diff --git "a/checkpoint-457920/trainer_state.json" "b/checkpoint-457920/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-457920/trainer_state.json"
@@ -0,0 +1,5437 @@
+{
+  "best_metric": 3.8715946674346924,
+  "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/outputs/existential-there-quantifier/transformer/4/checkpoints/checkpoint-457920",
+  "epoch": 1.0250006060157382,
+  "eval_steps": 10,
+  "global_step": 457920,
+  "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": 11.0368,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999161405248948e-05,
+      "loss": 6.8134,
+      "step": 512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.998322810497896e-05,
+      "loss": 6.1877,
+      "step": 1024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.997484215746844e-05,
+      "loss": 5.9788,
+      "step": 1536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.996645620995792e-05,
+      "loss": 5.8108,
+      "step": 2048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99580702624474e-05,
+      "loss": 5.7032,
+      "step": 2560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.994968431493688e-05,
+      "loss": 5.5999,
+      "step": 3072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.994129836742636e-05,
+      "loss": 5.5361,
+      "step": 3584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.993291241991584e-05,
+      "loss": 5.4453,
+      "step": 4096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.992452647240532e-05,
+      "loss": 5.4008,
+      "step": 4608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99161405248948e-05,
+      "loss": 5.3507,
+      "step": 5120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.990775457738428e-05,
+      "loss": 5.3098,
+      "step": 5632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.989936862987376e-05,
+      "loss": 5.2582,
+      "step": 6144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.989099906116697e-05,
+      "loss": 5.2103,
+      "step": 6656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.988261311365645e-05,
+      "loss": 5.1687,
+      "step": 7168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.987422716614593e-05,
+      "loss": 5.1357,
+      "step": 7680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.986584121863541e-05,
+      "loss": 5.1055,
+      "step": 8192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.985745527112489e-05,
+      "loss": 5.0901,
+      "step": 8704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.984906932361437e-05,
+      "loss": 5.0477,
+      "step": 9216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.984068337610385e-05,
+      "loss": 5.0274,
+      "step": 9728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.983229742859333e-05,
+      "loss": 5.0154,
+      "step": 10240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9823927859886547e-05,
+      "loss": 4.9707,
+      "step": 10752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9815541912376026e-05,
+      "loss": 4.9456,
+      "step": 11264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9807155964865506e-05,
+      "loss": 4.9307,
+      "step": 11776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9798770017354986e-05,
+      "loss": 4.9209,
+      "step": 12288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9790400448648195e-05,
+      "loss": 4.8868,
+      "step": 12800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9782030879941405e-05,
+      "loss": 4.8669,
+      "step": 13312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9773644932430885e-05,
+      "loss": 4.8606,
+      "step": 13824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9765258984920364e-05,
+      "loss": 4.8385,
+      "step": 14336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9756873037409844e-05,
+      "loss": 4.8117,
+      "step": 14848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9748487089899324e-05,
+      "loss": 4.7945,
+      "step": 15360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9740117521192533e-05,
+      "loss": 4.7903,
+      "step": 15872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9731731573682013e-05,
+      "loss": 4.7787,
+      "step": 16384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.97233456261715e-05,
+      "loss": 4.7747,
+      "step": 16896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.971495967866098e-05,
+      "loss": 4.7513,
+      "step": 17408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.970657373115046e-05,
+      "loss": 4.736,
+      "step": 17920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.969820416244367e-05,
+      "loss": 4.7379,
+      "step": 18432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.968981821493315e-05,
+      "loss": 4.7031,
+      "step": 18944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.968143226742263e-05,
+      "loss": 4.6961,
+      "step": 19456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.967304631991211e-05,
+      "loss": 4.6831,
+      "step": 19968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.966466037240159e-05,
+      "loss": 4.6832,
+      "step": 20480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.965627442489107e-05,
+      "loss": 4.6604,
+      "step": 20992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.964788847738054e-05,
+      "loss": 4.672,
+      "step": 21504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.963950252987002e-05,
+      "loss": 4.6425,
+      "step": 22016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.963113296116324e-05,
+      "loss": 4.6463,
+      "step": 22528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.962274701365272e-05,
+      "loss": 4.6366,
+      "step": 23040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.96143610661422e-05,
+      "loss": 4.6276,
+      "step": 23552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.960597511863168e-05,
+      "loss": 4.6242,
+      "step": 24064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.959758917112116e-05,
+      "loss": 4.5995,
+      "step": 24576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9589219602414374e-05,
+      "loss": 4.587,
+      "step": 25088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.958085003370758e-05,
+      "loss": 4.6012,
+      "step": 25600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.957246408619706e-05,
+      "loss": 4.5854,
+      "step": 26112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.956407813868654e-05,
+      "loss": 4.5687,
+      "step": 26624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9555692191176016e-05,
+      "loss": 4.5577,
+      "step": 27136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.954732262246923e-05,
+      "loss": 4.5648,
+      "step": 27648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.953893667495871e-05,
+      "loss": 4.5469,
+      "step": 28160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.953055072744819e-05,
+      "loss": 4.5599,
+      "step": 28672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.952216477993767e-05,
+      "loss": 4.5342,
+      "step": 29184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.951377883242715e-05,
+      "loss": 4.5313,
+      "step": 29696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.950539288491663e-05,
+      "loss": 4.535,
+      "step": 30208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.949700693740611e-05,
+      "loss": 4.5165,
+      "step": 30720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.948862098989559e-05,
+      "loss": 4.4989,
+      "step": 31232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.94802514211888e-05,
+      "loss": 4.5047,
+      "step": 31744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.947186547367828e-05,
+      "loss": 4.488,
+      "step": 32256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.946347952616776e-05,
+      "loss": 4.4898,
+      "step": 32768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.945509357865724e-05,
+      "loss": 4.4974,
+      "step": 33280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.944670763114672e-05,
+      "loss": 4.4911,
+      "step": 33792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.943833806243993e-05,
+      "loss": 4.4782,
+      "step": 34304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.942995211492941e-05,
+      "loss": 4.4663,
+      "step": 34816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.942156616741889e-05,
+      "loss": 4.4584,
+      "step": 35328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.941318021990837e-05,
+      "loss": 4.4699,
+      "step": 35840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9404810651201585e-05,
+      "loss": 4.466,
+      "step": 36352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9396441082494794e-05,
+      "loss": 4.4578,
+      "step": 36864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9388055134984274e-05,
+      "loss": 4.456,
+      "step": 37376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9379669187473754e-05,
+      "loss": 4.4582,
+      "step": 37888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9371283239963234e-05,
+      "loss": 4.4521,
+      "step": 38400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.936291367125644e-05,
+      "loss": 4.4261,
+      "step": 38912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.935452772374592e-05,
+      "loss": 4.4359,
+      "step": 39424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.93461417762354e-05,
+      "loss": 4.4321,
+      "step": 39936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.933775582872488e-05,
+      "loss": 4.4188,
+      "step": 40448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.932936988121436e-05,
+      "loss": 4.4303,
+      "step": 40960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.932098393370384e-05,
+      "loss": 4.4199,
+      "step": 41472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.931259798619332e-05,
+      "loss": 4.4254,
+      "step": 41984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.930422841748654e-05,
+      "loss": 4.4167,
+      "step": 42496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.929584246997602e-05,
+      "loss": 4.3887,
+      "step": 43008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.92874565224655e-05,
+      "loss": 4.3998,
+      "step": 43520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.927907057495498e-05,
+      "loss": 4.3993,
+      "step": 44032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.927068462744446e-05,
+      "loss": 4.3994,
+      "step": 44544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.926229867993394e-05,
+      "loss": 4.3916,
+      "step": 45056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.925392911122715e-05,
+      "loss": 4.3838,
+      "step": 45568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.924554316371663e-05,
+      "loss": 4.3862,
+      "step": 46080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.923715721620611e-05,
+      "loss": 4.375,
+      "step": 46592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.922878764749932e-05,
+      "loss": 4.368,
+      "step": 47104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.92204016999888e-05,
+      "loss": 4.3685,
+      "step": 47616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.921201575247828e-05,
+      "loss": 4.3654,
+      "step": 48128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.920362980496776e-05,
+      "loss": 4.3794,
+      "step": 48640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.919524385745724e-05,
+      "loss": 4.3545,
+      "step": 49152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.918685790994672e-05,
+      "loss": 4.3487,
+      "step": 49664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.91784719624362e-05,
+      "loss": 4.347,
+      "step": 50176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9170086014925676e-05,
+      "loss": 4.3576,
+      "step": 50688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9161700067415156e-05,
+      "loss": 4.3531,
+      "step": 51200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9153314119904636e-05,
+      "loss": 4.3398,
+      "step": 51712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9144928172394116e-05,
+      "loss": 4.3393,
+      "step": 52224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9136542224883596e-05,
+      "loss": 4.3214,
+      "step": 52736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9128172656176805e-05,
+      "loss": 4.3374,
+      "step": 53248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9119786708666285e-05,
+      "loss": 4.322,
+      "step": 53760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9111400761155765e-05,
+      "loss": 4.3206,
+      "step": 54272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.910303119244898e-05,
+      "loss": 4.3238,
+      "step": 54784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.909464524493846e-05,
+      "loss": 4.3228,
+      "step": 55296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.908625929742794e-05,
+      "loss": 4.3172,
+      "step": 55808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.907787334991742e-05,
+      "loss": 4.3153,
+      "step": 56320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.90694874024069e-05,
+      "loss": 4.3074,
+      "step": 56832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.906110145489638e-05,
+      "loss": 4.3136,
+      "step": 57344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.905273188618959e-05,
+      "loss": 4.3028,
+      "step": 57856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.904434593867907e-05,
+      "loss": 4.3023,
+      "step": 58368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.903595999116855e-05,
+      "loss": 4.3076,
+      "step": 58880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.902757404365803e-05,
+      "loss": 4.3048,
+      "step": 59392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.901920447495124e-05,
+      "loss": 4.3025,
+      "step": 59904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.901081852744072e-05,
+      "loss": 4.2937,
+      "step": 60416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.90024325799302e-05,
+      "loss": 4.2983,
+      "step": 60928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8994046632419686e-05,
+      "loss": 4.2932,
+      "step": 61440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8985660684909166e-05,
+      "loss": 4.2793,
+      "step": 61952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8977274737398646e-05,
+      "loss": 4.2973,
+      "step": 62464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8968888789888125e-05,
+      "loss": 4.2813,
+      "step": 62976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8960502842377605e-05,
+      "loss": 4.2894,
+      "step": 63488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8952133273670815e-05,
+      "loss": 4.2709,
+      "step": 64000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8943747326160294e-05,
+      "loss": 4.2661,
+      "step": 64512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8935361378649774e-05,
+      "loss": 4.2649,
+      "step": 65024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8926991809942984e-05,
+      "loss": 4.2682,
+      "step": 65536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8918605862432463e-05,
+      "loss": 4.258,
+      "step": 66048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8910219914921943e-05,
+      "loss": 4.2754,
+      "step": 66560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8901833967411423e-05,
+      "loss": 4.272,
+      "step": 67072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.88934480199009e-05,
+      "loss": 4.2555,
+      "step": 67584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.888506207239038e-05,
+      "loss": 4.2596,
+      "step": 68096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.887667612487986e-05,
+      "loss": 4.2631,
+      "step": 68608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.886829017736934e-05,
+      "loss": 4.2597,
+      "step": 69120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.885992060866256e-05,
+      "loss": 4.2637,
+      "step": 69632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.885155103995577e-05,
+      "loss": 4.2595,
+      "step": 70144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.884316509244525e-05,
+      "loss": 4.2466,
+      "step": 70656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.883477914493473e-05,
+      "loss": 4.2528,
+      "step": 71168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.882639319742421e-05,
+      "loss": 4.2464,
+      "step": 71680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.881800724991369e-05,
+      "loss": 4.24,
+      "step": 72192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.880962130240316e-05,
+      "loss": 4.2503,
+      "step": 72704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.880123535489264e-05,
+      "loss": 4.2374,
+      "step": 73216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.879286578618586e-05,
+      "loss": 4.2263,
+      "step": 73728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.878447983867534e-05,
+      "loss": 4.2481,
+      "step": 74240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.877609389116482e-05,
+      "loss": 4.2278,
+      "step": 74752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.87677079436543e-05,
+      "loss": 4.228,
+      "step": 75264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.875932199614378e-05,
+      "loss": 4.2307,
+      "step": 75776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.875093604863326e-05,
+      "loss": 4.2235,
+      "step": 76288
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.195763111114502,
+      "eval_runtime": 304.3449,
+      "eval_samples_per_second": 1253.811,
+      "eval_steps_per_second": 39.183,
+      "step": 76320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.874255010112274e-05,
+      "loss": 4.2068,
+      "step": 76800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.873416415361222e-05,
+      "loss": 4.2199,
+      "step": 77312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8725794584905426e-05,
+      "loss": 4.2339,
+      "step": 77824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8717408637394906e-05,
+      "loss": 4.2115,
+      "step": 78336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8709022689884386e-05,
+      "loss": 4.2262,
+      "step": 78848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8700636742373866e-05,
+      "loss": 4.2001,
+      "step": 79360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8692267173667075e-05,
+      "loss": 4.2133,
+      "step": 79872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8683881226156555e-05,
+      "loss": 4.1979,
+      "step": 80384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.867551165744977e-05,
+      "loss": 4.2083,
+      "step": 80896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.866712570993925e-05,
+      "loss": 4.2035,
+      "step": 81408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.865873976242873e-05,
+      "loss": 4.209,
+      "step": 81920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.865035381491821e-05,
+      "loss": 4.2051,
+      "step": 82432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.864198424621142e-05,
+      "loss": 4.1916,
+      "step": 82944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.86335982987009e-05,
+      "loss": 4.1833,
+      "step": 83456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.862521235119038e-05,
+      "loss": 4.1845,
+      "step": 83968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.861682640367986e-05,
+      "loss": 4.1852,
+      "step": 84480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.860844045616934e-05,
+      "loss": 4.1949,
+      "step": 84992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.860005450865882e-05,
+      "loss": 4.1823,
+      "step": 85504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.85916685611483e-05,
+      "loss": 4.1885,
+      "step": 86016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.858328261363778e-05,
+      "loss": 4.205,
+      "step": 86528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.857489666612726e-05,
+      "loss": 4.179,
+      "step": 87040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.856651071861674e-05,
+      "loss": 4.1732,
+      "step": 87552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8558124771106226e-05,
+      "loss": 4.1771,
+      "step": 88064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8549755202399435e-05,
+      "loss": 4.186,
+      "step": 88576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8541385633692644e-05,
+      "loss": 4.1671,
+      "step": 89088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8532999686182124e-05,
+      "loss": 4.171,
+      "step": 89600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8524613738671604e-05,
+      "loss": 4.1727,
+      "step": 90112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8516227791161084e-05,
+      "loss": 4.1665,
+      "step": 90624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8507841843650564e-05,
+      "loss": 4.1514,
+      "step": 91136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8499455896140044e-05,
+      "loss": 4.1567,
+      "step": 91648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8491069948629524e-05,
+      "loss": 4.1553,
+      "step": 92160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.848270037992273e-05,
+      "loss": 4.1666,
+      "step": 92672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.847431443241221e-05,
+      "loss": 4.1708,
+      "step": 93184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.846592848490169e-05,
+      "loss": 4.1529,
+      "step": 93696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.845754253739117e-05,
+      "loss": 4.1501,
+      "step": 94208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.844915658988065e-05,
+      "loss": 4.1648,
+      "step": 94720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.844077064237013e-05,
+      "loss": 4.1387,
+      "step": 95232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.843238469485961e-05,
+      "loss": 4.1448,
+      "step": 95744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.842401512615282e-05,
+      "loss": 4.1371,
+      "step": 96256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.84156291786423e-05,
+      "loss": 4.1433,
+      "step": 96768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.840724323113178e-05,
+      "loss": 4.1352,
+      "step": 97280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.839885728362126e-05,
+      "loss": 4.1532,
+      "step": 97792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.839048771491447e-05,
+      "loss": 4.1375,
+      "step": 98304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.838210176740395e-05,
+      "loss": 4.1386,
+      "step": 98816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.837371581989343e-05,
+      "loss": 4.1346,
+      "step": 99328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.836532987238292e-05,
+      "loss": 4.1408,
+      "step": 99840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.83569439248724e-05,
+      "loss": 4.133,
+      "step": 100352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.834855797736188e-05,
+      "loss": 4.1239,
+      "step": 100864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.834017202985136e-05,
+      "loss": 4.117,
+      "step": 101376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8331802461144566e-05,
+      "loss": 4.1385,
+      "step": 101888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8323416513634046e-05,
+      "loss": 4.1267,
+      "step": 102400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8315030566123526e-05,
+      "loss": 4.1169,
+      "step": 102912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8306644618613006e-05,
+      "loss": 4.1171,
+      "step": 103424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8298275049906215e-05,
+      "loss": 4.1231,
+      "step": 103936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8289889102395695e-05,
+      "loss": 4.112,
+      "step": 104448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8281503154885175e-05,
+      "loss": 4.13,
+      "step": 104960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8273117207374655e-05,
+      "loss": 4.113,
+      "step": 105472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8264731259864135e-05,
+      "loss": 4.1149,
+      "step": 105984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8256345312353615e-05,
+      "loss": 4.124,
+      "step": 106496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8247959364843095e-05,
+      "loss": 4.1038,
+      "step": 107008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.823957341733258e-05,
+      "loss": 4.1002,
+      "step": 107520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.823120384862579e-05,
+      "loss": 4.1103,
+      "step": 108032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.822281790111527e-05,
+      "loss": 4.0924,
+      "step": 108544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.821443195360475e-05,
+      "loss": 4.0985,
+      "step": 109056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.820604600609423e-05,
+      "loss": 4.1096,
+      "step": 109568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.819766005858371e-05,
+      "loss": 4.1122,
+      "step": 110080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8189274111073184e-05,
+      "loss": 4.1009,
+      "step": 110592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8180888163562664e-05,
+      "loss": 4.0957,
+      "step": 111104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.817253497365961e-05,
+      "loss": 4.0861,
+      "step": 111616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.816414902614909e-05,
+      "loss": 4.1051,
+      "step": 112128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.815576307863857e-05,
+      "loss": 4.1013,
+      "step": 112640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.814737713112805e-05,
+      "loss": 4.1016,
+      "step": 113152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8138991183617535e-05,
+      "loss": 4.102,
+      "step": 113664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.813060523610701e-05,
+      "loss": 4.1048,
+      "step": 114176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8122235667400224e-05,
+      "loss": 4.1031,
+      "step": 114688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8113849719889704e-05,
+      "loss": 4.0819,
+      "step": 115200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8105480151182914e-05,
+      "loss": 4.0935,
+      "step": 115712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8097094203672393e-05,
+      "loss": 4.095,
+      "step": 116224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8088708256161873e-05,
+      "loss": 4.0788,
+      "step": 116736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.808032230865135e-05,
+      "loss": 4.1021,
+      "step": 117248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8071936361140827e-05,
+      "loss": 4.0938,
+      "step": 117760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8063550413630307e-05,
+      "loss": 4.0971,
+      "step": 118272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8055164466119786e-05,
+      "loss": 4.0954,
+      "step": 118784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.804677851860927e-05,
+      "loss": 4.071,
+      "step": 119296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.803840894990248e-05,
+      "loss": 4.0829,
+      "step": 119808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.803002300239196e-05,
+      "loss": 4.0819,
+      "step": 120320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.802163705488144e-05,
+      "loss": 4.0849,
+      "step": 120832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.801325110737092e-05,
+      "loss": 4.0833,
+      "step": 121344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.800488153866413e-05,
+      "loss": 4.0746,
+      "step": 121856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.799649559115361e-05,
+      "loss": 4.084,
+      "step": 122368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.798810964364309e-05,
+      "loss": 4.0771,
+      "step": 122880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.797972369613257e-05,
+      "loss": 4.0647,
+      "step": 123392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.797133774862205e-05,
+      "loss": 4.0716,
+      "step": 123904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.796295180111153e-05,
+      "loss": 4.0769,
+      "step": 124416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.795458223240474e-05,
+      "loss": 4.0869,
+      "step": 124928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.794619628489423e-05,
+      "loss": 4.0682,
+      "step": 125440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.793781033738371e-05,
+      "loss": 4.0545,
+      "step": 125952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.792942438987319e-05,
+      "loss": 4.0634,
+      "step": 126464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.792103844236267e-05,
+      "loss": 4.074,
+      "step": 126976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.791265249485215e-05,
+      "loss": 4.071,
+      "step": 127488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.790426654734163e-05,
+      "loss": 4.0658,
+      "step": 128000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7895880599831107e-05,
+      "loss": 4.0598,
+      "step": 128512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7887494652320587e-05,
+      "loss": 4.0503,
+      "step": 129024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7879125083613796e-05,
+      "loss": 4.0659,
+      "step": 129536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7870739136103276e-05,
+      "loss": 4.0511,
+      "step": 130048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7862353188592756e-05,
+      "loss": 4.0559,
+      "step": 130560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7853983619885965e-05,
+      "loss": 4.0591,
+      "step": 131072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7845597672375445e-05,
+      "loss": 4.0593,
+      "step": 131584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7837211724864925e-05,
+      "loss": 4.0524,
+      "step": 132096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.782882577735441e-05,
+      "loss": 4.0583,
+      "step": 132608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.782043982984389e-05,
+      "loss": 4.0494,
+      "step": 133120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.781205388233337e-05,
+      "loss": 4.0549,
+      "step": 133632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7803667934822844e-05,
+      "loss": 4.0477,
+      "step": 134144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7795281987312324e-05,
+      "loss": 4.0458,
+      "step": 134656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.778691241860554e-05,
+      "loss": 4.0589,
+      "step": 135168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.777852647109502e-05,
+      "loss": 4.0574,
+      "step": 135680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.777015690238823e-05,
+      "loss": 4.053,
+      "step": 136192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.776177095487771e-05,
+      "loss": 4.0468,
+      "step": 136704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.775338500736719e-05,
+      "loss": 4.0515,
+      "step": 137216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.774499905985666e-05,
+      "loss": 4.0501,
+      "step": 137728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.773661311234615e-05,
+      "loss": 4.0367,
+      "step": 138240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.772822716483563e-05,
+      "loss": 4.057,
+      "step": 138752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.771984121732511e-05,
+      "loss": 4.041,
+      "step": 139264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.771145526981459e-05,
+      "loss": 4.0523,
+      "step": 139776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.77030857011078e-05,
+      "loss": 4.0341,
+      "step": 140288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.769469975359728e-05,
+      "loss": 4.0357,
+      "step": 140800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.768631380608676e-05,
+      "loss": 4.0319,
+      "step": 141312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.767792785857624e-05,
+      "loss": 4.0344,
+      "step": 141824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.766955828986945e-05,
+      "loss": 4.0316,
+      "step": 142336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.766117234235893e-05,
+      "loss": 4.0465,
+      "step": 142848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.765278639484841e-05,
+      "loss": 4.0479,
+      "step": 143360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.764440044733789e-05,
+      "loss": 4.0357,
+      "step": 143872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.763601449982737e-05,
+      "loss": 4.0289,
+      "step": 144384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.762764493112058e-05,
+      "loss": 4.0387,
+      "step": 144896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.761925898361006e-05,
+      "loss": 4.0371,
+      "step": 145408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.761087303609954e-05,
+      "loss": 4.0461,
+      "step": 145920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.760248708858902e-05,
+      "loss": 4.0374,
+      "step": 146432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.759411751988223e-05,
+      "loss": 4.0331,
+      "step": 146944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.758573157237171e-05,
+      "loss": 4.0284,
+      "step": 147456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.757734562486119e-05,
+      "loss": 4.0381,
+      "step": 147968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.756895967735067e-05,
+      "loss": 4.0264,
+      "step": 148480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.756059010864388e-05,
+      "loss": 4.0306,
+      "step": 148992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.755220416113336e-05,
+      "loss": 4.0283,
+      "step": 149504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.754381821362284e-05,
+      "loss": 4.0197,
+      "step": 150016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.753544864491605e-05,
+      "loss": 4.0409,
+      "step": 150528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7527062697405536e-05,
+      "loss": 4.0207,
+      "step": 151040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7518676749895016e-05,
+      "loss": 4.0199,
+      "step": 151552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7510290802384496e-05,
+      "loss": 4.0233,
+      "step": 152064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7501904854873976e-05,
+      "loss": 4.0188,
+      "step": 152576
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.028048515319824,
+      "eval_runtime": 306.2895,
+      "eval_samples_per_second": 1245.851,
+      "eval_steps_per_second": 38.934,
+      "step": 152640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7493518907363456e-05,
+      "loss": 4.0033,
+      "step": 153088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7485132959852936e-05,
+      "loss": 4.0187,
+      "step": 153600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7476747012342416e-05,
+      "loss": 4.0332,
+      "step": 154112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7468361064831896e-05,
+      "loss": 4.0153,
+      "step": 154624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7459975117321376e-05,
+      "loss": 4.0297,
+      "step": 155136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.745158916981085e-05,
+      "loss": 4.0079,
+      "step": 155648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.744320322230033e-05,
+      "loss": 4.021,
+      "step": 156160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.743481727478981e-05,
+      "loss": 3.9999,
+      "step": 156672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.742643132727929e-05,
+      "loss": 4.017,
+      "step": 157184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.741804537976877e-05,
+      "loss": 4.0142,
+      "step": 157696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.740965943225825e-05,
+      "loss": 4.0125,
+      "step": 158208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7401289863551465e-05,
+      "loss": 4.0213,
+      "step": 158720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.739292029484468e-05,
+      "loss": 4.0019,
+      "step": 159232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7384534347334154e-05,
+      "loss": 3.9977,
+      "step": 159744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7376148399823634e-05,
+      "loss": 3.996,
+      "step": 160256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7367762452313114e-05,
+      "loss": 4.004,
+      "step": 160768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7359376504802594e-05,
+      "loss": 4.0087,
+      "step": 161280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7350990557292074e-05,
+      "loss": 4.0022,
+      "step": 161792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7342604609781554e-05,
+      "loss": 4.0041,
+      "step": 162304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.733423504107476e-05,
+      "loss": 4.0239,
+      "step": 162816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.732584909356424e-05,
+      "loss": 3.9994,
+      "step": 163328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.731746314605372e-05,
+      "loss": 3.9946,
+      "step": 163840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.73090771985432e-05,
+      "loss": 4.0021,
+      "step": 164352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.730070762983642e-05,
+      "loss": 4.0077,
+      "step": 164864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.72923216823259e-05,
+      "loss": 3.9908,
+      "step": 165376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.728393573481538e-05,
+      "loss": 3.9969,
+      "step": 165888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.727554978730486e-05,
+      "loss": 3.9984,
+      "step": 166400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.726718021859807e-05,
+      "loss": 3.9948,
+      "step": 166912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.725879427108755e-05,
+      "loss": 3.9809,
+      "step": 167424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.725040832357703e-05,
+      "loss": 3.9865,
+      "step": 167936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.724202237606651e-05,
+      "loss": 3.9889,
+      "step": 168448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7233652807359716e-05,
+      "loss": 3.9934,
+      "step": 168960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7225266859849196e-05,
+      "loss": 4.0065,
+      "step": 169472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7216880912338676e-05,
+      "loss": 3.9845,
+      "step": 169984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7208494964828156e-05,
+      "loss": 3.9848,
+      "step": 170496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.720010901731764e-05,
+      "loss": 3.996,
+      "step": 171008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.719172306980712e-05,
+      "loss": 3.9793,
+      "step": 171520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.71833371222966e-05,
+      "loss": 3.9776,
+      "step": 172032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.717496755358981e-05,
+      "loss": 3.9745,
+      "step": 172544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.716658160607929e-05,
+      "loss": 3.9827,
+      "step": 173056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.715819565856877e-05,
+      "loss": 3.9709,
+      "step": 173568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.714980971105825e-05,
+      "loss": 3.9936,
+      "step": 174080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.714144014235146e-05,
+      "loss": 3.9745,
+      "step": 174592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.713305419484094e-05,
+      "loss": 3.9844,
+      "step": 175104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.712466824733042e-05,
+      "loss": 3.9788,
+      "step": 175616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.71162822998199e-05,
+      "loss": 3.9783,
+      "step": 176128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.710791273111311e-05,
+      "loss": 3.9774,
+      "step": 176640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.709952678360259e-05,
+      "loss": 3.9654,
+      "step": 177152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.709114083609208e-05,
+      "loss": 3.9623,
+      "step": 177664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.708275488858156e-05,
+      "loss": 3.9837,
+      "step": 178176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7074385319874766e-05,
+      "loss": 3.9745,
+      "step": 178688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7065999372364246e-05,
+      "loss": 3.9643,
+      "step": 179200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7057613424853726e-05,
+      "loss": 3.9668,
+      "step": 179712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7049227477343206e-05,
+      "loss": 3.9747,
+      "step": 180224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7040857908636415e-05,
+      "loss": 3.9541,
+      "step": 180736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7032471961125895e-05,
+      "loss": 3.9765,
+      "step": 181248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7024086013615375e-05,
+      "loss": 3.9645,
+      "step": 181760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7015700066104855e-05,
+      "loss": 3.9645,
+      "step": 182272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7007330497398064e-05,
+      "loss": 3.9803,
+      "step": 182784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6998944549887544e-05,
+      "loss": 3.9573,
+      "step": 183296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.699055860237703e-05,
+      "loss": 3.9535,
+      "step": 183808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.698217265486651e-05,
+      "loss": 3.9659,
+      "step": 184320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.697380308615972e-05,
+      "loss": 3.9483,
+      "step": 184832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.69654171386492e-05,
+      "loss": 3.9539,
+      "step": 185344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.695703119113868e-05,
+      "loss": 3.9648,
+      "step": 185856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.694864524362816e-05,
+      "loss": 3.9693,
+      "step": 186368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.694027567492137e-05,
+      "loss": 3.9537,
+      "step": 186880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.693188972741085e-05,
+      "loss": 3.9554,
+      "step": 187392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.692352015870406e-05,
+      "loss": 3.9434,
+      "step": 187904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.691513421119354e-05,
+      "loss": 3.9656,
+      "step": 188416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.690674826368302e-05,
+      "loss": 3.9581,
+      "step": 188928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.68983623161725e-05,
+      "loss": 3.9644,
+      "step": 189440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6889976368661984e-05,
+      "loss": 3.9602,
+      "step": 189952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.688159042115146e-05,
+      "loss": 3.9644,
+      "step": 190464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.687320447364094e-05,
+      "loss": 3.9693,
+      "step": 190976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.686481852613042e-05,
+      "loss": 3.9435,
+      "step": 191488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.685644895742363e-05,
+      "loss": 3.9572,
+      "step": 192000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6848063009913106e-05,
+      "loss": 3.9608,
+      "step": 192512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6839677062402586e-05,
+      "loss": 3.944,
+      "step": 193024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6831291114892066e-05,
+      "loss": 3.9648,
+      "step": 193536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6822905167381546e-05,
+      "loss": 3.9586,
+      "step": 194048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6814519219871026e-05,
+      "loss": 3.9626,
+      "step": 194560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6806133272360506e-05,
+      "loss": 3.9618,
+      "step": 195072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6797747324849986e-05,
+      "loss": 3.9387,
+      "step": 195584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.67893777561432e-05,
+      "loss": 3.9516,
+      "step": 196096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.678099180863268e-05,
+      "loss": 3.9484,
+      "step": 196608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.677260586112216e-05,
+      "loss": 3.9564,
+      "step": 197120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.676421991361164e-05,
+      "loss": 3.9478,
+      "step": 197632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.675585034490485e-05,
+      "loss": 3.9475,
+      "step": 198144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.674746439739433e-05,
+      "loss": 3.9522,
+      "step": 198656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.673907844988381e-05,
+      "loss": 3.9495,
+      "step": 199168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.673069250237329e-05,
+      "loss": 3.9351,
+      "step": 199680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.67223229336665e-05,
+      "loss": 3.9416,
+      "step": 200192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.671393698615598e-05,
+      "loss": 3.9505,
+      "step": 200704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.670556741744919e-05,
+      "loss": 3.9569,
+      "step": 201216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6697181469938675e-05,
+      "loss": 3.9474,
+      "step": 201728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6688795522428155e-05,
+      "loss": 3.9269,
+      "step": 202240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6680409574917635e-05,
+      "loss": 3.9391,
+      "step": 202752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6672023627407115e-05,
+      "loss": 3.9474,
+      "step": 203264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6663654058700324e-05,
+      "loss": 3.944,
+      "step": 203776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6655268111189804e-05,
+      "loss": 3.9414,
+      "step": 204288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6646882163679284e-05,
+      "loss": 3.9361,
+      "step": 204800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6638496216168764e-05,
+      "loss": 3.9264,
+      "step": 205312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6630110268658244e-05,
+      "loss": 3.9458,
+      "step": 205824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.662174069995145e-05,
+      "loss": 3.9211,
+      "step": 206336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.661335475244093e-05,
+      "loss": 3.9351,
+      "step": 206848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.660496880493041e-05,
+      "loss": 3.9315,
+      "step": 207360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.659658285741989e-05,
+      "loss": 3.943,
+      "step": 207872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.658821328871311e-05,
+      "loss": 3.9272,
+      "step": 208384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.657982734120259e-05,
+      "loss": 3.9401,
+      "step": 208896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.657144139369207e-05,
+      "loss": 3.9298,
+      "step": 209408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.656305544618155e-05,
+      "loss": 3.9292,
+      "step": 209920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.655466949867103e-05,
+      "loss": 3.9344,
+      "step": 210432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.654628355116051e-05,
+      "loss": 3.9229,
+      "step": 210944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.653789760364999e-05,
+      "loss": 3.9399,
+      "step": 211456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.652951165613947e-05,
+      "loss": 3.9386,
+      "step": 211968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.652115846623641e-05,
+      "loss": 3.9365,
+      "step": 212480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.651277251872589e-05,
+      "loss": 3.9297,
+      "step": 212992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.650438657121537e-05,
+      "loss": 3.9341,
+      "step": 213504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.649600062370485e-05,
+      "loss": 3.9345,
+      "step": 214016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.648761467619433e-05,
+      "loss": 3.9155,
+      "step": 214528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.647922872868381e-05,
+      "loss": 3.9416,
+      "step": 215040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6470842781173294e-05,
+      "loss": 3.9279,
+      "step": 215552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.646245683366277e-05,
+      "loss": 3.9346,
+      "step": 216064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.645408726495598e-05,
+      "loss": 3.9222,
+      "step": 216576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.644570131744546e-05,
+      "loss": 3.9244,
+      "step": 217088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.643731536993494e-05,
+      "loss": 3.9128,
+      "step": 217600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6428929422424416e-05,
+      "loss": 3.9204,
+      "step": 218112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.642055985371763e-05,
+      "loss": 3.919,
+      "step": 218624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.641217390620711e-05,
+      "loss": 3.9338,
+      "step": 219136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.640378795869659e-05,
+      "loss": 3.9356,
+      "step": 219648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6395402011186065e-05,
+      "loss": 3.9235,
+      "step": 220160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.638703244247928e-05,
+      "loss": 3.9154,
+      "step": 220672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.637864649496876e-05,
+      "loss": 3.9256,
+      "step": 221184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.637026054745824e-05,
+      "loss": 3.9266,
+      "step": 221696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.636187459994772e-05,
+      "loss": 3.9315,
+      "step": 222208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6353505031240936e-05,
+      "loss": 3.9275,
+      "step": 222720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6345119083730416e-05,
+      "loss": 3.9174,
+      "step": 223232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6336749515023625e-05,
+      "loss": 3.9248,
+      "step": 223744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6328363567513105e-05,
+      "loss": 3.9292,
+      "step": 224256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6319977620002585e-05,
+      "loss": 3.915,
+      "step": 224768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.631159167249206e-05,
+      "loss": 3.9225,
+      "step": 225280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.630320572498154e-05,
+      "loss": 3.9175,
+      "step": 225792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.629481977747102e-05,
+      "loss": 3.9107,
+      "step": 226304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6286450208764234e-05,
+      "loss": 3.9311,
+      "step": 226816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6278064261253714e-05,
+      "loss": 3.9154,
+      "step": 227328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6269678313743194e-05,
+      "loss": 3.912,
+      "step": 227840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6261292366232674e-05,
+      "loss": 3.9144,
+      "step": 228352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6252906418722154e-05,
+      "loss": 3.91,
+      "step": 228864
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.9538862705230713,
+      "eval_runtime": 302.4885,
+      "eval_samples_per_second": 1261.506,
+      "eval_steps_per_second": 39.423,
+      "step": 228960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.624453685001536e-05,
+      "loss": 3.9037,
+      "step": 229376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.623615090250484e-05,
+      "loss": 3.9107,
+      "step": 229888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.622776495499432e-05,
+      "loss": 3.9273,
+      "step": 230400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.62193790074838e-05,
+      "loss": 3.9119,
+      "step": 230912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.621099305997328e-05,
+      "loss": 3.9315,
+      "step": 231424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.620260711246276e-05,
+      "loss": 3.8984,
+      "step": 231936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.619422116495224e-05,
+      "loss": 3.9195,
+      "step": 232448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.618585159624545e-05,
+      "loss": 3.8957,
+      "step": 232960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.617746564873494e-05,
+      "loss": 3.9131,
+      "step": 233472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.616907970122442e-05,
+      "loss": 3.9148,
+      "step": 233984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.61606937537139e-05,
+      "loss": 3.9083,
+      "step": 234496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.615230780620338e-05,
+      "loss": 3.9165,
+      "step": 235008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.614393823749659e-05,
+      "loss": 3.8992,
+      "step": 235520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.613555228998607e-05,
+      "loss": 3.8971,
+      "step": 236032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.612716634247555e-05,
+      "loss": 3.8962,
+      "step": 236544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.611878039496503e-05,
+      "loss": 3.9008,
+      "step": 237056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6110410826258237e-05,
+      "loss": 3.9076,
+      "step": 237568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6102024878747717e-05,
+      "loss": 3.9012,
+      "step": 238080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6093638931237196e-05,
+      "loss": 3.9041,
+      "step": 238592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6085252983726676e-05,
+      "loss": 3.9225,
+      "step": 239104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.607688341501989e-05,
+      "loss": 3.9027,
+      "step": 239616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.606849746750937e-05,
+      "loss": 3.8997,
+      "step": 240128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.606011151999885e-05,
+      "loss": 3.9046,
+      "step": 240640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.605172557248833e-05,
+      "loss": 3.9062,
+      "step": 241152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.604335600378154e-05,
+      "loss": 3.8915,
+      "step": 241664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.603497005627102e-05,
+      "loss": 3.9013,
+      "step": 242176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.60265841087605e-05,
+      "loss": 3.9022,
+      "step": 242688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.601819816124998e-05,
+      "loss": 3.8943,
+      "step": 243200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.600981221373946e-05,
+      "loss": 3.8868,
+      "step": 243712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.600142626622894e-05,
+      "loss": 3.8853,
+      "step": 244224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.599305669752215e-05,
+      "loss": 3.8975,
+      "step": 244736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.598467075001163e-05,
+      "loss": 3.8959,
+      "step": 245248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.597628480250111e-05,
+      "loss": 3.9098,
+      "step": 245760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.596789885499059e-05,
+      "loss": 3.8971,
+      "step": 246272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.595951290748008e-05,
+      "loss": 3.8862,
+      "step": 246784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5951143338773286e-05,
+      "loss": 3.8985,
+      "step": 247296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5942757391262766e-05,
+      "loss": 3.889,
+      "step": 247808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5934371443752246e-05,
+      "loss": 3.8807,
+      "step": 248320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5925985496241726e-05,
+      "loss": 3.8831,
+      "step": 248832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.59175995487312e-05,
+      "loss": 3.887,
+      "step": 249344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.590921360122068e-05,
+      "loss": 3.8802,
+      "step": 249856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.590082765371016e-05,
+      "loss": 3.9009,
+      "step": 250368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.589245808500337e-05,
+      "loss": 3.8842,
+      "step": 250880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5884088516296584e-05,
+      "loss": 3.8912,
+      "step": 251392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5875702568786064e-05,
+      "loss": 3.8897,
+      "step": 251904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5867316621275544e-05,
+      "loss": 3.8834,
+      "step": 252416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5858930673765024e-05,
+      "loss": 3.8882,
+      "step": 252928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5850544726254504e-05,
+      "loss": 3.8727,
+      "step": 253440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5842158778743984e-05,
+      "loss": 3.8633,
+      "step": 253952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5833772831233464e-05,
+      "loss": 3.8964,
+      "step": 254464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.582540326252667e-05,
+      "loss": 3.8859,
+      "step": 254976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.581701731501615e-05,
+      "loss": 3.8793,
+      "step": 255488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.580863136750563e-05,
+      "loss": 3.8741,
+      "step": 256000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.580024541999511e-05,
+      "loss": 3.8816,
+      "step": 256512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.579187585128832e-05,
+      "loss": 3.8669,
+      "step": 257024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.57834899037778e-05,
+      "loss": 3.8855,
+      "step": 257536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.577510395626728e-05,
+      "loss": 3.8802,
+      "step": 258048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.576671800875677e-05,
+      "loss": 3.878,
+      "step": 258560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.575833206124625e-05,
+      "loss": 3.8896,
+      "step": 259072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.574994611373573e-05,
+      "loss": 3.8682,
+      "step": 259584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.574156016622521e-05,
+      "loss": 3.8664,
+      "step": 260096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.573319059751842e-05,
+      "loss": 3.8826,
+      "step": 260608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.57248046500079e-05,
+      "loss": 3.8575,
+      "step": 261120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.571641870249738e-05,
+      "loss": 3.863,
+      "step": 261632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.570803275498686e-05,
+      "loss": 3.8807,
+      "step": 262144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.569964680747634e-05,
+      "loss": 3.8832,
+      "step": 262656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.569126085996582e-05,
+      "loss": 3.8624,
+      "step": 263168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.56828749124553e-05,
+      "loss": 3.8739,
+      "step": 263680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.567448896494478e-05,
+      "loss": 3.8554,
+      "step": 264192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5666119396237986e-05,
+      "loss": 3.8799,
+      "step": 264704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5657733448727466e-05,
+      "loss": 3.8728,
+      "step": 265216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5649347501216946e-05,
+      "loss": 3.8802,
+      "step": 265728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.564096155370643e-05,
+      "loss": 3.8711,
+      "step": 266240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.563257560619591e-05,
+      "loss": 3.8786,
+      "step": 266752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.562420603748912e-05,
+      "loss": 3.8855,
+      "step": 267264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.56158200899786e-05,
+      "loss": 3.8558,
+      "step": 267776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.560743414246808e-05,
+      "loss": 3.8762,
+      "step": 268288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5599048194957555e-05,
+      "loss": 3.8783,
+      "step": 268800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5590662247447035e-05,
+      "loss": 3.8542,
+      "step": 269312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.558229267874025e-05,
+      "loss": 3.8836,
+      "step": 269824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.557390673122973e-05,
+      "loss": 3.8754,
+      "step": 270336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5565520783719204e-05,
+      "loss": 3.8765,
+      "step": 270848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5557134836208684e-05,
+      "loss": 3.8805,
+      "step": 271360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.554874888869817e-05,
+      "loss": 3.8586,
+      "step": 271872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.554036294118765e-05,
+      "loss": 3.8616,
+      "step": 272384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.553199337248086e-05,
+      "loss": 3.8662,
+      "step": 272896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.552360742497034e-05,
+      "loss": 3.8729,
+      "step": 273408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.551522147745982e-05,
+      "loss": 3.8638,
+      "step": 273920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.55068355299493e-05,
+      "loss": 3.8668,
+      "step": 274432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.549844958243878e-05,
+      "loss": 3.8659,
+      "step": 274944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.549008001373199e-05,
+      "loss": 3.8741,
+      "step": 275456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.548169406622147e-05,
+      "loss": 3.8544,
+      "step": 275968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.547330811871095e-05,
+      "loss": 3.8592,
+      "step": 276480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.546492217120043e-05,
+      "loss": 3.8692,
+      "step": 276992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.545655260249364e-05,
+      "loss": 3.8778,
+      "step": 277504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5448166654983124e-05,
+      "loss": 3.8687,
+      "step": 278016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5439780707472604e-05,
+      "loss": 3.8462,
+      "step": 278528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5431394759962084e-05,
+      "loss": 3.8575,
+      "step": 279040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.542302519125529e-05,
+      "loss": 3.8649,
+      "step": 279552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.541463924374477e-05,
+      "loss": 3.8661,
+      "step": 280064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.540625329623425e-05,
+      "loss": 3.8652,
+      "step": 280576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.539786734872373e-05,
+      "loss": 3.8569,
+      "step": 281088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.538949778001694e-05,
+      "loss": 3.8476,
+      "step": 281600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.538111183250642e-05,
+      "loss": 3.864,
+      "step": 282112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.53727258849959e-05,
+      "loss": 3.8461,
+      "step": 282624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.536433993748538e-05,
+      "loss": 3.8556,
+      "step": 283136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.535597036877859e-05,
+      "loss": 3.85,
+      "step": 283648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.534758442126808e-05,
+      "loss": 3.8645,
+      "step": 284160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.533919847375756e-05,
+      "loss": 3.8494,
+      "step": 284672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.533081252624704e-05,
+      "loss": 3.8662,
+      "step": 285184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.532242657873652e-05,
+      "loss": 3.8469,
+      "step": 285696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.531405701002973e-05,
+      "loss": 3.8539,
+      "step": 286208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.530567106251921e-05,
+      "loss": 3.8554,
+      "step": 286720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.529728511500869e-05,
+      "loss": 3.8511,
+      "step": 287232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5288899167498167e-05,
+      "loss": 3.8584,
+      "step": 287744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5280529598791376e-05,
+      "loss": 3.8647,
+      "step": 288256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5272143651280856e-05,
+      "loss": 3.8579,
+      "step": 288768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5263757703770336e-05,
+      "loss": 3.852,
+      "step": 289280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5255371756259816e-05,
+      "loss": 3.8574,
+      "step": 289792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5246985808749295e-05,
+      "loss": 3.858,
+      "step": 290304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.523861624004251e-05,
+      "loss": 3.8398,
+      "step": 290816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.523023029253199e-05,
+      "loss": 3.8636,
+      "step": 291328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.522184434502147e-05,
+      "loss": 3.8486,
+      "step": 291840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.521345839751095e-05,
+      "loss": 3.8621,
+      "step": 292352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.520507245000043e-05,
+      "loss": 3.85,
+      "step": 292864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.519670288129364e-05,
+      "loss": 3.8495,
+      "step": 293376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.518831693378312e-05,
+      "loss": 3.8367,
+      "step": 293888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.51799309862726e-05,
+      "loss": 3.8497,
+      "step": 294400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.517154503876208e-05,
+      "loss": 3.8454,
+      "step": 294912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.516317547005529e-05,
+      "loss": 3.8537,
+      "step": 295424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.515478952254477e-05,
+      "loss": 3.859,
+      "step": 295936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.514640357503425e-05,
+      "loss": 3.8537,
+      "step": 296448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.513801762752373e-05,
+      "loss": 3.8374,
+      "step": 296960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.512963168001321e-05,
+      "loss": 3.8531,
+      "step": 297472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5121262111306425e-05,
+      "loss": 3.8536,
+      "step": 297984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5112876163795905e-05,
+      "loss": 3.8569,
+      "step": 298496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5104490216285385e-05,
+      "loss": 3.8533,
+      "step": 299008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5096104268774865e-05,
+      "loss": 3.8444,
+      "step": 299520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5087734700068074e-05,
+      "loss": 3.8488,
+      "step": 300032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5079348752557554e-05,
+      "loss": 3.8579,
+      "step": 300544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5070962805047034e-05,
+      "loss": 3.8412,
+      "step": 301056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5062576857536514e-05,
+      "loss": 3.8505,
+      "step": 301568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.505420728882972e-05,
+      "loss": 3.8401,
+      "step": 302080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.50458213413192e-05,
+      "loss": 3.8439,
+      "step": 302592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.503743539380868e-05,
+      "loss": 3.8579,
+      "step": 303104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.502904944629816e-05,
+      "loss": 3.8443,
+      "step": 303616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.502066349878764e-05,
+      "loss": 3.8407,
+      "step": 304128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.501229393008086e-05,
+      "loss": 3.8382,
+      "step": 304640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.500390798257034e-05,
+      "loss": 3.842,
+      "step": 305152
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.9126405715942383,
+      "eval_runtime": 301.972,
+      "eval_samples_per_second": 1263.664,
+      "eval_steps_per_second": 39.49,
+      "step": 305280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.499552203505981e-05,
+      "loss": 3.8269,
+      "step": 305664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.498713608754929e-05,
+      "loss": 3.8402,
+      "step": 306176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.497875014003877e-05,
+      "loss": 3.8511,
+      "step": 306688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.497036419252825e-05,
+      "loss": 3.8433,
+      "step": 307200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.496197824501773e-05,
+      "loss": 3.8576,
+      "step": 307712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.495359229750721e-05,
+      "loss": 3.8348,
+      "step": 308224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.494520634999669e-05,
+      "loss": 3.8434,
+      "step": 308736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.493682040248617e-05,
+      "loss": 3.8299,
+      "step": 309248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.492845083377939e-05,
+      "loss": 3.8385,
+      "step": 309760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.492006488626887e-05,
+      "loss": 3.8445,
+      "step": 310272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.491167893875835e-05,
+      "loss": 3.8406,
+      "step": 310784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.490329299124783e-05,
+      "loss": 3.8472,
+      "step": 311296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4894923422541036e-05,
+      "loss": 3.8299,
+      "step": 311808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4886537475030516e-05,
+      "loss": 3.8302,
+      "step": 312320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4878151527519996e-05,
+      "loss": 3.8279,
+      "step": 312832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4869765580009476e-05,
+      "loss": 3.8321,
+      "step": 313344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4861396011302685e-05,
+      "loss": 3.84,
+      "step": 313856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4853010063792165e-05,
+      "loss": 3.8319,
+      "step": 314368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4844624116281645e-05,
+      "loss": 3.8362,
+      "step": 314880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4836238168771125e-05,
+      "loss": 3.8541,
+      "step": 315392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.482786860006434e-05,
+      "loss": 3.8363,
+      "step": 315904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.481948265255382e-05,
+      "loss": 3.8328,
+      "step": 316416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.48110967050433e-05,
+      "loss": 3.8348,
+      "step": 316928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.480271075753278e-05,
+      "loss": 3.8362,
+      "step": 317440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.479434118882599e-05,
+      "loss": 3.8243,
+      "step": 317952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.478595524131547e-05,
+      "loss": 3.8355,
+      "step": 318464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.477756929380495e-05,
+      "loss": 3.8307,
+      "step": 318976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.476918334629443e-05,
+      "loss": 3.8292,
+      "step": 319488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.476079739878391e-05,
+      "loss": 3.8197,
+      "step": 320000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.475242783007712e-05,
+      "loss": 3.8227,
+      "step": 320512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.47440418825666e-05,
+      "loss": 3.8267,
+      "step": 321024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.473565593505608e-05,
+      "loss": 3.8305,
+      "step": 321536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.472726998754556e-05,
+      "loss": 3.8398,
+      "step": 322048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4718900418838775e-05,
+      "loss": 3.8299,
+      "step": 322560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4710514471328255e-05,
+      "loss": 3.8175,
+      "step": 323072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4702128523817734e-05,
+      "loss": 3.8322,
+      "step": 323584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4693742576307214e-05,
+      "loss": 3.8296,
+      "step": 324096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4685373007600424e-05,
+      "loss": 3.8124,
+      "step": 324608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4676987060089903e-05,
+      "loss": 3.8148,
+      "step": 325120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4668601112579383e-05,
+      "loss": 3.824,
+      "step": 325632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.466021516506886e-05,
+      "loss": 3.8143,
+      "step": 326144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.465182921755834e-05,
+      "loss": 3.8357,
+      "step": 326656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.464345964885155e-05,
+      "loss": 3.8213,
+      "step": 327168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.463507370134103e-05,
+      "loss": 3.8225,
+      "step": 327680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.462668775383051e-05,
+      "loss": 3.8243,
+      "step": 328192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.461830180631999e-05,
+      "loss": 3.8252,
+      "step": 328704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.460993223761321e-05,
+      "loss": 3.8232,
+      "step": 329216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.460154629010269e-05,
+      "loss": 3.8089,
+      "step": 329728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.459316034259217e-05,
+      "loss": 3.7961,
+      "step": 330240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.458477439508165e-05,
+      "loss": 3.8367,
+      "step": 330752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.457640482637486e-05,
+      "loss": 3.817,
+      "step": 331264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.456801887886434e-05,
+      "loss": 3.8192,
+      "step": 331776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.455963293135382e-05,
+      "loss": 3.8147,
+      "step": 332288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.455124698384329e-05,
+      "loss": 3.8109,
+      "step": 332800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4542877415136506e-05,
+      "loss": 3.8052,
+      "step": 333312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4534491467625986e-05,
+      "loss": 3.8196,
+      "step": 333824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4526105520115466e-05,
+      "loss": 3.8166,
+      "step": 334336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4517719572604946e-05,
+      "loss": 3.8133,
+      "step": 334848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.450935000389816e-05,
+      "loss": 3.8315,
+      "step": 335360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.450096405638764e-05,
+      "loss": 3.8024,
+      "step": 335872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.449257810887712e-05,
+      "loss": 3.8005,
+      "step": 336384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4484192161366595e-05,
+      "loss": 3.8238,
+      "step": 336896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.447582259265981e-05,
+      "loss": 3.7922,
+      "step": 337408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.446743664514929e-05,
+      "loss": 3.7988,
+      "step": 337920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4459050697638764e-05,
+      "loss": 3.8222,
+      "step": 338432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4450664750128244e-05,
+      "loss": 3.8223,
+      "step": 338944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4442278802617724e-05,
+      "loss": 3.802,
+      "step": 339456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.443390923391094e-05,
+      "loss": 3.8124,
+      "step": 339968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.442552328640042e-05,
+      "loss": 3.7958,
+      "step": 340480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4417153717693636e-05,
+      "loss": 3.8144,
+      "step": 340992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4408767770183116e-05,
+      "loss": 3.8104,
+      "step": 341504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4400381822672595e-05,
+      "loss": 3.8161,
+      "step": 342016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.439199587516207e-05,
+      "loss": 3.8095,
+      "step": 342528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.438360992765155e-05,
+      "loss": 3.8183,
+      "step": 343040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.437522398014103e-05,
+      "loss": 3.825,
+      "step": 343552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.436683803263051e-05,
+      "loss": 3.7988,
+      "step": 344064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.435845208511999e-05,
+      "loss": 3.8108,
+      "step": 344576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.43500825164132e-05,
+      "loss": 3.8182,
+      "step": 345088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.434169656890268e-05,
+      "loss": 3.7938,
+      "step": 345600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.433331062139216e-05,
+      "loss": 3.8203,
+      "step": 346112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.432492467388164e-05,
+      "loss": 3.8186,
+      "step": 346624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.431655510517485e-05,
+      "loss": 3.8163,
+      "step": 347136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.430816915766433e-05,
+      "loss": 3.82,
+      "step": 347648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.429978321015381e-05,
+      "loss": 3.8021,
+      "step": 348160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.429139726264329e-05,
+      "loss": 3.7982,
+      "step": 348672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.42830276939365e-05,
+      "loss": 3.8073,
+      "step": 349184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.427464174642598e-05,
+      "loss": 3.817,
+      "step": 349696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.426625579891546e-05,
+      "loss": 3.8087,
+      "step": 350208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.425786985140494e-05,
+      "loss": 3.8042,
+      "step": 350720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.424950028269815e-05,
+      "loss": 3.8039,
+      "step": 351232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.424111433518763e-05,
+      "loss": 3.8148,
+      "step": 351744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.423272838767711e-05,
+      "loss": 3.7997,
+      "step": 352256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.422434244016659e-05,
+      "loss": 3.7994,
+      "step": 352768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.421595649265607e-05,
+      "loss": 3.808,
+      "step": 353280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4207603302753016e-05,
+      "loss": 3.8187,
+      "step": 353792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4199217355242496e-05,
+      "loss": 3.8094,
+      "step": 354304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4190831407731976e-05,
+      "loss": 3.7919,
+      "step": 354816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4182445460221456e-05,
+      "loss": 3.7978,
+      "step": 355328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4174059512710936e-05,
+      "loss": 3.8051,
+      "step": 355840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4165673565200416e-05,
+      "loss": 3.8091,
+      "step": 356352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4157287617689896e-05,
+      "loss": 3.805,
+      "step": 356864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4148901670179376e-05,
+      "loss": 3.8023,
+      "step": 357376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4140532101472585e-05,
+      "loss": 3.7899,
+      "step": 357888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4132146153962065e-05,
+      "loss": 3.8009,
+      "step": 358400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4123760206451545e-05,
+      "loss": 3.7899,
+      "step": 358912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4115374258941025e-05,
+      "loss": 3.8026,
+      "step": 359424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.410700469023424e-05,
+      "loss": 3.7852,
+      "step": 359936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.409861874272372e-05,
+      "loss": 3.8148,
+      "step": 360448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.40902327952132e-05,
+      "loss": 3.7922,
+      "step": 360960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.408184684770268e-05,
+      "loss": 3.8012,
+      "step": 361472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.407346090019216e-05,
+      "loss": 3.7946,
+      "step": 361984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.406509133148537e-05,
+      "loss": 3.7959,
+      "step": 362496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.405670538397485e-05,
+      "loss": 3.798,
+      "step": 363008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.404831943646433e-05,
+      "loss": 3.7952,
+      "step": 363520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.403993348895381e-05,
+      "loss": 3.7998,
+      "step": 364032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.403156392024702e-05,
+      "loss": 3.8074,
+      "step": 364544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.40231779727365e-05,
+      "loss": 3.7992,
+      "step": 365056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.401479202522598e-05,
+      "loss": 3.7976,
+      "step": 365568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4006406077715465e-05,
+      "loss": 3.7998,
+      "step": 366080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3998036509008674e-05,
+      "loss": 3.8053,
+      "step": 366592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3989650561498154e-05,
+      "loss": 3.7804,
+      "step": 367104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3981264613987634e-05,
+      "loss": 3.8073,
+      "step": 367616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3972878666477114e-05,
+      "loss": 3.791,
+      "step": 368128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.396450909777032e-05,
+      "loss": 3.805,
+      "step": 368640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.39561231502598e-05,
+      "loss": 3.7916,
+      "step": 369152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.394773720274928e-05,
+      "loss": 3.7958,
+      "step": 369664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.393935125523876e-05,
+      "loss": 3.7857,
+      "step": 370176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.393098168653197e-05,
+      "loss": 3.7916,
+      "step": 370688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.392259573902145e-05,
+      "loss": 3.7943,
+      "step": 371200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.391420979151093e-05,
+      "loss": 3.7971,
+      "step": 371712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.390582384400042e-05,
+      "loss": 3.7991,
+      "step": 372224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.38974378964899e-05,
+      "loss": 3.8028,
+      "step": 372736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.388906832778311e-05,
+      "loss": 3.7809,
+      "step": 373248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.388068238027259e-05,
+      "loss": 3.7966,
+      "step": 373760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.387229643276207e-05,
+      "loss": 3.8018,
+      "step": 374272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.386391048525155e-05,
+      "loss": 3.7996,
+      "step": 374784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.385554091654476e-05,
+      "loss": 3.795,
+      "step": 375296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.384715496903424e-05,
+      "loss": 3.7967,
+      "step": 375808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.383876902152372e-05,
+      "loss": 3.7903,
+      "step": 376320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.38303830740132e-05,
+      "loss": 3.8036,
+      "step": 376832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3822013505306406e-05,
+      "loss": 3.7849,
+      "step": 377344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3813627557795886e-05,
+      "loss": 3.7967,
+      "step": 377856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.380524161028537e-05,
+      "loss": 3.7866,
+      "step": 378368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3796855662774846e-05,
+      "loss": 3.7893,
+      "step": 378880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.378848609406806e-05,
+      "loss": 3.803,
+      "step": 379392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.378010014655754e-05,
+      "loss": 3.7929,
+      "step": 379904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.377171419904702e-05,
+      "loss": 3.7885,
+      "step": 380416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3763328251536495e-05,
+      "loss": 3.7802,
+      "step": 380928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.375495868282971e-05,
+      "loss": 3.7897,
+      "step": 381440
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 3.886935234069824,
+      "eval_runtime": 302.3733,
+      "eval_samples_per_second": 1261.986,
+      "eval_steps_per_second": 39.438,
+      "step": 381600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.374657273531919e-05,
+      "loss": 3.7834,
+      "step": 381952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.373818678780867e-05,
+      "loss": 3.7835,
+      "step": 382464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3729800840298144e-05,
+      "loss": 3.7993,
+      "step": 382976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.372143127159136e-05,
+      "loss": 3.7898,
+      "step": 383488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.371304532408084e-05,
+      "loss": 3.8058,
+      "step": 384000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.370465937657032e-05,
+      "loss": 3.7817,
+      "step": 384512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.36962734290598e-05,
+      "loss": 3.7842,
+      "step": 385024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.368788748154928e-05,
+      "loss": 3.7809,
+      "step": 385536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3679517912842495e-05,
+      "loss": 3.7821,
+      "step": 386048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3671148344135704e-05,
+      "loss": 3.7915,
+      "step": 386560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3662762396625184e-05,
+      "loss": 3.7913,
+      "step": 387072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3654376449114664e-05,
+      "loss": 3.7887,
+      "step": 387584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3645990501604144e-05,
+      "loss": 3.7831,
+      "step": 388096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.363760455409362e-05,
+      "loss": 3.7771,
+      "step": 388608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.36292186065831e-05,
+      "loss": 3.7753,
+      "step": 389120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.362083265907258e-05,
+      "loss": 3.7795,
+      "step": 389632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.361246309036579e-05,
+      "loss": 3.7855,
+      "step": 390144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.360407714285527e-05,
+      "loss": 3.7817,
+      "step": 390656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.359569119534475e-05,
+      "loss": 3.7854,
+      "step": 391168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.358730524783423e-05,
+      "loss": 3.7994,
+      "step": 391680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.357891930032371e-05,
+      "loss": 3.7879,
+      "step": 392192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.357054973161692e-05,
+      "loss": 3.7747,
+      "step": 392704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.35621637841064e-05,
+      "loss": 3.7834,
+      "step": 393216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.355377783659588e-05,
+      "loss": 3.7839,
+      "step": 393728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.354540826788909e-05,
+      "loss": 3.7748,
+      "step": 394240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.353702232037857e-05,
+      "loss": 3.7847,
+      "step": 394752
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.352863637286805e-05,
+      "loss": 3.7804,
+      "step": 395264
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.352025042535753e-05,
+      "loss": 3.7795,
+      "step": 395776
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.351186447784701e-05,
+      "loss": 3.7738,
+      "step": 396288
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.350349490914023e-05,
+      "loss": 3.7658,
+      "step": 396800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.349510896162971e-05,
+      "loss": 3.7762,
+      "step": 397312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3486723014119187e-05,
+      "loss": 3.7794,
+      "step": 397824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3478337066608667e-05,
+      "loss": 3.7873,
+      "step": 398336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3469967497901876e-05,
+      "loss": 3.7821,
+      "step": 398848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3461581550391356e-05,
+      "loss": 3.7656,
+      "step": 399360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3453195602880836e-05,
+      "loss": 3.786,
+      "step": 399872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3444809655370316e-05,
+      "loss": 3.7766,
+      "step": 400384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3436440086663525e-05,
+      "loss": 3.7583,
+      "step": 400896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3428054139153005e-05,
+      "loss": 3.7703,
+      "step": 401408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3419668191642485e-05,
+      "loss": 3.7704,
+      "step": 401920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.34112986229357e-05,
+      "loss": 3.7673,
+      "step": 402432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.340291267542518e-05,
+      "loss": 3.7827,
+      "step": 402944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.339452672791466e-05,
+      "loss": 3.7719,
+      "step": 403456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.338614078040414e-05,
+      "loss": 3.7708,
+      "step": 403968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.337777121169735e-05,
+      "loss": 3.7732,
+      "step": 404480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.336938526418683e-05,
+      "loss": 3.7735,
+      "step": 404992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.336099931667631e-05,
+      "loss": 3.7778,
+      "step": 405504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.335261336916579e-05,
+      "loss": 3.7597,
+      "step": 406016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3344243800459e-05,
+      "loss": 3.7421,
+      "step": 406528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.333585785294848e-05,
+      "loss": 3.789,
+      "step": 407040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.332747190543796e-05,
+      "loss": 3.7675,
+      "step": 407552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.331908595792744e-05,
+      "loss": 3.7734,
+      "step": 408064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3310716389220654e-05,
+      "loss": 3.7633,
+      "step": 408576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3302330441710134e-05,
+      "loss": 3.7625,
+      "step": 409088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3293944494199614e-05,
+      "loss": 3.7594,
+      "step": 409600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3285558546689094e-05,
+      "loss": 3.7667,
+      "step": 410112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.32771889779823e-05,
+      "loss": 3.7715,
+      "step": 410624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.326880303047178e-05,
+      "loss": 3.7601,
+      "step": 411136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.326041708296126e-05,
+      "loss": 3.7851,
+      "step": 411648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.325203113545074e-05,
+      "loss": 3.7592,
+      "step": 412160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.324366156674395e-05,
+      "loss": 3.7505,
+      "step": 412672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.323527561923343e-05,
+      "loss": 3.7731,
+      "step": 413184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.322688967172291e-05,
+      "loss": 3.7476,
+      "step": 413696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.321850372421239e-05,
+      "loss": 3.7483,
+      "step": 414208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.321013415550561e-05,
+      "loss": 3.7721,
+      "step": 414720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.320174820799509e-05,
+      "loss": 3.7703,
+      "step": 415232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.319336226048457e-05,
+      "loss": 3.7562,
+      "step": 415744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.318497631297405e-05,
+      "loss": 3.7657,
+      "step": 416256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.317660674426726e-05,
+      "loss": 3.7468,
+      "step": 416768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.316822079675674e-05,
+      "loss": 3.7676,
+      "step": 417280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3159834849246217e-05,
+      "loss": 3.7601,
+      "step": 417792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3151448901735697e-05,
+      "loss": 3.77,
+      "step": 418304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3143079333028906e-05,
+      "loss": 3.7633,
+      "step": 418816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3134693385518386e-05,
+      "loss": 3.7674,
+      "step": 419328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3126307438007866e-05,
+      "loss": 3.7756,
+      "step": 419840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3117921490497346e-05,
+      "loss": 3.7509,
+      "step": 420352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.310955192179056e-05,
+      "loss": 3.7653,
+      "step": 420864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.310116597428004e-05,
+      "loss": 3.7683,
+      "step": 421376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.309278002676952e-05,
+      "loss": 3.7503,
+      "step": 421888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3084394079259e-05,
+      "loss": 3.7712,
+      "step": 422400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.307602451055221e-05,
+      "loss": 3.7738,
+      "step": 422912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.306763856304169e-05,
+      "loss": 3.7675,
+      "step": 423424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.305925261553117e-05,
+      "loss": 3.7774,
+      "step": 423936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.305086666802065e-05,
+      "loss": 3.7553,
+      "step": 424448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.304249709931386e-05,
+      "loss": 3.7506,
+      "step": 424960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.303411115180334e-05,
+      "loss": 3.7622,
+      "step": 425472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.302572520429282e-05,
+      "loss": 3.7653,
+      "step": 425984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.30173392567823e-05,
+      "loss": 3.7673,
+      "step": 426496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3008969688075515e-05,
+      "loss": 3.7568,
+      "step": 427008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.3000583740564995e-05,
+      "loss": 3.756,
+      "step": 427520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2992197793054475e-05,
+      "loss": 3.7677,
+      "step": 428032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2983811845543955e-05,
+      "loss": 3.7546,
+      "step": 428544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2975442276837164e-05,
+      "loss": 3.7525,
+      "step": 429056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2967056329326644e-05,
+      "loss": 3.7597,
+      "step": 429568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2958670381816124e-05,
+      "loss": 3.7743,
+      "step": 430080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2950284434305604e-05,
+      "loss": 3.7643,
+      "step": 430592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.294191486559881e-05,
+      "loss": 3.7448,
+      "step": 431104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.293352891808829e-05,
+      "loss": 3.7527,
+      "step": 431616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.292514297057777e-05,
+      "loss": 3.7611,
+      "step": 432128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.291675702306725e-05,
+      "loss": 3.7636,
+      "step": 432640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.290838745436047e-05,
+      "loss": 3.7609,
+      "step": 433152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.290000150684995e-05,
+      "loss": 3.758,
+      "step": 433664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.289161555933943e-05,
+      "loss": 3.745,
+      "step": 434176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.288322961182891e-05,
+      "loss": 3.7525,
+      "step": 434688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.287486004312212e-05,
+      "loss": 3.7502,
+      "step": 435200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.28664740956116e-05,
+      "loss": 3.7549,
+      "step": 435712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.285808814810108e-05,
+      "loss": 3.7407,
+      "step": 436224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.284970220059056e-05,
+      "loss": 3.7701,
+      "step": 436736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.284133263188377e-05,
+      "loss": 3.7469,
+      "step": 437248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2832946684373247e-05,
+      "loss": 3.756,
+      "step": 437760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2824560736862727e-05,
+      "loss": 3.7449,
+      "step": 438272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2816174789352207e-05,
+      "loss": 3.758,
+      "step": 438784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.280780522064542e-05,
+      "loss": 3.7519,
+      "step": 439296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.27994192731349e-05,
+      "loss": 3.7471,
+      "step": 439808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.279103332562438e-05,
+      "loss": 3.7564,
+      "step": 440320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.278264737811386e-05,
+      "loss": 3.7646,
+      "step": 440832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.277427780940707e-05,
+      "loss": 3.749,
+      "step": 441344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.276589186189655e-05,
+      "loss": 3.7615,
+      "step": 441856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.275750591438603e-05,
+      "loss": 3.7485,
+      "step": 442368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.274911996687551e-05,
+      "loss": 3.7603,
+      "step": 442880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.274075039816872e-05,
+      "loss": 3.7413,
+      "step": 443392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.27323644506582e-05,
+      "loss": 3.7616,
+      "step": 443904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.272397850314768e-05,
+      "loss": 3.7457,
+      "step": 444416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.271559255563716e-05,
+      "loss": 3.7603,
+      "step": 444928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2707222986930376e-05,
+      "loss": 3.7503,
+      "step": 445440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2698837039419856e-05,
+      "loss": 3.75,
+      "step": 445952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2690451091909336e-05,
+      "loss": 3.7393,
+      "step": 446464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2682081523202545e-05,
+      "loss": 3.7495,
+      "step": 446976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2673695575692025e-05,
+      "loss": 3.75,
+      "step": 447488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2665309628181505e-05,
+      "loss": 3.756,
+      "step": 448000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2656923680670985e-05,
+      "loss": 3.7495,
+      "step": 448512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2648554111964194e-05,
+      "loss": 3.7606,
+      "step": 449024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2640168164453674e-05,
+      "loss": 3.7412,
+      "step": 449536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2631782216943154e-05,
+      "loss": 3.7522,
+      "step": 450048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2623396269432634e-05,
+      "loss": 3.7557,
+      "step": 450560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.261502670072584e-05,
+      "loss": 3.7589,
+      "step": 451072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.260664075321533e-05,
+      "loss": 3.7492,
+      "step": 451584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.259825480570481e-05,
+      "loss": 3.7521,
+      "step": 452096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.258986885819429e-05,
+      "loss": 3.7456,
+      "step": 452608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.25814992894875e-05,
+      "loss": 3.764,
+      "step": 453120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.257311334197698e-05,
+      "loss": 3.7415,
+      "step": 453632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.256472739446646e-05,
+      "loss": 3.7547,
+      "step": 454144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.255634144695594e-05,
+      "loss": 3.7442,
+      "step": 454656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.254797187824915e-05,
+      "loss": 3.7477,
+      "step": 455168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.253958593073863e-05,
+      "loss": 3.7567,
+      "step": 455680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.253119998322811e-05,
+      "loss": 3.7467,
+      "step": 456192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.252281403571758e-05,
+      "loss": 3.751,
+      "step": 456704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.25144444670108e-05,
+      "loss": 3.7341,
+      "step": 457216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.250605851950028e-05,
+      "loss": 3.7491,
+      "step": 457728
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 3.8715946674346924,
+      "eval_runtime": 302.2449,
+      "eval_samples_per_second": 1262.522,
+      "eval_steps_per_second": 39.455,
+      "step": 457920
+    }
+  ],
+  "logging_steps": 512,
+  "max_steps": 3052726,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 10,
+  "total_flos": 3.1738385361250714e+17,
+  "trial_name": null,
+  "trial_params": null
+}