{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 16.0,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 1.1046091318130493,
      "learning_rate": 5.319148936170213e-05,
      "loss": 0.6562,
      "step": 25
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4066573977470398,
      "learning_rate": 0.00010638297872340425,
      "loss": 0.5373,
      "step": 50
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.7340325713157654,
      "learning_rate": 0.00015957446808510637,
      "loss": 0.4995,
      "step": 75
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.134040117263794,
      "learning_rate": 0.0002127659574468085,
      "loss": 0.4871,
      "step": 100
    },
    {
      "epoch": 0.5,
      "grad_norm": 2.911975860595703,
      "learning_rate": 0.00026595744680851064,
      "loss": 0.4507,
      "step": 125
    },
    {
      "epoch": 0.6,
      "grad_norm": 2.040365219116211,
      "learning_rate": 0.00031914893617021275,
      "loss": 0.3944,
      "step": 150
    },
    {
      "epoch": 0.7,
      "grad_norm": 5.643661022186279,
      "learning_rate": 0.0003723404255319149,
      "loss": 0.3863,
      "step": 175
    },
    {
      "epoch": 0.8,
      "grad_norm": 3.882492780685425,
      "learning_rate": 0.0003992081821181128,
      "loss": 0.3875,
      "step": 200
    },
    {
      "epoch": 0.9,
      "grad_norm": 5.87437105178833,
      "learning_rate": 0.0003975585615308479,
      "loss": 0.3856,
      "step": 225
    },
    {
      "epoch": 1.0,
      "grad_norm": 4.611206531524658,
      "learning_rate": 0.00039590894094358297,
      "loss": 0.3866,
      "step": 250
    },
    {
      "epoch": 1.1,
      "grad_norm": 3.6404929161071777,
      "learning_rate": 0.0003942593203563181,
      "loss": 0.3437,
      "step": 275
    },
    {
      "epoch": 1.2,
      "grad_norm": 6.469444751739502,
      "learning_rate": 0.0003926096997690532,
      "loss": 0.3192,
      "step": 300
    },
    {
      "epoch": 1.3,
      "grad_norm": 3.6251447200775146,
      "learning_rate": 0.00039096007918178817,
      "loss": 0.349,
      "step": 325
    },
    {
      "epoch": 1.4,
      "grad_norm": 2.854794979095459,
      "learning_rate": 0.0003893104585945233,
      "loss": 0.3274,
      "step": 350
    },
    {
      "epoch": 1.5,
      "grad_norm": 4.805097579956055,
      "learning_rate": 0.0003876608380072583,
      "loss": 0.3279,
      "step": 375
    },
    {
      "epoch": 1.6,
      "grad_norm": 2.9891581535339355,
      "learning_rate": 0.00038601121741999343,
      "loss": 0.3166,
      "step": 400
    },
    {
      "epoch": 1.7,
      "grad_norm": 5.333752155303955,
      "learning_rate": 0.0003843615968327285,
      "loss": 0.2856,
      "step": 425
    },
    {
      "epoch": 1.8,
      "grad_norm": 2.615621566772461,
      "learning_rate": 0.0003827119762454636,
      "loss": 0.3286,
      "step": 450
    },
    {
      "epoch": 1.9,
      "grad_norm": 2.4228415489196777,
      "learning_rate": 0.00038106235565819863,
      "loss": 0.2812,
      "step": 475
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.6015517711639404,
      "learning_rate": 0.0003794127350709337,
      "loss": 0.2884,
      "step": 500
    },
    {
      "epoch": 2.1,
      "grad_norm": 2.7867860794067383,
      "learning_rate": 0.0003777631144836688,
      "loss": 0.2582,
      "step": 525
    },
    {
      "epoch": 2.2,
      "grad_norm": 10.079645156860352,
      "learning_rate": 0.00037611349389640383,
      "loss": 0.245,
      "step": 550
    },
    {
      "epoch": 2.3,
      "grad_norm": 2.475917100906372,
      "learning_rate": 0.00037446387330913894,
      "loss": 0.2615,
      "step": 575
    },
    {
      "epoch": 2.4,
      "grad_norm": 8.098592758178711,
      "learning_rate": 0.000372814252721874,
      "loss": 0.2484,
      "step": 600
    },
    {
      "epoch": 2.5,
      "grad_norm": 2.318723201751709,
      "learning_rate": 0.00037116463213460903,
      "loss": 0.2056,
      "step": 625
    },
    {
      "epoch": 2.6,
      "grad_norm": 3.616283893585205,
      "learning_rate": 0.00036951501154734414,
      "loss": 0.2358,
      "step": 650
    },
    {
      "epoch": 2.7,
      "grad_norm": 6.419492721557617,
      "learning_rate": 0.0003678653909600792,
      "loss": 0.2204,
      "step": 675
    },
    {
      "epoch": 2.8,
      "grad_norm": 2.9333388805389404,
      "learning_rate": 0.0003662157703728143,
      "loss": 0.1857,
      "step": 700
    },
    {
      "epoch": 2.9,
      "grad_norm": 7.691112995147705,
      "learning_rate": 0.00036456614978554934,
      "loss": 0.2054,
      "step": 725
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.908492565155029,
      "learning_rate": 0.00036291652919828444,
      "loss": 0.1956,
      "step": 750
    },
    {
      "epoch": 3.1,
      "grad_norm": 1.8923287391662598,
      "learning_rate": 0.00036126690861101944,
      "loss": 0.1708,
      "step": 775
    },
    {
      "epoch": 3.2,
      "grad_norm": 4.403504848480225,
      "learning_rate": 0.00035961728802375454,
      "loss": 0.1441,
      "step": 800
    },
    {
      "epoch": 3.3,
      "grad_norm": 4.117386817932129,
      "learning_rate": 0.0003579676674364896,
      "loss": 0.1586,
      "step": 825
    },
    {
      "epoch": 3.4,
      "grad_norm": 7.119816303253174,
      "learning_rate": 0.0003563180468492247,
      "loss": 0.1472,
      "step": 850
    },
    {
      "epoch": 3.5,
      "grad_norm": 3.5796430110931396,
      "learning_rate": 0.0003546684262619598,
      "loss": 0.1802,
      "step": 875
    },
    {
      "epoch": 3.6,
      "grad_norm": 2.97688889503479,
      "learning_rate": 0.00035301880567469485,
      "loss": 0.1578,
      "step": 900
    },
    {
      "epoch": 3.7,
      "grad_norm": 3.716148614883423,
      "learning_rate": 0.0003513691850874299,
      "loss": 0.1662,
      "step": 925
    },
    {
      "epoch": 3.8,
      "grad_norm": 6.1249566078186035,
      "learning_rate": 0.00034971956450016495,
      "loss": 0.159,
      "step": 950
    },
    {
      "epoch": 3.9,
      "grad_norm": 3.0592427253723145,
      "learning_rate": 0.00034806994391290005,
      "loss": 0.148,
      "step": 975
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.467265605926514,
      "learning_rate": 0.0003464203233256351,
      "loss": 0.1442,
      "step": 1000
    },
    {
      "epoch": 4.1,
      "grad_norm": 2.7223546504974365,
      "learning_rate": 0.0003447707027383702,
      "loss": 0.1339,
      "step": 1025
    },
    {
      "epoch": 4.2,
      "grad_norm": 3.698854923248291,
      "learning_rate": 0.00034312108215110525,
      "loss": 0.1381,
      "step": 1050
    },
    {
      "epoch": 4.3,
      "grad_norm": 7.418384552001953,
      "learning_rate": 0.0003414714615638403,
      "loss": 0.1158,
      "step": 1075
    },
    {
      "epoch": 4.4,
      "grad_norm": 1.2887814044952393,
      "learning_rate": 0.0003398218409765754,
      "loss": 0.0855,
      "step": 1100
    },
    {
      "epoch": 4.5,
      "grad_norm": 0.838731586933136,
      "learning_rate": 0.00033817222038931045,
      "loss": 0.1272,
      "step": 1125
    },
    {
      "epoch": 4.6,
      "grad_norm": 5.912592887878418,
      "learning_rate": 0.00033652259980204556,
      "loss": 0.1156,
      "step": 1150
    },
    {
      "epoch": 4.7,
      "grad_norm": 8.49521255493164,
      "learning_rate": 0.0003348729792147806,
      "loss": 0.1182,
      "step": 1175
    },
    {
      "epoch": 4.8,
      "grad_norm": 10.278315544128418,
      "learning_rate": 0.0003332233586275157,
      "loss": 0.1071,
      "step": 1200
    },
    {
      "epoch": 4.9,
      "grad_norm": 0.724703311920166,
      "learning_rate": 0.00033157373804025076,
      "loss": 0.11,
      "step": 1225
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.6808199882507324,
      "learning_rate": 0.0003299241174529858,
      "loss": 0.1101,
      "step": 1250
    },
    {
      "epoch": 5.1,
      "grad_norm": 0.1038585901260376,
      "learning_rate": 0.0003282744968657209,
      "loss": 0.0791,
      "step": 1275
    },
    {
      "epoch": 5.2,
      "grad_norm": 0.15866787731647491,
      "learning_rate": 0.00032662487627845596,
      "loss": 0.0638,
      "step": 1300
    },
    {
      "epoch": 5.3,
      "grad_norm": 6.748287200927734,
      "learning_rate": 0.00032497525569119106,
      "loss": 0.0961,
      "step": 1325
    },
    {
      "epoch": 5.4,
      "grad_norm": 1.5960336923599243,
      "learning_rate": 0.0003233256351039261,
      "loss": 0.1057,
      "step": 1350
    },
    {
      "epoch": 5.5,
      "grad_norm": 3.3678526878356934,
      "learning_rate": 0.00032167601451666116,
      "loss": 0.1092,
      "step": 1375
    },
    {
      "epoch": 5.6,
      "grad_norm": 4.743562698364258,
      "learning_rate": 0.00032002639392939627,
      "loss": 0.1054,
      "step": 1400
    },
    {
      "epoch": 5.7,
      "grad_norm": 12.860156059265137,
      "learning_rate": 0.0003183767733421313,
      "loss": 0.1009,
      "step": 1425
    },
    {
      "epoch": 5.8,
      "grad_norm": 0.8820792436599731,
      "learning_rate": 0.0003167271527548664,
      "loss": 0.1219,
      "step": 1450
    },
    {
      "epoch": 5.9,
      "grad_norm": 2.732572317123413,
      "learning_rate": 0.00031507753216760147,
      "loss": 0.0682,
      "step": 1475
    },
    {
      "epoch": 6.0,
      "grad_norm": 9.810711860656738,
      "learning_rate": 0.00031342791158033657,
      "loss": 0.1134,
      "step": 1500
    },
    {
      "epoch": 6.1,
      "grad_norm": 6.699489593505859,
      "learning_rate": 0.00031177829099307157,
      "loss": 0.0557,
      "step": 1525
    },
    {
      "epoch": 6.2,
      "grad_norm": 14.691123008728027,
      "learning_rate": 0.00031012867040580667,
      "loss": 0.0599,
      "step": 1550
    },
    {
      "epoch": 6.3,
      "grad_norm": 24.44135093688965,
      "learning_rate": 0.0003084790498185417,
      "loss": 0.0818,
      "step": 1575
    },
    {
      "epoch": 6.4,
      "grad_norm": 0.033995576202869415,
      "learning_rate": 0.0003068294292312768,
      "loss": 0.0692,
      "step": 1600
    },
    {
      "epoch": 6.5,
      "grad_norm": 6.722524642944336,
      "learning_rate": 0.0003051798086440119,
      "loss": 0.0702,
      "step": 1625
    },
    {
      "epoch": 6.6,
      "grad_norm": 2.9519436359405518,
      "learning_rate": 0.000303530188056747,
      "loss": 0.073,
      "step": 1650
    },
    {
      "epoch": 6.7,
      "grad_norm": 1.5118328332901,
      "learning_rate": 0.000301880567469482,
      "loss": 0.0926,
      "step": 1675
    },
    {
      "epoch": 6.8,
      "grad_norm": 2.1923201084136963,
      "learning_rate": 0.0003002309468822171,
      "loss": 0.0821,
      "step": 1700
    },
    {
      "epoch": 6.9,
      "grad_norm": 3.34309458732605,
      "learning_rate": 0.0002985813262949522,
      "loss": 0.0827,
      "step": 1725
    },
    {
      "epoch": 7.0,
      "grad_norm": 10.289947509765625,
      "learning_rate": 0.00029693170570768723,
      "loss": 0.1162,
      "step": 1750
    },
    {
      "epoch": 7.1,
      "grad_norm": 0.051936667412519455,
      "learning_rate": 0.00029528208512042233,
      "loss": 0.0626,
      "step": 1775
    },
    {
      "epoch": 7.2,
      "grad_norm": 1.8913339376449585,
      "learning_rate": 0.0002936324645331574,
      "loss": 0.0657,
      "step": 1800
    },
    {
      "epoch": 7.3,
      "grad_norm": 15.073695182800293,
      "learning_rate": 0.00029198284394589243,
      "loss": 0.0539,
      "step": 1825
    },
    {
      "epoch": 7.4,
      "grad_norm": 4.514949798583984,
      "learning_rate": 0.00029033322335862753,
      "loss": 0.056,
      "step": 1850
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.46094000339508057,
      "learning_rate": 0.0002886836027713626,
      "loss": 0.0667,
      "step": 1875
    },
    {
      "epoch": 7.6,
      "grad_norm": 4.682873725891113,
      "learning_rate": 0.0002870339821840977,
      "loss": 0.0673,
      "step": 1900
    },
    {
      "epoch": 7.7,
      "grad_norm": 3.581599235534668,
      "learning_rate": 0.00028538436159683273,
      "loss": 0.0699,
      "step": 1925
    },
    {
      "epoch": 7.8,
      "grad_norm": 3.2518703937530518,
      "learning_rate": 0.00028373474100956784,
      "loss": 0.0628,
      "step": 1950
    },
    {
      "epoch": 7.9,
      "grad_norm": 0.24071219563484192,
      "learning_rate": 0.0002820851204223029,
      "loss": 0.0747,
      "step": 1975
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.8545596599578857,
      "learning_rate": 0.00028043549983503794,
      "loss": 0.0688,
      "step": 2000
    },
    {
      "epoch": 8.1,
      "grad_norm": 0.049499694257974625,
      "learning_rate": 0.00027878587924777304,
      "loss": 0.0563,
      "step": 2025
    },
    {
      "epoch": 8.2,
      "grad_norm": 22.112478256225586,
      "learning_rate": 0.0002771362586605081,
      "loss": 0.0545,
      "step": 2050
    },
    {
      "epoch": 8.3,
      "grad_norm": 20.87473487854004,
      "learning_rate": 0.0002754866380732432,
      "loss": 0.0256,
      "step": 2075
    },
    {
      "epoch": 8.4,
      "grad_norm": 2.317768096923828,
      "learning_rate": 0.00027383701748597824,
      "loss": 0.0753,
      "step": 2100
    },
    {
      "epoch": 8.5,
      "grad_norm": 20.241336822509766,
      "learning_rate": 0.0002721873968987133,
      "loss": 0.0257,
      "step": 2125
    },
    {
      "epoch": 8.6,
      "grad_norm": 14.994575500488281,
      "learning_rate": 0.00027053777631144834,
      "loss": 0.0748,
      "step": 2150
    },
    {
      "epoch": 8.7,
      "grad_norm": 0.8268639445304871,
      "learning_rate": 0.00026888815572418344,
      "loss": 0.0805,
      "step": 2175
    },
    {
      "epoch": 8.8,
      "grad_norm": 0.031071001663804054,
      "learning_rate": 0.00026723853513691855,
      "loss": 0.0603,
      "step": 2200
    },
    {
      "epoch": 8.9,
      "grad_norm": 2.4572932720184326,
      "learning_rate": 0.0002655889145496536,
      "loss": 0.071,
      "step": 2225
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.21708321571350098,
      "learning_rate": 0.0002639392939623887,
      "loss": 0.0606,
      "step": 2250
    },
    {
      "epoch": 9.1,
      "grad_norm": 14.928559303283691,
      "learning_rate": 0.0002622896733751237,
      "loss": 0.0443,
      "step": 2275
    },
    {
      "epoch": 9.2,
      "grad_norm": 1.108826994895935,
      "learning_rate": 0.0002606400527878588,
      "loss": 0.0509,
      "step": 2300
    },
    {
      "epoch": 9.3,
      "grad_norm": 0.05721910670399666,
      "learning_rate": 0.00025899043220059385,
      "loss": 0.0459,
      "step": 2325
    },
    {
      "epoch": 9.4,
      "grad_norm": 7.81903600692749,
      "learning_rate": 0.00025734081161332895,
      "loss": 0.0535,
      "step": 2350
    },
    {
      "epoch": 9.5,
      "grad_norm": 0.01475981343537569,
      "learning_rate": 0.000255691191026064,
      "loss": 0.0399,
      "step": 2375
    },
    {
      "epoch": 9.6,
      "grad_norm": 5.496493339538574,
      "learning_rate": 0.0002540415704387991,
      "loss": 0.0744,
      "step": 2400
    },
    {
      "epoch": 9.7,
      "grad_norm": 0.05929339677095413,
      "learning_rate": 0.00025239194985153415,
      "loss": 0.0416,
      "step": 2425
    },
    {
      "epoch": 9.8,
      "grad_norm": 0.06961135566234589,
      "learning_rate": 0.0002507423292642692,
      "loss": 0.04,
      "step": 2450
    },
    {
      "epoch": 9.9,
      "grad_norm": 0.02165246568620205,
      "learning_rate": 0.0002490927086770043,
      "loss": 0.0984,
      "step": 2475
    },
    {
      "epoch": 10.0,
      "grad_norm": 25.155675888061523,
      "learning_rate": 0.00024744308808973936,
      "loss": 0.042,
      "step": 2500
    },
    {
      "epoch": 10.1,
      "grad_norm": 8.566617965698242,
      "learning_rate": 0.00024579346750247446,
      "loss": 0.0385,
      "step": 2525
    },
    {
      "epoch": 10.2,
      "grad_norm": 0.024992404505610466,
      "learning_rate": 0.0002441438469152095,
      "loss": 0.0327,
      "step": 2550
    },
    {
      "epoch": 10.3,
      "grad_norm": 0.22233247756958008,
      "learning_rate": 0.00024249422632794456,
      "loss": 0.0211,
      "step": 2575
    },
    {
      "epoch": 10.4,
      "grad_norm": 13.307366371154785,
      "learning_rate": 0.00024084460574067963,
      "loss": 0.0472,
      "step": 2600
    },
    {
      "epoch": 10.5,
      "grad_norm": 0.17787696421146393,
      "learning_rate": 0.0002391949851534147,
      "loss": 0.0346,
      "step": 2625
    },
    {
      "epoch": 10.6,
      "grad_norm": 1.2845549583435059,
      "learning_rate": 0.0002375453645661498,
      "loss": 0.0587,
      "step": 2650
    },
    {
      "epoch": 10.7,
      "grad_norm": 0.0603482760488987,
      "learning_rate": 0.00023589574397888486,
      "loss": 0.0394,
      "step": 2675
    },
    {
      "epoch": 10.8,
      "grad_norm": 0.031819943338632584,
      "learning_rate": 0.00023424612339161997,
      "loss": 0.0434,
      "step": 2700
    },
    {
      "epoch": 10.9,
      "grad_norm": 1.5738537311553955,
      "learning_rate": 0.000232596502804355,
      "loss": 0.0393,
      "step": 2725
    },
    {
      "epoch": 11.0,
      "grad_norm": 0.6862583756446838,
      "learning_rate": 0.00023094688221709007,
      "loss": 0.0183,
      "step": 2750
    },
    {
      "epoch": 11.1,
      "grad_norm": 0.0053489054553210735,
      "learning_rate": 0.00022929726162982514,
      "loss": 0.0277,
      "step": 2775
    },
    {
      "epoch": 11.2,
      "grad_norm": 0.04365606606006622,
      "learning_rate": 0.00022764764104256022,
      "loss": 0.0337,
      "step": 2800
    },
    {
      "epoch": 11.3,
      "grad_norm": 1.3348641395568848,
      "learning_rate": 0.0002259980204552953,
      "loss": 0.0219,
      "step": 2825
    },
    {
      "epoch": 11.4,
      "grad_norm": 3.3695547580718994,
      "learning_rate": 0.00022434839986803037,
      "loss": 0.0227,
      "step": 2850
    },
    {
      "epoch": 11.5,
      "grad_norm": 2.507455587387085,
      "learning_rate": 0.00022269877928076542,
      "loss": 0.0443,
      "step": 2875
    },
    {
      "epoch": 11.6,
      "grad_norm": 0.013451912440359592,
      "learning_rate": 0.0002210491586935005,
      "loss": 0.0486,
      "step": 2900
    },
    {
      "epoch": 11.7,
      "grad_norm": 0.031177405267953873,
      "learning_rate": 0.00021939953810623557,
      "loss": 0.0511,
      "step": 2925
    },
    {
      "epoch": 11.8,
      "grad_norm": 0.02148960903286934,
      "learning_rate": 0.00021774991751897065,
      "loss": 0.0269,
      "step": 2950
    },
    {
      "epoch": 11.9,
      "grad_norm": 0.01219563465565443,
      "learning_rate": 0.00021610029693170573,
      "loss": 0.042,
      "step": 2975
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.2062515765428543,
      "learning_rate": 0.0002144506763444408,
      "loss": 0.031,
      "step": 3000
    },
    {
      "epoch": 12.1,
      "grad_norm": 3.235403060913086,
      "learning_rate": 0.00021280105575717582,
      "loss": 0.0414,
      "step": 3025
    },
    {
      "epoch": 12.2,
      "grad_norm": 1.089821696281433,
      "learning_rate": 0.00021115143516991093,
      "loss": 0.0692,
      "step": 3050
    },
    {
      "epoch": 12.3,
      "grad_norm": 0.013664484024047852,
      "learning_rate": 0.000209501814582646,
      "loss": 0.0367,
      "step": 3075
    },
    {
      "epoch": 12.4,
      "grad_norm": 1.3101387023925781,
      "learning_rate": 0.00020785219399538108,
      "loss": 0.0277,
      "step": 3100
    },
    {
      "epoch": 12.5,
      "grad_norm": 3.244853973388672,
      "learning_rate": 0.00020620257340811616,
      "loss": 0.0373,
      "step": 3125
    },
    {
      "epoch": 12.6,
      "grad_norm": 0.013548053801059723,
      "learning_rate": 0.00020455295282085123,
      "loss": 0.027,
      "step": 3150
    },
    {
      "epoch": 12.7,
      "grad_norm": 0.004670978989452124,
      "learning_rate": 0.00020290333223358626,
      "loss": 0.0252,
      "step": 3175
    },
    {
      "epoch": 12.8,
      "grad_norm": 0.007080752402544022,
      "learning_rate": 0.00020125371164632133,
      "loss": 0.0187,
      "step": 3200
    },
    {
      "epoch": 12.9,
      "grad_norm": 0.005918944254517555,
      "learning_rate": 0.0001996040910590564,
      "loss": 0.0423,
      "step": 3225
    },
    {
      "epoch": 13.0,
      "grad_norm": 4.530269145965576,
      "learning_rate": 0.00019795447047179148,
      "loss": 0.0381,
      "step": 3250
    },
    {
      "epoch": 13.1,
      "grad_norm": 0.00533739197999239,
      "learning_rate": 0.0001963048498845266,
      "loss": 0.0097,
      "step": 3275
    },
    {
      "epoch": 13.2,
      "grad_norm": 0.7641226649284363,
      "learning_rate": 0.00019465522929726164,
      "loss": 0.0205,
      "step": 3300
    },
    {
      "epoch": 13.3,
      "grad_norm": 2.16174054145813,
      "learning_rate": 0.00019300560870999671,
      "loss": 0.0174,
      "step": 3325
    },
    {
      "epoch": 13.4,
      "grad_norm": 0.00814723875373602,
      "learning_rate": 0.0001913559881227318,
      "loss": 0.0215,
      "step": 3350
    },
    {
      "epoch": 13.5,
      "grad_norm": 10.500170707702637,
      "learning_rate": 0.00018970636753546684,
      "loss": 0.0226,
      "step": 3375
    },
    {
      "epoch": 13.6,
      "grad_norm": 2.612359046936035,
      "learning_rate": 0.00018805674694820192,
      "loss": 0.0314,
      "step": 3400
    },
    {
      "epoch": 13.7,
      "grad_norm": 0.11265736818313599,
      "learning_rate": 0.000186407126360937,
      "loss": 0.0234,
      "step": 3425
    },
    {
      "epoch": 13.8,
      "grad_norm": 0.06453213095664978,
      "learning_rate": 0.00018475750577367207,
      "loss": 0.0314,
      "step": 3450
    },
    {
      "epoch": 13.9,
      "grad_norm": 0.01650763861835003,
      "learning_rate": 0.00018310788518640715,
      "loss": 0.0403,
      "step": 3475
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.002764373552054167,
      "learning_rate": 0.00018145826459914222,
      "loss": 0.0137,
      "step": 3500
    },
    {
      "epoch": 14.1,
      "grad_norm": 0.0022430066019296646,
      "learning_rate": 0.00017980864401187727,
      "loss": 0.0084,
      "step": 3525
    },
    {
      "epoch": 14.2,
      "grad_norm": 0.007267679553478956,
      "learning_rate": 0.00017815902342461235,
      "loss": 0.0161,
      "step": 3550
    },
    {
      "epoch": 14.3,
      "grad_norm": 0.004844362381845713,
      "learning_rate": 0.00017650940283734742,
      "loss": 0.0399,
      "step": 3575
    },
    {
      "epoch": 14.4,
      "grad_norm": 0.05529040843248367,
      "learning_rate": 0.00017485978225008247,
      "loss": 0.0192,
      "step": 3600
    },
    {
      "epoch": 14.5,
      "grad_norm": 0.008248478174209595,
      "learning_rate": 0.00017321016166281755,
      "loss": 0.0079,
      "step": 3625
    },
    {
      "epoch": 14.6,
      "grad_norm": 0.02617962658405304,
      "learning_rate": 0.00017156054107555263,
      "loss": 0.0244,
      "step": 3650
    },
    {
      "epoch": 14.7,
      "grad_norm": 0.004497932270169258,
      "learning_rate": 0.0001699109204882877,
      "loss": 0.0464,
      "step": 3675
    },
    {
      "epoch": 14.8,
      "grad_norm": 0.008586671203374863,
      "learning_rate": 0.00016826129990102278,
      "loss": 0.0354,
      "step": 3700
    },
    {
      "epoch": 14.9,
      "grad_norm": 0.30177807807922363,
      "learning_rate": 0.00016661167931375785,
      "loss": 0.0147,
      "step": 3725
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.10390808433294296,
      "learning_rate": 0.0001649620587264929,
      "loss": 0.0078,
      "step": 3750
    },
    {
      "epoch": 15.1,
      "grad_norm": 1.584868311882019,
      "learning_rate": 0.00016331243813922798,
      "loss": 0.0059,
      "step": 3775
    },
    {
      "epoch": 15.2,
      "grad_norm": 0.0038078685756772757,
      "learning_rate": 0.00016166281755196306,
      "loss": 0.0143,
      "step": 3800
    },
    {
      "epoch": 15.3,
      "grad_norm": 0.017795555293560028,
      "learning_rate": 0.00016001319696469813,
      "loss": 0.0003,
      "step": 3825
    },
    {
      "epoch": 15.4,
      "grad_norm": 7.246474742889404,
      "learning_rate": 0.0001583635763774332,
      "loss": 0.0288,
      "step": 3850
    },
    {
      "epoch": 15.5,
      "grad_norm": 0.0014716258738189936,
      "learning_rate": 0.00015671395579016829,
      "loss": 0.0113,
      "step": 3875
    },
    {
      "epoch": 15.6,
      "grad_norm": 0.002159764291718602,
      "learning_rate": 0.00015506433520290334,
      "loss": 0.0138,
      "step": 3900
    },
    {
      "epoch": 15.7,
      "grad_norm": 27.16730308532715,
      "learning_rate": 0.0001534147146156384,
      "loss": 0.0048,
      "step": 3925
    },
    {
      "epoch": 15.8,
      "grad_norm": 0.004570333287119865,
      "learning_rate": 0.0001517650940283735,
      "loss": 0.0144,
      "step": 3950
    },
    {
      "epoch": 15.9,
      "grad_norm": 0.008330491371452808,
      "learning_rate": 0.00015011547344110854,
      "loss": 0.0371,
      "step": 3975
    },
    {
      "epoch": 16.0,
      "grad_norm": 15.75979232788086,
      "learning_rate": 0.00014846585285384361,
      "loss": 0.0207,
      "step": 4000
    }
  ],
  "logging_steps": 25,
  "max_steps": 6250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 203277926400000.0,
  "train_batch_size": 20,
  "trial_name": null,
  "trial_params": null
}