diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,8542 @@
+{
+  "best_global_step": 500,
+  "best_metric": 0.78649825,
+  "best_model_checkpoint": "/root/ms-swift/examples/train/liger/output/Qwen4SFTDeepResearch/v2-20250930-212021/checkpoint-400",
+  "epoch": 4.316546762589928,
+  "eval_steps": 100,
+  "global_step": 1200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0035971223021582736,
+      "grad_norm": 15.756150320288057,
+      "learning_rate": 9.999995011498398e-06,
+      "loss": 1.5632460117340088,
+      "step": 1
+    },
+    {
+      "epoch": 0.007194244604316547,
+      "grad_norm": 7.517349932297116,
+      "learning_rate": 9.999980046003546e-06,
+      "loss": 1.343105435371399,
+      "step": 2
+    },
+    {
+      "epoch": 0.01079136690647482,
+      "grad_norm": 3.1958830788674173,
+      "learning_rate": 9.999955103545307e-06,
+      "loss": 1.1570754051208496,
+      "step": 3
+    },
+    {
+      "epoch": 0.014388489208633094,
+      "grad_norm": 2.315374404024555,
+      "learning_rate": 9.999920184173449e-06,
+      "loss": 1.1298246383666992,
+      "step": 4
+    },
+    {
+      "epoch": 0.017985611510791366,
+      "grad_norm": 2.941589170867705,
+      "learning_rate": 9.999875287957652e-06,
+      "loss": 1.1169365644454956,
+      "step": 5
+    },
+    {
+      "epoch": 0.02158273381294964,
+      "grad_norm": 2.515381306131075,
+      "learning_rate": 9.999820414987502e-06,
+      "loss": 1.0570898056030273,
+      "step": 6
+    },
+    {
+      "epoch": 0.025179856115107913,
+      "grad_norm": 2.2826473676552625,
+      "learning_rate": 9.999755565372492e-06,
+      "loss": 1.0654137134552002,
+      "step": 7
+    },
+    {
+      "epoch": 0.02877697841726619,
+      "grad_norm": 1.5778221040103755,
+      "learning_rate": 9.999680739242022e-06,
+      "loss": 1.010119915008545,
+      "step": 8
+    },
+    {
+      "epoch": 0.03237410071942446,
+      "grad_norm": 1.2932029633476676,
+      "learning_rate": 9.999595936745401e-06,
+      "loss": 1.0208889245986938,
+      "step": 9
+    },
+    {
+      "epoch": 0.03597122302158273,
+      "grad_norm": 1.0284808012699043,
+      "learning_rate": 9.999501158051846e-06,
+      "loss": 0.9741397500038147,
+      "step": 10
+    },
+    {
+      "epoch": 0.039568345323741004,
+      "grad_norm": 0.9153376375520705,
+      "learning_rate": 9.999396403350476e-06,
+      "loss": 0.9674245119094849,
+      "step": 11
+    },
+    {
+      "epoch": 0.04316546762589928,
+      "grad_norm": 0.8405459688436054,
+      "learning_rate": 9.999281672850317e-06,
+      "loss": 0.956840991973877,
+      "step": 12
+    },
+    {
+      "epoch": 0.046762589928057555,
+      "grad_norm": 0.7954149229636902,
+      "learning_rate": 9.999156966780306e-06,
+      "loss": 0.9293943643569946,
+      "step": 13
+    },
+    {
+      "epoch": 0.050359712230215826,
+      "grad_norm": 0.7264144566060944,
+      "learning_rate": 9.99902228538928e-06,
+      "loss": 0.9339001774787903,
+      "step": 14
+    },
+    {
+      "epoch": 0.0539568345323741,
+      "grad_norm": 0.6989667251925004,
+      "learning_rate": 9.998877628945982e-06,
+      "loss": 0.9139963984489441,
+      "step": 15
+    },
+    {
+      "epoch": 0.05755395683453238,
+      "grad_norm": 0.7146203444147694,
+      "learning_rate": 9.99872299773906e-06,
+      "loss": 0.9200505018234253,
+      "step": 16
+    },
+    {
+      "epoch": 0.06115107913669065,
+      "grad_norm": 0.6965354357010114,
+      "learning_rate": 9.998558392077065e-06,
+      "loss": 0.9260781407356262,
+      "step": 17
+    },
+    {
+      "epoch": 0.06474820143884892,
+      "grad_norm": 0.7526497086024109,
+      "learning_rate": 9.998383812288451e-06,
+      "loss": 0.9045039415359497,
+      "step": 18
+    },
+    {
+      "epoch": 0.0683453237410072,
+      "grad_norm": 0.6111014847522879,
+      "learning_rate": 9.998199258721576e-06,
+      "loss": 0.8944211602210999,
+      "step": 19
+    },
+    {
+      "epoch": 0.07194244604316546,
+      "grad_norm": 0.5854801434764001,
+      "learning_rate": 9.998004731744696e-06,
+      "loss": 0.901360034942627,
+      "step": 20
+    },
+    {
+      "epoch": 0.07553956834532374,
+      "grad_norm": 0.5912075388434902,
+      "learning_rate": 9.997800231745973e-06,
+      "loss": 0.8743350505828857,
+      "step": 21
+    },
+    {
+      "epoch": 0.07913669064748201,
+      "grad_norm": 0.5644564808326691,
+      "learning_rate": 9.997585759133463e-06,
+      "loss": 0.8950492143630981,
+      "step": 22
+    },
+    {
+      "epoch": 0.08273381294964029,
+      "grad_norm": 0.6193963516298768,
+      "learning_rate": 9.99736131433513e-06,
+      "loss": 0.8587119579315186,
+      "step": 23
+    },
+    {
+      "epoch": 0.08633093525179857,
+      "grad_norm": 0.6319747604029503,
+      "learning_rate": 9.997126897798826e-06,
+      "loss": 0.8936506509780884,
+      "step": 24
+    },
+    {
+      "epoch": 0.08992805755395683,
+      "grad_norm": 0.5786372791735689,
+      "learning_rate": 9.996882509992307e-06,
+      "loss": 0.8633454442024231,
+      "step": 25
+    },
+    {
+      "epoch": 0.09352517985611511,
+      "grad_norm": 0.5786725523056919,
+      "learning_rate": 9.996628151403226e-06,
+      "loss": 0.8897263407707214,
+      "step": 26
+    },
+    {
+      "epoch": 0.09712230215827339,
+      "grad_norm": 0.5664869834871367,
+      "learning_rate": 9.99636382253913e-06,
+      "loss": 0.8681536316871643,
+      "step": 27
+    },
+    {
+      "epoch": 0.10071942446043165,
+      "grad_norm": 0.5515477526139622,
+      "learning_rate": 9.996089523927461e-06,
+      "loss": 0.8630425930023193,
+      "step": 28
+    },
+    {
+      "epoch": 0.10431654676258993,
+      "grad_norm": 0.5315269272765065,
+      "learning_rate": 9.995805256115555e-06,
+      "loss": 0.8772830367088318,
+      "step": 29
+    },
+    {
+      "epoch": 0.1079136690647482,
+      "grad_norm": 0.5437185163066915,
+      "learning_rate": 9.995511019670639e-06,
+      "loss": 0.8837330341339111,
+      "step": 30
+    },
+    {
+      "epoch": 0.11151079136690648,
+      "grad_norm": 0.5549255962064505,
+      "learning_rate": 9.995206815179833e-06,
+      "loss": 0.870895266532898,
+      "step": 31
+    },
+    {
+      "epoch": 0.11510791366906475,
+      "grad_norm": 0.5591701123937822,
+      "learning_rate": 9.994892643250147e-06,
+      "loss": 0.8615888953208923,
+      "step": 32
+    },
+    {
+      "epoch": 0.11870503597122302,
+      "grad_norm": 0.5452402301752459,
+      "learning_rate": 9.99456850450848e-06,
+      "loss": 0.8485289812088013,
+      "step": 33
+    },
+    {
+      "epoch": 0.1223021582733813,
+      "grad_norm": 0.5271304889146575,
+      "learning_rate": 9.99423439960162e-06,
+      "loss": 0.8544217944145203,
+      "step": 34
+    },
+    {
+      "epoch": 0.12589928057553956,
+      "grad_norm": 0.5465089936521315,
+      "learning_rate": 9.993890329196236e-06,
+      "loss": 0.8891148567199707,
+      "step": 35
+    },
+    {
+      "epoch": 0.12949640287769784,
+      "grad_norm": 0.545309807427878,
+      "learning_rate": 9.993536293978892e-06,
+      "loss": 0.8729881644248962,
+      "step": 36
+    },
+    {
+      "epoch": 0.13309352517985612,
+      "grad_norm": 0.5522501928569817,
+      "learning_rate": 9.993172294656024e-06,
+      "loss": 0.8602065443992615,
+      "step": 37
+    },
+    {
+      "epoch": 0.1366906474820144,
+      "grad_norm": 0.5384827461098874,
+      "learning_rate": 9.992798331953962e-06,
+      "loss": 0.8517411351203918,
+      "step": 38
+    },
+    {
+      "epoch": 0.14028776978417265,
+      "grad_norm": 0.5297608942892242,
+      "learning_rate": 9.99241440661891e-06,
+      "loss": 0.8619793653488159,
+      "step": 39
+    },
+    {
+      "epoch": 0.14388489208633093,
+      "grad_norm": 0.5347497959937362,
+      "learning_rate": 9.99202051941695e-06,
+      "loss": 0.8757240772247314,
+      "step": 40
+    },
+    {
+      "epoch": 0.1474820143884892,
+      "grad_norm": 0.5302181279741001,
+      "learning_rate": 9.991616671134045e-06,
+      "loss": 0.8662016987800598,
+      "step": 41
+    },
+    {
+      "epoch": 0.1510791366906475,
+      "grad_norm": 0.5110611514487727,
+      "learning_rate": 9.991202862576036e-06,
+      "loss": 0.8414884209632874,
+      "step": 42
+    },
+    {
+      "epoch": 0.15467625899280577,
+      "grad_norm": 0.5318960632271792,
+      "learning_rate": 9.99077909456864e-06,
+      "loss": 0.8459748029708862,
+      "step": 43
+    },
+    {
+      "epoch": 0.15827338129496402,
+      "grad_norm": 0.5509590436828911,
+      "learning_rate": 9.99034536795744e-06,
+      "loss": 0.8320192098617554,
+      "step": 44
+    },
+    {
+      "epoch": 0.1618705035971223,
+      "grad_norm": 0.5244253719817727,
+      "learning_rate": 9.989901683607893e-06,
+      "loss": 0.8553211688995361,
+      "step": 45
+    },
+    {
+      "epoch": 0.16546762589928057,
+      "grad_norm": 0.5323576490946048,
+      "learning_rate": 9.989448042405328e-06,
+      "loss": 0.8586495518684387,
+      "step": 46
+    },
+    {
+      "epoch": 0.16906474820143885,
+      "grad_norm": 0.5877802964746905,
+      "learning_rate": 9.988984445254944e-06,
+      "loss": 0.8607613444328308,
+      "step": 47
+    },
+    {
+      "epoch": 0.17266187050359713,
+      "grad_norm": 0.5495938798715749,
+      "learning_rate": 9.9885108930818e-06,
+      "loss": 0.8512550592422485,
+      "step": 48
+    },
+    {
+      "epoch": 0.17625899280575538,
+      "grad_norm": 0.5540112993478788,
+      "learning_rate": 9.988027386830825e-06,
+      "loss": 0.8515195846557617,
+      "step": 49
+    },
+    {
+      "epoch": 0.17985611510791366,
+      "grad_norm": 0.5806816109293668,
+      "learning_rate": 9.987533927466804e-06,
+      "loss": 0.8428958654403687,
+      "step": 50
+    },
+    {
+      "epoch": 0.18345323741007194,
+      "grad_norm": 0.5679037465356196,
+      "learning_rate": 9.987030515974389e-06,
+      "loss": 0.8490457534790039,
+      "step": 51
+    },
+    {
+      "epoch": 0.18705035971223022,
+      "grad_norm": 0.5649574806363405,
+      "learning_rate": 9.986517153358086e-06,
+      "loss": 0.8424092531204224,
+      "step": 52
+    },
+    {
+      "epoch": 0.1906474820143885,
+      "grad_norm": 0.5107378637113733,
+      "learning_rate": 9.985993840642261e-06,
+      "loss": 0.8451225161552429,
+      "step": 53
+    },
+    {
+      "epoch": 0.19424460431654678,
+      "grad_norm": 0.5266148919393265,
+      "learning_rate": 9.98546057887113e-06,
+      "loss": 0.8574127554893494,
+      "step": 54
+    },
+    {
+      "epoch": 0.19784172661870503,
+      "grad_norm": 0.4968321264322306,
+      "learning_rate": 9.984917369108767e-06,
+      "loss": 0.8196417093276978,
+      "step": 55
+    },
+    {
+      "epoch": 0.2014388489208633,
+      "grad_norm": 0.5301227993296292,
+      "learning_rate": 9.984364212439089e-06,
+      "loss": 0.8178601264953613,
+      "step": 56
+    },
+    {
+      "epoch": 0.20503597122302158,
+      "grad_norm": 0.5322541177201829,
+      "learning_rate": 9.98380110996587e-06,
+      "loss": 0.8680044412612915,
+      "step": 57
+    },
+    {
+      "epoch": 0.20863309352517986,
+      "grad_norm": 0.5617210521859881,
+      "learning_rate": 9.98322806281272e-06,
+      "loss": 0.8366754055023193,
+      "step": 58
+    },
+    {
+      "epoch": 0.21223021582733814,
+      "grad_norm": 0.5333079804740503,
+      "learning_rate": 9.9826450721231e-06,
+      "loss": 0.8472665548324585,
+      "step": 59
+    },
+    {
+      "epoch": 0.2158273381294964,
+      "grad_norm": 0.5670784495454859,
+      "learning_rate": 9.982052139060312e-06,
+      "loss": 0.8183034658432007,
+      "step": 60
+    },
+    {
+      "epoch": 0.21942446043165467,
+      "grad_norm": 0.522473762554814,
+      "learning_rate": 9.981449264807493e-06,
+      "loss": 0.8251825571060181,
+      "step": 61
+    },
+    {
+      "epoch": 0.22302158273381295,
+      "grad_norm": 0.5359962882882059,
+      "learning_rate": 9.980836450567619e-06,
+      "loss": 0.8035885095596313,
+      "step": 62
+    },
+    {
+      "epoch": 0.22661870503597123,
+      "grad_norm": 0.5452200763440471,
+      "learning_rate": 9.9802136975635e-06,
+      "loss": 0.850425124168396,
+      "step": 63
+    },
+    {
+      "epoch": 0.2302158273381295,
+      "grad_norm": 0.505908979785123,
+      "learning_rate": 9.979581007037776e-06,
+      "loss": 0.8475151062011719,
+      "step": 64
+    },
+    {
+      "epoch": 0.23381294964028776,
+      "grad_norm": 0.4973287153261601,
+      "learning_rate": 9.97893838025292e-06,
+      "loss": 0.8509107828140259,
+      "step": 65
+    },
+    {
+      "epoch": 0.23741007194244604,
+      "grad_norm": 0.5916298246777657,
+      "learning_rate": 9.978285818491232e-06,
+      "loss": 0.8484339714050293,
+      "step": 66
+    },
+    {
+      "epoch": 0.24100719424460432,
+      "grad_norm": 0.5448094767840429,
+      "learning_rate": 9.977623323054828e-06,
+      "loss": 0.8285648822784424,
+      "step": 67
+    },
+    {
+      "epoch": 0.2446043165467626,
+      "grad_norm": 0.5924423321400274,
+      "learning_rate": 9.976950895265657e-06,
+      "loss": 0.8395726680755615,
+      "step": 68
+    },
+    {
+      "epoch": 0.24820143884892087,
+      "grad_norm": 0.5664782512788527,
+      "learning_rate": 9.976268536465482e-06,
+      "loss": 0.8284764289855957,
+      "step": 69
+    },
+    {
+      "epoch": 0.2517985611510791,
+      "grad_norm": 0.5692368646880401,
+      "learning_rate": 9.975576248015878e-06,
+      "loss": 0.8533764481544495,
+      "step": 70
+    },
+    {
+      "epoch": 0.25539568345323743,
+      "grad_norm": 0.5573191956981842,
+      "learning_rate": 9.974874031298242e-06,
+      "loss": 0.8336201906204224,
+      "step": 71
+    },
+    {
+      "epoch": 0.2589928057553957,
+      "grad_norm": 0.5819733970676213,
+      "learning_rate": 9.974161887713775e-06,
+      "loss": 0.8206812143325806,
+      "step": 72
+    },
+    {
+      "epoch": 0.26258992805755393,
+      "grad_norm": 0.6358219302706564,
+      "learning_rate": 9.97343981868349e-06,
+      "loss": 0.8437105417251587,
+      "step": 73
+    },
+    {
+      "epoch": 0.26618705035971224,
+      "grad_norm": 0.5478398959411097,
+      "learning_rate": 9.972707825648204e-06,
+      "loss": 0.8165892362594604,
+      "step": 74
+    },
+    {
+      "epoch": 0.2697841726618705,
+      "grad_norm": 0.5383050403703055,
+      "learning_rate": 9.971965910068536e-06,
+      "loss": 0.8145566582679749,
+      "step": 75
+    },
+    {
+      "epoch": 0.2733812949640288,
+      "grad_norm": 0.5787407471208639,
+      "learning_rate": 9.971214073424906e-06,
+      "loss": 0.8369835019111633,
+      "step": 76
+    },
+    {
+      "epoch": 0.27697841726618705,
+      "grad_norm": 0.6060564743412846,
+      "learning_rate": 9.970452317217528e-06,
+      "loss": 0.8166014552116394,
+      "step": 77
+    },
+    {
+      "epoch": 0.2805755395683453,
+      "grad_norm": 0.5626076159082846,
+      "learning_rate": 9.969680642966409e-06,
+      "loss": 0.8157432079315186,
+      "step": 78
+    },
+    {
+      "epoch": 0.2841726618705036,
+      "grad_norm": 0.5633091616428877,
+      "learning_rate": 9.96889905221135e-06,
+      "loss": 0.8347904086112976,
+      "step": 79
+    },
+    {
+      "epoch": 0.28776978417266186,
+      "grad_norm": 0.5768068832703904,
+      "learning_rate": 9.968107546511942e-06,
+      "loss": 0.8339566588401794,
+      "step": 80
+    },
+    {
+      "epoch": 0.29136690647482016,
+      "grad_norm": 0.5328471089077459,
+      "learning_rate": 9.96730612744755e-06,
+      "loss": 0.8300585746765137,
+      "step": 81
+    },
+    {
+      "epoch": 0.2949640287769784,
+      "grad_norm": 0.5259481084278995,
+      "learning_rate": 9.966494796617328e-06,
+      "loss": 0.8587024211883545,
+      "step": 82
+    },
+    {
+      "epoch": 0.29856115107913667,
+      "grad_norm": 0.6263217869688971,
+      "learning_rate": 9.965673555640208e-06,
+      "loss": 0.7940714955329895,
+      "step": 83
+    },
+    {
+      "epoch": 0.302158273381295,
+      "grad_norm": 0.5672018351537288,
+      "learning_rate": 9.964842406154892e-06,
+      "loss": 0.8141801357269287,
+      "step": 84
+    },
+    {
+      "epoch": 0.3057553956834532,
+      "grad_norm": 0.534278403371781,
+      "learning_rate": 9.964001349819856e-06,
+      "loss": 0.8463040590286255,
+      "step": 85
+    },
+    {
+      "epoch": 0.30935251798561153,
+      "grad_norm": 0.5640353678708808,
+      "learning_rate": 9.963150388313347e-06,
+      "loss": 0.817227840423584,
+      "step": 86
+    },
+    {
+      "epoch": 0.3129496402877698,
+      "grad_norm": 0.5402262875034433,
+      "learning_rate": 9.962289523333372e-06,
+      "loss": 0.8320227861404419,
+      "step": 87
+    },
+    {
+      "epoch": 0.31654676258992803,
+      "grad_norm": 0.5559951550582068,
+      "learning_rate": 9.961418756597703e-06,
+      "loss": 0.8306981325149536,
+      "step": 88
+    },
+    {
+      "epoch": 0.32014388489208634,
+      "grad_norm": 0.575668186919706,
+      "learning_rate": 9.960538089843868e-06,
+      "loss": 0.8308272361755371,
+      "step": 89
+    },
+    {
+      "epoch": 0.3237410071942446,
+      "grad_norm": 0.5382001607426908,
+      "learning_rate": 9.959647524829148e-06,
+      "loss": 0.8604013323783875,
+      "step": 90
+    },
+    {
+      "epoch": 0.3273381294964029,
+      "grad_norm": 0.5604808001643791,
+      "learning_rate": 9.95874706333058e-06,
+      "loss": 0.8057097792625427,
+      "step": 91
+    },
+    {
+      "epoch": 0.33093525179856115,
+      "grad_norm": 0.5619174670637143,
+      "learning_rate": 9.957836707144943e-06,
+      "loss": 0.8586455583572388,
+      "step": 92
+    },
+    {
+      "epoch": 0.3345323741007194,
+      "grad_norm": 0.5431824517780087,
+      "learning_rate": 9.956916458088765e-06,
+      "loss": 0.8322749137878418,
+      "step": 93
+    },
+    {
+      "epoch": 0.3381294964028777,
+      "grad_norm": 0.5304521035196852,
+      "learning_rate": 9.95598631799831e-06,
+      "loss": 0.8248855471611023,
+      "step": 94
+    },
+    {
+      "epoch": 0.34172661870503596,
+      "grad_norm": 0.5606065890967215,
+      "learning_rate": 9.95504628872958e-06,
+      "loss": 0.8336918950080872,
+      "step": 95
+    },
+    {
+      "epoch": 0.34532374100719426,
+      "grad_norm": 0.5383743438248503,
+      "learning_rate": 9.95409637215831e-06,
+      "loss": 0.7947922945022583,
+      "step": 96
+    },
+    {
+      "epoch": 0.3489208633093525,
+      "grad_norm": 0.5318998589457373,
+      "learning_rate": 9.953136570179966e-06,
+      "loss": 0.8381302356719971,
+      "step": 97
+    },
+    {
+      "epoch": 0.35251798561151076,
+      "grad_norm": 0.5019982029325636,
+      "learning_rate": 9.952166884709735e-06,
+      "loss": 0.7833576202392578,
+      "step": 98
+    },
+    {
+      "epoch": 0.35611510791366907,
+      "grad_norm": 0.5382418331396808,
+      "learning_rate": 9.951187317682527e-06,
+      "loss": 0.8222751021385193,
+      "step": 99
+    },
+    {
+      "epoch": 0.3597122302158273,
+      "grad_norm": 0.5170694830246451,
+      "learning_rate": 9.950197871052974e-06,
+      "loss": 0.8138470649719238,
+      "step": 100
+    },
+    {
+      "epoch": 0.3597122302158273,
+      "eval_loss": 0.816231906414032,
+      "eval_runtime": 7.507,
+      "eval_samples_per_second": 11.722,
+      "eval_steps_per_second": 0.4,
+      "eval_token_acc": 0.74746085109322,
+      "step": 100
+    },
+    {
+      "epoch": 0.36330935251798563,
+      "grad_norm": 0.5673295296992458,
+      "learning_rate": 9.949198546795418e-06,
+      "loss": 0.8289770483970642,
+      "step": 101
+    },
+    {
+      "epoch": 0.3669064748201439,
+      "grad_norm": 0.563998064338557,
+      "learning_rate": 9.94818934690391e-06,
+      "loss": 0.8491591811180115,
+      "step": 102
+    },
+    {
+      "epoch": 0.37050359712230213,
+      "grad_norm": 0.5269425479847079,
+      "learning_rate": 9.947170273392206e-06,
+      "loss": 0.8284671902656555,
+      "step": 103
+    },
+    {
+      "epoch": 0.37410071942446044,
+      "grad_norm": 0.5473532877664629,
+      "learning_rate": 9.94614132829377e-06,
+      "loss": 0.8391440510749817,
+      "step": 104
+    },
+    {
+      "epoch": 0.3776978417266187,
+      "grad_norm": 0.5584379208820623,
+      "learning_rate": 9.945102513661758e-06,
+      "loss": 0.8177752494812012,
+      "step": 105
+    },
+    {
+      "epoch": 0.381294964028777,
+      "grad_norm": 0.5660179757557332,
+      "learning_rate": 9.94405383156902e-06,
+      "loss": 0.8083318471908569,
+      "step": 106
+    },
+    {
+      "epoch": 0.38489208633093525,
+      "grad_norm": 0.5366634741683651,
+      "learning_rate": 9.9429952841081e-06,
+      "loss": 0.826050877571106,
+      "step": 107
+    },
+    {
+      "epoch": 0.38848920863309355,
+      "grad_norm": 0.5282446109645809,
+      "learning_rate": 9.941926873391223e-06,
+      "loss": 0.8216806650161743,
+      "step": 108
+    },
+    {
+      "epoch": 0.3920863309352518,
+      "grad_norm": 0.5386018738453927,
+      "learning_rate": 9.940848601550295e-06,
+      "loss": 0.788651704788208,
+      "step": 109
+    },
+    {
+      "epoch": 0.39568345323741005,
+      "grad_norm": 0.5483359069173901,
+      "learning_rate": 9.939760470736902e-06,
+      "loss": 0.8089826107025146,
+      "step": 110
+    },
+    {
+      "epoch": 0.39928057553956836,
+      "grad_norm": 0.5617901623159829,
+      "learning_rate": 9.938662483122299e-06,
+      "loss": 0.7851020097732544,
+      "step": 111
+    },
+    {
+      "epoch": 0.4028776978417266,
+      "grad_norm": 0.591579269733334,
+      "learning_rate": 9.937554640897414e-06,
+      "loss": 0.8273585438728333,
+      "step": 112
+    },
+    {
+      "epoch": 0.4064748201438849,
+      "grad_norm": 0.557557042763066,
+      "learning_rate": 9.936436946272834e-06,
+      "loss": 0.8208383917808533,
+      "step": 113
+    },
+    {
+      "epoch": 0.41007194244604317,
+      "grad_norm": 0.548113632755496,
+      "learning_rate": 9.935309401478808e-06,
+      "loss": 0.8339937925338745,
+      "step": 114
+    },
+    {
+      "epoch": 0.4136690647482014,
+      "grad_norm": 0.5492991264279992,
+      "learning_rate": 9.93417200876524e-06,
+      "loss": 0.8161983489990234,
+      "step": 115
+    },
+    {
+      "epoch": 0.4172661870503597,
+      "grad_norm": 0.5883988065819238,
+      "learning_rate": 9.933024770401682e-06,
+      "loss": 0.8346230983734131,
+      "step": 116
+    },
+    {
+      "epoch": 0.420863309352518,
+      "grad_norm": 0.574684102157635,
+      "learning_rate": 9.931867688677338e-06,
+      "loss": 0.8311742544174194,
+      "step": 117
+    },
+    {
+      "epoch": 0.4244604316546763,
+      "grad_norm": 0.5340788781452219,
+      "learning_rate": 9.930700765901046e-06,
+      "loss": 0.8129891157150269,
+      "step": 118
+    },
+    {
+      "epoch": 0.42805755395683454,
+      "grad_norm": 0.5336075844649993,
+      "learning_rate": 9.92952400440129e-06,
+      "loss": 0.8301393985748291,
+      "step": 119
+    },
+    {
+      "epoch": 0.4316546762589928,
+      "grad_norm": 0.5416549418922793,
+      "learning_rate": 9.928337406526172e-06,
+      "loss": 0.8299385905265808,
+      "step": 120
+    },
+    {
+      "epoch": 0.4352517985611511,
+      "grad_norm": 0.5807637671155456,
+      "learning_rate": 9.927140974643438e-06,
+      "loss": 0.8278335332870483,
+      "step": 121
+    },
+    {
+      "epoch": 0.43884892086330934,
+      "grad_norm": 0.5707236799451327,
+      "learning_rate": 9.925934711140444e-06,
+      "loss": 0.8232138156890869,
+      "step": 122
+    },
+    {
+      "epoch": 0.44244604316546765,
+      "grad_norm": 0.5567248659200839,
+      "learning_rate": 9.924718618424172e-06,
+      "loss": 0.8311301469802856,
+      "step": 123
+    },
+    {
+      "epoch": 0.4460431654676259,
+      "grad_norm": 0.5594513308954804,
+      "learning_rate": 9.923492698921214e-06,
+      "loss": 0.8267166614532471,
+      "step": 124
+    },
+    {
+      "epoch": 0.44964028776978415,
+      "grad_norm": 0.546018461439583,
+      "learning_rate": 9.922256955077767e-06,
+      "loss": 0.8282893896102905,
+      "step": 125
+    },
+    {
+      "epoch": 0.45323741007194246,
+      "grad_norm": 0.5325942679065744,
+      "learning_rate": 9.921011389359638e-06,
+      "loss": 0.8170527219772339,
+      "step": 126
+    },
+    {
+      "epoch": 0.4568345323741007,
+      "grad_norm": 0.5480251944916799,
+      "learning_rate": 9.91975600425223e-06,
+      "loss": 0.8059794902801514,
+      "step": 127
+    },
+    {
+      "epoch": 0.460431654676259,
+      "grad_norm": 0.5395204540816534,
+      "learning_rate": 9.918490802260538e-06,
+      "loss": 0.81952303647995,
+      "step": 128
+    },
+    {
+      "epoch": 0.46402877697841727,
+      "grad_norm": 0.5392455596536625,
+      "learning_rate": 9.917215785909149e-06,
+      "loss": 0.8166015148162842,
+      "step": 129
+    },
+    {
+      "epoch": 0.4676258992805755,
+      "grad_norm": 0.5359787934095424,
+      "learning_rate": 9.915930957742228e-06,
+      "loss": 0.80816251039505,
+      "step": 130
+    },
+    {
+      "epoch": 0.4712230215827338,
+      "grad_norm": 0.5119849543094582,
+      "learning_rate": 9.914636320323525e-06,
+      "loss": 0.8249898552894592,
+      "step": 131
+    },
+    {
+      "epoch": 0.4748201438848921,
+      "grad_norm": 0.5921026203775647,
+      "learning_rate": 9.913331876236358e-06,
+      "loss": 0.826164186000824,
+      "step": 132
+    },
+    {
+      "epoch": 0.4784172661870504,
+      "grad_norm": 0.5533992224690237,
+      "learning_rate": 9.912017628083618e-06,
+      "loss": 0.808993935585022,
+      "step": 133
+    },
+    {
+      "epoch": 0.48201438848920863,
+      "grad_norm": 0.5509377136584458,
+      "learning_rate": 9.910693578487755e-06,
+      "loss": 0.8223838210105896,
+      "step": 134
+    },
+    {
+      "epoch": 0.4856115107913669,
+      "grad_norm": 0.5177279997027195,
+      "learning_rate": 9.909359730090777e-06,
+      "loss": 0.8290198445320129,
+      "step": 135
+    },
+    {
+      "epoch": 0.4892086330935252,
+      "grad_norm": 0.5583013977579655,
+      "learning_rate": 9.90801608555425e-06,
+      "loss": 0.8074214458465576,
+      "step": 136
+    },
+    {
+      "epoch": 0.49280575539568344,
+      "grad_norm": 0.5227084946155521,
+      "learning_rate": 9.906662647559279e-06,
+      "loss": 0.8390957713127136,
+      "step": 137
+    },
+    {
+      "epoch": 0.49640287769784175,
+      "grad_norm": 0.5406546058934267,
+      "learning_rate": 9.905299418806517e-06,
+      "loss": 0.7853333950042725,
+      "step": 138
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.5476922356088839,
+      "learning_rate": 9.903926402016153e-06,
+      "loss": 0.8493127822875977,
+      "step": 139
+    },
+    {
+      "epoch": 0.5035971223021583,
+      "grad_norm": 0.5005184529391463,
+      "learning_rate": 9.902543599927903e-06,
+      "loss": 0.8202897906303406,
+      "step": 140
+    },
+    {
+      "epoch": 0.5071942446043165,
+      "grad_norm": 0.5958907753430173,
+      "learning_rate": 9.901151015301012e-06,
+      "loss": 0.8216853737831116,
+      "step": 141
+    },
+    {
+      "epoch": 0.5107913669064749,
+      "grad_norm": 0.5599001635444122,
+      "learning_rate": 9.899748650914245e-06,
+      "loss": 0.8264304399490356,
+      "step": 142
+    },
+    {
+      "epoch": 0.5143884892086331,
+      "grad_norm": 0.5215334397603238,
+      "learning_rate": 9.89833650956588e-06,
+      "loss": 0.8092686533927917,
+      "step": 143
+    },
+    {
+      "epoch": 0.5179856115107914,
+      "grad_norm": 0.5542935554492459,
+      "learning_rate": 9.896914594073703e-06,
+      "loss": 0.798918604850769,
+      "step": 144
+    },
+    {
+      "epoch": 0.5215827338129496,
+      "grad_norm": 0.528969503072376,
+      "learning_rate": 9.895482907275009e-06,
+      "loss": 0.793848991394043,
+      "step": 145
+    },
+    {
+      "epoch": 0.5251798561151079,
+      "grad_norm": 0.5063983054470111,
+      "learning_rate": 9.894041452026584e-06,
+      "loss": 0.8367159962654114,
+      "step": 146
+    },
+    {
+      "epoch": 0.5287769784172662,
+      "grad_norm": 0.5543247794065501,
+      "learning_rate": 9.892590231204709e-06,
+      "loss": 0.8097944259643555,
+      "step": 147
+    },
+    {
+      "epoch": 0.5323741007194245,
+      "grad_norm": 0.539616362600845,
+      "learning_rate": 9.891129247705153e-06,
+      "loss": 0.809989333152771,
+      "step": 148
+    },
+    {
+      "epoch": 0.5359712230215827,
+      "grad_norm": 0.5113476170140943,
+      "learning_rate": 9.88965850444316e-06,
+      "loss": 0.8000381588935852,
+      "step": 149
+    },
+    {
+      "epoch": 0.539568345323741,
+      "grad_norm": 0.5424905593254367,
+      "learning_rate": 9.888178004353456e-06,
+      "loss": 0.7990270256996155,
+      "step": 150
+    },
+    {
+      "epoch": 0.5431654676258992,
+      "grad_norm": 0.5388457802161849,
+      "learning_rate": 9.88668775039023e-06,
+      "loss": 0.8318261504173279,
+      "step": 151
+    },
+    {
+      "epoch": 0.5467625899280576,
+      "grad_norm": 0.5485418764664756,
+      "learning_rate": 9.885187745527132e-06,
+      "loss": 0.8133857250213623,
+      "step": 152
+    },
+    {
+      "epoch": 0.5503597122302158,
+      "grad_norm": 0.544940799274823,
+      "learning_rate": 9.883677992757278e-06,
+      "loss": 0.8147492408752441,
+      "step": 153
+    },
+    {
+      "epoch": 0.5539568345323741,
+      "grad_norm": 0.5709632152106743,
+      "learning_rate": 9.88215849509323e-06,
+      "loss": 0.8083347082138062,
+      "step": 154
+    },
+    {
+      "epoch": 0.5575539568345323,
+      "grad_norm": 0.58028630349183,
+      "learning_rate": 9.880629255566988e-06,
+      "loss": 0.80954909324646,
+      "step": 155
+    },
+    {
+      "epoch": 0.5611510791366906,
+      "grad_norm": 0.5566497809950424,
+      "learning_rate": 9.879090277230005e-06,
+      "loss": 0.8241982460021973,
+      "step": 156
+    },
+    {
+      "epoch": 0.564748201438849,
+      "grad_norm": 0.5518771004831493,
+      "learning_rate": 9.877541563153157e-06,
+      "loss": 0.7945687174797058,
+      "step": 157
+    },
+    {
+      "epoch": 0.5683453237410072,
+      "grad_norm": 0.5367710065256509,
+      "learning_rate": 9.875983116426745e-06,
+      "loss": 0.8111982941627502,
+      "step": 158
+    },
+    {
+      "epoch": 0.5719424460431655,
+      "grad_norm": 0.5486988063810168,
+      "learning_rate": 9.874414940160501e-06,
+      "loss": 0.8155021667480469,
+      "step": 159
+    },
+    {
+      "epoch": 0.5755395683453237,
+      "grad_norm": 0.5480126518439215,
+      "learning_rate": 9.87283703748356e-06,
+      "loss": 0.7951115369796753,
+      "step": 160
+    },
+    {
+      "epoch": 0.579136690647482,
+      "grad_norm": 0.5717071424611518,
+      "learning_rate": 9.871249411544473e-06,
+      "loss": 0.8322063088417053,
+      "step": 161
+    },
+    {
+      "epoch": 0.5827338129496403,
+      "grad_norm": 0.528060657804702,
+      "learning_rate": 9.869652065511188e-06,
+      "loss": 0.8200956583023071,
+      "step": 162
+    },
+    {
+      "epoch": 0.5863309352517986,
+      "grad_norm": 0.5244238381398686,
+      "learning_rate": 9.868045002571052e-06,
+      "loss": 0.8096636533737183,
+      "step": 163
+    },
+    {
+      "epoch": 0.5899280575539568,
+      "grad_norm": 0.49919420997101305,
+      "learning_rate": 9.866428225930798e-06,
+      "loss": 0.7951469421386719,
+      "step": 164
+    },
+    {
+      "epoch": 0.5935251798561151,
+      "grad_norm": 0.535446981889461,
+      "learning_rate": 9.864801738816543e-06,
+      "loss": 0.8106690049171448,
+      "step": 165
+    },
+    {
+      "epoch": 0.5971223021582733,
+      "grad_norm": 0.5333985451516444,
+      "learning_rate": 9.863165544473781e-06,
+      "loss": 0.8486871719360352,
+      "step": 166
+    },
+    {
+      "epoch": 0.6007194244604317,
+      "grad_norm": 0.5458606068076399,
+      "learning_rate": 9.861519646167375e-06,
+      "loss": 0.7913973331451416,
+      "step": 167
+    },
+    {
+      "epoch": 0.60431654676259,
+      "grad_norm": 0.5489824115884749,
+      "learning_rate": 9.859864047181551e-06,
+      "loss": 0.8237329125404358,
+      "step": 168
+    },
+    {
+      "epoch": 0.6079136690647482,
+      "grad_norm": 0.5525766076515911,
+      "learning_rate": 9.858198750819896e-06,
+      "loss": 0.7945156097412109,
+      "step": 169
+    },
+    {
+      "epoch": 0.6115107913669064,
+      "grad_norm": 0.5434672326912037,
+      "learning_rate": 9.856523760405339e-06,
+      "loss": 0.826322078704834,
+      "step": 170
+    },
+    {
+      "epoch": 0.6151079136690647,
+      "grad_norm": 0.5272432795596584,
+      "learning_rate": 9.854839079280157e-06,
+      "loss": 0.8193082213401794,
+      "step": 171
+    },
+    {
+      "epoch": 0.6187050359712231,
+      "grad_norm": 0.5424514802280374,
+      "learning_rate": 9.853144710805966e-06,
+      "loss": 0.7865622639656067,
+      "step": 172
+    },
+    {
+      "epoch": 0.6223021582733813,
+      "grad_norm": 0.5354805511996885,
+      "learning_rate": 9.851440658363708e-06,
+      "loss": 0.8112363815307617,
+      "step": 173
+    },
+    {
+      "epoch": 0.6258992805755396,
+      "grad_norm": 0.5386838343839405,
+      "learning_rate": 9.849726925353655e-06,
+      "loss": 0.8229953646659851,
+      "step": 174
+    },
+    {
+      "epoch": 0.6294964028776978,
+      "grad_norm": 0.5531212837926246,
+      "learning_rate": 9.848003515195384e-06,
+      "loss": 0.8074018955230713,
+      "step": 175
+    },
+    {
+      "epoch": 0.6330935251798561,
+      "grad_norm": 0.5708673804346629,
+      "learning_rate": 9.846270431327793e-06,
+      "loss": 0.8243955373764038,
+      "step": 176
+    },
+    {
+      "epoch": 0.6366906474820144,
+      "grad_norm": 0.5576031502036335,
+      "learning_rate": 9.84452767720908e-06,
+      "loss": 0.8192329406738281,
+      "step": 177
+    },
+    {
+      "epoch": 0.6402877697841727,
+      "grad_norm": 0.5436357529700776,
+      "learning_rate": 9.842775256316732e-06,
+      "loss": 0.8275960683822632,
+      "step": 178
+    },
+    {
+      "epoch": 0.6438848920863309,
+      "grad_norm": 0.5913335099756953,
+      "learning_rate": 9.841013172147537e-06,
+      "loss": 0.8125853538513184,
+      "step": 179
+    },
+    {
+      "epoch": 0.6474820143884892,
+      "grad_norm": 0.5813220515343052,
+      "learning_rate": 9.839241428217553e-06,
+      "loss": 0.8220874071121216,
+      "step": 180
+    },
+    {
+      "epoch": 0.6510791366906474,
+      "grad_norm": 0.5298380673365016,
+      "learning_rate": 9.837460028062126e-06,
+      "loss": 0.8092227578163147,
+      "step": 181
+    },
+    {
+      "epoch": 0.6546762589928058,
+      "grad_norm": 0.5772475049484589,
+      "learning_rate": 9.835668975235857e-06,
+      "loss": 0.8138964176177979,
+      "step": 182
+    },
+    {
+      "epoch": 0.658273381294964,
+      "grad_norm": 0.5786451222409693,
+      "learning_rate": 9.833868273312617e-06,
+      "loss": 0.7935299277305603,
+      "step": 183
+    },
+    {
+      "epoch": 0.6618705035971223,
+      "grad_norm": 0.5552310441780026,
+      "learning_rate": 9.832057925885526e-06,
+      "loss": 0.7891409397125244,
+      "step": 184
+    },
+    {
+      "epoch": 0.6654676258992805,
+      "grad_norm": 0.5734949837391681,
+      "learning_rate": 9.830237936566953e-06,
+      "loss": 0.7957472801208496,
+      "step": 185
+    },
+    {
+      "epoch": 0.6690647482014388,
+      "grad_norm": 0.599694217341625,
+      "learning_rate": 9.828408308988506e-06,
+      "loss": 0.8263557553291321,
+      "step": 186
+    },
+    {
+      "epoch": 0.6726618705035972,
+      "grad_norm": 0.5799493702311006,
+      "learning_rate": 9.826569046801027e-06,
+      "loss": 0.7941974401473999,
+      "step": 187
+    },
+    {
+      "epoch": 0.6762589928057554,
+      "grad_norm": 0.5563225911482126,
+      "learning_rate": 9.824720153674578e-06,
+      "loss": 0.829971194267273,
+      "step": 188
+    },
+    {
+      "epoch": 0.6798561151079137,
+      "grad_norm": 0.5351288370080682,
+      "learning_rate": 9.822861633298442e-06,
+      "loss": 0.8155908584594727,
+      "step": 189
+    },
+    {
+      "epoch": 0.6834532374100719,
+      "grad_norm": 0.5331482850465583,
+      "learning_rate": 9.820993489381114e-06,
+      "loss": 0.8068544864654541,
+      "step": 190
+    },
+    {
+      "epoch": 0.6870503597122302,
+      "grad_norm": 0.552140047325023,
+      "learning_rate": 9.819115725650287e-06,
+      "loss": 0.8251328468322754,
+      "step": 191
+    },
+    {
+      "epoch": 0.6906474820143885,
+      "grad_norm": 0.5565602616626132,
+      "learning_rate": 9.817228345852853e-06,
+      "loss": 0.8075183629989624,
+      "step": 192
+    },
+    {
+      "epoch": 0.6942446043165468,
+      "grad_norm": 0.5424847223472784,
+      "learning_rate": 9.81533135375489e-06,
+      "loss": 0.7979018092155457,
+      "step": 193
+    },
+    {
+      "epoch": 0.697841726618705,
+      "grad_norm": 0.5341753227747238,
+      "learning_rate": 9.813424753141658e-06,
+      "loss": 0.8171260356903076,
+      "step": 194
+    },
+    {
+      "epoch": 0.7014388489208633,
+      "grad_norm": 0.5804615507655949,
+      "learning_rate": 9.81150854781759e-06,
+      "loss": 0.8082928657531738,
+      "step": 195
+    },
+    {
+      "epoch": 0.7050359712230215,
+      "grad_norm": 0.5509094640235682,
+      "learning_rate": 9.809582741606283e-06,
+      "loss": 0.8074817657470703,
+      "step": 196
+    },
+    {
+      "epoch": 0.7086330935251799,
+      "grad_norm": 0.5228610158009157,
+      "learning_rate": 9.80764733835049e-06,
+      "loss": 0.8101227283477783,
+      "step": 197
+    },
+    {
+      "epoch": 0.7122302158273381,
+      "grad_norm": 0.5359888946429793,
+      "learning_rate": 9.805702341912117e-06,
+      "loss": 0.8064497709274292,
+      "step": 198
+    },
+    {
+      "epoch": 0.7158273381294964,
+      "grad_norm": 0.5280348608576221,
+      "learning_rate": 9.803747756172214e-06,
+      "loss": 0.8218724727630615,
+      "step": 199
+    },
+    {
+      "epoch": 0.7194244604316546,
+      "grad_norm": 0.5264312906235269,
+      "learning_rate": 9.801783585030959e-06,
+      "loss": 0.7979246377944946,
+      "step": 200
+    },
+    {
+      "epoch": 0.7194244604316546,
+      "eval_loss": 0.7993502616882324,
+      "eval_runtime": 7.4862,
+      "eval_samples_per_second": 11.755,
+      "eval_steps_per_second": 0.401,
+      "eval_token_acc": 0.7509832726974058,
+      "step": 200
+    },
+    {
+      "epoch": 0.7230215827338129,
+      "grad_norm": 0.5604566185186979,
+      "learning_rate": 9.79980983240766e-06,
+      "loss": 0.8044528961181641,
+      "step": 201
+    },
+    {
+      "epoch": 0.7266187050359713,
+      "grad_norm": 0.5690323675931319,
+      "learning_rate": 9.797826502240746e-06,
+      "loss": 0.8124678134918213,
+      "step": 202
+    },
+    {
+      "epoch": 0.7302158273381295,
+      "grad_norm": 0.5295911758422164,
+      "learning_rate": 9.795833598487757e-06,
+      "loss": 0.7948049306869507,
+      "step": 203
+    },
+    {
+      "epoch": 0.7338129496402878,
+      "grad_norm": 0.5420116184769913,
+      "learning_rate": 9.79383112512533e-06,
+      "loss": 0.8087724447250366,
+      "step": 204
+    },
+    {
+      "epoch": 0.737410071942446,
+      "grad_norm": 0.52000926600773,
+      "learning_rate": 9.791819086149204e-06,
+      "loss": 0.7969048023223877,
+      "step": 205
+    },
+    {
+      "epoch": 0.7410071942446043,
+      "grad_norm": 0.5439164932565841,
+      "learning_rate": 9.789797485574204e-06,
+      "loss": 0.8051252365112305,
+      "step": 206
+    },
+    {
+      "epoch": 0.7446043165467626,
+      "grad_norm": 0.527493581148968,
+      "learning_rate": 9.787766327434231e-06,
+      "loss": 0.8121972680091858,
+      "step": 207
+    },
+    {
+      "epoch": 0.7482014388489209,
+      "grad_norm": 0.5316866063486113,
+      "learning_rate": 9.785725615782262e-06,
+      "loss": 0.7998046278953552,
+      "step": 208
+    },
+    {
+      "epoch": 0.7517985611510791,
+      "grad_norm": 0.5529173658462174,
+      "learning_rate": 9.783675354690331e-06,
+      "loss": 0.7912847399711609,
+      "step": 209
+    },
+    {
+      "epoch": 0.7553956834532374,
+      "grad_norm": 0.5434220395363016,
+      "learning_rate": 9.781615548249533e-06,
+      "loss": 0.8146765232086182,
+      "step": 210
+    },
+    {
+      "epoch": 0.7589928057553957,
+      "grad_norm": 0.5174228032873612,
+      "learning_rate": 9.779546200570005e-06,
+      "loss": 0.8111634850502014,
+      "step": 211
+    },
+    {
+      "epoch": 0.762589928057554,
+      "grad_norm": 0.5196942606957381,
+      "learning_rate": 9.777467315780926e-06,
+      "loss": 0.815482497215271,
+      "step": 212
+    },
+    {
+      "epoch": 0.7661870503597122,
+      "grad_norm": 0.5341280558565928,
+      "learning_rate": 9.775378898030503e-06,
+      "loss": 0.8034203052520752,
+      "step": 213
+    },
+    {
+      "epoch": 0.7697841726618705,
+      "grad_norm": 0.5284120510650635,
+      "learning_rate": 9.773280951485967e-06,
+      "loss": 0.8246862292289734,
+      "step": 214
+    },
+    {
+      "epoch": 0.7733812949640287,
+      "grad_norm": 0.5544183956098815,
+      "learning_rate": 9.771173480333562e-06,
+      "loss": 0.8020129799842834,
+      "step": 215
+    },
+    {
+      "epoch": 0.7769784172661871,
+      "grad_norm": 0.5229832751207113,
+      "learning_rate": 9.769056488778538e-06,
+      "loss": 0.8137690424919128,
+      "step": 216
+    },
+    {
+      "epoch": 0.7805755395683454,
+      "grad_norm": 0.5699698782617088,
+      "learning_rate": 9.766929981045139e-06,
+      "loss": 0.7960363030433655,
+      "step": 217
+    },
+    {
+      "epoch": 0.7841726618705036,
+      "grad_norm": 0.5417245877819069,
+      "learning_rate": 9.764793961376602e-06,
+      "loss": 0.8026304244995117,
+      "step": 218
+    },
+    {
+      "epoch": 0.7877697841726619,
+      "grad_norm": 0.5541975108855247,
+      "learning_rate": 9.76264843403514e-06,
+      "loss": 0.820999026298523,
+      "step": 219
+    },
+    {
+      "epoch": 0.7913669064748201,
+      "grad_norm": 0.5646050091020177,
+      "learning_rate": 9.760493403301941e-06,
+      "loss": 0.7787805199623108,
+      "step": 220
+    },
+    {
+      "epoch": 0.7949640287769785,
+      "grad_norm": 0.5632008003186489,
+      "learning_rate": 9.758328873477154e-06,
+      "loss": 0.805914044380188,
+      "step": 221
+    },
+    {
+      "epoch": 0.7985611510791367,
+      "grad_norm": 0.5508175428854032,
+      "learning_rate": 9.756154848879885e-06,
+      "loss": 0.8235931992530823,
+      "step": 222
+    },
+    {
+      "epoch": 0.802158273381295,
+      "grad_norm": 0.5655349496395068,
+      "learning_rate": 9.753971333848183e-06,
+      "loss": 0.8301273584365845,
+      "step": 223
+    },
+    {
+      "epoch": 0.8057553956834532,
+      "grad_norm": 0.5507470566009575,
+      "learning_rate": 9.751778332739033e-06,
+      "loss": 0.7969081401824951,
+      "step": 224
+    },
+    {
+      "epoch": 0.8093525179856115,
+      "grad_norm": 0.5581369046139822,
+      "learning_rate": 9.749575849928354e-06,
+      "loss": 0.802039384841919,
+      "step": 225
+    },
+    {
+      "epoch": 0.8129496402877698,
+      "grad_norm": 0.5481550959696141,
+      "learning_rate": 9.74736388981098e-06,
+      "loss": 0.7913751602172852,
+      "step": 226
+    },
+    {
+      "epoch": 0.8165467625899281,
+      "grad_norm": 0.5654210262128606,
+      "learning_rate": 9.74514245680066e-06,
+      "loss": 0.7987425327301025,
+      "step": 227
+    },
+    {
+      "epoch": 0.8201438848920863,
+      "grad_norm": 0.5265619857022084,
+      "learning_rate": 9.74291155533004e-06,
+      "loss": 0.8149737119674683,
+      "step": 228
+    },
+    {
+      "epoch": 0.8237410071942446,
+      "grad_norm": 0.5501005539202152,
+      "learning_rate": 9.740671189850664e-06,
+      "loss": 0.8153947591781616,
+      "step": 229
+    },
+    {
+      "epoch": 0.8273381294964028,
+      "grad_norm": 0.512562612677006,
+      "learning_rate": 9.738421364832956e-06,
+      "loss": 0.8032717704772949,
+      "step": 230
+    },
+    {
+      "epoch": 0.8309352517985612,
+      "grad_norm": 0.56695996273869,
+      "learning_rate": 9.73616208476622e-06,
+      "loss": 0.8103581070899963,
+      "step": 231
+    },
+    {
+      "epoch": 0.8345323741007195,
+      "grad_norm": 0.5176504269302804,
+      "learning_rate": 9.733893354158628e-06,
+      "loss": 0.797197699546814,
+      "step": 232
+    },
+    {
+      "epoch": 0.8381294964028777,
+      "grad_norm": 0.568000922091697,
+      "learning_rate": 9.7316151775372e-06,
+      "loss": 0.83544921875,
+      "step": 233
+    },
+    {
+      "epoch": 0.841726618705036,
+      "grad_norm": 0.5386142282078233,
+      "learning_rate": 9.729327559447817e-06,
+      "loss": 0.782428503036499,
+      "step": 234
+    },
+    {
+      "epoch": 0.8453237410071942,
+      "grad_norm": 0.520703942967087,
+      "learning_rate": 9.727030504455192e-06,
+      "loss": 0.8059707283973694,
+      "step": 235
+    },
+    {
+      "epoch": 0.8489208633093526,
+      "grad_norm": 0.5632290536976243,
+      "learning_rate": 9.724724017142869e-06,
+      "loss": 0.8086649179458618,
+      "step": 236
+    },
+    {
+      "epoch": 0.8525179856115108,
+      "grad_norm": 0.5422450922661384,
+      "learning_rate": 9.722408102113212e-06,
+      "loss": 0.7960709929466248,
+      "step": 237
+    },
+    {
+      "epoch": 0.8561151079136691,
+      "grad_norm": 0.5303594727396739,
+      "learning_rate": 9.720082763987404e-06,
+      "loss": 0.802277684211731,
+      "step": 238
+    },
+    {
+      "epoch": 0.8597122302158273,
+      "grad_norm": 0.5607106547711301,
+      "learning_rate": 9.717748007405422e-06,
+      "loss": 0.7961342930793762,
+      "step": 239
+    },
+    {
+      "epoch": 0.8633093525179856,
+      "grad_norm": 0.5600486650731337,
+      "learning_rate": 9.715403837026046e-06,
+      "loss": 0.7967619895935059,
+      "step": 240
+    },
+    {
+      "epoch": 0.8669064748201439,
+      "grad_norm": 0.5516198506118594,
+      "learning_rate": 9.71305025752683e-06,
+      "loss": 0.8042537569999695,
+      "step": 241
+    },
+    {
+      "epoch": 0.8705035971223022,
+      "grad_norm": 0.5402147757719346,
+      "learning_rate": 9.710687273604111e-06,
+      "loss": 0.8142298460006714,
+      "step": 242
+    },
+    {
+      "epoch": 0.8741007194244604,
+      "grad_norm": 0.5690473240624642,
+      "learning_rate": 9.708314889972988e-06,
+      "loss": 0.7934916019439697,
+      "step": 243
+    },
+    {
+      "epoch": 0.8776978417266187,
+      "grad_norm": 0.5277029008553127,
+      "learning_rate": 9.705933111367314e-06,
+      "loss": 0.7867487668991089,
+      "step": 244
+    },
+    {
+      "epoch": 0.8812949640287769,
+      "grad_norm": 0.540333896423015,
+      "learning_rate": 9.703541942539697e-06,
+      "loss": 0.8125045299530029,
+      "step": 245
+    },
+    {
+      "epoch": 0.8848920863309353,
+      "grad_norm": 0.5250035223657058,
+      "learning_rate": 9.701141388261471e-06,
+      "loss": 0.8113247156143188,
+      "step": 246
+    },
+    {
+      "epoch": 0.8884892086330936,
+      "grad_norm": 0.5215338961621239,
+      "learning_rate": 9.698731453322706e-06,
+      "loss": 0.8007864952087402,
+      "step": 247
+    },
+    {
+      "epoch": 0.8920863309352518,
+      "grad_norm": 0.5310159690708876,
+      "learning_rate": 9.69631214253219e-06,
+      "loss": 0.812530517578125,
+      "step": 248
+    },
+    {
+      "epoch": 0.89568345323741,
+      "grad_norm": 0.5340141074169955,
+      "learning_rate": 9.693883460717418e-06,
+      "loss": 0.7936503887176514,
+      "step": 249
+    },
+    {
+      "epoch": 0.8992805755395683,
+      "grad_norm": 0.5374886150627688,
+      "learning_rate": 9.691445412724576e-06,
+      "loss": 0.7722389101982117,
+      "step": 250
+    },
+    {
+      "epoch": 0.9028776978417267,
+      "grad_norm": 0.541501770110855,
+      "learning_rate": 9.688998003418553e-06,
+      "loss": 0.8209556937217712,
+      "step": 251
+    },
+    {
+      "epoch": 0.9064748201438849,
+      "grad_norm": 0.5240271320265909,
+      "learning_rate": 9.68654123768291e-06,
+      "loss": 0.7845938801765442,
+      "step": 252
+    },
+    {
+      "epoch": 0.9100719424460432,
+      "grad_norm": 0.5224977136818818,
+      "learning_rate": 9.684075120419876e-06,
+      "loss": 0.7874033451080322,
+      "step": 253
+    },
+    {
+      "epoch": 0.9136690647482014,
+      "grad_norm": 0.5310749021933068,
+      "learning_rate": 9.681599656550346e-06,
+      "loss": 0.8098745346069336,
+      "step": 254
+    },
+    {
+      "epoch": 0.9172661870503597,
+      "grad_norm": 0.5611962784041806,
+      "learning_rate": 9.679114851013862e-06,
+      "loss": 0.7997758388519287,
+      "step": 255
+    },
+    {
+      "epoch": 0.920863309352518,
+      "grad_norm": 0.552250838905918,
+      "learning_rate": 9.676620708768608e-06,
+      "loss": 0.7942588329315186,
+      "step": 256
+    },
+    {
+      "epoch": 0.9244604316546763,
+      "grad_norm": 0.4881842864225739,
+      "learning_rate": 9.674117234791391e-06,
+      "loss": 0.8118790984153748,
+      "step": 257
+    },
+    {
+      "epoch": 0.9280575539568345,
+      "grad_norm": 0.512794242246418,
+      "learning_rate": 9.67160443407765e-06,
+      "loss": 0.7766081094741821,
+      "step": 258
+    },
+    {
+      "epoch": 0.9316546762589928,
+      "grad_norm": 0.5200931171118809,
+      "learning_rate": 9.669082311641427e-06,
+      "loss": 0.7934583425521851,
+      "step": 259
+    },
+    {
+      "epoch": 0.935251798561151,
+      "grad_norm": 0.5033254279172306,
+      "learning_rate": 9.666550872515367e-06,
+      "loss": 0.8074806928634644,
+      "step": 260
+    },
+    {
+      "epoch": 0.9388489208633094,
+      "grad_norm": 0.5524080696367791,
+      "learning_rate": 9.664010121750708e-06,
+      "loss": 0.812640368938446,
+      "step": 261
+    },
+    {
+      "epoch": 0.9424460431654677,
+      "grad_norm": 0.5522305688677573,
+      "learning_rate": 9.661460064417263e-06,
+      "loss": 0.8016558885574341,
+      "step": 262
+    },
+    {
+      "epoch": 0.9460431654676259,
+      "grad_norm": 0.5023947284632377,
+      "learning_rate": 9.658900705603417e-06,
+      "loss": 0.7996955513954163,
+      "step": 263
+    },
+    {
+      "epoch": 0.9496402877697842,
+      "grad_norm": 0.5316226094004911,
+      "learning_rate": 9.656332050416118e-06,
+      "loss": 0.8242018818855286,
+      "step": 264
+    },
+    {
+      "epoch": 0.9532374100719424,
+      "grad_norm": 0.5432189986932476,
+      "learning_rate": 9.653754103980861e-06,
+      "loss": 0.80191969871521,
+      "step": 265
+    },
+    {
+      "epoch": 0.9568345323741008,
+      "grad_norm": 0.5391216754238493,
+      "learning_rate": 9.651166871441684e-06,
+      "loss": 0.8091660737991333,
+      "step": 266
+    },
+    {
+      "epoch": 0.960431654676259,
+      "grad_norm": 0.5053265004385964,
+      "learning_rate": 9.64857035796115e-06,
+      "loss": 0.8128654360771179,
+      "step": 267
+    },
+    {
+      "epoch": 0.9640287769784173,
+      "grad_norm": 0.5474674002836111,
+      "learning_rate": 9.645964568720345e-06,
+      "loss": 0.8065285682678223,
+      "step": 268
+    },
+    {
+      "epoch": 0.9676258992805755,
+      "grad_norm": 0.5546363834942746,
+      "learning_rate": 9.643349508918864e-06,
+      "loss": 0.8089965581893921,
+      "step": 269
+    },
+    {
+      "epoch": 0.9712230215827338,
+      "grad_norm": 0.5235854155432896,
+      "learning_rate": 9.640725183774797e-06,
+      "loss": 0.7981188297271729,
+      "step": 270
+    },
+    {
+      "epoch": 0.9748201438848921,
+      "grad_norm": 0.5097113985031559,
+      "learning_rate": 9.638091598524723e-06,
+      "loss": 0.7769266366958618,
+      "step": 271
+    },
+    {
+      "epoch": 0.9784172661870504,
+      "grad_norm": 0.5407878481684559,
+      "learning_rate": 9.635448758423703e-06,
+      "loss": 0.7945355176925659,
+      "step": 272
+    },
+    {
+      "epoch": 0.9820143884892086,
+      "grad_norm": 0.5292720741687515,
+      "learning_rate": 9.632796668745259e-06,
+      "loss": 0.7883850932121277,
+      "step": 273
+    },
+    {
+      "epoch": 0.9856115107913669,
+      "grad_norm": 0.5447108091170464,
+      "learning_rate": 9.630135334781373e-06,
+      "loss": 0.8052664995193481,
+      "step": 274
+    },
+    {
+      "epoch": 0.9892086330935251,
+      "grad_norm": 0.5034511826735798,
+      "learning_rate": 9.627464761842475e-06,
+      "loss": 0.7900129556655884,
+      "step": 275
+    },
+    {
+      "epoch": 0.9928057553956835,
+      "grad_norm": 0.5324839406585946,
+      "learning_rate": 9.624784955257423e-06,
+      "loss": 0.813730001449585,
+      "step": 276
+    },
+    {
+      "epoch": 0.9964028776978417,
+      "grad_norm": 0.5860885589822529,
+      "learning_rate": 9.62209592037351e-06,
+      "loss": 0.7950912714004517,
+      "step": 277
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.7374379136630071,
+      "learning_rate": 9.619397662556434e-06,
+      "loss": 0.7994446158409119,
+      "step": 278
+    },
+    {
+      "epoch": 1.0035971223021583,
+      "grad_norm": 0.6776588146804744,
+      "learning_rate": 9.616690187190304e-06,
+      "loss": 0.7538776397705078,
+      "step": 279
+    },
+    {
+      "epoch": 1.0071942446043165,
+      "grad_norm": 0.5879953650944638,
+      "learning_rate": 9.613973499677613e-06,
+      "loss": 0.7604464888572693,
+      "step": 280
+    },
+    {
+      "epoch": 1.0107913669064748,
+      "grad_norm": 0.5939581051805073,
+      "learning_rate": 9.611247605439246e-06,
+      "loss": 0.7662951946258545,
+      "step": 281
+    },
+    {
+      "epoch": 1.014388489208633,
+      "grad_norm": 0.6157303231020771,
+      "learning_rate": 9.608512509914453e-06,
+      "loss": 0.7262904644012451,
+      "step": 282
+    },
+    {
+      "epoch": 1.0179856115107913,
+      "grad_norm": 0.5893791108583288,
+      "learning_rate": 9.605768218560844e-06,
+      "loss": 0.7444915175437927,
+      "step": 283
+    },
+    {
+      "epoch": 1.0215827338129497,
+      "grad_norm": 0.5873809938610391,
+      "learning_rate": 9.60301473685438e-06,
+      "loss": 0.7208749055862427,
+      "step": 284
+    },
+    {
+      "epoch": 1.025179856115108,
+      "grad_norm": 0.6108756409032599,
+      "learning_rate": 9.60025207028936e-06,
+      "loss": 0.7390925884246826,
+      "step": 285
+    },
+    {
+      "epoch": 1.0287769784172662,
+      "grad_norm": 0.6137049233321895,
+      "learning_rate": 9.597480224378412e-06,
+      "loss": 0.7449308633804321,
+      "step": 286
+    },
+    {
+      "epoch": 1.0323741007194245,
+      "grad_norm": 0.5613813157036074,
+      "learning_rate": 9.594699204652478e-06,
+      "loss": 0.7346478700637817,
+      "step": 287
+    },
+    {
+      "epoch": 1.0359712230215827,
+      "grad_norm": 0.5939425307104599,
+      "learning_rate": 9.591909016660806e-06,
+      "loss": 0.7614048719406128,
+      "step": 288
+    },
+    {
+      "epoch": 1.039568345323741,
+      "grad_norm": 0.5610799009674011,
+      "learning_rate": 9.589109665970941e-06,
+      "loss": 0.7561317682266235,
+      "step": 289
+    },
+    {
+      "epoch": 1.0431654676258992,
+      "grad_norm": 0.6115021111353046,
+      "learning_rate": 9.586301158168706e-06,
+      "loss": 0.738190233707428,
+      "step": 290
+    },
+    {
+      "epoch": 1.0467625899280575,
+      "grad_norm": 0.6234487490114379,
+      "learning_rate": 9.583483498858202e-06,
+      "loss": 0.7404968738555908,
+      "step": 291
+    },
+    {
+      "epoch": 1.0503597122302157,
+      "grad_norm": 0.5779335573279631,
+      "learning_rate": 9.580656693661787e-06,
+      "loss": 0.7546930909156799,
+      "step": 292
+    },
+    {
+      "epoch": 1.0539568345323742,
+      "grad_norm": 0.5563644509280081,
+      "learning_rate": 9.57782074822007e-06,
+      "loss": 0.7298511266708374,
+      "step": 293
+    },
+    {
+      "epoch": 1.0575539568345325,
+      "grad_norm": 0.5403469768323708,
+      "learning_rate": 9.574975668191899e-06,
+      "loss": 0.7354592084884644,
+      "step": 294
+    },
+    {
+      "epoch": 1.0611510791366907,
+      "grad_norm": 0.5712568902968939,
+      "learning_rate": 9.572121459254349e-06,
+      "loss": 0.7321658134460449,
+      "step": 295
+    },
+    {
+      "epoch": 1.064748201438849,
+      "grad_norm": 0.5822296978334045,
+      "learning_rate": 9.569258127102708e-06,
+      "loss": 0.7498965263366699,
+      "step": 296
+    },
+    {
+      "epoch": 1.0683453237410072,
+      "grad_norm": 0.5189814814100225,
+      "learning_rate": 9.56638567745047e-06,
+      "loss": 0.7498589158058167,
+      "step": 297
+    },
+    {
+      "epoch": 1.0719424460431655,
+      "grad_norm": 0.5809627924078522,
+      "learning_rate": 9.56350411602933e-06,
+      "loss": 0.717056393623352,
+      "step": 298
+    },
+    {
+      "epoch": 1.0755395683453237,
+      "grad_norm": 0.6022908196879311,
+      "learning_rate": 9.560613448589149e-06,
+      "loss": 0.7413934469223022,
+      "step": 299
+    },
+    {
+      "epoch": 1.079136690647482,
+      "grad_norm": 0.546409548066647,
+      "learning_rate": 9.55771368089797e-06,
+      "loss": 0.7399191856384277,
+      "step": 300
+    },
+    {
+      "epoch": 1.079136690647482,
+      "eval_loss": 0.7952051162719727,
+      "eval_runtime": 7.7175,
+      "eval_samples_per_second": 11.403,
+      "eval_steps_per_second": 0.389,
+      "eval_token_acc": 0.7517150105187311,
+      "step": 300
+    },
+    {
+      "epoch": 1.0827338129496402,
+      "grad_norm": 0.5970255750138046,
+      "learning_rate": 9.554804818741994e-06,
+      "loss": 0.7446647882461548,
+      "step": 301
+    },
+    {
+      "epoch": 1.0863309352517985,
+      "grad_norm": 0.5541334916798,
+      "learning_rate": 9.551886867925565e-06,
+      "loss": 0.7263193130493164,
+      "step": 302
+    },
+    {
+      "epoch": 1.0899280575539567,
+      "grad_norm": 0.6369779920740517,
+      "learning_rate": 9.548959834271158e-06,
+      "loss": 0.7311091423034668,
+      "step": 303
+    },
+    {
+      "epoch": 1.0935251798561152,
+      "grad_norm": 0.5385079083621497,
+      "learning_rate": 9.546023723619387e-06,
+      "loss": 0.7392459511756897,
+      "step": 304
+    },
+    {
+      "epoch": 1.0971223021582734,
+      "grad_norm": 0.5523777755477719,
+      "learning_rate": 9.543078541828962e-06,
+      "loss": 0.7363039255142212,
+      "step": 305
+    },
+    {
+      "epoch": 1.1007194244604317,
+      "grad_norm": 0.61129080591158,
+      "learning_rate": 9.540124294776704e-06,
+      "loss": 0.7463387846946716,
+      "step": 306
+    },
+    {
+      "epoch": 1.10431654676259,
+      "grad_norm": 0.5704927319446069,
+      "learning_rate": 9.53716098835752e-06,
+      "loss": 0.7332433462142944,
+      "step": 307
+    },
+    {
+      "epoch": 1.1079136690647482,
+      "grad_norm": 0.5647463745104909,
+      "learning_rate": 9.534188628484391e-06,
+      "loss": 0.7626124620437622,
+      "step": 308
+    },
+    {
+      "epoch": 1.1115107913669064,
+      "grad_norm": 0.5634890166624024,
+      "learning_rate": 9.531207221088368e-06,
+      "loss": 0.7278907895088196,
+      "step": 309
+    },
+    {
+      "epoch": 1.1151079136690647,
+      "grad_norm": 0.5684506907730069,
+      "learning_rate": 9.528216772118553e-06,
+      "loss": 0.7007638812065125,
+      "step": 310
+    },
+    {
+      "epoch": 1.118705035971223,
+      "grad_norm": 0.569487505785454,
+      "learning_rate": 9.525217287542088e-06,
+      "loss": 0.73263019323349,
+      "step": 311
+    },
+    {
+      "epoch": 1.1223021582733812,
+      "grad_norm": 0.531149414984143,
+      "learning_rate": 9.522208773344147e-06,
+      "loss": 0.726309061050415,
+      "step": 312
+    },
+    {
+      "epoch": 1.1258992805755397,
+      "grad_norm": 0.5730769528941625,
+      "learning_rate": 9.519191235527923e-06,
+      "loss": 0.740322470664978,
+      "step": 313
+    },
+    {
+      "epoch": 1.129496402877698,
+      "grad_norm": 0.7272415428533405,
+      "learning_rate": 9.516164680114612e-06,
+      "loss": 0.7354894280433655,
+      "step": 314
+    },
+    {
+      "epoch": 1.1330935251798562,
+      "grad_norm": 0.5406965572808433,
+      "learning_rate": 9.513129113143405e-06,
+      "loss": 0.7542805671691895,
+      "step": 315
+    },
+    {
+      "epoch": 1.1366906474820144,
+      "grad_norm": 0.5384238572794032,
+      "learning_rate": 9.510084540671471e-06,
+      "loss": 0.7344615459442139,
+      "step": 316
+    },
+    {
+      "epoch": 1.1402877697841727,
+      "grad_norm": 0.5692959473422989,
+      "learning_rate": 9.507030968773956e-06,
+      "loss": 0.7373325824737549,
+      "step": 317
+    },
+    {
+      "epoch": 1.143884892086331,
+      "grad_norm": 0.5354506821538485,
+      "learning_rate": 9.50396840354396e-06,
+      "loss": 0.7316826581954956,
+      "step": 318
+    },
+    {
+      "epoch": 1.1474820143884892,
+      "grad_norm": 0.5262961196184975,
+      "learning_rate": 9.500896851092522e-06,
+      "loss": 0.7321020364761353,
+      "step": 319
+    },
+    {
+      "epoch": 1.1510791366906474,
+      "grad_norm": 0.5206041056150756,
+      "learning_rate": 9.497816317548625e-06,
+      "loss": 0.7204076051712036,
+      "step": 320
+    },
+    {
+      "epoch": 1.1546762589928057,
+      "grad_norm": 0.5198471869309446,
+      "learning_rate": 9.494726809059164e-06,
+      "loss": 0.7209657430648804,
+      "step": 321
+    },
+    {
+      "epoch": 1.158273381294964,
+      "grad_norm": 0.539020727608202,
+      "learning_rate": 9.491628331788949e-06,
+      "loss": 0.7441370487213135,
+      "step": 322
+    },
+    {
+      "epoch": 1.1618705035971222,
+      "grad_norm": 0.53720351049394,
+      "learning_rate": 9.488520891920683e-06,
+      "loss": 0.7360929250717163,
+      "step": 323
+    },
+    {
+      "epoch": 1.1654676258992807,
+      "grad_norm": 0.5307529143698696,
+      "learning_rate": 9.485404495654953e-06,
+      "loss": 0.7320318222045898,
+      "step": 324
+    },
+    {
+      "epoch": 1.169064748201439,
+      "grad_norm": 0.5586408968190051,
+      "learning_rate": 9.482279149210217e-06,
+      "loss": 0.7476272583007812,
+      "step": 325
+    },
+    {
+      "epoch": 1.1726618705035972,
+      "grad_norm": 0.5466847193799189,
+      "learning_rate": 9.479144858822795e-06,
+      "loss": 0.7348689436912537,
+      "step": 326
+    },
+    {
+      "epoch": 1.1762589928057554,
+      "grad_norm": 0.51531963600872,
+      "learning_rate": 9.476001630746852e-06,
+      "loss": 0.7275116443634033,
+      "step": 327
+    },
+    {
+      "epoch": 1.1798561151079137,
+      "grad_norm": 0.5483263021729351,
+      "learning_rate": 9.472849471254386e-06,
+      "loss": 0.7249739766120911,
+      "step": 328
+    },
+    {
+      "epoch": 1.183453237410072,
+      "grad_norm": 0.5392307946676005,
+      "learning_rate": 9.469688386635218e-06,
+      "loss": 0.7299447059631348,
+      "step": 329
+    },
+    {
+      "epoch": 1.1870503597122302,
+      "grad_norm": 0.54831249804049,
+      "learning_rate": 9.466518383196981e-06,
+      "loss": 0.7359820604324341,
+      "step": 330
+    },
+    {
+      "epoch": 1.1906474820143884,
+      "grad_norm": 0.5238958451400837,
+      "learning_rate": 9.4633394672651e-06,
+      "loss": 0.7344633340835571,
+      "step": 331
+    },
+    {
+      "epoch": 1.1942446043165469,
+      "grad_norm": 0.5489024189498876,
+      "learning_rate": 9.460151645182784e-06,
+      "loss": 0.749465823173523,
+      "step": 332
+    },
+    {
+      "epoch": 1.1978417266187051,
+      "grad_norm": 0.5493718850190237,
+      "learning_rate": 9.456954923311017e-06,
+      "loss": 0.7115287780761719,
+      "step": 333
+    },
+    {
+      "epoch": 1.2014388489208634,
+      "grad_norm": 0.5381363309369281,
+      "learning_rate": 9.453749308028542e-06,
+      "loss": 0.7070856690406799,
+      "step": 334
+    },
+    {
+      "epoch": 1.2050359712230216,
+      "grad_norm": 0.6270654982496251,
+      "learning_rate": 9.450534805731843e-06,
+      "loss": 0.7269983291625977,
+      "step": 335
+    },
+    {
+      "epoch": 1.20863309352518,
+      "grad_norm": 0.5892033078226898,
+      "learning_rate": 9.447311422835141e-06,
+      "loss": 0.7400103807449341,
+      "step": 336
+    },
+    {
+      "epoch": 1.2122302158273381,
+      "grad_norm": 0.5118100086223286,
+      "learning_rate": 9.444079165770376e-06,
+      "loss": 0.7445212602615356,
+      "step": 337
+    },
+    {
+      "epoch": 1.2158273381294964,
+      "grad_norm": 0.6003610021860912,
+      "learning_rate": 9.440838040987194e-06,
+      "loss": 0.7496727705001831,
+      "step": 338
+    },
+    {
+      "epoch": 1.2194244604316546,
+      "grad_norm": 0.6014272158963897,
+      "learning_rate": 9.437588054952943e-06,
+      "loss": 0.7463854551315308,
+      "step": 339
+    },
+    {
+      "epoch": 1.223021582733813,
+      "grad_norm": 0.5622909415736467,
+      "learning_rate": 9.43432921415264e-06,
+      "loss": 0.735248327255249,
+      "step": 340
+    },
+    {
+      "epoch": 1.2266187050359711,
+      "grad_norm": 0.5330204356289486,
+      "learning_rate": 9.431061525088982e-06,
+      "loss": 0.7185267806053162,
+      "step": 341
+    },
+    {
+      "epoch": 1.2302158273381294,
+      "grad_norm": 0.5744485810048889,
+      "learning_rate": 9.427784994282318e-06,
+      "loss": 0.7163181900978088,
+      "step": 342
+    },
+    {
+      "epoch": 1.2338129496402876,
+      "grad_norm": 0.6027911826984552,
+      "learning_rate": 9.42449962827064e-06,
+      "loss": 0.748116672039032,
+      "step": 343
+    },
+    {
+      "epoch": 1.2374100719424461,
+      "grad_norm": 0.5121609550128294,
+      "learning_rate": 9.421205433609568e-06,
+      "loss": 0.7265034914016724,
+      "step": 344
+    },
+    {
+      "epoch": 1.2410071942446044,
+      "grad_norm": 0.5503931962639481,
+      "learning_rate": 9.41790241687234e-06,
+      "loss": 0.7382141351699829,
+      "step": 345
+    },
+    {
+      "epoch": 1.2446043165467626,
+      "grad_norm": 0.5310640074137198,
+      "learning_rate": 9.414590584649798e-06,
+      "loss": 0.7352370023727417,
+      "step": 346
+    },
+    {
+      "epoch": 1.2482014388489209,
+      "grad_norm": 0.5583912500613908,
+      "learning_rate": 9.411269943550376e-06,
+      "loss": 0.7385387420654297,
+      "step": 347
+    },
+    {
+      "epoch": 1.2517985611510791,
+      "grad_norm": 0.5386372550063233,
+      "learning_rate": 9.407940500200082e-06,
+      "loss": 0.7346241474151611,
+      "step": 348
+    },
+    {
+      "epoch": 1.2553956834532374,
+      "grad_norm": 0.5725708975617698,
+      "learning_rate": 9.404602261242488e-06,
+      "loss": 0.7429782152175903,
+      "step": 349
+    },
+    {
+      "epoch": 1.2589928057553956,
+      "grad_norm": 0.5071301743038213,
+      "learning_rate": 9.40125523333872e-06,
+      "loss": 0.7370673418045044,
+      "step": 350
+    },
+    {
+      "epoch": 1.2625899280575539,
+      "grad_norm": 0.5927884076224349,
+      "learning_rate": 9.397899423167438e-06,
+      "loss": 0.7469853162765503,
+      "step": 351
+    },
+    {
+      "epoch": 1.2661870503597124,
+      "grad_norm": 0.548296598654736,
+      "learning_rate": 9.39453483742483e-06,
+      "loss": 0.7487526535987854,
+      "step": 352
+    },
+    {
+      "epoch": 1.2697841726618706,
+      "grad_norm": 0.5097163098936354,
+      "learning_rate": 9.391161482824591e-06,
+      "loss": 0.7192838191986084,
+      "step": 353
+    },
+    {
+      "epoch": 1.2733812949640289,
+      "grad_norm": 0.5552628434665424,
+      "learning_rate": 9.387779366097915e-06,
+      "loss": 0.7359743118286133,
+      "step": 354
+    },
+    {
+      "epoch": 1.276978417266187,
+      "grad_norm": 0.5419514485186101,
+      "learning_rate": 9.38438849399348e-06,
+      "loss": 0.7057188749313354,
+      "step": 355
+    },
+    {
+      "epoch": 1.2805755395683454,
+      "grad_norm": 0.5621822924024781,
+      "learning_rate": 9.380988873277436e-06,
+      "loss": 0.7457002401351929,
+      "step": 356
+    },
+    {
+      "epoch": 1.2841726618705036,
+      "grad_norm": 0.523519240727049,
+      "learning_rate": 9.377580510733386e-06,
+      "loss": 0.7411536574363708,
+      "step": 357
+    },
+    {
+      "epoch": 1.2877697841726619,
+      "grad_norm": 0.5491854800918223,
+      "learning_rate": 9.37416341316238e-06,
+      "loss": 0.7158464193344116,
+      "step": 358
+    },
+    {
+      "epoch": 1.29136690647482,
+      "grad_norm": 0.666484368385392,
+      "learning_rate": 9.370737587382896e-06,
+      "loss": 0.7320539951324463,
+      "step": 359
+    },
+    {
+      "epoch": 1.2949640287769784,
+      "grad_norm": 0.5365848280521327,
+      "learning_rate": 9.367303040230828e-06,
+      "loss": 0.7353171110153198,
+      "step": 360
+    },
+    {
+      "epoch": 1.2985611510791366,
+      "grad_norm": 0.5617185149633399,
+      "learning_rate": 9.363859778559476e-06,
+      "loss": 0.7434992790222168,
+      "step": 361
+    },
+    {
+      "epoch": 1.3021582733812949,
+      "grad_norm": 0.5274326997351031,
+      "learning_rate": 9.360407809239527e-06,
+      "loss": 0.7784279584884644,
+      "step": 362
+    },
+    {
+      "epoch": 1.3057553956834531,
+      "grad_norm": 0.5215842842673377,
+      "learning_rate": 9.35694713915904e-06,
+      "loss": 0.7144894599914551,
+      "step": 363
+    },
+    {
+      "epoch": 1.3093525179856116,
+      "grad_norm": 0.5382925072863687,
+      "learning_rate": 9.35347777522344e-06,
+      "loss": 0.7470029592514038,
+      "step": 364
+    },
+    {
+      "epoch": 1.3129496402877698,
+      "grad_norm": 0.5362900701795816,
+      "learning_rate": 9.349999724355496e-06,
+      "loss": 0.7410938739776611,
+      "step": 365
+    },
+    {
+      "epoch": 1.316546762589928,
+      "grad_norm": 0.5791880585481317,
+      "learning_rate": 9.346512993495315e-06,
+      "loss": 0.740675151348114,
+      "step": 366
+    },
+    {
+      "epoch": 1.3201438848920863,
+      "grad_norm": 0.5328457297104043,
+      "learning_rate": 9.343017589600322e-06,
+      "loss": 0.7332954406738281,
+      "step": 367
+    },
+    {
+      "epoch": 1.3237410071942446,
+      "grad_norm": 0.5270359471036749,
+      "learning_rate": 9.33951351964525e-06,
+      "loss": 0.7480515241622925,
+      "step": 368
+    },
+    {
+      "epoch": 1.3273381294964028,
+      "grad_norm": 0.5541952398439698,
+      "learning_rate": 9.336000790622117e-06,
+      "loss": 0.7403005361557007,
+      "step": 369
+    },
+    {
+      "epoch": 1.330935251798561,
+      "grad_norm": 0.5248778268205648,
+      "learning_rate": 9.33247940954023e-06,
+      "loss": 0.7062946557998657,
+      "step": 370
+    },
+    {
+      "epoch": 1.3345323741007193,
+      "grad_norm": 0.5341904855512976,
+      "learning_rate": 9.32894938342615e-06,
+      "loss": 0.7464907169342041,
+      "step": 371
+    },
+    {
+      "epoch": 1.3381294964028778,
+      "grad_norm": 0.5245639415447767,
+      "learning_rate": 9.3254107193237e-06,
+      "loss": 0.7266008853912354,
+      "step": 372
+    },
+    {
+      "epoch": 1.341726618705036,
+      "grad_norm": 0.5220543007037212,
+      "learning_rate": 9.321863424293926e-06,
+      "loss": 0.7336535453796387,
+      "step": 373
+    },
+    {
+      "epoch": 1.3453237410071943,
+      "grad_norm": 0.5504538542775073,
+      "learning_rate": 9.318307505415108e-06,
+      "loss": 0.7529866695404053,
+      "step": 374
+    },
+    {
+      "epoch": 1.3489208633093526,
+      "grad_norm": 0.5138351742769647,
+      "learning_rate": 9.314742969782725e-06,
+      "loss": 0.766657829284668,
+      "step": 375
+    },
+    {
+      "epoch": 1.3525179856115108,
+      "grad_norm": 0.5615878942041659,
+      "learning_rate": 9.311169824509454e-06,
+      "loss": 0.7512322664260864,
+      "step": 376
+    },
+    {
+      "epoch": 1.356115107913669,
+      "grad_norm": 0.5276166142542134,
+      "learning_rate": 9.307588076725153e-06,
+      "loss": 0.7312484979629517,
+      "step": 377
+    },
+    {
+      "epoch": 1.3597122302158273,
+      "grad_norm": 0.5728321595101409,
+      "learning_rate": 9.303997733576846e-06,
+      "loss": 0.759013831615448,
+      "step": 378
+    },
+    {
+      "epoch": 1.3633093525179856,
+      "grad_norm": 0.5903085328443332,
+      "learning_rate": 9.3003988022287e-06,
+      "loss": 0.7296818494796753,
+      "step": 379
+    },
+    {
+      "epoch": 1.3669064748201438,
+      "grad_norm": 0.5359969819862859,
+      "learning_rate": 9.296791289862031e-06,
+      "loss": 0.7522664666175842,
+      "step": 380
+    },
+    {
+      "epoch": 1.370503597122302,
+      "grad_norm": 0.5588759989046174,
+      "learning_rate": 9.293175203675268e-06,
+      "loss": 0.726894736289978,
+      "step": 381
+    },
+    {
+      "epoch": 1.3741007194244603,
+      "grad_norm": 0.577877679626679,
+      "learning_rate": 9.289550550883952e-06,
+      "loss": 0.7532353401184082,
+      "step": 382
+    },
+    {
+      "epoch": 1.3776978417266186,
+      "grad_norm": 0.5697600504827269,
+      "learning_rate": 9.285917338720719e-06,
+      "loss": 0.7286536693572998,
+      "step": 383
+    },
+    {
+      "epoch": 1.381294964028777,
+      "grad_norm": 0.5365518724533357,
+      "learning_rate": 9.28227557443528e-06,
+      "loss": 0.7433767318725586,
+      "step": 384
+    },
+    {
+      "epoch": 1.3848920863309353,
+      "grad_norm": 0.5739665857401817,
+      "learning_rate": 9.278625265294417e-06,
+      "loss": 0.7427285313606262,
+      "step": 385
+    },
+    {
+      "epoch": 1.3884892086330936,
+      "grad_norm": 0.5891308782340648,
+      "learning_rate": 9.274966418581958e-06,
+      "loss": 0.755325436592102,
+      "step": 386
+    },
+    {
+      "epoch": 1.3920863309352518,
+      "grad_norm": 0.5221581630154059,
+      "learning_rate": 9.271299041598768e-06,
+      "loss": 0.7211991548538208,
+      "step": 387
+    },
+    {
+      "epoch": 1.39568345323741,
+      "grad_norm": 0.5591826784203816,
+      "learning_rate": 9.267623141662734e-06,
+      "loss": 0.742611289024353,
+      "step": 388
+    },
+    {
+      "epoch": 1.3992805755395683,
+      "grad_norm": 0.6031525882590488,
+      "learning_rate": 9.263938726108747e-06,
+      "loss": 0.7558526992797852,
+      "step": 389
+    },
+    {
+      "epoch": 1.4028776978417266,
+      "grad_norm": 0.5434043181383015,
+      "learning_rate": 9.260245802288694e-06,
+      "loss": 0.7408391833305359,
+      "step": 390
+    },
+    {
+      "epoch": 1.406474820143885,
+      "grad_norm": 0.5318047677201455,
+      "learning_rate": 9.256544377571437e-06,
+      "loss": 0.7365950345993042,
+      "step": 391
+    },
+    {
+      "epoch": 1.4100719424460433,
+      "grad_norm": 0.5998492183556869,
+      "learning_rate": 9.252834459342801e-06,
+      "loss": 0.7364234924316406,
+      "step": 392
+    },
+    {
+      "epoch": 1.4136690647482015,
+      "grad_norm": 0.5498503102293683,
+      "learning_rate": 9.24911605500556e-06,
+      "loss": 0.7535672187805176,
+      "step": 393
+    },
+    {
+      "epoch": 1.4172661870503598,
+      "grad_norm": 0.5503794189317693,
+      "learning_rate": 9.24538917197942e-06,
+      "loss": 0.7519177198410034,
+      "step": 394
+    },
+    {
+      "epoch": 1.420863309352518,
+      "grad_norm": 0.6162738039028262,
+      "learning_rate": 9.241653817701006e-06,
+      "loss": 0.7295320630073547,
+      "step": 395
+    },
+    {
+      "epoch": 1.4244604316546763,
+      "grad_norm": 0.5555563314128995,
+      "learning_rate": 9.237909999623847e-06,
+      "loss": 0.7336751222610474,
+      "step": 396
+    },
+    {
+      "epoch": 1.4280575539568345,
+      "grad_norm": 0.5259597744806965,
+      "learning_rate": 9.234157725218355e-06,
+      "loss": 0.7189645171165466,
+      "step": 397
+    },
+    {
+      "epoch": 1.4316546762589928,
+      "grad_norm": 0.5630698969687603,
+      "learning_rate": 9.230397001971829e-06,
+      "loss": 0.73908531665802,
+      "step": 398
+    },
+    {
+      "epoch": 1.435251798561151,
+      "grad_norm": 0.5276648720445877,
+      "learning_rate": 9.226627837388413e-06,
+      "loss": 0.7388709783554077,
+      "step": 399
+    },
+    {
+      "epoch": 1.4388489208633093,
+      "grad_norm": 0.5536664986659428,
+      "learning_rate": 9.222850238989104e-06,
+      "loss": 0.7502380013465881,
+      "step": 400
+    },
+    {
+      "epoch": 1.4388489208633093,
+      "eval_loss": 0.7909448742866516,
+      "eval_runtime": 7.5526,
+      "eval_samples_per_second": 11.652,
+      "eval_steps_per_second": 0.397,
+      "eval_token_acc": 0.7528686175941879,
+      "step": 400
+    },
+    {
+      "epoch": 1.4424460431654675,
+      "grad_norm": 0.5311490565632518,
+      "learning_rate": 9.219064214311719e-06,
+      "loss": 0.7382062673568726,
+      "step": 401
+    },
+    {
+      "epoch": 1.4460431654676258,
+      "grad_norm": 0.5370721218852971,
+      "learning_rate": 9.215269770910897e-06,
+      "loss": 0.7413097620010376,
+      "step": 402
+    },
+    {
+      "epoch": 1.449640287769784,
+      "grad_norm": 0.5465408436928648,
+      "learning_rate": 9.211466916358074e-06,
+      "loss": 0.7206458449363708,
+      "step": 403
+    },
+    {
+      "epoch": 1.4532374100719425,
+      "grad_norm": 0.5656672911765269,
+      "learning_rate": 9.207655658241469e-06,
+      "loss": 0.7349211573600769,
+      "step": 404
+    },
+    {
+      "epoch": 1.4568345323741008,
+      "grad_norm": 0.5335944635099276,
+      "learning_rate": 9.203836004166064e-06,
+      "loss": 0.7555948495864868,
+      "step": 405
+    },
+    {
+      "epoch": 1.460431654676259,
+      "grad_norm": 0.5549021913248626,
+      "learning_rate": 9.200007961753605e-06,
+      "loss": 0.7555477619171143,
+      "step": 406
+    },
+    {
+      "epoch": 1.4640287769784173,
+      "grad_norm": 0.614098701080317,
+      "learning_rate": 9.196171538642567e-06,
+      "loss": 0.7401256561279297,
+      "step": 407
+    },
+    {
+      "epoch": 1.4676258992805755,
+      "grad_norm": 0.5418307481265139,
+      "learning_rate": 9.192326742488153e-06,
+      "loss": 0.7573046088218689,
+      "step": 408
+    },
+    {
+      "epoch": 1.4712230215827338,
+      "grad_norm": 0.5956224565062808,
+      "learning_rate": 9.18847358096227e-06,
+      "loss": 0.7176140546798706,
+      "step": 409
+    },
+    {
+      "epoch": 1.474820143884892,
+      "grad_norm": 0.570779754662452,
+      "learning_rate": 9.184612061753518e-06,
+      "loss": 0.7395877242088318,
+      "step": 410
+    },
+    {
+      "epoch": 1.4784172661870505,
+      "grad_norm": 0.5388215450729462,
+      "learning_rate": 9.180742192567177e-06,
+      "loss": 0.7351829409599304,
+      "step": 411
+    },
+    {
+      "epoch": 1.4820143884892087,
+      "grad_norm": 0.5727588774150603,
+      "learning_rate": 9.176863981125185e-06,
+      "loss": 0.7495085597038269,
+      "step": 412
+    },
+    {
+      "epoch": 1.485611510791367,
+      "grad_norm": 0.5697617446092769,
+      "learning_rate": 9.172977435166132e-06,
+      "loss": 0.7626926898956299,
+      "step": 413
+    },
+    {
+      "epoch": 1.4892086330935252,
+      "grad_norm": 0.5675953440673872,
+      "learning_rate": 9.169082562445228e-06,
+      "loss": 0.7373538017272949,
+      "step": 414
+    },
+    {
+      "epoch": 1.4928057553956835,
+      "grad_norm": 0.5479509775604315,
+      "learning_rate": 9.165179370734308e-06,
+      "loss": 0.7284923195838928,
+      "step": 415
+    },
+    {
+      "epoch": 1.4964028776978417,
+      "grad_norm": 0.5814710062165914,
+      "learning_rate": 9.161267867821802e-06,
+      "loss": 0.7396392822265625,
+      "step": 416
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 0.5857709804866665,
+      "learning_rate": 9.157348061512728e-06,
+      "loss": 0.7299116849899292,
+      "step": 417
+    },
+    {
+      "epoch": 1.5035971223021583,
+      "grad_norm": 0.5637392300431893,
+      "learning_rate": 9.153419959628666e-06,
+      "loss": 0.7442845106124878,
+      "step": 418
+    },
+    {
+      "epoch": 1.5071942446043165,
+      "grad_norm": 0.5477432560373958,
+      "learning_rate": 9.149483570007755e-06,
+      "loss": 0.7370480895042419,
+      "step": 419
+    },
+    {
+      "epoch": 1.5107913669064748,
+      "grad_norm": 0.5761456074550545,
+      "learning_rate": 9.14553890050467e-06,
+      "loss": 0.7441853284835815,
+      "step": 420
+    },
+    {
+      "epoch": 1.514388489208633,
+      "grad_norm": 0.5357942602104266,
+      "learning_rate": 9.141585958990606e-06,
+      "loss": 0.6985485553741455,
+      "step": 421
+    },
+    {
+      "epoch": 1.5179856115107913,
+      "grad_norm": 0.5689984573039134,
+      "learning_rate": 9.137624753353268e-06,
+      "loss": 0.726935625076294,
+      "step": 422
+    },
+    {
+      "epoch": 1.5215827338129495,
+      "grad_norm": 0.5402664519943788,
+      "learning_rate": 9.133655291496842e-06,
+      "loss": 0.7158384919166565,
+      "step": 423
+    },
+    {
+      "epoch": 1.5251798561151078,
+      "grad_norm": 0.5393049283880241,
+      "learning_rate": 9.129677581342e-06,
+      "loss": 0.7156991958618164,
+      "step": 424
+    },
+    {
+      "epoch": 1.5287769784172662,
+      "grad_norm": 0.5417741371232614,
+      "learning_rate": 9.125691630825867e-06,
+      "loss": 0.7445224523544312,
+      "step": 425
+    },
+    {
+      "epoch": 1.5323741007194245,
+      "grad_norm": 0.537883938976634,
+      "learning_rate": 9.121697447902006e-06,
+      "loss": 0.734693169593811,
+      "step": 426
+    },
+    {
+      "epoch": 1.5359712230215827,
+      "grad_norm": 0.5276159043279869,
+      "learning_rate": 9.117695040540422e-06,
+      "loss": 0.7173066735267639,
+      "step": 427
+    },
+    {
+      "epoch": 1.539568345323741,
+      "grad_norm": 0.5675455525239093,
+      "learning_rate": 9.113684416727511e-06,
+      "loss": 0.7307520508766174,
+      "step": 428
+    },
+    {
+      "epoch": 1.5431654676258992,
+      "grad_norm": 0.4933193822920981,
+      "learning_rate": 9.10966558446608e-06,
+      "loss": 0.7307450175285339,
+      "step": 429
+    },
+    {
+      "epoch": 1.5467625899280577,
+      "grad_norm": 0.485898122283688,
+      "learning_rate": 9.10563855177531e-06,
+      "loss": 0.6748644113540649,
+      "step": 430
+    },
+    {
+      "epoch": 1.550359712230216,
+      "grad_norm": 0.5745646730466706,
+      "learning_rate": 9.101603326690742e-06,
+      "loss": 0.7469466924667358,
+      "step": 431
+    },
+    {
+      "epoch": 1.5539568345323742,
+      "grad_norm": 0.534327855901563,
+      "learning_rate": 9.097559917264268e-06,
+      "loss": 0.7524317502975464,
+      "step": 432
+    },
+    {
+      "epoch": 1.5575539568345325,
+      "grad_norm": 0.5412259006260789,
+      "learning_rate": 9.093508331564107e-06,
+      "loss": 0.7445288300514221,
+      "step": 433
+    },
+    {
+      "epoch": 1.5611510791366907,
+      "grad_norm": 0.5058320181775204,
+      "learning_rate": 9.089448577674799e-06,
+      "loss": 0.7400709390640259,
+      "step": 434
+    },
+    {
+      "epoch": 1.564748201438849,
+      "grad_norm": 0.5193454182825687,
+      "learning_rate": 9.08538066369718e-06,
+      "loss": 0.7423480749130249,
+      "step": 435
+    },
+    {
+      "epoch": 1.5683453237410072,
+      "grad_norm": 0.5207804383383916,
+      "learning_rate": 9.081304597748366e-06,
+      "loss": 0.7210599184036255,
+      "step": 436
+    },
+    {
+      "epoch": 1.5719424460431655,
+      "grad_norm": 0.5175530796698581,
+      "learning_rate": 9.077220387961742e-06,
+      "loss": 0.6935808658599854,
+      "step": 437
+    },
+    {
+      "epoch": 1.5755395683453237,
+      "grad_norm": 0.515905293322507,
+      "learning_rate": 9.073128042486945e-06,
+      "loss": 0.7411532402038574,
+      "step": 438
+    },
+    {
+      "epoch": 1.579136690647482,
+      "grad_norm": 0.5425545076894603,
+      "learning_rate": 9.06902756948984e-06,
+      "loss": 0.735691249370575,
+      "step": 439
+    },
+    {
+      "epoch": 1.5827338129496402,
+      "grad_norm": 0.599077809988531,
+      "learning_rate": 9.064918977152517e-06,
+      "loss": 0.7355791926383972,
+      "step": 440
+    },
+    {
+      "epoch": 1.5863309352517985,
+      "grad_norm": 0.6082129897806806,
+      "learning_rate": 9.060802273673259e-06,
+      "loss": 0.7199162244796753,
+      "step": 441
+    },
+    {
+      "epoch": 1.5899280575539567,
+      "grad_norm": 0.5103649601151766,
+      "learning_rate": 9.056677467266543e-06,
+      "loss": 0.7381383180618286,
+      "step": 442
+    },
+    {
+      "epoch": 1.593525179856115,
+      "grad_norm": 0.5627707086300117,
+      "learning_rate": 9.052544566163011e-06,
+      "loss": 0.7414311766624451,
+      "step": 443
+    },
+    {
+      "epoch": 1.5971223021582732,
+      "grad_norm": 0.5505980295381908,
+      "learning_rate": 9.048403578609454e-06,
+      "loss": 0.7309621572494507,
+      "step": 444
+    },
+    {
+      "epoch": 1.6007194244604317,
+      "grad_norm": 0.5456865652811913,
+      "learning_rate": 9.044254512868802e-06,
+      "loss": 0.7410526871681213,
+      "step": 445
+    },
+    {
+      "epoch": 1.60431654676259,
+      "grad_norm": 0.528770536586938,
+      "learning_rate": 9.040097377220103e-06,
+      "loss": 0.7214845418930054,
+      "step": 446
+    },
+    {
+      "epoch": 1.6079136690647482,
+      "grad_norm": 0.5102313482024892,
+      "learning_rate": 9.03593217995851e-06,
+      "loss": 0.7267544269561768,
+      "step": 447
+    },
+    {
+      "epoch": 1.6115107913669064,
+      "grad_norm": 0.5357842837733169,
+      "learning_rate": 9.031758929395259e-06,
+      "loss": 0.749438464641571,
+      "step": 448
+    },
+    {
+      "epoch": 1.6151079136690647,
+      "grad_norm": 0.5183117043379116,
+      "learning_rate": 9.027577633857653e-06,
+      "loss": 0.7412010431289673,
+      "step": 449
+    },
+    {
+      "epoch": 1.6187050359712232,
+      "grad_norm": 0.544498698607748,
+      "learning_rate": 9.02338830168906e-06,
+      "loss": 0.7321810722351074,
+      "step": 450
+    },
+    {
+      "epoch": 1.6223021582733814,
+      "grad_norm": 0.5591180344098369,
+      "learning_rate": 9.019190941248868e-06,
+      "loss": 0.7417110800743103,
+      "step": 451
+    },
+    {
+      "epoch": 1.6258992805755397,
+      "grad_norm": 0.5356391251910433,
+      "learning_rate": 9.014985560912499e-06,
+      "loss": 0.7386839985847473,
+      "step": 452
+    },
+    {
+      "epoch": 1.629496402877698,
+      "grad_norm": 0.5220619414688078,
+      "learning_rate": 9.010772169071366e-06,
+      "loss": 0.734441876411438,
+      "step": 453
+    },
+    {
+      "epoch": 1.6330935251798562,
+      "grad_norm": 0.5601301117263874,
+      "learning_rate": 9.00655077413288e-06,
+      "loss": 0.7375533580780029,
+      "step": 454
+    },
+    {
+      "epoch": 1.6366906474820144,
+      "grad_norm": 0.6271941176180186,
+      "learning_rate": 9.00232138452041e-06,
+      "loss": 0.7252668142318726,
+      "step": 455
+    },
+    {
+      "epoch": 1.6402877697841727,
+      "grad_norm": 0.500079390514242,
+      "learning_rate": 8.998084008673284e-06,
+      "loss": 0.751512885093689,
+      "step": 456
+    },
+    {
+      "epoch": 1.643884892086331,
+      "grad_norm": 0.5295943834414766,
+      "learning_rate": 8.993838655046768e-06,
+      "loss": 0.7241981029510498,
+      "step": 457
+    },
+    {
+      "epoch": 1.6474820143884892,
+      "grad_norm": 0.5128645411360977,
+      "learning_rate": 8.989585332112039e-06,
+      "loss": 0.7336317300796509,
+      "step": 458
+    },
+    {
+      "epoch": 1.6510791366906474,
+      "grad_norm": 0.520422971407478,
+      "learning_rate": 8.985324048356182e-06,
+      "loss": 0.7330301403999329,
+      "step": 459
+    },
+    {
+      "epoch": 1.6546762589928057,
+      "grad_norm": 0.5486117483299725,
+      "learning_rate": 8.981054812282162e-06,
+      "loss": 0.7439688444137573,
+      "step": 460
+    },
+    {
+      "epoch": 1.658273381294964,
+      "grad_norm": 0.5474240792690704,
+      "learning_rate": 8.976777632408823e-06,
+      "loss": 0.7569353580474854,
+      "step": 461
+    },
+    {
+      "epoch": 1.6618705035971222,
+      "grad_norm": 0.5537282539324976,
+      "learning_rate": 8.972492517270845e-06,
+      "loss": 0.7264491319656372,
+      "step": 462
+    },
+    {
+      "epoch": 1.6654676258992804,
+      "grad_norm": 0.5210887026164476,
+      "learning_rate": 8.968199475418753e-06,
+      "loss": 0.7550468444824219,
+      "step": 463
+    },
+    {
+      "epoch": 1.6690647482014387,
+      "grad_norm": 0.5227914073582502,
+      "learning_rate": 8.963898515418885e-06,
+      "loss": 0.7275254130363464,
+      "step": 464
+    },
+    {
+      "epoch": 1.6726618705035972,
+      "grad_norm": 0.547248636298651,
+      "learning_rate": 8.959589645853379e-06,
+      "loss": 0.7321031093597412,
+      "step": 465
+    },
+    {
+      "epoch": 1.6762589928057554,
+      "grad_norm": 0.5635636922781485,
+      "learning_rate": 8.955272875320159e-06,
+      "loss": 0.7457851767539978,
+      "step": 466
+    },
+    {
+      "epoch": 1.6798561151079137,
+      "grad_norm": 0.5260209843292031,
+      "learning_rate": 8.950948212432904e-06,
+      "loss": 0.7447621822357178,
+      "step": 467
+    },
+    {
+      "epoch": 1.683453237410072,
+      "grad_norm": 0.5809358369922651,
+      "learning_rate": 8.946615665821059e-06,
+      "loss": 0.7280055284500122,
+      "step": 468
+    },
+    {
+      "epoch": 1.6870503597122302,
+      "grad_norm": 0.5551594345768393,
+      "learning_rate": 8.942275244129784e-06,
+      "loss": 0.7527322769165039,
+      "step": 469
+    },
+    {
+      "epoch": 1.6906474820143886,
+      "grad_norm": 0.5617975071761427,
+      "learning_rate": 8.937926956019957e-06,
+      "loss": 0.7513195276260376,
+      "step": 470
+    },
+    {
+      "epoch": 1.6942446043165469,
+      "grad_norm": 0.5112275243890037,
+      "learning_rate": 8.933570810168164e-06,
+      "loss": 0.6966029405593872,
+      "step": 471
+    },
+    {
+      "epoch": 1.6978417266187051,
+      "grad_norm": 0.5375351381393408,
+      "learning_rate": 8.929206815266653e-06,
+      "loss": 0.73940110206604,
+      "step": 472
+    },
+    {
+      "epoch": 1.7014388489208634,
+      "grad_norm": 0.554451465074595,
+      "learning_rate": 8.924834980023345e-06,
+      "loss": 0.7569516897201538,
+      "step": 473
+    },
+    {
+      "epoch": 1.7050359712230216,
+      "grad_norm": 0.5383238373862146,
+      "learning_rate": 8.920455313161801e-06,
+      "loss": 0.7401361465454102,
+      "step": 474
+    },
+    {
+      "epoch": 1.70863309352518,
+      "grad_norm": 0.5104340245269192,
+      "learning_rate": 8.916067823421217e-06,
+      "loss": 0.7461481094360352,
+      "step": 475
+    },
+    {
+      "epoch": 1.7122302158273381,
+      "grad_norm": 0.53560752921969,
+      "learning_rate": 8.911672519556386e-06,
+      "loss": 0.7368592619895935,
+      "step": 476
+    },
+    {
+      "epoch": 1.7158273381294964,
+      "grad_norm": 0.536053839213111,
+      "learning_rate": 8.907269410337704e-06,
+      "loss": 0.7240039706230164,
+      "step": 477
+    },
+    {
+      "epoch": 1.7194244604316546,
+      "grad_norm": 0.5478965091566458,
+      "learning_rate": 8.902858504551136e-06,
+      "loss": 0.7458293437957764,
+      "step": 478
+    },
+    {
+      "epoch": 1.723021582733813,
+      "grad_norm": 0.5564378779508917,
+      "learning_rate": 8.898439810998207e-06,
+      "loss": 0.7189388871192932,
+      "step": 479
+    },
+    {
+      "epoch": 1.7266187050359711,
+      "grad_norm": 0.5174245932797992,
+      "learning_rate": 8.89401333849598e-06,
+      "loss": 0.735579252243042,
+      "step": 480
+    },
+    {
+      "epoch": 1.7302158273381294,
+      "grad_norm": 0.5695227424690098,
+      "learning_rate": 8.889579095877045e-06,
+      "loss": 0.7450761795043945,
+      "step": 481
+    },
+    {
+      "epoch": 1.7338129496402876,
+      "grad_norm": 0.5787345149604068,
+      "learning_rate": 8.885137091989488e-06,
+      "loss": 0.7268454432487488,
+      "step": 482
+    },
+    {
+      "epoch": 1.737410071942446,
+      "grad_norm": 0.5394179165975326,
+      "learning_rate": 8.880687335696888e-06,
+      "loss": 0.7514413595199585,
+      "step": 483
+    },
+    {
+      "epoch": 1.7410071942446042,
+      "grad_norm": 0.5303547723039538,
+      "learning_rate": 8.87622983587829e-06,
+      "loss": 0.7547081708908081,
+      "step": 484
+    },
+    {
+      "epoch": 1.7446043165467626,
+      "grad_norm": 0.514637981124783,
+      "learning_rate": 8.871764601428197e-06,
+      "loss": 0.734933078289032,
+      "step": 485
+    },
+    {
+      "epoch": 1.7482014388489209,
+      "grad_norm": 0.5586343580104267,
+      "learning_rate": 8.867291641256534e-06,
+      "loss": 0.7437984347343445,
+      "step": 486
+    },
+    {
+      "epoch": 1.7517985611510791,
+      "grad_norm": 0.5998511954854915,
+      "learning_rate": 8.862810964288653e-06,
+      "loss": 0.7622708082199097,
+      "step": 487
+    },
+    {
+      "epoch": 1.7553956834532374,
+      "grad_norm": 0.49612184572197465,
+      "learning_rate": 8.8583225794653e-06,
+      "loss": 0.723395586013794,
+      "step": 488
+    },
+    {
+      "epoch": 1.7589928057553958,
+      "grad_norm": 0.528640028638296,
+      "learning_rate": 8.853826495742598e-06,
+      "loss": 0.7272807359695435,
+      "step": 489
+    },
+    {
+      "epoch": 1.762589928057554,
+      "grad_norm": 0.524042782338471,
+      "learning_rate": 8.849322722092037e-06,
+      "loss": 0.7300413846969604,
+      "step": 490
+    },
+    {
+      "epoch": 1.7661870503597124,
+      "grad_norm": 0.5318202821857035,
+      "learning_rate": 8.84481126750045e-06,
+      "loss": 0.7286820411682129,
+      "step": 491
+    },
+    {
+      "epoch": 1.7697841726618706,
+      "grad_norm": 0.5651665747593855,
+      "learning_rate": 8.840292140969995e-06,
+      "loss": 0.7338886857032776,
+      "step": 492
+    },
+    {
+      "epoch": 1.7733812949640289,
+      "grad_norm": 0.5205303749173067,
+      "learning_rate": 8.835765351518141e-06,
+      "loss": 0.7232467532157898,
+      "step": 493
+    },
+    {
+      "epoch": 1.776978417266187,
+      "grad_norm": 0.5577630630822902,
+      "learning_rate": 8.83123090817765e-06,
+      "loss": 0.7327965497970581,
+      "step": 494
+    },
+    {
+      "epoch": 1.7805755395683454,
+      "grad_norm": 0.5453661781921937,
+      "learning_rate": 8.826688819996547e-06,
+      "loss": 0.7516458034515381,
+      "step": 495
+    },
+    {
+      "epoch": 1.7841726618705036,
+      "grad_norm": 0.5303678797009448,
+      "learning_rate": 8.82213909603812e-06,
+      "loss": 0.7440846562385559,
+      "step": 496
+    },
+    {
+      "epoch": 1.7877697841726619,
+      "grad_norm": 0.5342645964789604,
+      "learning_rate": 8.817581745380892e-06,
+      "loss": 0.7308763265609741,
+      "step": 497
+    },
+    {
+      "epoch": 1.79136690647482,
+      "grad_norm": 0.5495259593847395,
+      "learning_rate": 8.813016777118604e-06,
+      "loss": 0.751781702041626,
+      "step": 498
+    },
+    {
+      "epoch": 1.7949640287769784,
+      "grad_norm": 0.5663289088779772,
+      "learning_rate": 8.808444200360193e-06,
+      "loss": 0.7341732978820801,
+      "step": 499
+    },
+    {
+      "epoch": 1.7985611510791366,
+      "grad_norm": 0.5581078677091702,
+      "learning_rate": 8.803864024229786e-06,
+      "loss": 0.7606270909309387,
+      "step": 500
+    },
+    {
+      "epoch": 1.7985611510791366,
+      "eval_loss": 0.7864982485771179,
+      "eval_runtime": 7.7709,
+      "eval_samples_per_second": 11.324,
+      "eval_steps_per_second": 0.386,
+      "eval_token_acc": 0.753477154659882,
+      "step": 500
+    },
+    {
+      "epoch": 1.8021582733812949,
+      "grad_norm": 0.5531324796536057,
+      "learning_rate": 8.799276257866668e-06,
+      "loss": 0.7583173513412476,
+      "step": 501
+    },
+    {
+      "epoch": 1.8057553956834531,
+      "grad_norm": 0.5395790483494515,
+      "learning_rate": 8.79468091042527e-06,
+      "loss": 0.7361318469047546,
+      "step": 502
+    },
+    {
+      "epoch": 1.8093525179856114,
+      "grad_norm": 0.5517392006475015,
+      "learning_rate": 8.790077991075152e-06,
+      "loss": 0.722428560256958,
+      "step": 503
+    },
+    {
+      "epoch": 1.8129496402877698,
+      "grad_norm": 0.5929611636500313,
+      "learning_rate": 8.78546750900098e-06,
+      "loss": 0.730574369430542,
+      "step": 504
+    },
+    {
+      "epoch": 1.816546762589928,
+      "grad_norm": 0.5448615482316708,
+      "learning_rate": 8.780849473402516e-06,
+      "loss": 0.7352193593978882,
+      "step": 505
+    },
+    {
+      "epoch": 1.8201438848920863,
+      "grad_norm": 0.5253273778616157,
+      "learning_rate": 8.77622389349459e-06,
+      "loss": 0.7241075038909912,
+      "step": 506
+    },
+    {
+      "epoch": 1.8237410071942446,
+      "grad_norm": 0.5257045541245243,
+      "learning_rate": 8.77159077850709e-06,
+      "loss": 0.7681543827056885,
+      "step": 507
+    },
+    {
+      "epoch": 1.8273381294964028,
+      "grad_norm": 0.6002720552753918,
+      "learning_rate": 8.766950137684929e-06,
+      "loss": 0.7483848929405212,
+      "step": 508
+    },
+    {
+      "epoch": 1.8309352517985613,
+      "grad_norm": 0.5756325346480011,
+      "learning_rate": 8.762301980288052e-06,
+      "loss": 0.7401927709579468,
+      "step": 509
+    },
+    {
+      "epoch": 1.8345323741007196,
+      "grad_norm": 0.5405458899100752,
+      "learning_rate": 8.757646315591391e-06,
+      "loss": 0.7275959849357605,
+      "step": 510
+    },
+    {
+      "epoch": 1.8381294964028778,
+      "grad_norm": 0.5055345910881511,
+      "learning_rate": 8.752983152884865e-06,
+      "loss": 0.7402202486991882,
+      "step": 511
+    },
+    {
+      "epoch": 1.841726618705036,
+      "grad_norm": 0.5294594291530128,
+      "learning_rate": 8.748312501473351e-06,
+      "loss": 0.7236826419830322,
+      "step": 512
+    },
+    {
+      "epoch": 1.8453237410071943,
+      "grad_norm": 0.525511680520969,
+      "learning_rate": 8.743634370676668e-06,
+      "loss": 0.7717915773391724,
+      "step": 513
+    },
+    {
+      "epoch": 1.8489208633093526,
+      "grad_norm": 0.5238277780917114,
+      "learning_rate": 8.738948769829565e-06,
+      "loss": 0.7489827871322632,
+      "step": 514
+    },
+    {
+      "epoch": 1.8525179856115108,
+      "grad_norm": 0.541734141362648,
+      "learning_rate": 8.734255708281689e-06,
+      "loss": 0.7424919605255127,
+      "step": 515
+    },
+    {
+      "epoch": 1.856115107913669,
+      "grad_norm": 0.5469220011947715,
+      "learning_rate": 8.72955519539758e-06,
+      "loss": 0.7435829639434814,
+      "step": 516
+    },
+    {
+      "epoch": 1.8597122302158273,
+      "grad_norm": 0.5189466053231159,
+      "learning_rate": 8.724847240556646e-06,
+      "loss": 0.7290617227554321,
+      "step": 517
+    },
+    {
+      "epoch": 1.8633093525179856,
+      "grad_norm": 0.5480359007192052,
+      "learning_rate": 8.72013185315314e-06,
+      "loss": 0.7334051132202148,
+      "step": 518
+    },
+    {
+      "epoch": 1.8669064748201438,
+      "grad_norm": 0.5417268446962726,
+      "learning_rate": 8.71540904259615e-06,
+      "loss": 0.7550539374351501,
+      "step": 519
+    },
+    {
+      "epoch": 1.870503597122302,
+      "grad_norm": 0.5167623466664647,
+      "learning_rate": 8.710678818309576e-06,
+      "loss": 0.7526887059211731,
+      "step": 520
+    },
+    {
+      "epoch": 1.8741007194244603,
+      "grad_norm": 0.5225951181805961,
+      "learning_rate": 8.705941189732112e-06,
+      "loss": 0.7233267426490784,
+      "step": 521
+    },
+    {
+      "epoch": 1.8776978417266186,
+      "grad_norm": 0.5083543003377556,
+      "learning_rate": 8.701196166317221e-06,
+      "loss": 0.7155008912086487,
+      "step": 522
+    },
+    {
+      "epoch": 1.8812949640287768,
+      "grad_norm": 0.5184841428097599,
+      "learning_rate": 8.696443757533128e-06,
+      "loss": 0.7537486553192139,
+      "step": 523
+    },
+    {
+      "epoch": 1.8848920863309353,
+      "grad_norm": 0.5355064843039317,
+      "learning_rate": 8.691683972862792e-06,
+      "loss": 0.7314133644104004,
+      "step": 524
+    },
+    {
+      "epoch": 1.8884892086330936,
+      "grad_norm": 0.5274575652182439,
+      "learning_rate": 8.686916821803892e-06,
+      "loss": 0.7118090391159058,
+      "step": 525
+    },
+    {
+      "epoch": 1.8920863309352518,
+      "grad_norm": 0.543918653349304,
+      "learning_rate": 8.682142313868802e-06,
+      "loss": 0.7229712605476379,
+      "step": 526
+    },
+    {
+      "epoch": 1.89568345323741,
+      "grad_norm": 0.5411862870781207,
+      "learning_rate": 8.677360458584581e-06,
+      "loss": 0.7486415505409241,
+      "step": 527
+    },
+    {
+      "epoch": 1.8992805755395683,
+      "grad_norm": 0.513452919974691,
+      "learning_rate": 8.672571265492944e-06,
+      "loss": 0.7225480079650879,
+      "step": 528
+    },
+    {
+      "epoch": 1.9028776978417268,
+      "grad_norm": 0.5490215040066515,
+      "learning_rate": 8.667774744150251e-06,
+      "loss": 0.7476856708526611,
+      "step": 529
+    },
+    {
+      "epoch": 1.906474820143885,
+      "grad_norm": 0.5169106694290708,
+      "learning_rate": 8.66297090412748e-06,
+      "loss": 0.7361697554588318,
+      "step": 530
+    },
+    {
+      "epoch": 1.9100719424460433,
+      "grad_norm": 0.5268875864918143,
+      "learning_rate": 8.658159755010223e-06,
+      "loss": 0.7248069047927856,
+      "step": 531
+    },
+    {
+      "epoch": 1.9136690647482015,
+      "grad_norm": 0.5403639981809885,
+      "learning_rate": 8.653341306398645e-06,
+      "loss": 0.7208993434906006,
+      "step": 532
+    },
+    {
+      "epoch": 1.9172661870503598,
+      "grad_norm": 0.513303533346121,
+      "learning_rate": 8.648515567907485e-06,
+      "loss": 0.7428863644599915,
+      "step": 533
+    },
+    {
+      "epoch": 1.920863309352518,
+      "grad_norm": 0.5531793218051347,
+      "learning_rate": 8.643682549166021e-06,
+      "loss": 0.7477810382843018,
+      "step": 534
+    },
+    {
+      "epoch": 1.9244604316546763,
+      "grad_norm": 0.5286725859333752,
+      "learning_rate": 8.638842259818064e-06,
+      "loss": 0.7352632284164429,
+      "step": 535
+    },
+    {
+      "epoch": 1.9280575539568345,
+      "grad_norm": 0.528661274771883,
+      "learning_rate": 8.63399470952193e-06,
+      "loss": 0.7406404614448547,
+      "step": 536
+    },
+    {
+      "epoch": 1.9316546762589928,
+      "grad_norm": 0.5541977935906626,
+      "learning_rate": 8.629139907950424e-06,
+      "loss": 0.7357466220855713,
+      "step": 537
+    },
+    {
+      "epoch": 1.935251798561151,
+      "grad_norm": 0.5491288214982604,
+      "learning_rate": 8.62427786479082e-06,
+      "loss": 0.7417570948600769,
+      "step": 538
+    },
+    {
+      "epoch": 1.9388489208633093,
+      "grad_norm": 0.5348837842686455,
+      "learning_rate": 8.619408589744842e-06,
+      "loss": 0.7317093014717102,
+      "step": 539
+    },
+    {
+      "epoch": 1.9424460431654675,
+      "grad_norm": 0.540958863951335,
+      "learning_rate": 8.614532092528645e-06,
+      "loss": 0.7405022382736206,
+      "step": 540
+    },
+    {
+      "epoch": 1.9460431654676258,
+      "grad_norm": 0.5163812629918846,
+      "learning_rate": 8.609648382872795e-06,
+      "loss": 0.741442084312439,
+      "step": 541
+    },
+    {
+      "epoch": 1.949640287769784,
+      "grad_norm": 0.5442615603715865,
+      "learning_rate": 8.604757470522247e-06,
+      "loss": 0.763242244720459,
+      "step": 542
+    },
+    {
+      "epoch": 1.9532374100719423,
+      "grad_norm": 0.566199733420676,
+      "learning_rate": 8.599859365236334e-06,
+      "loss": 0.7346137762069702,
+      "step": 543
+    },
+    {
+      "epoch": 1.9568345323741008,
+      "grad_norm": 0.5354204680310805,
+      "learning_rate": 8.594954076788736e-06,
+      "loss": 0.7331790924072266,
+      "step": 544
+    },
+    {
+      "epoch": 1.960431654676259,
+      "grad_norm": 0.537266301641817,
+      "learning_rate": 8.590041614967471e-06,
+      "loss": 0.7450471520423889,
+      "step": 545
+    },
+    {
+      "epoch": 1.9640287769784173,
+      "grad_norm": 0.5595286110764061,
+      "learning_rate": 8.585121989574865e-06,
+      "loss": 0.71414715051651,
+      "step": 546
+    },
+    {
+      "epoch": 1.9676258992805755,
+      "grad_norm": 0.5434065388365801,
+      "learning_rate": 8.580195210427544e-06,
+      "loss": 0.7267420887947083,
+      "step": 547
+    },
+    {
+      "epoch": 1.9712230215827338,
+      "grad_norm": 0.5823483147181996,
+      "learning_rate": 8.575261287356407e-06,
+      "loss": 0.7253061532974243,
+      "step": 548
+    },
+    {
+      "epoch": 1.9748201438848922,
+      "grad_norm": 0.5663754055488593,
+      "learning_rate": 8.570320230206606e-06,
+      "loss": 0.7432679533958435,
+      "step": 549
+    },
+    {
+      "epoch": 1.9784172661870505,
+      "grad_norm": 0.5301593265031658,
+      "learning_rate": 8.565372048837529e-06,
+      "loss": 0.72602379322052,
+      "step": 550
+    },
+    {
+      "epoch": 1.9820143884892087,
+      "grad_norm": 0.5919650898951195,
+      "learning_rate": 8.56041675312278e-06,
+      "loss": 0.7479984760284424,
+      "step": 551
+    },
+    {
+      "epoch": 1.985611510791367,
+      "grad_norm": 0.5725881545065112,
+      "learning_rate": 8.555454352950161e-06,
+      "loss": 0.7247596979141235,
+      "step": 552
+    },
+    {
+      "epoch": 1.9892086330935252,
+      "grad_norm": 0.545104484483576,
+      "learning_rate": 8.550484858221648e-06,
+      "loss": 0.7440117001533508,
+      "step": 553
+    },
+    {
+      "epoch": 1.9928057553956835,
+      "grad_norm": 0.5046621526349823,
+      "learning_rate": 8.545508278853373e-06,
+      "loss": 0.7102154493331909,
+      "step": 554
+    },
+    {
+      "epoch": 1.9964028776978417,
+      "grad_norm": 0.5827148240841841,
+      "learning_rate": 8.54052462477561e-06,
+      "loss": 0.7293152809143066,
+      "step": 555
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6796164196944195,
+      "learning_rate": 8.535533905932739e-06,
+      "loss": 0.7514892816543579,
+      "step": 556
+    },
+    {
+      "epoch": 2.0035971223021583,
+      "grad_norm": 0.6397885740084429,
+      "learning_rate": 8.530536132283247e-06,
+      "loss": 0.671186089515686,
+      "step": 557
+    },
+    {
+      "epoch": 2.0071942446043165,
+      "grad_norm": 0.5966614415471234,
+      "learning_rate": 8.525531313799693e-06,
+      "loss": 0.6722039580345154,
+      "step": 558
+    },
+    {
+      "epoch": 2.0107913669064748,
+      "grad_norm": 0.5579282766077329,
+      "learning_rate": 8.520519460468697e-06,
+      "loss": 0.690311074256897,
+      "step": 559
+    },
+    {
+      "epoch": 2.014388489208633,
+      "grad_norm": 0.5433122510327065,
+      "learning_rate": 8.515500582290914e-06,
+      "loss": 0.6836049556732178,
+      "step": 560
+    },
+    {
+      "epoch": 2.0179856115107913,
+      "grad_norm": 0.5538529733134517,
+      "learning_rate": 8.510474689281016e-06,
+      "loss": 0.6595236659049988,
+      "step": 561
+    },
+    {
+      "epoch": 2.0215827338129495,
+      "grad_norm": 0.5422413975220873,
+      "learning_rate": 8.505441791467674e-06,
+      "loss": 0.6806621551513672,
+      "step": 562
+    },
+    {
+      "epoch": 2.0251798561151078,
+      "grad_norm": 0.5578089222740708,
+      "learning_rate": 8.500401898893536e-06,
+      "loss": 0.6593084335327148,
+      "step": 563
+    },
+    {
+      "epoch": 2.028776978417266,
+      "grad_norm": 0.565177209250898,
+      "learning_rate": 8.495355021615204e-06,
+      "loss": 0.6911220550537109,
+      "step": 564
+    },
+    {
+      "epoch": 2.0323741007194243,
+      "grad_norm": 0.5119808699742511,
+      "learning_rate": 8.490301169703224e-06,
+      "loss": 0.6811946630477905,
+      "step": 565
+    },
+    {
+      "epoch": 2.0359712230215825,
+      "grad_norm": 0.5731400065818258,
+      "learning_rate": 8.48524035324205e-06,
+      "loss": 0.6667191982269287,
+      "step": 566
+    },
+    {
+      "epoch": 2.039568345323741,
+      "grad_norm": 0.5998221151013243,
+      "learning_rate": 8.480172582330044e-06,
+      "loss": 0.6908724904060364,
+      "step": 567
+    },
+    {
+      "epoch": 2.0431654676258995,
+      "grad_norm": 0.5527881634506817,
+      "learning_rate": 8.475097867079437e-06,
+      "loss": 0.6831552982330322,
+      "step": 568
+    },
+    {
+      "epoch": 2.0467625899280577,
+      "grad_norm": 0.5385272898977095,
+      "learning_rate": 8.470016217616319e-06,
+      "loss": 0.6840323209762573,
+      "step": 569
+    },
+    {
+      "epoch": 2.050359712230216,
+      "grad_norm": 0.5758971391452382,
+      "learning_rate": 8.464927644080617e-06,
+      "loss": 0.6829456090927124,
+      "step": 570
+    },
+    {
+      "epoch": 2.053956834532374,
+      "grad_norm": 0.5722853086808806,
+      "learning_rate": 8.459832156626073e-06,
+      "loss": 0.6708418130874634,
+      "step": 571
+    },
+    {
+      "epoch": 2.0575539568345325,
+      "grad_norm": 0.5267935272022233,
+      "learning_rate": 8.454729765420228e-06,
+      "loss": 0.6527084112167358,
+      "step": 572
+    },
+    {
+      "epoch": 2.0611510791366907,
+      "grad_norm": 0.5433510959719299,
+      "learning_rate": 8.449620480644393e-06,
+      "loss": 0.6686661243438721,
+      "step": 573
+    },
+    {
+      "epoch": 2.064748201438849,
+      "grad_norm": 0.5333455600158562,
+      "learning_rate": 8.444504312493642e-06,
+      "loss": 0.6683956384658813,
+      "step": 574
+    },
+    {
+      "epoch": 2.068345323741007,
+      "grad_norm": 0.5559059316065588,
+      "learning_rate": 8.439381271176779e-06,
+      "loss": 0.6839472055435181,
+      "step": 575
+    },
+    {
+      "epoch": 2.0719424460431655,
+      "grad_norm": 0.5686589857247856,
+      "learning_rate": 8.434251366916323e-06,
+      "loss": 0.677751898765564,
+      "step": 576
+    },
+    {
+      "epoch": 2.0755395683453237,
+      "grad_norm": 0.581140831766113,
+      "learning_rate": 8.429114609948488e-06,
+      "loss": 0.6839741468429565,
+      "step": 577
+    },
+    {
+      "epoch": 2.079136690647482,
+      "grad_norm": 0.6030439117251434,
+      "learning_rate": 8.423971010523165e-06,
+      "loss": 0.6848390698432922,
+      "step": 578
+    },
+    {
+      "epoch": 2.08273381294964,
+      "grad_norm": 0.5462427914093523,
+      "learning_rate": 8.418820578903891e-06,
+      "loss": 0.6663479804992676,
+      "step": 579
+    },
+    {
+      "epoch": 2.0863309352517985,
+      "grad_norm": 0.5542619430609473,
+      "learning_rate": 8.413663325367845e-06,
+      "loss": 0.6420421004295349,
+      "step": 580
+    },
+    {
+      "epoch": 2.0899280575539567,
+      "grad_norm": 0.565911591941745,
+      "learning_rate": 8.40849926020581e-06,
+      "loss": 0.6627936959266663,
+      "step": 581
+    },
+    {
+      "epoch": 2.093525179856115,
+      "grad_norm": 0.5222447784122286,
+      "learning_rate": 8.403328393722169e-06,
+      "loss": 0.675050675868988,
+      "step": 582
+    },
+    {
+      "epoch": 2.097122302158273,
+      "grad_norm": 0.5564681058103657,
+      "learning_rate": 8.39815073623487e-06,
+      "loss": 0.6631220579147339,
+      "step": 583
+    },
+    {
+      "epoch": 2.1007194244604315,
+      "grad_norm": 0.5701117496605299,
+      "learning_rate": 8.392966298075413e-06,
+      "loss": 0.6832946538925171,
+      "step": 584
+    },
+    {
+      "epoch": 2.1043165467625897,
+      "grad_norm": 0.5567281046765565,
+      "learning_rate": 8.387775089588833e-06,
+      "loss": 0.6763100624084473,
+      "step": 585
+    },
+    {
+      "epoch": 2.1079136690647484,
+      "grad_norm": 0.5141405571972344,
+      "learning_rate": 8.382577121133664e-06,
+      "loss": 0.6828013062477112,
+      "step": 586
+    },
+    {
+      "epoch": 2.1115107913669067,
+      "grad_norm": 0.5690795804417714,
+      "learning_rate": 8.377372403081943e-06,
+      "loss": 0.6641282439231873,
+      "step": 587
+    },
+    {
+      "epoch": 2.115107913669065,
+      "grad_norm": 0.5877720850231689,
+      "learning_rate": 8.372160945819164e-06,
+      "loss": 0.6777174472808838,
+      "step": 588
+    },
+    {
+      "epoch": 2.118705035971223,
+      "grad_norm": 0.5971069404183997,
+      "learning_rate": 8.366942759744274e-06,
+      "loss": 0.6761478185653687,
+      "step": 589
+    },
+    {
+      "epoch": 2.1223021582733814,
+      "grad_norm": 0.5374508547318486,
+      "learning_rate": 8.361717855269643e-06,
+      "loss": 0.6704797744750977,
+      "step": 590
+    },
+    {
+      "epoch": 2.1258992805755397,
+      "grad_norm": 0.575014664172825,
+      "learning_rate": 8.356486242821048e-06,
+      "loss": 0.6633048057556152,
+      "step": 591
+    },
+    {
+      "epoch": 2.129496402877698,
+      "grad_norm": 0.5630765161146606,
+      "learning_rate": 8.351247932837655e-06,
+      "loss": 0.6555532217025757,
+      "step": 592
+    },
+    {
+      "epoch": 2.133093525179856,
+      "grad_norm": 0.5540501457675682,
+      "learning_rate": 8.346002935771988e-06,
+      "loss": 0.6675060987472534,
+      "step": 593
+    },
+    {
+      "epoch": 2.1366906474820144,
+      "grad_norm": 0.5771881310936134,
+      "learning_rate": 8.34075126208992e-06,
+      "loss": 0.6720588803291321,
+      "step": 594
+    },
+    {
+      "epoch": 2.1402877697841727,
+      "grad_norm": 0.5516169446446759,
+      "learning_rate": 8.335492922270642e-06,
+      "loss": 0.6864827871322632,
+      "step": 595
+    },
+    {
+      "epoch": 2.143884892086331,
+      "grad_norm": 0.5537872437146788,
+      "learning_rate": 8.330227926806652e-06,
+      "loss": 0.675635814666748,
+      "step": 596
+    },
+    {
+      "epoch": 2.147482014388489,
+      "grad_norm": 0.5399664246709232,
+      "learning_rate": 8.324956286203717e-06,
+      "loss": 0.6826367378234863,
+      "step": 597
+    },
+    {
+      "epoch": 2.1510791366906474,
+      "grad_norm": 0.5763658067111991,
+      "learning_rate": 8.319678010980883e-06,
+      "loss": 0.6901789903640747,
+      "step": 598
+    },
+    {
+      "epoch": 2.1546762589928057,
+      "grad_norm": 0.4974273454797143,
+      "learning_rate": 8.314393111670419e-06,
+      "loss": 0.6468210220336914,
+      "step": 599
+    },
+    {
+      "epoch": 2.158273381294964,
+      "grad_norm": 0.5859989003860766,
+      "learning_rate": 8.309101598817812e-06,
+      "loss": 0.6841794848442078,
+      "step": 600
+    },
+    {
+      "epoch": 2.158273381294964,
+      "eval_loss": 0.7964679598808289,
+      "eval_runtime": 7.6714,
+      "eval_samples_per_second": 11.471,
+      "eval_steps_per_second": 0.391,
+      "eval_token_acc": 0.7527902171133316,
+      "step": 600
+    },
+    {
+      "epoch": 2.161870503597122,
+      "grad_norm": 0.5935282145282037,
+      "learning_rate": 8.303803482981758e-06,
+      "loss": 0.6777781248092651,
+      "step": 601
+    },
+    {
+      "epoch": 2.1654676258992804,
+      "grad_norm": 0.5646029208074659,
+      "learning_rate": 8.298498774734114e-06,
+      "loss": 0.6669676303863525,
+      "step": 602
+    },
+    {
+      "epoch": 2.1690647482014387,
+      "grad_norm": 0.5488370688552963,
+      "learning_rate": 8.293187484659904e-06,
+      "loss": 0.6794064044952393,
+      "step": 603
+    },
+    {
+      "epoch": 2.172661870503597,
+      "grad_norm": 0.5567754831392528,
+      "learning_rate": 8.287869623357275e-06,
+      "loss": 0.6805890798568726,
+      "step": 604
+    },
+    {
+      "epoch": 2.176258992805755,
+      "grad_norm": 0.5112276480668334,
+      "learning_rate": 8.282545201437493e-06,
+      "loss": 0.6340111494064331,
+      "step": 605
+    },
+    {
+      "epoch": 2.1798561151079134,
+      "grad_norm": 0.5892481231313883,
+      "learning_rate": 8.277214229524913e-06,
+      "loss": 0.6796945333480835,
+      "step": 606
+    },
+    {
+      "epoch": 2.183453237410072,
+      "grad_norm": 0.5771549281819284,
+      "learning_rate": 8.271876718256958e-06,
+      "loss": 0.6738216280937195,
+      "step": 607
+    },
+    {
+      "epoch": 2.1870503597122304,
+      "grad_norm": 0.5416360126424467,
+      "learning_rate": 8.266532678284103e-06,
+      "loss": 0.6745922565460205,
+      "step": 608
+    },
+    {
+      "epoch": 2.1906474820143886,
+      "grad_norm": 0.5232225408923397,
+      "learning_rate": 8.261182120269851e-06,
+      "loss": 0.6739056706428528,
+      "step": 609
+    },
+    {
+      "epoch": 2.194244604316547,
+      "grad_norm": 0.5868675716566435,
+      "learning_rate": 8.255825054890705e-06,
+      "loss": 0.6711397171020508,
+      "step": 610
+    },
+    {
+      "epoch": 2.197841726618705,
+      "grad_norm": 0.5539408109487034,
+      "learning_rate": 8.250461492836158e-06,
+      "loss": 0.6721957921981812,
+      "step": 611
+    },
+    {
+      "epoch": 2.2014388489208634,
+      "grad_norm": 0.5576072951558111,
+      "learning_rate": 8.245091444808663e-06,
+      "loss": 0.6724331378936768,
+      "step": 612
+    },
+    {
+      "epoch": 2.2050359712230216,
+      "grad_norm": 0.5583937132116154,
+      "learning_rate": 8.239714921523623e-06,
+      "loss": 0.6844789981842041,
+      "step": 613
+    },
+    {
+      "epoch": 2.20863309352518,
+      "grad_norm": 0.5596163368588887,
+      "learning_rate": 8.234331933709353e-06,
+      "loss": 0.6690089702606201,
+      "step": 614
+    },
+    {
+      "epoch": 2.212230215827338,
+      "grad_norm": 0.5537700483756164,
+      "learning_rate": 8.228942492107065e-06,
+      "loss": 0.6785811185836792,
+      "step": 615
+    },
+    {
+      "epoch": 2.2158273381294964,
+      "grad_norm": 0.5596429509210736,
+      "learning_rate": 8.223546607470863e-06,
+      "loss": 0.6860724687576294,
+      "step": 616
+    },
+    {
+      "epoch": 2.2194244604316546,
+      "grad_norm": 0.5615860198473355,
+      "learning_rate": 8.218144290567694e-06,
+      "loss": 0.6716651916503906,
+      "step": 617
+    },
+    {
+      "epoch": 2.223021582733813,
+      "grad_norm": 0.5828702845645434,
+      "learning_rate": 8.212735552177347e-06,
+      "loss": 0.6850781440734863,
+      "step": 618
+    },
+    {
+      "epoch": 2.226618705035971,
+      "grad_norm": 0.5414999227771777,
+      "learning_rate": 8.207320403092419e-06,
+      "loss": 0.6482763290405273,
+      "step": 619
+    },
+    {
+      "epoch": 2.2302158273381294,
+      "grad_norm": 0.5932037908152304,
+      "learning_rate": 8.201898854118301e-06,
+      "loss": 0.6657392382621765,
+      "step": 620
+    },
+    {
+      "epoch": 2.2338129496402876,
+      "grad_norm": 0.5342906139121061,
+      "learning_rate": 8.196470916073161e-06,
+      "loss": 0.6627857685089111,
+      "step": 621
+    },
+    {
+      "epoch": 2.237410071942446,
+      "grad_norm": 0.6026520242202834,
+      "learning_rate": 8.191036599787908e-06,
+      "loss": 0.6858589053153992,
+      "step": 622
+    },
+    {
+      "epoch": 2.241007194244604,
+      "grad_norm": 0.5530261201191066,
+      "learning_rate": 8.185595916106175e-06,
+      "loss": 0.6939041018486023,
+      "step": 623
+    },
+    {
+      "epoch": 2.2446043165467624,
+      "grad_norm": 0.5223090539726519,
+      "learning_rate": 8.18014887588431e-06,
+      "loss": 0.6830934286117554,
+      "step": 624
+    },
+    {
+      "epoch": 2.2482014388489207,
+      "grad_norm": 0.5211039805430082,
+      "learning_rate": 8.174695489991343e-06,
+      "loss": 0.6655054092407227,
+      "step": 625
+    },
+    {
+      "epoch": 2.2517985611510793,
+      "grad_norm": 0.5394362084185865,
+      "learning_rate": 8.169235769308959e-06,
+      "loss": 0.6696172952651978,
+      "step": 626
+    },
+    {
+      "epoch": 2.2553956834532376,
+      "grad_norm": 0.5667420557087554,
+      "learning_rate": 8.163769724731491e-06,
+      "loss": 0.6671138405799866,
+      "step": 627
+    },
+    {
+      "epoch": 2.258992805755396,
+      "grad_norm": 0.545382220641246,
+      "learning_rate": 8.158297367165885e-06,
+      "loss": 0.6894287467002869,
+      "step": 628
+    },
+    {
+      "epoch": 2.262589928057554,
+      "grad_norm": 0.5417774154601794,
+      "learning_rate": 8.152818707531691e-06,
+      "loss": 0.6864540576934814,
+      "step": 629
+    },
+    {
+      "epoch": 2.2661870503597124,
+      "grad_norm": 0.5614615667910734,
+      "learning_rate": 8.147333756761027e-06,
+      "loss": 0.668978214263916,
+      "step": 630
+    },
+    {
+      "epoch": 2.2697841726618706,
+      "grad_norm": 0.5488039597992234,
+      "learning_rate": 8.141842525798567e-06,
+      "loss": 0.6848398447036743,
+      "step": 631
+    },
+    {
+      "epoch": 2.273381294964029,
+      "grad_norm": 0.5723936696701672,
+      "learning_rate": 8.13634502560152e-06,
+      "loss": 0.6667437553405762,
+      "step": 632
+    },
+    {
+      "epoch": 2.276978417266187,
+      "grad_norm": 0.5733329856292084,
+      "learning_rate": 8.130841267139598e-06,
+      "loss": 0.6772172451019287,
+      "step": 633
+    },
+    {
+      "epoch": 2.2805755395683454,
+      "grad_norm": 0.5176511941575904,
+      "learning_rate": 8.125331261395004e-06,
+      "loss": 0.6722500324249268,
+      "step": 634
+    },
+    {
+      "epoch": 2.2841726618705036,
+      "grad_norm": 0.5418827338877102,
+      "learning_rate": 8.11981501936241e-06,
+      "loss": 0.6875349283218384,
+      "step": 635
+    },
+    {
+      "epoch": 2.287769784172662,
+      "grad_norm": 0.5714231681415783,
+      "learning_rate": 8.114292552048925e-06,
+      "loss": 0.6613736152648926,
+      "step": 636
+    },
+    {
+      "epoch": 2.29136690647482,
+      "grad_norm": 0.5938180697661647,
+      "learning_rate": 8.108763870474088e-06,
+      "loss": 0.6840851306915283,
+      "step": 637
+    },
+    {
+      "epoch": 2.2949640287769784,
+      "grad_norm": 0.5260002035466227,
+      "learning_rate": 8.10322898566983e-06,
+      "loss": 0.6638031601905823,
+      "step": 638
+    },
+    {
+      "epoch": 2.2985611510791366,
+      "grad_norm": 0.591475324476344,
+      "learning_rate": 8.097687908680467e-06,
+      "loss": 0.6879281401634216,
+      "step": 639
+    },
+    {
+      "epoch": 2.302158273381295,
+      "grad_norm": 0.5331036790988722,
+      "learning_rate": 8.092140650562665e-06,
+      "loss": 0.6734105348587036,
+      "step": 640
+    },
+    {
+      "epoch": 2.305755395683453,
+      "grad_norm": 0.5172033041212313,
+      "learning_rate": 8.086587222385428e-06,
+      "loss": 0.6683194041252136,
+      "step": 641
+    },
+    {
+      "epoch": 2.3093525179856114,
+      "grad_norm": 0.5473876959056185,
+      "learning_rate": 8.08102763523007e-06,
+      "loss": 0.6741974353790283,
+      "step": 642
+    },
+    {
+      "epoch": 2.3129496402877696,
+      "grad_norm": 0.5324084730756219,
+      "learning_rate": 8.075461900190193e-06,
+      "loss": 0.6834567785263062,
+      "step": 643
+    },
+    {
+      "epoch": 2.316546762589928,
+      "grad_norm": 0.5181323608798409,
+      "learning_rate": 8.069890028371672e-06,
+      "loss": 0.6760212182998657,
+      "step": 644
+    },
+    {
+      "epoch": 2.3201438848920866,
+      "grad_norm": 0.5435693010138409,
+      "learning_rate": 8.064312030892618e-06,
+      "loss": 0.6685901880264282,
+      "step": 645
+    },
+    {
+      "epoch": 2.3237410071942444,
+      "grad_norm": 0.545294047023693,
+      "learning_rate": 8.058727918883376e-06,
+      "loss": 0.6555389165878296,
+      "step": 646
+    },
+    {
+      "epoch": 2.327338129496403,
+      "grad_norm": 0.5410247736328141,
+      "learning_rate": 8.053137703486482e-06,
+      "loss": 0.6900992393493652,
+      "step": 647
+    },
+    {
+      "epoch": 2.3309352517985613,
+      "grad_norm": 0.5466463022120158,
+      "learning_rate": 8.047541395856661e-06,
+      "loss": 0.6867014765739441,
+      "step": 648
+    },
+    {
+      "epoch": 2.3345323741007196,
+      "grad_norm": 0.550847927547804,
+      "learning_rate": 8.041939007160784e-06,
+      "loss": 0.666527509689331,
+      "step": 649
+    },
+    {
+      "epoch": 2.338129496402878,
+      "grad_norm": 0.5421989902212895,
+      "learning_rate": 8.036330548577866e-06,
+      "loss": 0.6644178628921509,
+      "step": 650
+    },
+    {
+      "epoch": 2.341726618705036,
+      "grad_norm": 0.5236348296663158,
+      "learning_rate": 8.030716031299022e-06,
+      "loss": 0.685157299041748,
+      "step": 651
+    },
+    {
+      "epoch": 2.3453237410071943,
+      "grad_norm": 0.5176525641905038,
+      "learning_rate": 8.025095466527468e-06,
+      "loss": 0.6704069375991821,
+      "step": 652
+    },
+    {
+      "epoch": 2.3489208633093526,
+      "grad_norm": 0.5892074323237977,
+      "learning_rate": 8.019468865478483e-06,
+      "loss": 0.6835877895355225,
+      "step": 653
+    },
+    {
+      "epoch": 2.352517985611511,
+      "grad_norm": 0.5652692671595871,
+      "learning_rate": 8.013836239379388e-06,
+      "loss": 0.6817514896392822,
+      "step": 654
+    },
+    {
+      "epoch": 2.356115107913669,
+      "grad_norm": 0.5620911489748651,
+      "learning_rate": 8.008197599469529e-06,
+      "loss": 0.6783486604690552,
+      "step": 655
+    },
+    {
+      "epoch": 2.3597122302158273,
+      "grad_norm": 0.5936476741299256,
+      "learning_rate": 8.002552957000254e-06,
+      "loss": 0.6681350469589233,
+      "step": 656
+    },
+    {
+      "epoch": 2.3633093525179856,
+      "grad_norm": 0.5443626063476964,
+      "learning_rate": 7.996902323234883e-06,
+      "loss": 0.6636416912078857,
+      "step": 657
+    },
+    {
+      "epoch": 2.366906474820144,
+      "grad_norm": 0.6165820617546384,
+      "learning_rate": 7.991245709448697e-06,
+      "loss": 0.6662098169326782,
+      "step": 658
+    },
+    {
+      "epoch": 2.370503597122302,
+      "grad_norm": 0.5954448294054072,
+      "learning_rate": 7.985583126928904e-06,
+      "loss": 0.671536386013031,
+      "step": 659
+    },
+    {
+      "epoch": 2.3741007194244603,
+      "grad_norm": 0.5699739721187903,
+      "learning_rate": 7.979914586974628e-06,
+      "loss": 0.678837776184082,
+      "step": 660
+    },
+    {
+      "epoch": 2.3776978417266186,
+      "grad_norm": 0.5714872752111637,
+      "learning_rate": 7.974240100896874e-06,
+      "loss": 0.6746315956115723,
+      "step": 661
+    },
+    {
+      "epoch": 2.381294964028777,
+      "grad_norm": 0.8033103327542886,
+      "learning_rate": 7.968559680018518e-06,
+      "loss": 0.6854047775268555,
+      "step": 662
+    },
+    {
+      "epoch": 2.384892086330935,
+      "grad_norm": 0.5422166443828278,
+      "learning_rate": 7.962873335674272e-06,
+      "loss": 0.6693898439407349,
+      "step": 663
+    },
+    {
+      "epoch": 2.3884892086330938,
+      "grad_norm": 0.5495434573450082,
+      "learning_rate": 7.957181079210676e-06,
+      "loss": 0.656349778175354,
+      "step": 664
+    },
+    {
+      "epoch": 2.3920863309352516,
+      "grad_norm": 0.5929225873029843,
+      "learning_rate": 7.951482921986058e-06,
+      "loss": 0.6681583523750305,
+      "step": 665
+    },
+    {
+      "epoch": 2.3956834532374103,
+      "grad_norm": 0.530361021028339,
+      "learning_rate": 7.945778875370527e-06,
+      "loss": 0.6552359461784363,
+      "step": 666
+    },
+    {
+      "epoch": 2.3992805755395685,
+      "grad_norm": 0.5756796589055374,
+      "learning_rate": 7.940068950745941e-06,
+      "loss": 0.6794135570526123,
+      "step": 667
+    },
+    {
+      "epoch": 2.402877697841727,
+      "grad_norm": 0.564306119284853,
+      "learning_rate": 7.934353159505885e-06,
+      "loss": 0.6614569425582886,
+      "step": 668
+    },
+    {
+      "epoch": 2.406474820143885,
+      "grad_norm": 0.551021041442497,
+      "learning_rate": 7.928631513055656e-06,
+      "loss": 0.6986576318740845,
+      "step": 669
+    },
+    {
+      "epoch": 2.4100719424460433,
+      "grad_norm": 0.5624604572287012,
+      "learning_rate": 7.922904022812228e-06,
+      "loss": 0.6947466135025024,
+      "step": 670
+    },
+    {
+      "epoch": 2.4136690647482015,
+      "grad_norm": 0.5443760240676394,
+      "learning_rate": 7.91717070020424e-06,
+      "loss": 0.6702431440353394,
+      "step": 671
+    },
+    {
+      "epoch": 2.41726618705036,
+      "grad_norm": 0.556358152638375,
+      "learning_rate": 7.911431556671967e-06,
+      "loss": 0.6885404586791992,
+      "step": 672
+    },
+    {
+      "epoch": 2.420863309352518,
+      "grad_norm": 0.5278486204625064,
+      "learning_rate": 7.905686603667301e-06,
+      "loss": 0.7020502090454102,
+      "step": 673
+    },
+    {
+      "epoch": 2.4244604316546763,
+      "grad_norm": 0.6140900556871232,
+      "learning_rate": 7.899935852653723e-06,
+      "loss": 0.6749525666236877,
+      "step": 674
+    },
+    {
+      "epoch": 2.4280575539568345,
+      "grad_norm": 0.5175983027819278,
+      "learning_rate": 7.89417931510629e-06,
+      "loss": 0.6766486167907715,
+      "step": 675
+    },
+    {
+      "epoch": 2.431654676258993,
+      "grad_norm": 0.6029093317105096,
+      "learning_rate": 7.888417002511592e-06,
+      "loss": 0.6930079460144043,
+      "step": 676
+    },
+    {
+      "epoch": 2.435251798561151,
+      "grad_norm": 0.5675850568410075,
+      "learning_rate": 7.88264892636776e-06,
+      "loss": 0.6770628690719604,
+      "step": 677
+    },
+    {
+      "epoch": 2.4388489208633093,
+      "grad_norm": 0.5692480264597963,
+      "learning_rate": 7.876875098184411e-06,
+      "loss": 0.6888518333435059,
+      "step": 678
+    },
+    {
+      "epoch": 2.4424460431654675,
+      "grad_norm": 0.5297351357365194,
+      "learning_rate": 7.871095529482648e-06,
+      "loss": 0.6821136474609375,
+      "step": 679
+    },
+    {
+      "epoch": 2.446043165467626,
+      "grad_norm": 0.5529634859522956,
+      "learning_rate": 7.865310231795026e-06,
+      "loss": 0.6747450828552246,
+      "step": 680
+    },
+    {
+      "epoch": 2.449640287769784,
+      "grad_norm": 0.5932486024596301,
+      "learning_rate": 7.859519216665531e-06,
+      "loss": 0.691978931427002,
+      "step": 681
+    },
+    {
+      "epoch": 2.4532374100719423,
+      "grad_norm": 0.5514442445967912,
+      "learning_rate": 7.853722495649558e-06,
+      "loss": 0.6781147718429565,
+      "step": 682
+    },
+    {
+      "epoch": 2.4568345323741005,
+      "grad_norm": 0.5447933548480308,
+      "learning_rate": 7.84792008031389e-06,
+      "loss": 0.6920583844184875,
+      "step": 683
+    },
+    {
+      "epoch": 2.460431654676259,
+      "grad_norm": 0.5079458959170539,
+      "learning_rate": 7.842111982236669e-06,
+      "loss": 0.6551939249038696,
+      "step": 684
+    },
+    {
+      "epoch": 2.4640287769784175,
+      "grad_norm": 0.5459392427484495,
+      "learning_rate": 7.836298213007376e-06,
+      "loss": 0.6810588836669922,
+      "step": 685
+    },
+    {
+      "epoch": 2.4676258992805753,
+      "grad_norm": 0.5979754426371551,
+      "learning_rate": 7.83047878422681e-06,
+      "loss": 0.6689721941947937,
+      "step": 686
+    },
+    {
+      "epoch": 2.471223021582734,
+      "grad_norm": 0.5536269344910104,
+      "learning_rate": 7.824653707507066e-06,
+      "loss": 0.6804069876670837,
+      "step": 687
+    },
+    {
+      "epoch": 2.4748201438848922,
+      "grad_norm": 0.5379884636295093,
+      "learning_rate": 7.818822994471504e-06,
+      "loss": 0.6857604384422302,
+      "step": 688
+    },
+    {
+      "epoch": 2.4784172661870505,
+      "grad_norm": 0.6094243016024037,
+      "learning_rate": 7.812986656754734e-06,
+      "loss": 0.6841216087341309,
+      "step": 689
+    },
+    {
+      "epoch": 2.4820143884892087,
+      "grad_norm": 0.569511803474814,
+      "learning_rate": 7.807144706002582e-06,
+      "loss": 0.6696672439575195,
+      "step": 690
+    },
+    {
+      "epoch": 2.485611510791367,
+      "grad_norm": 0.5526697959007867,
+      "learning_rate": 7.80129715387209e-06,
+      "loss": 0.6752933263778687,
+      "step": 691
+    },
+    {
+      "epoch": 2.4892086330935252,
+      "grad_norm": 0.580107961423166,
+      "learning_rate": 7.79544401203146e-06,
+      "loss": 0.6760281920433044,
+      "step": 692
+    },
+    {
+      "epoch": 2.4928057553956835,
+      "grad_norm": 0.5861334677021053,
+      "learning_rate": 7.789585292160056e-06,
+      "loss": 0.6881128549575806,
+      "step": 693
+    },
+    {
+      "epoch": 2.4964028776978417,
+      "grad_norm": 0.547004214562634,
+      "learning_rate": 7.783721005948374e-06,
+      "loss": 0.6737272143363953,
+      "step": 694
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.560582389214819,
+      "learning_rate": 7.777851165098012e-06,
+      "loss": 0.6740151643753052,
+      "step": 695
+    },
+    {
+      "epoch": 2.5035971223021583,
+      "grad_norm": 0.5498672869869043,
+      "learning_rate": 7.771975781321655e-06,
+      "loss": 0.6854318380355835,
+      "step": 696
+    },
+    {
+      "epoch": 2.5071942446043165,
+      "grad_norm": 0.5913867717651166,
+      "learning_rate": 7.766094866343047e-06,
+      "loss": 0.6820253133773804,
+      "step": 697
+    },
+    {
+      "epoch": 2.5107913669064748,
+      "grad_norm": 0.5526401475935052,
+      "learning_rate": 7.760208431896971e-06,
+      "loss": 0.7144240736961365,
+      "step": 698
+    },
+    {
+      "epoch": 2.514388489208633,
+      "grad_norm": 0.551040934868349,
+      "learning_rate": 7.754316489729224e-06,
+      "loss": 0.6829532384872437,
+      "step": 699
+    },
+    {
+      "epoch": 2.5179856115107913,
+      "grad_norm": 0.5353368003261932,
+      "learning_rate": 7.748419051596586e-06,
+      "loss": 0.6781253814697266,
+      "step": 700
+    },
+    {
+      "epoch": 2.5179856115107913,
+      "eval_loss": 0.7946798205375671,
+      "eval_runtime": 7.5195,
+      "eval_samples_per_second": 11.703,
+      "eval_steps_per_second": 0.399,
+      "eval_token_acc": 0.7532400865391975,
+      "step": 700
+    },
+    {
+      "epoch": 2.5215827338129495,
+      "grad_norm": 0.5429962157842656,
+      "learning_rate": 7.74251612926681e-06,
+      "loss": 0.6926379203796387,
+      "step": 701
+    },
+    {
+      "epoch": 2.5251798561151078,
+      "grad_norm": 0.5529607867258931,
+      "learning_rate": 7.736607734518593e-06,
+      "loss": 0.6826739311218262,
+      "step": 702
+    },
+    {
+      "epoch": 2.528776978417266,
+      "grad_norm": 0.5709854889740096,
+      "learning_rate": 7.730693879141548e-06,
+      "loss": 0.6736966967582703,
+      "step": 703
+    },
+    {
+      "epoch": 2.5323741007194247,
+      "grad_norm": 0.5405832962269905,
+      "learning_rate": 7.72477457493619e-06,
+      "loss": 0.6709224581718445,
+      "step": 704
+    },
+    {
+      "epoch": 2.5359712230215825,
+      "grad_norm": 0.5470431841631396,
+      "learning_rate": 7.718849833713895e-06,
+      "loss": 0.6827516555786133,
+      "step": 705
+    },
+    {
+      "epoch": 2.539568345323741,
+      "grad_norm": 0.5587267021136716,
+      "learning_rate": 7.712919667296902e-06,
+      "loss": 0.6617273688316345,
+      "step": 706
+    },
+    {
+      "epoch": 2.543165467625899,
+      "grad_norm": 0.6011679551118603,
+      "learning_rate": 7.706984087518265e-06,
+      "loss": 0.6764623522758484,
+      "step": 707
+    },
+    {
+      "epoch": 2.5467625899280577,
+      "grad_norm": 0.5491196843814833,
+      "learning_rate": 7.701043106221847e-06,
+      "loss": 0.6807689070701599,
+      "step": 708
+    },
+    {
+      "epoch": 2.550359712230216,
+      "grad_norm": 0.5551355195436328,
+      "learning_rate": 7.695096735262284e-06,
+      "loss": 0.6934317350387573,
+      "step": 709
+    },
+    {
+      "epoch": 2.553956834532374,
+      "grad_norm": 0.5412804571062837,
+      "learning_rate": 7.689144986504966e-06,
+      "loss": 0.6880820989608765,
+      "step": 710
+    },
+    {
+      "epoch": 2.5575539568345325,
+      "grad_norm": 0.5104975508357923,
+      "learning_rate": 7.68318787182602e-06,
+      "loss": 0.6632246971130371,
+      "step": 711
+    },
+    {
+      "epoch": 2.5611510791366907,
+      "grad_norm": 0.5588315359311713,
+      "learning_rate": 7.677225403112277e-06,
+      "loss": 0.6966925263404846,
+      "step": 712
+    },
+    {
+      "epoch": 2.564748201438849,
+      "grad_norm": 0.5583593320200122,
+      "learning_rate": 7.671257592261247e-06,
+      "loss": 0.6910549402236938,
+      "step": 713
+    },
+    {
+      "epoch": 2.568345323741007,
+      "grad_norm": 0.5726244981625573,
+      "learning_rate": 7.665284451181106e-06,
+      "loss": 0.6583359241485596,
+      "step": 714
+    },
+    {
+      "epoch": 2.5719424460431655,
+      "grad_norm": 0.575772820092244,
+      "learning_rate": 7.659305991790663e-06,
+      "loss": 0.6736561059951782,
+      "step": 715
+    },
+    {
+      "epoch": 2.5755395683453237,
+      "grad_norm": 0.5446990409778754,
+      "learning_rate": 7.653322226019341e-06,
+      "loss": 0.6787956953048706,
+      "step": 716
+    },
+    {
+      "epoch": 2.579136690647482,
+      "grad_norm": 0.6357696348428653,
+      "learning_rate": 7.647333165807147e-06,
+      "loss": 0.673660159111023,
+      "step": 717
+    },
+    {
+      "epoch": 2.58273381294964,
+      "grad_norm": 0.5689245629147284,
+      "learning_rate": 7.64133882310466e-06,
+      "loss": 0.6633349657058716,
+      "step": 718
+    },
+    {
+      "epoch": 2.5863309352517985,
+      "grad_norm": 0.553093137270363,
+      "learning_rate": 7.63533920987299e-06,
+      "loss": 0.6687748432159424,
+      "step": 719
+    },
+    {
+      "epoch": 2.5899280575539567,
+      "grad_norm": 0.512706946176256,
+      "learning_rate": 7.629334338083774e-06,
+      "loss": 0.6984813809394836,
+      "step": 720
+    },
+    {
+      "epoch": 2.593525179856115,
+      "grad_norm": 0.5253240659400193,
+      "learning_rate": 7.623324219719134e-06,
+      "loss": 0.6692329049110413,
+      "step": 721
+    },
+    {
+      "epoch": 2.597122302158273,
+      "grad_norm": 0.5540251715820621,
+      "learning_rate": 7.617308866771667e-06,
+      "loss": 0.682288646697998,
+      "step": 722
+    },
+    {
+      "epoch": 2.600719424460432,
+      "grad_norm": 0.5415999732946063,
+      "learning_rate": 7.611288291244407e-06,
+      "loss": 0.6738604307174683,
+      "step": 723
+    },
+    {
+      "epoch": 2.6043165467625897,
+      "grad_norm": 0.5398258033309221,
+      "learning_rate": 7.605262505150819e-06,
+      "loss": 0.6791110634803772,
+      "step": 724
+    },
+    {
+      "epoch": 2.6079136690647484,
+      "grad_norm": 0.5870193785417733,
+      "learning_rate": 7.599231520514756e-06,
+      "loss": 0.6960330009460449,
+      "step": 725
+    },
+    {
+      "epoch": 2.6115107913669062,
+      "grad_norm": 0.5478681078185549,
+      "learning_rate": 7.5931953493704545e-06,
+      "loss": 0.6817760467529297,
+      "step": 726
+    },
+    {
+      "epoch": 2.615107913669065,
+      "grad_norm": 0.5643881705522068,
+      "learning_rate": 7.58715400376249e-06,
+      "loss": 0.6764886379241943,
+      "step": 727
+    },
+    {
+      "epoch": 2.618705035971223,
+      "grad_norm": 0.5311796813237344,
+      "learning_rate": 7.58110749574577e-06,
+      "loss": 0.681626558303833,
+      "step": 728
+    },
+    {
+      "epoch": 2.6223021582733814,
+      "grad_norm": 0.593290297861787,
+      "learning_rate": 7.575055837385497e-06,
+      "loss": 0.6835793256759644,
+      "step": 729
+    },
+    {
+      "epoch": 2.6258992805755397,
+      "grad_norm": 0.5433885872721661,
+      "learning_rate": 7.568999040757157e-06,
+      "loss": 0.692463219165802,
+      "step": 730
+    },
+    {
+      "epoch": 2.629496402877698,
+      "grad_norm": 0.566347769689796,
+      "learning_rate": 7.562937117946484e-06,
+      "loss": 0.6714706420898438,
+      "step": 731
+    },
+    {
+      "epoch": 2.633093525179856,
+      "grad_norm": 0.5788236428878424,
+      "learning_rate": 7.556870081049444e-06,
+      "loss": 0.6920543909072876,
+      "step": 732
+    },
+    {
+      "epoch": 2.6366906474820144,
+      "grad_norm": 0.5629361850310614,
+      "learning_rate": 7.550797942172207e-06,
+      "loss": 0.6735811233520508,
+      "step": 733
+    },
+    {
+      "epoch": 2.6402877697841727,
+      "grad_norm": 0.5238564692093962,
+      "learning_rate": 7.5447207134311195e-06,
+      "loss": 0.6693782210350037,
+      "step": 734
+    },
+    {
+      "epoch": 2.643884892086331,
+      "grad_norm": 0.5202480993459492,
+      "learning_rate": 7.538638406952692e-06,
+      "loss": 0.6865730285644531,
+      "step": 735
+    },
+    {
+      "epoch": 2.647482014388489,
+      "grad_norm": 0.5398235199330227,
+      "learning_rate": 7.532551034873558e-06,
+      "loss": 0.6806303262710571,
+      "step": 736
+    },
+    {
+      "epoch": 2.6510791366906474,
+      "grad_norm": 0.5169653355837797,
+      "learning_rate": 7.526458609340468e-06,
+      "loss": 0.6690386533737183,
+      "step": 737
+    },
+    {
+      "epoch": 2.6546762589928057,
+      "grad_norm": 0.5466767225574092,
+      "learning_rate": 7.520361142510248e-06,
+      "loss": 0.6830763816833496,
+      "step": 738
+    },
+    {
+      "epoch": 2.658273381294964,
+      "grad_norm": 0.5566750563025337,
+      "learning_rate": 7.5142586465497925e-06,
+      "loss": 0.6852663159370422,
+      "step": 739
+    },
+    {
+      "epoch": 2.661870503597122,
+      "grad_norm": 0.5280240519640393,
+      "learning_rate": 7.50815113363602e-06,
+      "loss": 0.6713100671768188,
+      "step": 740
+    },
+    {
+      "epoch": 2.6654676258992804,
+      "grad_norm": 0.5334063560685919,
+      "learning_rate": 7.5020386159558685e-06,
+      "loss": 0.6669318675994873,
+      "step": 741
+    },
+    {
+      "epoch": 2.6690647482014387,
+      "grad_norm": 0.5410901797634948,
+      "learning_rate": 7.49592110570626e-06,
+      "loss": 0.6787617206573486,
+      "step": 742
+    },
+    {
+      "epoch": 2.672661870503597,
+      "grad_norm": 0.5035182125942762,
+      "learning_rate": 7.489798615094077e-06,
+      "loss": 0.6858699917793274,
+      "step": 743
+    },
+    {
+      "epoch": 2.6762589928057556,
+      "grad_norm": 0.5272103112959701,
+      "learning_rate": 7.483671156336142e-06,
+      "loss": 0.6731346249580383,
+      "step": 744
+    },
+    {
+      "epoch": 2.6798561151079134,
+      "grad_norm": 0.5466411879626102,
+      "learning_rate": 7.4775387416591905e-06,
+      "loss": 0.6715824604034424,
+      "step": 745
+    },
+    {
+      "epoch": 2.683453237410072,
+      "grad_norm": 0.556601396124224,
+      "learning_rate": 7.471401383299847e-06,
+      "loss": 0.6648485660552979,
+      "step": 746
+    },
+    {
+      "epoch": 2.68705035971223,
+      "grad_norm": 0.5171685221567268,
+      "learning_rate": 7.465259093504598e-06,
+      "loss": 0.6769684553146362,
+      "step": 747
+    },
+    {
+      "epoch": 2.6906474820143886,
+      "grad_norm": 0.5606435302484213,
+      "learning_rate": 7.459111884529775e-06,
+      "loss": 0.6720460653305054,
+      "step": 748
+    },
+    {
+      "epoch": 2.694244604316547,
+      "grad_norm": 0.5842110032032478,
+      "learning_rate": 7.452959768641521e-06,
+      "loss": 0.677690863609314,
+      "step": 749
+    },
+    {
+      "epoch": 2.697841726618705,
+      "grad_norm": 0.5701430080753563,
+      "learning_rate": 7.446802758115775e-06,
+      "loss": 0.6708799600601196,
+      "step": 750
+    },
+    {
+      "epoch": 2.7014388489208634,
+      "grad_norm": 0.5581963729995891,
+      "learning_rate": 7.4406408652382355e-06,
+      "loss": 0.6710215210914612,
+      "step": 751
+    },
+    {
+      "epoch": 2.7050359712230216,
+      "grad_norm": 0.5658077696960466,
+      "learning_rate": 7.43447410230435e-06,
+      "loss": 0.6781923770904541,
+      "step": 752
+    },
+    {
+      "epoch": 2.70863309352518,
+      "grad_norm": 0.518577960318417,
+      "learning_rate": 7.428302481619281e-06,
+      "loss": 0.6796495318412781,
+      "step": 753
+    },
+    {
+      "epoch": 2.712230215827338,
+      "grad_norm": 0.5688235971525768,
+      "learning_rate": 7.422126015497884e-06,
+      "loss": 0.6821467280387878,
+      "step": 754
+    },
+    {
+      "epoch": 2.7158273381294964,
+      "grad_norm": 0.5480902682706095,
+      "learning_rate": 7.415944716264682e-06,
+      "loss": 0.6760743856430054,
+      "step": 755
+    },
+    {
+      "epoch": 2.7194244604316546,
+      "grad_norm": 0.6283165937878773,
+      "learning_rate": 7.409758596253849e-06,
+      "loss": 0.689350962638855,
+      "step": 756
+    },
+    {
+      "epoch": 2.723021582733813,
+      "grad_norm": 0.5425030285263174,
+      "learning_rate": 7.403567667809165e-06,
+      "loss": 0.6779059171676636,
+      "step": 757
+    },
+    {
+      "epoch": 2.726618705035971,
+      "grad_norm": 0.5682122278337587,
+      "learning_rate": 7.397371943284017e-06,
+      "loss": 0.6475615501403809,
+      "step": 758
+    },
+    {
+      "epoch": 2.7302158273381294,
+      "grad_norm": 0.547877297052162,
+      "learning_rate": 7.391171435041358e-06,
+      "loss": 0.6795694231987,
+      "step": 759
+    },
+    {
+      "epoch": 2.7338129496402876,
+      "grad_norm": 0.5533186641373539,
+      "learning_rate": 7.384966155453686e-06,
+      "loss": 0.6737759113311768,
+      "step": 760
+    },
+    {
+      "epoch": 2.737410071942446,
+      "grad_norm": 0.542436921949499,
+      "learning_rate": 7.378756116903018e-06,
+      "loss": 0.6761566996574402,
+      "step": 761
+    },
+    {
+      "epoch": 2.741007194244604,
+      "grad_norm": 0.5668917886571787,
+      "learning_rate": 7.372541331780871e-06,
+      "loss": 0.6677440404891968,
+      "step": 762
+    },
+    {
+      "epoch": 2.744604316546763,
+      "grad_norm": 0.5216933811477933,
+      "learning_rate": 7.366321812488228e-06,
+      "loss": 0.6678866147994995,
+      "step": 763
+    },
+    {
+      "epoch": 2.7482014388489207,
+      "grad_norm": 0.5317153536757551,
+      "learning_rate": 7.360097571435527e-06,
+      "loss": 0.6877514123916626,
+      "step": 764
+    },
+    {
+      "epoch": 2.7517985611510793,
+      "grad_norm": 0.5612769192367274,
+      "learning_rate": 7.353868621042617e-06,
+      "loss": 0.6856654286384583,
+      "step": 765
+    },
+    {
+      "epoch": 2.755395683453237,
+      "grad_norm": 0.5757162734183815,
+      "learning_rate": 7.347634973738753e-06,
+      "loss": 0.6839314699172974,
+      "step": 766
+    },
+    {
+      "epoch": 2.758992805755396,
+      "grad_norm": 0.5107418889568749,
+      "learning_rate": 7.3413966419625595e-06,
+      "loss": 0.6695917844772339,
+      "step": 767
+    },
+    {
+      "epoch": 2.762589928057554,
+      "grad_norm": 0.598400476477111,
+      "learning_rate": 7.335153638162005e-06,
+      "loss": 0.6794536113739014,
+      "step": 768
+    },
+    {
+      "epoch": 2.7661870503597124,
+      "grad_norm": 0.5997392739164393,
+      "learning_rate": 7.328905974794383e-06,
+      "loss": 0.6909158229827881,
+      "step": 769
+    },
+    {
+      "epoch": 2.7697841726618706,
+      "grad_norm": 0.5335666892122978,
+      "learning_rate": 7.322653664326289e-06,
+      "loss": 0.6530051827430725,
+      "step": 770
+    },
+    {
+      "epoch": 2.773381294964029,
+      "grad_norm": 0.5459019178889832,
+      "learning_rate": 7.3163967192335825e-06,
+      "loss": 0.6856040358543396,
+      "step": 771
+    },
+    {
+      "epoch": 2.776978417266187,
+      "grad_norm": 0.5439989869401481,
+      "learning_rate": 7.310135152001381e-06,
+      "loss": 0.6689339876174927,
+      "step": 772
+    },
+    {
+      "epoch": 2.7805755395683454,
+      "grad_norm": 0.5477532967369428,
+      "learning_rate": 7.303868975124014e-06,
+      "loss": 0.6802787780761719,
+      "step": 773
+    },
+    {
+      "epoch": 2.7841726618705036,
+      "grad_norm": 0.5874456113809587,
+      "learning_rate": 7.29759820110502e-06,
+      "loss": 0.7009991407394409,
+      "step": 774
+    },
+    {
+      "epoch": 2.787769784172662,
+      "grad_norm": 0.551919806400329,
+      "learning_rate": 7.291322842457103e-06,
+      "loss": 0.6680014133453369,
+      "step": 775
+    },
+    {
+      "epoch": 2.79136690647482,
+      "grad_norm": 0.5800724322668104,
+      "learning_rate": 7.285042911702116e-06,
+      "loss": 0.6827471852302551,
+      "step": 776
+    },
+    {
+      "epoch": 2.7949640287769784,
+      "grad_norm": 0.550034880832541,
+      "learning_rate": 7.278758421371041e-06,
+      "loss": 0.6836577653884888,
+      "step": 777
+    },
+    {
+      "epoch": 2.7985611510791366,
+      "grad_norm": 0.5525819839003926,
+      "learning_rate": 7.27246938400395e-06,
+      "loss": 0.6785211563110352,
+      "step": 778
+    },
+    {
+      "epoch": 2.802158273381295,
+      "grad_norm": 0.5439718572691205,
+      "learning_rate": 7.266175812149993e-06,
+      "loss": 0.6937276124954224,
+      "step": 779
+    },
+    {
+      "epoch": 2.805755395683453,
+      "grad_norm": 0.5432443235974502,
+      "learning_rate": 7.259877718367372e-06,
+      "loss": 0.6906551122665405,
+      "step": 780
+    },
+    {
+      "epoch": 2.8093525179856114,
+      "grad_norm": 0.5604293839707127,
+      "learning_rate": 7.2535751152233015e-06,
+      "loss": 0.6693239808082581,
+      "step": 781
+    },
+    {
+      "epoch": 2.81294964028777,
+      "grad_norm": 0.5334401786784696,
+      "learning_rate": 7.2472680152940015e-06,
+      "loss": 0.6608414649963379,
+      "step": 782
+    },
+    {
+      "epoch": 2.816546762589928,
+      "grad_norm": 0.5556108924839624,
+      "learning_rate": 7.240956431164664e-06,
+      "loss": 0.6791374087333679,
+      "step": 783
+    },
+    {
+      "epoch": 2.8201438848920866,
+      "grad_norm": 0.5595296433734097,
+      "learning_rate": 7.234640375429427e-06,
+      "loss": 0.6910173892974854,
+      "step": 784
+    },
+    {
+      "epoch": 2.8237410071942444,
+      "grad_norm": 0.6026988319042906,
+      "learning_rate": 7.228319860691354e-06,
+      "loss": 0.6745973825454712,
+      "step": 785
+    },
+    {
+      "epoch": 2.827338129496403,
+      "grad_norm": 0.5537401684593308,
+      "learning_rate": 7.2219948995624035e-06,
+      "loss": 0.6714953184127808,
+      "step": 786
+    },
+    {
+      "epoch": 2.8309352517985613,
+      "grad_norm": 0.5159995510628995,
+      "learning_rate": 7.215665504663405e-06,
+      "loss": 0.6651762127876282,
+      "step": 787
+    },
+    {
+      "epoch": 2.8345323741007196,
+      "grad_norm": 0.5478561040721687,
+      "learning_rate": 7.209331688624039e-06,
+      "loss": 0.6660327911376953,
+      "step": 788
+    },
+    {
+      "epoch": 2.838129496402878,
+      "grad_norm": 0.5379987836741197,
+      "learning_rate": 7.202993464082807e-06,
+      "loss": 0.6914957761764526,
+      "step": 789
+    },
+    {
+      "epoch": 2.841726618705036,
+      "grad_norm": 0.5253645674454258,
+      "learning_rate": 7.1966508436870044e-06,
+      "loss": 0.6743172407150269,
+      "step": 790
+    },
+    {
+      "epoch": 2.8453237410071943,
+      "grad_norm": 0.5598208338037115,
+      "learning_rate": 7.190303840092701e-06,
+      "loss": 0.6961389780044556,
+      "step": 791
+    },
+    {
+      "epoch": 2.8489208633093526,
+      "grad_norm": 0.5390983976675429,
+      "learning_rate": 7.183952465964711e-06,
+      "loss": 0.6708844900131226,
+      "step": 792
+    },
+    {
+      "epoch": 2.852517985611511,
+      "grad_norm": 0.5442089855190607,
+      "learning_rate": 7.177596733976571e-06,
+      "loss": 0.6630173921585083,
+      "step": 793
+    },
+    {
+      "epoch": 2.856115107913669,
+      "grad_norm": 0.671153831312209,
+      "learning_rate": 7.171236656810513e-06,
+      "loss": 0.6860195994377136,
+      "step": 794
+    },
+    {
+      "epoch": 2.8597122302158273,
+      "grad_norm": 0.5780688024521513,
+      "learning_rate": 7.1648722471574385e-06,
+      "loss": 0.6743713617324829,
+      "step": 795
+    },
+    {
+      "epoch": 2.8633093525179856,
+      "grad_norm": 0.5966815162498091,
+      "learning_rate": 7.158503517716894e-06,
+      "loss": 0.6660425066947937,
+      "step": 796
+    },
+    {
+      "epoch": 2.866906474820144,
+      "grad_norm": 0.5220025188437087,
+      "learning_rate": 7.152130481197048e-06,
+      "loss": 0.677868127822876,
+      "step": 797
+    },
+    {
+      "epoch": 2.870503597122302,
+      "grad_norm": 0.6530859738063051,
+      "learning_rate": 7.145753150314661e-06,
+      "loss": 0.6992630362510681,
+      "step": 798
+    },
+    {
+      "epoch": 2.8741007194244603,
+      "grad_norm": 0.5372141806985119,
+      "learning_rate": 7.139371537795063e-06,
+      "loss": 0.6782932281494141,
+      "step": 799
+    },
+    {
+      "epoch": 2.8776978417266186,
+      "grad_norm": 0.5594739478699063,
+      "learning_rate": 7.132985656372126e-06,
+      "loss": 0.6552035212516785,
+      "step": 800
+    },
+    {
+      "epoch": 2.8776978417266186,
+      "eval_loss": 0.7926490902900696,
+      "eval_runtime": 7.5343,
+      "eval_samples_per_second": 11.68,
+      "eval_steps_per_second": 0.398,
+      "eval_token_acc": 0.7537646230896883,
+      "step": 800
+    },
+    {
+      "epoch": 2.881294964028777,
+      "grad_norm": 0.4969988859148219,
+      "learning_rate": 7.126595518788244e-06,
+      "loss": 0.6519225835800171,
+      "step": 801
+    },
+    {
+      "epoch": 2.884892086330935,
+      "grad_norm": 0.5345813451985074,
+      "learning_rate": 7.120201137794301e-06,
+      "loss": 0.665630578994751,
+      "step": 802
+    },
+    {
+      "epoch": 2.8884892086330938,
+      "grad_norm": 0.5027605576333997,
+      "learning_rate": 7.11380252614965e-06,
+      "loss": 0.6709873676300049,
+      "step": 803
+    },
+    {
+      "epoch": 2.8920863309352516,
+      "grad_norm": 0.5884864926353124,
+      "learning_rate": 7.1073996966220835e-06,
+      "loss": 0.6509050130844116,
+      "step": 804
+    },
+    {
+      "epoch": 2.8956834532374103,
+      "grad_norm": 0.5531162956721329,
+      "learning_rate": 7.10099266198781e-06,
+      "loss": 0.6866740584373474,
+      "step": 805
+    },
+    {
+      "epoch": 2.899280575539568,
+      "grad_norm": 0.5162955485980054,
+      "learning_rate": 7.0945814350314346e-06,
+      "loss": 0.6785730123519897,
+      "step": 806
+    },
+    {
+      "epoch": 2.902877697841727,
+      "grad_norm": 0.520461442041789,
+      "learning_rate": 7.0881660285459205e-06,
+      "loss": 0.6749669313430786,
+      "step": 807
+    },
+    {
+      "epoch": 2.906474820143885,
+      "grad_norm": 0.5301790568209777,
+      "learning_rate": 7.0817464553325764e-06,
+      "loss": 0.6701935529708862,
+      "step": 808
+    },
+    {
+      "epoch": 2.9100719424460433,
+      "grad_norm": 0.5630040213851715,
+      "learning_rate": 7.07532272820102e-06,
+      "loss": 0.678869366645813,
+      "step": 809
+    },
+    {
+      "epoch": 2.9136690647482015,
+      "grad_norm": 0.5390394232246156,
+      "learning_rate": 7.068894859969162e-06,
+      "loss": 0.6616356372833252,
+      "step": 810
+    },
+    {
+      "epoch": 2.91726618705036,
+      "grad_norm": 0.5666314327761091,
+      "learning_rate": 7.062462863463176e-06,
+      "loss": 0.6806893348693848,
+      "step": 811
+    },
+    {
+      "epoch": 2.920863309352518,
+      "grad_norm": 0.5302133065489669,
+      "learning_rate": 7.0560267515174685e-06,
+      "loss": 0.669838547706604,
+      "step": 812
+    },
+    {
+      "epoch": 2.9244604316546763,
+      "grad_norm": 0.5765934775556447,
+      "learning_rate": 7.049586536974664e-06,
+      "loss": 0.6633042097091675,
+      "step": 813
+    },
+    {
+      "epoch": 2.9280575539568345,
+      "grad_norm": 0.5277877946096216,
+      "learning_rate": 7.043142232685572e-06,
+      "loss": 0.6631823778152466,
+      "step": 814
+    },
+    {
+      "epoch": 2.931654676258993,
+      "grad_norm": 0.5388691932839309,
+      "learning_rate": 7.036693851509158e-06,
+      "loss": 0.6837793588638306,
+      "step": 815
+    },
+    {
+      "epoch": 2.935251798561151,
+      "grad_norm": 0.5504190884085625,
+      "learning_rate": 7.030241406312528e-06,
+      "loss": 0.6746753454208374,
+      "step": 816
+    },
+    {
+      "epoch": 2.9388489208633093,
+      "grad_norm": 0.5366808043734594,
+      "learning_rate": 7.023784909970894e-06,
+      "loss": 0.6726278066635132,
+      "step": 817
+    },
+    {
+      "epoch": 2.9424460431654675,
+      "grad_norm": 0.5435443879201644,
+      "learning_rate": 7.0173243753675544e-06,
+      "loss": 0.6841650009155273,
+      "step": 818
+    },
+    {
+      "epoch": 2.946043165467626,
+      "grad_norm": 0.5542852746597902,
+      "learning_rate": 7.010859815393863e-06,
+      "loss": 0.6923326253890991,
+      "step": 819
+    },
+    {
+      "epoch": 2.949640287769784,
+      "grad_norm": 0.5289515792429416,
+      "learning_rate": 7.004391242949209e-06,
+      "loss": 0.6755713224411011,
+      "step": 820
+    },
+    {
+      "epoch": 2.9532374100719423,
+      "grad_norm": 0.537949591506382,
+      "learning_rate": 6.997918670940984e-06,
+      "loss": 0.6711938381195068,
+      "step": 821
+    },
+    {
+      "epoch": 2.956834532374101,
+      "grad_norm": 0.5728469119954346,
+      "learning_rate": 6.991442112284561e-06,
+      "loss": 0.6789706945419312,
+      "step": 822
+    },
+    {
+      "epoch": 2.960431654676259,
+      "grad_norm": 0.5356838759841633,
+      "learning_rate": 6.984961579903273e-06,
+      "loss": 0.6805518865585327,
+      "step": 823
+    },
+    {
+      "epoch": 2.9640287769784175,
+      "grad_norm": 0.5526149559188349,
+      "learning_rate": 6.978477086728375e-06,
+      "loss": 0.6938677430152893,
+      "step": 824
+    },
+    {
+      "epoch": 2.9676258992805753,
+      "grad_norm": 0.5717927051941512,
+      "learning_rate": 6.9719886456990306e-06,
+      "loss": 0.681894838809967,
+      "step": 825
+    },
+    {
+      "epoch": 2.971223021582734,
+      "grad_norm": 0.537402386834528,
+      "learning_rate": 6.96549626976228e-06,
+      "loss": 0.6860501766204834,
+      "step": 826
+    },
+    {
+      "epoch": 2.9748201438848922,
+      "grad_norm": 0.5826694738153739,
+      "learning_rate": 6.958999971873012e-06,
+      "loss": 0.6714417934417725,
+      "step": 827
+    },
+    {
+      "epoch": 2.9784172661870505,
+      "grad_norm": 0.6983420638788042,
+      "learning_rate": 6.952499764993945e-06,
+      "loss": 0.6649173498153687,
+      "step": 828
+    },
+    {
+      "epoch": 2.9820143884892087,
+      "grad_norm": 0.5186299633485753,
+      "learning_rate": 6.945995662095596e-06,
+      "loss": 0.6727434396743774,
+      "step": 829
+    },
+    {
+      "epoch": 2.985611510791367,
+      "grad_norm": 0.543769678807827,
+      "learning_rate": 6.9394876761562555e-06,
+      "loss": 0.6850245594978333,
+      "step": 830
+    },
+    {
+      "epoch": 2.9892086330935252,
+      "grad_norm": 0.5438520208221167,
+      "learning_rate": 6.932975820161965e-06,
+      "loss": 0.6802777051925659,
+      "step": 831
+    },
+    {
+      "epoch": 2.9928057553956835,
+      "grad_norm": 0.5462075560169861,
+      "learning_rate": 6.926460107106483e-06,
+      "loss": 0.690509021282196,
+      "step": 832
+    },
+    {
+      "epoch": 2.9964028776978417,
+      "grad_norm": 0.5333472735304643,
+      "learning_rate": 6.9199405499912675e-06,
+      "loss": 0.690028190612793,
+      "step": 833
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.6699724818892617,
+      "learning_rate": 6.913417161825449e-06,
+      "loss": 0.6984052658081055,
+      "step": 834
+    },
+    {
+      "epoch": 3.0035971223021583,
+      "grad_norm": 0.714752211223029,
+      "learning_rate": 6.906889955625801e-06,
+      "loss": 0.6158499121665955,
+      "step": 835
+    },
+    {
+      "epoch": 3.0071942446043165,
+      "grad_norm": 0.6514192010389133,
+      "learning_rate": 6.900358944416712e-06,
+      "loss": 0.6168891191482544,
+      "step": 836
+    },
+    {
+      "epoch": 3.0107913669064748,
+      "grad_norm": 0.64522746175661,
+      "learning_rate": 6.893824141230168e-06,
+      "loss": 0.6089640855789185,
+      "step": 837
+    },
+    {
+      "epoch": 3.014388489208633,
+      "grad_norm": 0.5626906528595018,
+      "learning_rate": 6.887285559105721e-06,
+      "loss": 0.6088428497314453,
+      "step": 838
+    },
+    {
+      "epoch": 3.0179856115107913,
+      "grad_norm": 0.5952893215965329,
+      "learning_rate": 6.880743211090457e-06,
+      "loss": 0.6181361079216003,
+      "step": 839
+    },
+    {
+      "epoch": 3.0215827338129495,
+      "grad_norm": 0.6112525299111327,
+      "learning_rate": 6.874197110238986e-06,
+      "loss": 0.6280969381332397,
+      "step": 840
+    },
+    {
+      "epoch": 3.0251798561151078,
+      "grad_norm": 0.6255439774080229,
+      "learning_rate": 6.8676472696134e-06,
+      "loss": 0.6228752732276917,
+      "step": 841
+    },
+    {
+      "epoch": 3.028776978417266,
+      "grad_norm": 0.5893354763922161,
+      "learning_rate": 6.861093702283254e-06,
+      "loss": 0.6126219630241394,
+      "step": 842
+    },
+    {
+      "epoch": 3.0323741007194243,
+      "grad_norm": 0.5519668238089376,
+      "learning_rate": 6.854536421325544e-06,
+      "loss": 0.6280915141105652,
+      "step": 843
+    },
+    {
+      "epoch": 3.0359712230215825,
+      "grad_norm": 0.5744905423808728,
+      "learning_rate": 6.847975439824669e-06,
+      "loss": 0.6058259606361389,
+      "step": 844
+    },
+    {
+      "epoch": 3.039568345323741,
+      "grad_norm": 0.6867335751974597,
+      "learning_rate": 6.841410770872418e-06,
+      "loss": 0.6044293642044067,
+      "step": 845
+    },
+    {
+      "epoch": 3.0431654676258995,
+      "grad_norm": 0.6004411853436984,
+      "learning_rate": 6.834842427567934e-06,
+      "loss": 0.6123199462890625,
+      "step": 846
+    },
+    {
+      "epoch": 3.0467625899280577,
+      "grad_norm": 0.6205986433996288,
+      "learning_rate": 6.828270423017694e-06,
+      "loss": 0.6331131458282471,
+      "step": 847
+    },
+    {
+      "epoch": 3.050359712230216,
+      "grad_norm": 0.584969288502554,
+      "learning_rate": 6.8216947703354815e-06,
+      "loss": 0.6207732558250427,
+      "step": 848
+    },
+    {
+      "epoch": 3.053956834532374,
+      "grad_norm": 0.5646862308680043,
+      "learning_rate": 6.815115482642358e-06,
+      "loss": 0.6176056265830994,
+      "step": 849
+    },
+    {
+      "epoch": 3.0575539568345325,
+      "grad_norm": 0.6211660655189948,
+      "learning_rate": 6.808532573066635e-06,
+      "loss": 0.6335552930831909,
+      "step": 850
+    },
+    {
+      "epoch": 3.0611510791366907,
+      "grad_norm": 0.5444740509963648,
+      "learning_rate": 6.801946054743858e-06,
+      "loss": 0.5941608548164368,
+      "step": 851
+    },
+    {
+      "epoch": 3.064748201438849,
+      "grad_norm": 0.6099555103213884,
+      "learning_rate": 6.795355940816768e-06,
+      "loss": 0.612288236618042,
+      "step": 852
+    },
+    {
+      "epoch": 3.068345323741007,
+      "grad_norm": 0.5819148236716216,
+      "learning_rate": 6.7887622444352826e-06,
+      "loss": 0.6323422193527222,
+      "step": 853
+    },
+    {
+      "epoch": 3.0719424460431655,
+      "grad_norm": 0.5943964905724924,
+      "learning_rate": 6.7821649787564706e-06,
+      "loss": 0.6132618188858032,
+      "step": 854
+    },
+    {
+      "epoch": 3.0755395683453237,
+      "grad_norm": 0.5760335005664527,
+      "learning_rate": 6.775564156944517e-06,
+      "loss": 0.6210411787033081,
+      "step": 855
+    },
+    {
+      "epoch": 3.079136690647482,
+      "grad_norm": 0.7508771443385528,
+      "learning_rate": 6.7689597921707065e-06,
+      "loss": 0.6110687255859375,
+      "step": 856
+    },
+    {
+      "epoch": 3.08273381294964,
+      "grad_norm": 0.565618301653177,
+      "learning_rate": 6.762351897613393e-06,
+      "loss": 0.6275750398635864,
+      "step": 857
+    },
+    {
+      "epoch": 3.0863309352517985,
+      "grad_norm": 0.5746230797567085,
+      "learning_rate": 6.755740486457973e-06,
+      "loss": 0.6221362352371216,
+      "step": 858
+    },
+    {
+      "epoch": 3.0899280575539567,
+      "grad_norm": 0.5958569000961971,
+      "learning_rate": 6.749125571896863e-06,
+      "loss": 0.6043376922607422,
+      "step": 859
+    },
+    {
+      "epoch": 3.093525179856115,
+      "grad_norm": 0.6241765945103184,
+      "learning_rate": 6.742507167129465e-06,
+      "loss": 0.6196264624595642,
+      "step": 860
+    },
+    {
+      "epoch": 3.097122302158273,
+      "grad_norm": 0.5663996849434653,
+      "learning_rate": 6.735885285362151e-06,
+      "loss": 0.608167290687561,
+      "step": 861
+    },
+    {
+      "epoch": 3.1007194244604315,
+      "grad_norm": 0.6078428622732656,
+      "learning_rate": 6.729259939808223e-06,
+      "loss": 0.6227314472198486,
+      "step": 862
+    },
+    {
+      "epoch": 3.1043165467625897,
+      "grad_norm": 0.5866739065637034,
+      "learning_rate": 6.722631143687904e-06,
+      "loss": 0.6130132079124451,
+      "step": 863
+    },
+    {
+      "epoch": 3.1079136690647484,
+      "grad_norm": 0.6096843816859564,
+      "learning_rate": 6.715998910228296e-06,
+      "loss": 0.6262279748916626,
+      "step": 864
+    },
+    {
+      "epoch": 3.1115107913669067,
+      "grad_norm": 0.5731998321265244,
+      "learning_rate": 6.7093632526633655e-06,
+      "loss": 0.6277258396148682,
+      "step": 865
+    },
+    {
+      "epoch": 3.115107913669065,
+      "grad_norm": 0.5829645067278316,
+      "learning_rate": 6.702724184233904e-06,
+      "loss": 0.6140638589859009,
+      "step": 866
+    },
+    {
+      "epoch": 3.118705035971223,
+      "grad_norm": 0.7914211484234219,
+      "learning_rate": 6.6960817181875146e-06,
+      "loss": 0.6318271160125732,
+      "step": 867
+    },
+    {
+      "epoch": 3.1223021582733814,
+      "grad_norm": 0.5308246774286106,
+      "learning_rate": 6.689435867778578e-06,
+      "loss": 0.6080624461174011,
+      "step": 868
+    },
+    {
+      "epoch": 3.1258992805755397,
+      "grad_norm": 0.5842607361037628,
+      "learning_rate": 6.682786646268226e-06,
+      "loss": 0.6224777698516846,
+      "step": 869
+    },
+    {
+      "epoch": 3.129496402877698,
+      "grad_norm": 0.5788476300864046,
+      "learning_rate": 6.676134066924325e-06,
+      "loss": 0.6108511090278625,
+      "step": 870
+    },
+    {
+      "epoch": 3.133093525179856,
+      "grad_norm": 0.6005172650285331,
+      "learning_rate": 6.66947814302143e-06,
+      "loss": 0.6167119741439819,
+      "step": 871
+    },
+    {
+      "epoch": 3.1366906474820144,
+      "grad_norm": 0.54491315413818,
+      "learning_rate": 6.6628188878407806e-06,
+      "loss": 0.6260868310928345,
+      "step": 872
+    },
+    {
+      "epoch": 3.1402877697841727,
+      "grad_norm": 0.5905361649715649,
+      "learning_rate": 6.656156314670257e-06,
+      "loss": 0.6211916208267212,
+      "step": 873
+    },
+    {
+      "epoch": 3.143884892086331,
+      "grad_norm": 0.5576927601793549,
+      "learning_rate": 6.64949043680436e-06,
+      "loss": 0.6212013363838196,
+      "step": 874
+    },
+    {
+      "epoch": 3.147482014388489,
+      "grad_norm": 0.5761586070224683,
+      "learning_rate": 6.642821267544189e-06,
+      "loss": 0.6145411729812622,
+      "step": 875
+    },
+    {
+      "epoch": 3.1510791366906474,
+      "grad_norm": 0.5805295146865893,
+      "learning_rate": 6.636148820197409e-06,
+      "loss": 0.627646803855896,
+      "step": 876
+    },
+    {
+      "epoch": 3.1546762589928057,
+      "grad_norm": 0.5660419934936849,
+      "learning_rate": 6.6294731080782236e-06,
+      "loss": 0.6196659803390503,
+      "step": 877
+    },
+    {
+      "epoch": 3.158273381294964,
+      "grad_norm": 0.5794080453638514,
+      "learning_rate": 6.622794144507356e-06,
+      "loss": 0.6331967115402222,
+      "step": 878
+    },
+    {
+      "epoch": 3.161870503597122,
+      "grad_norm": 0.57141434057609,
+      "learning_rate": 6.616111942812011e-06,
+      "loss": 0.6022119522094727,
+      "step": 879
+    },
+    {
+      "epoch": 3.1654676258992804,
+      "grad_norm": 0.570784712886501,
+      "learning_rate": 6.609426516325859e-06,
+      "loss": 0.6375267505645752,
+      "step": 880
+    },
+    {
+      "epoch": 3.1690647482014387,
+      "grad_norm": 0.5479852387476096,
+      "learning_rate": 6.6027378783890065e-06,
+      "loss": 0.6149478554725647,
+      "step": 881
+    },
+    {
+      "epoch": 3.172661870503597,
+      "grad_norm": 0.5384725050590596,
+      "learning_rate": 6.596046042347964e-06,
+      "loss": 0.604001522064209,
+      "step": 882
+    },
+    {
+      "epoch": 3.176258992805755,
+      "grad_norm": 0.5946710979785322,
+      "learning_rate": 6.589351021555626e-06,
+      "loss": 0.61385178565979,
+      "step": 883
+    },
+    {
+      "epoch": 3.1798561151079134,
+      "grad_norm": 0.5397188512181753,
+      "learning_rate": 6.58265282937124e-06,
+      "loss": 0.6094223856925964,
+      "step": 884
+    },
+    {
+      "epoch": 3.183453237410072,
+      "grad_norm": 0.5599086808031262,
+      "learning_rate": 6.575951479160385e-06,
+      "loss": 0.6294345855712891,
+      "step": 885
+    },
+    {
+      "epoch": 3.1870503597122304,
+      "grad_norm": 0.58204300377972,
+      "learning_rate": 6.569246984294938e-06,
+      "loss": 0.6208990812301636,
+      "step": 886
+    },
+    {
+      "epoch": 3.1906474820143886,
+      "grad_norm": 0.5531860200808557,
+      "learning_rate": 6.562539358153053e-06,
+      "loss": 0.6135451197624207,
+      "step": 887
+    },
+    {
+      "epoch": 3.194244604316547,
+      "grad_norm": 0.5471226603588754,
+      "learning_rate": 6.555828614119132e-06,
+      "loss": 0.6110199689865112,
+      "step": 888
+    },
+    {
+      "epoch": 3.197841726618705,
+      "grad_norm": 0.5562073314211422,
+      "learning_rate": 6.549114765583799e-06,
+      "loss": 0.6199094653129578,
+      "step": 889
+    },
+    {
+      "epoch": 3.2014388489208634,
+      "grad_norm": 0.584420712338533,
+      "learning_rate": 6.542397825943867e-06,
+      "loss": 0.602266252040863,
+      "step": 890
+    },
+    {
+      "epoch": 3.2050359712230216,
+      "grad_norm": 0.5664900474746994,
+      "learning_rate": 6.535677808602327e-06,
+      "loss": 0.634172797203064,
+      "step": 891
+    },
+    {
+      "epoch": 3.20863309352518,
+      "grad_norm": 0.5574653576860307,
+      "learning_rate": 6.528954726968302e-06,
+      "loss": 0.6266528367996216,
+      "step": 892
+    },
+    {
+      "epoch": 3.212230215827338,
+      "grad_norm": 0.6214080021118603,
+      "learning_rate": 6.522228594457036e-06,
+      "loss": 0.6186414957046509,
+      "step": 893
+    },
+    {
+      "epoch": 3.2158273381294964,
+      "grad_norm": 0.6278160972504896,
+      "learning_rate": 6.515499424489857e-06,
+      "loss": 0.6303264498710632,
+      "step": 894
+    },
+    {
+      "epoch": 3.2194244604316546,
+      "grad_norm": 0.5700913834752749,
+      "learning_rate": 6.508767230494155e-06,
+      "loss": 0.636367917060852,
+      "step": 895
+    },
+    {
+      "epoch": 3.223021582733813,
+      "grad_norm": 0.5539077159835792,
+      "learning_rate": 6.502032025903356e-06,
+      "loss": 0.6246867179870605,
+      "step": 896
+    },
+    {
+      "epoch": 3.226618705035971,
+      "grad_norm": 0.5722974248170735,
+      "learning_rate": 6.495293824156887e-06,
+      "loss": 0.6357770562171936,
+      "step": 897
+    },
+    {
+      "epoch": 3.2302158273381294,
+      "grad_norm": 0.55706090318583,
+      "learning_rate": 6.4885526387001654e-06,
+      "loss": 0.6066106557846069,
+      "step": 898
+    },
+    {
+      "epoch": 3.2338129496402876,
+      "grad_norm": 0.6139163839279302,
+      "learning_rate": 6.481808482984554e-06,
+      "loss": 0.6301413774490356,
+      "step": 899
+    },
+    {
+      "epoch": 3.237410071942446,
+      "grad_norm": 0.5821754793482845,
+      "learning_rate": 6.475061370467346e-06,
+      "loss": 0.6320218443870544,
+      "step": 900
+    },
+    {
+      "epoch": 3.237410071942446,
+      "eval_loss": 0.8133171200752258,
+      "eval_runtime": 7.8053,
+      "eval_samples_per_second": 11.274,
+      "eval_steps_per_second": 0.384,
+      "eval_token_acc": 0.7516627435314936,
+      "step": 900
+    },
+    {
+      "epoch": 3.241007194244604,
+      "grad_norm": 0.5433359774163814,
+      "learning_rate": 6.468311314611734e-06,
+      "loss": 0.6361285448074341,
+      "step": 901
+    },
+    {
+      "epoch": 3.2446043165467624,
+      "grad_norm": 0.5663609802203171,
+      "learning_rate": 6.461558328886786e-06,
+      "loss": 0.6213710308074951,
+      "step": 902
+    },
+    {
+      "epoch": 3.2482014388489207,
+      "grad_norm": 0.5485700818172271,
+      "learning_rate": 6.454802426767409e-06,
+      "loss": 0.6224983930587769,
+      "step": 903
+    },
+    {
+      "epoch": 3.2517985611510793,
+      "grad_norm": 0.6075827214980387,
+      "learning_rate": 6.4480436217343366e-06,
+      "loss": 0.6264669895172119,
+      "step": 904
+    },
+    {
+      "epoch": 3.2553956834532376,
+      "grad_norm": 0.5688295864017505,
+      "learning_rate": 6.4412819272740955e-06,
+      "loss": 0.6276412606239319,
+      "step": 905
+    },
+    {
+      "epoch": 3.258992805755396,
+      "grad_norm": 0.5275825790352916,
+      "learning_rate": 6.434517356878974e-06,
+      "loss": 0.6151010990142822,
+      "step": 906
+    },
+    {
+      "epoch": 3.262589928057554,
+      "grad_norm": 0.6096221075109286,
+      "learning_rate": 6.427749924046999e-06,
+      "loss": 0.6230264902114868,
+      "step": 907
+    },
+    {
+      "epoch": 3.2661870503597124,
+      "grad_norm": 0.612072472195744,
+      "learning_rate": 6.420979642281909e-06,
+      "loss": 0.6169232726097107,
+      "step": 908
+    },
+    {
+      "epoch": 3.2697841726618706,
+      "grad_norm": 0.5645657602010562,
+      "learning_rate": 6.4142065250931315e-06,
+      "loss": 0.6280378699302673,
+      "step": 909
+    },
+    {
+      "epoch": 3.273381294964029,
+      "grad_norm": 0.5445126890364774,
+      "learning_rate": 6.4074305859957475e-06,
+      "loss": 0.6251436471939087,
+      "step": 910
+    },
+    {
+      "epoch": 3.276978417266187,
+      "grad_norm": 0.6226293432430507,
+      "learning_rate": 6.40065183851047e-06,
+      "loss": 0.6066725850105286,
+      "step": 911
+    },
+    {
+      "epoch": 3.2805755395683454,
+      "grad_norm": 0.5532512858779444,
+      "learning_rate": 6.393870296163616e-06,
+      "loss": 0.6277509927749634,
+      "step": 912
+    },
+    {
+      "epoch": 3.2841726618705036,
+      "grad_norm": 0.5484197426412167,
+      "learning_rate": 6.387085972487082e-06,
+      "loss": 0.6050516366958618,
+      "step": 913
+    },
+    {
+      "epoch": 3.287769784172662,
+      "grad_norm": 0.553188339774859,
+      "learning_rate": 6.380298881018307e-06,
+      "loss": 0.6248925924301147,
+      "step": 914
+    },
+    {
+      "epoch": 3.29136690647482,
+      "grad_norm": 0.5838918969845017,
+      "learning_rate": 6.373509035300261e-06,
+      "loss": 0.6234239339828491,
+      "step": 915
+    },
+    {
+      "epoch": 3.2949640287769784,
+      "grad_norm": 0.6761408326081915,
+      "learning_rate": 6.366716448881407e-06,
+      "loss": 0.6067996025085449,
+      "step": 916
+    },
+    {
+      "epoch": 3.2985611510791366,
+      "grad_norm": 0.5789294408382947,
+      "learning_rate": 6.359921135315673e-06,
+      "loss": 0.6355147957801819,
+      "step": 917
+    },
+    {
+      "epoch": 3.302158273381295,
+      "grad_norm": 0.5703657815163174,
+      "learning_rate": 6.353123108162435e-06,
+      "loss": 0.6251250505447388,
+      "step": 918
+    },
+    {
+      "epoch": 3.305755395683453,
+      "grad_norm": 0.5705102674833985,
+      "learning_rate": 6.346322380986482e-06,
+      "loss": 0.6201918125152588,
+      "step": 919
+    },
+    {
+      "epoch": 3.3093525179856114,
+      "grad_norm": 0.5182484237658441,
+      "learning_rate": 6.339518967357985e-06,
+      "loss": 0.5968239307403564,
+      "step": 920
+    },
+    {
+      "epoch": 3.3129496402877696,
+      "grad_norm": 0.579106482961709,
+      "learning_rate": 6.332712880852483e-06,
+      "loss": 0.6315656900405884,
+      "step": 921
+    },
+    {
+      "epoch": 3.316546762589928,
+      "grad_norm": 0.5490619934499182,
+      "learning_rate": 6.325904135050844e-06,
+      "loss": 0.6207892298698425,
+      "step": 922
+    },
+    {
+      "epoch": 3.3201438848920866,
+      "grad_norm": 0.551984663606509,
+      "learning_rate": 6.319092743539244e-06,
+      "loss": 0.6102540493011475,
+      "step": 923
+    },
+    {
+      "epoch": 3.3237410071942444,
+      "grad_norm": 0.5709695659026294,
+      "learning_rate": 6.312278719909138e-06,
+      "loss": 0.6253349781036377,
+      "step": 924
+    },
+    {
+      "epoch": 3.327338129496403,
+      "grad_norm": 0.5903086917501752,
+      "learning_rate": 6.305462077757235e-06,
+      "loss": 0.6325236558914185,
+      "step": 925
+    },
+    {
+      "epoch": 3.3309352517985613,
+      "grad_norm": 0.5898347793539362,
+      "learning_rate": 6.298642830685464e-06,
+      "loss": 0.6274194717407227,
+      "step": 926
+    },
+    {
+      "epoch": 3.3345323741007196,
+      "grad_norm": 0.5419411991976139,
+      "learning_rate": 6.291820992300956e-06,
+      "loss": 0.6277546286582947,
+      "step": 927
+    },
+    {
+      "epoch": 3.338129496402878,
+      "grad_norm": 0.5990610317077253,
+      "learning_rate": 6.284996576216014e-06,
+      "loss": 0.6244072318077087,
+      "step": 928
+    },
+    {
+      "epoch": 3.341726618705036,
+      "grad_norm": 0.5900745326399537,
+      "learning_rate": 6.278169596048081e-06,
+      "loss": 0.628463625907898,
+      "step": 929
+    },
+    {
+      "epoch": 3.3453237410071943,
+      "grad_norm": 0.5480932820904987,
+      "learning_rate": 6.271340065419715e-06,
+      "loss": 0.6218065023422241,
+      "step": 930
+    },
+    {
+      "epoch": 3.3489208633093526,
+      "grad_norm": 0.6125479999978816,
+      "learning_rate": 6.2645079979585686e-06,
+      "loss": 0.6407708525657654,
+      "step": 931
+    },
+    {
+      "epoch": 3.352517985611511,
+      "grad_norm": 0.5531587910215395,
+      "learning_rate": 6.257673407297352e-06,
+      "loss": 0.6171718835830688,
+      "step": 932
+    },
+    {
+      "epoch": 3.356115107913669,
+      "grad_norm": 0.5486944909683307,
+      "learning_rate": 6.250836307073813e-06,
+      "loss": 0.6350491642951965,
+      "step": 933
+    },
+    {
+      "epoch": 3.3597122302158273,
+      "grad_norm": 0.5693874304619684,
+      "learning_rate": 6.243996710930705e-06,
+      "loss": 0.6411446332931519,
+      "step": 934
+    },
+    {
+      "epoch": 3.3633093525179856,
+      "grad_norm": 0.5756735464434138,
+      "learning_rate": 6.237154632515764e-06,
+      "loss": 0.6317604780197144,
+      "step": 935
+    },
+    {
+      "epoch": 3.366906474820144,
+      "grad_norm": 0.5792345382019256,
+      "learning_rate": 6.230310085481677e-06,
+      "loss": 0.6155068278312683,
+      "step": 936
+    },
+    {
+      "epoch": 3.370503597122302,
+      "grad_norm": 0.580886016903391,
+      "learning_rate": 6.223463083486055e-06,
+      "loss": 0.6173931360244751,
+      "step": 937
+    },
+    {
+      "epoch": 3.3741007194244603,
+      "grad_norm": 0.5970064971945658,
+      "learning_rate": 6.216613640191414e-06,
+      "loss": 0.6314873695373535,
+      "step": 938
+    },
+    {
+      "epoch": 3.3776978417266186,
+      "grad_norm": 0.5650813171755603,
+      "learning_rate": 6.209761769265133e-06,
+      "loss": 0.6283479332923889,
+      "step": 939
+    },
+    {
+      "epoch": 3.381294964028777,
+      "grad_norm": 0.5998949151653221,
+      "learning_rate": 6.2029074843794445e-06,
+      "loss": 0.6180216670036316,
+      "step": 940
+    },
+    {
+      "epoch": 3.384892086330935,
+      "grad_norm": 0.5913243467686229,
+      "learning_rate": 6.19605079921139e-06,
+      "loss": 0.621105432510376,
+      "step": 941
+    },
+    {
+      "epoch": 3.3884892086330938,
+      "grad_norm": 0.6160120518012295,
+      "learning_rate": 6.189191727442807e-06,
+      "loss": 0.639343798160553,
+      "step": 942
+    },
+    {
+      "epoch": 3.3920863309352516,
+      "grad_norm": 0.5522125574974834,
+      "learning_rate": 6.182330282760286e-06,
+      "loss": 0.651248037815094,
+      "step": 943
+    },
+    {
+      "epoch": 3.3956834532374103,
+      "grad_norm": 0.567072359350329,
+      "learning_rate": 6.175466478855161e-06,
+      "loss": 0.6345071792602539,
+      "step": 944
+    },
+    {
+      "epoch": 3.3992805755395685,
+      "grad_norm": 0.5859261607861899,
+      "learning_rate": 6.168600329423472e-06,
+      "loss": 0.6209615468978882,
+      "step": 945
+    },
+    {
+      "epoch": 3.402877697841727,
+      "grad_norm": 0.607665412086887,
+      "learning_rate": 6.1617318481659364e-06,
+      "loss": 0.6264888644218445,
+      "step": 946
+    },
+    {
+      "epoch": 3.406474820143885,
+      "grad_norm": 0.5896905998280686,
+      "learning_rate": 6.154861048787928e-06,
+      "loss": 0.6273084878921509,
+      "step": 947
+    },
+    {
+      "epoch": 3.4100719424460433,
+      "grad_norm": 0.5186817083485703,
+      "learning_rate": 6.147987944999441e-06,
+      "loss": 0.610866904258728,
+      "step": 948
+    },
+    {
+      "epoch": 3.4136690647482015,
+      "grad_norm": 0.6000730221149679,
+      "learning_rate": 6.1411125505150735e-06,
+      "loss": 0.6253103017807007,
+      "step": 949
+    },
+    {
+      "epoch": 3.41726618705036,
+      "grad_norm": 0.60799932528241,
+      "learning_rate": 6.13423487905399e-06,
+      "loss": 0.6207418441772461,
+      "step": 950
+    },
+    {
+      "epoch": 3.420863309352518,
+      "grad_norm": 0.5555330946593812,
+      "learning_rate": 6.127354944339905e-06,
+      "loss": 0.6155734658241272,
+      "step": 951
+    },
+    {
+      "epoch": 3.4244604316546763,
+      "grad_norm": 0.5767982008584605,
+      "learning_rate": 6.1204727601010396e-06,
+      "loss": 0.6276605725288391,
+      "step": 952
+    },
+    {
+      "epoch": 3.4280575539568345,
+      "grad_norm": 0.5748993900038213,
+      "learning_rate": 6.113588340070112e-06,
+      "loss": 0.6277233362197876,
+      "step": 953
+    },
+    {
+      "epoch": 3.431654676258993,
+      "grad_norm": 0.5772809250644908,
+      "learning_rate": 6.106701697984296e-06,
+      "loss": 0.6388263702392578,
+      "step": 954
+    },
+    {
+      "epoch": 3.435251798561151,
+      "grad_norm": 0.5769783761628331,
+      "learning_rate": 6.0998128475852025e-06,
+      "loss": 0.6354755163192749,
+      "step": 955
+    },
+    {
+      "epoch": 3.4388489208633093,
+      "grad_norm": 0.5424758329996289,
+      "learning_rate": 6.092921802618849e-06,
+      "loss": 0.6111853122711182,
+      "step": 956
+    },
+    {
+      "epoch": 3.4424460431654675,
+      "grad_norm": 0.5611679378503189,
+      "learning_rate": 6.0860285768356295e-06,
+      "loss": 0.6185018420219421,
+      "step": 957
+    },
+    {
+      "epoch": 3.446043165467626,
+      "grad_norm": 0.588002247832766,
+      "learning_rate": 6.079133183990292e-06,
+      "loss": 0.6287339925765991,
+      "step": 958
+    },
+    {
+      "epoch": 3.449640287769784,
+      "grad_norm": 0.7443287565839711,
+      "learning_rate": 6.072235637841909e-06,
+      "loss": 0.5987130403518677,
+      "step": 959
+    },
+    {
+      "epoch": 3.4532374100719423,
+      "grad_norm": 0.5583210492633316,
+      "learning_rate": 6.065335952153846e-06,
+      "loss": 0.6217409372329712,
+      "step": 960
+    },
+    {
+      "epoch": 3.4568345323741005,
+      "grad_norm": 0.5325265812852955,
+      "learning_rate": 6.058434140693741e-06,
+      "loss": 0.6196964979171753,
+      "step": 961
+    },
+    {
+      "epoch": 3.460431654676259,
+      "grad_norm": 0.5606450235116687,
+      "learning_rate": 6.051530217233473e-06,
+      "loss": 0.6245044469833374,
+      "step": 962
+    },
+    {
+      "epoch": 3.4640287769784175,
+      "grad_norm": 0.6416629534030027,
+      "learning_rate": 6.044624195549137e-06,
+      "loss": 0.6353148818016052,
+      "step": 963
+    },
+    {
+      "epoch": 3.4676258992805753,
+      "grad_norm": 0.549414675909668,
+      "learning_rate": 6.037716089421011e-06,
+      "loss": 0.6427045464515686,
+      "step": 964
+    },
+    {
+      "epoch": 3.471223021582734,
+      "grad_norm": 0.5653728079544905,
+      "learning_rate": 6.030805912633535e-06,
+      "loss": 0.6173840165138245,
+      "step": 965
+    },
+    {
+      "epoch": 3.4748201438848922,
+      "grad_norm": 0.5750387905226975,
+      "learning_rate": 6.023893678975281e-06,
+      "loss": 0.6136468052864075,
+      "step": 966
+    },
+    {
+      "epoch": 3.4784172661870505,
+      "grad_norm": 0.5562192543103848,
+      "learning_rate": 6.016979402238922e-06,
+      "loss": 0.6289895176887512,
+      "step": 967
+    },
+    {
+      "epoch": 3.4820143884892087,
+      "grad_norm": 0.6071855904365742,
+      "learning_rate": 6.010063096221215e-06,
+      "loss": 0.6055720448493958,
+      "step": 968
+    },
+    {
+      "epoch": 3.485611510791367,
+      "grad_norm": 0.5853664881319613,
+      "learning_rate": 6.003144774722956e-06,
+      "loss": 0.6191238164901733,
+      "step": 969
+    },
+    {
+      "epoch": 3.4892086330935252,
+      "grad_norm": 0.5836022611985129,
+      "learning_rate": 5.996224451548974e-06,
+      "loss": 0.6153513193130493,
+      "step": 970
+    },
+    {
+      "epoch": 3.4928057553956835,
+      "grad_norm": 0.5501626906244823,
+      "learning_rate": 5.989302140508081e-06,
+      "loss": 0.6205925941467285,
+      "step": 971
+    },
+    {
+      "epoch": 3.4964028776978417,
+      "grad_norm": 0.561858415194521,
+      "learning_rate": 5.982377855413063e-06,
+      "loss": 0.618821382522583,
+      "step": 972
+    },
+    {
+      "epoch": 3.5,
+      "grad_norm": 0.9366644791621612,
+      "learning_rate": 5.975451610080643e-06,
+      "loss": 0.6133970618247986,
+      "step": 973
+    },
+    {
+      "epoch": 3.5035971223021583,
+      "grad_norm": 0.5477008824327935,
+      "learning_rate": 5.968523418331453e-06,
+      "loss": 0.6296148300170898,
+      "step": 974
+    },
+    {
+      "epoch": 3.5071942446043165,
+      "grad_norm": 0.5208248633143249,
+      "learning_rate": 5.9615932939900154e-06,
+      "loss": 0.5915785431861877,
+      "step": 975
+    },
+    {
+      "epoch": 3.5107913669064748,
+      "grad_norm": 0.578508865290858,
+      "learning_rate": 5.954661250884704e-06,
+      "loss": 0.6324668526649475,
+      "step": 976
+    },
+    {
+      "epoch": 3.514388489208633,
+      "grad_norm": 0.5461540530255063,
+      "learning_rate": 5.947727302847719e-06,
+      "loss": 0.6380304098129272,
+      "step": 977
+    },
+    {
+      "epoch": 3.5179856115107913,
+      "grad_norm": 0.5362843251572532,
+      "learning_rate": 5.940791463715068e-06,
+      "loss": 0.6359459161758423,
+      "step": 978
+    },
+    {
+      "epoch": 3.5215827338129495,
+      "grad_norm": 0.5526432685264635,
+      "learning_rate": 5.933853747326525e-06,
+      "loss": 0.6332674026489258,
+      "step": 979
+    },
+    {
+      "epoch": 3.5251798561151078,
+      "grad_norm": 0.5128905428703581,
+      "learning_rate": 5.926914167525618e-06,
+      "loss": 0.6182379722595215,
+      "step": 980
+    },
+    {
+      "epoch": 3.528776978417266,
+      "grad_norm": 0.5869091984847182,
+      "learning_rate": 5.919972738159587e-06,
+      "loss": 0.6357277631759644,
+      "step": 981
+    },
+    {
+      "epoch": 3.5323741007194247,
+      "grad_norm": 0.5717959873717039,
+      "learning_rate": 5.913029473079365e-06,
+      "loss": 0.6151210069656372,
+      "step": 982
+    },
+    {
+      "epoch": 3.5359712230215825,
+      "grad_norm": 0.5775579121789572,
+      "learning_rate": 5.906084386139545e-06,
+      "loss": 0.6470330953598022,
+      "step": 983
+    },
+    {
+      "epoch": 3.539568345323741,
+      "grad_norm": 0.5474266720437985,
+      "learning_rate": 5.899137491198364e-06,
+      "loss": 0.6377556324005127,
+      "step": 984
+    },
+    {
+      "epoch": 3.543165467625899,
+      "grad_norm": 0.6664511062730858,
+      "learning_rate": 5.8921888021176545e-06,
+      "loss": 0.632561445236206,
+      "step": 985
+    },
+    {
+      "epoch": 3.5467625899280577,
+      "grad_norm": 0.5982625641511464,
+      "learning_rate": 5.885238332762837e-06,
+      "loss": 0.6204080581665039,
+      "step": 986
+    },
+    {
+      "epoch": 3.550359712230216,
+      "grad_norm": 0.5419504378083249,
+      "learning_rate": 5.878286097002884e-06,
+      "loss": 0.6141122579574585,
+      "step": 987
+    },
+    {
+      "epoch": 3.553956834532374,
+      "grad_norm": 0.526761312287397,
+      "learning_rate": 5.871332108710292e-06,
+      "loss": 0.6195293068885803,
+      "step": 988
+    },
+    {
+      "epoch": 3.5575539568345325,
+      "grad_norm": 0.5560617796758003,
+      "learning_rate": 5.86437638176105e-06,
+      "loss": 0.6226100921630859,
+      "step": 989
+    },
+    {
+      "epoch": 3.5611510791366907,
+      "grad_norm": 0.5147824798221119,
+      "learning_rate": 5.8574189300346215e-06,
+      "loss": 0.5921021103858948,
+      "step": 990
+    },
+    {
+      "epoch": 3.564748201438849,
+      "grad_norm": 0.5976606416916541,
+      "learning_rate": 5.8504597674139115e-06,
+      "loss": 0.6294490098953247,
+      "step": 991
+    },
+    {
+      "epoch": 3.568345323741007,
+      "grad_norm": 0.5632826758618952,
+      "learning_rate": 5.843498907785236e-06,
+      "loss": 0.6351437568664551,
+      "step": 992
+    },
+    {
+      "epoch": 3.5719424460431655,
+      "grad_norm": 0.5509722038236922,
+      "learning_rate": 5.8365363650383e-06,
+      "loss": 0.6352639198303223,
+      "step": 993
+    },
+    {
+      "epoch": 3.5755395683453237,
+      "grad_norm": 0.5524669379370264,
+      "learning_rate": 5.829572153066166e-06,
+      "loss": 0.6275777816772461,
+      "step": 994
+    },
+    {
+      "epoch": 3.579136690647482,
+      "grad_norm": 0.5675598136381055,
+      "learning_rate": 5.8226062857652254e-06,
+      "loss": 0.622998833656311,
+      "step": 995
+    },
+    {
+      "epoch": 3.58273381294964,
+      "grad_norm": 0.5104403543711602,
+      "learning_rate": 5.815638777035175e-06,
+      "loss": 0.6314291954040527,
+      "step": 996
+    },
+    {
+      "epoch": 3.5863309352517985,
+      "grad_norm": 0.5122673876452001,
+      "learning_rate": 5.808669640778986e-06,
+      "loss": 0.6079280376434326,
+      "step": 997
+    },
+    {
+      "epoch": 3.5899280575539567,
+      "grad_norm": 0.5581467016360934,
+      "learning_rate": 5.801698890902877e-06,
+      "loss": 0.6032590866088867,
+      "step": 998
+    },
+    {
+      "epoch": 3.593525179856115,
+      "grad_norm": 0.5438226946507215,
+      "learning_rate": 5.794726541316288e-06,
+      "loss": 0.6497535109519958,
+      "step": 999
+    },
+    {
+      "epoch": 3.597122302158273,
+      "grad_norm": 0.5832843430696965,
+      "learning_rate": 5.78775260593185e-06,
+      "loss": 0.6055078506469727,
+      "step": 1000
+    },
+    {
+      "epoch": 3.597122302158273,
+      "eval_loss": 0.8125540614128113,
+      "eval_runtime": 7.8509,
+      "eval_samples_per_second": 11.209,
+      "eval_steps_per_second": 0.382,
+      "eval_token_acc": 0.7516720769220718,
+      "step": 1000
+    },
+    {
+      "epoch": 3.600719424460432,
+      "grad_norm": 0.5387037791518224,
+      "learning_rate": 5.780777098665357e-06,
+      "loss": 0.6361275911331177,
+      "step": 1001
+    },
+    {
+      "epoch": 3.6043165467625897,
+      "grad_norm": 0.5931750766720936,
+      "learning_rate": 5.77380003343574e-06,
+      "loss": 0.6325686573982239,
+      "step": 1002
+    },
+    {
+      "epoch": 3.6079136690647484,
+      "grad_norm": 0.5683590800729738,
+      "learning_rate": 5.766821424165039e-06,
+      "loss": 0.6415737271308899,
+      "step": 1003
+    },
+    {
+      "epoch": 3.6115107913669062,
+      "grad_norm": 0.5261138010028708,
+      "learning_rate": 5.759841284778379e-06,
+      "loss": 0.6264912486076355,
+      "step": 1004
+    },
+    {
+      "epoch": 3.615107913669065,
+      "grad_norm": 0.544748136565225,
+      "learning_rate": 5.752859629203933e-06,
+      "loss": 0.6192924380302429,
+      "step": 1005
+    },
+    {
+      "epoch": 3.618705035971223,
+      "grad_norm": 0.6841535562931874,
+      "learning_rate": 5.7458764713729e-06,
+      "loss": 0.6135982275009155,
+      "step": 1006
+    },
+    {
+      "epoch": 3.6223021582733814,
+      "grad_norm": 0.5757825111921732,
+      "learning_rate": 5.738891825219477e-06,
+      "loss": 0.6414120197296143,
+      "step": 1007
+    },
+    {
+      "epoch": 3.6258992805755397,
+      "grad_norm": 0.834185194411395,
+      "learning_rate": 5.731905704680834e-06,
+      "loss": 0.6300562620162964,
+      "step": 1008
+    },
+    {
+      "epoch": 3.629496402877698,
+      "grad_norm": 0.5914648938614046,
+      "learning_rate": 5.7249181236970774e-06,
+      "loss": 0.6344118714332581,
+      "step": 1009
+    },
+    {
+      "epoch": 3.633093525179856,
+      "grad_norm": 0.5837957271398493,
+      "learning_rate": 5.717929096211232e-06,
+      "loss": 0.6232997179031372,
+      "step": 1010
+    },
+    {
+      "epoch": 3.6366906474820144,
+      "grad_norm": 0.6949256014787403,
+      "learning_rate": 5.710938636169211e-06,
+      "loss": 0.6331793069839478,
+      "step": 1011
+    },
+    {
+      "epoch": 3.6402877697841727,
+      "grad_norm": 0.533156956369847,
+      "learning_rate": 5.703946757519777e-06,
+      "loss": 0.6080916523933411,
+      "step": 1012
+    },
+    {
+      "epoch": 3.643884892086331,
+      "grad_norm": 0.571944227267364,
+      "learning_rate": 5.6969534742145325e-06,
+      "loss": 0.6211006045341492,
+      "step": 1013
+    },
+    {
+      "epoch": 3.647482014388489,
+      "grad_norm": 0.6397571875041864,
+      "learning_rate": 5.689958800207879e-06,
+      "loss": 0.6273400187492371,
+      "step": 1014
+    },
+    {
+      "epoch": 3.6510791366906474,
+      "grad_norm": 0.5689459230735638,
+      "learning_rate": 5.682962749456994e-06,
+      "loss": 0.6134950518608093,
+      "step": 1015
+    },
+    {
+      "epoch": 3.6546762589928057,
+      "grad_norm": 0.5569625906726352,
+      "learning_rate": 5.6759653359218e-06,
+      "loss": 0.611720085144043,
+      "step": 1016
+    },
+    {
+      "epoch": 3.658273381294964,
+      "grad_norm": 0.5368099011844354,
+      "learning_rate": 5.668966573564942e-06,
+      "loss": 0.6366114616394043,
+      "step": 1017
+    },
+    {
+      "epoch": 3.661870503597122,
+      "grad_norm": 0.5636975451024273,
+      "learning_rate": 5.661966476351756e-06,
+      "loss": 0.6357487440109253,
+      "step": 1018
+    },
+    {
+      "epoch": 3.6654676258992804,
+      "grad_norm": 0.5381342570956121,
+      "learning_rate": 5.6549650582502366e-06,
+      "loss": 0.621475875377655,
+      "step": 1019
+    },
+    {
+      "epoch": 3.6690647482014387,
+      "grad_norm": 0.5680718171486167,
+      "learning_rate": 5.647962333231021e-06,
+      "loss": 0.620261549949646,
+      "step": 1020
+    },
+    {
+      "epoch": 3.672661870503597,
+      "grad_norm": 0.5382743912552496,
+      "learning_rate": 5.640958315267351e-06,
+      "loss": 0.650659441947937,
+      "step": 1021
+    },
+    {
+      "epoch": 3.6762589928057556,
+      "grad_norm": 0.5569716148721183,
+      "learning_rate": 5.633953018335047e-06,
+      "loss": 0.6229879856109619,
+      "step": 1022
+    },
+    {
+      "epoch": 3.6798561151079134,
+      "grad_norm": 0.5749593410999209,
+      "learning_rate": 5.626946456412484e-06,
+      "loss": 0.6209127902984619,
+      "step": 1023
+    },
+    {
+      "epoch": 3.683453237410072,
+      "grad_norm": 0.5456231050875293,
+      "learning_rate": 5.6199386434805615e-06,
+      "loss": 0.6160396337509155,
+      "step": 1024
+    },
+    {
+      "epoch": 3.68705035971223,
+      "grad_norm": 0.5289471769079246,
+      "learning_rate": 5.612929593522671e-06,
+      "loss": 0.6104173064231873,
+      "step": 1025
+    },
+    {
+      "epoch": 3.6906474820143886,
+      "grad_norm": 0.5843671396747508,
+      "learning_rate": 5.605919320524677e-06,
+      "loss": 0.6159001588821411,
+      "step": 1026
+    },
+    {
+      "epoch": 3.694244604316547,
+      "grad_norm": 0.5651476004357376,
+      "learning_rate": 5.598907838474883e-06,
+      "loss": 0.6387501955032349,
+      "step": 1027
+    },
+    {
+      "epoch": 3.697841726618705,
+      "grad_norm": 0.5706908322156868,
+      "learning_rate": 5.591895161364006e-06,
+      "loss": 0.6199760437011719,
+      "step": 1028
+    },
+    {
+      "epoch": 3.7014388489208634,
+      "grad_norm": 0.5322410609869666,
+      "learning_rate": 5.584881303185142e-06,
+      "loss": 0.6229681372642517,
+      "step": 1029
+    },
+    {
+      "epoch": 3.7050359712230216,
+      "grad_norm": 0.5721084026652471,
+      "learning_rate": 5.577866277933752e-06,
+      "loss": 0.619432270526886,
+      "step": 1030
+    },
+    {
+      "epoch": 3.70863309352518,
+      "grad_norm": 0.5264586525019622,
+      "learning_rate": 5.570850099607621e-06,
+      "loss": 0.6281554698944092,
+      "step": 1031
+    },
+    {
+      "epoch": 3.712230215827338,
+      "grad_norm": 0.5603922534581942,
+      "learning_rate": 5.563832782206835e-06,
+      "loss": 0.6256821155548096,
+      "step": 1032
+    },
+    {
+      "epoch": 3.7158273381294964,
+      "grad_norm": 0.561218503029973,
+      "learning_rate": 5.556814339733754e-06,
+      "loss": 0.6172206401824951,
+      "step": 1033
+    },
+    {
+      "epoch": 3.7194244604316546,
+      "grad_norm": 0.5800927690024219,
+      "learning_rate": 5.549794786192984e-06,
+      "loss": 0.6275442838668823,
+      "step": 1034
+    },
+    {
+      "epoch": 3.723021582733813,
+      "grad_norm": 0.5703911890599143,
+      "learning_rate": 5.542774135591344e-06,
+      "loss": 0.6432952880859375,
+      "step": 1035
+    },
+    {
+      "epoch": 3.726618705035971,
+      "grad_norm": 0.5288906768093431,
+      "learning_rate": 5.535752401937846e-06,
+      "loss": 0.6199349164962769,
+      "step": 1036
+    },
+    {
+      "epoch": 3.7302158273381294,
+      "grad_norm": 0.5526986942953863,
+      "learning_rate": 5.528729599243662e-06,
+      "loss": 0.634512722492218,
+      "step": 1037
+    },
+    {
+      "epoch": 3.7338129496402876,
+      "grad_norm": 0.5338421783071607,
+      "learning_rate": 5.521705741522096e-06,
+      "loss": 0.6207709312438965,
+      "step": 1038
+    },
+    {
+      "epoch": 3.737410071942446,
+      "grad_norm": 0.5441886221006373,
+      "learning_rate": 5.51468084278856e-06,
+      "loss": 0.6194617748260498,
+      "step": 1039
+    },
+    {
+      "epoch": 3.741007194244604,
+      "grad_norm": 0.57751833674519,
+      "learning_rate": 5.507654917060541e-06,
+      "loss": 0.6412806510925293,
+      "step": 1040
+    },
+    {
+      "epoch": 3.744604316546763,
+      "grad_norm": 0.5480165246411622,
+      "learning_rate": 5.500627978357576e-06,
+      "loss": 0.6190841197967529,
+      "step": 1041
+    },
+    {
+      "epoch": 3.7482014388489207,
+      "grad_norm": 0.5486370930882859,
+      "learning_rate": 5.49360004070122e-06,
+      "loss": 0.6254152059555054,
+      "step": 1042
+    },
+    {
+      "epoch": 3.7517985611510793,
+      "grad_norm": 0.5879050372467775,
+      "learning_rate": 5.486571118115026e-06,
+      "loss": 0.6343721151351929,
+      "step": 1043
+    },
+    {
+      "epoch": 3.755395683453237,
+      "grad_norm": 0.5370445690132445,
+      "learning_rate": 5.4795412246245126e-06,
+      "loss": 0.6343209743499756,
+      "step": 1044
+    },
+    {
+      "epoch": 3.758992805755396,
+      "grad_norm": 0.5450154530597571,
+      "learning_rate": 5.472510374257133e-06,
+      "loss": 0.6196166276931763,
+      "step": 1045
+    },
+    {
+      "epoch": 3.762589928057554,
+      "grad_norm": 0.5343709806843335,
+      "learning_rate": 5.465478581042248e-06,
+      "loss": 0.6157248616218567,
+      "step": 1046
+    },
+    {
+      "epoch": 3.7661870503597124,
+      "grad_norm": 0.5628459523993931,
+      "learning_rate": 5.458445859011105e-06,
+      "loss": 0.6136109828948975,
+      "step": 1047
+    },
+    {
+      "epoch": 3.7697841726618706,
+      "grad_norm": 0.5565016825347403,
+      "learning_rate": 5.451412222196801e-06,
+      "loss": 0.6306936144828796,
+      "step": 1048
+    },
+    {
+      "epoch": 3.773381294964029,
+      "grad_norm": 0.5580290826020277,
+      "learning_rate": 5.44437768463426e-06,
+      "loss": 0.6189572215080261,
+      "step": 1049
+    },
+    {
+      "epoch": 3.776978417266187,
+      "grad_norm": 0.5431193983755793,
+      "learning_rate": 5.4373422603602025e-06,
+      "loss": 0.6176700592041016,
+      "step": 1050
+    },
+    {
+      "epoch": 3.7805755395683454,
+      "grad_norm": 0.5397695802779046,
+      "learning_rate": 5.4303059634131175e-06,
+      "loss": 0.6354320049285889,
+      "step": 1051
+    },
+    {
+      "epoch": 3.7841726618705036,
+      "grad_norm": 0.5417952730271903,
+      "learning_rate": 5.4232688078332384e-06,
+      "loss": 0.6268483400344849,
+      "step": 1052
+    },
+    {
+      "epoch": 3.787769784172662,
+      "grad_norm": 0.577453991046705,
+      "learning_rate": 5.416230807662509e-06,
+      "loss": 0.6346728205680847,
+      "step": 1053
+    },
+    {
+      "epoch": 3.79136690647482,
+      "grad_norm": 0.5315960494904669,
+      "learning_rate": 5.409191976944559e-06,
+      "loss": 0.6320599317550659,
+      "step": 1054
+    },
+    {
+      "epoch": 3.7949640287769784,
+      "grad_norm": 0.5438905695079264,
+      "learning_rate": 5.402152329724675e-06,
+      "loss": 0.6306254863739014,
+      "step": 1055
+    },
+    {
+      "epoch": 3.7985611510791366,
+      "grad_norm": 0.5530900019557347,
+      "learning_rate": 5.395111880049775e-06,
+      "loss": 0.6188409924507141,
+      "step": 1056
+    },
+    {
+      "epoch": 3.802158273381295,
+      "grad_norm": 0.6056617368798535,
+      "learning_rate": 5.388070641968376e-06,
+      "loss": 0.6206061840057373,
+      "step": 1057
+    },
+    {
+      "epoch": 3.805755395683453,
+      "grad_norm": 0.5472356989518676,
+      "learning_rate": 5.3810286295305715e-06,
+      "loss": 0.6444727182388306,
+      "step": 1058
+    },
+    {
+      "epoch": 3.8093525179856114,
+      "grad_norm": 0.5910357756320749,
+      "learning_rate": 5.373985856787993e-06,
+      "loss": 0.6342467069625854,
+      "step": 1059
+    },
+    {
+      "epoch": 3.81294964028777,
+      "grad_norm": 0.6026061719511084,
+      "learning_rate": 5.366942337793798e-06,
+      "loss": 0.6200609803199768,
+      "step": 1060
+    },
+    {
+      "epoch": 3.816546762589928,
+      "grad_norm": 0.5898403550967662,
+      "learning_rate": 5.359898086602625e-06,
+      "loss": 0.6159372329711914,
+      "step": 1061
+    },
+    {
+      "epoch": 3.8201438848920866,
+      "grad_norm": 0.5481084757987922,
+      "learning_rate": 5.352853117270581e-06,
+      "loss": 0.6328880190849304,
+      "step": 1062
+    },
+    {
+      "epoch": 3.8237410071942444,
+      "grad_norm": 0.5511071790400082,
+      "learning_rate": 5.3458074438552e-06,
+      "loss": 0.6281937956809998,
+      "step": 1063
+    },
+    {
+      "epoch": 3.827338129496403,
+      "grad_norm": 0.5714388926162269,
+      "learning_rate": 5.338761080415425e-06,
+      "loss": 0.6304974555969238,
+      "step": 1064
+    },
+    {
+      "epoch": 3.8309352517985613,
+      "grad_norm": 0.5751960650810334,
+      "learning_rate": 5.331714041011571e-06,
+      "loss": 0.6289598941802979,
+      "step": 1065
+    },
+    {
+      "epoch": 3.8345323741007196,
+      "grad_norm": 0.5602235046145158,
+      "learning_rate": 5.324666339705307e-06,
+      "loss": 0.6236809492111206,
+      "step": 1066
+    },
+    {
+      "epoch": 3.838129496402878,
+      "grad_norm": 0.600074702160604,
+      "learning_rate": 5.317617990559623e-06,
+      "loss": 0.6230597496032715,
+      "step": 1067
+    },
+    {
+      "epoch": 3.841726618705036,
+      "grad_norm": 0.5296383380122605,
+      "learning_rate": 5.310569007638795e-06,
+      "loss": 0.6161667704582214,
+      "step": 1068
+    },
+    {
+      "epoch": 3.8453237410071943,
+      "grad_norm": 0.5612028308860264,
+      "learning_rate": 5.303519405008373e-06,
+      "loss": 0.6226469874382019,
+      "step": 1069
+    },
+    {
+      "epoch": 3.8489208633093526,
+      "grad_norm": 0.5521493512307732,
+      "learning_rate": 5.296469196735134e-06,
+      "loss": 0.622602105140686,
+      "step": 1070
+    },
+    {
+      "epoch": 3.852517985611511,
+      "grad_norm": 0.5462130844151715,
+      "learning_rate": 5.28941839688707e-06,
+      "loss": 0.6175041198730469,
+      "step": 1071
+    },
+    {
+      "epoch": 3.856115107913669,
+      "grad_norm": 0.5856748118827945,
+      "learning_rate": 5.28236701953335e-06,
+      "loss": 0.6236345171928406,
+      "step": 1072
+    },
+    {
+      "epoch": 3.8597122302158273,
+      "grad_norm": 0.5596463937851168,
+      "learning_rate": 5.2753150787443e-06,
+      "loss": 0.631768524646759,
+      "step": 1073
+    },
+    {
+      "epoch": 3.8633093525179856,
+      "grad_norm": 0.5934140299026798,
+      "learning_rate": 5.268262588591364e-06,
+      "loss": 0.6321205496788025,
+      "step": 1074
+    },
+    {
+      "epoch": 3.866906474820144,
+      "grad_norm": 0.5178313892634216,
+      "learning_rate": 5.261209563147088e-06,
+      "loss": 0.6305399537086487,
+      "step": 1075
+    },
+    {
+      "epoch": 3.870503597122302,
+      "grad_norm": 0.5380795134593158,
+      "learning_rate": 5.254156016485082e-06,
+      "loss": 0.6334234476089478,
+      "step": 1076
+    },
+    {
+      "epoch": 3.8741007194244603,
+      "grad_norm": 0.5555020176810418,
+      "learning_rate": 5.247101962679998e-06,
+      "loss": 0.6309899091720581,
+      "step": 1077
+    },
+    {
+      "epoch": 3.8776978417266186,
+      "grad_norm": 0.5979700486904005,
+      "learning_rate": 5.240047415807499e-06,
+      "loss": 0.6348890066146851,
+      "step": 1078
+    },
+    {
+      "epoch": 3.881294964028777,
+      "grad_norm": 0.5496978553199497,
+      "learning_rate": 5.2329923899442325e-06,
+      "loss": 0.6270749568939209,
+      "step": 1079
+    },
+    {
+      "epoch": 3.884892086330935,
+      "grad_norm": 0.5400189981264519,
+      "learning_rate": 5.225936899167803e-06,
+      "loss": 0.6539863348007202,
+      "step": 1080
+    },
+    {
+      "epoch": 3.8884892086330938,
+      "grad_norm": 0.553113550788483,
+      "learning_rate": 5.21888095755674e-06,
+      "loss": 0.6327900886535645,
+      "step": 1081
+    },
+    {
+      "epoch": 3.8920863309352516,
+      "grad_norm": 0.5478545536226945,
+      "learning_rate": 5.211824579190473e-06,
+      "loss": 0.6178438663482666,
+      "step": 1082
+    },
+    {
+      "epoch": 3.8956834532374103,
+      "grad_norm": 0.547517202573347,
+      "learning_rate": 5.2047677781493035e-06,
+      "loss": 0.6254318952560425,
+      "step": 1083
+    },
+    {
+      "epoch": 3.899280575539568,
+      "grad_norm": 0.5502866887115144,
+      "learning_rate": 5.197710568514381e-06,
+      "loss": 0.6129550337791443,
+      "step": 1084
+    },
+    {
+      "epoch": 3.902877697841727,
+      "grad_norm": 0.5289111949450003,
+      "learning_rate": 5.190652964367662e-06,
+      "loss": 0.6353284120559692,
+      "step": 1085
+    },
+    {
+      "epoch": 3.906474820143885,
+      "grad_norm": 0.5697558816611187,
+      "learning_rate": 5.183594979791896e-06,
+      "loss": 0.6242085695266724,
+      "step": 1086
+    },
+    {
+      "epoch": 3.9100719424460433,
+      "grad_norm": 0.5276051579756038,
+      "learning_rate": 5.1765366288705896e-06,
+      "loss": 0.6136658191680908,
+      "step": 1087
+    },
+    {
+      "epoch": 3.9136690647482015,
+      "grad_norm": 0.5392013067435808,
+      "learning_rate": 5.169477925687981e-06,
+      "loss": 0.6261371374130249,
+      "step": 1088
+    },
+    {
+      "epoch": 3.91726618705036,
+      "grad_norm": 0.5252931018787899,
+      "learning_rate": 5.1624188843290115e-06,
+      "loss": 0.6383857727050781,
+      "step": 1089
+    },
+    {
+      "epoch": 3.920863309352518,
+      "grad_norm": 0.5447213918855934,
+      "learning_rate": 5.155359518879295e-06,
+      "loss": 0.6328091621398926,
+      "step": 1090
+    },
+    {
+      "epoch": 3.9244604316546763,
+      "grad_norm": 0.5302972618495836,
+      "learning_rate": 5.148299843425095e-06,
+      "loss": 0.6408435702323914,
+      "step": 1091
+    },
+    {
+      "epoch": 3.9280575539568345,
+      "grad_norm": 0.5190325944565576,
+      "learning_rate": 5.141239872053294e-06,
+      "loss": 0.6295177936553955,
+      "step": 1092
+    },
+    {
+      "epoch": 3.931654676258993,
+      "grad_norm": 0.5383200426474236,
+      "learning_rate": 5.134179618851361e-06,
+      "loss": 0.6353203058242798,
+      "step": 1093
+    },
+    {
+      "epoch": 3.935251798561151,
+      "grad_norm": 0.5315118968640762,
+      "learning_rate": 5.1271190979073296e-06,
+      "loss": 0.6318327188491821,
+      "step": 1094
+    },
+    {
+      "epoch": 3.9388489208633093,
+      "grad_norm": 0.517717860063696,
+      "learning_rate": 5.120058323309769e-06,
+      "loss": 0.6413143873214722,
+      "step": 1095
+    },
+    {
+      "epoch": 3.9424460431654675,
+      "grad_norm": 0.5448478306910934,
+      "learning_rate": 5.112997309147753e-06,
+      "loss": 0.6358555555343628,
+      "step": 1096
+    },
+    {
+      "epoch": 3.946043165467626,
+      "grad_norm": 0.5503703443156099,
+      "learning_rate": 5.105936069510835e-06,
+      "loss": 0.6255465745925903,
+      "step": 1097
+    },
+    {
+      "epoch": 3.949640287769784,
+      "grad_norm": 0.5332494995645366,
+      "learning_rate": 5.098874618489017e-06,
+      "loss": 0.6314849257469177,
+      "step": 1098
+    },
+    {
+      "epoch": 3.9532374100719423,
+      "grad_norm": 0.5462920590900435,
+      "learning_rate": 5.0918129701727205e-06,
+      "loss": 0.6063763499259949,
+      "step": 1099
+    },
+    {
+      "epoch": 3.956834532374101,
+      "grad_norm": 0.595376308125965,
+      "learning_rate": 5.084751138652764e-06,
+      "loss": 0.6303770542144775,
+      "step": 1100
+    },
+    {
+      "epoch": 3.956834532374101,
+      "eval_loss": 0.8109321594238281,
+      "eval_runtime": 7.8833,
+      "eval_samples_per_second": 11.163,
+      "eval_steps_per_second": 0.381,
+      "eval_token_acc": 0.751970745420572,
+      "step": 1100
+    },
+    {
+      "epoch": 3.960431654676259,
+      "grad_norm": 0.5545223372673511,
+      "learning_rate": 5.077689138020331e-06,
+      "loss": 0.636076807975769,
+      "step": 1101
+    },
+    {
+      "epoch": 3.9640287769784175,
+      "grad_norm": 0.5415888886715428,
+      "learning_rate": 5.070626982366943e-06,
+      "loss": 0.6309289932250977,
+      "step": 1102
+    },
+    {
+      "epoch": 3.9676258992805753,
+      "grad_norm": 0.530195768532205,
+      "learning_rate": 5.063564685784428e-06,
+      "loss": 0.6334230303764343,
+      "step": 1103
+    },
+    {
+      "epoch": 3.971223021582734,
+      "grad_norm": 0.5264957915760434,
+      "learning_rate": 5.0565022623649e-06,
+      "loss": 0.6209136843681335,
+      "step": 1104
+    },
+    {
+      "epoch": 3.9748201438848922,
+      "grad_norm": 0.5675717547938626,
+      "learning_rate": 5.04943972620072e-06,
+      "loss": 0.6466224193572998,
+      "step": 1105
+    },
+    {
+      "epoch": 3.9784172661870505,
+      "grad_norm": 0.546893892458039,
+      "learning_rate": 5.0423770913844795e-06,
+      "loss": 0.6209077835083008,
+      "step": 1106
+    },
+    {
+      "epoch": 3.9820143884892087,
+      "grad_norm": 0.566904391407611,
+      "learning_rate": 5.035314372008962e-06,
+      "loss": 0.6295535564422607,
+      "step": 1107
+    },
+    {
+      "epoch": 3.985611510791367,
+      "grad_norm": 0.5112849932312409,
+      "learning_rate": 5.028251582167126e-06,
+      "loss": 0.6494243144989014,
+      "step": 1108
+    },
+    {
+      "epoch": 3.9892086330935252,
+      "grad_norm": 0.5445209737712887,
+      "learning_rate": 5.021188735952065e-06,
+      "loss": 0.6363496780395508,
+      "step": 1109
+    },
+    {
+      "epoch": 3.9928057553956835,
+      "grad_norm": 0.5196093708153622,
+      "learning_rate": 5.0141258474569855e-06,
+      "loss": 0.6188162565231323,
+      "step": 1110
+    },
+    {
+      "epoch": 3.9964028776978417,
+      "grad_norm": 0.5349441127307514,
+      "learning_rate": 5.0070629307751816e-06,
+      "loss": 0.6086728572845459,
+      "step": 1111
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7632014833912022,
+      "learning_rate": 5e-06,
+      "loss": 0.6367975473403931,
+      "step": 1112
+    },
+    {
+      "epoch": 4.003597122302159,
+      "grad_norm": 0.6671812832280104,
+      "learning_rate": 4.992937069224818e-06,
+      "loss": 0.5751312971115112,
+      "step": 1113
+    },
+    {
+      "epoch": 4.0071942446043165,
+      "grad_norm": 0.6179882227804654,
+      "learning_rate": 4.985874152543015e-06,
+      "loss": 0.6045902967453003,
+      "step": 1114
+    },
+    {
+      "epoch": 4.010791366906475,
+      "grad_norm": 0.5961266178839438,
+      "learning_rate": 4.978811264047936e-06,
+      "loss": 0.5806431174278259,
+      "step": 1115
+    },
+    {
+      "epoch": 4.014388489208633,
+      "grad_norm": 0.5691235961486315,
+      "learning_rate": 4.971748417832876e-06,
+      "loss": 0.566252589225769,
+      "step": 1116
+    },
+    {
+      "epoch": 4.017985611510792,
+      "grad_norm": 0.5791093646172678,
+      "learning_rate": 4.964685627991038e-06,
+      "loss": 0.5949286818504333,
+      "step": 1117
+    },
+    {
+      "epoch": 4.0215827338129495,
+      "grad_norm": 0.5771818604656702,
+      "learning_rate": 4.957622908615522e-06,
+      "loss": 0.5870711803436279,
+      "step": 1118
+    },
+    {
+      "epoch": 4.025179856115108,
+      "grad_norm": 0.6035369572670033,
+      "learning_rate": 4.9505602737992805e-06,
+      "loss": 0.5918598175048828,
+      "step": 1119
+    },
+    {
+      "epoch": 4.028776978417266,
+      "grad_norm": 0.6065311525715938,
+      "learning_rate": 4.943497737635103e-06,
+      "loss": 0.5854548215866089,
+      "step": 1120
+    },
+    {
+      "epoch": 4.032374100719425,
+      "grad_norm": 0.5894501705087571,
+      "learning_rate": 4.936435314215573e-06,
+      "loss": 0.5900008082389832,
+      "step": 1121
+    },
+    {
+      "epoch": 4.0359712230215825,
+      "grad_norm": 0.5343328098237431,
+      "learning_rate": 4.92937301763306e-06,
+      "loss": 0.5490322113037109,
+      "step": 1122
+    },
+    {
+      "epoch": 4.039568345323741,
+      "grad_norm": 0.5705647249862055,
+      "learning_rate": 4.9223108619796705e-06,
+      "loss": 0.5875539779663086,
+      "step": 1123
+    },
+    {
+      "epoch": 4.043165467625899,
+      "grad_norm": 0.5943474698787913,
+      "learning_rate": 4.915248861347239e-06,
+      "loss": 0.5715224742889404,
+      "step": 1124
+    },
+    {
+      "epoch": 4.046762589928058,
+      "grad_norm": 0.5608640103822211,
+      "learning_rate": 4.908187029827282e-06,
+      "loss": 0.59614098072052,
+      "step": 1125
+    },
+    {
+      "epoch": 4.0503597122302155,
+      "grad_norm": 0.5736354228437628,
+      "learning_rate": 4.901125381510986e-06,
+      "loss": 0.5507819652557373,
+      "step": 1126
+    },
+    {
+      "epoch": 4.053956834532374,
+      "grad_norm": 0.5809586226091275,
+      "learning_rate": 4.894063930489166e-06,
+      "loss": 0.5653989315032959,
+      "step": 1127
+    },
+    {
+      "epoch": 4.057553956834532,
+      "grad_norm": 0.6151642417224333,
+      "learning_rate": 4.887002690852249e-06,
+      "loss": 0.5762919187545776,
+      "step": 1128
+    },
+    {
+      "epoch": 4.061151079136691,
+      "grad_norm": 0.5502446261477949,
+      "learning_rate": 4.8799416766902315e-06,
+      "loss": 0.5796780586242676,
+      "step": 1129
+    },
+    {
+      "epoch": 4.0647482014388485,
+      "grad_norm": 0.5658259411352802,
+      "learning_rate": 4.872880902092673e-06,
+      "loss": 0.5663573145866394,
+      "step": 1130
+    },
+    {
+      "epoch": 4.068345323741007,
+      "grad_norm": 0.5470757908122678,
+      "learning_rate": 4.865820381148641e-06,
+      "loss": 0.5743259191513062,
+      "step": 1131
+    },
+    {
+      "epoch": 4.071942446043165,
+      "grad_norm": 0.5785124023334323,
+      "learning_rate": 4.858760127946707e-06,
+      "loss": 0.5840204954147339,
+      "step": 1132
+    },
+    {
+      "epoch": 4.075539568345324,
+      "grad_norm": 0.59689209556221,
+      "learning_rate": 4.8517001565749065e-06,
+      "loss": 0.5999338626861572,
+      "step": 1133
+    },
+    {
+      "epoch": 4.079136690647482,
+      "grad_norm": 0.5647273424274459,
+      "learning_rate": 4.8446404811207055e-06,
+      "loss": 0.565536379814148,
+      "step": 1134
+    },
+    {
+      "epoch": 4.08273381294964,
+      "grad_norm": 0.561464772321604,
+      "learning_rate": 4.837581115670991e-06,
+      "loss": 0.588786244392395,
+      "step": 1135
+    },
+    {
+      "epoch": 4.086330935251799,
+      "grad_norm": 0.5804342083767994,
+      "learning_rate": 4.830522074312019e-06,
+      "loss": 0.5741912126541138,
+      "step": 1136
+    },
+    {
+      "epoch": 4.089928057553957,
+      "grad_norm": 0.5752775990581241,
+      "learning_rate": 4.823463371129411e-06,
+      "loss": 0.5821037292480469,
+      "step": 1137
+    },
+    {
+      "epoch": 4.093525179856115,
+      "grad_norm": 0.5494215057833093,
+      "learning_rate": 4.816405020208105e-06,
+      "loss": 0.5683557391166687,
+      "step": 1138
+    },
+    {
+      "epoch": 4.097122302158273,
+      "grad_norm": 0.5819120226688422,
+      "learning_rate": 4.80934703563234e-06,
+      "loss": 0.5606533885002136,
+      "step": 1139
+    },
+    {
+      "epoch": 4.100719424460432,
+      "grad_norm": 0.5665547666948522,
+      "learning_rate": 4.80228943148562e-06,
+      "loss": 0.559918999671936,
+      "step": 1140
+    },
+    {
+      "epoch": 4.10431654676259,
+      "grad_norm": 0.5671885384831782,
+      "learning_rate": 4.795232221850697e-06,
+      "loss": 0.574749767780304,
+      "step": 1141
+    },
+    {
+      "epoch": 4.107913669064748,
+      "grad_norm": 0.5802314387975677,
+      "learning_rate": 4.788175420809528e-06,
+      "loss": 0.5675938725471497,
+      "step": 1142
+    },
+    {
+      "epoch": 4.111510791366906,
+      "grad_norm": 0.5569385712485541,
+      "learning_rate": 4.781119042443263e-06,
+      "loss": 0.5684279203414917,
+      "step": 1143
+    },
+    {
+      "epoch": 4.115107913669065,
+      "grad_norm": 0.5543611755367137,
+      "learning_rate": 4.774063100832199e-06,
+      "loss": 0.5701501369476318,
+      "step": 1144
+    },
+    {
+      "epoch": 4.118705035971223,
+      "grad_norm": 0.545108517220662,
+      "learning_rate": 4.76700761005577e-06,
+      "loss": 0.5616727471351624,
+      "step": 1145
+    },
+    {
+      "epoch": 4.122302158273381,
+      "grad_norm": 0.5583632998054106,
+      "learning_rate": 4.759952584192502e-06,
+      "loss": 0.585227370262146,
+      "step": 1146
+    },
+    {
+      "epoch": 4.125899280575539,
+      "grad_norm": 0.5385461358182791,
+      "learning_rate": 4.752898037320004e-06,
+      "loss": 0.5916910767555237,
+      "step": 1147
+    },
+    {
+      "epoch": 4.129496402877698,
+      "grad_norm": 0.5411752970794544,
+      "learning_rate": 4.745843983514919e-06,
+      "loss": 0.5454212427139282,
+      "step": 1148
+    },
+    {
+      "epoch": 4.133093525179856,
+      "grad_norm": 0.5378753096176189,
+      "learning_rate": 4.738790436852914e-06,
+      "loss": 0.5978329181671143,
+      "step": 1149
+    },
+    {
+      "epoch": 4.136690647482014,
+      "grad_norm": 0.5559630783400712,
+      "learning_rate": 4.731737411408637e-06,
+      "loss": 0.555715799331665,
+      "step": 1150
+    },
+    {
+      "epoch": 4.140287769784172,
+      "grad_norm": 0.5920639013002785,
+      "learning_rate": 4.724684921255704e-06,
+      "loss": 0.5829776525497437,
+      "step": 1151
+    },
+    {
+      "epoch": 4.143884892086331,
+      "grad_norm": 0.5487788143272614,
+      "learning_rate": 4.717632980466652e-06,
+      "loss": 0.5738891363143921,
+      "step": 1152
+    },
+    {
+      "epoch": 4.14748201438849,
+      "grad_norm": 0.5979520376739319,
+      "learning_rate": 4.710581603112931e-06,
+      "loss": 0.5732789039611816,
+      "step": 1153
+    },
+    {
+      "epoch": 4.151079136690647,
+      "grad_norm": 0.5763606911373871,
+      "learning_rate": 4.703530803264868e-06,
+      "loss": 0.5686631798744202,
+      "step": 1154
+    },
+    {
+      "epoch": 4.154676258992806,
+      "grad_norm": 0.5661816906841558,
+      "learning_rate": 4.696480594991629e-06,
+      "loss": 0.5841742753982544,
+      "step": 1155
+    },
+    {
+      "epoch": 4.158273381294964,
+      "grad_norm": 0.5950331674109112,
+      "learning_rate": 4.6894309923612055e-06,
+      "loss": 0.5730827450752258,
+      "step": 1156
+    },
+    {
+      "epoch": 4.161870503597123,
+      "grad_norm": 0.5881173493949299,
+      "learning_rate": 4.682382009440378e-06,
+      "loss": 0.5937288999557495,
+      "step": 1157
+    },
+    {
+      "epoch": 4.16546762589928,
+      "grad_norm": 0.55290224725042,
+      "learning_rate": 4.675333660294693e-06,
+      "loss": 0.5837142467498779,
+      "step": 1158
+    },
+    {
+      "epoch": 4.169064748201439,
+      "grad_norm": 0.6889012765482136,
+      "learning_rate": 4.66828595898843e-06,
+      "loss": 0.5661232471466064,
+      "step": 1159
+    },
+    {
+      "epoch": 4.172661870503597,
+      "grad_norm": 0.5466375770495574,
+      "learning_rate": 4.661238919584578e-06,
+      "loss": 0.5689799785614014,
+      "step": 1160
+    },
+    {
+      "epoch": 4.176258992805756,
+      "grad_norm": 0.5568083020527735,
+      "learning_rate": 4.654192556144801e-06,
+      "loss": 0.5801382064819336,
+      "step": 1161
+    },
+    {
+      "epoch": 4.179856115107913,
+      "grad_norm": 0.5758747775029548,
+      "learning_rate": 4.647146882729422e-06,
+      "loss": 0.5707106590270996,
+      "step": 1162
+    },
+    {
+      "epoch": 4.183453237410072,
+      "grad_norm": 0.5763824084621219,
+      "learning_rate": 4.640101913397376e-06,
+      "loss": 0.5798112154006958,
+      "step": 1163
+    },
+    {
+      "epoch": 4.18705035971223,
+      "grad_norm": 0.5941164621193183,
+      "learning_rate": 4.633057662206205e-06,
+      "loss": 0.5752370357513428,
+      "step": 1164
+    },
+    {
+      "epoch": 4.190647482014389,
+      "grad_norm": 0.5625375092244381,
+      "learning_rate": 4.626014143212008e-06,
+      "loss": 0.5721004009246826,
+      "step": 1165
+    },
+    {
+      "epoch": 4.194244604316546,
+      "grad_norm": 0.5665818979516964,
+      "learning_rate": 4.618971370469431e-06,
+      "loss": 0.5877313613891602,
+      "step": 1166
+    },
+    {
+      "epoch": 4.197841726618705,
+      "grad_norm": 0.6795586900116738,
+      "learning_rate": 4.611929358031625e-06,
+      "loss": 0.5642364025115967,
+      "step": 1167
+    },
+    {
+      "epoch": 4.201438848920863,
+      "grad_norm": 0.5415548661428123,
+      "learning_rate": 4.6048881199502265e-06,
+      "loss": 0.5845080614089966,
+      "step": 1168
+    },
+    {
+      "epoch": 4.205035971223022,
+      "grad_norm": 0.5869059273185752,
+      "learning_rate": 4.597847670275327e-06,
+      "loss": 0.5940054655075073,
+      "step": 1169
+    },
+    {
+      "epoch": 4.2086330935251794,
+      "grad_norm": 0.5607873720318685,
+      "learning_rate": 4.590808023055444e-06,
+      "loss": 0.573835551738739,
+      "step": 1170
+    },
+    {
+      "epoch": 4.212230215827338,
+      "grad_norm": 0.6003731381877622,
+      "learning_rate": 4.583769192337493e-06,
+      "loss": 0.5652705430984497,
+      "step": 1171
+    },
+    {
+      "epoch": 4.215827338129497,
+      "grad_norm": 0.5739016573249696,
+      "learning_rate": 4.576731192166762e-06,
+      "loss": 0.5749368667602539,
+      "step": 1172
+    },
+    {
+      "epoch": 4.219424460431655,
+      "grad_norm": 0.5581888001105628,
+      "learning_rate": 4.569694036586883e-06,
+      "loss": 0.5914247035980225,
+      "step": 1173
+    },
+    {
+      "epoch": 4.223021582733813,
+      "grad_norm": 0.5559528710698097,
+      "learning_rate": 4.562657739639798e-06,
+      "loss": 0.5822369456291199,
+      "step": 1174
+    },
+    {
+      "epoch": 4.226618705035971,
+      "grad_norm": 0.5533112566404729,
+      "learning_rate": 4.555622315365741e-06,
+      "loss": 0.6120118498802185,
+      "step": 1175
+    },
+    {
+      "epoch": 4.23021582733813,
+      "grad_norm": 0.6197934095165797,
+      "learning_rate": 4.548587777803198e-06,
+      "loss": 0.5657411217689514,
+      "step": 1176
+    },
+    {
+      "epoch": 4.233812949640288,
+      "grad_norm": 0.5538805662869524,
+      "learning_rate": 4.541554140988896e-06,
+      "loss": 0.6044785976409912,
+      "step": 1177
+    },
+    {
+      "epoch": 4.237410071942446,
+      "grad_norm": 0.599310631889973,
+      "learning_rate": 4.534521418957753e-06,
+      "loss": 0.575440526008606,
+      "step": 1178
+    },
+    {
+      "epoch": 4.241007194244604,
+      "grad_norm": 0.6071682175544293,
+      "learning_rate": 4.5274896257428684e-06,
+      "loss": 0.5827499032020569,
+      "step": 1179
+    },
+    {
+      "epoch": 4.244604316546763,
+      "grad_norm": 0.5585723963020499,
+      "learning_rate": 4.520458775375488e-06,
+      "loss": 0.5806722640991211,
+      "step": 1180
+    },
+    {
+      "epoch": 4.248201438848921,
+      "grad_norm": 0.6680955934459385,
+      "learning_rate": 4.513428881884976e-06,
+      "loss": 0.5744950771331787,
+      "step": 1181
+    },
+    {
+      "epoch": 4.251798561151079,
+      "grad_norm": 0.5498839260681047,
+      "learning_rate": 4.506399959298781e-06,
+      "loss": 0.5698388814926147,
+      "step": 1182
+    },
+    {
+      "epoch": 4.255395683453237,
+      "grad_norm": 0.6069444284018976,
+      "learning_rate": 4.499372021642427e-06,
+      "loss": 0.5755789875984192,
+      "step": 1183
+    },
+    {
+      "epoch": 4.258992805755396,
+      "grad_norm": 0.5599727416246281,
+      "learning_rate": 4.49234508293946e-06,
+      "loss": 0.5962889194488525,
+      "step": 1184
+    },
+    {
+      "epoch": 4.262589928057554,
+      "grad_norm": 0.6342165286791643,
+      "learning_rate": 4.485319157211441e-06,
+      "loss": 0.5775662660598755,
+      "step": 1185
+    },
+    {
+      "epoch": 4.266187050359712,
+      "grad_norm": 0.5526730574177247,
+      "learning_rate": 4.4782942584779046e-06,
+      "loss": 0.5755195617675781,
+      "step": 1186
+    },
+    {
+      "epoch": 4.26978417266187,
+      "grad_norm": 0.6227375274845137,
+      "learning_rate": 4.47127040075634e-06,
+      "loss": 0.5777647495269775,
+      "step": 1187
+    },
+    {
+      "epoch": 4.273381294964029,
+      "grad_norm": 1.0110886173810598,
+      "learning_rate": 4.464247598062156e-06,
+      "loss": 0.5726004838943481,
+      "step": 1188
+    },
+    {
+      "epoch": 4.276978417266187,
+      "grad_norm": 0.5669341213587702,
+      "learning_rate": 4.457225864408659e-06,
+      "loss": 0.5730968117713928,
+      "step": 1189
+    },
+    {
+      "epoch": 4.280575539568345,
+      "grad_norm": 0.5763712469502784,
+      "learning_rate": 4.450205213807018e-06,
+      "loss": 0.595857560634613,
+      "step": 1190
+    },
+    {
+      "epoch": 4.284172661870503,
+      "grad_norm": 0.5946725282808688,
+      "learning_rate": 4.443185660266248e-06,
+      "loss": 0.5865065455436707,
+      "step": 1191
+    },
+    {
+      "epoch": 4.287769784172662,
+      "grad_norm": 0.5740713350476653,
+      "learning_rate": 4.436167217793167e-06,
+      "loss": 0.5740945935249329,
+      "step": 1192
+    },
+    {
+      "epoch": 4.2913669064748206,
+      "grad_norm": 0.5612912595120162,
+      "learning_rate": 4.429149900392379e-06,
+      "loss": 0.5785430669784546,
+      "step": 1193
+    },
+    {
+      "epoch": 4.294964028776978,
+      "grad_norm": 0.5740412978546652,
+      "learning_rate": 4.4221337220662495e-06,
+      "loss": 0.5770116448402405,
+      "step": 1194
+    },
+    {
+      "epoch": 4.298561151079137,
+      "grad_norm": 0.5795305629307095,
+      "learning_rate": 4.415118696814859e-06,
+      "loss": 0.5700653791427612,
+      "step": 1195
+    },
+    {
+      "epoch": 4.302158273381295,
+      "grad_norm": 0.5718157967814855,
+      "learning_rate": 4.408104838635996e-06,
+      "loss": 0.6017115116119385,
+      "step": 1196
+    },
+    {
+      "epoch": 4.305755395683454,
+      "grad_norm": 0.5892253053245672,
+      "learning_rate": 4.401092161525117e-06,
+      "loss": 0.5771710872650146,
+      "step": 1197
+    },
+    {
+      "epoch": 4.309352517985611,
+      "grad_norm": 0.6233608453273818,
+      "learning_rate": 4.394080679475324e-06,
+      "loss": 0.599836528301239,
+      "step": 1198
+    },
+    {
+      "epoch": 4.31294964028777,
+      "grad_norm": 0.5719492505819993,
+      "learning_rate": 4.38707040647733e-06,
+      "loss": 0.5776287913322449,
+      "step": 1199
+    },
+    {
+      "epoch": 4.316546762589928,
+      "grad_norm": 0.551585937328142,
+      "learning_rate": 4.38006135651944e-06,
+      "loss": 0.5568167567253113,
+      "step": 1200
+    },
+    {
+      "epoch": 4.316546762589928,
+      "eval_loss": 0.8378241658210754,
+      "eval_runtime": 7.7777,
+      "eval_samples_per_second": 11.314,
+      "eval_steps_per_second": 0.386,
+      "eval_token_acc": 0.7497680652441335,
+      "step": 1200
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 2224,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 899169880113152.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}