diff --git "a/checkpoint-1450080/trainer_state.json" "b/checkpoint-1450080/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-1450080/trainer_state.json"
@@ -0,0 +1,17169 @@
+{
+  "best_metric": 4.056296348571777,
+  "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/outputs/pp-mod-subj2/lstm/3/checkpoints/checkpoint-1450080",
+  "epoch": 0.025000606015738065,
+  "eval_steps": 10,
+  "global_step": 1450080,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999998362119627e-05,
+      "loss": 10.8194,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999161405248948e-05,
+      "loss": 7.5719,
+      "step": 512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.998322810497896e-05,
+      "loss": 7.067,
+      "step": 1024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.997484215746844e-05,
+      "loss": 6.9956,
+      "step": 1536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.996645620995792e-05,
+      "loss": 6.9424,
+      "step": 2048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99580702624474e-05,
+      "loss": 6.8558,
+      "step": 2560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.994968431493688e-05,
+      "loss": 6.6935,
+      "step": 3072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.994129836742636e-05,
+      "loss": 6.5846,
+      "step": 3584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.993291241991584e-05,
+      "loss": 6.4975,
+      "step": 4096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.992452647240532e-05,
+      "loss": 6.4297,
+      "step": 4608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99161405248948e-05,
+      "loss": 6.3811,
+      "step": 5120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.990775457738428e-05,
+      "loss": 6.3144,
+      "step": 5632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.989936862987376e-05,
+      "loss": 6.263,
+      "step": 6144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.989098268236324e-05,
+      "loss": 6.1904,
+      "step": 6656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.988261311365645e-05,
+      "loss": 6.136,
+      "step": 7168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.987422716614593e-05,
+      "loss": 6.067,
+      "step": 7680
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.986584121863541e-05,
+      "loss": 6.035,
+      "step": 8192
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.985745527112489e-05,
+      "loss": 5.9783,
+      "step": 8704
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.984906932361437e-05,
+      "loss": 5.9322,
+      "step": 9216
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.984068337610385e-05,
+      "loss": 5.8985,
+      "step": 9728
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.983229742859333e-05,
+      "loss": 5.8769,
+      "step": 10240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.982391148108281e-05,
+      "loss": 5.8304,
+      "step": 10752
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.981552553357229e-05,
+      "loss": 5.8075,
+      "step": 11264
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.980713958606178e-05,
+      "loss": 5.7724,
+      "step": 11776
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.979875363855125e-05,
+      "loss": 5.7408,
+      "step": 12288
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9790384069844466e-05,
+      "loss": 5.7232,
+      "step": 12800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9781998122333946e-05,
+      "loss": 5.6853,
+      "step": 13312
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9773612174823426e-05,
+      "loss": 5.673,
+      "step": 13824
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9765242606116635e-05,
+      "loss": 5.648,
+      "step": 14336
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9756856658606115e-05,
+      "loss": 5.6195,
+      "step": 14848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9748470711095595e-05,
+      "loss": 5.6054,
+      "step": 15360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.974008476358507e-05,
+      "loss": 5.5823,
+      "step": 15872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.973169881607455e-05,
+      "loss": 5.5682,
+      "step": 16384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.972331286856403e-05,
+      "loss": 5.5513,
+      "step": 16896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9714926921053515e-05,
+      "loss": 5.5308,
+      "step": 17408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9706540973542995e-05,
+      "loss": 5.5215,
+      "step": 17920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9698171404836204e-05,
+      "loss": 5.5139,
+      "step": 18432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9689785457325684e-05,
+      "loss": 5.4756,
+      "step": 18944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9681399509815164e-05,
+      "loss": 5.4657,
+      "step": 19456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9673013562304644e-05,
+      "loss": 5.4519,
+      "step": 19968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9664627614794124e-05,
+      "loss": 5.4357,
+      "step": 20480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9656241667283604e-05,
+      "loss": 5.4216,
+      "step": 20992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.964787209857681e-05,
+      "loss": 5.4229,
+      "step": 21504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.963948615106629e-05,
+      "loss": 5.4026,
+      "step": 22016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.963110020355577e-05,
+      "loss": 5.387,
+      "step": 22528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.962271425604525e-05,
+      "loss": 5.3803,
+      "step": 23040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.961434468733847e-05,
+      "loss": 5.3695,
+      "step": 23552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.960595873982795e-05,
+      "loss": 5.3622,
+      "step": 24064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.959758917112116e-05,
+      "loss": 5.3381,
+      "step": 24576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.958920322361064e-05,
+      "loss": 5.327,
+      "step": 25088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.958081727610012e-05,
+      "loss": 5.3347,
+      "step": 25600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.95724313285896e-05,
+      "loss": 5.3237,
+      "step": 26112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.956404538107908e-05,
+      "loss": 5.3118,
+      "step": 26624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.955565943356856e-05,
+      "loss": 5.2875,
+      "step": 27136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.954727348605804e-05,
+      "loss": 5.2966,
+      "step": 27648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.953888753854752e-05,
+      "loss": 5.275,
+      "step": 28160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9530501591037e-05,
+      "loss": 5.2905,
+      "step": 28672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9522132022330207e-05,
+      "loss": 5.2563,
+      "step": 29184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9513746074819686e-05,
+      "loss": 5.2599,
+      "step": 29696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9505360127309166e-05,
+      "loss": 5.2398,
+      "step": 30208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9496974179798646e-05,
+      "loss": 5.2303,
+      "step": 30720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.948858823228813e-05,
+      "loss": 5.2318,
+      "step": 31232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.948020228477761e-05,
+      "loss": 5.2149,
+      "step": 31744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.947183271607082e-05,
+      "loss": 5.21,
+      "step": 32256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.94634467685603e-05,
+      "loss": 5.2032,
+      "step": 32768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.945506082104978e-05,
+      "loss": 5.2125,
+      "step": 33280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9446674873539255e-05,
+      "loss": 5.1849,
+      "step": 33792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9438288926028735e-05,
+      "loss": 5.1752,
+      "step": 34304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9429902978518215e-05,
+      "loss": 5.1602,
+      "step": 34816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.942153340981143e-05,
+      "loss": 5.1583,
+      "step": 35328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9413147462300904e-05,
+      "loss": 5.1737,
+      "step": 35840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9404761514790384e-05,
+      "loss": 5.1619,
+      "step": 36352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.939637556727987e-05,
+      "loss": 5.1444,
+      "step": 36864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.938798961976935e-05,
+      "loss": 5.1404,
+      "step": 37376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.937960367225883e-05,
+      "loss": 5.143,
+      "step": 37888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9371250482355776e-05,
+      "loss": 5.1359,
+      "step": 38400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9362864534845256e-05,
+      "loss": 5.1131,
+      "step": 38912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.935447858733473e-05,
+      "loss": 5.118,
+      "step": 39424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.934609263982421e-05,
+      "loss": 5.1037,
+      "step": 39936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.933770669231369e-05,
+      "loss": 5.1055,
+      "step": 40448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.932932074480317e-05,
+      "loss": 5.1022,
+      "step": 40960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.932093479729265e-05,
+      "loss": 5.0905,
+      "step": 41472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.931254884978213e-05,
+      "loss": 5.0925,
+      "step": 41984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.930416290227161e-05,
+      "loss": 5.0832,
+      "step": 42496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9295793333564825e-05,
+      "loss": 5.0675,
+      "step": 43008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9287407386054305e-05,
+      "loss": 5.0665,
+      "step": 43520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9279021438543785e-05,
+      "loss": 5.0716,
+      "step": 44032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9270635491033264e-05,
+      "loss": 5.0642,
+      "step": 44544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9262249543522744e-05,
+      "loss": 5.0513,
+      "step": 45056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9253863596012224e-05,
+      "loss": 5.0386,
+      "step": 45568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9245494027305433e-05,
+      "loss": 5.0431,
+      "step": 46080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9237108079794913e-05,
+      "loss": 5.0302,
+      "step": 46592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.922873851108812e-05,
+      "loss": 5.0416,
+      "step": 47104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.92203525635776e-05,
+      "loss": 5.0192,
+      "step": 47616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.921196661606708e-05,
+      "loss": 5.0148,
+      "step": 48128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.920358066855656e-05,
+      "loss": 5.032,
+      "step": 48640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.919519472104604e-05,
+      "loss": 5.0004,
+      "step": 49152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.918680877353552e-05,
+      "loss": 4.9947,
+      "step": 49664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.917842282602501e-05,
+      "loss": 4.9991,
+      "step": 50176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.917003687851449e-05,
+      "loss": 4.9971,
+      "step": 50688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.916165093100397e-05,
+      "loss": 4.9973,
+      "step": 51200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.915328136229718e-05,
+      "loss": 4.9856,
+      "step": 51712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.914489541478666e-05,
+      "loss": 4.9793,
+      "step": 52224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.913650946727614e-05,
+      "loss": 4.9719,
+      "step": 52736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.912812351976562e-05,
+      "loss": 4.9706,
+      "step": 53248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.911973757225509e-05,
+      "loss": 4.9546,
+      "step": 53760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.911136800354831e-05,
+      "loss": 4.9543,
+      "step": 54272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.910298205603779e-05,
+      "loss": 4.9592,
+      "step": 54784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.909459610852727e-05,
+      "loss": 4.9467,
+      "step": 55296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.908621016101675e-05,
+      "loss": 4.9415,
+      "step": 55808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.907782421350623e-05,
+      "loss": 4.9358,
+      "step": 56320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.906943826599571e-05,
+      "loss": 4.9273,
+      "step": 56832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.906105231848519e-05,
+      "loss": 4.9364,
+      "step": 57344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.905266637097467e-05,
+      "loss": 4.922,
+      "step": 57856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9044296802267876e-05,
+      "loss": 4.9191,
+      "step": 58368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.903592723356109e-05,
+      "loss": 4.9196,
+      "step": 58880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9027541286050565e-05,
+      "loss": 4.9136,
+      "step": 59392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9019155338540045e-05,
+      "loss": 4.9097,
+      "step": 59904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9010769391029525e-05,
+      "loss": 4.9072,
+      "step": 60416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9002383443519005e-05,
+      "loss": 4.9064,
+      "step": 60928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.899403025361595e-05,
+      "loss": 4.9048,
+      "step": 61440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.898564430610543e-05,
+      "loss": 4.8852,
+      "step": 61952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.897725835859491e-05,
+      "loss": 4.8965,
+      "step": 62464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8968872411084396e-05,
+      "loss": 4.8864,
+      "step": 62976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.896048646357387e-05,
+      "loss": 4.8876,
+      "step": 63488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.895210051606335e-05,
+      "loss": 4.8696,
+      "step": 64000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.894371456855283e-05,
+      "loss": 4.8786,
+      "step": 64512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.893532862104231e-05,
+      "loss": 4.8627,
+      "step": 65024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.892694267353179e-05,
+      "loss": 4.8685,
+      "step": 65536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.891855672602127e-05,
+      "loss": 4.8463,
+      "step": 66048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.891017077851075e-05,
+      "loss": 4.869,
+      "step": 66560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.890178483100023e-05,
+      "loss": 4.874,
+      "step": 67072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.889343164109717e-05,
+      "loss": 4.8545,
+      "step": 67584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.888504569358665e-05,
+      "loss": 4.8404,
+      "step": 68096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8876659746076134e-05,
+      "loss": 4.8522,
+      "step": 68608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8868273798565614e-05,
+      "loss": 4.8459,
+      "step": 69120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8859887851055094e-05,
+      "loss": 4.8439,
+      "step": 69632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8851501903544574e-05,
+      "loss": 4.8368,
+      "step": 70144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8843115956034054e-05,
+      "loss": 4.8399,
+      "step": 70656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.8834730008523534e-05,
+      "loss": 4.8305,
+      "step": 71168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.882637681862047e-05,
+      "loss": 4.8263,
+      "step": 71680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.881799087110995e-05,
+      "loss": 4.8124,
+      "step": 72192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.880960492359943e-05,
+      "loss": 4.823,
+      "step": 72704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.880121897608891e-05,
+      "loss": 4.8183,
+      "step": 73216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.879283302857839e-05,
+      "loss": 4.8102,
+      "step": 73728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.878444708106787e-05,
+      "loss": 4.8228,
+      "step": 74240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.877606113355735e-05,
+      "loss": 4.8007,
+      "step": 74752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.876769156485057e-05,
+      "loss": 4.8102,
+      "step": 75264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.875930561734005e-05,
+      "loss": 4.8111,
+      "step": 75776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.875091966982953e-05,
+      "loss": 4.7998,
+      "step": 76288
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.812014102935791,
+      "eval_runtime": 329.0718,
+      "eval_samples_per_second": 1159.598,
+      "eval_steps_per_second": 36.238,
+      "step": 76320
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.874253372231901e-05,
+      "loss": 4.7881,
+      "step": 76800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.873414777480849e-05,
+      "loss": 4.794,
+      "step": 77312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.872576182729797e-05,
+      "loss": 4.7965,
+      "step": 77824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8717408637394906e-05,
+      "loss": 4.7873,
+      "step": 78336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8709022689884386e-05,
+      "loss": 4.788,
+      "step": 78848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8700636742373866e-05,
+      "loss": 4.7708,
+      "step": 79360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8692250794863346e-05,
+      "loss": 4.777,
+      "step": 79872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8683864847352826e-05,
+      "loss": 4.7661,
+      "step": 80384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8675478899842306e-05,
+      "loss": 4.7728,
+      "step": 80896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.866710933113552e-05,
+      "loss": 4.7638,
+      "step": 81408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.8658723383625e-05,
+      "loss": 4.7689,
+      "step": 81920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.865035381491821e-05,
+      "loss": 4.7706,
+      "step": 82432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.864196786740769e-05,
+      "loss": 4.7564,
+      "step": 82944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.863358191989717e-05,
+      "loss": 4.7541,
+      "step": 83456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.862519597238665e-05,
+      "loss": 4.7389,
+      "step": 83968
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.861681002487613e-05,
+      "loss": 4.7494,
+      "step": 84480
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.860842407736561e-05,
+      "loss": 4.7428,
+      "step": 84992
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.860003812985509e-05,
+      "loss": 4.7369,
+      "step": 85504
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.859165218234457e-05,
+      "loss": 4.7337,
+      "step": 86016
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.858326623483405e-05,
+      "loss": 4.7617,
+      "step": 86528
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.857488028732352e-05,
+      "loss": 4.7338,
+      "step": 87040
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.856649433981301e-05,
+      "loss": 4.7344,
+      "step": 87552
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.855810839230249e-05,
+      "loss": 4.7383,
+      "step": 88064
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.854972244479197e-05,
+      "loss": 4.7276,
+      "step": 88576
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.854135287608518e-05,
+      "loss": 4.7265,
+      "step": 89088
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.853296692857466e-05,
+      "loss": 4.7141,
+      "step": 89600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.852458098106414e-05,
+      "loss": 4.7234,
+      "step": 90112
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.851619503355362e-05,
+      "loss": 4.718,
+      "step": 90624
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.85078090860431e-05,
+      "loss": 4.7051,
+      "step": 91136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.849942313853258e-05,
+      "loss": 4.7132,
+      "step": 91648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.849103719102206e-05,
+      "loss": 4.7046,
+      "step": 92160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.848266762231527e-05,
+      "loss": 4.7136,
+      "step": 92672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.847428167480475e-05,
+      "loss": 4.7108,
+      "step": 93184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.846589572729423e-05,
+      "loss": 4.6974,
+      "step": 93696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.845750977978371e-05,
+      "loss": 4.706,
+      "step": 94208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8449123832273194e-05,
+      "loss": 4.7134,
+      "step": 94720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8440754263566404e-05,
+      "loss": 4.6825,
+      "step": 95232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8432368316055884e-05,
+      "loss": 4.6939,
+      "step": 95744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8423982368545363e-05,
+      "loss": 4.6896,
+      "step": 96256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8415596421034843e-05,
+      "loss": 4.6815,
+      "step": 96768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.840721047352432e-05,
+      "loss": 4.6821,
+      "step": 97280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.83988245260138e-05,
+      "loss": 4.6906,
+      "step": 97792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.839045495730701e-05,
+      "loss": 4.6842,
+      "step": 98304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.838206900979649e-05,
+      "loss": 4.678,
+      "step": 98816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.83736994410897e-05,
+      "loss": 4.6824,
+      "step": 99328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.836531349357918e-05,
+      "loss": 4.6748,
+      "step": 99840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.835692754606866e-05,
+      "loss": 4.679,
+      "step": 100352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.834854159855814e-05,
+      "loss": 4.6607,
+      "step": 100864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.834015565104763e-05,
+      "loss": 4.6578,
+      "step": 101376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.833178608234084e-05,
+      "loss": 4.674,
+      "step": 101888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.832340013483032e-05,
+      "loss": 4.6727,
+      "step": 102400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.83150141873198e-05,
+      "loss": 4.6614,
+      "step": 102912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.830662823980928e-05,
+      "loss": 4.652,
+      "step": 103424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.829824229229876e-05,
+      "loss": 4.6634,
+      "step": 103936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.828985634478824e-05,
+      "loss": 4.6453,
+      "step": 104448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.828147039727771e-05,
+      "loss": 4.6713,
+      "step": 104960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.827308444976719e-05,
+      "loss": 4.6471,
+      "step": 105472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.826469850225667e-05,
+      "loss": 4.6591,
+      "step": 105984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.825631255474615e-05,
+      "loss": 4.645,
+      "step": 106496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.824792660723563e-05,
+      "loss": 4.6375,
+      "step": 107008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.823954065972511e-05,
+      "loss": 4.6416,
+      "step": 107520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8231171091018326e-05,
+      "loss": 4.6462,
+      "step": 108032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8222785143507806e-05,
+      "loss": 4.6295,
+      "step": 108544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8214399195997286e-05,
+      "loss": 4.6322,
+      "step": 109056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8206013248486766e-05,
+      "loss": 4.6472,
+      "step": 109568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8197627300976246e-05,
+      "loss": 4.6271,
+      "step": 110080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8189241353465726e-05,
+      "loss": 4.6174,
+      "step": 110592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8180871784758935e-05,
+      "loss": 4.6184,
+      "step": 111104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8172502216052144e-05,
+      "loss": 4.6168,
+      "step": 111616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8164116268541624e-05,
+      "loss": 4.6382,
+      "step": 112128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8155730321031104e-05,
+      "loss": 4.6323,
+      "step": 112640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8147344373520584e-05,
+      "loss": 4.6167,
+      "step": 113152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.8138958426010064e-05,
+      "loss": 4.6206,
+      "step": 113664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.813057247849955e-05,
+      "loss": 4.6233,
+      "step": 114176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.812218653098903e-05,
+      "loss": 4.622,
+      "step": 114688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.811380058347851e-05,
+      "loss": 4.6052,
+      "step": 115200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.810543101477172e-05,
+      "loss": 4.6188,
+      "step": 115712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.80970450672612e-05,
+      "loss": 4.6063,
+      "step": 116224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.808865911975068e-05,
+      "loss": 4.6141,
+      "step": 116736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.808027317224016e-05,
+      "loss": 4.6061,
+      "step": 117248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.807188722472964e-05,
+      "loss": 4.6047,
+      "step": 117760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.806350127721912e-05,
+      "loss": 4.6082,
+      "step": 118272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.80551153297086e-05,
+      "loss": 4.6081,
+      "step": 118784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.804672938219807e-05,
+      "loss": 4.5911,
+      "step": 119296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.803835981349129e-05,
+      "loss": 4.5996,
+      "step": 119808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.802997386598077e-05,
+      "loss": 4.5972,
+      "step": 120320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.802158791847025e-05,
+      "loss": 4.6029,
+      "step": 120832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.801320197095973e-05,
+      "loss": 4.5948,
+      "step": 121344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.800481602344921e-05,
+      "loss": 4.5827,
+      "step": 121856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7996446454742424e-05,
+      "loss": 4.5923,
+      "step": 122368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.79880605072319e-05,
+      "loss": 4.5834,
+      "step": 122880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.797967455972138e-05,
+      "loss": 4.5957,
+      "step": 123392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.797128861221086e-05,
+      "loss": 4.577,
+      "step": 123904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.796290266470034e-05,
+      "loss": 4.5814,
+      "step": 124416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7954533095993546e-05,
+      "loss": 4.5918,
+      "step": 124928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7946147148483026e-05,
+      "loss": 4.5816,
+      "step": 125440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7937761200972506e-05,
+      "loss": 4.5627,
+      "step": 125952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7929375253461986e-05,
+      "loss": 4.5757,
+      "step": 126464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.792098930595147e-05,
+      "loss": 4.579,
+      "step": 126976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.791260335844095e-05,
+      "loss": 4.5809,
+      "step": 127488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.790421741093043e-05,
+      "loss": 4.5698,
+      "step": 128000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.789584784222364e-05,
+      "loss": 4.5701,
+      "step": 128512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.788746189471312e-05,
+      "loss": 4.5639,
+      "step": 129024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.78790759472026e-05,
+      "loss": 4.5699,
+      "step": 129536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.787068999969208e-05,
+      "loss": 4.5512,
+      "step": 130048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.786232043098529e-05,
+      "loss": 4.5595,
+      "step": 130560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.785393448347477e-05,
+      "loss": 4.566,
+      "step": 131072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.784554853596425e-05,
+      "loss": 4.5617,
+      "step": 131584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.783716258845373e-05,
+      "loss": 4.5464,
+      "step": 132096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.782877664094321e-05,
+      "loss": 4.555,
+      "step": 132608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.782040707223642e-05,
+      "loss": 4.5441,
+      "step": 133120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7812021124725906e-05,
+      "loss": 4.5602,
+      "step": 133632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7803635177215386e-05,
+      "loss": 4.5462,
+      "step": 134144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7795249229704866e-05,
+      "loss": 4.5378,
+      "step": 134656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7786863282194346e-05,
+      "loss": 4.5567,
+      "step": 135168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7778493713487555e-05,
+      "loss": 4.5497,
+      "step": 135680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7770107765977035e-05,
+      "loss": 4.5439,
+      "step": 136192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7761721818466515e-05,
+      "loss": 4.5462,
+      "step": 136704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7753335870955995e-05,
+      "loss": 4.544,
+      "step": 137216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7744966302249204e-05,
+      "loss": 4.549,
+      "step": 137728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7736580354738684e-05,
+      "loss": 4.5282,
+      "step": 138240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7728194407228164e-05,
+      "loss": 4.5482,
+      "step": 138752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7719808459717644e-05,
+      "loss": 4.5363,
+      "step": 139264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.771143889101086e-05,
+      "loss": 4.5396,
+      "step": 139776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.770305294350034e-05,
+      "loss": 4.5301,
+      "step": 140288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.769466699598982e-05,
+      "loss": 4.5355,
+      "step": 140800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.76862810484793e-05,
+      "loss": 4.5243,
+      "step": 141312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.767791147977251e-05,
+      "loss": 4.5326,
+      "step": 141824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.766952553226199e-05,
+      "loss": 4.5091,
+      "step": 142336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.766113958475147e-05,
+      "loss": 4.5433,
+      "step": 142848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.765275363724095e-05,
+      "loss": 4.5358,
+      "step": 143360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.764436768973043e-05,
+      "loss": 4.5304,
+      "step": 143872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.763599812102364e-05,
+      "loss": 4.5148,
+      "step": 144384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.762761217351312e-05,
+      "loss": 4.5308,
+      "step": 144896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.76192262260026e-05,
+      "loss": 4.5304,
+      "step": 145408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.761084027849208e-05,
+      "loss": 4.5232,
+      "step": 145920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.760245433098156e-05,
+      "loss": 4.5159,
+      "step": 146432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7594084762274773e-05,
+      "loss": 4.5234,
+      "step": 146944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.758571519356798e-05,
+      "loss": 4.5201,
+      "step": 147456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.757732924605746e-05,
+      "loss": 4.5264,
+      "step": 147968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.756894329854694e-05,
+      "loss": 4.5028,
+      "step": 148480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.756055735103642e-05,
+      "loss": 4.5184,
+      "step": 148992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.75521714035259e-05,
+      "loss": 4.5155,
+      "step": 149504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.754380183481911e-05,
+      "loss": 4.5093,
+      "step": 150016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.753541588730859e-05,
+      "loss": 4.5219,
+      "step": 150528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.752702993979807e-05,
+      "loss": 4.5025,
+      "step": 151040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.751864399228755e-05,
+      "loss": 4.5115,
+      "step": 151552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.751027442358077e-05,
+      "loss": 4.5097,
+      "step": 152064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.750188847607025e-05,
+      "loss": 4.5096,
+      "step": 152576
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.525084018707275,
+      "eval_runtime": 315.6819,
+      "eval_samples_per_second": 1208.783,
+      "eval_steps_per_second": 37.775,
+      "step": 152640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.749350252855973e-05,
+      "loss": 4.4977,
+      "step": 153088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.748511658104921e-05,
+      "loss": 4.5038,
+      "step": 153600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.747673063353868e-05,
+      "loss": 4.5097,
+      "step": 154112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.746834468602816e-05,
+      "loss": 4.5055,
+      "step": 154624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.745995873851764e-05,
+      "loss": 4.5043,
+      "step": 155136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.745157279100712e-05,
+      "loss": 4.4878,
+      "step": 155648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.74431868434966e-05,
+      "loss": 4.4965,
+      "step": 156160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.743480089598608e-05,
+      "loss": 4.4882,
+      "step": 156672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.742641494847556e-05,
+      "loss": 4.4984,
+      "step": 157184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.741802900096504e-05,
+      "loss": 4.4904,
+      "step": 157696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.740964305345452e-05,
+      "loss": 4.4977,
+      "step": 158208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7401257105944e-05,
+      "loss": 4.5,
+      "step": 158720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7392887537237216e-05,
+      "loss": 4.4825,
+      "step": 159232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7384501589726696e-05,
+      "loss": 4.4876,
+      "step": 159744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7376115642216176e-05,
+      "loss": 4.4701,
+      "step": 160256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7367729694705656e-05,
+      "loss": 4.4828,
+      "step": 160768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7359343747195135e-05,
+      "loss": 4.4828,
+      "step": 161280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7350957799684615e-05,
+      "loss": 4.4747,
+      "step": 161792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7342571852174095e-05,
+      "loss": 4.4752,
+      "step": 162304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.733418590466357e-05,
+      "loss": 4.5013,
+      "step": 162816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.732579995715305e-05,
+      "loss": 4.4784,
+      "step": 163328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.731741400964253e-05,
+      "loss": 4.4767,
+      "step": 163840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.730902806213201e-05,
+      "loss": 4.4837,
+      "step": 164352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.730065849342522e-05,
+      "loss": 4.4802,
+      "step": 164864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.7292288924718433e-05,
+      "loss": 4.4735,
+      "step": 165376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.728390297720791e-05,
+      "loss": 4.4641,
+      "step": 165888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.727551702969739e-05,
+      "loss": 4.4752,
+      "step": 166400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.726713108218687e-05,
+      "loss": 4.4674,
+      "step": 166912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.725876151348009e-05,
+      "loss": 4.4622,
+      "step": 167424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.725037556596957e-05,
+      "loss": 4.4703,
+      "step": 167936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.724198961845904e-05,
+      "loss": 4.4605,
+      "step": 168448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.723360367094852e-05,
+      "loss": 4.4724,
+      "step": 168960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.722523410224174e-05,
+      "loss": 4.4672,
+      "step": 169472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.721684815473121e-05,
+      "loss": 4.4577,
+      "step": 169984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.720846220722069e-05,
+      "loss": 4.4683,
+      "step": 170496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.720007625971017e-05,
+      "loss": 4.4789,
+      "step": 171008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.719169031219965e-05,
+      "loss": 4.4465,
+      "step": 171520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.71833371222966e-05,
+      "loss": 4.463,
+      "step": 172032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.717495117478608e-05,
+      "loss": 4.4526,
+      "step": 172544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.716656522727556e-05,
+      "loss": 4.4549,
+      "step": 173056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.715817927976504e-05,
+      "loss": 4.4495,
+      "step": 173568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7149793332254516e-05,
+      "loss": 4.464,
+      "step": 174080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7141407384743996e-05,
+      "loss": 4.4593,
+      "step": 174592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7133021437233476e-05,
+      "loss": 4.4522,
+      "step": 175104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7124635489722956e-05,
+      "loss": 4.4556,
+      "step": 175616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7116249542212436e-05,
+      "loss": 4.4476,
+      "step": 176128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7107863594701916e-05,
+      "loss": 4.4602,
+      "step": 176640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7099494025995125e-05,
+      "loss": 4.4401,
+      "step": 177152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.7091108078484605e-05,
+      "loss": 4.4399,
+      "step": 177664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.708272213097409e-05,
+      "loss": 4.4537,
+      "step": 178176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.707433618346357e-05,
+      "loss": 4.4514,
+      "step": 178688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.706595023595305e-05,
+      "loss": 4.4435,
+      "step": 179200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.705756428844253e-05,
+      "loss": 4.4392,
+      "step": 179712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.704917834093201e-05,
+      "loss": 4.4477,
+      "step": 180224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.704079239342149e-05,
+      "loss": 4.4303,
+      "step": 180736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.70324228247147e-05,
+      "loss": 4.4551,
+      "step": 181248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.702403687720418e-05,
+      "loss": 4.439,
+      "step": 181760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.701565092969366e-05,
+      "loss": 4.4472,
+      "step": 182272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.700726498218314e-05,
+      "loss": 4.4412,
+      "step": 182784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.699889541347635e-05,
+      "loss": 4.4306,
+      "step": 183296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.699050946596583e-05,
+      "loss": 4.4325,
+      "step": 183808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6982139897259045e-05,
+      "loss": 4.4406,
+      "step": 184320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6973753949748525e-05,
+      "loss": 4.4217,
+      "step": 184832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6965368002238005e-05,
+      "loss": 4.4287,
+      "step": 185344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6956982054727485e-05,
+      "loss": 4.4388,
+      "step": 185856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6948612486020694e-05,
+      "loss": 4.4302,
+      "step": 186368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6940226538510174e-05,
+      "loss": 4.4158,
+      "step": 186880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6931840590999654e-05,
+      "loss": 4.4181,
+      "step": 187392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6923454643489134e-05,
+      "loss": 4.4162,
+      "step": 187904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6915068695978614e-05,
+      "loss": 4.4424,
+      "step": 188416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6906682748468094e-05,
+      "loss": 4.4303,
+      "step": 188928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.68983131797613e-05,
+      "loss": 4.4207,
+      "step": 189440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.688992723225078e-05,
+      "loss": 4.4291,
+      "step": 189952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.688154128474026e-05,
+      "loss": 4.4251,
+      "step": 190464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.687315533722974e-05,
+      "loss": 4.4287,
+      "step": 190976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.686480214732669e-05,
+      "loss": 4.4129,
+      "step": 191488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.685641619981617e-05,
+      "loss": 4.4276,
+      "step": 192000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.684803025230565e-05,
+      "loss": 4.4204,
+      "step": 192512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.683964430479513e-05,
+      "loss": 4.4213,
+      "step": 193024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.683125835728461e-05,
+      "loss": 4.4146,
+      "step": 193536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.682287240977409e-05,
+      "loss": 4.4177,
+      "step": 194048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.681448646226357e-05,
+      "loss": 4.4211,
+      "step": 194560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.680610051475305e-05,
+      "loss": 4.4231,
+      "step": 195072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.679773094604626e-05,
+      "loss": 4.4079,
+      "step": 195584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.678934499853574e-05,
+      "loss": 4.4133,
+      "step": 196096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.678097542982895e-05,
+      "loss": 4.4108,
+      "step": 196608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.677258948231843e-05,
+      "loss": 4.4175,
+      "step": 197120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.676420353480791e-05,
+      "loss": 4.4158,
+      "step": 197632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.675581758729739e-05,
+      "loss": 4.397,
+      "step": 198144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.674743163978687e-05,
+      "loss": 4.4098,
+      "step": 198656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6739045692276346e-05,
+      "loss": 4.4053,
+      "step": 199168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6730659744765826e-05,
+      "loss": 4.4155,
+      "step": 199680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6722273797255305e-05,
+      "loss": 4.3952,
+      "step": 200192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.671392060735225e-05,
+      "loss": 4.4046,
+      "step": 200704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.670553465984173e-05,
+      "loss": 4.4156,
+      "step": 201216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.669714871233121e-05,
+      "loss": 4.4032,
+      "step": 201728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.668876276482069e-05,
+      "loss": 4.3873,
+      "step": 202240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.668037681731017e-05,
+      "loss": 4.4005,
+      "step": 202752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.667199086979965e-05,
+      "loss": 4.4062,
+      "step": 203264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.666360492228913e-05,
+      "loss": 4.4072,
+      "step": 203776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.665521897477861e-05,
+      "loss": 4.3998,
+      "step": 204288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.664684940607182e-05,
+      "loss": 4.3928,
+      "step": 204800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.66384634585613e-05,
+      "loss": 4.3914,
+      "step": 205312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.663007751105078e-05,
+      "loss": 4.3985,
+      "step": 205824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.662169156354026e-05,
+      "loss": 4.3828,
+      "step": 206336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.661332199483347e-05,
+      "loss": 4.3894,
+      "step": 206848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.660493604732295e-05,
+      "loss": 4.3959,
+      "step": 207360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.659655009981243e-05,
+      "loss": 4.3946,
+      "step": 207872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.658816415230191e-05,
+      "loss": 4.3771,
+      "step": 208384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.657977820479139e-05,
+      "loss": 4.3911,
+      "step": 208896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6571408636084604e-05,
+      "loss": 4.379,
+      "step": 209408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6563022688574084e-05,
+      "loss": 4.3901,
+      "step": 209920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6554636741063564e-05,
+      "loss": 4.3855,
+      "step": 210432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6546250793553044e-05,
+      "loss": 4.3776,
+      "step": 210944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6537864846042524e-05,
+      "loss": 4.3917,
+      "step": 211456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.652949527733573e-05,
+      "loss": 4.3914,
+      "step": 211968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.652110932982521e-05,
+      "loss": 4.3794,
+      "step": 212480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.651272338231469e-05,
+      "loss": 4.3851,
+      "step": 212992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.650433743480417e-05,
+      "loss": 4.3853,
+      "step": 213504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.649595148729365e-05,
+      "loss": 4.3855,
+      "step": 214016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.648756553978313e-05,
+      "loss": 4.3723,
+      "step": 214528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.647919597107634e-05,
+      "loss": 4.3868,
+      "step": 215040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.647081002356582e-05,
+      "loss": 4.3805,
+      "step": 215552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.646242407605531e-05,
+      "loss": 4.3813,
+      "step": 216064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.645403812854479e-05,
+      "loss": 4.3768,
+      "step": 216576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.644565218103427e-05,
+      "loss": 4.3836,
+      "step": 217088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.643728261232748e-05,
+      "loss": 4.3634,
+      "step": 217600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.642889666481696e-05,
+      "loss": 4.3777,
+      "step": 218112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.642051071730644e-05,
+      "loss": 4.358,
+      "step": 218624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.641212476979592e-05,
+      "loss": 4.3882,
+      "step": 219136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.64037388222854e-05,
+      "loss": 4.379,
+      "step": 219648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.639535287477488e-05,
+      "loss": 4.3786,
+      "step": 220160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6386983306068086e-05,
+      "loss": 4.3635,
+      "step": 220672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6378597358557566e-05,
+      "loss": 4.3795,
+      "step": 221184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6370211411047046e-05,
+      "loss": 4.3821,
+      "step": 221696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6361825463536526e-05,
+      "loss": 4.3709,
+      "step": 222208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6353439516026006e-05,
+      "loss": 4.3677,
+      "step": 222720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6345053568515486e-05,
+      "loss": 4.376,
+      "step": 223232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6336667621004966e-05,
+      "loss": 4.3697,
+      "step": 223744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6328281673494446e-05,
+      "loss": 4.3769,
+      "step": 224256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6319912104787655e-05,
+      "loss": 4.3581,
+      "step": 224768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6311526157277135e-05,
+      "loss": 4.3689,
+      "step": 225280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6303140209766615e-05,
+      "loss": 4.367,
+      "step": 225792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6294754262256095e-05,
+      "loss": 4.36,
+      "step": 226304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6286368314745575e-05,
+      "loss": 4.379,
+      "step": 226816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6277998746038784e-05,
+      "loss": 4.3548,
+      "step": 227328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6269612798528264e-05,
+      "loss": 4.3656,
+      "step": 227840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.6261226851017744e-05,
+      "loss": 4.3688,
+      "step": 228352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.625284090350723e-05,
+      "loss": 4.3616,
+      "step": 228864
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.390535354614258,
+      "eval_runtime": 309.1781,
+      "eval_samples_per_second": 1234.211,
+      "eval_steps_per_second": 38.57,
+      "step": 228960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.624445495599671e-05,
+      "loss": 4.3619,
+      "step": 229376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.623606900848619e-05,
+      "loss": 4.3586,
+      "step": 229888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.622768306097567e-05,
+      "loss": 4.3688,
+      "step": 230400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.621929711346515e-05,
+      "loss": 4.3645,
+      "step": 230912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.621094392356209e-05,
+      "loss": 4.3632,
+      "step": 231424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.620255797605157e-05,
+      "loss": 4.3512,
+      "step": 231936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.619417202854105e-05,
+      "loss": 4.3577,
+      "step": 232448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.618578608103053e-05,
+      "loss": 4.3412,
+      "step": 232960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.617740013352001e-05,
+      "loss": 4.3647,
+      "step": 233472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.616901418600949e-05,
+      "loss": 4.3509,
+      "step": 233984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.616062823849897e-05,
+      "loss": 4.3606,
+      "step": 234496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.615224229098845e-05,
+      "loss": 4.3617,
+      "step": 235008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.614390547988913e-05,
+      "loss": 4.3501,
+      "step": 235520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.61355195323786e-05,
+      "loss": 4.3451,
+      "step": 236032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.612713358486808e-05,
+      "loss": 4.3406,
+      "step": 236544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.611874763735756e-05,
+      "loss": 4.3426,
+      "step": 237056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.611036168984704e-05,
+      "loss": 4.3497,
+      "step": 237568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.610197574233652e-05,
+      "loss": 4.3393,
+      "step": 238080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6093589794826e-05,
+      "loss": 4.342,
+      "step": 238592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.608520384731548e-05,
+      "loss": 4.365,
+      "step": 239104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.607681789980496e-05,
+      "loss": 4.3468,
+      "step": 239616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.606843195229444e-05,
+      "loss": 4.3462,
+      "step": 240128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.606004600478392e-05,
+      "loss": 4.3521,
+      "step": 240640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.60516600572734e-05,
+      "loss": 4.3476,
+      "step": 241152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.604327410976288e-05,
+      "loss": 4.3424,
+      "step": 241664
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.603488816225236e-05,
+      "loss": 4.332,
+      "step": 242176
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.602650221474184e-05,
+      "loss": 4.3468,
+      "step": 242688
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.601813264603506e-05,
+      "loss": 4.3322,
+      "step": 243200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.600974669852454e-05,
+      "loss": 4.3356,
+      "step": 243712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.600136075101402e-05,
+      "loss": 4.3365,
+      "step": 244224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.599297480350349e-05,
+      "loss": 4.3327,
+      "step": 244736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.598458885599297e-05,
+      "loss": 4.3469,
+      "step": 245248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.597620290848245e-05,
+      "loss": 4.3376,
+      "step": 245760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.596781696097193e-05,
+      "loss": 4.3297,
+      "step": 246272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.595944739226514e-05,
+      "loss": 4.3409,
+      "step": 246784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.595106144475462e-05,
+      "loss": 4.3491,
+      "step": 247296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.59426754972441e-05,
+      "loss": 4.327,
+      "step": 247808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5934289549733587e-05,
+      "loss": 4.3376,
+      "step": 248320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5925903602223066e-05,
+      "loss": 4.3237,
+      "step": 248832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5917517654712546e-05,
+      "loss": 4.3376,
+      "step": 249344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5909131707202026e-05,
+      "loss": 4.3221,
+      "step": 249856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5900762138495235e-05,
+      "loss": 4.3409,
+      "step": 250368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5892376190984715e-05,
+      "loss": 4.3348,
+      "step": 250880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5883990243474195e-05,
+      "loss": 4.3297,
+      "step": 251392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5875604295963675e-05,
+      "loss": 4.3344,
+      "step": 251904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5867218348453155e-05,
+      "loss": 4.3246,
+      "step": 252416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5858832400942635e-05,
+      "loss": 4.3377,
+      "step": 252928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5850446453432115e-05,
+      "loss": 4.3206,
+      "step": 253440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5842076884725324e-05,
+      "loss": 4.3166,
+      "step": 253952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5833690937214804e-05,
+      "loss": 4.3336,
+      "step": 254464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.582532136850802e-05,
+      "loss": 4.3274,
+      "step": 254976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.58169354209975e-05,
+      "loss": 4.3286,
+      "step": 255488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.580854947348698e-05,
+      "loss": 4.3156,
+      "step": 256000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.580016352597646e-05,
+      "loss": 4.3279,
+      "step": 256512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.579177757846594e-05,
+      "loss": 4.3121,
+      "step": 257024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.578339163095542e-05,
+      "loss": 4.3341,
+      "step": 257536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.57750056834449e-05,
+      "loss": 4.3193,
+      "step": 258048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.576661973593438e-05,
+      "loss": 4.331,
+      "step": 258560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.575823378842385e-05,
+      "loss": 4.3242,
+      "step": 259072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.574984784091333e-05,
+      "loss": 4.3162,
+      "step": 259584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.574146189340281e-05,
+      "loss": 4.3129,
+      "step": 260096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.573309232469602e-05,
+      "loss": 4.3232,
+      "step": 260608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.572470637718551e-05,
+      "loss": 4.3068,
+      "step": 261120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.571632042967499e-05,
+      "loss": 4.3149,
+      "step": 261632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.570793448216447e-05,
+      "loss": 4.3203,
+      "step": 262144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.569954853465395e-05,
+      "loss": 4.3192,
+      "step": 262656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.569116258714343e-05,
+      "loss": 4.2983,
+      "step": 263168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.568277663963291e-05,
+      "loss": 4.3053,
+      "step": 263680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.567439069212239e-05,
+      "loss": 4.3011,
+      "step": 264192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.56660211234156e-05,
+      "loss": 4.3307,
+      "step": 264704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.565765155470881e-05,
+      "loss": 4.3155,
+      "step": 265216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.564928198600202e-05,
+      "loss": 4.3107,
+      "step": 265728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5640896038491496e-05,
+      "loss": 4.3162,
+      "step": 266240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.5632510090980976e-05,
+      "loss": 4.3112,
+      "step": 266752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.562412414347046e-05,
+      "loss": 4.3236,
+      "step": 267264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.561573819595994e-05,
+      "loss": 4.2978,
+      "step": 267776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.560735224844942e-05,
+      "loss": 4.3194,
+      "step": 268288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.55989663009389e-05,
+      "loss": 4.3067,
+      "step": 268800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.559058035342838e-05,
+      "loss": 4.3104,
+      "step": 269312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.558219440591786e-05,
+      "loss": 4.3086,
+      "step": 269824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.557380845840734e-05,
+      "loss": 4.3049,
+      "step": 270336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.556542251089682e-05,
+      "loss": 4.3138,
+      "step": 270848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.55570365633863e-05,
+      "loss": 4.3098,
+      "step": 271360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.554866699467951e-05,
+      "loss": 4.2979,
+      "step": 271872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.554028104716899e-05,
+      "loss": 4.3081,
+      "step": 272384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.553189509965847e-05,
+      "loss": 4.2969,
+      "step": 272896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.552350915214795e-05,
+      "loss": 4.3125,
+      "step": 273408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.551513958344116e-05,
+      "loss": 4.3055,
+      "step": 273920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.550675363593064e-05,
+      "loss": 4.2895,
+      "step": 274432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5498384067223856e-05,
+      "loss": 4.3056,
+      "step": 274944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5489998119713336e-05,
+      "loss": 4.2966,
+      "step": 275456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5481612172202816e-05,
+      "loss": 4.3059,
+      "step": 275968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5473226224692296e-05,
+      "loss": 4.2914,
+      "step": 276480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5464856655985505e-05,
+      "loss": 4.2989,
+      "step": 276992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5456470708474985e-05,
+      "loss": 4.3084,
+      "step": 277504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5448084760964465e-05,
+      "loss": 4.295,
+      "step": 278016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5439698813453945e-05,
+      "loss": 4.2869,
+      "step": 278528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5431312865943425e-05,
+      "loss": 4.2944,
+      "step": 279040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5422926918432905e-05,
+      "loss": 4.2938,
+      "step": 279552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5414540970922385e-05,
+      "loss": 4.3072,
+      "step": 280064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5406171402215594e-05,
+      "loss": 4.2941,
+      "step": 280576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.539778545470508e-05,
+      "loss": 4.2917,
+      "step": 281088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.538939950719456e-05,
+      "loss": 4.2891,
+      "step": 281600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.538101355968404e-05,
+      "loss": 4.2975,
+      "step": 282112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5372627612173514e-05,
+      "loss": 4.2756,
+      "step": 282624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5364241664662994e-05,
+      "loss": 4.2902,
+      "step": 283136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.535587209595621e-05,
+      "loss": 4.2913,
+      "step": 283648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.534748614844569e-05,
+      "loss": 4.2937,
+      "step": 284160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.533910020093516e-05,
+      "loss": 4.2715,
+      "step": 284672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.533071425342464e-05,
+      "loss": 4.2916,
+      "step": 285184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.532232830591412e-05,
+      "loss": 4.2846,
+      "step": 285696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.53139423584036e-05,
+      "loss": 4.2822,
+      "step": 286208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.530555641089308e-05,
+      "loss": 4.289,
+      "step": 286720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.52971868421863e-05,
+      "loss": 4.2737,
+      "step": 287232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.528880089467578e-05,
+      "loss": 4.296,
+      "step": 287744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.528043132596899e-05,
+      "loss": 4.29,
+      "step": 288256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.527204537845847e-05,
+      "loss": 4.2812,
+      "step": 288768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.526365943094795e-05,
+      "loss": 4.2888,
+      "step": 289280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.525527348343743e-05,
+      "loss": 4.2823,
+      "step": 289792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.524688753592691e-05,
+      "loss": 4.2887,
+      "step": 290304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.523850158841639e-05,
+      "loss": 4.271,
+      "step": 290816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.523011564090587e-05,
+      "loss": 4.2928,
+      "step": 291328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.522172969339535e-05,
+      "loss": 4.2794,
+      "step": 291840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5213376503492285e-05,
+      "loss": 4.2817,
+      "step": 292352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.520499055598177e-05,
+      "loss": 4.2806,
+      "step": 292864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.519660460847125e-05,
+      "loss": 4.2875,
+      "step": 293376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.518821866096073e-05,
+      "loss": 4.2726,
+      "step": 293888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.517983271345021e-05,
+      "loss": 4.2764,
+      "step": 294400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.517144676593969e-05,
+      "loss": 4.2649,
+      "step": 294912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.516306081842917e-05,
+      "loss": 4.2892,
+      "step": 295424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.515467487091865e-05,
+      "loss": 4.2832,
+      "step": 295936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.514632168101559e-05,
+      "loss": 4.2832,
+      "step": 296448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.513793573350507e-05,
+      "loss": 4.2689,
+      "step": 296960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.512954978599455e-05,
+      "loss": 4.2828,
+      "step": 297472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.512116383848403e-05,
+      "loss": 4.2855,
+      "step": 297984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.511277789097351e-05,
+      "loss": 4.2759,
+      "step": 298496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.510439194346299e-05,
+      "loss": 4.2721,
+      "step": 299008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.509600599595247e-05,
+      "loss": 4.2811,
+      "step": 299520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5087620048441956e-05,
+      "loss": 4.2792,
+      "step": 300032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5079234100931436e-05,
+      "loss": 4.2807,
+      "step": 300544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5070864532224645e-05,
+      "loss": 4.2649,
+      "step": 301056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5062478584714125e-05,
+      "loss": 4.2765,
+      "step": 301568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5054092637203605e-05,
+      "loss": 4.273,
+      "step": 302080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5045706689693085e-05,
+      "loss": 4.2722,
+      "step": 302592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5037337120986294e-05,
+      "loss": 4.2801,
+      "step": 303104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5028967552279503e-05,
+      "loss": 4.266,
+      "step": 303616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.5020581604768983e-05,
+      "loss": 4.2743,
+      "step": 304128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.501219565725846e-05,
+      "loss": 4.2713,
+      "step": 304640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.500380970974794e-05,
+      "loss": 4.2752,
+      "step": 305152
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.3080854415893555,
+      "eval_runtime": 304.2229,
+      "eval_samples_per_second": 1254.314,
+      "eval_steps_per_second": 39.198,
+      "step": 305280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.499542376223742e-05,
+      "loss": 4.2653,
+      "step": 305664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.498703781472691e-05,
+      "loss": 4.2706,
+      "step": 306176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.497865186721639e-05,
+      "loss": 4.2789,
+      "step": 306688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.497026591970587e-05,
+      "loss": 4.2699,
+      "step": 307200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.496187997219535e-05,
+      "loss": 4.2773,
+      "step": 307712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.495349402468482e-05,
+      "loss": 4.2591,
+      "step": 308224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.49451080771743e-05,
+      "loss": 4.2668,
+      "step": 308736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.493672212966378e-05,
+      "loss": 4.2546,
+      "step": 309248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.492833618215326e-05,
+      "loss": 4.2738,
+      "step": 309760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.491995023464274e-05,
+      "loss": 4.2671,
+      "step": 310272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.491156428713222e-05,
+      "loss": 4.2681,
+      "step": 310784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.49031783396217e-05,
+      "loss": 4.2699,
+      "step": 311296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.489479239211118e-05,
+      "loss": 4.2627,
+      "step": 311808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.488640644460066e-05,
+      "loss": 4.2572,
+      "step": 312320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.487802049709014e-05,
+      "loss": 4.2591,
+      "step": 312832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.486963454957962e-05,
+      "loss": 4.2483,
+      "step": 313344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.486126498087284e-05,
+      "loss": 4.2645,
+      "step": 313856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.485289541216605e-05,
+      "loss": 4.2522,
+      "step": 314368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.484450946465553e-05,
+      "loss": 4.2545,
+      "step": 314880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.483612351714501e-05,
+      "loss": 4.2834,
+      "step": 315392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.482773756963449e-05,
+      "loss": 4.2561,
+      "step": 315904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.481935162212397e-05,
+      "loss": 4.2623,
+      "step": 316416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.481096567461345e-05,
+      "loss": 4.2655,
+      "step": 316928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.480257972710293e-05,
+      "loss": 4.26,
+      "step": 317440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.479419377959241e-05,
+      "loss": 4.2588,
+      "step": 317952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.478580783208188e-05,
+      "loss": 4.2482,
+      "step": 318464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4777438263375096e-05,
+      "loss": 4.2638,
+      "step": 318976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4769052315864576e-05,
+      "loss": 4.2422,
+      "step": 319488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4760666368354056e-05,
+      "loss": 4.2549,
+      "step": 320000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.4752280420843536e-05,
+      "loss": 4.2478,
+      "step": 320512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4743894473333016e-05,
+      "loss": 4.2493,
+      "step": 321024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.473552490462623e-05,
+      "loss": 4.2625,
+      "step": 321536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.472713895711571e-05,
+      "loss": 4.2554,
+      "step": 322048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4718753009605185e-05,
+      "loss": 4.25,
+      "step": 322560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4710367062094665e-05,
+      "loss": 4.2574,
+      "step": 323072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.470199749338788e-05,
+      "loss": 4.2643,
+      "step": 323584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4693611545877354e-05,
+      "loss": 4.2468,
+      "step": 324096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4685225598366834e-05,
+      "loss": 4.2553,
+      "step": 324608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4676839650856314e-05,
+      "loss": 4.241,
+      "step": 325120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4668453703345794e-05,
+      "loss": 4.2555,
+      "step": 325632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.466006775583528e-05,
+      "loss": 4.2351,
+      "step": 326144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.465168180832476e-05,
+      "loss": 4.2617,
+      "step": 326656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.464329586081424e-05,
+      "loss": 4.2525,
+      "step": 327168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.463492629210745e-05,
+      "loss": 4.2485,
+      "step": 327680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.462654034459693e-05,
+      "loss": 4.2536,
+      "step": 328192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.461817077589014e-05,
+      "loss": 4.2452,
+      "step": 328704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.460978482837962e-05,
+      "loss": 4.2554,
+      "step": 329216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.46013988808691e-05,
+      "loss": 4.241,
+      "step": 329728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.459301293335858e-05,
+      "loss": 4.2359,
+      "step": 330240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.458464336465179e-05,
+      "loss": 4.2533,
+      "step": 330752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4576273795945004e-05,
+      "loss": 4.2476,
+      "step": 331264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4567887848434484e-05,
+      "loss": 4.2479,
+      "step": 331776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4559501900923964e-05,
+      "loss": 4.2372,
+      "step": 332288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4551115953413444e-05,
+      "loss": 4.245,
+      "step": 332800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4542730005902923e-05,
+      "loss": 4.2373,
+      "step": 333312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4534344058392403e-05,
+      "loss": 4.2514,
+      "step": 333824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4525958110881883e-05,
+      "loss": 4.2385,
+      "step": 334336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.451757216337136e-05,
+      "loss": 4.2531,
+      "step": 334848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.450918621586084e-05,
+      "loss": 4.2454,
+      "step": 335360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.450080026835032e-05,
+      "loss": 4.2371,
+      "step": 335872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.44924143208398e-05,
+      "loss": 4.2371,
+      "step": 336384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.448402837332928e-05,
+      "loss": 4.2445,
+      "step": 336896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.447565880462249e-05,
+      "loss": 4.2269,
+      "step": 337408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.446727285711197e-05,
+      "loss": 4.2382,
+      "step": 337920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.445888690960145e-05,
+      "loss": 4.2408,
+      "step": 338432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.445050096209093e-05,
+      "loss": 4.245,
+      "step": 338944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.444213139338415e-05,
+      "loss": 4.2199,
+      "step": 339456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.443374544587363e-05,
+      "loss": 4.2303,
+      "step": 339968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.442535949836311e-05,
+      "loss": 4.2225,
+      "step": 340480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.441698992965632e-05,
+      "loss": 4.2495,
+      "step": 340992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.44086039821458e-05,
+      "loss": 4.2372,
+      "step": 341504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.440021803463528e-05,
+      "loss": 4.2372,
+      "step": 342016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.439183208712476e-05,
+      "loss": 4.2379,
+      "step": 342528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4383462518417966e-05,
+      "loss": 4.2345,
+      "step": 343040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4375076570907446e-05,
+      "loss": 4.2512,
+      "step": 343552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4366690623396926e-05,
+      "loss": 4.2205,
+      "step": 344064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4358304675886406e-05,
+      "loss": 4.2399,
+      "step": 344576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4349918728375886e-05,
+      "loss": 4.2354,
+      "step": 345088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4341532780865366e-05,
+      "loss": 4.234,
+      "step": 345600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4333146833354846e-05,
+      "loss": 4.2357,
+      "step": 346112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.432477726464806e-05,
+      "loss": 4.231,
+      "step": 346624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.431639131713754e-05,
+      "loss": 4.2362,
+      "step": 347136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4308005369627015e-05,
+      "loss": 4.2368,
+      "step": 347648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4299619422116495e-05,
+      "loss": 4.2227,
+      "step": 348160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.4291233474605975e-05,
+      "loss": 4.236,
+      "step": 348672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.428286390589919e-05,
+      "loss": 4.2235,
+      "step": 349184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.42744943371924e-05,
+      "loss": 4.2372,
+      "step": 349696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.426610838968188e-05,
+      "loss": 4.2295,
+      "step": 350208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.425772244217136e-05,
+      "loss": 4.2179,
+      "step": 350720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.424933649466084e-05,
+      "loss": 4.2311,
+      "step": 351232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.424095054715032e-05,
+      "loss": 4.2242,
+      "step": 351744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.42325645996398e-05,
+      "loss": 4.2338,
+      "step": 352256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.422417865212928e-05,
+      "loss": 4.2154,
+      "step": 352768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.421579270461876e-05,
+      "loss": 4.23,
+      "step": 353280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.420742313591197e-05,
+      "loss": 4.241,
+      "step": 353792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.419903718840145e-05,
+      "loss": 4.2208,
+      "step": 354304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.419065124089093e-05,
+      "loss": 4.2103,
+      "step": 354816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.418226529338041e-05,
+      "loss": 4.2213,
+      "step": 355328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.417387934586989e-05,
+      "loss": 4.2252,
+      "step": 355840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.416549339835937e-05,
+      "loss": 4.2317,
+      "step": 356352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.415710745084885e-05,
+      "loss": 4.2254,
+      "step": 356864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.414872150333833e-05,
+      "loss": 4.2197,
+      "step": 357376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.414033555582781e-05,
+      "loss": 4.2159,
+      "step": 357888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4131965987121024e-05,
+      "loss": 4.2269,
+      "step": 358400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4123580039610504e-05,
+      "loss": 4.2033,
+      "step": 358912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4115194092099984e-05,
+      "loss": 4.2193,
+      "step": 359424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.410682452339319e-05,
+      "loss": 4.2197,
+      "step": 359936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.409843857588267e-05,
+      "loss": 4.2242,
+      "step": 360448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.409005262837215e-05,
+      "loss": 4.2015,
+      "step": 360960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.408166668086163e-05,
+      "loss": 4.221,
+      "step": 361472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.407329711215484e-05,
+      "loss": 4.2137,
+      "step": 361984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.406491116464432e-05,
+      "loss": 4.216,
+      "step": 362496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.40565252171338e-05,
+      "loss": 4.2153,
+      "step": 363008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.404813926962328e-05,
+      "loss": 4.2079,
+      "step": 363520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.403975332211276e-05,
+      "loss": 4.2216,
+      "step": 364032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.403138375340598e-05,
+      "loss": 4.2218,
+      "step": 364544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.402299780589546e-05,
+      "loss": 4.2105,
+      "step": 365056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.401461185838494e-05,
+      "loss": 4.2204,
+      "step": 365568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.400622591087442e-05,
+      "loss": 4.2106,
+      "step": 366080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.39978399633639e-05,
+      "loss": 4.2212,
+      "step": 366592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.398945401585338e-05,
+      "loss": 4.2046,
+      "step": 367104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.398106806834285e-05,
+      "loss": 4.2261,
+      "step": 367616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3972698499636066e-05,
+      "loss": 4.2055,
+      "step": 368128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3964312552125546e-05,
+      "loss": 4.2154,
+      "step": 368640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3955926604615026e-05,
+      "loss": 4.2068,
+      "step": 369152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.39475406571045e-05,
+      "loss": 4.2202,
+      "step": 369664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.393915470959398e-05,
+      "loss": 4.208,
+      "step": 370176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3930785140887195e-05,
+      "loss": 4.2106,
+      "step": 370688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3922399193376675e-05,
+      "loss": 4.1938,
+      "step": 371200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3914013245866155e-05,
+      "loss": 4.2255,
+      "step": 371712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3905627298355635e-05,
+      "loss": 4.2103,
+      "step": 372224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3897241350845115e-05,
+      "loss": 4.2129,
+      "step": 372736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3888871782138324e-05,
+      "loss": 4.1996,
+      "step": 373248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3880485834627804e-05,
+      "loss": 4.217,
+      "step": 373760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3872099887117284e-05,
+      "loss": 4.2212,
+      "step": 374272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3863713939606764e-05,
+      "loss": 4.2061,
+      "step": 374784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.3855327992096244e-05,
+      "loss": 4.2041,
+      "step": 375296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.384695842338945e-05,
+      "loss": 4.2137,
+      "step": 375808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.383857247587893e-05,
+      "loss": 4.2145,
+      "step": 376320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.383018652836842e-05,
+      "loss": 4.2155,
+      "step": 376832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.38218005808579e-05,
+      "loss": 4.1964,
+      "step": 377344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.381341463334738e-05,
+      "loss": 4.2098,
+      "step": 377856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.380502868583686e-05,
+      "loss": 4.2047,
+      "step": 378368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.379664273832634e-05,
+      "loss": 4.2081,
+      "step": 378880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.378825679081582e-05,
+      "loss": 4.2135,
+      "step": 379392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.377988722210903e-05,
+      "loss": 4.1974,
+      "step": 379904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.377150127459851e-05,
+      "loss": 4.2139,
+      "step": 380416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.376311532708799e-05,
+      "loss": 4.2026,
+      "step": 380928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.375472937957747e-05,
+      "loss": 4.2147,
+      "step": 381440
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.25103235244751,
+      "eval_runtime": 313.3193,
+      "eval_samples_per_second": 1217.898,
+      "eval_steps_per_second": 38.06,
+      "step": 381600
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.374634343206695e-05,
+      "loss": 4.1995,
+      "step": 381952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.373795748455643e-05,
+      "loss": 4.2076,
+      "step": 382464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.372957153704591e-05,
+      "loss": 4.2101,
+      "step": 382976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.372120196833912e-05,
+      "loss": 4.2042,
+      "step": 383488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.37128160208286e-05,
+      "loss": 4.2105,
+      "step": 384000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3704430073318084e-05,
+      "loss": 4.1967,
+      "step": 384512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3696044125807564e-05,
+      "loss": 4.2013,
+      "step": 385024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.368765817829704e-05,
+      "loss": 4.1911,
+      "step": 385536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.367927223078652e-05,
+      "loss": 4.2092,
+      "step": 386048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3670886283276e-05,
+      "loss": 4.2066,
+      "step": 386560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.366251671456921e-05,
+      "loss": 4.2014,
+      "step": 387072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3654130767058686e-05,
+      "loss": 4.207,
+      "step": 387584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.36457611983519e-05,
+      "loss": 4.1957,
+      "step": 388096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.363737525084138e-05,
+      "loss": 4.1964,
+      "step": 388608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.362898930333086e-05,
+      "loss": 4.1954,
+      "step": 389120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.3620603355820335e-05,
+      "loss": 4.1864,
+      "step": 389632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.361221740830982e-05,
+      "loss": 4.1983,
+      "step": 390144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.36038314607993e-05,
+      "loss": 4.1884,
+      "step": 390656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.359544551328878e-05,
+      "loss": 4.1919,
+      "step": 391168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.358705956577826e-05,
+      "loss": 4.2272,
+      "step": 391680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.357867361826774e-05,
+      "loss": 4.1905,
+      "step": 392192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.357030404956095e-05,
+      "loss": 4.2008,
+      "step": 392704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.356191810205043e-05,
+      "loss": 4.2017,
+      "step": 393216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.355353215453991e-05,
+      "loss": 4.1999,
+      "step": 393728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.354514620702939e-05,
+      "loss": 4.1919,
+      "step": 394240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.353676025951887e-05,
+      "loss": 4.1905,
+      "step": 394752
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.352837431200835e-05,
+      "loss": 4.1992,
+      "step": 395264
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.351998836449783e-05,
+      "loss": 4.1802,
+      "step": 395776
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.351160241698731e-05,
+      "loss": 4.1901,
+      "step": 396288
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.350323284828052e-05,
+      "loss": 4.1863,
+      "step": 396800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3494846900770006e-05,
+      "loss": 4.191,
+      "step": 397312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3486477332063216e-05,
+      "loss": 4.1973,
+      "step": 397824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3478091384552696e-05,
+      "loss": 4.1977,
+      "step": 398336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3469705437042175e-05,
+      "loss": 4.1899,
+      "step": 398848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3461319489531655e-05,
+      "loss": 4.1963,
+      "step": 399360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3452949920824865e-05,
+      "loss": 4.2009,
+      "step": 399872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3444563973314344e-05,
+      "loss": 4.1851,
+      "step": 400384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3436178025803824e-05,
+      "loss": 4.1988,
+      "step": 400896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3427792078293304e-05,
+      "loss": 4.1765,
+      "step": 401408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3419406130782784e-05,
+      "loss": 4.1999,
+      "step": 401920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3411020183272264e-05,
+      "loss": 4.1714,
+      "step": 402432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3402634235761744e-05,
+      "loss": 4.2034,
+      "step": 402944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3394248288251224e-05,
+      "loss": 4.1913,
+      "step": 403456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.338587871954444e-05,
+      "loss": 4.1893,
+      "step": 403968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.337750915083765e-05,
+      "loss": 4.1917,
+      "step": 404480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.336912320332713e-05,
+      "loss": 4.1877,
+      "step": 404992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.336073725581661e-05,
+      "loss": 4.191,
+      "step": 405504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.335235130830609e-05,
+      "loss": 4.18,
+      "step": 406016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.334396536079557e-05,
+      "loss": 4.1706,
+      "step": 406528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.333557941328505e-05,
+      "loss": 4.1992,
+      "step": 407040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.332719346577452e-05,
+      "loss": 4.1889,
+      "step": 407552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3318807518264e-05,
+      "loss": 4.1932,
+      "step": 408064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.331043794955722e-05,
+      "loss": 4.1783,
+      "step": 408576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.33020520020467e-05,
+      "loss": 4.1861,
+      "step": 409088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.329366605453618e-05,
+      "loss": 4.1793,
+      "step": 409600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.328528010702566e-05,
+      "loss": 4.1905,
+      "step": 410112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.327689415951514e-05,
+      "loss": 4.1805,
+      "step": 410624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.326850821200462e-05,
+      "loss": 4.1941,
+      "step": 411136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.32601222644941e-05,
+      "loss": 4.1922,
+      "step": 411648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.325175269578731e-05,
+      "loss": 4.1748,
+      "step": 412160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.324336674827679e-05,
+      "loss": 4.18,
+      "step": 412672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.323498080076627e-05,
+      "loss": 4.1884,
+      "step": 413184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.322659485325575e-05,
+      "loss": 4.1706,
+      "step": 413696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3218225284548956e-05,
+      "loss": 4.1772,
+      "step": 414208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3209839337038436e-05,
+      "loss": 4.184,
+      "step": 414720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3201453389527916e-05,
+      "loss": 4.1837,
+      "step": 415232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3193067442017396e-05,
+      "loss": 4.165,
+      "step": 415744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3184681494506876e-05,
+      "loss": 4.172,
+      "step": 416256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.317631192580009e-05,
+      "loss": 4.1657,
+      "step": 416768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.316792597828957e-05,
+      "loss": 4.1912,
+      "step": 417280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.315954003077905e-05,
+      "loss": 4.181,
+      "step": 417792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.315115408326853e-05,
+      "loss": 4.18,
+      "step": 418304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.314276813575801e-05,
+      "loss": 4.1849,
+      "step": 418816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.313439856705122e-05,
+      "loss": 4.1757,
+      "step": 419328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.312602899834443e-05,
+      "loss": 4.1951,
+      "step": 419840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.311764305083391e-05,
+      "loss": 4.1662,
+      "step": 420352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.310925710332339e-05,
+      "loss": 4.1823,
+      "step": 420864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.310087115581287e-05,
+      "loss": 4.1803,
+      "step": 421376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.309248520830235e-05,
+      "loss": 4.1776,
+      "step": 421888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.308409926079183e-05,
+      "loss": 4.182,
+      "step": 422400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3075713313281316e-05,
+      "loss": 4.1744,
+      "step": 422912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3067327365770796e-05,
+      "loss": 4.1783,
+      "step": 423424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3058941418260276e-05,
+      "loss": 4.1833,
+      "step": 423936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3050571849553485e-05,
+      "loss": 4.1647,
+      "step": 424448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3042185902042965e-05,
+      "loss": 4.1828,
+      "step": 424960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3033799954532445e-05,
+      "loss": 4.1687,
+      "step": 425472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3025430385825654e-05,
+      "loss": 4.1831,
+      "step": 425984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3017044438315134e-05,
+      "loss": 4.1713,
+      "step": 426496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.3008658490804614e-05,
+      "loss": 4.1661,
+      "step": 427008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.3000272543294094e-05,
+      "loss": 4.173,
+      "step": 427520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2991886595783574e-05,
+      "loss": 4.1741,
+      "step": 428032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2983500648273054e-05,
+      "loss": 4.1742,
+      "step": 428544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2975114700762534e-05,
+      "loss": 4.1605,
+      "step": 429056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2966728753252014e-05,
+      "loss": 4.1764,
+      "step": 429568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.295835918454523e-05,
+      "loss": 4.1847,
+      "step": 430080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.294997323703471e-05,
+      "loss": 4.1699,
+      "step": 430592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.294158728952418e-05,
+      "loss": 4.1553,
+      "step": 431104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.293320134201366e-05,
+      "loss": 4.1688,
+      "step": 431616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.292483177330688e-05,
+      "loss": 4.1712,
+      "step": 432128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.291644582579636e-05,
+      "loss": 4.1743,
+      "step": 432640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.290805987828583e-05,
+      "loss": 4.1751,
+      "step": 433152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.289967393077531e-05,
+      "loss": 4.1671,
+      "step": 433664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.289128798326479e-05,
+      "loss": 4.1627,
+      "step": 434176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.288290203575427e-05,
+      "loss": 4.1742,
+      "step": 434688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.287453246704749e-05,
+      "loss": 4.1506,
+      "step": 435200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.286614651953697e-05,
+      "loss": 4.1662,
+      "step": 435712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.285776057202645e-05,
+      "loss": 4.1647,
+      "step": 436224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.284937462451593e-05,
+      "loss": 4.1716,
+      "step": 436736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.284098867700541e-05,
+      "loss": 4.1516,
+      "step": 437248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.283260272949489e-05,
+      "loss": 4.1636,
+      "step": 437760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.282421678198437e-05,
+      "loss": 4.1634,
+      "step": 438272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.281583083447385e-05,
+      "loss": 4.161,
+      "step": 438784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2807461265767056e-05,
+      "loss": 4.1641,
+      "step": 439296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2799075318256536e-05,
+      "loss": 4.1534,
+      "step": 439808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2790689370746016e-05,
+      "loss": 4.172,
+      "step": 440320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2782319802039225e-05,
+      "loss": 4.1683,
+      "step": 440832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2773933854528705e-05,
+      "loss": 4.1572,
+      "step": 441344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.276554790701819e-05,
+      "loss": 4.1715,
+      "step": 441856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.275716195950767e-05,
+      "loss": 4.1564,
+      "step": 442368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.274877601199715e-05,
+      "loss": 4.1716,
+      "step": 442880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.274039006448663e-05,
+      "loss": 4.1566,
+      "step": 443392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.273200411697611e-05,
+      "loss": 4.1695,
+      "step": 443904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.272361816946559e-05,
+      "loss": 4.154,
+      "step": 444416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.27152486007588e-05,
+      "loss": 4.1647,
+      "step": 444928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.270687903205201e-05,
+      "loss": 4.1573,
+      "step": 445440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.269849308454149e-05,
+      "loss": 4.1671,
+      "step": 445952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.269010713703097e-05,
+      "loss": 4.1541,
+      "step": 446464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.268172118952045e-05,
+      "loss": 4.1612,
+      "step": 446976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.267333524200993e-05,
+      "loss": 4.1405,
+      "step": 447488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.266494929449941e-05,
+      "loss": 4.1775,
+      "step": 448000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.265656334698889e-05,
+      "loss": 4.1565,
+      "step": 448512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.264817739947837e-05,
+      "loss": 4.164,
+      "step": 449024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2639807830771585e-05,
+      "loss": 4.149,
+      "step": 449536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2631421883261065e-05,
+      "loss": 4.1679,
+      "step": 450048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2623035935750545e-05,
+      "loss": 4.1687,
+      "step": 450560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.261464998824002e-05,
+      "loss": 4.1563,
+      "step": 451072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.26062640407295e-05,
+      "loss": 4.1567,
+      "step": 451584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2597894472022714e-05,
+      "loss": 4.1635,
+      "step": 452096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.258950852451219e-05,
+      "loss": 4.1647,
+      "step": 452608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.258112257700167e-05,
+      "loss": 4.173,
+      "step": 453120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.257273662949115e-05,
+      "loss": 4.1437,
+      "step": 453632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.256435068198063e-05,
+      "loss": 4.1655,
+      "step": 454144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.255596473447011e-05,
+      "loss": 4.1523,
+      "step": 454656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2547578786959594e-05,
+      "loss": 4.1629,
+      "step": 455168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.2539192839449074e-05,
+      "loss": 4.1598,
+      "step": 455680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.253082327074228e-05,
+      "loss": 4.1461,
+      "step": 456192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.252243732323176e-05,
+      "loss": 4.1657,
+      "step": 456704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.251405137572124e-05,
+      "loss": 4.1575,
+      "step": 457216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.250568180701445e-05,
+      "loss": 4.1606,
+      "step": 457728
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.2099761962890625,
+      "eval_runtime": 300.7093,
+      "eval_samples_per_second": 1268.97,
+      "eval_steps_per_second": 39.656,
+      "step": 457920
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.249729585950393e-05,
+      "loss": 4.1547,
+      "step": 458240
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.248890991199341e-05,
+      "loss": 4.1545,
+      "step": 458752
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.248052396448289e-05,
+      "loss": 4.165,
+      "step": 459264
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.24721543957761e-05,
+      "loss": 4.1579,
+      "step": 459776
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.246376844826558e-05,
+      "loss": 4.1607,
+      "step": 460288
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.245538250075506e-05,
+      "loss": 4.1528,
+      "step": 460800
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.244699655324455e-05,
+      "loss": 4.1464,
+      "step": 461312
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.243861060573403e-05,
+      "loss": 4.1466,
+      "step": 461824
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.243022465822351e-05,
+      "loss": 4.1561,
+      "step": 462336
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.242183871071299e-05,
+      "loss": 4.1603,
+      "step": 462848
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.241345276320247e-05,
+      "loss": 4.1534,
+      "step": 463360
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.240508319449568e-05,
+      "loss": 4.1592,
+      "step": 463872
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2396713625788886e-05,
+      "loss": 4.1449,
+      "step": 464384
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2388327678278366e-05,
+      "loss": 4.1509,
+      "step": 464896
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2379941730767846e-05,
+      "loss": 4.1495,
+      "step": 465408
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2371555783257326e-05,
+      "loss": 4.1373,
+      "step": 465920
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2363169835746806e-05,
+      "loss": 4.1524,
+      "step": 466432
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2354783888236286e-05,
+      "loss": 4.1371,
+      "step": 466944
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2346397940725765e-05,
+      "loss": 4.1446,
+      "step": 467456
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.233802837201898e-05,
+      "loss": 4.1805,
+      "step": 467968
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.232964242450846e-05,
+      "loss": 4.1443,
+      "step": 468480
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.232125647699794e-05,
+      "loss": 4.1508,
+      "step": 468992
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.231287052948742e-05,
+      "loss": 4.1525,
+      "step": 469504
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.23044845819769e-05,
+      "loss": 4.1598,
+      "step": 470016
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2296098634466374e-05,
+      "loss": 4.1364,
+      "step": 470528
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2287712686955854e-05,
+      "loss": 4.1449,
+      "step": 471040
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2279326739445334e-05,
+      "loss": 4.1476,
+      "step": 471552
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2270940791934814e-05,
+      "loss": 4.139,
+      "step": 472064
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2262554844424294e-05,
+      "loss": 4.1447,
+      "step": 472576
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.2254168896913774e-05,
+      "loss": 4.1379,
+      "step": 473088
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.2245782949403254e-05,
+      "loss": 4.1455,
+      "step": 473600
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.223741338069647e-05,
+      "loss": 4.1518,
+      "step": 474112
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.222902743318595e-05,
+      "loss": 4.1495,
+      "step": 474624
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.222064148567543e-05,
+      "loss": 4.1425,
+      "step": 475136
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.221227191696864e-05,
+      "loss": 4.1482,
+      "step": 475648
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.220388596945812e-05,
+      "loss": 4.154,
+      "step": 476160
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.219551640075133e-05,
+      "loss": 4.1384,
+      "step": 476672
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.218713045324081e-05,
+      "loss": 4.1565,
+      "step": 477184
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.217874450573029e-05,
+      "loss": 4.1297,
+      "step": 477696
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.217035855821977e-05,
+      "loss": 4.151,
+      "step": 478208
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.216197261070925e-05,
+      "loss": 4.1278,
+      "step": 478720
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.215358666319873e-05,
+      "loss": 4.1583,
+      "step": 479232
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.214520071568821e-05,
+      "loss": 4.1427,
+      "step": 479744
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.213681476817769e-05,
+      "loss": 4.1473,
+      "step": 480256
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.212842882066717e-05,
+      "loss": 4.1445,
+      "step": 480768
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.212004287315665e-05,
+      "loss": 4.1453,
+      "step": 481280
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.2111656925646134e-05,
+      "loss": 4.1478,
+      "step": 481792
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.2103270978135614e-05,
+      "loss": 4.1384,
+      "step": 482304
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.2094901409428823e-05,
+      "loss": 4.1214,
+      "step": 482816
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.20865154619183e-05,
+      "loss": 4.1566,
+      "step": 483328
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.207812951440778e-05,
+      "loss": 4.1413,
+      "step": 483840
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.206974356689726e-05,
+      "loss": 4.1519,
+      "step": 484352
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.206137399819047e-05,
+      "loss": 4.1315,
+      "step": 484864
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.205298805067995e-05,
+      "loss": 4.1403,
+      "step": 485376
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.204460210316943e-05,
+      "loss": 4.1316,
+      "step": 485888
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.203621615565891e-05,
+      "loss": 4.1465,
+      "step": 486400
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.2027830208148385e-05,
+      "loss": 4.1407,
+      "step": 486912
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.201944426063787e-05,
+      "loss": 4.1492,
+      "step": 487424
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.201107469193109e-05,
+      "loss": 4.1465,
+      "step": 487936
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.200268874442057e-05,
+      "loss": 4.1293,
+      "step": 488448
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.199430279691004e-05,
+      "loss": 4.1339,
+      "step": 488960
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.198591684939952e-05,
+      "loss": 4.1505,
+      "step": 489472
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.1977530901889e-05,
+      "loss": 4.1258,
+      "step": 489984
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.196914495437848e-05,
+      "loss": 4.1333,
+      "step": 490496
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.196075900686796e-05,
+      "loss": 4.1373,
+      "step": 491008
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.195237305935744e-05,
+      "loss": 4.1463,
+      "step": 491520
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.1944019869454386e-05,
+      "loss": 4.1203,
+      "step": 492032
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.193563392194386e-05,
+      "loss": 4.1284,
+      "step": 492544
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.192724797443334e-05,
+      "loss": 4.1277,
+      "step": 493056
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.1918862026922826e-05,
+      "loss": 4.1361,
+      "step": 493568
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.1910476079412306e-05,
+      "loss": 4.1413,
+      "step": 494080
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.1902090131901786e-05,
+      "loss": 4.1373,
+      "step": 494592
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.1893704184391266e-05,
+      "loss": 4.137,
+      "step": 495104
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.1885318236880746e-05,
+      "loss": 4.1302,
+      "step": 495616
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.1876948668173955e-05,
+      "loss": 4.1539,
+      "step": 496128
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.1868579099467164e-05,
+      "loss": 4.1225,
+      "step": 496640
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.1860193151956644e-05,
+      "loss": 4.1372,
+      "step": 497152
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.185182358324986e-05,
+      "loss": 4.1379,
+      "step": 497664
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.184343763573933e-05,
+      "loss": 4.1331,
+      "step": 498176
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.183505168822881e-05,
+      "loss": 4.1388,
+      "step": 498688
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.182666574071829e-05,
+      "loss": 4.1354,
+      "step": 499200
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.181827979320778e-05,
+      "loss": 4.1299,
+      "step": 499712
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.180989384569726e-05,
+      "loss": 4.1427,
+      "step": 500224
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.180150789818674e-05,
+      "loss": 4.1229,
+      "step": 500736
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.179312195067622e-05,
+      "loss": 4.1363,
+      "step": 501248
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.17847360031657e-05,
+      "loss": 4.13,
+      "step": 501760
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.177635005565518e-05,
+      "loss": 4.1402,
+      "step": 502272
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.176796410814466e-05,
+      "loss": 4.1274,
+      "step": 502784
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.175957816063414e-05,
+      "loss": 4.1228,
+      "step": 503296
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.175120859192735e-05,
+      "loss": 4.1294,
+      "step": 503808
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.174282264441683e-05,
+      "loss": 4.1335,
+      "step": 504320
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.173445307571004e-05,
+      "loss": 4.1331,
+      "step": 504832
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.172606712819952e-05,
+      "loss": 4.118,
+      "step": 505344
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1717681180689e-05,
+      "loss": 4.1332,
+      "step": 505856
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.170929523317848e-05,
+      "loss": 4.1402,
+      "step": 506368
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1700909285667964e-05,
+      "loss": 4.1302,
+      "step": 506880
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1692523338157444e-05,
+      "loss": 4.1142,
+      "step": 507392
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.168415376945065e-05,
+      "loss": 4.1265,
+      "step": 507904
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.167576782194013e-05,
+      "loss": 4.1243,
+      "step": 508416
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.166738187442961e-05,
+      "loss": 4.1358,
+      "step": 508928
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.165899592691909e-05,
+      "loss": 4.135,
+      "step": 509440
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.165060997940857e-05,
+      "loss": 4.1215,
+      "step": 509952
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1642224031898046e-05,
+      "loss": 4.1212,
+      "step": 510464
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1633838084387526e-05,
+      "loss": 4.1325,
+      "step": 510976
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1625452136877006e-05,
+      "loss": 4.1102,
+      "step": 511488
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.161708256817022e-05,
+      "loss": 4.1274,
+      "step": 512000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.160871299946343e-05,
+      "loss": 4.1182,
+      "step": 512512
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.160032705195292e-05,
+      "loss": 4.1349,
+      "step": 513024
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.159195748324613e-05,
+      "loss": 4.1094,
+      "step": 513536
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.158357153573561e-05,
+      "loss": 4.1213,
+      "step": 514048
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1575185588225087e-05,
+      "loss": 4.1201,
+      "step": 514560
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1566799640714567e-05,
+      "loss": 4.1267,
+      "step": 515072
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1558413693204047e-05,
+      "loss": 4.1193,
+      "step": 515584
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.155002774569352e-05,
+      "loss": 4.1147,
+      "step": 516096
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1541641798183e-05,
+      "loss": 4.1271,
+      "step": 516608
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.153325585067248e-05,
+      "loss": 4.129,
+      "step": 517120
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1524902660769425e-05,
+      "loss": 4.1155,
+      "step": 517632
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1516516713258905e-05,
+      "loss": 4.1322,
+      "step": 518144
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1508130765748385e-05,
+      "loss": 4.1158,
+      "step": 518656
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1499744818237864e-05,
+      "loss": 4.1247,
+      "step": 519168
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1491358870727344e-05,
+      "loss": 4.1181,
+      "step": 519680
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1482972923216824e-05,
+      "loss": 4.1256,
+      "step": 520192
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1474586975706304e-05,
+      "loss": 4.1143,
+      "step": 520704
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1466201028195784e-05,
+      "loss": 4.1229,
+      "step": 521216
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1457831459488993e-05,
+      "loss": 4.1252,
+      "step": 521728
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.144944551197847e-05,
+      "loss": 4.1215,
+      "step": 522240
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.144105956446795e-05,
+      "loss": 4.1129,
+      "step": 522752
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.143267361695743e-05,
+      "loss": 4.1227,
+      "step": 523264
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.142428766944691e-05,
+      "loss": 4.1039,
+      "step": 523776
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.141591810074012e-05,
+      "loss": 4.1336,
+      "step": 524288
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.14075321532296e-05,
+      "loss": 4.1194,
+      "step": 524800
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.139914620571909e-05,
+      "loss": 4.1225,
+      "step": 525312
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.139076025820857e-05,
+      "loss": 4.1109,
+      "step": 525824
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.138237431069805e-05,
+      "loss": 4.1321,
+      "step": 526336
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.137398836318753e-05,
+      "loss": 4.1228,
+      "step": 526848
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.136560241567701e-05,
+      "loss": 4.1169,
+      "step": 527360
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.135721646816649e-05,
+      "loss": 4.1162,
+      "step": 527872
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.134883052065597e-05,
+      "loss": 4.1236,
+      "step": 528384
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.134046095194918e-05,
+      "loss": 4.1225,
+      "step": 528896
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.133207500443866e-05,
+      "loss": 4.1327,
+      "step": 529408
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.132370543573187e-05,
+      "loss": 4.1032,
+      "step": 529920
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.131531948822135e-05,
+      "loss": 4.1238,
+      "step": 530432
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.130693354071083e-05,
+      "loss": 4.1139,
+      "step": 530944
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.129854759320031e-05,
+      "loss": 4.1251,
+      "step": 531456
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.129016164568979e-05,
+      "loss": 4.1168,
+      "step": 531968
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1281775698179273e-05,
+      "loss": 4.1111,
+      "step": 532480
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.1273389750668753e-05,
+      "loss": 4.1251,
+      "step": 532992
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.126500380315823e-05,
+      "loss": 4.1208,
+      "step": 533504
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.125663423445144e-05,
+      "loss": 4.121,
+      "step": 534016
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.180113792419434,
+      "eval_runtime": 298.6947,
+      "eval_samples_per_second": 1277.529,
+      "eval_steps_per_second": 39.924,
+      "step": 534240
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.124824828694092e-05,
+      "loss": 4.1185,
+      "step": 534528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.12398623394304e-05,
+      "loss": 4.1143,
+      "step": 535040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.123147639191988e-05,
+      "loss": 4.1244,
+      "step": 535552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1223090444409356e-05,
+      "loss": 4.1223,
+      "step": 536064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1214704496898835e-05,
+      "loss": 4.1245,
+      "step": 536576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1206318549388315e-05,
+      "loss": 4.1102,
+      "step": 537088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1197932601877795e-05,
+      "loss": 4.1073,
+      "step": 537600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1189546654367275e-05,
+      "loss": 4.1069,
+      "step": 538112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1181160706856755e-05,
+      "loss": 4.1193,
+      "step": 538624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.117277475934624e-05,
+      "loss": 4.1254,
+      "step": 539136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.116438881183572e-05,
+      "loss": 4.1096,
+      "step": 539648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.11560028643252e-05,
+      "loss": 4.1215,
+      "step": 540160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.114763329561841e-05,
+      "loss": 4.1129,
+      "step": 540672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.113924734810789e-05,
+      "loss": 4.1112,
+      "step": 541184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.113086140059737e-05,
+      "loss": 4.1087,
+      "step": 541696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.112247545308685e-05,
+      "loss": 4.1035,
+      "step": 542208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.111408950557633e-05,
+      "loss": 4.1117,
+      "step": 542720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.110570355806581e-05,
+      "loss": 4.1045,
+      "step": 543232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.109731761055529e-05,
+      "loss": 4.1035,
+      "step": 543744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.108893166304477e-05,
+      "loss": 4.1378,
+      "step": 544256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1080545715534244e-05,
+      "loss": 4.1125,
+      "step": 544768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1072159768023724e-05,
+      "loss": 4.1116,
+      "step": 545280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.106377382051321e-05,
+      "loss": 4.1145,
+      "step": 545792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.105538787300269e-05,
+      "loss": 4.1179,
+      "step": 546304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.10470183042959e-05,
+      "loss": 4.1009,
+      "step": 546816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1038648735589115e-05,
+      "loss": 4.1095,
+      "step": 547328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.1030262788078595e-05,
+      "loss": 4.1121,
+      "step": 547840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.102187684056807e-05,
+      "loss": 4.1062,
+      "step": 548352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.101349089305755e-05,
+      "loss": 4.1026,
+      "step": 548864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.100510494554703e-05,
+      "loss": 4.1014,
+      "step": 549376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.099671899803651e-05,
+      "loss": 4.1088,
+      "step": 549888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.098833305052599e-05,
+      "loss": 4.1162,
+      "step": 550400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.09799634818192e-05,
+      "loss": 4.1119,
+      "step": 550912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.097157753430868e-05,
+      "loss": 4.1042,
+      "step": 551424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.096319158679816e-05,
+      "loss": 4.1115,
+      "step": 551936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0954805639287644e-05,
+      "loss": 4.1148,
+      "step": 552448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0946419691777124e-05,
+      "loss": 4.1074,
+      "step": 552960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0938033744266604e-05,
+      "loss": 4.1138,
+      "step": 553472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0929647796756084e-05,
+      "loss": 4.0968,
+      "step": 553984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0921261849245564e-05,
+      "loss": 4.1115,
+      "step": 554496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.091289228053877e-05,
+      "loss": 4.0931,
+      "step": 555008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.090450633302825e-05,
+      "loss": 4.117,
+      "step": 555520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.089613676432146e-05,
+      "loss": 4.1102,
+      "step": 556032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.088775081681094e-05,
+      "loss": 4.11,
+      "step": 556544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.087936486930042e-05,
+      "loss": 4.1115,
+      "step": 557056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.08709789217899e-05,
+      "loss": 4.1049,
+      "step": 557568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.086259297427938e-05,
+      "loss": 4.1115,
+      "step": 558080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.08542234055726e-05,
+      "loss": 4.1018,
+      "step": 558592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.084583745806208e-05,
+      "loss": 4.0807,
+      "step": 559104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.083745151055156e-05,
+      "loss": 4.1199,
+      "step": 559616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.082906556304104e-05,
+      "loss": 4.1047,
+      "step": 560128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.082067961553052e-05,
+      "loss": 4.1171,
+      "step": 560640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.081229366802e-05,
+      "loss": 4.0986,
+      "step": 561152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.080390772050948e-05,
+      "loss": 4.103,
+      "step": 561664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.079553815180269e-05,
+      "loss": 4.0926,
+      "step": 562176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.078715220429217e-05,
+      "loss": 4.112,
+      "step": 562688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0778766256781647e-05,
+      "loss": 4.1037,
+      "step": 563200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0770380309271127e-05,
+      "loss": 4.1093,
+      "step": 563712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0761994361760607e-05,
+      "loss": 4.1162,
+      "step": 564224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0753624793053816e-05,
+      "loss": 4.0952,
+      "step": 564736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0745238845543296e-05,
+      "loss": 4.0971,
+      "step": 565248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.073685289803278e-05,
+      "loss": 4.1155,
+      "step": 565760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0728466950522255e-05,
+      "loss": 4.092,
+      "step": 566272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0720081003011735e-05,
+      "loss": 4.0938,
+      "step": 566784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0711695055501215e-05,
+      "loss": 4.1037,
+      "step": 567296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.070332548679443e-05,
+      "loss": 4.1082,
+      "step": 567808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0694939539283904e-05,
+      "loss": 4.0886,
+      "step": 568320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0686553591773384e-05,
+      "loss": 4.0937,
+      "step": 568832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0678167644262864e-05,
+      "loss": 4.0875,
+      "step": 569344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0669781696752344e-05,
+      "loss": 4.102,
+      "step": 569856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0661395749241824e-05,
+      "loss": 4.1077,
+      "step": 570368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0653009801731304e-05,
+      "loss": 4.1024,
+      "step": 570880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.064464023302452e-05,
+      "loss": 4.0992,
+      "step": 571392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0636254285514e-05,
+      "loss": 4.0997,
+      "step": 571904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.062786833800348e-05,
+      "loss": 4.1174,
+      "step": 572416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.061948239049296e-05,
+      "loss": 4.0883,
+      "step": 572928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.061111282178617e-05,
+      "loss": 4.1054,
+      "step": 573440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.060272687427565e-05,
+      "loss": 4.102,
+      "step": 573952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.059434092676513e-05,
+      "loss": 4.0938,
+      "step": 574464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.058595497925461e-05,
+      "loss": 4.1079,
+      "step": 574976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.057756903174409e-05,
+      "loss": 4.0995,
+      "step": 575488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.056918308423357e-05,
+      "loss": 4.0973,
+      "step": 576000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.056079713672305e-05,
+      "loss": 4.1071,
+      "step": 576512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.055241118921253e-05,
+      "loss": 4.0877,
+      "step": 577024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.054402524170201e-05,
+      "loss": 4.0999,
+      "step": 577536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.053563929419149e-05,
+      "loss": 4.0958,
+      "step": 578048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.05272697254847e-05,
+      "loss": 4.1054,
+      "step": 578560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0518883777974184e-05,
+      "loss": 4.0917,
+      "step": 579072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.0510497830463664e-05,
+      "loss": 4.0933,
+      "step": 579584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0502111882953144e-05,
+      "loss": 4.0891,
+      "step": 580096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.049372593544262e-05,
+      "loss": 4.101,
+      "step": 580608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0485356366735833e-05,
+      "loss": 4.0995,
+      "step": 581120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0476970419225313e-05,
+      "loss": 4.0831,
+      "step": 581632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.046858447171479e-05,
+      "loss": 4.1031,
+      "step": 582144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0460198524204267e-05,
+      "loss": 4.1041,
+      "step": 582656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.045182895549748e-05,
+      "loss": 4.0946,
+      "step": 583168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.044344300798696e-05,
+      "loss": 4.0792,
+      "step": 583680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.043505706047644e-05,
+      "loss": 4.0916,
+      "step": 584192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.042667111296592e-05,
+      "loss": 4.0889,
+      "step": 584704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.04182851654554e-05,
+      "loss": 4.104,
+      "step": 585216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.040989921794488e-05,
+      "loss": 4.1,
+      "step": 585728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.040152964923809e-05,
+      "loss": 4.0896,
+      "step": 586240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.039314370172757e-05,
+      "loss": 4.0896,
+      "step": 586752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.038475775421705e-05,
+      "loss": 4.0919,
+      "step": 587264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.037637180670653e-05,
+      "loss": 4.0836,
+      "step": 587776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.036798585919601e-05,
+      "loss": 4.0906,
+      "step": 588288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.035959991168549e-05,
+      "loss": 4.0851,
+      "step": 588800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.035121396417497e-05,
+      "loss": 4.1032,
+      "step": 589312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.034282801666445e-05,
+      "loss": 4.0766,
+      "step": 589824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.033444206915393e-05,
+      "loss": 4.0874,
+      "step": 590336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.032607250044714e-05,
+      "loss": 4.0857,
+      "step": 590848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.031768655293662e-05,
+      "loss": 4.0952,
+      "step": 591360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.030930060542611e-05,
+      "loss": 4.0834,
+      "step": 591872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0300931036719316e-05,
+      "loss": 4.0825,
+      "step": 592384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0292545089208796e-05,
+      "loss": 4.0965,
+      "step": 592896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0284159141698276e-05,
+      "loss": 4.0959,
+      "step": 593408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0275773194187756e-05,
+      "loss": 4.0805,
+      "step": 593920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0267403625480965e-05,
+      "loss": 4.0966,
+      "step": 594432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0259017677970445e-05,
+      "loss": 4.0865,
+      "step": 594944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0250631730459925e-05,
+      "loss": 4.0925,
+      "step": 595456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0242245782949405e-05,
+      "loss": 4.0869,
+      "step": 595968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0233859835438885e-05,
+      "loss": 4.0908,
+      "step": 596480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0225473887928365e-05,
+      "loss": 4.0837,
+      "step": 596992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0217087940417845e-05,
+      "loss": 4.092,
+      "step": 597504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0208701992907324e-05,
+      "loss": 4.0942,
+      "step": 598016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.020033242420054e-05,
+      "loss": 4.0892,
+      "step": 598528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.019194647669002e-05,
+      "loss": 4.0822,
+      "step": 599040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.018357690798323e-05,
+      "loss": 4.0879,
+      "step": 599552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.017519096047271e-05,
+      "loss": 4.072,
+      "step": 600064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.016680501296219e-05,
+      "loss": 4.1044,
+      "step": 600576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.015841906545167e-05,
+      "loss": 4.0813,
+      "step": 601088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.015003311794115e-05,
+      "loss": 4.0914,
+      "step": 601600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.014164717043063e-05,
+      "loss": 4.0778,
+      "step": 602112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.013327760172384e-05,
+      "loss": 4.0983,
+      "step": 602624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.012489165421332e-05,
+      "loss": 4.0923,
+      "step": 603136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.01165057067028e-05,
+      "loss": 4.0841,
+      "step": 603648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.010811975919228e-05,
+      "loss": 4.0878,
+      "step": 604160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.009973381168176e-05,
+      "loss": 4.0869,
+      "step": 604672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.009134786417124e-05,
+      "loss": 4.0895,
+      "step": 605184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.008296191666072e-05,
+      "loss": 4.1051,
+      "step": 605696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.00745759691502e-05,
+      "loss": 4.0674,
+      "step": 606208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.006620640044341e-05,
+      "loss": 4.0965,
+      "step": 606720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.005782045293289e-05,
+      "loss": 4.0816,
+      "step": 607232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.004943450542237e-05,
+      "loss": 4.0916,
+      "step": 607744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.004104855791185e-05,
+      "loss": 4.0818,
+      "step": 608256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.003266261040133e-05,
+      "loss": 4.085,
+      "step": 608768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.0024293041694536e-05,
+      "loss": 4.0915,
+      "step": 609280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.001592347298775e-05,
+      "loss": 4.089,
+      "step": 609792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.000753752547723e-05,
+      "loss": 4.0874,
+      "step": 610304
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.156889915466309,
+      "eval_runtime": 280.1064,
+      "eval_samples_per_second": 1362.307,
+      "eval_steps_per_second": 42.573,
+      "step": 610560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.999915157796671e-05,
+      "loss": 4.0867,
+      "step": 610816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.999076563045619e-05,
+      "loss": 4.0824,
+      "step": 611328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.998237968294567e-05,
+      "loss": 4.09,
+      "step": 611840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.997399373543515e-05,
+      "loss": 4.0883,
+      "step": 612352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.996560778792463e-05,
+      "loss": 4.0944,
+      "step": 612864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.995723821921784e-05,
+      "loss": 4.0782,
+      "step": 613376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.994885227170732e-05,
+      "loss": 4.0774,
+      "step": 613888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.99404663241968e-05,
+      "loss": 4.0796,
+      "step": 614400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.993208037668628e-05,
+      "loss": 4.085,
+      "step": 614912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.992369442917576e-05,
+      "loss": 4.0928,
+      "step": 615424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.991532486046897e-05,
+      "loss": 4.0797,
+      "step": 615936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.990693891295845e-05,
+      "loss": 4.0871,
+      "step": 616448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.98985857230554e-05,
+      "loss": 4.0847,
+      "step": 616960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9890199775544875e-05,
+      "loss": 4.0767,
+      "step": 617472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9881813828034354e-05,
+      "loss": 4.0804,
+      "step": 617984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9873427880523834e-05,
+      "loss": 4.0684,
+      "step": 618496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9865041933013314e-05,
+      "loss": 4.0813,
+      "step": 619008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9856655985502794e-05,
+      "loss": 4.0702,
+      "step": 619520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9848270037992274e-05,
+      "loss": 4.075,
+      "step": 620032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9839884090481754e-05,
+      "loss": 4.1024,
+      "step": 620544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9831498142971234e-05,
+      "loss": 4.0866,
+      "step": 621056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9823112195460714e-05,
+      "loss": 4.0801,
+      "step": 621568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.981474262675392e-05,
+      "loss": 4.0817,
+      "step": 622080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.98063566792434e-05,
+      "loss": 4.0892,
+      "step": 622592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.979798711053662e-05,
+      "loss": 4.0646,
+      "step": 623104
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.97896011630261e-05,
+      "loss": 4.0817,
+      "step": 623616
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.978121521551558e-05,
+      "loss": 4.0831,
+      "step": 624128
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.977282926800506e-05,
+      "loss": 4.0708,
+      "step": 624640
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.976444332049454e-05,
+      "loss": 4.0706,
+      "step": 625152
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.975605737298402e-05,
+      "loss": 4.0727,
+      "step": 625664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.97476714254735e-05,
+      "loss": 4.077,
+      "step": 626176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.973928547796298e-05,
+      "loss": 4.0862,
+      "step": 626688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.973091590925619e-05,
+      "loss": 4.0801,
+      "step": 627200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.972252996174567e-05,
+      "loss": 4.0765,
+      "step": 627712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.971416039303888e-05,
+      "loss": 4.0803,
+      "step": 628224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.970577444552836e-05,
+      "loss": 4.0818,
+      "step": 628736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.969738849801784e-05,
+      "loss": 4.078,
+      "step": 629248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9689002550507324e-05,
+      "loss": 4.0816,
+      "step": 629760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9680616602996804e-05,
+      "loss": 4.0663,
+      "step": 630272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9672230655486283e-05,
+      "loss": 4.0787,
+      "step": 630784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.966386108677949e-05,
+      "loss": 4.0665,
+      "step": 631296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.965547513926897e-05,
+      "loss": 4.0874,
+      "step": 631808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.964708919175845e-05,
+      "loss": 4.0798,
+      "step": 632320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.963870324424793e-05,
+      "loss": 4.0763,
+      "step": 632832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9630317296737406e-05,
+      "loss": 4.0822,
+      "step": 633344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9621931349226886e-05,
+      "loss": 4.0762,
+      "step": 633856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9613545401716366e-05,
+      "loss": 4.0852,
+      "step": 634368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9605159454205845e-05,
+      "loss": 4.072,
+      "step": 634880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.959678988549906e-05,
+      "loss": 4.0464,
+      "step": 635392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.958840393798854e-05,
+      "loss": 4.0882,
+      "step": 635904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.958001799047802e-05,
+      "loss": 4.0749,
+      "step": 636416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.95716320429675e-05,
+      "loss": 4.0882,
+      "step": 636928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.956324609545698e-05,
+      "loss": 4.0702,
+      "step": 637440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.955486014794646e-05,
+      "loss": 4.07,
+      "step": 637952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.954647420043594e-05,
+      "loss": 4.0636,
+      "step": 638464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.953808825292542e-05,
+      "loss": 4.0789,
+      "step": 638976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.952971868421863e-05,
+      "loss": 4.0777,
+      "step": 639488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.952133273670811e-05,
+      "loss": 4.0761,
+      "step": 640000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.951294678919759e-05,
+      "loss": 4.0884,
+      "step": 640512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.950456084168707e-05,
+      "loss": 4.0679,
+      "step": 641024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.949617489417655e-05,
+      "loss": 4.0639,
+      "step": 641536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.948780532546976e-05,
+      "loss": 4.0855,
+      "step": 642048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9479419377959246e-05,
+      "loss": 4.0616,
+      "step": 642560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9471033430448726e-05,
+      "loss": 4.0664,
+      "step": 643072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9462647482938206e-05,
+      "loss": 4.0709,
+      "step": 643584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9454261535427686e-05,
+      "loss": 4.0811,
+      "step": 644096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9445875587917166e-05,
+      "loss": 4.0586,
+      "step": 644608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9437489640406646e-05,
+      "loss": 4.0656,
+      "step": 645120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9429103692896126e-05,
+      "loss": 4.0592,
+      "step": 645632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9420734124189335e-05,
+      "loss": 4.0705,
+      "step": 646144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9412364555482544e-05,
+      "loss": 4.0794,
+      "step": 646656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9403978607972024e-05,
+      "loss": 4.0708,
+      "step": 647168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9395592660461504e-05,
+      "loss": 4.0715,
+      "step": 647680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9387206712950984e-05,
+      "loss": 4.0703,
+      "step": 648192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9378820765440464e-05,
+      "loss": 4.0925,
+      "step": 648704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9370434817929943e-05,
+      "loss": 4.061,
+      "step": 649216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.936206524922316e-05,
+      "loss": 4.0728,
+      "step": 649728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.935367930171264e-05,
+      "loss": 4.0754,
+      "step": 650240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.934529335420212e-05,
+      "loss": 4.0669,
+      "step": 650752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.933690740669159e-05,
+      "loss": 4.0758,
+      "step": 651264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.932852145918107e-05,
+      "loss": 4.0741,
+      "step": 651776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.932013551167055e-05,
+      "loss": 4.0695,
+      "step": 652288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.931174956416003e-05,
+      "loss": 4.0753,
+      "step": 652800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.930336361664951e-05,
+      "loss": 4.0634,
+      "step": 653312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.929499404794272e-05,
+      "loss": 4.0672,
+      "step": 653824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.92866081004322e-05,
+      "loss": 4.0699,
+      "step": 654336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.927822215292168e-05,
+      "loss": 4.0769,
+      "step": 654848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.926983620541116e-05,
+      "loss": 4.0624,
+      "step": 655360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.926145025790065e-05,
+      "loss": 4.0646,
+      "step": 655872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.925306431039013e-05,
+      "loss": 4.0607,
+      "step": 656384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.924467836287961e-05,
+      "loss": 4.0751,
+      "step": 656896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.923630879417282e-05,
+      "loss": 4.0702,
+      "step": 657408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.92279228466623e-05,
+      "loss": 4.0572,
+      "step": 657920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.921953689915178e-05,
+      "loss": 4.0725,
+      "step": 658432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.921115095164126e-05,
+      "loss": 4.0758,
+      "step": 658944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9202781382934466e-05,
+      "loss": 4.067,
+      "step": 659456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9194395435423946e-05,
+      "loss": 4.0501,
+      "step": 659968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9186009487913426e-05,
+      "loss": 4.0643,
+      "step": 660480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9177623540402906e-05,
+      "loss": 4.0587,
+      "step": 660992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9169253971696115e-05,
+      "loss": 4.0752,
+      "step": 661504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.91608680241856e-05,
+      "loss": 4.0694,
+      "step": 662016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.915248207667508e-05,
+      "loss": 4.0684,
+      "step": 662528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.914409612916456e-05,
+      "loss": 4.0623,
+      "step": 663040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.913571018165404e-05,
+      "loss": 4.0618,
+      "step": 663552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.912732423414352e-05,
+      "loss": 4.0568,
+      "step": 664064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9118938286633e-05,
+      "loss": 4.0614,
+      "step": 664576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.911056871792621e-05,
+      "loss": 4.0563,
+      "step": 665088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.910218277041569e-05,
+      "loss": 4.0739,
+      "step": 665600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.909379682290517e-05,
+      "loss": 4.0503,
+      "step": 666112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.908541087539465e-05,
+      "loss": 4.0542,
+      "step": 666624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.907702492788413e-05,
+      "loss": 4.06,
+      "step": 667136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9068638980373604e-05,
+      "loss": 4.0615,
+      "step": 667648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9060253032863083e-05,
+      "loss": 4.061,
+      "step": 668160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.905186708535257e-05,
+      "loss": 4.0555,
+      "step": 668672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.904348113784205e-05,
+      "loss": 4.0664,
+      "step": 669184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.903511156913526e-05,
+      "loss": 4.0717,
+      "step": 669696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9026742000428475e-05,
+      "loss": 4.0522,
+      "step": 670208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9018372431721684e-05,
+      "loss": 4.0652,
+      "step": 670720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9009986484211164e-05,
+      "loss": 4.0594,
+      "step": 671232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9001600536700644e-05,
+      "loss": 4.0657,
+      "step": 671744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8993214589190124e-05,
+      "loss": 4.0614,
+      "step": 672256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8984828641679604e-05,
+      "loss": 4.0574,
+      "step": 672768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.897644269416908e-05,
+      "loss": 4.0567,
+      "step": 673280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.896805674665856e-05,
+      "loss": 4.0634,
+      "step": 673792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.895967079914804e-05,
+      "loss": 4.0664,
+      "step": 674304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8951284851637524e-05,
+      "loss": 4.0602,
+      "step": 674816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8942898904127004e-05,
+      "loss": 4.0548,
+      "step": 675328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8934512956616484e-05,
+      "loss": 4.0592,
+      "step": 675840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.892614338790969e-05,
+      "loss": 4.046,
+      "step": 676352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.891775744039917e-05,
+      "loss": 4.0747,
+      "step": 676864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.890937149288865e-05,
+      "loss": 4.0533,
+      "step": 677376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.890098554537813e-05,
+      "loss": 4.0683,
+      "step": 677888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.889259959786761e-05,
+      "loss": 4.0483,
+      "step": 678400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.888421365035709e-05,
+      "loss": 4.0715,
+      "step": 678912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.887582770284657e-05,
+      "loss": 4.0648,
+      "step": 679424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.886744175533605e-05,
+      "loss": 4.0574,
+      "step": 679936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.885907218662926e-05,
+      "loss": 4.0599,
+      "step": 680448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.885068623911874e-05,
+      "loss": 4.0606,
+      "step": 680960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.884230029160822e-05,
+      "loss": 4.0635,
+      "step": 681472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.88339143440977e-05,
+      "loss": 4.0753,
+      "step": 681984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.882554477539092e-05,
+      "loss": 4.0431,
+      "step": 682496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.88171588278804e-05,
+      "loss": 4.0695,
+      "step": 683008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.880877288036988e-05,
+      "loss": 4.0504,
+      "step": 683520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.880038693285936e-05,
+      "loss": 4.0653,
+      "step": 684032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8792017364152566e-05,
+      "loss": 4.0559,
+      "step": 684544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8783647795445775e-05,
+      "loss": 4.0553,
+      "step": 685056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8775261847935255e-05,
+      "loss": 4.0628,
+      "step": 685568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8766875900424735e-05,
+      "loss": 4.0648,
+      "step": 686080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.8758489952914215e-05,
+      "loss": 4.0594,
+      "step": 686592
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.136818885803223,
+      "eval_runtime": 280.1538,
+      "eval_samples_per_second": 1362.077,
+      "eval_steps_per_second": 42.566,
+      "step": 686880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8750104005403695e-05,
+      "loss": 4.0586,
+      "step": 687104
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8741718057893175e-05,
+      "loss": 4.0578,
+      "step": 687616
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8733332110382655e-05,
+      "loss": 4.0641,
+      "step": 688128
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.872494616287214e-05,
+      "loss": 4.0591,
+      "step": 688640
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8716560215361615e-05,
+      "loss": 4.065,
+      "step": 689152
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8708174267851095e-05,
+      "loss": 4.0541,
+      "step": 689664
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8699788320340575e-05,
+      "loss": 4.052,
+      "step": 690176
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8691402372830055e-05,
+      "loss": 4.0535,
+      "step": 690688
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8683016425319535e-05,
+      "loss": 4.057,
+      "step": 691200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8674630477809015e-05,
+      "loss": 4.0648,
+      "step": 691712
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8666244530298495e-05,
+      "loss": 4.0568,
+      "step": 692224
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8657858582787975e-05,
+      "loss": 4.0582,
+      "step": 692736
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8649489014081184e-05,
+      "loss": 4.0609,
+      "step": 693248
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8641103066570664e-05,
+      "loss": 4.0488,
+      "step": 693760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8632717119060144e-05,
+      "loss": 4.0543,
+      "step": 694272
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8624331171549624e-05,
+      "loss": 4.044,
+      "step": 694784
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.861594522403911e-05,
+      "loss": 4.0507,
+      "step": 695296
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.860755927652859e-05,
+      "loss": 4.0471,
+      "step": 695808
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.859917332901807e-05,
+      "loss": 4.0495,
+      "step": 696320
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.859078738150755e-05,
+      "loss": 4.0716,
+      "step": 696832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.858240143399703e-05,
+      "loss": 4.0617,
+      "step": 697344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8574015486486503e-05,
+      "loss": 4.0566,
+      "step": 697856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.8565629538975983e-05,
+      "loss": 4.0551,
+      "step": 698368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.85572599702692e-05,
+      "loss": 4.0642,
+      "step": 698880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.854887402275868e-05,
+      "loss": 4.0394,
+      "step": 699392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.854048807524815e-05,
+      "loss": 4.054,
+      "step": 699904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.853210212773763e-05,
+      "loss": 4.0565,
+      "step": 700416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.852371618022711e-05,
+      "loss": 4.0403,
+      "step": 700928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.851533023271659e-05,
+      "loss": 4.0508,
+      "step": 701440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.850694428520608e-05,
+      "loss": 4.0444,
+      "step": 701952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.849855833769556e-05,
+      "loss": 4.0511,
+      "step": 702464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.849018876898877e-05,
+      "loss": 4.0542,
+      "step": 702976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.848180282147825e-05,
+      "loss": 4.0531,
+      "step": 703488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.847341687396773e-05,
+      "loss": 4.0578,
+      "step": 704000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.846503092645721e-05,
+      "loss": 4.0479,
+      "step": 704512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.845664497894669e-05,
+      "loss": 4.0564,
+      "step": 705024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.84482754102399e-05,
+      "loss": 4.0595,
+      "step": 705536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.843988946272938e-05,
+      "loss": 4.0481,
+      "step": 706048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.843150351521886e-05,
+      "loss": 4.0419,
+      "step": 706560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8423133946512066e-05,
+      "loss": 4.0529,
+      "step": 707072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8414747999001546e-05,
+      "loss": 4.042,
+      "step": 707584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.840636205149103e-05,
+      "loss": 4.0623,
+      "step": 708096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.839797610398051e-05,
+      "loss": 4.0553,
+      "step": 708608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.838959015646999e-05,
+      "loss": 4.0486,
+      "step": 709120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.838120420895947e-05,
+      "loss": 4.0564,
+      "step": 709632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.837281826144895e-05,
+      "loss": 4.0483,
+      "step": 710144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.836443231393843e-05,
+      "loss": 4.0574,
+      "step": 710656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.835604636642791e-05,
+      "loss": 4.0535,
+      "step": 711168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.834766041891739e-05,
+      "loss": 4.0183,
+      "step": 711680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8339274471406866e-05,
+      "loss": 4.0621,
+      "step": 712192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8330888523896346e-05,
+      "loss": 4.0505,
+      "step": 712704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.832253533399329e-05,
+      "loss": 4.0576,
+      "step": 713216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.831414938648277e-05,
+      "loss": 4.0441,
+      "step": 713728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.830576343897225e-05,
+      "loss": 4.0444,
+      "step": 714240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.829737749146173e-05,
+      "loss": 4.0387,
+      "step": 714752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.828899154395121e-05,
+      "loss": 4.053,
+      "step": 715264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.82806055964407e-05,
+      "loss": 4.0516,
+      "step": 715776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.827221964893017e-05,
+      "loss": 4.0481,
+      "step": 716288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.826383370141965e-05,
+      "loss": 4.0669,
+      "step": 716800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8255464132712866e-05,
+      "loss": 4.045,
+      "step": 717312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.824707818520234e-05,
+      "loss": 4.0348,
+      "step": 717824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.823869223769182e-05,
+      "loss": 4.0639,
+      "step": 718336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.82303062901813e-05,
+      "loss": 4.0357,
+      "step": 718848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8221936721474515e-05,
+      "loss": 4.0335,
+      "step": 719360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.821355077396399e-05,
+      "loss": 4.0511,
+      "step": 719872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8205181205257204e-05,
+      "loss": 4.0523,
+      "step": 720384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8196795257746684e-05,
+      "loss": 4.0367,
+      "step": 720896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8188409310236164e-05,
+      "loss": 4.0384,
+      "step": 721408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8180023362725644e-05,
+      "loss": 4.0313,
+      "step": 721920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8171637415215124e-05,
+      "loss": 4.0437,
+      "step": 722432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8163251467704604e-05,
+      "loss": 4.0547,
+      "step": 722944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8154865520194084e-05,
+      "loss": 4.0456,
+      "step": 723456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8146479572683564e-05,
+      "loss": 4.0476,
+      "step": 723968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8138093625173044e-05,
+      "loss": 4.0408,
+      "step": 724480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8129707677662524e-05,
+      "loss": 4.0637,
+      "step": 724992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.812133810895573e-05,
+      "loss": 4.0378,
+      "step": 725504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.811295216144521e-05,
+      "loss": 4.0512,
+      "step": 726016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.810456621393469e-05,
+      "loss": 4.0472,
+      "step": 726528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.809618026642417e-05,
+      "loss": 4.0436,
+      "step": 727040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.808779431891365e-05,
+      "loss": 4.049,
+      "step": 727552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.807940837140313e-05,
+      "loss": 4.0528,
+      "step": 728064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.807102242389262e-05,
+      "loss": 4.0422,
+      "step": 728576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.80626364763821e-05,
+      "loss": 4.0528,
+      "step": 729088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.805426690767531e-05,
+      "loss": 4.0404,
+      "step": 729600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.804588096016479e-05,
+      "loss": 4.0421,
+      "step": 730112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8037511391458e-05,
+      "loss": 4.0447,
+      "step": 730624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8029141822751207e-05,
+      "loss": 4.0503,
+      "step": 731136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8020772254044416e-05,
+      "loss": 4.037,
+      "step": 731648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.8012386306533896e-05,
+      "loss": 4.042,
+      "step": 732160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8004000359023376e-05,
+      "loss": 4.0381,
+      "step": 732672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7995614411512856e-05,
+      "loss": 4.0464,
+      "step": 733184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.798722846400234e-05,
+      "loss": 4.0463,
+      "step": 733696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.797884251649182e-05,
+      "loss": 4.0299,
+      "step": 734208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.79704565689813e-05,
+      "loss": 4.0488,
+      "step": 734720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.796207062147078e-05,
+      "loss": 4.0462,
+      "step": 735232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.795368467396026e-05,
+      "loss": 4.0488,
+      "step": 735744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.794529872644974e-05,
+      "loss": 4.0256,
+      "step": 736256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.793691277893922e-05,
+      "loss": 4.0369,
+      "step": 736768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.79285268314287e-05,
+      "loss": 4.0351,
+      "step": 737280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7920140883918175e-05,
+      "loss": 4.0508,
+      "step": 737792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.791177131521139e-05,
+      "loss": 4.0445,
+      "step": 738304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.790338536770087e-05,
+      "loss": 4.0469,
+      "step": 738816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.789499942019035e-05,
+      "loss": 4.0413,
+      "step": 739328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7886613472679824e-05,
+      "loss": 4.0335,
+      "step": 739840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.787824390397304e-05,
+      "loss": 4.031,
+      "step": 740352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.786985795646253e-05,
+      "loss": 4.04,
+      "step": 740864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7861472008952e-05,
+      "loss": 4.0323,
+      "step": 741376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.785308606144148e-05,
+      "loss": 4.0428,
+      "step": 741888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.784470011393096e-05,
+      "loss": 4.0316,
+      "step": 742400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.783631416642044e-05,
+      "loss": 4.0281,
+      "step": 742912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.782794459771365e-05,
+      "loss": 4.0351,
+      "step": 743424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.781955865020313e-05,
+      "loss": 4.0396,
+      "step": 743936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.781117270269261e-05,
+      "loss": 4.0354,
+      "step": 744448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.780278675518209e-05,
+      "loss": 4.0301,
+      "step": 744960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.779440080767157e-05,
+      "loss": 4.0453,
+      "step": 745472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.778601486016105e-05,
+      "loss": 4.0465,
+      "step": 745984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7777645291454264e-05,
+      "loss": 4.0313,
+      "step": 746496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7769259343943744e-05,
+      "loss": 4.0395,
+      "step": 747008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7760873396433224e-05,
+      "loss": 4.0389,
+      "step": 747520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7752487448922704e-05,
+      "loss": 4.039,
+      "step": 748032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7744101501412184e-05,
+      "loss": 4.0391,
+      "step": 748544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7735715553901664e-05,
+      "loss": 4.0347,
+      "step": 749056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7727329606391144e-05,
+      "loss": 4.0356,
+      "step": 749568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7718943658880624e-05,
+      "loss": 4.0359,
+      "step": 750080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.771057409017383e-05,
+      "loss": 4.0429,
+      "step": 750592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.770218814266331e-05,
+      "loss": 4.0374,
+      "step": 751104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.769381857395652e-05,
+      "loss": 4.028,
+      "step": 751616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7685432626446e-05,
+      "loss": 4.0377,
+      "step": 752128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.767704667893548e-05,
+      "loss": 4.0213,
+      "step": 752640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.766866073142496e-05,
+      "loss": 4.051,
+      "step": 753152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.766027478391444e-05,
+      "loss": 4.0302,
+      "step": 753664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.765188883640393e-05,
+      "loss": 4.045,
+      "step": 754176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.764350288889341e-05,
+      "loss": 4.0256,
+      "step": 754688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.763511694138289e-05,
+      "loss": 4.0473,
+      "step": 755200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.76267473726761e-05,
+      "loss": 4.0435,
+      "step": 755712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.761836142516558e-05,
+      "loss": 4.0313,
+      "step": 756224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.760999185645879e-05,
+      "loss": 4.0367,
+      "step": 756736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.760160590894827e-05,
+      "loss": 4.0334,
+      "step": 757248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.759321996143775e-05,
+      "loss": 4.0422,
+      "step": 757760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.758483401392723e-05,
+      "loss": 4.0478,
+      "step": 758272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.757644806641671e-05,
+      "loss": 4.0222,
+      "step": 758784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7568078497709916e-05,
+      "loss": 4.0406,
+      "step": 759296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7559692550199396e-05,
+      "loss": 4.0323,
+      "step": 759808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.755130660268888e-05,
+      "loss": 4.0426,
+      "step": 760320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.754292065517836e-05,
+      "loss": 4.0326,
+      "step": 760832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.7534534707667836e-05,
+      "loss": 4.0366,
+      "step": 761344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.752616513896105e-05,
+      "loss": 4.0367,
+      "step": 761856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.751777919145053e-05,
+      "loss": 4.0402,
+      "step": 762368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.750939324394001e-05,
+      "loss": 4.0388,
+      "step": 762880
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.1215620040893555,
+      "eval_runtime": 294.8761,
+      "eval_samples_per_second": 1294.072,
+      "eval_steps_per_second": 40.441,
+      "step": 763200
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7501007296429485e-05,
+      "loss": 4.0309,
+      "step": 763392
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7492621348918965e-05,
+      "loss": 4.033,
+      "step": 763904
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7484235401408445e-05,
+      "loss": 4.041,
+      "step": 764416
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7475849453897924e-05,
+      "loss": 4.0387,
+      "step": 764928
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7467463506387404e-05,
+      "loss": 4.0421,
+      "step": 765440
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.745909393768062e-05,
+      "loss": 4.0314,
+      "step": 765952
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.74507079901701e-05,
+      "loss": 4.0294,
+      "step": 766464
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.744232204265958e-05,
+      "loss": 4.0312,
+      "step": 766976
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.743393609514906e-05,
+      "loss": 4.0296,
+      "step": 767488
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.742555014763854e-05,
+      "loss": 4.0444,
+      "step": 768000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.741718057893175e-05,
+      "loss": 4.0306,
+      "step": 768512
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.740879463142123e-05,
+      "loss": 4.0323,
+      "step": 769024
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7400441441518174e-05,
+      "loss": 4.0404,
+      "step": 769536
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7392055494007654e-05,
+      "loss": 4.0251,
+      "step": 770048
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.738366954649713e-05,
+      "loss": 4.0338,
+      "step": 770560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.737528359898661e-05,
+      "loss": 4.021,
+      "step": 771072
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.736689765147609e-05,
+      "loss": 4.0274,
+      "step": 771584
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7358511703965574e-05,
+      "loss": 4.028,
+      "step": 772096
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7350125756455054e-05,
+      "loss": 4.0222,
+      "step": 772608
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.7341739808944534e-05,
+      "loss": 4.0456,
+      "step": 773120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.733337024023774e-05,
+      "loss": 4.0419,
+      "step": 773632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.732498429272722e-05,
+      "loss": 4.0334,
+      "step": 774144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.73165983452167e-05,
+      "loss": 4.0338,
+      "step": 774656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.730821239770618e-05,
+      "loss": 4.0373,
+      "step": 775168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.729984282899939e-05,
+      "loss": 4.0221,
+      "step": 775680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.729145688148887e-05,
+      "loss": 4.032,
+      "step": 776192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.728307093397835e-05,
+      "loss": 4.0311,
+      "step": 776704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.727468498646783e-05,
+      "loss": 4.0166,
+      "step": 777216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.726629903895731e-05,
+      "loss": 4.0276,
+      "step": 777728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.725791309144679e-05,
+      "loss": 4.0221,
+      "step": 778240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.724952714393627e-05,
+      "loss": 4.0328,
+      "step": 778752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.724114119642576e-05,
+      "loss": 4.0295,
+      "step": 779264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.723277162771897e-05,
+      "loss": 4.0315,
+      "step": 779776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.722438568020845e-05,
+      "loss": 4.0343,
+      "step": 780288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7216016111501657e-05,
+      "loss": 4.0223,
+      "step": 780800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7207630163991137e-05,
+      "loss": 4.0356,
+      "step": 781312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7199244216480616e-05,
+      "loss": 4.0388,
+      "step": 781824
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7190858268970096e-05,
+      "loss": 4.0235,
+      "step": 782336
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7182472321459576e-05,
+      "loss": 4.0256,
+      "step": 782848
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7174086373949056e-05,
+      "loss": 4.0254,
+      "step": 783360
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7165700426438536e-05,
+      "loss": 4.0216,
+      "step": 783872
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7157314478928016e-05,
+      "loss": 4.0377,
+      "step": 784384
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7148928531417496e-05,
+      "loss": 4.0324,
+      "step": 784896
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7140542583906976e-05,
+      "loss": 4.028,
+      "step": 785408
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7132156636396456e-05,
+      "loss": 4.0315,
+      "step": 785920
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7123770688885936e-05,
+      "loss": 4.0291,
+      "step": 786432
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7115401120179145e-05,
+      "loss": 4.0325,
+      "step": 786944
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7107015172668625e-05,
+      "loss": 4.0331,
+      "step": 787456
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.709864560396184e-05,
+      "loss": 3.9988,
+      "step": 787968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.709025965645132e-05,
+      "loss": 4.0381,
+      "step": 788480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7081873708940794e-05,
+      "loss": 4.0251,
+      "step": 788992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7073487761430274e-05,
+      "loss": 4.0361,
+      "step": 789504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7065101813919754e-05,
+      "loss": 4.0267,
+      "step": 790016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7056715866409234e-05,
+      "loss": 4.0207,
+      "step": 790528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7048329918898714e-05,
+      "loss": 4.0186,
+      "step": 791040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.7039943971388194e-05,
+      "loss": 4.0299,
+      "step": 791552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.703157440268141e-05,
+      "loss": 4.0311,
+      "step": 792064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.702318845517089e-05,
+      "loss": 4.026,
+      "step": 792576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.70148188864641e-05,
+      "loss": 4.0432,
+      "step": 793088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.700643293895358e-05,
+      "loss": 4.0293,
+      "step": 793600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.699804699144306e-05,
+      "loss": 4.0076,
+      "step": 794112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.698966104393254e-05,
+      "loss": 4.0392,
+      "step": 794624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.698127509642202e-05,
+      "loss": 4.0165,
+      "step": 795136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.69728891489115e-05,
+      "loss": 4.0164,
+      "step": 795648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.696450320140098e-05,
+      "loss": 4.0318,
+      "step": 796160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.695611725389046e-05,
+      "loss": 4.0289,
+      "step": 796672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.694774768518367e-05,
+      "loss": 4.0158,
+      "step": 797184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.693936173767315e-05,
+      "loss": 4.0187,
+      "step": 797696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6930992168966363e-05,
+      "loss": 4.0108,
+      "step": 798208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6922606221455843e-05,
+      "loss": 4.0211,
+      "step": 798720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.691422027394532e-05,
+      "loss": 4.0309,
+      "step": 799232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.69058343264348e-05,
+      "loss": 4.0267,
+      "step": 799744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.689746475772801e-05,
+      "loss": 4.0268,
+      "step": 800256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.688907881021749e-05,
+      "loss": 4.0188,
+      "step": 800768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.688069286270697e-05,
+      "loss": 4.0414,
+      "step": 801280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.687232329400018e-05,
+      "loss": 4.0204,
+      "step": 801792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.686393734648966e-05,
+      "loss": 4.03,
+      "step": 802304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.685555139897914e-05,
+      "loss": 4.023,
+      "step": 802816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.684716545146862e-05,
+      "loss": 4.0254,
+      "step": 803328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.68387795039581e-05,
+      "loss": 4.028,
+      "step": 803840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.683039355644758e-05,
+      "loss": 4.0306,
+      "step": 804352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.682200760893707e-05,
+      "loss": 4.02,
+      "step": 804864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.681362166142655e-05,
+      "loss": 4.0357,
+      "step": 805376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.680523571391603e-05,
+      "loss": 4.0123,
+      "step": 805888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.679684976640551e-05,
+      "loss": 4.0224,
+      "step": 806400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.678846381889498e-05,
+      "loss": 4.0244,
+      "step": 806912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.678007787138446e-05,
+      "loss": 4.0309,
+      "step": 807424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.677169192387394e-05,
+      "loss": 4.0176,
+      "step": 807936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6763338733970886e-05,
+      "loss": 4.0211,
+      "step": 808448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.6754952786460366e-05,
+      "loss": 4.0155,
+      "step": 808960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6746566838949846e-05,
+      "loss": 4.024,
+      "step": 809472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6738180891439326e-05,
+      "loss": 4.0267,
+      "step": 809984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6729794943928806e-05,
+      "loss": 4.0101,
+      "step": 810496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6721408996418286e-05,
+      "loss": 4.027,
+      "step": 811008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6713023048907766e-05,
+      "loss": 4.0286,
+      "step": 811520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6704637101397246e-05,
+      "loss": 4.0282,
+      "step": 812032
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6696267532690455e-05,
+      "loss": 4.0012,
+      "step": 812544
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6687881585179935e-05,
+      "loss": 4.0197,
+      "step": 813056
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6679495637669415e-05,
+      "loss": 4.0175,
+      "step": 813568
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6671126068962624e-05,
+      "loss": 4.0265,
+      "step": 814080
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6662740121452104e-05,
+      "loss": 4.0273,
+      "step": 814592
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6654354173941584e-05,
+      "loss": 4.0205,
+      "step": 815104
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6645968226431064e-05,
+      "loss": 4.018,
+      "step": 815616
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6637582278920544e-05,
+      "loss": 4.0195,
+      "step": 816128
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6629196331410023e-05,
+      "loss": 4.0108,
+      "step": 816640
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6620810383899503e-05,
+      "loss": 4.0193,
+      "step": 817152
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6612424436388983e-05,
+      "loss": 4.0141,
+      "step": 817664
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.66040548676822e-05,
+      "loss": 4.0239,
+      "step": 818176
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.659566892017168e-05,
+      "loss": 4.0131,
+      "step": 818688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.658728297266116e-05,
+      "loss": 4.0087,
+      "step": 819200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.657889702515064e-05,
+      "loss": 4.0105,
+      "step": 819712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.657052745644385e-05,
+      "loss": 4.0159,
+      "step": 820224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.656214150893333e-05,
+      "loss": 4.0198,
+      "step": 820736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.655377194022654e-05,
+      "loss": 4.0111,
+      "step": 821248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.654538599271602e-05,
+      "loss": 4.0224,
+      "step": 821760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.65370000452055e-05,
+      "loss": 4.023,
+      "step": 822272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.652861409769498e-05,
+      "loss": 4.0146,
+      "step": 822784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.652022815018446e-05,
+      "loss": 4.0225,
+      "step": 823296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.651184220267394e-05,
+      "loss": 4.0174,
+      "step": 823808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.650347263396715e-05,
+      "loss": 4.0199,
+      "step": 824320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.649508668645663e-05,
+      "loss": 4.0165,
+      "step": 824832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.648670073894611e-05,
+      "loss": 4.0128,
+      "step": 825344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.647831479143559e-05,
+      "loss": 4.0205,
+      "step": 825856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.646992884392507e-05,
+      "loss": 4.0122,
+      "step": 826368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.646154289641455e-05,
+      "loss": 4.0235,
+      "step": 826880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.645315694890403e-05,
+      "loss": 4.021,
+      "step": 827392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.644477100139351e-05,
+      "loss": 4.0084,
+      "step": 827904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.643640143268672e-05,
+      "loss": 4.014,
+      "step": 828416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.642803186397993e-05,
+      "loss": 4.0035,
+      "step": 828928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.641964591646941e-05,
+      "loss": 4.0291,
+      "step": 829440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.641125996895889e-05,
+      "loss": 4.0096,
+      "step": 829952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.640287402144838e-05,
+      "loss": 4.0229,
+      "step": 830464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.639448807393786e-05,
+      "loss": 4.0103,
+      "step": 830976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6386118505231067e-05,
+      "loss": 4.0253,
+      "step": 831488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6377748936524276e-05,
+      "loss": 4.0238,
+      "step": 832000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6369362989013756e-05,
+      "loss": 4.0121,
+      "step": 832512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6360977041503236e-05,
+      "loss": 4.0205,
+      "step": 833024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6352591093992715e-05,
+      "loss": 4.0112,
+      "step": 833536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6344205146482195e-05,
+      "loss": 4.0267,
+      "step": 834048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6335819198971675e-05,
+      "loss": 4.0248,
+      "step": 834560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6327433251461155e-05,
+      "loss": 4.009,
+      "step": 835072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6319047303950635e-05,
+      "loss": 4.0157,
+      "step": 835584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6310661356440115e-05,
+      "loss": 4.0163,
+      "step": 836096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6302275408929595e-05,
+      "loss": 4.0247,
+      "step": 836608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6293889461419075e-05,
+      "loss": 4.0106,
+      "step": 837120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6285519892712284e-05,
+      "loss": 4.0218,
+      "step": 837632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6277133945201764e-05,
+      "loss": 4.0128,
+      "step": 838144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6268747997691244e-05,
+      "loss": 4.0249,
+      "step": 838656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.6260362050180724e-05,
+      "loss": 4.016,
+      "step": 839168
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.1087212562561035,
+      "eval_runtime": 287.4834,
+      "eval_samples_per_second": 1327.35,
+      "eval_steps_per_second": 41.481,
+      "step": 839520
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6251976102670204e-05,
+      "loss": 4.0139,
+      "step": 839680
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6243590155159684e-05,
+      "loss": 4.0137,
+      "step": 840192
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6235204207649164e-05,
+      "loss": 4.0183,
+      "step": 840704
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.6226818260138644e-05,
+      "loss": 4.0217,
+      "step": 841216
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.621844869143185e-05,
+      "loss": 4.0251,
+      "step": 841728
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.621006274392133e-05,
+      "loss": 4.0113,
+      "step": 842240
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.620169317521455e-05,
+      "loss": 4.0112,
+      "step": 842752
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.619330722770403e-05,
+      "loss": 4.0139,
+      "step": 843264
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.618492128019351e-05,
+      "loss": 4.0142,
+      "step": 843776
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.617653533268299e-05,
+      "loss": 4.0198,
+      "step": 844288
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.616814938517247e-05,
+      "loss": 4.0106,
+      "step": 844800
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.615977981646568e-05,
+      "loss": 4.0143,
+      "step": 845312
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.615141024775889e-05,
+      "loss": 4.0254,
+      "step": 845824
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.614302430024837e-05,
+      "loss": 4.0068,
+      "step": 846336
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.613463835273785e-05,
+      "loss": 4.009,
+      "step": 846848
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.612625240522733e-05,
+      "loss": 4.0012,
+      "step": 847360
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.611786645771681e-05,
+      "loss": 4.0066,
+      "step": 847872
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.610948051020629e-05,
+      "loss": 4.0106,
+      "step": 848384
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.610109456269577e-05,
+      "loss": 4.0081,
+      "step": 848896
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.609270861518525e-05,
+      "loss": 4.0236,
+      "step": 849408
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.608432266767473e-05,
+      "loss": 4.0236,
+      "step": 849920
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.607595309896794e-05,
+      "loss": 4.0148,
+      "step": 850432
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.606756715145742e-05,
+      "loss": 4.0156,
+      "step": 850944
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.605919758275063e-05,
+      "loss": 4.0187,
+      "step": 851456
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.605081163524011e-05,
+      "loss": 3.999,
+      "step": 851968
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.604242568772959e-05,
+      "loss": 4.0178,
+      "step": 852480
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.603403974021907e-05,
+      "loss": 4.0116,
+      "step": 852992
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.602565379270855e-05,
+      "loss": 3.9969,
+      "step": 853504
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.601726784519803e-05,
+      "loss": 4.0078,
+      "step": 854016
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.600888189768751e-05,
+      "loss": 4.0084,
+      "step": 854528
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.600049595017699e-05,
+      "loss": 4.008,
+      "step": 855040
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.599211000266647e-05,
+      "loss": 4.0103,
+      "step": 855552
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.598374043395969e-05,
+      "loss": 4.0171,
+      "step": 856064
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.597535448644917e-05,
+      "loss": 4.012,
+      "step": 856576
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.596696853893865e-05,
+      "loss": 4.0028,
+      "step": 857088
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5958598970231856e-05,
+      "loss": 4.0186,
+      "step": 857600
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5950213022721336e-05,
+      "loss": 4.0198,
+      "step": 858112
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5941827075210816e-05,
+      "loss": 4.0037,
+      "step": 858624
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5933441127700296e-05,
+      "loss": 4.0079,
+      "step": 859136
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.592505518018977e-05,
+      "loss": 4.0062,
+      "step": 859648
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.591666923267925e-05,
+      "loss": 4.0006,
+      "step": 860160
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.590828328516873e-05,
+      "loss": 4.017,
+      "step": 860672
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.589989733765821e-05,
+      "loss": 4.0134,
+      "step": 861184
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.589151139014769e-05,
+      "loss": 4.0162,
+      "step": 861696
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.588312544263717e-05,
+      "loss": 4.0101,
+      "step": 862208
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5874755873930385e-05,
+      "loss": 4.0143,
+      "step": 862720
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5866369926419865e-05,
+      "loss": 4.0136,
+      "step": 863232
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5857983978909345e-05,
+      "loss": 4.0126,
+      "step": 863744
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5849598031398825e-05,
+      "loss": 3.9852,
+      "step": 864256
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5841212083888305e-05,
+      "loss": 4.0185,
+      "step": 864768
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5832842515181514e-05,
+      "loss": 4.0069,
+      "step": 865280
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5824456567670994e-05,
+      "loss": 4.0214,
+      "step": 865792
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5816070620160474e-05,
+      "loss": 4.0062,
+      "step": 866304
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5807684672649953e-05,
+      "loss": 4.0025,
+      "step": 866816
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.579931510394316e-05,
+      "loss": 3.9993,
+      "step": 867328
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.579092915643264e-05,
+      "loss": 4.0075,
+      "step": 867840
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.578254320892212e-05,
+      "loss": 4.0173,
+      "step": 868352
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.577415726141161e-05,
+      "loss": 4.0058,
+      "step": 868864
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.576577131390109e-05,
+      "loss": 4.0248,
+      "step": 869376
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.575738536639057e-05,
+      "loss": 4.0109,
+      "step": 869888
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.574901579768378e-05,
+      "loss": 3.9907,
+      "step": 870400
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.574062985017326e-05,
+      "loss": 4.0204,
+      "step": 870912
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.573224390266274e-05,
+      "loss": 4.0005,
+      "step": 871424
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.572385795515222e-05,
+      "loss": 4.0004,
+      "step": 871936
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.57154720076417e-05,
+      "loss": 4.0105,
+      "step": 872448
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.570710243893491e-05,
+      "loss": 4.0128,
+      "step": 872960
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.569871649142439e-05,
+      "loss": 3.9935,
+      "step": 873472
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.569033054391387e-05,
+      "loss": 4.0018,
+      "step": 873984
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.568194459640335e-05,
+      "loss": 3.9909,
+      "step": 874496
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.567355864889283e-05,
+      "loss": 4.0058,
+      "step": 875008
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.566517270138231e-05,
+      "loss": 4.0096,
+      "step": 875520
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.565678675387179e-05,
+      "loss": 4.0058,
+      "step": 876032
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.564840080636127e-05,
+      "loss": 4.0116,
+      "step": 876544
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.564001485885075e-05,
+      "loss": 3.9988,
+      "step": 877056
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5631645290143956e-05,
+      "loss": 4.0202,
+      "step": 877568
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5623259342633436e-05,
+      "loss": 4.0071,
+      "step": 878080
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5614873395122916e-05,
+      "loss": 4.0112,
+      "step": 878592
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5606487447612396e-05,
+      "loss": 4.0033,
+      "step": 879104
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5598101500101876e-05,
+      "loss": 4.0071,
+      "step": 879616
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5589715552591356e-05,
+      "loss": 4.0107,
+      "step": 880128
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5581329605080836e-05,
+      "loss": 4.0122,
+      "step": 880640
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5572943657570316e-05,
+      "loss": 4.0043,
+      "step": 881152
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.5564557710059796e-05,
+      "loss": 4.0159,
+      "step": 881664
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.555618814135301e-05,
+      "loss": 3.9945,
+      "step": 882176
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.554780219384249e-05,
+      "loss": 4.0021,
+      "step": 882688
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.553941624633197e-05,
+      "loss": 4.0058,
+      "step": 883200
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.553103029882145e-05,
+      "loss": 4.0145,
+      "step": 883712
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.552266073011466e-05,
+      "loss": 3.9977,
+      "step": 884224
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.551429116140787e-05,
+      "loss": 4.0015,
+      "step": 884736
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.550590521389735e-05,
+      "loss": 4.0011,
+      "step": 885248
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.549751926638683e-05,
+      "loss": 4.0051,
+      "step": 885760
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.548913331887631e-05,
+      "loss": 4.0078,
+      "step": 886272
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.548074737136579e-05,
+      "loss": 3.9936,
+      "step": 886784
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.547236142385527e-05,
+      "loss": 4.0065,
+      "step": 887296
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.546397547634475e-05,
+      "loss": 4.0116,
+      "step": 887808
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.545558952883423e-05,
+      "loss": 4.0112,
+      "step": 888320
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5447236338931174e-05,
+      "loss": 3.9856,
+      "step": 888832
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5438850391420654e-05,
+      "loss": 3.9987,
+      "step": 889344
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5430464443910134e-05,
+      "loss": 3.9945,
+      "step": 889856
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5422078496399614e-05,
+      "loss": 4.0111,
+      "step": 890368
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5413692548889094e-05,
+      "loss": 4.0118,
+      "step": 890880
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5405306601378574e-05,
+      "loss": 4.0008,
+      "step": 891392
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5396920653868054e-05,
+      "loss": 3.9997,
+      "step": 891904
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5388534706357534e-05,
+      "loss": 4.0015,
+      "step": 892416
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.538016513765074e-05,
+      "loss": 3.9939,
+      "step": 892928
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.537177919014022e-05,
+      "loss": 4.0008,
+      "step": 893440
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.536340962143343e-05,
+      "loss": 3.9974,
+      "step": 893952
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.535502367392292e-05,
+      "loss": 4.0045,
+      "step": 894464
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.53466377264124e-05,
+      "loss": 3.9958,
+      "step": 894976
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.533825177890188e-05,
+      "loss": 3.9907,
+      "step": 895488
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.532986583139136e-05,
+      "loss": 3.9945,
+      "step": 896000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.532149626268457e-05,
+      "loss": 3.9959,
+      "step": 896512
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.531311031517405e-05,
+      "loss": 4.003,
+      "step": 897024
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.530472436766353e-05,
+      "loss": 3.9937,
+      "step": 897536
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.529633842015301e-05,
+      "loss": 4.0047,
+      "step": 898048
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.528795247264249e-05,
+      "loss": 3.9999,
+      "step": 898560
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.52795829039357e-05,
+      "loss": 4.0014,
+      "step": 899072
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5271196956425177e-05,
+      "loss": 4.0036,
+      "step": 899584
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5262811008914657e-05,
+      "loss": 3.9967,
+      "step": 900096
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5254425061404136e-05,
+      "loss": 4.0041,
+      "step": 900608
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5246039113893616e-05,
+      "loss": 3.9979,
+      "step": 901120
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5237653166383096e-05,
+      "loss": 3.9944,
+      "step": 901632
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.522928359767631e-05,
+      "loss": 4.0025,
+      "step": 902144
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.522089765016579e-05,
+      "loss": 3.9977,
+      "step": 902656
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5212511702655265e-05,
+      "loss": 4.0041,
+      "step": 903168
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5204125755144745e-05,
+      "loss": 3.9998,
+      "step": 903680
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5195739807634225e-05,
+      "loss": 3.9941,
+      "step": 904192
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5187353860123705e-05,
+      "loss": 3.9939,
+      "step": 904704
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5178967912613185e-05,
+      "loss": 3.988,
+      "step": 905216
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5170581965102665e-05,
+      "loss": 4.006,
+      "step": 905728
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5162196017592145e-05,
+      "loss": 3.997,
+      "step": 906240
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5153826448885354e-05,
+      "loss": 4.0062,
+      "step": 906752
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.514544050137484e-05,
+      "loss": 3.9916,
+      "step": 907264
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.513705455386432e-05,
+      "loss": 4.0047,
+      "step": 907776
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5128701363961266e-05,
+      "loss": 4.0071,
+      "step": 908288
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.512031541645074e-05,
+      "loss": 3.9963,
+      "step": 908800
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.511192946894022e-05,
+      "loss": 3.9988,
+      "step": 909312
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.51035435214297e-05,
+      "loss": 3.9944,
+      "step": 909824
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.509515757391918e-05,
+      "loss": 4.0084,
+      "step": 910336
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.508677162640866e-05,
+      "loss": 4.0087,
+      "step": 910848
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.507838567889814e-05,
+      "loss": 3.9954,
+      "step": 911360
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.506999973138762e-05,
+      "loss": 3.9928,
+      "step": 911872
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.50616137838771e-05,
+      "loss": 4.0015,
+      "step": 912384
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.505322783636658e-05,
+      "loss": 4.0046,
+      "step": 912896
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.504484188885606e-05,
+      "loss": 3.9904,
+      "step": 913408
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5036472320149275e-05,
+      "loss": 4.006,
+      "step": 913920
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5028086372638755e-05,
+      "loss": 3.994,
+      "step": 914432
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5019700425128234e-05,
+      "loss": 4.0062,
+      "step": 914944
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.5011314477617714e-05,
+      "loss": 4.0008,
+      "step": 915456
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.0983662605285645,
+      "eval_runtime": 285.4733,
+      "eval_samples_per_second": 1336.696,
+      "eval_steps_per_second": 41.773,
+      "step": 915840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.5002928530107194e-05,
+      "loss": 4.0071,
+      "step": 915968
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4994542582596674e-05,
+      "loss": 3.9921,
+      "step": 916480
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4986156635086154e-05,
+      "loss": 3.9993,
+      "step": 916992
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.497777068757563e-05,
+      "loss": 4.005,
+      "step": 917504
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.496938474006511e-05,
+      "loss": 4.0057,
+      "step": 918016
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.496099879255459e-05,
+      "loss": 3.9969,
+      "step": 918528
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.495261284504407e-05,
+      "loss": 3.9923,
+      "step": 919040
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.494422689753355e-05,
+      "loss": 3.9966,
+      "step": 919552
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.493584095002303e-05,
+      "loss": 3.9941,
+      "step": 920064
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.492745500251251e-05,
+      "loss": 3.9982,
+      "step": 920576
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.491906905500199e-05,
+      "loss": 3.9971,
+      "step": 921088
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4910683107491474e-05,
+      "loss": 3.9965,
+      "step": 921600
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.490231353878468e-05,
+      "loss": 4.0106,
+      "step": 922112
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.489392759127416e-05,
+      "loss": 3.9875,
+      "step": 922624
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.488554164376364e-05,
+      "loss": 3.9918,
+      "step": 923136
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.487715569625312e-05,
+      "loss": 3.9865,
+      "step": 923648
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.48687697487426e-05,
+      "loss": 3.9872,
+      "step": 924160
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.486038380123208e-05,
+      "loss": 3.9933,
+      "step": 924672
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.485199785372156e-05,
+      "loss": 3.9877,
+      "step": 925184
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.484361190621104e-05,
+      "loss": 4.0033,
+      "step": 925696
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.4835225958700516e-05,
+      "loss": 4.008,
+      "step": 926208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.482685638999373e-05,
+      "loss": 3.996,
+      "step": 926720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.481847044248321e-05,
+      "loss": 3.9973,
+      "step": 927232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.481008449497269e-05,
+      "loss": 4.0029,
+      "step": 927744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.480169854746217e-05,
+      "loss": 3.9806,
+      "step": 928256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.479331259995165e-05,
+      "loss": 4.0035,
+      "step": 928768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.478494303124487e-05,
+      "loss": 3.99,
+      "step": 929280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.477655708373434e-05,
+      "loss": 3.9844,
+      "step": 929792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.476817113622382e-05,
+      "loss": 3.9913,
+      "step": 930304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.47597851887133e-05,
+      "loss": 3.9867,
+      "step": 930816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.475139924120278e-05,
+      "loss": 3.9922,
+      "step": 931328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.474301329369226e-05,
+      "loss": 3.9884,
+      "step": 931840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.473462734618174e-05,
+      "loss": 4.0025,
+      "step": 932352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.472625777747495e-05,
+      "loss": 3.9909,
+      "step": 932864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.471787182996443e-05,
+      "loss": 3.988,
+      "step": 933376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4709502261257645e-05,
+      "loss": 3.9987,
+      "step": 933888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4701116313747125e-05,
+      "loss": 4.0041,
+      "step": 934400
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4692730366236605e-05,
+      "loss": 3.9878,
+      "step": 934912
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4684344418726085e-05,
+      "loss": 3.988,
+      "step": 935424
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4675958471215565e-05,
+      "loss": 3.9891,
+      "step": 935936
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4667572523705045e-05,
+      "loss": 3.9862,
+      "step": 936448
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4659186576194525e-05,
+      "loss": 3.9964,
+      "step": 936960
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4650800628684005e-05,
+      "loss": 3.995,
+      "step": 937472
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4642414681173485e-05,
+      "loss": 4.0028,
+      "step": 937984
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4634028733662965e-05,
+      "loss": 3.9934,
+      "step": 938496
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4625642786152445e-05,
+      "loss": 3.9946,
+      "step": 939008
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4617273217445654e-05,
+      "loss": 3.9946,
+      "step": 939520
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4608887269935134e-05,
+      "loss": 3.9956,
+      "step": 940032
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.460051770122835e-05,
+      "loss": 3.9688,
+      "step": 940544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.459213175371783e-05,
+      "loss": 3.9986,
+      "step": 941056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.458374580620731e-05,
+      "loss": 3.99,
+      "step": 941568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.457535985869679e-05,
+      "loss": 4.0027,
+      "step": 942080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.456697391118627e-05,
+      "loss": 3.9868,
+      "step": 942592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.455860434247948e-05,
+      "loss": 3.9908,
+      "step": 943104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.455021839496896e-05,
+      "loss": 3.9793,
+      "step": 943616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.454183244745844e-05,
+      "loss": 3.9871,
+      "step": 944128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.453344649994792e-05,
+      "loss": 4.0046,
+      "step": 944640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.45250605524374e-05,
+      "loss": 3.9893,
+      "step": 945152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.451667460492687e-05,
+      "loss": 4.0062,
+      "step": 945664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.450830503622009e-05,
+      "loss": 3.994,
+      "step": 946176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.449991908870957e-05,
+      "loss": 3.9697,
+      "step": 946688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.449153314119905e-05,
+      "loss": 4.0078,
+      "step": 947200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.448314719368853e-05,
+      "loss": 3.9813,
+      "step": 947712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.447476124617801e-05,
+      "loss": 3.9806,
+      "step": 948224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.446639167747122e-05,
+      "loss": 3.9961,
+      "step": 948736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.44580057299607e-05,
+      "loss": 3.9929,
+      "step": 949248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4449619782450176e-05,
+      "loss": 3.9831,
+      "step": 949760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4441233834939656e-05,
+      "loss": 3.9835,
+      "step": 950272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4432847887429136e-05,
+      "loss": 3.9773,
+      "step": 950784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4424461939918616e-05,
+      "loss": 3.982,
+      "step": 951296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4416075992408096e-05,
+      "loss": 3.9972,
+      "step": 951808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4407690044897576e-05,
+      "loss": 3.9902,
+      "step": 952320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4399320476190785e-05,
+      "loss": 3.9905,
+      "step": 952832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.439093452868027e-05,
+      "loss": 3.9821,
+      "step": 953344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.438254858116975e-05,
+      "loss": 4.0021,
+      "step": 953856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.437417901246296e-05,
+      "loss": 3.9903,
+      "step": 954368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.436580944375618e-05,
+      "loss": 3.9957,
+      "step": 954880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.435742349624565e-05,
+      "loss": 3.9856,
+      "step": 955392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.434903754873513e-05,
+      "loss": 3.9906,
+      "step": 955904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.434065160122461e-05,
+      "loss": 3.9944,
+      "step": 956416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.433226565371409e-05,
+      "loss": 3.9918,
+      "step": 956928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.43238960850073e-05,
+      "loss": 3.9896,
+      "step": 957440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.431551013749678e-05,
+      "loss": 3.9986,
+      "step": 957952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.430712418998626e-05,
+      "loss": 3.9817,
+      "step": 958464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.429873824247574e-05,
+      "loss": 3.9836,
+      "step": 958976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4290368673768955e-05,
+      "loss": 3.9877,
+      "step": 959488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.428199910506217e-05,
+      "loss": 3.9948,
+      "step": 960000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.427361315755165e-05,
+      "loss": 3.9834,
+      "step": 960512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4265227210041124e-05,
+      "loss": 3.9891,
+      "step": 961024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.4256841262530604e-05,
+      "loss": 3.9833,
+      "step": 961536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4248455315020084e-05,
+      "loss": 3.9868,
+      "step": 962048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4240069367509564e-05,
+      "loss": 3.9895,
+      "step": 962560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4231683419999044e-05,
+      "loss": 3.9799,
+      "step": 963072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4223297472488524e-05,
+      "loss": 3.9895,
+      "step": 963584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.421492790378173e-05,
+      "loss": 3.9952,
+      "step": 964096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.420654195627121e-05,
+      "loss": 3.9979,
+      "step": 964608
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.419815600876069e-05,
+      "loss": 3.965,
+      "step": 965120
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.418977006125017e-05,
+      "loss": 3.9848,
+      "step": 965632
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.418138411373966e-05,
+      "loss": 3.978,
+      "step": 966144
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.417299816622914e-05,
+      "loss": 3.9938,
+      "step": 966656
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.416461221871862e-05,
+      "loss": 3.9976,
+      "step": 967168
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.41562262712081e-05,
+      "loss": 3.9837,
+      "step": 967680
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.414784032369758e-05,
+      "loss": 3.9851,
+      "step": 968192
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.413947075499079e-05,
+      "loss": 3.9847,
+      "step": 968704
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.413108480748027e-05,
+      "loss": 3.9801,
+      "step": 969216
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.412269885996975e-05,
+      "loss": 3.9818,
+      "step": 969728
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.411431291245923e-05,
+      "loss": 3.9791,
+      "step": 970240
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.410594334375244e-05,
+      "loss": 3.9884,
+      "step": 970752
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.409755739624192e-05,
+      "loss": 3.9799,
+      "step": 971264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.40891714487314e-05,
+      "loss": 3.9753,
+      "step": 971776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.408078550122088e-05,
+      "loss": 3.9799,
+      "step": 972288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.407241593251409e-05,
+      "loss": 3.9757,
+      "step": 972800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.406402998500357e-05,
+      "loss": 3.9887,
+      "step": 973312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.405564403749305e-05,
+      "loss": 3.9735,
+      "step": 973824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.404725808998253e-05,
+      "loss": 3.9887,
+      "step": 974336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4038872142472006e-05,
+      "loss": 3.9869,
+      "step": 974848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.403050257376522e-05,
+      "loss": 3.9856,
+      "step": 975360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.402213300505843e-05,
+      "loss": 3.9848,
+      "step": 975872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.401374705754791e-05,
+      "loss": 3.9831,
+      "step": 976384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.400536111003739e-05,
+      "loss": 3.9903,
+      "step": 976896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.399697516252687e-05,
+      "loss": 3.9813,
+      "step": 977408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.398860559382008e-05,
+      "loss": 3.9774,
+      "step": 977920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.398021964630957e-05,
+      "loss": 3.9864,
+      "step": 978432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.397183369879905e-05,
+      "loss": 3.9827,
+      "step": 978944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3963447751288527e-05,
+      "loss": 3.9875,
+      "step": 979456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3955061803778007e-05,
+      "loss": 3.9814,
+      "step": 979968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3946692235071216e-05,
+      "loss": 3.9779,
+      "step": 980480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3938306287560696e-05,
+      "loss": 3.9807,
+      "step": 980992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3929920340050176e-05,
+      "loss": 3.9758,
+      "step": 981504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3921534392539655e-05,
+      "loss": 3.9848,
+      "step": 982016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.391314844502913e-05,
+      "loss": 3.9838,
+      "step": 982528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.390476249751861e-05,
+      "loss": 3.9937,
+      "step": 983040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.389637655000809e-05,
+      "loss": 3.9767,
+      "step": 983552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.388799060249757e-05,
+      "loss": 3.9849,
+      "step": 984064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3879621033790784e-05,
+      "loss": 3.9915,
+      "step": 984576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3871235086280264e-05,
+      "loss": 3.9835,
+      "step": 985088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.386286551757348e-05,
+      "loss": 3.978,
+      "step": 985600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3854479570062953e-05,
+      "loss": 3.9815,
+      "step": 986112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.384609362255243e-05,
+      "loss": 3.9928,
+      "step": 986624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.383770767504191e-05,
+      "loss": 3.9906,
+      "step": 987136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.382933810633513e-05,
+      "loss": 3.9809,
+      "step": 987648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.38209521588246e-05,
+      "loss": 3.9758,
+      "step": 988160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.381256621131408e-05,
+      "loss": 3.9867,
+      "step": 988672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.380418026380356e-05,
+      "loss": 3.9891,
+      "step": 989184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.379579431629304e-05,
+      "loss": 3.9728,
+      "step": 989696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.378740836878252e-05,
+      "loss": 3.9976,
+      "step": 990208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.3779022421272e-05,
+      "loss": 3.9752,
+      "step": 990720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.377063647376149e-05,
+      "loss": 3.9903,
+      "step": 991232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.37622669050547e-05,
+      "loss": 3.9863,
+      "step": 991744
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.088757514953613,
+      "eval_runtime": 302.492,
+      "eval_samples_per_second": 1261.491,
+      "eval_steps_per_second": 39.423,
+      "step": 992160
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.375388095754418e-05,
+      "loss": 3.985,
+      "step": 992256
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.374549501003366e-05,
+      "loss": 3.9786,
+      "step": 992768
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.373710906252314e-05,
+      "loss": 3.9808,
+      "step": 993280
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.372873949381635e-05,
+      "loss": 3.9928,
+      "step": 993792
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.372035354630583e-05,
+      "loss": 3.9849,
+      "step": 994304
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.371196759879531e-05,
+      "loss": 3.9854,
+      "step": 994816
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.370358165128479e-05,
+      "loss": 3.9751,
+      "step": 995328
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.369519570377427e-05,
+      "loss": 3.9777,
+      "step": 995840
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.368680975626375e-05,
+      "loss": 3.9794,
+      "step": 996352
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.367842380875323e-05,
+      "loss": 3.9873,
+      "step": 996864
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.367003786124271e-05,
+      "loss": 3.9789,
+      "step": 997376
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3661651913732187e-05,
+      "loss": 3.9877,
+      "step": 997888
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.36532823450254e-05,
+      "loss": 3.9915,
+      "step": 998400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.364489639751488e-05,
+      "loss": 3.9708,
+      "step": 998912
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.363651045000436e-05,
+      "loss": 3.9809,
+      "step": 999424
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.362812450249384e-05,
+      "loss": 3.9667,
+      "step": 999936
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3619738554983316e-05,
+      "loss": 3.9693,
+      "step": 1000448
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3611352607472795e-05,
+      "loss": 3.9839,
+      "step": 1000960
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3602966659962275e-05,
+      "loss": 3.9706,
+      "step": 1001472
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3594580712451755e-05,
+      "loss": 3.9869,
+      "step": 1001984
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3586194764941235e-05,
+      "loss": 3.9937,
+      "step": 1002496
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3577825196234444e-05,
+      "loss": 3.9803,
+      "step": 1003008
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3569439248723924e-05,
+      "loss": 3.9855,
+      "step": 1003520
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3561053301213404e-05,
+      "loss": 3.9866,
+      "step": 1004032
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.355266735370289e-05,
+      "loss": 3.9662,
+      "step": 1004544
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.35442977849961e-05,
+      "loss": 3.9845,
+      "step": 1005056
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.353591183748558e-05,
+      "loss": 3.9786,
+      "step": 1005568
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.352752588997506e-05,
+      "loss": 3.9707,
+      "step": 1006080
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.351913994246454e-05,
+      "loss": 3.9784,
+      "step": 1006592
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.351075399495402e-05,
+      "loss": 3.9685,
+      "step": 1007104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3502400805050965e-05,
+      "loss": 3.9747,
+      "step": 1007616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.349401485754044e-05,
+      "loss": 3.9783,
+      "step": 1008128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.348562891002992e-05,
+      "loss": 3.9853,
+      "step": 1008640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.34772429625194e-05,
+      "loss": 3.9774,
+      "step": 1009152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.346885701500888e-05,
+      "loss": 3.9733,
+      "step": 1009664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3460487446302094e-05,
+      "loss": 3.9797,
+      "step": 1010176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3452101498791574e-05,
+      "loss": 3.9904,
+      "step": 1010688
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3443715551281054e-05,
+      "loss": 3.9745,
+      "step": 1011200
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3435329603770534e-05,
+      "loss": 3.9752,
+      "step": 1011712
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3426943656260014e-05,
+      "loss": 3.973,
+      "step": 1012224
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3418557708749494e-05,
+      "loss": 3.9748,
+      "step": 1012736
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.34101881400427e-05,
+      "loss": 3.9764,
+      "step": 1013248
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.340180219253218e-05,
+      "loss": 3.9833,
+      "step": 1013760
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.339341624502166e-05,
+      "loss": 3.9837,
+      "step": 1014272
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.338503029751114e-05,
+      "loss": 3.9789,
+      "step": 1014784
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.337664435000062e-05,
+      "loss": 3.9791,
+      "step": 1015296
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.33682584024901e-05,
+      "loss": 3.9775,
+      "step": 1015808
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.335987245497958e-05,
+      "loss": 3.9808,
+      "step": 1016320
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.335148650746906e-05,
+      "loss": 3.9622,
+      "step": 1016832
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.334311693876228e-05,
+      "loss": 3.9795,
+      "step": 1017344
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.333474737005549e-05,
+      "loss": 3.9767,
+      "step": 1017856
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.332636142254497e-05,
+      "loss": 3.9863,
+      "step": 1018368
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.331797547503445e-05,
+      "loss": 3.9707,
+      "step": 1018880
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.330958952752393e-05,
+      "loss": 3.9734,
+      "step": 1019392
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.330120358001341e-05,
+      "loss": 3.9714,
+      "step": 1019904
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.329281763250289e-05,
+      "loss": 3.9685,
+      "step": 1020416
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3284448063796096e-05,
+      "loss": 3.9904,
+      "step": 1020928
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3276062116285576e-05,
+      "loss": 3.9714,
+      "step": 1021440
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3267676168775056e-05,
+      "loss": 3.9971,
+      "step": 1021952
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3259290221264536e-05,
+      "loss": 3.9754,
+      "step": 1022464
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.325092065255775e-05,
+      "loss": 3.9596,
+      "step": 1022976
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.324253470504723e-05,
+      "loss": 3.9944,
+      "step": 1023488
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.323414875753671e-05,
+      "loss": 3.9684,
+      "step": 1024000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.322576281002619e-05,
+      "loss": 3.964,
+      "step": 1024512
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.32173932413194e-05,
+      "loss": 3.9748,
+      "step": 1025024
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.320900729380888e-05,
+      "loss": 3.984,
+      "step": 1025536
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.320062134629836e-05,
+      "loss": 3.9654,
+      "step": 1026048
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.319223539878784e-05,
+      "loss": 3.9714,
+      "step": 1026560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.318384945127732e-05,
+      "loss": 3.9635,
+      "step": 1027072
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.317549626137426e-05,
+      "loss": 3.9654,
+      "step": 1027584
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.316711031386374e-05,
+      "loss": 3.986,
+      "step": 1028096
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.315872436635322e-05,
+      "loss": 3.978,
+      "step": 1028608
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3150338418842706e-05,
+      "loss": 3.9768,
+      "step": 1029120
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3141952471332186e-05,
+      "loss": 3.9682,
+      "step": 1029632
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3133566523821666e-05,
+      "loss": 3.9863,
+      "step": 1030144
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3125180576311146e-05,
+      "loss": 3.9779,
+      "step": 1030656
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3116794628800626e-05,
+      "loss": 3.983,
+      "step": 1031168
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.31084086812901e-05,
+      "loss": 3.9709,
+      "step": 1031680
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3100039112583315e-05,
+      "loss": 3.9747,
+      "step": 1032192
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3091653165072795e-05,
+      "loss": 3.9821,
+      "step": 1032704
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3083267217562275e-05,
+      "loss": 3.9754,
+      "step": 1033216
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.307488127005175e-05,
+      "loss": 3.9753,
+      "step": 1033728
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.306649532254123e-05,
+      "loss": 3.9848,
+      "step": 1034240
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.305810937503071e-05,
+      "loss": 3.9741,
+      "step": 1034752
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3049739806323923e-05,
+      "loss": 3.9658,
+      "step": 1035264
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3041353858813403e-05,
+      "loss": 3.9743,
+      "step": 1035776
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.3032967911302883e-05,
+      "loss": 3.981,
+      "step": 1036288
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.302458196379236e-05,
+      "loss": 3.9723,
+      "step": 1036800
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.301619601628184e-05,
+      "loss": 3.9729,
+      "step": 1037312
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.300782644757505e-05,
+      "loss": 3.967,
+      "step": 1037824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.299944050006453e-05,
+      "loss": 3.972,
+      "step": 1038336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.299107093135775e-05,
+      "loss": 3.9738,
+      "step": 1038848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.298268498384722e-05,
+      "loss": 3.9694,
+      "step": 1039360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.29742990363367e-05,
+      "loss": 3.9764,
+      "step": 1039872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.296591308882618e-05,
+      "loss": 3.9795,
+      "step": 1040384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.295752714131566e-05,
+      "loss": 3.9848,
+      "step": 1040896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.294914119380514e-05,
+      "loss": 3.951,
+      "step": 1041408
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.294075524629462e-05,
+      "loss": 3.9688,
+      "step": 1041920
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.293236929878411e-05,
+      "loss": 3.967,
+      "step": 1042432
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.292398335127359e-05,
+      "loss": 3.9773,
+      "step": 1042944
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.29156137825668e-05,
+      "loss": 3.9831,
+      "step": 1043456
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.290722783505628e-05,
+      "loss": 3.9702,
+      "step": 1043968
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.289884188754576e-05,
+      "loss": 3.9697,
+      "step": 1044480
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.289045594003524e-05,
+      "loss": 3.9703,
+      "step": 1044992
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.288206999252472e-05,
+      "loss": 3.97,
+      "step": 1045504
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.28736840450142e-05,
+      "loss": 3.9666,
+      "step": 1046016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2865314476307406e-05,
+      "loss": 3.9671,
+      "step": 1046528
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2856928528796886e-05,
+      "loss": 3.9735,
+      "step": 1047040
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2848542581286366e-05,
+      "loss": 3.9644,
+      "step": 1047552
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2840156633775846e-05,
+      "loss": 3.9638,
+      "step": 1048064
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2831770686265326e-05,
+      "loss": 3.9685,
+      "step": 1048576
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2823384738754806e-05,
+      "loss": 3.9572,
+      "step": 1049088
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.281501517004802e-05,
+      "loss": 3.9735,
+      "step": 1049600
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.28066292225375e-05,
+      "loss": 3.9667,
+      "step": 1050112
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.279824327502698e-05,
+      "loss": 3.9695,
+      "step": 1050624
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.278985732751646e-05,
+      "loss": 3.9748,
+      "step": 1051136
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2781471380005935e-05,
+      "loss": 3.9741,
+      "step": 1051648
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.277310181129915e-05,
+      "loss": 3.9696,
+      "step": 1052160
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.276471586378863e-05,
+      "loss": 3.9661,
+      "step": 1052672
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2756329916278104e-05,
+      "loss": 3.9784,
+      "step": 1053184
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.274796034757132e-05,
+      "loss": 3.9683,
+      "step": 1053696
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.27395744000608e-05,
+      "loss": 3.9625,
+      "step": 1054208
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.273118845255028e-05,
+      "loss": 3.9738,
+      "step": 1054720
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.272280250503976e-05,
+      "loss": 3.97,
+      "step": 1055232
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.271441655752924e-05,
+      "loss": 3.9696,
+      "step": 1055744
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.270603061001872e-05,
+      "loss": 3.9671,
+      "step": 1056256
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.26976446625082e-05,
+      "loss": 3.9706,
+      "step": 1056768
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.268925871499768e-05,
+      "loss": 3.9679,
+      "step": 1057280
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.268087276748716e-05,
+      "loss": 3.9641,
+      "step": 1057792
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.267248681997664e-05,
+      "loss": 3.9615,
+      "step": 1058304
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.266410087246612e-05,
+      "loss": 3.9756,
+      "step": 1058816
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.26557149249556e-05,
+      "loss": 3.9808,
+      "step": 1059328
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.264734535624881e-05,
+      "loss": 3.9607,
+      "step": 1059840
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.263897578754202e-05,
+      "loss": 3.9685,
+      "step": 1060352
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.26305898400315e-05,
+      "loss": 3.9829,
+      "step": 1060864
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2622203892520984e-05,
+      "loss": 3.9725,
+      "step": 1061376
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.2613817945010464e-05,
+      "loss": 3.9638,
+      "step": 1061888
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.260544837630367e-05,
+      "loss": 3.9674,
+      "step": 1062400
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.259706242879315e-05,
+      "loss": 3.9835,
+      "step": 1062912
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.258867648128263e-05,
+      "loss": 3.9791,
+      "step": 1063424
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.258029053377211e-05,
+      "loss": 3.9672,
+      "step": 1063936
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.257190458626159e-05,
+      "loss": 3.9608,
+      "step": 1064448
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.25635350175548e-05,
+      "loss": 3.9741,
+      "step": 1064960
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.255514907004428e-05,
+      "loss": 3.9746,
+      "step": 1065472
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.254676312253376e-05,
+      "loss": 3.9599,
+      "step": 1065984
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.253837717502324e-05,
+      "loss": 3.9823,
+      "step": 1066496
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.252999122751272e-05,
+      "loss": 3.966,
+      "step": 1067008
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.25216052800022e-05,
+      "loss": 3.9755,
+      "step": 1067520
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.251321933249168e-05,
+      "loss": 3.9719,
+      "step": 1068032
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.080809593200684,
+      "eval_runtime": 304.1609,
+      "eval_samples_per_second": 1254.57,
+      "eval_steps_per_second": 39.206,
+      "step": 1068480
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.250483338498116e-05,
+      "loss": 3.9699,
+      "step": 1068544
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.249644743747065e-05,
+      "loss": 3.9629,
+      "step": 1069056
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.248806148996012e-05,
+      "loss": 3.9694,
+      "step": 1069568
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.24796755424496e-05,
+      "loss": 3.9785,
+      "step": 1070080
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.247128959493908e-05,
+      "loss": 3.9731,
+      "step": 1070592
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.24629200262323e-05,
+      "loss": 3.9705,
+      "step": 1071104
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.245453407872177e-05,
+      "loss": 3.966,
+      "step": 1071616
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.244614813121125e-05,
+      "loss": 3.9641,
+      "step": 1072128
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2437778562504466e-05,
+      "loss": 3.9647,
+      "step": 1072640
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.242939261499394e-05,
+      "loss": 3.9736,
+      "step": 1073152
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.242100666748342e-05,
+      "loss": 3.9661,
+      "step": 1073664
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.24126207199729e-05,
+      "loss": 3.9766,
+      "step": 1074176
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2404251151266115e-05,
+      "loss": 3.9788,
+      "step": 1074688
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2395865203755595e-05,
+      "loss": 3.957,
+      "step": 1075200
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2387479256245075e-05,
+      "loss": 3.9674,
+      "step": 1075712
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2379093308734555e-05,
+      "loss": 3.9536,
+      "step": 1076224
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2370707361224035e-05,
+      "loss": 3.9588,
+      "step": 1076736
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2362321413713515e-05,
+      "loss": 3.9675,
+      "step": 1077248
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2353935466202995e-05,
+      "loss": 3.9565,
+      "step": 1077760
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2345549518692475e-05,
+      "loss": 3.9701,
+      "step": 1078272
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.2337163571181955e-05,
+      "loss": 3.9836,
+      "step": 1078784
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.232881038127889e-05,
+      "loss": 3.9641,
+      "step": 1079296
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.232042443376837e-05,
+      "loss": 3.9722,
+      "step": 1079808
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.231203848625785e-05,
+      "loss": 3.9745,
+      "step": 1080320
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.230365253874734e-05,
+      "loss": 3.9553,
+      "step": 1080832
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.229526659123682e-05,
+      "loss": 3.9731,
+      "step": 1081344
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.22868806437263e-05,
+      "loss": 3.9597,
+      "step": 1081856
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.227849469621578e-05,
+      "loss": 3.9614,
+      "step": 1082368
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.227010874870526e-05,
+      "loss": 3.9641,
+      "step": 1082880
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.226172280119474e-05,
+      "loss": 3.9548,
+      "step": 1083392
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.225335323248795e-05,
+      "loss": 3.9594,
+      "step": 1083904
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.224498366378116e-05,
+      "loss": 3.9678,
+      "step": 1084416
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.223661409507437e-05,
+      "loss": 3.9723,
+      "step": 1084928
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.222822814756385e-05,
+      "loss": 3.9644,
+      "step": 1085440
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.221984220005333e-05,
+      "loss": 3.9578,
+      "step": 1085952
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.221145625254281e-05,
+      "loss": 3.9642,
+      "step": 1086464
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.220307030503229e-05,
+      "loss": 3.9816,
+      "step": 1086976
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.219468435752177e-05,
+      "loss": 3.9578,
+      "step": 1087488
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.218629841001125e-05,
+      "loss": 3.9644,
+      "step": 1088000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.217791246250073e-05,
+      "loss": 3.9622,
+      "step": 1088512
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.216952651499021e-05,
+      "loss": 3.9572,
+      "step": 1089024
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.216114056747969e-05,
+      "loss": 3.9668,
+      "step": 1089536
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.215275461996917e-05,
+      "loss": 3.9694,
+      "step": 1090048
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.214436867245865e-05,
+      "loss": 3.974,
+      "step": 1090560
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.213599910375186e-05,
+      "loss": 3.9671,
+      "step": 1091072
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.212761315624134e-05,
+      "loss": 3.9644,
+      "step": 1091584
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.211922720873082e-05,
+      "loss": 3.9614,
+      "step": 1092096
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.21108412612203e-05,
+      "loss": 3.9743,
+      "step": 1092608
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.2102455313709775e-05,
+      "loss": 3.9438,
+      "step": 1093120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.209408574500299e-05,
+      "loss": 3.9645,
+      "step": 1093632
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.208569979749248e-05,
+      "loss": 3.9688,
+      "step": 1094144
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.207731384998196e-05,
+      "loss": 3.972,
+      "step": 1094656
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.206892790247143e-05,
+      "loss": 3.9607,
+      "step": 1095168
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.206054195496091e-05,
+      "loss": 3.9567,
+      "step": 1095680
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.205215600745039e-05,
+      "loss": 3.9632,
+      "step": 1096192
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.204377005993987e-05,
+      "loss": 3.9527,
+      "step": 1096704
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.203538411242935e-05,
+      "loss": 3.9758,
+      "step": 1097216
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.202699816491883e-05,
+      "loss": 3.9616,
+      "step": 1097728
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.201862859621204e-05,
+      "loss": 3.9779,
+      "step": 1098240
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.201025902750525e-05,
+      "loss": 3.9636,
+      "step": 1098752
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.200187307999473e-05,
+      "loss": 3.9469,
+      "step": 1099264
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1993487132484216e-05,
+      "loss": 3.9771,
+      "step": 1099776
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1985101184973696e-05,
+      "loss": 3.9548,
+      "step": 1100288
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1976715237463175e-05,
+      "loss": 3.9522,
+      "step": 1100800
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1968329289952655e-05,
+      "loss": 3.9638,
+      "step": 1101312
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1959943342442135e-05,
+      "loss": 3.9697,
+      "step": 1101824
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1951573773735344e-05,
+      "loss": 3.9516,
+      "step": 1102336
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1943187826224824e-05,
+      "loss": 3.9559,
+      "step": 1102848
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1934801878714304e-05,
+      "loss": 3.9537,
+      "step": 1103360
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1926415931203784e-05,
+      "loss": 3.9532,
+      "step": 1103872
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1918029983693264e-05,
+      "loss": 3.9723,
+      "step": 1104384
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1909644036182744e-05,
+      "loss": 3.9646,
+      "step": 1104896
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1901258088672224e-05,
+      "loss": 3.9602,
+      "step": 1105408
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.189288851996543e-05,
+      "loss": 3.9589,
+      "step": 1105920
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.188450257245491e-05,
+      "loss": 3.9712,
+      "step": 1106432
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.187611662494439e-05,
+      "loss": 3.9644,
+      "step": 1106944
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.186774705623761e-05,
+      "loss": 3.9665,
+      "step": 1107456
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.185936110872709e-05,
+      "loss": 3.9607,
+      "step": 1107968
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.185097516121657e-05,
+      "loss": 3.9638,
+      "step": 1108480
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.184258921370605e-05,
+      "loss": 3.9658,
+      "step": 1108992
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.183421964499926e-05,
+      "loss": 3.9605,
+      "step": 1109504
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.182583369748874e-05,
+      "loss": 3.9616,
+      "step": 1110016
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.181744774997822e-05,
+      "loss": 3.9708,
+      "step": 1110528
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.18090618024677e-05,
+      "loss": 3.9667,
+      "step": 1111040
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.180067585495718e-05,
+      "loss": 3.9489,
+      "step": 1111552
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.179230628625039e-05,
+      "loss": 3.9601,
+      "step": 1112064
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.178392033873987e-05,
+      "loss": 3.9699,
+      "step": 1112576
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.177553439122935e-05,
+      "loss": 3.9576,
+      "step": 1113088
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1767148443718834e-05,
+      "loss": 3.9603,
+      "step": 1113600
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 3.1758762496208314e-05,
+      "loss": 3.9553,
+      "step": 1114112
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1750376548697794e-05,
+      "loss": 3.9602,
+      "step": 1114624
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1742006979991e-05,
+      "loss": 3.9592,
+      "step": 1115136
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.173362103248048e-05,
+      "loss": 3.9594,
+      "step": 1115648
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.172523508496996e-05,
+      "loss": 3.9623,
+      "step": 1116160
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1716849137459436e-05,
+      "loss": 3.96,
+      "step": 1116672
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1708463189948916e-05,
+      "loss": 3.974,
+      "step": 1117184
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1700077242438396e-05,
+      "loss": 3.9375,
+      "step": 1117696
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1691691294927876e-05,
+      "loss": 3.9591,
+      "step": 1118208
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1683321726221085e-05,
+      "loss": 3.9529,
+      "step": 1118720
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.167493577871057e-05,
+      "loss": 3.9644,
+      "step": 1119232
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.166654983120005e-05,
+      "loss": 3.9686,
+      "step": 1119744
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.165816388368953e-05,
+      "loss": 3.9582,
+      "step": 1120256
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.164977793617901e-05,
+      "loss": 3.9549,
+      "step": 1120768
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.164139198866849e-05,
+      "loss": 3.9571,
+      "step": 1121280
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.163300604115797e-05,
+      "loss": 3.9611,
+      "step": 1121792
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.162463647245118e-05,
+      "loss": 3.9517,
+      "step": 1122304
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.161625052494066e-05,
+      "loss": 3.9539,
+      "step": 1122816
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.160786457743014e-05,
+      "loss": 3.9578,
+      "step": 1123328
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.159947862991962e-05,
+      "loss": 3.9592,
+      "step": 1123840
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.15910926824091e-05,
+      "loss": 3.9479,
+      "step": 1124352
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.158272311370231e-05,
+      "loss": 3.957,
+      "step": 1124864
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.157433716619179e-05,
+      "loss": 3.9448,
+      "step": 1125376
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.156595121868127e-05,
+      "loss": 3.9627,
+      "step": 1125888
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1557565271170756e-05,
+      "loss": 3.9501,
+      "step": 1126400
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1549195702463965e-05,
+      "loss": 3.9549,
+      "step": 1126912
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1540809754953445e-05,
+      "loss": 3.9609,
+      "step": 1127424
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1532423807442925e-05,
+      "loss": 3.963,
+      "step": 1127936
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1524054238736134e-05,
+      "loss": 3.9572,
+      "step": 1128448
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1515668291225614e-05,
+      "loss": 3.9561,
+      "step": 1128960
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1507282343715094e-05,
+      "loss": 3.9632,
+      "step": 1129472
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1498896396204574e-05,
+      "loss": 3.9522,
+      "step": 1129984
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1490510448694054e-05,
+      "loss": 3.9499,
+      "step": 1130496
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1482124501183534e-05,
+      "loss": 3.9615,
+      "step": 1131008
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1473738553673014e-05,
+      "loss": 3.9607,
+      "step": 1131520
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1465352606162494e-05,
+      "loss": 3.9602,
+      "step": 1132032
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.145698303745571e-05,
+      "loss": 3.9544,
+      "step": 1132544
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.144859708994519e-05,
+      "loss": 3.9568,
+      "step": 1133056
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.144021114243467e-05,
+      "loss": 3.952,
+      "step": 1133568
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.143182519492415e-05,
+      "loss": 3.9497,
+      "step": 1134080
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.142343924741362e-05,
+      "loss": 3.9472,
+      "step": 1134592
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.14150532999031e-05,
+      "loss": 3.9671,
+      "step": 1135104
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.140666735239258e-05,
+      "loss": 3.9664,
+      "step": 1135616
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.139828140488206e-05,
+      "loss": 3.9508,
+      "step": 1136128
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.138991183617527e-05,
+      "loss": 3.954,
+      "step": 1136640
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.138152588866475e-05,
+      "loss": 3.9642,
+      "step": 1137152
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.137313994115423e-05,
+      "loss": 3.9594,
+      "step": 1137664
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.136475399364371e-05,
+      "loss": 3.9556,
+      "step": 1138176
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.135638442493693e-05,
+      "loss": 3.9566,
+      "step": 1138688
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.134799847742641e-05,
+      "loss": 3.9695,
+      "step": 1139200
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.133961252991589e-05,
+      "loss": 3.9634,
+      "step": 1139712
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.133125934001283e-05,
+      "loss": 3.9552,
+      "step": 1140224
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.132287339250231e-05,
+      "loss": 3.9485,
+      "step": 1140736
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.131448744499179e-05,
+      "loss": 3.9605,
+      "step": 1141248
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.130610149748127e-05,
+      "loss": 3.9603,
+      "step": 1141760
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1297715549970745e-05,
+      "loss": 3.9481,
+      "step": 1142272
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1289329602460225e-05,
+      "loss": 3.9713,
+      "step": 1142784
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1280943654949705e-05,
+      "loss": 3.9532,
+      "step": 1143296
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1272557707439185e-05,
+      "loss": 3.9632,
+      "step": 1143808
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 3.1264171759928665e-05,
+      "loss": 3.9568,
+      "step": 1144320
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.074085235595703,
+      "eval_runtime": 302.3145,
+      "eval_samples_per_second": 1262.232,
+      "eval_steps_per_second": 39.446,
+      "step": 1144800
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.125580219122188e-05,
+      "loss": 3.9572,
+      "step": 1144832
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.124741624371136e-05,
+      "loss": 3.9544,
+      "step": 1145344
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.123903029620084e-05,
+      "loss": 3.957,
+      "step": 1145856
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.123064434869032e-05,
+      "loss": 3.9655,
+      "step": 1146368
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.12222584011798e-05,
+      "loss": 3.9609,
+      "step": 1146880
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.121387245366928e-05,
+      "loss": 3.9586,
+      "step": 1147392
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.120548650615876e-05,
+      "loss": 3.9518,
+      "step": 1147904
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.119710055864824e-05,
+      "loss": 3.9529,
+      "step": 1148416
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.118871461113772e-05,
+      "loss": 3.9542,
+      "step": 1148928
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.11803286636272e-05,
+      "loss": 3.9631,
+      "step": 1149440
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.117194271611668e-05,
+      "loss": 3.9541,
+      "step": 1149952
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.116355676860616e-05,
+      "loss": 3.9589,
+      "step": 1150464
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.115518719989937e-05,
+      "loss": 3.9673,
+      "step": 1150976
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.114680125238885e-05,
+      "loss": 3.9466,
+      "step": 1151488
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.113841530487833e-05,
+      "loss": 3.9532,
+      "step": 1152000
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.113002935736781e-05,
+      "loss": 3.9396,
+      "step": 1152512
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.112164340985729e-05,
+      "loss": 3.9471,
+      "step": 1153024
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.111325746234677e-05,
+      "loss": 3.9553,
+      "step": 1153536
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.110487151483625e-05,
+      "loss": 3.9445,
+      "step": 1154048
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.109648556732573e-05,
+      "loss": 3.9555,
+      "step": 1154560
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.108809961981521e-05,
+      "loss": 3.9724,
+      "step": 1155072
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.1079746429912154e-05,
+      "loss": 3.951,
+      "step": 1155584
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.1071360482401634e-05,
+      "loss": 3.9582,
+      "step": 1156096
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.106297453489111e-05,
+      "loss": 3.9635,
+      "step": 1156608
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.105458858738059e-05,
+      "loss": 3.9495,
+      "step": 1157120
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.104620263987007e-05,
+      "loss": 3.9556,
+      "step": 1157632
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.103781669235955e-05,
+      "loss": 3.9436,
+      "step": 1158144
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.1029430744849034e-05,
+      "loss": 3.9553,
+      "step": 1158656
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.1021044797338514e-05,
+      "loss": 3.9525,
+      "step": 1159168
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 3.101267522863172e-05,
+      "loss": 3.9425,
+      "step": 1159680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.10042892811212e-05,
+      "loss": 3.9526,
+      "step": 1160192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.099590333361068e-05,
+      "loss": 3.9472,
+      "step": 1160704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.098751738610016e-05,
+      "loss": 3.9609,
+      "step": 1161216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.097913143858964e-05,
+      "loss": 3.9592,
+      "step": 1161728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.097074549107912e-05,
+      "loss": 3.9467,
+      "step": 1162240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.096237592237233e-05,
+      "loss": 3.9513,
+      "step": 1162752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.095398997486181e-05,
+      "loss": 3.9702,
+      "step": 1163264
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.094560402735129e-05,
+      "loss": 3.9443,
+      "step": 1163776
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.093721807984077e-05,
+      "loss": 3.9531,
+      "step": 1164288
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.092883213233025e-05,
+      "loss": 3.9483,
+      "step": 1164800
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.092044618481973e-05,
+      "loss": 3.9483,
+      "step": 1165312
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.091206023730921e-05,
+      "loss": 3.9484,
+      "step": 1165824
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.09036742897987e-05,
+      "loss": 3.964,
+      "step": 1166336
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.089530472109191e-05,
+      "loss": 3.9595,
+      "step": 1166848
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.088691877358139e-05,
+      "loss": 3.9545,
+      "step": 1167360
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.087853282607087e-05,
+      "loss": 3.9544,
+      "step": 1167872
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.087014687856035e-05,
+      "loss": 3.9465,
+      "step": 1168384
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0861777309853556e-05,
+      "loss": 3.9623,
+      "step": 1168896
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0853391362343036e-05,
+      "loss": 3.9327,
+      "step": 1169408
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0845005414832516e-05,
+      "loss": 3.9515,
+      "step": 1169920
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0836619467321996e-05,
+      "loss": 3.9556,
+      "step": 1170432
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.082823351981147e-05,
+      "loss": 3.9618,
+      "step": 1170944
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.081984757230095e-05,
+      "loss": 3.9516,
+      "step": 1171456
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0811461624790436e-05,
+      "loss": 3.9401,
+      "step": 1171968
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0803075677279916e-05,
+      "loss": 3.9527,
+      "step": 1172480
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0794706108573125e-05,
+      "loss": 3.9413,
+      "step": 1172992
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0786320161062605e-05,
+      "loss": 3.9681,
+      "step": 1173504
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0777934213552085e-05,
+      "loss": 3.948,
+      "step": 1174016
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0769548266041565e-05,
+      "loss": 3.9677,
+      "step": 1174528
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0761178697334774e-05,
+      "loss": 3.9498,
+      "step": 1175040
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0752792749824254e-05,
+      "loss": 3.9438,
+      "step": 1175552
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.074442318111747e-05,
+      "loss": 3.9609,
+      "step": 1176064
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.073603723360694e-05,
+      "loss": 3.9417,
+      "step": 1176576
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.072765128609642e-05,
+      "loss": 3.9478,
+      "step": 1177088
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.07192653385859e-05,
+      "loss": 3.9459,
+      "step": 1177600
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.071089576987912e-05,
+      "loss": 3.9603,
+      "step": 1178112
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.07025098223686e-05,
+      "loss": 3.9414,
+      "step": 1178624
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.069412387485808e-05,
+      "loss": 3.9464,
+      "step": 1179136
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.068573792734756e-05,
+      "loss": 3.9397,
+      "step": 1179648
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.067735197983704e-05,
+      "loss": 3.9416,
+      "step": 1180160
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.066896603232652e-05,
+      "loss": 3.9615,
+      "step": 1180672
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0660580084816e-05,
+      "loss": 3.9579,
+      "step": 1181184
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.065219413730548e-05,
+      "loss": 3.9466,
+      "step": 1181696
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.064382456859869e-05,
+      "loss": 3.95,
+      "step": 1182208
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.06354549998919e-05,
+      "loss": 3.957,
+      "step": 1182720
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.062706905238138e-05,
+      "loss": 3.9551,
+      "step": 1183232
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.061868310487086e-05,
+      "loss": 3.9521,
+      "step": 1183744
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0610297157360343e-05,
+      "loss": 3.9508,
+      "step": 1184256
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0601911209849823e-05,
+      "loss": 3.9541,
+      "step": 1184768
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.059354164114303e-05,
+      "loss": 3.9578,
+      "step": 1185280
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.058515569363251e-05,
+      "loss": 3.9492,
+      "step": 1185792
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.057676974612199e-05,
+      "loss": 3.949,
+      "step": 1186304
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.056838379861147e-05,
+      "loss": 3.9632,
+      "step": 1186816
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.055999785110095e-05,
+      "loss": 3.9538,
+      "step": 1187328
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.055162828239416e-05,
+      "loss": 3.9369,
+      "step": 1187840
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.054324233488364e-05,
+      "loss": 3.952,
+      "step": 1188352
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.053485638737312e-05,
+      "loss": 3.9556,
+      "step": 1188864
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.05264704398626e-05,
+      "loss": 3.9503,
+      "step": 1189376
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.051810087115581e-05,
+      "loss": 3.9487,
+      "step": 1189888
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3.0509714923645294e-05,
+      "loss": 3.9433,
+      "step": 1190400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0501328976134774e-05,
+      "loss": 3.9485,
+      "step": 1190912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0492943028624254e-05,
+      "loss": 3.946,
+      "step": 1191424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0484557081113734e-05,
+      "loss": 3.9465,
+      "step": 1191936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0476171133603214e-05,
+      "loss": 3.9517,
+      "step": 1192448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0467785186092697e-05,
+      "loss": 3.9475,
+      "step": 1192960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0459399238582177e-05,
+      "loss": 3.9643,
+      "step": 1193472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0451013291071657e-05,
+      "loss": 3.9325,
+      "step": 1193984
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0442643722364866e-05,
+      "loss": 3.9418,
+      "step": 1194496
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0434257774854346e-05,
+      "loss": 3.9486,
+      "step": 1195008
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0425871827343826e-05,
+      "loss": 3.9487,
+      "step": 1195520
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0417485879833302e-05,
+      "loss": 3.9569,
+      "step": 1196032
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0409099932322782e-05,
+      "loss": 3.9517,
+      "step": 1196544
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0400730363615998e-05,
+      "loss": 3.9422,
+      "step": 1197056
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0392344416105478e-05,
+      "loss": 3.9465,
+      "step": 1197568
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.038395846859495e-05,
+      "loss": 3.9484,
+      "step": 1198080
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0375572521084435e-05,
+      "loss": 3.9387,
+      "step": 1198592
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0367186573573915e-05,
+      "loss": 3.9399,
+      "step": 1199104
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.035881700486713e-05,
+      "loss": 3.9479,
+      "step": 1199616
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0350431057356604e-05,
+      "loss": 3.95,
+      "step": 1200128
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0342045109846084e-05,
+      "loss": 3.9365,
+      "step": 1200640
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0333659162335564e-05,
+      "loss": 3.9438,
+      "step": 1201152
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0325273214825044e-05,
+      "loss": 3.934,
+      "step": 1201664
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0316903646118256e-05,
+      "loss": 3.9519,
+      "step": 1202176
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0308517698607736e-05,
+      "loss": 3.9406,
+      "step": 1202688
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0300131751097216e-05,
+      "loss": 3.944,
+      "step": 1203200
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0291745803586696e-05,
+      "loss": 3.9487,
+      "step": 1203712
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0283359856076176e-05,
+      "loss": 3.9515,
+      "step": 1204224
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.027499028736939e-05,
+      "loss": 3.9448,
+      "step": 1204736
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.026660433985887e-05,
+      "loss": 3.9443,
+      "step": 1205248
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0258218392348348e-05,
+      "loss": 3.9497,
+      "step": 1205760
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0249832444837828e-05,
+      "loss": 3.9433,
+      "step": 1206272
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0241446497327308e-05,
+      "loss": 3.9365,
+      "step": 1206784
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0233076928620517e-05,
+      "loss": 3.949,
+      "step": 1207296
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0224690981109997e-05,
+      "loss": 3.9444,
+      "step": 1207808
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.021630503359948e-05,
+      "loss": 3.9554,
+      "step": 1208320
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.020791908608896e-05,
+      "loss": 3.9396,
+      "step": 1208832
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.019954951738217e-05,
+      "loss": 3.9462,
+      "step": 1209344
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.019116356987165e-05,
+      "loss": 3.9402,
+      "step": 1209856
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.018279400116486e-05,
+      "loss": 3.9444,
+      "step": 1210368
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0174408053654342e-05,
+      "loss": 3.9337,
+      "step": 1210880
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0166022106143822e-05,
+      "loss": 3.9546,
+      "step": 1211392
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0157636158633302e-05,
+      "loss": 3.9595,
+      "step": 1211904
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0149250211122782e-05,
+      "loss": 3.9441,
+      "step": 1212416
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0140864263612262e-05,
+      "loss": 3.939,
+      "step": 1212928
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0132478316101742e-05,
+      "loss": 3.9522,
+      "step": 1213440
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0124092368591222e-05,
+      "loss": 3.9499,
+      "step": 1213952
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0115722799884434e-05,
+      "loss": 3.9427,
+      "step": 1214464
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0107336852373914e-05,
+      "loss": 3.9464,
+      "step": 1214976
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0098950904863394e-05,
+      "loss": 3.9572,
+      "step": 1215488
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0090564957352874e-05,
+      "loss": 3.9506,
+      "step": 1216000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0082195388646083e-05,
+      "loss": 3.9462,
+      "step": 1216512
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0073809441135563e-05,
+      "loss": 3.9386,
+      "step": 1217024
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0065423493625043e-05,
+      "loss": 3.9515,
+      "step": 1217536
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0057037546114526e-05,
+      "loss": 3.9485,
+      "step": 1218048
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0048651598604006e-05,
+      "loss": 3.9364,
+      "step": 1218560
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0040282029897216e-05,
+      "loss": 3.9609,
+      "step": 1219072
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0031896082386695e-05,
+      "loss": 3.9422,
+      "step": 1219584
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0023510134876175e-05,
+      "loss": 3.9503,
+      "step": 1220096
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.0015124187365655e-05,
+      "loss": 3.9518,
+      "step": 1220608
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 3.0006738239855135e-05,
+      "loss": 3.9403,
+      "step": 1221120
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.069087505340576,
+      "eval_runtime": 303.4734,
+      "eval_samples_per_second": 1257.412,
+      "eval_steps_per_second": 39.295,
+      "step": 1221120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9998352292344612e-05,
+      "loss": 3.9442,
+      "step": 1221632
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9989966344834092e-05,
+      "loss": 3.943,
+      "step": 1222144
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9981580397323572e-05,
+      "loss": 3.9566,
+      "step": 1222656
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9973210828616788e-05,
+      "loss": 3.9469,
+      "step": 1223168
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9964824881106264e-05,
+      "loss": 3.95,
+      "step": 1223680
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9956438933595744e-05,
+      "loss": 3.9392,
+      "step": 1224192
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.994806936488896e-05,
+      "loss": 3.9427,
+      "step": 1224704
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.993968341737844e-05,
+      "loss": 3.9423,
+      "step": 1225216
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9931297469867913e-05,
+      "loss": 3.9526,
+      "step": 1225728
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9922911522357393e-05,
+      "loss": 3.9409,
+      "step": 1226240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9914525574846873e-05,
+      "loss": 3.9525,
+      "step": 1226752
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9906156006140086e-05,
+      "loss": 3.9562,
+      "step": 1227264
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9897770058629566e-05,
+      "loss": 3.9394,
+      "step": 1227776
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9889384111119046e-05,
+      "loss": 3.9403,
+      "step": 1228288
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9880998163608525e-05,
+      "loss": 3.9301,
+      "step": 1228800
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9872612216098005e-05,
+      "loss": 3.9377,
+      "step": 1229312
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9864226268587485e-05,
+      "loss": 3.9436,
+      "step": 1229824
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9855840321076965e-05,
+      "loss": 3.9372,
+      "step": 1230336
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9847454373566445e-05,
+      "loss": 3.9432,
+      "step": 1230848
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9839084804859658e-05,
+      "loss": 3.9632,
+      "step": 1231360
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9830715236152867e-05,
+      "loss": 3.9405,
+      "step": 1231872
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9822329288642347e-05,
+      "loss": 3.9487,
+      "step": 1232384
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9813943341131827e-05,
+      "loss": 3.9472,
+      "step": 1232896
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9805557393621307e-05,
+      "loss": 3.9421,
+      "step": 1233408
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.979717144611079e-05,
+      "loss": 3.9438,
+      "step": 1233920
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.9788801877404e-05,
+      "loss": 3.9349,
+      "step": 1234432
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.978041592989348e-05,
+      "loss": 3.9402,
+      "step": 1234944
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.977202998238296e-05,
+      "loss": 3.945,
+      "step": 1235456
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.976364403487244e-05,
+      "loss": 3.9308,
+      "step": 1235968
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.975525808736192e-05,
+      "loss": 3.9413,
+      "step": 1236480
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.97468721398514e-05,
+      "loss": 3.9384,
+      "step": 1236992
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.973850257114461e-05,
+      "loss": 3.9491,
+      "step": 1237504
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.973011662363409e-05,
+      "loss": 3.944,
+      "step": 1238016
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.972173067612357e-05,
+      "loss": 3.9364,
+      "step": 1238528
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.971336110741678e-05,
+      "loss": 3.9427,
+      "step": 1239040
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.970497515990626e-05,
+      "loss": 3.9592,
+      "step": 1239552
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9696589212395744e-05,
+      "loss": 3.9335,
+      "step": 1240064
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9688203264885224e-05,
+      "loss": 3.9423,
+      "step": 1240576
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9679817317374704e-05,
+      "loss": 3.94,
+      "step": 1241088
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9671431369864184e-05,
+      "loss": 3.9328,
+      "step": 1241600
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9663045422353664e-05,
+      "loss": 3.9404,
+      "step": 1242112
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9654659474843144e-05,
+      "loss": 3.9548,
+      "step": 1242624
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9646273527332624e-05,
+      "loss": 3.9496,
+      "step": 1243136
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9637903958625836e-05,
+      "loss": 3.9401,
+      "step": 1243648
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9629518011115316e-05,
+      "loss": 3.9469,
+      "step": 1244160
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9621132063604796e-05,
+      "loss": 3.9384,
+      "step": 1244672
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.961274611609427e-05,
+      "loss": 3.9509,
+      "step": 1245184
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.960436016858375e-05,
+      "loss": 3.9249,
+      "step": 1245696
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.959597422107323e-05,
+      "loss": 3.9385,
+      "step": 1246208
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9587588273562712e-05,
+      "loss": 3.9489,
+      "step": 1246720
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9579202326052192e-05,
+      "loss": 3.9473,
+      "step": 1247232
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.95708327573454e-05,
+      "loss": 3.9426,
+      "step": 1247744
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.956244680983488e-05,
+      "loss": 3.9308,
+      "step": 1248256
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9554077241128097e-05,
+      "loss": 3.9435,
+      "step": 1248768
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9545691293617574e-05,
+      "loss": 3.929,
+      "step": 1249280
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9537305346107054e-05,
+      "loss": 3.9548,
+      "step": 1249792
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9528919398596534e-05,
+      "loss": 3.9355,
+      "step": 1250304
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9520549829889743e-05,
+      "loss": 3.9533,
+      "step": 1250816
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9512163882379223e-05,
+      "loss": 3.9415,
+      "step": 1251328
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9503777934868703e-05,
+      "loss": 3.9349,
+      "step": 1251840
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9495391987358183e-05,
+      "loss": 3.9453,
+      "step": 1252352
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9487006039847666e-05,
+      "loss": 3.9372,
+      "step": 1252864
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9478620092337146e-05,
+      "loss": 3.938,
+      "step": 1253376
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9470234144826626e-05,
+      "loss": 3.9335,
+      "step": 1253888
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9461848197316106e-05,
+      "loss": 3.9496,
+      "step": 1254400
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9453462249805586e-05,
+      "loss": 3.933,
+      "step": 1254912
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9445076302295066e-05,
+      "loss": 3.9348,
+      "step": 1255424
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9436706733588275e-05,
+      "loss": 3.9258,
+      "step": 1255936
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9428320786077758e-05,
+      "loss": 3.93,
+      "step": 1256448
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9419934838567238e-05,
+      "loss": 3.9508,
+      "step": 1256960
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9411548891056718e-05,
+      "loss": 3.9483,
+      "step": 1257472
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9403162943546198e-05,
+      "loss": 3.9371,
+      "step": 1257984
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9394776996035678e-05,
+      "loss": 3.9398,
+      "step": 1258496
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9386391048525158e-05,
+      "loss": 3.9452,
+      "step": 1259008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9378021479818367e-05,
+      "loss": 3.9483,
+      "step": 1259520
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9369635532307847e-05,
+      "loss": 3.9397,
+      "step": 1260032
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.936124958479733e-05,
+      "loss": 3.9426,
+      "step": 1260544
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.935286363728681e-05,
+      "loss": 3.9416,
+      "step": 1261056
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9344477689776284e-05,
+      "loss": 3.9452,
+      "step": 1261568
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.93361081210695e-05,
+      "loss": 3.9397,
+      "step": 1262080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.932772217355898e-05,
+      "loss": 3.9416,
+      "step": 1262592
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.931933622604846e-05,
+      "loss": 3.9511,
+      "step": 1263104
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9310950278537936e-05,
+      "loss": 3.9415,
+      "step": 1263616
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9302564331027416e-05,
+      "loss": 3.925,
+      "step": 1264128
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9294194762320632e-05,
+      "loss": 3.9404,
+      "step": 1264640
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9285808814810105e-05,
+      "loss": 3.9422,
+      "step": 1265152
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9277422867299585e-05,
+      "loss": 3.9433,
+      "step": 1265664
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9269036919789068e-05,
+      "loss": 3.9373,
+      "step": 1266176
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.9260650972278548e-05,
+      "loss": 3.9325,
+      "step": 1266688
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9252265024768028e-05,
+      "loss": 3.941,
+      "step": 1267200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9243895456061237e-05,
+      "loss": 3.9317,
+      "step": 1267712
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9235509508550717e-05,
+      "loss": 3.9394,
+      "step": 1268224
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9227123561040197e-05,
+      "loss": 3.9363,
+      "step": 1268736
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9218737613529677e-05,
+      "loss": 3.9334,
+      "step": 1269248
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.921035166601916e-05,
+      "loss": 3.9565,
+      "step": 1269760
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.920196571850864e-05,
+      "loss": 3.9264,
+      "step": 1270272
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.919359614980185e-05,
+      "loss": 3.9256,
+      "step": 1270784
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.918521020229133e-05,
+      "loss": 3.9412,
+      "step": 1271296
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.917682425478081e-05,
+      "loss": 3.9411,
+      "step": 1271808
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.916843830727029e-05,
+      "loss": 3.9462,
+      "step": 1272320
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.916005235975977e-05,
+      "loss": 3.937,
+      "step": 1272832
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9151666412249253e-05,
+      "loss": 3.9315,
+      "step": 1273344
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9143280464738733e-05,
+      "loss": 3.934,
+      "step": 1273856
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9134894517228213e-05,
+      "loss": 3.944,
+      "step": 1274368
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.912652494852142e-05,
+      "loss": 3.9278,
+      "step": 1274880
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.911815537981463e-05,
+      "loss": 3.9291,
+      "step": 1275392
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9109769432304114e-05,
+      "loss": 3.9408,
+      "step": 1275904
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9101383484793594e-05,
+      "loss": 3.9376,
+      "step": 1276416
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9092997537283074e-05,
+      "loss": 3.9284,
+      "step": 1276928
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9084611589772554e-05,
+      "loss": 3.9302,
+      "step": 1277440
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9076225642262034e-05,
+      "loss": 3.9296,
+      "step": 1277952
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9067839694751514e-05,
+      "loss": 3.9412,
+      "step": 1278464
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9059453747240994e-05,
+      "loss": 3.9307,
+      "step": 1278976
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9051084178534206e-05,
+      "loss": 3.9358,
+      "step": 1279488
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9042698231023686e-05,
+      "loss": 3.9385,
+      "step": 1280000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9034312283513166e-05,
+      "loss": 3.9421,
+      "step": 1280512
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9025942714806375e-05,
+      "loss": 3.9319,
+      "step": 1281024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9017556767295855e-05,
+      "loss": 3.9367,
+      "step": 1281536
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9009170819785335e-05,
+      "loss": 3.9412,
+      "step": 1282048
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.9000784872274815e-05,
+      "loss": 3.9376,
+      "step": 1282560
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8992398924764292e-05,
+      "loss": 3.9241,
+      "step": 1283072
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.898401297725377e-05,
+      "loss": 3.94,
+      "step": 1283584
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.897562702974325e-05,
+      "loss": 3.9357,
+      "step": 1284096
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.896724108223273e-05,
+      "loss": 3.9419,
+      "step": 1284608
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.895885513472221e-05,
+      "loss": 3.9284,
+      "step": 1285120
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.895046918721169e-05,
+      "loss": 3.941,
+      "step": 1285632
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8942099618504904e-05,
+      "loss": 3.9302,
+      "step": 1286144
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8933713670994384e-05,
+      "loss": 3.9344,
+      "step": 1286656
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8925327723483864e-05,
+      "loss": 3.918,
+      "step": 1287168
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8916941775973344e-05,
+      "loss": 3.9492,
+      "step": 1287680
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8908555828462824e-05,
+      "loss": 3.9472,
+      "step": 1288192
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8900169880952304e-05,
+      "loss": 3.9328,
+      "step": 1288704
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8891783933441784e-05,
+      "loss": 3.9247,
+      "step": 1289216
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8883397985931264e-05,
+      "loss": 3.949,
+      "step": 1289728
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8875028417224476e-05,
+      "loss": 3.9369,
+      "step": 1290240
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8866642469713956e-05,
+      "loss": 3.9371,
+      "step": 1290752
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8858256522203436e-05,
+      "loss": 3.9347,
+      "step": 1291264
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8849870574692916e-05,
+      "loss": 3.9455,
+      "step": 1291776
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8841484627182396e-05,
+      "loss": 3.9378,
+      "step": 1292288
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8833098679671876e-05,
+      "loss": 3.9363,
+      "step": 1292800
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8824712732161356e-05,
+      "loss": 3.9279,
+      "step": 1293312
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8816326784650832e-05,
+      "loss": 3.9411,
+      "step": 1293824
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.880795721594405e-05,
+      "loss": 3.9387,
+      "step": 1294336
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8799571268433528e-05,
+      "loss": 3.9277,
+      "step": 1294848
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8791201699726737e-05,
+      "loss": 3.9499,
+      "step": 1295360
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8782815752216217e-05,
+      "loss": 3.931,
+      "step": 1295872
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.87744298047057e-05,
+      "loss": 3.9397,
+      "step": 1296384
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.876604385719518e-05,
+      "loss": 3.9407,
+      "step": 1296896
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.8757657909684654e-05,
+      "loss": 3.9322,
+      "step": 1297408
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 4.06406307220459,
+      "eval_runtime": 304.3479,
+      "eval_samples_per_second": 1253.799,
+      "eval_steps_per_second": 39.182,
+      "step": 1297440
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8749271962174134e-05,
+      "loss": 3.9313,
+      "step": 1297920
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8740886014663614e-05,
+      "loss": 3.9353,
+      "step": 1298432
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.873251644595683e-05,
+      "loss": 3.9447,
+      "step": 1298944
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8724130498446306e-05,
+      "loss": 3.9388,
+      "step": 1299456
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8715744550935786e-05,
+      "loss": 3.9425,
+      "step": 1299968
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8707358603425266e-05,
+      "loss": 3.9289,
+      "step": 1300480
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8698989034718482e-05,
+      "loss": 3.9325,
+      "step": 1300992
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8690603087207955e-05,
+      "loss": 3.9295,
+      "step": 1301504
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.868221713969744e-05,
+      "loss": 3.9427,
+      "step": 1302016
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.867383119218692e-05,
+      "loss": 3.9325,
+      "step": 1302528
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8665461623480128e-05,
+      "loss": 3.9393,
+      "step": 1303040
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8657075675969607e-05,
+      "loss": 3.9452,
+      "step": 1303552
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8648706107262823e-05,
+      "loss": 3.9293,
+      "step": 1304064
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8640320159752303e-05,
+      "loss": 3.9308,
+      "step": 1304576
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.863193421224178e-05,
+      "loss": 3.921,
+      "step": 1305088
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.862354826473126e-05,
+      "loss": 3.929,
+      "step": 1305600
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.861516231722074e-05,
+      "loss": 3.9303,
+      "step": 1306112
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.860677636971022e-05,
+      "loss": 3.9287,
+      "step": 1306624
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.85983904221997e-05,
+      "loss": 3.9289,
+      "step": 1307136
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.859000447468918e-05,
+      "loss": 3.955,
+      "step": 1307648
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.858161852717866e-05,
+      "loss": 3.93,
+      "step": 1308160
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.857323257966814e-05,
+      "loss": 3.9351,
+      "step": 1308672
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8564863010961352e-05,
+      "loss": 3.9425,
+      "step": 1309184
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.855649344225456e-05,
+      "loss": 3.9349,
+      "step": 1309696
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.854810749474404e-05,
+      "loss": 3.9301,
+      "step": 1310208
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.853972154723352e-05,
+      "loss": 3.925,
+      "step": 1310720
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8531335599723e-05,
+      "loss": 3.9295,
+      "step": 1311232
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8522949652212484e-05,
+      "loss": 3.9324,
+      "step": 1311744
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 2.8514563704701964e-05,
+      "loss": 3.9214,
+      "step": 1312256
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8506177757191444e-05,
+      "loss": 3.9353,
+      "step": 1312768
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8497791809680924e-05,
+      "loss": 3.9245,
+      "step": 1313280
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8489405862170404e-05,
+      "loss": 3.9383,
+      "step": 1313792
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8481019914659884e-05,
+      "loss": 3.937,
+      "step": 1314304
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8472633967149364e-05,
+      "loss": 3.9222,
+      "step": 1314816
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8464280777246306e-05,
+      "loss": 3.9379,
+      "step": 1315328
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8455894829735786e-05,
+      "loss": 3.9481,
+      "step": 1315840
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8447508882225266e-05,
+      "loss": 3.9217,
+      "step": 1316352
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8439122934714746e-05,
+      "loss": 3.9368,
+      "step": 1316864
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8430736987204226e-05,
+      "loss": 3.9293,
+      "step": 1317376
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8422351039693706e-05,
+      "loss": 3.9235,
+      "step": 1317888
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8413965092183185e-05,
+      "loss": 3.9311,
+      "step": 1318400
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.840557914467267e-05,
+      "loss": 3.9433,
+      "step": 1318912
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8397193197162142e-05,
+      "loss": 3.9371,
+      "step": 1319424
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8388807249651622e-05,
+      "loss": 3.93,
+      "step": 1319936
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8380421302141102e-05,
+      "loss": 3.9401,
+      "step": 1320448
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8372035354630582e-05,
+      "loss": 3.9298,
+      "step": 1320960
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8363665785923794e-05,
+      "loss": 3.9375,
+      "step": 1321472
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8355279838413274e-05,
+      "loss": 3.9177,
+      "step": 1321984
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.834691026970649e-05,
+      "loss": 3.9231,
+      "step": 1322496
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8338524322195963e-05,
+      "loss": 3.9385,
+      "step": 1323008
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8330138374685443e-05,
+      "loss": 3.941,
+      "step": 1323520
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8321752427174923e-05,
+      "loss": 3.9281,
+      "step": 1324032
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8313366479664407e-05,
+      "loss": 3.9208,
+      "step": 1324544
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8304996910957616e-05,
+      "loss": 3.9362,
+      "step": 1325056
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8296610963447096e-05,
+      "loss": 3.9164,
+      "step": 1325568
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8288225015936576e-05,
+      "loss": 3.9433,
+      "step": 1326080
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8279839068426056e-05,
+      "loss": 3.931,
+      "step": 1326592
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8271453120915536e-05,
+      "loss": 3.9443,
+      "step": 1327104
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8263067173405015e-05,
+      "loss": 3.9318,
+      "step": 1327616
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8254697604698228e-05,
+      "loss": 3.9249,
+      "step": 1328128
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8246311657187708e-05,
+      "loss": 3.9314,
+      "step": 1328640
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8237925709677188e-05,
+      "loss": 3.9332,
+      "step": 1329152
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8229539762166668e-05,
+      "loss": 3.9235,
+      "step": 1329664
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8221170193459877e-05,
+      "loss": 3.9249,
+      "step": 1330176
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8212784245949357e-05,
+      "loss": 3.9397,
+      "step": 1330688
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.820439829843884e-05,
+      "loss": 3.9283,
+      "step": 1331200
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.819601235092832e-05,
+      "loss": 3.9198,
+      "step": 1331712
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.81876264034178e-05,
+      "loss": 3.9208,
+      "step": 1332224
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.817924045590728e-05,
+      "loss": 3.9211,
+      "step": 1332736
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.817085450839676e-05,
+      "loss": 3.94,
+      "step": 1333248
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.816246856088624e-05,
+      "loss": 3.9404,
+      "step": 1333760
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.815408261337572e-05,
+      "loss": 3.9263,
+      "step": 1334272
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8145713044668932e-05,
+      "loss": 3.929,
+      "step": 1334784
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8137327097158412e-05,
+      "loss": 3.9349,
+      "step": 1335296
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.812895752845162e-05,
+      "loss": 3.9353,
+      "step": 1335808
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.81205715809411e-05,
+      "loss": 3.9272,
+      "step": 1336320
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.811220201223431e-05,
+      "loss": 3.9344,
+      "step": 1336832
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8103816064723794e-05,
+      "loss": 3.9321,
+      "step": 1337344
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8095430117213274e-05,
+      "loss": 3.9408,
+      "step": 1337856
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8087044169702754e-05,
+      "loss": 3.9268,
+      "step": 1338368
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8078658222192234e-05,
+      "loss": 3.9324,
+      "step": 1338880
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8070272274681714e-05,
+      "loss": 3.9373,
+      "step": 1339392
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8061886327171194e-05,
+      "loss": 3.9352,
+      "step": 1339904
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8053500379660674e-05,
+      "loss": 3.9177,
+      "step": 1340416
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8045130810953886e-05,
+      "loss": 3.9318,
+      "step": 1340928
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8036744863443366e-05,
+      "loss": 3.9284,
+      "step": 1341440
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8028358915932846e-05,
+      "loss": 3.9334,
+      "step": 1341952
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.8019972968422326e-05,
+      "loss": 3.9308,
+      "step": 1342464
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.80115870209118e-05,
+      "loss": 3.921,
+      "step": 1342976
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.800320107340128e-05,
+      "loss": 3.9301,
+      "step": 1343488
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7994831504694495e-05,
+      "loss": 3.9229,
+      "step": 1344000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.798644555718397e-05,
+      "loss": 3.9336,
+      "step": 1344512
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.797805960967345e-05,
+      "loss": 3.9227,
+      "step": 1345024
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7969690040966667e-05,
+      "loss": 3.9282,
+      "step": 1345536
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7961304093456147e-05,
+      "loss": 3.9405,
+      "step": 1346048
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7952918145945624e-05,
+      "loss": 3.9298,
+      "step": 1346560
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7944532198435104e-05,
+      "loss": 3.9127,
+      "step": 1347072
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7936146250924584e-05,
+      "loss": 3.9283,
+      "step": 1347584
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7927760303414064e-05,
+      "loss": 3.9319,
+      "step": 1348096
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7919374355903544e-05,
+      "loss": 3.9347,
+      "step": 1348608
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7910988408393024e-05,
+      "loss": 3.9269,
+      "step": 1349120
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7902602460882504e-05,
+      "loss": 3.9241,
+      "step": 1349632
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7894232892175716e-05,
+      "loss": 3.9228,
+      "step": 1350144
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7885846944665196e-05,
+      "loss": 3.9348,
+      "step": 1350656
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7877460997154676e-05,
+      "loss": 3.9143,
+      "step": 1351168
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7869075049644156e-05,
+      "loss": 3.9232,
+      "step": 1351680
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7860689102133636e-05,
+      "loss": 3.9314,
+      "step": 1352192
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7852303154623116e-05,
+      "loss": 3.9309,
+      "step": 1352704
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7843917207112596e-05,
+      "loss": 3.9133,
+      "step": 1353216
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7835531259602076e-05,
+      "loss": 3.9274,
+      "step": 1353728
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7827161690895288e-05,
+      "loss": 3.9157,
+      "step": 1354240
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7818775743384768e-05,
+      "loss": 3.9316,
+      "step": 1354752
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7810389795874248e-05,
+      "loss": 3.9223,
+      "step": 1355264
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7802003848363728e-05,
+      "loss": 3.9175,
+      "step": 1355776
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7793617900853208e-05,
+      "loss": 3.9369,
+      "step": 1356288
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7785248332146417e-05,
+      "loss": 3.9322,
+      "step": 1356800
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7776862384635897e-05,
+      "loss": 3.9223,
+      "step": 1357312
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.776847643712538e-05,
+      "loss": 3.9283,
+      "step": 1357824
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.776009048961486e-05,
+      "loss": 3.9256,
+      "step": 1358336
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.775172092090807e-05,
+      "loss": 3.9307,
+      "step": 1358848
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.774333497339755e-05,
+      "loss": 3.9115,
+      "step": 1359360
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.773494902588703e-05,
+      "loss": 3.9355,
+      "step": 1359872
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.772656307837651e-05,
+      "loss": 3.9234,
+      "step": 1360384
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7718177130865986e-05,
+      "loss": 3.9315,
+      "step": 1360896
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7709791183355466e-05,
+      "loss": 3.9209,
+      "step": 1361408
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7701405235844946e-05,
+      "loss": 3.9287,
+      "step": 1361920
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7693019288334426e-05,
+      "loss": 3.9211,
+      "step": 1362432
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7684649719627635e-05,
+      "loss": 3.9259,
+      "step": 1362944
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.767626377211712e-05,
+      "loss": 3.9083,
+      "step": 1363456
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7667877824606598e-05,
+      "loss": 3.941,
+      "step": 1363968
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7659491877096078e-05,
+      "loss": 3.9343,
+      "step": 1364480
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7651122308389287e-05,
+      "loss": 3.9277,
+      "step": 1364992
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7642736360878767e-05,
+      "loss": 3.9178,
+      "step": 1365504
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7634350413368247e-05,
+      "loss": 3.9361,
+      "step": 1366016
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7625964465857727e-05,
+      "loss": 3.9329,
+      "step": 1366528
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.761757851834721e-05,
+      "loss": 3.9242,
+      "step": 1367040
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.760920894964042e-05,
+      "loss": 3.9226,
+      "step": 1367552
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.76008230021299e-05,
+      "loss": 3.9335,
+      "step": 1368064
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.759243705461938e-05,
+      "loss": 3.9284,
+      "step": 1368576
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.758405110710886e-05,
+      "loss": 3.9357,
+      "step": 1369088
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.757566515959834e-05,
+      "loss": 3.9149,
+      "step": 1369600
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7567295590891552e-05,
+      "loss": 3.9321,
+      "step": 1370112
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7558909643381032e-05,
+      "loss": 3.9286,
+      "step": 1370624
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.755054007467424e-05,
+      "loss": 3.9208,
+      "step": 1371136
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.754215412716372e-05,
+      "loss": 3.941,
+      "step": 1371648
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.75337681796532e-05,
+      "loss": 3.9235,
+      "step": 1372160
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.752538223214268e-05,
+      "loss": 3.9292,
+      "step": 1372672
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7516996284632164e-05,
+      "loss": 3.9276,
+      "step": 1373184
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.7508610337121644e-05,
+      "loss": 3.9259,
+      "step": 1373696
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 4.059935569763184,
+      "eval_runtime": 304.0784,
+      "eval_samples_per_second": 1254.91,
+      "eval_steps_per_second": 39.217,
+      "step": 1373760
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7500224389611124e-05,
+      "loss": 3.9216,
+      "step": 1374208
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7491838442100604e-05,
+      "loss": 3.9299,
+      "step": 1374720
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7483452494590084e-05,
+      "loss": 3.9334,
+      "step": 1375232
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7475066547079564e-05,
+      "loss": 3.9292,
+      "step": 1375744
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7466680599569044e-05,
+      "loss": 3.9364,
+      "step": 1376256
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.745829465205852e-05,
+      "loss": 3.9154,
+      "step": 1376768
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7449908704548e-05,
+      "loss": 3.9213,
+      "step": 1377280
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.744152275703748e-05,
+      "loss": 3.9212,
+      "step": 1377792
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.743313680952696e-05,
+      "loss": 3.9342,
+      "step": 1378304
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.742475086201644e-05,
+      "loss": 3.9251,
+      "step": 1378816
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.741636491450592e-05,
+      "loss": 3.9312,
+      "step": 1379328
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.74079789669954e-05,
+      "loss": 3.9345,
+      "step": 1379840
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7399609398288613e-05,
+      "loss": 3.9207,
+      "step": 1380352
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7391223450778093e-05,
+      "loss": 3.9232,
+      "step": 1380864
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7382837503267573e-05,
+      "loss": 3.9104,
+      "step": 1381376
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7374451555757053e-05,
+      "loss": 3.9224,
+      "step": 1381888
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7366065608246533e-05,
+      "loss": 3.9231,
+      "step": 1382400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7357679660736012e-05,
+      "loss": 3.9182,
+      "step": 1382912
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7349293713225492e-05,
+      "loss": 3.92,
+      "step": 1383424
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7340924144518705e-05,
+      "loss": 3.9457,
+      "step": 1383936
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7332538197008185e-05,
+      "loss": 3.9225,
+      "step": 1384448
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7324152249497665e-05,
+      "loss": 3.9258,
+      "step": 1384960
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7315766301987145e-05,
+      "loss": 3.9324,
+      "step": 1385472
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7307380354476625e-05,
+      "loss": 3.9288,
+      "step": 1385984
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7298994406966105e-05,
+      "loss": 3.9188,
+      "step": 1386496
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7290608459455585e-05,
+      "loss": 3.9155,
+      "step": 1387008
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.728222251194506e-05,
+      "loss": 3.9247,
+      "step": 1387520
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.727383656443454e-05,
+      "loss": 3.9187,
+      "step": 1388032
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.726545061692402e-05,
+      "loss": 3.9135,
+      "step": 1388544
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.72570646694135e-05,
+      "loss": 3.9256,
+      "step": 1389056
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.724867872190298e-05,
+      "loss": 3.9181,
+      "step": 1389568
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.724030915319619e-05,
+      "loss": 3.929,
+      "step": 1390080
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7231923205685673e-05,
+      "loss": 3.9251,
+      "step": 1390592
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7223537258175153e-05,
+      "loss": 3.9148,
+      "step": 1391104
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7215167689468363e-05,
+      "loss": 3.928,
+      "step": 1391616
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7206781741957842e-05,
+      "loss": 3.9416,
+      "step": 1392128
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.719841217325106e-05,
+      "loss": 3.9096,
+      "step": 1392640
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7190026225740535e-05,
+      "loss": 3.9302,
+      "step": 1393152
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7181640278230015e-05,
+      "loss": 3.9156,
+      "step": 1393664
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7173254330719495e-05,
+      "loss": 3.9189,
+      "step": 1394176
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7164868383208975e-05,
+      "loss": 3.9203,
+      "step": 1394688
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7156482435698455e-05,
+      "loss": 3.9363,
+      "step": 1395200
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7148096488187935e-05,
+      "loss": 3.9259,
+      "step": 1395712
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7139710540677415e-05,
+      "loss": 3.9225,
+      "step": 1396224
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7131324593166895e-05,
+      "loss": 3.9283,
+      "step": 1396736
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7122938645656375e-05,
+      "loss": 3.9189,
+      "step": 1397248
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7114552698145858e-05,
+      "loss": 3.9312,
+      "step": 1397760
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7106166750635338e-05,
+      "loss": 3.913,
+      "step": 1398272
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7097797181928547e-05,
+      "loss": 3.9152,
+      "step": 1398784
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7089427613221756e-05,
+      "loss": 3.9284,
+      "step": 1399296
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7081041665711236e-05,
+      "loss": 3.9287,
+      "step": 1399808
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.707265571820072e-05,
+      "loss": 3.9197,
+      "step": 1400320
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.70642697706902e-05,
+      "loss": 3.9152,
+      "step": 1400832
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.705590020198341e-05,
+      "loss": 3.9255,
+      "step": 1401344
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.704751425447289e-05,
+      "loss": 3.9085,
+      "step": 1401856
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.703912830696237e-05,
+      "loss": 3.9306,
+      "step": 1402368
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7030742359451848e-05,
+      "loss": 3.9243,
+      "step": 1402880
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.7022356411941328e-05,
+      "loss": 3.9332,
+      "step": 1403392
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.701397046443081e-05,
+      "loss": 3.9282,
+      "step": 1403904
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.700558451692029e-05,
+      "loss": 3.9164,
+      "step": 1404416
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.699719856940977e-05,
+      "loss": 3.9206,
+      "step": 1404928
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.698882900070298e-05,
+      "loss": 3.926,
+      "step": 1405440
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.698044305319246e-05,
+      "loss": 3.9112,
+      "step": 1405952
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.697205710568194e-05,
+      "loss": 3.9168,
+      "step": 1406464
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.696367115817142e-05,
+      "loss": 3.927,
+      "step": 1406976
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6955285210660897e-05,
+      "loss": 3.9206,
+      "step": 1407488
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6946899263150377e-05,
+      "loss": 3.9091,
+      "step": 1408000
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6938513315639857e-05,
+      "loss": 3.9112,
+      "step": 1408512
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6930127368129337e-05,
+      "loss": 3.9118,
+      "step": 1409024
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.692175779942255e-05,
+      "loss": 3.931,
+      "step": 1409536
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.691337185191203e-05,
+      "loss": 3.9279,
+      "step": 1410048
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.690498590440151e-05,
+      "loss": 3.9187,
+      "step": 1410560
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.689661633569472e-05,
+      "loss": 3.9245,
+      "step": 1411072
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.68882303881842e-05,
+      "loss": 3.9196,
+      "step": 1411584
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.687984444067368e-05,
+      "loss": 3.9289,
+      "step": 1412096
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6871458493163158e-05,
+      "loss": 3.9161,
+      "step": 1412608
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.686307254565264e-05,
+      "loss": 3.9271,
+      "step": 1413120
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.685470297694585e-05,
+      "loss": 3.926,
+      "step": 1413632
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.684631702943533e-05,
+      "loss": 3.928,
+      "step": 1414144
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.683793108192481e-05,
+      "loss": 3.9194,
+      "step": 1414656
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.682954513441429e-05,
+      "loss": 3.923,
+      "step": 1415168
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.682115918690377e-05,
+      "loss": 3.927,
+      "step": 1415680
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.681277323939325e-05,
+      "loss": 3.9264,
+      "step": 1416192
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.680438729188273e-05,
+      "loss": 3.9142,
+      "step": 1416704
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6796001344372214e-05,
+      "loss": 3.9212,
+      "step": 1417216
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6787631775665423e-05,
+      "loss": 3.92,
+      "step": 1417728
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6779245828154903e-05,
+      "loss": 3.9256,
+      "step": 1418240
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6770859880644383e-05,
+      "loss": 3.9248,
+      "step": 1418752
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6762473933133863e-05,
+      "loss": 3.9085,
+      "step": 1419264
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6754104364427075e-05,
+      "loss": 3.92,
+      "step": 1419776
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6745718416916555e-05,
+      "loss": 3.9179,
+      "step": 1420288
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6737332469406035e-05,
+      "loss": 3.9265,
+      "step": 1420800
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6728946521895515e-05,
+      "loss": 3.9098,
+      "step": 1421312
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6720560574384995e-05,
+      "loss": 3.921,
+      "step": 1421824
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6712174626874475e-05,
+      "loss": 3.9312,
+      "step": 1422336
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6703788679363955e-05,
+      "loss": 3.9196,
+      "step": 1422848
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.669540273185343e-05,
+      "loss": 3.9041,
+      "step": 1423360
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.668701678434291e-05,
+      "loss": 3.9179,
+      "step": 1423872
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6678647215636127e-05,
+      "loss": 3.9241,
+      "step": 1424384
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6670261268125607e-05,
+      "loss": 3.9258,
+      "step": 1424896
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6661891699418816e-05,
+      "loss": 3.9222,
+      "step": 1425408
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6653505751908296e-05,
+      "loss": 3.9132,
+      "step": 1425920
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6645119804397776e-05,
+      "loss": 3.9121,
+      "step": 1426432
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.663675023569099e-05,
+      "loss": 3.9246,
+      "step": 1426944
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.662836428818047e-05,
+      "loss": 3.9081,
+      "step": 1427456
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.661997834066995e-05,
+      "loss": 3.9143,
+      "step": 1427968
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.661159239315943e-05,
+      "loss": 3.9195,
+      "step": 1428480
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6603206445648905e-05,
+      "loss": 3.9229,
+      "step": 1428992
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6594820498138385e-05,
+      "loss": 3.901,
+      "step": 1429504
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6586434550627865e-05,
+      "loss": 3.9204,
+      "step": 1430016
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6578048603117345e-05,
+      "loss": 3.9085,
+      "step": 1430528
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6569662655606825e-05,
+      "loss": 3.9207,
+      "step": 1431040
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6561276708096305e-05,
+      "loss": 3.9168,
+      "step": 1431552
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6552890760585785e-05,
+      "loss": 3.9107,
+      "step": 1432064
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6544504813075265e-05,
+      "loss": 3.9265,
+      "step": 1432576
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6536135244368477e-05,
+      "loss": 3.9249,
+      "step": 1433088
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6527782054465422e-05,
+      "loss": 3.9113,
+      "step": 1433600
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6519396106954902e-05,
+      "loss": 3.9182,
+      "step": 1434112
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6511010159444376e-05,
+      "loss": 3.9206,
+      "step": 1434624
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.650262421193386e-05,
+      "loss": 3.919,
+      "step": 1435136
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.649423826442334e-05,
+      "loss": 3.9062,
+      "step": 1435648
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.648585231691282e-05,
+      "loss": 3.9226,
+      "step": 1436160
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.64774663694023e-05,
+      "loss": 3.9169,
+      "step": 1436672
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.646908042189178e-05,
+      "loss": 3.9211,
+      "step": 1437184
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.646069447438126e-05,
+      "loss": 3.9157,
+      "step": 1437696
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.645230852687074e-05,
+      "loss": 3.9211,
+      "step": 1438208
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.644392257936022e-05,
+      "loss": 3.9083,
+      "step": 1438720
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.64355366318497e-05,
+      "loss": 3.9163,
+      "step": 1439232
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6427150684339182e-05,
+      "loss": 3.9007,
+      "step": 1439744
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.641878111563239e-05,
+      "loss": 3.931,
+      "step": 1440256
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.641039516812187e-05,
+      "loss": 3.9238,
+      "step": 1440768
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.640200922061135e-05,
+      "loss": 3.9195,
+      "step": 1441280
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.639363965190456e-05,
+      "loss": 3.909,
+      "step": 1441792
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6385253704394043e-05,
+      "loss": 3.9264,
+      "step": 1442304
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6376867756883523e-05,
+      "loss": 3.9262,
+      "step": 1442816
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6368481809373003e-05,
+      "loss": 3.9141,
+      "step": 1443328
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6360095861862483e-05,
+      "loss": 3.9158,
+      "step": 1443840
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6351726293155692e-05,
+      "loss": 3.924,
+      "step": 1444352
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6343340345645172e-05,
+      "loss": 3.9197,
+      "step": 1444864
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6334954398134652e-05,
+      "loss": 3.9277,
+      "step": 1445376
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6326568450624136e-05,
+      "loss": 3.9116,
+      "step": 1445888
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6318198881917345e-05,
+      "loss": 3.924,
+      "step": 1446400
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6309812934406825e-05,
+      "loss": 3.9177,
+      "step": 1446912
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6301426986896305e-05,
+      "loss": 3.9112,
+      "step": 1447424
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6293041039385785e-05,
+      "loss": 3.9332,
+      "step": 1447936
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6284655091875264e-05,
+      "loss": 3.9139,
+      "step": 1448448
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.627626914436474e-05,
+      "loss": 3.9204,
+      "step": 1448960
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6267899575657957e-05,
+      "loss": 3.9211,
+      "step": 1449472
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.6259513628147437e-05,
+      "loss": 3.9149,
+      "step": 1449984
+    },
+    {
+      "epoch": 0.03,
+      "eval_loss": 4.056296348571777,
+      "eval_runtime": 310.4986,
+      "eval_samples_per_second": 1228.962,
+      "eval_steps_per_second": 38.406,
+      "step": 1450080
+    }
+  ],
+  "logging_steps": 512,
+  "max_steps": 3052726,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 10,
+  "total_flos": 5.5850208657184115e+17,
+  "trial_name": null,
+  "trial_params": null
+}