{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9999059177721329,
  "eval_steps": 20,
  "global_step": 5314,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0037632891146862357,
      "grad_norm": 0.15685315430164337,
      "learning_rate": 0.0002,
      "loss": 0.4618,
      "step": 20
    },
    {
      "epoch": 0.0075265782293724715,
      "grad_norm": 0.10632659494876862,
      "learning_rate": 0.0002,
      "loss": 0.2985,
      "step": 40
    },
    {
      "epoch": 0.011289867344058707,
      "grad_norm": 0.12228264659643173,
      "learning_rate": 0.0002,
      "loss": 0.2689,
      "step": 60
    },
    {
      "epoch": 0.015053156458744943,
      "grad_norm": 0.09272768348455429,
      "learning_rate": 0.0002,
      "loss": 0.2249,
      "step": 80
    },
    {
      "epoch": 0.01881644557343118,
      "grad_norm": 0.11586301028728485,
      "learning_rate": 0.0002,
      "loss": 0.23,
      "step": 100
    },
    {
      "epoch": 0.022579734688117414,
      "grad_norm": 0.08657937496900558,
      "learning_rate": 0.0002,
      "loss": 0.2189,
      "step": 120
    },
    {
      "epoch": 0.02634302380280365,
      "grad_norm": 0.08374184370040894,
      "learning_rate": 0.0002,
      "loss": 0.1987,
      "step": 140
    },
    {
      "epoch": 0.030106312917489886,
      "grad_norm": 0.08528616279363632,
      "learning_rate": 0.0002,
      "loss": 0.2065,
      "step": 160
    },
    {
      "epoch": 0.03386960203217612,
      "grad_norm": 0.17550894618034363,
      "learning_rate": 0.0002,
      "loss": 0.1904,
      "step": 180
    },
    {
      "epoch": 0.03763289114686236,
      "grad_norm": 0.09562012553215027,
      "learning_rate": 0.0002,
      "loss": 0.1824,
      "step": 200
    },
    {
      "epoch": 0.04139618026154859,
      "grad_norm": 0.12333519756793976,
      "learning_rate": 0.0002,
      "loss": 0.1702,
      "step": 220
    },
    {
      "epoch": 0.04515946937623483,
      "grad_norm": 0.11404936760663986,
      "learning_rate": 0.0002,
      "loss": 0.18,
      "step": 240
    },
    {
      "epoch": 0.048922758490921064,
      "grad_norm": 0.08656694740056992,
      "learning_rate": 0.0002,
      "loss": 0.1749,
      "step": 260
    },
    {
      "epoch": 0.0526860476056073,
      "grad_norm": 0.09797225147485733,
      "learning_rate": 0.0002,
      "loss": 0.1731,
      "step": 280
    },
    {
      "epoch": 0.056449336720293536,
      "grad_norm": 0.09765412658452988,
      "learning_rate": 0.0002,
      "loss": 0.1578,
      "step": 300
    },
    {
      "epoch": 0.06021262583497977,
      "grad_norm": 0.07540671527385712,
      "learning_rate": 0.0002,
      "loss": 0.1693,
      "step": 320
    },
    {
      "epoch": 0.06397591494966601,
      "grad_norm": 0.08590289205312729,
      "learning_rate": 0.0002,
      "loss": 0.1607,
      "step": 340
    },
    {
      "epoch": 0.06773920406435224,
      "grad_norm": 0.09767664223909378,
      "learning_rate": 0.0002,
      "loss": 0.1419,
      "step": 360
    },
    {
      "epoch": 0.07150249317903848,
      "grad_norm": 0.10479151457548141,
      "learning_rate": 0.0002,
      "loss": 0.1629,
      "step": 380
    },
    {
      "epoch": 0.07526578229372471,
      "grad_norm": 0.08791118115186691,
      "learning_rate": 0.0002,
      "loss": 0.1484,
      "step": 400
    },
    {
      "epoch": 0.07902907140841095,
      "grad_norm": 0.10221686214208603,
      "learning_rate": 0.0002,
      "loss": 0.1499,
      "step": 420
    },
    {
      "epoch": 0.08279236052309719,
      "grad_norm": 0.09131903946399689,
      "learning_rate": 0.0002,
      "loss": 0.1485,
      "step": 440
    },
    {
      "epoch": 0.08655564963778342,
      "grad_norm": 0.10372031480073929,
      "learning_rate": 0.0002,
      "loss": 0.1441,
      "step": 460
    },
    {
      "epoch": 0.09031893875246966,
      "grad_norm": 0.09649350494146347,
      "learning_rate": 0.0002,
      "loss": 0.147,
      "step": 480
    },
    {
      "epoch": 0.0940822278671559,
      "grad_norm": 0.09961670637130737,
      "learning_rate": 0.0002,
      "loss": 0.1465,
      "step": 500
    },
    {
      "epoch": 0.09784551698184213,
      "grad_norm": 0.08490657806396484,
      "learning_rate": 0.0002,
      "loss": 0.132,
      "step": 520
    },
    {
      "epoch": 0.10160880609652836,
      "grad_norm": 0.08765380829572678,
      "learning_rate": 0.0002,
      "loss": 0.1283,
      "step": 540
    },
    {
      "epoch": 0.1053720952112146,
      "grad_norm": 0.09319768846035004,
      "learning_rate": 0.0002,
      "loss": 0.1382,
      "step": 560
    },
    {
      "epoch": 0.10913538432590084,
      "grad_norm": 0.08941628783941269,
      "learning_rate": 0.0002,
      "loss": 0.1335,
      "step": 580
    },
    {
      "epoch": 0.11289867344058707,
      "grad_norm": 0.0971933901309967,
      "learning_rate": 0.0002,
      "loss": 0.137,
      "step": 600
    },
    {
      "epoch": 0.11666196255527331,
      "grad_norm": 0.07488075643777847,
      "learning_rate": 0.0002,
      "loss": 0.1283,
      "step": 620
    },
    {
      "epoch": 0.12042525166995954,
      "grad_norm": 0.08711710572242737,
      "learning_rate": 0.0002,
      "loss": 0.1292,
      "step": 640
    },
    {
      "epoch": 0.12418854078464578,
      "grad_norm": 0.08043856918811798,
      "learning_rate": 0.0002,
      "loss": 0.1269,
      "step": 660
    },
    {
      "epoch": 0.12795182989933201,
      "grad_norm": 0.07097792625427246,
      "learning_rate": 0.0002,
      "loss": 0.1364,
      "step": 680
    },
    {
      "epoch": 0.13171511901401825,
      "grad_norm": 0.0742156058549881,
      "learning_rate": 0.0002,
      "loss": 0.1277,
      "step": 700
    },
    {
      "epoch": 0.1354784081287045,
      "grad_norm": 0.09956187009811401,
      "learning_rate": 0.0002,
      "loss": 0.1314,
      "step": 720
    },
    {
      "epoch": 0.13924169724339072,
      "grad_norm": 0.08661571145057678,
      "learning_rate": 0.0002,
      "loss": 0.124,
      "step": 740
    },
    {
      "epoch": 0.14300498635807696,
      "grad_norm": 0.08070897310972214,
      "learning_rate": 0.0002,
      "loss": 0.1147,
      "step": 760
    },
    {
      "epoch": 0.1467682754727632,
      "grad_norm": 0.09516704082489014,
      "learning_rate": 0.0002,
      "loss": 0.1168,
      "step": 780
    },
    {
      "epoch": 0.15053156458744943,
      "grad_norm": 0.09076276421546936,
      "learning_rate": 0.0002,
      "loss": 0.1178,
      "step": 800
    },
    {
      "epoch": 0.15429485370213566,
      "grad_norm": 0.0896017774939537,
      "learning_rate": 0.0002,
      "loss": 0.1192,
      "step": 820
    },
    {
      "epoch": 0.1580581428168219,
      "grad_norm": 0.07477965205907822,
      "learning_rate": 0.0002,
      "loss": 0.1127,
      "step": 840
    },
    {
      "epoch": 0.16182143193150814,
      "grad_norm": 0.08405464887619019,
      "learning_rate": 0.0002,
      "loss": 0.119,
      "step": 860
    },
    {
      "epoch": 0.16558472104619437,
      "grad_norm": 0.07539790868759155,
      "learning_rate": 0.0002,
      "loss": 0.1213,
      "step": 880
    },
    {
      "epoch": 0.1693480101608806,
      "grad_norm": 0.08806908130645752,
      "learning_rate": 0.0002,
      "loss": 0.1192,
      "step": 900
    },
    {
      "epoch": 0.17311129927556684,
      "grad_norm": 0.08064749836921692,
      "learning_rate": 0.0002,
      "loss": 0.1141,
      "step": 920
    },
    {
      "epoch": 0.17687458839025308,
      "grad_norm": 0.09680119901895523,
      "learning_rate": 0.0002,
      "loss": 0.1042,
      "step": 940
    },
    {
      "epoch": 0.18063787750493931,
      "grad_norm": 0.09092500060796738,
      "learning_rate": 0.0002,
      "loss": 0.1067,
      "step": 960
    },
    {
      "epoch": 0.18440116661962555,
      "grad_norm": 0.07870171219110489,
      "learning_rate": 0.0002,
      "loss": 0.1101,
      "step": 980
    },
    {
      "epoch": 0.1881644557343118,
      "grad_norm": 0.06842092424631119,
      "learning_rate": 0.0002,
      "loss": 0.1045,
      "step": 1000
    },
    {
      "epoch": 0.19192774484899802,
      "grad_norm": 0.08229291439056396,
      "learning_rate": 0.0002,
      "loss": 0.114,
      "step": 1020
    },
    {
      "epoch": 0.19569103396368426,
      "grad_norm": 0.07617371529340744,
      "learning_rate": 0.0002,
      "loss": 0.1077,
      "step": 1040
    },
    {
      "epoch": 0.1994543230783705,
      "grad_norm": 0.09529408067464828,
      "learning_rate": 0.0002,
      "loss": 0.101,
      "step": 1060
    },
    {
      "epoch": 0.20321761219305673,
      "grad_norm": 0.07952335476875305,
      "learning_rate": 0.0002,
      "loss": 0.1065,
      "step": 1080
    },
    {
      "epoch": 0.20698090130774296,
      "grad_norm": 0.07625720649957657,
      "learning_rate": 0.0002,
      "loss": 0.0987,
      "step": 1100
    },
    {
      "epoch": 0.2107441904224292,
      "grad_norm": 0.07906854152679443,
      "learning_rate": 0.0002,
      "loss": 0.0986,
      "step": 1120
    },
    {
      "epoch": 0.21450747953711544,
      "grad_norm": 0.08430849015712738,
      "learning_rate": 0.0002,
      "loss": 0.1085,
      "step": 1140
    },
    {
      "epoch": 0.21827076865180167,
      "grad_norm": 0.0861297994852066,
      "learning_rate": 0.0002,
      "loss": 0.1066,
      "step": 1160
    },
    {
      "epoch": 0.2220340577664879,
      "grad_norm": 0.07576191425323486,
      "learning_rate": 0.0002,
      "loss": 0.1051,
      "step": 1180
    },
    {
      "epoch": 0.22579734688117414,
      "grad_norm": 0.08132428675889969,
      "learning_rate": 0.0002,
      "loss": 0.1036,
      "step": 1200
    },
    {
      "epoch": 0.22956063599586038,
      "grad_norm": 0.06960251182317734,
      "learning_rate": 0.0002,
      "loss": 0.1039,
      "step": 1220
    },
    {
      "epoch": 0.23332392511054661,
      "grad_norm": 0.08230841159820557,
      "learning_rate": 0.0002,
      "loss": 0.1046,
      "step": 1240
    },
    {
      "epoch": 0.23708721422523285,
      "grad_norm": 0.07119760662317276,
      "learning_rate": 0.0002,
      "loss": 0.1064,
      "step": 1260
    },
    {
      "epoch": 0.2408505033399191,
      "grad_norm": 0.06965576857328415,
      "learning_rate": 0.0002,
      "loss": 0.0964,
      "step": 1280
    },
    {
      "epoch": 0.24461379245460532,
      "grad_norm": 0.07533243298530579,
      "learning_rate": 0.0002,
      "loss": 0.0973,
      "step": 1300
    },
    {
      "epoch": 0.24837708156929156,
      "grad_norm": 0.07530753314495087,
      "learning_rate": 0.0002,
      "loss": 0.1002,
      "step": 1320
    },
    {
      "epoch": 0.2521403706839778,
      "grad_norm": 0.0701604038476944,
      "learning_rate": 0.0002,
      "loss": 0.1014,
      "step": 1340
    },
    {
      "epoch": 0.25590365979866403,
      "grad_norm": 0.08768032491207123,
      "learning_rate": 0.0002,
      "loss": 0.099,
      "step": 1360
    },
    {
      "epoch": 0.2596669489133503,
      "grad_norm": 0.0789860337972641,
      "learning_rate": 0.0002,
      "loss": 0.0947,
      "step": 1380
    },
    {
      "epoch": 0.2634302380280365,
      "grad_norm": 0.09132009744644165,
      "learning_rate": 0.0002,
      "loss": 0.0975,
      "step": 1400
    },
    {
      "epoch": 0.26719352714272276,
      "grad_norm": 0.07386859506368637,
      "learning_rate": 0.0002,
      "loss": 0.0997,
      "step": 1420
    },
    {
      "epoch": 0.270956816257409,
      "grad_norm": 0.07243089377880096,
      "learning_rate": 0.0002,
      "loss": 0.0956,
      "step": 1440
    },
    {
      "epoch": 0.27472010537209524,
      "grad_norm": 0.0814971774816513,
      "learning_rate": 0.0002,
      "loss": 0.0869,
      "step": 1460
    },
    {
      "epoch": 0.27848339448678144,
      "grad_norm": 0.07246191054582596,
      "learning_rate": 0.0002,
      "loss": 0.0958,
      "step": 1480
    },
    {
      "epoch": 0.2822466836014677,
      "grad_norm": 0.08997531235218048,
      "learning_rate": 0.0002,
      "loss": 0.0875,
      "step": 1500
    },
    {
      "epoch": 0.2860099727161539,
      "grad_norm": 0.08498572558164597,
      "learning_rate": 0.0002,
      "loss": 0.0872,
      "step": 1520
    },
    {
      "epoch": 0.2897732618308402,
      "grad_norm": 0.08210768550634384,
      "learning_rate": 0.0002,
      "loss": 0.085,
      "step": 1540
    },
    {
      "epoch": 0.2935365509455264,
      "grad_norm": 0.09801402688026428,
      "learning_rate": 0.0002,
      "loss": 0.0937,
      "step": 1560
    },
    {
      "epoch": 0.29729984006021265,
      "grad_norm": 0.07310175150632858,
      "learning_rate": 0.0002,
      "loss": 0.0898,
      "step": 1580
    },
    {
      "epoch": 0.30106312917489886,
      "grad_norm": 0.0679851844906807,
      "learning_rate": 0.0002,
      "loss": 0.0853,
      "step": 1600
    },
    {
      "epoch": 0.3048264182895851,
      "grad_norm": 0.0764717161655426,
      "learning_rate": 0.0002,
      "loss": 0.0966,
      "step": 1620
    },
    {
      "epoch": 0.30858970740427133,
      "grad_norm": 0.0813748836517334,
      "learning_rate": 0.0002,
      "loss": 0.0914,
      "step": 1640
    },
    {
      "epoch": 0.3123529965189576,
      "grad_norm": 0.07560984790325165,
      "learning_rate": 0.0002,
      "loss": 0.0889,
      "step": 1660
    },
    {
      "epoch": 0.3161162856336438,
      "grad_norm": 0.07131631672382355,
      "learning_rate": 0.0002,
      "loss": 0.0862,
      "step": 1680
    },
    {
      "epoch": 0.31987957474833006,
      "grad_norm": 0.07922184467315674,
      "learning_rate": 0.0002,
      "loss": 0.0908,
      "step": 1700
    },
    {
      "epoch": 0.3236428638630163,
      "grad_norm": 0.09043222665786743,
      "learning_rate": 0.0002,
      "loss": 0.0887,
      "step": 1720
    },
    {
      "epoch": 0.32740615297770254,
      "grad_norm": 0.08574160188436508,
      "learning_rate": 0.0002,
      "loss": 0.0895,
      "step": 1740
    },
    {
      "epoch": 0.33116944209238874,
      "grad_norm": 0.0885438472032547,
      "learning_rate": 0.0002,
      "loss": 0.0886,
      "step": 1760
    },
    {
      "epoch": 0.334932731207075,
      "grad_norm": 0.08546678721904755,
      "learning_rate": 0.0002,
      "loss": 0.0871,
      "step": 1780
    },
    {
      "epoch": 0.3386960203217612,
      "grad_norm": 0.0922817662358284,
      "learning_rate": 0.0002,
      "loss": 0.0877,
      "step": 1800
    },
    {
      "epoch": 0.3424593094364475,
      "grad_norm": 0.08879639208316803,
      "learning_rate": 0.0002,
      "loss": 0.0877,
      "step": 1820
    },
    {
      "epoch": 0.3462225985511337,
      "grad_norm": 0.09202056378126144,
      "learning_rate": 0.0002,
      "loss": 0.0827,
      "step": 1840
    },
    {
      "epoch": 0.34998588766581995,
      "grad_norm": 0.0998295396566391,
      "learning_rate": 0.0002,
      "loss": 0.0835,
      "step": 1860
    },
    {
      "epoch": 0.35374917678050616,
      "grad_norm": 0.09490591287612915,
      "learning_rate": 0.0002,
      "loss": 0.0862,
      "step": 1880
    },
    {
      "epoch": 0.3575124658951924,
      "grad_norm": 0.08920489251613617,
      "learning_rate": 0.0002,
      "loss": 0.0859,
      "step": 1900
    },
    {
      "epoch": 0.36127575500987863,
      "grad_norm": 0.0803561583161354,
      "learning_rate": 0.0002,
      "loss": 0.081,
      "step": 1920
    },
    {
      "epoch": 0.3650390441245649,
      "grad_norm": 0.08496900647878647,
      "learning_rate": 0.0002,
      "loss": 0.0835,
      "step": 1940
    },
    {
      "epoch": 0.3688023332392511,
      "grad_norm": 0.08658849447965622,
      "learning_rate": 0.0002,
      "loss": 0.0883,
      "step": 1960
    },
    {
      "epoch": 0.37256562235393736,
      "grad_norm": 0.08497461676597595,
      "learning_rate": 0.0002,
      "loss": 0.0811,
      "step": 1980
    },
    {
      "epoch": 0.3763289114686236,
      "grad_norm": 0.06599749624729156,
      "learning_rate": 0.0002,
      "loss": 0.0714,
      "step": 2000
    },
    {
      "epoch": 0.38009220058330984,
      "grad_norm": 0.07988911867141724,
      "learning_rate": 0.0002,
      "loss": 0.0874,
      "step": 2020
    },
    {
      "epoch": 0.38385548969799604,
      "grad_norm": 0.07323348522186279,
      "learning_rate": 0.0002,
      "loss": 0.0809,
      "step": 2040
    },
    {
      "epoch": 0.3876187788126823,
      "grad_norm": 0.07455869019031525,
      "learning_rate": 0.0002,
      "loss": 0.0819,
      "step": 2060
    },
    {
      "epoch": 0.3913820679273685,
      "grad_norm": 0.08383121341466904,
      "learning_rate": 0.0002,
      "loss": 0.0811,
      "step": 2080
    },
    {
      "epoch": 0.3951453570420548,
      "grad_norm": 0.07616332918405533,
      "learning_rate": 0.0002,
      "loss": 0.0802,
      "step": 2100
    },
    {
      "epoch": 0.398908646156741,
      "grad_norm": 0.08373293280601501,
      "learning_rate": 0.0002,
      "loss": 0.0878,
      "step": 2120
    },
    {
      "epoch": 0.40267193527142725,
      "grad_norm": 0.07459353655576706,
      "learning_rate": 0.0002,
      "loss": 0.0806,
      "step": 2140
    },
    {
      "epoch": 0.40643522438611346,
      "grad_norm": 0.08267400413751602,
      "learning_rate": 0.0002,
      "loss": 0.0799,
      "step": 2160
    },
    {
      "epoch": 0.4101985135007997,
      "grad_norm": 0.07844484597444534,
      "learning_rate": 0.0002,
      "loss": 0.0843,
      "step": 2180
    },
    {
      "epoch": 0.41396180261548593,
      "grad_norm": 0.07998470962047577,
      "learning_rate": 0.0002,
      "loss": 0.0821,
      "step": 2200
    },
    {
      "epoch": 0.4177250917301722,
      "grad_norm": 0.08860679715871811,
      "learning_rate": 0.0002,
      "loss": 0.0819,
      "step": 2220
    },
    {
      "epoch": 0.4214883808448584,
      "grad_norm": 0.07845838367938995,
      "learning_rate": 0.0002,
      "loss": 0.0802,
      "step": 2240
    },
    {
      "epoch": 0.42525166995954466,
      "grad_norm": 0.08801402896642685,
      "learning_rate": 0.0002,
      "loss": 0.0768,
      "step": 2260
    },
    {
      "epoch": 0.4290149590742309,
      "grad_norm": 0.06149598956108093,
      "learning_rate": 0.0002,
      "loss": 0.0806,
      "step": 2280
    },
    {
      "epoch": 0.43277824818891714,
      "grad_norm": 0.07813749462366104,
      "learning_rate": 0.0002,
      "loss": 0.0775,
      "step": 2300
    },
    {
      "epoch": 0.43654153730360334,
      "grad_norm": 0.08169027417898178,
      "learning_rate": 0.0002,
      "loss": 0.0774,
      "step": 2320
    },
    {
      "epoch": 0.4403048264182896,
      "grad_norm": 0.06748262792825699,
      "learning_rate": 0.0002,
      "loss": 0.0721,
      "step": 2340
    },
    {
      "epoch": 0.4440681155329758,
      "grad_norm": 0.07497742772102356,
      "learning_rate": 0.0002,
      "loss": 0.0687,
      "step": 2360
    },
    {
      "epoch": 0.4478314046476621,
      "grad_norm": 0.08851341158151627,
      "learning_rate": 0.0002,
      "loss": 0.0809,
      "step": 2380
    },
    {
      "epoch": 0.4515946937623483,
      "grad_norm": 0.06805267184972763,
      "learning_rate": 0.0002,
      "loss": 0.0783,
      "step": 2400
    },
    {
      "epoch": 0.45535798287703455,
      "grad_norm": 0.07998234033584595,
      "learning_rate": 0.0002,
      "loss": 0.0816,
      "step": 2420
    },
    {
      "epoch": 0.45912127199172076,
      "grad_norm": 0.07358496636152267,
      "learning_rate": 0.0002,
      "loss": 0.0715,
      "step": 2440
    },
    {
      "epoch": 0.462884561106407,
      "grad_norm": 0.08105491101741791,
      "learning_rate": 0.0002,
      "loss": 0.076,
      "step": 2460
    },
    {
      "epoch": 0.46664785022109323,
      "grad_norm": 0.08625109493732452,
      "learning_rate": 0.0002,
      "loss": 0.0731,
      "step": 2480
    },
    {
      "epoch": 0.4704111393357795,
      "grad_norm": 0.07986035943031311,
      "learning_rate": 0.0002,
      "loss": 0.0726,
      "step": 2500
    },
    {
      "epoch": 0.4741744284504657,
      "grad_norm": 0.07247906923294067,
      "learning_rate": 0.0002,
      "loss": 0.0763,
      "step": 2520
    },
    {
      "epoch": 0.47793771756515196,
      "grad_norm": 0.08120191842317581,
      "learning_rate": 0.0002,
      "loss": 0.0764,
      "step": 2540
    },
    {
      "epoch": 0.4817010066798382,
      "grad_norm": 0.09049087017774582,
      "learning_rate": 0.0002,
      "loss": 0.0799,
      "step": 2560
    },
    {
      "epoch": 0.48546429579452444,
      "grad_norm": 0.06859074532985687,
      "learning_rate": 0.0002,
      "loss": 0.0733,
      "step": 2580
    },
    {
      "epoch": 0.48922758490921064,
      "grad_norm": 0.07872766256332397,
      "learning_rate": 0.0002,
      "loss": 0.0724,
      "step": 2600
    },
    {
      "epoch": 0.4929908740238969,
      "grad_norm": 0.0663181021809578,
      "learning_rate": 0.0002,
      "loss": 0.0744,
      "step": 2620
    },
    {
      "epoch": 0.4967541631385831,
      "grad_norm": 0.07880811393260956,
      "learning_rate": 0.0002,
      "loss": 0.0759,
      "step": 2640
    },
    {
      "epoch": 0.5005174522532694,
      "grad_norm": 0.12468737363815308,
      "learning_rate": 0.0002,
      "loss": 0.0747,
      "step": 2660
    },
    {
      "epoch": 0.5042807413679556,
      "grad_norm": 0.08328507095575333,
      "learning_rate": 0.0002,
      "loss": 0.0754,
      "step": 2680
    },
    {
      "epoch": 0.5080440304826418,
      "grad_norm": 0.06665025651454926,
      "learning_rate": 0.0002,
      "loss": 0.0721,
      "step": 2700
    },
    {
      "epoch": 0.5118073195973281,
      "grad_norm": 0.07980209589004517,
      "learning_rate": 0.0002,
      "loss": 0.0733,
      "step": 2720
    },
    {
      "epoch": 0.5155706087120143,
      "grad_norm": 0.07951670140028,
      "learning_rate": 0.0002,
      "loss": 0.0792,
      "step": 2740
    },
    {
      "epoch": 0.5193338978267006,
      "grad_norm": 0.07515596598386765,
      "learning_rate": 0.0002,
      "loss": 0.0724,
      "step": 2760
    },
    {
      "epoch": 0.5230971869413867,
      "grad_norm": 0.08162270486354828,
      "learning_rate": 0.0002,
      "loss": 0.0714,
      "step": 2780
    },
    {
      "epoch": 0.526860476056073,
      "grad_norm": 0.08002326637506485,
      "learning_rate": 0.0002,
      "loss": 0.077,
      "step": 2800
    },
    {
      "epoch": 0.5306237651707593,
      "grad_norm": 0.07449716329574585,
      "learning_rate": 0.0002,
      "loss": 0.0707,
      "step": 2820
    },
    {
      "epoch": 0.5343870542854455,
      "grad_norm": 0.07990019023418427,
      "learning_rate": 0.0002,
      "loss": 0.071,
      "step": 2840
    },
    {
      "epoch": 0.5381503434001317,
      "grad_norm": 0.08538304269313812,
      "learning_rate": 0.0002,
      "loss": 0.0707,
      "step": 2860
    },
    {
      "epoch": 0.541913632514818,
      "grad_norm": 0.073255255818367,
      "learning_rate": 0.0002,
      "loss": 0.0699,
      "step": 2880
    },
    {
      "epoch": 0.5456769216295042,
      "grad_norm": 0.09101062268018723,
      "learning_rate": 0.0002,
      "loss": 0.0708,
      "step": 2900
    },
    {
      "epoch": 0.5494402107441905,
      "grad_norm": 0.0757337138056755,
      "learning_rate": 0.0002,
      "loss": 0.0702,
      "step": 2920
    },
    {
      "epoch": 0.5532034998588766,
      "grad_norm": 0.06553716212511063,
      "learning_rate": 0.0002,
      "loss": 0.0698,
      "step": 2940
    },
    {
      "epoch": 0.5569667889735629,
      "grad_norm": 0.08524072915315628,
      "learning_rate": 0.0002,
      "loss": 0.0738,
      "step": 2960
    },
    {
      "epoch": 0.5607300780882492,
      "grad_norm": 0.07361309975385666,
      "learning_rate": 0.0002,
      "loss": 0.0695,
      "step": 2980
    },
    {
      "epoch": 0.5644933672029354,
      "grad_norm": 0.07155182212591171,
      "learning_rate": 0.0002,
      "loss": 0.0667,
      "step": 3000
    },
    {
      "epoch": 0.5682566563176216,
      "grad_norm": 0.08088400214910507,
      "learning_rate": 0.0002,
      "loss": 0.0695,
      "step": 3020
    },
    {
      "epoch": 0.5720199454323078,
      "grad_norm": 0.08842818439006805,
      "learning_rate": 0.0002,
      "loss": 0.0656,
      "step": 3040
    },
    {
      "epoch": 0.5757832345469941,
      "grad_norm": 0.07135680317878723,
      "learning_rate": 0.0002,
      "loss": 0.0673,
      "step": 3060
    },
    {
      "epoch": 0.5795465236616804,
      "grad_norm": 0.08081484586000443,
      "learning_rate": 0.0002,
      "loss": 0.0709,
      "step": 3080
    },
    {
      "epoch": 0.5833098127763665,
      "grad_norm": 0.06999066472053528,
      "learning_rate": 0.0002,
      "loss": 0.0684,
      "step": 3100
    },
    {
      "epoch": 0.5870731018910528,
      "grad_norm": 0.06368447095155716,
      "learning_rate": 0.0002,
      "loss": 0.0691,
      "step": 3120
    },
    {
      "epoch": 0.590836391005739,
      "grad_norm": 0.08371565490961075,
      "learning_rate": 0.0002,
      "loss": 0.0641,
      "step": 3140
    },
    {
      "epoch": 0.5945996801204253,
      "grad_norm": 0.0675625279545784,
      "learning_rate": 0.0002,
      "loss": 0.069,
      "step": 3160
    },
    {
      "epoch": 0.5983629692351115,
      "grad_norm": 0.08121863007545471,
      "learning_rate": 0.0002,
      "loss": 0.065,
      "step": 3180
    },
    {
      "epoch": 0.6021262583497977,
      "grad_norm": 0.06689095497131348,
      "learning_rate": 0.0002,
      "loss": 0.0675,
      "step": 3200
    },
    {
      "epoch": 0.605889547464484,
      "grad_norm": 0.06634503602981567,
      "learning_rate": 0.0002,
      "loss": 0.064,
      "step": 3220
    },
    {
      "epoch": 0.6096528365791702,
      "grad_norm": 0.07062681019306183,
      "learning_rate": 0.0002,
      "loss": 0.0698,
      "step": 3240
    },
    {
      "epoch": 0.6134161256938564,
      "grad_norm": 0.07908321917057037,
      "learning_rate": 0.0002,
      "loss": 0.0674,
      "step": 3260
    },
    {
      "epoch": 0.6171794148085427,
      "grad_norm": 0.08132966607809067,
      "learning_rate": 0.0002,
      "loss": 0.0694,
      "step": 3280
    },
    {
      "epoch": 0.6209427039232289,
      "grad_norm": 0.081205353140831,
      "learning_rate": 0.0002,
      "loss": 0.0627,
      "step": 3300
    },
    {
      "epoch": 0.6247059930379152,
      "grad_norm": 0.07323930412530899,
      "learning_rate": 0.0002,
      "loss": 0.0645,
      "step": 3320
    },
    {
      "epoch": 0.6284692821526013,
      "grad_norm": 0.07979920506477356,
      "learning_rate": 0.0002,
      "loss": 0.0601,
      "step": 3340
    },
    {
      "epoch": 0.6322325712672876,
      "grad_norm": 0.06519381701946259,
      "learning_rate": 0.0002,
      "loss": 0.0677,
      "step": 3360
    },
    {
      "epoch": 0.6359958603819739,
      "grad_norm": 0.07861756533384323,
      "learning_rate": 0.0002,
      "loss": 0.0648,
      "step": 3380
    },
    {
      "epoch": 0.6397591494966601,
      "grad_norm": 0.09498826414346695,
      "learning_rate": 0.0002,
      "loss": 0.0644,
      "step": 3400
    },
    {
      "epoch": 0.6435224386113463,
      "grad_norm": 0.07069331407546997,
      "learning_rate": 0.0002,
      "loss": 0.0683,
      "step": 3420
    },
    {
      "epoch": 0.6472857277260325,
      "grad_norm": 0.07212232798337936,
      "learning_rate": 0.0002,
      "loss": 0.061,
      "step": 3440
    },
    {
      "epoch": 0.6510490168407188,
      "grad_norm": 0.06527985632419586,
      "learning_rate": 0.0002,
      "loss": 0.063,
      "step": 3460
    },
    {
      "epoch": 0.6548123059554051,
      "grad_norm": 0.09721993654966354,
      "learning_rate": 0.0002,
      "loss": 0.0699,
      "step": 3480
    },
    {
      "epoch": 0.6585755950700912,
      "grad_norm": 0.06381040811538696,
      "learning_rate": 0.0002,
      "loss": 0.0604,
      "step": 3500
    },
    {
      "epoch": 0.6623388841847775,
      "grad_norm": 0.07853078097105026,
      "learning_rate": 0.0002,
      "loss": 0.0649,
      "step": 3520
    },
    {
      "epoch": 0.6661021732994638,
      "grad_norm": 0.09212848544120789,
      "learning_rate": 0.0002,
      "loss": 0.0588,
      "step": 3540
    },
    {
      "epoch": 0.66986546241415,
      "grad_norm": 0.07097720354795456,
      "learning_rate": 0.0002,
      "loss": 0.0635,
      "step": 3560
    },
    {
      "epoch": 0.6736287515288362,
      "grad_norm": 0.08024760335683823,
      "learning_rate": 0.0002,
      "loss": 0.068,
      "step": 3580
    },
    {
      "epoch": 0.6773920406435224,
      "grad_norm": 0.07371170073747635,
      "learning_rate": 0.0002,
      "loss": 0.0652,
      "step": 3600
    },
    {
      "epoch": 0.6811553297582087,
      "grad_norm": 0.0894666537642479,
      "learning_rate": 0.0002,
      "loss": 0.0659,
      "step": 3620
    },
    {
      "epoch": 0.684918618872895,
      "grad_norm": 0.07533634454011917,
      "learning_rate": 0.0002,
      "loss": 0.0656,
      "step": 3640
    },
    {
      "epoch": 0.6886819079875811,
      "grad_norm": 0.09198255091905594,
      "learning_rate": 0.0002,
      "loss": 0.0645,
      "step": 3660
    },
    {
      "epoch": 0.6924451971022674,
      "grad_norm": 0.07213396579027176,
      "learning_rate": 0.0002,
      "loss": 0.0634,
      "step": 3680
    },
    {
      "epoch": 0.6962084862169536,
      "grad_norm": 0.06533551216125488,
      "learning_rate": 0.0002,
      "loss": 0.0666,
      "step": 3700
    },
    {
      "epoch": 0.6999717753316399,
      "grad_norm": 0.09535999596118927,
      "learning_rate": 0.0002,
      "loss": 0.0651,
      "step": 3720
    },
    {
      "epoch": 0.703735064446326,
      "grad_norm": 0.07631547003984451,
      "learning_rate": 0.0002,
      "loss": 0.0678,
      "step": 3740
    },
    {
      "epoch": 0.7074983535610123,
      "grad_norm": 0.09607011079788208,
      "learning_rate": 0.0002,
      "loss": 0.0681,
      "step": 3760
    },
    {
      "epoch": 0.7112616426756986,
      "grad_norm": 0.07209835946559906,
      "learning_rate": 0.0002,
      "loss": 0.068,
      "step": 3780
    },
    {
      "epoch": 0.7150249317903848,
      "grad_norm": 0.09137856960296631,
      "learning_rate": 0.0002,
      "loss": 0.0598,
      "step": 3800
    },
    {
      "epoch": 0.718788220905071,
      "grad_norm": 0.09964293241500854,
      "learning_rate": 0.0002,
      "loss": 0.0656,
      "step": 3820
    },
    {
      "epoch": 0.7225515100197573,
      "grad_norm": 0.07041902840137482,
      "learning_rate": 0.0002,
      "loss": 0.0617,
      "step": 3840
    },
    {
      "epoch": 0.7263147991344435,
      "grad_norm": 0.05418103560805321,
      "learning_rate": 0.0002,
      "loss": 0.0686,
      "step": 3860
    },
    {
      "epoch": 0.7300780882491298,
      "grad_norm": 0.07180003076791763,
      "learning_rate": 0.0002,
      "loss": 0.0627,
      "step": 3880
    },
    {
      "epoch": 0.7338413773638159,
      "grad_norm": 0.07342009246349335,
      "learning_rate": 0.0002,
      "loss": 0.0651,
      "step": 3900
    },
    {
      "epoch": 0.7376046664785022,
      "grad_norm": 0.08151030540466309,
      "learning_rate": 0.0002,
      "loss": 0.0645,
      "step": 3920
    },
    {
      "epoch": 0.7413679555931885,
      "grad_norm": 0.09977607429027557,
      "learning_rate": 0.0002,
      "loss": 0.0606,
      "step": 3940
    },
    {
      "epoch": 0.7451312447078747,
      "grad_norm": 0.06991346180438995,
      "learning_rate": 0.0002,
      "loss": 0.064,
      "step": 3960
    },
    {
      "epoch": 0.7488945338225609,
      "grad_norm": 0.06797617673873901,
      "learning_rate": 0.0002,
      "loss": 0.0595,
      "step": 3980
    },
    {
      "epoch": 0.7526578229372471,
      "grad_norm": 0.06056559830904007,
      "learning_rate": 0.0002,
      "loss": 0.0595,
      "step": 4000
    },
    {
      "epoch": 0.7564211120519334,
      "grad_norm": 0.06775388866662979,
      "learning_rate": 0.0002,
      "loss": 0.0624,
      "step": 4020
    },
    {
      "epoch": 0.7601844011666197,
      "grad_norm": 0.08091945201158524,
      "learning_rate": 0.0002,
      "loss": 0.0657,
      "step": 4040
    },
    {
      "epoch": 0.7639476902813058,
      "grad_norm": 0.1335289478302002,
      "learning_rate": 0.0002,
      "loss": 0.061,
      "step": 4060
    },
    {
      "epoch": 0.7677109793959921,
      "grad_norm": 0.07901336252689362,
      "learning_rate": 0.0002,
      "loss": 0.0635,
      "step": 4080
    },
    {
      "epoch": 0.7714742685106784,
      "grad_norm": 0.07874254137277603,
      "learning_rate": 0.0002,
      "loss": 0.066,
      "step": 4100
    },
    {
      "epoch": 0.7752375576253646,
      "grad_norm": 0.07547064125537872,
      "learning_rate": 0.0002,
      "loss": 0.061,
      "step": 4120
    },
    {
      "epoch": 0.7790008467400508,
      "grad_norm": 0.06392517685890198,
      "learning_rate": 0.0002,
      "loss": 0.0595,
      "step": 4140
    },
    {
      "epoch": 0.782764135854737,
      "grad_norm": 0.08115773648023605,
      "learning_rate": 0.0002,
      "loss": 0.0633,
      "step": 4160
    },
    {
      "epoch": 0.7865274249694233,
      "grad_norm": 0.07906658202409744,
      "learning_rate": 0.0002,
      "loss": 0.06,
      "step": 4180
    },
    {
      "epoch": 0.7902907140841096,
      "grad_norm": 0.8131846189498901,
      "learning_rate": 0.0002,
      "loss": 0.0642,
      "step": 4200
    },
    {
      "epoch": 0.7940540031987957,
      "grad_norm": 0.08816728740930557,
      "learning_rate": 0.0002,
      "loss": 0.0683,
      "step": 4220
    },
    {
      "epoch": 0.797817292313482,
      "grad_norm": 0.08981813490390778,
      "learning_rate": 0.0002,
      "loss": 0.0654,
      "step": 4240
    },
    {
      "epoch": 0.8015805814281682,
      "grad_norm": 0.08015542477369308,
      "learning_rate": 0.0002,
      "loss": 0.0623,
      "step": 4260
    },
    {
      "epoch": 0.8053438705428545,
      "grad_norm": 0.08604927361011505,
      "learning_rate": 0.0002,
      "loss": 0.0616,
      "step": 4280
    },
    {
      "epoch": 0.8091071596575407,
      "grad_norm": 0.08916622400283813,
      "learning_rate": 0.0002,
      "loss": 0.0651,
      "step": 4300
    },
    {
      "epoch": 0.8128704487722269,
      "grad_norm": 0.06978412717580795,
      "learning_rate": 0.0002,
      "loss": 0.0577,
      "step": 4320
    },
    {
      "epoch": 0.8166337378869132,
      "grad_norm": 0.089443139731884,
      "learning_rate": 0.0002,
      "loss": 0.0651,
      "step": 4340
    },
    {
      "epoch": 0.8203970270015994,
      "grad_norm": 0.08108926564455032,
      "learning_rate": 0.0002,
      "loss": 0.0584,
      "step": 4360
    },
    {
      "epoch": 0.8241603161162856,
      "grad_norm": 0.0913953110575676,
      "learning_rate": 0.0002,
      "loss": 0.0611,
      "step": 4380
    },
    {
      "epoch": 0.8279236052309719,
      "grad_norm": 0.07268033176660538,
      "learning_rate": 0.0002,
      "loss": 0.0601,
      "step": 4400
    },
    {
      "epoch": 0.8316868943456581,
      "grad_norm": 0.07602666318416595,
      "learning_rate": 0.0002,
      "loss": 0.0642,
      "step": 4420
    },
    {
      "epoch": 0.8354501834603444,
      "grad_norm": 0.06899157911539078,
      "learning_rate": 0.0002,
      "loss": 0.0603,
      "step": 4440
    },
    {
      "epoch": 0.8392134725750305,
      "grad_norm": 0.07310888916254044,
      "learning_rate": 0.0002,
      "loss": 0.0614,
      "step": 4460
    },
    {
      "epoch": 0.8429767616897168,
      "grad_norm": 0.086586132645607,
      "learning_rate": 0.0002,
      "loss": 0.0636,
      "step": 4480
    },
    {
      "epoch": 0.8467400508044031,
      "grad_norm": 0.04938528314232826,
      "learning_rate": 0.0002,
      "loss": 0.0577,
      "step": 4500
    },
    {
      "epoch": 0.8505033399190893,
      "grad_norm": 0.07192698121070862,
      "learning_rate": 0.0002,
      "loss": 0.0574,
      "step": 4520
    },
    {
      "epoch": 0.8542666290337755,
      "grad_norm": 0.06893257051706314,
      "learning_rate": 0.0002,
      "loss": 0.0604,
      "step": 4540
    },
    {
      "epoch": 0.8580299181484617,
      "grad_norm": 0.06691521406173706,
      "learning_rate": 0.0002,
      "loss": 0.0649,
      "step": 4560
    },
    {
      "epoch": 0.861793207263148,
      "grad_norm": 0.09420821070671082,
      "learning_rate": 0.0002,
      "loss": 0.0598,
      "step": 4580
    },
    {
      "epoch": 0.8655564963778343,
      "grad_norm": 0.07227174192667007,
      "learning_rate": 0.0002,
      "loss": 0.062,
      "step": 4600
    },
    {
      "epoch": 0.8693197854925204,
      "grad_norm": 0.07998435199260712,
      "learning_rate": 0.0002,
      "loss": 0.0585,
      "step": 4620
    },
    {
      "epoch": 0.8730830746072067,
      "grad_norm": 0.07850491255521774,
      "learning_rate": 0.0002,
      "loss": 0.0586,
      "step": 4640
    },
    {
      "epoch": 0.876846363721893,
      "grad_norm": 0.09450593590736389,
      "learning_rate": 0.0002,
      "loss": 0.0625,
      "step": 4660
    },
    {
      "epoch": 0.8806096528365792,
      "grad_norm": 0.08073689788579941,
      "learning_rate": 0.0002,
      "loss": 0.062,
      "step": 4680
    },
    {
      "epoch": 0.8843729419512654,
      "grad_norm": 0.06979519873857498,
      "learning_rate": 0.0002,
      "loss": 0.0584,
      "step": 4700
    },
    {
      "epoch": 0.8881362310659516,
      "grad_norm": 0.06407754868268967,
      "learning_rate": 0.0002,
      "loss": 0.0626,
      "step": 4720
    },
    {
      "epoch": 0.8918995201806379,
      "grad_norm": 0.2963426411151886,
      "learning_rate": 0.0002,
      "loss": 0.0625,
      "step": 4740
    },
    {
      "epoch": 0.8956628092953242,
      "grad_norm": 0.08902490139007568,
      "learning_rate": 0.0002,
      "loss": 0.0621,
      "step": 4760
    },
    {
      "epoch": 0.8994260984100103,
      "grad_norm": 0.08447249978780746,
      "learning_rate": 0.0002,
      "loss": 0.0638,
      "step": 4780
    },
    {
      "epoch": 0.9031893875246966,
      "grad_norm": 0.07621373981237411,
      "learning_rate": 0.0002,
      "loss": 0.0618,
      "step": 4800
    },
    {
      "epoch": 0.9069526766393828,
      "grad_norm": 0.07609863579273224,
      "learning_rate": 0.0002,
      "loss": 0.0627,
      "step": 4820
    },
    {
      "epoch": 0.9107159657540691,
      "grad_norm": 0.0785209983587265,
      "learning_rate": 0.0002,
      "loss": 0.0606,
      "step": 4840
    },
    {
      "epoch": 0.9144792548687553,
      "grad_norm": 0.08515089750289917,
      "learning_rate": 0.0002,
      "loss": 0.0577,
      "step": 4860
    },
    {
      "epoch": 0.9182425439834415,
      "grad_norm": 0.0810500979423523,
      "learning_rate": 0.0002,
      "loss": 0.0578,
      "step": 4880
    },
    {
      "epoch": 0.9220058330981278,
      "grad_norm": 0.08074364811182022,
      "learning_rate": 0.0002,
      "loss": 0.0556,
      "step": 4900
    },
    {
      "epoch": 0.925769122212814,
      "grad_norm": 0.06328209489583969,
      "learning_rate": 0.0002,
      "loss": 0.0555,
      "step": 4920
    },
    {
      "epoch": 0.9295324113275002,
      "grad_norm": 0.06556906551122665,
      "learning_rate": 0.0002,
      "loss": 0.056,
      "step": 4940
    },
    {
      "epoch": 0.9332957004421865,
      "grad_norm": 0.0747644379734993,
      "learning_rate": 0.0002,
      "loss": 0.0608,
      "step": 4960
    },
    {
      "epoch": 0.9370589895568727,
      "grad_norm": 0.07274675369262695,
      "learning_rate": 0.0002,
      "loss": 0.057,
      "step": 4980
    },
    {
      "epoch": 0.940822278671559,
      "grad_norm": 0.061261050403118134,
      "learning_rate": 0.0002,
      "loss": 0.0587,
      "step": 5000
    },
    {
      "epoch": 0.9445855677862451,
      "grad_norm": 0.08277834206819534,
      "learning_rate": 0.0002,
      "loss": 0.0608,
      "step": 5020
    },
    {
      "epoch": 0.9483488569009314,
      "grad_norm": 0.08663126826286316,
      "learning_rate": 0.0002,
      "loss": 0.0595,
      "step": 5040
    },
    {
      "epoch": 0.9521121460156177,
      "grad_norm": 0.06938447803258896,
      "learning_rate": 0.0002,
      "loss": 0.0567,
      "step": 5060
    },
    {
      "epoch": 0.9558754351303039,
      "grad_norm": 0.07865908741950989,
      "learning_rate": 0.0002,
      "loss": 0.0588,
      "step": 5080
    },
    {
      "epoch": 0.9596387242449901,
      "grad_norm": 0.061116304248571396,
      "learning_rate": 0.0002,
      "loss": 0.0574,
      "step": 5100
    },
    {
      "epoch": 0.9634020133596763,
      "grad_norm": 0.05689473822712898,
      "learning_rate": 0.0002,
      "loss": 0.0533,
      "step": 5120
    },
    {
      "epoch": 0.9671653024743626,
      "grad_norm": 0.07368452847003937,
      "learning_rate": 0.0002,
      "loss": 0.0591,
      "step": 5140
    },
    {
      "epoch": 0.9709285915890489,
      "grad_norm": 0.06758634001016617,
      "learning_rate": 0.0002,
      "loss": 0.0572,
      "step": 5160
    },
    {
      "epoch": 0.974691880703735,
      "grad_norm": 0.07348816096782684,
      "learning_rate": 0.0002,
      "loss": 0.0577,
      "step": 5180
    },
    {
      "epoch": 0.9784551698184213,
      "grad_norm": 0.07613357156515121,
      "learning_rate": 0.0002,
      "loss": 0.0566,
      "step": 5200
    },
    {
      "epoch": 0.9822184589331076,
      "grad_norm": 0.062316812574863434,
      "learning_rate": 0.0002,
      "loss": 0.0621,
      "step": 5220
    },
    {
      "epoch": 0.9859817480477938,
      "grad_norm": 0.08600709587335587,
      "learning_rate": 0.0002,
      "loss": 0.0576,
      "step": 5240
    },
    {
      "epoch": 0.98974503716248,
      "grad_norm": 0.06687742471694946,
      "learning_rate": 0.0002,
      "loss": 0.0571,
      "step": 5260
    },
    {
      "epoch": 0.9935083262771662,
      "grad_norm": 0.07486743479967117,
      "learning_rate": 0.0002,
      "loss": 0.0594,
      "step": 5280
    },
    {
      "epoch": 0.9972716153918525,
      "grad_norm": 0.08373595029115677,
      "learning_rate": 0.0002,
      "loss": 0.0579,
      "step": 5300
    },
    {
      "epoch": 0.9999059177721329,
      "eval_loss": 0.20169740915298462,
      "eval_runtime": 422.4603,
      "eval_samples_per_second": 4.1,
      "eval_steps_per_second": 0.514,
      "step": 5314
    }
  ],
  "logging_steps": 20,
  "max_steps": 14000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 77,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.423221588013613e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}