diff --git "a/checkpoint-2823840/trainer_state.json" "b/checkpoint-2823840/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-2823840/trainer_state.json"
@@ -0,0 +1,33411 @@
+{
+  "best_metric": 3.9746415615081787,
+  "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/outputs/binding-domain/lstm/0/checkpoints/checkpoint-2823840",
+  "epoch": 0.025000606015738065,
+  "eval_steps": 10,
+  "global_step": 2823840,
+  "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": 10.8182,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999161405248948e-05,
+      "loss": 7.555,
+      "step": 512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.998322810497896e-05,
+      "loss": 7.061,
+      "step": 1024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.997484215746844e-05,
+      "loss": 6.9947,
+      "step": 1536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.996645620995792e-05,
+      "loss": 6.9465,
+      "step": 2048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99580702624474e-05,
+      "loss": 6.929,
+      "step": 2560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.994968431493688e-05,
+      "loss": 6.7614,
+      "step": 3072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.994129836742636e-05,
+      "loss": 6.6484,
+      "step": 3584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.993291241991584e-05,
+      "loss": 6.5484,
+      "step": 4096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.992452647240532e-05,
+      "loss": 6.4747,
+      "step": 4608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99161405248948e-05,
+      "loss": 6.4091,
+      "step": 5120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.990775457738428e-05,
+      "loss": 6.3454,
+      "step": 5632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.989936862987376e-05,
+      "loss": 6.2704,
+      "step": 6144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.989099906116697e-05,
+      "loss": 6.203,
+      "step": 6656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.988261311365645e-05,
+      "loss": 6.1436,
+      "step": 7168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.987422716614593e-05,
+      "loss": 6.0788,
+      "step": 7680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.986585759743914e-05,
+      "loss": 6.0283,
+      "step": 8192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.985747164992862e-05,
+      "loss": 5.9861,
+      "step": 8704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.98490857024181e-05,
+      "loss": 5.9469,
+      "step": 9216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.984069975490758e-05,
+      "loss": 5.9076,
+      "step": 9728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.983231380739706e-05,
+      "loss": 5.8715,
+      "step": 10240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9823927859886547e-05,
+      "loss": 5.829,
+      "step": 10752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9815541912376026e-05,
+      "loss": 5.8047,
+      "step": 11264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9807155964865506e-05,
+      "loss": 5.7666,
+      "step": 11776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9798770017354986e-05,
+      "loss": 5.745,
+      "step": 12288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9790384069844466e-05,
+      "loss": 5.7159,
+      "step": 12800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9781998122333946e-05,
+      "loss": 5.6932,
+      "step": 13312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9773612174823426e-05,
+      "loss": 5.6635,
+      "step": 13824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.97652262273129e-05,
+      "loss": 5.6382,
+      "step": 14336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.975684027980238e-05,
+      "loss": 5.6195,
+      "step": 14848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.974845433229186e-05,
+      "loss": 5.5953,
+      "step": 15360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.974008476358507e-05,
+      "loss": 5.5887,
+      "step": 15872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.973169881607455e-05,
+      "loss": 5.5587,
+      "step": 16384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.972331286856403e-05,
+      "loss": 5.5501,
+      "step": 16896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9714926921053515e-05,
+      "loss": 5.5406,
+      "step": 17408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9706557352346724e-05,
+      "loss": 5.52,
+      "step": 17920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9698171404836204e-05,
+      "loss": 5.5063,
+      "step": 18432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9689785457325684e-05,
+      "loss": 5.4725,
+      "step": 18944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9681399509815164e-05,
+      "loss": 5.468,
+      "step": 19456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.967302994110837e-05,
+      "loss": 5.4457,
+      "step": 19968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.966466037240159e-05,
+      "loss": 5.4408,
+      "step": 20480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.965627442489107e-05,
+      "loss": 5.4234,
+      "step": 20992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.964788847738054e-05,
+      "loss": 5.4148,
+      "step": 21504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.963950252987002e-05,
+      "loss": 5.3846,
+      "step": 22016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.963113296116324e-05,
+      "loss": 5.3958,
+      "step": 22528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.962274701365272e-05,
+      "loss": 5.3854,
+      "step": 23040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.96143610661422e-05,
+      "loss": 5.3759,
+      "step": 23552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.960597511863168e-05,
+      "loss": 5.3638,
+      "step": 24064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.959762192872862e-05,
+      "loss": 5.3371,
+      "step": 24576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.95892359812181e-05,
+      "loss": 5.3301,
+      "step": 25088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.958085003370758e-05,
+      "loss": 5.3326,
+      "step": 25600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.957246408619706e-05,
+      "loss": 5.3175,
+      "step": 26112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.956407813868654e-05,
+      "loss": 5.3069,
+      "step": 26624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9555692191176016e-05,
+      "loss": 5.2817,
+      "step": 27136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9547306243665496e-05,
+      "loss": 5.2819,
+      "step": 27648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9538920296154976e-05,
+      "loss": 5.2701,
+      "step": 28160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9530534348644456e-05,
+      "loss": 5.2957,
+      "step": 28672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.952216477993767e-05,
+      "loss": 5.2485,
+      "step": 29184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.951377883242715e-05,
+      "loss": 5.2529,
+      "step": 29696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.950539288491663e-05,
+      "loss": 5.2476,
+      "step": 30208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.949700693740611e-05,
+      "loss": 5.2275,
+      "step": 30720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.948863736869932e-05,
+      "loss": 5.2286,
+      "step": 31232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.94802514211888e-05,
+      "loss": 5.2126,
+      "step": 31744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.947186547367828e-05,
+      "loss": 5.1987,
+      "step": 32256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.946347952616776e-05,
+      "loss": 5.1974,
+      "step": 32768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.945509357865724e-05,
+      "loss": 5.2144,
+      "step": 33280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.944670763114672e-05,
+      "loss": 5.1869,
+      "step": 33792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.94383216836362e-05,
+      "loss": 5.1787,
+      "step": 34304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.942993573612568e-05,
+      "loss": 5.1617,
+      "step": 34816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.942156616741889e-05,
+      "loss": 5.1635,
+      "step": 35328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.941318021990837e-05,
+      "loss": 5.1727,
+      "step": 35840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9404794272397856e-05,
+      "loss": 5.1679,
+      "step": 36352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9396408324887336e-05,
+      "loss": 5.1546,
+      "step": 36864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9388022377376816e-05,
+      "loss": 5.1624,
+      "step": 37376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9379636429866296e-05,
+      "loss": 5.1552,
+      "step": 37888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9371250482355776e-05,
+      "loss": 5.1435,
+      "step": 38400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9362880913648985e-05,
+      "loss": 5.124,
+      "step": 38912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9354494966138465e-05,
+      "loss": 5.1145,
+      "step": 39424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9346109018627945e-05,
+      "loss": 5.1123,
+      "step": 39936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9337723071117425e-05,
+      "loss": 5.1054,
+      "step": 40448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9329353502410634e-05,
+      "loss": 5.1007,
+      "step": 40960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9320967554900114e-05,
+      "loss": 5.0968,
+      "step": 41472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9312581607389594e-05,
+      "loss": 5.099,
+      "step": 41984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.930421203868281e-05,
+      "loss": 5.0907,
+      "step": 42496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.929582609117229e-05,
+      "loss": 5.0649,
+      "step": 43008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.928744014366177e-05,
+      "loss": 5.0602,
+      "step": 43520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.927905419615125e-05,
+      "loss": 5.0706,
+      "step": 44032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.927066824864073e-05,
+      "loss": 5.0711,
+      "step": 44544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.92622823011302e-05,
+      "loss": 5.0481,
+      "step": 45056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.925389635361968e-05,
+      "loss": 5.0507,
+      "step": 45568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.924551040610916e-05,
+      "loss": 5.0505,
+      "step": 46080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.923714083740238e-05,
+      "loss": 5.0368,
+      "step": 46592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.922877126869559e-05,
+      "loss": 5.0294,
+      "step": 47104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.922038532118507e-05,
+      "loss": 5.0291,
+      "step": 47616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.921199937367455e-05,
+      "loss": 5.0182,
+      "step": 48128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.920361342616403e-05,
+      "loss": 5.0223,
+      "step": 48640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.919522747865351e-05,
+      "loss": 5.0116,
+      "step": 49152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.918684153114299e-05,
+      "loss": 4.9974,
+      "step": 49664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.917845558363247e-05,
+      "loss": 4.9961,
+      "step": 50176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.917006963612195e-05,
+      "loss": 4.9958,
+      "step": 50688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9161700067415156e-05,
+      "loss": 4.9915,
+      "step": 51200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9153314119904636e-05,
+      "loss": 4.9916,
+      "step": 51712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9144928172394116e-05,
+      "loss": 4.9811,
+      "step": 52224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9136542224883596e-05,
+      "loss": 4.9778,
+      "step": 52736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9128156277373076e-05,
+      "loss": 4.9698,
+      "step": 53248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9119770329862556e-05,
+      "loss": 4.9498,
+      "step": 53760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9111400761155765e-05,
+      "loss": 4.9543,
+      "step": 54272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9103014813645245e-05,
+      "loss": 4.9529,
+      "step": 54784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.909462886613473e-05,
+      "loss": 4.9491,
+      "step": 55296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.908624291862421e-05,
+      "loss": 4.9413,
+      "step": 55808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.907785697111369e-05,
+      "loss": 4.936,
+      "step": 56320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.906947102360317e-05,
+      "loss": 4.9287,
+      "step": 56832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.906110145489638e-05,
+      "loss": 4.9295,
+      "step": 57344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.905273188618959e-05,
+      "loss": 4.9237,
+      "step": 57856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.904434593867907e-05,
+      "loss": 4.9239,
+      "step": 58368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.903595999116855e-05,
+      "loss": 4.916,
+      "step": 58880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.902757404365803e-05,
+      "loss": 4.9144,
+      "step": 59392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.901918809614751e-05,
+      "loss": 4.9161,
+      "step": 59904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.901080214863699e-05,
+      "loss": 4.9065,
+      "step": 60416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.900241620112647e-05,
+      "loss": 4.9127,
+      "step": 60928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8994046632419686e-05,
+      "loss": 4.9107,
+      "step": 61440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8985660684909166e-05,
+      "loss": 4.8952,
+      "step": 61952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8977274737398646e-05,
+      "loss": 4.8999,
+      "step": 62464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8968888789888125e-05,
+      "loss": 4.8849,
+      "step": 62976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8960502842377605e-05,
+      "loss": 4.889,
+      "step": 63488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8952116894867085e-05,
+      "loss": 4.8647,
+      "step": 64000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8943730947356565e-05,
+      "loss": 4.8766,
+      "step": 64512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.893534499984604e-05,
+      "loss": 4.8628,
+      "step": 65024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8926975431139254e-05,
+      "loss": 4.8639,
+      "step": 65536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8918589483628734e-05,
+      "loss": 4.8456,
+      "step": 66048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8910203536118214e-05,
+      "loss": 4.878,
+      "step": 66560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8901833967411423e-05,
+      "loss": 4.8645,
+      "step": 67072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.88934480199009e-05,
+      "loss": 4.8526,
+      "step": 67584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.888506207239038e-05,
+      "loss": 4.8494,
+      "step": 68096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.887667612487986e-05,
+      "loss": 4.846,
+      "step": 68608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.886829017736934e-05,
+      "loss": 4.854,
+      "step": 69120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.885990422985882e-05,
+      "loss": 4.8478,
+      "step": 69632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.88515182823483e-05,
+      "loss": 4.8366,
+      "step": 70144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.884313233483778e-05,
+      "loss": 4.8424,
+      "step": 70656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.883474638732726e-05,
+      "loss": 4.8285,
+      "step": 71168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.882636043981674e-05,
+      "loss": 4.838,
+      "step": 71680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.881797449230622e-05,
+      "loss": 4.8203,
+      "step": 72192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.880962130240316e-05,
+      "loss": 4.8227,
+      "step": 72704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.880123535489264e-05,
+      "loss": 4.8204,
+      "step": 73216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.879284940738212e-05,
+      "loss": 4.8086,
+      "step": 73728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.87844634598716e-05,
+      "loss": 4.8114,
+      "step": 74240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.877607751236109e-05,
+      "loss": 4.8116,
+      "step": 74752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.876769156485057e-05,
+      "loss": 4.806,
+      "step": 75264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.875932199614378e-05,
+      "loss": 4.8099,
+      "step": 75776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.875093604863326e-05,
+      "loss": 4.7977,
+      "step": 76288
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.761485576629639,
+      "eval_runtime": 294.2132,
+      "eval_samples_per_second": 1296.988,
+      "eval_steps_per_second": 40.532,
+      "step": 76320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.874255010112274e-05,
+      "loss": 4.7878,
+      "step": 76800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.873416415361222e-05,
+      "loss": 4.783,
+      "step": 77312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8725794584905426e-05,
+      "loss": 4.7999,
+      "step": 77824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8717408637394906e-05,
+      "loss": 4.782,
+      "step": 78336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8709022689884386e-05,
+      "loss": 4.7984,
+      "step": 78848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8700636742373866e-05,
+      "loss": 4.769,
+      "step": 79360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8692250794863346e-05,
+      "loss": 4.7811,
+      "step": 79872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8683881226156555e-05,
+      "loss": 4.7585,
+      "step": 80384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.867549527864604e-05,
+      "loss": 4.7713,
+      "step": 80896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.866710933113552e-05,
+      "loss": 4.7688,
+      "step": 81408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8658723383625e-05,
+      "loss": 4.7729,
+      "step": 81920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.865035381491821e-05,
+      "loss": 4.7676,
+      "step": 82432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.864198424621142e-05,
+      "loss": 4.7497,
+      "step": 82944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.86335982987009e-05,
+      "loss": 4.7519,
+      "step": 83456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.862521235119038e-05,
+      "loss": 4.7467,
+      "step": 83968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.861682640367986e-05,
+      "loss": 4.7414,
+      "step": 84480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.860844045616934e-05,
+      "loss": 4.7514,
+      "step": 84992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.860005450865882e-05,
+      "loss": 4.7457,
+      "step": 85504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.85916685611483e-05,
+      "loss": 4.7424,
+      "step": 86016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.858328261363778e-05,
+      "loss": 4.7573,
+      "step": 86528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8574913044930995e-05,
+      "loss": 4.7315,
+      "step": 87040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8566527097420475e-05,
+      "loss": 4.7403,
+      "step": 87552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8558141149909955e-05,
+      "loss": 4.7348,
+      "step": 88064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8549755202399435e-05,
+      "loss": 4.7426,
+      "step": 88576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8541369254888915e-05,
+      "loss": 4.719,
+      "step": 89088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8532999686182124e-05,
+      "loss": 4.7273,
+      "step": 89600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8524613738671604e-05,
+      "loss": 4.7184,
+      "step": 90112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8516227791161084e-05,
+      "loss": 4.7268,
+      "step": 90624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8507841843650564e-05,
+      "loss": 4.7062,
+      "step": 91136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8499455896140044e-05,
+      "loss": 4.7104,
+      "step": 91648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.849110270623698e-05,
+      "loss": 4.7193,
+      "step": 92160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.848271675872646e-05,
+      "loss": 4.7102,
+      "step": 92672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.847433081121595e-05,
+      "loss": 4.7099,
+      "step": 93184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.846594486370543e-05,
+      "loss": 4.7162,
+      "step": 93696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.845755891619491e-05,
+      "loss": 4.709,
+      "step": 94208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.844917296868439e-05,
+      "loss": 4.7093,
+      "step": 94720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.844078702117387e-05,
+      "loss": 4.6903,
+      "step": 95232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.843240107366335e-05,
+      "loss": 4.7001,
+      "step": 95744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.842401512615282e-05,
+      "loss": 4.6835,
+      "step": 96256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.84156291786423e-05,
+      "loss": 4.6934,
+      "step": 96768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.840724323113178e-05,
+      "loss": 4.6902,
+      "step": 97280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8398873662425e-05,
+      "loss": 4.6873,
+      "step": 97792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.839048771491447e-05,
+      "loss": 4.6741,
+      "step": 98304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.838210176740395e-05,
+      "loss": 4.6899,
+      "step": 98816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.837371581989343e-05,
+      "loss": 4.6891,
+      "step": 99328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.836532987238292e-05,
+      "loss": 4.6807,
+      "step": 99840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8356960303676126e-05,
+      "loss": 4.6873,
+      "step": 100352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.834859073496934e-05,
+      "loss": 4.6571,
+      "step": 100864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.834020478745882e-05,
+      "loss": 4.6598,
+      "step": 101376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8331818839948295e-05,
+      "loss": 4.6779,
+      "step": 101888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8323432892437775e-05,
+      "loss": 4.6659,
+      "step": 102400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8315046944927255e-05,
+      "loss": 4.6641,
+      "step": 102912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8306660997416735e-05,
+      "loss": 4.6444,
+      "step": 103424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8298275049906215e-05,
+      "loss": 4.6512,
+      "step": 103936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8289889102395695e-05,
+      "loss": 4.6467,
+      "step": 104448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8281503154885175e-05,
+      "loss": 4.6726,
+      "step": 104960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8273133586178384e-05,
+      "loss": 4.6412,
+      "step": 105472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.826474763866787e-05,
+      "loss": 4.6495,
+      "step": 105984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.825636169115735e-05,
+      "loss": 4.655,
+      "step": 106496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.824797574364683e-05,
+      "loss": 4.6354,
+      "step": 107008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.823958979613631e-05,
+      "loss": 4.643,
+      "step": 107520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.823120384862579e-05,
+      "loss": 4.6359,
+      "step": 108032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.822281790111527e-05,
+      "loss": 4.6264,
+      "step": 108544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.821444833240848e-05,
+      "loss": 4.631,
+      "step": 109056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.820606238489796e-05,
+      "loss": 4.6456,
+      "step": 109568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.819767643738744e-05,
+      "loss": 4.6334,
+      "step": 110080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.818929048987692e-05,
+      "loss": 4.6234,
+      "step": 110592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.81809045423664e-05,
+      "loss": 4.6214,
+      "step": 111104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.817251859485588e-05,
+      "loss": 4.6211,
+      "step": 111616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.816414902614909e-05,
+      "loss": 4.6378,
+      "step": 112128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.815576307863857e-05,
+      "loss": 4.6378,
+      "step": 112640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.814737713112805e-05,
+      "loss": 4.6278,
+      "step": 113152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8138991183617535e-05,
+      "loss": 4.6405,
+      "step": 113664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.813060523610701e-05,
+      "loss": 4.6375,
+      "step": 114176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.812221928859649e-05,
+      "loss": 4.6325,
+      "step": 114688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.811383334108597e-05,
+      "loss": 4.6174,
+      "step": 115200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.810544739357545e-05,
+      "loss": 4.6164,
+      "step": 115712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.809707782486866e-05,
+      "loss": 4.6128,
+      "step": 116224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.808869187735814e-05,
+      "loss": 4.6126,
+      "step": 116736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.808030592984762e-05,
+      "loss": 4.6087,
+      "step": 117248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.80719199823371e-05,
+      "loss": 4.6149,
+      "step": 117760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.806353403482658e-05,
+      "loss": 4.6206,
+      "step": 118272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.805514808731606e-05,
+      "loss": 4.6108,
+      "step": 118784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.804676213980554e-05,
+      "loss": 4.5963,
+      "step": 119296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.803837619229502e-05,
+      "loss": 4.5948,
+      "step": 119808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.803000662358823e-05,
+      "loss": 4.6015,
+      "step": 120320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.802162067607771e-05,
+      "loss": 4.6128,
+      "step": 120832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.801323472856719e-05,
+      "loss": 4.5931,
+      "step": 121344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.80048651598604e-05,
+      "loss": 4.5998,
+      "step": 121856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.799647921234988e-05,
+      "loss": 4.6035,
+      "step": 122368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.798809326483936e-05,
+      "loss": 4.5946,
+      "step": 122880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.797970731732884e-05,
+      "loss": 4.589,
+      "step": 123392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.797133774862205e-05,
+      "loss": 4.5875,
+      "step": 123904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.796295180111153e-05,
+      "loss": 4.5891,
+      "step": 124416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.795456585360101e-05,
+      "loss": 4.5897,
+      "step": 124928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.794617990609049e-05,
+      "loss": 4.5912,
+      "step": 125440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.793779395857997e-05,
+      "loss": 4.5757,
+      "step": 125952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.792940801106946e-05,
+      "loss": 4.5822,
+      "step": 126464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.792103844236267e-05,
+      "loss": 4.5791,
+      "step": 126976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.791265249485215e-05,
+      "loss": 4.5864,
+      "step": 127488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.790426654734163e-05,
+      "loss": 4.5814,
+      "step": 128000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7895880599831107e-05,
+      "loss": 4.5804,
+      "step": 128512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7887494652320587e-05,
+      "loss": 4.5775,
+      "step": 129024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7879108704810067e-05,
+      "loss": 4.5747,
+      "step": 129536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7870722757299546e-05,
+      "loss": 4.5586,
+      "step": 130048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.786233680978902e-05,
+      "loss": 4.5619,
+      "step": 130560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7853983619885965e-05,
+      "loss": 4.5682,
+      "step": 131072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7845597672375445e-05,
+      "loss": 4.563,
+      "step": 131584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7837211724864925e-05,
+      "loss": 4.5601,
+      "step": 132096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.782882577735441e-05,
+      "loss": 4.5579,
+      "step": 132608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.782043982984389e-05,
+      "loss": 4.5532,
+      "step": 133120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.78120702611371e-05,
+      "loss": 4.5573,
+      "step": 133632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.780368431362658e-05,
+      "loss": 4.5558,
+      "step": 134144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.779529836611606e-05,
+      "loss": 4.5541,
+      "step": 134656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.778691241860554e-05,
+      "loss": 4.5563,
+      "step": 135168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.777852647109502e-05,
+      "loss": 4.5563,
+      "step": 135680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.777014052358449e-05,
+      "loss": 4.5575,
+      "step": 136192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.776175457607397e-05,
+      "loss": 4.5515,
+      "step": 136704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.775336862856345e-05,
+      "loss": 4.5608,
+      "step": 137216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.774499905985666e-05,
+      "loss": 4.5602,
+      "step": 137728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.773661311234615e-05,
+      "loss": 4.5476,
+      "step": 138240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.772822716483563e-05,
+      "loss": 4.5576,
+      "step": 138752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.771984121732511e-05,
+      "loss": 4.544,
+      "step": 139264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.771147164861832e-05,
+      "loss": 4.5521,
+      "step": 139776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.77030857011078e-05,
+      "loss": 4.5263,
+      "step": 140288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.769469975359728e-05,
+      "loss": 4.544,
+      "step": 140800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.768631380608676e-05,
+      "loss": 4.5275,
+      "step": 141312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.767792785857624e-05,
+      "loss": 4.5346,
+      "step": 141824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.766954191106572e-05,
+      "loss": 4.5197,
+      "step": 142336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.76611559635552e-05,
+      "loss": 4.5531,
+      "step": 142848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.765278639484841e-05,
+      "loss": 4.543,
+      "step": 143360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7644416826141616e-05,
+      "loss": 4.5374,
+      "step": 143872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7636030878631096e-05,
+      "loss": 4.5288,
+      "step": 144384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.762764493112058e-05,
+      "loss": 4.5333,
+      "step": 144896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.761925898361006e-05,
+      "loss": 4.5411,
+      "step": 145408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.761087303609954e-05,
+      "loss": 4.5363,
+      "step": 145920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.760248708858902e-05,
+      "loss": 4.5296,
+      "step": 146432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.75941011410785e-05,
+      "loss": 4.5322,
+      "step": 146944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.758571519356798e-05,
+      "loss": 4.5249,
+      "step": 147456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.757732924605746e-05,
+      "loss": 4.5374,
+      "step": 147968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.756895967735067e-05,
+      "loss": 4.5198,
+      "step": 148480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.756057372984015e-05,
+      "loss": 4.5234,
+      "step": 148992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.755218778232963e-05,
+      "loss": 4.5223,
+      "step": 149504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.754380183481911e-05,
+      "loss": 4.5155,
+      "step": 150016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.753541588730859e-05,
+      "loss": 4.5194,
+      "step": 150528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.752702993979807e-05,
+      "loss": 4.5187,
+      "step": 151040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.751866037109128e-05,
+      "loss": 4.5192,
+      "step": 151552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.751027442358077e-05,
+      "loss": 4.5175,
+      "step": 152064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.750188847607025e-05,
+      "loss": 4.5122,
+      "step": 152576
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.480395793914795,
+      "eval_runtime": 293.1861,
+      "eval_samples_per_second": 1301.532,
+      "eval_steps_per_second": 40.674,
+      "step": 152640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.749350252855973e-05,
+      "loss": 4.5036,
+      "step": 153088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.748511658104921e-05,
+      "loss": 4.4978,
+      "step": 153600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.747673063353868e-05,
+      "loss": 4.5172,
+      "step": 154112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.746834468602816e-05,
+      "loss": 4.5029,
+      "step": 154624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.745995873851764e-05,
+      "loss": 4.5231,
+      "step": 155136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.745157279100712e-05,
+      "loss": 4.4938,
+      "step": 155648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.74431868434966e-05,
+      "loss": 4.506,
+      "step": 156160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.743480089598608e-05,
+      "loss": 4.4878,
+      "step": 156672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.742641494847556e-05,
+      "loss": 4.5046,
+      "step": 157184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.741802900096504e-05,
+      "loss": 4.5015,
+      "step": 157696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.740964305345452e-05,
+      "loss": 4.5054,
+      "step": 158208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7401257105944e-05,
+      "loss": 4.5033,
+      "step": 158720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7392887537237216e-05,
+      "loss": 4.4849,
+      "step": 159232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7384501589726696e-05,
+      "loss": 4.4943,
+      "step": 159744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7376115642216176e-05,
+      "loss": 4.4892,
+      "step": 160256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7367729694705656e-05,
+      "loss": 4.4865,
+      "step": 160768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7359360125998865e-05,
+      "loss": 4.4888,
+      "step": 161280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7350974178488345e-05,
+      "loss": 4.4906,
+      "step": 161792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7342588230977825e-05,
+      "loss": 4.4867,
+      "step": 162304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7334202283467305e-05,
+      "loss": 4.5055,
+      "step": 162816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7325816335956784e-05,
+      "loss": 4.481,
+      "step": 163328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7317430388446264e-05,
+      "loss": 4.4875,
+      "step": 163840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.730904444093574e-05,
+      "loss": 4.4869,
+      "step": 164352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.730065849342522e-05,
+      "loss": 4.4958,
+      "step": 164864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7292288924718433e-05,
+      "loss": 4.4746,
+      "step": 165376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.728390297720791e-05,
+      "loss": 4.4864,
+      "step": 165888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.727551702969739e-05,
+      "loss": 4.4762,
+      "step": 166400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.726713108218687e-05,
+      "loss": 4.4816,
+      "step": 166912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.725874513467635e-05,
+      "loss": 4.4683,
+      "step": 167424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.725035918716583e-05,
+      "loss": 4.4694,
+      "step": 167936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.724198961845904e-05,
+      "loss": 4.4787,
+      "step": 168448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.723360367094852e-05,
+      "loss": 4.4783,
+      "step": 168960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7225217723438e-05,
+      "loss": 4.4761,
+      "step": 169472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.721683177592748e-05,
+      "loss": 4.4807,
+      "step": 169984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.720846220722069e-05,
+      "loss": 4.4746,
+      "step": 170496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.720007625971017e-05,
+      "loss": 4.4802,
+      "step": 171008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.719169031219965e-05,
+      "loss": 4.4593,
+      "step": 171520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.718330436468914e-05,
+      "loss": 4.4754,
+      "step": 172032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.717493479598235e-05,
+      "loss": 4.4567,
+      "step": 172544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.716654884847183e-05,
+      "loss": 4.4672,
+      "step": 173056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.715816290096131e-05,
+      "loss": 4.4634,
+      "step": 173568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.714977695345079e-05,
+      "loss": 4.4673,
+      "step": 174080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.714139100594027e-05,
+      "loss": 4.4592,
+      "step": 174592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.713300505842975e-05,
+      "loss": 4.4624,
+      "step": 175104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.712461911091923e-05,
+      "loss": 4.4703,
+      "step": 175616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7116249542212436e-05,
+      "loss": 4.4636,
+      "step": 176128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7107863594701916e-05,
+      "loss": 4.466,
+      "step": 176640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7099477647191396e-05,
+      "loss": 4.4472,
+      "step": 177152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7091091699680876e-05,
+      "loss": 4.4398,
+      "step": 177664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.708272213097409e-05,
+      "loss": 4.4579,
+      "step": 178176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.707433618346357e-05,
+      "loss": 4.4519,
+      "step": 178688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.706595023595305e-05,
+      "loss": 4.4518,
+      "step": 179200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.705756428844253e-05,
+      "loss": 4.4352,
+      "step": 179712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.704917834093201e-05,
+      "loss": 4.4407,
+      "step": 180224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.704079239342149e-05,
+      "loss": 4.4413,
+      "step": 180736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.70324228247147e-05,
+      "loss": 4.4558,
+      "step": 181248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.702405325600791e-05,
+      "loss": 4.4346,
+      "step": 181760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.701566730849739e-05,
+      "loss": 4.4438,
+      "step": 182272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.700728136098687e-05,
+      "loss": 4.4475,
+      "step": 182784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.699889541347635e-05,
+      "loss": 4.4346,
+      "step": 183296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.699050946596583e-05,
+      "loss": 4.4384,
+      "step": 183808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.698212351845531e-05,
+      "loss": 4.4336,
+      "step": 184320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.697373757094479e-05,
+      "loss": 4.4233,
+      "step": 184832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6965351623434276e-05,
+      "loss": 4.4364,
+      "step": 185344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6956965675923756e-05,
+      "loss": 4.4385,
+      "step": 185856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.694857972841323e-05,
+      "loss": 4.4423,
+      "step": 186368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.694019378090271e-05,
+      "loss": 4.4197,
+      "step": 186880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.693180783339219e-05,
+      "loss": 4.4264,
+      "step": 187392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.69234382646854e-05,
+      "loss": 4.4247,
+      "step": 187904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.691505231717488e-05,
+      "loss": 4.4504,
+      "step": 188416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.690666636966436e-05,
+      "loss": 4.4361,
+      "step": 188928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6898296800957574e-05,
+      "loss": 4.436,
+      "step": 189440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.688991085344705e-05,
+      "loss": 4.449,
+      "step": 189952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.688152490593653e-05,
+      "loss": 4.4472,
+      "step": 190464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6873138958426014e-05,
+      "loss": 4.4406,
+      "step": 190976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.686476938971923e-05,
+      "loss": 4.4281,
+      "step": 191488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.68563834422087e-05,
+      "loss": 4.4299,
+      "step": 192000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.684799749469818e-05,
+      "loss": 4.4262,
+      "step": 192512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.683961154718766e-05,
+      "loss": 4.4232,
+      "step": 193024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.683122559967714e-05,
+      "loss": 4.4246,
+      "step": 193536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.682285603097035e-05,
+      "loss": 4.4264,
+      "step": 194048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.681447008345983e-05,
+      "loss": 4.4347,
+      "step": 194560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.680608413594931e-05,
+      "loss": 4.4248,
+      "step": 195072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.679769818843879e-05,
+      "loss": 4.4159,
+      "step": 195584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.678931224092827e-05,
+      "loss": 4.4106,
+      "step": 196096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.678094267222148e-05,
+      "loss": 4.4204,
+      "step": 196608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.677255672471097e-05,
+      "loss": 4.4278,
+      "step": 197120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.676417077720045e-05,
+      "loss": 4.4185,
+      "step": 197632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.675578482968993e-05,
+      "loss": 4.4173,
+      "step": 198144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6747415260983136e-05,
+      "loss": 4.4263,
+      "step": 198656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6739029313472616e-05,
+      "loss": 4.4068,
+      "step": 199168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6730643365962096e-05,
+      "loss": 4.4164,
+      "step": 199680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6722257418451576e-05,
+      "loss": 4.4109,
+      "step": 200192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6713871470941056e-05,
+      "loss": 4.4182,
+      "step": 200704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6705485523430536e-05,
+      "loss": 4.4143,
+      "step": 201216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6697099575920016e-05,
+      "loss": 4.4116,
+      "step": 201728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6688713628409496e-05,
+      "loss": 4.4037,
+      "step": 202240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6680327680898976e-05,
+      "loss": 4.4126,
+      "step": 202752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6671958112192185e-05,
+      "loss": 4.4019,
+      "step": 203264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.66635885434854e-05,
+      "loss": 4.4189,
+      "step": 203776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.665520259597488e-05,
+      "loss": 4.4108,
+      "step": 204288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.664681664846436e-05,
+      "loss": 4.4093,
+      "step": 204800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.663843070095384e-05,
+      "loss": 4.4034,
+      "step": 205312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.663004475344332e-05,
+      "loss": 4.4121,
+      "step": 205824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.66216588059328e-05,
+      "loss": 4.3872,
+      "step": 206336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.661327285842228e-05,
+      "loss": 4.3945,
+      "step": 206848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.660488691091176e-05,
+      "loss": 4.3991,
+      "step": 207360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6596500963401234e-05,
+      "loss": 4.3993,
+      "step": 207872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6588115015890714e-05,
+      "loss": 4.3917,
+      "step": 208384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.657974544718393e-05,
+      "loss": 4.3954,
+      "step": 208896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.657135949967341e-05,
+      "loss": 4.386,
+      "step": 209408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.656297355216288e-05,
+      "loss": 4.394,
+      "step": 209920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.65546039834561e-05,
+      "loss": 4.3921,
+      "step": 210432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6546218035945586e-05,
+      "loss": 4.3883,
+      "step": 210944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6537832088435065e-05,
+      "loss": 4.3993,
+      "step": 211456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.652944614092454e-05,
+      "loss": 4.3986,
+      "step": 211968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.652106019341402e-05,
+      "loss": 4.4006,
+      "step": 212480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.65126742459035e-05,
+      "loss": 4.3893,
+      "step": 212992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.650428829839298e-05,
+      "loss": 4.3998,
+      "step": 213504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.649590235088246e-05,
+      "loss": 4.3968,
+      "step": 214016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.648753278217567e-05,
+      "loss": 4.3901,
+      "step": 214528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.647914683466515e-05,
+      "loss": 4.4049,
+      "step": 215040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.647076088715463e-05,
+      "loss": 4.3837,
+      "step": 215552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.646237493964411e-05,
+      "loss": 4.395,
+      "step": 216064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.645398899213359e-05,
+      "loss": 4.3754,
+      "step": 216576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.644560304462307e-05,
+      "loss": 4.3898,
+      "step": 217088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6437217097112554e-05,
+      "loss": 4.3706,
+      "step": 217600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6428831149602034e-05,
+      "loss": 4.3818,
+      "step": 218112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.642046158089524e-05,
+      "loss": 4.3707,
+      "step": 218624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.641207563338472e-05,
+      "loss": 4.3973,
+      "step": 219136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.640370606467793e-05,
+      "loss": 4.3887,
+      "step": 219648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.639532011716741e-05,
+      "loss": 4.3881,
+      "step": 220160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.638695054846062e-05,
+      "loss": 4.3774,
+      "step": 220672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.63785646009501e-05,
+      "loss": 4.3801,
+      "step": 221184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.637017865343958e-05,
+      "loss": 4.3956,
+      "step": 221696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.636180908473279e-05,
+      "loss": 4.3829,
+      "step": 222208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.635342313722228e-05,
+      "loss": 4.3784,
+      "step": 222720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.634503718971176e-05,
+      "loss": 4.3882,
+      "step": 223232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.633665124220124e-05,
+      "loss": 4.3776,
+      "step": 223744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.632826529469072e-05,
+      "loss": 4.3853,
+      "step": 224256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.63198793471802e-05,
+      "loss": 4.3722,
+      "step": 224768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.631149339966968e-05,
+      "loss": 4.3789,
+      "step": 225280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.630310745215916e-05,
+      "loss": 4.3746,
+      "step": 225792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.629472150464864e-05,
+      "loss": 4.3745,
+      "step": 226304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.628633555713812e-05,
+      "loss": 4.3704,
+      "step": 226816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6277949609627597e-05,
+      "loss": 4.3785,
+      "step": 227328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.626956366211707e-05,
+      "loss": 4.3688,
+      "step": 227840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.626117771460655e-05,
+      "loss": 4.3728,
+      "step": 228352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.625279176709603e-05,
+      "loss": 4.3702,
+      "step": 228864
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.345595836639404,
+      "eval_runtime": 311.3159,
+      "eval_samples_per_second": 1225.736,
+      "eval_steps_per_second": 38.305,
+      "step": 228960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6244422198389246e-05,
+      "loss": 4.3665,
+      "step": 229376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.623605262968246e-05,
+      "loss": 4.3566,
+      "step": 229888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.622766668217194e-05,
+      "loss": 4.3757,
+      "step": 230400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.621928073466142e-05,
+      "loss": 4.3601,
+      "step": 230912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6210894787150895e-05,
+      "loss": 4.3856,
+      "step": 231424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6202508839640374e-05,
+      "loss": 4.352,
+      "step": 231936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6194122892129854e-05,
+      "loss": 4.3651,
+      "step": 232448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6185736944619334e-05,
+      "loss": 4.3496,
+      "step": 232960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6177350997108814e-05,
+      "loss": 4.3659,
+      "step": 233472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6168965049598294e-05,
+      "loss": 4.3666,
+      "step": 233984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6160579102087774e-05,
+      "loss": 4.3644,
+      "step": 234496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6152193154577254e-05,
+      "loss": 4.3671,
+      "step": 235008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.614382358587046e-05,
+      "loss": 4.3479,
+      "step": 235520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.613543763835994e-05,
+      "loss": 4.3586,
+      "step": 236032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.612705169084942e-05,
+      "loss": 4.3539,
+      "step": 236544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.611866574333891e-05,
+      "loss": 4.3524,
+      "step": 237056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.611029617463212e-05,
+      "loss": 4.3534,
+      "step": 237568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.61019102271216e-05,
+      "loss": 4.3564,
+      "step": 238080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.609352427961108e-05,
+      "loss": 4.3549,
+      "step": 238592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.608513833210056e-05,
+      "loss": 4.3679,
+      "step": 239104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.607675238459004e-05,
+      "loss": 4.35,
+      "step": 239616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.606836643707952e-05,
+      "loss": 4.3555,
+      "step": 240128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.605999686837273e-05,
+      "loss": 4.3552,
+      "step": 240640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.605161092086221e-05,
+      "loss": 4.3654,
+      "step": 241152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.604322497335169e-05,
+      "loss": 4.3444,
+      "step": 241664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.603483902584117e-05,
+      "loss": 4.3569,
+      "step": 242176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.602646945713438e-05,
+      "loss": 4.3449,
+      "step": 242688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6018083509623864e-05,
+      "loss": 4.348,
+      "step": 243200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6009697562113344e-05,
+      "loss": 4.3367,
+      "step": 243712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.6001311614602824e-05,
+      "loss": 4.3468,
+      "step": 244224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.599294204589603e-05,
+      "loss": 4.3459,
+      "step": 244736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.598455609838551e-05,
+      "loss": 4.3524,
+      "step": 245248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.597617015087499e-05,
+      "loss": 4.3473,
+      "step": 245760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.596778420336447e-05,
+      "loss": 4.353,
+      "step": 246272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.595939825585395e-05,
+      "loss": 4.346,
+      "step": 246784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.595101230834343e-05,
+      "loss": 4.3521,
+      "step": 247296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.594265911844037e-05,
+      "loss": 4.3375,
+      "step": 247808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.593427317092985e-05,
+      "loss": 4.3476,
+      "step": 248320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.592588722341933e-05,
+      "loss": 4.3319,
+      "step": 248832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.591750127590882e-05,
+      "loss": 4.3425,
+      "step": 249344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5909131707202026e-05,
+      "loss": 4.3408,
+      "step": 249856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5900745759691506e-05,
+      "loss": 4.3438,
+      "step": 250368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5892359812180986e-05,
+      "loss": 4.3342,
+      "step": 250880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5883973864670466e-05,
+      "loss": 4.341,
+      "step": 251392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5875587917159946e-05,
+      "loss": 4.3449,
+      "step": 251904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5867201969649426e-05,
+      "loss": 4.3407,
+      "step": 252416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5858816022138906e-05,
+      "loss": 4.3421,
+      "step": 252928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.585043007462838e-05,
+      "loss": 4.3323,
+      "step": 253440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.584204412711786e-05,
+      "loss": 4.3125,
+      "step": 253952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5833674558411075e-05,
+      "loss": 4.3381,
+      "step": 254464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5825288610900555e-05,
+      "loss": 4.3317,
+      "step": 254976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5816902663390035e-05,
+      "loss": 4.3276,
+      "step": 255488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5808516715879515e-05,
+      "loss": 4.3224,
+      "step": 256000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5800130768368995e-05,
+      "loss": 4.3204,
+      "step": 256512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5791744820858475e-05,
+      "loss": 4.317,
+      "step": 257024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5783375252151684e-05,
+      "loss": 4.3385,
+      "step": 257536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5774989304641164e-05,
+      "loss": 4.3178,
+      "step": 258048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5766603357130644e-05,
+      "loss": 4.3247,
+      "step": 258560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.575823378842385e-05,
+      "loss": 4.3348,
+      "step": 259072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.574984784091333e-05,
+      "loss": 4.3179,
+      "step": 259584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.574146189340281e-05,
+      "loss": 4.3188,
+      "step": 260096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.573307594589229e-05,
+      "loss": 4.3226,
+      "step": 260608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.572468999838177e-05,
+      "loss": 4.304,
+      "step": 261120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.571630405087125e-05,
+      "loss": 4.3215,
+      "step": 261632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.570791810336074e-05,
+      "loss": 4.3237,
+      "step": 262144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.569953215585022e-05,
+      "loss": 4.3237,
+      "step": 262656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.569116258714343e-05,
+      "loss": 4.3076,
+      "step": 263168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.568277663963291e-05,
+      "loss": 4.3151,
+      "step": 263680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.567439069212239e-05,
+      "loss": 4.3098,
+      "step": 264192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.566600474461187e-05,
+      "loss": 4.3309,
+      "step": 264704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.565763517590508e-05,
+      "loss": 4.3236,
+      "step": 265216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.564924922839456e-05,
+      "loss": 4.3249,
+      "step": 265728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.564086328088404e-05,
+      "loss": 4.334,
+      "step": 266240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.563247733337352e-05,
+      "loss": 4.3355,
+      "step": 266752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5624107764666726e-05,
+      "loss": 4.332,
+      "step": 267264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5615721817156206e-05,
+      "loss": 4.3118,
+      "step": 267776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.560733586964569e-05,
+      "loss": 4.3188,
+      "step": 268288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.559894992213517e-05,
+      "loss": 4.3206,
+      "step": 268800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.559056397462465e-05,
+      "loss": 4.3109,
+      "step": 269312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.558217802711413e-05,
+      "loss": 4.3188,
+      "step": 269824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.557380845840734e-05,
+      "loss": 4.3141,
+      "step": 270336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.556542251089682e-05,
+      "loss": 4.3229,
+      "step": 270848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.55570365633863e-05,
+      "loss": 4.3184,
+      "step": 271360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.554865061587578e-05,
+      "loss": 4.2996,
+      "step": 271872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.554028104716899e-05,
+      "loss": 4.3052,
+      "step": 272384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.553189509965847e-05,
+      "loss": 4.3133,
+      "step": 272896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.552350915214795e-05,
+      "loss": 4.317,
+      "step": 273408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.551512320463743e-05,
+      "loss": 4.3096,
+      "step": 273920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.550673725712691e-05,
+      "loss": 4.3086,
+      "step": 274432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.549836768842013e-05,
+      "loss": 4.3164,
+      "step": 274944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.548998174090961e-05,
+      "loss": 4.3023,
+      "step": 275456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.548159579339909e-05,
+      "loss": 4.3045,
+      "step": 275968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.547320984588857e-05,
+      "loss": 4.3062,
+      "step": 276480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.546482389837804e-05,
+      "loss": 4.3127,
+      "step": 276992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.545643795086752e-05,
+      "loss": 4.3115,
+      "step": 277504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5448052003357e-05,
+      "loss": 4.3039,
+      "step": 278016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.543966605584648e-05,
+      "loss": 4.2939,
+      "step": 278528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.543129648713969e-05,
+      "loss": 4.3101,
+      "step": 279040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.542291053962917e-05,
+      "loss": 4.2957,
+      "step": 279552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5414540970922385e-05,
+      "loss": 4.3168,
+      "step": 280064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5406155023411865e-05,
+      "loss": 4.3087,
+      "step": 280576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5397769075901345e-05,
+      "loss": 4.3032,
+      "step": 281088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5389383128390824e-05,
+      "loss": 4.2987,
+      "step": 281600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5380997180880304e-05,
+      "loss": 4.3094,
+      "step": 282112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5372627612173514e-05,
+      "loss": 4.2857,
+      "step": 282624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5364241664662994e-05,
+      "loss": 4.2965,
+      "step": 283136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5355855717152473e-05,
+      "loss": 4.2914,
+      "step": 283648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5347469769641953e-05,
+      "loss": 4.2974,
+      "step": 284160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.533910020093516e-05,
+      "loss": 4.2899,
+      "step": 284672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.533071425342464e-05,
+      "loss": 4.2971,
+      "step": 285184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.532232830591412e-05,
+      "loss": 4.2895,
+      "step": 285696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.53139423584036e-05,
+      "loss": 4.2857,
+      "step": 286208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.530557278969682e-05,
+      "loss": 4.2946,
+      "step": 286720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.52971868421863e-05,
+      "loss": 4.2856,
+      "step": 287232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5288817273479514e-05,
+      "loss": 4.3024,
+      "step": 287744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.528043132596899e-05,
+      "loss": 4.2986,
+      "step": 288256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.527204537845847e-05,
+      "loss": 4.2997,
+      "step": 288768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.526365943094795e-05,
+      "loss": 4.2907,
+      "step": 289280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.525527348343743e-05,
+      "loss": 4.2996,
+      "step": 289792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.524688753592691e-05,
+      "loss": 4.3011,
+      "step": 290304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5238517967220116e-05,
+      "loss": 4.2909,
+      "step": 290816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5230132019709596e-05,
+      "loss": 4.3038,
+      "step": 291328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5221746072199076e-05,
+      "loss": 4.288,
+      "step": 291840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5213376503492285e-05,
+      "loss": 4.2953,
+      "step": 292352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.520499055598177e-05,
+      "loss": 4.2812,
+      "step": 292864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.519660460847125e-05,
+      "loss": 4.2923,
+      "step": 293376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.518821866096073e-05,
+      "loss": 4.2745,
+      "step": 293888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.517983271345021e-05,
+      "loss": 4.2859,
+      "step": 294400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.517144676593969e-05,
+      "loss": 4.2765,
+      "step": 294912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.516306081842917e-05,
+      "loss": 4.2952,
+      "step": 295424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.515467487091865e-05,
+      "loss": 4.2889,
+      "step": 295936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.514628892340813e-05,
+      "loss": 4.2937,
+      "step": 296448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.513790297589761e-05,
+      "loss": 4.2811,
+      "step": 296960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.512951702838709e-05,
+      "loss": 4.2883,
+      "step": 297472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.512113108087657e-05,
+      "loss": 4.2974,
+      "step": 297984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5112745133366045e-05,
+      "loss": 4.2888,
+      "step": 298496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.510439194346299e-05,
+      "loss": 4.2852,
+      "step": 299008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.509600599595247e-05,
+      "loss": 4.2943,
+      "step": 299520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5087620048441956e-05,
+      "loss": 4.2821,
+      "step": 300032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5079234100931436e-05,
+      "loss": 4.2935,
+      "step": 300544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5070848153420916e-05,
+      "loss": 4.2754,
+      "step": 301056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5062462205910396e-05,
+      "loss": 4.287,
+      "step": 301568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5054076258399876e-05,
+      "loss": 4.2803,
+      "step": 302080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.504569031088935e-05,
+      "loss": 4.2796,
+      "step": 302592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.503730436337883e-05,
+      "loss": 4.2827,
+      "step": 303104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.502891841586831e-05,
+      "loss": 4.2796,
+      "step": 303616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.502054884716152e-05,
+      "loss": 4.2786,
+      "step": 304128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5012162899651e-05,
+      "loss": 4.2795,
+      "step": 304640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.500377695214048e-05,
+      "loss": 4.2828,
+      "step": 305152
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.262184143066406,
+      "eval_runtime": 304.8532,
+      "eval_samples_per_second": 1251.72,
+      "eval_steps_per_second": 39.117,
+      "step": 305280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.499539100462996e-05,
+      "loss": 4.2711,
+      "step": 305664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.498700505711944e-05,
+      "loss": 4.2633,
+      "step": 306176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4978619109608925e-05,
+      "loss": 4.2846,
+      "step": 306688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4970233162098405e-05,
+      "loss": 4.27,
+      "step": 307200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4961847214587885e-05,
+      "loss": 4.2908,
+      "step": 307712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4953461267077365e-05,
+      "loss": 4.2707,
+      "step": 308224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4945075319566845e-05,
+      "loss": 4.269,
+      "step": 308736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4936689372056325e-05,
+      "loss": 4.2596,
+      "step": 309248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4928303424545805e-05,
+      "loss": 4.2752,
+      "step": 309760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4919933855839014e-05,
+      "loss": 4.2804,
+      "step": 310272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4911547908328494e-05,
+      "loss": 4.2731,
+      "step": 310784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4903161960817974e-05,
+      "loss": 4.2722,
+      "step": 311296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.489479239211118e-05,
+      "loss": 4.2626,
+      "step": 311808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.488640644460066e-05,
+      "loss": 4.2703,
+      "step": 312320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.487802049709014e-05,
+      "loss": 4.2676,
+      "step": 312832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.486963454957962e-05,
+      "loss": 4.2596,
+      "step": 313344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.48612486020691e-05,
+      "loss": 4.2676,
+      "step": 313856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.485286265455859e-05,
+      "loss": 4.2717,
+      "step": 314368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.484447670704806e-05,
+      "loss": 4.2651,
+      "step": 314880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.483609075953754e-05,
+      "loss": 4.2775,
+      "step": 315392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.482770481202702e-05,
+      "loss": 4.269,
+      "step": 315904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.48193188645165e-05,
+      "loss": 4.2701,
+      "step": 316416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.481093291700598e-05,
+      "loss": 4.2683,
+      "step": 316928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.480254696949546e-05,
+      "loss": 4.2792,
+      "step": 317440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.479416102198494e-05,
+      "loss": 4.2579,
+      "step": 317952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.478577507447442e-05,
+      "loss": 4.2675,
+      "step": 318464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.477740550576763e-05,
+      "loss": 4.2641,
+      "step": 318976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.476901955825711e-05,
+      "loss": 4.2593,
+      "step": 319488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.476063361074659e-05,
+      "loss": 4.2532,
+      "step": 320000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.475224766323607e-05,
+      "loss": 4.2592,
+      "step": 320512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.474386171572556e-05,
+      "loss": 4.2667,
+      "step": 321024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.473547576821504e-05,
+      "loss": 4.2661,
+      "step": 321536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.472708982070452e-05,
+      "loss": 4.2659,
+      "step": 322048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.471872025199773e-05,
+      "loss": 4.2652,
+      "step": 322560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.471033430448721e-05,
+      "loss": 4.263,
+      "step": 323072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.470194835697669e-05,
+      "loss": 4.2677,
+      "step": 323584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.469356240946617e-05,
+      "loss": 4.2589,
+      "step": 324096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.468517646195565e-05,
+      "loss": 4.2607,
+      "step": 324608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4676806893248856e-05,
+      "loss": 4.2526,
+      "step": 325120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4668420945738336e-05,
+      "loss": 4.2592,
+      "step": 325632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4660034998227816e-05,
+      "loss": 4.2566,
+      "step": 326144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4651649050717296e-05,
+      "loss": 4.2621,
+      "step": 326656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4643263103206776e-05,
+      "loss": 4.2546,
+      "step": 327168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4634877155696256e-05,
+      "loss": 4.2588,
+      "step": 327680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4626491208185736e-05,
+      "loss": 4.2661,
+      "step": 328192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4618105260675215e-05,
+      "loss": 4.2565,
+      "step": 328704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4609735691968425e-05,
+      "loss": 4.2581,
+      "step": 329216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.460136612326164e-05,
+      "loss": 4.2502,
+      "step": 329728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.459298017575112e-05,
+      "loss": 4.2318,
+      "step": 330240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4584594228240594e-05,
+      "loss": 4.2607,
+      "step": 330752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4576208280730074e-05,
+      "loss": 4.251,
+      "step": 331264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4567822333219554e-05,
+      "loss": 4.2482,
+      "step": 331776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4559436385709033e-05,
+      "loss": 4.2365,
+      "step": 332288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4551050438198513e-05,
+      "loss": 4.2449,
+      "step": 332800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.454266449068799e-05,
+      "loss": 4.2356,
+      "step": 333312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.453429492198121e-05,
+      "loss": 4.2583,
+      "step": 333824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.452590897447069e-05,
+      "loss": 4.2404,
+      "step": 334336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.451752302696017e-05,
+      "loss": 4.2438,
+      "step": 334848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.450913707944965e-05,
+      "loss": 4.2551,
+      "step": 335360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.450075113193913e-05,
+      "loss": 4.2383,
+      "step": 335872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.449238156323234e-05,
+      "loss": 4.24,
+      "step": 336384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.448399561572182e-05,
+      "loss": 4.2466,
+      "step": 336896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.44756096682113e-05,
+      "loss": 4.2249,
+      "step": 337408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.446722372070078e-05,
+      "loss": 4.2444,
+      "step": 337920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.445883777319026e-05,
+      "loss": 4.2412,
+      "step": 338432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.445045182567974e-05,
+      "loss": 4.249,
+      "step": 338944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.444208225697295e-05,
+      "loss": 4.2311,
+      "step": 339456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.443369630946243e-05,
+      "loss": 4.2383,
+      "step": 339968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4425310361951914e-05,
+      "loss": 4.2288,
+      "step": 340480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4416924414441394e-05,
+      "loss": 4.2541,
+      "step": 340992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.44085548457346e-05,
+      "loss": 4.2464,
+      "step": 341504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.440016889822408e-05,
+      "loss": 4.2451,
+      "step": 342016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.439178295071356e-05,
+      "loss": 4.2594,
+      "step": 342528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.438339700320304e-05,
+      "loss": 4.2543,
+      "step": 343040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.437501105569252e-05,
+      "loss": 4.2637,
+      "step": 343552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.436664148698573e-05,
+      "loss": 4.2368,
+      "step": 344064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.435825553947521e-05,
+      "loss": 4.2421,
+      "step": 344576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.434986959196469e-05,
+      "loss": 4.245,
+      "step": 345088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.434148364445417e-05,
+      "loss": 4.2327,
+      "step": 345600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.433311407574738e-05,
+      "loss": 4.2446,
+      "step": 346112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.432472812823687e-05,
+      "loss": 4.2387,
+      "step": 346624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.431634218072635e-05,
+      "loss": 4.246,
+      "step": 347136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.430795623321583e-05,
+      "loss": 4.2427,
+      "step": 347648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.429957028570531e-05,
+      "loss": 4.2284,
+      "step": 348160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.429118433819479e-05,
+      "loss": 4.2398,
+      "step": 348672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4282814769487996e-05,
+      "loss": 4.2296,
+      "step": 349184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4274428821977476e-05,
+      "loss": 4.2452,
+      "step": 349696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4266042874466956e-05,
+      "loss": 4.238,
+      "step": 350208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.425765692695643e-05,
+      "loss": 4.233,
+      "step": 350720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.424927097944591e-05,
+      "loss": 4.2423,
+      "step": 351232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.424088503193539e-05,
+      "loss": 4.2344,
+      "step": 351744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4232531842032334e-05,
+      "loss": 4.2274,
+      "step": 352256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.422414589452182e-05,
+      "loss": 4.2324,
+      "step": 352768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.42157599470113e-05,
+      "loss": 4.2428,
+      "step": 353280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.420737399950078e-05,
+      "loss": 4.238,
+      "step": 353792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.419898805199026e-05,
+      "loss": 4.2342,
+      "step": 354304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4190602104479734e-05,
+      "loss": 4.2212,
+      "step": 354816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4182216156969214e-05,
+      "loss": 4.2358,
+      "step": 355328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4173830209458694e-05,
+      "loss": 4.2264,
+      "step": 355840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4165444261948174e-05,
+      "loss": 4.2421,
+      "step": 356352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4157058314437654e-05,
+      "loss": 4.2406,
+      "step": 356864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4148672366927134e-05,
+      "loss": 4.2321,
+      "step": 357376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4140286419416614e-05,
+      "loss": 4.2272,
+      "step": 357888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.413191685070982e-05,
+      "loss": 4.2426,
+      "step": 358400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.412354728200304e-05,
+      "loss": 4.2118,
+      "step": 358912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.411516133449252e-05,
+      "loss": 4.2265,
+      "step": 359424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4106775386982e-05,
+      "loss": 4.2177,
+      "step": 359936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.409838943947148e-05,
+      "loss": 4.2331,
+      "step": 360448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.409000349196096e-05,
+      "loss": 4.2147,
+      "step": 360960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.408161754445044e-05,
+      "loss": 4.2242,
+      "step": 361472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.407324797574365e-05,
+      "loss": 4.2194,
+      "step": 361984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.406486202823313e-05,
+      "loss": 4.2169,
+      "step": 362496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.405647608072261e-05,
+      "loss": 4.2223,
+      "step": 363008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.404809013321209e-05,
+      "loss": 4.2194,
+      "step": 363520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.403970418570157e-05,
+      "loss": 4.2322,
+      "step": 364032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.403131823819105e-05,
+      "loss": 4.2299,
+      "step": 364544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4022948669484257e-05,
+      "loss": 4.2281,
+      "step": 365056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.401456272197374e-05,
+      "loss": 4.2207,
+      "step": 365568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.400617677446322e-05,
+      "loss": 4.2309,
+      "step": 366080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.39977908269527e-05,
+      "loss": 4.2289,
+      "step": 366592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.398940487944218e-05,
+      "loss": 4.2209,
+      "step": 367104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.398101893193166e-05,
+      "loss": 4.2373,
+      "step": 367616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.397264936322487e-05,
+      "loss": 4.2203,
+      "step": 368128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.396426341571435e-05,
+      "loss": 4.2248,
+      "step": 368640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.395587746820383e-05,
+      "loss": 4.2142,
+      "step": 369152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.394750789949704e-05,
+      "loss": 4.2247,
+      "step": 369664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.393912195198652e-05,
+      "loss": 4.2099,
+      "step": 370176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3930736004476e-05,
+      "loss": 4.2159,
+      "step": 370688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.392235005696548e-05,
+      "loss": 4.2133,
+      "step": 371200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.391396410945496e-05,
+      "loss": 4.2282,
+      "step": 371712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.390557816194444e-05,
+      "loss": 4.2206,
+      "step": 372224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.389719221443392e-05,
+      "loss": 4.225,
+      "step": 372736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.38888062669234e-05,
+      "loss": 4.2121,
+      "step": 373248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.388042031941288e-05,
+      "loss": 4.2246,
+      "step": 373760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.387203437190236e-05,
+      "loss": 4.2327,
+      "step": 374272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.386364842439184e-05,
+      "loss": 4.2213,
+      "step": 374784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.385526247688132e-05,
+      "loss": 4.2187,
+      "step": 375296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.384689290817453e-05,
+      "loss": 4.2268,
+      "step": 375808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.383850696066401e-05,
+      "loss": 4.2153,
+      "step": 376320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.383012101315349e-05,
+      "loss": 4.2259,
+      "step": 376832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.382173506564297e-05,
+      "loss": 4.2111,
+      "step": 377344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.381336549693618e-05,
+      "loss": 4.2205,
+      "step": 377856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.380497954942566e-05,
+      "loss": 4.2128,
+      "step": 378368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3796593601915145e-05,
+      "loss": 4.2153,
+      "step": 378880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3788207654404625e-05,
+      "loss": 4.2164,
+      "step": 379392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3779821706894105e-05,
+      "loss": 4.214,
+      "step": 379904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3771435759383585e-05,
+      "loss": 4.2167,
+      "step": 380416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3763049811873065e-05,
+      "loss": 4.2095,
+      "step": 380928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3754680243166274e-05,
+      "loss": 4.2219,
+      "step": 381440
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.205575466156006,
+      "eval_runtime": 295.4756,
+      "eval_samples_per_second": 1291.447,
+      "eval_steps_per_second": 40.359,
+      "step": 381600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3746294295655754e-05,
+      "loss": 4.2119,
+      "step": 381952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3737908348145234e-05,
+      "loss": 4.1986,
+      "step": 382464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3729522400634714e-05,
+      "loss": 4.2211,
+      "step": 382976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3721136453124194e-05,
+      "loss": 4.2086,
+      "step": 383488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3712750505613674e-05,
+      "loss": 4.2241,
+      "step": 384000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.370438093690688e-05,
+      "loss": 4.2101,
+      "step": 384512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.369599498939636e-05,
+      "loss": 4.1993,
+      "step": 385024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.368760904188584e-05,
+      "loss": 4.2044,
+      "step": 385536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.367923947317906e-05,
+      "loss": 4.2057,
+      "step": 386048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.367085352566854e-05,
+      "loss": 4.2166,
+      "step": 386560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.366246757815802e-05,
+      "loss": 4.2089,
+      "step": 387072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.36540816306475e-05,
+      "loss": 4.2103,
+      "step": 387584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.364571206194071e-05,
+      "loss": 4.2024,
+      "step": 388096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.363732611443019e-05,
+      "loss": 4.2012,
+      "step": 388608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.362894016691967e-05,
+      "loss": 4.208,
+      "step": 389120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.362055421940915e-05,
+      "loss": 4.1976,
+      "step": 389632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.361216827189863e-05,
+      "loss": 4.2032,
+      "step": 390144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.36037823243881e-05,
+      "loss": 4.2086,
+      "step": 390656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.359539637687758e-05,
+      "loss": 4.2066,
+      "step": 391168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.358701042936707e-05,
+      "loss": 4.212,
+      "step": 391680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3578640860660284e-05,
+      "loss": 4.2079,
+      "step": 392192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.357025491314976e-05,
+      "loss": 4.2085,
+      "step": 392704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.356186896563924e-05,
+      "loss": 4.2071,
+      "step": 393216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.355348301812872e-05,
+      "loss": 4.2135,
+      "step": 393728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3545113449421926e-05,
+      "loss": 4.1967,
+      "step": 394240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3536727501911406e-05,
+      "loss": 4.2112,
+      "step": 394752
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3528341554400886e-05,
+      "loss": 4.1968,
+      "step": 395264
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3519955606890366e-05,
+      "loss": 4.1978,
+      "step": 395776
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3511586038183575e-05,
+      "loss": 4.1932,
+      "step": 396288
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3503200090673055e-05,
+      "loss": 4.1979,
+      "step": 396800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3494814143162535e-05,
+      "loss": 4.203,
+      "step": 397312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.348642819565202e-05,
+      "loss": 4.2033,
+      "step": 397824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.34780422481415e-05,
+      "loss": 4.2081,
+      "step": 398336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.346965630063098e-05,
+      "loss": 4.209,
+      "step": 398848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.346127035312046e-05,
+      "loss": 4.1985,
+      "step": 399360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.345288440560994e-05,
+      "loss": 4.206,
+      "step": 399872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.344451483690315e-05,
+      "loss": 4.2003,
+      "step": 400384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.343614526819636e-05,
+      "loss": 4.1995,
+      "step": 400896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.342775932068584e-05,
+      "loss": 4.1945,
+      "step": 401408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.341937337317532e-05,
+      "loss": 4.1949,
+      "step": 401920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.34109874256648e-05,
+      "loss": 4.202,
+      "step": 402432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.340261785695801e-05,
+      "loss": 4.2054,
+      "step": 402944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.339423190944749e-05,
+      "loss": 4.1927,
+      "step": 403456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3385845961936975e-05,
+      "loss": 4.1984,
+      "step": 403968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3377460014426455e-05,
+      "loss": 4.2048,
+      "step": 404480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3369074066915935e-05,
+      "loss": 4.1995,
+      "step": 404992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3360704498209144e-05,
+      "loss": 4.2006,
+      "step": 405504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3352318550698624e-05,
+      "loss": 4.1933,
+      "step": 406016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3343932603188104e-05,
+      "loss": 4.1688,
+      "step": 406528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3335546655677584e-05,
+      "loss": 4.2057,
+      "step": 407040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3327160708167064e-05,
+      "loss": 4.1923,
+      "step": 407552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.331879113946027e-05,
+      "loss": 4.1925,
+      "step": 408064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.331040519194975e-05,
+      "loss": 4.1833,
+      "step": 408576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.330201924443923e-05,
+      "loss": 4.1784,
+      "step": 409088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.329363329692871e-05,
+      "loss": 4.1817,
+      "step": 409600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.328526372822193e-05,
+      "loss": 4.1948,
+      "step": 410112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.327687778071141e-05,
+      "loss": 4.1833,
+      "step": 410624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.326849183320089e-05,
+      "loss": 4.1886,
+      "step": 411136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.326010588569037e-05,
+      "loss": 4.1945,
+      "step": 411648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.325173631698358e-05,
+      "loss": 4.1825,
+      "step": 412160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.324335036947306e-05,
+      "loss": 4.1829,
+      "step": 412672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.323496442196254e-05,
+      "loss": 4.1918,
+      "step": 413184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.322657847445202e-05,
+      "loss": 4.1706,
+      "step": 413696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.32181925269415e-05,
+      "loss": 4.1822,
+      "step": 414208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.320980657943098e-05,
+      "loss": 4.1833,
+      "step": 414720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.320142063192046e-05,
+      "loss": 4.1946,
+      "step": 415232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.319303468440994e-05,
+      "loss": 4.1706,
+      "step": 415744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3184665115703146e-05,
+      "loss": 4.1832,
+      "step": 416256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3176279168192626e-05,
+      "loss": 4.1709,
+      "step": 416768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3167893220682106e-05,
+      "loss": 4.1955,
+      "step": 417280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3159507273171586e-05,
+      "loss": 4.1906,
+      "step": 417792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.31511377044648e-05,
+      "loss": 4.1888,
+      "step": 418304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.314275175695428e-05,
+      "loss": 4.1994,
+      "step": 418816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.313436580944376e-05,
+      "loss": 4.2005,
+      "step": 419328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3125979861933235e-05,
+      "loss": 4.2073,
+      "step": 419840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.311762667203018e-05,
+      "loss": 4.1824,
+      "step": 420352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.310924072451966e-05,
+      "loss": 4.187,
+      "step": 420864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.310085477700914e-05,
+      "loss": 4.186,
+      "step": 421376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.309246882949862e-05,
+      "loss": 4.1745,
+      "step": 421888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.308409926079183e-05,
+      "loss": 4.1912,
+      "step": 422400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3075713313281316e-05,
+      "loss": 4.1826,
+      "step": 422912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3067327365770796e-05,
+      "loss": 4.1914,
+      "step": 423424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3058941418260276e-05,
+      "loss": 4.1881,
+      "step": 423936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3050555470749756e-05,
+      "loss": 4.1732,
+      "step": 424448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3042169523239236e-05,
+      "loss": 4.1842,
+      "step": 424960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.303378357572871e-05,
+      "loss": 4.1739,
+      "step": 425472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3025414007021925e-05,
+      "loss": 4.1897,
+      "step": 425984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3017028059511405e-05,
+      "loss": 4.1868,
+      "step": 426496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3008658490804614e-05,
+      "loss": 4.1738,
+      "step": 427008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.3000272543294094e-05,
+      "loss": 4.1913,
+      "step": 427520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2991886595783574e-05,
+      "loss": 4.1794,
+      "step": 428032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2983500648273054e-05,
+      "loss": 4.1746,
+      "step": 428544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2975114700762534e-05,
+      "loss": 4.1779,
+      "step": 429056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2966728753252014e-05,
+      "loss": 4.1874,
+      "step": 429568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2958342805741494e-05,
+      "loss": 4.1824,
+      "step": 430080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2949956858230974e-05,
+      "loss": 4.186,
+      "step": 430592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.294158728952418e-05,
+      "loss": 4.1675,
+      "step": 431104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.293320134201366e-05,
+      "loss": 4.1802,
+      "step": 431616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.292481539450314e-05,
+      "loss": 4.1738,
+      "step": 432128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.291642944699262e-05,
+      "loss": 4.1873,
+      "step": 432640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.29080434994821e-05,
+      "loss": 4.1887,
+      "step": 433152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.289965755197158e-05,
+      "loss": 4.1755,
+      "step": 433664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.289127160446106e-05,
+      "loss": 4.1791,
+      "step": 434176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.288288565695054e-05,
+      "loss": 4.1853,
+      "step": 434688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.287453246704749e-05,
+      "loss": 4.1604,
+      "step": 435200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.286614651953697e-05,
+      "loss": 4.1718,
+      "step": 435712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.285776057202645e-05,
+      "loss": 4.1643,
+      "step": 436224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.284937462451593e-05,
+      "loss": 4.1771,
+      "step": 436736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.284098867700541e-05,
+      "loss": 4.1666,
+      "step": 437248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.283260272949489e-05,
+      "loss": 4.1749,
+      "step": 437760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.282421678198437e-05,
+      "loss": 4.1659,
+      "step": 438272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.281583083447385e-05,
+      "loss": 4.1659,
+      "step": 438784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2807461265767056e-05,
+      "loss": 4.1722,
+      "step": 439296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2799075318256536e-05,
+      "loss": 4.1648,
+      "step": 439808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2790705749549745e-05,
+      "loss": 4.1809,
+      "step": 440320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2782319802039225e-05,
+      "loss": 4.1759,
+      "step": 440832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2773933854528705e-05,
+      "loss": 4.1778,
+      "step": 441344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.276554790701819e-05,
+      "loss": 4.1679,
+      "step": 441856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.27571783383114e-05,
+      "loss": 4.177,
+      "step": 442368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.274879239080088e-05,
+      "loss": 4.1798,
+      "step": 442880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.274040644329036e-05,
+      "loss": 4.1679,
+      "step": 443392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.273202049577984e-05,
+      "loss": 4.1877,
+      "step": 443904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.272363454826932e-05,
+      "loss": 4.1649,
+      "step": 444416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.27152486007588e-05,
+      "loss": 4.175,
+      "step": 444928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.270686265324828e-05,
+      "loss": 4.16,
+      "step": 445440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.269847670573776e-05,
+      "loss": 4.1762,
+      "step": 445952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.269010713703097e-05,
+      "loss": 4.158,
+      "step": 446464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.268173756832418e-05,
+      "loss": 4.1671,
+      "step": 446976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2673367999617395e-05,
+      "loss": 4.1582,
+      "step": 447488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2664982052106875e-05,
+      "loss": 4.1819,
+      "step": 448000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2656596104596355e-05,
+      "loss": 4.1669,
+      "step": 448512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2648226535889564e-05,
+      "loss": 4.174,
+      "step": 449024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2639840588379044e-05,
+      "loss": 4.1617,
+      "step": 449536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2631454640868524e-05,
+      "loss": 4.175,
+      "step": 450048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2623068693358004e-05,
+      "loss": 4.1793,
+      "step": 450560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2614682745847484e-05,
+      "loss": 4.1723,
+      "step": 451072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2606296798336964e-05,
+      "loss": 4.1726,
+      "step": 451584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2597910850826443e-05,
+      "loss": 4.1723,
+      "step": 452096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2589524903315923e-05,
+      "loss": 4.1674,
+      "step": 452608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2581138955805403e-05,
+      "loss": 4.1783,
+      "step": 453120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.257275300829488e-05,
+      "loss": 4.1611,
+      "step": 453632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.256436706078436e-05,
+      "loss": 4.1703,
+      "step": 454144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.255598111327384e-05,
+      "loss": 4.1659,
+      "step": 454656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.254759516576332e-05,
+      "loss": 4.1659,
+      "step": 455168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.253922559705654e-05,
+      "loss": 4.166,
+      "step": 455680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.253083964954602e-05,
+      "loss": 4.1596,
+      "step": 456192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.252245370203549e-05,
+      "loss": 4.1745,
+      "step": 456704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.251406775452497e-05,
+      "loss": 4.155,
+      "step": 457216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.250568180701445e-05,
+      "loss": 4.1741,
+      "step": 457728
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.163763046264648,
+      "eval_runtime": 289.5615,
+      "eval_samples_per_second": 1317.824,
+      "eval_steps_per_second": 41.183,
+      "step": 457920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.249729585950393e-05,
+      "loss": 4.1635,
+      "step": 458240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.248890991199341e-05,
+      "loss": 4.1493,
+      "step": 458752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.248052396448289e-05,
+      "loss": 4.1697,
+      "step": 459264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.247213801697237e-05,
+      "loss": 4.1625,
+      "step": 459776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.246375206946185e-05,
+      "loss": 4.1728,
+      "step": 460288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.245536612195133e-05,
+      "loss": 4.1616,
+      "step": 460800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.244698017444081e-05,
+      "loss": 4.152,
+      "step": 461312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.243859422693029e-05,
+      "loss": 4.1524,
+      "step": 461824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.243020827941978e-05,
+      "loss": 4.1599,
+      "step": 462336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.242182233190926e-05,
+      "loss": 4.1684,
+      "step": 462848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.241343638439873e-05,
+      "loss": 4.1619,
+      "step": 463360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.240505043688821e-05,
+      "loss": 4.158,
+      "step": 463872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.239668086818143e-05,
+      "loss": 4.1607,
+      "step": 464384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.238829492067091e-05,
+      "loss": 4.1534,
+      "step": 464896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.237990897316038e-05,
+      "loss": 4.1578,
+      "step": 465408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.237152302564986e-05,
+      "loss": 4.1451,
+      "step": 465920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.236313707813934e-05,
+      "loss": 4.1575,
+      "step": 466432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.235475113062882e-05,
+      "loss": 4.1575,
+      "step": 466944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.23463651831183e-05,
+      "loss": 4.1579,
+      "step": 467456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.233797923560778e-05,
+      "loss": 4.1654,
+      "step": 467968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.232959328809726e-05,
+      "loss": 4.1664,
+      "step": 468480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2321223719390476e-05,
+      "loss": 4.1574,
+      "step": 468992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2312837771879956e-05,
+      "loss": 4.1569,
+      "step": 469504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2304451824369436e-05,
+      "loss": 4.1688,
+      "step": 470016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2296065876858916e-05,
+      "loss": 4.145,
+      "step": 470528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2287679929348396e-05,
+      "loss": 4.1691,
+      "step": 471040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2279293981837876e-05,
+      "loss": 4.1492,
+      "step": 471552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2270908034327356e-05,
+      "loss": 4.1483,
+      "step": 472064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2262538465620565e-05,
+      "loss": 4.147,
+      "step": 472576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.2254152518110045e-05,
+      "loss": 4.1541,
+      "step": 473088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2245766570599525e-05,
+      "loss": 4.1536,
+      "step": 473600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2237380623089005e-05,
+      "loss": 4.1597,
+      "step": 474112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2228994675578485e-05,
+      "loss": 4.159,
+      "step": 474624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2220608728067965e-05,
+      "loss": 4.162,
+      "step": 475136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2212222780557445e-05,
+      "loss": 4.1495,
+      "step": 475648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2203836833046925e-05,
+      "loss": 4.1624,
+      "step": 476160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.219548364314387e-05,
+      "loss": 4.1579,
+      "step": 476672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.218709769563335e-05,
+      "loss": 4.1479,
+      "step": 477184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.217871174812283e-05,
+      "loss": 4.1476,
+      "step": 477696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.217032580061231e-05,
+      "loss": 4.1523,
+      "step": 478208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.216193985310179e-05,
+      "loss": 4.1512,
+      "step": 478720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.215355390559126e-05,
+      "loss": 4.1602,
+      "step": 479232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.214516795808074e-05,
+      "loss": 4.1457,
+      "step": 479744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.213679838937396e-05,
+      "loss": 4.1516,
+      "step": 480256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.212842882066717e-05,
+      "loss": 4.1606,
+      "step": 480768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.212004287315665e-05,
+      "loss": 4.1498,
+      "step": 481280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2111656925646134e-05,
+      "loss": 4.1581,
+      "step": 481792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2103270978135614e-05,
+      "loss": 4.1473,
+      "step": 482304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2094885030625094e-05,
+      "loss": 4.1171,
+      "step": 482816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.20865154619183e-05,
+      "loss": 4.1638,
+      "step": 483328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.207812951440778e-05,
+      "loss": 4.1459,
+      "step": 483840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.206974356689726e-05,
+      "loss": 4.1492,
+      "step": 484352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2061357619386736e-05,
+      "loss": 4.1394,
+      "step": 484864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2052971671876216e-05,
+      "loss": 4.1285,
+      "step": 485376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2044585724365696e-05,
+      "loss": 4.1399,
+      "step": 485888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.203621615565891e-05,
+      "loss": 4.1461,
+      "step": 486400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2027830208148385e-05,
+      "loss": 4.1404,
+      "step": 486912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.201944426063787e-05,
+      "loss": 4.1408,
+      "step": 487424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.201105831312735e-05,
+      "loss": 4.1535,
+      "step": 487936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.200267236561683e-05,
+      "loss": 4.1359,
+      "step": 488448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.199428641810631e-05,
+      "loss": 4.134,
+      "step": 488960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.198590047059579e-05,
+      "loss": 4.1493,
+      "step": 489472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.197751452308527e-05,
+      "loss": 4.1232,
+      "step": 489984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.196914495437848e-05,
+      "loss": 4.1384,
+      "step": 490496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.196075900686796e-05,
+      "loss": 4.1394,
+      "step": 491008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.195237305935744e-05,
+      "loss": 4.1528,
+      "step": 491520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.194398711184692e-05,
+      "loss": 4.1254,
+      "step": 492032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.19356011643364e-05,
+      "loss": 4.1425,
+      "step": 492544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.192721521682588e-05,
+      "loss": 4.1245,
+      "step": 493056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.191884564811909e-05,
+      "loss": 4.1525,
+      "step": 493568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.191045970060857e-05,
+      "loss": 4.1491,
+      "step": 494080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1902073753098057e-05,
+      "loss": 4.1453,
+      "step": 494592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1893687805587537e-05,
+      "loss": 4.1513,
+      "step": 495104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1885301858077017e-05,
+      "loss": 4.1554,
+      "step": 495616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1876932289370226e-05,
+      "loss": 4.1646,
+      "step": 496128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1868562720663435e-05,
+      "loss": 4.1392,
+      "step": 496640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1860176773152915e-05,
+      "loss": 4.1448,
+      "step": 497152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1851790825642395e-05,
+      "loss": 4.1461,
+      "step": 497664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1843404878131875e-05,
+      "loss": 4.1298,
+      "step": 498176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1835018930621355e-05,
+      "loss": 4.145,
+      "step": 498688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1826632983110834e-05,
+      "loss": 4.1416,
+      "step": 499200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1818247035600314e-05,
+      "loss": 4.1467,
+      "step": 499712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1809861088089794e-05,
+      "loss": 4.1538,
+      "step": 500224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1801475140579274e-05,
+      "loss": 4.1308,
+      "step": 500736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1793089193068754e-05,
+      "loss": 4.135,
+      "step": 501248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1784703245558234e-05,
+      "loss": 4.1347,
+      "step": 501760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.1776317298047714e-05,
+      "loss": 4.1475,
+      "step": 502272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.176794772934093e-05,
+      "loss": 4.1407,
+      "step": 502784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.17595617818304e-05,
+      "loss": 4.1375,
+      "step": 503296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.175119221312362e-05,
+      "loss": 4.1356,
+      "step": 503808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.17428062656131e-05,
+      "loss": 4.1438,
+      "step": 504320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.173442031810257e-05,
+      "loss": 4.1338,
+      "step": 504832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.172603437059205e-05,
+      "loss": 4.136,
+      "step": 505344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.171764842308153e-05,
+      "loss": 4.15,
+      "step": 505856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.170926247557101e-05,
+      "loss": 4.1373,
+      "step": 506368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.170087652806049e-05,
+      "loss": 4.1408,
+      "step": 506880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.169249058054998e-05,
+      "loss": 4.125,
+      "step": 507392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.168412101184319e-05,
+      "loss": 4.138,
+      "step": 507904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.16757514431364e-05,
+      "loss": 4.1328,
+      "step": 508416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.166736549562588e-05,
+      "loss": 4.1447,
+      "step": 508928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.165897954811536e-05,
+      "loss": 4.1437,
+      "step": 509440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.165059360060484e-05,
+      "loss": 4.1403,
+      "step": 509952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.164220765309432e-05,
+      "loss": 4.1386,
+      "step": 510464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.16338217055838e-05,
+      "loss": 4.1407,
+      "step": 510976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.162543575807328e-05,
+      "loss": 4.1205,
+      "step": 511488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1617066189366486e-05,
+      "loss": 4.1328,
+      "step": 512000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1608680241855966e-05,
+      "loss": 4.1181,
+      "step": 512512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1600294294345446e-05,
+      "loss": 4.1383,
+      "step": 513024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1591908346834926e-05,
+      "loss": 4.1253,
+      "step": 513536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.158352239932441e-05,
+      "loss": 4.1309,
+      "step": 514048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.157515283061762e-05,
+      "loss": 4.1232,
+      "step": 514560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.15667668831071e-05,
+      "loss": 4.1262,
+      "step": 515072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.155838093559658e-05,
+      "loss": 4.1317,
+      "step": 515584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.154999498808606e-05,
+      "loss": 4.1233,
+      "step": 516096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.154162541937927e-05,
+      "loss": 4.1355,
+      "step": 516608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.153325585067248e-05,
+      "loss": 4.1405,
+      "step": 517120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.152486990316196e-05,
+      "loss": 4.1318,
+      "step": 517632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.151648395565144e-05,
+      "loss": 4.1358,
+      "step": 518144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.150809800814092e-05,
+      "loss": 4.131,
+      "step": 518656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.14997120606304e-05,
+      "loss": 4.1371,
+      "step": 519168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.149132611311988e-05,
+      "loss": 4.1335,
+      "step": 519680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1482940165609366e-05,
+      "loss": 4.1424,
+      "step": 520192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1474554218098846e-05,
+      "loss": 4.1248,
+      "step": 520704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1466168270588326e-05,
+      "loss": 4.1361,
+      "step": 521216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1457782323077806e-05,
+      "loss": 4.1241,
+      "step": 521728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1449396375567286e-05,
+      "loss": 4.1278,
+      "step": 522240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.144101042805676e-05,
+      "loss": 4.1237,
+      "step": 522752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1432640859349975e-05,
+      "loss": 4.1254,
+      "step": 523264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1424271290643184e-05,
+      "loss": 4.1203,
+      "step": 523776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1415885343132664e-05,
+      "loss": 4.1366,
+      "step": 524288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1407499395622144e-05,
+      "loss": 4.1293,
+      "step": 524800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1399113448111624e-05,
+      "loss": 4.1334,
+      "step": 525312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1390727500601104e-05,
+      "loss": 4.1202,
+      "step": 525824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.138235793189432e-05,
+      "loss": 4.1373,
+      "step": 526336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.13739719843838e-05,
+      "loss": 4.1372,
+      "step": 526848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.136558603687328e-05,
+      "loss": 4.1344,
+      "step": 527360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.135720008936276e-05,
+      "loss": 4.1342,
+      "step": 527872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.134881414185223e-05,
+      "loss": 4.1291,
+      "step": 528384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.134042819434171e-05,
+      "loss": 4.1294,
+      "step": 528896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.133204224683119e-05,
+      "loss": 4.1407,
+      "step": 529408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.132365629932067e-05,
+      "loss": 4.1193,
+      "step": 529920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.131528673061388e-05,
+      "loss": 4.1289,
+      "step": 530432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.130690078310336e-05,
+      "loss": 4.1303,
+      "step": 530944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.129853121439658e-05,
+      "loss": 4.1267,
+      "step": 531456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.129014526688606e-05,
+      "loss": 4.1196,
+      "step": 531968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1281775698179273e-05,
+      "loss": 4.1196,
+      "step": 532480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1273389750668753e-05,
+      "loss": 4.1358,
+      "step": 532992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.126500380315823e-05,
+      "loss": 4.1195,
+      "step": 533504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1256617855647707e-05,
+      "loss": 4.1361,
+      "step": 534016
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.132443904876709,
+      "eval_runtime": 339.1955,
+      "eval_samples_per_second": 1124.988,
+      "eval_steps_per_second": 35.157,
+      "step": 534240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.124824828694092e-05,
+      "loss": 4.1229,
+      "step": 534528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.12398623394304e-05,
+      "loss": 4.111,
+      "step": 535040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.123147639191988e-05,
+      "loss": 4.1319,
+      "step": 535552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1223090444409356e-05,
+      "loss": 4.1196,
+      "step": 536064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1214704496898835e-05,
+      "loss": 4.131,
+      "step": 536576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1206318549388315e-05,
+      "loss": 4.1273,
+      "step": 537088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1197932601877795e-05,
+      "loss": 4.1174,
+      "step": 537600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1189546654367275e-05,
+      "loss": 4.1149,
+      "step": 538112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1181160706856755e-05,
+      "loss": 4.1207,
+      "step": 538624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.117277475934624e-05,
+      "loss": 4.1288,
+      "step": 539136
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.116438881183572e-05,
+      "loss": 4.1208,
+      "step": 539648
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.11560028643252e-05,
+      "loss": 4.1196,
+      "step": 540160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.114763329561841e-05,
+      "loss": 4.119,
+      "step": 540672
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.113924734810789e-05,
+      "loss": 4.1161,
+      "step": 541184
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.113086140059737e-05,
+      "loss": 4.1196,
+      "step": 541696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.112247545308685e-05,
+      "loss": 4.1107,
+      "step": 542208
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.111408950557633e-05,
+      "loss": 4.1172,
+      "step": 542720
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.110571993686954e-05,
+      "loss": 4.1203,
+      "step": 543232
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.109733398935902e-05,
+      "loss": 4.1204,
+      "step": 543744
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.10889480418485e-05,
+      "loss": 4.123,
+      "step": 544256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.108056209433798e-05,
+      "loss": 4.1313,
+      "step": 544768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1072208904434925e-05,
+      "loss": 4.1216,
+      "step": 545280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1063822956924405e-05,
+      "loss": 4.1198,
+      "step": 545792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1055437009413885e-05,
+      "loss": 4.126,
+      "step": 546304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1047051061903365e-05,
+      "loss": 4.1141,
+      "step": 546816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.1038681493196574e-05,
+      "loss": 4.1274,
+      "step": 547328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.103031192448978e-05,
+      "loss": 4.1112,
+      "step": 547840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.102192597697926e-05,
+      "loss": 4.117,
+      "step": 548352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.101354002946874e-05,
+      "loss": 4.1123,
+      "step": 548864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.100515408195822e-05,
+      "loss": 4.1133,
+      "step": 549376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.09967681344477e-05,
+      "loss": 4.1135,
+      "step": 549888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.098838218693718e-05,
+      "loss": 4.121,
+      "step": 550400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.097999623942666e-05,
+      "loss": 4.1205,
+      "step": 550912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.097161029191615e-05,
+      "loss": 4.1235,
+      "step": 551424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.096322434440563e-05,
+      "loss": 4.1124,
+      "step": 551936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.095485477569884e-05,
+      "loss": 4.1272,
+      "step": 552448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.094646882818832e-05,
+      "loss": 4.1212,
+      "step": 552960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.09380828806778e-05,
+      "loss": 4.1073,
+      "step": 553472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.092969693316728e-05,
+      "loss": 4.1151,
+      "step": 553984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.092131098565676e-05,
+      "loss": 4.1129,
+      "step": 554496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.091292503814624e-05,
+      "loss": 4.1125,
+      "step": 555008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.090455546943945e-05,
+      "loss": 4.1203,
+      "step": 555520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.089616952192893e-05,
+      "loss": 4.1109,
+      "step": 556032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.088778357441841e-05,
+      "loss": 4.1153,
+      "step": 556544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.087939762690789e-05,
+      "loss": 4.1233,
+      "step": 557056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0871028058201096e-05,
+      "loss": 4.1144,
+      "step": 557568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.086264211069058e-05,
+      "loss": 4.117,
+      "step": 558080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.085425616318006e-05,
+      "loss": 4.1158,
+      "step": 558592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.084587021566954e-05,
+      "loss": 4.0774,
+      "step": 559104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0837484268159016e-05,
+      "loss": 4.1289,
+      "step": 559616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.082911469945223e-05,
+      "loss": 4.107,
+      "step": 560128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.082072875194171e-05,
+      "loss": 4.1169,
+      "step": 560640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.081234280443119e-05,
+      "loss": 4.0995,
+      "step": 561152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0803956856920665e-05,
+      "loss": 4.0947,
+      "step": 561664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0795570909410145e-05,
+      "loss": 4.1019,
+      "step": 562176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.078720134070336e-05,
+      "loss": 4.1079,
+      "step": 562688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0778815393192834e-05,
+      "loss": 4.1078,
+      "step": 563200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.077042944568232e-05,
+      "loss": 4.1018,
+      "step": 563712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.07620434981718e-05,
+      "loss": 4.1211,
+      "step": 564224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.075365755066128e-05,
+      "loss": 4.1023,
+      "step": 564736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.074527160315076e-05,
+      "loss": 4.0987,
+      "step": 565248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.073688565564024e-05,
+      "loss": 4.1138,
+      "step": 565760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.072849970812972e-05,
+      "loss": 4.0886,
+      "step": 566272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.072013013942293e-05,
+      "loss": 4.0983,
+      "step": 566784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.071174419191241e-05,
+      "loss": 4.1059,
+      "step": 567296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.070335824440189e-05,
+      "loss": 4.1157,
+      "step": 567808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.06949886756951e-05,
+      "loss": 4.0911,
+      "step": 568320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.068660272818458e-05,
+      "loss": 4.1061,
+      "step": 568832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.067821678067406e-05,
+      "loss": 4.0901,
+      "step": 569344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.066983083316354e-05,
+      "loss": 4.1171,
+      "step": 569856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.066144488565302e-05,
+      "loss": 4.1108,
+      "step": 570368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0653058938142505e-05,
+      "loss": 4.1116,
+      "step": 570880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0644672990631985e-05,
+      "loss": 4.1135,
+      "step": 571392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.0636303421925194e-05,
+      "loss": 4.1217,
+      "step": 571904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.06279338532184e-05,
+      "loss": 4.1288,
+      "step": 572416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.061956428451161e-05,
+      "loss": 4.1053,
+      "step": 572928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.061117833700109e-05,
+      "loss": 4.108,
+      "step": 573440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.060279238949057e-05,
+      "loss": 4.1129,
+      "step": 573952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.059440644198005e-05,
+      "loss": 4.0939,
+      "step": 574464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.058602049446953e-05,
+      "loss": 4.1059,
+      "step": 574976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.057763454695901e-05,
+      "loss": 4.1083,
+      "step": 575488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.056924859944849e-05,
+      "loss": 4.1126,
+      "step": 576000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.056086265193797e-05,
+      "loss": 4.1148,
+      "step": 576512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.055247670442746e-05,
+      "loss": 4.0989,
+      "step": 577024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.054409075691694e-05,
+      "loss": 4.0985,
+      "step": 577536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.053570480940642e-05,
+      "loss": 4.1021,
+      "step": 578048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.05273188618959e-05,
+      "loss": 4.1122,
+      "step": 578560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.051893291438538e-05,
+      "loss": 4.1074,
+      "step": 579072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.051056334567859e-05,
+      "loss": 4.1017,
+      "step": 579584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.050217739816807e-05,
+      "loss": 4.1001,
+      "step": 580096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.049379145065755e-05,
+      "loss": 4.1098,
+      "step": 580608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.048540550314702e-05,
+      "loss": 4.1005,
+      "step": 581120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.04770195556365e-05,
+      "loss": 4.1001,
+      "step": 581632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.046863360812598e-05,
+      "loss": 4.1153,
+      "step": 582144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.046024766061546e-05,
+      "loss": 4.1037,
+      "step": 582656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.045186171310494e-05,
+      "loss": 4.1046,
+      "step": 583168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0443492144398157e-05,
+      "loss": 4.0956,
+      "step": 583680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0435106196887637e-05,
+      "loss": 4.0999,
+      "step": 584192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0426720249377116e-05,
+      "loss": 4.0973,
+      "step": 584704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0418334301866596e-05,
+      "loss": 4.113,
+      "step": 585216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0409964733159806e-05,
+      "loss": 4.1078,
+      "step": 585728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0401578785649285e-05,
+      "loss": 4.1069,
+      "step": 586240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0393192838138765e-05,
+      "loss": 4.1067,
+      "step": 586752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0384823269431975e-05,
+      "loss": 4.1036,
+      "step": 587264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0376437321921454e-05,
+      "loss": 4.0917,
+      "step": 587776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0368051374410934e-05,
+      "loss": 4.0922,
+      "step": 588288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0359665426900414e-05,
+      "loss": 4.0866,
+      "step": 588800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0351279479389894e-05,
+      "loss": 4.1055,
+      "step": 589312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0342893531879374e-05,
+      "loss": 4.0916,
+      "step": 589824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.033450758436886e-05,
+      "loss": 4.0955,
+      "step": 590336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.032613801566207e-05,
+      "loss": 4.0876,
+      "step": 590848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.031775206815155e-05,
+      "loss": 4.0971,
+      "step": 591360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.030936612064103e-05,
+      "loss": 4.0966,
+      "step": 591872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.030098017313051e-05,
+      "loss": 4.088,
+      "step": 592384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.029259422561999e-05,
+      "loss": 4.1078,
+      "step": 592896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.028420827810947e-05,
+      "loss": 4.1025,
+      "step": 593408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.027582233059895e-05,
+      "loss": 4.0984,
+      "step": 593920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.026743638308843e-05,
+      "loss": 4.1042,
+      "step": 594432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.025906681438164e-05,
+      "loss": 4.0979,
+      "step": 594944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.025069724567485e-05,
+      "loss": 4.1013,
+      "step": 595456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.024231129816433e-05,
+      "loss": 4.1037,
+      "step": 595968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0233925350653815e-05,
+      "loss": 4.1045,
+      "step": 596480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0225539403143295e-05,
+      "loss": 4.0937,
+      "step": 596992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0217153455632775e-05,
+      "loss": 4.1015,
+      "step": 597504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0208767508122255e-05,
+      "loss": 4.0943,
+      "step": 598016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0200397939415464e-05,
+      "loss": 4.0955,
+      "step": 598528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0192011991904944e-05,
+      "loss": 4.0867,
+      "step": 599040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0183626044394424e-05,
+      "loss": 4.0928,
+      "step": 599552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0175240096883904e-05,
+      "loss": 4.0884,
+      "step": 600064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0166854149373383e-05,
+      "loss": 4.1025,
+      "step": 600576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.015846820186286e-05,
+      "loss": 4.0949,
+      "step": 601088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.015008225435234e-05,
+      "loss": 4.1009,
+      "step": 601600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0141696306841817e-05,
+      "loss": 4.084,
+      "step": 602112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0133310359331297e-05,
+      "loss": 4.1088,
+      "step": 602624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.012492441182078e-05,
+      "loss": 4.1011,
+      "step": 603136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.011655484311399e-05,
+      "loss": 4.1018,
+      "step": 603648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.010816889560347e-05,
+      "loss": 4.1018,
+      "step": 604160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.009978294809295e-05,
+      "loss": 4.097,
+      "step": 604672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.009139700058243e-05,
+      "loss": 4.0944,
+      "step": 605184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.008302743187564e-05,
+      "loss": 4.1064,
+      "step": 605696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.007464148436512e-05,
+      "loss": 4.0882,
+      "step": 606208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.00662555368546e-05,
+      "loss": 4.0976,
+      "step": 606720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.005786958934408e-05,
+      "loss": 4.0987,
+      "step": 607232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.004950002063729e-05,
+      "loss": 4.0937,
+      "step": 607744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.004111407312677e-05,
+      "loss": 4.0867,
+      "step": 608256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.003272812561625e-05,
+      "loss": 4.0879,
+      "step": 608768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0024358556909466e-05,
+      "loss": 4.1021,
+      "step": 609280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0015972609398946e-05,
+      "loss": 4.0894,
+      "step": 609792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.0007586661888426e-05,
+      "loss": 4.1019,
+      "step": 610304
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.107419490814209,
+      "eval_runtime": 325.8697,
+      "eval_samples_per_second": 1170.993,
+      "eval_steps_per_second": 36.594,
+      "step": 610560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9999200714377906e-05,
+      "loss": 4.0853,
+      "step": 610816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9990814766867386e-05,
+      "loss": 4.0836,
+      "step": 611328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9982428819356866e-05,
+      "loss": 4.0917,
+      "step": 611840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9974042871846346e-05,
+      "loss": 4.0905,
+      "step": 612352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9965656924335826e-05,
+      "loss": 4.1006,
+      "step": 612864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9957270976825306e-05,
+      "loss": 4.0942,
+      "step": 613376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9948885029314786e-05,
+      "loss": 4.0848,
+      "step": 613888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9940499081804266e-05,
+      "loss": 4.0847,
+      "step": 614400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9932113134293746e-05,
+      "loss": 4.0863,
+      "step": 614912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.992372718678322e-05,
+      "loss": 4.0982,
+      "step": 615424
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9915341239272706e-05,
+      "loss": 4.0884,
+      "step": 615936
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9906955291762185e-05,
+      "loss": 4.0861,
+      "step": 616448
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9898569344251665e-05,
+      "loss": 4.0968,
+      "step": 616960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9890183396741145e-05,
+      "loss": 4.0772,
+      "step": 617472
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9881797449230625e-05,
+      "loss": 4.0839,
+      "step": 617984
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9873411501720105e-05,
+      "loss": 4.0811,
+      "step": 618496
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9865025554209585e-05,
+      "loss": 4.0861,
+      "step": 619008
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9856639606699065e-05,
+      "loss": 4.0888,
+      "step": 619520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9848253659188545e-05,
+      "loss": 4.0887,
+      "step": 620032
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9839884090481754e-05,
+      "loss": 4.0888,
+      "step": 620544
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.9831498142971234e-05,
+      "loss": 4.1026,
+      "step": 621056
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.982312857426444e-05,
+      "loss": 4.0896,
+      "step": 621568
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.981474262675392e-05,
+      "loss": 4.0913,
+      "step": 622080
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.98063566792434e-05,
+      "loss": 4.0895,
+      "step": 622592
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.979797073173288e-05,
+      "loss": 4.0863,
+      "step": 623104
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.978958478422237e-05,
+      "loss": 4.0946,
+      "step": 623616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.978119883671185e-05,
+      "loss": 4.0824,
+      "step": 624128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.977281288920133e-05,
+      "loss": 4.0862,
+      "step": 624640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.976442694169081e-05,
+      "loss": 4.0752,
+      "step": 625152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.975604099418029e-05,
+      "loss": 4.0828,
+      "step": 625664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.97476714254735e-05,
+      "loss": 4.0802,
+      "step": 626176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.973928547796298e-05,
+      "loss": 4.0884,
+      "step": 626688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.973089953045246e-05,
+      "loss": 4.0928,
+      "step": 627200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.972251358294193e-05,
+      "loss": 4.0943,
+      "step": 627712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.971412763543141e-05,
+      "loss": 4.0836,
+      "step": 628224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.970574168792089e-05,
+      "loss": 4.0897,
+      "step": 628736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.969735574041037e-05,
+      "loss": 4.0933,
+      "step": 629248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.968896979289985e-05,
+      "loss": 4.0806,
+      "step": 629760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.968060022419307e-05,
+      "loss": 4.081,
+      "step": 630272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.967221427668255e-05,
+      "loss": 4.0826,
+      "step": 630784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.966382832917203e-05,
+      "loss": 4.0867,
+      "step": 631296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.965544238166151e-05,
+      "loss": 4.0825,
+      "step": 631808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9647072812954717e-05,
+      "loss": 4.0871,
+      "step": 632320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9638686865444197e-05,
+      "loss": 4.0853,
+      "step": 632832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9630300917933676e-05,
+      "loss": 4.09,
+      "step": 633344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9621914970423156e-05,
+      "loss": 4.0854,
+      "step": 633856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9613545401716366e-05,
+      "loss": 4.0857,
+      "step": 634368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9605159454205845e-05,
+      "loss": 4.085,
+      "step": 634880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9596773506695325e-05,
+      "loss": 4.0481,
+      "step": 635392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9588387559184805e-05,
+      "loss": 4.0961,
+      "step": 635904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.958001799047802e-05,
+      "loss": 4.0763,
+      "step": 636416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.957164842177124e-05,
+      "loss": 4.0876,
+      "step": 636928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.956326247426071e-05,
+      "loss": 4.0714,
+      "step": 637440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.955487652675019e-05,
+      "loss": 4.0659,
+      "step": 637952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.954649057923967e-05,
+      "loss": 4.068,
+      "step": 638464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.953812101053288e-05,
+      "loss": 4.0768,
+      "step": 638976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.952973506302236e-05,
+      "loss": 4.077,
+      "step": 639488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.952134911551184e-05,
+      "loss": 4.0733,
+      "step": 640000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.951296316800132e-05,
+      "loss": 4.0864,
+      "step": 640512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.95045772204908e-05,
+      "loss": 4.0761,
+      "step": 641024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.949619127298028e-05,
+      "loss": 4.0682,
+      "step": 641536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.948780532546976e-05,
+      "loss": 4.0823,
+      "step": 642048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9479419377959246e-05,
+      "loss": 4.0619,
+      "step": 642560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9471049809252455e-05,
+      "loss": 4.0665,
+      "step": 643072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9462663861741935e-05,
+      "loss": 4.0763,
+      "step": 643584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9454277914231415e-05,
+      "loss": 4.083,
+      "step": 644096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9445891966720895e-05,
+      "loss": 4.0655,
+      "step": 644608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.9437522398014104e-05,
+      "loss": 4.0742,
+      "step": 645120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.942915282930731e-05,
+      "loss": 4.0588,
+      "step": 645632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.942076688179679e-05,
+      "loss": 4.084,
+      "step": 646144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.941238093428627e-05,
+      "loss": 4.0831,
+      "step": 646656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.940399498677575e-05,
+      "loss": 4.0795,
+      "step": 647168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.939560903926523e-05,
+      "loss": 4.085,
+      "step": 647680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.938722309175471e-05,
+      "loss": 4.0898,
+      "step": 648192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.93788371442442e-05,
+      "loss": 4.1005,
+      "step": 648704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.937046757553741e-05,
+      "loss": 4.0772,
+      "step": 649216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.936208162802689e-05,
+      "loss": 4.0784,
+      "step": 649728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.93537120593201e-05,
+      "loss": 4.0777,
+      "step": 650240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.934532611180958e-05,
+      "loss": 4.0717,
+      "step": 650752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.933694016429906e-05,
+      "loss": 4.0734,
+      "step": 651264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.932855421678854e-05,
+      "loss": 4.0817,
+      "step": 651776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.932016826927802e-05,
+      "loss": 4.0836,
+      "step": 652288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.93117823217675e-05,
+      "loss": 4.0893,
+      "step": 652800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.930339637425698e-05,
+      "loss": 4.0678,
+      "step": 653312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.929501042674646e-05,
+      "loss": 4.0676,
+      "step": 653824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.928662447923594e-05,
+      "loss": 4.0717,
+      "step": 654336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.927823853172542e-05,
+      "loss": 4.0857,
+      "step": 654848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.92698525842149e-05,
+      "loss": 4.0808,
+      "step": 655360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.926148301550811e-05,
+      "loss": 4.0687,
+      "step": 655872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.925309706799759e-05,
+      "loss": 4.0715,
+      "step": 656384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9244711120487066e-05,
+      "loss": 4.0804,
+      "step": 656896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9236325172976546e-05,
+      "loss": 4.0737,
+      "step": 657408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9227939225466026e-05,
+      "loss": 4.0685,
+      "step": 657920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9219553277955506e-05,
+      "loss": 4.0857,
+      "step": 658432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9211167330444986e-05,
+      "loss": 4.078,
+      "step": 658944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9202781382934466e-05,
+      "loss": 4.0754,
+      "step": 659456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9194395435423946e-05,
+      "loss": 4.0689,
+      "step": 659968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9186025866717155e-05,
+      "loss": 4.07,
+      "step": 660480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9177639919206635e-05,
+      "loss": 4.0695,
+      "step": 660992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9169253971696115e-05,
+      "loss": 4.0822,
+      "step": 661504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.91608680241856e-05,
+      "loss": 4.0803,
+      "step": 662016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.915249845547881e-05,
+      "loss": 4.0761,
+      "step": 662528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.914411250796829e-05,
+      "loss": 4.0805,
+      "step": 663040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.913572656045777e-05,
+      "loss": 4.0712,
+      "step": 663552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.912734061294725e-05,
+      "loss": 4.0709,
+      "step": 664064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.911897104424046e-05,
+      "loss": 4.0629,
+      "step": 664576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.911058509672994e-05,
+      "loss": 4.0568,
+      "step": 665088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.910219914921942e-05,
+      "loss": 4.079,
+      "step": 665600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.90938132017089e-05,
+      "loss": 4.0661,
+      "step": 666112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.908542725419838e-05,
+      "loss": 4.063,
+      "step": 666624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.907704130668786e-05,
+      "loss": 4.0602,
+      "step": 667136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.906867173798107e-05,
+      "loss": 4.0613,
+      "step": 667648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9060285790470555e-05,
+      "loss": 4.0722,
+      "step": 668160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9051899842960035e-05,
+      "loss": 4.0641,
+      "step": 668672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9043513895449515e-05,
+      "loss": 4.0766,
+      "step": 669184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9035144326742724e-05,
+      "loss": 4.0748,
+      "step": 669696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9026758379232204e-05,
+      "loss": 4.0743,
+      "step": 670208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9018372431721684e-05,
+      "loss": 4.0742,
+      "step": 670720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9009986484211164e-05,
+      "loss": 4.0702,
+      "step": 671232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.900161691550437e-05,
+      "loss": 4.0724,
+      "step": 671744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.899323096799385e-05,
+      "loss": 4.0775,
+      "step": 672256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.898484502048333e-05,
+      "loss": 4.0753,
+      "step": 672768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.897645907297281e-05,
+      "loss": 4.0685,
+      "step": 673280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.896808950426602e-05,
+      "loss": 4.0724,
+      "step": 673792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.895970355675551e-05,
+      "loss": 4.0677,
+      "step": 674304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.895133398804872e-05,
+      "loss": 4.0671,
+      "step": 674816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.89429480405382e-05,
+      "loss": 4.0586,
+      "step": 675328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.893456209302768e-05,
+      "loss": 4.0639,
+      "step": 675840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.892617614551716e-05,
+      "loss": 4.0595,
+      "step": 676352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.891779019800664e-05,
+      "loss": 4.0738,
+      "step": 676864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.890940425049612e-05,
+      "loss": 4.0658,
+      "step": 677376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.89010183029856e-05,
+      "loss": 4.0786,
+      "step": 677888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.889263235547508e-05,
+      "loss": 4.056,
+      "step": 678400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.888424640796455e-05,
+      "loss": 4.0806,
+      "step": 678912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.887586046045403e-05,
+      "loss": 4.0741,
+      "step": 679424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.886747451294351e-05,
+      "loss": 4.0764,
+      "step": 679936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.885910494423673e-05,
+      "loss": 4.0732,
+      "step": 680448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.885071899672621e-05,
+      "loss": 4.0687,
+      "step": 680960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.884233304921569e-05,
+      "loss": 4.0692,
+      "step": 681472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8833947101705167e-05,
+      "loss": 4.0803,
+      "step": 681984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8825561154194647e-05,
+      "loss": 4.0588,
+      "step": 682496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8817175206684127e-05,
+      "loss": 4.0722,
+      "step": 683008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.880882201678107e-05,
+      "loss": 4.0696,
+      "step": 683520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.880043606927055e-05,
+      "loss": 4.0655,
+      "step": 684032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8792050121760025e-05,
+      "loss": 4.0645,
+      "step": 684544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8783664174249505e-05,
+      "loss": 4.0622,
+      "step": 685056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.877529460554272e-05,
+      "loss": 4.0743,
+      "step": 685568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.87669086580322e-05,
+      "loss": 4.0658,
+      "step": 686080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.875852271052168e-05,
+      "loss": 4.0734,
+      "step": 686592
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.087892532348633,
+      "eval_runtime": 294.9945,
+      "eval_samples_per_second": 1293.553,
+      "eval_steps_per_second": 40.424,
+      "step": 686880
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.875013676301116e-05,
+      "loss": 4.0577,
+      "step": 687104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.874175081550064e-05,
+      "loss": 4.0592,
+      "step": 687616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.873336486799012e-05,
+      "loss": 4.0623,
+      "step": 688128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.87249789204796e-05,
+      "loss": 4.0642,
+      "step": 688640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.871660935177281e-05,
+      "loss": 4.077,
+      "step": 689152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.870822340426229e-05,
+      "loss": 4.0662,
+      "step": 689664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.869983745675177e-05,
+      "loss": 4.06,
+      "step": 690176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.869145150924125e-05,
+      "loss": 4.0613,
+      "step": 690688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.868306556173073e-05,
+      "loss": 4.0575,
+      "step": 691200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.867469599302394e-05,
+      "loss": 4.0715,
+      "step": 691712
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.866631004551342e-05,
+      "loss": 4.0618,
+      "step": 692224
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.86579240980029e-05,
+      "loss": 4.0612,
+      "step": 692736
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.864953815049238e-05,
+      "loss": 4.0698,
+      "step": 693248
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.8641152202981865e-05,
+      "loss": 4.0509,
+      "step": 693760
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.8632782634275074e-05,
+      "loss": 4.0565,
+      "step": 694272
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.8624396686764554e-05,
+      "loss": 4.0534,
+      "step": 694784
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.8616010739254034e-05,
+      "loss": 4.0622,
+      "step": 695296
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.8607624791743514e-05,
+      "loss": 4.0582,
+      "step": 695808
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.8599238844232994e-05,
+      "loss": 4.066,
+      "step": 696320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.85908692755262e-05,
+      "loss": 4.0631,
+      "step": 696832
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.858248332801568e-05,
+      "loss": 4.076,
+      "step": 697344
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.857409738050516e-05,
+      "loss": 4.0627,
+      "step": 697856
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.856571143299464e-05,
+      "loss": 4.067,
+      "step": 698368
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.855732548548412e-05,
+      "loss": 4.0602,
+      "step": 698880
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.85489395379736e-05,
+      "loss": 4.0643,
+      "step": 699392
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.854055359046308e-05,
+      "loss": 4.066,
+      "step": 699904
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.85321840217563e-05,
+      "loss": 4.0533,
+      "step": 700416
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.852379807424578e-05,
+      "loss": 4.0647,
+      "step": 700928
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.851541212673526e-05,
+      "loss": 4.0501,
+      "step": 701440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.850702617922474e-05,
+      "loss": 4.0568,
+      "step": 701952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.849864023171421e-05,
+      "loss": 4.0562,
+      "step": 702464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.849025428420369e-05,
+      "loss": 4.0601,
+      "step": 702976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.848186833669317e-05,
+      "loss": 4.0696,
+      "step": 703488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.847348238918265e-05,
+      "loss": 4.0675,
+      "step": 704000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.846509644167213e-05,
+      "loss": 4.0598,
+      "step": 704512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8456743251769076e-05,
+      "loss": 4.0601,
+      "step": 705024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8448357304258556e-05,
+      "loss": 4.072,
+      "step": 705536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8439971356748036e-05,
+      "loss": 4.055,
+      "step": 706048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8431585409237516e-05,
+      "loss": 4.0542,
+      "step": 706560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8423199461726996e-05,
+      "loss": 4.0559,
+      "step": 707072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8414813514216476e-05,
+      "loss": 4.0617,
+      "step": 707584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8406427566705956e-05,
+      "loss": 4.058,
+      "step": 708096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8398041619195436e-05,
+      "loss": 4.0575,
+      "step": 708608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8389655671684916e-05,
+      "loss": 4.0616,
+      "step": 709120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8381286102978125e-05,
+      "loss": 4.0606,
+      "step": 709632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8372900155467605e-05,
+      "loss": 4.0619,
+      "step": 710144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8364514207957085e-05,
+      "loss": 4.0606,
+      "step": 710656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8356144639250294e-05,
+      "loss": 4.0609,
+      "step": 711168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8347758691739774e-05,
+      "loss": 4.0247,
+      "step": 711680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8339372744229254e-05,
+      "loss": 4.067,
+      "step": 712192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.833098679671874e-05,
+      "loss": 4.0507,
+      "step": 712704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.832260084920822e-05,
+      "loss": 4.0648,
+      "step": 713216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.83142149016977e-05,
+      "loss": 4.0472,
+      "step": 713728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.830582895418718e-05,
+      "loss": 4.0409,
+      "step": 714240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.829745938548039e-05,
+      "loss": 4.0439,
+      "step": 714752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.828907343796987e-05,
+      "loss": 4.0508,
+      "step": 715264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.828068749045935e-05,
+      "loss": 4.0605,
+      "step": 715776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.827230154294883e-05,
+      "loss": 4.0429,
+      "step": 716288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.826391559543831e-05,
+      "loss": 4.0634,
+      "step": 716800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.825552964792779e-05,
+      "loss": 4.053,
+      "step": 717312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.824714370041727e-05,
+      "loss": 4.0383,
+      "step": 717824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.823875775290675e-05,
+      "loss": 4.0575,
+      "step": 718336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.823038818419996e-05,
+      "loss": 4.0417,
+      "step": 718848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8222018615493174e-05,
+      "loss": 4.0395,
+      "step": 719360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8213632667982654e-05,
+      "loss": 4.0536,
+      "step": 719872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8205246720472134e-05,
+      "loss": 4.0595,
+      "step": 720384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8196860772961614e-05,
+      "loss": 4.0433,
+      "step": 720896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8188474825451094e-05,
+      "loss": 4.048,
+      "step": 721408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8180088877940574e-05,
+      "loss": 4.0386,
+      "step": 721920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.817170293043005e-05,
+      "loss": 4.0572,
+      "step": 722432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.816331698291953e-05,
+      "loss": 4.0596,
+      "step": 722944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.815494741421274e-05,
+      "loss": 4.0543,
+      "step": 723456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.814656146670222e-05,
+      "loss": 4.0614,
+      "step": 723968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8138175519191696e-05,
+      "loss": 4.0655,
+      "step": 724480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.812980595048491e-05,
+      "loss": 4.074,
+      "step": 724992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.812142000297439e-05,
+      "loss": 4.0544,
+      "step": 725504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.811303405546387e-05,
+      "loss": 4.0561,
+      "step": 726016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.810464810795335e-05,
+      "loss": 4.0501,
+      "step": 726528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.809626216044283e-05,
+      "loss": 4.0519,
+      "step": 727040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.808787621293231e-05,
+      "loss": 4.0469,
+      "step": 727552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.807950664422552e-05,
+      "loss": 4.0589,
+      "step": 728064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8071120696715e-05,
+      "loss": 4.0572,
+      "step": 728576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.806273474920448e-05,
+      "loss": 4.0662,
+      "step": 729088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.805434880169396e-05,
+      "loss": 4.0411,
+      "step": 729600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.804596285418344e-05,
+      "loss": 4.0442,
+      "step": 730112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.803759328547665e-05,
+      "loss": 4.0506,
+      "step": 730624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.802920733796613e-05,
+      "loss": 4.0547,
+      "step": 731136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.802082139045561e-05,
+      "loss": 4.0573,
+      "step": 731648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.8012435442945097e-05,
+      "loss": 4.0497,
+      "step": 732160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8004049495434577e-05,
+      "loss": 4.0472,
+      "step": 732672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7995679926727786e-05,
+      "loss": 4.0561,
+      "step": 733184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7987293979217266e-05,
+      "loss": 4.0488,
+      "step": 733696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7978908031706746e-05,
+      "loss": 4.043,
+      "step": 734208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7970522084196225e-05,
+      "loss": 4.0614,
+      "step": 734720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7962152515489435e-05,
+      "loss": 4.0533,
+      "step": 735232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7953766567978915e-05,
+      "loss": 4.0541,
+      "step": 735744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7945380620468395e-05,
+      "loss": 4.0437,
+      "step": 736256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7936994672957874e-05,
+      "loss": 4.0449,
+      "step": 736768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7928608725447354e-05,
+      "loss": 4.0456,
+      "step": 737280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7920222777936834e-05,
+      "loss": 4.0578,
+      "step": 737792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.791185320923005e-05,
+      "loss": 4.0611,
+      "step": 738304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.790346726171953e-05,
+      "loss": 4.0492,
+      "step": 738816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.789508131420901e-05,
+      "loss": 4.0584,
+      "step": 739328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.788669536669849e-05,
+      "loss": 4.0479,
+      "step": 739840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.787830941918797e-05,
+      "loss": 4.05,
+      "step": 740352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.786993985048118e-05,
+      "loss": 4.038,
+      "step": 740864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.786155390297066e-05,
+      "loss": 4.0377,
+      "step": 741376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.785316795546014e-05,
+      "loss": 4.0488,
+      "step": 741888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.784478200794962e-05,
+      "loss": 4.0484,
+      "step": 742400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.78363960604391e-05,
+      "loss": 4.0363,
+      "step": 742912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.782801011292858e-05,
+      "loss": 4.0383,
+      "step": 743424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.781962416541805e-05,
+      "loss": 4.0391,
+      "step": 743936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.781123821790753e-05,
+      "loss": 4.0482,
+      "step": 744448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.780286864920075e-05,
+      "loss": 4.0401,
+      "step": 744960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7794482701690235e-05,
+      "loss": 4.0525,
+      "step": 745472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7786113132983444e-05,
+      "loss": 4.0517,
+      "step": 745984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7777727185472924e-05,
+      "loss": 4.0502,
+      "step": 746496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7769341237962404e-05,
+      "loss": 4.0454,
+      "step": 747008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7760955290451884e-05,
+      "loss": 4.0517,
+      "step": 747520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.775256934294136e-05,
+      "loss": 4.0481,
+      "step": 748032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.774418339543084e-05,
+      "loss": 4.0555,
+      "step": 748544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.773579744792032e-05,
+      "loss": 4.0525,
+      "step": 749056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.77274115004098e-05,
+      "loss": 4.0435,
+      "step": 749568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7719041931703006e-05,
+      "loss": 4.0494,
+      "step": 750080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7710655984192486e-05,
+      "loss": 4.0474,
+      "step": 750592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.770227003668197e-05,
+      "loss": 4.0405,
+      "step": 751104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.769388408917145e-05,
+      "loss": 4.0395,
+      "step": 751616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.768549814166093e-05,
+      "loss": 4.0422,
+      "step": 752128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.767714495175788e-05,
+      "loss": 4.0335,
+      "step": 752640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.766875900424736e-05,
+      "loss": 4.0525,
+      "step": 753152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.766037305673683e-05,
+      "loss": 4.0433,
+      "step": 753664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.765198710922631e-05,
+      "loss": 4.0532,
+      "step": 754176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.764360116171579e-05,
+      "loss": 4.0337,
+      "step": 754688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.763521521420527e-05,
+      "loss": 4.0531,
+      "step": 755200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.762684564549848e-05,
+      "loss": 4.0532,
+      "step": 755712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.761845969798796e-05,
+      "loss": 4.0527,
+      "step": 756224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.761007375047744e-05,
+      "loss": 4.0499,
+      "step": 756736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7601687802966926e-05,
+      "loss": 4.0441,
+      "step": 757248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7593301855456406e-05,
+      "loss": 4.0513,
+      "step": 757760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7584932286749615e-05,
+      "loss": 4.0547,
+      "step": 758272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7576546339239095e-05,
+      "loss": 4.0381,
+      "step": 758784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7568160391728575e-05,
+      "loss": 4.0492,
+      "step": 759296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7559790823021784e-05,
+      "loss": 4.0485,
+      "step": 759808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7551404875511264e-05,
+      "loss": 4.0437,
+      "step": 760320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7543018928000744e-05,
+      "loss": 4.0411,
+      "step": 760832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7534632980490224e-05,
+      "loss": 4.0374,
+      "step": 761344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7526247032979704e-05,
+      "loss": 4.0498,
+      "step": 761856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7517861085469184e-05,
+      "loss": 4.0458,
+      "step": 762368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.7509475137958664e-05,
+      "loss": 4.0511,
+      "step": 762880
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.071868896484375,
+      "eval_runtime": 297.0603,
+      "eval_samples_per_second": 1284.557,
+      "eval_steps_per_second": 40.143,
+      "step": 763200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7501089190448144e-05,
+      "loss": 4.0278,
+      "step": 763392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7492703242937624e-05,
+      "loss": 4.0381,
+      "step": 763904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7484317295427104e-05,
+      "loss": 4.0386,
+      "step": 764416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.747593134791659e-05,
+      "loss": 4.0418,
+      "step": 764928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.746754540040607e-05,
+      "loss": 4.0532,
+      "step": 765440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7459159452895544e-05,
+      "loss": 4.0487,
+      "step": 765952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7450773505385024e-05,
+      "loss": 4.0349,
+      "step": 766464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7442387557874504e-05,
+      "loss": 4.039,
+      "step": 766976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7434001610363984e-05,
+      "loss": 4.0355,
+      "step": 767488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7425615662853463e-05,
+      "loss": 4.0452,
+      "step": 768000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7417229715342943e-05,
+      "loss": 4.0411,
+      "step": 768512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7408843767832423e-05,
+      "loss": 4.0417,
+      "step": 769024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.74004578203219e-05,
+      "loss": 4.0471,
+      "step": 769536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.739207187281138e-05,
+      "loss": 4.0261,
+      "step": 770048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.738368592530086e-05,
+      "loss": 4.036,
+      "step": 770560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.737529997779034e-05,
+      "loss": 4.0296,
+      "step": 771072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.736693040908356e-05,
+      "loss": 4.045,
+      "step": 771584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.735854446157304e-05,
+      "loss": 4.0358,
+      "step": 772096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.735015851406252e-05,
+      "loss": 4.0421,
+      "step": 772608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7341772566552e-05,
+      "loss": 4.0396,
+      "step": 773120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.733338661904148e-05,
+      "loss": 4.0534,
+      "step": 773632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.732501705033469e-05,
+      "loss": 4.0393,
+      "step": 774144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.731663110282417e-05,
+      "loss": 4.0493,
+      "step": 774656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.730824515531365e-05,
+      "loss": 4.0414,
+      "step": 775168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.729985920780313e-05,
+      "loss": 4.0413,
+      "step": 775680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.72914732602926e-05,
+      "loss": 4.0422,
+      "step": 776192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7283120070389546e-05,
+      "loss": 4.0305,
+      "step": 776704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.7274734122879026e-05,
+      "loss": 4.0402,
+      "step": 777216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.726634817536851e-05,
+      "loss": 4.0326,
+      "step": 777728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.725796222785799e-05,
+      "loss": 4.0334,
+      "step": 778240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.724957628034747e-05,
+      "loss": 4.0371,
+      "step": 778752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.724119033283695e-05,
+      "loss": 4.0388,
+      "step": 779264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.723280438532643e-05,
+      "loss": 4.0412,
+      "step": 779776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7224418437815906e-05,
+      "loss": 4.0512,
+      "step": 780288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7216032490305386e-05,
+      "loss": 4.0358,
+      "step": 780800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.720767930040233e-05,
+      "loss": 4.0377,
+      "step": 781312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.719929335289181e-05,
+      "loss": 4.052,
+      "step": 781824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.719090740538129e-05,
+      "loss": 4.0309,
+      "step": 782336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.718252145787077e-05,
+      "loss": 4.0326,
+      "step": 782848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.717413551036025e-05,
+      "loss": 4.0373,
+      "step": 783360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.716574956284973e-05,
+      "loss": 4.0334,
+      "step": 783872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.715736361533921e-05,
+      "loss": 4.0391,
+      "step": 784384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.714897766782869e-05,
+      "loss": 4.0323,
+      "step": 784896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.714059172031817e-05,
+      "loss": 4.0436,
+      "step": 785408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.713222215161138e-05,
+      "loss": 4.0355,
+      "step": 785920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.712383620410086e-05,
+      "loss": 4.0409,
+      "step": 786432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.711545025659034e-05,
+      "loss": 4.0363,
+      "step": 786944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.710706430907982e-05,
+      "loss": 4.0433,
+      "step": 787456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.709869474037303e-05,
+      "loss": 4.0055,
+      "step": 787968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7090325171666244e-05,
+      "loss": 4.0456,
+      "step": 788480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7081939224155724e-05,
+      "loss": 4.0261,
+      "step": 788992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7073553276645204e-05,
+      "loss": 4.0416,
+      "step": 789504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7065167329134684e-05,
+      "loss": 4.026,
+      "step": 790016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7056781381624164e-05,
+      "loss": 4.0204,
+      "step": 790528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7048395434113644e-05,
+      "loss": 4.0207,
+      "step": 791040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7040009486603124e-05,
+      "loss": 4.0293,
+      "step": 791552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7031623539092604e-05,
+      "loss": 4.0374,
+      "step": 792064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7023237591582084e-05,
+      "loss": 4.0222,
+      "step": 792576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7014851644071564e-05,
+      "loss": 4.0433,
+      "step": 793088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.7006465696561044e-05,
+      "loss": 4.0359,
+      "step": 793600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.699809612785425e-05,
+      "loss": 4.0143,
+      "step": 794112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.698971018034373e-05,
+      "loss": 4.0373,
+      "step": 794624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.698132423283321e-05,
+      "loss": 4.0181,
+      "step": 795136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.697293828532269e-05,
+      "loss": 4.0191,
+      "step": 795648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.69645687166159e-05,
+      "loss": 4.0331,
+      "step": 796160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.695618276910538e-05,
+      "loss": 4.037,
+      "step": 796672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.694779682159487e-05,
+      "loss": 4.0176,
+      "step": 797184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.693941087408435e-05,
+      "loss": 4.033,
+      "step": 797696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.693102492657383e-05,
+      "loss": 4.0178,
+      "step": 798208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.692263897906331e-05,
+      "loss": 4.0299,
+      "step": 798720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.691426941035652e-05,
+      "loss": 4.0364,
+      "step": 799232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.690589984164973e-05,
+      "loss": 4.0348,
+      "step": 799744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.689751389413921e-05,
+      "loss": 4.0411,
+      "step": 800256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6889127946628687e-05,
+      "loss": 4.0399,
+      "step": 800768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6880741999118167e-05,
+      "loss": 4.0552,
+      "step": 801280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6872356051607646e-05,
+      "loss": 4.0346,
+      "step": 801792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6863970104097126e-05,
+      "loss": 4.0391,
+      "step": 802304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6855584156586606e-05,
+      "loss": 4.0256,
+      "step": 802816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6847198209076086e-05,
+      "loss": 4.0305,
+      "step": 803328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6838812261565566e-05,
+      "loss": 4.027,
+      "step": 803840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.683044269285878e-05,
+      "loss": 4.035,
+      "step": 804352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.682205674534826e-05,
+      "loss": 4.0396,
+      "step": 804864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6813670797837735e-05,
+      "loss": 4.0445,
+      "step": 805376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6805284850327215e-05,
+      "loss": 4.0203,
+      "step": 805888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6796898902816695e-05,
+      "loss": 4.0262,
+      "step": 806400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.678852933410991e-05,
+      "loss": 4.0273,
+      "step": 806912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6780143386599384e-05,
+      "loss": 4.0348,
+      "step": 807424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6771757439088864e-05,
+      "loss": 4.0341,
+      "step": 807936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.6763371491578344e-05,
+      "loss": 4.0283,
+      "step": 808448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.675500192287156e-05,
+      "loss": 4.0291,
+      "step": 808960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6746632354164776e-05,
+      "loss": 4.0327,
+      "step": 809472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6738246406654256e-05,
+      "loss": 4.0271,
+      "step": 809984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6729860459143736e-05,
+      "loss": 4.0231,
+      "step": 810496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.672147451163321e-05,
+      "loss": 4.0401,
+      "step": 811008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.671308856412269e-05,
+      "loss": 4.0365,
+      "step": 811520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.670470261661217e-05,
+      "loss": 4.0343,
+      "step": 812032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.669631666910165e-05,
+      "loss": 4.0222,
+      "step": 812544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.668793072159113e-05,
+      "loss": 4.0239,
+      "step": 813056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.667954477408061e-05,
+      "loss": 4.0267,
+      "step": 813568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.667117520537382e-05,
+      "loss": 4.0345,
+      "step": 814080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.66627892578633e-05,
+      "loss": 4.0419,
+      "step": 814592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.665440331035278e-05,
+      "loss": 4.0242,
+      "step": 815104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.664601736284226e-05,
+      "loss": 4.0406,
+      "step": 815616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6637631415331745e-05,
+      "loss": 4.0297,
+      "step": 816128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6629261846624954e-05,
+      "loss": 4.0264,
+      "step": 816640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.662089227791816e-05,
+      "loss": 4.0191,
+      "step": 817152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.661250633040764e-05,
+      "loss": 4.0169,
+      "step": 817664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.660412038289712e-05,
+      "loss": 4.0264,
+      "step": 818176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.65957344353866e-05,
+      "loss": 4.0254,
+      "step": 818688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.658734848787608e-05,
+      "loss": 4.0187,
+      "step": 819200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.657896254036556e-05,
+      "loss": 4.0197,
+      "step": 819712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.657057659285504e-05,
+      "loss": 4.0126,
+      "step": 820224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.656219064534452e-05,
+      "loss": 4.0283,
+      "step": 820736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6553804697834e-05,
+      "loss": 4.022,
+      "step": 821248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.654543512912721e-05,
+      "loss": 4.0296,
+      "step": 821760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.65370491816167e-05,
+      "loss": 4.0322,
+      "step": 822272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.652866323410618e-05,
+      "loss": 4.0291,
+      "step": 822784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.652027728659566e-05,
+      "loss": 4.0275,
+      "step": 823296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.651189133908514e-05,
+      "loss": 4.0323,
+      "step": 823808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.650350539157462e-05,
+      "loss": 4.0276,
+      "step": 824320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.64951194440641e-05,
+      "loss": 4.0385,
+      "step": 824832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.648673349655357e-05,
+      "loss": 4.0288,
+      "step": 825344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.647836392784679e-05,
+      "loss": 4.027,
+      "step": 825856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.646997798033627e-05,
+      "loss": 4.0276,
+      "step": 826368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6461608411629476e-05,
+      "loss": 4.026,
+      "step": 826880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6453222464118956e-05,
+      "loss": 4.0211,
+      "step": 827392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6444836516608436e-05,
+      "loss": 4.0203,
+      "step": 827904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6436450569097916e-05,
+      "loss": 4.018,
+      "step": 828416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.642808100039113e-05,
+      "loss": 4.0159,
+      "step": 828928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.641969505288061e-05,
+      "loss": 4.0326,
+      "step": 829440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.641130910537009e-05,
+      "loss": 4.0249,
+      "step": 829952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.640292315785957e-05,
+      "loss": 4.0318,
+      "step": 830464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.639455358915278e-05,
+      "loss": 4.0183,
+      "step": 830976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.638616764164226e-05,
+      "loss": 4.0329,
+      "step": 831488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.637778169413174e-05,
+      "loss": 4.0329,
+      "step": 832000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.636939574662122e-05,
+      "loss": 4.0305,
+      "step": 832512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6361009799110694e-05,
+      "loss": 4.0287,
+      "step": 833024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.635264023040391e-05,
+      "loss": 4.0253,
+      "step": 833536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.634425428289339e-05,
+      "loss": 4.0349,
+      "step": 834048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.633586833538287e-05,
+      "loss": 4.0376,
+      "step": 834560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.632748238787235e-05,
+      "loss": 4.0181,
+      "step": 835072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.631909644036183e-05,
+      "loss": 4.0268,
+      "step": 835584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.631071049285131e-05,
+      "loss": 4.0291,
+      "step": 836096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.630232454534079e-05,
+      "loss": 4.0227,
+      "step": 836608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6293954976634e-05,
+      "loss": 4.0209,
+      "step": 837120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.628556902912348e-05,
+      "loss": 4.0222,
+      "step": 837632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.627718308161296e-05,
+      "loss": 4.0284,
+      "step": 838144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.626879713410244e-05,
+      "loss": 4.025,
+      "step": 838656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.626042756539565e-05,
+      "loss": 4.0305,
+      "step": 839168
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.05925989151001,
+      "eval_runtime": 302.8737,
+      "eval_samples_per_second": 1259.901,
+      "eval_steps_per_second": 39.373,
+      "step": 839520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.625204161788513e-05,
+      "loss": 4.0157,
+      "step": 839680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.6243672049178337e-05,
+      "loss": 4.017,
+      "step": 840192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.623528610166782e-05,
+      "loss": 4.0185,
+      "step": 840704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.62269001541573e-05,
+      "loss": 4.0247,
+      "step": 841216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.621851420664678e-05,
+      "loss": 4.0319,
+      "step": 841728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.621012825913626e-05,
+      "loss": 4.03,
+      "step": 842240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.620174231162574e-05,
+      "loss": 4.0173,
+      "step": 842752
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.619335636411522e-05,
+      "loss": 4.0215,
+      "step": 843264
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.61849704166047e-05,
+      "loss": 4.0138,
+      "step": 843776
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.617660084789791e-05,
+      "loss": 4.0223,
+      "step": 844288
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.616821490038739e-05,
+      "loss": 4.028,
+      "step": 844800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.615986171048434e-05,
+      "loss": 4.0227,
+      "step": 845312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.615147576297381e-05,
+      "loss": 4.0258,
+      "step": 845824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.614308981546329e-05,
+      "loss": 4.0105,
+      "step": 846336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.613470386795278e-05,
+      "loss": 4.0186,
+      "step": 846848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.612631792044226e-05,
+      "loss": 4.0047,
+      "step": 847360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.611793197293174e-05,
+      "loss": 4.0267,
+      "step": 847872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.610954602542122e-05,
+      "loss": 4.0193,
+      "step": 848384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.61011600779107e-05,
+      "loss": 4.021,
+      "step": 848896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.609277413040018e-05,
+      "loss": 4.0193,
+      "step": 849408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.608438818288966e-05,
+      "loss": 4.039,
+      "step": 849920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.607600223537914e-05,
+      "loss": 4.0163,
+      "step": 850432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.6067616287868617e-05,
+      "loss": 4.0304,
+      "step": 850944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.6059230340358097e-05,
+      "loss": 4.0197,
+      "step": 851456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.6050844392847576e-05,
+      "loss": 4.0236,
+      "step": 851968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.6042474824140786e-05,
+      "loss": 4.0261,
+      "step": 852480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.6034088876630266e-05,
+      "loss": 4.0098,
+      "step": 852992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.6025702929119745e-05,
+      "loss": 4.0197,
+      "step": 853504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.6017316981609225e-05,
+      "loss": 4.0132,
+      "step": 854016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.600894741290244e-05,
+      "loss": 4.0141,
+      "step": 854528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.600056146539192e-05,
+      "loss": 4.0186,
+      "step": 855040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.59921755178814e-05,
+      "loss": 4.0192,
+      "step": 855552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.598378957037088e-05,
+      "loss": 4.017,
+      "step": 856064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5975403622860354e-05,
+      "loss": 4.0362,
+      "step": 856576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5967017675349834e-05,
+      "loss": 4.0194,
+      "step": 857088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.595864810664305e-05,
+      "loss": 4.0185,
+      "step": 857600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.595026215913253e-05,
+      "loss": 4.0294,
+      "step": 858112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5941876211622e-05,
+      "loss": 4.0104,
+      "step": 858624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.593349026411148e-05,
+      "loss": 4.0157,
+      "step": 859136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.592510431660096e-05,
+      "loss": 4.0224,
+      "step": 859648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.591673474789418e-05,
+      "loss": 4.0142,
+      "step": 860160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.590834880038366e-05,
+      "loss": 4.0192,
+      "step": 860672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.589996285287314e-05,
+      "loss": 4.0204,
+      "step": 861184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.589157690536262e-05,
+      "loss": 4.0168,
+      "step": 861696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.588320733665583e-05,
+      "loss": 4.0222,
+      "step": 862208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.587482138914531e-05,
+      "loss": 4.0218,
+      "step": 862720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.586643544163479e-05,
+      "loss": 4.0173,
+      "step": 863232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.585804949412427e-05,
+      "loss": 4.0236,
+      "step": 863744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.584966354661375e-05,
+      "loss": 3.9887,
+      "step": 864256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.584127759910323e-05,
+      "loss": 4.022,
+      "step": 864768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.583289165159271e-05,
+      "loss": 4.0115,
+      "step": 865280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.582450570408219e-05,
+      "loss": 4.0218,
+      "step": 865792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.581611975657167e-05,
+      "loss": 4.0092,
+      "step": 866304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5807750187864884e-05,
+      "loss": 3.9977,
+      "step": 866816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5799364240354364e-05,
+      "loss": 4.0057,
+      "step": 867328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.579099467164757e-05,
+      "loss": 4.0091,
+      "step": 867840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.578260872413705e-05,
+      "loss": 4.0173,
+      "step": 868352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.577422277662653e-05,
+      "loss": 4.0,
+      "step": 868864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.576583682911601e-05,
+      "loss": 4.0259,
+      "step": 869376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.575745088160549e-05,
+      "loss": 4.02,
+      "step": 869888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.574906493409497e-05,
+      "loss": 3.9936,
+      "step": 870400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.574069536538818e-05,
+      "loss": 4.0163,
+      "step": 870912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.573230941787766e-05,
+      "loss": 4.0012,
+      "step": 871424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.572392347036714e-05,
+      "loss": 3.9994,
+      "step": 871936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.571553752285662e-05,
+      "loss": 4.0107,
+      "step": 872448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.57071515753461e-05,
+      "loss": 4.0226,
+      "step": 872960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.569878200663932e-05,
+      "loss": 3.999,
+      "step": 873472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.56903960591288e-05,
+      "loss": 4.0162,
+      "step": 873984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.568201011161828e-05,
+      "loss": 3.9953,
+      "step": 874496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.567362416410776e-05,
+      "loss": 4.0128,
+      "step": 875008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.566523821659724e-05,
+      "loss": 4.0173,
+      "step": 875520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.565685226908672e-05,
+      "loss": 4.0209,
+      "step": 876032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.564846632157619e-05,
+      "loss": 4.0197,
+      "step": 876544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.564008037406567e-05,
+      "loss": 4.0211,
+      "step": 877056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5631710805358886e-05,
+      "loss": 4.0372,
+      "step": 877568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5623341236652095e-05,
+      "loss": 4.0157,
+      "step": 878080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5614955289141575e-05,
+      "loss": 4.0214,
+      "step": 878592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5606569341631055e-05,
+      "loss": 4.0059,
+      "step": 879104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5598183394120535e-05,
+      "loss": 4.0116,
+      "step": 879616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5589797446610015e-05,
+      "loss": 4.0094,
+      "step": 880128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5581411499099495e-05,
+      "loss": 4.012,
+      "step": 880640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5573025551588975e-05,
+      "loss": 4.0219,
+      "step": 881152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5564639604078455e-05,
+      "loss": 4.0239,
+      "step": 881664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5556270035371664e-05,
+      "loss": 4.008,
+      "step": 882176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5547884087861144e-05,
+      "loss": 4.0077,
+      "step": 882688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5539498140350624e-05,
+      "loss": 4.0025,
+      "step": 883200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5531112192840104e-05,
+      "loss": 4.0237,
+      "step": 883712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.5522726245329584e-05,
+      "loss": 4.0129,
+      "step": 884224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.551435667662279e-05,
+      "loss": 4.0081,
+      "step": 884736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.550597072911227e-05,
+      "loss": 4.0159,
+      "step": 885248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.549758478160175e-05,
+      "loss": 4.0123,
+      "step": 885760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.548919883409124e-05,
+      "loss": 4.0096,
+      "step": 886272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.548081288658072e-05,
+      "loss": 4.0072,
+      "step": 886784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.54724269390702e-05,
+      "loss": 4.0202,
+      "step": 887296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.546404099155968e-05,
+      "loss": 4.0151,
+      "step": 887808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.545565504404916e-05,
+      "loss": 4.0182,
+      "step": 888320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.544728547534237e-05,
+      "loss": 4.0031,
+      "step": 888832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.543889952783185e-05,
+      "loss": 4.0066,
+      "step": 889344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.543051358032133e-05,
+      "loss": 4.0103,
+      "step": 889856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.542214401161454e-05,
+      "loss": 4.0178,
+      "step": 890368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.541375806410402e-05,
+      "loss": 4.0191,
+      "step": 890880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.54053721165935e-05,
+      "loss": 4.0121,
+      "step": 891392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.539698616908298e-05,
+      "loss": 4.0182,
+      "step": 891904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.538860022157246e-05,
+      "loss": 4.0133,
+      "step": 892416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.538021427406194e-05,
+      "loss": 4.0106,
+      "step": 892928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.537184470535515e-05,
+      "loss": 4.0002,
+      "step": 893440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.536345875784463e-05,
+      "loss": 4.0028,
+      "step": 893952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.535507281033411e-05,
+      "loss": 4.0035,
+      "step": 894464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.534668686282359e-05,
+      "loss": 4.0082,
+      "step": 894976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.53383172941168e-05,
+      "loss": 3.9993,
+      "step": 895488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.532993134660628e-05,
+      "loss": 4.0004,
+      "step": 896000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.532154539909576e-05,
+      "loss": 3.999,
+      "step": 896512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.531315945158524e-05,
+      "loss": 4.0069,
+      "step": 897024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.530477350407472e-05,
+      "loss": 4.0079,
+      "step": 897536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.529640393536793e-05,
+      "loss": 4.0082,
+      "step": 898048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.528801798785741e-05,
+      "loss": 4.0133,
+      "step": 898560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.527963204034689e-05,
+      "loss": 4.015,
+      "step": 899072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.527124609283637e-05,
+      "loss": 4.0092,
+      "step": 899584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.526287652412959e-05,
+      "loss": 4.016,
+      "step": 900096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5254490576619067e-05,
+      "loss": 4.0096,
+      "step": 900608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5246104629108547e-05,
+      "loss": 4.0201,
+      "step": 901120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5237718681598027e-05,
+      "loss": 4.0118,
+      "step": 901632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.52293327340875e-05,
+      "loss": 4.0082,
+      "step": 902144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.522094678657698e-05,
+      "loss": 4.007,
+      "step": 902656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.521256083906646e-05,
+      "loss": 4.0114,
+      "step": 903168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.520417489155594e-05,
+      "loss": 4.0029,
+      "step": 903680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.519578894404542e-05,
+      "loss": 4.005,
+      "step": 904192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.51874029965349e-05,
+      "loss": 3.9948,
+      "step": 904704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.517903342782811e-05,
+      "loss": 4.0028,
+      "step": 905216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5170647480317595e-05,
+      "loss": 4.0123,
+      "step": 905728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5162261532807075e-05,
+      "loss": 4.0084,
+      "step": 906240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5153875585296555e-05,
+      "loss": 4.0132,
+      "step": 906752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5145489637786035e-05,
+      "loss": 3.995,
+      "step": 907264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5137120069079244e-05,
+      "loss": 4.0204,
+      "step": 907776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5128734121568724e-05,
+      "loss": 4.0161,
+      "step": 908288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5120348174058204e-05,
+      "loss": 4.0172,
+      "step": 908800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5111962226547684e-05,
+      "loss": 4.003,
+      "step": 909312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.5103576279037164e-05,
+      "loss": 4.0099,
+      "step": 909824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.509520671033037e-05,
+      "loss": 4.0179,
+      "step": 910336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.508682076281985e-05,
+      "loss": 4.019,
+      "step": 910848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.507843481530933e-05,
+      "loss": 4.0029,
+      "step": 911360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.507004886779881e-05,
+      "loss": 4.0044,
+      "step": 911872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.506166292028829e-05,
+      "loss": 4.0111,
+      "step": 912384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.505327697277778e-05,
+      "loss": 4.0114,
+      "step": 912896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.504489102526726e-05,
+      "loss": 4.0023,
+      "step": 913408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.503650507775674e-05,
+      "loss": 4.0072,
+      "step": 913920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.502813550904995e-05,
+      "loss": 4.0069,
+      "step": 914432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.501974956153943e-05,
+      "loss": 4.0076,
+      "step": 914944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.501137999283264e-05,
+      "loss": 4.0145,
+      "step": 915456
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.047857761383057,
+      "eval_runtime": 297.363,
+      "eval_samples_per_second": 1283.25,
+      "eval_steps_per_second": 40.102,
+      "step": 915840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.500299404532212e-05,
+      "loss": 4.0043,
+      "step": 915968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.49946080978116e-05,
+      "loss": 3.9938,
+      "step": 916480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.498622215030108e-05,
+      "loss": 4.0047,
+      "step": 916992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.497783620279056e-05,
+      "loss": 4.0094,
+      "step": 917504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.496945025528003e-05,
+      "loss": 4.0058,
+      "step": 918016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.496106430776952e-05,
+      "loss": 4.0164,
+      "step": 918528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4952678360259e-05,
+      "loss": 4.0022,
+      "step": 919040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.494429241274848e-05,
+      "loss": 4.0024,
+      "step": 919552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.493590646523796e-05,
+      "loss": 3.9937,
+      "step": 920064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.492752051772744e-05,
+      "loss": 4.0056,
+      "step": 920576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.491913457021692e-05,
+      "loss": 4.0097,
+      "step": 921088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.49107486227064e-05,
+      "loss": 4.0041,
+      "step": 921600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4902379053999606e-05,
+      "loss": 4.0141,
+      "step": 922112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4893993106489086e-05,
+      "loss": 3.9882,
+      "step": 922624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4885607158978566e-05,
+      "loss": 3.9975,
+      "step": 923136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4877221211468046e-05,
+      "loss": 3.9924,
+      "step": 923648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4868835263957526e-05,
+      "loss": 4.0086,
+      "step": 924160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4860449316447006e-05,
+      "loss": 4.0015,
+      "step": 924672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4852063368936486e-05,
+      "loss": 4.0019,
+      "step": 925184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4843677421425966e-05,
+      "loss": 3.999,
+      "step": 925696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4835291473915446e-05,
+      "loss": 4.0217,
+      "step": 926208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4826905526404926e-05,
+      "loss": 3.9997,
+      "step": 926720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4818519578894406e-05,
+      "loss": 4.0123,
+      "step": 927232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4810133631383886e-05,
+      "loss": 4.0035,
+      "step": 927744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.48017640626771e-05,
+      "loss": 4.0086,
+      "step": 928256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4793378115166575e-05,
+      "loss": 4.0081,
+      "step": 928768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.478500854645979e-05,
+      "loss": 3.9917,
+      "step": 929280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.477662259894927e-05,
+      "loss": 4.0011,
+      "step": 929792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4768236651438744e-05,
+      "loss": 3.9952,
+      "step": 930304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4759850703928224e-05,
+      "loss": 3.9964,
+      "step": 930816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4751464756417704e-05,
+      "loss": 4.0047,
+      "step": 931328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.474309518771092e-05,
+      "loss": 4.0023,
+      "step": 931840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.47347092402004e-05,
+      "loss": 4.0026,
+      "step": 932352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.472632329268988e-05,
+      "loss": 4.017,
+      "step": 932864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.471793734517936e-05,
+      "loss": 4.0016,
+      "step": 933376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.470955139766884e-05,
+      "loss": 4.0039,
+      "step": 933888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.470118182896205e-05,
+      "loss": 4.0133,
+      "step": 934400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.469279588145153e-05,
+      "loss": 3.9906,
+      "step": 934912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.468440993394101e-05,
+      "loss": 4.0001,
+      "step": 935424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.467602398643049e-05,
+      "loss": 4.0017,
+      "step": 935936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.466763803891997e-05,
+      "loss": 3.9976,
+      "step": 936448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.465926847021318e-05,
+      "loss": 4.0058,
+      "step": 936960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.465088252270266e-05,
+      "loss": 4.0064,
+      "step": 937472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.464249657519214e-05,
+      "loss": 3.9983,
+      "step": 937984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.463411062768162e-05,
+      "loss": 4.0075,
+      "step": 938496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.462574105897483e-05,
+      "loss": 4.0036,
+      "step": 939008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.461735511146431e-05,
+      "loss": 4.0004,
+      "step": 939520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.460896916395379e-05,
+      "loss": 4.0061,
+      "step": 940032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.460058321644327e-05,
+      "loss": 3.9754,
+      "step": 940544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.459219726893275e-05,
+      "loss": 4.0006,
+      "step": 941056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.458382770022596e-05,
+      "loss": 4.0008,
+      "step": 941568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.457544175271544e-05,
+      "loss": 4.0057,
+      "step": 942080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.456705580520492e-05,
+      "loss": 3.9909,
+      "step": 942592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.45586698576944e-05,
+      "loss": 3.9825,
+      "step": 943104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.455028391018388e-05,
+      "loss": 3.9861,
+      "step": 943616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.454191434147709e-05,
+      "loss": 3.989,
+      "step": 944128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.453352839396657e-05,
+      "loss": 4.0067,
+      "step": 944640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.452514244645606e-05,
+      "loss": 3.9791,
+      "step": 945152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.451675649894554e-05,
+      "loss": 4.0098,
+      "step": 945664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.450837055143502e-05,
+      "loss": 4.0012,
+      "step": 946176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.44999846039245e-05,
+      "loss": 3.9798,
+      "step": 946688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.449161503521771e-05,
+      "loss": 3.9998,
+      "step": 947200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.448322908770719e-05,
+      "loss": 3.9832,
+      "step": 947712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.447484314019667e-05,
+      "loss": 3.9852,
+      "step": 948224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.446645719268615e-05,
+      "loss": 3.9895,
+      "step": 948736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4458071245175627e-05,
+      "loss": 4.0047,
+      "step": 949248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4449685297665107e-05,
+      "loss": 3.988,
+      "step": 949760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4441315728958316e-05,
+      "loss": 4.0001,
+      "step": 950272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4432929781447796e-05,
+      "loss": 3.9757,
+      "step": 950784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4424543833937276e-05,
+      "loss": 3.995,
+      "step": 951296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4416157886426756e-05,
+      "loss": 4.0014,
+      "step": 951808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4407771938916235e-05,
+      "loss": 4.0042,
+      "step": 952320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4399385991405715e-05,
+      "loss": 4.0033,
+      "step": 952832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4391000043895195e-05,
+      "loss": 4.0056,
+      "step": 953344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4382614096384675e-05,
+      "loss": 4.0166,
+      "step": 953856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4374244527677884e-05,
+      "loss": 4.0017,
+      "step": 954368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4365858580167364e-05,
+      "loss": 4.0046,
+      "step": 954880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4357472632656844e-05,
+      "loss": 3.9888,
+      "step": 955392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4349086685146324e-05,
+      "loss": 3.9941,
+      "step": 955904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4340717116439533e-05,
+      "loss": 3.9924,
+      "step": 956416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.433233116892901e-05,
+      "loss": 3.995,
+      "step": 956928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.432394522141849e-05,
+      "loss": 4.0054,
+      "step": 957440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.431557565271171e-05,
+      "loss": 4.0091,
+      "step": 957952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.430718970520119e-05,
+      "loss": 3.997,
+      "step": 958464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.429880375769067e-05,
+      "loss": 3.9872,
+      "step": 958976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.429041781018015e-05,
+      "loss": 3.9886,
+      "step": 959488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.428203186266963e-05,
+      "loss": 4.0054,
+      "step": 960000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.427366229396284e-05,
+      "loss": 3.9958,
+      "step": 960512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.426527634645232e-05,
+      "loss": 3.9925,
+      "step": 961024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.42568903989418e-05,
+      "loss": 3.999,
+      "step": 961536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.424850445143128e-05,
+      "loss": 3.9966,
+      "step": 962048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.424013488272449e-05,
+      "loss": 3.9927,
+      "step": 962560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.423174893521397e-05,
+      "loss": 3.9897,
+      "step": 963072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.422336298770345e-05,
+      "loss": 4.0027,
+      "step": 963584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4214977040192934e-05,
+      "loss": 4.0037,
+      "step": 964096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4206591092682414e-05,
+      "loss": 4.0017,
+      "step": 964608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4198205145171894e-05,
+      "loss": 3.987,
+      "step": 965120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4189819197661374e-05,
+      "loss": 3.9926,
+      "step": 965632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4181433250150854e-05,
+      "loss": 3.9951,
+      "step": 966144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.417306368144406e-05,
+      "loss": 3.9998,
+      "step": 966656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.416467773393354e-05,
+      "loss": 4.0039,
+      "step": 967168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.415630816522675e-05,
+      "loss": 3.9937,
+      "step": 967680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.414792221771623e-05,
+      "loss": 4.0017,
+      "step": 968192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.413953627020571e-05,
+      "loss": 3.9974,
+      "step": 968704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.413115032269519e-05,
+      "loss": 3.9952,
+      "step": 969216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.41227807539884e-05,
+      "loss": 3.9854,
+      "step": 969728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.411439480647789e-05,
+      "loss": 3.9873,
+      "step": 970240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.410600885896737e-05,
+      "loss": 3.9899,
+      "step": 970752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.409762291145685e-05,
+      "loss": 3.9924,
+      "step": 971264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.408923696394633e-05,
+      "loss": 3.9826,
+      "step": 971776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.408085101643581e-05,
+      "loss": 3.9874,
+      "step": 972288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.407246506892529e-05,
+      "loss": 3.9781,
+      "step": 972800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.406407912141477e-05,
+      "loss": 3.9961,
+      "step": 973312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.405569317390424e-05,
+      "loss": 3.9924,
+      "step": 973824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.404730722639372e-05,
+      "loss": 3.9938,
+      "step": 974336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.40389212788832e-05,
+      "loss": 3.9943,
+      "step": 974848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.403053533137268e-05,
+      "loss": 3.9981,
+      "step": 975360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.402216576266589e-05,
+      "loss": 3.9939,
+      "step": 975872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.401377981515537e-05,
+      "loss": 3.9945,
+      "step": 976384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.400539386764485e-05,
+      "loss": 3.9978,
+      "step": 976896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3997007920134336e-05,
+      "loss": 4.0013,
+      "step": 977408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3988638351427545e-05,
+      "loss": 3.9964,
+      "step": 977920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3980252403917025e-05,
+      "loss": 3.9948,
+      "step": 978432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3971866456406505e-05,
+      "loss": 3.9928,
+      "step": 978944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3963480508895985e-05,
+      "loss": 3.9925,
+      "step": 979456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.395512731899293e-05,
+      "loss": 3.9868,
+      "step": 979968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.394674137148241e-05,
+      "loss": 3.9894,
+      "step": 980480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.393835542397189e-05,
+      "loss": 3.9774,
+      "step": 980992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.392996947646136e-05,
+      "loss": 3.9901,
+      "step": 981504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.392158352895084e-05,
+      "loss": 3.9929,
+      "step": 982016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.391319758144032e-05,
+      "loss": 3.9957,
+      "step": 982528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.390482801273354e-05,
+      "loss": 3.9935,
+      "step": 983040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.389644206522302e-05,
+      "loss": 3.9861,
+      "step": 983552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.38880561177125e-05,
+      "loss": 3.9996,
+      "step": 984064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.387967017020198e-05,
+      "loss": 3.9997,
+      "step": 984576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.387130060149519e-05,
+      "loss": 3.9957,
+      "step": 985088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.386291465398467e-05,
+      "loss": 3.9916,
+      "step": 985600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.385452870647415e-05,
+      "loss": 3.995,
+      "step": 986112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.384614275896363e-05,
+      "loss": 4.0032,
+      "step": 986624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.383775681145311e-05,
+      "loss": 3.9999,
+      "step": 987136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.382937086394259e-05,
+      "loss": 3.9894,
+      "step": 987648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.382098491643207e-05,
+      "loss": 3.9884,
+      "step": 988160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.381259896892155e-05,
+      "loss": 3.9952,
+      "step": 988672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.380421302141103e-05,
+      "loss": 3.9929,
+      "step": 989184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.379582707390051e-05,
+      "loss": 3.9884,
+      "step": 989696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.378744112638999e-05,
+      "loss": 3.9907,
+      "step": 990208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3779055178879474e-05,
+      "loss": 3.9908,
+      "step": 990720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.377068561017268e-05,
+      "loss": 3.9941,
+      "step": 991232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.376229966266216e-05,
+      "loss": 3.9977,
+      "step": 991744
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.038654804229736,
+      "eval_runtime": 296.7732,
+      "eval_samples_per_second": 1285.8,
+      "eval_steps_per_second": 40.182,
+      "step": 992160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.375391371515164e-05,
+      "loss": 3.9896,
+      "step": 992256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.374552776764112e-05,
+      "loss": 3.977,
+      "step": 992768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.373715819893433e-05,
+      "loss": 3.9846,
+      "step": 993280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.372877225142381e-05,
+      "loss": 3.998,
+      "step": 993792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.372038630391329e-05,
+      "loss": 3.9927,
+      "step": 994304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.371200035640277e-05,
+      "loss": 3.9996,
+      "step": 994816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.370363078769598e-05,
+      "loss": 3.9832,
+      "step": 995328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.369524484018546e-05,
+      "loss": 3.9887,
+      "step": 995840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.368687527147868e-05,
+      "loss": 3.9773,
+      "step": 996352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.367848932396816e-05,
+      "loss": 3.994,
+      "step": 996864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.367010337645764e-05,
+      "loss": 3.9899,
+      "step": 997376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.366171742894712e-05,
+      "loss": 3.992,
+      "step": 997888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3653347860240326e-05,
+      "loss": 3.9986,
+      "step": 998400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3644961912729806e-05,
+      "loss": 3.9723,
+      "step": 998912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3636575965219286e-05,
+      "loss": 3.9871,
+      "step": 999424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3628190017708766e-05,
+      "loss": 3.9746,
+      "step": 999936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3619804070198246e-05,
+      "loss": 3.992,
+      "step": 1000448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3611418122687726e-05,
+      "loss": 3.9833,
+      "step": 1000960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3603048553980935e-05,
+      "loss": 3.9907,
+      "step": 1001472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3594662606470415e-05,
+      "loss": 3.9823,
+      "step": 1001984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3586276658959895e-05,
+      "loss": 4.0052,
+      "step": 1002496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3577890711449375e-05,
+      "loss": 3.9872,
+      "step": 1003008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3569504763938855e-05,
+      "loss": 3.9936,
+      "step": 1003520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3561118816428334e-05,
+      "loss": 3.9906,
+      "step": 1004032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3552732868917814e-05,
+      "loss": 3.9933,
+      "step": 1004544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3544363300211024e-05,
+      "loss": 3.9896,
+      "step": 1005056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3535977352700503e-05,
+      "loss": 3.979,
+      "step": 1005568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3527591405189983e-05,
+      "loss": 3.9894,
+      "step": 1006080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3519205457679463e-05,
+      "loss": 3.9787,
+      "step": 1006592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.351083588897267e-05,
+      "loss": 3.9808,
+      "step": 1007104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.350244994146215e-05,
+      "loss": 3.986,
+      "step": 1007616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.349406399395163e-05,
+      "loss": 3.9832,
+      "step": 1008128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.348567804644111e-05,
+      "loss": 3.9881,
+      "step": 1008640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.34772920989306e-05,
+      "loss": 4.0004,
+      "step": 1009152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.346890615142008e-05,
+      "loss": 3.9878,
+      "step": 1009664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.346053658271329e-05,
+      "loss": 3.9928,
+      "step": 1010176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.345215063520277e-05,
+      "loss": 3.9941,
+      "step": 1010688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.344376468769225e-05,
+      "loss": 3.9777,
+      "step": 1011200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.343537874018173e-05,
+      "loss": 3.9906,
+      "step": 1011712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.342699279267121e-05,
+      "loss": 3.9832,
+      "step": 1012224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.341862322396442e-05,
+      "loss": 3.9808,
+      "step": 1012736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.34102372764539e-05,
+      "loss": 3.9911,
+      "step": 1013248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3401867707747106e-05,
+      "loss": 3.9928,
+      "step": 1013760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3393481760236586e-05,
+      "loss": 3.9787,
+      "step": 1014272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3385095812726066e-05,
+      "loss": 3.9923,
+      "step": 1014784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.337670986521555e-05,
+      "loss": 3.9908,
+      "step": 1015296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.336832391770503e-05,
+      "loss": 3.9842,
+      "step": 1015808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.335993797019451e-05,
+      "loss": 3.9925,
+      "step": 1016320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.335155202268399e-05,
+      "loss": 3.9594,
+      "step": 1016832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.334316607517347e-05,
+      "loss": 3.9857,
+      "step": 1017344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.333478012766295e-05,
+      "loss": 3.9844,
+      "step": 1017856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.332639418015243e-05,
+      "loss": 3.9898,
+      "step": 1018368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.331800823264191e-05,
+      "loss": 3.9779,
+      "step": 1018880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3309622285131386e-05,
+      "loss": 3.9688,
+      "step": 1019392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.33012527164246e-05,
+      "loss": 3.9694,
+      "step": 1019904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.329289952652154e-05,
+      "loss": 3.9724,
+      "step": 1020416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.328451357901102e-05,
+      "loss": 3.9929,
+      "step": 1020928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3276127631500506e-05,
+      "loss": 3.9681,
+      "step": 1021440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3267741683989986e-05,
+      "loss": 3.9918,
+      "step": 1021952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3259355736479466e-05,
+      "loss": 3.9831,
+      "step": 1022464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3250969788968946e-05,
+      "loss": 3.9719,
+      "step": 1022976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3242583841458426e-05,
+      "loss": 3.9822,
+      "step": 1023488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3234197893947906e-05,
+      "loss": 3.9696,
+      "step": 1024000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3225811946437386e-05,
+      "loss": 3.9706,
+      "step": 1024512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.321742599892686e-05,
+      "loss": 3.9743,
+      "step": 1025024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.320904005141634e-05,
+      "loss": 3.9901,
+      "step": 1025536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.320065410390582e-05,
+      "loss": 3.9745,
+      "step": 1026048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.319228453519903e-05,
+      "loss": 3.9772,
+      "step": 1026560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.318389858768851e-05,
+      "loss": 3.9673,
+      "step": 1027072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.317551264017799e-05,
+      "loss": 3.9775,
+      "step": 1027584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3167143071471204e-05,
+      "loss": 3.989,
+      "step": 1028096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3158757123960684e-05,
+      "loss": 3.9889,
+      "step": 1028608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3150371176450164e-05,
+      "loss": 3.9878,
+      "step": 1029120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3141985228939644e-05,
+      "loss": 3.9888,
+      "step": 1029632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3133599281429124e-05,
+      "loss": 4.0015,
+      "step": 1030144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.312522971272233e-05,
+      "loss": 3.9877,
+      "step": 1030656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.311684376521181e-05,
+      "loss": 3.9918,
+      "step": 1031168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.310845781770129e-05,
+      "loss": 3.9784,
+      "step": 1031680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.310007187019077e-05,
+      "loss": 3.9825,
+      "step": 1032192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.309170230148398e-05,
+      "loss": 3.9775,
+      "step": 1032704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.308331635397346e-05,
+      "loss": 3.9782,
+      "step": 1033216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.307493040646294e-05,
+      "loss": 3.9903,
+      "step": 1033728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.306654445895243e-05,
+      "loss": 3.993,
+      "step": 1034240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.305817489024564e-05,
+      "loss": 3.9889,
+      "step": 1034752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.304978894273512e-05,
+      "loss": 3.968,
+      "step": 1035264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.30414029952246e-05,
+      "loss": 3.9737,
+      "step": 1035776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.303301704771408e-05,
+      "loss": 3.9886,
+      "step": 1036288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.302463110020356e-05,
+      "loss": 3.9836,
+      "step": 1036800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.301624515269304e-05,
+      "loss": 3.9791,
+      "step": 1037312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.300787558398625e-05,
+      "loss": 3.9814,
+      "step": 1037824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2999489636475727e-05,
+      "loss": 3.9836,
+      "step": 1038336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2991103688965207e-05,
+      "loss": 3.9795,
+      "step": 1038848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2982717741454686e-05,
+      "loss": 3.9788,
+      "step": 1039360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2974331793944166e-05,
+      "loss": 3.9905,
+      "step": 1039872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2965945846433646e-05,
+      "loss": 3.984,
+      "step": 1040384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.295757627772686e-05,
+      "loss": 3.9874,
+      "step": 1040896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.294919033021634e-05,
+      "loss": 3.9702,
+      "step": 1041408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.294080438270582e-05,
+      "loss": 3.9751,
+      "step": 1041920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.29324184351953e-05,
+      "loss": 3.9833,
+      "step": 1042432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.292404886648851e-05,
+      "loss": 3.9852,
+      "step": 1042944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.291566291897799e-05,
+      "loss": 3.9896,
+      "step": 1043456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.290727697146747e-05,
+      "loss": 3.9815,
+      "step": 1043968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.289889102395695e-05,
+      "loss": 3.9844,
+      "step": 1044480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.289052145525016e-05,
+      "loss": 3.9798,
+      "step": 1044992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.288213550773964e-05,
+      "loss": 3.982,
+      "step": 1045504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.287374956022912e-05,
+      "loss": 3.9718,
+      "step": 1046016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.28653636127186e-05,
+      "loss": 3.9706,
+      "step": 1046528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2856994044011816e-05,
+      "loss": 3.9793,
+      "step": 1047040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2848608096501296e-05,
+      "loss": 3.9786,
+      "step": 1047552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2840222148990776e-05,
+      "loss": 3.9694,
+      "step": 1048064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2831836201480256e-05,
+      "loss": 3.9702,
+      "step": 1048576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2823466632773465e-05,
+      "loss": 3.9664,
+      "step": 1049088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2815080685262945e-05,
+      "loss": 3.9755,
+      "step": 1049600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2806694737752425e-05,
+      "loss": 3.9794,
+      "step": 1050112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2798308790241905e-05,
+      "loss": 3.9762,
+      "step": 1050624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2789922842731385e-05,
+      "loss": 3.9791,
+      "step": 1051136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2781536895220865e-05,
+      "loss": 3.9873,
+      "step": 1051648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.277315094771034e-05,
+      "loss": 3.9788,
+      "step": 1052160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2764781379003554e-05,
+      "loss": 3.98,
+      "step": 1052672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2756395431493034e-05,
+      "loss": 3.987,
+      "step": 1053184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2748009483982514e-05,
+      "loss": 3.9856,
+      "step": 1053696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.273963991527573e-05,
+      "loss": 3.9796,
+      "step": 1054208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.273125396776521e-05,
+      "loss": 3.9844,
+      "step": 1054720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.272286802025469e-05,
+      "loss": 3.9799,
+      "step": 1055232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.271448207274416e-05,
+      "loss": 3.9749,
+      "step": 1055744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.270609612523364e-05,
+      "loss": 3.9683,
+      "step": 1056256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.269771017772312e-05,
+      "loss": 3.981,
+      "step": 1056768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.26893242302126e-05,
+      "loss": 3.9713,
+      "step": 1057280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.268093828270208e-05,
+      "loss": 3.9768,
+      "step": 1057792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.267256871399529e-05,
+      "loss": 3.9701,
+      "step": 1058304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.266418276648477e-05,
+      "loss": 3.9834,
+      "step": 1058816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.265579681897425e-05,
+      "loss": 3.9783,
+      "step": 1059328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.264741087146374e-05,
+      "loss": 3.9702,
+      "step": 1059840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.263902492395322e-05,
+      "loss": 3.9853,
+      "step": 1060352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.26306389764427e-05,
+      "loss": 3.9853,
+      "step": 1060864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.262225302893218e-05,
+      "loss": 3.9855,
+      "step": 1061376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.261388346022539e-05,
+      "loss": 3.9803,
+      "step": 1061888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.260549751271487e-05,
+      "loss": 3.9757,
+      "step": 1062400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.259711156520435e-05,
+      "loss": 3.9917,
+      "step": 1062912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.258872561769383e-05,
+      "loss": 3.9781,
+      "step": 1063424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.258033967018331e-05,
+      "loss": 3.982,
+      "step": 1063936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.257195372267279e-05,
+      "loss": 3.972,
+      "step": 1064448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2563584153965996e-05,
+      "loss": 3.9798,
+      "step": 1064960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2555198206455476e-05,
+      "loss": 3.9761,
+      "step": 1065472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2546812258944956e-05,
+      "loss": 3.9752,
+      "step": 1065984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2538426311434436e-05,
+      "loss": 3.9797,
+      "step": 1066496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.253004036392392e-05,
+      "loss": 3.9777,
+      "step": 1067008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.25216544164134e-05,
+      "loss": 3.981,
+      "step": 1067520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.251328484770661e-05,
+      "loss": 3.9796,
+      "step": 1068032
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.03078031539917,
+      "eval_runtime": 299.8455,
+      "eval_samples_per_second": 1272.625,
+      "eval_steps_per_second": 39.77,
+      "step": 1068480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.250489890019609e-05,
+      "loss": 3.9645,
+      "step": 1068544
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.249651295268557e-05,
+      "loss": 3.9671,
+      "step": 1069056
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.248812700517505e-05,
+      "loss": 3.9692,
+      "step": 1069568
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2479741057664525e-05,
+      "loss": 3.9839,
+      "step": 1070080
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2471355110154005e-05,
+      "loss": 3.9788,
+      "step": 1070592
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2462969162643485e-05,
+      "loss": 3.9879,
+      "step": 1071104
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2454583215132965e-05,
+      "loss": 3.9697,
+      "step": 1071616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2446197267622445e-05,
+      "loss": 3.9745,
+      "step": 1072128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2437811320111924e-05,
+      "loss": 3.966,
+      "step": 1072640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2429425372601404e-05,
+      "loss": 3.9774,
+      "step": 1073152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.2421039425090884e-05,
+      "loss": 3.9769,
+      "step": 1073664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.241265347758037e-05,
+      "loss": 3.9763,
+      "step": 1074176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.240428390887358e-05,
+      "loss": 3.9821,
+      "step": 1074688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.239589796136306e-05,
+      "loss": 3.9626,
+      "step": 1075200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.238751201385254e-05,
+      "loss": 3.9683,
+      "step": 1075712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.237912606634202e-05,
+      "loss": 3.9644,
+      "step": 1076224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.23707401188315e-05,
+      "loss": 3.9786,
+      "step": 1076736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.236235417132098e-05,
+      "loss": 3.9693,
+      "step": 1077248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.235396822381046e-05,
+      "loss": 3.9783,
+      "step": 1077760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.234558227629994e-05,
+      "loss": 3.9671,
+      "step": 1078272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.233719632878941e-05,
+      "loss": 3.9903,
+      "step": 1078784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.232882676008263e-05,
+      "loss": 3.9745,
+      "step": 1079296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.232044081257211e-05,
+      "loss": 3.9769,
+      "step": 1079808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.231205486506159e-05,
+      "loss": 3.9761,
+      "step": 1080320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.230366891755107e-05,
+      "loss": 3.9815,
+      "step": 1080832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.229528297004055e-05,
+      "loss": 3.9764,
+      "step": 1081344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.228689702253003e-05,
+      "loss": 3.965,
+      "step": 1081856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.227851107501951e-05,
+      "loss": 3.9778,
+      "step": 1082368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.227012512750899e-05,
+      "loss": 3.9611,
+      "step": 1082880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.22617555588022e-05,
+      "loss": 3.9671,
+      "step": 1083392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2253385990095414e-05,
+      "loss": 3.9704,
+      "step": 1083904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.224500004258489e-05,
+      "loss": 3.9697,
+      "step": 1084416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.223661409507437e-05,
+      "loss": 3.9764,
+      "step": 1084928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.222822814756385e-05,
+      "loss": 3.9843,
+      "step": 1085440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.221984220005333e-05,
+      "loss": 3.9764,
+      "step": 1085952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.221145625254281e-05,
+      "loss": 3.9797,
+      "step": 1086464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.220308668383602e-05,
+      "loss": 3.9802,
+      "step": 1086976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.21947007363255e-05,
+      "loss": 3.9621,
+      "step": 1087488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.218631478881498e-05,
+      "loss": 3.9791,
+      "step": 1088000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.217792884130446e-05,
+      "loss": 3.9681,
+      "step": 1088512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.216954289379394e-05,
+      "loss": 3.969,
+      "step": 1089024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.216115694628342e-05,
+      "loss": 3.9751,
+      "step": 1089536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.21527709987729e-05,
+      "loss": 3.9809,
+      "step": 1090048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.214438505126238e-05,
+      "loss": 3.9615,
+      "step": 1090560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.213601548255559e-05,
+      "loss": 3.9839,
+      "step": 1091072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.212762953504507e-05,
+      "loss": 3.9723,
+      "step": 1091584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.211925996633828e-05,
+      "loss": 3.973,
+      "step": 1092096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.211087401882776e-05,
+      "loss": 3.9776,
+      "step": 1092608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.210248807131725e-05,
+      "loss": 3.9444,
+      "step": 1093120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.209410212380673e-05,
+      "loss": 3.9692,
+      "step": 1093632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.208571617629621e-05,
+      "loss": 3.9776,
+      "step": 1094144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.207733022878569e-05,
+      "loss": 3.9742,
+      "step": 1094656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.206894428127517e-05,
+      "loss": 3.9613,
+      "step": 1095168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2060574712568376e-05,
+      "loss": 3.9559,
+      "step": 1095680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2052188765057856e-05,
+      "loss": 3.9543,
+      "step": 1096192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2043802817547336e-05,
+      "loss": 3.9578,
+      "step": 1096704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2035416870036816e-05,
+      "loss": 3.9835,
+      "step": 1097216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2027030922526296e-05,
+      "loss": 3.9526,
+      "step": 1097728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2018644975015776e-05,
+      "loss": 3.9802,
+      "step": 1098240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.201025902750525e-05,
+      "loss": 3.9698,
+      "step": 1098752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.2001889458798465e-05,
+      "loss": 3.9561,
+      "step": 1099264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1993503511287945e-05,
+      "loss": 3.9685,
+      "step": 1099776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.198511756377743e-05,
+      "loss": 3.9579,
+      "step": 1100288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1976731616266905e-05,
+      "loss": 3.9598,
+      "step": 1100800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.196836204756012e-05,
+      "loss": 3.9551,
+      "step": 1101312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.19599761000496e-05,
+      "loss": 3.9756,
+      "step": 1101824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1951590152539074e-05,
+      "loss": 3.9655,
+      "step": 1102336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1943204205028554e-05,
+      "loss": 3.9613,
+      "step": 1102848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1934818257518034e-05,
+      "loss": 3.9566,
+      "step": 1103360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.1926432310007514e-05,
+      "loss": 3.9621,
+      "step": 1103872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.191806274130072e-05,
+      "loss": 3.9745,
+      "step": 1104384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.19096767937902e-05,
+      "loss": 3.9759,
+      "step": 1104896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.190129084627968e-05,
+      "loss": 3.972,
+      "step": 1105408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.189290489876917e-05,
+      "loss": 3.9753,
+      "step": 1105920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.188451895125865e-05,
+      "loss": 3.9871,
+      "step": 1106432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.187614938255186e-05,
+      "loss": 3.9735,
+      "step": 1106944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.186776343504134e-05,
+      "loss": 3.9775,
+      "step": 1107456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.185937748753082e-05,
+      "loss": 3.9654,
+      "step": 1107968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.18509915400203e-05,
+      "loss": 3.9702,
+      "step": 1108480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.184260559250978e-05,
+      "loss": 3.9643,
+      "step": 1108992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.183423602380299e-05,
+      "loss": 3.9613,
+      "step": 1109504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.182585007629247e-05,
+      "loss": 3.9785,
+      "step": 1110016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.181746412878195e-05,
+      "loss": 3.9801,
+      "step": 1110528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.180907818127143e-05,
+      "loss": 3.9756,
+      "step": 1111040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.180069223376091e-05,
+      "loss": 3.9511,
+      "step": 1111552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.179230628625039e-05,
+      "loss": 3.9613,
+      "step": 1112064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.178392033873987e-05,
+      "loss": 3.9743,
+      "step": 1112576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.177553439122935e-05,
+      "loss": 3.9713,
+      "step": 1113088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.176716482252256e-05,
+      "loss": 3.9646,
+      "step": 1113600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.175877887501204e-05,
+      "loss": 3.9715,
+      "step": 1114112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.175039292750152e-05,
+      "loss": 3.9672,
+      "step": 1114624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.174203973759846e-05,
+      "loss": 3.9649,
+      "step": 1115136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.173365379008794e-05,
+      "loss": 3.9643,
+      "step": 1115648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.172526784257742e-05,
+      "loss": 3.9763,
+      "step": 1116160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.17168818950669e-05,
+      "loss": 3.9667,
+      "step": 1116672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.170849594755638e-05,
+      "loss": 3.9784,
+      "step": 1117184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.170011000004586e-05,
+      "loss": 3.9585,
+      "step": 1117696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.169172405253534e-05,
+      "loss": 3.9623,
+      "step": 1118208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.168333810502482e-05,
+      "loss": 3.9715,
+      "step": 1118720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.16749521575143e-05,
+      "loss": 3.9725,
+      "step": 1119232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.166656621000379e-05,
+      "loss": 3.9748,
+      "step": 1119744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.165818026249327e-05,
+      "loss": 3.9692,
+      "step": 1120256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.164979431498274e-05,
+      "loss": 3.9708,
+      "step": 1120768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1641424746275956e-05,
+      "loss": 3.9673,
+      "step": 1121280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1633038798765436e-05,
+      "loss": 3.9722,
+      "step": 1121792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1624669230058645e-05,
+      "loss": 3.9538,
+      "step": 1122304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1616283282548125e-05,
+      "loss": 3.9617,
+      "step": 1122816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1607897335037605e-05,
+      "loss": 3.9609,
+      "step": 1123328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1599511387527085e-05,
+      "loss": 3.9639,
+      "step": 1123840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.159112544001656e-05,
+      "loss": 3.9603,
+      "step": 1124352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.158273949250604e-05,
+      "loss": 3.956,
+      "step": 1124864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1574353544995525e-05,
+      "loss": 3.9491,
+      "step": 1125376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1565983976288734e-05,
+      "loss": 3.9673,
+      "step": 1125888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1557598028778214e-05,
+      "loss": 3.9638,
+      "step": 1126400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1549212081267694e-05,
+      "loss": 3.9647,
+      "step": 1126912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1540826133757174e-05,
+      "loss": 3.9668,
+      "step": 1127424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1532440186246654e-05,
+      "loss": 3.9731,
+      "step": 1127936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1524054238736134e-05,
+      "loss": 3.9668,
+      "step": 1128448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.151568467002934e-05,
+      "loss": 3.9632,
+      "step": 1128960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.150729872251882e-05,
+      "loss": 3.9745,
+      "step": 1129472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.14989127750083e-05,
+      "loss": 3.9698,
+      "step": 1129984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.149052682749778e-05,
+      "loss": 3.9602,
+      "step": 1130496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.148214087998726e-05,
+      "loss": 3.9705,
+      "step": 1131008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.147375493247674e-05,
+      "loss": 3.9662,
+      "step": 1131520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.146536898496622e-05,
+      "loss": 3.9657,
+      "step": 1132032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.145698303745571e-05,
+      "loss": 3.9531,
+      "step": 1132544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.144861346874892e-05,
+      "loss": 3.9673,
+      "step": 1133056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.14402275212384e-05,
+      "loss": 3.9578,
+      "step": 1133568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.143184157372788e-05,
+      "loss": 3.9635,
+      "step": 1134080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.142347200502109e-05,
+      "loss": 3.9511,
+      "step": 1134592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.141508605751057e-05,
+      "loss": 3.9771,
+      "step": 1135104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.140671648880378e-05,
+      "loss": 3.9671,
+      "step": 1135616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.139833054129326e-05,
+      "loss": 3.9548,
+      "step": 1136128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1389944593782737e-05,
+      "loss": 3.9728,
+      "step": 1136640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1381558646272217e-05,
+      "loss": 3.9703,
+      "step": 1137152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1373172698761697e-05,
+      "loss": 3.9734,
+      "step": 1137664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1364786751251176e-05,
+      "loss": 3.9656,
+      "step": 1138176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1356400803740656e-05,
+      "loss": 3.9672,
+      "step": 1138688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.134801485623014e-05,
+      "loss": 3.9777,
+      "step": 1139200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.133962890871962e-05,
+      "loss": 3.9644,
+      "step": 1139712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1331242961209096e-05,
+      "loss": 3.9691,
+      "step": 1140224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1322857013698576e-05,
+      "loss": 3.9593,
+      "step": 1140736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.131448744499179e-05,
+      "loss": 3.9673,
+      "step": 1141248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.130610149748127e-05,
+      "loss": 3.9598,
+      "step": 1141760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1297715549970745e-05,
+      "loss": 3.9586,
+      "step": 1142272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1289329602460225e-05,
+      "loss": 3.9663,
+      "step": 1142784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.1280943654949705e-05,
+      "loss": 3.967,
+      "step": 1143296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.127257408624292e-05,
+      "loss": 3.9711,
+      "step": 1143808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.126420451753613e-05,
+      "loss": 3.9665,
+      "step": 1144320
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.022707939147949,
+      "eval_runtime": 293.8732,
+      "eval_samples_per_second": 1298.489,
+      "eval_steps_per_second": 40.579,
+      "step": 1144800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.125581857002561e-05,
+      "loss": 3.9593,
+      "step": 1144832
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.12474326225151e-05,
+      "loss": 3.9595,
+      "step": 1145344
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.123904667500457e-05,
+      "loss": 3.957,
+      "step": 1145856
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.123066072749405e-05,
+      "loss": 3.9691,
+      "step": 1146368
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.122227477998353e-05,
+      "loss": 3.9642,
+      "step": 1146880
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.121388883247301e-05,
+      "loss": 3.9763,
+      "step": 1147392
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.120550288496249e-05,
+      "loss": 3.9553,
+      "step": 1147904
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.119711693745197e-05,
+      "loss": 3.9609,
+      "step": 1148416
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.118873098994145e-05,
+      "loss": 3.9526,
+      "step": 1148928
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.118034504243093e-05,
+      "loss": 3.9667,
+      "step": 1149440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.117197547372414e-05,
+      "loss": 3.9596,
+      "step": 1149952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.116358952621362e-05,
+      "loss": 3.9669,
+      "step": 1150464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1155219957506835e-05,
+      "loss": 3.9697,
+      "step": 1150976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1146834009996315e-05,
+      "loss": 3.9516,
+      "step": 1151488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1138448062485795e-05,
+      "loss": 3.953,
+      "step": 1152000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1130062114975274e-05,
+      "loss": 3.9522,
+      "step": 1152512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1121676167464754e-05,
+      "loss": 3.9632,
+      "step": 1153024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1113290219954234e-05,
+      "loss": 3.9595,
+      "step": 1153536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1104904272443714e-05,
+      "loss": 3.9611,
+      "step": 1154048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1096518324933194e-05,
+      "loss": 3.9548,
+      "step": 1154560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1088132377422674e-05,
+      "loss": 3.9755,
+      "step": 1155072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.1079746429912154e-05,
+      "loss": 3.9582,
+      "step": 1155584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.107137686120536e-05,
+      "loss": 3.9644,
+      "step": 1156096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.106299091369484e-05,
+      "loss": 3.9667,
+      "step": 1156608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.105460496618432e-05,
+      "loss": 3.9707,
+      "step": 1157120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.10462190186738e-05,
+      "loss": 3.9627,
+      "step": 1157632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.103783307116328e-05,
+      "loss": 3.9536,
+      "step": 1158144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.102944712365276e-05,
+      "loss": 3.9624,
+      "step": 1158656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.102106117614224e-05,
+      "loss": 3.9464,
+      "step": 1159168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.101267522863172e-05,
+      "loss": 3.9581,
+      "step": 1159680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.100430565992493e-05,
+      "loss": 3.9534,
+      "step": 1160192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.099593609121815e-05,
+      "loss": 3.9633,
+      "step": 1160704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.098755014370763e-05,
+      "loss": 3.9606,
+      "step": 1161216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.097916419619711e-05,
+      "loss": 3.9675,
+      "step": 1161728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.097077824868658e-05,
+      "loss": 3.9672,
+      "step": 1162240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.09624086799798e-05,
+      "loss": 3.9664,
+      "step": 1162752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.095402273246928e-05,
+      "loss": 3.9658,
+      "step": 1163264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.094563678495876e-05,
+      "loss": 3.9511,
+      "step": 1163776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.093725083744824e-05,
+      "loss": 3.9667,
+      "step": 1164288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.092886488993772e-05,
+      "loss": 3.9527,
+      "step": 1164800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.09204789424272e-05,
+      "loss": 3.9549,
+      "step": 1165312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0912109373720406e-05,
+      "loss": 3.9639,
+      "step": 1165824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0903723426209886e-05,
+      "loss": 3.9661,
+      "step": 1166336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0895337478699366e-05,
+      "loss": 3.9536,
+      "step": 1166848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0886951531188846e-05,
+      "loss": 3.9676,
+      "step": 1167360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0878565583678326e-05,
+      "loss": 3.9603,
+      "step": 1167872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0870179636167806e-05,
+      "loss": 3.9601,
+      "step": 1168384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0861793688657286e-05,
+      "loss": 3.9674,
+      "step": 1168896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0853407741146765e-05,
+      "loss": 3.9363,
+      "step": 1169408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0845038172439975e-05,
+      "loss": 3.9519,
+      "step": 1169920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0836652224929455e-05,
+      "loss": 3.9661,
+      "step": 1170432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.082826627741894e-05,
+      "loss": 3.9613,
+      "step": 1170944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.081988032990842e-05,
+      "loss": 3.9504,
+      "step": 1171456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.081151076120163e-05,
+      "loss": 3.9401,
+      "step": 1171968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.080312481369111e-05,
+      "loss": 3.9475,
+      "step": 1172480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.079473886618059e-05,
+      "loss": 3.9387,
+      "step": 1172992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.078635291867007e-05,
+      "loss": 3.975,
+      "step": 1173504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.077798334996328e-05,
+      "loss": 3.9389,
+      "step": 1174016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.076959740245276e-05,
+      "loss": 3.9664,
+      "step": 1174528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.076121145494224e-05,
+      "loss": 3.9577,
+      "step": 1175040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.075282550743172e-05,
+      "loss": 3.9474,
+      "step": 1175552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.07444395599212e-05,
+      "loss": 3.9534,
+      "step": 1176064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.073605361241068e-05,
+      "loss": 3.9482,
+      "step": 1176576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.072768404370389e-05,
+      "loss": 3.9448,
+      "step": 1177088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0719298096193375e-05,
+      "loss": 3.944,
+      "step": 1177600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0710912148682855e-05,
+      "loss": 3.967,
+      "step": 1178112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0702526201172335e-05,
+      "loss": 3.952,
+      "step": 1178624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0694156632465544e-05,
+      "loss": 3.949,
+      "step": 1179136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0685770684955024e-05,
+      "loss": 3.9416,
+      "step": 1179648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0677384737444504e-05,
+      "loss": 3.9513,
+      "step": 1180160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0668998789933984e-05,
+      "loss": 3.9621,
+      "step": 1180672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0660612842423464e-05,
+      "loss": 3.9656,
+      "step": 1181184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0652226894912944e-05,
+      "loss": 3.9636,
+      "step": 1181696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.064384094740242e-05,
+      "loss": 3.9606,
+      "step": 1182208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.063547137869563e-05,
+      "loss": 3.9746,
+      "step": 1182720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.062708543118511e-05,
+      "loss": 3.963,
+      "step": 1183232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.061869948367459e-05,
+      "loss": 3.963,
+      "step": 1183744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.061031353616407e-05,
+      "loss": 3.9525,
+      "step": 1184256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.060192758865355e-05,
+      "loss": 3.9579,
+      "step": 1184768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.059354164114303e-05,
+      "loss": 3.9486,
+      "step": 1185280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.058515569363251e-05,
+      "loss": 3.9543,
+      "step": 1185792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.057676974612199e-05,
+      "loss": 3.9651,
+      "step": 1186304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.05684001774152e-05,
+      "loss": 3.9697,
+      "step": 1186816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.056003060870842e-05,
+      "loss": 3.9603,
+      "step": 1187328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.055164466119789e-05,
+      "loss": 3.9448,
+      "step": 1187840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.054325871368737e-05,
+      "loss": 3.9492,
+      "step": 1188352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.053487276617685e-05,
+      "loss": 3.9575,
+      "step": 1188864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.052648681866633e-05,
+      "loss": 3.9651,
+      "step": 1189376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.051810087115581e-05,
+      "loss": 3.9494,
+      "step": 1189888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.0509714923645294e-05,
+      "loss": 3.9595,
+      "step": 1190400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0501328976134774e-05,
+      "loss": 3.9587,
+      "step": 1190912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0492959407427986e-05,
+      "loss": 3.95,
+      "step": 1191424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0484573459917466e-05,
+      "loss": 3.9533,
+      "step": 1191936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0476187512406946e-05,
+      "loss": 3.9634,
+      "step": 1192448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0467801564896426e-05,
+      "loss": 3.9536,
+      "step": 1192960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0459415617385906e-05,
+      "loss": 3.9667,
+      "step": 1193472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0451029669875386e-05,
+      "loss": 3.9501,
+      "step": 1193984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0442643722364866e-05,
+      "loss": 3.9491,
+      "step": 1194496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0434257774854346e-05,
+      "loss": 3.955,
+      "step": 1195008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.042588820614756e-05,
+      "loss": 3.964,
+      "step": 1195520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.041750225863704e-05,
+      "loss": 3.9598,
+      "step": 1196032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0409132689930247e-05,
+      "loss": 3.9587,
+      "step": 1196544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0400746742419727e-05,
+      "loss": 3.9607,
+      "step": 1197056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0392360794909207e-05,
+      "loss": 3.9576,
+      "step": 1197568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0383974847398687e-05,
+      "loss": 3.9567,
+      "step": 1198080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0375588899888167e-05,
+      "loss": 3.9408,
+      "step": 1198592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.036720295237765e-05,
+      "loss": 3.9479,
+      "step": 1199104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.035881700486713e-05,
+      "loss": 3.9549,
+      "step": 1199616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0350431057356604e-05,
+      "loss": 3.9503,
+      "step": 1200128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.034206148864982e-05,
+      "loss": 3.9484,
+      "step": 1200640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.03336755411393e-05,
+      "loss": 3.946,
+      "step": 1201152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0325289593628776e-05,
+      "loss": 3.9406,
+      "step": 1201664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0316903646118256e-05,
+      "loss": 3.9541,
+      "step": 1202176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0308534077411472e-05,
+      "loss": 3.949,
+      "step": 1202688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.030016450870468e-05,
+      "loss": 3.9571,
+      "step": 1203200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.029177856119416e-05,
+      "loss": 3.9522,
+      "step": 1203712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.028339261368364e-05,
+      "loss": 3.9623,
+      "step": 1204224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.027500666617312e-05,
+      "loss": 3.956,
+      "step": 1204736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0266620718662604e-05,
+      "loss": 3.9524,
+      "step": 1205248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0258234771152077e-05,
+      "loss": 3.9598,
+      "step": 1205760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0249848823641557e-05,
+      "loss": 3.9591,
+      "step": 1206272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0241462876131037e-05,
+      "loss": 3.9549,
+      "step": 1206784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0233076928620517e-05,
+      "loss": 3.9588,
+      "step": 1207296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0224690981109997e-05,
+      "loss": 3.954,
+      "step": 1207808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.021630503359948e-05,
+      "loss": 3.9549,
+      "step": 1208320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.020791908608896e-05,
+      "loss": 3.9458,
+      "step": 1208832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.019953313857844e-05,
+      "loss": 3.9504,
+      "step": 1209344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.019116356987165e-05,
+      "loss": 3.9477,
+      "step": 1209856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.018277762236113e-05,
+      "loss": 3.9491,
+      "step": 1210368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.017439167485061e-05,
+      "loss": 3.9388,
+      "step": 1210880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.016600572734009e-05,
+      "loss": 3.9698,
+      "step": 1211392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0157619779829573e-05,
+      "loss": 3.9543,
+      "step": 1211904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0149233832319053e-05,
+      "loss": 3.9502,
+      "step": 1212416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0140847884808533e-05,
+      "loss": 3.9548,
+      "step": 1212928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0132461937298013e-05,
+      "loss": 3.9559,
+      "step": 1213440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0124092368591222e-05,
+      "loss": 3.9656,
+      "step": 1213952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0115706421080702e-05,
+      "loss": 3.9545,
+      "step": 1214464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.010732047357018e-05,
+      "loss": 3.9533,
+      "step": 1214976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0098934526059665e-05,
+      "loss": 3.9637,
+      "step": 1215488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0090564957352874e-05,
+      "loss": 3.9499,
+      "step": 1216000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0082195388646083e-05,
+      "loss": 3.9593,
+      "step": 1216512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0073809441135563e-05,
+      "loss": 3.9501,
+      "step": 1217024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0065423493625043e-05,
+      "loss": 3.9609,
+      "step": 1217536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0057037546114526e-05,
+      "loss": 3.9544,
+      "step": 1218048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0048651598604006e-05,
+      "loss": 3.9427,
+      "step": 1218560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0040265651093486e-05,
+      "loss": 3.9573,
+      "step": 1219072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0031879703582966e-05,
+      "loss": 3.9558,
+      "step": 1219584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.002349375607244e-05,
+      "loss": 3.9553,
+      "step": 1220096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.0015124187365655e-05,
+      "loss": 3.9536,
+      "step": 1220608
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.0006738239855135e-05,
+      "loss": 3.953,
+      "step": 1221120
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.016727447509766,
+      "eval_runtime": 289.1819,
+      "eval_samples_per_second": 1319.554,
+      "eval_steps_per_second": 41.237,
+      "step": 1221120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9998352292344612e-05,
+      "loss": 3.9455,
+      "step": 1221632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9989966344834092e-05,
+      "loss": 3.9419,
+      "step": 1222144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9981580397323572e-05,
+      "loss": 3.959,
+      "step": 1222656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9973194449813052e-05,
+      "loss": 3.9546,
+      "step": 1223168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9964808502302532e-05,
+      "loss": 3.965,
+      "step": 1223680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9956422554792012e-05,
+      "loss": 3.9469,
+      "step": 1224192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.994803660728149e-05,
+      "loss": 3.9482,
+      "step": 1224704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9939650659770975e-05,
+      "loss": 3.9389,
+      "step": 1225216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9931264712260455e-05,
+      "loss": 3.9579,
+      "step": 1225728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9922878764749935e-05,
+      "loss": 3.9488,
+      "step": 1226240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9914492817239415e-05,
+      "loss": 3.9575,
+      "step": 1226752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9906106869728895e-05,
+      "loss": 3.9557,
+      "step": 1227264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9897737301022104e-05,
+      "loss": 3.9459,
+      "step": 1227776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9889351353511584e-05,
+      "loss": 3.9418,
+      "step": 1228288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9880965406001067e-05,
+      "loss": 3.9431,
+      "step": 1228800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9872579458490547e-05,
+      "loss": 3.9444,
+      "step": 1229312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9864193510980027e-05,
+      "loss": 3.9517,
+      "step": 1229824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.98558075634695e-05,
+      "loss": 3.9482,
+      "step": 1230336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.984742161595898e-05,
+      "loss": 3.9427,
+      "step": 1230848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.983903566844846e-05,
+      "loss": 3.9681,
+      "step": 1231360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9830649720937944e-05,
+      "loss": 3.947,
+      "step": 1231872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9822280152231153e-05,
+      "loss": 3.9571,
+      "step": 1232384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9813894204720633e-05,
+      "loss": 3.9541,
+      "step": 1232896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9805508257210113e-05,
+      "loss": 3.9588,
+      "step": 1233408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9797122309699593e-05,
+      "loss": 3.9465,
+      "step": 1233920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9788752740992805e-05,
+      "loss": 3.9486,
+      "step": 1234432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9780366793482285e-05,
+      "loss": 3.9459,
+      "step": 1234944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9771980845971765e-05,
+      "loss": 3.9439,
+      "step": 1235456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.9763611277264974e-05,
+      "loss": 3.9437,
+      "step": 1235968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9755225329754454e-05,
+      "loss": 3.9407,
+      "step": 1236480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9746839382243934e-05,
+      "loss": 3.9525,
+      "step": 1236992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9738453434733414e-05,
+      "loss": 3.9515,
+      "step": 1237504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9730067487222897e-05,
+      "loss": 3.9544,
+      "step": 1238016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9721681539712377e-05,
+      "loss": 3.9574,
+      "step": 1238528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9713311971005586e-05,
+      "loss": 3.9543,
+      "step": 1239040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9704926023495066e-05,
+      "loss": 3.9533,
+      "step": 1239552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9696540075984546e-05,
+      "loss": 3.9456,
+      "step": 1240064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9688154128474026e-05,
+      "loss": 3.9501,
+      "step": 1240576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9679768180963506e-05,
+      "loss": 3.9456,
+      "step": 1241088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.967138223345299e-05,
+      "loss": 3.9499,
+      "step": 1241600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.966299628594247e-05,
+      "loss": 3.9496,
+      "step": 1242112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.965461033843195e-05,
+      "loss": 3.9548,
+      "step": 1242624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.964624076972516e-05,
+      "loss": 3.9415,
+      "step": 1243136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.963785482221464e-05,
+      "loss": 3.9546,
+      "step": 1243648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.962946887470412e-05,
+      "loss": 3.9529,
+      "step": 1244160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.96210829271936e-05,
+      "loss": 3.9531,
+      "step": 1244672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.961271335848681e-05,
+      "loss": 3.9528,
+      "step": 1245184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.960432741097629e-05,
+      "loss": 3.9262,
+      "step": 1245696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.959594146346577e-05,
+      "loss": 3.9408,
+      "step": 1246208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.958755551595525e-05,
+      "loss": 3.9538,
+      "step": 1246720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.957916956844473e-05,
+      "loss": 3.951,
+      "step": 1247232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.957078362093421e-05,
+      "loss": 3.9407,
+      "step": 1247744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9562397673423687e-05,
+      "loss": 3.9288,
+      "step": 1248256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9554011725913167e-05,
+      "loss": 3.937,
+      "step": 1248768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9545642157206383e-05,
+      "loss": 3.9308,
+      "step": 1249280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9537256209695863e-05,
+      "loss": 3.9622,
+      "step": 1249792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9528886640989072e-05,
+      "loss": 3.9334,
+      "step": 1250304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9520500693478552e-05,
+      "loss": 3.9518,
+      "step": 1250816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9512114745968032e-05,
+      "loss": 3.9475,
+      "step": 1251328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9503728798457515e-05,
+      "loss": 3.9332,
+      "step": 1251840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9495359229750724e-05,
+      "loss": 3.9447,
+      "step": 1252352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9486973282240204e-05,
+      "loss": 3.9396,
+      "step": 1252864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9478587334729684e-05,
+      "loss": 3.9303,
+      "step": 1253376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.947020138721916e-05,
+      "loss": 3.9325,
+      "step": 1253888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.946181543970864e-05,
+      "loss": 3.9587,
+      "step": 1254400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.945342949219812e-05,
+      "loss": 3.9458,
+      "step": 1254912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.94450435446876e-05,
+      "loss": 3.9372,
+      "step": 1255424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.943665759717708e-05,
+      "loss": 3.9322,
+      "step": 1255936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.942827164966656e-05,
+      "loss": 3.9388,
+      "step": 1256448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.941988570215604e-05,
+      "loss": 3.9505,
+      "step": 1256960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.941149975464552e-05,
+      "loss": 3.956,
+      "step": 1257472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9403113807135e-05,
+      "loss": 3.9494,
+      "step": 1257984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9394744238428213e-05,
+      "loss": 3.9554,
+      "step": 1258496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9386374669721422e-05,
+      "loss": 3.9614,
+      "step": 1259008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9377988722210902e-05,
+      "loss": 3.9533,
+      "step": 1259520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9369602774700382e-05,
+      "loss": 3.9496,
+      "step": 1260032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9361216827189862e-05,
+      "loss": 3.9432,
+      "step": 1260544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9352830879679345e-05,
+      "loss": 3.9489,
+      "step": 1261056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9344444932168825e-05,
+      "loss": 3.9436,
+      "step": 1261568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9336075363462034e-05,
+      "loss": 3.9423,
+      "step": 1262080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9327689415951514e-05,
+      "loss": 3.9508,
+      "step": 1262592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9319303468440994e-05,
+      "loss": 3.9568,
+      "step": 1263104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9310917520930474e-05,
+      "loss": 3.9521,
+      "step": 1263616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9302547952223687e-05,
+      "loss": 3.9331,
+      "step": 1264128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9294162004713167e-05,
+      "loss": 3.9354,
+      "step": 1264640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9285776057202647e-05,
+      "loss": 3.9486,
+      "step": 1265152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9277406488495856e-05,
+      "loss": 3.9533,
+      "step": 1265664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9269020540985336e-05,
+      "loss": 3.9395,
+      "step": 1266176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9260634593474816e-05,
+      "loss": 3.9475,
+      "step": 1266688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.92522486459643e-05,
+      "loss": 3.9521,
+      "step": 1267200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.924386269845378e-05,
+      "loss": 3.9402,
+      "step": 1267712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.923547675094326e-05,
+      "loss": 3.9414,
+      "step": 1268224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.922709080343274e-05,
+      "loss": 3.9476,
+      "step": 1268736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.921870485592222e-05,
+      "loss": 3.9447,
+      "step": 1269248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.92103189084117e-05,
+      "loss": 3.9543,
+      "step": 1269760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9201932960901175e-05,
+      "loss": 3.9452,
+      "step": 1270272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9193547013390655e-05,
+      "loss": 3.9339,
+      "step": 1270784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9185161065880135e-05,
+      "loss": 3.9498,
+      "step": 1271296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9176791497173344e-05,
+      "loss": 3.9513,
+      "step": 1271808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.916842192846656e-05,
+      "loss": 3.9483,
+      "step": 1272320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.916003598095604e-05,
+      "loss": 3.9517,
+      "step": 1272832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.915165003344552e-05,
+      "loss": 3.9462,
+      "step": 1273344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9143264085934997e-05,
+      "loss": 3.9475,
+      "step": 1273856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9134878138424477e-05,
+      "loss": 3.9515,
+      "step": 1274368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9126492190913957e-05,
+      "loss": 3.9301,
+      "step": 1274880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9118122622207172e-05,
+      "loss": 3.9345,
+      "step": 1275392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9109736674696646e-05,
+      "loss": 3.946,
+      "step": 1275904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.910135072718613e-05,
+      "loss": 3.9437,
+      "step": 1276416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.909296477967561e-05,
+      "loss": 3.9342,
+      "step": 1276928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.908457883216509e-05,
+      "loss": 3.9359,
+      "step": 1277440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.907619288465457e-05,
+      "loss": 3.9335,
+      "step": 1277952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.906780693714405e-05,
+      "loss": 3.9424,
+      "step": 1278464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9059437368437258e-05,
+      "loss": 3.9394,
+      "step": 1278976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.9051051420926738e-05,
+      "loss": 3.9435,
+      "step": 1279488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.904266547341622e-05,
+      "loss": 3.9425,
+      "step": 1280000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.90342795259057e-05,
+      "loss": 3.9501,
+      "step": 1280512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.902589357839518e-05,
+      "loss": 3.9471,
+      "step": 1281024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.901750763088466e-05,
+      "loss": 3.947,
+      "step": 1281536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.900913806217787e-05,
+      "loss": 3.9509,
+      "step": 1282048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.900075211466735e-05,
+      "loss": 3.9503,
+      "step": 1282560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.899236616715683e-05,
+      "loss": 3.9383,
+      "step": 1283072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8983980219646313e-05,
+      "loss": 3.9508,
+      "step": 1283584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8975594272135793e-05,
+      "loss": 3.9403,
+      "step": 1284096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8967224703429002e-05,
+      "loss": 3.9512,
+      "step": 1284608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8958838755918482e-05,
+      "loss": 3.93,
+      "step": 1285120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.895046918721169e-05,
+      "loss": 3.9455,
+      "step": 1285632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8942083239701175e-05,
+      "loss": 3.9331,
+      "step": 1286144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8933697292190655e-05,
+      "loss": 3.9384,
+      "step": 1286656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8925311344680135e-05,
+      "loss": 3.9283,
+      "step": 1287168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8916925397169615e-05,
+      "loss": 3.9589,
+      "step": 1287680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8908539449659095e-05,
+      "loss": 3.9508,
+      "step": 1288192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8900169880952304e-05,
+      "loss": 3.9373,
+      "step": 1288704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8891783933441784e-05,
+      "loss": 3.942,
+      "step": 1289216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8883397985931264e-05,
+      "loss": 3.9462,
+      "step": 1289728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8875012038420747e-05,
+      "loss": 3.9546,
+      "step": 1290240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8866642469713956e-05,
+      "loss": 3.9489,
+      "step": 1290752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8858256522203436e-05,
+      "loss": 3.9434,
+      "step": 1291264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8849870574692916e-05,
+      "loss": 3.9535,
+      "step": 1291776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8841484627182396e-05,
+      "loss": 3.937,
+      "step": 1292288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8833098679671876e-05,
+      "loss": 3.9524,
+      "step": 1292800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.882472911096509e-05,
+      "loss": 3.9427,
+      "step": 1293312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.881634316345457e-05,
+      "loss": 3.9455,
+      "step": 1293824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.880795721594405e-05,
+      "loss": 3.9445,
+      "step": 1294336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8799571268433528e-05,
+      "loss": 3.9359,
+      "step": 1294848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8791185320923008e-05,
+      "loss": 3.9462,
+      "step": 1295360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8782799373412485e-05,
+      "loss": 3.9452,
+      "step": 1295872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8774413425901965e-05,
+      "loss": 3.9464,
+      "step": 1296384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.876604385719518e-05,
+      "loss": 3.9441,
+      "step": 1296896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.8757657909684654e-05,
+      "loss": 3.9407,
+      "step": 1297408
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.011997699737549,
+      "eval_runtime": 299.7104,
+      "eval_samples_per_second": 1273.199,
+      "eval_steps_per_second": 39.788,
+      "step": 1297440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8749271962174134e-05,
+      "loss": 3.9351,
+      "step": 1297920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8740886014663614e-05,
+      "loss": 3.9352,
+      "step": 1298432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.873251644595683e-05,
+      "loss": 3.9504,
+      "step": 1298944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8724130498446306e-05,
+      "loss": 3.939,
+      "step": 1299456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8715744550935786e-05,
+      "loss": 3.9578,
+      "step": 1299968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8707358603425266e-05,
+      "loss": 3.9331,
+      "step": 1300480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8698972655914746e-05,
+      "loss": 3.94,
+      "step": 1300992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8690586708404226e-05,
+      "loss": 3.9282,
+      "step": 1301504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8682200760893706e-05,
+      "loss": 3.9469,
+      "step": 1302016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8673814813383186e-05,
+      "loss": 3.9413,
+      "step": 1302528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.866542886587267e-05,
+      "loss": 3.9457,
+      "step": 1303040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.865704291836215e-05,
+      "loss": 3.9454,
+      "step": 1303552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.864867334965536e-05,
+      "loss": 3.9335,
+      "step": 1304064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8640287402144838e-05,
+      "loss": 3.9329,
+      "step": 1304576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8631901454634318e-05,
+      "loss": 3.9342,
+      "step": 1305088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.862353188592753e-05,
+      "loss": 3.9316,
+      "step": 1305600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.861514593841701e-05,
+      "loss": 3.9422,
+      "step": 1306112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.860675999090649e-05,
+      "loss": 3.9388,
+      "step": 1306624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.859837404339597e-05,
+      "loss": 3.9368,
+      "step": 1307136
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.858998809588545e-05,
+      "loss": 3.9595,
+      "step": 1307648
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.858160214837493e-05,
+      "loss": 3.9339,
+      "step": 1308160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.857321620086441e-05,
+      "loss": 3.9473,
+      "step": 1308672
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.856483025335389e-05,
+      "loss": 3.9465,
+      "step": 1309184
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8556444305843367e-05,
+      "loss": 3.9526,
+      "step": 1309696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8548058358332847e-05,
+      "loss": 3.9311,
+      "step": 1310208
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8539672410822327e-05,
+      "loss": 3.9388,
+      "step": 1310720
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8531286463311807e-05,
+      "loss": 3.9342,
+      "step": 1311232
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.8522916894605016e-05,
+      "loss": 3.9425,
+      "step": 1311744
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.85145309470945e-05,
+      "loss": 3.9274,
+      "step": 1312256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8506161378387715e-05,
+      "loss": 3.9333,
+      "step": 1312768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8497775430877195e-05,
+      "loss": 3.9433,
+      "step": 1313280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8489389483366668e-05,
+      "loss": 3.942,
+      "step": 1313792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8481003535856148e-05,
+      "loss": 3.9422,
+      "step": 1314304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8472617588345628e-05,
+      "loss": 3.9496,
+      "step": 1314816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.846424801963884e-05,
+      "loss": 3.9442,
+      "step": 1315328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.845586207212832e-05,
+      "loss": 3.9421,
+      "step": 1315840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.84474761246178e-05,
+      "loss": 3.9351,
+      "step": 1316352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.843909017710728e-05,
+      "loss": 3.9464,
+      "step": 1316864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.843070422959676e-05,
+      "loss": 3.9317,
+      "step": 1317376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.842233466088997e-05,
+      "loss": 3.9381,
+      "step": 1317888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8413948713379453e-05,
+      "loss": 3.9439,
+      "step": 1318400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8405562765868933e-05,
+      "loss": 3.9442,
+      "step": 1318912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8397176818358413e-05,
+      "loss": 3.9315,
+      "step": 1319424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8388790870847893e-05,
+      "loss": 3.9408,
+      "step": 1319936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8380404923337373e-05,
+      "loss": 3.9467,
+      "step": 1320448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8372035354630582e-05,
+      "loss": 3.9417,
+      "step": 1320960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8363649407120062e-05,
+      "loss": 3.9453,
+      "step": 1321472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8355263459609542e-05,
+      "loss": 3.9168,
+      "step": 1321984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8346877512099025e-05,
+      "loss": 3.9283,
+      "step": 1322496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8338491564588505e-05,
+      "loss": 3.9467,
+      "step": 1323008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8330105617077985e-05,
+      "loss": 3.9408,
+      "step": 1323520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8321736048371194e-05,
+      "loss": 3.9354,
+      "step": 1324032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8313350100860674e-05,
+      "loss": 3.9201,
+      "step": 1324544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8304964153350154e-05,
+      "loss": 3.926,
+      "step": 1325056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8296578205839634e-05,
+      "loss": 3.9214,
+      "step": 1325568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8288208637132846e-05,
+      "loss": 3.9479,
+      "step": 1326080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8279822689622326e-05,
+      "loss": 3.925,
+      "step": 1326592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8271436742111806e-05,
+      "loss": 3.9407,
+      "step": 1327104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8263050794601286e-05,
+      "loss": 3.9411,
+      "step": 1327616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8254664847090766e-05,
+      "loss": 3.9252,
+      "step": 1328128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8246278899580246e-05,
+      "loss": 3.933,
+      "step": 1328640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8237892952069726e-05,
+      "loss": 3.9281,
+      "step": 1329152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.822952338336294e-05,
+      "loss": 3.925,
+      "step": 1329664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.822113743585242e-05,
+      "loss": 3.9261,
+      "step": 1330176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.82127514883419e-05,
+      "loss": 3.9437,
+      "step": 1330688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.820436554083138e-05,
+      "loss": 3.9375,
+      "step": 1331200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8195979593320855e-05,
+      "loss": 3.925,
+      "step": 1331712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.818761002461407e-05,
+      "loss": 3.9248,
+      "step": 1332224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.817922407710355e-05,
+      "loss": 3.9287,
+      "step": 1332736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.817083812959303e-05,
+      "loss": 3.9408,
+      "step": 1333248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8162452182082504e-05,
+      "loss": 3.9472,
+      "step": 1333760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8154066234571984e-05,
+      "loss": 3.9397,
+      "step": 1334272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8145680287061464e-05,
+      "loss": 3.949,
+      "step": 1334784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8137294339550947e-05,
+      "loss": 3.9474,
+      "step": 1335296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8128924770844156e-05,
+      "loss": 3.9468,
+      "step": 1335808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8120538823333636e-05,
+      "loss": 3.9381,
+      "step": 1336320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8112169254626852e-05,
+      "loss": 3.9367,
+      "step": 1336832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8103783307116325e-05,
+      "loss": 3.9404,
+      "step": 1337344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.809539735960581e-05,
+      "loss": 3.9355,
+      "step": 1337856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.808701141209529e-05,
+      "loss": 3.9335,
+      "step": 1338368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.807862546458477e-05,
+      "loss": 3.9412,
+      "step": 1338880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.807023951707425e-05,
+      "loss": 3.9471,
+      "step": 1339392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.806185356956373e-05,
+      "loss": 3.9397,
+      "step": 1339904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8053484000856938e-05,
+      "loss": 3.9273,
+      "step": 1340416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.8045098053346418e-05,
+      "loss": 3.927,
+      "step": 1340928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.80367121058359e-05,
+      "loss": 3.9361,
+      "step": 1341440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.802832615832538e-05,
+      "loss": 3.9452,
+      "step": 1341952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.801994021081486e-05,
+      "loss": 3.934,
+      "step": 1342464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.801157064210807e-05,
+      "loss": 3.9378,
+      "step": 1342976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.800318469459755e-05,
+      "loss": 3.9413,
+      "step": 1343488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.799479874708703e-05,
+      "loss": 3.9339,
+      "step": 1344000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.798641279957651e-05,
+      "loss": 3.9325,
+      "step": 1344512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7978026852065993e-05,
+      "loss": 3.9328,
+      "step": 1345024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7969657283359202e-05,
+      "loss": 3.9392,
+      "step": 1345536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7961271335848682e-05,
+      "loss": 3.9413,
+      "step": 1346048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7952885388338162e-05,
+      "loss": 3.94,
+      "step": 1346560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7944499440827642e-05,
+      "loss": 3.9262,
+      "step": 1347072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7936129872120855e-05,
+      "loss": 3.9403,
+      "step": 1347584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7927743924610335e-05,
+      "loss": 3.9351,
+      "step": 1348096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7919357977099815e-05,
+      "loss": 3.9454,
+      "step": 1348608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7910972029589295e-05,
+      "loss": 3.9383,
+      "step": 1349120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7902586082078774e-05,
+      "loss": 3.9404,
+      "step": 1349632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7894216513371984e-05,
+      "loss": 3.9389,
+      "step": 1350144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7885830565861464e-05,
+      "loss": 3.9418,
+      "step": 1350656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7877444618350947e-05,
+      "loss": 3.9219,
+      "step": 1351168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7869058670840427e-05,
+      "loss": 3.9281,
+      "step": 1351680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7860672723329907e-05,
+      "loss": 3.9349,
+      "step": 1352192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7852286775819387e-05,
+      "loss": 3.9326,
+      "step": 1352704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.784390082830886e-05,
+      "loss": 3.9276,
+      "step": 1353216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.783551488079834e-05,
+      "loss": 3.9274,
+      "step": 1353728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7827128933287823e-05,
+      "loss": 3.924,
+      "step": 1354240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7818742985777303e-05,
+      "loss": 3.9335,
+      "step": 1354752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7810357038266783e-05,
+      "loss": 3.9306,
+      "step": 1355264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7801971090756263e-05,
+      "loss": 3.9311,
+      "step": 1355776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7793617900853208e-05,
+      "loss": 3.936,
+      "step": 1356288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7785231953342688e-05,
+      "loss": 3.9408,
+      "step": 1356800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7776846005832165e-05,
+      "loss": 3.9379,
+      "step": 1357312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.776847643712538e-05,
+      "loss": 3.9338,
+      "step": 1357824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.776009048961486e-05,
+      "loss": 3.9432,
+      "step": 1358336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7751704542104334e-05,
+      "loss": 3.9418,
+      "step": 1358848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7743318594593814e-05,
+      "loss": 3.9306,
+      "step": 1359360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7734932647083294e-05,
+      "loss": 3.9437,
+      "step": 1359872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7726546699572774e-05,
+      "loss": 3.9341,
+      "step": 1360384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7718177130865986e-05,
+      "loss": 3.9429,
+      "step": 1360896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7709791183355466e-05,
+      "loss": 3.9199,
+      "step": 1361408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7701405235844946e-05,
+      "loss": 3.94,
+      "step": 1361920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7693019288334426e-05,
+      "loss": 3.92,
+      "step": 1362432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7684649719627635e-05,
+      "loss": 3.9287,
+      "step": 1362944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.767626377211712e-05,
+      "loss": 3.9209,
+      "step": 1363456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7667877824606598e-05,
+      "loss": 3.9501,
+      "step": 1363968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7659491877096078e-05,
+      "loss": 3.9417,
+      "step": 1364480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7651105929585558e-05,
+      "loss": 3.933,
+      "step": 1364992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7642719982075038e-05,
+      "loss": 3.9303,
+      "step": 1365504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7634334034564518e-05,
+      "loss": 3.9377,
+      "step": 1366016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7625948087053998e-05,
+      "loss": 3.9436,
+      "step": 1366528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7617562139543478e-05,
+      "loss": 3.9392,
+      "step": 1367040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7609176192032958e-05,
+      "loss": 3.9357,
+      "step": 1367552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.760079024452244e-05,
+      "loss": 3.9415,
+      "step": 1368064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.759240429701192e-05,
+      "loss": 3.9326,
+      "step": 1368576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.75840183495014e-05,
+      "loss": 3.9453,
+      "step": 1369088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7575632401990874e-05,
+      "loss": 3.9312,
+      "step": 1369600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7567246454480354e-05,
+      "loss": 3.9374,
+      "step": 1370112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7558860506969834e-05,
+      "loss": 3.9338,
+      "step": 1370624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7550507317066783e-05,
+      "loss": 3.9292,
+      "step": 1371136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7542121369556263e-05,
+      "loss": 3.9366,
+      "step": 1371648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7533751800849472e-05,
+      "loss": 3.9365,
+      "step": 1372160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7525365853338952e-05,
+      "loss": 3.939,
+      "step": 1372672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.751697990582843e-05,
+      "loss": 3.9339,
+      "step": 1373184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.750859395831791e-05,
+      "loss": 3.9301,
+      "step": 1373696
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.008100509643555,
+      "eval_runtime": 300.6516,
+      "eval_samples_per_second": 1269.213,
+      "eval_steps_per_second": 39.664,
+      "step": 1373760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7500208010807395e-05,
+      "loss": 3.928,
+      "step": 1374208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7491822063296875e-05,
+      "loss": 3.9272,
+      "step": 1374720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7483436115786348e-05,
+      "loss": 3.9397,
+      "step": 1375232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7475050168275828e-05,
+      "loss": 3.9294,
+      "step": 1375744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7466664220765308e-05,
+      "loss": 3.9519,
+      "step": 1376256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7458278273254788e-05,
+      "loss": 3.9226,
+      "step": 1376768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.744989232574427e-05,
+      "loss": 3.9338,
+      "step": 1377280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.744150637823375e-05,
+      "loss": 3.9199,
+      "step": 1377792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.743312043072323e-05,
+      "loss": 3.9382,
+      "step": 1378304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.742473448321271e-05,
+      "loss": 3.9315,
+      "step": 1378816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.741634853570219e-05,
+      "loss": 3.936,
+      "step": 1379328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.740796258819167e-05,
+      "loss": 3.9353,
+      "step": 1379840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.739959301948488e-05,
+      "loss": 3.9232,
+      "step": 1380352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7391207071974364e-05,
+      "loss": 3.9266,
+      "step": 1380864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7382821124463843e-05,
+      "loss": 3.9285,
+      "step": 1381376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7374435176953323e-05,
+      "loss": 3.925,
+      "step": 1381888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7366049229442803e-05,
+      "loss": 3.9292,
+      "step": 1382400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7357663281932283e-05,
+      "loss": 3.9318,
+      "step": 1382912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7349277334421763e-05,
+      "loss": 3.928,
+      "step": 1383424
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.734089138691124e-05,
+      "loss": 3.9505,
+      "step": 1383936
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.733250543940072e-05,
+      "loss": 3.9261,
+      "step": 1384448
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.73241194918902e-05,
+      "loss": 3.9351,
+      "step": 1384960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.731574992318341e-05,
+      "loss": 3.9374,
+      "step": 1385472
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.730736397567289e-05,
+      "loss": 3.9411,
+      "step": 1385984
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.729897802816237e-05,
+      "loss": 3.9235,
+      "step": 1386496
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.729059208065185e-05,
+      "loss": 3.9345,
+      "step": 1387008
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.728220613314133e-05,
+      "loss": 3.9255,
+      "step": 1387520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7273820185630812e-05,
+      "loss": 3.932,
+      "step": 1388032
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7265434238120292e-05,
+      "loss": 3.9201,
+      "step": 1388544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7257048290609772e-05,
+      "loss": 3.9221,
+      "step": 1389056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7248662343099252e-05,
+      "loss": 3.9338,
+      "step": 1389568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.724029277439246e-05,
+      "loss": 3.9356,
+      "step": 1390080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.723190682688194e-05,
+      "loss": 3.936,
+      "step": 1390592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7223537258175153e-05,
+      "loss": 3.9401,
+      "step": 1391104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7215151310664633e-05,
+      "loss": 3.9326,
+      "step": 1391616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7206765363154113e-05,
+      "loss": 3.9358,
+      "step": 1392128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7198379415643593e-05,
+      "loss": 3.9238,
+      "step": 1392640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7189993468133073e-05,
+      "loss": 3.9392,
+      "step": 1393152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7181607520622553e-05,
+      "loss": 3.9217,
+      "step": 1393664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7173237951915766e-05,
+      "loss": 3.9301,
+      "step": 1394176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7164852004405246e-05,
+      "loss": 3.9299,
+      "step": 1394688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7156466056894726e-05,
+      "loss": 3.9376,
+      "step": 1395200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7148080109384206e-05,
+      "loss": 3.9277,
+      "step": 1395712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7139710540677415e-05,
+      "loss": 3.9285,
+      "step": 1396224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7131324593166895e-05,
+      "loss": 3.9382,
+      "step": 1396736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7122938645656375e-05,
+      "loss": 3.9347,
+      "step": 1397248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7114552698145858e-05,
+      "loss": 3.9353,
+      "step": 1397760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7106166750635338e-05,
+      "loss": 3.9168,
+      "step": 1398272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7097780803124818e-05,
+      "loss": 3.9161,
+      "step": 1398784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7089411234418027e-05,
+      "loss": 3.9332,
+      "step": 1399296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7081025286907507e-05,
+      "loss": 3.931,
+      "step": 1399808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7072639339396987e-05,
+      "loss": 3.9273,
+      "step": 1400320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7064253391886467e-05,
+      "loss": 3.9132,
+      "step": 1400832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.705586744437595e-05,
+      "loss": 3.9161,
+      "step": 1401344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7047481496865423e-05,
+      "loss": 3.9199,
+      "step": 1401856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.703911192815864e-05,
+      "loss": 3.9317,
+      "step": 1402368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.703072598064812e-05,
+      "loss": 3.9171,
+      "step": 1402880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.70223400331376e-05,
+      "loss": 3.9318,
+      "step": 1403392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7013954085627076e-05,
+      "loss": 3.9286,
+      "step": 1403904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.700558451692029e-05,
+      "loss": 3.9205,
+      "step": 1404416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.699719856940977e-05,
+      "loss": 3.9233,
+      "step": 1404928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6988812621899245e-05,
+      "loss": 3.9211,
+      "step": 1405440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.698044305319246e-05,
+      "loss": 3.9123,
+      "step": 1405952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.697205710568194e-05,
+      "loss": 3.924,
+      "step": 1406464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.696367115817142e-05,
+      "loss": 3.9267,
+      "step": 1406976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6955285210660897e-05,
+      "loss": 3.9357,
+      "step": 1407488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6946899263150377e-05,
+      "loss": 3.9121,
+      "step": 1408000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6938513315639857e-05,
+      "loss": 3.9153,
+      "step": 1408512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6930127368129337e-05,
+      "loss": 3.9166,
+      "step": 1409024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6921741420618817e-05,
+      "loss": 3.9387,
+      "step": 1409536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6913355473108297e-05,
+      "loss": 3.9317,
+      "step": 1410048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.690496952559778e-05,
+      "loss": 3.93,
+      "step": 1410560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.689658357808726e-05,
+      "loss": 3.943,
+      "step": 1411072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.688819763057674e-05,
+      "loss": 3.9388,
+      "step": 1411584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.687984444067368e-05,
+      "loss": 3.938,
+      "step": 1412096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6871458493163158e-05,
+      "loss": 3.9278,
+      "step": 1412608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.686307254565264e-05,
+      "loss": 3.9319,
+      "step": 1413120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.685468659814212e-05,
+      "loss": 3.9289,
+      "step": 1413632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.68463006506316e-05,
+      "loss": 3.9243,
+      "step": 1414144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.683791470312108e-05,
+      "loss": 3.9284,
+      "step": 1414656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.682952875561056e-05,
+      "loss": 3.9294,
+      "step": 1415168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.682115918690377e-05,
+      "loss": 3.9385,
+      "step": 1415680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.681277323939325e-05,
+      "loss": 3.9294,
+      "step": 1416192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.680438729188273e-05,
+      "loss": 3.9229,
+      "step": 1416704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6796001344372214e-05,
+      "loss": 3.9161,
+      "step": 1417216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6787631775665423e-05,
+      "loss": 3.9277,
+      "step": 1417728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6779245828154903e-05,
+      "loss": 3.9337,
+      "step": 1418240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6770859880644383e-05,
+      "loss": 3.9269,
+      "step": 1418752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6762473933133863e-05,
+      "loss": 3.926,
+      "step": 1419264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6754087985623343e-05,
+      "loss": 3.9341,
+      "step": 1419776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6745718416916555e-05,
+      "loss": 3.9179,
+      "step": 1420288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6737332469406035e-05,
+      "loss": 3.929,
+      "step": 1420800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6728962900699244e-05,
+      "loss": 3.9232,
+      "step": 1421312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6720593331992457e-05,
+      "loss": 3.9334,
+      "step": 1421824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6712207384481937e-05,
+      "loss": 3.9335,
+      "step": 1422336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6703821436971417e-05,
+      "loss": 3.9272,
+      "step": 1422848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6695435489460897e-05,
+      "loss": 3.9187,
+      "step": 1423360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6687049541950377e-05,
+      "loss": 3.933,
+      "step": 1423872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6678663594439856e-05,
+      "loss": 3.9219,
+      "step": 1424384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6670277646929336e-05,
+      "loss": 3.9421,
+      "step": 1424896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6661891699418816e-05,
+      "loss": 3.9298,
+      "step": 1425408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6653505751908296e-05,
+      "loss": 3.9314,
+      "step": 1425920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6645119804397776e-05,
+      "loss": 3.9261,
+      "step": 1426432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.663673385688726e-05,
+      "loss": 3.9382,
+      "step": 1426944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6628347909376733e-05,
+      "loss": 3.9116,
+      "step": 1427456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.661997834066995e-05,
+      "loss": 3.9196,
+      "step": 1427968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.661159239315943e-05,
+      "loss": 3.9248,
+      "step": 1428480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6603206445648905e-05,
+      "loss": 3.9249,
+      "step": 1428992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6594820498138385e-05,
+      "loss": 3.9165,
+      "step": 1429504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6586434550627865e-05,
+      "loss": 3.9218,
+      "step": 1430016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6578048603117345e-05,
+      "loss": 3.9147,
+      "step": 1430528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6569662655606825e-05,
+      "loss": 3.9245,
+      "step": 1431040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6561276708096305e-05,
+      "loss": 3.9218,
+      "step": 1431552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6552890760585785e-05,
+      "loss": 3.9193,
+      "step": 1432064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6544521191878997e-05,
+      "loss": 3.9323,
+      "step": 1432576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6536135244368477e-05,
+      "loss": 3.932,
+      "step": 1433088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6527765675661687e-05,
+      "loss": 3.932,
+      "step": 1433600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6519396106954902e-05,
+      "loss": 3.9223,
+      "step": 1434112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6511010159444376e-05,
+      "loss": 3.9332,
+      "step": 1434624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.650262421193386e-05,
+      "loss": 3.9289,
+      "step": 1435136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.649423826442334e-05,
+      "loss": 3.9226,
+      "step": 1435648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.648585231691282e-05,
+      "loss": 3.9394,
+      "step": 1436160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.64774663694023e-05,
+      "loss": 3.9213,
+      "step": 1436672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.646908042189178e-05,
+      "loss": 3.9331,
+      "step": 1437184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.646069447438126e-05,
+      "loss": 3.913,
+      "step": 1437696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.645230852687074e-05,
+      "loss": 3.9303,
+      "step": 1438208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.644393895816395e-05,
+      "loss": 3.9121,
+      "step": 1438720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.643555301065343e-05,
+      "loss": 3.922,
+      "step": 1439232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.642716706314291e-05,
+      "loss": 3.9134,
+      "step": 1439744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.641878111563239e-05,
+      "loss": 3.937,
+      "step": 1440256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.64104115469256e-05,
+      "loss": 3.9298,
+      "step": 1440768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.640202559941508e-05,
+      "loss": 3.927,
+      "step": 1441280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.639363965190456e-05,
+      "loss": 3.9222,
+      "step": 1441792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6385253704394043e-05,
+      "loss": 3.9267,
+      "step": 1442304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6376867756883523e-05,
+      "loss": 3.9409,
+      "step": 1442816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6368481809373003e-05,
+      "loss": 3.9267,
+      "step": 1443328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6360112240666212e-05,
+      "loss": 3.9268,
+      "step": 1443840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6351726293155692e-05,
+      "loss": 3.9358,
+      "step": 1444352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6343340345645172e-05,
+      "loss": 3.9277,
+      "step": 1444864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6334954398134652e-05,
+      "loss": 3.9331,
+      "step": 1445376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6326584829427865e-05,
+      "loss": 3.9214,
+      "step": 1445888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6318198881917345e-05,
+      "loss": 3.928,
+      "step": 1446400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6309812934406825e-05,
+      "loss": 3.9251,
+      "step": 1446912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6301426986896305e-05,
+      "loss": 3.9245,
+      "step": 1447424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6293057418189514e-05,
+      "loss": 3.9234,
+      "step": 1447936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6284671470678997e-05,
+      "loss": 3.9335,
+      "step": 1448448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6276285523168477e-05,
+      "loss": 3.9229,
+      "step": 1448960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6267899575657957e-05,
+      "loss": 3.9264,
+      "step": 1449472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6259530006951166e-05,
+      "loss": 3.9233,
+      "step": 1449984
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.003993988037109,
+      "eval_runtime": 294.7522,
+      "eval_samples_per_second": 1294.616,
+      "eval_steps_per_second": 40.458,
+      "step": 1450080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6251144059440646e-05,
+      "loss": 3.924,
+      "step": 1450496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6242758111930126e-05,
+      "loss": 3.9197,
+      "step": 1451008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6234372164419606e-05,
+      "loss": 3.93,
+      "step": 1451520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6225986216909086e-05,
+      "loss": 3.9184,
+      "step": 1452032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6217600269398562e-05,
+      "loss": 3.9473,
+      "step": 1452544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6209230700691778e-05,
+      "loss": 3.9128,
+      "step": 1453056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6200844753181258e-05,
+      "loss": 3.9243,
+      "step": 1453568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6192458805670738e-05,
+      "loss": 3.9122,
+      "step": 1454080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6184072858160215e-05,
+      "loss": 3.9276,
+      "step": 1454592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6175686910649695e-05,
+      "loss": 3.9253,
+      "step": 1455104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6167300963139175e-05,
+      "loss": 3.9225,
+      "step": 1455616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.615893139443239e-05,
+      "loss": 3.9279,
+      "step": 1456128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.61505618257256e-05,
+      "loss": 3.915,
+      "step": 1456640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.614217587821508e-05,
+      "loss": 3.9203,
+      "step": 1457152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.613378993070456e-05,
+      "loss": 3.9192,
+      "step": 1457664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6125403983194036e-05,
+      "loss": 3.9188,
+      "step": 1458176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6117018035683516e-05,
+      "loss": 3.9181,
+      "step": 1458688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6108632088172996e-05,
+      "loss": 3.9228,
+      "step": 1459200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6100246140662476e-05,
+      "loss": 3.9229,
+      "step": 1459712
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6091860193151956e-05,
+      "loss": 3.9375,
+      "step": 1460224
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6083474245641436e-05,
+      "loss": 3.9189,
+      "step": 1460736
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.607510467693465e-05,
+      "loss": 3.9247,
+      "step": 1461248
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.606671872942413e-05,
+      "loss": 3.9281,
+      "step": 1461760
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6058332781913608e-05,
+      "loss": 3.9345,
+      "step": 1462272
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6049946834403088e-05,
+      "loss": 3.9151,
+      "step": 1462784
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6041577265696297e-05,
+      "loss": 3.9265,
+      "step": 1463296
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.603320769698951e-05,
+      "loss": 3.9167,
+      "step": 1463808
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.602482174947899e-05,
+      "loss": 3.9215,
+      "step": 1464320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.601643580196847e-05,
+      "loss": 3.91,
+      "step": 1464832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.600804985445795e-05,
+      "loss": 3.9197,
+      "step": 1465344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.599966390694743e-05,
+      "loss": 3.9204,
+      "step": 1465856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.599127795943691e-05,
+      "loss": 3.9281,
+      "step": 1466368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.598289201192639e-05,
+      "loss": 3.9245,
+      "step": 1466880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.597450606441587e-05,
+      "loss": 3.9322,
+      "step": 1467392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5966120116905353e-05,
+      "loss": 3.9219,
+      "step": 1467904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5957750548198562e-05,
+      "loss": 3.9286,
+      "step": 1468416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.594938097949177e-05,
+      "loss": 3.9166,
+      "step": 1468928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.594099503198125e-05,
+      "loss": 3.9294,
+      "step": 1469440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.593260908447073e-05,
+      "loss": 3.9123,
+      "step": 1469952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5924223136960214e-05,
+      "loss": 3.922,
+      "step": 1470464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5915837189449694e-05,
+      "loss": 3.9221,
+      "step": 1470976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5907451241939174e-05,
+      "loss": 3.93,
+      "step": 1471488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5899065294428654e-05,
+      "loss": 3.9198,
+      "step": 1472000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5890695725721863e-05,
+      "loss": 3.9207,
+      "step": 1472512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5882309778211343e-05,
+      "loss": 3.9285,
+      "step": 1473024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5873923830700823e-05,
+      "loss": 3.9252,
+      "step": 1473536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5865537883190307e-05,
+      "loss": 3.9261,
+      "step": 1474048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5857151935679786e-05,
+      "loss": 3.915,
+      "step": 1474560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5848765988169266e-05,
+      "loss": 3.8996,
+      "step": 1475072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5840380040658746e-05,
+      "loss": 3.9247,
+      "step": 1475584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5831994093148226e-05,
+      "loss": 3.9206,
+      "step": 1476096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5823624524441435e-05,
+      "loss": 3.9153,
+      "step": 1476608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5815238576930915e-05,
+      "loss": 3.9123,
+      "step": 1477120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.58068526294204e-05,
+      "loss": 3.9093,
+      "step": 1477632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5798466681909872e-05,
+      "loss": 3.908,
+      "step": 1478144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5790080734399352e-05,
+      "loss": 3.9249,
+      "step": 1478656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5781711165692568e-05,
+      "loss": 3.9119,
+      "step": 1479168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5773325218182048e-05,
+      "loss": 3.9188,
+      "step": 1479680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.576495564947526e-05,
+      "loss": 3.9246,
+      "step": 1480192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.575656970196474e-05,
+      "loss": 3.9119,
+      "step": 1480704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.574818375445422e-05,
+      "loss": 3.9124,
+      "step": 1481216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.57397978069437e-05,
+      "loss": 3.9185,
+      "step": 1481728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5731411859433173e-05,
+      "loss": 3.8993,
+      "step": 1482240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.572304229072639e-05,
+      "loss": 3.9174,
+      "step": 1482752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.571465634321587e-05,
+      "loss": 3.921,
+      "step": 1483264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5706270395705346e-05,
+      "loss": 3.9246,
+      "step": 1483776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5697884448194826e-05,
+      "loss": 3.9064,
+      "step": 1484288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5689498500684306e-05,
+      "loss": 3.9117,
+      "step": 1484800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5681112553173786e-05,
+      "loss": 3.9059,
+      "step": 1485312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5672726605663265e-05,
+      "loss": 3.9257,
+      "step": 1485824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5664357036956478e-05,
+      "loss": 3.923,
+      "step": 1486336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5655971089445958e-05,
+      "loss": 3.9239,
+      "step": 1486848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5647585141935438e-05,
+      "loss": 3.9332,
+      "step": 1487360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5639199194424918e-05,
+      "loss": 3.9309,
+      "step": 1487872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5630813246914398e-05,
+      "loss": 3.9322,
+      "step": 1488384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5622443678207607e-05,
+      "loss": 3.9141,
+      "step": 1488896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.561405773069709e-05,
+      "loss": 3.9233,
+      "step": 1489408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.560567178318657e-05,
+      "loss": 3.9255,
+      "step": 1489920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.559728583567605e-05,
+      "loss": 3.9154,
+      "step": 1490432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.558889988816553e-05,
+      "loss": 3.9245,
+      "step": 1490944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.558051394065501e-05,
+      "loss": 3.9197,
+      "step": 1491456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.557212799314449e-05,
+      "loss": 3.9286,
+      "step": 1491968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.556374204563397e-05,
+      "loss": 3.9228,
+      "step": 1492480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.555535609812345e-05,
+      "loss": 3.9068,
+      "step": 1492992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.554697015061293e-05,
+      "loss": 3.9117,
+      "step": 1493504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.553858420310241e-05,
+      "loss": 3.9214,
+      "step": 1494016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5530198255591886e-05,
+      "loss": 3.9261,
+      "step": 1494528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5521828686885102e-05,
+      "loss": 3.9184,
+      "step": 1495040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.5513442739374582e-05,
+      "loss": 3.9163,
+      "step": 1495552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5505056791864055e-05,
+      "loss": 3.9246,
+      "step": 1496064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.549667084435354e-05,
+      "loss": 3.9129,
+      "step": 1496576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.548828489684302e-05,
+      "loss": 3.9151,
+      "step": 1497088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.54798989493325e-05,
+      "loss": 3.9178,
+      "step": 1497600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5471529380625708e-05,
+      "loss": 3.9256,
+      "step": 1498112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5463143433115188e-05,
+      "loss": 3.9274,
+      "step": 1498624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5454757485604668e-05,
+      "loss": 3.9182,
+      "step": 1499136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5446371538094148e-05,
+      "loss": 3.9054,
+      "step": 1499648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.543798559058363e-05,
+      "loss": 3.9284,
+      "step": 1500160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.542959964307311e-05,
+      "loss": 3.9113,
+      "step": 1500672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.542121369556259e-05,
+      "loss": 3.935,
+      "step": 1501184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.541282774805207e-05,
+      "loss": 3.924,
+      "step": 1501696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.540445817934528e-05,
+      "loss": 3.921,
+      "step": 1502208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.539607223183476e-05,
+      "loss": 3.9176,
+      "step": 1502720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.538768628432424e-05,
+      "loss": 3.9296,
+      "step": 1503232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5379300336813723e-05,
+      "loss": 3.9038,
+      "step": 1503744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5370930768106932e-05,
+      "loss": 3.9137,
+      "step": 1504256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.536256119940014e-05,
+      "loss": 3.9106,
+      "step": 1504768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.535417525188962e-05,
+      "loss": 3.9172,
+      "step": 1505280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.53457893043791e-05,
+      "loss": 3.9074,
+      "step": 1505792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5337403356868585e-05,
+      "loss": 3.9157,
+      "step": 1506304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5329017409358065e-05,
+      "loss": 3.9102,
+      "step": 1506816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5320631461847545e-05,
+      "loss": 3.9095,
+      "step": 1507328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5312261893140754e-05,
+      "loss": 3.9156,
+      "step": 1507840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5303875945630234e-05,
+      "loss": 3.908,
+      "step": 1508352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5295489998119714e-05,
+      "loss": 3.9272,
+      "step": 1508864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5287104050609193e-05,
+      "loss": 3.9225,
+      "step": 1509376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5278718103098677e-05,
+      "loss": 3.9235,
+      "step": 1509888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5270348534391886e-05,
+      "loss": 3.9144,
+      "step": 1510400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5261962586881366e-05,
+      "loss": 3.9222,
+      "step": 1510912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5253576639370846e-05,
+      "loss": 3.9254,
+      "step": 1511424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5245190691860326e-05,
+      "loss": 3.9124,
+      "step": 1511936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5236804744349806e-05,
+      "loss": 3.9276,
+      "step": 1512448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5228435175643018e-05,
+      "loss": 3.9145,
+      "step": 1512960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5220049228132498e-05,
+      "loss": 3.9247,
+      "step": 1513472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5211663280621978e-05,
+      "loss": 3.9088,
+      "step": 1513984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5203293711915187e-05,
+      "loss": 3.9194,
+      "step": 1514496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5194907764404667e-05,
+      "loss": 3.9044,
+      "step": 1515008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5186521816894147e-05,
+      "loss": 3.9146,
+      "step": 1515520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.517813586938363e-05,
+      "loss": 3.9093,
+      "step": 1516032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.516974992187311e-05,
+      "loss": 3.9244,
+      "step": 1516544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.516136397436259e-05,
+      "loss": 3.9189,
+      "step": 1517056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.515297802685207e-05,
+      "loss": 3.9242,
+      "step": 1517568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.514460845814528e-05,
+      "loss": 3.9142,
+      "step": 1518080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.513622251063476e-05,
+      "loss": 3.9214,
+      "step": 1518592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.512783656312424e-05,
+      "loss": 3.931,
+      "step": 1519104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5119450615613716e-05,
+      "loss": 3.9185,
+      "step": 1519616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5111064668103196e-05,
+      "loss": 3.9201,
+      "step": 1520128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5102678720592676e-05,
+      "loss": 3.9265,
+      "step": 1520640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5094292773082156e-05,
+      "loss": 3.9169,
+      "step": 1521152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5085906825571636e-05,
+      "loss": 3.9279,
+      "step": 1521664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5077520878061116e-05,
+      "loss": 3.9104,
+      "step": 1522176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.50691349305506e-05,
+      "loss": 3.9225,
+      "step": 1522688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5060781740647544e-05,
+      "loss": 3.916,
+      "step": 1523200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5052395793137017e-05,
+      "loss": 3.9156,
+      "step": 1523712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5044009845626497e-05,
+      "loss": 3.9194,
+      "step": 1524224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.5035623898115977e-05,
+      "loss": 3.919,
+      "step": 1524736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.502723795060546e-05,
+      "loss": 3.9166,
+      "step": 1525248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.501885200309494e-05,
+      "loss": 3.9145,
+      "step": 1525760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.501046605558442e-05,
+      "loss": 3.919,
+      "step": 1526272
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.0003581047058105,
+      "eval_runtime": 294.1306,
+      "eval_samples_per_second": 1297.352,
+      "eval_steps_per_second": 40.543,
+      "step": 1526400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.50020801080739e-05,
+      "loss": 3.912,
+      "step": 1526784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.499369416056338e-05,
+      "loss": 3.9094,
+      "step": 1527296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.498530821305286e-05,
+      "loss": 3.9229,
+      "step": 1527808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.497692226554234e-05,
+      "loss": 3.9107,
+      "step": 1528320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4968536318031817e-05,
+      "loss": 3.9352,
+      "step": 1528832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.49601503705213e-05,
+      "loss": 3.913,
+      "step": 1529344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.495176442301078e-05,
+      "loss": 3.9126,
+      "step": 1529856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.494337847550026e-05,
+      "loss": 3.9022,
+      "step": 1530368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.493499252798974e-05,
+      "loss": 3.9181,
+      "step": 1530880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.492660658047922e-05,
+      "loss": 3.9219,
+      "step": 1531392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4918220632968697e-05,
+      "loss": 3.9139,
+      "step": 1531904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4909834685458176e-05,
+      "loss": 3.9163,
+      "step": 1532416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4901465116751392e-05,
+      "loss": 3.9087,
+      "step": 1532928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4893079169240872e-05,
+      "loss": 3.9128,
+      "step": 1533440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.488469322173035e-05,
+      "loss": 3.9132,
+      "step": 1533952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.487630727421983e-05,
+      "loss": 3.9076,
+      "step": 1534464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.486793770551304e-05,
+      "loss": 3.9116,
+      "step": 1534976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.485955175800252e-05,
+      "loss": 3.9159,
+      "step": 1535488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4851165810492e-05,
+      "loss": 3.9122,
+      "step": 1536000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.484277986298148e-05,
+      "loss": 3.9278,
+      "step": 1536512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.483439391547096e-05,
+      "loss": 3.9153,
+      "step": 1537024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.482600796796044e-05,
+      "loss": 3.9178,
+      "step": 1537536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.481762202044992e-05,
+      "loss": 3.9199,
+      "step": 1538048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.48092360729394e-05,
+      "loss": 3.9254,
+      "step": 1538560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4800850125428878e-05,
+      "loss": 3.9075,
+      "step": 1539072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4792464177918358e-05,
+      "loss": 3.9166,
+      "step": 1539584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.478407823040784e-05,
+      "loss": 3.9133,
+      "step": 1540096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.4775708661701053e-05,
+      "loss": 3.9106,
+      "step": 1540608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.476732271419053e-05,
+      "loss": 3.9023,
+      "step": 1541120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.475893676668001e-05,
+      "loss": 3.9096,
+      "step": 1541632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.475055081916949e-05,
+      "loss": 3.9179,
+      "step": 1542144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.474216487165897e-05,
+      "loss": 3.9194,
+      "step": 1542656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.473377892414845e-05,
+      "loss": 3.9184,
+      "step": 1543168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4725409355441662e-05,
+      "loss": 3.921,
+      "step": 1543680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4717023407931142e-05,
+      "loss": 3.9159,
+      "step": 1544192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4708637460420622e-05,
+      "loss": 3.9199,
+      "step": 1544704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4700251512910102e-05,
+      "loss": 3.9127,
+      "step": 1545216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.469188194420331e-05,
+      "loss": 3.9174,
+      "step": 1545728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4683495996692795e-05,
+      "loss": 3.9085,
+      "step": 1546240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4675110049182275e-05,
+      "loss": 3.9138,
+      "step": 1546752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4666724101671754e-05,
+      "loss": 3.914,
+      "step": 1547264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4658338154161234e-05,
+      "loss": 3.9213,
+      "step": 1547776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.464995220665071e-05,
+      "loss": 3.9109,
+      "step": 1548288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4641582637943923e-05,
+      "loss": 3.913,
+      "step": 1548800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4633196690433403e-05,
+      "loss": 3.9231,
+      "step": 1549312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4624810742922887e-05,
+      "loss": 3.9137,
+      "step": 1549824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4616424795412363e-05,
+      "loss": 3.9168,
+      "step": 1550336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4608055226705576e-05,
+      "loss": 3.9062,
+      "step": 1550848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4599669279195056e-05,
+      "loss": 3.8928,
+      "step": 1551360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4591283331684532e-05,
+      "loss": 3.9193,
+      "step": 1551872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4582897384174016e-05,
+      "loss": 3.9145,
+      "step": 1552384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4574511436663496e-05,
+      "loss": 3.9112,
+      "step": 1552896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4566125489152976e-05,
+      "loss": 3.8988,
+      "step": 1553408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4557739541642456e-05,
+      "loss": 3.9032,
+      "step": 1553920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4549353594131936e-05,
+      "loss": 3.8975,
+      "step": 1554432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4540984025425145e-05,
+      "loss": 3.9168,
+      "step": 1554944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4532598077914625e-05,
+      "loss": 3.9039,
+      "step": 1555456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4524212130404105e-05,
+      "loss": 3.9076,
+      "step": 1555968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4515826182893588e-05,
+      "loss": 3.9186,
+      "step": 1556480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4507456614186797e-05,
+      "loss": 3.904,
+      "step": 1556992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4499070666676277e-05,
+      "loss": 3.9059,
+      "step": 1557504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4490684719165757e-05,
+      "loss": 3.9119,
+      "step": 1558016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4482298771655237e-05,
+      "loss": 3.8906,
+      "step": 1558528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.447392920294845e-05,
+      "loss": 3.9089,
+      "step": 1559040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.446554325543793e-05,
+      "loss": 3.9102,
+      "step": 1559552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.445715730792741e-05,
+      "loss": 3.9171,
+      "step": 1560064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.444877136041689e-05,
+      "loss": 3.8984,
+      "step": 1560576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4440401791710098e-05,
+      "loss": 3.9049,
+      "step": 1561088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4432015844199578e-05,
+      "loss": 3.8945,
+      "step": 1561600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4423629896689058e-05,
+      "loss": 3.9201,
+      "step": 1562112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4415243949178538e-05,
+      "loss": 3.9151,
+      "step": 1562624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4406858001668018e-05,
+      "loss": 3.9131,
+      "step": 1563136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4398472054157498e-05,
+      "loss": 3.9268,
+      "step": 1563648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.439010248545071e-05,
+      "loss": 3.9204,
+      "step": 1564160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4381716537940187e-05,
+      "loss": 3.9339,
+      "step": 1564672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.437333059042967e-05,
+      "loss": 3.9067,
+      "step": 1565184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.436494464291915e-05,
+      "loss": 3.9131,
+      "step": 1565696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.435655869540863e-05,
+      "loss": 3.9182,
+      "step": 1566208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.434817274789811e-05,
+      "loss": 3.9048,
+      "step": 1566720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.433978680038759e-05,
+      "loss": 3.9181,
+      "step": 1567232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.43314172316808e-05,
+      "loss": 3.9121,
+      "step": 1567744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.432303128417028e-05,
+      "loss": 3.9194,
+      "step": 1568256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.431464533665976e-05,
+      "loss": 3.9147,
+      "step": 1568768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4306259389149243e-05,
+      "loss": 3.9016,
+      "step": 1569280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.429787344163872e-05,
+      "loss": 3.9131,
+      "step": 1569792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.42894874941282e-05,
+      "loss": 3.9041,
+      "step": 1570304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.428111792542141e-05,
+      "loss": 3.92,
+      "step": 1570816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.427273197791089e-05,
+      "loss": 3.9132,
+      "step": 1571328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.426434603040037e-05,
+      "loss": 3.9065,
+      "step": 1571840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.425596008288985e-05,
+      "loss": 3.9167,
+      "step": 1572352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.424757413537933e-05,
+      "loss": 3.9097,
+      "step": 1572864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4239204566672544e-05,
+      "loss": 3.9057,
+      "step": 1573376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4230834997965753e-05,
+      "loss": 3.9098,
+      "step": 1573888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4222449050455233e-05,
+      "loss": 3.9217,
+      "step": 1574400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4214063102944713e-05,
+      "loss": 3.9194,
+      "step": 1574912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4205677155434193e-05,
+      "loss": 3.912,
+      "step": 1575424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4197291207923673e-05,
+      "loss": 3.8986,
+      "step": 1575936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4188905260413153e-05,
+      "loss": 3.9151,
+      "step": 1576448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4180519312902633e-05,
+      "loss": 3.9053,
+      "step": 1576960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4172133365392113e-05,
+      "loss": 3.9237,
+      "step": 1577472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4163747417881593e-05,
+      "loss": 3.9193,
+      "step": 1577984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4155377849174805e-05,
+      "loss": 3.9147,
+      "step": 1578496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4146991901664285e-05,
+      "loss": 3.9108,
+      "step": 1579008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4138605954153765e-05,
+      "loss": 3.9242,
+      "step": 1579520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4130220006643245e-05,
+      "loss": 3.8945,
+      "step": 1580032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4121834059132725e-05,
+      "loss": 3.9067,
+      "step": 1580544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4113464490425934e-05,
+      "loss": 3.8997,
+      "step": 1581056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4105078542915417e-05,
+      "loss": 3.9164,
+      "step": 1581568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4096692595404897e-05,
+      "loss": 3.8977,
+      "step": 1582080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4088306647894374e-05,
+      "loss": 3.9083,
+      "step": 1582592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4079920700383854e-05,
+      "loss": 3.9047,
+      "step": 1583104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4071534752873334e-05,
+      "loss": 3.9022,
+      "step": 1583616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4063165184166546e-05,
+      "loss": 3.9075,
+      "step": 1584128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4054779236656026e-05,
+      "loss": 3.9058,
+      "step": 1584640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4046393289145506e-05,
+      "loss": 3.9187,
+      "step": 1585152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4038007341634986e-05,
+      "loss": 3.9171,
+      "step": 1585664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4029621394124466e-05,
+      "loss": 3.9143,
+      "step": 1586176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4021235446613946e-05,
+      "loss": 3.9054,
+      "step": 1586688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4012865877907155e-05,
+      "loss": 3.9171,
+      "step": 1587200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.4004479930396635e-05,
+      "loss": 3.9138,
+      "step": 1587712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.399609398288612e-05,
+      "loss": 3.9045,
+      "step": 1588224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3987724414179328e-05,
+      "loss": 3.9243,
+      "step": 1588736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3979338466668808e-05,
+      "loss": 3.9057,
+      "step": 1589248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3970952519158288e-05,
+      "loss": 3.9144,
+      "step": 1589760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3962566571647767e-05,
+      "loss": 3.9021,
+      "step": 1590272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3954180624137247e-05,
+      "loss": 3.9143,
+      "step": 1590784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3945794676626727e-05,
+      "loss": 3.9012,
+      "step": 1591296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.393742510791994e-05,
+      "loss": 3.9042,
+      "step": 1591808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.392903916040942e-05,
+      "loss": 3.9045,
+      "step": 1592320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.39206532128989e-05,
+      "loss": 3.916,
+      "step": 1592832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3912267265388376e-05,
+      "loss": 3.9107,
+      "step": 1593344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3903881317877856e-05,
+      "loss": 3.9151,
+      "step": 1593856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3895495370367336e-05,
+      "loss": 3.9036,
+      "step": 1594368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.388710942285682e-05,
+      "loss": 3.9174,
+      "step": 1594880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.38787234753463e-05,
+      "loss": 3.9222,
+      "step": 1595392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.387033752783578e-05,
+      "loss": 3.9142,
+      "step": 1595904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.386195158032526e-05,
+      "loss": 3.9117,
+      "step": 1596416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3853565632814736e-05,
+      "loss": 3.918,
+      "step": 1596928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.384519606410795e-05,
+      "loss": 3.909,
+      "step": 1597440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.383681011659743e-05,
+      "loss": 3.9203,
+      "step": 1597952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3828424169086912e-05,
+      "loss": 3.9051,
+      "step": 1598464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.382003822157639e-05,
+      "loss": 3.9149,
+      "step": 1598976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.38116686528696e-05,
+      "loss": 3.9082,
+      "step": 1599488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.380328270535908e-05,
+      "loss": 3.9111,
+      "step": 1600000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3794896757848557e-05,
+      "loss": 3.9107,
+      "step": 1600512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3786527189141773e-05,
+      "loss": 3.9122,
+      "step": 1601024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3778141241631253e-05,
+      "loss": 3.9106,
+      "step": 1601536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.3769755294120733e-05,
+      "loss": 3.9029,
+      "step": 1602048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.376136934661021e-05,
+      "loss": 3.9145,
+      "step": 1602560
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.9975011348724365,
+      "eval_runtime": 290.9243,
+      "eval_samples_per_second": 1311.651,
+      "eval_steps_per_second": 40.99,
+      "step": 1602720
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3752999777903422e-05,
+      "loss": 3.908,
+      "step": 1603072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3744613830392902e-05,
+      "loss": 3.9008,
+      "step": 1603584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3736227882882382e-05,
+      "loss": 3.9169,
+      "step": 1604096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3727841935371862e-05,
+      "loss": 3.9058,
+      "step": 1604608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3719455987861342e-05,
+      "loss": 3.9235,
+      "step": 1605120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3711070040350822e-05,
+      "loss": 3.9096,
+      "step": 1605632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3702684092840302e-05,
+      "loss": 3.8979,
+      "step": 1606144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3694298145329782e-05,
+      "loss": 3.902,
+      "step": 1606656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3685912197819262e-05,
+      "loss": 3.9053,
+      "step": 1607168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3677526250308742e-05,
+      "loss": 3.914,
+      "step": 1607680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3669140302798222e-05,
+      "loss": 3.9052,
+      "step": 1608192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3660770734091434e-05,
+      "loss": 3.9097,
+      "step": 1608704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3652401165384643e-05,
+      "loss": 3.9026,
+      "step": 1609216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3644015217874123e-05,
+      "loss": 3.8999,
+      "step": 1609728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3635629270363603e-05,
+      "loss": 3.9104,
+      "step": 1610240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3627243322853083e-05,
+      "loss": 3.8993,
+      "step": 1610752
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3618857375342567e-05,
+      "loss": 3.903,
+      "step": 1611264
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3610471427832043e-05,
+      "loss": 3.9077,
+      "step": 1611776
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3602085480321523e-05,
+      "loss": 3.9094,
+      "step": 1612288
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3593699532811003e-05,
+      "loss": 3.9172,
+      "step": 1612800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3585329964104212e-05,
+      "loss": 3.9123,
+      "step": 1613312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3576960395397428e-05,
+      "loss": 3.9116,
+      "step": 1613824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3568574447886908e-05,
+      "loss": 3.9123,
+      "step": 1614336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3560188500376388e-05,
+      "loss": 3.9147,
+      "step": 1614848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3551802552865865e-05,
+      "loss": 3.8993,
+      "step": 1615360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3543416605355344e-05,
+      "loss": 3.9144,
+      "step": 1615872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3535047036648557e-05,
+      "loss": 3.9022,
+      "step": 1616384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3526661089138037e-05,
+      "loss": 3.9022,
+      "step": 1616896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3518275141627517e-05,
+      "loss": 3.8994,
+      "step": 1617408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.3509889194116997e-05,
+      "loss": 3.9017,
+      "step": 1617920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.350151962541021e-05,
+      "loss": 3.9099,
+      "step": 1618432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3493133677899686e-05,
+      "loss": 3.9096,
+      "step": 1618944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3484747730389166e-05,
+      "loss": 3.9151,
+      "step": 1619456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.347636178287865e-05,
+      "loss": 3.9157,
+      "step": 1619968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.346797583536813e-05,
+      "loss": 3.9046,
+      "step": 1620480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.345958988785761e-05,
+      "loss": 3.9115,
+      "step": 1620992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.345120394034709e-05,
+      "loss": 3.9058,
+      "step": 1621504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.344281799283657e-05,
+      "loss": 3.9074,
+      "step": 1622016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3434432045326046e-05,
+      "loss": 3.9015,
+      "step": 1622528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3426046097815526e-05,
+      "loss": 3.9017,
+      "step": 1623040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3417660150305005e-05,
+      "loss": 3.9091,
+      "step": 1623552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.340927420279449e-05,
+      "loss": 3.9156,
+      "step": 1624064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.340088825528397e-05,
+      "loss": 3.9021,
+      "step": 1624576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.339250230777345e-05,
+      "loss": 3.9063,
+      "step": 1625088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3384132739066658e-05,
+      "loss": 3.9146,
+      "step": 1625600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3375746791556138e-05,
+      "loss": 3.9086,
+      "step": 1626112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3367360844045618e-05,
+      "loss": 3.91,
+      "step": 1626624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3358974896535098e-05,
+      "loss": 3.902,
+      "step": 1627136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3350588949024578e-05,
+      "loss": 3.881,
+      "step": 1627648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3342203001514058e-05,
+      "loss": 3.9168,
+      "step": 1628160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3333817054003538e-05,
+      "loss": 3.9054,
+      "step": 1628672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3325431106493018e-05,
+      "loss": 3.9052,
+      "step": 1629184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3317045158982497e-05,
+      "loss": 3.8976,
+      "step": 1629696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3308659211471977e-05,
+      "loss": 3.89,
+      "step": 1630208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.330028964276519e-05,
+      "loss": 3.8913,
+      "step": 1630720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.329190369525467e-05,
+      "loss": 3.9044,
+      "step": 1631232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.328353412654788e-05,
+      "loss": 3.8974,
+      "step": 1631744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.327514817903736e-05,
+      "loss": 3.9055,
+      "step": 1632256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.326676223152684e-05,
+      "loss": 3.9084,
+      "step": 1632768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.325837628401632e-05,
+      "loss": 3.8983,
+      "step": 1633280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.325000671530953e-05,
+      "loss": 3.8988,
+      "step": 1633792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.324162076779901e-05,
+      "loss": 3.9073,
+      "step": 1634304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.323323482028849e-05,
+      "loss": 3.8878,
+      "step": 1634816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.322484887277797e-05,
+      "loss": 3.8982,
+      "step": 1635328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.321646292526745e-05,
+      "loss": 3.9009,
+      "step": 1635840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.320809335656066e-05,
+      "loss": 3.9124,
+      "step": 1636352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3199707409050144e-05,
+      "loss": 3.8891,
+      "step": 1636864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3191321461539624e-05,
+      "loss": 3.9009,
+      "step": 1637376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3182935514029103e-05,
+      "loss": 3.8887,
+      "step": 1637888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.317454956651858e-05,
+      "loss": 3.9107,
+      "step": 1638400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3166179997811793e-05,
+      "loss": 3.9073,
+      "step": 1638912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3157794050301272e-05,
+      "loss": 3.9054,
+      "step": 1639424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3149408102790752e-05,
+      "loss": 3.9165,
+      "step": 1639936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3141022155280232e-05,
+      "loss": 3.9131,
+      "step": 1640448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3132636207769712e-05,
+      "loss": 3.925,
+      "step": 1640960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3124266639062925e-05,
+      "loss": 3.9022,
+      "step": 1641472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3115880691552405e-05,
+      "loss": 3.9071,
+      "step": 1641984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.310749474404188e-05,
+      "loss": 3.9063,
+      "step": 1642496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.309910879653136e-05,
+      "loss": 3.895,
+      "step": 1643008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3090722849020845e-05,
+      "loss": 3.9134,
+      "step": 1643520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3082353280314054e-05,
+      "loss": 3.9044,
+      "step": 1644032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3073967332803534e-05,
+      "loss": 3.9123,
+      "step": 1644544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3065581385293014e-05,
+      "loss": 3.9081,
+      "step": 1645056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3057195437782494e-05,
+      "loss": 3.8956,
+      "step": 1645568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3048825869075706e-05,
+      "loss": 3.905,
+      "step": 1646080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3040439921565186e-05,
+      "loss": 3.8968,
+      "step": 1646592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3032053974054666e-05,
+      "loss": 3.9124,
+      "step": 1647104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.302368440534788e-05,
+      "loss": 3.9092,
+      "step": 1647616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.3015298457837355e-05,
+      "loss": 3.8949,
+      "step": 1648128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.3006912510326835e-05,
+      "loss": 3.9144,
+      "step": 1648640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2998526562816315e-05,
+      "loss": 3.9033,
+      "step": 1649152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.29901406153058e-05,
+      "loss": 3.8989,
+      "step": 1649664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2981754667795278e-05,
+      "loss": 3.9003,
+      "step": 1650176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2973368720284758e-05,
+      "loss": 3.9145,
+      "step": 1650688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2964982772774235e-05,
+      "loss": 3.9097,
+      "step": 1651200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2956596825263715e-05,
+      "loss": 3.9111,
+      "step": 1651712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2948210877753195e-05,
+      "loss": 3.8918,
+      "step": 1652224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2939824930242675e-05,
+      "loss": 3.9078,
+      "step": 1652736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2931438982732155e-05,
+      "loss": 3.9011,
+      "step": 1653248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2923069414025367e-05,
+      "loss": 3.9151,
+      "step": 1653760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.291469984531858e-05,
+      "loss": 3.9122,
+      "step": 1654272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.290631389780806e-05,
+      "loss": 3.9027,
+      "step": 1654784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2897927950297536e-05,
+      "loss": 3.9078,
+      "step": 1655296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2889542002787016e-05,
+      "loss": 3.9134,
+      "step": 1655808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.28811560552765e-05,
+      "loss": 3.8897,
+      "step": 1656320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.287280286537344e-05,
+      "loss": 3.8999,
+      "step": 1656832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.286441691786292e-05,
+      "loss": 3.8924,
+      "step": 1657344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.28560309703524e-05,
+      "loss": 3.9055,
+      "step": 1657856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.284764502284188e-05,
+      "loss": 3.8962,
+      "step": 1658368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.283925907533136e-05,
+      "loss": 3.9053,
+      "step": 1658880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.283087312782084e-05,
+      "loss": 3.8959,
+      "step": 1659392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.282248718031032e-05,
+      "loss": 3.8969,
+      "step": 1659904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.28141012327998e-05,
+      "loss": 3.9025,
+      "step": 1660416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.280571528528928e-05,
+      "loss": 3.8966,
+      "step": 1660928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.279732933777876e-05,
+      "loss": 3.9111,
+      "step": 1661440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2788943390268237e-05,
+      "loss": 3.9076,
+      "step": 1661952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.278055744275772e-05,
+      "loss": 3.9095,
+      "step": 1662464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2772187874050933e-05,
+      "loss": 3.8982,
+      "step": 1662976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2763801926540413e-05,
+      "loss": 3.9077,
+      "step": 1663488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.275541597902989e-05,
+      "loss": 3.9101,
+      "step": 1664000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.274703003151937e-05,
+      "loss": 3.8974,
+      "step": 1664512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.273864408400885e-05,
+      "loss": 3.9192,
+      "step": 1665024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.273025813649833e-05,
+      "loss": 3.8946,
+      "step": 1665536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2721872188987813e-05,
+      "loss": 3.9087,
+      "step": 1666048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2713502620281022e-05,
+      "loss": 3.8918,
+      "step": 1666560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2705116672770502e-05,
+      "loss": 3.9097,
+      "step": 1667072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.269674710406371e-05,
+      "loss": 3.8917,
+      "step": 1667584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.268836115655319e-05,
+      "loss": 3.8995,
+      "step": 1668096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2679975209042674e-05,
+      "loss": 3.8924,
+      "step": 1668608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2671589261532154e-05,
+      "loss": 3.9141,
+      "step": 1669120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2663203314021634e-05,
+      "loss": 3.9005,
+      "step": 1669632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2654817366511114e-05,
+      "loss": 3.9084,
+      "step": 1670144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2646431419000594e-05,
+      "loss": 3.8964,
+      "step": 1670656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.263804547149007e-05,
+      "loss": 3.9111,
+      "step": 1671168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.262965952397955e-05,
+      "loss": 3.9129,
+      "step": 1671680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.262127357646903e-05,
+      "loss": 3.9081,
+      "step": 1672192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2612887628958514e-05,
+      "loss": 3.908,
+      "step": 1672704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2604501681447994e-05,
+      "loss": 3.9073,
+      "step": 1673216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2596148491544935e-05,
+      "loss": 3.9039,
+      "step": 1673728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2587762544034415e-05,
+      "loss": 3.9128,
+      "step": 1674240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2579376596523892e-05,
+      "loss": 3.8968,
+      "step": 1674752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2570990649013375e-05,
+      "loss": 3.9069,
+      "step": 1675264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2562621080306588e-05,
+      "loss": 3.9034,
+      "step": 1675776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2554235132796068e-05,
+      "loss": 3.905,
+      "step": 1676288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2545849185285544e-05,
+      "loss": 3.9043,
+      "step": 1676800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2537463237775024e-05,
+      "loss": 3.9004,
+      "step": 1677312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2529077290264504e-05,
+      "loss": 3.9119,
+      "step": 1677824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2520707721557717e-05,
+      "loss": 3.8911,
+      "step": 1678336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.2512321774047197e-05,
+      "loss": 3.9119,
+      "step": 1678848
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.9947001934051514,
+      "eval_runtime": 293.7078,
+      "eval_samples_per_second": 1299.22,
+      "eval_steps_per_second": 40.602,
+      "step": 1679040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2503935826536677e-05,
+      "loss": 3.9009,
+      "step": 1679360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2495549879026157e-05,
+      "loss": 3.893,
+      "step": 1679872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2487163931515637e-05,
+      "loss": 3.9076,
+      "step": 1680384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2478777984005117e-05,
+      "loss": 3.9018,
+      "step": 1680896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2470392036494596e-05,
+      "loss": 3.9146,
+      "step": 1681408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2462006088984076e-05,
+      "loss": 3.9028,
+      "step": 1681920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2453620141473556e-05,
+      "loss": 3.8944,
+      "step": 1682432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2445234193963036e-05,
+      "loss": 3.8925,
+      "step": 1682944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2436848246452516e-05,
+      "loss": 3.9004,
+      "step": 1683456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2428462298941996e-05,
+      "loss": 3.9073,
+      "step": 1683968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2420076351431476e-05,
+      "loss": 3.9015,
+      "step": 1684480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2411690403920956e-05,
+      "loss": 3.9016,
+      "step": 1684992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.240332083521417e-05,
+      "loss": 3.9013,
+      "step": 1685504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.239493488770365e-05,
+      "loss": 3.8944,
+      "step": 1686016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.238654894019313e-05,
+      "loss": 3.8997,
+      "step": 1686528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.237816299268261e-05,
+      "loss": 3.8879,
+      "step": 1687040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2369777045172085e-05,
+      "loss": 3.8995,
+      "step": 1687552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2361391097661565e-05,
+      "loss": 3.8986,
+      "step": 1688064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2353005150151045e-05,
+      "loss": 3.8996,
+      "step": 1688576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2344619202640525e-05,
+      "loss": 3.9105,
+      "step": 1689088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2336233255130008e-05,
+      "loss": 3.9106,
+      "step": 1689600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2327847307619488e-05,
+      "loss": 3.9006,
+      "step": 1690112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2319477738912697e-05,
+      "loss": 3.9035,
+      "step": 1690624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2311091791402177e-05,
+      "loss": 3.9107,
+      "step": 1691136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2302705843891657e-05,
+      "loss": 3.8891,
+      "step": 1691648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2294319896381137e-05,
+      "loss": 3.9119,
+      "step": 1692160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.2285933948870617e-05,
+      "loss": 3.895,
+      "step": 1692672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.227756438016383e-05,
+      "loss": 3.8919,
+      "step": 1693184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.226917843265331e-05,
+      "loss": 3.8925,
+      "step": 1693696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.226079248514279e-05,
+      "loss": 3.8982,
+      "step": 1694208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2252406537632266e-05,
+      "loss": 3.8995,
+      "step": 1694720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.224403696892548e-05,
+      "loss": 3.9052,
+      "step": 1695232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2235651021414962e-05,
+      "loss": 3.9038,
+      "step": 1695744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.222726507390444e-05,
+      "loss": 3.9096,
+      "step": 1696256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.221887912639392e-05,
+      "loss": 3.8969,
+      "step": 1696768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.22104931788834e-05,
+      "loss": 3.9085,
+      "step": 1697280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.220210723137288e-05,
+      "loss": 3.905,
+      "step": 1697792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.219373766266609e-05,
+      "loss": 3.8973,
+      "step": 1698304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.218535171515557e-05,
+      "loss": 3.8955,
+      "step": 1698816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.217696576764505e-05,
+      "loss": 3.8993,
+      "step": 1699328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.216857982013453e-05,
+      "loss": 3.8998,
+      "step": 1699840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.216019387262401e-05,
+      "loss": 3.9107,
+      "step": 1700352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.215180792511349e-05,
+      "loss": 3.8955,
+      "step": 1700864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2143421977602967e-05,
+      "loss": 3.898,
+      "step": 1701376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.213505240889618e-05,
+      "loss": 3.9094,
+      "step": 1701888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2126666461385663e-05,
+      "loss": 3.8983,
+      "step": 1702400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2118280513875143e-05,
+      "loss": 3.9063,
+      "step": 1702912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.210989456636462e-05,
+      "loss": 3.8979,
+      "step": 1703424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.21015086188541e-05,
+      "loss": 3.8661,
+      "step": 1703936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.209312267134358e-05,
+      "loss": 3.9135,
+      "step": 1704448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2084753102636792e-05,
+      "loss": 3.898,
+      "step": 1704960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2076367155126272e-05,
+      "loss": 3.8983,
+      "step": 1705472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2067981207615752e-05,
+      "loss": 3.8913,
+      "step": 1705984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2059595260105232e-05,
+      "loss": 3.8809,
+      "step": 1706496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2051209312594712e-05,
+      "loss": 3.89,
+      "step": 1707008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2042823365084192e-05,
+      "loss": 3.8977,
+      "step": 1707520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.20344537963774e-05,
+      "loss": 3.8939,
+      "step": 1708032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2026067848866884e-05,
+      "loss": 3.8942,
+      "step": 1708544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2017681901356364e-05,
+      "loss": 3.9041,
+      "step": 1709056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2009295953845844e-05,
+      "loss": 3.8892,
+      "step": 1709568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2000910006335324e-05,
+      "loss": 3.8895,
+      "step": 1710080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.19925240588248e-05,
+      "loss": 3.9041,
+      "step": 1710592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.198413811131428e-05,
+      "loss": 3.8764,
+      "step": 1711104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.197575216380376e-05,
+      "loss": 3.8909,
+      "step": 1711616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1967382595096976e-05,
+      "loss": 3.8951,
+      "step": 1712128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1958996647586453e-05,
+      "loss": 3.9077,
+      "step": 1712640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1950627078879665e-05,
+      "loss": 3.8798,
+      "step": 1713152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1942241131369145e-05,
+      "loss": 3.8972,
+      "step": 1713664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1933855183858622e-05,
+      "loss": 3.8774,
+      "step": 1714176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1925469236348102e-05,
+      "loss": 3.9056,
+      "step": 1714688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1917083288837585e-05,
+      "loss": 3.9035,
+      "step": 1715200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1908697341327065e-05,
+      "loss": 3.899,
+      "step": 1715712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1900311393816545e-05,
+      "loss": 3.9052,
+      "step": 1716224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1891925446306025e-05,
+      "loss": 3.9082,
+      "step": 1716736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1883555877599234e-05,
+      "loss": 3.9215,
+      "step": 1717248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1875186308892447e-05,
+      "loss": 3.895,
+      "step": 1717760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1866800361381927e-05,
+      "loss": 3.9025,
+      "step": 1718272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1858414413871407e-05,
+      "loss": 3.9022,
+      "step": 1718784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1850028466360887e-05,
+      "loss": 3.8877,
+      "step": 1719296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1841642518850367e-05,
+      "loss": 3.9012,
+      "step": 1719808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1833272950143576e-05,
+      "loss": 3.8996,
+      "step": 1720320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1824887002633056e-05,
+      "loss": 3.9045,
+      "step": 1720832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.181650105512254e-05,
+      "loss": 3.9099,
+      "step": 1721344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.180811510761202e-05,
+      "loss": 3.8882,
+      "step": 1721856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.17997291601015e-05,
+      "loss": 3.8922,
+      "step": 1722368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.179134321259098e-05,
+      "loss": 3.8913,
+      "step": 1722880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1782957265080455e-05,
+      "loss": 3.9064,
+      "step": 1723392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1774571317569935e-05,
+      "loss": 3.9005,
+      "step": 1723904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.1766201748863148e-05,
+      "loss": 3.8937,
+      "step": 1724416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.175781580135263e-05,
+      "loss": 3.8936,
+      "step": 1724928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1749429853842108e-05,
+      "loss": 3.9032,
+      "step": 1725440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1741043906331588e-05,
+      "loss": 3.8927,
+      "step": 1725952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1732657958821068e-05,
+      "loss": 3.8947,
+      "step": 1726464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1724272011310548e-05,
+      "loss": 3.9109,
+      "step": 1726976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1715902442603757e-05,
+      "loss": 3.902,
+      "step": 1727488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.170751649509324e-05,
+      "loss": 3.9026,
+      "step": 1728000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.169913054758272e-05,
+      "loss": 3.8884,
+      "step": 1728512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.169076097887593e-05,
+      "loss": 3.8998,
+      "step": 1729024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.168237503136541e-05,
+      "loss": 3.8939,
+      "step": 1729536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.167398908385489e-05,
+      "loss": 3.9065,
+      "step": 1730048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.166560313634437e-05,
+      "loss": 3.9044,
+      "step": 1730560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.165721718883385e-05,
+      "loss": 3.9031,
+      "step": 1731072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1648831241323332e-05,
+      "loss": 3.9021,
+      "step": 1731584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.164044529381281e-05,
+      "loss": 3.9052,
+      "step": 1732096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.163205934630229e-05,
+      "loss": 3.8839,
+      "step": 1732608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.16236897775955e-05,
+      "loss": 3.8952,
+      "step": 1733120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.161530383008498e-05,
+      "loss": 3.8825,
+      "step": 1733632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.160691788257446e-05,
+      "loss": 3.9017,
+      "step": 1734144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.159853193506394e-05,
+      "loss": 3.8884,
+      "step": 1734656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1590162366357154e-05,
+      "loss": 3.8951,
+      "step": 1735168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1581776418846634e-05,
+      "loss": 3.8882,
+      "step": 1735680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.157339047133611e-05,
+      "loss": 3.8907,
+      "step": 1736192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.156500452382559e-05,
+      "loss": 3.8945,
+      "step": 1736704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.155661857631507e-05,
+      "loss": 3.8915,
+      "step": 1737216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1548249007608283e-05,
+      "loss": 3.899,
+      "step": 1737728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1539863060097762e-05,
+      "loss": 3.9063,
+      "step": 1738240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1531477112587242e-05,
+      "loss": 3.8983,
+      "step": 1738752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1523091165076722e-05,
+      "loss": 3.8995,
+      "step": 1739264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1514705217566202e-05,
+      "loss": 3.8965,
+      "step": 1739776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.150633564885941e-05,
+      "loss": 3.9025,
+      "step": 1740288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1497949701348895e-05,
+      "loss": 3.8956,
+      "step": 1740800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1489563753838375e-05,
+      "loss": 3.9079,
+      "step": 1741312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1481177806327855e-05,
+      "loss": 3.8891,
+      "step": 1741824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1472791858817335e-05,
+      "loss": 3.9035,
+      "step": 1742336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1464405911306815e-05,
+      "loss": 3.8903,
+      "step": 1742848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.145601996379629e-05,
+      "loss": 3.8939,
+      "step": 1743360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.144763401628577e-05,
+      "loss": 3.8912,
+      "step": 1743872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1439248068775254e-05,
+      "loss": 3.8916,
+      "step": 1744384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1430878500068464e-05,
+      "loss": 3.8877,
+      "step": 1744896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1422492552557944e-05,
+      "loss": 3.9033,
+      "step": 1745408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1414106605047423e-05,
+      "loss": 3.8978,
+      "step": 1745920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1405737036340636e-05,
+      "loss": 3.9012,
+      "step": 1746432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1397351088830116e-05,
+      "loss": 3.889,
+      "step": 1746944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1388965141319596e-05,
+      "loss": 3.9065,
+      "step": 1747456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1380579193809076e-05,
+      "loss": 3.905,
+      "step": 1747968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1372193246298556e-05,
+      "loss": 3.9031,
+      "step": 1748480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1363823677591765e-05,
+      "loss": 3.9012,
+      "step": 1748992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1355437730081245e-05,
+      "loss": 3.897,
+      "step": 1749504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1347051782570725e-05,
+      "loss": 3.8987,
+      "step": 1750016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1338682213863937e-05,
+      "loss": 3.9103,
+      "step": 1750528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1330296266353417e-05,
+      "loss": 3.8879,
+      "step": 1751040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1321910318842897e-05,
+      "loss": 3.8974,
+      "step": 1751552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1313524371332377e-05,
+      "loss": 3.902,
+      "step": 1752064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1305138423821857e-05,
+      "loss": 3.8999,
+      "step": 1752576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1296752476311337e-05,
+      "loss": 3.8891,
+      "step": 1753088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1288366528800817e-05,
+      "loss": 3.894,
+      "step": 1753600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1279980581290297e-05,
+      "loss": 3.9075,
+      "step": 1754112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1271594633779777e-05,
+      "loss": 3.8882,
+      "step": 1754624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.1263208686269257e-05,
+      "loss": 3.9048,
+      "step": 1755136
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.9918816089630127,
+      "eval_runtime": 301.5302,
+      "eval_samples_per_second": 1265.515,
+      "eval_steps_per_second": 39.548,
+      "step": 1755360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1254822738758737e-05,
+      "loss": 3.8945,
+      "step": 1755648
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1246436791248217e-05,
+      "loss": 3.8861,
+      "step": 1756160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1238050843737697e-05,
+      "loss": 3.9031,
+      "step": 1756672
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.122968127503091e-05,
+      "loss": 3.8908,
+      "step": 1757184
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.122129532752039e-05,
+      "loss": 3.9058,
+      "step": 1757696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.121290938000987e-05,
+      "loss": 3.8994,
+      "step": 1758208
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.120452343249935e-05,
+      "loss": 3.8912,
+      "step": 1758720
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1196153863792558e-05,
+      "loss": 3.8881,
+      "step": 1759232
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1187767916282038e-05,
+      "loss": 3.8962,
+      "step": 1759744
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1179381968771518e-05,
+      "loss": 3.8989,
+      "step": 1760256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1170996021261e-05,
+      "loss": 3.8914,
+      "step": 1760768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.116262645255421e-05,
+      "loss": 3.8932,
+      "step": 1761280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.115425688384742e-05,
+      "loss": 3.8919,
+      "step": 1761792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.11458709363369e-05,
+      "loss": 3.8885,
+      "step": 1762304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.113748498882638e-05,
+      "loss": 3.8932,
+      "step": 1762816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1129099041315863e-05,
+      "loss": 3.884,
+      "step": 1763328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1120713093805343e-05,
+      "loss": 3.8908,
+      "step": 1763840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1112327146294823e-05,
+      "loss": 3.8935,
+      "step": 1764352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.11039411987843e-05,
+      "loss": 3.8933,
+      "step": 1764864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.109555525127378e-05,
+      "loss": 3.899,
+      "step": 1765376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.108716930376326e-05,
+      "loss": 3.9065,
+      "step": 1765888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.107878335625274e-05,
+      "loss": 3.896,
+      "step": 1766400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.107039740874222e-05,
+      "loss": 3.8964,
+      "step": 1766912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1062011461231703e-05,
+      "loss": 3.8992,
+      "step": 1767424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1053625513721183e-05,
+      "loss": 3.8887,
+      "step": 1767936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.104525594501439e-05,
+      "loss": 3.9026,
+      "step": 1768448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.103686999750387e-05,
+      "loss": 3.8873,
+      "step": 1768960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.102848404999335e-05,
+      "loss": 3.8934,
+      "step": 1769472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.1020098102482828e-05,
+      "loss": 3.889,
+      "step": 1769984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.101171215497231e-05,
+      "loss": 3.8892,
+      "step": 1770496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.100332620746179e-05,
+      "loss": 3.8919,
+      "step": 1771008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0994956638755004e-05,
+      "loss": 3.9002,
+      "step": 1771520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.098657069124448e-05,
+      "loss": 3.8967,
+      "step": 1772032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.097818474373396e-05,
+      "loss": 3.901,
+      "step": 1772544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.096979879622344e-05,
+      "loss": 3.89,
+      "step": 1773056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.096141284871292e-05,
+      "loss": 3.9045,
+      "step": 1773568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0953043280006133e-05,
+      "loss": 3.8983,
+      "step": 1774080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0944657332495613e-05,
+      "loss": 3.889,
+      "step": 1774592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0936271384985093e-05,
+      "loss": 3.8924,
+      "step": 1775104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0927885437474573e-05,
+      "loss": 3.8915,
+      "step": 1775616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0919499489964053e-05,
+      "loss": 3.8914,
+      "step": 1776128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0911113542453533e-05,
+      "loss": 3.9001,
+      "step": 1776640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0902727594943013e-05,
+      "loss": 3.8906,
+      "step": 1777152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0894341647432492e-05,
+      "loss": 3.8931,
+      "step": 1777664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0885972078725705e-05,
+      "loss": 3.9031,
+      "step": 1778176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0877586131215185e-05,
+      "loss": 3.8925,
+      "step": 1778688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0869216562508394e-05,
+      "loss": 3.8962,
+      "step": 1779200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0860830614997874e-05,
+      "loss": 3.8948,
+      "step": 1779712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0852444667487357e-05,
+      "loss": 3.858,
+      "step": 1780224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0844058719976837e-05,
+      "loss": 3.9089,
+      "step": 1780736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0835689151270046e-05,
+      "loss": 3.8899,
+      "step": 1781248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0827303203759526e-05,
+      "loss": 3.8966,
+      "step": 1781760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0818917256249006e-05,
+      "loss": 3.8803,
+      "step": 1782272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0810531308738486e-05,
+      "loss": 3.8753,
+      "step": 1782784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0802145361227966e-05,
+      "loss": 3.8817,
+      "step": 1783296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.079377579252118e-05,
+      "loss": 3.8876,
+      "step": 1783808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0785406223814388e-05,
+      "loss": 3.8886,
+      "step": 1784320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0777020276303868e-05,
+      "loss": 3.8836,
+      "step": 1784832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0768634328793348e-05,
+      "loss": 3.9031,
+      "step": 1785344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.076026476008656e-05,
+      "loss": 3.886,
+      "step": 1785856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.075187881257604e-05,
+      "loss": 3.8831,
+      "step": 1786368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.074349286506552e-05,
+      "loss": 3.8977,
+      "step": 1786880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0735106917555e-05,
+      "loss": 3.8705,
+      "step": 1787392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.072672097004448e-05,
+      "loss": 3.8808,
+      "step": 1787904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0718335022533957e-05,
+      "loss": 3.8899,
+      "step": 1788416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0709949075023437e-05,
+      "loss": 3.901,
+      "step": 1788928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.070156312751292e-05,
+      "loss": 3.8741,
+      "step": 1789440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.06931771800024e-05,
+      "loss": 3.8898,
+      "step": 1789952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.068479123249188e-05,
+      "loss": 3.8728,
+      "step": 1790464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.067640528498136e-05,
+      "loss": 3.8991,
+      "step": 1790976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.066801933747084e-05,
+      "loss": 3.8945,
+      "step": 1791488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0659633389960316e-05,
+      "loss": 3.896,
+      "step": 1792000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0651247442449796e-05,
+      "loss": 3.897,
+      "step": 1792512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.064286149493928e-05,
+      "loss": 3.9022,
+      "step": 1793024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0634491926232492e-05,
+      "loss": 3.9101,
+      "step": 1793536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.062610597872197e-05,
+      "loss": 3.8908,
+      "step": 1794048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.061772003121145e-05,
+      "loss": 3.8943,
+      "step": 1794560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.060933408370093e-05,
+      "loss": 3.8991,
+      "step": 1795072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.060094813619041e-05,
+      "loss": 3.8805,
+      "step": 1795584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.059256218867989e-05,
+      "loss": 3.8924,
+      "step": 1796096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.05841926199731e-05,
+      "loss": 3.8955,
+      "step": 1796608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.057580667246258e-05,
+      "loss": 3.8982,
+      "step": 1797120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.056742072495206e-05,
+      "loss": 3.8992,
+      "step": 1797632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.055905115624527e-05,
+      "loss": 3.8851,
+      "step": 1798144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.055066520873475e-05,
+      "loss": 3.8827,
+      "step": 1798656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0542279261224233e-05,
+      "loss": 3.8885,
+      "step": 1799168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0533893313713713e-05,
+      "loss": 3.9005,
+      "step": 1799680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0525507366203193e-05,
+      "loss": 3.894,
+      "step": 1800192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.0517121418692673e-05,
+      "loss": 3.8867,
+      "step": 1800704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.050873547118215e-05,
+      "loss": 3.8859,
+      "step": 1801216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.050034952367163e-05,
+      "loss": 3.8974,
+      "step": 1801728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0491979954964842e-05,
+      "loss": 3.8888,
+      "step": 1802240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0483594007454322e-05,
+      "loss": 3.888,
+      "step": 1802752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0475208059943802e-05,
+      "loss": 3.9043,
+      "step": 1803264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0466822112433282e-05,
+      "loss": 3.8919,
+      "step": 1803776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0458436164922762e-05,
+      "loss": 3.8944,
+      "step": 1804288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0450050217412242e-05,
+      "loss": 3.883,
+      "step": 1804800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0441664269901722e-05,
+      "loss": 3.8905,
+      "step": 1805312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0433278322391202e-05,
+      "loss": 3.8862,
+      "step": 1805824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0424908753684414e-05,
+      "loss": 3.9029,
+      "step": 1806336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0416539184977623e-05,
+      "loss": 3.895,
+      "step": 1806848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0408153237467103e-05,
+      "loss": 3.899,
+      "step": 1807360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0399767289956583e-05,
+      "loss": 3.8964,
+      "step": 1807872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0391381342446063e-05,
+      "loss": 3.896,
+      "step": 1808384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0382995394935543e-05,
+      "loss": 3.8792,
+      "step": 1808896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0374609447425027e-05,
+      "loss": 3.8845,
+      "step": 1809408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0366223499914503e-05,
+      "loss": 3.8796,
+      "step": 1809920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0357837552403983e-05,
+      "loss": 3.8971,
+      "step": 1810432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0349451604893463e-05,
+      "loss": 3.8819,
+      "step": 1810944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0341082036186675e-05,
+      "loss": 3.8876,
+      "step": 1811456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0332696088676152e-05,
+      "loss": 3.8793,
+      "step": 1811968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0324310141165635e-05,
+      "loss": 3.8888,
+      "step": 1812480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0315924193655115e-05,
+      "loss": 3.8885,
+      "step": 1812992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0307538246144595e-05,
+      "loss": 3.8826,
+      "step": 1813504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0299168677437804e-05,
+      "loss": 3.8988,
+      "step": 1814016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0290782729927284e-05,
+      "loss": 3.8972,
+      "step": 1814528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0282396782416764e-05,
+      "loss": 3.8913,
+      "step": 1815040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0274010834906244e-05,
+      "loss": 3.8953,
+      "step": 1815552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0265641266199457e-05,
+      "loss": 3.8899,
+      "step": 1816064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0257255318688937e-05,
+      "loss": 3.8913,
+      "step": 1816576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0248869371178417e-05,
+      "loss": 3.8934,
+      "step": 1817088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0240483423667897e-05,
+      "loss": 3.8968,
+      "step": 1817600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0232097476157377e-05,
+      "loss": 3.8842,
+      "step": 1818112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0223711528646857e-05,
+      "loss": 3.8952,
+      "step": 1818624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.021534195994007e-05,
+      "loss": 3.8873,
+      "step": 1819136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.020695601242955e-05,
+      "loss": 3.8896,
+      "step": 1819648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.019857006491903e-05,
+      "loss": 3.8827,
+      "step": 1820160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0190184117408505e-05,
+      "loss": 3.8846,
+      "step": 1820672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0181814548701718e-05,
+      "loss": 3.882,
+      "step": 1821184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0173428601191198e-05,
+      "loss": 3.8963,
+      "step": 1821696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.016504265368068e-05,
+      "loss": 3.8906,
+      "step": 1822208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0156656706170158e-05,
+      "loss": 3.8966,
+      "step": 1822720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0148270758659638e-05,
+      "loss": 3.8814,
+      "step": 1823232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0139884811149118e-05,
+      "loss": 3.9025,
+      "step": 1823744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0131498863638598e-05,
+      "loss": 3.8938,
+      "step": 1824256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0123129294931807e-05,
+      "loss": 3.896,
+      "step": 1824768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.011474334742129e-05,
+      "loss": 3.8962,
+      "step": 1825280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.010635739991077e-05,
+      "loss": 3.8919,
+      "step": 1825792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.009797145240025e-05,
+      "loss": 3.891,
+      "step": 1826304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.008960188369346e-05,
+      "loss": 3.903,
+      "step": 1826816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.008121593618294e-05,
+      "loss": 3.8829,
+      "step": 1827328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.007282998867242e-05,
+      "loss": 3.8948,
+      "step": 1827840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.00644440411619e-05,
+      "loss": 3.8945,
+      "step": 1828352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0056058093651382e-05,
+      "loss": 3.8901,
+      "step": 1828864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.0047672146140862e-05,
+      "loss": 3.8866,
+      "step": 1829376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.003930257743407e-05,
+      "loss": 3.8887,
+      "step": 1829888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.003091662992355e-05,
+      "loss": 3.9,
+      "step": 1830400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.002253068241303e-05,
+      "loss": 3.8862,
+      "step": 1830912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.001414473490251e-05,
+      "loss": 3.8974,
+      "step": 1831424
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.989259958267212,
+      "eval_runtime": 299.5619,
+      "eval_samples_per_second": 1273.83,
+      "eval_steps_per_second": 39.808,
+      "step": 1831680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.000575878739199e-05,
+      "loss": 3.8839,
+      "step": 1831936
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.999737283988147e-05,
+      "loss": 3.8844,
+      "step": 1832448
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.998898689237095e-05,
+      "loss": 3.8894,
+      "step": 1832960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.998060094486043e-05,
+      "loss": 3.888,
+      "step": 1833472
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.997221499734991e-05,
+      "loss": 3.9,
+      "step": 1833984
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.996382904983939e-05,
+      "loss": 3.891,
+      "step": 1834496
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.9955443102328868e-05,
+      "loss": 3.8858,
+      "step": 1835008
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.994705715481835e-05,
+      "loss": 3.8837,
+      "step": 1835520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.993867120730783e-05,
+      "loss": 3.8845,
+      "step": 1836032
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.993028525979731e-05,
+      "loss": 3.8971,
+      "step": 1836544
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.992189931228679e-05,
+      "loss": 3.8838,
+      "step": 1837056
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.991351336477627e-05,
+      "loss": 3.8858,
+      "step": 1837568
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.990514379606948e-05,
+      "loss": 3.895,
+      "step": 1838080
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.989675784855896e-05,
+      "loss": 3.8767,
+      "step": 1838592
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.9888371901048443e-05,
+      "loss": 3.8824,
+      "step": 1839104
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.9879985953537923e-05,
+      "loss": 3.8837,
+      "step": 1839616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.98716000060274e-05,
+      "loss": 3.8854,
+      "step": 1840128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.986321405851688e-05,
+      "loss": 3.8873,
+      "step": 1840640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.985482811100636e-05,
+      "loss": 3.8889,
+      "step": 1841152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.984644216349584e-05,
+      "loss": 3.8901,
+      "step": 1841664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.983805621598532e-05,
+      "loss": 3.9033,
+      "step": 1842176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.9829686647278532e-05,
+      "loss": 3.8906,
+      "step": 1842688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.9821300699768012e-05,
+      "loss": 3.894,
+      "step": 1843200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.9812914752257492e-05,
+      "loss": 3.8882,
+      "step": 1843712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.9804528804746972e-05,
+      "loss": 3.887,
+      "step": 1844224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.979615923604018e-05,
+      "loss": 3.8948,
+      "step": 1844736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.978777328852966e-05,
+      "loss": 3.8824,
+      "step": 1845248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.9779387341019144e-05,
+      "loss": 3.8872,
+      "step": 1845760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.9771001393508624e-05,
+      "loss": 3.8781,
+      "step": 1846272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.9762615445998104e-05,
+      "loss": 3.8849,
+      "step": 1846784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.975422949848758e-05,
+      "loss": 3.8811,
+      "step": 1847296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9745859929780793e-05,
+      "loss": 3.8918,
+      "step": 1847808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9737473982270273e-05,
+      "loss": 3.8954,
+      "step": 1848320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9729088034759753e-05,
+      "loss": 3.8955,
+      "step": 1848832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9720702087249233e-05,
+      "loss": 3.8865,
+      "step": 1849344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9712332518542446e-05,
+      "loss": 3.893,
+      "step": 1849856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9703946571031926e-05,
+      "loss": 3.8945,
+      "step": 1850368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9695560623521405e-05,
+      "loss": 3.8829,
+      "step": 1850880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9687174676010882e-05,
+      "loss": 3.8839,
+      "step": 1851392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9678805107304098e-05,
+      "loss": 3.884,
+      "step": 1851904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9670419159793578e-05,
+      "loss": 3.889,
+      "step": 1852416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9662033212283054e-05,
+      "loss": 3.8883,
+      "step": 1852928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9653647264772534e-05,
+      "loss": 3.8913,
+      "step": 1853440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9645261317262014e-05,
+      "loss": 3.8868,
+      "step": 1853952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9636891748555227e-05,
+      "loss": 3.8955,
+      "step": 1854464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9628505801044707e-05,
+      "loss": 3.8909,
+      "step": 1854976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9620119853534187e-05,
+      "loss": 3.8895,
+      "step": 1855488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9611733906023667e-05,
+      "loss": 3.8885,
+      "step": 1856000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.960336433731688e-05,
+      "loss": 3.8528,
+      "step": 1856512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9594978389806356e-05,
+      "loss": 3.9023,
+      "step": 1857024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9586608821099568e-05,
+      "loss": 3.8836,
+      "step": 1857536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.957822287358905e-05,
+      "loss": 3.8904,
+      "step": 1858048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9569836926078528e-05,
+      "loss": 3.8769,
+      "step": 1858560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9561450978568008e-05,
+      "loss": 3.8714,
+      "step": 1859072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9553065031057488e-05,
+      "loss": 3.874,
+      "step": 1859584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.95446954623507e-05,
+      "loss": 3.8791,
+      "step": 1860096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9536309514840177e-05,
+      "loss": 3.8843,
+      "step": 1860608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.952792356732966e-05,
+      "loss": 3.8793,
+      "step": 1861120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.951953761981914e-05,
+      "loss": 3.8933,
+      "step": 1861632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.951115167230862e-05,
+      "loss": 3.8829,
+      "step": 1862144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.950278210360183e-05,
+      "loss": 3.8764,
+      "step": 1862656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.949439615609131e-05,
+      "loss": 3.8899,
+      "step": 1863168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.948601020858079e-05,
+      "loss": 3.868,
+      "step": 1863680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9477640639874002e-05,
+      "loss": 3.873,
+      "step": 1864192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9469254692363482e-05,
+      "loss": 3.8859,
+      "step": 1864704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9460868744852962e-05,
+      "loss": 3.8905,
+      "step": 1865216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9452482797342442e-05,
+      "loss": 3.8731,
+      "step": 1865728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.944409684983192e-05,
+      "loss": 3.8824,
+      "step": 1866240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.94357109023214e-05,
+      "loss": 3.8679,
+      "step": 1866752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.942732495481088e-05,
+      "loss": 3.8898,
+      "step": 1867264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.941893900730036e-05,
+      "loss": 3.8899,
+      "step": 1867776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.941055305978984e-05,
+      "loss": 3.8863,
+      "step": 1868288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.940216711227932e-05,
+      "loss": 3.8925,
+      "step": 1868800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.93937811647688e-05,
+      "loss": 3.8957,
+      "step": 1869312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.938539521725828e-05,
+      "loss": 3.9061,
+      "step": 1869824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.937702564855149e-05,
+      "loss": 3.8845,
+      "step": 1870336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.936863970104097e-05,
+      "loss": 3.8889,
+      "step": 1870848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9360253753530454e-05,
+      "loss": 3.8858,
+      "step": 1871360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9351867806019934e-05,
+      "loss": 3.8825,
+      "step": 1871872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9343481858509414e-05,
+      "loss": 3.8842,
+      "step": 1872384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.933509591099889e-05,
+      "loss": 3.8911,
+      "step": 1872896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.932670996348837e-05,
+      "loss": 3.893,
+      "step": 1873408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.931832401597785e-05,
+      "loss": 3.8984,
+      "step": 1873920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.930993806846733e-05,
+      "loss": 3.8776,
+      "step": 1874432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9301552120956813e-05,
+      "loss": 3.8774,
+      "step": 1874944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9293166173446293e-05,
+      "loss": 3.8804,
+      "step": 1875456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9284796604739502e-05,
+      "loss": 3.8956,
+      "step": 1875968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.927642703603271e-05,
+      "loss": 3.8916,
+      "step": 1876480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.926804108852219e-05,
+      "loss": 3.8781,
+      "step": 1876992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9259655141011675e-05,
+      "loss": 3.881,
+      "step": 1877504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9251269193501155e-05,
+      "loss": 3.892,
+      "step": 1878016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9242899624794364e-05,
+      "loss": 3.8835,
+      "step": 1878528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9234513677283844e-05,
+      "loss": 3.88,
+      "step": 1879040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9226127729773324e-05,
+      "loss": 3.8957,
+      "step": 1879552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9217741782262804e-05,
+      "loss": 3.8898,
+      "step": 1880064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9209372213556016e-05,
+      "loss": 3.8894,
+      "step": 1880576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9200986266045496e-05,
+      "loss": 3.8799,
+      "step": 1881088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9192600318534976e-05,
+      "loss": 3.8823,
+      "step": 1881600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9184214371024456e-05,
+      "loss": 3.8826,
+      "step": 1882112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9175828423513936e-05,
+      "loss": 3.894,
+      "step": 1882624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9167442476003416e-05,
+      "loss": 3.8904,
+      "step": 1883136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9159056528492893e-05,
+      "loss": 3.8878,
+      "step": 1883648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.915068695978611e-05,
+      "loss": 3.8942,
+      "step": 1884160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.914230101227559e-05,
+      "loss": 3.8835,
+      "step": 1884672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.913391506476507e-05,
+      "loss": 3.8841,
+      "step": 1885184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9125529117254545e-05,
+      "loss": 3.8755,
+      "step": 1885696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9117143169744025e-05,
+      "loss": 3.8702,
+      "step": 1886208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9108757222233505e-05,
+      "loss": 3.8934,
+      "step": 1886720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9100371274722985e-05,
+      "loss": 3.8804,
+      "step": 1887232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9091985327212468e-05,
+      "loss": 3.8779,
+      "step": 1887744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9083615758505677e-05,
+      "loss": 3.876,
+      "step": 1888256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9075229810995157e-05,
+      "loss": 3.8766,
+      "step": 1888768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9066843863484637e-05,
+      "loss": 3.8854,
+      "step": 1889280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9058474294777846e-05,
+      "loss": 3.8781,
+      "step": 1889792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.905008834726733e-05,
+      "loss": 3.892,
+      "step": 1890304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.904170239975681e-05,
+      "loss": 3.8901,
+      "step": 1890816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.903331645224629e-05,
+      "loss": 3.8883,
+      "step": 1891328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.902493050473577e-05,
+      "loss": 3.8888,
+      "step": 1891840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.901656093602898e-05,
+      "loss": 3.8825,
+      "step": 1892352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.900817498851846e-05,
+      "loss": 3.8856,
+      "step": 1892864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.899978904100794e-05,
+      "loss": 3.8886,
+      "step": 1893376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8991403093497422e-05,
+      "loss": 3.8894,
+      "step": 1893888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8983017145986902e-05,
+      "loss": 3.8828,
+      "step": 1894400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.897464757728011e-05,
+      "loss": 3.89,
+      "step": 1894912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.896627800857332e-05,
+      "loss": 3.8835,
+      "step": 1895424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.89578920610628e-05,
+      "loss": 3.8824,
+      "step": 1895936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8949506113552283e-05,
+      "loss": 3.8739,
+      "step": 1896448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8941120166041763e-05,
+      "loss": 3.8792,
+      "step": 1896960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8932734218531243e-05,
+      "loss": 3.877,
+      "step": 1897472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8924348271020723e-05,
+      "loss": 3.8922,
+      "step": 1897984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.89159623235102e-05,
+      "loss": 3.8818,
+      "step": 1898496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.890757637599968e-05,
+      "loss": 3.8937,
+      "step": 1899008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.889919042848916e-05,
+      "loss": 3.8736,
+      "step": 1899520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8890820859782372e-05,
+      "loss": 3.8955,
+      "step": 1900032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8882434912271852e-05,
+      "loss": 3.8899,
+      "step": 1900544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8874048964761332e-05,
+      "loss": 3.8901,
+      "step": 1901056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8865663017250812e-05,
+      "loss": 3.8907,
+      "step": 1901568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.885729344854402e-05,
+      "loss": 3.8868,
+      "step": 1902080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.88489075010335e-05,
+      "loss": 3.8878,
+      "step": 1902592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8840521553522984e-05,
+      "loss": 3.8976,
+      "step": 1903104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8832135606012464e-05,
+      "loss": 3.8759,
+      "step": 1903616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8823749658501944e-05,
+      "loss": 3.8888,
+      "step": 1904128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8815363710991424e-05,
+      "loss": 3.8868,
+      "step": 1904640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8806977763480904e-05,
+      "loss": 3.8851,
+      "step": 1905152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8798608194774113e-05,
+      "loss": 3.8835,
+      "step": 1905664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8790222247263593e-05,
+      "loss": 3.8801,
+      "step": 1906176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8781836299753077e-05,
+      "loss": 3.8933,
+      "step": 1906688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8773466731046286e-05,
+      "loss": 3.8823,
+      "step": 1907200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.8765080783535766e-05,
+      "loss": 3.8909,
+      "step": 1907712
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.9871957302093506,
+      "eval_runtime": 291.3606,
+      "eval_samples_per_second": 1309.686,
+      "eval_steps_per_second": 40.929,
+      "step": 1908000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8756694836025246e-05,
+      "loss": 3.8762,
+      "step": 1908224
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8748308888514726e-05,
+      "loss": 3.8811,
+      "step": 1908736
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8739922941004202e-05,
+      "loss": 3.8821,
+      "step": 1909248
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8731536993493686e-05,
+      "loss": 3.8842,
+      "step": 1909760
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8723151045983165e-05,
+      "loss": 3.8975,
+      "step": 1910272
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8714765098472645e-05,
+      "loss": 3.8873,
+      "step": 1910784
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8706395529765855e-05,
+      "loss": 3.8812,
+      "step": 1911296
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8698009582255334e-05,
+      "loss": 3.8803,
+      "step": 1911808
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8689623634744814e-05,
+      "loss": 3.877,
+      "step": 1912320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8681237687234294e-05,
+      "loss": 3.892,
+      "step": 1912832
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8672851739723778e-05,
+      "loss": 3.8807,
+      "step": 1913344
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8664465792213258e-05,
+      "loss": 3.8828,
+      "step": 1913856
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8656096223506467e-05,
+      "loss": 3.8892,
+      "step": 1914368
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8647710275995947e-05,
+      "loss": 3.873,
+      "step": 1914880
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8639324328485427e-05,
+      "loss": 3.8772,
+      "step": 1915392
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8630938380974907e-05,
+      "loss": 3.8757,
+      "step": 1915904
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8622552433464387e-05,
+      "loss": 3.8812,
+      "step": 1916416
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8614166485953867e-05,
+      "loss": 3.8771,
+      "step": 1916928
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8605780538443347e-05,
+      "loss": 3.8858,
+      "step": 1917440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.859741096973656e-05,
+      "loss": 3.8853,
+      "step": 1917952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8589025022226036e-05,
+      "loss": 3.8959,
+      "step": 1918464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8580639074715516e-05,
+      "loss": 3.8828,
+      "step": 1918976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8572253127204995e-05,
+      "loss": 3.8899,
+      "step": 1919488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.856386717969448e-05,
+      "loss": 3.8809,
+      "step": 1920000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.855548123218396e-05,
+      "loss": 3.8849,
+      "step": 1920512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.854709528467344e-05,
+      "loss": 3.8879,
+      "step": 1921024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8538709337162915e-05,
+      "loss": 3.8748,
+      "step": 1921536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8530323389652395e-05,
+      "loss": 3.8865,
+      "step": 1922048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8521937442141875e-05,
+      "loss": 3.8723,
+      "step": 1922560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8513551494631355e-05,
+      "loss": 3.8794,
+      "step": 1923072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8505181925924568e-05,
+      "loss": 3.8761,
+      "step": 1923584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8496795978414048e-05,
+      "loss": 3.8837,
+      "step": 1924096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8488410030903528e-05,
+      "loss": 3.8904,
+      "step": 1924608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8480024083393008e-05,
+      "loss": 3.8916,
+      "step": 1925120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8471654514686217e-05,
+      "loss": 3.883,
+      "step": 1925632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.84632685671757e-05,
+      "loss": 3.8806,
+      "step": 1926144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8454898998468912e-05,
+      "loss": 3.8944,
+      "step": 1926656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.844651305095839e-05,
+      "loss": 3.8782,
+      "step": 1927168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.843812710344787e-05,
+      "loss": 3.8797,
+      "step": 1927680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.842974115593735e-05,
+      "loss": 3.8796,
+      "step": 1928192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.842135520842683e-05,
+      "loss": 3.8858,
+      "step": 1928704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.841296926091631e-05,
+      "loss": 3.8818,
+      "step": 1929216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.840459969220952e-05,
+      "loss": 3.8837,
+      "step": 1929728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8396213744699e-05,
+      "loss": 3.8832,
+      "step": 1930240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.838782779718848e-05,
+      "loss": 3.8863,
+      "step": 1930752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.837944184967796e-05,
+      "loss": 3.886,
+      "step": 1931264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.837105590216744e-05,
+      "loss": 3.8827,
+      "step": 1931776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.836266995465692e-05,
+      "loss": 3.8849,
+      "step": 1932288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8354300385950134e-05,
+      "loss": 3.8486,
+      "step": 1932800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8345914438439614e-05,
+      "loss": 3.8939,
+      "step": 1933312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8337528490929093e-05,
+      "loss": 3.8764,
+      "step": 1933824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.832914254341857e-05,
+      "loss": 3.8881,
+      "step": 1934336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.832075659590805e-05,
+      "loss": 3.8731,
+      "step": 1934848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.831237064839753e-05,
+      "loss": 3.8652,
+      "step": 1935360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.830398470088701e-05,
+      "loss": 3.867,
+      "step": 1935872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8295598753376493e-05,
+      "loss": 3.8761,
+      "step": 1936384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8287212805865973e-05,
+      "loss": 3.8838,
+      "step": 1936896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8278826858355453e-05,
+      "loss": 3.87,
+      "step": 1937408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.827044091084493e-05,
+      "loss": 3.8895,
+      "step": 1937920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8262071342138142e-05,
+      "loss": 3.8813,
+      "step": 1938432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8253685394627622e-05,
+      "loss": 3.8655,
+      "step": 1938944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8245299447117102e-05,
+      "loss": 3.8853,
+      "step": 1939456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8236913499606582e-05,
+      "loss": 3.8676,
+      "step": 1939968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8228527552096062e-05,
+      "loss": 3.866,
+      "step": 1940480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8220141604585542e-05,
+      "loss": 3.8802,
+      "step": 1940992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8211755657075022e-05,
+      "loss": 3.8862,
+      "step": 1941504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8203369709564502e-05,
+      "loss": 3.8712,
+      "step": 1942016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.819500014085771e-05,
+      "loss": 3.875,
+      "step": 1942528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8186614193347194e-05,
+      "loss": 3.8646,
+      "step": 1943040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8178228245836674e-05,
+      "loss": 3.8814,
+      "step": 1943552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8169842298326154e-05,
+      "loss": 3.8851,
+      "step": 1944064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8161472729619363e-05,
+      "loss": 3.8821,
+      "step": 1944576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8153103160912572e-05,
+      "loss": 3.885,
+      "step": 1945088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8144717213402056e-05,
+      "loss": 3.8911,
+      "step": 1945600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8136331265891536e-05,
+      "loss": 3.9005,
+      "step": 1946112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8127945318381016e-05,
+      "loss": 3.8833,
+      "step": 1946624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8119559370870496e-05,
+      "loss": 3.8862,
+      "step": 1947136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8111173423359976e-05,
+      "loss": 3.8804,
+      "step": 1947648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8102787475849456e-05,
+      "loss": 3.8806,
+      "step": 1948160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8094434285946397e-05,
+      "loss": 3.8747,
+      "step": 1948672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8086048338435877e-05,
+      "loss": 3.8881,
+      "step": 1949184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8077662390925357e-05,
+      "loss": 3.8844,
+      "step": 1949696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.806929282221857e-05,
+      "loss": 3.8956,
+      "step": 1950208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8060906874708046e-05,
+      "loss": 3.8704,
+      "step": 1950720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.8052520927197526e-05,
+      "loss": 3.8724,
+      "step": 1951232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.804413497968701e-05,
+      "loss": 3.8786,
+      "step": 1951744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.803574903217649e-05,
+      "loss": 3.8868,
+      "step": 1952256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.802736308466597e-05,
+      "loss": 3.8867,
+      "step": 1952768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.801897713715545e-05,
+      "loss": 3.877,
+      "step": 1953280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.801059118964493e-05,
+      "loss": 3.8754,
+      "step": 1953792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.800222162093814e-05,
+      "loss": 3.8846,
+      "step": 1954304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.799383567342762e-05,
+      "loss": 3.8803,
+      "step": 1954816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.79854497259171e-05,
+      "loss": 3.8735,
+      "step": 1955328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.797706377840658e-05,
+      "loss": 3.8914,
+      "step": 1955840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7968677830896058e-05,
+      "loss": 3.8851,
+      "step": 1956352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7960291883385538e-05,
+      "loss": 3.8848,
+      "step": 1956864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7951905935875018e-05,
+      "loss": 3.8739,
+      "step": 1957376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7943519988364498e-05,
+      "loss": 3.8766,
+      "step": 1957888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.793515041965771e-05,
+      "loss": 3.8761,
+      "step": 1958400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.792676447214719e-05,
+      "loss": 3.8892,
+      "step": 1958912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.791837852463667e-05,
+      "loss": 3.8926,
+      "step": 1959424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.790999257712615e-05,
+      "loss": 3.8795,
+      "step": 1959936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.790160662961563e-05,
+      "loss": 3.8888,
+      "step": 1960448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.789322068210511e-05,
+      "loss": 3.8815,
+      "step": 1960960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7884834734594587e-05,
+      "loss": 3.882,
+      "step": 1961472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.787644878708407e-05,
+      "loss": 3.8697,
+      "step": 1961984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7868079218377283e-05,
+      "loss": 3.8682,
+      "step": 1962496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7859709649670492e-05,
+      "loss": 3.8811,
+      "step": 1963008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7851323702159972e-05,
+      "loss": 3.8787,
+      "step": 1963520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7842937754649452e-05,
+      "loss": 3.8704,
+      "step": 1964032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7834551807138932e-05,
+      "loss": 3.8714,
+      "step": 1964544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.782616585962841e-05,
+      "loss": 3.8717,
+      "step": 1965056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7817796290921624e-05,
+      "loss": 3.8794,
+      "step": 1965568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7809410343411104e-05,
+      "loss": 3.8718,
+      "step": 1966080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7801024395900584e-05,
+      "loss": 3.8859,
+      "step": 1966592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.779263844839006e-05,
+      "loss": 3.8849,
+      "step": 1967104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7784268879683273e-05,
+      "loss": 3.8848,
+      "step": 1967616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7775882932172756e-05,
+      "loss": 3.8783,
+      "step": 1968128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7767496984662236e-05,
+      "loss": 3.8829,
+      "step": 1968640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7759111037151713e-05,
+      "loss": 3.8793,
+      "step": 1969152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7750725089641193e-05,
+      "loss": 3.8872,
+      "step": 1969664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7742339142130673e-05,
+      "loss": 3.886,
+      "step": 1970176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7733953194620153e-05,
+      "loss": 3.8765,
+      "step": 1970688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7725567247109633e-05,
+      "loss": 3.8837,
+      "step": 1971200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7717197678402845e-05,
+      "loss": 3.8844,
+      "step": 1971712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7708811730892325e-05,
+      "loss": 3.8734,
+      "step": 1972224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7700442162185534e-05,
+      "loss": 3.8738,
+      "step": 1972736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7692056214675014e-05,
+      "loss": 3.8748,
+      "step": 1973248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7683670267164494e-05,
+      "loss": 3.8689,
+      "step": 1973760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7675284319653974e-05,
+      "loss": 3.8889,
+      "step": 1974272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7666898372143458e-05,
+      "loss": 3.8768,
+      "step": 1974784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.76585451822404e-05,
+      "loss": 3.8882,
+      "step": 1975296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.765015923472988e-05,
+      "loss": 3.8688,
+      "step": 1975808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7641773287219356e-05,
+      "loss": 3.8879,
+      "step": 1976320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7633387339708836e-05,
+      "loss": 3.8864,
+      "step": 1976832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.762500139219832e-05,
+      "loss": 3.8868,
+      "step": 1977344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.76166154446878e-05,
+      "loss": 3.8843,
+      "step": 1977856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.760822949717728e-05,
+      "loss": 3.8778,
+      "step": 1978368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7599859928470488e-05,
+      "loss": 3.8855,
+      "step": 1978880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7591473980959968e-05,
+      "loss": 3.8901,
+      "step": 1979392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7583088033449448e-05,
+      "loss": 3.8752,
+      "step": 1979904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7574702085938928e-05,
+      "loss": 3.8842,
+      "step": 1980416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.756631613842841e-05,
+      "loss": 3.8831,
+      "step": 1980928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.755793019091789e-05,
+      "loss": 3.8782,
+      "step": 1981440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7549544243407368e-05,
+      "loss": 3.8781,
+      "step": 1981952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7541158295896848e-05,
+      "loss": 3.8758,
+      "step": 1982464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.753280510599379e-05,
+      "loss": 3.8863,
+      "step": 1982976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7524419158483273e-05,
+      "loss": 3.8796,
+      "step": 1983488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.7516033210972753e-05,
+      "loss": 3.8852,
+      "step": 1984000
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.9859120845794678,
+      "eval_runtime": 290.8286,
+      "eval_samples_per_second": 1312.082,
+      "eval_steps_per_second": 41.004,
+      "step": 1984320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7507647263462233e-05,
+      "loss": 3.8677,
+      "step": 1984512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7499261315951713e-05,
+      "loss": 3.8767,
+      "step": 1985024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.749087536844119e-05,
+      "loss": 3.8754,
+      "step": 1985536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.748248942093067e-05,
+      "loss": 3.8796,
+      "step": 1986048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.747410347342015e-05,
+      "loss": 3.8934,
+      "step": 1986560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.746571752590963e-05,
+      "loss": 3.8847,
+      "step": 1987072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7457331578399112e-05,
+      "loss": 3.875,
+      "step": 1987584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7448945630888592e-05,
+      "loss": 3.8748,
+      "step": 1988096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.74405760621818e-05,
+      "loss": 3.873,
+      "step": 1988608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.743219011467128e-05,
+      "loss": 3.8817,
+      "step": 1989120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.742380416716076e-05,
+      "loss": 3.877,
+      "step": 1989632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.741541821965024e-05,
+      "loss": 3.8808,
+      "step": 1990144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.740703227213972e-05,
+      "loss": 3.8845,
+      "step": 1990656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.73986463246292e-05,
+      "loss": 3.8638,
+      "step": 1991168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.739026037711868e-05,
+      "loss": 3.8727,
+      "step": 1991680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.738187442960816e-05,
+      "loss": 3.8674,
+      "step": 1992192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.737348848209764e-05,
+      "loss": 3.8822,
+      "step": 1992704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.736510253458712e-05,
+      "loss": 3.874,
+      "step": 1993216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.73567165870766e-05,
+      "loss": 3.8789,
+      "step": 1993728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.734833063956608e-05,
+      "loss": 3.8786,
+      "step": 1994240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7339961070859293e-05,
+      "loss": 3.8908,
+      "step": 1994752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7331575123348773e-05,
+      "loss": 3.8772,
+      "step": 1995264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.732318917583825e-05,
+      "loss": 3.8868,
+      "step": 1995776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.731480322832773e-05,
+      "loss": 3.8768,
+      "step": 1996288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7306433659620942e-05,
+      "loss": 3.8792,
+      "step": 1996800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7298047712110426e-05,
+      "loss": 3.8826,
+      "step": 1997312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7289661764599902e-05,
+      "loss": 3.8679,
+      "step": 1997824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7281292195893115e-05,
+      "loss": 3.8776,
+      "step": 1998336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7272906248382595e-05,
+      "loss": 3.8693,
+      "step": 1998848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.7264520300872075e-05,
+      "loss": 3.8722,
+      "step": 1999360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.725613435336155e-05,
+      "loss": 3.8741,
+      "step": 1999872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7247748405851035e-05,
+      "loss": 3.8805,
+      "step": 2000384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7239362458340514e-05,
+      "loss": 3.88,
+      "step": 2000896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7230976510829994e-05,
+      "loss": 3.89,
+      "step": 2001408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7222590563319474e-05,
+      "loss": 3.876,
+      "step": 2001920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7214204615808954e-05,
+      "loss": 3.8783,
+      "step": 2002432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.720581866829843e-05,
+      "loss": 3.8912,
+      "step": 2002944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7197449099591643e-05,
+      "loss": 3.8734,
+      "step": 2003456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7189063152081127e-05,
+      "loss": 3.8734,
+      "step": 2003968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7180677204570607e-05,
+      "loss": 3.8771,
+      "step": 2004480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7172291257060083e-05,
+      "loss": 3.875,
+      "step": 2004992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7163921688353296e-05,
+      "loss": 3.8795,
+      "step": 2005504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7155535740842776e-05,
+      "loss": 3.8768,
+      "step": 2006016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7147166172135988e-05,
+      "loss": 3.884,
+      "step": 2006528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7138780224625468e-05,
+      "loss": 3.8774,
+      "step": 2007040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7130394277114948e-05,
+      "loss": 3.8838,
+      "step": 2007552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7122008329604428e-05,
+      "loss": 3.8784,
+      "step": 2008064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7113622382093905e-05,
+      "loss": 3.8826,
+      "step": 2008576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7105252813387117e-05,
+      "loss": 3.8446,
+      "step": 2009088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7096866865876597e-05,
+      "loss": 3.8888,
+      "step": 2009600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.708848091836608e-05,
+      "loss": 3.8686,
+      "step": 2010112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7080094970855557e-05,
+      "loss": 3.8837,
+      "step": 2010624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7071709023345037e-05,
+      "loss": 3.8668,
+      "step": 2011136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7063323075834517e-05,
+      "loss": 3.8629,
+      "step": 2011648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7054937128323997e-05,
+      "loss": 3.8628,
+      "step": 2012160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.7046567559617206e-05,
+      "loss": 3.8695,
+      "step": 2012672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.703818161210669e-05,
+      "loss": 3.8806,
+      "step": 2013184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.702979566459617e-05,
+      "loss": 3.8646,
+      "step": 2013696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.702140971708565e-05,
+      "loss": 3.8851,
+      "step": 2014208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.701302376957513e-05,
+      "loss": 3.878,
+      "step": 2014720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.700463782206461e-05,
+      "loss": 3.8586,
+      "step": 2015232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6996251874554086e-05,
+      "loss": 3.8784,
+      "step": 2015744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6987865927043566e-05,
+      "loss": 3.8597,
+      "step": 2016256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6979479979533046e-05,
+      "loss": 3.8613,
+      "step": 2016768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.697111041082626e-05,
+      "loss": 3.8763,
+      "step": 2017280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6962724463315738e-05,
+      "loss": 3.8795,
+      "step": 2017792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.695435489460895e-05,
+      "loss": 3.8621,
+      "step": 2018304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.694596894709843e-05,
+      "loss": 3.8752,
+      "step": 2018816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6937582999587907e-05,
+      "loss": 3.8622,
+      "step": 2019328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.692919705207739e-05,
+      "loss": 3.8731,
+      "step": 2019840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.692081110456687e-05,
+      "loss": 3.8798,
+      "step": 2020352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.691242515705635e-05,
+      "loss": 3.8785,
+      "step": 2020864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.690403920954583e-05,
+      "loss": 3.8832,
+      "step": 2021376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.689565326203531e-05,
+      "loss": 3.8819,
+      "step": 2021888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.688726731452479e-05,
+      "loss": 3.8982,
+      "step": 2022400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6878897745818e-05,
+      "loss": 3.8798,
+      "step": 2022912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6870511798307483e-05,
+      "loss": 3.8842,
+      "step": 2023424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6862125850796963e-05,
+      "loss": 3.8706,
+      "step": 2023936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6853739903286443e-05,
+      "loss": 3.8757,
+      "step": 2024448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.684535395577592e-05,
+      "loss": 3.8729,
+      "step": 2024960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.683698438706913e-05,
+      "loss": 3.8802,
+      "step": 2025472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.682859843955861e-05,
+      "loss": 3.882,
+      "step": 2025984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.682021249204809e-05,
+      "loss": 3.8876,
+      "step": 2026496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6811842923341304e-05,
+      "loss": 3.8659,
+      "step": 2027008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6803456975830784e-05,
+      "loss": 3.8707,
+      "step": 2027520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6795071028320264e-05,
+      "loss": 3.8708,
+      "step": 2028032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.678668508080974e-05,
+      "loss": 3.8815,
+      "step": 2028544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.677829913329922e-05,
+      "loss": 3.8786,
+      "step": 2029056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6769929564592436e-05,
+      "loss": 3.874,
+      "step": 2029568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6761543617081916e-05,
+      "loss": 3.8735,
+      "step": 2030080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6753157669571393e-05,
+      "loss": 3.8775,
+      "step": 2030592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6744771722060873e-05,
+      "loss": 3.8742,
+      "step": 2031104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6736385774550353e-05,
+      "loss": 3.8697,
+      "step": 2031616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6727999827039833e-05,
+      "loss": 3.8833,
+      "step": 2032128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6719613879529313e-05,
+      "loss": 3.8827,
+      "step": 2032640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6711227932018793e-05,
+      "loss": 3.8815,
+      "step": 2033152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6702841984508273e-05,
+      "loss": 3.8673,
+      "step": 2033664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6694472415801485e-05,
+      "loss": 3.8703,
+      "step": 2034176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6686086468290965e-05,
+      "loss": 3.8744,
+      "step": 2034688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6677700520780445e-05,
+      "loss": 3.882,
+      "step": 2035200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.666931457326992e-05,
+      "loss": 3.8876,
+      "step": 2035712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6660928625759405e-05,
+      "loss": 3.872,
+      "step": 2036224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6652542678248885e-05,
+      "loss": 3.8841,
+      "step": 2036736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6644173109542097e-05,
+      "loss": 3.8782,
+      "step": 2037248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6635787162031574e-05,
+      "loss": 3.8757,
+      "step": 2037760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6627401214521054e-05,
+      "loss": 3.8655,
+      "step": 2038272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6619015267010534e-05,
+      "loss": 3.8638,
+      "step": 2038784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6610629319500014e-05,
+      "loss": 3.8737,
+      "step": 2039296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6602243371989497e-05,
+      "loss": 3.876,
+      "step": 2039808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6593857424478977e-05,
+      "loss": 3.8674,
+      "step": 2040320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6585471476968454e-05,
+      "loss": 3.8675,
+      "step": 2040832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6577085529457934e-05,
+      "loss": 3.8613,
+      "step": 2041344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6568732339554875e-05,
+      "loss": 3.8767,
+      "step": 2041856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.656036277084809e-05,
+      "loss": 3.8724,
+      "step": 2042368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.655197682333757e-05,
+      "loss": 3.8779,
+      "step": 2042880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6543590875827048e-05,
+      "loss": 3.8823,
+      "step": 2043392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6535204928316528e-05,
+      "loss": 3.8781,
+      "step": 2043904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6526818980806007e-05,
+      "loss": 3.8743,
+      "step": 2044416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6518433033295487e-05,
+      "loss": 3.882,
+      "step": 2044928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.65100634645887e-05,
+      "loss": 3.8756,
+      "step": 2045440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.650167751707818e-05,
+      "loss": 3.8845,
+      "step": 2045952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.649329156956766e-05,
+      "loss": 3.8765,
+      "step": 2046464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.648490562205714e-05,
+      "loss": 3.8747,
+      "step": 2046976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.647653605335035e-05,
+      "loss": 3.8763,
+      "step": 2047488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.646815010583983e-05,
+      "loss": 3.8789,
+      "step": 2048000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6459764158329312e-05,
+      "loss": 3.8697,
+      "step": 2048512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6451378210818792e-05,
+      "loss": 3.8699,
+      "step": 2049024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6442992263308272e-05,
+      "loss": 3.8676,
+      "step": 2049536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6434606315797752e-05,
+      "loss": 3.8673,
+      "step": 2050048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.642622036828723e-05,
+      "loss": 3.8831,
+      "step": 2050560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.641783442077671e-05,
+      "loss": 3.8722,
+      "step": 2051072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.640944847326619e-05,
+      "loss": 3.8836,
+      "step": 2051584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.64010789045594e-05,
+      "loss": 3.8675,
+      "step": 2052096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6392709335852613e-05,
+      "loss": 3.8825,
+      "step": 2052608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6384323388342093e-05,
+      "loss": 3.8798,
+      "step": 2053120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6375937440831573e-05,
+      "loss": 3.8787,
+      "step": 2053632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.636755149332105e-05,
+      "loss": 3.8773,
+      "step": 2054144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.635916554581053e-05,
+      "loss": 3.8742,
+      "step": 2054656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6350779598300013e-05,
+      "loss": 3.8857,
+      "step": 2055168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6342393650789493e-05,
+      "loss": 3.8875,
+      "step": 2055680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6334007703278973e-05,
+      "loss": 3.868,
+      "step": 2056192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6325638134572182e-05,
+      "loss": 3.8766,
+      "step": 2056704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6317252187061662e-05,
+      "loss": 3.8801,
+      "step": 2057216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6308866239551142e-05,
+      "loss": 3.8736,
+      "step": 2057728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6300480292040622e-05,
+      "loss": 3.8735,
+      "step": 2058240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6292094344530105e-05,
+      "loss": 3.8736,
+      "step": 2058752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6283708397019582e-05,
+      "loss": 3.8795,
+      "step": 2059264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6275322449509062e-05,
+      "loss": 3.8775,
+      "step": 2059776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.6266952880802274e-05,
+      "loss": 3.8798,
+      "step": 2060288
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.984645366668701,
+      "eval_runtime": 291.7343,
+      "eval_samples_per_second": 1308.009,
+      "eval_steps_per_second": 40.876,
+      "step": 2060640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6258566933291754e-05,
+      "loss": 3.8665,
+      "step": 2060800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.625018098578123e-05,
+      "loss": 3.8694,
+      "step": 2061312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6241795038270714e-05,
+      "loss": 3.8717,
+      "step": 2061824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6233409090760194e-05,
+      "loss": 3.8762,
+      "step": 2062336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6225039522053403e-05,
+      "loss": 3.8861,
+      "step": 2062848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6216653574542883e-05,
+      "loss": 3.8817,
+      "step": 2063360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6208267627032363e-05,
+      "loss": 3.8717,
+      "step": 2063872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6199898058325576e-05,
+      "loss": 3.8737,
+      "step": 2064384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6191512110815056e-05,
+      "loss": 3.8661,
+      "step": 2064896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6183126163304536e-05,
+      "loss": 3.8731,
+      "step": 2065408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6174740215794016e-05,
+      "loss": 3.8782,
+      "step": 2065920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6166354268283496e-05,
+      "loss": 3.8743,
+      "step": 2066432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6157968320772976e-05,
+      "loss": 3.8779,
+      "step": 2066944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6149582373262456e-05,
+      "loss": 3.8605,
+      "step": 2067456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6141196425751935e-05,
+      "loss": 3.871,
+      "step": 2067968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6132826857045148e-05,
+      "loss": 3.8594,
+      "step": 2068480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6124457288338357e-05,
+      "loss": 3.8777,
+      "step": 2068992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6116071340827837e-05,
+      "loss": 3.869,
+      "step": 2069504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6107685393317317e-05,
+      "loss": 3.8724,
+      "step": 2070016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6099299445806797e-05,
+      "loss": 3.8746,
+      "step": 2070528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6090913498296277e-05,
+      "loss": 3.8904,
+      "step": 2071040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.608252755078576e-05,
+      "loss": 3.8696,
+      "step": 2071552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6074141603275237e-05,
+      "loss": 3.8843,
+      "step": 2072064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6065755655764717e-05,
+      "loss": 3.872,
+      "step": 2072576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6057369708254197e-05,
+      "loss": 3.8762,
+      "step": 2073088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6048983760743677e-05,
+      "loss": 3.8808,
+      "step": 2073600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.6040614192036886e-05,
+      "loss": 3.8633,
+      "step": 2074112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.603222824452637e-05,
+      "loss": 3.873,
+      "step": 2074624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.602384229701585e-05,
+      "loss": 3.8676,
+      "step": 2075136
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.601545634950533e-05,
+      "loss": 3.8665,
+      "step": 2075648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.600707040199481e-05,
+      "loss": 3.8713,
+      "step": 2076160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5998700833288018e-05,
+      "loss": 3.8763,
+      "step": 2076672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5990314885777498e-05,
+      "loss": 3.8703,
+      "step": 2077184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5981928938266978e-05,
+      "loss": 3.8903,
+      "step": 2077696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.597355936956019e-05,
+      "loss": 3.8734,
+      "step": 2078208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.596517342204967e-05,
+      "loss": 3.8714,
+      "step": 2078720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.595678747453915e-05,
+      "loss": 3.8838,
+      "step": 2079232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.594840152702863e-05,
+      "loss": 3.8669,
+      "step": 2079744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.594003195832184e-05,
+      "loss": 3.8681,
+      "step": 2080256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5931646010811323e-05,
+      "loss": 3.8762,
+      "step": 2080768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5923260063300803e-05,
+      "loss": 3.8684,
+      "step": 2081280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5914874115790283e-05,
+      "loss": 3.8747,
+      "step": 2081792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5906488168279763e-05,
+      "loss": 3.8777,
+      "step": 2082304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.589810222076924e-05,
+      "loss": 3.8732,
+      "step": 2082816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.588971627325872e-05,
+      "loss": 3.8773,
+      "step": 2083328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.58813303257482e-05,
+      "loss": 3.8767,
+      "step": 2083840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.587294437823768e-05,
+      "loss": 3.8723,
+      "step": 2084352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5864558430727162e-05,
+      "loss": 3.8784,
+      "step": 2084864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.585618886202037e-05,
+      "loss": 3.846,
+      "step": 2085376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.584780291450985e-05,
+      "loss": 3.8784,
+      "step": 2085888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.583941696699933e-05,
+      "loss": 3.8687,
+      "step": 2086400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.583103101948881e-05,
+      "loss": 3.8789,
+      "step": 2086912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.582264507197829e-05,
+      "loss": 3.8645,
+      "step": 2087424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.581425912446777e-05,
+      "loss": 3.8556,
+      "step": 2087936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5805889555760984e-05,
+      "loss": 3.8611,
+      "step": 2088448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5797503608250464e-05,
+      "loss": 3.8652,
+      "step": 2088960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5789117660739944e-05,
+      "loss": 3.8752,
+      "step": 2089472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.578073171322942e-05,
+      "loss": 3.8568,
+      "step": 2089984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.57723457657189e-05,
+      "loss": 3.8843,
+      "step": 2090496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5763959818208384e-05,
+      "loss": 3.876,
+      "step": 2091008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5755573870697864e-05,
+      "loss": 3.8506,
+      "step": 2091520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5747187923187343e-05,
+      "loss": 3.8747,
+      "step": 2092032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5738818354480553e-05,
+      "loss": 3.8593,
+      "step": 2092544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5730432406970033e-05,
+      "loss": 3.8562,
+      "step": 2093056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5722046459459512e-05,
+      "loss": 3.8669,
+      "step": 2093568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5713660511948992e-05,
+      "loss": 3.8792,
+      "step": 2094080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5705274564438476e-05,
+      "loss": 3.856,
+      "step": 2094592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5696904995731685e-05,
+      "loss": 3.8731,
+      "step": 2095104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5688519048221165e-05,
+      "loss": 3.8539,
+      "step": 2095616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5680133100710645e-05,
+      "loss": 3.8689,
+      "step": 2096128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5671747153200125e-05,
+      "loss": 3.8739,
+      "step": 2096640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.566339396329707e-05,
+      "loss": 3.8781,
+      "step": 2097152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5655008015786546e-05,
+      "loss": 3.8766,
+      "step": 2097664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5646622068276026e-05,
+      "loss": 3.8784,
+      "step": 2098176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5638236120765506e-05,
+      "loss": 3.8931,
+      "step": 2098688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5629850173254986e-05,
+      "loss": 3.8745,
+      "step": 2099200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.56214806045482e-05,
+      "loss": 3.8777,
+      "step": 2099712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.561309465703768e-05,
+      "loss": 3.8646,
+      "step": 2100224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.560470870952716e-05,
+      "loss": 3.8709,
+      "step": 2100736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.559632276201664e-05,
+      "loss": 3.8685,
+      "step": 2101248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.558793681450612e-05,
+      "loss": 3.873,
+      "step": 2101760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.557958362460306e-05,
+      "loss": 3.8807,
+      "step": 2102272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.557119767709254e-05,
+      "loss": 3.8817,
+      "step": 2102784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.556281172958202e-05,
+      "loss": 3.8645,
+      "step": 2103296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.55544257820715e-05,
+      "loss": 3.8645,
+      "step": 2103808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.554603983456098e-05,
+      "loss": 3.8628,
+      "step": 2104320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.553767026585419e-05,
+      "loss": 3.8834,
+      "step": 2104832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.552928431834367e-05,
+      "loss": 3.8726,
+      "step": 2105344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5520898370833152e-05,
+      "loss": 3.8648,
+      "step": 2105856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5512512423322632e-05,
+      "loss": 3.8754,
+      "step": 2106368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.550414285461584e-05,
+      "loss": 3.8718,
+      "step": 2106880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.549575690710532e-05,
+      "loss": 3.8691,
+      "step": 2107392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.54873709595948e-05,
+      "loss": 3.8683,
+      "step": 2107904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.547898501208428e-05,
+      "loss": 3.8798,
+      "step": 2108416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.547059906457376e-05,
+      "loss": 3.8756,
+      "step": 2108928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.546221311706324e-05,
+      "loss": 3.8803,
+      "step": 2109440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5453827169552725e-05,
+      "loss": 3.8638,
+      "step": 2109952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.54454412220422e-05,
+      "loss": 3.8681,
+      "step": 2110464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5437071653335414e-05,
+      "loss": 3.8716,
+      "step": 2110976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5428685705824894e-05,
+      "loss": 3.8766,
+      "step": 2111488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5420316137118103e-05,
+      "loss": 3.8798,
+      "step": 2112000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5411930189607586e-05,
+      "loss": 3.8734,
+      "step": 2112512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5403544242097066e-05,
+      "loss": 3.8785,
+      "step": 2113024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5395158294586546e-05,
+      "loss": 3.8762,
+      "step": 2113536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5386772347076022e-05,
+      "loss": 3.8715,
+      "step": 2114048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5378386399565502e-05,
+      "loss": 3.8605,
+      "step": 2114560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5370000452054982e-05,
+      "loss": 3.8638,
+      "step": 2115072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5361614504544462e-05,
+      "loss": 3.8651,
+      "step": 2115584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5353244935837675e-05,
+      "loss": 3.8705,
+      "step": 2116096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5344875367130887e-05,
+      "loss": 3.8607,
+      "step": 2116608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5336489419620367e-05,
+      "loss": 3.8614,
+      "step": 2117120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5328103472109844e-05,
+      "loss": 3.8595,
+      "step": 2117632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5319717524599324e-05,
+      "loss": 3.8683,
+      "step": 2118144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5311331577088807e-05,
+      "loss": 3.871,
+      "step": 2118656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.530296200838202e-05,
+      "loss": 3.8715,
+      "step": 2119168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5294576060871496e-05,
+      "loss": 3.8749,
+      "step": 2119680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5286190113360976e-05,
+      "loss": 3.8765,
+      "step": 2120192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5277804165850456e-05,
+      "loss": 3.8684,
+      "step": 2120704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5269418218339936e-05,
+      "loss": 3.8774,
+      "step": 2121216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5261032270829416e-05,
+      "loss": 3.8702,
+      "step": 2121728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5252646323318898e-05,
+      "loss": 3.879,
+      "step": 2122240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5244260375808378e-05,
+      "loss": 3.8739,
+      "step": 2122752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5235874428297856e-05,
+      "loss": 3.8719,
+      "step": 2123264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5227504859591068e-05,
+      "loss": 3.8708,
+      "step": 2123776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5219118912080548e-05,
+      "loss": 3.8741,
+      "step": 2124288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5210732964570027e-05,
+      "loss": 3.8649,
+      "step": 2124800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5202347017059506e-05,
+      "loss": 3.8681,
+      "step": 2125312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5193961069548986e-05,
+      "loss": 3.8589,
+      "step": 2125824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.51855915008422e-05,
+      "loss": 3.8658,
+      "step": 2126336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5177205553331677e-05,
+      "loss": 3.8773,
+      "step": 2126848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5168819605821157e-05,
+      "loss": 3.872,
+      "step": 2127360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5160433658310639e-05,
+      "loss": 3.8784,
+      "step": 2127872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5152047710800119e-05,
+      "loss": 3.8581,
+      "step": 2128384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5143661763289599e-05,
+      "loss": 3.8825,
+      "step": 2128896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5135275815779079e-05,
+      "loss": 3.8786,
+      "step": 2129408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5126889868268557e-05,
+      "loss": 3.8788,
+      "step": 2129920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.511852029956177e-05,
+      "loss": 3.8645,
+      "step": 2130432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5110167109658713e-05,
+      "loss": 3.8732,
+      "step": 2130944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5101781162148193e-05,
+      "loss": 3.8823,
+      "step": 2131456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5093395214637673e-05,
+      "loss": 3.8822,
+      "step": 2131968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5085009267127151e-05,
+      "loss": 3.8659,
+      "step": 2132480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5076623319616631e-05,
+      "loss": 3.869,
+      "step": 2132992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.506823737210611e-05,
+      "loss": 3.8734,
+      "step": 2133504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5059851424595592e-05,
+      "loss": 3.8755,
+      "step": 2134016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5051465477085072e-05,
+      "loss": 3.8683,
+      "step": 2134528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5043079529574552e-05,
+      "loss": 3.8738,
+      "step": 2135040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5034709960867763e-05,
+      "loss": 3.8714,
+      "step": 2135552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5026324013357243e-05,
+      "loss": 3.8711,
+      "step": 2136064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5017938065846723e-05,
+      "loss": 3.8767,
+      "step": 2136576
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.9830987453460693,
+      "eval_runtime": 297.463,
+      "eval_samples_per_second": 1282.818,
+      "eval_steps_per_second": 40.089,
+      "step": 2136960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.5009552118336203e-05,
+      "loss": 3.8673,
+      "step": 2137088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.5001166170825681e-05,
+      "loss": 3.8609,
+      "step": 2137600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4992780223315161e-05,
+      "loss": 3.8692,
+      "step": 2138112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4984394275804641e-05,
+      "loss": 3.8739,
+      "step": 2138624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4976008328294123e-05,
+      "loss": 3.8744,
+      "step": 2139136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4967622380783603e-05,
+      "loss": 3.8795,
+      "step": 2139648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4959236433273083e-05,
+      "loss": 3.8685,
+      "step": 2140160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4950850485762563e-05,
+      "loss": 3.8665,
+      "step": 2140672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4942464538252041e-05,
+      "loss": 3.8593,
+      "step": 2141184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4934078590741521e-05,
+      "loss": 3.8717,
+      "step": 2141696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4925692643231001e-05,
+      "loss": 3.8709,
+      "step": 2142208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4917306695720481e-05,
+      "loss": 3.8701,
+      "step": 2142720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4908937127013692e-05,
+      "loss": 3.8786,
+      "step": 2143232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4900551179503172e-05,
+      "loss": 3.853,
+      "step": 2143744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4892165231992653e-05,
+      "loss": 3.8645,
+      "step": 2144256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4883779284482133e-05,
+      "loss": 3.857,
+      "step": 2144768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4875393336971613e-05,
+      "loss": 3.8735,
+      "step": 2145280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4867007389461093e-05,
+      "loss": 3.8661,
+      "step": 2145792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4858621441950571e-05,
+      "loss": 3.8681,
+      "step": 2146304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4850235494440051e-05,
+      "loss": 3.8642,
+      "step": 2146816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4841865925733264e-05,
+      "loss": 3.8864,
+      "step": 2147328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4833479978222744e-05,
+      "loss": 3.8658,
+      "step": 2147840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4825094030712222e-05,
+      "loss": 3.8773,
+      "step": 2148352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4816708083201702e-05,
+      "loss": 3.8687,
+      "step": 2148864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4808338514494916e-05,
+      "loss": 3.873,
+      "step": 2149376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4799952566984393e-05,
+      "loss": 3.8742,
+      "step": 2149888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4791566619473873e-05,
+      "loss": 3.859,
+      "step": 2150400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4783180671963354e-05,
+      "loss": 3.8652,
+      "step": 2150912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4774794724452834e-05,
+      "loss": 3.8615,
+      "step": 2151424
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4766408776942314e-05,
+      "loss": 3.8647,
+      "step": 2151936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4758039208235525e-05,
+      "loss": 3.8696,
+      "step": 2152448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4749653260725005e-05,
+      "loss": 3.8716,
+      "step": 2152960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4741283692018217e-05,
+      "loss": 3.8666,
+      "step": 2153472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4732897744507696e-05,
+      "loss": 3.884,
+      "step": 2153984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4724511796997176e-05,
+      "loss": 3.8697,
+      "step": 2154496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4716125849486656e-05,
+      "loss": 3.8708,
+      "step": 2155008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4707756280779866e-05,
+      "loss": 3.879,
+      "step": 2155520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4699370333269346e-05,
+      "loss": 3.8591,
+      "step": 2156032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4690984385758826e-05,
+      "loss": 3.8678,
+      "step": 2156544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4682598438248308e-05,
+      "loss": 3.8682,
+      "step": 2157056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4674212490737788e-05,
+      "loss": 3.8629,
+      "step": 2157568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4665826543227268e-05,
+      "loss": 3.875,
+      "step": 2158080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4657440595716748e-05,
+      "loss": 3.8732,
+      "step": 2158592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4649054648206226e-05,
+      "loss": 3.8644,
+      "step": 2159104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4640668700695706e-05,
+      "loss": 3.8751,
+      "step": 2159616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4632282753185186e-05,
+      "loss": 3.8723,
+      "step": 2160128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.46239131844784e-05,
+      "loss": 3.8677,
+      "step": 2160640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4615527236967877e-05,
+      "loss": 3.8737,
+      "step": 2161152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4607141289457357e-05,
+      "loss": 3.8426,
+      "step": 2161664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4598755341946838e-05,
+      "loss": 3.8688,
+      "step": 2162176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4590369394436318e-05,
+      "loss": 3.8682,
+      "step": 2162688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4581983446925798e-05,
+      "loss": 3.8732,
+      "step": 2163200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4573597499415278e-05,
+      "loss": 3.861,
+      "step": 2163712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4565211551904757e-05,
+      "loss": 3.8495,
+      "step": 2164224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4556825604394236e-05,
+      "loss": 3.8566,
+      "step": 2164736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4548456035687449e-05,
+      "loss": 3.8576,
+      "step": 2165248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4540070088176929e-05,
+      "loss": 3.875,
+      "step": 2165760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4531684140666407e-05,
+      "loss": 3.8495,
+      "step": 2166272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4523298193155887e-05,
+      "loss": 3.8786,
+      "step": 2166784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4514912245645367e-05,
+      "loss": 3.8713,
+      "step": 2167296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4506542676938578e-05,
+      "loss": 3.8495,
+      "step": 2167808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4498156729428058e-05,
+      "loss": 3.868,
+      "step": 2168320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.448977078191754e-05,
+      "loss": 3.8559,
+      "step": 2168832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.448138483440702e-05,
+      "loss": 3.856,
+      "step": 2169344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4473015265700229e-05,
+      "loss": 3.8578,
+      "step": 2169856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.446462931818971e-05,
+      "loss": 3.8717,
+      "step": 2170368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.445624337067919e-05,
+      "loss": 3.8579,
+      "step": 2170880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.444785742316867e-05,
+      "loss": 3.8681,
+      "step": 2171392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4439487854461881e-05,
+      "loss": 3.8485,
+      "step": 2171904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4431101906951361e-05,
+      "loss": 3.8655,
+      "step": 2172416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.442271595944084e-05,
+      "loss": 3.869,
+      "step": 2172928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.441433001193032e-05,
+      "loss": 3.8741,
+      "step": 2173440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4405944064419802e-05,
+      "loss": 3.8719,
+      "step": 2173952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4397558116909282e-05,
+      "loss": 3.8744,
+      "step": 2174464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4389172169398759e-05,
+      "loss": 3.8863,
+      "step": 2174976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4380802600691973e-05,
+      "loss": 3.8733,
+      "step": 2175488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4372416653181453e-05,
+      "loss": 3.8752,
+      "step": 2176000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4364030705670933e-05,
+      "loss": 3.86,
+      "step": 2176512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4355644758160411e-05,
+      "loss": 3.8654,
+      "step": 2177024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4347258810649891e-05,
+      "loss": 3.8639,
+      "step": 2177536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4338889241943104e-05,
+      "loss": 3.8688,
+      "step": 2178048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4330519673236315e-05,
+      "loss": 3.8764,
+      "step": 2178560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4322133725725794e-05,
+      "loss": 3.8783,
+      "step": 2179072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4313747778215274e-05,
+      "loss": 3.8671,
+      "step": 2179584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4305361830704756e-05,
+      "loss": 3.8569,
+      "step": 2180096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4296975883194233e-05,
+      "loss": 3.8583,
+      "step": 2180608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4288589935683713e-05,
+      "loss": 3.8748,
+      "step": 2181120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4280203988173194e-05,
+      "loss": 3.8684,
+      "step": 2181632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4271818040662674e-05,
+      "loss": 3.8637,
+      "step": 2182144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4263448471955885e-05,
+      "loss": 3.8695,
+      "step": 2182656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4255062524445365e-05,
+      "loss": 3.8676,
+      "step": 2183168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4246676576934845e-05,
+      "loss": 3.8644,
+      "step": 2183680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4238290629424325e-05,
+      "loss": 3.8635,
+      "step": 2184192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4229904681913805e-05,
+      "loss": 3.8754,
+      "step": 2184704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4221518734403286e-05,
+      "loss": 3.8755,
+      "step": 2185216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4213132786892766e-05,
+      "loss": 3.8746,
+      "step": 2185728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4204763218185976e-05,
+      "loss": 3.856,
+      "step": 2186240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4196377270675457e-05,
+      "loss": 3.8644,
+      "step": 2186752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4188007701968666e-05,
+      "loss": 3.8682,
+      "step": 2187264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4179621754458148e-05,
+      "loss": 3.8704,
+      "step": 2187776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4171235806947628e-05,
+      "loss": 3.8772,
+      "step": 2188288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4162849859437108e-05,
+      "loss": 3.8659,
+      "step": 2188800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4154463911926588e-05,
+      "loss": 3.872,
+      "step": 2189312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4146077964416066e-05,
+      "loss": 3.8705,
+      "step": 2189824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4137692016905546e-05,
+      "loss": 3.8681,
+      "step": 2190336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4129306069395026e-05,
+      "loss": 3.8569,
+      "step": 2190848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.412093650068824e-05,
+      "loss": 3.8597,
+      "step": 2191360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4112550553177717e-05,
+      "loss": 3.8632,
+      "step": 2191872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4104164605667197e-05,
+      "loss": 3.8671,
+      "step": 2192384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4095778658156678e-05,
+      "loss": 3.8547,
+      "step": 2192896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4087392710646158e-05,
+      "loss": 3.8596,
+      "step": 2193408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4079023141939367e-05,
+      "loss": 3.8506,
+      "step": 2193920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4070637194428849e-05,
+      "loss": 3.8689,
+      "step": 2194432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4062251246918329e-05,
+      "loss": 3.8645,
+      "step": 2194944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.4053865299407809e-05,
+      "loss": 3.8666,
+      "step": 2195456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.404549573070102e-05,
+      "loss": 3.8679,
+      "step": 2195968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.40371097831905e-05,
+      "loss": 3.8721,
+      "step": 2196480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.402872383567998e-05,
+      "loss": 3.8673,
+      "step": 2196992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.402033788816946e-05,
+      "loss": 3.8668,
+      "step": 2197504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.401196831946267e-05,
+      "loss": 3.8706,
+      "step": 2198016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.400358237195215e-05,
+      "loss": 3.8736,
+      "step": 2198528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3995196424441632e-05,
+      "loss": 3.8687,
+      "step": 2199040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3986810476931112e-05,
+      "loss": 3.868,
+      "step": 2199552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3978440908224321e-05,
+      "loss": 3.8676,
+      "step": 2200064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3970054960713803e-05,
+      "loss": 3.8677,
+      "step": 2200576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3961669013203283e-05,
+      "loss": 3.8602,
+      "step": 2201088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3953283065692763e-05,
+      "loss": 3.8643,
+      "step": 2201600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3944897118182243e-05,
+      "loss": 3.8528,
+      "step": 2202112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.393651117067172e-05,
+      "loss": 3.8653,
+      "step": 2202624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3928141601964933e-05,
+      "loss": 3.8681,
+      "step": 2203136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3919755654454413e-05,
+      "loss": 3.8702,
+      "step": 2203648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3911369706943892e-05,
+      "loss": 3.8685,
+      "step": 2204160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3902983759433371e-05,
+      "loss": 3.861,
+      "step": 2204672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3894597811922851e-05,
+      "loss": 3.876,
+      "step": 2205184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3886244622019795e-05,
+      "loss": 3.8757,
+      "step": 2205696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3877858674509275e-05,
+      "loss": 3.8709,
+      "step": 2206208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3869472726998756e-05,
+      "loss": 3.8658,
+      "step": 2206720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3861086779488236e-05,
+      "loss": 3.8696,
+      "step": 2207232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3852700831977716e-05,
+      "loss": 3.8788,
+      "step": 2207744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3844314884467195e-05,
+      "loss": 3.875,
+      "step": 2208256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3835928936956674e-05,
+      "loss": 3.8645,
+      "step": 2208768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3827542989446154e-05,
+      "loss": 3.8649,
+      "step": 2209280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3819173420739365e-05,
+      "loss": 3.8685,
+      "step": 2209792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3810787473228845e-05,
+      "loss": 3.867,
+      "step": 2210304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3802417904522058e-05,
+      "loss": 3.8645,
+      "step": 2210816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3794031957011538e-05,
+      "loss": 3.8671,
+      "step": 2211328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3785646009501016e-05,
+      "loss": 3.868,
+      "step": 2211840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3777260061990496e-05,
+      "loss": 3.8699,
+      "step": 2212352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3768874114479976e-05,
+      "loss": 3.8729,
+      "step": 2212864
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.981768846511841,
+      "eval_runtime": 295.7618,
+      "eval_samples_per_second": 1290.197,
+      "eval_steps_per_second": 40.32,
+      "step": 2213280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3760488166969457e-05,
+      "loss": 3.8656,
+      "step": 2213376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3752118598262667e-05,
+      "loss": 3.8545,
+      "step": 2213888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3743732650752148e-05,
+      "loss": 3.864,
+      "step": 2214400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3735346703241628e-05,
+      "loss": 3.8726,
+      "step": 2214912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3726960755731108e-05,
+      "loss": 3.8703,
+      "step": 2215424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3718574808220588e-05,
+      "loss": 3.8771,
+      "step": 2215936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3710188860710068e-05,
+      "loss": 3.8617,
+      "step": 2216448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3701802913199546e-05,
+      "loss": 3.8645,
+      "step": 2216960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3693416965689026e-05,
+      "loss": 3.8555,
+      "step": 2217472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3685031018178506e-05,
+      "loss": 3.8703,
+      "step": 2217984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.367666144947172e-05,
+      "loss": 3.8638,
+      "step": 2218496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3668275501961197e-05,
+      "loss": 3.8677,
+      "step": 2219008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3659889554450679e-05,
+      "loss": 3.8751,
+      "step": 2219520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3651503606940159e-05,
+      "loss": 3.8472,
+      "step": 2220032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3643117659429638e-05,
+      "loss": 3.8623,
+      "step": 2220544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.363474809072285e-05,
+      "loss": 3.8511,
+      "step": 2221056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.362636214321233e-05,
+      "loss": 3.8676,
+      "step": 2221568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.361797619570181e-05,
+      "loss": 3.8601,
+      "step": 2222080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3609590248191289e-05,
+      "loss": 3.8665,
+      "step": 2222592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3601204300680769e-05,
+      "loss": 3.8583,
+      "step": 2223104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.359283473197398e-05,
+      "loss": 3.8826,
+      "step": 2223616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.358444878446346e-05,
+      "loss": 3.8648,
+      "step": 2224128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3576062836952941e-05,
+      "loss": 3.8728,
+      "step": 2224640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3567676889442421e-05,
+      "loss": 3.8675,
+      "step": 2225152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3559290941931901e-05,
+      "loss": 3.8709,
+      "step": 2225664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3550921373225112e-05,
+      "loss": 3.867,
+      "step": 2226176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3542535425714592e-05,
+      "loss": 3.8563,
+      "step": 2226688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3534149478204072e-05,
+      "loss": 3.8662,
+      "step": 2227200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3525779909497283e-05,
+      "loss": 3.854,
+      "step": 2227712
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.3517393961986763e-05,
+      "loss": 3.8585,
+      "step": 2228224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3509008014476243e-05,
+      "loss": 3.8653,
+      "step": 2228736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3500622066965723e-05,
+      "loss": 3.8631,
+      "step": 2229248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3492252498258934e-05,
+      "loss": 3.8652,
+      "step": 2229760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3483866550748414e-05,
+      "loss": 3.8788,
+      "step": 2230272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3475480603237895e-05,
+      "loss": 3.8638,
+      "step": 2230784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3467094655727375e-05,
+      "loss": 3.8711,
+      "step": 2231296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3458708708216852e-05,
+      "loss": 3.872,
+      "step": 2231808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3450322760706333e-05,
+      "loss": 3.8541,
+      "step": 2232320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3441936813195813e-05,
+      "loss": 3.8684,
+      "step": 2232832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3433550865685293e-05,
+      "loss": 3.8602,
+      "step": 2233344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3425164918174773e-05,
+      "loss": 3.8602,
+      "step": 2233856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3416778970664253e-05,
+      "loss": 3.8699,
+      "step": 2234368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3408393023153735e-05,
+      "loss": 3.8716,
+      "step": 2234880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3400023454446944e-05,
+      "loss": 3.8573,
+      "step": 2235392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3391637506936426e-05,
+      "loss": 3.8695,
+      "step": 2235904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3383251559425906e-05,
+      "loss": 3.8702,
+      "step": 2236416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3374865611915382e-05,
+      "loss": 3.8641,
+      "step": 2236928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3366479664404862e-05,
+      "loss": 3.8713,
+      "step": 2237440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3358110095698076e-05,
+      "loss": 3.8376,
+      "step": 2237952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3349740526991287e-05,
+      "loss": 3.864,
+      "step": 2238464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3341354579480767e-05,
+      "loss": 3.8647,
+      "step": 2238976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3332968631970247e-05,
+      "loss": 3.8695,
+      "step": 2239488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3324582684459727e-05,
+      "loss": 3.8573,
+      "step": 2240000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3316196736949205e-05,
+      "loss": 3.8489,
+      "step": 2240512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3307810789438685e-05,
+      "loss": 3.8487,
+      "step": 2241024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3299424841928165e-05,
+      "loss": 3.8516,
+      "step": 2241536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3291038894417645e-05,
+      "loss": 3.8726,
+      "step": 2242048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3282652946907127e-05,
+      "loss": 3.8499,
+      "step": 2242560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3274283378200336e-05,
+      "loss": 3.8724,
+      "step": 2243072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3265897430689816e-05,
+      "loss": 3.8621,
+      "step": 2243584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3257511483179297e-05,
+      "loss": 3.8529,
+      "step": 2244096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3249125535668777e-05,
+      "loss": 3.8632,
+      "step": 2244608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3240739588158257e-05,
+      "loss": 3.8484,
+      "step": 2245120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3232353640647737e-05,
+      "loss": 3.8518,
+      "step": 2245632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3223967693137215e-05,
+      "loss": 3.8556,
+      "step": 2246144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3215598124430428e-05,
+      "loss": 3.8708,
+      "step": 2246656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3207212176919908e-05,
+      "loss": 3.8552,
+      "step": 2247168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3198826229409386e-05,
+      "loss": 3.858,
+      "step": 2247680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3190440281898866e-05,
+      "loss": 3.8478,
+      "step": 2248192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.318207071319208e-05,
+      "loss": 3.8575,
+      "step": 2248704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.317368476568156e-05,
+      "loss": 3.8685,
+      "step": 2249216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3165298818171037e-05,
+      "loss": 3.8717,
+      "step": 2249728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3156912870660518e-05,
+      "loss": 3.8681,
+      "step": 2250240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3148526923149998e-05,
+      "loss": 3.8692,
+      "step": 2250752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3140140975639478e-05,
+      "loss": 3.8824,
+      "step": 2251264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3131755028128958e-05,
+      "loss": 3.8682,
+      "step": 2251776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3123385459422169e-05,
+      "loss": 3.8727,
+      "step": 2252288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3114999511911649e-05,
+      "loss": 3.8592,
+      "step": 2252800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3106613564401129e-05,
+      "loss": 3.8626,
+      "step": 2253312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3098227616890609e-05,
+      "loss": 3.8603,
+      "step": 2253824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.308984166938009e-05,
+      "loss": 3.8618,
+      "step": 2254336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.30814721006733e-05,
+      "loss": 3.8709,
+      "step": 2254848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3073086153162781e-05,
+      "loss": 3.8733,
+      "step": 2255360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3064700205652261e-05,
+      "loss": 3.8699,
+      "step": 2255872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3056314258141741e-05,
+      "loss": 3.8494,
+      "step": 2256384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.304792831063122e-05,
+      "loss": 3.8551,
+      "step": 2256896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.30395423631207e-05,
+      "loss": 3.8691,
+      "step": 2257408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.303115641561018e-05,
+      "loss": 3.8644,
+      "step": 2257920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.302277046809966e-05,
+      "loss": 3.8601,
+      "step": 2258432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.3014433657000334e-05,
+      "loss": 3.8649,
+      "step": 2258944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.3006047709489814e-05,
+      "loss": 3.8631,
+      "step": 2259456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2997661761979294e-05,
+      "loss": 3.8612,
+      "step": 2259968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2989275814468773e-05,
+      "loss": 3.8636,
+      "step": 2260480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2980889866958253e-05,
+      "loss": 3.8698,
+      "step": 2260992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2972503919447735e-05,
+      "loss": 3.8668,
+      "step": 2261504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2964117971937215e-05,
+      "loss": 3.8719,
+      "step": 2262016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2955732024426692e-05,
+      "loss": 3.8539,
+      "step": 2262528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2947346076916173e-05,
+      "loss": 3.8609,
+      "step": 2263040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2938960129405653e-05,
+      "loss": 3.8662,
+      "step": 2263552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2930574181895133e-05,
+      "loss": 3.8673,
+      "step": 2264064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2922188234384613e-05,
+      "loss": 3.8729,
+      "step": 2264576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2913818665677824e-05,
+      "loss": 3.8638,
+      "step": 2265088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2905432718167304e-05,
+      "loss": 3.8677,
+      "step": 2265600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2897046770656784e-05,
+      "loss": 3.8658,
+      "step": 2266112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2888660823146265e-05,
+      "loss": 3.8646,
+      "step": 2266624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2880291254439475e-05,
+      "loss": 3.8526,
+      "step": 2267136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2871905306928955e-05,
+      "loss": 3.8536,
+      "step": 2267648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2863519359418436e-05,
+      "loss": 3.8607,
+      "step": 2268160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2855133411907916e-05,
+      "loss": 3.8636,
+      "step": 2268672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2846747464397396e-05,
+      "loss": 3.8523,
+      "step": 2269184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2838361516886874e-05,
+      "loss": 3.8537,
+      "step": 2269696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2829975569376354e-05,
+      "loss": 3.8494,
+      "step": 2270208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2821589621865834e-05,
+      "loss": 3.8595,
+      "step": 2270720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2813220053159047e-05,
+      "loss": 3.8637,
+      "step": 2271232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2804834105648525e-05,
+      "loss": 3.8596,
+      "step": 2271744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2796448158138005e-05,
+      "loss": 3.8624,
+      "step": 2272256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2788062210627485e-05,
+      "loss": 3.8716,
+      "step": 2272768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2779676263116967e-05,
+      "loss": 3.8615,
+      "step": 2273280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2771306694410176e-05,
+      "loss": 3.864,
+      "step": 2273792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2762920746899656e-05,
+      "loss": 3.8686,
+      "step": 2274304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2754534799389137e-05,
+      "loss": 3.8664,
+      "step": 2274816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2746148851878617e-05,
+      "loss": 3.8626,
+      "step": 2275328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2737762904368097e-05,
+      "loss": 3.8682,
+      "step": 2275840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2729376956857577e-05,
+      "loss": 3.8656,
+      "step": 2276352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2721007388150788e-05,
+      "loss": 3.8606,
+      "step": 2276864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2712621440640268e-05,
+      "loss": 3.8517,
+      "step": 2277376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2704235493129748e-05,
+      "loss": 3.8647,
+      "step": 2277888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2695849545619226e-05,
+      "loss": 3.8561,
+      "step": 2278400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2687479976912439e-05,
+      "loss": 3.8608,
+      "step": 2278912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.267909402940192e-05,
+      "loss": 3.8569,
+      "step": 2279424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.26707080818914e-05,
+      "loss": 3.8679,
+      "step": 2279936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2662322134380877e-05,
+      "loss": 3.8635,
+      "step": 2280448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2653936186870358e-05,
+      "loss": 3.8575,
+      "step": 2280960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2645550239359838e-05,
+      "loss": 3.8705,
+      "step": 2281472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.263718067065305e-05,
+      "loss": 3.8701,
+      "step": 2281984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2628794723142529e-05,
+      "loss": 3.8712,
+      "step": 2282496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2620408775632009e-05,
+      "loss": 3.8638,
+      "step": 2283008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2612022828121489e-05,
+      "loss": 3.8615,
+      "step": 2283520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.26036532594147e-05,
+      "loss": 3.8778,
+      "step": 2284032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.259526731190418e-05,
+      "loss": 3.863,
+      "step": 2284544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.258688136439366e-05,
+      "loss": 3.8668,
+      "step": 2285056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.257849541688314e-05,
+      "loss": 3.8585,
+      "step": 2285568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.257012584817635e-05,
+      "loss": 3.8646,
+      "step": 2286080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.256173990066583e-05,
+      "loss": 3.8619,
+      "step": 2286592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2553353953155312e-05,
+      "loss": 3.8604,
+      "step": 2287104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2544968005644792e-05,
+      "loss": 3.8664,
+      "step": 2287616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2536582058134272e-05,
+      "loss": 3.864,
+      "step": 2288128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2528196110623752e-05,
+      "loss": 3.8668,
+      "step": 2288640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.2519826541916963e-05,
+      "loss": 3.8653,
+      "step": 2289152
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.980762243270874,
+      "eval_runtime": 294.783,
+      "eval_samples_per_second": 1294.481,
+      "eval_steps_per_second": 40.453,
+      "step": 2289600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2511440594406443e-05,
+      "loss": 3.848,
+      "step": 2289664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2503054646895923e-05,
+      "loss": 3.8551,
+      "step": 2290176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2494668699385403e-05,
+      "loss": 3.8575,
+      "step": 2290688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2486282751874883e-05,
+      "loss": 3.8705,
+      "step": 2291200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2477896804364362e-05,
+      "loss": 3.8664,
+      "step": 2291712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.246951085685384e-05,
+      "loss": 3.8739,
+      "step": 2292224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2461124909343322e-05,
+      "loss": 3.8567,
+      "step": 2292736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2452738961832802e-05,
+      "loss": 3.8589,
+      "step": 2293248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.244435301432228e-05,
+      "loss": 3.853,
+      "step": 2293760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2435967066811762e-05,
+      "loss": 3.8648,
+      "step": 2294272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2427581119301242e-05,
+      "loss": 3.8637,
+      "step": 2294784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2419195171790722e-05,
+      "loss": 3.8635,
+      "step": 2295296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2410825603083933e-05,
+      "loss": 3.8682,
+      "step": 2295808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2402439655573413e-05,
+      "loss": 3.848,
+      "step": 2296320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2394053708062893e-05,
+      "loss": 3.8553,
+      "step": 2296832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2385667760552371e-05,
+      "loss": 3.85,
+      "step": 2297344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2377298191845584e-05,
+      "loss": 3.8652,
+      "step": 2297856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2368912244335064e-05,
+      "loss": 3.8542,
+      "step": 2298368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2360526296824544e-05,
+      "loss": 3.8639,
+      "step": 2298880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2352140349314023e-05,
+      "loss": 3.8543,
+      "step": 2299392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2343754401803503e-05,
+      "loss": 3.8775,
+      "step": 2299904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2335368454292983e-05,
+      "loss": 3.8612,
+      "step": 2300416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2326998885586196e-05,
+      "loss": 3.8632,
+      "step": 2300928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2318612938075674e-05,
+      "loss": 3.8652,
+      "step": 2301440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2310226990565154e-05,
+      "loss": 3.8686,
+      "step": 2301952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2301841043054636e-05,
+      "loss": 3.8634,
+      "step": 2302464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2293455095544114e-05,
+      "loss": 3.854,
+      "step": 2302976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2285069148033594e-05,
+      "loss": 3.863,
+      "step": 2303488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2276683200523074e-05,
+      "loss": 3.851,
+      "step": 2304000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2268297253012554e-05,
+      "loss": 3.8553,
+      "step": 2304512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2259927684305765e-05,
+      "loss": 3.8588,
+      "step": 2305024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2251541736795245e-05,
+      "loss": 3.8591,
+      "step": 2305536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2243155789284726e-05,
+      "loss": 3.8645,
+      "step": 2306048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2234769841774205e-05,
+      "loss": 3.8697,
+      "step": 2306560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2226400273067417e-05,
+      "loss": 3.8633,
+      "step": 2307072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2218014325556897e-05,
+      "loss": 3.8667,
+      "step": 2307584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2209628378046375e-05,
+      "loss": 3.8665,
+      "step": 2308096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2201258809339588e-05,
+      "loss": 3.852,
+      "step": 2308608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2192872861829068e-05,
+      "loss": 3.8683,
+      "step": 2309120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2184486914318548e-05,
+      "loss": 3.856,
+      "step": 2309632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2176100966808026e-05,
+      "loss": 3.8558,
+      "step": 2310144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2167715019297508e-05,
+      "loss": 3.865,
+      "step": 2310656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2159329071786988e-05,
+      "loss": 3.8711,
+      "step": 2311168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2150943124276466e-05,
+      "loss": 3.8518,
+      "step": 2311680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2142573555569678e-05,
+      "loss": 3.8709,
+      "step": 2312192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2134187608059158e-05,
+      "loss": 3.8621,
+      "step": 2312704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2125801660548638e-05,
+      "loss": 3.862,
+      "step": 2313216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2117415713038118e-05,
+      "loss": 3.8654,
+      "step": 2313728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2109029765527598e-05,
+      "loss": 3.8347,
+      "step": 2314240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2100643818017078e-05,
+      "loss": 3.8596,
+      "step": 2314752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2092257870506556e-05,
+      "loss": 3.867,
+      "step": 2315264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2083871922996038e-05,
+      "loss": 3.8621,
+      "step": 2315776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2075485975485518e-05,
+      "loss": 3.8505,
+      "step": 2316288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2067116406778729e-05,
+      "loss": 3.8455,
+      "step": 2316800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2058730459268209e-05,
+      "loss": 3.8442,
+      "step": 2317312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2050344511757689e-05,
+      "loss": 3.8476,
+      "step": 2317824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2041958564247169e-05,
+      "loss": 3.8733,
+      "step": 2318336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2033588995540381e-05,
+      "loss": 3.8432,
+      "step": 2318848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.202520304802986e-05,
+      "loss": 3.8678,
+      "step": 2319360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.201681710051934e-05,
+      "loss": 3.86,
+      "step": 2319872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.200843115300882e-05,
+      "loss": 3.8464,
+      "step": 2320384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.20000452054983e-05,
+      "loss": 3.858,
+      "step": 2320896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1991659257987779e-05,
+      "loss": 3.8464,
+      "step": 2321408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1983273310477259e-05,
+      "loss": 3.8495,
+      "step": 2321920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1974903741770472e-05,
+      "loss": 3.8472,
+      "step": 2322432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.196651779425995e-05,
+      "loss": 3.8648,
+      "step": 2322944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.195813184674943e-05,
+      "loss": 3.8556,
+      "step": 2323456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1949745899238911e-05,
+      "loss": 3.8521,
+      "step": 2323968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.194135995172839e-05,
+      "loss": 3.846,
+      "step": 2324480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1932990383021602e-05,
+      "loss": 3.8518,
+      "step": 2324992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1924604435511082e-05,
+      "loss": 3.8629,
+      "step": 2325504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1916218488000562e-05,
+      "loss": 3.8687,
+      "step": 2326016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.190783254049004e-05,
+      "loss": 3.8628,
+      "step": 2326528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1899446592979522e-05,
+      "loss": 3.8657,
+      "step": 2327040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1891060645469002e-05,
+      "loss": 3.8771,
+      "step": 2327552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.188267469795848e-05,
+      "loss": 3.8632,
+      "step": 2328064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1874305129251693e-05,
+      "loss": 3.8688,
+      "step": 2328576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1865935560544903e-05,
+      "loss": 3.8574,
+      "step": 2329088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1857549613034383e-05,
+      "loss": 3.86,
+      "step": 2329600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1849163665523863e-05,
+      "loss": 3.8567,
+      "step": 2330112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1840777718013343e-05,
+      "loss": 3.8535,
+      "step": 2330624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1832408149306554e-05,
+      "loss": 3.869,
+      "step": 2331136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1824022201796036e-05,
+      "loss": 3.8718,
+      "step": 2331648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1815636254285514e-05,
+      "loss": 3.8674,
+      "step": 2332160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1807250306774994e-05,
+      "loss": 3.842,
+      "step": 2332672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1798864359264476e-05,
+      "loss": 3.8515,
+      "step": 2333184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1790478411753954e-05,
+      "loss": 3.8645,
+      "step": 2333696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1782092464243434e-05,
+      "loss": 3.8624,
+      "step": 2334208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1773706516732914e-05,
+      "loss": 3.8539,
+      "step": 2334720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.1765320569222394e-05,
+      "loss": 3.8638,
+      "step": 2335232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1756951000515605e-05,
+      "loss": 3.8587,
+      "step": 2335744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1748565053005085e-05,
+      "loss": 3.857,
+      "step": 2336256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1740179105494566e-05,
+      "loss": 3.857,
+      "step": 2336768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1731809536787775e-05,
+      "loss": 3.8675,
+      "step": 2337280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1723423589277257e-05,
+      "loss": 3.8581,
+      "step": 2337792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1715037641766737e-05,
+      "loss": 3.87,
+      "step": 2338304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1706651694256217e-05,
+      "loss": 3.8479,
+      "step": 2338816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1698282125549428e-05,
+      "loss": 3.8544,
+      "step": 2339328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1689896178038908e-05,
+      "loss": 3.8628,
+      "step": 2339840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1681510230528388e-05,
+      "loss": 3.8643,
+      "step": 2340352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1673124283017866e-05,
+      "loss": 3.8678,
+      "step": 2340864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1664738335507347e-05,
+      "loss": 3.8623,
+      "step": 2341376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1656352387996827e-05,
+      "loss": 3.8632,
+      "step": 2341888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1647966440486307e-05,
+      "loss": 3.8621,
+      "step": 2342400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1639580492975787e-05,
+      "loss": 3.8644,
+      "step": 2342912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1631194545465267e-05,
+      "loss": 3.8468,
+      "step": 2343424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1622808597954747e-05,
+      "loss": 3.8532,
+      "step": 2343936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1614422650444226e-05,
+      "loss": 3.8535,
+      "step": 2344448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1606053081737438e-05,
+      "loss": 3.8571,
+      "step": 2344960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1597667134226918e-05,
+      "loss": 3.8523,
+      "step": 2345472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1589281186716396e-05,
+      "loss": 3.85,
+      "step": 2345984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1580895239205878e-05,
+      "loss": 3.8442,
+      "step": 2346496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1572525670499089e-05,
+      "loss": 3.862,
+      "step": 2347008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1564139722988569e-05,
+      "loss": 3.8577,
+      "step": 2347520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1555753775478049e-05,
+      "loss": 3.8559,
+      "step": 2348032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1547367827967529e-05,
+      "loss": 3.8606,
+      "step": 2348544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.153899825926074e-05,
+      "loss": 3.8661,
+      "step": 2349056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1530612311750221e-05,
+      "loss": 3.8586,
+      "step": 2349568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.15222263642397e-05,
+      "loss": 3.8568,
+      "step": 2350080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.151384041672918e-05,
+      "loss": 3.8673,
+      "step": 2350592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.150545446921866e-05,
+      "loss": 3.8636,
+      "step": 2351104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.149708490051187e-05,
+      "loss": 3.8545,
+      "step": 2351616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.148869895300135e-05,
+      "loss": 3.8648,
+      "step": 2352128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1480313005490832e-05,
+      "loss": 3.8602,
+      "step": 2352640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1471927057980311e-05,
+      "loss": 3.8616,
+      "step": 2353152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.146354111046979e-05,
+      "loss": 3.8482,
+      "step": 2353664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1455171541763002e-05,
+      "loss": 3.8622,
+      "step": 2354176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1446785594252482e-05,
+      "loss": 3.8505,
+      "step": 2354688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1438416025545693e-05,
+      "loss": 3.8584,
+      "step": 2355200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1430030078035173e-05,
+      "loss": 3.8462,
+      "step": 2355712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1421644130524653e-05,
+      "loss": 3.8693,
+      "step": 2356224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1413258183014133e-05,
+      "loss": 3.8623,
+      "step": 2356736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1404888614307344e-05,
+      "loss": 3.8505,
+      "step": 2357248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1396502666796824e-05,
+      "loss": 3.8677,
+      "step": 2357760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1388116719286304e-05,
+      "loss": 3.8647,
+      "step": 2358272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1379730771775785e-05,
+      "loss": 3.868,
+      "step": 2358784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1371344824265263e-05,
+      "loss": 3.8589,
+      "step": 2359296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1362975255558474e-05,
+      "loss": 3.8623,
+      "step": 2359808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1354589308047956e-05,
+      "loss": 3.8726,
+      "step": 2360320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1346203360537434e-05,
+      "loss": 3.8575,
+      "step": 2360832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1337817413026914e-05,
+      "loss": 3.8638,
+      "step": 2361344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1329431465516396e-05,
+      "loss": 3.8556,
+      "step": 2361856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1321045518005876e-05,
+      "loss": 3.8628,
+      "step": 2362368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1312659570495354e-05,
+      "loss": 3.8555,
+      "step": 2362880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1304273622984834e-05,
+      "loss": 3.8564,
+      "step": 2363392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1295887675474316e-05,
+      "loss": 3.8603,
+      "step": 2363904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1287518106767525e-05,
+      "loss": 3.8608,
+      "step": 2364416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1279148538060737e-05,
+      "loss": 3.8647,
+      "step": 2364928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.1270762590550217e-05,
+      "loss": 3.86,
+      "step": 2365440
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.9795572757720947,
+      "eval_runtime": 298.1592,
+      "eval_samples_per_second": 1279.823,
+      "eval_steps_per_second": 39.995,
+      "step": 2365920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1262376643039697e-05,
+      "loss": 3.8541,
+      "step": 2365952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1253990695529177e-05,
+      "loss": 3.8537,
+      "step": 2366464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1245604748018657e-05,
+      "loss": 3.8534,
+      "step": 2366976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1237218800508137e-05,
+      "loss": 3.8626,
+      "step": 2367488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1228832852997615e-05,
+      "loss": 3.8597,
+      "step": 2368000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1220446905487097e-05,
+      "loss": 3.8736,
+      "step": 2368512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1212060957976577e-05,
+      "loss": 3.8508,
+      "step": 2369024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1203675010466057e-05,
+      "loss": 3.8543,
+      "step": 2369536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1195305441759268e-05,
+      "loss": 3.8494,
+      "step": 2370048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1186919494248748e-05,
+      "loss": 3.8619,
+      "step": 2370560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1178533546738227e-05,
+      "loss": 3.8543,
+      "step": 2371072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1170147599227707e-05,
+      "loss": 3.8637,
+      "step": 2371584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1161778030520918e-05,
+      "loss": 3.8658,
+      "step": 2372096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1153392083010398e-05,
+      "loss": 3.8464,
+      "step": 2372608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1145006135499878e-05,
+      "loss": 3.8484,
+      "step": 2373120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1136620187989358e-05,
+      "loss": 3.849,
+      "step": 2373632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1128234240478838e-05,
+      "loss": 3.8572,
+      "step": 2374144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.111986467177205e-05,
+      "loss": 3.8551,
+      "step": 2374656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.111147872426153e-05,
+      "loss": 3.8586,
+      "step": 2375168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1103092776751009e-05,
+      "loss": 3.8514,
+      "step": 2375680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1094723208044221e-05,
+      "loss": 3.8712,
+      "step": 2376192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1086337260533701e-05,
+      "loss": 3.8558,
+      "step": 2376704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.107795131302318e-05,
+      "loss": 3.8615,
+      "step": 2377216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.106956536551266e-05,
+      "loss": 3.8614,
+      "step": 2377728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1061179418002141e-05,
+      "loss": 3.8659,
+      "step": 2378240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1052793470491621e-05,
+      "loss": 3.8591,
+      "step": 2378752
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1044423901784832e-05,
+      "loss": 3.8516,
+      "step": 2379264
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1036037954274312e-05,
+      "loss": 3.8572,
+      "step": 2379776
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1027652006763792e-05,
+      "loss": 3.8454,
+      "step": 2380288
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.101926605925327e-05,
+      "loss": 3.8554,
+      "step": 2380800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1010880111742752e-05,
+      "loss": 3.851,
+      "step": 2381312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.1002494164232232e-05,
+      "loss": 3.86,
+      "step": 2381824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.099410821672171e-05,
+      "loss": 3.8571,
+      "step": 2382336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.098572226921119e-05,
+      "loss": 3.864,
+      "step": 2382848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0977336321700671e-05,
+      "loss": 3.8655,
+      "step": 2383360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0968966752993882e-05,
+      "loss": 3.8615,
+      "step": 2383872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0960580805483362e-05,
+      "loss": 3.8636,
+      "step": 2384384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0952211236776573e-05,
+      "loss": 3.8501,
+      "step": 2384896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0943825289266053e-05,
+      "loss": 3.865,
+      "step": 2385408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0935439341755533e-05,
+      "loss": 3.8505,
+      "step": 2385920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0927053394245013e-05,
+      "loss": 3.8534,
+      "step": 2386432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0918667446734493e-05,
+      "loss": 3.8616,
+      "step": 2386944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0910281499223973e-05,
+      "loss": 3.8643,
+      "step": 2387456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0901895551713453e-05,
+      "loss": 3.8504,
+      "step": 2387968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0893509604202933e-05,
+      "loss": 3.8636,
+      "step": 2388480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0885123656692413e-05,
+      "loss": 3.8594,
+      "step": 2388992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0876754087985625e-05,
+      "loss": 3.8585,
+      "step": 2389504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0868368140475103e-05,
+      "loss": 3.865,
+      "step": 2390016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0859982192964583e-05,
+      "loss": 3.8346,
+      "step": 2390528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0851596245454063e-05,
+      "loss": 3.8505,
+      "step": 2391040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0843226676747274e-05,
+      "loss": 3.8647,
+      "step": 2391552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0834840729236754e-05,
+      "loss": 3.8608,
+      "step": 2392064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0826454781726236e-05,
+      "loss": 3.8476,
+      "step": 2392576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0818068834215716e-05,
+      "loss": 3.8382,
+      "step": 2393088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0809682886705194e-05,
+      "loss": 3.847,
+      "step": 2393600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0801296939194674e-05,
+      "loss": 3.8386,
+      "step": 2394112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0792910991684155e-05,
+      "loss": 3.8726,
+      "step": 2394624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0784541422977365e-05,
+      "loss": 3.8376,
+      "step": 2395136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0776155475466845e-05,
+      "loss": 3.863,
+      "step": 2395648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0767769527956326e-05,
+      "loss": 3.854,
+      "step": 2396160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0759383580445806e-05,
+      "loss": 3.8452,
+      "step": 2396672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0750997632935284e-05,
+      "loss": 3.8533,
+      "step": 2397184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0742611685424764e-05,
+      "loss": 3.8487,
+      "step": 2397696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0734242116717977e-05,
+      "loss": 3.8428,
+      "step": 2398208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0725856169207455e-05,
+      "loss": 3.8413,
+      "step": 2398720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0717470221696937e-05,
+      "loss": 3.8652,
+      "step": 2399232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0709084274186417e-05,
+      "loss": 3.8508,
+      "step": 2399744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0700714705479628e-05,
+      "loss": 3.8472,
+      "step": 2400256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0692328757969107e-05,
+      "loss": 3.8413,
+      "step": 2400768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0683942810458587e-05,
+      "loss": 3.8477,
+      "step": 2401280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0675556862948067e-05,
+      "loss": 3.8597,
+      "step": 2401792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.066718729424128e-05,
+      "loss": 3.8656,
+      "step": 2402304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0658801346730758e-05,
+      "loss": 3.8607,
+      "step": 2402816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0650415399220238e-05,
+      "loss": 3.8596,
+      "step": 2403328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0642029451709718e-05,
+      "loss": 3.8724,
+      "step": 2403840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0633643504199198e-05,
+      "loss": 3.8635,
+      "step": 2404352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0625273935492409e-05,
+      "loss": 3.8619,
+      "step": 2404864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.061688798798189e-05,
+      "loss": 3.8517,
+      "step": 2405376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.060850204047137e-05,
+      "loss": 3.859,
+      "step": 2405888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0600116092960849e-05,
+      "loss": 3.8511,
+      "step": 2406400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0591730145450329e-05,
+      "loss": 3.8551,
+      "step": 2406912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.058334419793981e-05,
+      "loss": 3.8632,
+      "step": 2407424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0574958250429289e-05,
+      "loss": 3.8684,
+      "step": 2407936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0566588681722501e-05,
+      "loss": 3.8596,
+      "step": 2408448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0558219113015712e-05,
+      "loss": 3.8436,
+      "step": 2408960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0549833165505192e-05,
+      "loss": 3.8497,
+      "step": 2409472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0541447217994672e-05,
+      "loss": 3.8557,
+      "step": 2409984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0533061270484152e-05,
+      "loss": 3.8649,
+      "step": 2410496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.0524675322973632e-05,
+      "loss": 3.8478,
+      "step": 2411008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.051628937546311e-05,
+      "loss": 3.8611,
+      "step": 2411520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0507903427952592e-05,
+      "loss": 3.8575,
+      "step": 2412032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0499517480442071e-05,
+      "loss": 3.8507,
+      "step": 2412544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0491131532931551e-05,
+      "loss": 3.8523,
+      "step": 2413056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0482761964224762e-05,
+      "loss": 3.8634,
+      "step": 2413568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0474376016714242e-05,
+      "loss": 3.8522,
+      "step": 2414080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0465990069203722e-05,
+      "loss": 3.8667,
+      "step": 2414592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0457604121693202e-05,
+      "loss": 3.8491,
+      "step": 2415104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0449234552986413e-05,
+      "loss": 3.8499,
+      "step": 2415616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0440848605475893e-05,
+      "loss": 3.8565,
+      "step": 2416128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0432462657965374e-05,
+      "loss": 3.8641,
+      "step": 2416640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0424076710454853e-05,
+      "loss": 3.8618,
+      "step": 2417152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0415690762944333e-05,
+      "loss": 3.8597,
+      "step": 2417664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0407304815433813e-05,
+      "loss": 3.8602,
+      "step": 2418176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0398935246727023e-05,
+      "loss": 3.8615,
+      "step": 2418688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0390549299216503e-05,
+      "loss": 3.8579,
+      "step": 2419200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0382179730509716e-05,
+      "loss": 3.8412,
+      "step": 2419712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0373793782999196e-05,
+      "loss": 3.85,
+      "step": 2420224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0365407835488674e-05,
+      "loss": 3.8562,
+      "step": 2420736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0357021887978156e-05,
+      "loss": 3.8523,
+      "step": 2421248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0348635940467636e-05,
+      "loss": 3.8505,
+      "step": 2421760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0340249992957114e-05,
+      "loss": 3.8475,
+      "step": 2422272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0331864045446594e-05,
+      "loss": 3.841,
+      "step": 2422784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0323478097936076e-05,
+      "loss": 3.8553,
+      "step": 2423296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0315092150425556e-05,
+      "loss": 3.8515,
+      "step": 2423808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0306706202915034e-05,
+      "loss": 3.8577,
+      "step": 2424320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0298320255404514e-05,
+      "loss": 3.8554,
+      "step": 2424832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0289934307893995e-05,
+      "loss": 3.8633,
+      "step": 2425344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0281548360383474e-05,
+      "loss": 3.858,
+      "step": 2425856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0273162412872954e-05,
+      "loss": 3.8543,
+      "step": 2426368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0264776465362434e-05,
+      "loss": 3.8609,
+      "step": 2426880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0256390517851914e-05,
+      "loss": 3.8591,
+      "step": 2427392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0248020949145124e-05,
+      "loss": 3.855,
+      "step": 2427904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0239635001634606e-05,
+      "loss": 3.8603,
+      "step": 2428416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0231249054124086e-05,
+      "loss": 3.8568,
+      "step": 2428928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0222863106613564e-05,
+      "loss": 3.8584,
+      "step": 2429440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0214493537906777e-05,
+      "loss": 3.8485,
+      "step": 2429952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0206107590396257e-05,
+      "loss": 3.8515,
+      "step": 2430464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0197721642885737e-05,
+      "loss": 3.8478,
+      "step": 2430976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0189335695375215e-05,
+      "loss": 3.8522,
+      "step": 2431488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0180966126668427e-05,
+      "loss": 3.8419,
+      "step": 2432000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.017261293676537e-05,
+      "loss": 3.8706,
+      "step": 2432512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.016422698925485e-05,
+      "loss": 3.8581,
+      "step": 2433024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0155841041744329e-05,
+      "loss": 3.8528,
+      "step": 2433536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.014745509423381e-05,
+      "loss": 3.8574,
+      "step": 2434048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.013906914672329e-05,
+      "loss": 3.8578,
+      "step": 2434560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0130683199212769e-05,
+      "loss": 3.8672,
+      "step": 2435072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0122297251702249e-05,
+      "loss": 3.8581,
+      "step": 2435584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.011391130419173e-05,
+      "loss": 3.856,
+      "step": 2436096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.010552535668121e-05,
+      "loss": 3.8673,
+      "step": 2436608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0097155787974421e-05,
+      "loss": 3.8513,
+      "step": 2437120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0088769840463901e-05,
+      "loss": 3.8613,
+      "step": 2437632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0080383892953381e-05,
+      "loss": 3.8522,
+      "step": 2438144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0072014324246592e-05,
+      "loss": 3.8601,
+      "step": 2438656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0063628376736072e-05,
+      "loss": 3.8562,
+      "step": 2439168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0055242429225552e-05,
+      "loss": 3.8455,
+      "step": 2439680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0046856481715032e-05,
+      "loss": 3.8594,
+      "step": 2440192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0038470534204512e-05,
+      "loss": 3.8577,
+      "step": 2440704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0030084586693992e-05,
+      "loss": 3.8578,
+      "step": 2441216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.0021698639183472e-05,
+      "loss": 3.8552,
+      "step": 2441728
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.001331269167295e-05,
+      "loss": 3.8572,
+      "step": 2442240
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.978581666946411,
+      "eval_runtime": 290.6167,
+      "eval_samples_per_second": 1313.039,
+      "eval_steps_per_second": 41.033,
+      "step": 2442240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.0004926744162431e-05,
+      "loss": 3.8492,
+      "step": 2442752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.996540796651911e-06,
+      "loss": 3.8456,
+      "step": 2443264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.988154849141391e-06,
+      "loss": 3.8621,
+      "step": 2443776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.97976890163087e-06,
+      "loss": 3.859,
+      "step": 2444288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.971382954120351e-06,
+      "loss": 3.8688,
+      "step": 2444800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.962997006609831e-06,
+      "loss": 3.8497,
+      "step": 2445312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.95461105909931e-06,
+      "loss": 3.8514,
+      "step": 2445824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.946225111588791e-06,
+      "loss": 3.8424,
+      "step": 2446336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.937839164078271e-06,
+      "loss": 3.862,
+      "step": 2446848
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.92945321656775e-06,
+      "loss": 3.8507,
+      "step": 2447360
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.92106726905723e-06,
+      "loss": 3.8591,
+      "step": 2447872
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.912681321546711e-06,
+      "loss": 3.861,
+      "step": 2448384
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.904311752839922e-06,
+      "loss": 3.848,
+      "step": 2448896
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.8959258053294e-06,
+      "loss": 3.8445,
+      "step": 2449408
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.887539857818882e-06,
+      "loss": 3.8473,
+      "step": 2449920
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.879153910308362e-06,
+      "loss": 3.8478,
+      "step": 2450432
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.87076796279784e-06,
+      "loss": 3.8547,
+      "step": 2450944
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.86238201528732e-06,
+      "loss": 3.8542,
+      "step": 2451456
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.853996067776801e-06,
+      "loss": 3.8463,
+      "step": 2451968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.845626499070012e-06,
+      "loss": 3.8736,
+      "step": 2452480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.837240551559492e-06,
+      "loss": 3.8511,
+      "step": 2452992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.828870982852703e-06,
+      "loss": 3.8614,
+      "step": 2453504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.820485035342183e-06,
+      "loss": 3.8578,
+      "step": 2454016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.812099087831663e-06,
+      "loss": 3.8633,
+      "step": 2454528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.803713140321143e-06,
+      "loss": 3.8513,
+      "step": 2455040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.795327192810623e-06,
+      "loss": 3.8535,
+      "step": 2455552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.786941245300103e-06,
+      "loss": 3.8497,
+      "step": 2456064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.778555297789583e-06,
+      "loss": 3.8497,
+      "step": 2456576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 9.770169350279063e-06,
+      "loss": 3.8474,
+      "step": 2457088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.761783402768543e-06,
+      "loss": 3.8459,
+      "step": 2457600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.753413834061753e-06,
+      "loss": 3.8571,
+      "step": 2458112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.745027886551233e-06,
+      "loss": 3.8563,
+      "step": 2458624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.736641939040713e-06,
+      "loss": 3.8575,
+      "step": 2459136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.728255991530193e-06,
+      "loss": 3.8629,
+      "step": 2459648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.719886422823404e-06,
+      "loss": 3.8594,
+      "step": 2460160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.711500475312884e-06,
+      "loss": 3.8571,
+      "step": 2460672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.703114527802366e-06,
+      "loss": 3.8512,
+      "step": 2461184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.694728580291844e-06,
+      "loss": 3.8559,
+      "step": 2461696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.686342632781324e-06,
+      "loss": 3.8467,
+      "step": 2462208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.677973064074536e-06,
+      "loss": 3.8541,
+      "step": 2462720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.669603495367747e-06,
+      "loss": 3.8551,
+      "step": 2463232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.661217547857227e-06,
+      "loss": 3.8611,
+      "step": 2463744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.652831600346707e-06,
+      "loss": 3.8481,
+      "step": 2464256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.644445652836187e-06,
+      "loss": 3.859,
+      "step": 2464768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.636059705325667e-06,
+      "loss": 3.8597,
+      "step": 2465280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.627673757815147e-06,
+      "loss": 3.8587,
+      "step": 2465792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.619287810304627e-06,
+      "loss": 3.8574,
+      "step": 2466304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.610901862794107e-06,
+      "loss": 3.8298,
+      "step": 2466816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.602515915283585e-06,
+      "loss": 3.8467,
+      "step": 2467328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.594146346576798e-06,
+      "loss": 3.8614,
+      "step": 2467840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.585760399066278e-06,
+      "loss": 3.8576,
+      "step": 2468352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.577374451555758e-06,
+      "loss": 3.8457,
+      "step": 2468864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.568988504045237e-06,
+      "loss": 3.834,
+      "step": 2469376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.560602556534717e-06,
+      "loss": 3.8433,
+      "step": 2469888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.552232987827928e-06,
+      "loss": 3.8379,
+      "step": 2470400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.543847040317408e-06,
+      "loss": 3.8649,
+      "step": 2470912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.535477471610619e-06,
+      "loss": 3.839,
+      "step": 2471424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.5270915241001e-06,
+      "loss": 3.858,
+      "step": 2471936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.51870557658958e-06,
+      "loss": 3.8543,
+      "step": 2472448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.510319629079059e-06,
+      "loss": 3.8391,
+      "step": 2472960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.501933681568539e-06,
+      "loss": 3.8509,
+      "step": 2473472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.49354773405802e-06,
+      "loss": 3.8461,
+      "step": 2473984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.485161786547499e-06,
+      "loss": 3.8356,
+      "step": 2474496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.476775839036979e-06,
+      "loss": 3.839,
+      "step": 2475008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.468389891526459e-06,
+      "loss": 3.8636,
+      "step": 2475520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.46000394401594e-06,
+      "loss": 3.852,
+      "step": 2476032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.45163437530915e-06,
+      "loss": 3.8432,
+      "step": 2476544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.443248427798631e-06,
+      "loss": 3.8384,
+      "step": 2477056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.434862480288111e-06,
+      "loss": 3.8461,
+      "step": 2477568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.42647653277759e-06,
+      "loss": 3.8564,
+      "step": 2478080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.41809058526707e-06,
+      "loss": 3.8627,
+      "step": 2478592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.409721016560282e-06,
+      "loss": 3.8547,
+      "step": 2479104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.401335069049762e-06,
+      "loss": 3.8599,
+      "step": 2479616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.39294912153924e-06,
+      "loss": 3.8675,
+      "step": 2480128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.384563174028722e-06,
+      "loss": 3.8598,
+      "step": 2480640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.376193605321932e-06,
+      "loss": 3.8583,
+      "step": 2481152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.367807657811412e-06,
+      "loss": 3.8504,
+      "step": 2481664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.359421710300892e-06,
+      "loss": 3.8578,
+      "step": 2482176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.351035762790372e-06,
+      "loss": 3.8507,
+      "step": 2482688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.342649815279852e-06,
+      "loss": 3.8506,
+      "step": 2483200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.334280246573063e-06,
+      "loss": 3.8559,
+      "step": 2483712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.325894299062543e-06,
+      "loss": 3.8629,
+      "step": 2484224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.317508351552023e-06,
+      "loss": 3.8607,
+      "step": 2484736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.309122404041505e-06,
+      "loss": 3.839,
+      "step": 2485248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.300752835334714e-06,
+      "loss": 3.8433,
+      "step": 2485760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.292366887824194e-06,
+      "loss": 3.8547,
+      "step": 2486272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.283980940313675e-06,
+      "loss": 3.8613,
+      "step": 2486784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.275594992803153e-06,
+      "loss": 3.8466,
+      "step": 2487296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 9.267209045292633e-06,
+      "loss": 3.8536,
+      "step": 2487808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.258823097782113e-06,
+      "loss": 3.8575,
+      "step": 2488320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.250437150271593e-06,
+      "loss": 3.8484,
+      "step": 2488832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.242051202761073e-06,
+      "loss": 3.8494,
+      "step": 2489344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.233665255250553e-06,
+      "loss": 3.8545,
+      "step": 2489856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.225279307740033e-06,
+      "loss": 3.8515,
+      "step": 2490368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.216893360229513e-06,
+      "loss": 3.8611,
+      "step": 2490880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.208523791522724e-06,
+      "loss": 3.8531,
+      "step": 2491392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.200137844012206e-06,
+      "loss": 3.843,
+      "step": 2491904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.191751896501684e-06,
+      "loss": 3.8555,
+      "step": 2492416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.183365948991164e-06,
+      "loss": 3.8582,
+      "step": 2492928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.174980001480644e-06,
+      "loss": 3.8554,
+      "step": 2493440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.166594053970125e-06,
+      "loss": 3.8598,
+      "step": 2493952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.158208106459604e-06,
+      "loss": 3.8542,
+      "step": 2494464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.149838537752816e-06,
+      "loss": 3.8568,
+      "step": 2494976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.141452590242296e-06,
+      "loss": 3.858,
+      "step": 2495488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.133066642731774e-06,
+      "loss": 3.8383,
+      "step": 2496000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.124680695221254e-06,
+      "loss": 3.8421,
+      "step": 2496512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.116294747710736e-06,
+      "loss": 3.8528,
+      "step": 2497024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.107908800200216e-06,
+      "loss": 3.8515,
+      "step": 2497536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.099539231493425e-06,
+      "loss": 3.8426,
+      "step": 2498048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.091153283982907e-06,
+      "loss": 3.8449,
+      "step": 2498560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.082783715276117e-06,
+      "loss": 3.8417,
+      "step": 2499072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.074397767765597e-06,
+      "loss": 3.85,
+      "step": 2499584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.066011820255077e-06,
+      "loss": 3.8485,
+      "step": 2500096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.057625872744557e-06,
+      "loss": 3.8505,
+      "step": 2500608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.049239925234037e-06,
+      "loss": 3.851,
+      "step": 2501120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.040853977723517e-06,
+      "loss": 3.8572,
+      "step": 2501632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.032484409016728e-06,
+      "loss": 3.8537,
+      "step": 2502144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.024098461506208e-06,
+      "loss": 3.8535,
+      "step": 2502656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.01571251399569e-06,
+      "loss": 3.8569,
+      "step": 2503168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 9.007326566485168e-06,
+      "loss": 3.8565,
+      "step": 2503680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.998940618974648e-06,
+      "loss": 3.8472,
+      "step": 2504192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.990554671464128e-06,
+      "loss": 3.8592,
+      "step": 2504704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.982168723953608e-06,
+      "loss": 3.8502,
+      "step": 2505216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.973782776443088e-06,
+      "loss": 3.8599,
+      "step": 2505728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.965396828932568e-06,
+      "loss": 3.839,
+      "step": 2506240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.95702726022578e-06,
+      "loss": 3.8544,
+      "step": 2506752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.948641312715258e-06,
+      "loss": 3.8394,
+      "step": 2507264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.940255365204738e-06,
+      "loss": 3.8472,
+      "step": 2507776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.931885796497951e-06,
+      "loss": 3.8386,
+      "step": 2508288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.923499848987429e-06,
+      "loss": 3.8674,
+      "step": 2508800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.915113901476909e-06,
+      "loss": 3.8584,
+      "step": 2509312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.90672795396639e-06,
+      "loss": 3.8467,
+      "step": 2509824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.89834200645587e-06,
+      "loss": 3.8515,
+      "step": 2510336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.889956058945349e-06,
+      "loss": 3.8545,
+      "step": 2510848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.881570111434829e-06,
+      "loss": 3.8631,
+      "step": 2511360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.873200542728041e-06,
+      "loss": 3.8577,
+      "step": 2511872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.86481459521752e-06,
+      "loss": 3.8517,
+      "step": 2512384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.856428647707001e-06,
+      "loss": 3.8644,
+      "step": 2512896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.848042700196481e-06,
+      "loss": 3.8484,
+      "step": 2513408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.839656752685961e-06,
+      "loss": 3.861,
+      "step": 2513920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.83127080517544e-06,
+      "loss": 3.8519,
+      "step": 2514432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.822884857664921e-06,
+      "loss": 3.8545,
+      "step": 2514944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.814498910154401e-06,
+      "loss": 3.8549,
+      "step": 2515456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.80611296264388e-06,
+      "loss": 3.8473,
+      "step": 2515968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.797743393937092e-06,
+      "loss": 3.8565,
+      "step": 2516480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.789357446426572e-06,
+      "loss": 3.8568,
+      "step": 2516992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.780971498916052e-06,
+      "loss": 3.8547,
+      "step": 2517504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.772618309012993e-06,
+      "loss": 3.8543,
+      "step": 2518016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.764232361502473e-06,
+      "loss": 3.8499,
+      "step": 2518528
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.977595329284668,
+      "eval_runtime": 293.6827,
+      "eval_samples_per_second": 1299.331,
+      "eval_steps_per_second": 40.605,
+      "step": 2518560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.755846413991953e-06,
+      "loss": 3.8453,
+      "step": 2519040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.747460466481435e-06,
+      "loss": 3.8451,
+      "step": 2519552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.739074518970913e-06,
+      "loss": 3.8605,
+      "step": 2520064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.730688571460393e-06,
+      "loss": 3.8506,
+      "step": 2520576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.722302623949873e-06,
+      "loss": 3.8693,
+      "step": 2521088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.713916676439353e-06,
+      "loss": 3.8432,
+      "step": 2521600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.705530728928833e-06,
+      "loss": 3.8512,
+      "step": 2522112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.697144781418313e-06,
+      "loss": 3.8392,
+      "step": 2522624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.688758833907793e-06,
+      "loss": 3.8566,
+      "step": 2523136
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.680372886397273e-06,
+      "loss": 3.8502,
+      "step": 2523648
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.672003317690484e-06,
+      "loss": 3.855,
+      "step": 2524160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.663617370179965e-06,
+      "loss": 3.8561,
+      "step": 2524672
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.655247801473174e-06,
+      "loss": 3.8453,
+      "step": 2525184
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.646861853962656e-06,
+      "loss": 3.8447,
+      "step": 2525696
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.638475906452136e-06,
+      "loss": 3.8451,
+      "step": 2526208
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.630089958941614e-06,
+      "loss": 3.8411,
+      "step": 2526720
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.621704011431094e-06,
+      "loss": 3.8518,
+      "step": 2527232
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.613318063920576e-06,
+      "loss": 3.8493,
+      "step": 2527744
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.604932116410056e-06,
+      "loss": 3.8475,
+      "step": 2528256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.596546168899534e-06,
+      "loss": 3.8726,
+      "step": 2528768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.588160221389014e-06,
+      "loss": 3.8448,
+      "step": 2529280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.579790652682227e-06,
+      "loss": 3.8578,
+      "step": 2529792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.571404705171705e-06,
+      "loss": 3.8555,
+      "step": 2530304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.563018757661185e-06,
+      "loss": 3.8637,
+      "step": 2530816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.554632810150666e-06,
+      "loss": 3.8405,
+      "step": 2531328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.546263241443877e-06,
+      "loss": 3.851,
+      "step": 2531840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.537877293933357e-06,
+      "loss": 3.8452,
+      "step": 2532352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.529507725226568e-06,
+      "loss": 3.853,
+      "step": 2532864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 8.521121777716048e-06,
+      "loss": 3.8391,
+      "step": 2533376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.51273583020553e-06,
+      "loss": 3.8431,
+      "step": 2533888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.504349882695008e-06,
+      "loss": 3.8542,
+      "step": 2534400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.495963935184488e-06,
+      "loss": 3.8542,
+      "step": 2534912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.487577987673968e-06,
+      "loss": 3.8521,
+      "step": 2535424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.479192040163448e-06,
+      "loss": 3.8592,
+      "step": 2535936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.470806092652928e-06,
+      "loss": 3.8572,
+      "step": 2536448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.462420145142408e-06,
+      "loss": 3.8539,
+      "step": 2536960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.45405057643562e-06,
+      "loss": 3.8455,
+      "step": 2537472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.445664628925098e-06,
+      "loss": 3.8584,
+      "step": 2537984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.43729506021831e-06,
+      "loss": 3.8423,
+      "step": 2538496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.42890911270779e-06,
+      "loss": 3.8497,
+      "step": 2539008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.420523165197269e-06,
+      "loss": 3.8544,
+      "step": 2539520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.412137217686749e-06,
+      "loss": 3.8574,
+      "step": 2540032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.40375127017623e-06,
+      "loss": 3.8439,
+      "step": 2540544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.39536532266571e-06,
+      "loss": 3.853,
+      "step": 2541056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.386979375155189e-06,
+      "loss": 3.8599,
+      "step": 2541568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.378593427644669e-06,
+      "loss": 3.855,
+      "step": 2542080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.370223858937881e-06,
+      "loss": 3.8556,
+      "step": 2542592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.36183791142736e-06,
+      "loss": 3.8277,
+      "step": 2543104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.353451963916841e-06,
+      "loss": 3.8383,
+      "step": 2543616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.345066016406321e-06,
+      "loss": 3.8604,
+      "step": 2544128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.336680068895801e-06,
+      "loss": 3.8511,
+      "step": 2544640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.32829412138528e-06,
+      "loss": 3.8479,
+      "step": 2545152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.319908173874761e-06,
+      "loss": 3.8338,
+      "step": 2545664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.311522226364241e-06,
+      "loss": 3.8393,
+      "step": 2546176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.30313627885372e-06,
+      "loss": 3.8338,
+      "step": 2546688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.294766710146932e-06,
+      "loss": 3.8585,
+      "step": 2547200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.286397141440143e-06,
+      "loss": 3.8374,
+      "step": 2547712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.278011193929623e-06,
+      "loss": 3.8526,
+      "step": 2548224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.269625246419102e-06,
+      "loss": 3.854,
+      "step": 2548736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.261239298908582e-06,
+      "loss": 3.8375,
+      "step": 2549248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.252853351398062e-06,
+      "loss": 3.8432,
+      "step": 2549760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.244467403887542e-06,
+      "loss": 3.8402,
+      "step": 2550272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.236081456377022e-06,
+      "loss": 3.8358,
+      "step": 2550784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.227695508866502e-06,
+      "loss": 3.8387,
+      "step": 2551296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.219342318963446e-06,
+      "loss": 3.8552,
+      "step": 2551808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.210956371452924e-06,
+      "loss": 3.8496,
+      "step": 2552320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.202570423942404e-06,
+      "loss": 3.8357,
+      "step": 2552832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.194184476431885e-06,
+      "loss": 3.8356,
+      "step": 2553344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.185798528921365e-06,
+      "loss": 3.8382,
+      "step": 2553856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.177412581410844e-06,
+      "loss": 3.853,
+      "step": 2554368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.169043012704056e-06,
+      "loss": 3.8591,
+      "step": 2554880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.160657065193536e-06,
+      "loss": 3.852,
+      "step": 2555392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.152271117683014e-06,
+      "loss": 3.8606,
+      "step": 2555904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.143885170172496e-06,
+      "loss": 3.8599,
+      "step": 2556416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.135515601465707e-06,
+      "loss": 3.86,
+      "step": 2556928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.127129653955187e-06,
+      "loss": 3.8512,
+      "step": 2557440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.118743706444667e-06,
+      "loss": 3.8498,
+      "step": 2557952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.110357758934147e-06,
+      "loss": 3.8516,
+      "step": 2558464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.101971811423627e-06,
+      "loss": 3.8492,
+      "step": 2558976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.093585863913105e-06,
+      "loss": 3.8474,
+      "step": 2559488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.085199916402587e-06,
+      "loss": 3.8538,
+      "step": 2560000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.076830347695797e-06,
+      "loss": 3.8591,
+      "step": 2560512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.068444400185277e-06,
+      "loss": 3.8537,
+      "step": 2561024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.060058452674757e-06,
+      "loss": 3.8399,
+      "step": 2561536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.051672505164237e-06,
+      "loss": 3.8396,
+      "step": 2562048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.043286557653717e-06,
+      "loss": 3.8499,
+      "step": 2562560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.034900610143197e-06,
+      "loss": 3.8586,
+      "step": 2563072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.026514662632677e-06,
+      "loss": 3.846,
+      "step": 2563584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 8.018128715122157e-06,
+      "loss": 3.8505,
+      "step": 2564096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 8.009742767611635e-06,
+      "loss": 3.8541,
+      "step": 2564608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 8.001373198904848e-06,
+      "loss": 3.8471,
+      "step": 2565120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.992987251394328e-06,
+      "loss": 3.8473,
+      "step": 2565632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.984601303883808e-06,
+      "loss": 3.8476,
+      "step": 2566144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.976215356373288e-06,
+      "loss": 3.8516,
+      "step": 2566656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.967829408862768e-06,
+      "loss": 3.8553,
+      "step": 2567168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.959443461352248e-06,
+      "loss": 3.855,
+      "step": 2567680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.951057513841727e-06,
+      "loss": 3.8388,
+      "step": 2568192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.942704323938669e-06,
+      "loss": 3.8548,
+      "step": 2568704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.93431837642815e-06,
+      "loss": 3.8507,
+      "step": 2569216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.92593242891763e-06,
+      "loss": 3.8592,
+      "step": 2569728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.917546481407109e-06,
+      "loss": 3.8523,
+      "step": 2570240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.909160533896589e-06,
+      "loss": 3.8526,
+      "step": 2570752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.90077458638607e-06,
+      "loss": 3.8543,
+      "step": 2571264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.89238863887555e-06,
+      "loss": 3.856,
+      "step": 2571776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.884002691365029e-06,
+      "loss": 3.8363,
+      "step": 2572288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.875616743854509e-06,
+      "loss": 3.8414,
+      "step": 2572800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.867247175147721e-06,
+      "loss": 3.8496,
+      "step": 2573312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.8588612276372e-06,
+      "loss": 3.8463,
+      "step": 2573824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.850475280126681e-06,
+      "loss": 3.8421,
+      "step": 2574336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.842089332616161e-06,
+      "loss": 3.8436,
+      "step": 2574848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.833719763909372e-06,
+      "loss": 3.8386,
+      "step": 2575360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.825333816398852e-06,
+      "loss": 3.8469,
+      "step": 2575872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.816947868888332e-06,
+      "loss": 3.8445,
+      "step": 2576384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.808561921377812e-06,
+      "loss": 3.845,
+      "step": 2576896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.80017597386729e-06,
+      "loss": 3.8505,
+      "step": 2577408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.791806405160503e-06,
+      "loss": 3.8547,
+      "step": 2577920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.783420457649982e-06,
+      "loss": 3.8511,
+      "step": 2578432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.775034510139462e-06,
+      "loss": 3.8478,
+      "step": 2578944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.766648562628942e-06,
+      "loss": 3.8559,
+      "step": 2579456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.758278993922153e-06,
+      "loss": 3.8546,
+      "step": 2579968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.749893046411635e-06,
+      "loss": 3.8433,
+      "step": 2580480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.741523477704846e-06,
+      "loss": 3.8582,
+      "step": 2580992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.733137530194324e-06,
+      "loss": 3.8471,
+      "step": 2581504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.724767961487536e-06,
+      "loss": 3.8576,
+      "step": 2582016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.716382013977016e-06,
+      "loss": 3.8344,
+      "step": 2582528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.707996066466496e-06,
+      "loss": 3.855,
+      "step": 2583040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.699610118955976e-06,
+      "loss": 3.8366,
+      "step": 2583552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.691224171445456e-06,
+      "loss": 3.8435,
+      "step": 2584064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.682838223934936e-06,
+      "loss": 3.8345,
+      "step": 2584576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.674452276424416e-06,
+      "loss": 3.8635,
+      "step": 2585088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.666066328913896e-06,
+      "loss": 3.8551,
+      "step": 2585600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.657680381403376e-06,
+      "loss": 3.8477,
+      "step": 2586112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.649294433892854e-06,
+      "loss": 3.8454,
+      "step": 2586624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.640908486382336e-06,
+      "loss": 3.8542,
+      "step": 2587136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.632522538871816e-06,
+      "loss": 3.8574,
+      "step": 2587648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.624136591361296e-06,
+      "loss": 3.8523,
+      "step": 2588160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.615767022654506e-06,
+      "loss": 3.8522,
+      "step": 2588672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.6073810751439866e-06,
+      "loss": 3.8551,
+      "step": 2589184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.5989951276334665e-06,
+      "loss": 3.8484,
+      "step": 2589696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.590609180122946e-06,
+      "loss": 3.8623,
+      "step": 2590208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.582239611416157e-06,
+      "loss": 3.8457,
+      "step": 2590720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.573853663905637e-06,
+      "loss": 3.8499,
+      "step": 2591232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.565467716395118e-06,
+      "loss": 3.8511,
+      "step": 2591744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.557081768884597e-06,
+      "loss": 3.8437,
+      "step": 2592256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.548695821374077e-06,
+      "loss": 3.8549,
+      "step": 2592768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.540326252667289e-06,
+      "loss": 3.8523,
+      "step": 2593280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.5319403051567695e-06,
+      "loss": 3.8549,
+      "step": 2593792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.5235707364499795e-06,
+      "loss": 3.8485,
+      "step": 2594304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 7.5151847889394594e-06,
+      "loss": 3.8451,
+      "step": 2594816
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.976717472076416,
+      "eval_runtime": 293.5342,
+      "eval_samples_per_second": 1299.988,
+      "eval_steps_per_second": 40.626,
+      "step": 2594880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.50679884142894e-06,
+      "loss": 3.8418,
+      "step": 2595328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.498412893918419e-06,
+      "loss": 3.8435,
+      "step": 2595840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.490026946407899e-06,
+      "loss": 3.8549,
+      "step": 2596352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.481640998897379e-06,
+      "loss": 3.8462,
+      "step": 2596864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.473255051386858e-06,
+      "loss": 3.8684,
+      "step": 2597376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.464869103876339e-06,
+      "loss": 3.8396,
+      "step": 2597888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.456483156365819e-06,
+      "loss": 3.8483,
+      "step": 2598400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.448097208855299e-06,
+      "loss": 3.8373,
+      "step": 2598912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.439711261344778e-06,
+      "loss": 3.8546,
+      "step": 2599424
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.431325313834259e-06,
+      "loss": 3.8472,
+      "step": 2599936
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.422939366323739e-06,
+      "loss": 3.8505,
+      "step": 2600448
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.414553418813218e-06,
+      "loss": 3.85,
+      "step": 2600960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.40618385010643e-06,
+      "loss": 3.84,
+      "step": 2601472
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.39779790259591e-06,
+      "loss": 3.8436,
+      "step": 2601984
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.3894119550853904e-06,
+      "loss": 3.8441,
+      "step": 2602496
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.381026007574869e-06,
+      "loss": 3.8398,
+      "step": 2603008
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.3726400600643495e-06,
+      "loss": 3.8448,
+      "step": 2603520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.3642541125538295e-06,
+      "loss": 3.8466,
+      "step": 2604032
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.35588454384704e-06,
+      "loss": 3.8428,
+      "step": 2604544
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.34749859633652e-06,
+      "loss": 3.8684,
+      "step": 2605056
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.339112648826001e-06,
+      "loss": 3.8436,
+      "step": 2605568
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.330726701315481e-06,
+      "loss": 3.8495,
+      "step": 2606080
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.32234075380496e-06,
+      "loss": 3.8531,
+      "step": 2606592
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.31395480629444e-06,
+      "loss": 3.8589,
+      "step": 2607104
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.305568858783921e-06,
+      "loss": 3.8389,
+      "step": 2607616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.297199290077131e-06,
+      "loss": 3.8518,
+      "step": 2608128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.288813342566611e-06,
+      "loss": 3.8423,
+      "step": 2608640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.2804273950560915e-06,
+      "loss": 3.8477,
+      "step": 2609152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 7.2720414475455715e-06,
+      "loss": 3.8361,
+      "step": 2609664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.263655500035051e-06,
+      "loss": 3.8396,
+      "step": 2610176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.2552695525245306e-06,
+      "loss": 3.8501,
+      "step": 2610688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.246883605014011e-06,
+      "loss": 3.8513,
+      "step": 2611200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.2384976575034905e-06,
+      "loss": 3.8511,
+      "step": 2611712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.23011170999297e-06,
+      "loss": 3.8571,
+      "step": 2612224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.221758520089914e-06,
+      "loss": 3.8488,
+      "step": 2612736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.213372572579394e-06,
+      "loss": 3.854,
+      "step": 2613248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.204986625068873e-06,
+      "loss": 3.841,
+      "step": 2613760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.196600677558353e-06,
+      "loss": 3.8568,
+      "step": 2614272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.1882147300478336e-06,
+      "loss": 3.8374,
+      "step": 2614784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.179828782537313e-06,
+      "loss": 3.8476,
+      "step": 2615296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.171442835026793e-06,
+      "loss": 3.8467,
+      "step": 2615808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.163073266320004e-06,
+      "loss": 3.8557,
+      "step": 2616320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.154687318809484e-06,
+      "loss": 3.847,
+      "step": 2616832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.146301371298963e-06,
+      "loss": 3.8466,
+      "step": 2617344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.137915423788444e-06,
+      "loss": 3.8562,
+      "step": 2617856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.129529476277924e-06,
+      "loss": 3.8522,
+      "step": 2618368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.121143528767403e-06,
+      "loss": 3.8516,
+      "step": 2618880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.112757581256883e-06,
+      "loss": 3.8347,
+      "step": 2619392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.104388012550095e-06,
+      "loss": 3.8319,
+      "step": 2619904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.0960184438433064e-06,
+      "loss": 3.8505,
+      "step": 2620416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.0876324963327855e-06,
+      "loss": 3.8488,
+      "step": 2620928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.0792465488222655e-06,
+      "loss": 3.8465,
+      "step": 2621440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.070860601311746e-06,
+      "loss": 3.8308,
+      "step": 2621952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.062474653801226e-06,
+      "loss": 3.8364,
+      "step": 2622464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.054088706290705e-06,
+      "loss": 3.8376,
+      "step": 2622976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.045702758780186e-06,
+      "loss": 3.8486,
+      "step": 2623488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.037316811269666e-06,
+      "loss": 3.8361,
+      "step": 2624000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.028930863759145e-06,
+      "loss": 3.8505,
+      "step": 2624512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.020544916248625e-06,
+      "loss": 3.8464,
+      "step": 2625024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.012158968738106e-06,
+      "loss": 3.8378,
+      "step": 2625536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 7.003773021227584e-06,
+      "loss": 3.8406,
+      "step": 2626048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.995387073717065e-06,
+      "loss": 3.8395,
+      "step": 2626560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.987017505010277e-06,
+      "loss": 3.8306,
+      "step": 2627072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.978631557499757e-06,
+      "loss": 3.8418,
+      "step": 2627584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.970245609989236e-06,
+      "loss": 3.8454,
+      "step": 2628096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.961859662478716e-06,
+      "loss": 3.8543,
+      "step": 2628608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.9534737149681965e-06,
+      "loss": 3.8295,
+      "step": 2629120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.945087767457676e-06,
+      "loss": 3.8346,
+      "step": 2629632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.936718198750887e-06,
+      "loss": 3.833,
+      "step": 2630144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.928332251240367e-06,
+      "loss": 3.8572,
+      "step": 2630656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.919962682533579e-06,
+      "loss": 3.8465,
+      "step": 2631168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.911576735023058e-06,
+      "loss": 3.8485,
+      "step": 2631680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.903190787512538e-06,
+      "loss": 3.859,
+      "step": 2632192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.894804840002019e-06,
+      "loss": 3.8586,
+      "step": 2632704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.88643527129523e-06,
+      "loss": 3.8569,
+      "step": 2633216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.8780493237847095e-06,
+      "loss": 3.8469,
+      "step": 2633728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.869663376274189e-06,
+      "loss": 3.8519,
+      "step": 2634240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.861277428763669e-06,
+      "loss": 3.8485,
+      "step": 2634752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.8528914812531485e-06,
+      "loss": 3.8428,
+      "step": 2635264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.844505533742629e-06,
+      "loss": 3.8467,
+      "step": 2635776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.836135965035841e-06,
+      "loss": 3.8487,
+      "step": 2636288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.827750017525321e-06,
+      "loss": 3.8583,
+      "step": 2636800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.8193640700148e-06,
+      "loss": 3.8483,
+      "step": 2637312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.81097812250428e-06,
+      "loss": 3.8413,
+      "step": 2637824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.802592174993761e-06,
+      "loss": 3.8348,
+      "step": 2638336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.794206227483239e-06,
+      "loss": 3.8498,
+      "step": 2638848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.78582027997272e-06,
+      "loss": 3.8539,
+      "step": 2639360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.7774343324622e-06,
+      "loss": 3.8447,
+      "step": 2639872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 6.769048384951679e-06,
+      "loss": 3.8438,
+      "step": 2640384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.760662437441159e-06,
+      "loss": 3.8541,
+      "step": 2640896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.752292868734371e-06,
+      "loss": 3.8355,
+      "step": 2641408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.743906921223851e-06,
+      "loss": 3.8487,
+      "step": 2641920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.73552097371333e-06,
+      "loss": 3.8424,
+      "step": 2642432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.72713502620281e-06,
+      "loss": 3.8525,
+      "step": 2642944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.718749078692291e-06,
+      "loss": 3.8541,
+      "step": 2643456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.710363131181769e-06,
+      "loss": 3.8474,
+      "step": 2643968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.70197718367125e-06,
+      "loss": 3.8385,
+      "step": 2644480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.69359123616073e-06,
+      "loss": 3.8537,
+      "step": 2644992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.685221667453942e-06,
+      "loss": 3.8416,
+      "step": 2645504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.676835719943421e-06,
+      "loss": 3.8616,
+      "step": 2646016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.668449772432901e-06,
+      "loss": 3.8501,
+      "step": 2646528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.660063824922382e-06,
+      "loss": 3.8508,
+      "step": 2647040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.651694256215593e-06,
+      "loss": 3.8479,
+      "step": 2647552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.643308308705072e-06,
+      "loss": 3.8552,
+      "step": 2648064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.634922361194552e-06,
+      "loss": 3.8308,
+      "step": 2648576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.626536413684032e-06,
+      "loss": 3.8383,
+      "step": 2649088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.618150466173511e-06,
+      "loss": 3.8437,
+      "step": 2649600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.609780897466723e-06,
+      "loss": 3.8443,
+      "step": 2650112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.601394949956204e-06,
+      "loss": 3.8368,
+      "step": 2650624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.593009002445684e-06,
+      "loss": 3.8411,
+      "step": 2651136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.584623054935163e-06,
+      "loss": 3.8351,
+      "step": 2651648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.576237107424643e-06,
+      "loss": 3.8441,
+      "step": 2652160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.567851159914124e-06,
+      "loss": 3.8423,
+      "step": 2652672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.559465212403603e-06,
+      "loss": 3.8384,
+      "step": 2653184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.551079264893083e-06,
+      "loss": 3.8534,
+      "step": 2653696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.542693317382563e-06,
+      "loss": 3.8527,
+      "step": 2654208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.534307369872042e-06,
+      "loss": 3.8521,
+      "step": 2654720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.5259378011652534e-06,
+      "loss": 3.8426,
+      "step": 2655232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.517551853654734e-06,
+      "loss": 3.8524,
+      "step": 2655744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.509165906144214e-06,
+      "loss": 3.8501,
+      "step": 2656256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.500779958633693e-06,
+      "loss": 3.8428,
+      "step": 2656768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.492394011123173e-06,
+      "loss": 3.8591,
+      "step": 2657280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.484024442416385e-06,
+      "loss": 3.8406,
+      "step": 2657792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.475638494905866e-06,
+      "loss": 3.8545,
+      "step": 2658304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.467252547395344e-06,
+      "loss": 3.8355,
+      "step": 2658816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.458866599884825e-06,
+      "loss": 3.8504,
+      "step": 2659328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.450480652374305e-06,
+      "loss": 3.8321,
+      "step": 2659840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.4421110836675155e-06,
+      "loss": 3.841,
+      "step": 2660352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.4337251361569955e-06,
+      "loss": 3.8343,
+      "step": 2660864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.425339188646476e-06,
+      "loss": 3.8591,
+      "step": 2661376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.416953241135956e-06,
+      "loss": 3.8522,
+      "step": 2661888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.408567293625435e-06,
+      "loss": 3.8498,
+      "step": 2662400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.400197724918647e-06,
+      "loss": 3.8408,
+      "step": 2662912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.391811777408127e-06,
+      "loss": 3.8471,
+      "step": 2663424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.383425829897606e-06,
+      "loss": 3.8616,
+      "step": 2663936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.375039882387086e-06,
+      "loss": 3.8463,
+      "step": 2664448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.366670313680298e-06,
+      "loss": 3.8477,
+      "step": 2664960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.3582843661697784e-06,
+      "loss": 3.8545,
+      "step": 2665472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.3498984186592575e-06,
+      "loss": 3.849,
+      "step": 2665984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.3415124711487375e-06,
+      "loss": 3.8532,
+      "step": 2666496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.333142902441949e-06,
+      "loss": 3.842,
+      "step": 2667008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.324756954931428e-06,
+      "loss": 3.8506,
+      "step": 2667520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.316371007420908e-06,
+      "loss": 3.8467,
+      "step": 2668032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.307985059910389e-06,
+      "loss": 3.8463,
+      "step": 2668544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.299599112399869e-06,
+      "loss": 3.8464,
+      "step": 2669056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.291229543693079e-06,
+      "loss": 3.8529,
+      "step": 2669568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.28284359618256e-06,
+      "loss": 3.8442,
+      "step": 2670080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.27445764867204e-06,
+      "loss": 3.8465,
+      "step": 2670592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.266071701161519e-06,
+      "loss": 3.8444,
+      "step": 2671104
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.976051092147827,
+      "eval_runtime": 326.9996,
+      "eval_samples_per_second": 1166.947,
+      "eval_steps_per_second": 36.468,
+      "step": 2671200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.257685753651e-06,
+      "loss": 3.8426,
+      "step": 2671616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.2492998061404795e-06,
+      "loss": 3.8416,
+      "step": 2672128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.2409138586299595e-06,
+      "loss": 3.8515,
+      "step": 2672640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.2325279111194394e-06,
+      "loss": 3.8416,
+      "step": 2673152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.22415834241265e-06,
+      "loss": 3.869,
+      "step": 2673664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.215772394902131e-06,
+      "loss": 3.8351,
+      "step": 2674176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.20738644739161e-06,
+      "loss": 3.8462,
+      "step": 2674688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.199016878684822e-06,
+      "loss": 3.833,
+      "step": 2675200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.190630931174302e-06,
+      "loss": 3.8488,
+      "step": 2675712
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.182244983663781e-06,
+      "loss": 3.848,
+      "step": 2676224
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.173859036153262e-06,
+      "loss": 3.8441,
+      "step": 2676736
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.165473088642741e-06,
+      "loss": 3.85,
+      "step": 2677248
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.157103519935952e-06,
+      "loss": 3.8361,
+      "step": 2677760
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.148717572425432e-06,
+      "loss": 3.8425,
+      "step": 2678272
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.140331624914913e-06,
+      "loss": 3.842,
+      "step": 2678784
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.131945677404392e-06,
+      "loss": 3.8419,
+      "step": 2679296
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.123559729893872e-06,
+      "loss": 3.8409,
+      "step": 2679808
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.115173782383352e-06,
+      "loss": 3.8427,
+      "step": 2680320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.106804213676563e-06,
+      "loss": 3.8441,
+      "step": 2680832
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.098418266166044e-06,
+      "loss": 3.8615,
+      "step": 2681344
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.090032318655523e-06,
+      "loss": 3.8427,
+      "step": 2681856
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.081646371145004e-06,
+      "loss": 3.8465,
+      "step": 2682368
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.073260423634483e-06,
+      "loss": 3.8501,
+      "step": 2682880
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.064874476123963e-06,
+      "loss": 3.8574,
+      "step": 2683392
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.056488528613443e-06,
+      "loss": 3.8377,
+      "step": 2683904
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.048118959906654e-06,
+      "loss": 3.8499,
+      "step": 2684416
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.039733012396134e-06,
+      "loss": 3.8398,
+      "step": 2684928
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.031347064885614e-06,
+      "loss": 3.8419,
+      "step": 2685440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 6.022961117375093e-06,
+      "loss": 3.833,
+      "step": 2685952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.014575169864574e-06,
+      "loss": 3.843,
+      "step": 2686464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 6.006189222354053e-06,
+      "loss": 3.8434,
+      "step": 2686976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.997819653647265e-06,
+      "loss": 3.8533,
+      "step": 2687488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.989433706136745e-06,
+      "loss": 3.8465,
+      "step": 2688000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.981047758626225e-06,
+      "loss": 3.8549,
+      "step": 2688512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.972661811115705e-06,
+      "loss": 3.8448,
+      "step": 2689024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.964275863605185e-06,
+      "loss": 3.8515,
+      "step": 2689536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.955889916094665e-06,
+      "loss": 3.8395,
+      "step": 2690048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.947503968584145e-06,
+      "loss": 3.8538,
+      "step": 2690560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.939118021073625e-06,
+      "loss": 3.8339,
+      "step": 2691072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.9307320735631045e-06,
+      "loss": 3.846,
+      "step": 2691584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.922346126052584e-06,
+      "loss": 3.8458,
+      "step": 2692096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.913992936149527e-06,
+      "loss": 3.8526,
+      "step": 2692608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.905606988639007e-06,
+      "loss": 3.8421,
+      "step": 2693120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.897221041128487e-06,
+      "loss": 3.844,
+      "step": 2693632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.888835093617966e-06,
+      "loss": 3.8511,
+      "step": 2694144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.880449146107447e-06,
+      "loss": 3.8481,
+      "step": 2694656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.872079577400658e-06,
+      "loss": 3.8484,
+      "step": 2695168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.8636936298901375e-06,
+      "loss": 3.8382,
+      "step": 2695680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.8553076823796175e-06,
+      "loss": 3.8244,
+      "step": 2696192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.8469217348690975e-06,
+      "loss": 3.8493,
+      "step": 2696704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.838535787358577e-06,
+      "loss": 3.845,
+      "step": 2697216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.830166218651789e-06,
+      "loss": 3.8382,
+      "step": 2697728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.821780271141269e-06,
+      "loss": 3.8341,
+      "step": 2698240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.813394323630748e-06,
+      "loss": 3.8327,
+      "step": 2698752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.805008376120229e-06,
+      "loss": 3.8295,
+      "step": 2699264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.796622428609708e-06,
+      "loss": 3.846,
+      "step": 2699776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.788236481099189e-06,
+      "loss": 3.8349,
+      "step": 2700288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.779850533588668e-06,
+      "loss": 3.8416,
+      "step": 2700800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.771464586078148e-06,
+      "loss": 3.8477,
+      "step": 2701312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.763078638567628e-06,
+      "loss": 3.8359,
+      "step": 2701824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.754692691057108e-06,
+      "loss": 3.8352,
+      "step": 2702336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.746306743546588e-06,
+      "loss": 3.8418,
+      "step": 2702848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.737920796036068e-06,
+      "loss": 3.8231,
+      "step": 2703360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.729567606133011e-06,
+      "loss": 3.8419,
+      "step": 2703872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.72118165862249e-06,
+      "loss": 3.8449,
+      "step": 2704384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.712795711111971e-06,
+      "loss": 3.848,
+      "step": 2704896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.70440976360145e-06,
+      "loss": 3.8289,
+      "step": 2705408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.69602381609093e-06,
+      "loss": 3.8352,
+      "step": 2705920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.687654247384142e-06,
+      "loss": 3.8303,
+      "step": 2706432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.679268299873621e-06,
+      "loss": 3.8499,
+      "step": 2706944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.6708823523631016e-06,
+      "loss": 3.8476,
+      "step": 2707456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.6624964048525815e-06,
+      "loss": 3.8484,
+      "step": 2707968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.654126836145793e-06,
+      "loss": 3.8546,
+      "step": 2708480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.645740888635272e-06,
+      "loss": 3.8551,
+      "step": 2708992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.637354941124753e-06,
+      "loss": 3.8577,
+      "step": 2709504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.628968993614232e-06,
+      "loss": 3.84,
+      "step": 2710016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.620583046103712e-06,
+      "loss": 3.8469,
+      "step": 2710528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.612197098593192e-06,
+      "loss": 3.8499,
+      "step": 2711040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.603811151082672e-06,
+      "loss": 3.8395,
+      "step": 2711552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.595425203572152e-06,
+      "loss": 3.8484,
+      "step": 2712064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.587055634865363e-06,
+      "loss": 3.8434,
+      "step": 2712576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.578669687354844e-06,
+      "loss": 3.8534,
+      "step": 2713088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.570300118648054e-06,
+      "loss": 3.8481,
+      "step": 2713600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.561914171137534e-06,
+      "loss": 3.831,
+      "step": 2714112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.553528223627014e-06,
+      "loss": 3.8365,
+      "step": 2714624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.545142276116494e-06,
+      "loss": 3.8452,
+      "step": 2715136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.536756328605974e-06,
+      "loss": 3.8501,
+      "step": 2715648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.528370381095454e-06,
+      "loss": 3.8439,
+      "step": 2716160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 5.519984433584934e-06,
+      "loss": 3.8411,
+      "step": 2716672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.511598486074414e-06,
+      "loss": 3.8494,
+      "step": 2717184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.503228917367626e-06,
+      "loss": 3.838,
+      "step": 2717696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.494842969857105e-06,
+      "loss": 3.8423,
+      "step": 2718208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.486457022346585e-06,
+      "loss": 3.8439,
+      "step": 2718720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.478071074836065e-06,
+      "loss": 3.851,
+      "step": 2719232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.469685127325545e-06,
+      "loss": 3.8538,
+      "step": 2719744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.461315558618756e-06,
+      "loss": 3.8441,
+      "step": 2720256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.452929611108236e-06,
+      "loss": 3.8321,
+      "step": 2720768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.444543663597715e-06,
+      "loss": 3.8521,
+      "step": 2721280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.436157716087196e-06,
+      "loss": 3.8368,
+      "step": 2721792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.427771768576675e-06,
+      "loss": 3.8585,
+      "step": 2722304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.419402199869887e-06,
+      "loss": 3.8498,
+      "step": 2722816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.411016252359367e-06,
+      "loss": 3.8469,
+      "step": 2723328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.402630304848847e-06,
+      "loss": 3.8437,
+      "step": 2723840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.394244357338327e-06,
+      "loss": 3.8525,
+      "step": 2724352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.385858409827806e-06,
+      "loss": 3.8304,
+      "step": 2724864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.377472462317287e-06,
+      "loss": 3.8399,
+      "step": 2725376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.3691028936104975e-06,
+      "loss": 3.8354,
+      "step": 2725888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.3607169460999775e-06,
+      "loss": 3.8422,
+      "step": 2726400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.352330998589457e-06,
+      "loss": 3.8333,
+      "step": 2726912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.343945051078938e-06,
+      "loss": 3.842,
+      "step": 2727424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.335575482372149e-06,
+      "loss": 3.8358,
+      "step": 2727936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.327189534861629e-06,
+      "loss": 3.834,
+      "step": 2728448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.318803587351109e-06,
+      "loss": 3.8413,
+      "step": 2728960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.310417639840588e-06,
+      "loss": 3.8347,
+      "step": 2729472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.3020480711338e-06,
+      "loss": 3.8527,
+      "step": 2729984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.29366212362328e-06,
+      "loss": 3.8481,
+      "step": 2730496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.2852761761127596e-06,
+      "loss": 3.8495,
+      "step": 2731008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.2768902286022395e-06,
+      "loss": 3.8393,
+      "step": 2731520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.268520659895451e-06,
+      "loss": 3.8477,
+      "step": 2732032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.260134712384931e-06,
+      "loss": 3.849,
+      "step": 2732544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.251765143678143e-06,
+      "loss": 3.8385,
+      "step": 2733056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.243379196167622e-06,
+      "loss": 3.8538,
+      "step": 2733568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.234993248657103e-06,
+      "loss": 3.841,
+      "step": 2734080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.226607301146582e-06,
+      "loss": 3.8503,
+      "step": 2734592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.218221353636062e-06,
+      "loss": 3.8348,
+      "step": 2735104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.209835406125542e-06,
+      "loss": 3.8448,
+      "step": 2735616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.2014658374187525e-06,
+      "loss": 3.8306,
+      "step": 2736128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.193079889908233e-06,
+      "loss": 3.8409,
+      "step": 2736640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.184693942397713e-06,
+      "loss": 3.8333,
+      "step": 2737152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.176307994887193e-06,
+      "loss": 3.8504,
+      "step": 2737664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.167922047376673e-06,
+      "loss": 3.8447,
+      "step": 2738176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.159536099866152e-06,
+      "loss": 3.8484,
+      "step": 2738688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.151150152355633e-06,
+      "loss": 3.8385,
+      "step": 2739200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.142764204845112e-06,
+      "loss": 3.848,
+      "step": 2739712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.134378257334593e-06,
+      "loss": 3.8561,
+      "step": 2740224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.125992309824072e-06,
+      "loss": 3.8452,
+      "step": 2740736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.117639119921015e-06,
+      "loss": 3.8464,
+      "step": 2741248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.1092531724104945e-06,
+      "loss": 3.8524,
+      "step": 2741760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.100867224899975e-06,
+      "loss": 3.8435,
+      "step": 2742272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.0924812773894544e-06,
+      "loss": 3.8544,
+      "step": 2742784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.084095329878934e-06,
+      "loss": 3.8368,
+      "step": 2743296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.075709382368414e-06,
+      "loss": 3.8503,
+      "step": 2743808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.067323434857894e-06,
+      "loss": 3.8439,
+      "step": 2744320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.058953866151106e-06,
+      "loss": 3.8423,
+      "step": 2744832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.050567918640586e-06,
+      "loss": 3.8471,
+      "step": 2745344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.042181971130066e-06,
+      "loss": 3.8474,
+      "step": 2745856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.033796023619546e-06,
+      "loss": 3.8435,
+      "step": 2746368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.0254264549127574e-06,
+      "loss": 3.8418,
+      "step": 2746880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 5.0170405074022365e-06,
+      "loss": 3.8457,
+      "step": 2747392
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.9752745628356934,
+      "eval_runtime": 316.9946,
+      "eval_samples_per_second": 1203.778,
+      "eval_steps_per_second": 37.619,
+      "step": 2747520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.0086545598917165e-06,
+      "loss": 3.8378,
+      "step": 2747904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 5.0002686123811965e-06,
+      "loss": 3.8379,
+      "step": 2748416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.991882664870676e-06,
+      "loss": 3.8497,
+      "step": 2748928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.983496717360156e-06,
+      "loss": 3.8386,
+      "step": 2749440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.975110769849636e-06,
+      "loss": 3.8625,
+      "step": 2749952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.966724822339116e-06,
+      "loss": 3.8408,
+      "step": 2750464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.958338874828596e-06,
+      "loss": 3.8407,
+      "step": 2750976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.949952927318076e-06,
+      "loss": 3.8296,
+      "step": 2751488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.941566979807556e-06,
+      "loss": 3.8461,
+      "step": 2752000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.933181032297036e-06,
+      "loss": 3.8505,
+      "step": 2752512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.924795084786515e-06,
+      "loss": 3.8416,
+      "step": 2753024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.916409137275996e-06,
+      "loss": 3.8423,
+      "step": 2753536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.908039568569207e-06,
+      "loss": 3.8361,
+      "step": 2754048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.899653621058687e-06,
+      "loss": 3.8391,
+      "step": 2754560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.891267673548167e-06,
+      "loss": 3.84,
+      "step": 2755072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.882881726037647e-06,
+      "loss": 3.837,
+      "step": 2755584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.874495778527127e-06,
+      "loss": 3.8387,
+      "step": 2756096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.8661098310166066e-06,
+      "loss": 3.8459,
+      "step": 2756608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.8577238835060865e-06,
+      "loss": 3.8397,
+      "step": 2757120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.8493379359955665e-06,
+      "loss": 3.856,
+      "step": 2757632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.840951988485046e-06,
+      "loss": 3.8447,
+      "step": 2758144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.832566040974526e-06,
+      "loss": 3.8466,
+      "step": 2758656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.8241800934640055e-06,
+      "loss": 3.8469,
+      "step": 2759168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.815794145953486e-06,
+      "loss": 3.853,
+      "step": 2759680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.807424577246697e-06,
+      "loss": 3.8358,
+      "step": 2760192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.799038629736177e-06,
+      "loss": 3.845,
+      "step": 2760704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.790669061029388e-06,
+      "loss": 3.841,
+      "step": 2761216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.782283113518869e-06,
+      "loss": 3.837,
+      "step": 2761728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.773897166008348e-06,
+      "loss": 3.8314,
+      "step": 2762240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.765527597301559e-06,
+      "loss": 3.8375,
+      "step": 2762752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.757141649791039e-06,
+      "loss": 3.8462,
+      "step": 2763264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.748755702280519e-06,
+      "loss": 3.8496,
+      "step": 2763776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.740369754769999e-06,
+      "loss": 3.8447,
+      "step": 2764288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.731983807259479e-06,
+      "loss": 3.8492,
+      "step": 2764800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.723597859748959e-06,
+      "loss": 3.8435,
+      "step": 2765312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.715211912238439e-06,
+      "loss": 3.8493,
+      "step": 2765824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.706825964727918e-06,
+      "loss": 3.8415,
+      "step": 2766336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.69845639602113e-06,
+      "loss": 3.8469,
+      "step": 2766848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.69007044851061e-06,
+      "loss": 3.8349,
+      "step": 2767360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.68168450100009e-06,
+      "loss": 3.8425,
+      "step": 2767872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.67329855348957e-06,
+      "loss": 3.8421,
+      "step": 2768384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.664928984782781e-06,
+      "loss": 3.8513,
+      "step": 2768896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.656559416075992e-06,
+      "loss": 3.8418,
+      "step": 2769408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.648189847369204e-06,
+      "loss": 3.8435,
+      "step": 2769920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.639803899858684e-06,
+      "loss": 3.8543,
+      "step": 2770432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.631417952348164e-06,
+      "loss": 3.8423,
+      "step": 2770944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.623032004837644e-06,
+      "loss": 3.845,
+      "step": 2771456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.614646057327124e-06,
+      "loss": 3.8355,
+      "step": 2771968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.606260109816604e-06,
+      "loss": 3.8209,
+      "step": 2772480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.597874162306083e-06,
+      "loss": 3.8504,
+      "step": 2772992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.5894882147955635e-06,
+      "loss": 3.8435,
+      "step": 2773504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.581102267285043e-06,
+      "loss": 3.8392,
+      "step": 2774016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.572716319774523e-06,
+      "loss": 3.8262,
+      "step": 2774528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.564330372264003e-06,
+      "loss": 3.8332,
+      "step": 2775040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.5559444247534825e-06,
+      "loss": 3.8272,
+      "step": 2775552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.547558477242963e-06,
+      "loss": 3.8432,
+      "step": 2776064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.539188908536174e-06,
+      "loss": 3.8328,
+      "step": 2776576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.530802961025654e-06,
+      "loss": 3.8384,
+      "step": 2777088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.522417013515134e-06,
+      "loss": 3.8458,
+      "step": 2777600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.514031066004614e-06,
+      "loss": 3.8339,
+      "step": 2778112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.505645118494094e-06,
+      "loss": 3.8334,
+      "step": 2778624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.497259170983573e-06,
+      "loss": 3.8429,
+      "step": 2779136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.488889602276785e-06,
+      "loss": 3.8213,
+      "step": 2779648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.480520033569996e-06,
+      "loss": 3.8379,
+      "step": 2780160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.472134086059476e-06,
+      "loss": 3.8405,
+      "step": 2780672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.463748138548956e-06,
+      "loss": 3.8455,
+      "step": 2781184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.455362191038436e-06,
+      "loss": 3.8267,
+      "step": 2781696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.446976243527916e-06,
+      "loss": 3.8343,
+      "step": 2782208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.438606674821128e-06,
+      "loss": 3.8231,
+      "step": 2782720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.430220727310607e-06,
+      "loss": 3.851,
+      "step": 2783232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4218511586038185e-06,
+      "loss": 3.8445,
+      "step": 2783744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4134652110932984e-06,
+      "loss": 3.8419,
+      "step": 2784256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.405079263582778e-06,
+      "loss": 3.8549,
+      "step": 2784768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.396693316072258e-06,
+      "loss": 3.8495,
+      "step": 2785280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.388307368561738e-06,
+      "loss": 3.8618,
+      "step": 2785792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.379921421051218e-06,
+      "loss": 3.8371,
+      "step": 2786304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.371535473540698e-06,
+      "loss": 3.8432,
+      "step": 2786816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.363149526030178e-06,
+      "loss": 3.8486,
+      "step": 2787328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.354763578519658e-06,
+      "loss": 3.8341,
+      "step": 2787840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.346377631009137e-06,
+      "loss": 3.8481,
+      "step": 2788352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.337991683498618e-06,
+      "loss": 3.8418,
+      "step": 2788864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.329605735988097e-06,
+      "loss": 3.8488,
+      "step": 2789376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.321219788477578e-06,
+      "loss": 3.8449,
+      "step": 2789888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.312833840967057e-06,
+      "loss": 3.8324,
+      "step": 2790400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.304464272260269e-06,
+      "loss": 3.8413,
+      "step": 2790912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.296078324749749e-06,
+      "loss": 3.8344,
+      "step": 2791424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.287692377239228e-06,
+      "loss": 3.8498,
+      "step": 2791936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.2793064297287086e-06,
+      "loss": 3.8418,
+      "step": 2792448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.270936861021919e-06,
+      "loss": 3.8375,
+      "step": 2792960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2625509135114e-06,
+      "loss": 3.8473,
+      "step": 2793472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.254164966000879e-06,
+      "loss": 3.8409,
+      "step": 2793984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.245779018490359e-06,
+      "loss": 3.8352,
+      "step": 2794496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.237393070979839e-06,
+      "loss": 3.8391,
+      "step": 2795008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.229007123469319e-06,
+      "loss": 3.8505,
+      "step": 2795520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.220637554762531e-06,
+      "loss": 3.8485,
+      "step": 2796032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.21225160725201e-06,
+      "loss": 3.845,
+      "step": 2796544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.203865659741491e-06,
+      "loss": 3.8295,
+      "step": 2797056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.19547971223097e-06,
+      "loss": 3.8474,
+      "step": 2797568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.18709376472045e-06,
+      "loss": 3.8351,
+      "step": 2798080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.17870781720993e-06,
+      "loss": 3.8537,
+      "step": 2798592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.17032186969941e-06,
+      "loss": 3.8492,
+      "step": 2799104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.16193592218889e-06,
+      "loss": 3.8456,
+      "step": 2799616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1535499746783696e-06,
+      "loss": 3.8409,
+      "step": 2800128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1451640271678495e-06,
+      "loss": 3.8536,
+      "step": 2800640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1367780796573295e-06,
+      "loss": 3.8244,
+      "step": 2801152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.12840851095054e-06,
+      "loss": 3.8385,
+      "step": 2801664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.120022563440021e-06,
+      "loss": 3.8316,
+      "step": 2802176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.1116366159295e-06,
+      "loss": 3.8463,
+      "step": 2802688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.103250668418981e-06,
+      "loss": 3.8275,
+      "step": 2803200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.09486472090846e-06,
+      "loss": 3.8387,
+      "step": 2803712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.08647877339794e-06,
+      "loss": 3.8351,
+      "step": 2804224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.078109204691152e-06,
+      "loss": 3.833,
+      "step": 2804736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.069723257180632e-06,
+      "loss": 3.8371,
+      "step": 2805248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.061337309670112e-06,
+      "loss": 3.8352,
+      "step": 2805760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0529513621595915e-06,
+      "loss": 3.8489,
+      "step": 2806272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0445654146490715e-06,
+      "loss": 3.8476,
+      "step": 2806784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0361794671385515e-06,
+      "loss": 3.8447,
+      "step": 2807296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0277935196280306e-06,
+      "loss": 3.8379,
+      "step": 2807808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.019423950921242e-06,
+      "loss": 3.8489,
+      "step": 2808320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.011038003410722e-06,
+      "loss": 3.8428,
+      "step": 2808832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.002668434703934e-06,
+      "loss": 3.8357,
+      "step": 2809344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.994282487193413e-06,
+      "loss": 3.8537,
+      "step": 2809856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.985896539682894e-06,
+      "loss": 3.8386,
+      "step": 2810368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.977510592172373e-06,
+      "loss": 3.8465,
+      "step": 2810880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.969124644661854e-06,
+      "loss": 3.8321,
+      "step": 2811392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.960738697151333e-06,
+      "loss": 3.8447,
+      "step": 2811904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.952352749640813e-06,
+      "loss": 3.8327,
+      "step": 2812416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.9439668021302935e-06,
+      "loss": 3.8352,
+      "step": 2812928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.935597233423504e-06,
+      "loss": 3.8361,
+      "step": 2813440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.927211285912984e-06,
+      "loss": 3.8478,
+      "step": 2813952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.918825338402464e-06,
+      "loss": 3.8418,
+      "step": 2814464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.910439390891944e-06,
+      "loss": 3.847,
+      "step": 2814976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.902053443381424e-06,
+      "loss": 3.834,
+      "step": 2815488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.893683874674636e-06,
+      "loss": 3.8463,
+      "step": 2816000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.885297927164115e-06,
+      "loss": 3.8521,
+      "step": 2816512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.876911979653595e-06,
+      "loss": 3.8442,
+      "step": 2817024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8685424109468064e-06,
+      "loss": 3.8421,
+      "step": 2817536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.860156463436286e-06,
+      "loss": 3.8493,
+      "step": 2818048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.851770515925766e-06,
+      "loss": 3.8414,
+      "step": 2818560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.843384568415246e-06,
+      "loss": 3.8507,
+      "step": 2819072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.834998620904726e-06,
+      "loss": 3.8342,
+      "step": 2819584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.826612673394206e-06,
+      "loss": 3.8462,
+      "step": 2820096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.818226725883685e-06,
+      "loss": 3.8393,
+      "step": 2820608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8098407783731657e-06,
+      "loss": 3.8415,
+      "step": 2821120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8014548308626457e-06,
+      "loss": 3.8421,
+      "step": 2821632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7930688833521256e-06,
+      "loss": 3.8438,
+      "step": 2822144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.784699314645337e-06,
+      "loss": 3.8444,
+      "step": 2822656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.776313367134817e-06,
+      "loss": 3.8347,
+      "step": 2823168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7679274196242967e-06,
+      "loss": 3.8482,
+      "step": 2823680
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.9746415615081787,
+      "eval_runtime": 296.5987,
+      "eval_samples_per_second": 1286.557,
+      "eval_steps_per_second": 40.206,
+      "step": 2823840
+    }
+  ],
+  "logging_steps": 512,
+  "max_steps": 3052726,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 10,
+  "total_flos": 1.1657855111792026e+18,
+  "trial_name": null,
+  "trial_params": null
+}