loss": 3.8387,
+      "step": 2549760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.244713085943515e-06,
+      "loss": 3.8608,
+      "step": 2550272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.236327138432995e-06,
+      "loss": 3.8377,
+      "step": 2550784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.227941190922473e-06,
+      "loss": 3.8505,
+      "step": 2551296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.219555243411955e-06,
+      "loss": 3.8606,
+      "step": 2551808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.211185674705166e-06,
+      "loss": 3.861,
+      "step": 2552320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.202799727194646e-06,
+      "loss": 3.8426,
+      "step": 2552832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.194413779684126e-06,
+      "loss": 3.8449,
+      "step": 2553344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.186027832173606e-06,
+      "loss": 3.8429,
+      "step": 2553856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.177641884663086e-06,
+      "loss": 3.8513,
+      "step": 2554368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.169255937152564e-06,
+      "loss": 3.8558,
+      "step": 2554880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.160869989642045e-06,
+      "loss": 3.8528,
+      "step": 2555392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.152484042131525e-06,
+      "loss": 3.8592,
+      "step": 2555904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.144130852228469e-06,
+      "loss": 3.8623,
+      "step": 2556416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.135744904717947e-06,
+      "loss": 3.8617,
+      "step": 2556928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.12737533601116e-06,
+      "loss": 3.8457,
+      "step": 2557440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.11898938850064e-06,
+      "loss": 3.8585,
+      "step": 2557952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.11060344099012e-06,
+      "loss": 3.8597,
+      "step": 2558464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.102217493479598e-06,
+      "loss": 3.8428,
+      "step": 2558976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.09384792477281e-06,
+      "loss": 3.8608,
+      "step": 2559488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.08546197726229e-06,
+      "loss": 3.8471,
+      "step": 2560000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.07707602975177e-06,
+      "loss": 3.8707,
+      "step": 2560512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.06869008224125e-06,
+      "loss": 3.8577,
+      "step": 2561024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.06032051353446e-06,
+      "loss": 3.8389,
+      "step": 2561536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.05193456602394e-06,
+      "loss": 3.8535,
+      "step": 2562048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.04354861851342e-06,
+      "loss": 3.8462,
+      "step": 2562560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.0351626710029e-06,
+      "loss": 3.8615,
+      "step": 2563072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.02677672349238e-06,
+      "loss": 3.8539,
+      "step": 2563584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.01839077598186e-06,
+      "loss": 3.8534,
+      "step": 2564096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 8.01000482847134e-06,
+      "loss": 3.8531,
+      "step": 2564608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 8.001635259764551e-06,
+      "loss": 3.8472,
+      "step": 2565120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.993249312254033e-06,
+      "loss": 3.8407,
+      "step": 2565632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.984863364743511e-06,
+      "loss": 3.8403,
+      "step": 2566144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.976477417232991e-06,
+      "loss": 3.8515,
+      "step": 2566656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.968091469722471e-06,
+      "loss": 3.8602,
+      "step": 2567168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.959705522211951e-06,
+      "loss": 3.852,
+      "step": 2567680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.951319574701431e-06,
+      "loss": 3.8433,
+      "step": 2568192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.942933627190911e-06,
+      "loss": 3.8469,
+      "step": 2568704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.934564058484123e-06,
+      "loss": 3.8522,
+      "step": 2569216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.926178110973602e-06,
+      "loss": 3.8622,
+      "step": 2569728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.917808542266814e-06,
+      "loss": 3.8514,
+      "step": 2570240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.909422594756294e-06,
+      "loss": 3.852,
+      "step": 2570752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.901036647245774e-06,
+      "loss": 3.8516,
+      "step": 2571264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.892650699735252e-06,
+      "loss": 3.8593,
+      "step": 2571776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.884264752224734e-06,
+      "loss": 3.8392,
+      "step": 2572288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.875878804714214e-06,
+      "loss": 3.851,
+      "step": 2572800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.867492857203692e-06,
+      "loss": 3.8556,
+      "step": 2573312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.859106909693172e-06,
+      "loss": 3.8495,
+      "step": 2573824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.850720962182654e-06,
+      "loss": 3.8423,
+      "step": 2574336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.842335014672132e-06,
+      "loss": 3.8485,
+      "step": 2574848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.833949067161612e-06,
+      "loss": 3.8486,
+      "step": 2575360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.825563119651092e-06,
+      "loss": 3.85,
+      "step": 2575872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.817193550944305e-06,
+      "loss": 3.8526,
+      "step": 2576384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.808807603433783e-06,
+      "loss": 3.8496,
+      "step": 2576896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.800421655923264e-06,
+      "loss": 3.8588,
+      "step": 2577408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.792035708412744e-06,
+      "loss": 3.8613,
+      "step": 2577920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.783666139705955e-06,
+      "loss": 3.853,
+      "step": 2578432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.775280192195435e-06,
+      "loss": 3.8572,
+      "step": 2578944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.766894244684915e-06,
+      "loss": 3.8555,
+      "step": 2579456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.758508297174395e-06,
+      "loss": 3.8515,
+      "step": 2579968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.750122349663875e-06,
+      "loss": 3.8461,
+      "step": 2580480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.741769159760817e-06,
+      "loss": 3.8488,
+      "step": 2580992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.733383212250298e-06,
+      "loss": 3.8533,
+      "step": 2581504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.724997264739778e-06,
+      "loss": 3.8565,
+      "step": 2582016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.716611317229256e-06,
+      "loss": 3.8537,
+      "step": 2582528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.708225369718736e-06,
+      "loss": 3.854,
+      "step": 2583040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.699855801011949e-06,
+      "loss": 3.8508,
+      "step": 2583552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.691469853501427e-06,
+      "loss": 3.8516,
+      "step": 2584064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.683083905990907e-06,
+      "loss": 3.8507,
+      "step": 2584576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.674697958480389e-06,
+      "loss": 3.859,
+      "step": 2585088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.666312010969869e-06,
+      "loss": 3.8463,
+      "step": 2585600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.657926063459347e-06,
+      "loss": 3.8368,
+      "step": 2586112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.64955649475256e-06,
+      "loss": 3.8459,
+      "step": 2586624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.64117054724204e-06,
+      "loss": 3.8603,
+      "step": 2587136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.632784599731518e-06,
+      "loss": 3.8625,
+      "step": 2587648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.6243986522209985e-06,
+      "loss": 3.8531,
+      "step": 2588160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.616012704710479e-06,
+      "loss": 3.8467,
+      "step": 2588672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.607626757199959e-06,
+      "loss": 3.8617,
+      "step": 2589184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.599257188493169e-06,
+      "loss": 3.8593,
+      "step": 2589696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.59087124098265e-06,
+      "loss": 3.8528,
+      "step": 2590208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.58248529347213e-06,
+      "loss": 3.849,
+      "step": 2590720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.574099345961609e-06,
+      "loss": 3.8531,
+      "step": 2591232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.565713398451089e-06,
+      "loss": 3.8543,
+      "step": 2591744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.55732745094057e-06,
+      "loss": 3.8481,
+      "step": 2592256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.5489578822337815e-06,
+      "loss": 3.8626,
+      "step": 2592768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.540571934723261e-06,
+      "loss": 3.8482,
+      "step": 2593280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.5321859872127405e-06,
+      "loss": 3.8612,
+      "step": 2593792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.523800039702221e-06,
+      "loss": 3.8548,
+      "step": 2594304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.5154140921917e-06,
+      "loss": 3.8445,
+      "step": 2594816
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.9923651218414307,
+      "eval_runtime": 294.323,
+      "eval_samples_per_second": 1296.504,
+      "eval_steps_per_second": 40.517,
+      "step": 2594880
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.507044523484911e-06,
+      "loss": 3.8579,
+      "step": 2595328
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.498658575974392e-06,
+      "loss": 3.8532,
+      "step": 2595840
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.490272628463872e-06,
+      "loss": 3.8572,
+      "step": 2596352
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.481886680953351e-06,
+      "loss": 3.8544,
+      "step": 2596864
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.473500733442831e-06,
+      "loss": 3.8641,
+      "step": 2597376
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.465114785932312e-06,
+      "loss": 3.8483,
+      "step": 2597888
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.456728838421791e-06,
+      "loss": 3.8472,
+      "step": 2598400
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.448342890911271e-06,
+      "loss": 3.8364,
+      "step": 2598912
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.4399733222044826e-06,
+      "loss": 3.8704,
+      "step": 2599424
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.4315873746939625e-06,
+      "loss": 3.8566,
+      "step": 2599936
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.423201427183442e-06,
+      "loss": 3.8488,
+      "step": 2600448
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.4148154796729224e-06,
+      "loss": 3.8654,
+      "step": 2600960
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.406445910966134e-06,
+      "loss": 3.8498,
+      "step": 2601472
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.398059963455614e-06,
+      "loss": 3.8533,
+      "step": 2601984
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.389690394748825e-06,
+      "loss": 3.8444,
+      "step": 2602496
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.381304447238305e-06,
+      "loss": 3.8471,
+      "step": 2603008
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.372934878531516e-06,
+      "loss": 3.8532,
+      "step": 2603520
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.364548931020996e-06,
+      "loss": 3.8443,
+      "step": 2604032
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.3561629835104755e-06,
+      "loss": 3.8623,
+      "step": 2604544
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.347777035999956e-06,
+      "loss": 3.8673,
+      "step": 2605056
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.339391088489436e-06,
+      "loss": 3.8555,
+      "step": 2605568
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.331005140978915e-06,
+      "loss": 3.8506,
+      "step": 2606080
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.322619193468395e-06,
+      "loss": 3.8544,
+      "step": 2606592
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.314233245957876e-06,
+      "loss": 3.8553,
+      "step": 2607104
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.305847298447354e-06,
+      "loss": 3.8473,
+      "step": 2607616
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.297461350936835e-06,
+      "loss": 3.8471,
+      "step": 2608128
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.289075403426315e-06,
+      "loss": 3.8393,
+      "step": 2608640
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.280689455915794e-06,
+      "loss": 3.8615,
+      "step": 2609152
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.272303508405274e-06,
+      "loss": 3.8344,
+      "step": 2609664
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.263917560894755e-06,
+      "loss": 3.8466,
+      "step": 2610176
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.255547992187967e-06,
+      "loss": 3.8554,
+      "step": 2610688
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.247162044677446e-06,
+      "loss": 3.8586,
+      "step": 2611200
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.238776097166926e-06,
+      "loss": 3.8558,
+      "step": 2611712
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.2303901496564065e-06,
+      "loss": 3.855,
+      "step": 2612224
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.222036959753349e-06,
+      "loss": 3.852,
+      "step": 2612736
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.213651012242828e-06,
+      "loss": 3.8604,
+      "step": 2613248
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.205265064732308e-06,
+      "loss": 3.8458,
+      "step": 2613760
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.196879117221789e-06,
+      "loss": 3.8503,
+      "step": 2614272
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.188493169711268e-06,
+      "loss": 3.8386,
+      "step": 2614784
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.180107222200748e-06,
+      "loss": 3.8463,
+      "step": 2615296
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.171721274690228e-06,
+      "loss": 3.8465,
+      "step": 2615808
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.163335327179709e-06,
+      "loss": 3.8571,
+      "step": 2616320
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.154965758472919e-06,
+      "loss": 3.8504,
+      "step": 2616832
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.146579810962399e-06,
+      "loss": 3.846,
+      "step": 2617344
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.138193863451879e-06,
+      "loss": 3.8528,
+      "step": 2617856
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.1298079159413585e-06,
+      "loss": 3.8524,
+      "step": 2618368
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.121421968430838e-06,
+      "loss": 3.8543,
+      "step": 2618880
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.113036020920319e-06,
+      "loss": 3.8447,
+      "step": 2619392
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.104650073409799e-06,
+      "loss": 3.8367,
+      "step": 2619904
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.096264125899278e-06,
+      "loss": 3.8615,
+      "step": 2620416
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.08789455719249e-06,
+      "loss": 3.8579,
+      "step": 2620928
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.07950860968197e-06,
+      "loss": 3.8483,
+      "step": 2621440
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.071122662171449e-06,
+      "loss": 3.8509,
+      "step": 2621952
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.062753093464661e-06,
+      "loss": 3.8444,
+      "step": 2622464
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.0543671459541414e-06,
+      "loss": 3.8454,
+      "step": 2622976
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.045997577247353e-06,
+      "loss": 3.8497,
+      "step": 2623488
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.037611629736831e-06,
+      "loss": 3.8404,
+      "step": 2624000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.029225682226312e-06,
+      "loss": 3.8557,
+      "step": 2624512
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.020839734715792e-06,
+      "loss": 3.85,
+      "step": 2625024
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.012453787205273e-06,
+      "loss": 3.8469,
+      "step": 2625536
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 7.004067839694751e-06,
+      "loss": 3.8393,
+      "step": 2626048
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.995681892184232e-06,
+      "loss": 3.8575,
+      "step": 2626560
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.987295944673712e-06,
+      "loss": 3.8369,
+      "step": 2627072
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.978909997163191e-06,
+      "loss": 3.8481,
+      "step": 2627584
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.970524049652671e-06,
+      "loss": 3.8526,
+      "step": 2628096
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.962154480945883e-06,
+      "loss": 3.8594,
+      "step": 2628608
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.953768533435363e-06,
+      "loss": 3.8389,
+      "step": 2629120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.9453825859248425e-06,
+      "loss": 3.8459,
+      "step": 2629632
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.9369966384143225e-06,
+      "loss": 3.8318,
+      "step": 2630144
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.928610690903803e-06,
+      "loss": 3.8515,
+      "step": 2630656
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.9202247433932815e-06,
+      "loss": 3.8538,
+      "step": 2631168
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.911838795882762e-06,
+      "loss": 3.8525,
+      "step": 2631680
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.903469227175974e-06,
+      "loss": 3.8535,
+      "step": 2632192
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.895083279665454e-06,
+      "loss": 3.8576,
+      "step": 2632704
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.886697332154933e-06,
+      "loss": 3.8626,
+      "step": 2633216
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.878311384644413e-06,
+      "loss": 3.8428,
+      "step": 2633728
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.869925437133894e-06,
+      "loss": 3.8551,
+      "step": 2634240
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.861555868427104e-06,
+      "loss": 3.8614,
+      "step": 2634752
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.8531699209165845e-06,
+      "loss": 3.8418,
+      "step": 2635264
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.8447839734060645e-06,
+      "loss": 3.8541,
+      "step": 2635776
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.8363980258955445e-06,
+      "loss": 3.8431,
+      "step": 2636288
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.8280120783850236e-06,
+      "loss": 3.8681,
+      "step": 2636800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.819626130874504e-06,
+      "loss": 3.8566,
+      "step": 2637312
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.811256562167716e-06,
+      "loss": 3.8348,
+      "step": 2637824
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.802870614657194e-06,
+      "loss": 3.8479,
+      "step": 2638336
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.794484667146675e-06,
+      "loss": 3.85,
+      "step": 2638848
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.786098719636155e-06,
+      "loss": 3.8577,
+      "step": 2639360
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.777712772125636e-06,
+      "loss": 3.8536,
+      "step": 2639872
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 6.769343203418846e-06,
+      "loss": 3.8503,
+      "step": 2640384
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.7609572559083266e-06,
+      "loss": 3.85,
+      "step": 2640896
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.7525713083978065e-06,
+      "loss": 3.8468,
+      "step": 2641408
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.744185360887286e-06,
+      "loss": 3.8435,
+      "step": 2641920
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.735799413376766e-06,
+      "loss": 3.8333,
+      "step": 2642432
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.727413465866246e-06,
+      "loss": 3.8505,
+      "step": 2642944
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.719043897159458e-06,
+      "loss": 3.8579,
+      "step": 2643456
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.710657949648936e-06,
+      "loss": 3.8506,
+      "step": 2643968
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.702272002138417e-06,
+      "loss": 3.8412,
+      "step": 2644480
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.693886054627897e-06,
+      "loss": 3.8473,
+      "step": 2644992
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.685500107117376e-06,
+      "loss": 3.8461,
+      "step": 2645504
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.677114159606856e-06,
+      "loss": 3.8619,
+      "step": 2646016
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.668744590900068e-06,
+      "loss": 3.8473,
+      "step": 2646528
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.6603586433895486e-06,
+      "loss": 3.8489,
+      "step": 2647040
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.651972695879028e-06,
+      "loss": 3.845,
+      "step": 2647552
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.643586748368508e-06,
+      "loss": 3.8559,
+      "step": 2648064
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.6352008008579884e-06,
+      "loss": 3.8352,
+      "step": 2648576
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.626831232151198e-06,
+      "loss": 3.8515,
+      "step": 2649088
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.618445284640678e-06,
+      "loss": 3.8478,
+      "step": 2649600
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.610059337130159e-06,
+      "loss": 3.8496,
+      "step": 2650112
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.601673389619639e-06,
+      "loss": 3.8407,
+      "step": 2650624
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.593287442109118e-06,
+      "loss": 3.8464,
+      "step": 2651136
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.584901494598598e-06,
+      "loss": 3.8478,
+      "step": 2651648
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.57653192589181e-06,
+      "loss": 3.8461,
+      "step": 2652160
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.568145978381289e-06,
+      "loss": 3.8504,
+      "step": 2652672
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.55976003087077e-06,
+      "loss": 3.8497,
+      "step": 2653184
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.55137408336025e-06,
+      "loss": 3.8547,
+      "step": 2653696
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.54298813584973e-06,
+      "loss": 3.8581,
+      "step": 2654208
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.534602188339209e-06,
+      "loss": 3.8534,
+      "step": 2654720
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.5262162408286895e-06,
+      "loss": 3.8549,
+      "step": 2655232
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.5178302933181695e-06,
+      "loss": 3.8516,
+      "step": 2655744
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.509460724611379e-06,
+      "loss": 3.8449,
+      "step": 2656256
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.50107477710086e-06,
+      "loss": 3.8463,
+      "step": 2656768
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.492705208394072e-06,
+      "loss": 3.848,
+      "step": 2657280
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.484319260883552e-06,
+      "loss": 3.8449,
+      "step": 2657792
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.475933313373031e-06,
+      "loss": 3.8558,
+      "step": 2658304
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.467547365862512e-06,
+      "loss": 3.848,
+      "step": 2658816
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.459177797155723e-06,
+      "loss": 3.8532,
+      "step": 2659328
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.450791849645203e-06,
+      "loss": 3.8476,
+      "step": 2659840
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.4424059021346824e-06,
+      "loss": 3.8452,
+      "step": 2660352
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.434019954624162e-06,
+      "loss": 3.8455,
+      "step": 2660864
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.425634007113643e-06,
+      "loss": 3.8598,
+      "step": 2661376
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.417264438406853e-06,
+      "loss": 3.8433,
+      "step": 2661888
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.408878490896333e-06,
+      "loss": 3.8419,
+      "step": 2662400
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.400492543385814e-06,
+      "loss": 3.8386,
+      "step": 2662912
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.392106595875294e-06,
+      "loss": 3.8545,
+      "step": 2663424
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.383720648364773e-06,
+      "loss": 3.8572,
+      "step": 2663936
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.375351079657985e-06,
+      "loss": 3.8543,
+      "step": 2664448
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.3669651321474645e-06,
+      "loss": 3.8456,
+      "step": 2664960
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.358579184636944e-06,
+      "loss": 3.8601,
+      "step": 2665472
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.3501932371264245e-06,
+      "loss": 3.8514,
+      "step": 2665984
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.341807289615904e-06,
+      "loss": 3.8537,
+      "step": 2666496
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.333421342105385e-06,
+      "loss": 3.8459,
+      "step": 2667008
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.325051773398595e-06,
+      "loss": 3.8525,
+      "step": 2667520
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.316665825888075e-06,
+      "loss": 3.8541,
+      "step": 2668032
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.308279878377556e-06,
+      "loss": 3.842,
+      "step": 2668544
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.299893930867035e-06,
+      "loss": 3.8609,
+      "step": 2669056
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.291507983356515e-06,
+      "loss": 3.8517,
+      "step": 2669568
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.283122035845995e-06,
+      "loss": 3.8552,
+      "step": 2670080
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.2747524671392066e-06,
+      "loss": 3.8525,
+      "step": 2670592
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 6.266366519628686e-06,
+      "loss": 3.8404,
+      "step": 2671104
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 3.9919233322143555,
+      "eval_runtime": 290.2835,
+      "eval_samples_per_second": 1314.546,
+      "eval_steps_per_second": 41.081,
+      "step": 2671200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.2579805721181665e-06,
+      "loss": 3.8543,
+      "step": 2671616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.249594624607646e-06,
+      "loss": 3.8539,
+      "step": 2672128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.241208677097126e-06,
+      "loss": 3.8538,
+      "step": 2672640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.2328227295866055e-06,
+      "loss": 3.8495,
+      "step": 2673152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.224436782076086e-06,
+      "loss": 3.8666,
+      "step": 2673664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.216050834565565e-06,
+      "loss": 3.8453,
+      "step": 2674176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.207664887055045e-06,
+      "loss": 3.8422,
+      "step": 2674688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.199278939544525e-06,
+      "loss": 3.8371,
+      "step": 2675200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.190892992034005e-06,
+      "loss": 3.862,
+      "step": 2675712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.182507044523485e-06,
+      "loss": 3.8585,
+      "step": 2676224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.174121097012965e-06,
+      "loss": 3.8426,
+      "step": 2676736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.165735149502444e-06,
+      "loss": 3.8611,
+      "step": 2677248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.157365580795657e-06,
+      "loss": 3.8488,
+      "step": 2677760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.148979633285136e-06,
+      "loss": 3.8489,
+      "step": 2678272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.140593685774617e-06,
+      "loss": 3.8452,
+      "step": 2678784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.132207738264096e-06,
+      "loss": 3.8456,
+      "step": 2679296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.123821790753577e-06,
+      "loss": 3.8496,
+      "step": 2679808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.115435843243056e-06,
+      "loss": 3.8454,
+      "step": 2680320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.107049895732536e-06,
+      "loss": 3.8576,
+      "step": 2680832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.098680327025747e-06,
+      "loss": 3.8642,
+      "step": 2681344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.090294379515226e-06,
+      "loss": 3.857,
+      "step": 2681856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.081908432004707e-06,
+      "loss": 3.8483,
+      "step": 2682368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.073522484494186e-06,
+      "loss": 3.8471,
+      "step": 2682880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.065152915787398e-06,
+      "loss": 3.8571,
+      "step": 2683392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.056766968276878e-06,
+      "loss": 3.8437,
+      "step": 2683904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.048381020766358e-06,
+      "loss": 3.8467,
+      "step": 2684416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.039995073255838e-06,
+      "loss": 3.839,
+      "step": 2684928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.031609125745318e-06,
+      "loss": 3.8566,
+      "step": 2685440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.023223178234798e-06,
+      "loss": 3.8356,
+      "step": 2685952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.014837230724278e-06,
+      "loss": 3.8436,
+      "step": 2686464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.006451283213758e-06,
+      "loss": 3.8539,
+      "step": 2686976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.998065335703238e-06,
+      "loss": 3.8527,
+      "step": 2687488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.989679388192717e-06,
+      "loss": 3.8547,
+      "step": 2688000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9812934406821975e-06,
+      "loss": 3.8564,
+      "step": 2688512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.97294025077914e-06,
+      "loss": 3.8513,
+      "step": 2689024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.96455430326862e-06,
+      "loss": 3.8569,
+      "step": 2689536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9561683557581e-06,
+      "loss": 3.8464,
+      "step": 2690048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.94778240824758e-06,
+      "loss": 3.8418,
+      "step": 2690560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.93939646073706e-06,
+      "loss": 3.8401,
+      "step": 2691072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9310268920302715e-06,
+      "loss": 3.8448,
+      "step": 2691584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9226409445197506e-06,
+      "loss": 3.8433,
+      "step": 2692096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.914254997009231e-06,
+      "loss": 3.8576,
+      "step": 2692608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9058690494987105e-06,
+      "loss": 3.8458,
+      "step": 2693120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.8974831019881904e-06,
+      "loss": 3.845,
+      "step": 2693632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.88909715447767e-06,
+      "loss": 3.8544,
+      "step": 2694144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.88071120696715e-06,
+      "loss": 3.8453,
+      "step": 2694656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.87232525945663e-06,
+      "loss": 3.8582,
+      "step": 2695168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.863955690749841e-06,
+      "loss": 3.8431,
+      "step": 2695680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.855569743239322e-06,
+      "loss": 3.8276,
+      "step": 2696192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.847183795728802e-06,
+      "loss": 3.8635,
+      "step": 2696704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.838797848218281e-06,
+      "loss": 3.8524,
+      "step": 2697216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.830411900707762e-06,
+      "loss": 3.8501,
+      "step": 2697728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.822025953197241e-06,
+      "loss": 3.8467,
+      "step": 2698240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.813640005686721e-06,
+      "loss": 3.843,
+      "step": 2698752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.805254058176201e-06,
+      "loss": 3.8434,
+      "step": 2699264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.7968844894694116e-06,
+      "loss": 3.8464,
+      "step": 2699776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.788498541958892e-06,
+      "loss": 3.8379,
+      "step": 2700288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.7801125944483715e-06,
+      "loss": 3.8476,
+      "step": 2700800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.771726646937852e-06,
+      "loss": 3.8536,
+      "step": 2701312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.763373457034795e-06,
+      "loss": 3.8421,
+      "step": 2701824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.754987509524275e-06,
+      "loss": 3.8348,
+      "step": 2702336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.746601562013755e-06,
+      "loss": 3.8568,
+      "step": 2702848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.738215614503235e-06,
+      "loss": 3.8347,
+      "step": 2703360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.729846045796445e-06,
+      "loss": 3.8414,
+      "step": 2703872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.721460098285926e-06,
+      "loss": 3.8542,
+      "step": 2704384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.713074150775405e-06,
+      "loss": 3.8563,
+      "step": 2704896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.704688203264885e-06,
+      "loss": 3.8371,
+      "step": 2705408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.696302255754365e-06,
+      "loss": 3.8439,
+      "step": 2705920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.687932687047577e-06,
+      "loss": 3.8278,
+      "step": 2706432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.679546739537057e-06,
+      "loss": 3.8541,
+      "step": 2706944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.671160792026537e-06,
+      "loss": 3.8492,
+      "step": 2707456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.662774844516017e-06,
+      "loss": 3.8514,
+      "step": 2707968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.654388897005497e-06,
+      "loss": 3.8487,
+      "step": 2708480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.646002949494976e-06,
+      "loss": 3.8571,
+      "step": 2708992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.637617001984457e-06,
+      "loss": 3.8614,
+      "step": 2709504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.629247433277667e-06,
+      "loss": 3.8403,
+      "step": 2710016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.620861485767147e-06,
+      "loss": 3.8512,
+      "step": 2710528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.612475538256627e-06,
+      "loss": 3.8582,
+      "step": 2711040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.604089590746107e-06,
+      "loss": 3.8396,
+      "step": 2711552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.595720022039318e-06,
+      "loss": 3.854,
+      "step": 2712064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.58735045333253e-06,
+      "loss": 3.8444,
+      "step": 2712576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.57896450582201e-06,
+      "loss": 3.8591,
+      "step": 2713088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.57057855831149e-06,
+      "loss": 3.8556,
+      "step": 2713600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.5621926108009696e-06,
+      "loss": 3.8353,
+      "step": 2714112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.5538066632904495e-06,
+      "loss": 3.8465,
+      "step": 2714624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.5454207157799295e-06,
+      "loss": 3.8435,
+      "step": 2715136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.5370347682694094e-06,
+      "loss": 3.8573,
+      "step": 2715648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.528648820758889e-06,
+      "loss": 3.8487,
+      "step": 2716160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.520262873248369e-06,
+      "loss": 3.8457,
+      "step": 2716672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.511893304541581e-06,
+      "loss": 3.8455,
+      "step": 2717184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.50350735703106e-06,
+      "loss": 3.8467,
+      "step": 2717696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.49512140952054e-06,
+      "loss": 3.842,
+      "step": 2718208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.48673546201002e-06,
+      "loss": 3.8333,
+      "step": 2718720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.478365893303232e-06,
+      "loss": 3.847,
+      "step": 2719232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.469996324596443e-06,
+      "loss": 3.8564,
+      "step": 2719744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.461610377085922e-06,
+      "loss": 3.8554,
+      "step": 2720256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.453224429575403e-06,
+      "loss": 3.8329,
+      "step": 2720768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.444838482064882e-06,
+      "loss": 3.8459,
+      "step": 2721280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.436452534554363e-06,
+      "loss": 3.8429,
+      "step": 2721792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.428066587043842e-06,
+      "loss": 3.8563,
+      "step": 2722304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.419680639533322e-06,
+      "loss": 3.8457,
+      "step": 2722816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.411294692022802e-06,
+      "loss": 3.8491,
+      "step": 2723328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.402908744512282e-06,
+      "loss": 3.842,
+      "step": 2723840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.394522797001762e-06,
+      "loss": 3.8571,
+      "step": 2724352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.386136849491242e-06,
+      "loss": 3.833,
+      "step": 2724864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.377767280784454e-06,
+      "loss": 3.8488,
+      "step": 2725376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.3693813332739336e-06,
+      "loss": 3.8437,
+      "step": 2725888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.360995385763413e-06,
+      "loss": 3.8468,
+      "step": 2726400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.3526094382528935e-06,
+      "loss": 3.8383,
+      "step": 2726912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.344223490742373e-06,
+      "loss": 3.8493,
+      "step": 2727424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.335837543231853e-06,
+      "loss": 3.8405,
+      "step": 2727936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.327467974525064e-06,
+      "loss": 3.8442,
+      "step": 2728448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.319082027014544e-06,
+      "loss": 3.8457,
+      "step": 2728960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.310696079504024e-06,
+      "loss": 3.8479,
+      "step": 2729472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.302310131993503e-06,
+      "loss": 3.8542,
+      "step": 2729984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.293924184482984e-06,
+      "loss": 3.8553,
+      "step": 2730496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.285538236972463e-06,
+      "loss": 3.8495,
+      "step": 2731008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.277168668265675e-06,
+      "loss": 3.8541,
+      "step": 2731520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.268782720755155e-06,
+      "loss": 3.8475,
+      "step": 2732032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.260396773244635e-06,
+      "loss": 3.8423,
+      "step": 2732544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.252010825734115e-06,
+      "loss": 3.8426,
+      "step": 2733056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.243641257027326e-06,
+      "loss": 3.8459,
+      "step": 2733568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.235271688320537e-06,
+      "loss": 3.846,
+      "step": 2734080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.226885740810018e-06,
+      "loss": 3.8528,
+      "step": 2734592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.218499793299497e-06,
+      "loss": 3.844,
+      "step": 2735104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.210113845788977e-06,
+      "loss": 3.851,
+      "step": 2735616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.201727898278457e-06,
+      "loss": 3.8436,
+      "step": 2736128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.193341950767937e-06,
+      "loss": 3.8442,
+      "step": 2736640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.184956003257417e-06,
+      "loss": 3.8439,
+      "step": 2737152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.176570055746897e-06,
+      "loss": 3.8586,
+      "step": 2737664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.168184108236377e-06,
+      "loss": 3.8436,
+      "step": 2738176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.159798160725857e-06,
+      "loss": 3.8369,
+      "step": 2738688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.151412213215337e-06,
+      "loss": 3.8333,
+      "step": 2739200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.143042644508548e-06,
+      "loss": 3.8504,
+      "step": 2739712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.134656696998027e-06,
+      "loss": 3.8556,
+      "step": 2740224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.126270749487507e-06,
+      "loss": 3.8524,
+      "step": 2740736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.117884801976987e-06,
+      "loss": 3.8483,
+      "step": 2741248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.109498854466467e-06,
+      "loss": 3.8595,
+      "step": 2741760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.101129285759679e-06,
+      "loss": 3.8492,
+      "step": 2742272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.092743338249158e-06,
+      "loss": 3.8497,
+      "step": 2742784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.084357390738639e-06,
+      "loss": 3.8454,
+      "step": 2743296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.075971443228118e-06,
+      "loss": 3.8504,
+      "step": 2743808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.067585495717598e-06,
+      "loss": 3.8521,
+      "step": 2744320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.059199548207079e-06,
+      "loss": 3.8388,
+      "step": 2744832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.0508299795002894e-06,
+      "loss": 3.8572,
+      "step": 2745344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.042444031989769e-06,
+      "loss": 3.8535,
+      "step": 2745856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.034058084479249e-06,
+      "loss": 3.8467,
+      "step": 2746368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.025672136968729e-06,
+      "loss": 3.8493,
+      "step": 2746880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5.017318947065672e-06,
+      "loss": 3.8438,
+      "step": 2747392
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.9912352561950684,
+      "eval_runtime": 312.8585,
+      "eval_samples_per_second": 1219.692,
+      "eval_steps_per_second": 38.116,
+      "step": 2747520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.008932999555152e-06,
+      "loss": 3.8513,
+      "step": 2747904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.000547052044632e-06,
+      "loss": 3.8524,
+      "step": 2748416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.992161104534112e-06,
+      "loss": 3.8545,
+      "step": 2748928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.983775157023592e-06,
+      "loss": 3.845,
+      "step": 2749440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9753892095130715e-06,
+      "loss": 3.8626,
+      "step": 2749952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9670032620025515e-06,
+      "loss": 3.8474,
+      "step": 2750464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9586173144920315e-06,
+      "loss": 3.8394,
+      "step": 2750976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.950231366981511e-06,
+      "loss": 3.8392,
+      "step": 2751488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.941845419470991e-06,
+      "loss": 3.8545,
+      "step": 2752000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9334594719604705e-06,
+      "loss": 3.854,
+      "step": 2752512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.925073524449951e-06,
+      "loss": 3.8445,
+      "step": 2753024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.91668757693943e-06,
+      "loss": 3.8566,
+      "step": 2753536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.908318008232642e-06,
+      "loss": 3.8485,
+      "step": 2754048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.899932060722122e-06,
+      "loss": 3.8466,
+      "step": 2754560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.891562492015334e-06,
+      "loss": 3.8416,
+      "step": 2755072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.8831765445048136e-06,
+      "loss": 3.8428,
+      "step": 2755584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.8747905969942935e-06,
+      "loss": 3.8441,
+      "step": 2756096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.8664046494837735e-06,
+      "loss": 3.8424,
+      "step": 2756608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.858018701973253e-06,
+      "loss": 3.8569,
+      "step": 2757120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.849632754462733e-06,
+      "loss": 3.8599,
+      "step": 2757632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.8412468069522125e-06,
+      "loss": 3.8561,
+      "step": 2758144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.832860859441693e-06,
+      "loss": 3.8474,
+      "step": 2758656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.824474911931172e-06,
+      "loss": 3.8447,
+      "step": 2759168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.816088964420652e-06,
+      "loss": 3.8512,
+      "step": 2759680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.807703016910132e-06,
+      "loss": 3.8446,
+      "step": 2760192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.799317069399612e-06,
+      "loss": 3.8446,
+      "step": 2760704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.790931121889092e-06,
+      "loss": 3.8355,
+      "step": 2761216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.782545174378572e-06,
+      "loss": 3.8505,
+      "step": 2761728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.774159226868051e-06,
+      "loss": 3.8397,
+      "step": 2762240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.765773279357532e-06,
+      "loss": 3.8362,
+      "step": 2762752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.757403710650743e-06,
+      "loss": 3.8529,
+      "step": 2763264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.749017763140224e-06,
+      "loss": 3.8468,
+      "step": 2763776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7406481944334345e-06,
+      "loss": 3.8527,
+      "step": 2764288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7322622469229144e-06,
+      "loss": 3.8545,
+      "step": 2764800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.723892678216125e-06,
+      "loss": 3.8454,
+      "step": 2765312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.715506730705606e-06,
+      "loss": 3.8569,
+      "step": 2765824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.707120783195085e-06,
+      "loss": 3.8473,
+      "step": 2766336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.698734835684566e-06,
+      "loss": 3.837,
+      "step": 2766848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.690348888174045e-06,
+      "loss": 3.8339,
+      "step": 2767360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.681962940663525e-06,
+      "loss": 3.8416,
+      "step": 2767872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.673576993153005e-06,
+      "loss": 3.8423,
+      "step": 2768384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.665191045642485e-06,
+      "loss": 3.8548,
+      "step": 2768896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.656805098131965e-06,
+      "loss": 3.8449,
+      "step": 2769408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.648419150621445e-06,
+      "loss": 3.8424,
+      "step": 2769920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.640033203110925e-06,
+      "loss": 3.8544,
+      "step": 2770432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6316636344041364e-06,
+      "loss": 3.8456,
+      "step": 2770944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6232776868936155e-06,
+      "loss": 3.8514,
+      "step": 2771456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.614908118186827e-06,
+      "loss": 3.839,
+      "step": 2771968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.606522170676307e-06,
+      "loss": 3.8236,
+      "step": 2772480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.598136223165787e-06,
+      "loss": 3.8592,
+      "step": 2772992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.589750275655267e-06,
+      "loss": 3.8547,
+      "step": 2773504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.581380706948479e-06,
+      "loss": 3.849,
+      "step": 2774016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.572994759437959e-06,
+      "loss": 3.8421,
+      "step": 2774528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.564608811927438e-06,
+      "loss": 3.8399,
+      "step": 2775040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.5562228644169185e-06,
+      "loss": 3.8384,
+      "step": 2775552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.547836916906398e-06,
+      "loss": 3.8421,
+      "step": 2776064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.5394509693958785e-06,
+      "loss": 3.8408,
+      "step": 2776576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.531081400689089e-06,
+      "loss": 3.8424,
+      "step": 2777088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.522695453178569e-06,
+      "loss": 3.852,
+      "step": 2777600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.514309505668049e-06,
+      "loss": 3.8449,
+      "step": 2778112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.505923558157528e-06,
+      "loss": 3.8327,
+      "step": 2778624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.497537610647009e-06,
+      "loss": 3.8568,
+      "step": 2779136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.489151663136488e-06,
+      "loss": 3.8317,
+      "step": 2779648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.480765715625969e-06,
+      "loss": 3.8353,
+      "step": 2780160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.472379768115448e-06,
+      "loss": 3.8551,
+      "step": 2780672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.463993820604928e-06,
+      "loss": 3.8569,
+      "step": 2781184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.45562425189814e-06,
+      "loss": 3.8333,
+      "step": 2781696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.44723830438762e-06,
+      "loss": 3.843,
+      "step": 2782208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4388523568771e-06,
+      "loss": 3.8266,
+      "step": 2782720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4304664093665795e-06,
+      "loss": 3.8498,
+      "step": 2783232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4220804618560595e-06,
+      "loss": 3.8509,
+      "step": 2783744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4136945143455395e-06,
+      "loss": 3.842,
+      "step": 2784256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4053085668350186e-06,
+      "loss": 3.8483,
+      "step": 2784768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.39693899812823e-06,
+      "loss": 3.8531,
+      "step": 2785280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.388569429421442e-06,
+      "loss": 3.8647,
+      "step": 2785792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.380183481910922e-06,
+      "loss": 3.8354,
+      "step": 2786304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.371797534400402e-06,
+      "loss": 3.8495,
+      "step": 2786816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.363411586889882e-06,
+      "loss": 3.8546,
+      "step": 2787328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.355025639379362e-06,
+      "loss": 3.8344,
+      "step": 2787840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.346639691868842e-06,
+      "loss": 3.853,
+      "step": 2788352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3382537443583216e-06,
+      "loss": 3.8428,
+      "step": 2788864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.329867796847801e-06,
+      "loss": 3.8564,
+      "step": 2789376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.3214818493372815e-06,
+      "loss": 3.8541,
+      "step": 2789888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.313128659434224e-06,
+      "loss": 3.835,
+      "step": 2790400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.304742711923704e-06,
+      "loss": 3.8396,
+      "step": 2790912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.296356764413183e-06,
+      "loss": 3.8422,
+      "step": 2791424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.287970816902664e-06,
+      "loss": 3.8596,
+      "step": 2791936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.279584869392144e-06,
+      "loss": 3.8453,
+      "step": 2792448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.271198921881624e-06,
+      "loss": 3.8424,
+      "step": 2792960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.262812974371104e-06,
+      "loss": 3.8436,
+      "step": 2793472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.254427026860583e-06,
+      "loss": 3.8481,
+      "step": 2793984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.246041079350064e-06,
+      "loss": 3.8409,
+      "step": 2794496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.237687889447006e-06,
+      "loss": 3.8287,
+      "step": 2795008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.229301941936486e-06,
+      "loss": 3.845,
+      "step": 2795520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.220915994425965e-06,
+      "loss": 3.8521,
+      "step": 2796032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.212530046915446e-06,
+      "loss": 3.8523,
+      "step": 2796544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.204144099404925e-06,
+      "loss": 3.8296,
+      "step": 2797056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.195758151894406e-06,
+      "loss": 3.8428,
+      "step": 2797568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.187372204383885e-06,
+      "loss": 3.8422,
+      "step": 2798080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.178986256873365e-06,
+      "loss": 3.8502,
+      "step": 2798592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1706166881665766e-06,
+      "loss": 3.8482,
+      "step": 2799104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1622307406560565e-06,
+      "loss": 3.8469,
+      "step": 2799616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1538447931455365e-06,
+      "loss": 3.8404,
+      "step": 2800128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.145458845635016e-06,
+      "loss": 3.8536,
+      "step": 2800640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.137072898124496e-06,
+      "loss": 3.8301,
+      "step": 2801152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.128686950613976e-06,
+      "loss": 3.8463,
+      "step": 2801664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1203010031034554e-06,
+      "loss": 3.8379,
+      "step": 2802176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.111915055592936e-06,
+      "loss": 3.85,
+      "step": 2802688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.103545486886147e-06,
+      "loss": 3.8322,
+      "step": 2803200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.095175918179359e-06,
+      "loss": 3.8484,
+      "step": 2803712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.086789970668839e-06,
+      "loss": 3.8391,
+      "step": 2804224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.078404023158319e-06,
+      "loss": 3.8416,
+      "step": 2804736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0700180756477985e-06,
+      "loss": 3.8431,
+      "step": 2805248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.061664885744741e-06,
+      "loss": 3.8462,
+      "step": 2805760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.053278938234221e-06,
+      "loss": 3.8492,
+      "step": 2806272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.044892990723701e-06,
+      "loss": 3.8563,
+      "step": 2806784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.036507043213181e-06,
+      "loss": 3.8482,
+      "step": 2807296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.02812109570266e-06,
+      "loss": 3.8509,
+      "step": 2807808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.019735148192141e-06,
+      "loss": 3.8456,
+      "step": 2808320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.01134920068162e-06,
+      "loss": 3.8397,
+      "step": 2808832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.002963253171101e-06,
+      "loss": 3.839,
+      "step": 2809344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.99457730566058e-06,
+      "loss": 3.8496,
+      "step": 2809856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.986207736953792e-06,
+      "loss": 3.8416,
+      "step": 2810368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.977821789443271e-06,
+      "loss": 3.8503,
+      "step": 2810880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.969435841932751e-06,
+      "loss": 3.8426,
+      "step": 2811392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.961049894422231e-06,
+      "loss": 3.8474,
+      "step": 2811904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.952663946911711e-06,
+      "loss": 3.8439,
+      "step": 2812416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.944294378204923e-06,
+      "loss": 3.8434,
+      "step": 2812928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.935908430694402e-06,
+      "loss": 3.8387,
+      "step": 2813440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.927522483183883e-06,
+      "loss": 3.8559,
+      "step": 2813952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.919136535673362e-06,
+      "loss": 3.8437,
+      "step": 2814464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.910750588162842e-06,
+      "loss": 3.8379,
+      "step": 2814976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.902364640652322e-06,
+      "loss": 3.8298,
+      "step": 2815488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.893978693141802e-06,
+      "loss": 3.8501,
+      "step": 2816000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8856091244350134e-06,
+      "loss": 3.8526,
+      "step": 2816512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.877223176924493e-06,
+      "loss": 3.8511,
+      "step": 2817024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.868837229413973e-06,
+      "loss": 3.8469,
+      "step": 2817536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.860451281903453e-06,
+      "loss": 3.8524,
+      "step": 2818048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.852065334392932e-06,
+      "loss": 3.8454,
+      "step": 2818560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.843679386882413e-06,
+      "loss": 3.8519,
+      "step": 2819072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.835293439371892e-06,
+      "loss": 3.8404,
+      "step": 2819584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.826923870665104e-06,
+      "loss": 3.8489,
+      "step": 2820096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.818537923154584e-06,
+      "loss": 3.8459,
+      "step": 2820608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8101519756440643e-06,
+      "loss": 3.8412,
+      "step": 2821120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.801766028133544e-06,
+      "loss": 3.8543,
+      "step": 2821632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7933800806230234e-06,
+      "loss": 3.8506,
+      "step": 2822144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.785010511916235e-06,
+      "loss": 3.8456,
+      "step": 2822656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7766245644057145e-06,
+      "loss": 3.8439,
+      "step": 2823168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.768238616895195e-06,
+      "loss": 3.845,
+      "step": 2823680
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.99065899848938,
+      "eval_runtime": 282.1563,
+      "eval_samples_per_second": 1352.41,
+      "eval_steps_per_second": 42.264,
+      "step": 2823840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7598526693846744e-06,
+      "loss": 3.8511,
+      "step": 2824192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.751466721874155e-06,
+      "loss": 3.85,
+      "step": 2824704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7430807743636343e-06,
+      "loss": 3.8527,
+      "step": 2825216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7346948268531143e-06,
+      "loss": 3.8427,
+      "step": 2825728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7263088793425943e-06,
+      "loss": 3.8598,
+      "step": 2826240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7179229318320742e-06,
+      "loss": 3.8458,
+      "step": 2826752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7095369843215546e-06,
+      "loss": 3.8387,
+      "step": 2827264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.701151036811034e-06,
+      "loss": 3.836,
+      "step": 2827776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6927650893005137e-06,
+      "loss": 3.8521,
+      "step": 2828288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.684379141789994e-06,
+      "loss": 3.8532,
+      "step": 2828800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6759931942794736e-06,
+      "loss": 3.8453,
+      "step": 2829312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.667607246768954e-06,
+      "loss": 3.8533,
+      "step": 2829824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6592376780621647e-06,
+      "loss": 3.8483,
+      "step": 2830336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.650851730551645e-06,
+      "loss": 3.845,
+      "step": 2830848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6424657830411246e-06,
+      "loss": 3.8418,
+      "step": 2831360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.634079835530604e-06,
+      "loss": 3.8377,
+      "step": 2831872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6256938880200846e-06,
+      "loss": 3.8446,
+      "step": 2832384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6173079405095645e-06,
+      "loss": 3.8381,
+      "step": 2832896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6089219929990445e-06,
+      "loss": 3.8537,
+      "step": 2833408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.6005360454885244e-06,
+      "loss": 3.858,
+      "step": 2833920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5921828555854673e-06,
+      "loss": 3.8555,
+      "step": 2834432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.583796908074947e-06,
+      "loss": 3.8473,
+      "step": 2834944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5754109605644272e-06,
+      "loss": 3.8454,
+      "step": 2835456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5670250130539068e-06,
+      "loss": 3.8473,
+      "step": 2835968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5586390655433863e-06,
+      "loss": 3.8413,
+      "step": 2836480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5502531180328667e-06,
+      "loss": 3.8432,
+      "step": 2836992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.541867170522346e-06,
+      "loss": 3.8301,
+      "step": 2837504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5334812230118266e-06,
+      "loss": 3.8502,
+      "step": 2838016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.525095275501306e-06,
+      "loss": 3.8389,
+      "step": 2838528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.516709327990786e-06,
+      "loss": 3.8348,
+      "step": 2839040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.5083397592839977e-06,
+      "loss": 3.8454,
+      "step": 2839552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4999538117734772e-06,
+      "loss": 3.8501,
+      "step": 2840064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4915678642629576e-06,
+      "loss": 3.8543,
+      "step": 2840576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.483181916752437e-06,
+      "loss": 3.8493,
+      "step": 2841088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.474812348045649e-06,
+      "loss": 3.8429,
+      "step": 2841600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4664264005351283e-06,
+      "loss": 3.8539,
+      "step": 2842112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.458040453024608e-06,
+      "loss": 3.8475,
+      "step": 2842624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4496545055140882e-06,
+      "loss": 3.8322,
+      "step": 2843136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4412685580035678e-06,
+      "loss": 3.8349,
+      "step": 2843648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.432882610493048e-06,
+      "loss": 3.8341,
+      "step": 2844160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4245130417862594e-06,
+      "loss": 3.8417,
+      "step": 2844672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4161270942757393e-06,
+      "loss": 3.8511,
+      "step": 2845184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4077411467652193e-06,
+      "loss": 3.8447,
+      "step": 2845696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.399355199254699e-06,
+      "loss": 3.8372,
+      "step": 2846208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.390969251744179e-06,
+      "loss": 3.8511,
+      "step": 2846720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3825833042336587e-06,
+      "loss": 3.8403,
+      "step": 2847232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3742137355268704e-06,
+      "loss": 3.8546,
+      "step": 2847744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.365844166820081e-06,
+      "loss": 3.8367,
+      "step": 2848256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3574582193095615e-06,
+      "loss": 3.8189,
+      "step": 2848768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.349072271799041e-06,
+      "loss": 3.8609,
+      "step": 2849280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3406863242885214e-06,
+      "loss": 3.8489,
+      "step": 2849792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.332300376778001e-06,
+      "loss": 3.8487,
+      "step": 2850304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.323914429267481e-06,
+      "loss": 3.8425,
+      "step": 2850816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.3155284817569613e-06,
+      "loss": 3.8388,
+      "step": 2851328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.307142534246441e-06,
+      "loss": 3.837,
+      "step": 2851840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2987565867359212e-06,
+      "loss": 3.8404,
+      "step": 2852352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.290387018029132e-06,
+      "loss": 3.8382,
+      "step": 2852864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2820174493223436e-06,
+      "loss": 3.8409,
+      "step": 2853376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.273631501811823e-06,
+      "loss": 3.8482,
+      "step": 2853888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2652455543013036e-06,
+      "loss": 3.8426,
+      "step": 2854400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.256859606790783e-06,
+      "loss": 3.8321,
+      "step": 2854912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2484736592802626e-06,
+      "loss": 3.8536,
+      "step": 2855424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.240087711769743e-06,
+      "loss": 3.8329,
+      "step": 2855936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.231701764259223e-06,
+      "loss": 3.8327,
+      "step": 2856448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.223315816748703e-06,
+      "loss": 3.8522,
+      "step": 2856960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.214929869238183e-06,
+      "loss": 3.8524,
+      "step": 2857472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2065439217276624e-06,
+      "loss": 3.8312,
+      "step": 2857984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.198174353020874e-06,
+      "loss": 3.8412,
+      "step": 2858496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1897884055103536e-06,
+      "loss": 3.8261,
+      "step": 2859008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.181418836803565e-06,
+      "loss": 3.8459,
+      "step": 2859520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1730328892930447e-06,
+      "loss": 3.847,
+      "step": 2860032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.164646941782525e-06,
+      "loss": 3.8438,
+      "step": 2860544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1562609942720046e-06,
+      "loss": 3.8437,
+      "step": 2861056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.147875046761485e-06,
+      "loss": 3.851,
+      "step": 2861568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1394890992509646e-06,
+      "loss": 3.859,
+      "step": 2862080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1311031517404445e-06,
+      "loss": 3.8349,
+      "step": 2862592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1227172042299245e-06,
+      "loss": 3.8518,
+      "step": 2863104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1143312567194044e-06,
+      "loss": 3.8507,
+      "step": 2863616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1059453092088844e-06,
+      "loss": 3.8305,
+      "step": 2864128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0975757405020956e-06,
+      "loss": 3.8513,
+      "step": 2864640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0891897929915755e-06,
+      "loss": 3.84,
+      "step": 2865152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0808038454810555e-06,
+      "loss": 3.8542,
+      "step": 2865664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0724178979705355e-06,
+      "loss": 3.8521,
+      "step": 2866176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0640483292637467e-06,
+      "loss": 3.8374,
+      "step": 2866688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0556623817532266e-06,
+      "loss": 3.8339,
+      "step": 2867200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0472764342427066e-06,
+      "loss": 3.84,
+      "step": 2867712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.038890486732186e-06,
+      "loss": 3.8558,
+      "step": 2868224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.030504539221666e-06,
+      "loss": 3.842,
+      "step": 2868736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.022118591711146e-06,
+      "loss": 3.8426,
+      "step": 2869248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.013732644200626e-06,
+      "loss": 3.8406,
+      "step": 2869760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.005346696690106e-06,
+      "loss": 3.8453,
+      "step": 2870272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.996960749179586e-06,
+      "loss": 3.8397,
+      "step": 2870784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.988574801669066e-06,
+      "loss": 3.8254,
+      "step": 2871296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.980188854158546e-06,
+      "loss": 3.8433,
+      "step": 2871808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9718356642554883e-06,
+      "loss": 3.8512,
+      "step": 2872320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9634660955487e-06,
+      "loss": 3.851,
+      "step": 2872832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9550801480381795e-06,
+      "loss": 3.8319,
+      "step": 2873344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9466942005276594e-06,
+      "loss": 3.8396,
+      "step": 2873856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9383082530171394e-06,
+      "loss": 3.8397,
+      "step": 2874368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9299223055066197e-06,
+      "loss": 3.8464,
+      "step": 2874880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9215363579960997e-06,
+      "loss": 3.8507,
+      "step": 2875392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.913166789289311e-06,
+      "loss": 3.8382,
+      "step": 2875904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.904780841778791e-06,
+      "loss": 3.8439,
+      "step": 2876416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8963948942682704e-06,
+      "loss": 3.8495,
+      "step": 2876928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8880089467577504e-06,
+      "loss": 3.8304,
+      "step": 2877440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8796229992472303e-06,
+      "loss": 3.8418,
+      "step": 2877952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8712370517367103e-06,
+      "loss": 3.8347,
+      "step": 2878464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8628511042261902e-06,
+      "loss": 3.8469,
+      "step": 2878976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8544651567156698e-06,
+      "loss": 3.832,
+      "step": 2879488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8460792092051497e-06,
+      "loss": 3.8456,
+      "step": 2880000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8376932616946297e-06,
+      "loss": 3.837,
+      "step": 2880512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8293073141841096e-06,
+      "loss": 3.8423,
+      "step": 2881024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8209213666735896e-06,
+      "loss": 3.8404,
+      "step": 2881536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8125354191630695e-06,
+      "loss": 3.8452,
+      "step": 2882048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.804165850456281e-06,
+      "loss": 3.8424,
+      "step": 2882560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7957799029457607e-06,
+      "loss": 3.8571,
+      "step": 2883072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7873939554352407e-06,
+      "loss": 3.8472,
+      "step": 2883584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7790080079247206e-06,
+      "loss": 3.849,
+      "step": 2884096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7706220604142006e-06,
+      "loss": 3.8395,
+      "step": 2884608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.76223611290368e-06,
+      "loss": 3.8454,
+      "step": 2885120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.75385016539316e-06,
+      "loss": 3.837,
+      "step": 2885632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.745496975490103e-06,
+      "loss": 3.8423,
+      "step": 2886144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.737111027979583e-06,
+      "loss": 3.8409,
+      "step": 2886656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.728725080469063e-06,
+      "loss": 3.8462,
+      "step": 2887168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.720339132958543e-06,
+      "loss": 3.8433,
+      "step": 2887680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7119531854480228e-06,
+      "loss": 3.8405,
+      "step": 2888192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7035672379375027e-06,
+      "loss": 3.842,
+      "step": 2888704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6951812904269827e-06,
+      "loss": 3.8445,
+      "step": 2889216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6867953429164622e-06,
+      "loss": 3.8375,
+      "step": 2889728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.678425774209674e-06,
+      "loss": 3.8475,
+      "step": 2890240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6700398266991534e-06,
+      "loss": 3.8418,
+      "step": 2890752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6616538791886333e-06,
+      "loss": 3.8382,
+      "step": 2891264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6532679316781133e-06,
+      "loss": 3.8277,
+      "step": 2891776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6448819841675933e-06,
+      "loss": 3.8505,
+      "step": 2892288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.636496036657073e-06,
+      "loss": 3.8517,
+      "step": 2892800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6281100891465527e-06,
+      "loss": 3.8484,
+      "step": 2893312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6197241416360327e-06,
+      "loss": 3.8457,
+      "step": 2893824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6113545729292443e-06,
+      "loss": 3.8509,
+      "step": 2894336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6029686254187243e-06,
+      "loss": 3.843,
+      "step": 2894848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5945990567119355e-06,
+      "loss": 3.8546,
+      "step": 2895360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5862131092014155e-06,
+      "loss": 3.8374,
+      "step": 2895872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5778271616908954e-06,
+      "loss": 3.844,
+      "step": 2896384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5694412141803754e-06,
+      "loss": 3.8425,
+      "step": 2896896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5610552666698553e-06,
+      "loss": 3.8411,
+      "step": 2897408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5526856979630665e-06,
+      "loss": 3.8544,
+      "step": 2897920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5442997504525465e-06,
+      "loss": 3.8453,
+      "step": 2898432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5359138029420265e-06,
+      "loss": 3.8502,
+      "step": 2898944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5275278554315064e-06,
+      "loss": 3.8403,
+      "step": 2899456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.5191419079209864e-06,
+      "loss": 3.8445,
+      "step": 2899968
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.990034580230713,
+      "eval_runtime": 294.1007,
+      "eval_samples_per_second": 1297.484,
+      "eval_steps_per_second": 40.547,
+      "step": 2900160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.5107559604104663e-06,
+      "loss": 3.852,
+      "step": 2900480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.5023700128999463e-06,
+      "loss": 3.8472,
+      "step": 2900992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.493984065389426e-06,
+      "loss": 3.8488,
+      "step": 2901504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4855981178789058e-06,
+      "loss": 3.8384,
+      "step": 2902016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4772121703683857e-06,
+      "loss": 3.8569,
+      "step": 2902528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4688262228578657e-06,
+      "loss": 3.8437,
+      "step": 2903040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.460440275347345e-06,
+      "loss": 3.841,
+      "step": 2903552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.452054327836825e-06,
+      "loss": 3.8333,
+      "step": 2904064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.443668380326305e-06,
+      "loss": 3.8513,
+      "step": 2904576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.435282432815785e-06,
+      "loss": 3.8523,
+      "step": 2905088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.426896485305265e-06,
+      "loss": 3.8431,
+      "step": 2905600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4185105377947446e-06,
+      "loss": 3.8489,
+      "step": 2906112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.410140969087956e-06,
+      "loss": 3.845,
+      "step": 2906624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.401755021577436e-06,
+      "loss": 3.8448,
+      "step": 2907136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.393369074066916e-06,
+      "loss": 3.8362,
+      "step": 2907648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.384983126556396e-06,
+      "loss": 3.8346,
+      "step": 2908160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.376597179045876e-06,
+      "loss": 3.8407,
+      "step": 2908672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.368211231535356e-06,
+      "loss": 3.8409,
+      "step": 2909184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3598252840248355e-06,
+      "loss": 3.8489,
+      "step": 2909696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3514393365143155e-06,
+      "loss": 3.8562,
+      "step": 2910208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3430533890037954e-06,
+      "loss": 3.8562,
+      "step": 2910720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3346838202970066e-06,
+      "loss": 3.8449,
+      "step": 2911232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3262978727864866e-06,
+      "loss": 3.843,
+      "step": 2911744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3179119252759665e-06,
+      "loss": 3.8423,
+      "step": 2912256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.3095259777654465e-06,
+      "loss": 3.8418,
+      "step": 2912768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.301140030254926e-06,
+      "loss": 3.8421,
+      "step": 2913280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2927540827444064e-06,
+      "loss": 3.8256,
+      "step": 2913792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2843845140376176e-06,
+      "loss": 3.8498,
+      "step": 2914304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2759985665270976e-06,
+      "loss": 3.8416,
+      "step": 2914816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2676126190165775e-06,
+      "loss": 3.8331,
+      "step": 2915328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2592266715060575e-06,
+      "loss": 3.8394,
+      "step": 2915840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2508571027992687e-06,
+      "loss": 3.8501,
+      "step": 2916352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2424711552887487e-06,
+      "loss": 3.8477,
+      "step": 2916864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2340852077782286e-06,
+      "loss": 3.848,
+      "step": 2917376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.225699260267708e-06,
+      "loss": 3.8403,
+      "step": 2917888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2173296915609198e-06,
+      "loss": 3.8526,
+      "step": 2918400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2089437440503997e-06,
+      "loss": 3.8451,
+      "step": 2918912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2005577965398797e-06,
+      "loss": 3.8318,
+      "step": 2919424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1921718490293596e-06,
+      "loss": 3.8343,
+      "step": 2919936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1837859015188396e-06,
+      "loss": 3.8318,
+      "step": 2920448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.175416332812051e-06,
+      "loss": 3.8416,
+      "step": 2920960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1670303853015308e-06,
+      "loss": 3.8499,
+      "step": 2921472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1586444377910103e-06,
+      "loss": 3.8412,
+      "step": 2921984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1502584902804903e-06,
+      "loss": 3.8383,
+      "step": 2922496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1418725427699702e-06,
+      "loss": 3.8446,
+      "step": 2923008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.13348659525945e-06,
+      "loss": 3.8369,
+      "step": 2923520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.12510064774893e-06,
+      "loss": 3.8527,
+      "step": 2924032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1167147002384097e-06,
+      "loss": 3.8388,
+      "step": 2924544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1083287527278896e-06,
+      "loss": 3.8148,
+      "step": 2925056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0999428052173696e-06,
+      "loss": 3.8559,
+      "step": 2925568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0915568577068495e-06,
+      "loss": 3.8482,
+      "step": 2926080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0831709101963295e-06,
+      "loss": 3.8469,
+      "step": 2926592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.074801341489541e-06,
+      "loss": 3.8412,
+      "step": 2927104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0664317727827523e-06,
+      "loss": 3.838,
+      "step": 2927616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0580458252722323e-06,
+      "loss": 3.8346,
+      "step": 2928128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0496598777617122e-06,
+      "loss": 3.8398,
+      "step": 2928640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0412739302511918e-06,
+      "loss": 3.8362,
+      "step": 2929152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0329207403481347e-06,
+      "loss": 3.8378,
+      "step": 2929664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0245347928376146e-06,
+      "loss": 3.8454,
+      "step": 2930176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0161488453270946e-06,
+      "loss": 3.8428,
+      "step": 2930688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0077628978165745e-06,
+      "loss": 3.8292,
+      "step": 2931200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.999393329109786e-06,
+      "loss": 3.8507,
+      "step": 2931712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9910073815992657e-06,
+      "loss": 3.83,
+      "step": 2932224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9826214340887457e-06,
+      "loss": 3.8335,
+      "step": 2932736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9742354865782256e-06,
+      "loss": 3.8514,
+      "step": 2933248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9658495390677056e-06,
+      "loss": 3.8492,
+      "step": 2933760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9574635915571855e-06,
+      "loss": 3.8279,
+      "step": 2934272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9490940228503967e-06,
+      "loss": 3.8388,
+      "step": 2934784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9407080753398767e-06,
+      "loss": 3.8236,
+      "step": 2935296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9323221278293562e-06,
+      "loss": 3.8459,
+      "step": 2935808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.923936180318836e-06,
+      "loss": 3.8453,
+      "step": 2936320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.915550232808316e-06,
+      "loss": 3.8438,
+      "step": 2936832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9071642852977963e-06,
+      "loss": 3.8421,
+      "step": 2937344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8987783377872763e-06,
+      "loss": 3.848,
+      "step": 2937856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.890392390276756e-06,
+      "loss": 3.8589,
+      "step": 2938368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8820228215699677e-06,
+      "loss": 3.8351,
+      "step": 2938880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8736368740594472e-06,
+      "loss": 3.8478,
+      "step": 2939392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8652509265489271e-06,
+      "loss": 3.8488,
+      "step": 2939904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.856864979038407e-06,
+      "loss": 3.832,
+      "step": 2940416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.848479031527887e-06,
+      "loss": 3.8475,
+      "step": 2940928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8400930840173668e-06,
+      "loss": 3.8445,
+      "step": 2941440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8317071365068468e-06,
+      "loss": 3.8469,
+      "step": 2941952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.823337567800058e-06,
+      "loss": 3.8526,
+      "step": 2942464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.814951620289538e-06,
+      "loss": 3.8336,
+      "step": 2942976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8065656727790179e-06,
+      "loss": 3.8328,
+      "step": 2943488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7981797252684978e-06,
+      "loss": 3.8395,
+      "step": 2944000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7897937777579778e-06,
+      "loss": 3.8489,
+      "step": 2944512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7814078302474575e-06,
+      "loss": 3.8431,
+      "step": 2945024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7730218827369375e-06,
+      "loss": 3.8399,
+      "step": 2945536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7646523140301487e-06,
+      "loss": 3.8391,
+      "step": 2946048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7562663665196287e-06,
+      "loss": 3.8454,
+      "step": 2946560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7478804190091086e-06,
+      "loss": 3.8376,
+      "step": 2947072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7394944714985886e-06,
+      "loss": 3.8218,
+      "step": 2947584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7311085239880685e-06,
+      "loss": 3.8415,
+      "step": 2948096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7227225764775483e-06,
+      "loss": 3.8473,
+      "step": 2948608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.71435300777076e-06,
+      "loss": 3.8486,
+      "step": 2949120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.7059670602602394e-06,
+      "loss": 3.8311,
+      "step": 2949632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6975811127497194e-06,
+      "loss": 3.8374,
+      "step": 2950144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6891951652391993e-06,
+      "loss": 3.8374,
+      "step": 2950656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6808092177286793e-06,
+      "loss": 3.8432,
+      "step": 2951168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6724232702181593e-06,
+      "loss": 3.8501,
+      "step": 2951680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.664037322707639e-06,
+      "loss": 3.8392,
+      "step": 2952192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6556677540008506e-06,
+      "loss": 3.842,
+      "step": 2952704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6472818064903302e-06,
+      "loss": 3.8493,
+      "step": 2953216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6388958589798101e-06,
+      "loss": 3.8278,
+      "step": 2953728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.63050991146929e-06,
+      "loss": 3.841,
+      "step": 2954240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.62212396395877e-06,
+      "loss": 3.8274,
+      "step": 2954752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6137380164482502e-06,
+      "loss": 3.8502,
+      "step": 2955264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6053684477414614e-06,
+      "loss": 3.8298,
+      "step": 2955776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.596982500230941e-06,
+      "loss": 3.8428,
+      "step": 2956288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.588596552720421e-06,
+      "loss": 3.8373,
+      "step": 2956800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5802269840136323e-06,
+      "loss": 3.8385,
+      "step": 2957312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5718410365031123e-06,
+      "loss": 3.8397,
+      "step": 2957824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5634550889925922e-06,
+      "loss": 3.8447,
+      "step": 2958336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5550691414820722e-06,
+      "loss": 3.8403,
+      "step": 2958848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.546683193971552e-06,
+      "loss": 3.8552,
+      "step": 2959360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.538297246461032e-06,
+      "loss": 3.8393,
+      "step": 2959872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5299112989505119e-06,
+      "loss": 3.8552,
+      "step": 2960384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.5215253514399918e-06,
+      "loss": 3.839,
+      "step": 2960896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.513155782733203e-06,
+      "loss": 3.8387,
+      "step": 2961408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.504769835222683e-06,
+      "loss": 3.8406,
+      "step": 2961920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.496383887712163e-06,
+      "loss": 3.8379,
+      "step": 2962432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4879979402016427e-06,
+      "loss": 3.8374,
+      "step": 2962944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4796119926911226e-06,
+      "loss": 3.8459,
+      "step": 2963456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4712260451806026e-06,
+      "loss": 3.8393,
+      "step": 2963968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4628400976700825e-06,
+      "loss": 3.8413,
+      "step": 2964480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4544541501595625e-06,
+      "loss": 3.8418,
+      "step": 2964992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4460682026490422e-06,
+      "loss": 3.8431,
+      "step": 2965504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4376822551385222e-06,
+      "loss": 3.8401,
+      "step": 2966016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.429296307628002e-06,
+      "loss": 3.8414,
+      "step": 2966528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.420910360117482e-06,
+      "loss": 3.8385,
+      "step": 2967040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4125571702144248e-06,
+      "loss": 3.8386,
+      "step": 2967552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4041712227039048e-06,
+      "loss": 3.8236,
+      "step": 2968064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3957852751933845e-06,
+      "loss": 3.8487,
+      "step": 2968576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3873993276828645e-06,
+      "loss": 3.8473,
+      "step": 2969088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3790297589760759e-06,
+      "loss": 3.8498,
+      "step": 2969600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3706438114655558e-06,
+      "loss": 3.8398,
+      "step": 2970112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3622578639550356e-06,
+      "loss": 3.8501,
+      "step": 2970624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3538719164445155e-06,
+      "loss": 3.8436,
+      "step": 2971136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3454859689339955e-06,
+      "loss": 3.8536,
+      "step": 2971648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3371000214234752e-06,
+      "loss": 3.8335,
+      "step": 2972160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3287140739129552e-06,
+      "loss": 3.8376,
+      "step": 2972672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3203445052061666e-06,
+      "loss": 3.8442,
+      "step": 2973184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3119585576956466e-06,
+      "loss": 3.8419,
+      "step": 2973696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3035726101851263e-06,
+      "loss": 3.8472,
+      "step": 2974208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2951866626746063e-06,
+      "loss": 3.8448,
+      "step": 2974720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.2868007151640862e-06,
+      "loss": 3.8464,
+      "step": 2975232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.278414767653566e-06,
+      "loss": 3.8373,
+      "step": 2975744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.270028820143046e-06,
+      "loss": 3.8503,
+      "step": 2976256
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.989668369293213,
+      "eval_runtime": 296.3495,
+      "eval_samples_per_second": 1287.638,
+      "eval_steps_per_second": 40.24,
+      "step": 2976480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2616428726325259e-06,
+      "loss": 3.8391,
+      "step": 2976768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2532733039257373e-06,
+      "loss": 3.85,
+      "step": 2977280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.244887356415217e-06,
+      "loss": 3.843,
+      "step": 2977792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2365177877084285e-06,
+      "loss": 3.8384,
+      "step": 2978304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2281318401979084e-06,
+      "loss": 3.8529,
+      "step": 2978816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2197458926873884e-06,
+      "loss": 3.8454,
+      "step": 2979328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.2113599451768681e-06,
+      "loss": 3.8389,
+      "step": 2979840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.202973997666348e-06,
+      "loss": 3.8345,
+      "step": 2980352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.194588050155828e-06,
+      "loss": 3.8415,
+      "step": 2980864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1862184814490395e-06,
+      "loss": 3.8535,
+      "step": 2981376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1778325339385194e-06,
+      "loss": 3.8389,
+      "step": 2981888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1694629652317306e-06,
+      "loss": 3.8484,
+      "step": 2982400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1610933965249419e-06,
+      "loss": 3.8497,
+      "step": 2982912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.152707449014422e-06,
+      "loss": 3.8382,
+      "step": 2983424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.144321501503902e-06,
+      "loss": 3.837,
+      "step": 2983936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1359355539933817e-06,
+      "loss": 3.8345,
+      "step": 2984448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1275496064828617e-06,
+      "loss": 3.8406,
+      "step": 2984960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1191636589723414e-06,
+      "loss": 3.8377,
+      "step": 2985472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1107777114618214e-06,
+      "loss": 3.8429,
+      "step": 2985984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1023917639513013e-06,
+      "loss": 3.8552,
+      "step": 2986496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.094005816440781e-06,
+      "loss": 3.8534,
+      "step": 2987008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.085619868930261e-06,
+      "loss": 3.8449,
+      "step": 2987520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.077233921419741e-06,
+      "loss": 3.8402,
+      "step": 2988032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.068847973909221e-06,
+      "loss": 3.8393,
+      "step": 2988544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.0604620263987007e-06,
+      "loss": 3.8418,
+      "step": 2989056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.0520924576919121e-06,
+      "loss": 3.8409,
+      "step": 2989568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.0437065101813919e-06,
+      "loss": 3.8265,
+      "step": 2990080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.0353205626708718e-06,
+      "loss": 3.8477,
+      "step": 2990592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.026934615160352e-06,
+      "loss": 3.8399,
+      "step": 2991104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.0185486676498317e-06,
+      "loss": 3.8305,
+      "step": 2991616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0101627201393117e-06,
+      "loss": 3.8329,
+      "step": 2992128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0017931514325229e-06,
+      "loss": 3.8516,
+      "step": 2992640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.934072039220028e-07,
+      "loss": 3.8485,
+      "step": 2993152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.850212564114828e-07,
+      "loss": 3.8453,
+      "step": 2993664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.766353089009628e-07,
+      "loss": 3.8404,
+      "step": 2994176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.68265740194174e-07,
+      "loss": 3.8475,
+      "step": 2994688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.59879792683654e-07,
+      "loss": 3.848,
+      "step": 2995200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.514938451731339e-07,
+      "loss": 3.8319,
+      "step": 2995712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.431078976626137e-07,
+      "loss": 3.8321,
+      "step": 2996224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.347219501520937e-07,
+      "loss": 3.831,
+      "step": 2996736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.263360026415734e-07,
+      "loss": 3.84,
+      "step": 2997248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.179500551310534e-07,
+      "loss": 3.8445,
+      "step": 2997760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.095641076205332e-07,
+      "loss": 3.8446,
+      "step": 2998272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 9.011945389137447e-07,
+      "loss": 3.8347,
+      "step": 2998784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.928085914032245e-07,
+      "loss": 3.842,
+      "step": 2999296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.844226438927045e-07,
+      "loss": 3.8369,
+      "step": 2999808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.760366963821844e-07,
+      "loss": 3.8488,
+      "step": 3000320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.676671276753957e-07,
+      "loss": 3.8414,
+      "step": 3000832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.592811801648757e-07,
+      "loss": 3.8137,
+      "step": 3001344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.508952326543554e-07,
+      "loss": 3.8526,
+      "step": 3001856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.425092851438355e-07,
+      "loss": 3.8457,
+      "step": 3002368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.341233376333152e-07,
+      "loss": 3.8494,
+      "step": 3002880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.257373901227952e-07,
+      "loss": 3.8362,
+      "step": 3003392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.173678214160065e-07,
+      "loss": 3.8322,
+      "step": 3003904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.089818739054865e-07,
+      "loss": 3.8348,
+      "step": 3004416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.005959263949664e-07,
+      "loss": 3.8384,
+      "step": 3004928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.922099788844463e-07,
+      "loss": 3.8409,
+      "step": 3005440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.838240313739262e-07,
+      "loss": 3.8296,
+      "step": 3005952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.754380838634061e-07,
+      "loss": 3.8434,
+      "step": 3006464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.670521363528859e-07,
+      "loss": 3.8418,
+      "step": 3006976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.586661888423659e-07,
+      "loss": 3.8262,
+      "step": 3007488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.502966201355772e-07,
+      "loss": 3.8482,
+      "step": 3008000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.419106726250572e-07,
+      "loss": 3.8297,
+      "step": 3008512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.33524725114537e-07,
+      "loss": 3.8285,
+      "step": 3009024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.251387776040169e-07,
+      "loss": 3.8469,
+      "step": 3009536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.167528300934967e-07,
+      "loss": 3.8468,
+      "step": 3010048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 7.083832613867081e-07,
+      "loss": 3.8301,
+      "step": 3010560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.99997313876188e-07,
+      "loss": 3.8375,
+      "step": 3011072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.916277451693994e-07,
+      "loss": 3.8199,
+      "step": 3011584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.832417976588793e-07,
+      "loss": 3.8427,
+      "step": 3012096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.748558501483592e-07,
+      "loss": 3.8429,
+      "step": 3012608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.664699026378392e-07,
+      "loss": 3.8432,
+      "step": 3013120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.58083955127319e-07,
+      "loss": 3.8377,
+      "step": 3013632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.497143864205305e-07,
+      "loss": 3.8463,
+      "step": 3014144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.413284389100103e-07,
+      "loss": 3.8558,
+      "step": 3014656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.329424913994902e-07,
+      "loss": 3.8311,
+      "step": 3015168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.245565438889701e-07,
+      "loss": 3.8458,
+      "step": 3015680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.1617059637845e-07,
+      "loss": 3.8481,
+      "step": 3016192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.077846488679298e-07,
+      "loss": 3.8311,
+      "step": 3016704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.993987013574098e-07,
+      "loss": 3.845,
+      "step": 3017216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.910127538468897e-07,
+      "loss": 3.8417,
+      "step": 3017728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.826268063363696e-07,
+      "loss": 3.8448,
+      "step": 3018240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.742408588258494e-07,
+      "loss": 3.853,
+      "step": 3018752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.658549113153293e-07,
+      "loss": 3.8273,
+      "step": 3019264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.574689638048092e-07,
+      "loss": 3.8335,
+      "step": 3019776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.490993950980205e-07,
+      "loss": 3.8372,
+      "step": 3020288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.40729826391232e-07,
+      "loss": 3.8462,
+      "step": 3020800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.323438788807118e-07,
+      "loss": 3.8369,
+      "step": 3021312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.239579313701918e-07,
+      "loss": 3.8434,
+      "step": 3021824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.155719838596717e-07,
+      "loss": 3.8417,
+      "step": 3022336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.071860363491516e-07,
+      "loss": 3.8357,
+      "step": 3022848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.98816467642363e-07,
+      "loss": 3.8385,
+      "step": 3023360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.904305201318429e-07,
+      "loss": 3.8202,
+      "step": 3023872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.820445726213227e-07,
+      "loss": 3.8379,
+      "step": 3024384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.736586251108026e-07,
+      "loss": 3.844,
+      "step": 3024896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.652726776002825e-07,
+      "loss": 3.8475,
+      "step": 3025408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5688673008976236e-07,
+      "loss": 3.8294,
+      "step": 3025920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.485007825792423e-07,
+      "loss": 3.8311,
+      "step": 3026432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.4013121387245364e-07,
+      "loss": 3.8382,
+      "step": 3026944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.317452663619336e-07,
+      "loss": 3.8428,
+      "step": 3027456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.233593188514135e-07,
+      "loss": 3.8449,
+      "step": 3027968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.149733713408934e-07,
+      "loss": 3.8404,
+      "step": 3028480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0658742383037325e-07,
+      "loss": 3.8413,
+      "step": 3028992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.982178551235846e-07,
+      "loss": 3.8421,
+      "step": 3029504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8983190761306453e-07,
+      "loss": 3.831,
+      "step": 3030016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8144596010254443e-07,
+      "loss": 3.8345,
+      "step": 3030528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7306001259202433e-07,
+      "loss": 3.8283,
+      "step": 3031040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6467406508150423e-07,
+      "loss": 3.8495,
+      "step": 3031552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.563044963747156e-07,
+      "loss": 3.8281,
+      "step": 3032064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.479185488641955e-07,
+      "loss": 3.8393,
+      "step": 3032576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.395326013536754e-07,
+      "loss": 3.836,
+      "step": 3033088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.3114665384315526e-07,
+      "loss": 3.8419,
+      "step": 3033600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2276070633263517e-07,
+      "loss": 3.8352,
+      "step": 3034112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.1439113762584654e-07,
+      "loss": 3.8418,
+      "step": 3034624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0600519011532644e-07,
+      "loss": 3.8367,
+      "step": 3035136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9761924260480634e-07,
+      "loss": 3.8529,
+      "step": 3035648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8923329509428625e-07,
+      "loss": 3.8356,
+      "step": 3036160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.808473475837661e-07,
+      "loss": 3.8515,
+      "step": 3036672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7246140007324605e-07,
+      "loss": 3.8369,
+      "step": 3037184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.640754525627259e-07,
+      "loss": 3.8388,
+      "step": 3037696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.556895050522058e-07,
+      "loss": 3.8374,
+      "step": 3038208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.4733631514914865e-07,
+      "loss": 3.8416,
+      "step": 3038720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3895036763862855e-07,
+      "loss": 3.8337,
+      "step": 3039232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3056442012810848e-07,
+      "loss": 3.8418,
+      "step": 3039744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2217847261758836e-07,
+      "loss": 3.8334,
+      "step": 3040256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.1379252510706823e-07,
+      "loss": 3.8409,
+      "step": 3040768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0542295640027963e-07,
+      "loss": 3.8431,
+      "step": 3041280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.970370088897595e-07,
+      "loss": 3.8372,
+      "step": 3041792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.8865106137923944e-07,
+      "loss": 3.8409,
+      "step": 3042304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.802651138687193e-07,
+      "loss": 3.8434,
+      "step": 3042816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7187916635819922e-07,
+      "loss": 3.834,
+      "step": 3043328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.6349321884767912e-07,
+      "loss": 3.8399,
+      "step": 3043840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.55107271337159e-07,
+      "loss": 3.8207,
+      "step": 3044352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4673770263037037e-07,
+      "loss": 3.847,
+      "step": 3044864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3835175511985027e-07,
+      "loss": 3.8464,
+      "step": 3045376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2996580760933017e-07,
+      "loss": 3.8483,
+      "step": 3045888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2157986009881008e-07,
+      "loss": 3.8383,
+      "step": 3046400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.1319391258828994e-07,
+      "loss": 3.8452,
+      "step": 3046912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0480796507776984e-07,
+      "loss": 3.8431,
+      "step": 3047424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 9.642201756724973e-08,
+      "loss": 3.8489,
+      "step": 3047936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 8.805244886046112e-08,
+      "loss": 3.8328,
+      "step": 3048448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.968288015367249e-08,
+      "loss": 3.8386,
+      "step": 3048960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.129693264315238e-08,
+      "loss": 3.8411,
+      "step": 3049472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 6.291098513263228e-08,
+      "loss": 3.845,
+      "step": 3049984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.452503762211217e-08,
+      "loss": 3.8426,
+      "step": 3050496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.613909011159207e-08,
+      "loss": 3.8396,
+      "step": 3051008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.775314260107196e-08,
+      "loss": 3.848,
+      "step": 3051520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.936719509055186e-08,
+      "loss": 3.8335,
+      "step": 3052032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.098124758003175e-08,
+      "loss": 3.8497,
+      "step": 3052544
+    },
+    {
+      "epoch": 1.02,
+      "eval_loss": 3.9892959594726562,
+      "eval_runtime": 292.3449,
+      "eval_samples_per_second": 1305.277,
+      "eval_steps_per_second": 40.791,
+      "step": 3052726
+    }
+  ],
+  "logging_steps": 512,
+  "max_steps": 3052726,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 10,
+  "total_flos": 1.2104594866057178e+18,
+  "trial_name": null,
+  "trial_params": null
+}