diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,4250 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 37,
+  "global_step": 584,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00684931506849315,
+      "grad_norm": 6.781628733365755,
+      "learning_rate": 2.0000000000000002e-07,
+      "loss": 2.5668,
+      "step": 1
+    },
+    {
+      "epoch": 0.00684931506849315,
+      "eval_loss": 2.5806074142456055,
+      "eval_runtime": 6.5651,
+      "eval_samples_per_second": 47.524,
+      "eval_steps_per_second": 3.046,
+      "step": 1
+    },
+    {
+      "epoch": 0.0136986301369863,
+      "grad_norm": 6.86619348498488,
+      "learning_rate": 4.0000000000000003e-07,
+      "loss": 2.6025,
+      "step": 2
+    },
+    {
+      "epoch": 0.02054794520547945,
+      "grad_norm": 6.661187717958318,
+      "learning_rate": 6.000000000000001e-07,
+      "loss": 2.6316,
+      "step": 3
+    },
+    {
+      "epoch": 0.0273972602739726,
+      "grad_norm": 6.861551727845244,
+      "learning_rate": 8.000000000000001e-07,
+      "loss": 2.5801,
+      "step": 4
+    },
+    {
+      "epoch": 0.03424657534246575,
+      "grad_norm": 6.69117874797922,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 2.5335,
+      "step": 5
+    },
+    {
+      "epoch": 0.0410958904109589,
+      "grad_norm": 5.875380376726272,
+      "learning_rate": 1.2000000000000002e-06,
+      "loss": 2.5241,
+      "step": 6
+    },
+    {
+      "epoch": 0.04794520547945205,
+      "grad_norm": 5.7978399687358255,
+      "learning_rate": 1.4000000000000001e-06,
+      "loss": 2.4675,
+      "step": 7
+    },
+    {
+      "epoch": 0.0547945205479452,
+      "grad_norm": 3.954765219007858,
+      "learning_rate": 1.6000000000000001e-06,
+      "loss": 2.5811,
+      "step": 8
+    },
+    {
+      "epoch": 0.06164383561643835,
+      "grad_norm": 3.8615688857265162,
+      "learning_rate": 1.8000000000000001e-06,
+      "loss": 2.4886,
+      "step": 9
+    },
+    {
+      "epoch": 0.0684931506849315,
+      "grad_norm": 3.1019873404806138,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 2.5337,
+      "step": 10
+    },
+    {
+      "epoch": 0.07534246575342465,
+      "grad_norm": 1.653000190239802,
+      "learning_rate": 2.2e-06,
+      "loss": 2.493,
+      "step": 11
+    },
+    {
+      "epoch": 0.0821917808219178,
+      "grad_norm": 1.6653405065701912,
+      "learning_rate": 2.4000000000000003e-06,
+      "loss": 2.4862,
+      "step": 12
+    },
+    {
+      "epoch": 0.08904109589041095,
+      "grad_norm": 1.9193066268748051,
+      "learning_rate": 2.6e-06,
+      "loss": 2.5172,
+      "step": 13
+    },
+    {
+      "epoch": 0.0958904109589041,
+      "grad_norm": 1.4108595092412395,
+      "learning_rate": 2.8000000000000003e-06,
+      "loss": 2.4411,
+      "step": 14
+    },
+    {
+      "epoch": 0.10273972602739725,
+      "grad_norm": 2.0335186868451713,
+      "learning_rate": 3e-06,
+      "loss": 2.5062,
+      "step": 15
+    },
+    {
+      "epoch": 0.1095890410958904,
+      "grad_norm": 1.9834579085161663,
+      "learning_rate": 3.2000000000000003e-06,
+      "loss": 2.406,
+      "step": 16
+    },
+    {
+      "epoch": 0.11643835616438356,
+      "grad_norm": 1.7772826080066895,
+      "learning_rate": 3.4000000000000005e-06,
+      "loss": 2.4608,
+      "step": 17
+    },
+    {
+      "epoch": 0.1232876712328767,
+      "grad_norm": 1.3962245609921142,
+      "learning_rate": 3.6000000000000003e-06,
+      "loss": 2.3551,
+      "step": 18
+    },
+    {
+      "epoch": 0.13013698630136986,
+      "grad_norm": 1.190039563451697,
+      "learning_rate": 3.8000000000000005e-06,
+      "loss": 2.4188,
+      "step": 19
+    },
+    {
+      "epoch": 0.136986301369863,
+      "grad_norm": 1.0109005845053234,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 2.4981,
+      "step": 20
+    },
+    {
+      "epoch": 0.14383561643835616,
+      "grad_norm": 1.1671978967591745,
+      "learning_rate": 4.2000000000000004e-06,
+      "loss": 2.319,
+      "step": 21
+    },
+    {
+      "epoch": 0.1506849315068493,
+      "grad_norm": 1.0658877053683267,
+      "learning_rate": 4.4e-06,
+      "loss": 2.3067,
+      "step": 22
+    },
+    {
+      "epoch": 0.15753424657534246,
+      "grad_norm": 1.0578186226484172,
+      "learning_rate": 4.600000000000001e-06,
+      "loss": 2.4644,
+      "step": 23
+    },
+    {
+      "epoch": 0.1643835616438356,
+      "grad_norm": 0.9493987331518986,
+      "learning_rate": 4.800000000000001e-06,
+      "loss": 2.4179,
+      "step": 24
+    },
+    {
+      "epoch": 0.17123287671232876,
+      "grad_norm": 0.8876721262354197,
+      "learning_rate": 5e-06,
+      "loss": 2.4548,
+      "step": 25
+    },
+    {
+      "epoch": 0.1780821917808219,
+      "grad_norm": 0.9016648720831166,
+      "learning_rate": 4.999960519285878e-06,
+      "loss": 2.4203,
+      "step": 26
+    },
+    {
+      "epoch": 0.18493150684931506,
+      "grad_norm": 0.8594911221513131,
+      "learning_rate": 4.999842078390492e-06,
+      "loss": 2.4445,
+      "step": 27
+    },
+    {
+      "epoch": 0.1917808219178082,
+      "grad_norm": 0.8294185847621298,
+      "learning_rate": 4.9996446810547464e-06,
+      "loss": 2.4286,
+      "step": 28
+    },
+    {
+      "epoch": 0.19863013698630136,
+      "grad_norm": 0.829981358086145,
+      "learning_rate": 4.999368333513354e-06,
+      "loss": 2.3941,
+      "step": 29
+    },
+    {
+      "epoch": 0.2054794520547945,
+      "grad_norm": 0.8259390860771237,
+      "learning_rate": 4.99901304449463e-06,
+      "loss": 2.2873,
+      "step": 30
+    },
+    {
+      "epoch": 0.21232876712328766,
+      "grad_norm": 0.7780493406506134,
+      "learning_rate": 4.998578825220228e-06,
+      "loss": 2.4414,
+      "step": 31
+    },
+    {
+      "epoch": 0.2191780821917808,
+      "grad_norm": 0.8139739891431346,
+      "learning_rate": 4.9980656894047776e-06,
+      "loss": 2.426,
+      "step": 32
+    },
+    {
+      "epoch": 0.22602739726027396,
+      "grad_norm": 0.7967892569249858,
+      "learning_rate": 4.9974736532554525e-06,
+      "loss": 2.436,
+      "step": 33
+    },
+    {
+      "epoch": 0.2328767123287671,
+      "grad_norm": 0.7428657720222388,
+      "learning_rate": 4.996802735471461e-06,
+      "loss": 2.3185,
+      "step": 34
+    },
+    {
+      "epoch": 0.23972602739726026,
+      "grad_norm": 0.7499845773485214,
+      "learning_rate": 4.996052957243455e-06,
+      "loss": 2.4402,
+      "step": 35
+    },
+    {
+      "epoch": 0.2465753424657534,
+      "grad_norm": 0.7359931004513611,
+      "learning_rate": 4.995224342252856e-06,
+      "loss": 2.3618,
+      "step": 36
+    },
+    {
+      "epoch": 0.2534246575342466,
+      "grad_norm": 0.7073950294074208,
+      "learning_rate": 4.994316916671115e-06,
+      "loss": 2.455,
+      "step": 37
+    },
+    {
+      "epoch": 0.2534246575342466,
+      "eval_loss": 2.4449844360351562,
+      "eval_runtime": 5.9369,
+      "eval_samples_per_second": 52.553,
+      "eval_steps_per_second": 3.369,
+      "step": 37
+    },
+    {
+      "epoch": 0.2602739726027397,
+      "grad_norm": 0.6796375355430841,
+      "learning_rate": 4.993330709158879e-06,
+      "loss": 2.4065,
+      "step": 38
+    },
+    {
+      "epoch": 0.2671232876712329,
+      "grad_norm": 0.7250510771173403,
+      "learning_rate": 4.992265750865091e-06,
+      "loss": 2.4527,
+      "step": 39
+    },
+    {
+      "epoch": 0.273972602739726,
+      "grad_norm": 0.739133472992406,
+      "learning_rate": 4.991122075426001e-06,
+      "loss": 2.4787,
+      "step": 40
+    },
+    {
+      "epoch": 0.2808219178082192,
+      "grad_norm": 0.7026481518003141,
+      "learning_rate": 4.989899718964108e-06,
+      "loss": 2.4464,
+      "step": 41
+    },
+    {
+      "epoch": 0.2876712328767123,
+      "grad_norm": 0.7034188124515799,
+      "learning_rate": 4.988598720087015e-06,
+      "loss": 2.4366,
+      "step": 42
+    },
+    {
+      "epoch": 0.2945205479452055,
+      "grad_norm": 0.7114177997700272,
+      "learning_rate": 4.9872191198862166e-06,
+      "loss": 2.3954,
+      "step": 43
+    },
+    {
+      "epoch": 0.3013698630136986,
+      "grad_norm": 0.7489678406861063,
+      "learning_rate": 4.985760961935791e-06,
+      "loss": 2.4092,
+      "step": 44
+    },
+    {
+      "epoch": 0.3082191780821918,
+      "grad_norm": 0.721624023056601,
+      "learning_rate": 4.984224292291035e-06,
+      "loss": 2.395,
+      "step": 45
+    },
+    {
+      "epoch": 0.3150684931506849,
+      "grad_norm": 0.6976242893548902,
+      "learning_rate": 4.982609159486998e-06,
+      "loss": 2.3621,
+      "step": 46
+    },
+    {
+      "epoch": 0.3219178082191781,
+      "grad_norm": 0.6979047705556364,
+      "learning_rate": 4.980915614536957e-06,
+      "loss": 2.3956,
+      "step": 47
+    },
+    {
+      "epoch": 0.3287671232876712,
+      "grad_norm": 0.6638188971561942,
+      "learning_rate": 4.979143710930805e-06,
+      "loss": 2.4447,
+      "step": 48
+    },
+    {
+      "epoch": 0.3356164383561644,
+      "grad_norm": 0.6550861221813643,
+      "learning_rate": 4.977293504633357e-06,
+      "loss": 2.3803,
+      "step": 49
+    },
+    {
+      "epoch": 0.3424657534246575,
+      "grad_norm": 0.6637160251936499,
+      "learning_rate": 4.975365054082586e-06,
+      "loss": 2.5123,
+      "step": 50
+    },
+    {
+      "epoch": 0.3493150684931507,
+      "grad_norm": 0.6657032864997552,
+      "learning_rate": 4.973358420187776e-06,
+      "loss": 2.4373,
+      "step": 51
+    },
+    {
+      "epoch": 0.3561643835616438,
+      "grad_norm": 0.6671449801461397,
+      "learning_rate": 4.971273666327598e-06,
+      "loss": 2.4515,
+      "step": 52
+    },
+    {
+      "epoch": 0.363013698630137,
+      "grad_norm": 0.6574174542409758,
+      "learning_rate": 4.969110858348108e-06,
+      "loss": 2.3797,
+      "step": 53
+    },
+    {
+      "epoch": 0.3698630136986301,
+      "grad_norm": 0.7037780907575699,
+      "learning_rate": 4.96687006456067e-06,
+      "loss": 2.4465,
+      "step": 54
+    },
+    {
+      "epoch": 0.3767123287671233,
+      "grad_norm": 0.7244746896553337,
+      "learning_rate": 4.964551355739796e-06,
+      "loss": 2.4265,
+      "step": 55
+    },
+    {
+      "epoch": 0.3835616438356164,
+      "grad_norm": 0.6581463626850778,
+      "learning_rate": 4.962154805120908e-06,
+      "loss": 2.4168,
+      "step": 56
+    },
+    {
+      "epoch": 0.3904109589041096,
+      "grad_norm": 0.6289298559777078,
+      "learning_rate": 4.959680488398031e-06,
+      "loss": 2.3328,
+      "step": 57
+    },
+    {
+      "epoch": 0.3972602739726027,
+      "grad_norm": 0.724536921071452,
+      "learning_rate": 4.957128483721398e-06,
+      "loss": 2.4319,
+      "step": 58
+    },
+    {
+      "epoch": 0.4041095890410959,
+      "grad_norm": 0.7119106876660016,
+      "learning_rate": 4.9544988716949825e-06,
+      "loss": 2.4339,
+      "step": 59
+    },
+    {
+      "epoch": 0.410958904109589,
+      "grad_norm": 0.6561228716087687,
+      "learning_rate": 4.951791735373953e-06,
+      "loss": 2.4057,
+      "step": 60
+    },
+    {
+      "epoch": 0.4178082191780822,
+      "grad_norm": 0.7138586667149742,
+      "learning_rate": 4.949007160262049e-06,
+      "loss": 2.3573,
+      "step": 61
+    },
+    {
+      "epoch": 0.4246575342465753,
+      "grad_norm": 0.6884519352750037,
+      "learning_rate": 4.946145234308884e-06,
+      "loss": 2.3747,
+      "step": 62
+    },
+    {
+      "epoch": 0.4315068493150685,
+      "grad_norm": 0.7403756155139569,
+      "learning_rate": 4.943206047907159e-06,
+      "loss": 2.2729,
+      "step": 63
+    },
+    {
+      "epoch": 0.4383561643835616,
+      "grad_norm": 0.6656934708441304,
+      "learning_rate": 4.940189693889819e-06,
+      "loss": 2.3805,
+      "step": 64
+    },
+    {
+      "epoch": 0.4452054794520548,
+      "grad_norm": 0.7251476112884715,
+      "learning_rate": 4.937096267527111e-06,
+      "loss": 2.3104,
+      "step": 65
+    },
+    {
+      "epoch": 0.4520547945205479,
+      "grad_norm": 0.6921252166237588,
+      "learning_rate": 4.9339258665235815e-06,
+      "loss": 2.3499,
+      "step": 66
+    },
+    {
+      "epoch": 0.4589041095890411,
+      "grad_norm": 0.685877730442195,
+      "learning_rate": 4.930678591014986e-06,
+      "loss": 2.4331,
+      "step": 67
+    },
+    {
+      "epoch": 0.4657534246575342,
+      "grad_norm": 0.6641111783418014,
+      "learning_rate": 4.927354543565131e-06,
+      "loss": 2.3789,
+      "step": 68
+    },
+    {
+      "epoch": 0.4726027397260274,
+      "grad_norm": 0.6657747916085803,
+      "learning_rate": 4.9239538291626285e-06,
+      "loss": 2.4542,
+      "step": 69
+    },
+    {
+      "epoch": 0.4794520547945205,
+      "grad_norm": 0.6544468951826022,
+      "learning_rate": 4.920476555217586e-06,
+      "loss": 2.4277,
+      "step": 70
+    },
+    {
+      "epoch": 0.4863013698630137,
+      "grad_norm": 0.6670629357129911,
+      "learning_rate": 4.91692283155821e-06,
+      "loss": 2.3881,
+      "step": 71
+    },
+    {
+      "epoch": 0.4931506849315068,
+      "grad_norm": 0.6559422824562596,
+      "learning_rate": 4.913292770427338e-06,
+      "loss": 2.4225,
+      "step": 72
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.6786459570539399,
+      "learning_rate": 4.909586486478897e-06,
+      "loss": 2.4963,
+      "step": 73
+    },
+    {
+      "epoch": 0.5068493150684932,
+      "grad_norm": 0.6892419553245916,
+      "learning_rate": 4.905804096774274e-06,
+      "loss": 2.4115,
+      "step": 74
+    },
+    {
+      "epoch": 0.5068493150684932,
+      "eval_loss": 2.4323267936706543,
+      "eval_runtime": 6.0665,
+      "eval_samples_per_second": 51.43,
+      "eval_steps_per_second": 3.297,
+      "step": 74
+    },
+    {
+      "epoch": 0.5136986301369864,
+      "grad_norm": 0.6768758807890537,
+      "learning_rate": 4.901945720778627e-06,
+      "loss": 2.3938,
+      "step": 75
+    },
+    {
+      "epoch": 0.5205479452054794,
+      "grad_norm": 0.7339799079732157,
+      "learning_rate": 4.898011480357109e-06,
+      "loss": 2.2883,
+      "step": 76
+    },
+    {
+      "epoch": 0.5273972602739726,
+      "grad_norm": 0.7397662592480156,
+      "learning_rate": 4.894001499771015e-06,
+      "loss": 2.4598,
+      "step": 77
+    },
+    {
+      "epoch": 0.5342465753424658,
+      "grad_norm": 0.6819769884160725,
+      "learning_rate": 4.889915905673865e-06,
+      "loss": 2.4188,
+      "step": 78
+    },
+    {
+      "epoch": 0.541095890410959,
+      "grad_norm": 0.6772539201070739,
+      "learning_rate": 4.885754827107395e-06,
+      "loss": 2.4142,
+      "step": 79
+    },
+    {
+      "epoch": 0.547945205479452,
+      "grad_norm": 0.659763812226515,
+      "learning_rate": 4.88151839549749e-06,
+      "loss": 2.298,
+      "step": 80
+    },
+    {
+      "epoch": 0.5547945205479452,
+      "grad_norm": 0.7286093280405468,
+      "learning_rate": 4.877206744650024e-06,
+      "loss": 2.3898,
+      "step": 81
+    },
+    {
+      "epoch": 0.5616438356164384,
+      "grad_norm": 0.6986932306550346,
+      "learning_rate": 4.8728200107466415e-06,
+      "loss": 2.3598,
+      "step": 82
+    },
+    {
+      "epoch": 0.5684931506849316,
+      "grad_norm": 0.6842367412456468,
+      "learning_rate": 4.868358332340451e-06,
+      "loss": 2.3763,
+      "step": 83
+    },
+    {
+      "epoch": 0.5753424657534246,
+      "grad_norm": 0.6724304892107736,
+      "learning_rate": 4.863821850351655e-06,
+      "loss": 2.4028,
+      "step": 84
+    },
+    {
+      "epoch": 0.5821917808219178,
+      "grad_norm": 0.6674281958373836,
+      "learning_rate": 4.859210708063091e-06,
+      "loss": 2.4226,
+      "step": 85
+    },
+    {
+      "epoch": 0.589041095890411,
+      "grad_norm": 0.6761850508539816,
+      "learning_rate": 4.854525051115711e-06,
+      "loss": 2.2755,
+      "step": 86
+    },
+    {
+      "epoch": 0.5958904109589042,
+      "grad_norm": 0.7049437160925219,
+      "learning_rate": 4.8497650275039795e-06,
+      "loss": 2.3633,
+      "step": 87
+    },
+    {
+      "epoch": 0.6027397260273972,
+      "grad_norm": 0.6675473282240035,
+      "learning_rate": 4.844930787571204e-06,
+      "loss": 2.4161,
+      "step": 88
+    },
+    {
+      "epoch": 0.6095890410958904,
+      "grad_norm": 0.6893380001896178,
+      "learning_rate": 4.84002248400478e-06,
+      "loss": 2.3261,
+      "step": 89
+    },
+    {
+      "epoch": 0.6164383561643836,
+      "grad_norm": 0.6557076444481973,
+      "learning_rate": 4.835040271831371e-06,
+      "loss": 2.4113,
+      "step": 90
+    },
+    {
+      "epoch": 0.6232876712328768,
+      "grad_norm": 0.7106326693649547,
+      "learning_rate": 4.829984308412011e-06,
+      "loss": 2.3468,
+      "step": 91
+    },
+    {
+      "epoch": 0.6301369863013698,
+      "grad_norm": 0.6780886154634945,
+      "learning_rate": 4.82485475343714e-06,
+      "loss": 2.4227,
+      "step": 92
+    },
+    {
+      "epoch": 0.636986301369863,
+      "grad_norm": 0.6848177542457299,
+      "learning_rate": 4.819651768921552e-06,
+      "loss": 2.4455,
+      "step": 93
+    },
+    {
+      "epoch": 0.6438356164383562,
+      "grad_norm": 0.6537386212526299,
+      "learning_rate": 4.814375519199281e-06,
+      "loss": 2.359,
+      "step": 94
+    },
+    {
+      "epoch": 0.6506849315068494,
+      "grad_norm": 0.692282068736651,
+      "learning_rate": 4.809026170918414e-06,
+      "loss": 2.4523,
+      "step": 95
+    },
+    {
+      "epoch": 0.6575342465753424,
+      "grad_norm": 0.6623529936148479,
+      "learning_rate": 4.803603893035822e-06,
+      "loss": 2.4321,
+      "step": 96
+    },
+    {
+      "epoch": 0.6643835616438356,
+      "grad_norm": 0.6842008308899095,
+      "learning_rate": 4.798108856811828e-06,
+      "loss": 2.415,
+      "step": 97
+    },
+    {
+      "epoch": 0.6712328767123288,
+      "grad_norm": 0.6761407885852002,
+      "learning_rate": 4.7925412358047965e-06,
+      "loss": 2.4099,
+      "step": 98
+    },
+    {
+      "epoch": 0.678082191780822,
+      "grad_norm": 0.695106737467069,
+      "learning_rate": 4.786901205865647e-06,
+      "loss": 2.3944,
+      "step": 99
+    },
+    {
+      "epoch": 0.684931506849315,
+      "grad_norm": 0.7154498681645167,
+      "learning_rate": 4.781188945132311e-06,
+      "loss": 2.4261,
+      "step": 100
+    },
+    {
+      "epoch": 0.6917808219178082,
+      "grad_norm": 0.708416595063437,
+      "learning_rate": 4.775404634024093e-06,
+      "loss": 2.4236,
+      "step": 101
+    },
+    {
+      "epoch": 0.6986301369863014,
+      "grad_norm": 0.6561586213988884,
+      "learning_rate": 4.769548455235979e-06,
+      "loss": 2.2712,
+      "step": 102
+    },
+    {
+      "epoch": 0.7054794520547946,
+      "grad_norm": 0.710907523788328,
+      "learning_rate": 4.763620593732867e-06,
+      "loss": 2.3401,
+      "step": 103
+    },
+    {
+      "epoch": 0.7123287671232876,
+      "grad_norm": 0.6765968692107874,
+      "learning_rate": 4.75762123674372e-06,
+      "loss": 2.4381,
+      "step": 104
+    },
+    {
+      "epoch": 0.7191780821917808,
+      "grad_norm": 0.9946667705374828,
+      "learning_rate": 4.751550573755658e-06,
+      "loss": 2.4825,
+      "step": 105
+    },
+    {
+      "epoch": 0.726027397260274,
+      "grad_norm": 0.6808703038492471,
+      "learning_rate": 4.745408796507968e-06,
+      "loss": 2.4417,
+      "step": 106
+    },
+    {
+      "epoch": 0.7328767123287672,
+      "grad_norm": 0.6511402815615133,
+      "learning_rate": 4.73919609898605e-06,
+      "loss": 2.412,
+      "step": 107
+    },
+    {
+      "epoch": 0.7397260273972602,
+      "grad_norm": 0.6806168774586489,
+      "learning_rate": 4.7329126774152945e-06,
+      "loss": 2.4047,
+      "step": 108
+    },
+    {
+      "epoch": 0.7465753424657534,
+      "grad_norm": 0.6581235881106364,
+      "learning_rate": 4.726558730254876e-06,
+      "loss": 2.3624,
+      "step": 109
+    },
+    {
+      "epoch": 0.7534246575342466,
+      "grad_norm": 0.6858279965463718,
+      "learning_rate": 4.720134458191494e-06,
+      "loss": 2.4046,
+      "step": 110
+    },
+    {
+      "epoch": 0.7602739726027398,
+      "grad_norm": 0.7044328584380498,
+      "learning_rate": 4.7136400641330245e-06,
+      "loss": 2.3298,
+      "step": 111
+    },
+    {
+      "epoch": 0.7602739726027398,
+      "eval_loss": 2.422318935394287,
+      "eval_runtime": 6.1327,
+      "eval_samples_per_second": 50.874,
+      "eval_steps_per_second": 3.261,
+      "step": 111
+    },
+    {
+      "epoch": 0.7671232876712328,
+      "grad_norm": 0.6949341182078216,
+      "learning_rate": 4.707075753202123e-06,
+      "loss": 2.4219,
+      "step": 112
+    },
+    {
+      "epoch": 0.773972602739726,
+      "grad_norm": 0.6883254066631505,
+      "learning_rate": 4.700441732729733e-06,
+      "loss": 2.4841,
+      "step": 113
+    },
+    {
+      "epoch": 0.7808219178082192,
+      "grad_norm": 0.688766337379056,
+      "learning_rate": 4.693738212248549e-06,
+      "loss": 2.3722,
+      "step": 114
+    },
+    {
+      "epoch": 0.7876712328767124,
+      "grad_norm": 0.6887580535922757,
+      "learning_rate": 4.68696540348639e-06,
+      "loss": 2.3637,
+      "step": 115
+    },
+    {
+      "epoch": 0.7945205479452054,
+      "grad_norm": 0.6719005239018524,
+      "learning_rate": 4.68012352035952e-06,
+      "loss": 2.4465,
+      "step": 116
+    },
+    {
+      "epoch": 0.8013698630136986,
+      "grad_norm": 0.6327678432767432,
+      "learning_rate": 4.673212778965881e-06,
+      "loss": 2.4219,
+      "step": 117
+    },
+    {
+      "epoch": 0.8082191780821918,
+      "grad_norm": 0.6572338268581427,
+      "learning_rate": 4.66623339757828e-06,
+      "loss": 2.4384,
+      "step": 118
+    },
+    {
+      "epoch": 0.815068493150685,
+      "grad_norm": 0.6774201762455547,
+      "learning_rate": 4.659185596637485e-06,
+      "loss": 2.4289,
+      "step": 119
+    },
+    {
+      "epoch": 0.821917808219178,
+      "grad_norm": 0.6765522988548955,
+      "learning_rate": 4.652069598745267e-06,
+      "loss": 2.3849,
+      "step": 120
+    },
+    {
+      "epoch": 0.8287671232876712,
+      "grad_norm": 0.6553606050310139,
+      "learning_rate": 4.644885628657369e-06,
+      "loss": 2.3953,
+      "step": 121
+    },
+    {
+      "epoch": 0.8356164383561644,
+      "grad_norm": 0.6817645269442179,
+      "learning_rate": 4.637633913276406e-06,
+      "loss": 2.4534,
+      "step": 122
+    },
+    {
+      "epoch": 0.8424657534246576,
+      "grad_norm": 0.6842353980823837,
+      "learning_rate": 4.630314681644701e-06,
+      "loss": 2.3376,
+      "step": 123
+    },
+    {
+      "epoch": 0.8493150684931506,
+      "grad_norm": 0.6938393939968492,
+      "learning_rate": 4.622928164937046e-06,
+      "loss": 2.3795,
+      "step": 124
+    },
+    {
+      "epoch": 0.8561643835616438,
+      "grad_norm": 0.6503903813017006,
+      "learning_rate": 4.615474596453406e-06,
+      "loss": 2.3801,
+      "step": 125
+    },
+    {
+      "epoch": 0.863013698630137,
+      "grad_norm": 0.6690065166974256,
+      "learning_rate": 4.607954211611543e-06,
+      "loss": 2.4771,
+      "step": 126
+    },
+    {
+      "epoch": 0.8698630136986302,
+      "grad_norm": 0.6565369111054283,
+      "learning_rate": 4.600367247939592e-06,
+      "loss": 2.4468,
+      "step": 127
+    },
+    {
+      "epoch": 0.8767123287671232,
+      "grad_norm": 0.6590600861747526,
+      "learning_rate": 4.5927139450685455e-06,
+      "loss": 2.2973,
+      "step": 128
+    },
+    {
+      "epoch": 0.8835616438356164,
+      "grad_norm": 0.6642768664806619,
+      "learning_rate": 4.584994544724695e-06,
+      "loss": 2.4463,
+      "step": 129
+    },
+    {
+      "epoch": 0.8904109589041096,
+      "grad_norm": 0.7254289901639452,
+      "learning_rate": 4.577209290721991e-06,
+      "loss": 2.2925,
+      "step": 130
+    },
+    {
+      "epoch": 0.8972602739726028,
+      "grad_norm": 0.6264601593599859,
+      "learning_rate": 4.569358428954343e-06,
+      "loss": 2.3624,
+      "step": 131
+    },
+    {
+      "epoch": 0.9041095890410958,
+      "grad_norm": 0.6932555262208767,
+      "learning_rate": 4.561442207387854e-06,
+      "loss": 2.3315,
+      "step": 132
+    },
+    {
+      "epoch": 0.910958904109589,
+      "grad_norm": 0.6857220601161248,
+      "learning_rate": 4.55346087605299e-06,
+      "loss": 2.372,
+      "step": 133
+    },
+    {
+      "epoch": 0.9178082191780822,
+      "grad_norm": 0.701001912101975,
+      "learning_rate": 4.5454146870366775e-06,
+      "loss": 2.4026,
+      "step": 134
+    },
+    {
+      "epoch": 0.9246575342465754,
+      "grad_norm": 0.6340272298295848,
+      "learning_rate": 4.537303894474349e-06,
+      "loss": 2.4131,
+      "step": 135
+    },
+    {
+      "epoch": 0.9315068493150684,
+      "grad_norm": 0.6136778043717291,
+      "learning_rate": 4.529128754541909e-06,
+      "loss": 2.297,
+      "step": 136
+    },
+    {
+      "epoch": 0.9383561643835616,
+      "grad_norm": 0.6322021005066626,
+      "learning_rate": 4.5208895254476495e-06,
+      "loss": 2.4084,
+      "step": 137
+    },
+    {
+      "epoch": 0.9452054794520548,
+      "grad_norm": 0.6445887057946764,
+      "learning_rate": 4.512586467424087e-06,
+      "loss": 2.3645,
+      "step": 138
+    },
+    {
+      "epoch": 0.952054794520548,
+      "grad_norm": 0.7062860383104262,
+      "learning_rate": 4.504219842719752e-06,
+      "loss": 2.3655,
+      "step": 139
+    },
+    {
+      "epoch": 0.958904109589041,
+      "grad_norm": 0.6781212322165483,
+      "learning_rate": 4.4957899155908954e-06,
+      "loss": 2.4285,
+      "step": 140
+    },
+    {
+      "epoch": 0.9657534246575342,
+      "grad_norm": 0.6678298874200482,
+      "learning_rate": 4.487296952293156e-06,
+      "loss": 2.38,
+      "step": 141
+    },
+    {
+      "epoch": 0.9726027397260274,
+      "grad_norm": 0.6575931921755482,
+      "learning_rate": 4.478741221073136e-06,
+      "loss": 2.4094,
+      "step": 142
+    },
+    {
+      "epoch": 0.9794520547945206,
+      "grad_norm": 0.6868011511324458,
+      "learning_rate": 4.470122992159938e-06,
+      "loss": 2.4096,
+      "step": 143
+    },
+    {
+      "epoch": 0.9863013698630136,
+      "grad_norm": 0.6854773154104912,
+      "learning_rate": 4.461442537756629e-06,
+      "loss": 2.4249,
+      "step": 144
+    },
+    {
+      "epoch": 0.9931506849315068,
+      "grad_norm": 0.708510432833667,
+      "learning_rate": 4.452700132031639e-06,
+      "loss": 2.3724,
+      "step": 145
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.6566297487438831,
+      "learning_rate": 4.443896051110105e-06,
+      "loss": 2.391,
+      "step": 146
+    },
+    {
+      "epoch": 1.0068493150684932,
+      "grad_norm": 0.7629369425964684,
+      "learning_rate": 4.435030573065148e-06,
+      "loss": 2.3448,
+      "step": 147
+    },
+    {
+      "epoch": 1.0136986301369864,
+      "grad_norm": 0.7384812144528767,
+      "learning_rate": 4.426103977909094e-06,
+      "loss": 2.322,
+      "step": 148
+    },
+    {
+      "epoch": 1.0136986301369864,
+      "eval_loss": 2.417771339416504,
+      "eval_runtime": 53.3919,
+      "eval_samples_per_second": 5.844,
+      "eval_steps_per_second": 0.375,
+      "step": 148
+    },
+    {
+      "epoch": 1.0205479452054795,
+      "grad_norm": 0.6662500395219653,
+      "learning_rate": 4.417116547584621e-06,
+      "loss": 2.35,
+      "step": 149
+    },
+    {
+      "epoch": 1.0273972602739727,
+      "grad_norm": 0.6789888617564587,
+      "learning_rate": 4.408068565955864e-06,
+      "loss": 2.3845,
+      "step": 150
+    },
+    {
+      "epoch": 1.0342465753424657,
+      "grad_norm": 0.6998134909721523,
+      "learning_rate": 4.398960318799446e-06,
+      "loss": 2.35,
+      "step": 151
+    },
+    {
+      "epoch": 1.0410958904109588,
+      "grad_norm": 0.7216093013270933,
+      "learning_rate": 4.389792093795444e-06,
+      "loss": 2.3223,
+      "step": 152
+    },
+    {
+      "epoch": 1.047945205479452,
+      "grad_norm": 0.7403493735812074,
+      "learning_rate": 4.380564180518318e-06,
+      "loss": 2.3253,
+      "step": 153
+    },
+    {
+      "epoch": 1.0547945205479452,
+      "grad_norm": 0.7218865625666383,
+      "learning_rate": 4.3712768704277535e-06,
+      "loss": 2.3106,
+      "step": 154
+    },
+    {
+      "epoch": 1.0616438356164384,
+      "grad_norm": 0.7197974746780744,
+      "learning_rate": 4.361930456859455e-06,
+      "loss": 2.3033,
+      "step": 155
+    },
+    {
+      "epoch": 1.0684931506849316,
+      "grad_norm": 0.7249006057182749,
+      "learning_rate": 4.35252523501589e-06,
+      "loss": 2.3888,
+      "step": 156
+    },
+    {
+      "epoch": 1.0753424657534247,
+      "grad_norm": 0.7037919961575326,
+      "learning_rate": 4.343061501956959e-06,
+      "loss": 2.3402,
+      "step": 157
+    },
+    {
+      "epoch": 1.0821917808219177,
+      "grad_norm": 0.6996471940866854,
+      "learning_rate": 4.3335395565906115e-06,
+      "loss": 2.3194,
+      "step": 158
+    },
+    {
+      "epoch": 1.0890410958904109,
+      "grad_norm": 0.717826362193492,
+      "learning_rate": 4.323959699663412e-06,
+      "loss": 2.3852,
+      "step": 159
+    },
+    {
+      "epoch": 1.095890410958904,
+      "grad_norm": 0.708765230517044,
+      "learning_rate": 4.314322233751034e-06,
+      "loss": 2.3206,
+      "step": 160
+    },
+    {
+      "epoch": 1.1027397260273972,
+      "grad_norm": 0.6776217729604187,
+      "learning_rate": 4.304627463248706e-06,
+      "loss": 2.3081,
+      "step": 161
+    },
+    {
+      "epoch": 1.1095890410958904,
+      "grad_norm": 0.6853081994090057,
+      "learning_rate": 4.294875694361599e-06,
+      "loss": 2.3176,
+      "step": 162
+    },
+    {
+      "epoch": 1.1164383561643836,
+      "grad_norm": 0.6830768891525606,
+      "learning_rate": 4.285067235095152e-06,
+      "loss": 2.3364,
+      "step": 163
+    },
+    {
+      "epoch": 1.1232876712328768,
+      "grad_norm": 0.7089445206356011,
+      "learning_rate": 4.275202395245346e-06,
+      "loss": 2.3418,
+      "step": 164
+    },
+    {
+      "epoch": 1.13013698630137,
+      "grad_norm": 0.7823709313759815,
+      "learning_rate": 4.26528148638892e-06,
+      "loss": 2.3798,
+      "step": 165
+    },
+    {
+      "epoch": 1.1369863013698631,
+      "grad_norm": 0.6709932434280309,
+      "learning_rate": 4.255304821873526e-06,
+      "loss": 2.2267,
+      "step": 166
+    },
+    {
+      "epoch": 1.143835616438356,
+      "grad_norm": 0.6628107638339615,
+      "learning_rate": 4.245272716807834e-06,
+      "loss": 2.3308,
+      "step": 167
+    },
+    {
+      "epoch": 1.1506849315068493,
+      "grad_norm": 0.7184114487855815,
+      "learning_rate": 4.2351854880515856e-06,
+      "loss": 2.3017,
+      "step": 168
+    },
+    {
+      "epoch": 1.1575342465753424,
+      "grad_norm": 0.7166834764618115,
+      "learning_rate": 4.225043454205573e-06,
+      "loss": 2.3551,
+      "step": 169
+    },
+    {
+      "epoch": 1.1643835616438356,
+      "grad_norm": 0.7234116110718222,
+      "learning_rate": 4.2148469356015895e-06,
+      "loss": 2.3205,
+      "step": 170
+    },
+    {
+      "epoch": 1.1712328767123288,
+      "grad_norm": 0.6873414003582445,
+      "learning_rate": 4.204596254292303e-06,
+      "loss": 2.3675,
+      "step": 171
+    },
+    {
+      "epoch": 1.178082191780822,
+      "grad_norm": 0.7161572137012578,
+      "learning_rate": 4.194291734041088e-06,
+      "loss": 2.2708,
+      "step": 172
+    },
+    {
+      "epoch": 1.1849315068493151,
+      "grad_norm": 0.7142712637206968,
+      "learning_rate": 4.183933700311801e-06,
+      "loss": 2.2617,
+      "step": 173
+    },
+    {
+      "epoch": 1.191780821917808,
+      "grad_norm": 0.698584695168926,
+      "learning_rate": 4.173522480258494e-06,
+      "loss": 2.3686,
+      "step": 174
+    },
+    {
+      "epoch": 1.1986301369863013,
+      "grad_norm": 0.7275575279831493,
+      "learning_rate": 4.163058402715092e-06,
+      "loss": 2.3088,
+      "step": 175
+    },
+    {
+      "epoch": 1.2054794520547945,
+      "grad_norm": 0.7066892597231554,
+      "learning_rate": 4.152541798184995e-06,
+      "loss": 2.3273,
+      "step": 176
+    },
+    {
+      "epoch": 1.2123287671232876,
+      "grad_norm": 0.6692377666271583,
+      "learning_rate": 4.141972998830651e-06,
+      "loss": 2.368,
+      "step": 177
+    },
+    {
+      "epoch": 1.2191780821917808,
+      "grad_norm": 0.7519738253706634,
+      "learning_rate": 4.1313523384630565e-06,
+      "loss": 2.2878,
+      "step": 178
+    },
+    {
+      "epoch": 1.226027397260274,
+      "grad_norm": 0.7029718569317863,
+      "learning_rate": 4.120680152531214e-06,
+      "loss": 2.2833,
+      "step": 179
+    },
+    {
+      "epoch": 1.2328767123287672,
+      "grad_norm": 0.7334026090190444,
+      "learning_rate": 4.109956778111544e-06,
+      "loss": 2.3322,
+      "step": 180
+    },
+    {
+      "epoch": 1.2397260273972603,
+      "grad_norm": 0.6951014303600972,
+      "learning_rate": 4.099182553897228e-06,
+      "loss": 2.2781,
+      "step": 181
+    },
+    {
+      "epoch": 1.2465753424657535,
+      "grad_norm": 0.6885637550568241,
+      "learning_rate": 4.088357820187521e-06,
+      "loss": 2.2814,
+      "step": 182
+    },
+    {
+      "epoch": 1.2534246575342465,
+      "grad_norm": 0.6917922559068629,
+      "learning_rate": 4.077482918876995e-06,
+      "loss": 2.378,
+      "step": 183
+    },
+    {
+      "epoch": 1.2602739726027397,
+      "grad_norm": 0.6912354421485389,
+      "learning_rate": 4.066558193444746e-06,
+      "loss": 2.3087,
+      "step": 184
+    },
+    {
+      "epoch": 1.2671232876712328,
+      "grad_norm": 0.7423647896303166,
+      "learning_rate": 4.0555839889435444e-06,
+      "loss": 2.2661,
+      "step": 185
+    },
+    {
+      "epoch": 1.2671232876712328,
+      "eval_loss": 2.4178218841552734,
+      "eval_runtime": 5.9059,
+      "eval_samples_per_second": 52.828,
+      "eval_steps_per_second": 3.386,
+      "step": 185
+    },
+    {
+      "epoch": 1.273972602739726,
+      "grad_norm": 0.7478496634954042,
+      "learning_rate": 4.044560651988933e-06,
+      "loss": 2.3361,
+      "step": 186
+    },
+    {
+      "epoch": 1.2808219178082192,
+      "grad_norm": 0.7065399264251201,
+      "learning_rate": 4.033488530748285e-06,
+      "loss": 2.2644,
+      "step": 187
+    },
+    {
+      "epoch": 1.2876712328767124,
+      "grad_norm": 0.726480536340863,
+      "learning_rate": 4.022367974929803e-06,
+      "loss": 2.1882,
+      "step": 188
+    },
+    {
+      "epoch": 1.2945205479452055,
+      "grad_norm": 0.6827946001097096,
+      "learning_rate": 4.0111993357714755e-06,
+      "loss": 2.2887,
+      "step": 189
+    },
+    {
+      "epoch": 1.3013698630136985,
+      "grad_norm": 0.6814103107109724,
+      "learning_rate": 3.999982966029981e-06,
+      "loss": 2.3315,
+      "step": 190
+    },
+    {
+      "epoch": 1.308219178082192,
+      "grad_norm": 0.65846390651437,
+      "learning_rate": 3.98871921996955e-06,
+      "loss": 2.2595,
+      "step": 191
+    },
+    {
+      "epoch": 1.3150684931506849,
+      "grad_norm": 0.6664034822825207,
+      "learning_rate": 3.977408453350774e-06,
+      "loss": 2.3524,
+      "step": 192
+    },
+    {
+      "epoch": 1.321917808219178,
+      "grad_norm": 0.7146251642801748,
+      "learning_rate": 3.966051023419366e-06,
+      "loss": 2.3562,
+      "step": 193
+    },
+    {
+      "epoch": 1.3287671232876712,
+      "grad_norm": 0.7013378563213172,
+      "learning_rate": 3.9546472888948825e-06,
+      "loss": 2.2865,
+      "step": 194
+    },
+    {
+      "epoch": 1.3356164383561644,
+      "grad_norm": 0.6884022789569789,
+      "learning_rate": 3.943197609959389e-06,
+      "loss": 2.3293,
+      "step": 195
+    },
+    {
+      "epoch": 1.3424657534246576,
+      "grad_norm": 0.7263022758504937,
+      "learning_rate": 3.931702348246087e-06,
+      "loss": 2.2485,
+      "step": 196
+    },
+    {
+      "epoch": 1.3493150684931507,
+      "grad_norm": 0.6879494874948079,
+      "learning_rate": 3.92016186682789e-06,
+      "loss": 2.2652,
+      "step": 197
+    },
+    {
+      "epoch": 1.356164383561644,
+      "grad_norm": 0.7035221618072612,
+      "learning_rate": 3.9085765302059556e-06,
+      "loss": 2.3413,
+      "step": 198
+    },
+    {
+      "epoch": 1.3630136986301369,
+      "grad_norm": 0.7042222964511125,
+      "learning_rate": 3.8969467042981725e-06,
+      "loss": 2.3202,
+      "step": 199
+    },
+    {
+      "epoch": 1.36986301369863,
+      "grad_norm": 0.6961599428087603,
+      "learning_rate": 3.885272756427609e-06,
+      "loss": 2.2134,
+      "step": 200
+    },
+    {
+      "epoch": 1.3767123287671232,
+      "grad_norm": 0.7445990308351226,
+      "learning_rate": 3.873555055310902e-06,
+      "loss": 2.3083,
+      "step": 201
+    },
+    {
+      "epoch": 1.3835616438356164,
+      "grad_norm": 0.7272556327744341,
+      "learning_rate": 3.86179397104662e-06,
+      "loss": 2.2611,
+      "step": 202
+    },
+    {
+      "epoch": 1.3904109589041096,
+      "grad_norm": 0.6734004584059651,
+      "learning_rate": 3.849989875103566e-06,
+      "loss": 2.3396,
+      "step": 203
+    },
+    {
+      "epoch": 1.3972602739726028,
+      "grad_norm": 0.702149629138006,
+      "learning_rate": 3.83814314030905e-06,
+      "loss": 2.2206,
+      "step": 204
+    },
+    {
+      "epoch": 1.404109589041096,
+      "grad_norm": 0.7668660946577405,
+      "learning_rate": 3.826254140837111e-06,
+      "loss": 2.3745,
+      "step": 205
+    },
+    {
+      "epoch": 1.410958904109589,
+      "grad_norm": 0.6972264397617832,
+      "learning_rate": 3.8143232521967023e-06,
+      "loss": 2.2376,
+      "step": 206
+    },
+    {
+      "epoch": 1.4178082191780823,
+      "grad_norm": 0.6802719750908277,
+      "learning_rate": 3.802350851219826e-06,
+      "loss": 2.3348,
+      "step": 207
+    },
+    {
+      "epoch": 1.4246575342465753,
+      "grad_norm": 0.679939282001721,
+      "learning_rate": 3.7903373160496342e-06,
+      "loss": 2.3184,
+      "step": 208
+    },
+    {
+      "epoch": 1.4315068493150684,
+      "grad_norm": 0.708335225525305,
+      "learning_rate": 3.778283026128485e-06,
+      "loss": 2.2065,
+      "step": 209
+    },
+    {
+      "epoch": 1.4383561643835616,
+      "grad_norm": 0.6821337057973484,
+      "learning_rate": 3.7661883621859584e-06,
+      "loss": 2.3815,
+      "step": 210
+    },
+    {
+      "epoch": 1.4452054794520548,
+      "grad_norm": 0.6864170898125791,
+      "learning_rate": 3.754053706226829e-06,
+      "loss": 2.2901,
+      "step": 211
+    },
+    {
+      "epoch": 1.452054794520548,
+      "grad_norm": 0.690989555098226,
+      "learning_rate": 3.741879441519004e-06,
+      "loss": 2.3053,
+      "step": 212
+    },
+    {
+      "epoch": 1.4589041095890412,
+      "grad_norm": 0.7326752735961912,
+      "learning_rate": 3.729665952581415e-06,
+      "loss": 2.2915,
+      "step": 213
+    },
+    {
+      "epoch": 1.4657534246575343,
+      "grad_norm": 0.6795313294527453,
+      "learning_rate": 3.7174136251718735e-06,
+      "loss": 2.3548,
+      "step": 214
+    },
+    {
+      "epoch": 1.4726027397260273,
+      "grad_norm": 0.7026030269068815,
+      "learning_rate": 3.705122846274889e-06,
+      "loss": 2.339,
+      "step": 215
+    },
+    {
+      "epoch": 1.4794520547945205,
+      "grad_norm": 0.6968149329541087,
+      "learning_rate": 3.6927940040894427e-06,
+      "loss": 2.324,
+      "step": 216
+    },
+    {
+      "epoch": 1.4863013698630136,
+      "grad_norm": 0.681949385138759,
+      "learning_rate": 3.680427488016731e-06,
+      "loss": 2.2522,
+      "step": 217
+    },
+    {
+      "epoch": 1.4931506849315068,
+      "grad_norm": 0.7101393810809886,
+      "learning_rate": 3.6680236886478627e-06,
+      "loss": 2.3273,
+      "step": 218
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 0.681265285699898,
+      "learning_rate": 3.6555829977515213e-06,
+      "loss": 2.3807,
+      "step": 219
+    },
+    {
+      "epoch": 1.5068493150684932,
+      "grad_norm": 0.6714546516939787,
+      "learning_rate": 3.6431058082615966e-06,
+      "loss": 2.3566,
+      "step": 220
+    },
+    {
+      "epoch": 1.5136986301369864,
+      "grad_norm": 0.677995114165458,
+      "learning_rate": 3.6305925142647704e-06,
+      "loss": 2.3523,
+      "step": 221
+    },
+    {
+      "epoch": 1.5205479452054793,
+      "grad_norm": 0.6347599230305153,
+      "learning_rate": 3.6180435109880677e-06,
+      "loss": 2.2482,
+      "step": 222
+    },
+    {
+      "epoch": 1.5205479452054793,
+      "eval_loss": 2.4155406951904297,
+      "eval_runtime": 5.9256,
+      "eval_samples_per_second": 52.653,
+      "eval_steps_per_second": 3.375,
+      "step": 222
+    },
+    {
+      "epoch": 1.5273972602739727,
+      "grad_norm": 0.7076904051784869,
+      "learning_rate": 3.6054591947863782e-06,
+      "loss": 2.3036,
+      "step": 223
+    },
+    {
+      "epoch": 1.5342465753424657,
+      "grad_norm": 0.6907845815242056,
+      "learning_rate": 3.592839963129934e-06,
+      "loss": 2.3765,
+      "step": 224
+    },
+    {
+      "epoch": 1.541095890410959,
+      "grad_norm": 0.698669011142619,
+      "learning_rate": 3.5801862145917565e-06,
+      "loss": 2.3535,
+      "step": 225
+    },
+    {
+      "epoch": 1.547945205479452,
+      "grad_norm": 0.6660369882110091,
+      "learning_rate": 3.5674983488350695e-06,
+      "loss": 2.3089,
+      "step": 226
+    },
+    {
+      "epoch": 1.5547945205479452,
+      "grad_norm": 0.6870900476918825,
+      "learning_rate": 3.5547767666006735e-06,
+      "loss": 2.1687,
+      "step": 227
+    },
+    {
+      "epoch": 1.5616438356164384,
+      "grad_norm": 0.6973754014334647,
+      "learning_rate": 3.542021869694289e-06,
+      "loss": 2.3408,
+      "step": 228
+    },
+    {
+      "epoch": 1.5684931506849316,
+      "grad_norm": 0.6883965513486622,
+      "learning_rate": 3.5292340609738667e-06,
+      "loss": 2.2992,
+      "step": 229
+    },
+    {
+      "epoch": 1.5753424657534247,
+      "grad_norm": 0.6818556003782501,
+      "learning_rate": 3.516413744336863e-06,
+      "loss": 2.3865,
+      "step": 230
+    },
+    {
+      "epoch": 1.5821917808219177,
+      "grad_norm": 0.702979144019098,
+      "learning_rate": 3.503561324707484e-06,
+      "loss": 2.2278,
+      "step": 231
+    },
+    {
+      "epoch": 1.589041095890411,
+      "grad_norm": 0.6806404665338951,
+      "learning_rate": 3.4906772080238925e-06,
+      "loss": 2.3231,
+      "step": 232
+    },
+    {
+      "epoch": 1.595890410958904,
+      "grad_norm": 0.685606147526874,
+      "learning_rate": 3.47776180122539e-06,
+      "loss": 2.3128,
+      "step": 233
+    },
+    {
+      "epoch": 1.6027397260273972,
+      "grad_norm": 0.6332716110587268,
+      "learning_rate": 3.4648155122395653e-06,
+      "loss": 2.3087,
+      "step": 234
+    },
+    {
+      "epoch": 1.6095890410958904,
+      "grad_norm": 0.7202386902926047,
+      "learning_rate": 3.4518387499694038e-06,
+      "loss": 2.3599,
+      "step": 235
+    },
+    {
+      "epoch": 1.6164383561643836,
+      "grad_norm": 0.6832286968858357,
+      "learning_rate": 3.438831924280381e-06,
+      "loss": 2.335,
+      "step": 236
+    },
+    {
+      "epoch": 1.6232876712328768,
+      "grad_norm": 0.6520482069071943,
+      "learning_rate": 3.425795445987508e-06,
+      "loss": 2.3355,
+      "step": 237
+    },
+    {
+      "epoch": 1.6301369863013697,
+      "grad_norm": 0.7019446668563534,
+      "learning_rate": 3.4127297268423637e-06,
+      "loss": 2.3111,
+      "step": 238
+    },
+    {
+      "epoch": 1.6369863013698631,
+      "grad_norm": 0.677088641753582,
+      "learning_rate": 3.3996351795200865e-06,
+      "loss": 2.3463,
+      "step": 239
+    },
+    {
+      "epoch": 1.643835616438356,
+      "grad_norm": 0.7234982640691922,
+      "learning_rate": 3.386512217606339e-06,
+      "loss": 2.3195,
+      "step": 240
+    },
+    {
+      "epoch": 1.6506849315068495,
+      "grad_norm": 0.6944442709324709,
+      "learning_rate": 3.373361255584249e-06,
+      "loss": 2.4005,
+      "step": 241
+    },
+    {
+      "epoch": 1.6575342465753424,
+      "grad_norm": 0.7269476410749787,
+      "learning_rate": 3.3601827088213124e-06,
+      "loss": 2.2849,
+      "step": 242
+    },
+    {
+      "epoch": 1.6643835616438356,
+      "grad_norm": 0.6933396138061356,
+      "learning_rate": 3.3469769935562798e-06,
+      "loss": 2.3302,
+      "step": 243
+    },
+    {
+      "epoch": 1.6712328767123288,
+      "grad_norm": 0.6951788312862931,
+      "learning_rate": 3.3337445268860065e-06,
+      "loss": 2.3419,
+      "step": 244
+    },
+    {
+      "epoch": 1.678082191780822,
+      "grad_norm": 0.7102864990853207,
+      "learning_rate": 3.3204857267522782e-06,
+      "loss": 2.2889,
+      "step": 245
+    },
+    {
+      "epoch": 1.6849315068493151,
+      "grad_norm": 0.6734448441851548,
+      "learning_rate": 3.3072010119286156e-06,
+      "loss": 2.3191,
+      "step": 246
+    },
+    {
+      "epoch": 1.691780821917808,
+      "grad_norm": 0.710004987174813,
+      "learning_rate": 3.2938908020070403e-06,
+      "loss": 2.2627,
+      "step": 247
+    },
+    {
+      "epoch": 1.6986301369863015,
+      "grad_norm": 0.7011446050991217,
+      "learning_rate": 3.280555517384825e-06,
+      "loss": 2.3647,
+      "step": 248
+    },
+    {
+      "epoch": 1.7054794520547945,
+      "grad_norm": 0.7319541321421087,
+      "learning_rate": 3.267195579251219e-06,
+      "loss": 2.3177,
+      "step": 249
+    },
+    {
+      "epoch": 1.7123287671232876,
+      "grad_norm": 0.6626335956989253,
+      "learning_rate": 3.2538114095741412e-06,
+      "loss": 2.3775,
+      "step": 250
+    },
+    {
+      "epoch": 1.7191780821917808,
+      "grad_norm": 0.6662336469102511,
+      "learning_rate": 3.2404034310868524e-06,
+      "loss": 2.3887,
+      "step": 251
+    },
+    {
+      "epoch": 1.726027397260274,
+      "grad_norm": 0.6904872645618079,
+      "learning_rate": 3.2269720672746045e-06,
+      "loss": 2.2411,
+      "step": 252
+    },
+    {
+      "epoch": 1.7328767123287672,
+      "grad_norm": 0.6786858149868691,
+      "learning_rate": 3.2135177423612668e-06,
+      "loss": 2.3343,
+      "step": 253
+    },
+    {
+      "epoch": 1.7397260273972601,
+      "grad_norm": 0.8070210101589392,
+      "learning_rate": 3.200040881295922e-06,
+      "loss": 2.3579,
+      "step": 254
+    },
+    {
+      "epoch": 1.7465753424657535,
+      "grad_norm": 0.6859306383539353,
+      "learning_rate": 3.186541909739452e-06,
+      "loss": 2.371,
+      "step": 255
+    },
+    {
+      "epoch": 1.7534246575342465,
+      "grad_norm": 0.6853782148050659,
+      "learning_rate": 3.1730212540510835e-06,
+      "loss": 2.117,
+      "step": 256
+    },
+    {
+      "epoch": 1.7602739726027399,
+      "grad_norm": 0.9960196249104211,
+      "learning_rate": 3.1594793412749316e-06,
+      "loss": 2.3843,
+      "step": 257
+    },
+    {
+      "epoch": 1.7671232876712328,
+      "grad_norm": 0.6757685094130473,
+      "learning_rate": 3.145916599126506e-06,
+      "loss": 2.3745,
+      "step": 258
+    },
+    {
+      "epoch": 1.773972602739726,
+      "grad_norm": 0.7253543893242833,
+      "learning_rate": 3.132333455979202e-06,
+      "loss": 2.3707,
+      "step": 259
+    },
+    {
+      "epoch": 1.773972602739726,
+      "eval_loss": 2.4115374088287354,
+      "eval_runtime": 6.1962,
+      "eval_samples_per_second": 50.354,
+      "eval_steps_per_second": 3.228,
+      "step": 259
+    },
+    {
+      "epoch": 1.7808219178082192,
+      "grad_norm": 0.6634255128578609,
+      "learning_rate": 3.118730340850774e-06,
+      "loss": 2.2806,
+      "step": 260
+    },
+    {
+      "epoch": 1.7876712328767124,
+      "grad_norm": 0.6969726139618965,
+      "learning_rate": 3.1051076833897814e-06,
+      "loss": 2.3148,
+      "step": 261
+    },
+    {
+      "epoch": 1.7945205479452055,
+      "grad_norm": 0.7393377224615634,
+      "learning_rate": 3.091465913862019e-06,
+      "loss": 2.3359,
+      "step": 262
+    },
+    {
+      "epoch": 1.8013698630136985,
+      "grad_norm": 0.6900014881921968,
+      "learning_rate": 3.077805463136931e-06,
+      "loss": 2.2974,
+      "step": 263
+    },
+    {
+      "epoch": 1.808219178082192,
+      "grad_norm": 0.6902244001587043,
+      "learning_rate": 3.0641267626739946e-06,
+      "loss": 2.3219,
+      "step": 264
+    },
+    {
+      "epoch": 1.8150684931506849,
+      "grad_norm": 0.6810057389366492,
+      "learning_rate": 3.050430244509103e-06,
+      "loss": 2.3185,
+      "step": 265
+    },
+    {
+      "epoch": 1.821917808219178,
+      "grad_norm": 0.7557527768095192,
+      "learning_rate": 3.0367163412409084e-06,
+      "loss": 2.2613,
+      "step": 266
+    },
+    {
+      "epoch": 1.8287671232876712,
+      "grad_norm": 0.7085704364858588,
+      "learning_rate": 3.0229854860171666e-06,
+      "loss": 2.3049,
+      "step": 267
+    },
+    {
+      "epoch": 1.8356164383561644,
+      "grad_norm": 0.7044887748838028,
+      "learning_rate": 3.009238112521054e-06,
+      "loss": 2.3123,
+      "step": 268
+    },
+    {
+      "epoch": 1.8424657534246576,
+      "grad_norm": 0.6900781076774555,
+      "learning_rate": 2.9954746549574696e-06,
+      "loss": 2.2775,
+      "step": 269
+    },
+    {
+      "epoch": 1.8493150684931505,
+      "grad_norm": 0.71668864808382,
+      "learning_rate": 2.981695548039319e-06,
+      "loss": 2.3295,
+      "step": 270
+    },
+    {
+      "epoch": 1.856164383561644,
+      "grad_norm": 0.667887574033749,
+      "learning_rate": 2.967901226973787e-06,
+      "loss": 2.3287,
+      "step": 271
+    },
+    {
+      "epoch": 1.8630136986301369,
+      "grad_norm": 0.6819433212131332,
+      "learning_rate": 2.9540921274485913e-06,
+      "loss": 2.3693,
+      "step": 272
+    },
+    {
+      "epoch": 1.8698630136986303,
+      "grad_norm": 0.9698629158570198,
+      "learning_rate": 2.9402686856182204e-06,
+      "loss": 2.2706,
+      "step": 273
+    },
+    {
+      "epoch": 1.8767123287671232,
+      "grad_norm": 0.6898650487034594,
+      "learning_rate": 2.9264313380901586e-06,
+      "loss": 2.3726,
+      "step": 274
+    },
+    {
+      "epoch": 1.8835616438356164,
+      "grad_norm": 0.6770885016618345,
+      "learning_rate": 2.912580521911095e-06,
+      "loss": 2.3218,
+      "step": 275
+    },
+    {
+      "epoch": 1.8904109589041096,
+      "grad_norm": 0.697939657290891,
+      "learning_rate": 2.898716674553121e-06,
+      "loss": 2.4074,
+      "step": 276
+    },
+    {
+      "epoch": 1.8972602739726028,
+      "grad_norm": 0.7048161669777684,
+      "learning_rate": 2.8848402338999116e-06,
+      "loss": 2.3547,
+      "step": 277
+    },
+    {
+      "epoch": 1.904109589041096,
+      "grad_norm": 0.7550887476234565,
+      "learning_rate": 2.870951638232896e-06,
+      "loss": 2.3242,
+      "step": 278
+    },
+    {
+      "epoch": 1.910958904109589,
+      "grad_norm": 0.674336958631372,
+      "learning_rate": 2.8570513262174155e-06,
+      "loss": 2.3511,
+      "step": 279
+    },
+    {
+      "epoch": 1.9178082191780823,
+      "grad_norm": 0.7168968714166564,
+      "learning_rate": 2.843139736888864e-06,
+      "loss": 2.3232,
+      "step": 280
+    },
+    {
+      "epoch": 1.9246575342465753,
+      "grad_norm": 0.7236665241621988,
+      "learning_rate": 2.829217309638828e-06,
+      "loss": 2.3281,
+      "step": 281
+    },
+    {
+      "epoch": 1.9315068493150684,
+      "grad_norm": 0.6931721284135243,
+      "learning_rate": 2.8152844842012034e-06,
+      "loss": 2.3643,
+      "step": 282
+    },
+    {
+      "epoch": 1.9383561643835616,
+      "grad_norm": 0.6928469669838125,
+      "learning_rate": 2.8013417006383078e-06,
+      "loss": 2.2855,
+      "step": 283
+    },
+    {
+      "epoch": 1.9452054794520548,
+      "grad_norm": 0.6982709588867243,
+      "learning_rate": 2.787389399326984e-06,
+      "loss": 2.2552,
+      "step": 284
+    },
+    {
+      "epoch": 1.952054794520548,
+      "grad_norm": 0.6890157309338247,
+      "learning_rate": 2.773428020944687e-06,
+      "loss": 2.1769,
+      "step": 285
+    },
+    {
+      "epoch": 1.958904109589041,
+      "grad_norm": 0.6890167227071454,
+      "learning_rate": 2.7594580064555664e-06,
+      "loss": 2.2975,
+      "step": 286
+    },
+    {
+      "epoch": 1.9657534246575343,
+      "grad_norm": 0.6603722163357953,
+      "learning_rate": 2.745479797096543e-06,
+      "loss": 2.3294,
+      "step": 287
+    },
+    {
+      "epoch": 1.9726027397260273,
+      "grad_norm": 0.673631868721191,
+      "learning_rate": 2.7314938343633656e-06,
+      "loss": 2.3338,
+      "step": 288
+    },
+    {
+      "epoch": 1.9794520547945207,
+      "grad_norm": 0.6838024849491231,
+      "learning_rate": 2.717500559996672e-06,
+      "loss": 2.2469,
+      "step": 289
+    },
+    {
+      "epoch": 1.9863013698630136,
+      "grad_norm": 0.6915378189577142,
+      "learning_rate": 2.7035004159680332e-06,
+      "loss": 2.3446,
+      "step": 290
+    },
+    {
+      "epoch": 1.9931506849315068,
+      "grad_norm": 0.7308967059013186,
+      "learning_rate": 2.6894938444659972e-06,
+      "loss": 2.3048,
+      "step": 291
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6670727912036851,
+      "learning_rate": 2.675481287882121e-06,
+      "loss": 2.3657,
+      "step": 292
+    },
+    {
+      "epoch": 2.006849315068493,
+      "grad_norm": 0.7842325057992966,
+      "learning_rate": 2.661463188796996e-06,
+      "loss": 2.2781,
+      "step": 293
+    },
+    {
+      "epoch": 2.0136986301369864,
+      "grad_norm": 0.7703049180632918,
+      "learning_rate": 2.647439989966272e-06,
+      "loss": 2.2511,
+      "step": 294
+    },
+    {
+      "epoch": 2.0205479452054793,
+      "grad_norm": 0.7618588659706208,
+      "learning_rate": 2.633412134306672e-06,
+      "loss": 2.3002,
+      "step": 295
+    },
+    {
+      "epoch": 2.0273972602739727,
+      "grad_norm": 0.7698611238860366,
+      "learning_rate": 2.6193800648820056e-06,
+      "loss": 2.293,
+      "step": 296
+    },
+    {
+      "epoch": 2.0273972602739727,
+      "eval_loss": 2.413179397583008,
+      "eval_runtime": 5.9465,
+      "eval_samples_per_second": 52.468,
+      "eval_steps_per_second": 3.363,
+      "step": 296
+    },
+    {
+      "epoch": 2.0342465753424657,
+      "grad_norm": 0.744217752384927,
+      "learning_rate": 2.605344224889167e-06,
+      "loss": 2.2864,
+      "step": 297
+    },
+    {
+      "epoch": 2.041095890410959,
+      "grad_norm": 0.7308267298605848,
+      "learning_rate": 2.591305057644148e-06,
+      "loss": 2.2549,
+      "step": 298
+    },
+    {
+      "epoch": 2.047945205479452,
+      "grad_norm": 0.7364683383887598,
+      "learning_rate": 2.5772630065680247e-06,
+      "loss": 2.2634,
+      "step": 299
+    },
+    {
+      "epoch": 2.0547945205479454,
+      "grad_norm": 0.6947110181620351,
+      "learning_rate": 2.563218515172962e-06,
+      "loss": 2.2756,
+      "step": 300
+    },
+    {
+      "epoch": 2.0616438356164384,
+      "grad_norm": 0.7441894255170367,
+      "learning_rate": 2.5491720270481995e-06,
+      "loss": 2.2196,
+      "step": 301
+    },
+    {
+      "epoch": 2.0684931506849313,
+      "grad_norm": 0.7308523751420212,
+      "learning_rate": 2.5351239858460425e-06,
+      "loss": 2.1325,
+      "step": 302
+    },
+    {
+      "epoch": 2.0753424657534247,
+      "grad_norm": 0.723559375401077,
+      "learning_rate": 2.521074835267851e-06,
+      "loss": 2.2826,
+      "step": 303
+    },
+    {
+      "epoch": 2.0821917808219177,
+      "grad_norm": 0.7107824928376044,
+      "learning_rate": 2.507025019050022e-06,
+      "loss": 2.3009,
+      "step": 304
+    },
+    {
+      "epoch": 2.089041095890411,
+      "grad_norm": 0.7407695595687628,
+      "learning_rate": 2.4929749809499788e-06,
+      "loss": 2.294,
+      "step": 305
+    },
+    {
+      "epoch": 2.095890410958904,
+      "grad_norm": 0.9861121146811986,
+      "learning_rate": 2.4789251647321498e-06,
+      "loss": 2.3456,
+      "step": 306
+    },
+    {
+      "epoch": 2.1027397260273974,
+      "grad_norm": 0.7154622282506935,
+      "learning_rate": 2.464876014153958e-06,
+      "loss": 2.3047,
+      "step": 307
+    },
+    {
+      "epoch": 2.1095890410958904,
+      "grad_norm": 0.764573333314849,
+      "learning_rate": 2.4508279729518013e-06,
+      "loss": 2.2963,
+      "step": 308
+    },
+    {
+      "epoch": 2.1164383561643834,
+      "grad_norm": 0.7200447494905048,
+      "learning_rate": 2.4367814848270386e-06,
+      "loss": 2.3017,
+      "step": 309
+    },
+    {
+      "epoch": 2.1232876712328768,
+      "grad_norm": 0.7331208423596353,
+      "learning_rate": 2.422736993431976e-06,
+      "loss": 2.2193,
+      "step": 310
+    },
+    {
+      "epoch": 2.1301369863013697,
+      "grad_norm": 0.7181268238018352,
+      "learning_rate": 2.408694942355853e-06,
+      "loss": 2.2816,
+      "step": 311
+    },
+    {
+      "epoch": 2.136986301369863,
+      "grad_norm": 0.71696981098615,
+      "learning_rate": 2.3946557751108333e-06,
+      "loss": 2.2556,
+      "step": 312
+    },
+    {
+      "epoch": 2.143835616438356,
+      "grad_norm": 0.7496868820684847,
+      "learning_rate": 2.3806199351179948e-06,
+      "loss": 2.2647,
+      "step": 313
+    },
+    {
+      "epoch": 2.1506849315068495,
+      "grad_norm": 0.7170898258087575,
+      "learning_rate": 2.3665878656933285e-06,
+      "loss": 2.2258,
+      "step": 314
+    },
+    {
+      "epoch": 2.1575342465753424,
+      "grad_norm": 0.7349227307009416,
+      "learning_rate": 2.3525600100337294e-06,
+      "loss": 2.3019,
+      "step": 315
+    },
+    {
+      "epoch": 2.1643835616438354,
+      "grad_norm": 0.7339023774015392,
+      "learning_rate": 2.3385368112030054e-06,
+      "loss": 2.2243,
+      "step": 316
+    },
+    {
+      "epoch": 2.171232876712329,
+      "grad_norm": 0.7094170245428788,
+      "learning_rate": 2.3245187121178804e-06,
+      "loss": 2.2296,
+      "step": 317
+    },
+    {
+      "epoch": 2.1780821917808217,
+      "grad_norm": 0.7201765979542566,
+      "learning_rate": 2.310506155534003e-06,
+      "loss": 2.3034,
+      "step": 318
+    },
+    {
+      "epoch": 2.184931506849315,
+      "grad_norm": 0.6843911581159334,
+      "learning_rate": 2.296499584031967e-06,
+      "loss": 2.3191,
+      "step": 319
+    },
+    {
+      "epoch": 2.191780821917808,
+      "grad_norm": 0.715427807427916,
+      "learning_rate": 2.282499440003329e-06,
+      "loss": 2.2465,
+      "step": 320
+    },
+    {
+      "epoch": 2.1986301369863015,
+      "grad_norm": 0.7344624489275093,
+      "learning_rate": 2.268506165636635e-06,
+      "loss": 2.2288,
+      "step": 321
+    },
+    {
+      "epoch": 2.2054794520547945,
+      "grad_norm": 0.7296105473696515,
+      "learning_rate": 2.254520202903458e-06,
+      "loss": 2.237,
+      "step": 322
+    },
+    {
+      "epoch": 2.212328767123288,
+      "grad_norm": 0.7740678634111837,
+      "learning_rate": 2.240541993544434e-06,
+      "loss": 2.2005,
+      "step": 323
+    },
+    {
+      "epoch": 2.219178082191781,
+      "grad_norm": 0.6876899720253623,
+      "learning_rate": 2.2265719790553147e-06,
+      "loss": 2.0113,
+      "step": 324
+    },
+    {
+      "epoch": 2.2260273972602738,
+      "grad_norm": 0.7385389732044396,
+      "learning_rate": 2.212610600673017e-06,
+      "loss": 2.2027,
+      "step": 325
+    },
+    {
+      "epoch": 2.232876712328767,
+      "grad_norm": 0.7237738352415183,
+      "learning_rate": 2.1986582993616926e-06,
+      "loss": 2.3329,
+      "step": 326
+    },
+    {
+      "epoch": 2.23972602739726,
+      "grad_norm": 0.7224811508262542,
+      "learning_rate": 2.1847155157987975e-06,
+      "loss": 2.3061,
+      "step": 327
+    },
+    {
+      "epoch": 2.2465753424657535,
+      "grad_norm": 0.7141960930180122,
+      "learning_rate": 2.170782690361173e-06,
+      "loss": 2.249,
+      "step": 328
+    },
+    {
+      "epoch": 2.2534246575342465,
+      "grad_norm": 0.7210265725669223,
+      "learning_rate": 2.1568602631111362e-06,
+      "loss": 2.3222,
+      "step": 329
+    },
+    {
+      "epoch": 2.26027397260274,
+      "grad_norm": 0.7622124004386663,
+      "learning_rate": 2.1429486737825857e-06,
+      "loss": 2.2205,
+      "step": 330
+    },
+    {
+      "epoch": 2.267123287671233,
+      "grad_norm": 0.7131919831150063,
+      "learning_rate": 2.129048361767104e-06,
+      "loss": 2.2562,
+      "step": 331
+    },
+    {
+      "epoch": 2.2739726027397262,
+      "grad_norm": 0.716232761563867,
+      "learning_rate": 2.1151597661000884e-06,
+      "loss": 2.1547,
+      "step": 332
+    },
+    {
+      "epoch": 2.280821917808219,
+      "grad_norm": 0.7370885639371628,
+      "learning_rate": 2.1012833254468803e-06,
+      "loss": 2.3085,
+      "step": 333
+    },
+    {
+      "epoch": 2.280821917808219,
+      "eval_loss": 2.413699150085449,
+      "eval_runtime": 5.9363,
+      "eval_samples_per_second": 52.558,
+      "eval_steps_per_second": 3.369,
+      "step": 333
+    },
+    {
+      "epoch": 2.287671232876712,
+      "grad_norm": 0.6858773863616128,
+      "learning_rate": 2.087419478088906e-06,
+      "loss": 2.1914,
+      "step": 334
+    },
+    {
+      "epoch": 2.2945205479452055,
+      "grad_norm": 0.7071367211538865,
+      "learning_rate": 2.073568661909842e-06,
+      "loss": 2.2382,
+      "step": 335
+    },
+    {
+      "epoch": 2.3013698630136985,
+      "grad_norm": 0.7099731184047182,
+      "learning_rate": 2.0597313143817804e-06,
+      "loss": 2.2705,
+      "step": 336
+    },
+    {
+      "epoch": 2.308219178082192,
+      "grad_norm": 0.7132760269247956,
+      "learning_rate": 2.045907872551409e-06,
+      "loss": 2.2967,
+      "step": 337
+    },
+    {
+      "epoch": 2.315068493150685,
+      "grad_norm": 0.7026265666695738,
+      "learning_rate": 2.0320987730262136e-06,
+      "loss": 2.1903,
+      "step": 338
+    },
+    {
+      "epoch": 2.3219178082191783,
+      "grad_norm": 0.7121256173220049,
+      "learning_rate": 2.018304451960682e-06,
+      "loss": 2.2352,
+      "step": 339
+    },
+    {
+      "epoch": 2.328767123287671,
+      "grad_norm": 0.716016616197666,
+      "learning_rate": 2.0045253450425308e-06,
+      "loss": 2.2246,
+      "step": 340
+    },
+    {
+      "epoch": 2.3356164383561646,
+      "grad_norm": 0.702268794797935,
+      "learning_rate": 1.9907618874789463e-06,
+      "loss": 2.1513,
+      "step": 341
+    },
+    {
+      "epoch": 2.3424657534246576,
+      "grad_norm": 0.7211508092099944,
+      "learning_rate": 1.9770145139828334e-06,
+      "loss": 2.2313,
+      "step": 342
+    },
+    {
+      "epoch": 2.3493150684931505,
+      "grad_norm": 0.6875475522410895,
+      "learning_rate": 1.963283658759093e-06,
+      "loss": 2.1937,
+      "step": 343
+    },
+    {
+      "epoch": 2.356164383561644,
+      "grad_norm": 0.7100032015503472,
+      "learning_rate": 1.9495697554908984e-06,
+      "loss": 2.2032,
+      "step": 344
+    },
+    {
+      "epoch": 2.363013698630137,
+      "grad_norm": 0.7090312927179646,
+      "learning_rate": 1.935873237326006e-06,
+      "loss": 2.3233,
+      "step": 345
+    },
+    {
+      "epoch": 2.3698630136986303,
+      "grad_norm": 0.7321348941655884,
+      "learning_rate": 1.9221945368630703e-06,
+      "loss": 2.3082,
+      "step": 346
+    },
+    {
+      "epoch": 2.3767123287671232,
+      "grad_norm": 0.7176395955608619,
+      "learning_rate": 1.9085340861379815e-06,
+      "loss": 2.2542,
+      "step": 347
+    },
+    {
+      "epoch": 2.383561643835616,
+      "grad_norm": 0.7470287281593275,
+      "learning_rate": 1.8948923166102192e-06,
+      "loss": 2.2732,
+      "step": 348
+    },
+    {
+      "epoch": 2.3904109589041096,
+      "grad_norm": 0.7438948749235669,
+      "learning_rate": 1.8812696591492265e-06,
+      "loss": 2.2342,
+      "step": 349
+    },
+    {
+      "epoch": 2.3972602739726026,
+      "grad_norm": 0.7307563711082086,
+      "learning_rate": 1.8676665440207982e-06,
+      "loss": 2.3099,
+      "step": 350
+    },
+    {
+      "epoch": 2.404109589041096,
+      "grad_norm": 0.7505670286099257,
+      "learning_rate": 1.8540834008734944e-06,
+      "loss": 2.1853,
+      "step": 351
+    },
+    {
+      "epoch": 2.410958904109589,
+      "grad_norm": 0.7136239479111697,
+      "learning_rate": 1.840520658725069e-06,
+      "loss": 2.2143,
+      "step": 352
+    },
+    {
+      "epoch": 2.4178082191780823,
+      "grad_norm": 0.7027371598115614,
+      "learning_rate": 1.8269787459489174e-06,
+      "loss": 2.2046,
+      "step": 353
+    },
+    {
+      "epoch": 2.4246575342465753,
+      "grad_norm": 0.707254420656309,
+      "learning_rate": 1.8134580902605491e-06,
+      "loss": 2.29,
+      "step": 354
+    },
+    {
+      "epoch": 2.4315068493150687,
+      "grad_norm": 0.7170888804392873,
+      "learning_rate": 1.7999591187040782e-06,
+      "loss": 2.2643,
+      "step": 355
+    },
+    {
+      "epoch": 2.4383561643835616,
+      "grad_norm": 0.6639457176222473,
+      "learning_rate": 1.786482257638734e-06,
+      "loss": 2.2946,
+      "step": 356
+    },
+    {
+      "epoch": 2.4452054794520546,
+      "grad_norm": 0.6791585701285383,
+      "learning_rate": 1.7730279327253964e-06,
+      "loss": 2.3087,
+      "step": 357
+    },
+    {
+      "epoch": 2.452054794520548,
+      "grad_norm": 0.6825147247611417,
+      "learning_rate": 1.7595965689131484e-06,
+      "loss": 2.2252,
+      "step": 358
+    },
+    {
+      "epoch": 2.458904109589041,
+      "grad_norm": 0.7066098521292989,
+      "learning_rate": 1.7461885904258592e-06,
+      "loss": 2.2848,
+      "step": 359
+    },
+    {
+      "epoch": 2.4657534246575343,
+      "grad_norm": 0.7168260488520454,
+      "learning_rate": 1.732804420748781e-06,
+      "loss": 2.2381,
+      "step": 360
+    },
+    {
+      "epoch": 2.4726027397260273,
+      "grad_norm": 0.6940716557572651,
+      "learning_rate": 1.7194444826151753e-06,
+      "loss": 2.229,
+      "step": 361
+    },
+    {
+      "epoch": 2.4794520547945207,
+      "grad_norm": 0.6907305923666783,
+      "learning_rate": 1.7061091979929611e-06,
+      "loss": 2.2814,
+      "step": 362
+    },
+    {
+      "epoch": 2.4863013698630136,
+      "grad_norm": 0.7177474380488151,
+      "learning_rate": 1.6927989880713852e-06,
+      "loss": 2.2602,
+      "step": 363
+    },
+    {
+      "epoch": 2.493150684931507,
+      "grad_norm": 0.7581488681368201,
+      "learning_rate": 1.6795142732477222e-06,
+      "loss": 2.317,
+      "step": 364
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.7320361051298075,
+      "learning_rate": 1.6662554731139945e-06,
+      "loss": 2.2549,
+      "step": 365
+    },
+    {
+      "epoch": 2.506849315068493,
+      "grad_norm": 0.686393513195622,
+      "learning_rate": 1.6530230064437213e-06,
+      "loss": 2.2856,
+      "step": 366
+    },
+    {
+      "epoch": 2.5136986301369864,
+      "grad_norm": 0.7545957049643752,
+      "learning_rate": 1.6398172911786883e-06,
+      "loss": 2.2899,
+      "step": 367
+    },
+    {
+      "epoch": 2.5205479452054793,
+      "grad_norm": 0.7106340176915894,
+      "learning_rate": 1.6266387444157519e-06,
+      "loss": 2.2894,
+      "step": 368
+    },
+    {
+      "epoch": 2.5273972602739727,
+      "grad_norm": 0.7389989197034765,
+      "learning_rate": 1.613487782393661e-06,
+      "loss": 2.2875,
+      "step": 369
+    },
+    {
+      "epoch": 2.5342465753424657,
+      "grad_norm": 0.7649624949493309,
+      "learning_rate": 1.600364820479914e-06,
+      "loss": 2.1902,
+      "step": 370
+    },
+    {
+      "epoch": 2.5342465753424657,
+      "eval_loss": 2.4123480319976807,
+      "eval_runtime": 6.0173,
+      "eval_samples_per_second": 51.85,
+      "eval_steps_per_second": 3.324,
+      "step": 370
+    },
+    {
+      "epoch": 2.541095890410959,
+      "grad_norm": 0.7223441929466774,
+      "learning_rate": 1.5872702731576374e-06,
+      "loss": 2.269,
+      "step": 371
+    },
+    {
+      "epoch": 2.547945205479452,
+      "grad_norm": 0.7067038601638844,
+      "learning_rate": 1.5742045540124932e-06,
+      "loss": 2.2439,
+      "step": 372
+    },
+    {
+      "epoch": 2.5547945205479454,
+      "grad_norm": 0.7198264845513052,
+      "learning_rate": 1.5611680757196198e-06,
+      "loss": 2.3084,
+      "step": 373
+    },
+    {
+      "epoch": 2.5616438356164384,
+      "grad_norm": 0.7048209878339,
+      "learning_rate": 1.5481612500305964e-06,
+      "loss": 2.2518,
+      "step": 374
+    },
+    {
+      "epoch": 2.5684931506849313,
+      "grad_norm": 0.7345257609929702,
+      "learning_rate": 1.5351844877604356e-06,
+      "loss": 2.2796,
+      "step": 375
+    },
+    {
+      "epoch": 2.5753424657534247,
+      "grad_norm": 0.713666987740628,
+      "learning_rate": 1.5222381987746104e-06,
+      "loss": 2.2713,
+      "step": 376
+    },
+    {
+      "epoch": 2.5821917808219177,
+      "grad_norm": 0.7150783105673093,
+      "learning_rate": 1.5093227919761084e-06,
+      "loss": 2.2099,
+      "step": 377
+    },
+    {
+      "epoch": 2.589041095890411,
+      "grad_norm": 0.704487566383048,
+      "learning_rate": 1.4964386752925165e-06,
+      "loss": 2.3054,
+      "step": 378
+    },
+    {
+      "epoch": 2.595890410958904,
+      "grad_norm": 0.7522139473067558,
+      "learning_rate": 1.4835862556631369e-06,
+      "loss": 2.2186,
+      "step": 379
+    },
+    {
+      "epoch": 2.602739726027397,
+      "grad_norm": 0.6891880183627714,
+      "learning_rate": 1.4707659390261337e-06,
+      "loss": 2.3112,
+      "step": 380
+    },
+    {
+      "epoch": 2.6095890410958904,
+      "grad_norm": 0.7087422764378963,
+      "learning_rate": 1.4579781303057123e-06,
+      "loss": 2.2926,
+      "step": 381
+    },
+    {
+      "epoch": 2.616438356164384,
+      "grad_norm": 0.8486535470180528,
+      "learning_rate": 1.4452232333993272e-06,
+      "loss": 2.2778,
+      "step": 382
+    },
+    {
+      "epoch": 2.6232876712328768,
+      "grad_norm": 0.7768366805945564,
+      "learning_rate": 1.4325016511649315e-06,
+      "loss": 2.2441,
+      "step": 383
+    },
+    {
+      "epoch": 2.6301369863013697,
+      "grad_norm": 0.759828305906503,
+      "learning_rate": 1.4198137854082443e-06,
+      "loss": 2.2788,
+      "step": 384
+    },
+    {
+      "epoch": 2.636986301369863,
+      "grad_norm": 0.6859934839331832,
+      "learning_rate": 1.4071600368700669e-06,
+      "loss": 2.2792,
+      "step": 385
+    },
+    {
+      "epoch": 2.643835616438356,
+      "grad_norm": 0.7385417901666564,
+      "learning_rate": 1.3945408052136222e-06,
+      "loss": 2.2977,
+      "step": 386
+    },
+    {
+      "epoch": 2.6506849315068495,
+      "grad_norm": 0.7287504655916882,
+      "learning_rate": 1.3819564890119325e-06,
+      "loss": 2.2535,
+      "step": 387
+    },
+    {
+      "epoch": 2.6575342465753424,
+      "grad_norm": 0.6590047061716495,
+      "learning_rate": 1.3694074857352306e-06,
+      "loss": 2.2968,
+      "step": 388
+    },
+    {
+      "epoch": 2.6643835616438354,
+      "grad_norm": 0.6943988283660237,
+      "learning_rate": 1.3568941917384038e-06,
+      "loss": 2.2821,
+      "step": 389
+    },
+    {
+      "epoch": 2.671232876712329,
+      "grad_norm": 0.715573529250739,
+      "learning_rate": 1.3444170022484803e-06,
+      "loss": 2.2114,
+      "step": 390
+    },
+    {
+      "epoch": 2.678082191780822,
+      "grad_norm": 0.7016875788114846,
+      "learning_rate": 1.331976311352139e-06,
+      "loss": 2.2775,
+      "step": 391
+    },
+    {
+      "epoch": 2.684931506849315,
+      "grad_norm": 0.7436040372608549,
+      "learning_rate": 1.3195725119832693e-06,
+      "loss": 2.1983,
+      "step": 392
+    },
+    {
+      "epoch": 2.691780821917808,
+      "grad_norm": 0.6959008156370489,
+      "learning_rate": 1.307205995910557e-06,
+      "loss": 2.2541,
+      "step": 393
+    },
+    {
+      "epoch": 2.6986301369863015,
+      "grad_norm": 0.6701075429596892,
+      "learning_rate": 1.294877153725112e-06,
+      "loss": 2.2895,
+      "step": 394
+    },
+    {
+      "epoch": 2.7054794520547945,
+      "grad_norm": 0.7133265486183699,
+      "learning_rate": 1.2825863748281269e-06,
+      "loss": 2.311,
+      "step": 395
+    },
+    {
+      "epoch": 2.712328767123288,
+      "grad_norm": 0.731649203575663,
+      "learning_rate": 1.2703340474185856e-06,
+      "loss": 2.214,
+      "step": 396
+    },
+    {
+      "epoch": 2.719178082191781,
+      "grad_norm": 0.6951302895005032,
+      "learning_rate": 1.258120558480996e-06,
+      "loss": 2.2798,
+      "step": 397
+    },
+    {
+      "epoch": 2.7260273972602738,
+      "grad_norm": 0.7342864394765313,
+      "learning_rate": 1.245946293773171e-06,
+      "loss": 2.2428,
+      "step": 398
+    },
+    {
+      "epoch": 2.732876712328767,
+      "grad_norm": 0.6748430142696247,
+      "learning_rate": 1.2338116378140424e-06,
+      "loss": 2.1684,
+      "step": 399
+    },
+    {
+      "epoch": 2.73972602739726,
+      "grad_norm": 0.706147657425132,
+      "learning_rate": 1.2217169738715162e-06,
+      "loss": 2.2685,
+      "step": 400
+    },
+    {
+      "epoch": 2.7465753424657535,
+      "grad_norm": 0.7029833217334348,
+      "learning_rate": 1.2096626839503666e-06,
+      "loss": 2.257,
+      "step": 401
+    },
+    {
+      "epoch": 2.7534246575342465,
+      "grad_norm": 0.6915579514085184,
+      "learning_rate": 1.1976491487801747e-06,
+      "loss": 2.3553,
+      "step": 402
+    },
+    {
+      "epoch": 2.76027397260274,
+      "grad_norm": 0.6935009825915379,
+      "learning_rate": 1.1856767478032978e-06,
+      "loss": 2.3177,
+      "step": 403
+    },
+    {
+      "epoch": 2.767123287671233,
+      "grad_norm": 0.776510135143567,
+      "learning_rate": 1.1737458591628898e-06,
+      "loss": 2.1945,
+      "step": 404
+    },
+    {
+      "epoch": 2.7739726027397262,
+      "grad_norm": 0.6785739385871513,
+      "learning_rate": 1.1618568596909514e-06,
+      "loss": 2.3425,
+      "step": 405
+    },
+    {
+      "epoch": 2.780821917808219,
+      "grad_norm": 0.7244494372149703,
+      "learning_rate": 1.150010124896435e-06,
+      "loss": 2.3191,
+      "step": 406
+    },
+    {
+      "epoch": 2.787671232876712,
+      "grad_norm": 0.723165620037459,
+      "learning_rate": 1.1382060289533805e-06,
+      "loss": 2.216,
+      "step": 407
+    },
+    {
+      "epoch": 2.787671232876712,
+      "eval_loss": 2.4112062454223633,
+      "eval_runtime": 6.0106,
+      "eval_samples_per_second": 51.908,
+      "eval_steps_per_second": 3.327,
+      "step": 407
+    },
+    {
+      "epoch": 2.7945205479452055,
+      "grad_norm": 0.716573732956154,
+      "learning_rate": 1.1264449446890976e-06,
+      "loss": 2.3031,
+      "step": 408
+    },
+    {
+      "epoch": 2.8013698630136985,
+      "grad_norm": 0.6990575425388178,
+      "learning_rate": 1.114727243572392e-06,
+      "loss": 2.2865,
+      "step": 409
+    },
+    {
+      "epoch": 2.808219178082192,
+      "grad_norm": 0.7033064610318523,
+      "learning_rate": 1.1030532957018288e-06,
+      "loss": 2.2547,
+      "step": 410
+    },
+    {
+      "epoch": 2.815068493150685,
+      "grad_norm": 0.7066376790442048,
+      "learning_rate": 1.0914234697940465e-06,
+      "loss": 2.2383,
+      "step": 411
+    },
+    {
+      "epoch": 2.821917808219178,
+      "grad_norm": 0.7220557474397004,
+      "learning_rate": 1.079838133172111e-06,
+      "loss": 2.18,
+      "step": 412
+    },
+    {
+      "epoch": 2.828767123287671,
+      "grad_norm": 0.710206848234572,
+      "learning_rate": 1.068297651753913e-06,
+      "loss": 2.2733,
+      "step": 413
+    },
+    {
+      "epoch": 2.8356164383561646,
+      "grad_norm": 0.7327379146961583,
+      "learning_rate": 1.0568023900406108e-06,
+      "loss": 2.1961,
+      "step": 414
+    },
+    {
+      "epoch": 2.8424657534246576,
+      "grad_norm": 0.6842961302424317,
+      "learning_rate": 1.0453527111051183e-06,
+      "loss": 2.1984,
+      "step": 415
+    },
+    {
+      "epoch": 2.8493150684931505,
+      "grad_norm": 0.7658479578627203,
+      "learning_rate": 1.0339489765806347e-06,
+      "loss": 2.2618,
+      "step": 416
+    },
+    {
+      "epoch": 2.856164383561644,
+      "grad_norm": 0.6961196544731081,
+      "learning_rate": 1.0225915466492268e-06,
+      "loss": 2.2644,
+      "step": 417
+    },
+    {
+      "epoch": 2.863013698630137,
+      "grad_norm": 0.6960320474171482,
+      "learning_rate": 1.0112807800304499e-06,
+      "loss": 2.1823,
+      "step": 418
+    },
+    {
+      "epoch": 2.8698630136986303,
+      "grad_norm": 0.7327958977392377,
+      "learning_rate": 1.00001703397002e-06,
+      "loss": 2.2894,
+      "step": 419
+    },
+    {
+      "epoch": 2.8767123287671232,
+      "grad_norm": 0.7175112056018497,
+      "learning_rate": 9.888006642285255e-07,
+      "loss": 2.3192,
+      "step": 420
+    },
+    {
+      "epoch": 2.883561643835616,
+      "grad_norm": 0.6951758710131047,
+      "learning_rate": 9.776320250701983e-07,
+      "loss": 2.2736,
+      "step": 421
+    },
+    {
+      "epoch": 2.8904109589041096,
+      "grad_norm": 0.8651114201507224,
+      "learning_rate": 9.665114692517157e-07,
+      "loss": 2.1695,
+      "step": 422
+    },
+    {
+      "epoch": 2.897260273972603,
+      "grad_norm": 0.7095105670222126,
+      "learning_rate": 9.554393480110678e-07,
+      "loss": 2.3053,
+      "step": 423
+    },
+    {
+      "epoch": 2.904109589041096,
+      "grad_norm": 0.721617667719065,
+      "learning_rate": 9.444160110564563e-07,
+      "loss": 2.2036,
+      "step": 424
+    },
+    {
+      "epoch": 2.910958904109589,
+      "grad_norm": 0.7086651454995305,
+      "learning_rate": 9.334418065552539e-07,
+      "loss": 2.3177,
+      "step": 425
+    },
+    {
+      "epoch": 2.9178082191780823,
+      "grad_norm": 0.725835606196638,
+      "learning_rate": 9.225170811230058e-07,
+      "loss": 2.3067,
+      "step": 426
+    },
+    {
+      "epoch": 2.9246575342465753,
+      "grad_norm": 0.6776112672757169,
+      "learning_rate": 9.116421798124794e-07,
+      "loss": 2.2832,
+      "step": 427
+    },
+    {
+      "epoch": 2.9315068493150687,
+      "grad_norm": 0.7033324367359345,
+      "learning_rate": 9.008174461027724e-07,
+      "loss": 2.1292,
+      "step": 428
+    },
+    {
+      "epoch": 2.9383561643835616,
+      "grad_norm": 0.6838016967986851,
+      "learning_rate": 8.900432218884567e-07,
+      "loss": 2.2064,
+      "step": 429
+    },
+    {
+      "epoch": 2.9452054794520546,
+      "grad_norm": 0.7010714046350294,
+      "learning_rate": 8.793198474687861e-07,
+      "loss": 2.2926,
+      "step": 430
+    },
+    {
+      "epoch": 2.952054794520548,
+      "grad_norm": 0.7329321374936051,
+      "learning_rate": 8.686476615369452e-07,
+      "loss": 2.1508,
+      "step": 431
+    },
+    {
+      "epoch": 2.958904109589041,
+      "grad_norm": 0.7131247907500139,
+      "learning_rate": 8.580270011693498e-07,
+      "loss": 2.2816,
+      "step": 432
+    },
+    {
+      "epoch": 2.9657534246575343,
+      "grad_norm": 0.7307979149023858,
+      "learning_rate": 8.474582018150054e-07,
+      "loss": 2.2547,
+      "step": 433
+    },
+    {
+      "epoch": 2.9726027397260273,
+      "grad_norm": 0.72846919982668,
+      "learning_rate": 8.369415972849087e-07,
+      "loss": 2.2389,
+      "step": 434
+    },
+    {
+      "epoch": 2.9794520547945207,
+      "grad_norm": 0.7303659540542966,
+      "learning_rate": 8.264775197415054e-07,
+      "loss": 2.2541,
+      "step": 435
+    },
+    {
+      "epoch": 2.9863013698630136,
+      "grad_norm": 0.7122425099335468,
+      "learning_rate": 8.160662996881996e-07,
+      "loss": 2.1782,
+      "step": 436
+    },
+    {
+      "epoch": 2.993150684931507,
+      "grad_norm": 0.7278886861085458,
+      "learning_rate": 8.057082659589116e-07,
+      "loss": 2.2729,
+      "step": 437
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.6862919290272542,
+      "learning_rate": 7.95403745707698e-07,
+      "loss": 2.26,
+      "step": 438
+    },
+    {
+      "epoch": 3.006849315068493,
+      "grad_norm": 0.7049119587815889,
+      "learning_rate": 7.851530643984112e-07,
+      "loss": 2.2465,
+      "step": 439
+    },
+    {
+      "epoch": 3.0136986301369864,
+      "grad_norm": 0.7526125166918066,
+      "learning_rate": 7.749565457944274e-07,
+      "loss": 2.2431,
+      "step": 440
+    },
+    {
+      "epoch": 3.0205479452054793,
+      "grad_norm": 0.723575654733539,
+      "learning_rate": 7.648145119484152e-07,
+      "loss": 2.2098,
+      "step": 441
+    },
+    {
+      "epoch": 3.0273972602739727,
+      "grad_norm": 0.7289118780284567,
+      "learning_rate": 7.547272831921665e-07,
+      "loss": 2.1957,
+      "step": 442
+    },
+    {
+      "epoch": 3.0342465753424657,
+      "grad_norm": 0.6949718275268681,
+      "learning_rate": 7.446951781264755e-07,
+      "loss": 2.2454,
+      "step": 443
+    },
+    {
+      "epoch": 3.041095890410959,
+      "grad_norm": 0.7196631653126768,
+      "learning_rate": 7.347185136110808e-07,
+      "loss": 2.3081,
+      "step": 444
+    },
+    {
+      "epoch": 3.041095890410959,
+      "eval_loss": 2.41233491897583,
+      "eval_runtime": 6.0379,
+      "eval_samples_per_second": 51.673,
+      "eval_steps_per_second": 3.312,
+      "step": 444
+    },
+    {
+      "epoch": 3.047945205479452,
+      "grad_norm": 0.7112349059017333,
+      "learning_rate": 7.24797604754654e-07,
+      "loss": 2.2418,
+      "step": 445
+    },
+    {
+      "epoch": 3.0547945205479454,
+      "grad_norm": 0.6958734173109298,
+      "learning_rate": 7.149327649048482e-07,
+      "loss": 2.1848,
+      "step": 446
+    },
+    {
+      "epoch": 3.0616438356164384,
+      "grad_norm": 0.6860159935805402,
+      "learning_rate": 7.051243056384016e-07,
+      "loss": 2.2345,
+      "step": 447
+    },
+    {
+      "epoch": 3.0684931506849313,
+      "grad_norm": 0.6960027631751124,
+      "learning_rate": 6.953725367512951e-07,
+      "loss": 2.1578,
+      "step": 448
+    },
+    {
+      "epoch": 3.0753424657534247,
+      "grad_norm": 0.7250494251501289,
+      "learning_rate": 6.856777662489669e-07,
+      "loss": 2.2682,
+      "step": 449
+    },
+    {
+      "epoch": 3.0821917808219177,
+      "grad_norm": 0.7329529785743732,
+      "learning_rate": 6.760403003365884e-07,
+      "loss": 2.2279,
+      "step": 450
+    },
+    {
+      "epoch": 3.089041095890411,
+      "grad_norm": 0.7376472824938267,
+      "learning_rate": 6.664604434093886e-07,
+      "loss": 2.227,
+      "step": 451
+    },
+    {
+      "epoch": 3.095890410958904,
+      "grad_norm": 0.7566044596383341,
+      "learning_rate": 6.569384980430416e-07,
+      "loss": 2.205,
+      "step": 452
+    },
+    {
+      "epoch": 3.1027397260273974,
+      "grad_norm": 0.7292054250308317,
+      "learning_rate": 6.474747649841104e-07,
+      "loss": 2.206,
+      "step": 453
+    },
+    {
+      "epoch": 3.1095890410958904,
+      "grad_norm": 0.712518081107752,
+      "learning_rate": 6.380695431405453e-07,
+      "loss": 2.205,
+      "step": 454
+    },
+    {
+      "epoch": 3.1164383561643834,
+      "grad_norm": 0.7536284239063324,
+      "learning_rate": 6.28723129572247e-07,
+      "loss": 2.2662,
+      "step": 455
+    },
+    {
+      "epoch": 3.1232876712328768,
+      "grad_norm": 0.7401865952989628,
+      "learning_rate": 6.194358194816813e-07,
+      "loss": 2.1479,
+      "step": 456
+    },
+    {
+      "epoch": 3.1301369863013697,
+      "grad_norm": 0.6936777496577218,
+      "learning_rate": 6.10207906204556e-07,
+      "loss": 2.2103,
+      "step": 457
+    },
+    {
+      "epoch": 3.136986301369863,
+      "grad_norm": 0.6883084739971798,
+      "learning_rate": 6.010396812005554e-07,
+      "loss": 2.2406,
+      "step": 458
+    },
+    {
+      "epoch": 3.143835616438356,
+      "grad_norm": 0.7021480171737262,
+      "learning_rate": 5.919314340441362e-07,
+      "loss": 2.2287,
+      "step": 459
+    },
+    {
+      "epoch": 3.1506849315068495,
+      "grad_norm": 0.7006207646547893,
+      "learning_rate": 5.828834524153795e-07,
+      "loss": 2.2729,
+      "step": 460
+    },
+    {
+      "epoch": 3.1575342465753424,
+      "grad_norm": 0.7229748958970227,
+      "learning_rate": 5.738960220909067e-07,
+      "loss": 2.2681,
+      "step": 461
+    },
+    {
+      "epoch": 3.1643835616438354,
+      "grad_norm": 0.701712318725158,
+      "learning_rate": 5.649694269348516e-07,
+      "loss": 2.266,
+      "step": 462
+    },
+    {
+      "epoch": 3.171232876712329,
+      "grad_norm": 0.7131095848134887,
+      "learning_rate": 5.561039488898959e-07,
+      "loss": 2.175,
+      "step": 463
+    },
+    {
+      "epoch": 3.1780821917808217,
+      "grad_norm": 0.7157421024649682,
+      "learning_rate": 5.472998679683619e-07,
+      "loss": 2.2373,
+      "step": 464
+    },
+    {
+      "epoch": 3.184931506849315,
+      "grad_norm": 0.6989091813195526,
+      "learning_rate": 5.385574622433715e-07,
+      "loss": 2.2671,
+      "step": 465
+    },
+    {
+      "epoch": 3.191780821917808,
+      "grad_norm": 0.7100047458690754,
+      "learning_rate": 5.298770078400628e-07,
+      "loss": 2.2431,
+      "step": 466
+    },
+    {
+      "epoch": 3.1986301369863015,
+      "grad_norm": 0.7221330810980103,
+      "learning_rate": 5.21258778926865e-07,
+      "loss": 2.1897,
+      "step": 467
+    },
+    {
+      "epoch": 3.2054794520547945,
+      "grad_norm": 0.7225424671384507,
+      "learning_rate": 5.127030477068446e-07,
+      "loss": 2.2322,
+      "step": 468
+    },
+    {
+      "epoch": 3.212328767123288,
+      "grad_norm": 0.6991689104341696,
+      "learning_rate": 5.04210084409105e-07,
+      "loss": 2.3075,
+      "step": 469
+    },
+    {
+      "epoch": 3.219178082191781,
+      "grad_norm": 0.6875008540842864,
+      "learning_rate": 4.957801572802495e-07,
+      "loss": 2.2509,
+      "step": 470
+    },
+    {
+      "epoch": 3.2260273972602738,
+      "grad_norm": 0.710321387958742,
+      "learning_rate": 4.874135325759133e-07,
+      "loss": 2.2283,
+      "step": 471
+    },
+    {
+      "epoch": 3.232876712328767,
+      "grad_norm": 0.7848622442448815,
+      "learning_rate": 4.791104745523509e-07,
+      "loss": 2.1495,
+      "step": 472
+    },
+    {
+      "epoch": 3.23972602739726,
+      "grad_norm": 0.6876021495166095,
+      "learning_rate": 4.708712454580905e-07,
+      "loss": 2.2525,
+      "step": 473
+    },
+    {
+      "epoch": 3.2465753424657535,
+      "grad_norm": 0.7059110406730769,
+      "learning_rate": 4.6269610552565153e-07,
+      "loss": 2.2432,
+      "step": 474
+    },
+    {
+      "epoch": 3.2534246575342465,
+      "grad_norm": 0.713960613803224,
+      "learning_rate": 4.5458531296332267e-07,
+      "loss": 2.3112,
+      "step": 475
+    },
+    {
+      "epoch": 3.26027397260274,
+      "grad_norm": 0.7145155485011042,
+      "learning_rate": 4.465391239470113e-07,
+      "loss": 2.1603,
+      "step": 476
+    },
+    {
+      "epoch": 3.267123287671233,
+      "grad_norm": 0.7510837858210794,
+      "learning_rate": 4.385577926121465e-07,
+      "loss": 2.1269,
+      "step": 477
+    },
+    {
+      "epoch": 3.2739726027397262,
+      "grad_norm": 0.7042639926048964,
+      "learning_rate": 4.306415710456577e-07,
+      "loss": 2.2198,
+      "step": 478
+    },
+    {
+      "epoch": 3.280821917808219,
+      "grad_norm": 0.746593356064073,
+      "learning_rate": 4.227907092780095e-07,
+      "loss": 2.1076,
+      "step": 479
+    },
+    {
+      "epoch": 3.287671232876712,
+      "grad_norm": 0.681218067460257,
+      "learning_rate": 4.150054552753055e-07,
+      "loss": 2.1681,
+      "step": 480
+    },
+    {
+      "epoch": 3.2945205479452055,
+      "grad_norm": 0.730138040035468,
+      "learning_rate": 4.07286054931455e-07,
+      "loss": 2.1989,
+      "step": 481
+    },
+    {
+      "epoch": 3.2945205479452055,
+      "eval_loss": 2.4142165184020996,
+      "eval_runtime": 5.917,
+      "eval_samples_per_second": 52.729,
+      "eval_steps_per_second": 3.38,
+      "step": 481
+    },
+    {
+      "epoch": 3.3013698630136985,
+      "grad_norm": 0.7712164896594699,
+      "learning_rate": 3.9963275206040876e-07,
+      "loss": 2.2356,
+      "step": 482
+    },
+    {
+      "epoch": 3.308219178082192,
+      "grad_norm": 0.7039183878730115,
+      "learning_rate": 3.920457883884571e-07,
+      "loss": 2.2611,
+      "step": 483
+    },
+    {
+      "epoch": 3.315068493150685,
+      "grad_norm": 0.6955665234984989,
+      "learning_rate": 3.845254035465951e-07,
+      "loss": 2.2646,
+      "step": 484
+    },
+    {
+      "epoch": 3.3219178082191783,
+      "grad_norm": 0.7099012837873744,
+      "learning_rate": 3.770718350629543e-07,
+      "loss": 2.2907,
+      "step": 485
+    },
+    {
+      "epoch": 3.328767123287671,
+      "grad_norm": 0.7786241217797155,
+      "learning_rate": 3.696853183552998e-07,
+      "loss": 2.2216,
+      "step": 486
+    },
+    {
+      "epoch": 3.3356164383561646,
+      "grad_norm": 0.7411561630337811,
+      "learning_rate": 3.6236608672359454e-07,
+      "loss": 2.1726,
+      "step": 487
+    },
+    {
+      "epoch": 3.3424657534246576,
+      "grad_norm": 0.7134345658926926,
+      "learning_rate": 3.5511437134263196e-07,
+      "loss": 2.2528,
+      "step": 488
+    },
+    {
+      "epoch": 3.3493150684931505,
+      "grad_norm": 0.7355272001792791,
+      "learning_rate": 3.479304012547338e-07,
+      "loss": 2.2795,
+      "step": 489
+    },
+    {
+      "epoch": 3.356164383561644,
+      "grad_norm": 0.714793937484987,
+      "learning_rate": 3.408144033625163e-07,
+      "loss": 2.2244,
+      "step": 490
+    },
+    {
+      "epoch": 3.363013698630137,
+      "grad_norm": 0.7048287291783248,
+      "learning_rate": 3.3376660242172093e-07,
+      "loss": 2.2781,
+      "step": 491
+    },
+    {
+      "epoch": 3.3698630136986303,
+      "grad_norm": 0.7772137652019824,
+      "learning_rate": 3.267872210341194e-07,
+      "loss": 2.1204,
+      "step": 492
+    },
+    {
+      "epoch": 3.3767123287671232,
+      "grad_norm": 0.7609396724890849,
+      "learning_rate": 3.1987647964048075e-07,
+      "loss": 2.1819,
+      "step": 493
+    },
+    {
+      "epoch": 3.383561643835616,
+      "grad_norm": 0.7160447015330966,
+      "learning_rate": 3.1303459651361027e-07,
+      "loss": 2.2686,
+      "step": 494
+    },
+    {
+      "epoch": 3.3904109589041096,
+      "grad_norm": 0.7002741574191857,
+      "learning_rate": 3.0626178775145175e-07,
+      "loss": 2.31,
+      "step": 495
+    },
+    {
+      "epoch": 3.3972602739726026,
+      "grad_norm": 0.7321963136731753,
+      "learning_rate": 2.995582672702679e-07,
+      "loss": 2.2098,
+      "step": 496
+    },
+    {
+      "epoch": 3.404109589041096,
+      "grad_norm": 0.7388217600508439,
+      "learning_rate": 2.9292424679787825e-07,
+      "loss": 2.259,
+      "step": 497
+    },
+    {
+      "epoch": 3.410958904109589,
+      "grad_norm": 0.7129862504829712,
+      "learning_rate": 2.8635993586697555e-07,
+      "loss": 2.2573,
+      "step": 498
+    },
+    {
+      "epoch": 3.4178082191780823,
+      "grad_norm": 0.7166684509813519,
+      "learning_rate": 2.7986554180850666e-07,
+      "loss": 2.1755,
+      "step": 499
+    },
+    {
+      "epoch": 3.4246575342465753,
+      "grad_norm": 0.700743323964397,
+      "learning_rate": 2.734412697451236e-07,
+      "loss": 2.2432,
+      "step": 500
+    },
+    {
+      "epoch": 3.4315068493150687,
+      "grad_norm": 0.7305544630087399,
+      "learning_rate": 2.670873225847062e-07,
+      "loss": 2.3166,
+      "step": 501
+    },
+    {
+      "epoch": 3.4383561643835616,
+      "grad_norm": 0.7014354116705241,
+      "learning_rate": 2.6080390101395044e-07,
+      "loss": 2.2137,
+      "step": 502
+    },
+    {
+      "epoch": 3.4452054794520546,
+      "grad_norm": 0.7810916976244717,
+      "learning_rate": 2.545912034920331e-07,
+      "loss": 2.2429,
+      "step": 503
+    },
+    {
+      "epoch": 3.452054794520548,
+      "grad_norm": 0.7667288256979875,
+      "learning_rate": 2.484494262443429e-07,
+      "loss": 2.057,
+      "step": 504
+    },
+    {
+      "epoch": 3.458904109589041,
+      "grad_norm": 0.7005192897641623,
+      "learning_rate": 2.423787632562802e-07,
+      "loss": 2.2204,
+      "step": 505
+    },
+    {
+      "epoch": 3.4657534246575343,
+      "grad_norm": 0.7565336210321355,
+      "learning_rate": 2.3637940626713346e-07,
+      "loss": 2.1146,
+      "step": 506
+    },
+    {
+      "epoch": 3.4726027397260273,
+      "grad_norm": 0.7673059104886998,
+      "learning_rate": 2.3045154476402154e-07,
+      "loss": 2.2906,
+      "step": 507
+    },
+    {
+      "epoch": 3.4794520547945207,
+      "grad_norm": 0.6961794195903263,
+      "learning_rate": 2.2459536597590785e-07,
+      "loss": 2.2579,
+      "step": 508
+    },
+    {
+      "epoch": 3.4863013698630136,
+      "grad_norm": 0.7197371965925977,
+      "learning_rate": 2.1881105486768943e-07,
+      "loss": 2.2471,
+      "step": 509
+    },
+    {
+      "epoch": 3.493150684931507,
+      "grad_norm": 0.7482356824909315,
+      "learning_rate": 2.1309879413435291e-07,
+      "loss": 2.2682,
+      "step": 510
+    },
+    {
+      "epoch": 3.5,
+      "grad_norm": 0.7135265971194484,
+      "learning_rate": 2.0745876419520446e-07,
+      "loss": 2.2263,
+      "step": 511
+    },
+    {
+      "epoch": 3.506849315068493,
+      "grad_norm": 0.7323968706612133,
+      "learning_rate": 2.0189114318817232e-07,
+      "loss": 2.273,
+      "step": 512
+    },
+    {
+      "epoch": 3.5136986301369864,
+      "grad_norm": 0.7399704016989775,
+      "learning_rate": 1.96396106964179e-07,
+      "loss": 2.2872,
+      "step": 513
+    },
+    {
+      "epoch": 3.5205479452054793,
+      "grad_norm": 0.7166487593603337,
+      "learning_rate": 1.9097382908158713e-07,
+      "loss": 2.2448,
+      "step": 514
+    },
+    {
+      "epoch": 3.5273972602739727,
+      "grad_norm": 0.6946094346915361,
+      "learning_rate": 1.856244808007196e-07,
+      "loss": 2.2628,
+      "step": 515
+    },
+    {
+      "epoch": 3.5342465753424657,
+      "grad_norm": 0.7027218079715242,
+      "learning_rate": 1.803482310784488e-07,
+      "loss": 2.228,
+      "step": 516
+    },
+    {
+      "epoch": 3.541095890410959,
+      "grad_norm": 0.7371132166155652,
+      "learning_rate": 1.751452465628603e-07,
+      "loss": 2.207,
+      "step": 517
+    },
+    {
+      "epoch": 3.547945205479452,
+      "grad_norm": 0.7085190460590639,
+      "learning_rate": 1.7001569158798915e-07,
+      "loss": 2.2527,
+      "step": 518
+    },
+    {
+      "epoch": 3.547945205479452,
+      "eval_loss": 2.414205312728882,
+      "eval_runtime": 6.2446,
+      "eval_samples_per_second": 49.963,
+      "eval_steps_per_second": 3.203,
+      "step": 518
+    },
+    {
+      "epoch": 3.5547945205479454,
+      "grad_norm": 0.7115503219623986,
+      "learning_rate": 1.649597281686302e-07,
+      "loss": 2.2244,
+      "step": 519
+    },
+    {
+      "epoch": 3.5616438356164384,
+      "grad_norm": 0.7393802383503587,
+      "learning_rate": 1.5997751599522048e-07,
+      "loss": 2.2057,
+      "step": 520
+    },
+    {
+      "epoch": 3.5684931506849313,
+      "grad_norm": 0.6954311400989135,
+      "learning_rate": 1.5506921242879613e-07,
+      "loss": 2.1232,
+      "step": 521
+    },
+    {
+      "epoch": 3.5753424657534247,
+      "grad_norm": 0.7058695725256162,
+      "learning_rate": 1.5023497249602088e-07,
+      "loss": 2.2073,
+      "step": 522
+    },
+    {
+      "epoch": 3.5821917808219177,
+      "grad_norm": 0.7372037350987035,
+      "learning_rate": 1.4547494888429074e-07,
+      "loss": 2.2279,
+      "step": 523
+    },
+    {
+      "epoch": 3.589041095890411,
+      "grad_norm": 0.7429665678283818,
+      "learning_rate": 1.4078929193691e-07,
+      "loss": 2.2386,
+      "step": 524
+    },
+    {
+      "epoch": 3.595890410958904,
+      "grad_norm": 0.686731529836446,
+      "learning_rate": 1.3617814964834525e-07,
+      "loss": 2.2043,
+      "step": 525
+    },
+    {
+      "epoch": 3.602739726027397,
+      "grad_norm": 0.7331622181608642,
+      "learning_rate": 1.316416676595486e-07,
+      "loss": 2.3034,
+      "step": 526
+    },
+    {
+      "epoch": 3.6095890410958904,
+      "grad_norm": 0.6824749210822239,
+      "learning_rate": 1.2717998925335927e-07,
+      "loss": 2.2717,
+      "step": 527
+    },
+    {
+      "epoch": 3.616438356164384,
+      "grad_norm": 0.7017615036181578,
+      "learning_rate": 1.2279325534997695e-07,
+      "loss": 2.2533,
+      "step": 528
+    },
+    {
+      "epoch": 3.6232876712328768,
+      "grad_norm": 0.7781861871902247,
+      "learning_rate": 1.1848160450251084e-07,
+      "loss": 2.2429,
+      "step": 529
+    },
+    {
+      "epoch": 3.6301369863013697,
+      "grad_norm": 0.746950853170954,
+      "learning_rate": 1.14245172892605e-07,
+      "loss": 2.2331,
+      "step": 530
+    },
+    {
+      "epoch": 3.636986301369863,
+      "grad_norm": 0.6763109907352243,
+      "learning_rate": 1.1008409432613526e-07,
+      "loss": 2.1877,
+      "step": 531
+    },
+    {
+      "epoch": 3.643835616438356,
+      "grad_norm": 0.7083895479311662,
+      "learning_rate": 1.0599850022898539e-07,
+      "loss": 2.0281,
+      "step": 532
+    },
+    {
+      "epoch": 3.6506849315068495,
+      "grad_norm": 0.7320813086291221,
+      "learning_rate": 1.0198851964289186e-07,
+      "loss": 2.2632,
+      "step": 533
+    },
+    {
+      "epoch": 3.6575342465753424,
+      "grad_norm": 0.7190075069996767,
+      "learning_rate": 9.805427922137373e-08,
+      "loss": 2.238,
+      "step": 534
+    },
+    {
+      "epoch": 3.6643835616438354,
+      "grad_norm": 0.7395924773928994,
+      "learning_rate": 9.419590322572725e-08,
+      "loss": 2.2408,
+      "step": 535
+    },
+    {
+      "epoch": 3.671232876712329,
+      "grad_norm": 0.7096579236402969,
+      "learning_rate": 9.041351352110428e-08,
+      "loss": 2.1901,
+      "step": 536
+    },
+    {
+      "epoch": 3.678082191780822,
+      "grad_norm": 0.7107976212854911,
+      "learning_rate": 8.670722957266231e-08,
+      "loss": 2.2501,
+      "step": 537
+    },
+    {
+      "epoch": 3.684931506849315,
+      "grad_norm": 0.6989979571761717,
+      "learning_rate": 8.307716844179081e-08,
+      "loss": 2.2474,
+      "step": 538
+    },
+    {
+      "epoch": 3.691780821917808,
+      "grad_norm": 0.693230240829913,
+      "learning_rate": 7.952344478241503e-08,
+      "loss": 2.0538,
+      "step": 539
+    },
+    {
+      "epoch": 3.6986301369863015,
+      "grad_norm": 0.7135983260649856,
+      "learning_rate": 7.60461708373722e-08,
+      "loss": 2.2146,
+      "step": 540
+    },
+    {
+      "epoch": 3.7054794520547945,
+      "grad_norm": 0.7029895057595396,
+      "learning_rate": 7.264545643486997e-08,
+      "loss": 2.2476,
+      "step": 541
+    },
+    {
+      "epoch": 3.712328767123288,
+      "grad_norm": 0.7565443276211162,
+      "learning_rate": 6.932140898501472e-08,
+      "loss": 2.2183,
+      "step": 542
+    },
+    {
+      "epoch": 3.719178082191781,
+      "grad_norm": 0.72358920752058,
+      "learning_rate": 6.60741334764195e-08,
+      "loss": 2.2229,
+      "step": 543
+    },
+    {
+      "epoch": 3.7260273972602738,
+      "grad_norm": 0.7176993155844448,
+      "learning_rate": 6.290373247289012e-08,
+      "loss": 2.2194,
+      "step": 544
+    },
+    {
+      "epoch": 3.732876712328767,
+      "grad_norm": 0.7849805871126758,
+      "learning_rate": 5.981030611018235e-08,
+      "loss": 2.2008,
+      "step": 545
+    },
+    {
+      "epoch": 3.73972602739726,
+      "grad_norm": 0.7348775068699039,
+      "learning_rate": 5.6793952092841776e-08,
+      "loss": 2.2368,
+      "step": 546
+    },
+    {
+      "epoch": 3.7465753424657535,
+      "grad_norm": 0.7061879454542815,
+      "learning_rate": 5.3854765691117306e-08,
+      "loss": 2.177,
+      "step": 547
+    },
+    {
+      "epoch": 3.7534246575342465,
+      "grad_norm": 0.7139675704777505,
+      "learning_rate": 5.099283973795111e-08,
+      "loss": 2.2418,
+      "step": 548
+    },
+    {
+      "epoch": 3.76027397260274,
+      "grad_norm": 0.6881741924415452,
+      "learning_rate": 4.8208264626047883e-08,
+      "loss": 2.2589,
+      "step": 549
+    },
+    {
+      "epoch": 3.767123287671233,
+      "grad_norm": 0.678483134148144,
+      "learning_rate": 4.550112830501802e-08,
+      "loss": 2.1943,
+      "step": 550
+    },
+    {
+      "epoch": 3.7739726027397262,
+      "grad_norm": 0.7322395480739227,
+      "learning_rate": 4.287151627860281e-08,
+      "loss": 2.2337,
+      "step": 551
+    },
+    {
+      "epoch": 3.780821917808219,
+      "grad_norm": 0.6991597575458696,
+      "learning_rate": 4.0319511601969454e-08,
+      "loss": 2.2939,
+      "step": 552
+    },
+    {
+      "epoch": 3.787671232876712,
+      "grad_norm": 0.7395716381497617,
+      "learning_rate": 3.784519487909255e-08,
+      "loss": 2.1576,
+      "step": 553
+    },
+    {
+      "epoch": 3.7945205479452055,
+      "grad_norm": 0.708999098284432,
+      "learning_rate": 3.544864426020478e-08,
+      "loss": 2.1984,
+      "step": 554
+    },
+    {
+      "epoch": 3.8013698630136985,
+      "grad_norm": 0.7188186346127752,
+      "learning_rate": 3.3129935439329965e-08,
+      "loss": 2.2419,
+      "step": 555
+    },
+    {
+      "epoch": 3.8013698630136985,
+      "eval_loss": 2.4143056869506836,
+      "eval_runtime": 6.0604,
+      "eval_samples_per_second": 51.481,
+      "eval_steps_per_second": 3.3,
+      "step": 555
+    },
+    {
+      "epoch": 3.808219178082192,
+      "grad_norm": 0.6933486224482055,
+      "learning_rate": 3.08891416518925e-08,
+      "loss": 2.2608,
+      "step": 556
+    },
+    {
+      "epoch": 3.815068493150685,
+      "grad_norm": 0.7189668885334565,
+      "learning_rate": 2.8726333672402794e-08,
+      "loss": 2.2518,
+      "step": 557
+    },
+    {
+      "epoch": 3.821917808219178,
+      "grad_norm": 0.7270031341616389,
+      "learning_rate": 2.6641579812224373e-08,
+      "loss": 2.143,
+      "step": 558
+    },
+    {
+      "epoch": 3.828767123287671,
+      "grad_norm": 0.737409588056891,
+      "learning_rate": 2.4634945917414166e-08,
+      "loss": 2.276,
+      "step": 559
+    },
+    {
+      "epoch": 3.8356164383561646,
+      "grad_norm": 0.7123900534788905,
+      "learning_rate": 2.27064953666431e-08,
+      "loss": 2.2219,
+      "step": 560
+    },
+    {
+      "epoch": 3.8424657534246576,
+      "grad_norm": 0.705372834056662,
+      "learning_rate": 2.0856289069195456e-08,
+      "loss": 2.1854,
+      "step": 561
+    },
+    {
+      "epoch": 3.8493150684931505,
+      "grad_norm": 0.6919438600438174,
+      "learning_rate": 1.9084385463043476e-08,
+      "loss": 2.2125,
+      "step": 562
+    },
+    {
+      "epoch": 3.856164383561644,
+      "grad_norm": 0.6986694089843037,
+      "learning_rate": 1.7390840513003548e-08,
+      "loss": 2.1738,
+      "step": 563
+    },
+    {
+      "epoch": 3.863013698630137,
+      "grad_norm": 0.687054683313373,
+      "learning_rate": 1.5775707708966247e-08,
+      "loss": 2.253,
+      "step": 564
+    },
+    {
+      "epoch": 3.8698630136986303,
+      "grad_norm": 0.7211573427191927,
+      "learning_rate": 1.4239038064209342e-08,
+      "loss": 2.2823,
+      "step": 565
+    },
+    {
+      "epoch": 3.8767123287671232,
+      "grad_norm": 0.7516891005428765,
+      "learning_rate": 1.2780880113784366e-08,
+      "loss": 2.2602,
+      "step": 566
+    },
+    {
+      "epoch": 3.883561643835616,
+      "grad_norm": 0.7289591089911002,
+      "learning_rate": 1.1401279912985342e-08,
+      "loss": 2.1369,
+      "step": 567
+    },
+    {
+      "epoch": 3.8904109589041096,
+      "grad_norm": 0.7077741511239809,
+      "learning_rate": 1.0100281035893277e-08,
+      "loss": 2.116,
+      "step": 568
+    },
+    {
+      "epoch": 3.897260273972603,
+      "grad_norm": 0.7449785250693013,
+      "learning_rate": 8.877924573999497e-09,
+      "loss": 2.2017,
+      "step": 569
+    },
+    {
+      "epoch": 3.904109589041096,
+      "grad_norm": 0.7175340509888491,
+      "learning_rate": 7.734249134909166e-09,
+      "loss": 2.2039,
+      "step": 570
+    },
+    {
+      "epoch": 3.910958904109589,
+      "grad_norm": 0.7465450819011743,
+      "learning_rate": 6.66929084112089e-09,
+      "loss": 2.2041,
+      "step": 571
+    },
+    {
+      "epoch": 3.9178082191780823,
+      "grad_norm": 0.7505837621338696,
+      "learning_rate": 5.683083328885397e-09,
+      "loss": 2.2363,
+      "step": 572
+    },
+    {
+      "epoch": 3.9246575342465753,
+      "grad_norm": 0.7012438494498914,
+      "learning_rate": 4.775657747144447e-09,
+      "loss": 2.2756,
+      "step": 573
+    },
+    {
+      "epoch": 3.9315068493150687,
+      "grad_norm": 0.6931662637268987,
+      "learning_rate": 3.94704275654606e-09,
+      "loss": 2.3161,
+      "step": 574
+    },
+    {
+      "epoch": 3.9383561643835616,
+      "grad_norm": 0.6811252463311278,
+      "learning_rate": 3.1972645285391344e-09,
+      "loss": 2.2709,
+      "step": 575
+    },
+    {
+      "epoch": 3.9452054794520546,
+      "grad_norm": 0.7228500421816013,
+      "learning_rate": 2.5263467445479916e-09,
+      "loss": 2.229,
+      "step": 576
+    },
+    {
+      "epoch": 3.952054794520548,
+      "grad_norm": 0.6973647416169606,
+      "learning_rate": 1.934310595222977e-09,
+      "loss": 2.1544,
+      "step": 577
+    },
+    {
+      "epoch": 3.958904109589041,
+      "grad_norm": 0.7092268184710049,
+      "learning_rate": 1.421174779772383e-09,
+      "loss": 2.125,
+      "step": 578
+    },
+    {
+      "epoch": 3.9657534246575343,
+      "grad_norm": 0.7130433388265456,
+      "learning_rate": 9.869555053704239e-10,
+      "loss": 2.2239,
+      "step": 579
+    },
+    {
+      "epoch": 3.9726027397260273,
+      "grad_norm": 0.7016357741558538,
+      "learning_rate": 6.316664866470867e-10,
+      "loss": 2.2333,
+      "step": 580
+    },
+    {
+      "epoch": 3.9794520547945207,
+      "grad_norm": 0.7133790437761107,
+      "learning_rate": 3.5531894525375753e-10,
+      "loss": 2.2426,
+      "step": 581
+    },
+    {
+      "epoch": 3.9863013698630136,
+      "grad_norm": 0.6940646116522826,
+      "learning_rate": 1.579216095087821e-10,
+      "loss": 2.2718,
+      "step": 582
+    },
+    {
+      "epoch": 3.993150684931507,
+      "grad_norm": 0.6894067976681987,
+      "learning_rate": 3.94807141224085e-11,
+      "loss": 2.2517,
+      "step": 583
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7502824336631434,
+      "learning_rate": 0.0,
+      "loss": 2.2361,
+      "step": 584
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 584,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 73,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 366833156751360.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}