diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,5144 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 730,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0013698630136986301,
+      "grad_norm": 0.9543317556381226,
+      "learning_rate": 0.0,
+      "loss": 1.289506435394287,
+      "step": 1
+    },
+    {
+      "epoch": 0.0027397260273972603,
+      "grad_norm": 1.102232575416565,
+      "learning_rate": 4.5454545454545457e-07,
+      "loss": 1.419710397720337,
+      "step": 2
+    },
+    {
+      "epoch": 0.00410958904109589,
+      "grad_norm": 1.055787205696106,
+      "learning_rate": 9.090909090909091e-07,
+      "loss": 1.2828402519226074,
+      "step": 3
+    },
+    {
+      "epoch": 0.005479452054794521,
+      "grad_norm": 1.1042431592941284,
+      "learning_rate": 1.3636363636363636e-06,
+      "loss": 1.3904743194580078,
+      "step": 4
+    },
+    {
+      "epoch": 0.00684931506849315,
+      "grad_norm": 0.9779807925224304,
+      "learning_rate": 1.8181818181818183e-06,
+      "loss": 1.307753086090088,
+      "step": 5
+    },
+    {
+      "epoch": 0.00821917808219178,
+      "grad_norm": 0.9276996850967407,
+      "learning_rate": 2.2727272727272728e-06,
+      "loss": 1.3470945358276367,
+      "step": 6
+    },
+    {
+      "epoch": 0.009589041095890411,
+      "grad_norm": 0.8735243678092957,
+      "learning_rate": 2.7272727272727272e-06,
+      "loss": 1.2755506038665771,
+      "step": 7
+    },
+    {
+      "epoch": 0.010958904109589041,
+      "grad_norm": 0.7032255530357361,
+      "learning_rate": 3.181818181818182e-06,
+      "loss": 1.2992908954620361,
+      "step": 8
+    },
+    {
+      "epoch": 0.012328767123287671,
+      "grad_norm": 0.6715389490127563,
+      "learning_rate": 3.6363636363636366e-06,
+      "loss": 1.2155871391296387,
+      "step": 9
+    },
+    {
+      "epoch": 0.0136986301369863,
+      "grad_norm": 0.5294431447982788,
+      "learning_rate": 4.0909090909090915e-06,
+      "loss": 1.0370020866394043,
+      "step": 10
+    },
+    {
+      "epoch": 0.015068493150684932,
+      "grad_norm": 0.45838990807533264,
+      "learning_rate": 4.5454545454545455e-06,
+      "loss": 1.0069210529327393,
+      "step": 11
+    },
+    {
+      "epoch": 0.01643835616438356,
+      "grad_norm": 0.40803831815719604,
+      "learning_rate": 5e-06,
+      "loss": 1.0808029174804688,
+      "step": 12
+    },
+    {
+      "epoch": 0.01780821917808219,
+      "grad_norm": 0.40420499444007874,
+      "learning_rate": 5.4545454545454545e-06,
+      "loss": 1.0166305303573608,
+      "step": 13
+    },
+    {
+      "epoch": 0.019178082191780823,
+      "grad_norm": 0.3624785244464874,
+      "learning_rate": 5.90909090909091e-06,
+      "loss": 0.9620136618614197,
+      "step": 14
+    },
+    {
+      "epoch": 0.02054794520547945,
+      "grad_norm": 0.3782968521118164,
+      "learning_rate": 6.363636363636364e-06,
+      "loss": 1.0190021991729736,
+      "step": 15
+    },
+    {
+      "epoch": 0.021917808219178082,
+      "grad_norm": 0.3190377652645111,
+      "learning_rate": 6.818181818181818e-06,
+      "loss": 0.8993116617202759,
+      "step": 16
+    },
+    {
+      "epoch": 0.023287671232876714,
+      "grad_norm": 0.3656272888183594,
+      "learning_rate": 7.272727272727273e-06,
+      "loss": 1.0050773620605469,
+      "step": 17
+    },
+    {
+      "epoch": 0.024657534246575342,
+      "grad_norm": 0.2802956998348236,
+      "learning_rate": 7.727272727272727e-06,
+      "loss": 0.9016516804695129,
+      "step": 18
+    },
+    {
+      "epoch": 0.026027397260273973,
+      "grad_norm": 0.26026973128318787,
+      "learning_rate": 8.181818181818183e-06,
+      "loss": 0.8167034387588501,
+      "step": 19
+    },
+    {
+      "epoch": 0.0273972602739726,
+      "grad_norm": 0.2704644203186035,
+      "learning_rate": 8.636363636363637e-06,
+      "loss": 0.9548209309577942,
+      "step": 20
+    },
+    {
+      "epoch": 0.028767123287671233,
+      "grad_norm": 0.22691063582897186,
+      "learning_rate": 9.090909090909091e-06,
+      "loss": 0.8790600299835205,
+      "step": 21
+    },
+    {
+      "epoch": 0.030136986301369864,
+      "grad_norm": 0.22116470336914062,
+      "learning_rate": 9.545454545454547e-06,
+      "loss": 0.8698333501815796,
+      "step": 22
+    },
+    {
+      "epoch": 0.031506849315068496,
+      "grad_norm": 0.21417191624641418,
+      "learning_rate": 1e-05,
+      "loss": 0.8363385200500488,
+      "step": 23
+    },
+    {
+      "epoch": 0.03287671232876712,
+      "grad_norm": 0.24161672592163086,
+      "learning_rate": 1.0454545454545455e-05,
+      "loss": 0.9342292547225952,
+      "step": 24
+    },
+    {
+      "epoch": 0.03424657534246575,
+      "grad_norm": 0.21060842275619507,
+      "learning_rate": 1.0909090909090909e-05,
+      "loss": 0.8269792795181274,
+      "step": 25
+    },
+    {
+      "epoch": 0.03561643835616438,
+      "grad_norm": 0.2038818746805191,
+      "learning_rate": 1.1363636363636366e-05,
+      "loss": 0.7519025802612305,
+      "step": 26
+    },
+    {
+      "epoch": 0.036986301369863014,
+      "grad_norm": 0.2230827510356903,
+      "learning_rate": 1.181818181818182e-05,
+      "loss": 0.8076636791229248,
+      "step": 27
+    },
+    {
+      "epoch": 0.038356164383561646,
+      "grad_norm": 0.18758194148540497,
+      "learning_rate": 1.2272727272727274e-05,
+      "loss": 0.7615480422973633,
+      "step": 28
+    },
+    {
+      "epoch": 0.03972602739726028,
+      "grad_norm": 0.195050448179245,
+      "learning_rate": 1.2727272727272728e-05,
+      "loss": 0.8048295378684998,
+      "step": 29
+    },
+    {
+      "epoch": 0.0410958904109589,
+      "grad_norm": 0.1838914453983307,
+      "learning_rate": 1.3181818181818183e-05,
+      "loss": 0.7350910902023315,
+      "step": 30
+    },
+    {
+      "epoch": 0.04246575342465753,
+      "grad_norm": 0.20437878370285034,
+      "learning_rate": 1.3636363636363637e-05,
+      "loss": 0.8430003523826599,
+      "step": 31
+    },
+    {
+      "epoch": 0.043835616438356165,
+      "grad_norm": 0.20855139195919037,
+      "learning_rate": 1.4090909090909092e-05,
+      "loss": 0.752582848072052,
+      "step": 32
+    },
+    {
+      "epoch": 0.045205479452054796,
+      "grad_norm": 0.21374471485614777,
+      "learning_rate": 1.4545454545454546e-05,
+      "loss": 0.7721849679946899,
+      "step": 33
+    },
+    {
+      "epoch": 0.04657534246575343,
+      "grad_norm": 0.19961094856262207,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.7583755850791931,
+      "step": 34
+    },
+    {
+      "epoch": 0.04794520547945205,
+      "grad_norm": 0.17798861861228943,
+      "learning_rate": 1.5454545454545454e-05,
+      "loss": 0.7548345327377319,
+      "step": 35
+    },
+    {
+      "epoch": 0.049315068493150684,
+      "grad_norm": 0.19771678745746613,
+      "learning_rate": 1.590909090909091e-05,
+      "loss": 0.7899003028869629,
+      "step": 36
+    },
+    {
+      "epoch": 0.050684931506849315,
+      "grad_norm": 0.18502211570739746,
+      "learning_rate": 1.6363636363636366e-05,
+      "loss": 0.7351113557815552,
+      "step": 37
+    },
+    {
+      "epoch": 0.052054794520547946,
+      "grad_norm": 0.2032163441181183,
+      "learning_rate": 1.681818181818182e-05,
+      "loss": 0.809609055519104,
+      "step": 38
+    },
+    {
+      "epoch": 0.05342465753424658,
+      "grad_norm": 0.17767126858234406,
+      "learning_rate": 1.7272727272727274e-05,
+      "loss": 0.6977474689483643,
+      "step": 39
+    },
+    {
+      "epoch": 0.0547945205479452,
+      "grad_norm": 0.1885196417570114,
+      "learning_rate": 1.772727272727273e-05,
+      "loss": 0.7495503425598145,
+      "step": 40
+    },
+    {
+      "epoch": 0.056164383561643834,
+      "grad_norm": 0.19037273526191711,
+      "learning_rate": 1.8181818181818182e-05,
+      "loss": 0.7687135338783264,
+      "step": 41
+    },
+    {
+      "epoch": 0.057534246575342465,
+      "grad_norm": 0.20152418315410614,
+      "learning_rate": 1.8636363636363638e-05,
+      "loss": 0.7619070410728455,
+      "step": 42
+    },
+    {
+      "epoch": 0.0589041095890411,
+      "grad_norm": 0.17043539881706238,
+      "learning_rate": 1.9090909090909094e-05,
+      "loss": 0.7616786360740662,
+      "step": 43
+    },
+    {
+      "epoch": 0.06027397260273973,
+      "grad_norm": 0.193017840385437,
+      "learning_rate": 1.9545454545454546e-05,
+      "loss": 0.6300766468048096,
+      "step": 44
+    },
+    {
+      "epoch": 0.06164383561643835,
+      "grad_norm": 0.17710553109645844,
+      "learning_rate": 2e-05,
+      "loss": 0.7160718441009521,
+      "step": 45
+    },
+    {
+      "epoch": 0.06301369863013699,
+      "grad_norm": 0.17088574171066284,
+      "learning_rate": 1.9999975388217704e-05,
+      "loss": 0.6980305314064026,
+      "step": 46
+    },
+    {
+      "epoch": 0.06438356164383562,
+      "grad_norm": 0.17629273235797882,
+      "learning_rate": 1.9999901552991966e-05,
+      "loss": 0.7771427035331726,
+      "step": 47
+    },
+    {
+      "epoch": 0.06575342465753424,
+      "grad_norm": 0.18566235899925232,
+      "learning_rate": 1.9999778494686226e-05,
+      "loss": 0.6564210057258606,
+      "step": 48
+    },
+    {
+      "epoch": 0.06712328767123288,
+      "grad_norm": 0.15922220051288605,
+      "learning_rate": 1.999960621390622e-05,
+      "loss": 0.6578161716461182,
+      "step": 49
+    },
+    {
+      "epoch": 0.0684931506849315,
+      "grad_norm": 0.17897970974445343,
+      "learning_rate": 1.999938471149998e-05,
+      "loss": 0.7257359623908997,
+      "step": 50
+    },
+    {
+      "epoch": 0.06986301369863014,
+      "grad_norm": 0.17765460908412933,
+      "learning_rate": 1.999911398855782e-05,
+      "loss": 0.7241641879081726,
+      "step": 51
+    },
+    {
+      "epoch": 0.07123287671232877,
+      "grad_norm": 0.1779472678899765,
+      "learning_rate": 1.999879404641233e-05,
+      "loss": 0.696056604385376,
+      "step": 52
+    },
+    {
+      "epoch": 0.07260273972602739,
+      "grad_norm": 0.18799914419651031,
+      "learning_rate": 1.9998424886638382e-05,
+      "loss": 0.7142607569694519,
+      "step": 53
+    },
+    {
+      "epoch": 0.07397260273972603,
+      "grad_norm": 0.17217737436294556,
+      "learning_rate": 1.999800651105311e-05,
+      "loss": 0.7063297033309937,
+      "step": 54
+    },
+    {
+      "epoch": 0.07534246575342465,
+      "grad_norm": 0.18024404346942902,
+      "learning_rate": 1.999753892171591e-05,
+      "loss": 0.741089940071106,
+      "step": 55
+    },
+    {
+      "epoch": 0.07671232876712329,
+      "grad_norm": 0.16077029705047607,
+      "learning_rate": 1.9997022120928423e-05,
+      "loss": 0.7010071277618408,
+      "step": 56
+    },
+    {
+      "epoch": 0.07808219178082192,
+      "grad_norm": 0.1703995168209076,
+      "learning_rate": 1.999645611123453e-05,
+      "loss": 0.6677719354629517,
+      "step": 57
+    },
+    {
+      "epoch": 0.07945205479452055,
+      "grad_norm": 0.16977134346961975,
+      "learning_rate": 1.9995840895420323e-05,
+      "loss": 0.7329045534133911,
+      "step": 58
+    },
+    {
+      "epoch": 0.08082191780821918,
+      "grad_norm": 0.17480944097042084,
+      "learning_rate": 1.999517647651412e-05,
+      "loss": 0.7148158550262451,
+      "step": 59
+    },
+    {
+      "epoch": 0.0821917808219178,
+      "grad_norm": 0.1859695464372635,
+      "learning_rate": 1.9994462857786433e-05,
+      "loss": 0.6804654598236084,
+      "step": 60
+    },
+    {
+      "epoch": 0.08356164383561644,
+      "grad_norm": 0.1692071110010147,
+      "learning_rate": 1.9993700042749937e-05,
+      "loss": 0.6517329216003418,
+      "step": 61
+    },
+    {
+      "epoch": 0.08493150684931507,
+      "grad_norm": 0.16654057800769806,
+      "learning_rate": 1.9992888035159484e-05,
+      "loss": 0.6754597425460815,
+      "step": 62
+    },
+    {
+      "epoch": 0.0863013698630137,
+      "grad_norm": 0.16556744277477264,
+      "learning_rate": 1.999202683901207e-05,
+      "loss": 0.6628267765045166,
+      "step": 63
+    },
+    {
+      "epoch": 0.08767123287671233,
+      "grad_norm": 0.17565493285655975,
+      "learning_rate": 1.99911164585468e-05,
+      "loss": 0.5974063873291016,
+      "step": 64
+    },
+    {
+      "epoch": 0.08904109589041095,
+      "grad_norm": 0.17702734470367432,
+      "learning_rate": 1.99901568982449e-05,
+      "loss": 0.7042257189750671,
+      "step": 65
+    },
+    {
+      "epoch": 0.09041095890410959,
+      "grad_norm": 0.18799234926700592,
+      "learning_rate": 1.9989148162829663e-05,
+      "loss": 0.6061130166053772,
+      "step": 66
+    },
+    {
+      "epoch": 0.09178082191780822,
+      "grad_norm": 0.1799335926771164,
+      "learning_rate": 1.9988090257266442e-05,
+      "loss": 0.6730984449386597,
+      "step": 67
+    },
+    {
+      "epoch": 0.09315068493150686,
+      "grad_norm": 0.18967068195343018,
+      "learning_rate": 1.998698318676263e-05,
+      "loss": 0.7190222144126892,
+      "step": 68
+    },
+    {
+      "epoch": 0.09452054794520548,
+      "grad_norm": 0.18353059887886047,
+      "learning_rate": 1.998582695676762e-05,
+      "loss": 0.660584568977356,
+      "step": 69
+    },
+    {
+      "epoch": 0.0958904109589041,
+      "grad_norm": 0.1780116707086563,
+      "learning_rate": 1.998462157297279e-05,
+      "loss": 0.7072656154632568,
+      "step": 70
+    },
+    {
+      "epoch": 0.09726027397260274,
+      "grad_norm": 0.16367685794830322,
+      "learning_rate": 1.998336704131147e-05,
+      "loss": 0.6226505041122437,
+      "step": 71
+    },
+    {
+      "epoch": 0.09863013698630137,
+      "grad_norm": 0.18768785893917084,
+      "learning_rate": 1.9982063367958907e-05,
+      "loss": 0.6319016218185425,
+      "step": 72
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.17008233070373535,
+      "learning_rate": 1.9980710559332248e-05,
+      "loss": 0.6461281776428223,
+      "step": 73
+    },
+    {
+      "epoch": 0.10136986301369863,
+      "grad_norm": 0.18165260553359985,
+      "learning_rate": 1.9979308622090503e-05,
+      "loss": 0.6717009544372559,
+      "step": 74
+    },
+    {
+      "epoch": 0.10273972602739725,
+      "grad_norm": 0.17221364378929138,
+      "learning_rate": 1.9977857563134503e-05,
+      "loss": 0.6939575672149658,
+      "step": 75
+    },
+    {
+      "epoch": 0.10410958904109589,
+      "grad_norm": 0.16935478150844574,
+      "learning_rate": 1.9976357389606878e-05,
+      "loss": 0.6323124170303345,
+      "step": 76
+    },
+    {
+      "epoch": 0.10547945205479452,
+      "grad_norm": 0.18027523159980774,
+      "learning_rate": 1.9974808108892017e-05,
+      "loss": 0.6536504626274109,
+      "step": 77
+    },
+    {
+      "epoch": 0.10684931506849316,
+      "grad_norm": 0.17837466299533844,
+      "learning_rate": 1.9973209728616032e-05,
+      "loss": 0.6764005422592163,
+      "step": 78
+    },
+    {
+      "epoch": 0.10821917808219178,
+      "grad_norm": 0.17923392355442047,
+      "learning_rate": 1.997156225664672e-05,
+      "loss": 0.64453125,
+      "step": 79
+    },
+    {
+      "epoch": 0.1095890410958904,
+      "grad_norm": 0.17504896223545074,
+      "learning_rate": 1.996986570109353e-05,
+      "loss": 0.7095593214035034,
+      "step": 80
+    },
+    {
+      "epoch": 0.11095890410958904,
+      "grad_norm": 0.16979750990867615,
+      "learning_rate": 1.9968120070307503e-05,
+      "loss": 0.7190498113632202,
+      "step": 81
+    },
+    {
+      "epoch": 0.11232876712328767,
+      "grad_norm": 0.19454807043075562,
+      "learning_rate": 1.9966325372881265e-05,
+      "loss": 0.7216591238975525,
+      "step": 82
+    },
+    {
+      "epoch": 0.1136986301369863,
+      "grad_norm": 0.1744404286146164,
+      "learning_rate": 1.9964481617648954e-05,
+      "loss": 0.6529302597045898,
+      "step": 83
+    },
+    {
+      "epoch": 0.11506849315068493,
+      "grad_norm": 0.17792807519435883,
+      "learning_rate": 1.9962588813686187e-05,
+      "loss": 0.6740435361862183,
+      "step": 84
+    },
+    {
+      "epoch": 0.11643835616438356,
+      "grad_norm": 0.18751150369644165,
+      "learning_rate": 1.9960646970310027e-05,
+      "loss": 0.713350236415863,
+      "step": 85
+    },
+    {
+      "epoch": 0.1178082191780822,
+      "grad_norm": 0.1781667172908783,
+      "learning_rate": 1.9958656097078912e-05,
+      "loss": 0.649537205696106,
+      "step": 86
+    },
+    {
+      "epoch": 0.11917808219178082,
+      "grad_norm": 0.171499565243721,
+      "learning_rate": 1.9956616203792636e-05,
+      "loss": 0.6647717952728271,
+      "step": 87
+    },
+    {
+      "epoch": 0.12054794520547946,
+      "grad_norm": 0.1776268035173416,
+      "learning_rate": 1.9954527300492277e-05,
+      "loss": 0.6820968389511108,
+      "step": 88
+    },
+    {
+      "epoch": 0.12191780821917808,
+      "grad_norm": 0.1685784012079239,
+      "learning_rate": 1.995238939746016e-05,
+      "loss": 0.6538463234901428,
+      "step": 89
+    },
+    {
+      "epoch": 0.1232876712328767,
+      "grad_norm": 0.1900128871202469,
+      "learning_rate": 1.9950202505219808e-05,
+      "loss": 0.6783770322799683,
+      "step": 90
+    },
+    {
+      "epoch": 0.12465753424657534,
+      "grad_norm": 0.17134107649326324,
+      "learning_rate": 1.9947966634535887e-05,
+      "loss": 0.6427039504051208,
+      "step": 91
+    },
+    {
+      "epoch": 0.12602739726027398,
+      "grad_norm": 0.17828185856342316,
+      "learning_rate": 1.9945681796414146e-05,
+      "loss": 0.6815697550773621,
+      "step": 92
+    },
+    {
+      "epoch": 0.1273972602739726,
+      "grad_norm": 0.18462392687797546,
+      "learning_rate": 1.9943348002101374e-05,
+      "loss": 0.6666452288627625,
+      "step": 93
+    },
+    {
+      "epoch": 0.12876712328767123,
+      "grad_norm": 0.17112930119037628,
+      "learning_rate": 1.9940965263085335e-05,
+      "loss": 0.6950005292892456,
+      "step": 94
+    },
+    {
+      "epoch": 0.13013698630136986,
+      "grad_norm": 0.16788211464881897,
+      "learning_rate": 1.9938533591094728e-05,
+      "loss": 0.6137875318527222,
+      "step": 95
+    },
+    {
+      "epoch": 0.13150684931506848,
+      "grad_norm": 0.1813577562570572,
+      "learning_rate": 1.99360529980991e-05,
+      "loss": 0.7134063243865967,
+      "step": 96
+    },
+    {
+      "epoch": 0.13287671232876713,
+      "grad_norm": 0.17547398805618286,
+      "learning_rate": 1.993352349630882e-05,
+      "loss": 0.6599196791648865,
+      "step": 97
+    },
+    {
+      "epoch": 0.13424657534246576,
+      "grad_norm": 0.16298982501029968,
+      "learning_rate": 1.9930945098174994e-05,
+      "loss": 0.5956114530563354,
+      "step": 98
+    },
+    {
+      "epoch": 0.13561643835616438,
+      "grad_norm": 0.1693538874387741,
+      "learning_rate": 1.9928317816389416e-05,
+      "loss": 0.6117294430732727,
+      "step": 99
+    },
+    {
+      "epoch": 0.136986301369863,
+      "grad_norm": 0.17330867052078247,
+      "learning_rate": 1.992564166388451e-05,
+      "loss": 0.6786526441574097,
+      "step": 100
+    },
+    {
+      "epoch": 0.13835616438356163,
+      "grad_norm": 0.17630691826343536,
+      "learning_rate": 1.992291665383325e-05,
+      "loss": 0.7339305877685547,
+      "step": 101
+    },
+    {
+      "epoch": 0.13972602739726028,
+      "grad_norm": 0.18277214467525482,
+      "learning_rate": 1.9920142799649098e-05,
+      "loss": 0.6647192239761353,
+      "step": 102
+    },
+    {
+      "epoch": 0.1410958904109589,
+      "grad_norm": 0.1676325649023056,
+      "learning_rate": 1.9917320114985964e-05,
+      "loss": 0.7023746967315674,
+      "step": 103
+    },
+    {
+      "epoch": 0.14246575342465753,
+      "grad_norm": 0.17900800704956055,
+      "learning_rate": 1.9914448613738107e-05,
+      "loss": 0.6510761976242065,
+      "step": 104
+    },
+    {
+      "epoch": 0.14383561643835616,
+      "grad_norm": 0.15803074836730957,
+      "learning_rate": 1.9911528310040073e-05,
+      "loss": 0.6323767900466919,
+      "step": 105
+    },
+    {
+      "epoch": 0.14520547945205478,
+      "grad_norm": 0.190176859498024,
+      "learning_rate": 1.9908559218266644e-05,
+      "loss": 0.7440241575241089,
+      "step": 106
+    },
+    {
+      "epoch": 0.14657534246575343,
+      "grad_norm": 0.1805161088705063,
+      "learning_rate": 1.9905541353032744e-05,
+      "loss": 0.6736125946044922,
+      "step": 107
+    },
+    {
+      "epoch": 0.14794520547945206,
+      "grad_norm": 0.16393084824085236,
+      "learning_rate": 1.9902474729193385e-05,
+      "loss": 0.639750063419342,
+      "step": 108
+    },
+    {
+      "epoch": 0.14931506849315068,
+      "grad_norm": 0.1559421420097351,
+      "learning_rate": 1.989935936184358e-05,
+      "loss": 0.6098452210426331,
+      "step": 109
+    },
+    {
+      "epoch": 0.1506849315068493,
+      "grad_norm": 0.1748615801334381,
+      "learning_rate": 1.989619526631828e-05,
+      "loss": 0.6362285614013672,
+      "step": 110
+    },
+    {
+      "epoch": 0.15205479452054796,
+      "grad_norm": 0.15956446528434753,
+      "learning_rate": 1.9892982458192286e-05,
+      "loss": 0.6010555028915405,
+      "step": 111
+    },
+    {
+      "epoch": 0.15342465753424658,
+      "grad_norm": 0.17703907191753387,
+      "learning_rate": 1.988972095328019e-05,
+      "loss": 0.6596381664276123,
+      "step": 112
+    },
+    {
+      "epoch": 0.1547945205479452,
+      "grad_norm": 0.17541560530662537,
+      "learning_rate": 1.9886410767636284e-05,
+      "loss": 0.6365323066711426,
+      "step": 113
+    },
+    {
+      "epoch": 0.15616438356164383,
+      "grad_norm": 0.1705840677022934,
+      "learning_rate": 1.9883051917554473e-05,
+      "loss": 0.6401112079620361,
+      "step": 114
+    },
+    {
+      "epoch": 0.15753424657534246,
+      "grad_norm": 0.18547984957695007,
+      "learning_rate": 1.987964441956822e-05,
+      "loss": 0.680147647857666,
+      "step": 115
+    },
+    {
+      "epoch": 0.1589041095890411,
+      "grad_norm": 0.157182514667511,
+      "learning_rate": 1.9876188290450445e-05,
+      "loss": 0.6003460884094238,
+      "step": 116
+    },
+    {
+      "epoch": 0.16027397260273973,
+      "grad_norm": 0.1699901968240738,
+      "learning_rate": 1.9872683547213446e-05,
+      "loss": 0.5792864561080933,
+      "step": 117
+    },
+    {
+      "epoch": 0.16164383561643836,
+      "grad_norm": 0.17358699440956116,
+      "learning_rate": 1.986913020710882e-05,
+      "loss": 0.6301059722900391,
+      "step": 118
+    },
+    {
+      "epoch": 0.16301369863013698,
+      "grad_norm": 0.18124189972877502,
+      "learning_rate": 1.9865528287627372e-05,
+      "loss": 0.6473714113235474,
+      "step": 119
+    },
+    {
+      "epoch": 0.1643835616438356,
+      "grad_norm": 0.1570880264043808,
+      "learning_rate": 1.9861877806499033e-05,
+      "loss": 0.6518967747688293,
+      "step": 120
+    },
+    {
+      "epoch": 0.16575342465753426,
+      "grad_norm": 0.16372060775756836,
+      "learning_rate": 1.9858178781692777e-05,
+      "loss": 0.5721158981323242,
+      "step": 121
+    },
+    {
+      "epoch": 0.16712328767123288,
+      "grad_norm": 0.1550905704498291,
+      "learning_rate": 1.9854431231416515e-05,
+      "loss": 0.6409134864807129,
+      "step": 122
+    },
+    {
+      "epoch": 0.1684931506849315,
+      "grad_norm": 0.1671498715877533,
+      "learning_rate": 1.9850635174117033e-05,
+      "loss": 0.6450335383415222,
+      "step": 123
+    },
+    {
+      "epoch": 0.16986301369863013,
+      "grad_norm": 0.1591179221868515,
+      "learning_rate": 1.9846790628479874e-05,
+      "loss": 0.6999216079711914,
+      "step": 124
+    },
+    {
+      "epoch": 0.17123287671232876,
+      "grad_norm": 0.17138256132602692,
+      "learning_rate": 1.984289761342926e-05,
+      "loss": 0.6327433586120605,
+      "step": 125
+    },
+    {
+      "epoch": 0.1726027397260274,
+      "grad_norm": 0.18257306516170502,
+      "learning_rate": 1.9838956148128004e-05,
+      "loss": 0.6688971519470215,
+      "step": 126
+    },
+    {
+      "epoch": 0.17397260273972603,
+      "grad_norm": 0.15964193642139435,
+      "learning_rate": 1.98349662519774e-05,
+      "loss": 0.6297944188117981,
+      "step": 127
+    },
+    {
+      "epoch": 0.17534246575342466,
+      "grad_norm": 0.17521503567695618,
+      "learning_rate": 1.983092794461714e-05,
+      "loss": 0.6212207674980164,
+      "step": 128
+    },
+    {
+      "epoch": 0.17671232876712328,
+      "grad_norm": 0.15988564491271973,
+      "learning_rate": 1.982684124592521e-05,
+      "loss": 0.5915682315826416,
+      "step": 129
+    },
+    {
+      "epoch": 0.1780821917808219,
+      "grad_norm": 0.16260051727294922,
+      "learning_rate": 1.9822706176017802e-05,
+      "loss": 0.6029642820358276,
+      "step": 130
+    },
+    {
+      "epoch": 0.17945205479452056,
+      "grad_norm": 0.16412433981895447,
+      "learning_rate": 1.9818522755249203e-05,
+      "loss": 0.6668821573257446,
+      "step": 131
+    },
+    {
+      "epoch": 0.18082191780821918,
+      "grad_norm": 0.16716210544109344,
+      "learning_rate": 1.9814291004211695e-05,
+      "loss": 0.5958414077758789,
+      "step": 132
+    },
+    {
+      "epoch": 0.1821917808219178,
+      "grad_norm": 0.17294858396053314,
+      "learning_rate": 1.981001094373548e-05,
+      "loss": 0.6482897996902466,
+      "step": 133
+    },
+    {
+      "epoch": 0.18356164383561643,
+      "grad_norm": 0.17001552879810333,
+      "learning_rate": 1.980568259488852e-05,
+      "loss": 0.7122180461883545,
+      "step": 134
+    },
+    {
+      "epoch": 0.18493150684931506,
+      "grad_norm": 0.16195915639400482,
+      "learning_rate": 1.980130597897651e-05,
+      "loss": 0.584160327911377,
+      "step": 135
+    },
+    {
+      "epoch": 0.1863013698630137,
+      "grad_norm": 0.17711922526359558,
+      "learning_rate": 1.9796881117542703e-05,
+      "loss": 0.620893657207489,
+      "step": 136
+    },
+    {
+      "epoch": 0.18767123287671234,
+      "grad_norm": 0.1668933480978012,
+      "learning_rate": 1.979240803236785e-05,
+      "loss": 0.6142734885215759,
+      "step": 137
+    },
+    {
+      "epoch": 0.18904109589041096,
+      "grad_norm": 0.16164347529411316,
+      "learning_rate": 1.9787886745470067e-05,
+      "loss": 0.6293851137161255,
+      "step": 138
+    },
+    {
+      "epoch": 0.19041095890410958,
+      "grad_norm": 0.15106764435768127,
+      "learning_rate": 1.9783317279104736e-05,
+      "loss": 0.6398227214813232,
+      "step": 139
+    },
+    {
+      "epoch": 0.1917808219178082,
+      "grad_norm": 0.16852658987045288,
+      "learning_rate": 1.977869965576441e-05,
+      "loss": 0.6542407274246216,
+      "step": 140
+    },
+    {
+      "epoch": 0.19315068493150686,
+      "grad_norm": 0.16625036299228668,
+      "learning_rate": 1.9774033898178668e-05,
+      "loss": 0.6504607200622559,
+      "step": 141
+    },
+    {
+      "epoch": 0.19452054794520549,
+      "grad_norm": 0.16894933581352234,
+      "learning_rate": 1.9769320029314037e-05,
+      "loss": 0.6176894903182983,
+      "step": 142
+    },
+    {
+      "epoch": 0.1958904109589041,
+      "grad_norm": 0.15120083093643188,
+      "learning_rate": 1.9764558072373855e-05,
+      "loss": 0.566133975982666,
+      "step": 143
+    },
+    {
+      "epoch": 0.19726027397260273,
+      "grad_norm": 0.1807292252779007,
+      "learning_rate": 1.9759748050798176e-05,
+      "loss": 0.6475929021835327,
+      "step": 144
+    },
+    {
+      "epoch": 0.19863013698630136,
+      "grad_norm": 0.18464243412017822,
+      "learning_rate": 1.975488998826364e-05,
+      "loss": 0.6212013959884644,
+      "step": 145
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.18034766614437103,
+      "learning_rate": 1.9749983908683362e-05,
+      "loss": 0.6008216142654419,
+      "step": 146
+    },
+    {
+      "epoch": 0.20136986301369864,
+      "grad_norm": 0.1733216792345047,
+      "learning_rate": 1.9745029836206813e-05,
+      "loss": 0.654267430305481,
+      "step": 147
+    },
+    {
+      "epoch": 0.20273972602739726,
+      "grad_norm": 0.1807546615600586,
+      "learning_rate": 1.9740027795219706e-05,
+      "loss": 0.6405000686645508,
+      "step": 148
+    },
+    {
+      "epoch": 0.20410958904109588,
+      "grad_norm": 0.17537952959537506,
+      "learning_rate": 1.9734977810343868e-05,
+      "loss": 0.6626770496368408,
+      "step": 149
+    },
+    {
+      "epoch": 0.2054794520547945,
+      "grad_norm": 0.16498853266239166,
+      "learning_rate": 1.9729879906437124e-05,
+      "loss": 0.6191319227218628,
+      "step": 150
+    },
+    {
+      "epoch": 0.20684931506849316,
+      "grad_norm": 0.1608179807662964,
+      "learning_rate": 1.972473410859318e-05,
+      "loss": 0.6513334512710571,
+      "step": 151
+    },
+    {
+      "epoch": 0.20821917808219179,
+      "grad_norm": 0.16782471537590027,
+      "learning_rate": 1.9719540442141477e-05,
+      "loss": 0.6400517225265503,
+      "step": 152
+    },
+    {
+      "epoch": 0.2095890410958904,
+      "grad_norm": 0.16582812368869781,
+      "learning_rate": 1.97142989326471e-05,
+      "loss": 0.6530036926269531,
+      "step": 153
+    },
+    {
+      "epoch": 0.21095890410958903,
+      "grad_norm": 0.1592300981283188,
+      "learning_rate": 1.9709009605910628e-05,
+      "loss": 0.5819565653800964,
+      "step": 154
+    },
+    {
+      "epoch": 0.21232876712328766,
+      "grad_norm": 0.1471843123435974,
+      "learning_rate": 1.970367248796801e-05,
+      "loss": 0.6289366483688354,
+      "step": 155
+    },
+    {
+      "epoch": 0.2136986301369863,
+      "grad_norm": 0.15616214275360107,
+      "learning_rate": 1.969828760509044e-05,
+      "loss": 0.6071575880050659,
+      "step": 156
+    },
+    {
+      "epoch": 0.21506849315068494,
+      "grad_norm": 0.15616996586322784,
+      "learning_rate": 1.9692854983784235e-05,
+      "loss": 0.570655345916748,
+      "step": 157
+    },
+    {
+      "epoch": 0.21643835616438356,
+      "grad_norm": 0.16667386889457703,
+      "learning_rate": 1.9687374650790693e-05,
+      "loss": 0.6458644866943359,
+      "step": 158
+    },
+    {
+      "epoch": 0.21780821917808219,
+      "grad_norm": 0.15973800420761108,
+      "learning_rate": 1.9681846633085968e-05,
+      "loss": 0.6385851502418518,
+      "step": 159
+    },
+    {
+      "epoch": 0.2191780821917808,
+      "grad_norm": 0.15689517557621002,
+      "learning_rate": 1.9676270957880928e-05,
+      "loss": 0.6954727172851562,
+      "step": 160
+    },
+    {
+      "epoch": 0.22054794520547946,
+      "grad_norm": 0.16799283027648926,
+      "learning_rate": 1.9670647652621044e-05,
+      "loss": 0.6535817384719849,
+      "step": 161
+    },
+    {
+      "epoch": 0.2219178082191781,
+      "grad_norm": 0.1637307107448578,
+      "learning_rate": 1.966497674498622e-05,
+      "loss": 0.6275250315666199,
+      "step": 162
+    },
+    {
+      "epoch": 0.2232876712328767,
+      "grad_norm": 0.1446286141872406,
+      "learning_rate": 1.9659258262890683e-05,
+      "loss": 0.6088711023330688,
+      "step": 163
+    },
+    {
+      "epoch": 0.22465753424657534,
+      "grad_norm": 0.16091319918632507,
+      "learning_rate": 1.965349223448285e-05,
+      "loss": 0.5717509984970093,
+      "step": 164
+    },
+    {
+      "epoch": 0.22602739726027396,
+      "grad_norm": 0.14229550957679749,
+      "learning_rate": 1.9647678688145163e-05,
+      "loss": 0.5792914628982544,
+      "step": 165
+    },
+    {
+      "epoch": 0.2273972602739726,
+      "grad_norm": 0.14515039324760437,
+      "learning_rate": 1.9641817652493964e-05,
+      "loss": 0.5595550537109375,
+      "step": 166
+    },
+    {
+      "epoch": 0.22876712328767124,
+      "grad_norm": 0.1727345734834671,
+      "learning_rate": 1.9635909156379373e-05,
+      "loss": 0.665512204170227,
+      "step": 167
+    },
+    {
+      "epoch": 0.23013698630136986,
+      "grad_norm": 0.14354562759399414,
+      "learning_rate": 1.96299532288851e-05,
+      "loss": 0.5523260235786438,
+      "step": 168
+    },
+    {
+      "epoch": 0.23150684931506849,
+      "grad_norm": 0.1551981419324875,
+      "learning_rate": 1.9623949899328352e-05,
+      "loss": 0.6456075310707092,
+      "step": 169
+    },
+    {
+      "epoch": 0.2328767123287671,
+      "grad_norm": 0.1689499169588089,
+      "learning_rate": 1.961789919725965e-05,
+      "loss": 0.621225118637085,
+      "step": 170
+    },
+    {
+      "epoch": 0.23424657534246576,
+      "grad_norm": 0.16959482431411743,
+      "learning_rate": 1.9611801152462715e-05,
+      "loss": 0.649752140045166,
+      "step": 171
+    },
+    {
+      "epoch": 0.2356164383561644,
+      "grad_norm": 0.1623905897140503,
+      "learning_rate": 1.960565579495429e-05,
+      "loss": 0.6222062110900879,
+      "step": 172
+    },
+    {
+      "epoch": 0.236986301369863,
+      "grad_norm": 0.16222091019153595,
+      "learning_rate": 1.959946315498402e-05,
+      "loss": 0.6512526869773865,
+      "step": 173
+    },
+    {
+      "epoch": 0.23835616438356164,
+      "grad_norm": 0.16222915053367615,
+      "learning_rate": 1.9593223263034288e-05,
+      "loss": 0.5998870134353638,
+      "step": 174
+    },
+    {
+      "epoch": 0.23972602739726026,
+      "grad_norm": 0.16737569868564606,
+      "learning_rate": 1.9586936149820056e-05,
+      "loss": 0.6641021966934204,
+      "step": 175
+    },
+    {
+      "epoch": 0.2410958904109589,
+      "grad_norm": 0.16463534533977509,
+      "learning_rate": 1.9580601846288747e-05,
+      "loss": 0.5590486526489258,
+      "step": 176
+    },
+    {
+      "epoch": 0.24246575342465754,
+      "grad_norm": 0.14511965215206146,
+      "learning_rate": 1.9574220383620054e-05,
+      "loss": 0.5980298519134521,
+      "step": 177
+    },
+    {
+      "epoch": 0.24383561643835616,
+      "grad_norm": 0.17374759912490845,
+      "learning_rate": 1.9567791793225818e-05,
+      "loss": 0.6144176721572876,
+      "step": 178
+    },
+    {
+      "epoch": 0.2452054794520548,
+      "grad_norm": 0.1671595722436905,
+      "learning_rate": 1.9561316106749847e-05,
+      "loss": 0.6108953952789307,
+      "step": 179
+    },
+    {
+      "epoch": 0.2465753424657534,
+      "grad_norm": 0.1550654023885727,
+      "learning_rate": 1.955479335606778e-05,
+      "loss": 0.563538670539856,
+      "step": 180
+    },
+    {
+      "epoch": 0.24794520547945206,
+      "grad_norm": 0.15254142880439758,
+      "learning_rate": 1.954822357328692e-05,
+      "loss": 0.605830192565918,
+      "step": 181
+    },
+    {
+      "epoch": 0.2493150684931507,
+      "grad_norm": 0.17636969685554504,
+      "learning_rate": 1.9541606790746082e-05,
+      "loss": 0.649669885635376,
+      "step": 182
+    },
+    {
+      "epoch": 0.25068493150684934,
+      "grad_norm": 0.16506697237491608,
+      "learning_rate": 1.9534943041015425e-05,
+      "loss": 0.6465010643005371,
+      "step": 183
+    },
+    {
+      "epoch": 0.25205479452054796,
+      "grad_norm": 0.16332168877124786,
+      "learning_rate": 1.95282323568963e-05,
+      "loss": 0.6338891386985779,
+      "step": 184
+    },
+    {
+      "epoch": 0.2534246575342466,
+      "grad_norm": 0.15076348185539246,
+      "learning_rate": 1.9521474771421093e-05,
+      "loss": 0.6196539998054504,
+      "step": 185
+    },
+    {
+      "epoch": 0.2547945205479452,
+      "grad_norm": 0.15440316498279572,
+      "learning_rate": 1.9514670317853043e-05,
+      "loss": 0.5998337864875793,
+      "step": 186
+    },
+    {
+      "epoch": 0.25616438356164384,
+      "grad_norm": 0.16012319922447205,
+      "learning_rate": 1.9507819029686094e-05,
+      "loss": 0.6014530658721924,
+      "step": 187
+    },
+    {
+      "epoch": 0.25753424657534246,
+      "grad_norm": 0.15418842434883118,
+      "learning_rate": 1.9500920940644734e-05,
+      "loss": 0.6077374219894409,
+      "step": 188
+    },
+    {
+      "epoch": 0.2589041095890411,
+      "grad_norm": 0.16678105294704437,
+      "learning_rate": 1.9493976084683814e-05,
+      "loss": 0.6204032897949219,
+      "step": 189
+    },
+    {
+      "epoch": 0.2602739726027397,
+      "grad_norm": 0.15462566912174225,
+      "learning_rate": 1.948698449598839e-05,
+      "loss": 0.5938429236412048,
+      "step": 190
+    },
+    {
+      "epoch": 0.26164383561643834,
+      "grad_norm": 0.15948761999607086,
+      "learning_rate": 1.9479946208973554e-05,
+      "loss": 0.5870481729507446,
+      "step": 191
+    },
+    {
+      "epoch": 0.26301369863013696,
+      "grad_norm": 0.17368458211421967,
+      "learning_rate": 1.947286125828426e-05,
+      "loss": 0.6516101360321045,
+      "step": 192
+    },
+    {
+      "epoch": 0.26438356164383564,
+      "grad_norm": 0.16198550164699554,
+      "learning_rate": 1.946572967879517e-05,
+      "loss": 0.5894348621368408,
+      "step": 193
+    },
+    {
+      "epoch": 0.26575342465753427,
+      "grad_norm": 0.15812379121780396,
+      "learning_rate": 1.9458551505610452e-05,
+      "loss": 0.6530876755714417,
+      "step": 194
+    },
+    {
+      "epoch": 0.2671232876712329,
+      "grad_norm": 0.15484529733657837,
+      "learning_rate": 1.9451326774063636e-05,
+      "loss": 0.5291624069213867,
+      "step": 195
+    },
+    {
+      "epoch": 0.2684931506849315,
+      "grad_norm": 0.14930173754692078,
+      "learning_rate": 1.944405551971743e-05,
+      "loss": 0.5648593306541443,
+      "step": 196
+    },
+    {
+      "epoch": 0.26986301369863014,
+      "grad_norm": 0.16413788497447968,
+      "learning_rate": 1.9436737778363526e-05,
+      "loss": 0.5979588031768799,
+      "step": 197
+    },
+    {
+      "epoch": 0.27123287671232876,
+      "grad_norm": 0.16401752829551697,
+      "learning_rate": 1.9429373586022472e-05,
+      "loss": 0.6155828237533569,
+      "step": 198
+    },
+    {
+      "epoch": 0.2726027397260274,
+      "grad_norm": 0.15918703377246857,
+      "learning_rate": 1.9421962978943442e-05,
+      "loss": 0.6395516991615295,
+      "step": 199
+    },
+    {
+      "epoch": 0.273972602739726,
+      "grad_norm": 0.16323615610599518,
+      "learning_rate": 1.9414505993604083e-05,
+      "loss": 0.6528345346450806,
+      "step": 200
+    },
+    {
+      "epoch": 0.27534246575342464,
+      "grad_norm": 0.17287138104438782,
+      "learning_rate": 1.9407002666710334e-05,
+      "loss": 0.6492872834205627,
+      "step": 201
+    },
+    {
+      "epoch": 0.27671232876712326,
+      "grad_norm": 0.17330391705036163,
+      "learning_rate": 1.939945303519625e-05,
+      "loss": 0.6085129380226135,
+      "step": 202
+    },
+    {
+      "epoch": 0.27808219178082194,
+      "grad_norm": 0.1582794040441513,
+      "learning_rate": 1.93918571362238e-05,
+      "loss": 0.6202000379562378,
+      "step": 203
+    },
+    {
+      "epoch": 0.27945205479452057,
+      "grad_norm": 0.15793833136558533,
+      "learning_rate": 1.938421500718272e-05,
+      "loss": 0.5806069374084473,
+      "step": 204
+    },
+    {
+      "epoch": 0.2808219178082192,
+      "grad_norm": 0.16570699214935303,
+      "learning_rate": 1.937652668569028e-05,
+      "loss": 0.6103482246398926,
+      "step": 205
+    },
+    {
+      "epoch": 0.2821917808219178,
+      "grad_norm": 0.1538430154323578,
+      "learning_rate": 1.9368792209591145e-05,
+      "loss": 0.5779464244842529,
+      "step": 206
+    },
+    {
+      "epoch": 0.28356164383561644,
+      "grad_norm": 0.1592852622270584,
+      "learning_rate": 1.9361011616957165e-05,
+      "loss": 0.6240794658660889,
+      "step": 207
+    },
+    {
+      "epoch": 0.28493150684931506,
+      "grad_norm": 0.1604919135570526,
+      "learning_rate": 1.935318494608718e-05,
+      "loss": 0.5771029591560364,
+      "step": 208
+    },
+    {
+      "epoch": 0.2863013698630137,
+      "grad_norm": 0.1573483794927597,
+      "learning_rate": 1.934531223550687e-05,
+      "loss": 0.6847480535507202,
+      "step": 209
+    },
+    {
+      "epoch": 0.2876712328767123,
+      "grad_norm": 0.15778696537017822,
+      "learning_rate": 1.933739352396851e-05,
+      "loss": 0.6099205017089844,
+      "step": 210
+    },
+    {
+      "epoch": 0.28904109589041094,
+      "grad_norm": 0.15076380968093872,
+      "learning_rate": 1.932942885045083e-05,
+      "loss": 0.5610252618789673,
+      "step": 211
+    },
+    {
+      "epoch": 0.29041095890410956,
+      "grad_norm": 0.1657320261001587,
+      "learning_rate": 1.9321418254158782e-05,
+      "loss": 0.6706528067588806,
+      "step": 212
+    },
+    {
+      "epoch": 0.29178082191780824,
+      "grad_norm": 0.1694263070821762,
+      "learning_rate": 1.9313361774523387e-05,
+      "loss": 0.6531721949577332,
+      "step": 213
+    },
+    {
+      "epoch": 0.29315068493150687,
+      "grad_norm": 0.15693512558937073,
+      "learning_rate": 1.9305259451201502e-05,
+      "loss": 0.6321283578872681,
+      "step": 214
+    },
+    {
+      "epoch": 0.2945205479452055,
+      "grad_norm": 0.16327035427093506,
+      "learning_rate": 1.929711132407565e-05,
+      "loss": 0.6443174481391907,
+      "step": 215
+    },
+    {
+      "epoch": 0.2958904109589041,
+      "grad_norm": 0.16162200272083282,
+      "learning_rate": 1.9288917433253823e-05,
+      "loss": 0.6530073881149292,
+      "step": 216
+    },
+    {
+      "epoch": 0.29726027397260274,
+      "grad_norm": 0.15862113237380981,
+      "learning_rate": 1.9280677819069273e-05,
+      "loss": 0.5875598788261414,
+      "step": 217
+    },
+    {
+      "epoch": 0.29863013698630136,
+      "grad_norm": 0.15702694654464722,
+      "learning_rate": 1.927239252208031e-05,
+      "loss": 0.5955379605293274,
+      "step": 218
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.15328216552734375,
+      "learning_rate": 1.9264061583070126e-05,
+      "loss": 0.663190484046936,
+      "step": 219
+    },
+    {
+      "epoch": 0.3013698630136986,
+      "grad_norm": 0.17017608880996704,
+      "learning_rate": 1.9255685043046572e-05,
+      "loss": 0.6066758036613464,
+      "step": 220
+    },
+    {
+      "epoch": 0.30273972602739724,
+      "grad_norm": 0.1607287973165512,
+      "learning_rate": 1.924726294324196e-05,
+      "loss": 0.5664497017860413,
+      "step": 221
+    },
+    {
+      "epoch": 0.3041095890410959,
+      "grad_norm": 0.1684304177761078,
+      "learning_rate": 1.9238795325112867e-05,
+      "loss": 0.6179485321044922,
+      "step": 222
+    },
+    {
+      "epoch": 0.30547945205479454,
+      "grad_norm": 0.1446060985326767,
+      "learning_rate": 1.9230282230339936e-05,
+      "loss": 0.581000030040741,
+      "step": 223
+    },
+    {
+      "epoch": 0.30684931506849317,
+      "grad_norm": 0.16912856698036194,
+      "learning_rate": 1.9221723700827646e-05,
+      "loss": 0.6572670936584473,
+      "step": 224
+    },
+    {
+      "epoch": 0.3082191780821918,
+      "grad_norm": 0.16073641180992126,
+      "learning_rate": 1.921311977870413e-05,
+      "loss": 0.5523173809051514,
+      "step": 225
+    },
+    {
+      "epoch": 0.3095890410958904,
+      "grad_norm": 0.16843587160110474,
+      "learning_rate": 1.9204470506320968e-05,
+      "loss": 0.6724870204925537,
+      "step": 226
+    },
+    {
+      "epoch": 0.31095890410958904,
+      "grad_norm": 0.16108570992946625,
+      "learning_rate": 1.9195775926252952e-05,
+      "loss": 0.6259427070617676,
+      "step": 227
+    },
+    {
+      "epoch": 0.31232876712328766,
+      "grad_norm": 0.15122473239898682,
+      "learning_rate": 1.9187036081297907e-05,
+      "loss": 0.5658705234527588,
+      "step": 228
+    },
+    {
+      "epoch": 0.3136986301369863,
+      "grad_norm": 0.1597052663564682,
+      "learning_rate": 1.917825101447647e-05,
+      "loss": 0.6100412607192993,
+      "step": 229
+    },
+    {
+      "epoch": 0.3150684931506849,
+      "grad_norm": 0.18265262246131897,
+      "learning_rate": 1.9169420769031863e-05,
+      "loss": 0.5868899822235107,
+      "step": 230
+    },
+    {
+      "epoch": 0.31643835616438354,
+      "grad_norm": 0.15701507031917572,
+      "learning_rate": 1.916054538842971e-05,
+      "loss": 0.6006147265434265,
+      "step": 231
+    },
+    {
+      "epoch": 0.3178082191780822,
+      "grad_norm": 0.15414513647556305,
+      "learning_rate": 1.9151624916357793e-05,
+      "loss": 0.6486912965774536,
+      "step": 232
+    },
+    {
+      "epoch": 0.31917808219178084,
+      "grad_norm": 0.15333938598632812,
+      "learning_rate": 1.9142659396725862e-05,
+      "loss": 0.6258531808853149,
+      "step": 233
+    },
+    {
+      "epoch": 0.32054794520547947,
+      "grad_norm": 0.14656788110733032,
+      "learning_rate": 1.913364887366539e-05,
+      "loss": 0.5718272924423218,
+      "step": 234
+    },
+    {
+      "epoch": 0.3219178082191781,
+      "grad_norm": 0.16382336616516113,
+      "learning_rate": 1.912459339152939e-05,
+      "loss": 0.5820138454437256,
+      "step": 235
+    },
+    {
+      "epoch": 0.3232876712328767,
+      "grad_norm": 0.1630294770002365,
+      "learning_rate": 1.9115492994892172e-05,
+      "loss": 0.600603461265564,
+      "step": 236
+    },
+    {
+      "epoch": 0.32465753424657534,
+      "grad_norm": 0.14892295002937317,
+      "learning_rate": 1.9106347728549134e-05,
+      "loss": 0.5658777952194214,
+      "step": 237
+    },
+    {
+      "epoch": 0.32602739726027397,
+      "grad_norm": 0.16635073721408844,
+      "learning_rate": 1.9097157637516532e-05,
+      "loss": 0.5740564465522766,
+      "step": 238
+    },
+    {
+      "epoch": 0.3273972602739726,
+      "grad_norm": 0.16513577103614807,
+      "learning_rate": 1.9087922767031276e-05,
+      "loss": 0.6348574757575989,
+      "step": 239
+    },
+    {
+      "epoch": 0.3287671232876712,
+      "grad_norm": 0.16342297196388245,
+      "learning_rate": 1.9078643162550686e-05,
+      "loss": 0.6322500109672546,
+      "step": 240
+    },
+    {
+      "epoch": 0.33013698630136984,
+      "grad_norm": 0.1535942405462265,
+      "learning_rate": 1.906931886975228e-05,
+      "loss": 0.53902268409729,
+      "step": 241
+    },
+    {
+      "epoch": 0.3315068493150685,
+      "grad_norm": 0.14455796778202057,
+      "learning_rate": 1.905994993453356e-05,
+      "loss": 0.5630102157592773,
+      "step": 242
+    },
+    {
+      "epoch": 0.33287671232876714,
+      "grad_norm": 0.1663145124912262,
+      "learning_rate": 1.905053640301176e-05,
+      "loss": 0.6214050054550171,
+      "step": 243
+    },
+    {
+      "epoch": 0.33424657534246577,
+      "grad_norm": 0.1636534482240677,
+      "learning_rate": 1.9041078321523637e-05,
+      "loss": 0.6704033613204956,
+      "step": 244
+    },
+    {
+      "epoch": 0.3356164383561644,
+      "grad_norm": 0.1620258390903473,
+      "learning_rate": 1.903157573662524e-05,
+      "loss": 0.6345779895782471,
+      "step": 245
+    },
+    {
+      "epoch": 0.336986301369863,
+      "grad_norm": 0.15679852664470673,
+      "learning_rate": 1.9022028695091678e-05,
+      "loss": 0.5873825550079346,
+      "step": 246
+    },
+    {
+      "epoch": 0.33835616438356164,
+      "grad_norm": 0.15328912436962128,
+      "learning_rate": 1.9012437243916895e-05,
+      "loss": 0.6029560565948486,
+      "step": 247
+    },
+    {
+      "epoch": 0.33972602739726027,
+      "grad_norm": 0.14538195729255676,
+      "learning_rate": 1.900280143031343e-05,
+      "loss": 0.5177454948425293,
+      "step": 248
+    },
+    {
+      "epoch": 0.3410958904109589,
+      "grad_norm": 0.16042332351207733,
+      "learning_rate": 1.8993121301712194e-05,
+      "loss": 0.6091316342353821,
+      "step": 249
+    },
+    {
+      "epoch": 0.3424657534246575,
+      "grad_norm": 0.1434759795665741,
+      "learning_rate": 1.8983396905762232e-05,
+      "loss": 0.5697845220565796,
+      "step": 250
+    },
+    {
+      "epoch": 0.34383561643835614,
+      "grad_norm": 0.14016889035701752,
+      "learning_rate": 1.8973628290330484e-05,
+      "loss": 0.5575758218765259,
+      "step": 251
+    },
+    {
+      "epoch": 0.3452054794520548,
+      "grad_norm": 0.14783437550067902,
+      "learning_rate": 1.896381550350156e-05,
+      "loss": 0.5810775756835938,
+      "step": 252
+    },
+    {
+      "epoch": 0.34657534246575344,
+      "grad_norm": 0.15758486092090607,
+      "learning_rate": 1.895395859357749e-05,
+      "loss": 0.613425076007843,
+      "step": 253
+    },
+    {
+      "epoch": 0.34794520547945207,
+      "grad_norm": 0.15336844325065613,
+      "learning_rate": 1.8944057609077506e-05,
+      "loss": 0.5932927131652832,
+      "step": 254
+    },
+    {
+      "epoch": 0.3493150684931507,
+      "grad_norm": 0.1474757194519043,
+      "learning_rate": 1.8934112598737777e-05,
+      "loss": 0.5725477337837219,
+      "step": 255
+    },
+    {
+      "epoch": 0.3506849315068493,
+      "grad_norm": 0.14386139810085297,
+      "learning_rate": 1.892412361151119e-05,
+      "loss": 0.6071690320968628,
+      "step": 256
+    },
+    {
+      "epoch": 0.35205479452054794,
+      "grad_norm": 0.17297492921352386,
+      "learning_rate": 1.8914090696567104e-05,
+      "loss": 0.6448274850845337,
+      "step": 257
+    },
+    {
+      "epoch": 0.35342465753424657,
+      "grad_norm": 0.1554916501045227,
+      "learning_rate": 1.89040139032911e-05,
+      "loss": 0.6005516052246094,
+      "step": 258
+    },
+    {
+      "epoch": 0.3547945205479452,
+      "grad_norm": 0.15703563392162323,
+      "learning_rate": 1.8893893281284744e-05,
+      "loss": 0.5192084312438965,
+      "step": 259
+    },
+    {
+      "epoch": 0.3561643835616438,
+      "grad_norm": 0.1578558385372162,
+      "learning_rate": 1.8883728880365347e-05,
+      "loss": 0.6114726066589355,
+      "step": 260
+    },
+    {
+      "epoch": 0.35753424657534244,
+      "grad_norm": 0.16833758354187012,
+      "learning_rate": 1.8873520750565716e-05,
+      "loss": 0.6224363446235657,
+      "step": 261
+    },
+    {
+      "epoch": 0.3589041095890411,
+      "grad_norm": 0.1642412692308426,
+      "learning_rate": 1.8863268942133905e-05,
+      "loss": 0.6578483581542969,
+      "step": 262
+    },
+    {
+      "epoch": 0.36027397260273974,
+      "grad_norm": 0.16325688362121582,
+      "learning_rate": 1.8852973505532963e-05,
+      "loss": 0.6258366107940674,
+      "step": 263
+    },
+    {
+      "epoch": 0.36164383561643837,
+      "grad_norm": 0.16593174636363983,
+      "learning_rate": 1.8842634491440704e-05,
+      "loss": 0.6312078237533569,
+      "step": 264
+    },
+    {
+      "epoch": 0.363013698630137,
+      "grad_norm": 0.1558452844619751,
+      "learning_rate": 1.8832251950749443e-05,
+      "loss": 0.5959346294403076,
+      "step": 265
+    },
+    {
+      "epoch": 0.3643835616438356,
+      "grad_norm": 0.16329628229141235,
+      "learning_rate": 1.882182593456574e-05,
+      "loss": 0.5636122226715088,
+      "step": 266
+    },
+    {
+      "epoch": 0.36575342465753424,
+      "grad_norm": 0.1444038301706314,
+      "learning_rate": 1.8811356494210166e-05,
+      "loss": 0.5212979316711426,
+      "step": 267
+    },
+    {
+      "epoch": 0.36712328767123287,
+      "grad_norm": 0.16624850034713745,
+      "learning_rate": 1.880084368121704e-05,
+      "loss": 0.6462916135787964,
+      "step": 268
+    },
+    {
+      "epoch": 0.3684931506849315,
+      "grad_norm": 0.15940053761005402,
+      "learning_rate": 1.8790287547334178e-05,
+      "loss": 0.6105808615684509,
+      "step": 269
+    },
+    {
+      "epoch": 0.3698630136986301,
+      "grad_norm": 0.15389667451381683,
+      "learning_rate": 1.8779688144522625e-05,
+      "loss": 0.5553721785545349,
+      "step": 270
+    },
+    {
+      "epoch": 0.37123287671232874,
+      "grad_norm": 0.15304426848888397,
+      "learning_rate": 1.876904552495642e-05,
+      "loss": 0.6068332195281982,
+      "step": 271
+    },
+    {
+      "epoch": 0.3726027397260274,
+      "grad_norm": 0.1442302167415619,
+      "learning_rate": 1.875835974102234e-05,
+      "loss": 0.5685116052627563,
+      "step": 272
+    },
+    {
+      "epoch": 0.37397260273972605,
+      "grad_norm": 0.14790469408035278,
+      "learning_rate": 1.874763084531961e-05,
+      "loss": 0.5844176411628723,
+      "step": 273
+    },
+    {
+      "epoch": 0.37534246575342467,
+      "grad_norm": 0.14410603046417236,
+      "learning_rate": 1.8736858890659696e-05,
+      "loss": 0.5909885168075562,
+      "step": 274
+    },
+    {
+      "epoch": 0.3767123287671233,
+      "grad_norm": 0.17475005984306335,
+      "learning_rate": 1.8726043930065985e-05,
+      "loss": 0.6046225428581238,
+      "step": 275
+    },
+    {
+      "epoch": 0.3780821917808219,
+      "grad_norm": 0.15059982240200043,
+      "learning_rate": 1.871518601677357e-05,
+      "loss": 0.5838688015937805,
+      "step": 276
+    },
+    {
+      "epoch": 0.37945205479452054,
+      "grad_norm": 0.17032840847969055,
+      "learning_rate": 1.8704285204228973e-05,
+      "loss": 0.6261906623840332,
+      "step": 277
+    },
+    {
+      "epoch": 0.38082191780821917,
+      "grad_norm": 0.15075044333934784,
+      "learning_rate": 1.8693341546089878e-05,
+      "loss": 0.6051650047302246,
+      "step": 278
+    },
+    {
+      "epoch": 0.3821917808219178,
+      "grad_norm": 0.1473180502653122,
+      "learning_rate": 1.8682355096224873e-05,
+      "loss": 0.5551210641860962,
+      "step": 279
+    },
+    {
+      "epoch": 0.3835616438356164,
+      "grad_norm": 0.15265880525112152,
+      "learning_rate": 1.8671325908713182e-05,
+      "loss": 0.570478081703186,
+      "step": 280
+    },
+    {
+      "epoch": 0.38493150684931504,
+      "grad_norm": 0.16518573462963104,
+      "learning_rate": 1.866025403784439e-05,
+      "loss": 0.6145535707473755,
+      "step": 281
+    },
+    {
+      "epoch": 0.3863013698630137,
+      "grad_norm": 0.15201589465141296,
+      "learning_rate": 1.8649139538118196e-05,
+      "loss": 0.5385153293609619,
+      "step": 282
+    },
+    {
+      "epoch": 0.38767123287671235,
+      "grad_norm": 0.14805659651756287,
+      "learning_rate": 1.863798246424413e-05,
+      "loss": 0.628183126449585,
+      "step": 283
+    },
+    {
+      "epoch": 0.38904109589041097,
+      "grad_norm": 0.14637944102287292,
+      "learning_rate": 1.862678287114129e-05,
+      "loss": 0.62421715259552,
+      "step": 284
+    },
+    {
+      "epoch": 0.3904109589041096,
+      "grad_norm": 0.1630602329969406,
+      "learning_rate": 1.8615540813938063e-05,
+      "loss": 0.6616168022155762,
+      "step": 285
+    },
+    {
+      "epoch": 0.3917808219178082,
+      "grad_norm": 0.1550370305776596,
+      "learning_rate": 1.860425634797186e-05,
+      "loss": 0.5864919424057007,
+      "step": 286
+    },
+    {
+      "epoch": 0.39315068493150684,
+      "grad_norm": 0.14578670263290405,
+      "learning_rate": 1.8592929528788844e-05,
+      "loss": 0.545616626739502,
+      "step": 287
+    },
+    {
+      "epoch": 0.39452054794520547,
+      "grad_norm": 0.14833520352840424,
+      "learning_rate": 1.8581560412143663e-05,
+      "loss": 0.5978792905807495,
+      "step": 288
+    },
+    {
+      "epoch": 0.3958904109589041,
+      "grad_norm": 0.1507352888584137,
+      "learning_rate": 1.8570149053999154e-05,
+      "loss": 0.579099178314209,
+      "step": 289
+    },
+    {
+      "epoch": 0.3972602739726027,
+      "grad_norm": 0.14446204900741577,
+      "learning_rate": 1.8558695510526092e-05,
+      "loss": 0.5924318432807922,
+      "step": 290
+    },
+    {
+      "epoch": 0.39863013698630134,
+      "grad_norm": 0.15075939893722534,
+      "learning_rate": 1.8547199838102904e-05,
+      "loss": 0.5611657500267029,
+      "step": 291
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.15933777391910553,
+      "learning_rate": 1.8535662093315386e-05,
+      "loss": 0.5471000671386719,
+      "step": 292
+    },
+    {
+      "epoch": 0.40136986301369865,
+      "grad_norm": 0.15519365668296814,
+      "learning_rate": 1.852408233295643e-05,
+      "loss": 0.6255986094474792,
+      "step": 293
+    },
+    {
+      "epoch": 0.40273972602739727,
+      "grad_norm": 0.15151312947273254,
+      "learning_rate": 1.851246061402574e-05,
+      "loss": 0.5669071674346924,
+      "step": 294
+    },
+    {
+      "epoch": 0.4041095890410959,
+      "grad_norm": 0.16434764862060547,
+      "learning_rate": 1.8500796993729563e-05,
+      "loss": 0.5679120421409607,
+      "step": 295
+    },
+    {
+      "epoch": 0.4054794520547945,
+      "grad_norm": 0.164402574300766,
+      "learning_rate": 1.8489091529480398e-05,
+      "loss": 0.6179913282394409,
+      "step": 296
+    },
+    {
+      "epoch": 0.40684931506849314,
+      "grad_norm": 0.15723305940628052,
+      "learning_rate": 1.8477344278896708e-05,
+      "loss": 0.5922874212265015,
+      "step": 297
+    },
+    {
+      "epoch": 0.40821917808219177,
+      "grad_norm": 0.1540336310863495,
+      "learning_rate": 1.8465555299802655e-05,
+      "loss": 0.5662035346031189,
+      "step": 298
+    },
+    {
+      "epoch": 0.4095890410958904,
+      "grad_norm": 0.17283470928668976,
+      "learning_rate": 1.8453724650227788e-05,
+      "loss": 0.6761048436164856,
+      "step": 299
+    },
+    {
+      "epoch": 0.410958904109589,
+      "grad_norm": 0.16036930680274963,
+      "learning_rate": 1.8441852388406788e-05,
+      "loss": 0.5548433065414429,
+      "step": 300
+    },
+    {
+      "epoch": 0.4123287671232877,
+      "grad_norm": 0.14102788269519806,
+      "learning_rate": 1.8429938572779154e-05,
+      "loss": 0.5084259510040283,
+      "step": 301
+    },
+    {
+      "epoch": 0.4136986301369863,
+      "grad_norm": 0.15470127761363983,
+      "learning_rate": 1.8417983261988933e-05,
+      "loss": 0.6092421412467957,
+      "step": 302
+    },
+    {
+      "epoch": 0.41506849315068495,
+      "grad_norm": 0.1494375318288803,
+      "learning_rate": 1.840598651488443e-05,
+      "loss": 0.5430501699447632,
+      "step": 303
+    },
+    {
+      "epoch": 0.41643835616438357,
+      "grad_norm": 0.14817532896995544,
+      "learning_rate": 1.8393948390517915e-05,
+      "loss": 0.5703598260879517,
+      "step": 304
+    },
+    {
+      "epoch": 0.4178082191780822,
+      "grad_norm": 0.15214185416698456,
+      "learning_rate": 1.8381868948145317e-05,
+      "loss": 0.6218011379241943,
+      "step": 305
+    },
+    {
+      "epoch": 0.4191780821917808,
+      "grad_norm": 0.14840349555015564,
+      "learning_rate": 1.8369748247225965e-05,
+      "loss": 0.6339513659477234,
+      "step": 306
+    },
+    {
+      "epoch": 0.42054794520547945,
+      "grad_norm": 0.14049845933914185,
+      "learning_rate": 1.8357586347422266e-05,
+      "loss": 0.5735560059547424,
+      "step": 307
+    },
+    {
+      "epoch": 0.42191780821917807,
+      "grad_norm": 0.14952462911605835,
+      "learning_rate": 1.8345383308599425e-05,
+      "loss": 0.5752905607223511,
+      "step": 308
+    },
+    {
+      "epoch": 0.4232876712328767,
+      "grad_norm": 0.1464746594429016,
+      "learning_rate": 1.833313919082515e-05,
+      "loss": 0.5919328927993774,
+      "step": 309
+    },
+    {
+      "epoch": 0.4246575342465753,
+      "grad_norm": 0.16054148972034454,
+      "learning_rate": 1.8320854054369354e-05,
+      "loss": 0.6318283081054688,
+      "step": 310
+    },
+    {
+      "epoch": 0.426027397260274,
+      "grad_norm": 0.14838211238384247,
+      "learning_rate": 1.8308527959703864e-05,
+      "loss": 0.5462095737457275,
+      "step": 311
+    },
+    {
+      "epoch": 0.4273972602739726,
+      "grad_norm": 0.14827798306941986,
+      "learning_rate": 1.82961609675021e-05,
+      "loss": 0.587713897228241,
+      "step": 312
+    },
+    {
+      "epoch": 0.42876712328767125,
+      "grad_norm": 0.14198598265647888,
+      "learning_rate": 1.828375313863881e-05,
+      "loss": 0.5431427955627441,
+      "step": 313
+    },
+    {
+      "epoch": 0.4301369863013699,
+      "grad_norm": 0.15106132626533508,
+      "learning_rate": 1.8271304534189754e-05,
+      "loss": 0.5700148940086365,
+      "step": 314
+    },
+    {
+      "epoch": 0.4315068493150685,
+      "grad_norm": 0.14311951398849487,
+      "learning_rate": 1.8258815215431395e-05,
+      "loss": 0.5781939625740051,
+      "step": 315
+    },
+    {
+      "epoch": 0.4328767123287671,
+      "grad_norm": 0.15375271439552307,
+      "learning_rate": 1.8246285243840617e-05,
+      "loss": 0.5642305612564087,
+      "step": 316
+    },
+    {
+      "epoch": 0.43424657534246575,
+      "grad_norm": 0.1475384384393692,
+      "learning_rate": 1.8233714681094405e-05,
+      "loss": 0.5396713018417358,
+      "step": 317
+    },
+    {
+      "epoch": 0.43561643835616437,
+      "grad_norm": 0.14523953199386597,
+      "learning_rate": 1.8221103589069553e-05,
+      "loss": 0.6184444427490234,
+      "step": 318
+    },
+    {
+      "epoch": 0.436986301369863,
+      "grad_norm": 0.1605875939130783,
+      "learning_rate": 1.8208452029842342e-05,
+      "loss": 0.5515637397766113,
+      "step": 319
+    },
+    {
+      "epoch": 0.4383561643835616,
+      "grad_norm": 0.14436954259872437,
+      "learning_rate": 1.8195760065688266e-05,
+      "loss": 0.6023062467575073,
+      "step": 320
+    },
+    {
+      "epoch": 0.4397260273972603,
+      "grad_norm": 0.13907724618911743,
+      "learning_rate": 1.818302775908169e-05,
+      "loss": 0.6034039258956909,
+      "step": 321
+    },
+    {
+      "epoch": 0.4410958904109589,
+      "grad_norm": 0.15003663301467896,
+      "learning_rate": 1.8170255172695572e-05,
+      "loss": 0.608570396900177,
+      "step": 322
+    },
+    {
+      "epoch": 0.44246575342465755,
+      "grad_norm": 0.16264384984970093,
+      "learning_rate": 1.815744236940113e-05,
+      "loss": 0.6284328699111938,
+      "step": 323
+    },
+    {
+      "epoch": 0.4438356164383562,
+      "grad_norm": 0.15702727437019348,
+      "learning_rate": 1.814458941226755e-05,
+      "loss": 0.5362198352813721,
+      "step": 324
+    },
+    {
+      "epoch": 0.4452054794520548,
+      "grad_norm": 0.1596728265285492,
+      "learning_rate": 1.813169636456167e-05,
+      "loss": 0.575636088848114,
+      "step": 325
+    },
+    {
+      "epoch": 0.4465753424657534,
+      "grad_norm": 0.1574542075395584,
+      "learning_rate": 1.8118763289747666e-05,
+      "loss": 0.6510027647018433,
+      "step": 326
+    },
+    {
+      "epoch": 0.44794520547945205,
+      "grad_norm": 0.14724481105804443,
+      "learning_rate": 1.810579025148674e-05,
+      "loss": 0.5525646209716797,
+      "step": 327
+    },
+    {
+      "epoch": 0.44931506849315067,
+      "grad_norm": 0.1698918640613556,
+      "learning_rate": 1.8092777313636813e-05,
+      "loss": 0.660180926322937,
+      "step": 328
+    },
+    {
+      "epoch": 0.4506849315068493,
+      "grad_norm": 0.15697503089904785,
+      "learning_rate": 1.8079724540252204e-05,
+      "loss": 0.5749319791793823,
+      "step": 329
+    },
+    {
+      "epoch": 0.4520547945205479,
+      "grad_norm": 0.1613071709871292,
+      "learning_rate": 1.8066631995583318e-05,
+      "loss": 0.5470437407493591,
+      "step": 330
+    },
+    {
+      "epoch": 0.4534246575342466,
+      "grad_norm": 0.14156115055084229,
+      "learning_rate": 1.8053499744076323e-05,
+      "loss": 0.5292378664016724,
+      "step": 331
+    },
+    {
+      "epoch": 0.4547945205479452,
+      "grad_norm": 0.14653272926807404,
+      "learning_rate": 1.8040327850372847e-05,
+      "loss": 0.6091462969779968,
+      "step": 332
+    },
+    {
+      "epoch": 0.45616438356164385,
+      "grad_norm": 0.16075585782527924,
+      "learning_rate": 1.8027116379309637e-05,
+      "loss": 0.6253776550292969,
+      "step": 333
+    },
+    {
+      "epoch": 0.4575342465753425,
+      "grad_norm": 0.1504659652709961,
+      "learning_rate": 1.801386539591827e-05,
+      "loss": 0.5656317472457886,
+      "step": 334
+    },
+    {
+      "epoch": 0.4589041095890411,
+      "grad_norm": 0.14729438722133636,
+      "learning_rate": 1.8000574965424813e-05,
+      "loss": 0.5766185522079468,
+      "step": 335
+    },
+    {
+      "epoch": 0.4602739726027397,
+      "grad_norm": 0.1585109382867813,
+      "learning_rate": 1.7987245153249496e-05,
+      "loss": 0.556631326675415,
+      "step": 336
+    },
+    {
+      "epoch": 0.46164383561643835,
+      "grad_norm": 0.15586094558238983,
+      "learning_rate": 1.7973876025006407e-05,
+      "loss": 0.5819780826568604,
+      "step": 337
+    },
+    {
+      "epoch": 0.46301369863013697,
+      "grad_norm": 0.14767439663410187,
+      "learning_rate": 1.796046764650316e-05,
+      "loss": 0.5321399569511414,
+      "step": 338
+    },
+    {
+      "epoch": 0.4643835616438356,
+      "grad_norm": 0.15002456307411194,
+      "learning_rate": 1.7947020083740575e-05,
+      "loss": 0.5469136834144592,
+      "step": 339
+    },
+    {
+      "epoch": 0.4657534246575342,
+      "grad_norm": 0.14220520853996277,
+      "learning_rate": 1.7933533402912354e-05,
+      "loss": 0.6022108793258667,
+      "step": 340
+    },
+    {
+      "epoch": 0.4671232876712329,
+      "grad_norm": 0.1343364119529724,
+      "learning_rate": 1.7920007670404738e-05,
+      "loss": 0.5860722064971924,
+      "step": 341
+    },
+    {
+      "epoch": 0.4684931506849315,
+      "grad_norm": 0.1453264057636261,
+      "learning_rate": 1.7906442952796212e-05,
+      "loss": 0.5470801591873169,
+      "step": 342
+    },
+    {
+      "epoch": 0.46986301369863015,
+      "grad_norm": 0.15289293229579926,
+      "learning_rate": 1.7892839316857146e-05,
+      "loss": 0.5834886431694031,
+      "step": 343
+    },
+    {
+      "epoch": 0.4712328767123288,
+      "grad_norm": 0.15057221055030823,
+      "learning_rate": 1.787919682954949e-05,
+      "loss": 0.5459625720977783,
+      "step": 344
+    },
+    {
+      "epoch": 0.4726027397260274,
+      "grad_norm": 0.14852501451969147,
+      "learning_rate": 1.786551555802643e-05,
+      "loss": 0.57718825340271,
+      "step": 345
+    },
+    {
+      "epoch": 0.473972602739726,
+      "grad_norm": 0.14916345477104187,
+      "learning_rate": 1.785179556963205e-05,
+      "loss": 0.5912124514579773,
+      "step": 346
+    },
+    {
+      "epoch": 0.47534246575342465,
+      "grad_norm": 0.14343911409378052,
+      "learning_rate": 1.7838036931901033e-05,
+      "loss": 0.5384149551391602,
+      "step": 347
+    },
+    {
+      "epoch": 0.4767123287671233,
+      "grad_norm": 0.13747356832027435,
+      "learning_rate": 1.7824239712558303e-05,
+      "loss": 0.5556282997131348,
+      "step": 348
+    },
+    {
+      "epoch": 0.4780821917808219,
+      "grad_norm": 0.153511181473732,
+      "learning_rate": 1.781040397951868e-05,
+      "loss": 0.5881044864654541,
+      "step": 349
+    },
+    {
+      "epoch": 0.4794520547945205,
+      "grad_norm": 0.1565730720758438,
+      "learning_rate": 1.7796529800886583e-05,
+      "loss": 0.564147412776947,
+      "step": 350
+    },
+    {
+      "epoch": 0.4808219178082192,
+      "grad_norm": 0.14074577391147614,
+      "learning_rate": 1.778261724495566e-05,
+      "loss": 0.5963798761367798,
+      "step": 351
+    },
+    {
+      "epoch": 0.4821917808219178,
+      "grad_norm": 0.1508810669183731,
+      "learning_rate": 1.7768666380208474e-05,
+      "loss": 0.6377789378166199,
+      "step": 352
+    },
+    {
+      "epoch": 0.48356164383561645,
+      "grad_norm": 0.14624693989753723,
+      "learning_rate": 1.775467727531615e-05,
+      "loss": 0.6010425090789795,
+      "step": 353
+    },
+    {
+      "epoch": 0.4849315068493151,
+      "grad_norm": 0.1480599194765091,
+      "learning_rate": 1.774064999913805e-05,
+      "loss": 0.5669881701469421,
+      "step": 354
+    },
+    {
+      "epoch": 0.4863013698630137,
+      "grad_norm": 0.1533871442079544,
+      "learning_rate": 1.772658462072143e-05,
+      "loss": 0.6225383877754211,
+      "step": 355
+    },
+    {
+      "epoch": 0.4876712328767123,
+      "grad_norm": 0.16357164084911346,
+      "learning_rate": 1.7712481209301094e-05,
+      "loss": 0.6383495926856995,
+      "step": 356
+    },
+    {
+      "epoch": 0.48904109589041095,
+      "grad_norm": 0.136616513133049,
+      "learning_rate": 1.7698339834299064e-05,
+      "loss": 0.5601040124893188,
+      "step": 357
+    },
+    {
+      "epoch": 0.4904109589041096,
+      "grad_norm": 0.15099939703941345,
+      "learning_rate": 1.7684160565324222e-05,
+      "loss": 0.5624288320541382,
+      "step": 358
+    },
+    {
+      "epoch": 0.4917808219178082,
+      "grad_norm": 0.14297816157341003,
+      "learning_rate": 1.766994347217199e-05,
+      "loss": 0.6051282286643982,
+      "step": 359
+    },
+    {
+      "epoch": 0.4931506849315068,
+      "grad_norm": 0.13573159277439117,
+      "learning_rate": 1.765568862482397e-05,
+      "loss": 0.5469596982002258,
+      "step": 360
+    },
+    {
+      "epoch": 0.4945205479452055,
+      "grad_norm": 0.1435568928718567,
+      "learning_rate": 1.764139609344759e-05,
+      "loss": 0.5686632394790649,
+      "step": 361
+    },
+    {
+      "epoch": 0.4958904109589041,
+      "grad_norm": 0.1436147540807724,
+      "learning_rate": 1.7627065948395795e-05,
+      "loss": 0.5006036758422852,
+      "step": 362
+    },
+    {
+      "epoch": 0.49726027397260275,
+      "grad_norm": 0.16107958555221558,
+      "learning_rate": 1.7612698260206668e-05,
+      "loss": 0.6271241903305054,
+      "step": 363
+    },
+    {
+      "epoch": 0.4986301369863014,
+      "grad_norm": 0.15104009211063385,
+      "learning_rate": 1.7598293099603087e-05,
+      "loss": 0.5952401161193848,
+      "step": 364
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.1506115049123764,
+      "learning_rate": 1.7583850537492386e-05,
+      "loss": 0.567657470703125,
+      "step": 365
+    },
+    {
+      "epoch": 0.5013698630136987,
+      "grad_norm": 0.16260883212089539,
+      "learning_rate": 1.7569370644966007e-05,
+      "loss": 0.6197588443756104,
+      "step": 366
+    },
+    {
+      "epoch": 0.5027397260273972,
+      "grad_norm": 0.16213645040988922,
+      "learning_rate": 1.7554853493299142e-05,
+      "loss": 0.6336584687232971,
+      "step": 367
+    },
+    {
+      "epoch": 0.5041095890410959,
+      "grad_norm": 0.15891368687152863,
+      "learning_rate": 1.7540299153950386e-05,
+      "loss": 0.6455602645874023,
+      "step": 368
+    },
+    {
+      "epoch": 0.5054794520547945,
+      "grad_norm": 0.14739622175693512,
+      "learning_rate": 1.7525707698561383e-05,
+      "loss": 0.5473576188087463,
+      "step": 369
+    },
+    {
+      "epoch": 0.5068493150684932,
+      "grad_norm": 0.1491318643093109,
+      "learning_rate": 1.7511079198956487e-05,
+      "loss": 0.5691012144088745,
+      "step": 370
+    },
+    {
+      "epoch": 0.5082191780821917,
+      "grad_norm": 0.15580914914608002,
+      "learning_rate": 1.7496413727142375e-05,
+      "loss": 0.5861622095108032,
+      "step": 371
+    },
+    {
+      "epoch": 0.5095890410958904,
+      "grad_norm": 0.1470382958650589,
+      "learning_rate": 1.7481711355307735e-05,
+      "loss": 0.6198167204856873,
+      "step": 372
+    },
+    {
+      "epoch": 0.510958904109589,
+      "grad_norm": 0.1570439487695694,
+      "learning_rate": 1.746697215582288e-05,
+      "loss": 0.5717829465866089,
+      "step": 373
+    },
+    {
+      "epoch": 0.5123287671232877,
+      "grad_norm": 0.16358670592308044,
+      "learning_rate": 1.74521962012394e-05,
+      "loss": 0.5932368040084839,
+      "step": 374
+    },
+    {
+      "epoch": 0.5136986301369864,
+      "grad_norm": 0.15532813966274261,
+      "learning_rate": 1.7437383564289816e-05,
+      "loss": 0.5695762038230896,
+      "step": 375
+    },
+    {
+      "epoch": 0.5150684931506849,
+      "grad_norm": 0.15465544164180756,
+      "learning_rate": 1.7422534317887204e-05,
+      "loss": 0.6023073196411133,
+      "step": 376
+    },
+    {
+      "epoch": 0.5164383561643836,
+      "grad_norm": 0.1464887410402298,
+      "learning_rate": 1.740764853512485e-05,
+      "loss": 0.5693724155426025,
+      "step": 377
+    },
+    {
+      "epoch": 0.5178082191780822,
+      "grad_norm": 0.1357395201921463,
+      "learning_rate": 1.739272628927588e-05,
+      "loss": 0.536882221698761,
+      "step": 378
+    },
+    {
+      "epoch": 0.5191780821917809,
+      "grad_norm": 0.15610602498054504,
+      "learning_rate": 1.7377767653792915e-05,
+      "loss": 0.6062382459640503,
+      "step": 379
+    },
+    {
+      "epoch": 0.5205479452054794,
+      "grad_norm": 0.1343219131231308,
+      "learning_rate": 1.7362772702307677e-05,
+      "loss": 0.5632308125495911,
+      "step": 380
+    },
+    {
+      "epoch": 0.5219178082191781,
+      "grad_norm": 0.15014639496803284,
+      "learning_rate": 1.7347741508630673e-05,
+      "loss": 0.5424991250038147,
+      "step": 381
+    },
+    {
+      "epoch": 0.5232876712328767,
+      "grad_norm": 0.13980846107006073,
+      "learning_rate": 1.7332674146750796e-05,
+      "loss": 0.5513417720794678,
+      "step": 382
+    },
+    {
+      "epoch": 0.5246575342465754,
+      "grad_norm": 0.1392579823732376,
+      "learning_rate": 1.731757069083497e-05,
+      "loss": 0.5692585110664368,
+      "step": 383
+    },
+    {
+      "epoch": 0.5260273972602739,
+      "grad_norm": 0.1480334848165512,
+      "learning_rate": 1.7302431215227782e-05,
+      "loss": 0.540658712387085,
+      "step": 384
+    },
+    {
+      "epoch": 0.5273972602739726,
+      "grad_norm": 0.15066419541835785,
+      "learning_rate": 1.728725579445114e-05,
+      "loss": 0.5327310562133789,
+      "step": 385
+    },
+    {
+      "epoch": 0.5287671232876713,
+      "grad_norm": 0.14655138552188873,
+      "learning_rate": 1.727204450320387e-05,
+      "loss": 0.5699806213378906,
+      "step": 386
+    },
+    {
+      "epoch": 0.5301369863013699,
+      "grad_norm": 0.13703569769859314,
+      "learning_rate": 1.725679741636136e-05,
+      "loss": 0.5863059759140015,
+      "step": 387
+    },
+    {
+      "epoch": 0.5315068493150685,
+      "grad_norm": 0.1514001339673996,
+      "learning_rate": 1.7241514608975224e-05,
+      "loss": 0.5537671446800232,
+      "step": 388
+    },
+    {
+      "epoch": 0.5328767123287671,
+      "grad_norm": 0.14395487308502197,
+      "learning_rate": 1.7226196156272873e-05,
+      "loss": 0.60512775182724,
+      "step": 389
+    },
+    {
+      "epoch": 0.5342465753424658,
+      "grad_norm": 0.143971249461174,
+      "learning_rate": 1.7210842133657197e-05,
+      "loss": 0.5473227500915527,
+      "step": 390
+    },
+    {
+      "epoch": 0.5356164383561643,
+      "grad_norm": 0.15576298534870148,
+      "learning_rate": 1.719545261670617e-05,
+      "loss": 0.601518452167511,
+      "step": 391
+    },
+    {
+      "epoch": 0.536986301369863,
+      "grad_norm": 0.14324787259101868,
+      "learning_rate": 1.718002768117248e-05,
+      "loss": 0.5695725679397583,
+      "step": 392
+    },
+    {
+      "epoch": 0.5383561643835616,
+      "grad_norm": 0.14859525859355927,
+      "learning_rate": 1.7164567402983153e-05,
+      "loss": 0.6340662240982056,
+      "step": 393
+    },
+    {
+      "epoch": 0.5397260273972603,
+      "grad_norm": 0.13690751791000366,
+      "learning_rate": 1.7149071858239195e-05,
+      "loss": 0.5283052325248718,
+      "step": 394
+    },
+    {
+      "epoch": 0.541095890410959,
+      "grad_norm": 0.1423366665840149,
+      "learning_rate": 1.7133541123215196e-05,
+      "loss": 0.5996649265289307,
+      "step": 395
+    },
+    {
+      "epoch": 0.5424657534246575,
+      "grad_norm": 0.15085336565971375,
+      "learning_rate": 1.7117975274358975e-05,
+      "loss": 0.5998474955558777,
+      "step": 396
+    },
+    {
+      "epoch": 0.5438356164383562,
+      "grad_norm": 0.15080514550209045,
+      "learning_rate": 1.7102374388291182e-05,
+      "loss": 0.5708898901939392,
+      "step": 397
+    },
+    {
+      "epoch": 0.5452054794520548,
+      "grad_norm": 0.131844162940979,
+      "learning_rate": 1.7086738541804947e-05,
+      "loss": 0.5706642866134644,
+      "step": 398
+    },
+    {
+      "epoch": 0.5465753424657535,
+      "grad_norm": 0.1624789983034134,
+      "learning_rate": 1.7071067811865477e-05,
+      "loss": 0.5648869872093201,
+      "step": 399
+    },
+    {
+      "epoch": 0.547945205479452,
+      "grad_norm": 0.14977680146694183,
+      "learning_rate": 1.7055362275609687e-05,
+      "loss": 0.5466816425323486,
+      "step": 400
+    },
+    {
+      "epoch": 0.5493150684931507,
+      "grad_norm": 0.12936653196811676,
+      "learning_rate": 1.7039622010345832e-05,
+      "loss": 0.5807684659957886,
+      "step": 401
+    },
+    {
+      "epoch": 0.5506849315068493,
+      "grad_norm": 0.15499675273895264,
+      "learning_rate": 1.70238470935531e-05,
+      "loss": 0.5646220445632935,
+      "step": 402
+    },
+    {
+      "epoch": 0.552054794520548,
+      "grad_norm": 0.15363413095474243,
+      "learning_rate": 1.700803760288126e-05,
+      "loss": 0.5537654161453247,
+      "step": 403
+    },
+    {
+      "epoch": 0.5534246575342465,
+      "grad_norm": 0.14746566116809845,
+      "learning_rate": 1.699219361615026e-05,
+      "loss": 0.5865470170974731,
+      "step": 404
+    },
+    {
+      "epoch": 0.5547945205479452,
+      "grad_norm": 0.14045830070972443,
+      "learning_rate": 1.6976315211349848e-05,
+      "loss": 0.5434695482254028,
+      "step": 405
+    },
+    {
+      "epoch": 0.5561643835616439,
+      "grad_norm": 0.15868021547794342,
+      "learning_rate": 1.6960402466639195e-05,
+      "loss": 0.5476789474487305,
+      "step": 406
+    },
+    {
+      "epoch": 0.5575342465753425,
+      "grad_norm": 0.1462065726518631,
+      "learning_rate": 1.6944455460346503e-05,
+      "loss": 0.5196582078933716,
+      "step": 407
+    },
+    {
+      "epoch": 0.5589041095890411,
+      "grad_norm": 0.14001627266407013,
+      "learning_rate": 1.692847427096862e-05,
+      "loss": 0.6024141311645508,
+      "step": 408
+    },
+    {
+      "epoch": 0.5602739726027397,
+      "grad_norm": 0.15260235965251923,
+      "learning_rate": 1.6912458977170655e-05,
+      "loss": 0.6341221332550049,
+      "step": 409
+    },
+    {
+      "epoch": 0.5616438356164384,
+      "grad_norm": 0.14107351005077362,
+      "learning_rate": 1.6896409657785598e-05,
+      "loss": 0.5811542272567749,
+      "step": 410
+    },
+    {
+      "epoch": 0.563013698630137,
+      "grad_norm": 0.1302739977836609,
+      "learning_rate": 1.6880326391813917e-05,
+      "loss": 0.544140100479126,
+      "step": 411
+    },
+    {
+      "epoch": 0.5643835616438356,
+      "grad_norm": 0.1420375555753708,
+      "learning_rate": 1.6864209258423177e-05,
+      "loss": 0.534217119216919,
+      "step": 412
+    },
+    {
+      "epoch": 0.5657534246575342,
+      "grad_norm": 0.14836066961288452,
+      "learning_rate": 1.6848058336947657e-05,
+      "loss": 0.6192389726638794,
+      "step": 413
+    },
+    {
+      "epoch": 0.5671232876712329,
+      "grad_norm": 0.16043499112129211,
+      "learning_rate": 1.683187370688795e-05,
+      "loss": 0.5258642435073853,
+      "step": 414
+    },
+    {
+      "epoch": 0.5684931506849316,
+      "grad_norm": 0.14773382246494293,
+      "learning_rate": 1.6815655447910574e-05,
+      "loss": 0.5696762800216675,
+      "step": 415
+    },
+    {
+      "epoch": 0.5698630136986301,
+      "grad_norm": 0.14542289078235626,
+      "learning_rate": 1.6799403639847582e-05,
+      "loss": 0.5968167185783386,
+      "step": 416
+    },
+    {
+      "epoch": 0.5712328767123288,
+      "grad_norm": 0.14496225118637085,
+      "learning_rate": 1.6783118362696162e-05,
+      "loss": 0.5987216830253601,
+      "step": 417
+    },
+    {
+      "epoch": 0.5726027397260274,
+      "grad_norm": 0.14567093551158905,
+      "learning_rate": 1.6766799696618257e-05,
+      "loss": 0.5913962125778198,
+      "step": 418
+    },
+    {
+      "epoch": 0.5739726027397261,
+      "grad_norm": 0.15489841997623444,
+      "learning_rate": 1.6750447721940157e-05,
+      "loss": 0.6150340437889099,
+      "step": 419
+    },
+    {
+      "epoch": 0.5753424657534246,
+      "grad_norm": 0.13718822598457336,
+      "learning_rate": 1.6734062519152113e-05,
+      "loss": 0.5184439420700073,
+      "step": 420
+    },
+    {
+      "epoch": 0.5767123287671233,
+      "grad_norm": 0.14923328161239624,
+      "learning_rate": 1.671764416890793e-05,
+      "loss": 0.6138299703598022,
+      "step": 421
+    },
+    {
+      "epoch": 0.5780821917808219,
+      "grad_norm": 0.16161896288394928,
+      "learning_rate": 1.670119275202458e-05,
+      "loss": 0.6450883150100708,
+      "step": 422
+    },
+    {
+      "epoch": 0.5794520547945206,
+      "grad_norm": 0.16076239943504333,
+      "learning_rate": 1.6684708349481808e-05,
+      "loss": 0.6587648391723633,
+      "step": 423
+    },
+    {
+      "epoch": 0.5808219178082191,
+      "grad_norm": 0.1475391983985901,
+      "learning_rate": 1.666819104242171e-05,
+      "loss": 0.5829423069953918,
+      "step": 424
+    },
+    {
+      "epoch": 0.5821917808219178,
+      "grad_norm": 0.159859299659729,
+      "learning_rate": 1.6651640912148362e-05,
+      "loss": 0.5809226632118225,
+      "step": 425
+    },
+    {
+      "epoch": 0.5835616438356165,
+      "grad_norm": 0.1477399319410324,
+      "learning_rate": 1.6635058040127408e-05,
+      "loss": 0.6071959733963013,
+      "step": 426
+    },
+    {
+      "epoch": 0.584931506849315,
+      "grad_norm": 0.1518445909023285,
+      "learning_rate": 1.661844250798565e-05,
+      "loss": 0.5572406649589539,
+      "step": 427
+    },
+    {
+      "epoch": 0.5863013698630137,
+      "grad_norm": 0.14316509664058685,
+      "learning_rate": 1.6601794397510666e-05,
+      "loss": 0.5702495574951172,
+      "step": 428
+    },
+    {
+      "epoch": 0.5876712328767123,
+      "grad_norm": 0.13820317387580872,
+      "learning_rate": 1.658511379065039e-05,
+      "loss": 0.6022301912307739,
+      "step": 429
+    },
+    {
+      "epoch": 0.589041095890411,
+      "grad_norm": 0.1447012573480606,
+      "learning_rate": 1.6568400769512708e-05,
+      "loss": 0.5637190341949463,
+      "step": 430
+    },
+    {
+      "epoch": 0.5904109589041096,
+      "grad_norm": 0.15339450538158417,
+      "learning_rate": 1.6551655416365072e-05,
+      "loss": 0.5606421232223511,
+      "step": 431
+    },
+    {
+      "epoch": 0.5917808219178082,
+      "grad_norm": 0.1346173882484436,
+      "learning_rate": 1.653487781363408e-05,
+      "loss": 0.5061419606208801,
+      "step": 432
+    },
+    {
+      "epoch": 0.5931506849315068,
+      "grad_norm": 0.14965689182281494,
+      "learning_rate": 1.6518068043905075e-05,
+      "loss": 0.5600637197494507,
+      "step": 433
+    },
+    {
+      "epoch": 0.5945205479452055,
+      "grad_norm": 0.13996529579162598,
+      "learning_rate": 1.650122618992173e-05,
+      "loss": 0.5642286539077759,
+      "step": 434
+    },
+    {
+      "epoch": 0.5958904109589042,
+      "grad_norm": 0.14598211646080017,
+      "learning_rate": 1.6484352334585654e-05,
+      "loss": 0.5623951554298401,
+      "step": 435
+    },
+    {
+      "epoch": 0.5972602739726027,
+      "grad_norm": 0.15535297989845276,
+      "learning_rate": 1.6467446560955982e-05,
+      "loss": 0.5812850594520569,
+      "step": 436
+    },
+    {
+      "epoch": 0.5986301369863014,
+      "grad_norm": 0.15179552137851715,
+      "learning_rate": 1.6450508952248957e-05,
+      "loss": 0.5285508632659912,
+      "step": 437
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.13990969955921173,
+      "learning_rate": 1.6433539591837527e-05,
+      "loss": 0.4808741807937622,
+      "step": 438
+    },
+    {
+      "epoch": 0.6013698630136987,
+      "grad_norm": 0.13985474407672882,
+      "learning_rate": 1.641653856325093e-05,
+      "loss": 0.526069164276123,
+      "step": 439
+    },
+    {
+      "epoch": 0.6027397260273972,
+      "grad_norm": 0.14668938517570496,
+      "learning_rate": 1.639950595017429e-05,
+      "loss": 0.55939781665802,
+      "step": 440
+    },
+    {
+      "epoch": 0.6041095890410959,
+      "grad_norm": 0.1409137099981308,
+      "learning_rate": 1.6382441836448203e-05,
+      "loss": 0.5505657196044922,
+      "step": 441
+    },
+    {
+      "epoch": 0.6054794520547945,
+      "grad_norm": 0.13390517234802246,
+      "learning_rate": 1.6365346306068315e-05,
+      "loss": 0.5604054927825928,
+      "step": 442
+    },
+    {
+      "epoch": 0.6068493150684932,
+      "grad_norm": 0.1515313684940338,
+      "learning_rate": 1.6348219443184927e-05,
+      "loss": 0.600440263748169,
+      "step": 443
+    },
+    {
+      "epoch": 0.6082191780821918,
+      "grad_norm": 0.15308818221092224,
+      "learning_rate": 1.633106133210255e-05,
+      "loss": 0.5832364559173584,
+      "step": 444
+    },
+    {
+      "epoch": 0.6095890410958904,
+      "grad_norm": 0.162045419216156,
+      "learning_rate": 1.6313872057279536e-05,
+      "loss": 0.577210009098053,
+      "step": 445
+    },
+    {
+      "epoch": 0.6109589041095891,
+      "grad_norm": 0.14119122922420502,
+      "learning_rate": 1.629665170332762e-05,
+      "loss": 0.5397899746894836,
+      "step": 446
+    },
+    {
+      "epoch": 0.6123287671232877,
+      "grad_norm": 0.13274981081485748,
+      "learning_rate": 1.627940035501152e-05,
+      "loss": 0.4862290024757385,
+      "step": 447
+    },
+    {
+      "epoch": 0.6136986301369863,
+      "grad_norm": 0.14096492528915405,
+      "learning_rate": 1.6262118097248524e-05,
+      "loss": 0.5487799048423767,
+      "step": 448
+    },
+    {
+      "epoch": 0.6150684931506849,
+      "grad_norm": 0.12473436444997787,
+      "learning_rate": 1.624480501510806e-05,
+      "loss": 0.505416989326477,
+      "step": 449
+    },
+    {
+      "epoch": 0.6164383561643836,
+      "grad_norm": 0.16270937025547028,
+      "learning_rate": 1.62274611938113e-05,
+      "loss": 0.5941034555435181,
+      "step": 450
+    },
+    {
+      "epoch": 0.6178082191780822,
+      "grad_norm": 0.1519913524389267,
+      "learning_rate": 1.62100867187307e-05,
+      "loss": 0.589063286781311,
+      "step": 451
+    },
+    {
+      "epoch": 0.6191780821917808,
+      "grad_norm": 0.1484452784061432,
+      "learning_rate": 1.619268167538963e-05,
+      "loss": 0.6294279098510742,
+      "step": 452
+    },
+    {
+      "epoch": 0.6205479452054794,
+      "grad_norm": 0.1418556571006775,
+      "learning_rate": 1.617524614946192e-05,
+      "loss": 0.6004430055618286,
+      "step": 453
+    },
+    {
+      "epoch": 0.6219178082191781,
+      "grad_norm": 0.14398352801799774,
+      "learning_rate": 1.6157780226771437e-05,
+      "loss": 0.5689352750778198,
+      "step": 454
+    },
+    {
+      "epoch": 0.6232876712328768,
+      "grad_norm": 0.14114904403686523,
+      "learning_rate": 1.614028399329168e-05,
+      "loss": 0.571104884147644,
+      "step": 455
+    },
+    {
+      "epoch": 0.6246575342465753,
+      "grad_norm": 0.14283505082130432,
+      "learning_rate": 1.6122757535145346e-05,
+      "loss": 0.5134957432746887,
+      "step": 456
+    },
+    {
+      "epoch": 0.626027397260274,
+      "grad_norm": 0.14088678359985352,
+      "learning_rate": 1.6105200938603917e-05,
+      "loss": 0.6189490556716919,
+      "step": 457
+    },
+    {
+      "epoch": 0.6273972602739726,
+      "grad_norm": 0.1482560932636261,
+      "learning_rate": 1.608761429008721e-05,
+      "loss": 0.5793185234069824,
+      "step": 458
+    },
+    {
+      "epoch": 0.6287671232876713,
+      "grad_norm": 0.16769099235534668,
+      "learning_rate": 1.606999767616298e-05,
+      "loss": 0.5795351266860962,
+      "step": 459
+    },
+    {
+      "epoch": 0.6301369863013698,
+      "grad_norm": 0.14694984257221222,
+      "learning_rate": 1.6052351183546478e-05,
+      "loss": 0.6406443119049072,
+      "step": 460
+    },
+    {
+      "epoch": 0.6315068493150685,
+      "grad_norm": 0.1542738825082779,
+      "learning_rate": 1.603467489910004e-05,
+      "loss": 0.5569589138031006,
+      "step": 461
+    },
+    {
+      "epoch": 0.6328767123287671,
+      "grad_norm": 0.1484406739473343,
+      "learning_rate": 1.6016968909832632e-05,
+      "loss": 0.563373863697052,
+      "step": 462
+    },
+    {
+      "epoch": 0.6342465753424658,
+      "grad_norm": 0.14980265498161316,
+      "learning_rate": 1.5999233302899444e-05,
+      "loss": 0.5926274657249451,
+      "step": 463
+    },
+    {
+      "epoch": 0.6356164383561644,
+      "grad_norm": 0.14588281512260437,
+      "learning_rate": 1.5981468165601462e-05,
+      "loss": 0.5282426476478577,
+      "step": 464
+    },
+    {
+      "epoch": 0.636986301369863,
+      "grad_norm": 0.14488136768341064,
+      "learning_rate": 1.5963673585385016e-05,
+      "loss": 0.5637092590332031,
+      "step": 465
+    },
+    {
+      "epoch": 0.6383561643835617,
+      "grad_norm": 0.14770545065402985,
+      "learning_rate": 1.594584964984138e-05,
+      "loss": 0.5751156806945801,
+      "step": 466
+    },
+    {
+      "epoch": 0.6397260273972603,
+      "grad_norm": 0.14717961847782135,
+      "learning_rate": 1.5927996446706308e-05,
+      "loss": 0.5032989382743835,
+      "step": 467
+    },
+    {
+      "epoch": 0.6410958904109589,
+      "grad_norm": 0.1413588970899582,
+      "learning_rate": 1.591011406385964e-05,
+      "loss": 0.5489969253540039,
+      "step": 468
+    },
+    {
+      "epoch": 0.6424657534246575,
+      "grad_norm": 0.13199076056480408,
+      "learning_rate": 1.5892202589324835e-05,
+      "loss": 0.5251445770263672,
+      "step": 469
+    },
+    {
+      "epoch": 0.6438356164383562,
+      "grad_norm": 0.14698363840579987,
+      "learning_rate": 1.587426211126856e-05,
+      "loss": 0.5794821977615356,
+      "step": 470
+    },
+    {
+      "epoch": 0.6452054794520548,
+      "grad_norm": 0.139966681599617,
+      "learning_rate": 1.5856292718000235e-05,
+      "loss": 0.5483118891716003,
+      "step": 471
+    },
+    {
+      "epoch": 0.6465753424657534,
+      "grad_norm": 0.1384710967540741,
+      "learning_rate": 1.5838294497971622e-05,
+      "loss": 0.5625190734863281,
+      "step": 472
+    },
+    {
+      "epoch": 0.647945205479452,
+      "grad_norm": 0.15267488360404968,
+      "learning_rate": 1.582026753977638e-05,
+      "loss": 0.5484758615493774,
+      "step": 473
+    },
+    {
+      "epoch": 0.6493150684931507,
+      "grad_norm": 0.15189437568187714,
+      "learning_rate": 1.5802211932149614e-05,
+      "loss": 0.5427678823471069,
+      "step": 474
+    },
+    {
+      "epoch": 0.6506849315068494,
+      "grad_norm": 0.1514596790075302,
+      "learning_rate": 1.5784127763967473e-05,
+      "loss": 0.5608749389648438,
+      "step": 475
+    },
+    {
+      "epoch": 0.6520547945205479,
+      "grad_norm": 0.13448850810527802,
+      "learning_rate": 1.5766015124246664e-05,
+      "loss": 0.551884114742279,
+      "step": 476
+    },
+    {
+      "epoch": 0.6534246575342466,
+      "grad_norm": 0.13851968944072723,
+      "learning_rate": 1.5747874102144073e-05,
+      "loss": 0.5595521330833435,
+      "step": 477
+    },
+    {
+      "epoch": 0.6547945205479452,
+      "grad_norm": 0.14666830003261566,
+      "learning_rate": 1.5729704786956262e-05,
+      "loss": 0.5839216113090515,
+      "step": 478
+    },
+    {
+      "epoch": 0.6561643835616439,
+      "grad_norm": 0.15627364814281464,
+      "learning_rate": 1.5711507268119087e-05,
+      "loss": 0.5988858938217163,
+      "step": 479
+    },
+    {
+      "epoch": 0.6575342465753424,
+      "grad_norm": 0.1463092565536499,
+      "learning_rate": 1.5693281635207214e-05,
+      "loss": 0.537287175655365,
+      "step": 480
+    },
+    {
+      "epoch": 0.6589041095890411,
+      "grad_norm": 0.1374531239271164,
+      "learning_rate": 1.567502797793372e-05,
+      "loss": 0.5598975419998169,
+      "step": 481
+    },
+    {
+      "epoch": 0.6602739726027397,
+      "grad_norm": 0.12984918057918549,
+      "learning_rate": 1.5656746386149596e-05,
+      "loss": 0.5133352279663086,
+      "step": 482
+    },
+    {
+      "epoch": 0.6616438356164384,
+      "grad_norm": 0.15310728549957275,
+      "learning_rate": 1.563843694984336e-05,
+      "loss": 0.570267915725708,
+      "step": 483
+    },
+    {
+      "epoch": 0.663013698630137,
+      "grad_norm": 0.16105115413665771,
+      "learning_rate": 1.5620099759140587e-05,
+      "loss": 0.6069197654724121,
+      "step": 484
+    },
+    {
+      "epoch": 0.6643835616438356,
+      "grad_norm": 0.14937542378902435,
+      "learning_rate": 1.560173490430346e-05,
+      "loss": 0.582807719707489,
+      "step": 485
+    },
+    {
+      "epoch": 0.6657534246575343,
+      "grad_norm": 0.14819355309009552,
+      "learning_rate": 1.558334247573035e-05,
+      "loss": 0.4967902898788452,
+      "step": 486
+    },
+    {
+      "epoch": 0.6671232876712329,
+      "grad_norm": 0.13898441195487976,
+      "learning_rate": 1.5564922563955337e-05,
+      "loss": 0.5023531317710876,
+      "step": 487
+    },
+    {
+      "epoch": 0.6684931506849315,
+      "grad_norm": 0.1410035490989685,
+      "learning_rate": 1.5546475259647795e-05,
+      "loss": 0.5132713317871094,
+      "step": 488
+    },
+    {
+      "epoch": 0.6698630136986301,
+      "grad_norm": 0.1309722661972046,
+      "learning_rate": 1.5528000653611935e-05,
+      "loss": 0.5644882321357727,
+      "step": 489
+    },
+    {
+      "epoch": 0.6712328767123288,
+      "grad_norm": 0.13525459170341492,
+      "learning_rate": 1.550949883678635e-05,
+      "loss": 0.5111340284347534,
+      "step": 490
+    },
+    {
+      "epoch": 0.6726027397260274,
+      "grad_norm": 0.14471714198589325,
+      "learning_rate": 1.549096990024358e-05,
+      "loss": 0.5402161478996277,
+      "step": 491
+    },
+    {
+      "epoch": 0.673972602739726,
+      "grad_norm": 0.1380838304758072,
+      "learning_rate": 1.5472413935189656e-05,
+      "loss": 0.57719486951828,
+      "step": 492
+    },
+    {
+      "epoch": 0.6753424657534246,
+      "grad_norm": 0.1449180692434311,
+      "learning_rate": 1.5453831032963648e-05,
+      "loss": 0.4582573175430298,
+      "step": 493
+    },
+    {
+      "epoch": 0.6767123287671233,
+      "grad_norm": 0.16058871150016785,
+      "learning_rate": 1.543522128503723e-05,
+      "loss": 0.6209670305252075,
+      "step": 494
+    },
+    {
+      "epoch": 0.678082191780822,
+      "grad_norm": 0.1415194571018219,
+      "learning_rate": 1.541658478301421e-05,
+      "loss": 0.542889416217804,
+      "step": 495
+    },
+    {
+      "epoch": 0.6794520547945205,
+      "grad_norm": 0.13594785332679749,
+      "learning_rate": 1.53979216186301e-05,
+      "loss": 0.5576387047767639,
+      "step": 496
+    },
+    {
+      "epoch": 0.6808219178082192,
+      "grad_norm": 0.16646577417850494,
+      "learning_rate": 1.537923188375164e-05,
+      "loss": 0.6171765327453613,
+      "step": 497
+    },
+    {
+      "epoch": 0.6821917808219178,
+      "grad_norm": 0.14226126670837402,
+      "learning_rate": 1.5360515670376373e-05,
+      "loss": 0.5574626922607422,
+      "step": 498
+    },
+    {
+      "epoch": 0.6835616438356165,
+      "grad_norm": 0.14714594185352325,
+      "learning_rate": 1.5341773070632172e-05,
+      "loss": 0.5930618643760681,
+      "step": 499
+    },
+    {
+      "epoch": 0.684931506849315,
+      "grad_norm": 0.12718486785888672,
+      "learning_rate": 1.5323004176776796e-05,
+      "loss": 0.49652647972106934,
+      "step": 500
+    },
+    {
+      "epoch": 0.6863013698630137,
+      "grad_norm": 0.13479647040367126,
+      "learning_rate": 1.5304209081197425e-05,
+      "loss": 0.5470659732818604,
+      "step": 501
+    },
+    {
+      "epoch": 0.6876712328767123,
+      "grad_norm": 0.1462414562702179,
+      "learning_rate": 1.5285387876410224e-05,
+      "loss": 0.558161735534668,
+      "step": 502
+    },
+    {
+      "epoch": 0.689041095890411,
+      "grad_norm": 0.1436636745929718,
+      "learning_rate": 1.5266540655059872e-05,
+      "loss": 0.5366897583007812,
+      "step": 503
+    },
+    {
+      "epoch": 0.6904109589041096,
+      "grad_norm": 0.15740245580673218,
+      "learning_rate": 1.5247667509919104e-05,
+      "loss": 0.5611869096755981,
+      "step": 504
+    },
+    {
+      "epoch": 0.6917808219178082,
+      "grad_norm": 0.1479470580816269,
+      "learning_rate": 1.5228768533888278e-05,
+      "loss": 0.5563034415245056,
+      "step": 505
+    },
+    {
+      "epoch": 0.6931506849315069,
+      "grad_norm": 0.14391250908374786,
+      "learning_rate": 1.520984381999488e-05,
+      "loss": 0.5359901189804077,
+      "step": 506
+    },
+    {
+      "epoch": 0.6945205479452055,
+      "grad_norm": 0.14507880806922913,
+      "learning_rate": 1.5190893461393108e-05,
+      "loss": 0.5985116958618164,
+      "step": 507
+    },
+    {
+      "epoch": 0.6958904109589041,
+      "grad_norm": 0.12985731661319733,
+      "learning_rate": 1.5171917551363376e-05,
+      "loss": 0.5191558003425598,
+      "step": 508
+    },
+    {
+      "epoch": 0.6972602739726027,
+      "grad_norm": 0.13191670179367065,
+      "learning_rate": 1.5152916183311881e-05,
+      "loss": 0.5371098518371582,
+      "step": 509
+    },
+    {
+      "epoch": 0.6986301369863014,
+      "grad_norm": 0.15381883084774017,
+      "learning_rate": 1.5133889450770122e-05,
+      "loss": 0.582601010799408,
+      "step": 510
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.13812492787837982,
+      "learning_rate": 1.5114837447394469e-05,
+      "loss": 0.5940638780593872,
+      "step": 511
+    },
+    {
+      "epoch": 0.7013698630136986,
+      "grad_norm": 0.14924225211143494,
+      "learning_rate": 1.5095760266965668e-05,
+      "loss": 0.5940001010894775,
+      "step": 512
+    },
+    {
+      "epoch": 0.7027397260273973,
+      "grad_norm": 0.14933045208454132,
+      "learning_rate": 1.50766580033884e-05,
+      "loss": 0.62128746509552,
+      "step": 513
+    },
+    {
+      "epoch": 0.7041095890410959,
+      "grad_norm": 0.15632203221321106,
+      "learning_rate": 1.5057530750690822e-05,
+      "loss": 0.5551106929779053,
+      "step": 514
+    },
+    {
+      "epoch": 0.7054794520547946,
+      "grad_norm": 0.13439548015594482,
+      "learning_rate": 1.5038378603024085e-05,
+      "loss": 0.5119856595993042,
+      "step": 515
+    },
+    {
+      "epoch": 0.7068493150684931,
+      "grad_norm": 0.14251737296581268,
+      "learning_rate": 1.5019201654661886e-05,
+      "loss": 0.5376251935958862,
+      "step": 516
+    },
+    {
+      "epoch": 0.7082191780821918,
+      "grad_norm": 0.1485617607831955,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.5784671306610107,
+      "step": 517
+    },
+    {
+      "epoch": 0.7095890410958904,
+      "grad_norm": 0.1372535526752472,
+      "learning_rate": 1.4980773733555823e-05,
+      "loss": 0.558871865272522,
+      "step": 518
+    },
+    {
+      "epoch": 0.7109589041095891,
+      "grad_norm": 0.1411508321762085,
+      "learning_rate": 1.4961522949967887e-05,
+      "loss": 0.5023566484451294,
+      "step": 519
+    },
+    {
+      "epoch": 0.7123287671232876,
+      "grad_norm": 0.15227437019348145,
+      "learning_rate": 1.4942247743995407e-05,
+      "loss": 0.5967291593551636,
+      "step": 520
+    },
+    {
+      "epoch": 0.7136986301369863,
+      "grad_norm": 0.15094850957393646,
+      "learning_rate": 1.4922948210517822e-05,
+      "loss": 0.6140634417533875,
+      "step": 521
+    },
+    {
+      "epoch": 0.7150684931506849,
+      "grad_norm": 0.1562165468931198,
+      "learning_rate": 1.4903624444534317e-05,
+      "loss": 0.5756810307502747,
+      "step": 522
+    },
+    {
+      "epoch": 0.7164383561643836,
+      "grad_norm": 0.14749285578727722,
+      "learning_rate": 1.4884276541163351e-05,
+      "loss": 0.5795880556106567,
+      "step": 523
+    },
+    {
+      "epoch": 0.7178082191780822,
+      "grad_norm": 0.13962800800800323,
+      "learning_rate": 1.4864904595642206e-05,
+      "loss": 0.5142415761947632,
+      "step": 524
+    },
+    {
+      "epoch": 0.7191780821917808,
+      "grad_norm": 0.14828866720199585,
+      "learning_rate": 1.4845508703326504e-05,
+      "loss": 0.5461621284484863,
+      "step": 525
+    },
+    {
+      "epoch": 0.7205479452054795,
+      "grad_norm": 0.14395621418952942,
+      "learning_rate": 1.4826088959689734e-05,
+      "loss": 0.6380258202552795,
+      "step": 526
+    },
+    {
+      "epoch": 0.7219178082191781,
+      "grad_norm": 0.15022049844264984,
+      "learning_rate": 1.4806645460322804e-05,
+      "loss": 0.5415664315223694,
+      "step": 527
+    },
+    {
+      "epoch": 0.7232876712328767,
+      "grad_norm": 0.14218366146087646,
+      "learning_rate": 1.4787178300933543e-05,
+      "loss": 0.5103275775909424,
+      "step": 528
+    },
+    {
+      "epoch": 0.7246575342465753,
+      "grad_norm": 0.15063804388046265,
+      "learning_rate": 1.4767687577346253e-05,
+      "loss": 0.5823582410812378,
+      "step": 529
+    },
+    {
+      "epoch": 0.726027397260274,
+      "grad_norm": 0.14951996505260468,
+      "learning_rate": 1.4748173385501222e-05,
+      "loss": 0.5916295051574707,
+      "step": 530
+    },
+    {
+      "epoch": 0.7273972602739726,
+      "grad_norm": 0.15501154959201813,
+      "learning_rate": 1.4728635821454255e-05,
+      "loss": 0.5518443584442139,
+      "step": 531
+    },
+    {
+      "epoch": 0.7287671232876712,
+      "grad_norm": 0.15150362253189087,
+      "learning_rate": 1.470907498137621e-05,
+      "loss": 0.5295286178588867,
+      "step": 532
+    },
+    {
+      "epoch": 0.7301369863013699,
+      "grad_norm": 0.14389854669570923,
+      "learning_rate": 1.4689490961552513e-05,
+      "loss": 0.573415994644165,
+      "step": 533
+    },
+    {
+      "epoch": 0.7315068493150685,
+      "grad_norm": 0.14579284191131592,
+      "learning_rate": 1.4669883858382689e-05,
+      "loss": 0.5175173282623291,
+      "step": 534
+    },
+    {
+      "epoch": 0.7328767123287672,
+      "grad_norm": 0.16025300323963165,
+      "learning_rate": 1.4650253768379896e-05,
+      "loss": 0.5780712366104126,
+      "step": 535
+    },
+    {
+      "epoch": 0.7342465753424657,
+      "grad_norm": 0.15502531826496124,
+      "learning_rate": 1.4630600788170429e-05,
+      "loss": 0.5766080617904663,
+      "step": 536
+    },
+    {
+      "epoch": 0.7356164383561644,
+      "grad_norm": 0.1452580839395523,
+      "learning_rate": 1.461092501449326e-05,
+      "loss": 0.5627033710479736,
+      "step": 537
+    },
+    {
+      "epoch": 0.736986301369863,
+      "grad_norm": 0.14218607544898987,
+      "learning_rate": 1.4591226544199564e-05,
+      "loss": 0.5479629635810852,
+      "step": 538
+    },
+    {
+      "epoch": 0.7383561643835617,
+      "grad_norm": 0.14063885807991028,
+      "learning_rate": 1.4571505474252234e-05,
+      "loss": 0.5578610301017761,
+      "step": 539
+    },
+    {
+      "epoch": 0.7397260273972602,
+      "grad_norm": 0.14620015025138855,
+      "learning_rate": 1.4551761901725402e-05,
+      "loss": 0.5858629941940308,
+      "step": 540
+    },
+    {
+      "epoch": 0.7410958904109589,
+      "grad_norm": 0.14063887298107147,
+      "learning_rate": 1.4531995923803974e-05,
+      "loss": 0.5905371904373169,
+      "step": 541
+    },
+    {
+      "epoch": 0.7424657534246575,
+      "grad_norm": 0.13075634837150574,
+      "learning_rate": 1.4512207637783138e-05,
+      "loss": 0.5407640337944031,
+      "step": 542
+    },
+    {
+      "epoch": 0.7438356164383562,
+      "grad_norm": 0.13585825264453888,
+      "learning_rate": 1.4492397141067888e-05,
+      "loss": 0.5683429837226868,
+      "step": 543
+    },
+    {
+      "epoch": 0.7452054794520548,
+      "grad_norm": 0.14631445705890656,
+      "learning_rate": 1.4472564531172558e-05,
+      "loss": 0.5532336235046387,
+      "step": 544
+    },
+    {
+      "epoch": 0.7465753424657534,
+      "grad_norm": 0.15100950002670288,
+      "learning_rate": 1.4452709905720314e-05,
+      "loss": 0.5505513548851013,
+      "step": 545
+    },
+    {
+      "epoch": 0.7479452054794521,
+      "grad_norm": 0.15109765529632568,
+      "learning_rate": 1.4432833362442708e-05,
+      "loss": 0.5818567276000977,
+      "step": 546
+    },
+    {
+      "epoch": 0.7493150684931507,
+      "grad_norm": 0.15869656205177307,
+      "learning_rate": 1.4412934999179169e-05,
+      "loss": 0.591871976852417,
+      "step": 547
+    },
+    {
+      "epoch": 0.7506849315068493,
+      "grad_norm": 0.13581599295139313,
+      "learning_rate": 1.4393014913876528e-05,
+      "loss": 0.5127702355384827,
+      "step": 548
+    },
+    {
+      "epoch": 0.7520547945205479,
+      "grad_norm": 0.1417117863893509,
+      "learning_rate": 1.4373073204588556e-05,
+      "loss": 0.5655680894851685,
+      "step": 549
+    },
+    {
+      "epoch": 0.7534246575342466,
+      "grad_norm": 0.15534058213233948,
+      "learning_rate": 1.4353109969475447e-05,
+      "loss": 0.577724814414978,
+      "step": 550
+    },
+    {
+      "epoch": 0.7547945205479452,
+      "grad_norm": 0.1513068974018097,
+      "learning_rate": 1.4333125306803362e-05,
+      "loss": 0.5496415495872498,
+      "step": 551
+    },
+    {
+      "epoch": 0.7561643835616438,
+      "grad_norm": 0.14230354130268097,
+      "learning_rate": 1.4313119314943933e-05,
+      "loss": 0.525494396686554,
+      "step": 552
+    },
+    {
+      "epoch": 0.7575342465753425,
+      "grad_norm": 0.13738034665584564,
+      "learning_rate": 1.4293092092373786e-05,
+      "loss": 0.519819974899292,
+      "step": 553
+    },
+    {
+      "epoch": 0.7589041095890411,
+      "grad_norm": 0.13675081729888916,
+      "learning_rate": 1.4273043737674048e-05,
+      "loss": 0.5584299564361572,
+      "step": 554
+    },
+    {
+      "epoch": 0.7602739726027398,
+      "grad_norm": 0.1386847048997879,
+      "learning_rate": 1.4252974349529871e-05,
+      "loss": 0.5083411335945129,
+      "step": 555
+    },
+    {
+      "epoch": 0.7616438356164383,
+      "grad_norm": 0.14538908004760742,
+      "learning_rate": 1.4232884026729927e-05,
+      "loss": 0.6105762720108032,
+      "step": 556
+    },
+    {
+      "epoch": 0.763013698630137,
+      "grad_norm": 0.15406113862991333,
+      "learning_rate": 1.4212772868165957e-05,
+      "loss": 0.5772764086723328,
+      "step": 557
+    },
+    {
+      "epoch": 0.7643835616438356,
+      "grad_norm": 0.16195538640022278,
+      "learning_rate": 1.419264097283225e-05,
+      "loss": 0.6081218719482422,
+      "step": 558
+    },
+    {
+      "epoch": 0.7657534246575343,
+      "grad_norm": 0.12385565787553787,
+      "learning_rate": 1.4172488439825167e-05,
+      "loss": 0.4937037229537964,
+      "step": 559
+    },
+    {
+      "epoch": 0.7671232876712328,
+      "grad_norm": 0.15183141827583313,
+      "learning_rate": 1.4152315368342661e-05,
+      "loss": 0.5616321563720703,
+      "step": 560
+    },
+    {
+      "epoch": 0.7684931506849315,
+      "grad_norm": 0.13611982762813568,
+      "learning_rate": 1.4132121857683782e-05,
+      "loss": 0.5686037540435791,
+      "step": 561
+    },
+    {
+      "epoch": 0.7698630136986301,
+      "grad_norm": 0.1663551926612854,
+      "learning_rate": 1.4111908007248188e-05,
+      "loss": 0.6417747735977173,
+      "step": 562
+    },
+    {
+      "epoch": 0.7712328767123288,
+      "grad_norm": 0.14242985844612122,
+      "learning_rate": 1.4091673916535655e-05,
+      "loss": 0.5676991939544678,
+      "step": 563
+    },
+    {
+      "epoch": 0.7726027397260274,
+      "grad_norm": 0.14829294383525848,
+      "learning_rate": 1.4071419685145587e-05,
+      "loss": 0.5915441513061523,
+      "step": 564
+    },
+    {
+      "epoch": 0.773972602739726,
+      "grad_norm": 0.15141542255878448,
+      "learning_rate": 1.4051145412776536e-05,
+      "loss": 0.5690219402313232,
+      "step": 565
+    },
+    {
+      "epoch": 0.7753424657534247,
+      "grad_norm": 0.13884322345256805,
+      "learning_rate": 1.4030851199225697e-05,
+      "loss": 0.5482669472694397,
+      "step": 566
+    },
+    {
+      "epoch": 0.7767123287671233,
+      "grad_norm": 0.141120046377182,
+      "learning_rate": 1.4010537144388416e-05,
+      "loss": 0.5446934103965759,
+      "step": 567
+    },
+    {
+      "epoch": 0.7780821917808219,
+      "grad_norm": 0.14071466028690338,
+      "learning_rate": 1.3990203348257722e-05,
+      "loss": 0.5528297424316406,
+      "step": 568
+    },
+    {
+      "epoch": 0.7794520547945205,
+      "grad_norm": 0.1462251991033554,
+      "learning_rate": 1.3969849910923803e-05,
+      "loss": 0.5360151529312134,
+      "step": 569
+    },
+    {
+      "epoch": 0.7808219178082192,
+      "grad_norm": 0.14809444546699524,
+      "learning_rate": 1.3949476932573531e-05,
+      "loss": 0.5697877407073975,
+      "step": 570
+    },
+    {
+      "epoch": 0.7821917808219178,
+      "grad_norm": 0.1515805721282959,
+      "learning_rate": 1.3929084513489972e-05,
+      "loss": 0.5751916170120239,
+      "step": 571
+    },
+    {
+      "epoch": 0.7835616438356164,
+      "grad_norm": 0.14923812448978424,
+      "learning_rate": 1.3908672754051877e-05,
+      "loss": 0.6014054417610168,
+      "step": 572
+    },
+    {
+      "epoch": 0.7849315068493151,
+      "grad_norm": 0.15840941667556763,
+      "learning_rate": 1.388824175473321e-05,
+      "loss": 0.5876986980438232,
+      "step": 573
+    },
+    {
+      "epoch": 0.7863013698630137,
+      "grad_norm": 0.14389535784721375,
+      "learning_rate": 1.386779161610262e-05,
+      "loss": 0.5680856704711914,
+      "step": 574
+    },
+    {
+      "epoch": 0.7876712328767124,
+      "grad_norm": 0.1493615210056305,
+      "learning_rate": 1.3847322438822988e-05,
+      "loss": 0.5676318407058716,
+      "step": 575
+    },
+    {
+      "epoch": 0.7890410958904109,
+      "grad_norm": 0.1369982659816742,
+      "learning_rate": 1.3826834323650899e-05,
+      "loss": 0.5286301970481873,
+      "step": 576
+    },
+    {
+      "epoch": 0.7904109589041096,
+      "grad_norm": 0.15035495162010193,
+      "learning_rate": 1.3806327371436159e-05,
+      "loss": 0.5285563468933105,
+      "step": 577
+    },
+    {
+      "epoch": 0.7917808219178082,
+      "grad_norm": 0.14841550588607788,
+      "learning_rate": 1.3785801683121296e-05,
+      "loss": 0.520169734954834,
+      "step": 578
+    },
+    {
+      "epoch": 0.7931506849315069,
+      "grad_norm": 0.14321400225162506,
+      "learning_rate": 1.3765257359741065e-05,
+      "loss": 0.5244216918945312,
+      "step": 579
+    },
+    {
+      "epoch": 0.7945205479452054,
+      "grad_norm": 0.15067681670188904,
+      "learning_rate": 1.3744694502421949e-05,
+      "loss": 0.6115273237228394,
+      "step": 580
+    },
+    {
+      "epoch": 0.7958904109589041,
+      "grad_norm": 0.14685063064098358,
+      "learning_rate": 1.372411321238166e-05,
+      "loss": 0.5616559982299805,
+      "step": 581
+    },
+    {
+      "epoch": 0.7972602739726027,
+      "grad_norm": 0.14456450939178467,
+      "learning_rate": 1.3703513590928647e-05,
+      "loss": 0.5625962018966675,
+      "step": 582
+    },
+    {
+      "epoch": 0.7986301369863014,
+      "grad_norm": 0.14497993886470795,
+      "learning_rate": 1.3682895739461588e-05,
+      "loss": 0.5844519138336182,
+      "step": 583
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.1386929750442505,
+      "learning_rate": 1.3662259759468896e-05,
+      "loss": 0.5730550289154053,
+      "step": 584
+    },
+    {
+      "epoch": 0.8013698630136986,
+      "grad_norm": 0.13944153487682343,
+      "learning_rate": 1.3641605752528225e-05,
+      "loss": 0.5775356888771057,
+      "step": 585
+    },
+    {
+      "epoch": 0.8027397260273973,
+      "grad_norm": 0.14761903882026672,
+      "learning_rate": 1.3620933820305953e-05,
+      "loss": 0.5440607666969299,
+      "step": 586
+    },
+    {
+      "epoch": 0.8041095890410959,
+      "grad_norm": 0.14507032930850983,
+      "learning_rate": 1.3600244064556702e-05,
+      "loss": 0.5188732147216797,
+      "step": 587
+    },
+    {
+      "epoch": 0.8054794520547945,
+      "grad_norm": 0.14819754660129547,
+      "learning_rate": 1.3579536587122828e-05,
+      "loss": 0.5256346464157104,
+      "step": 588
+    },
+    {
+      "epoch": 0.8068493150684931,
+      "grad_norm": 0.14310747385025024,
+      "learning_rate": 1.3558811489933909e-05,
+      "loss": 0.5244719386100769,
+      "step": 589
+    },
+    {
+      "epoch": 0.8082191780821918,
+      "grad_norm": 0.1424933671951294,
+      "learning_rate": 1.3538068875006272e-05,
+      "loss": 0.520290732383728,
+      "step": 590
+    },
+    {
+      "epoch": 0.8095890410958904,
+      "grad_norm": 0.13971343636512756,
+      "learning_rate": 1.351730884444245e-05,
+      "loss": 0.5435692071914673,
+      "step": 591
+    },
+    {
+      "epoch": 0.810958904109589,
+      "grad_norm": 0.13619816303253174,
+      "learning_rate": 1.3496531500430723e-05,
+      "loss": 0.5121521949768066,
+      "step": 592
+    },
+    {
+      "epoch": 0.8123287671232877,
+      "grad_norm": 0.14922301471233368,
+      "learning_rate": 1.3475736945244575e-05,
+      "loss": 0.5281805396080017,
+      "step": 593
+    },
+    {
+      "epoch": 0.8136986301369863,
+      "grad_norm": 0.14446088671684265,
+      "learning_rate": 1.3454925281242225e-05,
+      "loss": 0.6063831448554993,
+      "step": 594
+    },
+    {
+      "epoch": 0.815068493150685,
+      "grad_norm": 0.14300888776779175,
+      "learning_rate": 1.3434096610866106e-05,
+      "loss": 0.5407198071479797,
+      "step": 595
+    },
+    {
+      "epoch": 0.8164383561643835,
+      "grad_norm": 0.13160032033920288,
+      "learning_rate": 1.3413251036642349e-05,
+      "loss": 0.5404701232910156,
+      "step": 596
+    },
+    {
+      "epoch": 0.8178082191780822,
+      "grad_norm": 0.13810713589191437,
+      "learning_rate": 1.3392388661180303e-05,
+      "loss": 0.5198618769645691,
+      "step": 597
+    },
+    {
+      "epoch": 0.8191780821917808,
+      "grad_norm": 0.14353394508361816,
+      "learning_rate": 1.3371509587172023e-05,
+      "loss": 0.513703465461731,
+      "step": 598
+    },
+    {
+      "epoch": 0.8205479452054795,
+      "grad_norm": 0.1333569884300232,
+      "learning_rate": 1.3350613917391749e-05,
+      "loss": 0.5600924491882324,
+      "step": 599
+    },
+    {
+      "epoch": 0.821917808219178,
+      "grad_norm": 0.15609651803970337,
+      "learning_rate": 1.3329701754695412e-05,
+      "loss": 0.6079143285751343,
+      "step": 600
+    },
+    {
+      "epoch": 0.8232876712328767,
+      "grad_norm": 0.13941015303134918,
+      "learning_rate": 1.330877320202014e-05,
+      "loss": 0.5294625759124756,
+      "step": 601
+    },
+    {
+      "epoch": 0.8246575342465754,
+      "grad_norm": 0.14244301617145538,
+      "learning_rate": 1.3287828362383724e-05,
+      "loss": 0.6567379236221313,
+      "step": 602
+    },
+    {
+      "epoch": 0.826027397260274,
+      "grad_norm": 0.1582559496164322,
+      "learning_rate": 1.3266867338884131e-05,
+      "loss": 0.5663045644760132,
+      "step": 603
+    },
+    {
+      "epoch": 0.8273972602739726,
+      "grad_norm": 0.16126839816570282,
+      "learning_rate": 1.3245890234698993e-05,
+      "loss": 0.5470207333564758,
+      "step": 604
+    },
+    {
+      "epoch": 0.8287671232876712,
+      "grad_norm": 0.14516150951385498,
+      "learning_rate": 1.322489715308509e-05,
+      "loss": 0.5473780035972595,
+      "step": 605
+    },
+    {
+      "epoch": 0.8301369863013699,
+      "grad_norm": 0.13455289602279663,
+      "learning_rate": 1.3203888197377857e-05,
+      "loss": 0.5864403247833252,
+      "step": 606
+    },
+    {
+      "epoch": 0.8315068493150685,
+      "grad_norm": 0.1354503482580185,
+      "learning_rate": 1.318286347099086e-05,
+      "loss": 0.5317245721817017,
+      "step": 607
+    },
+    {
+      "epoch": 0.8328767123287671,
+      "grad_norm": 0.14837303757667542,
+      "learning_rate": 1.3161823077415299e-05,
+      "loss": 0.5762374401092529,
+      "step": 608
+    },
+    {
+      "epoch": 0.8342465753424657,
+      "grad_norm": 0.14379873871803284,
+      "learning_rate": 1.314076712021949e-05,
+      "loss": 0.5885000228881836,
+      "step": 609
+    },
+    {
+      "epoch": 0.8356164383561644,
+      "grad_norm": 0.1361761838197708,
+      "learning_rate": 1.3119695703048358e-05,
+      "loss": 0.43967384099960327,
+      "step": 610
+    },
+    {
+      "epoch": 0.836986301369863,
+      "grad_norm": 0.13545174896717072,
+      "learning_rate": 1.3098608929622934e-05,
+      "loss": 0.4924185276031494,
+      "step": 611
+    },
+    {
+      "epoch": 0.8383561643835616,
+      "grad_norm": 0.13373355567455292,
+      "learning_rate": 1.3077506903739829e-05,
+      "loss": 0.536264181137085,
+      "step": 612
+    },
+    {
+      "epoch": 0.8397260273972603,
+      "grad_norm": 0.15098173916339874,
+      "learning_rate": 1.305638972927074e-05,
+      "loss": 0.6109551787376404,
+      "step": 613
+    },
+    {
+      "epoch": 0.8410958904109589,
+      "grad_norm": 0.13542239367961884,
+      "learning_rate": 1.3035257510161925e-05,
+      "loss": 0.4766671657562256,
+      "step": 614
+    },
+    {
+      "epoch": 0.8424657534246576,
+      "grad_norm": 0.1498875766992569,
+      "learning_rate": 1.30141103504337e-05,
+      "loss": 0.598709225654602,
+      "step": 615
+    },
+    {
+      "epoch": 0.8438356164383561,
+      "grad_norm": 0.14445774257183075,
+      "learning_rate": 1.2992948354179919e-05,
+      "loss": 0.5338215827941895,
+      "step": 616
+    },
+    {
+      "epoch": 0.8452054794520548,
+      "grad_norm": 0.1474190205335617,
+      "learning_rate": 1.297177162556748e-05,
+      "loss": 0.518726110458374,
+      "step": 617
+    },
+    {
+      "epoch": 0.8465753424657534,
+      "grad_norm": 0.14172446727752686,
+      "learning_rate": 1.2950580268835784e-05,
+      "loss": 0.5487723350524902,
+      "step": 618
+    },
+    {
+      "epoch": 0.8479452054794521,
+      "grad_norm": 0.14646317064762115,
+      "learning_rate": 1.2929374388296241e-05,
+      "loss": 0.545096218585968,
+      "step": 619
+    },
+    {
+      "epoch": 0.8493150684931506,
+      "grad_norm": 0.14491736888885498,
+      "learning_rate": 1.2908154088331757e-05,
+      "loss": 0.5572469830513,
+      "step": 620
+    },
+    {
+      "epoch": 0.8506849315068493,
+      "grad_norm": 0.14330042898654938,
+      "learning_rate": 1.2886919473396212e-05,
+      "loss": 0.5331091284751892,
+      "step": 621
+    },
+    {
+      "epoch": 0.852054794520548,
+      "grad_norm": 0.13608761131763458,
+      "learning_rate": 1.2865670648013951e-05,
+      "loss": 0.5101725459098816,
+      "step": 622
+    },
+    {
+      "epoch": 0.8534246575342466,
+      "grad_norm": 0.15225724875926971,
+      "learning_rate": 1.2844407716779266e-05,
+      "loss": 0.5518755912780762,
+      "step": 623
+    },
+    {
+      "epoch": 0.8547945205479452,
+      "grad_norm": 0.14294537901878357,
+      "learning_rate": 1.2823130784355882e-05,
+      "loss": 0.48980948328971863,
+      "step": 624
+    },
+    {
+      "epoch": 0.8561643835616438,
+      "grad_norm": 0.13224340975284576,
+      "learning_rate": 1.2801839955476444e-05,
+      "loss": 0.4973132610321045,
+      "step": 625
+    },
+    {
+      "epoch": 0.8575342465753425,
+      "grad_norm": 0.1523551344871521,
+      "learning_rate": 1.2780535334942011e-05,
+      "loss": 0.5488629341125488,
+      "step": 626
+    },
+    {
+      "epoch": 0.8589041095890411,
+      "grad_norm": 0.12789522111415863,
+      "learning_rate": 1.2759217027621507e-05,
+      "loss": 0.5876277685165405,
+      "step": 627
+    },
+    {
+      "epoch": 0.8602739726027397,
+      "grad_norm": 0.12671145796775818,
+      "learning_rate": 1.2737885138451246e-05,
+      "loss": 0.5252135992050171,
+      "step": 628
+    },
+    {
+      "epoch": 0.8616438356164383,
+      "grad_norm": 0.13611386716365814,
+      "learning_rate": 1.2716539772434389e-05,
+      "loss": 0.5474650859832764,
+      "step": 629
+    },
+    {
+      "epoch": 0.863013698630137,
+      "grad_norm": 0.13836732506752014,
+      "learning_rate": 1.2695181034640435e-05,
+      "loss": 0.5324945449829102,
+      "step": 630
+    },
+    {
+      "epoch": 0.8643835616438356,
+      "grad_norm": 0.13409800827503204,
+      "learning_rate": 1.267380903020471e-05,
+      "loss": 0.5111215710639954,
+      "step": 631
+    },
+    {
+      "epoch": 0.8657534246575342,
+      "grad_norm": 0.14573852717876434,
+      "learning_rate": 1.2652423864327833e-05,
+      "loss": 0.551156759262085,
+      "step": 632
+    },
+    {
+      "epoch": 0.8671232876712329,
+      "grad_norm": 0.1373865008354187,
+      "learning_rate": 1.2631025642275212e-05,
+      "loss": 0.5151498317718506,
+      "step": 633
+    },
+    {
+      "epoch": 0.8684931506849315,
+      "grad_norm": 0.15578457713127136,
+      "learning_rate": 1.2609614469376532e-05,
+      "loss": 0.5745099782943726,
+      "step": 634
+    },
+    {
+      "epoch": 0.8698630136986302,
+      "grad_norm": 0.14487750828266144,
+      "learning_rate": 1.2588190451025209e-05,
+      "loss": 0.591609001159668,
+      "step": 635
+    },
+    {
+      "epoch": 0.8712328767123287,
+      "grad_norm": 0.1382559835910797,
+      "learning_rate": 1.2566753692677902e-05,
+      "loss": 0.5301614999771118,
+      "step": 636
+    },
+    {
+      "epoch": 0.8726027397260274,
+      "grad_norm": 0.1299545019865036,
+      "learning_rate": 1.2545304299853977e-05,
+      "loss": 0.4942835569381714,
+      "step": 637
+    },
+    {
+      "epoch": 0.873972602739726,
+      "grad_norm": 0.13769906759262085,
+      "learning_rate": 1.2523842378134991e-05,
+      "loss": 0.5325908660888672,
+      "step": 638
+    },
+    {
+      "epoch": 0.8753424657534247,
+      "grad_norm": 0.13713912665843964,
+      "learning_rate": 1.2502368033164176e-05,
+      "loss": 0.5291322469711304,
+      "step": 639
+    },
+    {
+      "epoch": 0.8767123287671232,
+      "grad_norm": 0.14469754695892334,
+      "learning_rate": 1.2480881370645906e-05,
+      "loss": 0.5346277952194214,
+      "step": 640
+    },
+    {
+      "epoch": 0.8780821917808219,
+      "grad_norm": 0.1554187834262848,
+      "learning_rate": 1.2459382496345199e-05,
+      "loss": 0.5714985728263855,
+      "step": 641
+    },
+    {
+      "epoch": 0.8794520547945206,
+      "grad_norm": 0.14312691986560822,
+      "learning_rate": 1.2437871516087174e-05,
+      "loss": 0.543634831905365,
+      "step": 642
+    },
+    {
+      "epoch": 0.8808219178082192,
+      "grad_norm": 0.14826460182666779,
+      "learning_rate": 1.2416348535756543e-05,
+      "loss": 0.5458000898361206,
+      "step": 643
+    },
+    {
+      "epoch": 0.8821917808219178,
+      "grad_norm": 0.144753098487854,
+      "learning_rate": 1.2394813661297093e-05,
+      "loss": 0.5427224636077881,
+      "step": 644
+    },
+    {
+      "epoch": 0.8835616438356164,
+      "grad_norm": 0.17040959000587463,
+      "learning_rate": 1.2373266998711152e-05,
+      "loss": 0.6165363192558289,
+      "step": 645
+    },
+    {
+      "epoch": 0.8849315068493151,
+      "grad_norm": 0.1435997039079666,
+      "learning_rate": 1.2351708654059065e-05,
+      "loss": 0.49909958243370056,
+      "step": 646
+    },
+    {
+      "epoch": 0.8863013698630137,
+      "grad_norm": 0.139425128698349,
+      "learning_rate": 1.2330138733458693e-05,
+      "loss": 0.5313612222671509,
+      "step": 647
+    },
+    {
+      "epoch": 0.8876712328767123,
+      "grad_norm": 0.13428698480129242,
+      "learning_rate": 1.2308557343084881e-05,
+      "loss": 0.5440371632575989,
+      "step": 648
+    },
+    {
+      "epoch": 0.8890410958904109,
+      "grad_norm": 0.14464199542999268,
+      "learning_rate": 1.2286964589168917e-05,
+      "loss": 0.6058000326156616,
+      "step": 649
+    },
+    {
+      "epoch": 0.8904109589041096,
+      "grad_norm": 0.13464415073394775,
+      "learning_rate": 1.2265360577998038e-05,
+      "loss": 0.5244266986846924,
+      "step": 650
+    },
+    {
+      "epoch": 0.8917808219178082,
+      "grad_norm": 0.17547836899757385,
+      "learning_rate": 1.2243745415914882e-05,
+      "loss": 0.5788463354110718,
+      "step": 651
+    },
+    {
+      "epoch": 0.8931506849315068,
+      "grad_norm": 0.1391308307647705,
+      "learning_rate": 1.2222119209316984e-05,
+      "loss": 0.5580483078956604,
+      "step": 652
+    },
+    {
+      "epoch": 0.8945205479452055,
+      "grad_norm": 0.15701870620250702,
+      "learning_rate": 1.220048206465625e-05,
+      "loss": 0.6042839288711548,
+      "step": 653
+    },
+    {
+      "epoch": 0.8958904109589041,
+      "grad_norm": 0.12679646909236908,
+      "learning_rate": 1.2178834088438404e-05,
+      "loss": 0.5429613590240479,
+      "step": 654
+    },
+    {
+      "epoch": 0.8972602739726028,
+      "grad_norm": 0.13499225676059723,
+      "learning_rate": 1.2157175387222516e-05,
+      "loss": 0.576964795589447,
+      "step": 655
+    },
+    {
+      "epoch": 0.8986301369863013,
+      "grad_norm": 0.12962238490581512,
+      "learning_rate": 1.2135506067620421e-05,
+      "loss": 0.4955563247203827,
+      "step": 656
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.14396265149116516,
+      "learning_rate": 1.2113826236296245e-05,
+      "loss": 0.5154756903648376,
+      "step": 657
+    },
+    {
+      "epoch": 0.9013698630136986,
+      "grad_norm": 0.1490243673324585,
+      "learning_rate": 1.209213599996584e-05,
+      "loss": 0.6035890579223633,
+      "step": 658
+    },
+    {
+      "epoch": 0.9027397260273973,
+      "grad_norm": 0.15239585936069489,
+      "learning_rate": 1.2070435465396283e-05,
+      "loss": 0.587848424911499,
+      "step": 659
+    },
+    {
+      "epoch": 0.9041095890410958,
+      "grad_norm": 0.1464146375656128,
+      "learning_rate": 1.2048724739405337e-05,
+      "loss": 0.6095236539840698,
+      "step": 660
+    },
+    {
+      "epoch": 0.9054794520547945,
+      "grad_norm": 0.12767134606838226,
+      "learning_rate": 1.2027003928860936e-05,
+      "loss": 0.4720427691936493,
+      "step": 661
+    },
+    {
+      "epoch": 0.9068493150684932,
+      "grad_norm": 0.14553502202033997,
+      "learning_rate": 1.2005273140680656e-05,
+      "loss": 0.5657269358634949,
+      "step": 662
+    },
+    {
+      "epoch": 0.9082191780821918,
+      "grad_norm": 0.13808585703372955,
+      "learning_rate": 1.1983532481831179e-05,
+      "loss": 0.4915253818035126,
+      "step": 663
+    },
+    {
+      "epoch": 0.9095890410958904,
+      "grad_norm": 0.14350081980228424,
+      "learning_rate": 1.196178205932778e-05,
+      "loss": 0.5666084289550781,
+      "step": 664
+    },
+    {
+      "epoch": 0.910958904109589,
+      "grad_norm": 0.142891526222229,
+      "learning_rate": 1.1940021980233784e-05,
+      "loss": 0.5144902467727661,
+      "step": 665
+    },
+    {
+      "epoch": 0.9123287671232877,
+      "grad_norm": 0.13920342922210693,
+      "learning_rate": 1.1918252351660066e-05,
+      "loss": 0.5250864028930664,
+      "step": 666
+    },
+    {
+      "epoch": 0.9136986301369863,
+      "grad_norm": 0.13702407479286194,
+      "learning_rate": 1.1896473280764498e-05,
+      "loss": 0.49932771921157837,
+      "step": 667
+    },
+    {
+      "epoch": 0.915068493150685,
+      "grad_norm": 0.15072178840637207,
+      "learning_rate": 1.187468487475142e-05,
+      "loss": 0.5834764242172241,
+      "step": 668
+    },
+    {
+      "epoch": 0.9164383561643835,
+      "grad_norm": 0.1453569233417511,
+      "learning_rate": 1.1852887240871145e-05,
+      "loss": 0.5249761343002319,
+      "step": 669
+    },
+    {
+      "epoch": 0.9178082191780822,
+      "grad_norm": 0.1490030437707901,
+      "learning_rate": 1.1831080486419388e-05,
+      "loss": 0.5667267441749573,
+      "step": 670
+    },
+    {
+      "epoch": 0.9191780821917809,
+      "grad_norm": 0.1355256885290146,
+      "learning_rate": 1.1809264718736775e-05,
+      "loss": 0.5361207723617554,
+      "step": 671
+    },
+    {
+      "epoch": 0.9205479452054794,
+      "grad_norm": 0.1491296887397766,
+      "learning_rate": 1.1787440045208287e-05,
+      "loss": 0.5815398693084717,
+      "step": 672
+    },
+    {
+      "epoch": 0.9219178082191781,
+      "grad_norm": 0.1443261355161667,
+      "learning_rate": 1.1765606573262745e-05,
+      "loss": 0.5183050036430359,
+      "step": 673
+    },
+    {
+      "epoch": 0.9232876712328767,
+      "grad_norm": 0.13231635093688965,
+      "learning_rate": 1.1743764410372283e-05,
+      "loss": 0.570467472076416,
+      "step": 674
+    },
+    {
+      "epoch": 0.9246575342465754,
+      "grad_norm": 0.1384153813123703,
+      "learning_rate": 1.1721913664051814e-05,
+      "loss": 0.544106125831604,
+      "step": 675
+    },
+    {
+      "epoch": 0.9260273972602739,
+      "grad_norm": 0.15208759903907776,
+      "learning_rate": 1.1700054441858497e-05,
+      "loss": 0.5341042876243591,
+      "step": 676
+    },
+    {
+      "epoch": 0.9273972602739726,
+      "grad_norm": 0.14940430223941803,
+      "learning_rate": 1.1678186851391218e-05,
+      "loss": 0.5699517726898193,
+      "step": 677
+    },
+    {
+      "epoch": 0.9287671232876712,
+      "grad_norm": 0.14947940409183502,
+      "learning_rate": 1.165631100029005e-05,
+      "loss": 0.5162645578384399,
+      "step": 678
+    },
+    {
+      "epoch": 0.9301369863013699,
+      "grad_norm": 0.12951557338237762,
+      "learning_rate": 1.1634426996235732e-05,
+      "loss": 0.5086703300476074,
+      "step": 679
+    },
+    {
+      "epoch": 0.9315068493150684,
+      "grad_norm": 0.14504235982894897,
+      "learning_rate": 1.1612534946949134e-05,
+      "loss": 0.475946307182312,
+      "step": 680
+    },
+    {
+      "epoch": 0.9328767123287671,
+      "grad_norm": 0.13910368084907532,
+      "learning_rate": 1.1590634960190722e-05,
+      "loss": 0.5487220883369446,
+      "step": 681
+    },
+    {
+      "epoch": 0.9342465753424658,
+      "grad_norm": 0.15137788653373718,
+      "learning_rate": 1.1568727143760042e-05,
+      "loss": 0.5802314281463623,
+      "step": 682
+    },
+    {
+      "epoch": 0.9356164383561644,
+      "grad_norm": 0.13175715506076813,
+      "learning_rate": 1.154681160549517e-05,
+      "loss": 0.5751426219940186,
+      "step": 683
+    },
+    {
+      "epoch": 0.936986301369863,
+      "grad_norm": 0.1453649252653122,
+      "learning_rate": 1.15248884532722e-05,
+      "loss": 0.563701868057251,
+      "step": 684
+    },
+    {
+      "epoch": 0.9383561643835616,
+      "grad_norm": 0.14549468457698822,
+      "learning_rate": 1.1502957795004706e-05,
+      "loss": 0.5319613218307495,
+      "step": 685
+    },
+    {
+      "epoch": 0.9397260273972603,
+      "grad_norm": 0.14660926163196564,
+      "learning_rate": 1.1481019738643202e-05,
+      "loss": 0.5583406686782837,
+      "step": 686
+    },
+    {
+      "epoch": 0.9410958904109589,
+      "grad_norm": 0.15349538624286652,
+      "learning_rate": 1.1459074392174619e-05,
+      "loss": 0.5592145919799805,
+      "step": 687
+    },
+    {
+      "epoch": 0.9424657534246575,
+      "grad_norm": 0.15075471997261047,
+      "learning_rate": 1.1437121863621776e-05,
+      "loss": 0.5834563374519348,
+      "step": 688
+    },
+    {
+      "epoch": 0.9438356164383561,
+      "grad_norm": 0.13851697742938995,
+      "learning_rate": 1.1415162261042849e-05,
+      "loss": 0.5394259691238403,
+      "step": 689
+    },
+    {
+      "epoch": 0.9452054794520548,
+      "grad_norm": 0.13821455836296082,
+      "learning_rate": 1.1393195692530823e-05,
+      "loss": 0.5597106218338013,
+      "step": 690
+    },
+    {
+      "epoch": 0.9465753424657535,
+      "grad_norm": 0.13941477239131927,
+      "learning_rate": 1.1371222266212983e-05,
+      "loss": 0.5378660559654236,
+      "step": 691
+    },
+    {
+      "epoch": 0.947945205479452,
+      "grad_norm": 0.15957188606262207,
+      "learning_rate": 1.134924209025036e-05,
+      "loss": 0.559799075126648,
+      "step": 692
+    },
+    {
+      "epoch": 0.9493150684931507,
+      "grad_norm": 0.14969097077846527,
+      "learning_rate": 1.1327255272837221e-05,
+      "loss": 0.5599546432495117,
+      "step": 693
+    },
+    {
+      "epoch": 0.9506849315068493,
+      "grad_norm": 0.1421581506729126,
+      "learning_rate": 1.130526192220052e-05,
+      "loss": 0.5460138320922852,
+      "step": 694
+    },
+    {
+      "epoch": 0.952054794520548,
+      "grad_norm": 0.1384434700012207,
+      "learning_rate": 1.128326214659936e-05,
+      "loss": 0.5014693140983582,
+      "step": 695
+    },
+    {
+      "epoch": 0.9534246575342465,
+      "grad_norm": 0.13422176241874695,
+      "learning_rate": 1.1261256054324488e-05,
+      "loss": 0.4789910912513733,
+      "step": 696
+    },
+    {
+      "epoch": 0.9547945205479452,
+      "grad_norm": 0.15614454448223114,
+      "learning_rate": 1.1239243753697728e-05,
+      "loss": 0.5928708910942078,
+      "step": 697
+    },
+    {
+      "epoch": 0.9561643835616438,
+      "grad_norm": 0.13170309364795685,
+      "learning_rate": 1.1217225353071473e-05,
+      "loss": 0.5444527268409729,
+      "step": 698
+    },
+    {
+      "epoch": 0.9575342465753425,
+      "grad_norm": 0.143415629863739,
+      "learning_rate": 1.1195200960828138e-05,
+      "loss": 0.5303205847740173,
+      "step": 699
+    },
+    {
+      "epoch": 0.958904109589041,
+      "grad_norm": 0.14177711308002472,
+      "learning_rate": 1.1173170685379635e-05,
+      "loss": 0.5031203627586365,
+      "step": 700
+    },
+    {
+      "epoch": 0.9602739726027397,
+      "grad_norm": 0.15743711590766907,
+      "learning_rate": 1.115113463516683e-05,
+      "loss": 0.5337619781494141,
+      "step": 701
+    },
+    {
+      "epoch": 0.9616438356164384,
+      "grad_norm": 0.14667122066020966,
+      "learning_rate": 1.1129092918659019e-05,
+      "loss": 0.48960891366004944,
+      "step": 702
+    },
+    {
+      "epoch": 0.963013698630137,
+      "grad_norm": 0.1391826868057251,
+      "learning_rate": 1.1107045644353388e-05,
+      "loss": 0.5426533222198486,
+      "step": 703
+    },
+    {
+      "epoch": 0.9643835616438357,
+      "grad_norm": 0.15035362541675568,
+      "learning_rate": 1.1084992920774478e-05,
+      "loss": 0.4937519431114197,
+      "step": 704
+    },
+    {
+      "epoch": 0.9657534246575342,
+      "grad_norm": 0.14533023536205292,
+      "learning_rate": 1.1062934856473655e-05,
+      "loss": 0.5189172029495239,
+      "step": 705
+    },
+    {
+      "epoch": 0.9671232876712329,
+      "grad_norm": 0.12609876692295074,
+      "learning_rate": 1.1040871560028576e-05,
+      "loss": 0.5009540915489197,
+      "step": 706
+    },
+    {
+      "epoch": 0.9684931506849315,
+      "grad_norm": 0.1392073929309845,
+      "learning_rate": 1.1018803140042651e-05,
+      "loss": 0.5657528638839722,
+      "step": 707
+    },
+    {
+      "epoch": 0.9698630136986301,
+      "grad_norm": 0.1347883641719818,
+      "learning_rate": 1.099672970514451e-05,
+      "loss": 0.5511807799339294,
+      "step": 708
+    },
+    {
+      "epoch": 0.9712328767123287,
+      "grad_norm": 0.15112189948558807,
+      "learning_rate": 1.0974651363987465e-05,
+      "loss": 0.5117864608764648,
+      "step": 709
+    },
+    {
+      "epoch": 0.9726027397260274,
+      "grad_norm": 0.1492203027009964,
+      "learning_rate": 1.0952568225248983e-05,
+      "loss": 0.5602498054504395,
+      "step": 710
+    },
+    {
+      "epoch": 0.9739726027397261,
+      "grad_norm": 0.13084228336811066,
+      "learning_rate": 1.0930480397630146e-05,
+      "loss": 0.503504753112793,
+      "step": 711
+    },
+    {
+      "epoch": 0.9753424657534246,
+      "grad_norm": 0.12569855153560638,
+      "learning_rate": 1.0908387989855114e-05,
+      "loss": 0.49362748861312866,
+      "step": 712
+    },
+    {
+      "epoch": 0.9767123287671233,
+      "grad_norm": 0.14552345871925354,
+      "learning_rate": 1.0886291110670592e-05,
+      "loss": 0.5039536952972412,
+      "step": 713
+    },
+    {
+      "epoch": 0.9780821917808219,
+      "grad_norm": 0.13342419266700745,
+      "learning_rate": 1.0864189868845296e-05,
+      "loss": 0.5050915479660034,
+      "step": 714
+    },
+    {
+      "epoch": 0.9794520547945206,
+      "grad_norm": 0.15216471254825592,
+      "learning_rate": 1.0842084373169417e-05,
+      "loss": 0.5598911046981812,
+      "step": 715
+    },
+    {
+      "epoch": 0.9808219178082191,
+      "grad_norm": 0.12561655044555664,
+      "learning_rate": 1.0819974732454086e-05,
+      "loss": 0.47096988558769226,
+      "step": 716
+    },
+    {
+      "epoch": 0.9821917808219178,
+      "grad_norm": 0.1414027214050293,
+      "learning_rate": 1.0797861055530832e-05,
+      "loss": 0.5470199584960938,
+      "step": 717
+    },
+    {
+      "epoch": 0.9835616438356164,
+      "grad_norm": 0.14046591520309448,
+      "learning_rate": 1.077574345125106e-05,
+      "loss": 0.5058327913284302,
+      "step": 718
+    },
+    {
+      "epoch": 0.9849315068493151,
+      "grad_norm": 0.13157160580158234,
+      "learning_rate": 1.0753622028485498e-05,
+      "loss": 0.4962088465690613,
+      "step": 719
+    },
+    {
+      "epoch": 0.9863013698630136,
+      "grad_norm": 0.1477247178554535,
+      "learning_rate": 1.0731496896123676e-05,
+      "loss": 0.611939549446106,
+      "step": 720
+    },
+    {
+      "epoch": 0.9876712328767123,
+      "grad_norm": 0.1309177726507187,
+      "learning_rate": 1.0709368163073385e-05,
+      "loss": 0.5082958936691284,
+      "step": 721
+    },
+    {
+      "epoch": 0.989041095890411,
+      "grad_norm": 0.13591881096363068,
+      "learning_rate": 1.0687235938260134e-05,
+      "loss": 0.5050320625305176,
+      "step": 722
+    },
+    {
+      "epoch": 0.9904109589041096,
+      "grad_norm": 0.1385071724653244,
+      "learning_rate": 1.0665100330626625e-05,
+      "loss": 0.5245853066444397,
+      "step": 723
+    },
+    {
+      "epoch": 0.9917808219178083,
+      "grad_norm": 0.14319604635238647,
+      "learning_rate": 1.0642961449132203e-05,
+      "loss": 0.47970306873321533,
+      "step": 724
+    },
+    {
+      "epoch": 0.9931506849315068,
+      "grad_norm": 0.131264328956604,
+      "learning_rate": 1.062081940275234e-05,
+      "loss": 0.4639028310775757,
+      "step": 725
+    },
+    {
+      "epoch": 0.9945205479452055,
+      "grad_norm": 0.14266684651374817,
+      "learning_rate": 1.0598674300478085e-05,
+      "loss": 0.521947979927063,
+      "step": 726
+    },
+    {
+      "epoch": 0.9958904109589041,
+      "grad_norm": 0.13719575107097626,
+      "learning_rate": 1.0576526251315515e-05,
+      "loss": 0.5097373723983765,
+      "step": 727
+    },
+    {
+      "epoch": 0.9972602739726028,
+      "grad_norm": 0.13589054346084595,
+      "learning_rate": 1.055437536428523e-05,
+      "loss": 0.5200362205505371,
+      "step": 728
+    },
+    {
+      "epoch": 0.9986301369863013,
+      "grad_norm": 0.14577971398830414,
+      "learning_rate": 1.0532221748421786e-05,
+      "loss": 0.5406281352043152,
+      "step": 729
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.1317002773284912,
+      "learning_rate": 1.0510065512773184e-05,
+      "loss": 0.4858514964580536,
+      "step": 730
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 1460,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 7.658595503088599e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}