{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.0,
  "eval_steps": 5030,
  "global_step": 150876,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.019883878151594685,
      "grad_norm": 2.3147895336151123,
      "learning_rate": 0.0001994694115774394,
      "loss": 0.594,
      "step": 500
    },
    {
      "epoch": 0.03976775630318937,
      "grad_norm": 2.2017321586608887,
      "learning_rate": 0.0001988061760492386,
      "loss": 0.4251,
      "step": 1000
    },
    {
      "epoch": 0.05965163445478406,
      "grad_norm": 2.0374693870544434,
      "learning_rate": 0.00019814294052103784,
      "loss": 0.386,
      "step": 1500
    },
    {
      "epoch": 0.07953551260637874,
      "grad_norm": 1.3447909355163574,
      "learning_rate": 0.00019747970499283706,
      "loss": 0.3585,
      "step": 2000
    },
    {
      "epoch": 0.09941939075797343,
      "grad_norm": 1.2458261251449585,
      "learning_rate": 0.00019682044887780548,
      "loss": 0.3545,
      "step": 2500
    },
    {
      "epoch": 0.11930326890956812,
      "grad_norm": 1.844347357749939,
      "learning_rate": 0.00019615721334960472,
      "loss": 0.3329,
      "step": 3000
    },
    {
      "epoch": 0.1391871470611628,
      "grad_norm": 1.3815340995788574,
      "learning_rate": 0.00019549397782140396,
      "loss": 0.3182,
      "step": 3500
    },
    {
      "epoch": 0.15907102521275748,
      "grad_norm": 2.2595863342285156,
      "learning_rate": 0.00019483074229320317,
      "loss": 0.2985,
      "step": 4000
    },
    {
      "epoch": 0.17895490336435219,
      "grad_norm": 1.8043708801269531,
      "learning_rate": 0.00019416750676500238,
      "loss": 0.3068,
      "step": 4500
    },
    {
      "epoch": 0.19883878151594686,
      "grad_norm": 1.8568594455718994,
      "learning_rate": 0.00019350427123680162,
      "loss": 0.3044,
      "step": 5000
    },
    {
      "epoch": 0.20003181420504254,
      "eval_loss": 0.29235559701919556,
      "eval_runtime": 31.6112,
      "eval_samples_per_second": 15.817,
      "eval_steps_per_second": 7.909,
      "step": 5030
    },
    {
      "epoch": 0.21872265966754156,
      "grad_norm": 1.780515193939209,
      "learning_rate": 0.00019284103570860084,
      "loss": 0.2884,
      "step": 5500
    },
    {
      "epoch": 0.23860653781913624,
      "grad_norm": 1.3652188777923584,
      "learning_rate": 0.00019217780018040008,
      "loss": 0.2924,
      "step": 6000
    },
    {
      "epoch": 0.2584904159707309,
      "grad_norm": 1.5280920267105103,
      "learning_rate": 0.0001915158911232557,
      "loss": 0.2714,
      "step": 6500
    },
    {
      "epoch": 0.2783742941223256,
      "grad_norm": 1.8563861846923828,
      "learning_rate": 0.00019085398206611135,
      "loss": 0.2729,
      "step": 7000
    },
    {
      "epoch": 0.2982581722739203,
      "grad_norm": 1.0958434343338013,
      "learning_rate": 0.00019019074653791056,
      "loss": 0.2735,
      "step": 7500
    },
    {
      "epoch": 0.31814205042551497,
      "grad_norm": 1.145859956741333,
      "learning_rate": 0.00018952751100970977,
      "loss": 0.2737,
      "step": 8000
    },
    {
      "epoch": 0.33802592857710967,
      "grad_norm": 1.2784571647644043,
      "learning_rate": 0.000188864275481509,
      "loss": 0.2611,
      "step": 8500
    },
    {
      "epoch": 0.35790980672870437,
      "grad_norm": 1.3518733978271484,
      "learning_rate": 0.00018820103995330822,
      "loss": 0.2612,
      "step": 9000
    },
    {
      "epoch": 0.3777936848802991,
      "grad_norm": 1.4692599773406982,
      "learning_rate": 0.00018753780442510743,
      "loss": 0.2609,
      "step": 9500
    },
    {
      "epoch": 0.3976775630318937,
      "grad_norm": 2.682018518447876,
      "learning_rate": 0.00018687456889690667,
      "loss": 0.2643,
      "step": 10000
    },
    {
      "epoch": 0.4000636284100851,
      "eval_loss": 0.2585141062736511,
      "eval_runtime": 31.9094,
      "eval_samples_per_second": 15.669,
      "eval_steps_per_second": 7.835,
      "step": 10060
    },
    {
      "epoch": 0.4175614411834884,
      "grad_norm": 1.7332996129989624,
      "learning_rate": 0.00018621133336870591,
      "loss": 0.2523,
      "step": 10500
    },
    {
      "epoch": 0.4374453193350831,
      "grad_norm": 2.1415417194366455,
      "learning_rate": 0.00018554942431156152,
      "loss": 0.2585,
      "step": 11000
    },
    {
      "epoch": 0.4573291974866778,
      "grad_norm": 1.2841447591781616,
      "learning_rate": 0.00018488618878336073,
      "loss": 0.2557,
      "step": 11500
    },
    {
      "epoch": 0.4772130756382725,
      "grad_norm": 3.2847816944122314,
      "learning_rate": 0.00018422295325516,
      "loss": 0.2609,
      "step": 12000
    },
    {
      "epoch": 0.4970969537898672,
      "grad_norm": 0.7331926822662354,
      "learning_rate": 0.00018355971772695921,
      "loss": 0.2477,
      "step": 12500
    },
    {
      "epoch": 0.5169808319414618,
      "grad_norm": 1.4415650367736816,
      "learning_rate": 0.00018289648219875843,
      "loss": 0.2482,
      "step": 13000
    },
    {
      "epoch": 0.5368647100930566,
      "grad_norm": 1.7630778551101685,
      "learning_rate": 0.00018223457314161406,
      "loss": 0.2421,
      "step": 13500
    },
    {
      "epoch": 0.5567485882446512,
      "grad_norm": 2.0509490966796875,
      "learning_rate": 0.0001815713376134133,
      "loss": 0.2444,
      "step": 14000
    },
    {
      "epoch": 0.5766324663962459,
      "grad_norm": 1.959215760231018,
      "learning_rate": 0.0001809081020852125,
      "loss": 0.2398,
      "step": 14500
    },
    {
      "epoch": 0.5965163445478406,
      "grad_norm": 1.6416336297988892,
      "learning_rate": 0.00018024486655701173,
      "loss": 0.2396,
      "step": 15000
    },
    {
      "epoch": 0.6000954426151277,
      "eval_loss": 0.23102714121341705,
      "eval_runtime": 32.4432,
      "eval_samples_per_second": 15.412,
      "eval_steps_per_second": 7.706,
      "step": 15090
    },
    {
      "epoch": 0.6164002226994353,
      "grad_norm": 0.5961350798606873,
      "learning_rate": 0.00017958163102881097,
      "loss": 0.2291,
      "step": 15500
    },
    {
      "epoch": 0.6362841008510299,
      "grad_norm": 1.3288564682006836,
      "learning_rate": 0.00017891972197166657,
      "loss": 0.2368,
      "step": 16000
    },
    {
      "epoch": 0.6561679790026247,
      "grad_norm": 1.7699204683303833,
      "learning_rate": 0.0001782564864434658,
      "loss": 0.2397,
      "step": 16500
    },
    {
      "epoch": 0.6760518571542193,
      "grad_norm": 1.3349616527557373,
      "learning_rate": 0.00017759325091526505,
      "loss": 0.2353,
      "step": 17000
    },
    {
      "epoch": 0.6959357353058141,
      "grad_norm": 1.6228617429733276,
      "learning_rate": 0.00017693001538706426,
      "loss": 0.2314,
      "step": 17500
    },
    {
      "epoch": 0.7158196134574087,
      "grad_norm": 1.349108338356018,
      "learning_rate": 0.00017626677985886348,
      "loss": 0.2339,
      "step": 18000
    },
    {
      "epoch": 0.7357034916090034,
      "grad_norm": 2.2128469944000244,
      "learning_rate": 0.00017560354433066272,
      "loss": 0.2252,
      "step": 18500
    },
    {
      "epoch": 0.7555873697605981,
      "grad_norm": 2.038177013397217,
      "learning_rate": 0.00017494163527351835,
      "loss": 0.2242,
      "step": 19000
    },
    {
      "epoch": 0.7754712479121928,
      "grad_norm": 1.1377464532852173,
      "learning_rate": 0.00017427839974531756,
      "loss": 0.2197,
      "step": 19500
    },
    {
      "epoch": 0.7953551260637874,
      "grad_norm": 1.2686738967895508,
      "learning_rate": 0.00017361516421711678,
      "loss": 0.2277,
      "step": 20000
    },
    {
      "epoch": 0.8001272568201702,
      "eval_loss": 0.21075737476348877,
      "eval_runtime": 32.819,
      "eval_samples_per_second": 15.235,
      "eval_steps_per_second": 7.618,
      "step": 20120
    },
    {
      "epoch": 0.8152390042153822,
      "grad_norm": 2.1816446781158447,
      "learning_rate": 0.00017295192868891602,
      "loss": 0.2245,
      "step": 20500
    },
    {
      "epoch": 0.8351228823669768,
      "grad_norm": 2.224292755126953,
      "learning_rate": 0.00017229001963177165,
      "loss": 0.2176,
      "step": 21000
    },
    {
      "epoch": 0.8550067605185715,
      "grad_norm": 0.7720803618431091,
      "learning_rate": 0.00017162811057462726,
      "loss": 0.2097,
      "step": 21500
    },
    {
      "epoch": 0.8748906386701663,
      "grad_norm": 1.716422200202942,
      "learning_rate": 0.0001709648750464265,
      "loss": 0.2104,
      "step": 22000
    },
    {
      "epoch": 0.8947745168217609,
      "grad_norm": 1.5487083196640015,
      "learning_rate": 0.00017030163951822574,
      "loss": 0.2179,
      "step": 22500
    },
    {
      "epoch": 0.9146583949733555,
      "grad_norm": 1.0968928337097168,
      "learning_rate": 0.00016963840399002495,
      "loss": 0.2194,
      "step": 23000
    },
    {
      "epoch": 0.9345422731249503,
      "grad_norm": 0.36388519406318665,
      "learning_rate": 0.00016897516846182416,
      "loss": 0.2259,
      "step": 23500
    },
    {
      "epoch": 0.954426151276545,
      "grad_norm": 1.2857156991958618,
      "learning_rate": 0.0001683119329336234,
      "loss": 0.2195,
      "step": 24000
    },
    {
      "epoch": 0.9743100294281397,
      "grad_norm": 1.7136821746826172,
      "learning_rate": 0.00016764869740542261,
      "loss": 0.2036,
      "step": 24500
    },
    {
      "epoch": 0.9941939075797344,
      "grad_norm": 2.1068880558013916,
      "learning_rate": 0.00016698546187722183,
      "loss": 0.2103,
      "step": 25000
    },
    {
      "epoch": 1.0001590710252128,
      "eval_loss": 0.2087584286928177,
      "eval_runtime": 34.3797,
      "eval_samples_per_second": 14.543,
      "eval_steps_per_second": 7.272,
      "step": 25150
    },
    {
      "epoch": 1.014077785731329,
      "grad_norm": 0.9411464929580688,
      "learning_rate": 0.0001663222263490211,
      "loss": 0.1902,
      "step": 25500
    },
    {
      "epoch": 1.0339616638829237,
      "grad_norm": 1.8499776124954224,
      "learning_rate": 0.0001656589908208203,
      "loss": 0.1961,
      "step": 26000
    },
    {
      "epoch": 1.0538455420345185,
      "grad_norm": 2.804515838623047,
      "learning_rate": 0.00016499708176367591,
      "loss": 0.1975,
      "step": 26500
    },
    {
      "epoch": 1.0737294201861132,
      "grad_norm": 1.0124012231826782,
      "learning_rate": 0.00016433384623547515,
      "loss": 0.1965,
      "step": 27000
    },
    {
      "epoch": 1.0936132983377078,
      "grad_norm": 2.1755409240722656,
      "learning_rate": 0.0001636706107072744,
      "loss": 0.2032,
      "step": 27500
    },
    {
      "epoch": 1.1134971764893025,
      "grad_norm": 1.9173059463500977,
      "learning_rate": 0.0001630073751790736,
      "loss": 0.1998,
      "step": 28000
    },
    {
      "epoch": 1.1333810546408971,
      "grad_norm": 2.6213159561157227,
      "learning_rate": 0.0001623454661219292,
      "loss": 0.1944,
      "step": 28500
    },
    {
      "epoch": 1.1532649327924918,
      "grad_norm": 1.979064702987671,
      "learning_rate": 0.00016168223059372845,
      "loss": 0.1947,
      "step": 29000
    },
    {
      "epoch": 1.1731488109440866,
      "grad_norm": 3.864307403564453,
      "learning_rate": 0.00016101899506552767,
      "loss": 0.1948,
      "step": 29500
    },
    {
      "epoch": 1.1930326890956813,
      "grad_norm": 1.3037413358688354,
      "learning_rate": 0.0001603557595373269,
      "loss": 0.1942,
      "step": 30000
    },
    {
      "epoch": 1.2001908852302554,
      "eval_loss": 0.20310936868190765,
      "eval_runtime": 33.1759,
      "eval_samples_per_second": 15.071,
      "eval_steps_per_second": 7.536,
      "step": 30180
    },
    {
      "epoch": 1.212916567247276,
      "grad_norm": 1.4742060899734497,
      "learning_rate": 0.00015969385048018254,
      "loss": 0.1873,
      "step": 30500
    },
    {
      "epoch": 1.2328004453988706,
      "grad_norm": 0.8921416401863098,
      "learning_rate": 0.00015903061495198175,
      "loss": 0.1888,
      "step": 31000
    },
    {
      "epoch": 1.2526843235504652,
      "grad_norm": 2.814047336578369,
      "learning_rate": 0.00015836737942378096,
      "loss": 0.1933,
      "step": 31500
    },
    {
      "epoch": 1.27256820170206,
      "grad_norm": 1.6011271476745605,
      "learning_rate": 0.0001577041438955802,
      "loss": 0.1943,
      "step": 32000
    },
    {
      "epoch": 1.2924520798536547,
      "grad_norm": 2.1085023880004883,
      "learning_rate": 0.00015704090836737944,
      "loss": 0.1939,
      "step": 32500
    },
    {
      "epoch": 1.3123359580052494,
      "grad_norm": 1.4254250526428223,
      "learning_rate": 0.00015637899931023505,
      "loss": 0.1898,
      "step": 33000
    },
    {
      "epoch": 1.332219836156844,
      "grad_norm": 2.6085307598114014,
      "learning_rate": 0.00015571576378203426,
      "loss": 0.1869,
      "step": 33500
    },
    {
      "epoch": 1.3521037143084387,
      "grad_norm": 1.4234007596969604,
      "learning_rate": 0.00015505252825383353,
      "loss": 0.1924,
      "step": 34000
    },
    {
      "epoch": 1.3719875924600333,
      "grad_norm": 1.5566816329956055,
      "learning_rate": 0.00015439061919668914,
      "loss": 0.1932,
      "step": 34500
    },
    {
      "epoch": 1.391871470611628,
      "grad_norm": 8.811753273010254,
      "learning_rate": 0.00015372738366848835,
      "loss": 0.1846,
      "step": 35000
    },
    {
      "epoch": 1.400222699435298,
      "eval_loss": 0.19551397860050201,
      "eval_runtime": 35.3626,
      "eval_samples_per_second": 14.139,
      "eval_steps_per_second": 7.07,
      "step": 35210
    },
    {
      "epoch": 1.4117553487632228,
      "grad_norm": 1.463742971420288,
      "learning_rate": 0.0001530641481402876,
      "loss": 0.1924,
      "step": 35500
    },
    {
      "epoch": 1.4316392269148175,
      "grad_norm": 2.8610801696777344,
      "learning_rate": 0.00015240091261208683,
      "loss": 0.1938,
      "step": 36000
    },
    {
      "epoch": 1.4515231050664121,
      "grad_norm": 1.2369511127471924,
      "learning_rate": 0.00015173767708388604,
      "loss": 0.1817,
      "step": 36500
    },
    {
      "epoch": 1.4714069832180068,
      "grad_norm": 1.4233882427215576,
      "learning_rate": 0.00015107444155568526,
      "loss": 0.1817,
      "step": 37000
    },
    {
      "epoch": 1.4912908613696017,
      "grad_norm": 1.5385404825210571,
      "learning_rate": 0.0001504112060274845,
      "loss": 0.1849,
      "step": 37500
    },
    {
      "epoch": 1.5111747395211963,
      "grad_norm": 1.2022610902786255,
      "learning_rate": 0.0001497479704992837,
      "loss": 0.183,
      "step": 38000
    },
    {
      "epoch": 1.531058617672791,
      "grad_norm": 1.5988807678222656,
      "learning_rate": 0.00014908606144213934,
      "loss": 0.1821,
      "step": 38500
    },
    {
      "epoch": 1.5509424958243856,
      "grad_norm": 1.480351209640503,
      "learning_rate": 0.00014842282591393855,
      "loss": 0.1787,
      "step": 39000
    },
    {
      "epoch": 1.5708263739759802,
      "grad_norm": 1.9051077365875244,
      "learning_rate": 0.0001477595903857378,
      "loss": 0.1826,
      "step": 39500
    },
    {
      "epoch": 1.5907102521275749,
      "grad_norm": 2.0775561332702637,
      "learning_rate": 0.000147096354857537,
      "loss": 0.1814,
      "step": 40000
    },
    {
      "epoch": 1.6002545136403405,
      "eval_loss": 0.18265177309513092,
      "eval_runtime": 31.3228,
      "eval_samples_per_second": 15.963,
      "eval_steps_per_second": 7.981,
      "step": 40240
    },
    {
      "epoch": 1.6105941302791695,
      "grad_norm": 2.4270079135894775,
      "learning_rate": 0.00014643444580039264,
      "loss": 0.1728,
      "step": 40500
    },
    {
      "epoch": 1.6304780084307642,
      "grad_norm": 1.4693158864974976,
      "learning_rate": 0.00014577121027219188,
      "loss": 0.1835,
      "step": 41000
    },
    {
      "epoch": 1.650361886582359,
      "grad_norm": 2.014503240585327,
      "learning_rate": 0.0001451079747439911,
      "loss": 0.1825,
      "step": 41500
    },
    {
      "epoch": 1.6702457647339537,
      "grad_norm": 2.4731264114379883,
      "learning_rate": 0.0001444447392157903,
      "loss": 0.1803,
      "step": 42000
    },
    {
      "epoch": 1.6901296428855483,
      "grad_norm": 1.043967366218567,
      "learning_rate": 0.00014378150368758955,
      "loss": 0.1831,
      "step": 42500
    },
    {
      "epoch": 1.7100135210371432,
      "grad_norm": 2.0062990188598633,
      "learning_rate": 0.00014311826815938879,
      "loss": 0.1851,
      "step": 43000
    },
    {
      "epoch": 1.7298973991887379,
      "grad_norm": 2.958763837814331,
      "learning_rate": 0.000142455032631188,
      "loss": 0.1825,
      "step": 43500
    },
    {
      "epoch": 1.7497812773403325,
      "grad_norm": 1.6624678373336792,
      "learning_rate": 0.0001417917971029872,
      "loss": 0.1834,
      "step": 44000
    },
    {
      "epoch": 1.7696651554919272,
      "grad_norm": 1.1731349229812622,
      "learning_rate": 0.00014112856157478645,
      "loss": 0.1775,
      "step": 44500
    },
    {
      "epoch": 1.7895490336435218,
      "grad_norm": 1.2256383895874023,
      "learning_rate": 0.00014046665251764206,
      "loss": 0.1688,
      "step": 45000
    },
    {
      "epoch": 1.8002863278453831,
      "eval_loss": 0.1788606494665146,
      "eval_runtime": 31.0353,
      "eval_samples_per_second": 16.111,
      "eval_steps_per_second": 8.055,
      "step": 45270
    },
    {
      "epoch": 1.8094329117951164,
      "grad_norm": 5.187135219573975,
      "learning_rate": 0.0001398034169894413,
      "loss": 0.1757,
      "step": 45500
    },
    {
      "epoch": 1.829316789946711,
      "grad_norm": 3.0221340656280518,
      "learning_rate": 0.00013914018146124054,
      "loss": 0.169,
      "step": 46000
    },
    {
      "epoch": 1.8492006680983057,
      "grad_norm": 1.780038833618164,
      "learning_rate": 0.00013847694593303975,
      "loss": 0.1815,
      "step": 46500
    },
    {
      "epoch": 1.8690845462499006,
      "grad_norm": 1.5590816736221313,
      "learning_rate": 0.00013781503687589536,
      "loss": 0.1754,
      "step": 47000
    },
    {
      "epoch": 1.8889684244014953,
      "grad_norm": 2.1302263736724854,
      "learning_rate": 0.0001371518013476946,
      "loss": 0.1737,
      "step": 47500
    },
    {
      "epoch": 1.90885230255309,
      "grad_norm": 2.173957109451294,
      "learning_rate": 0.00013648856581949384,
      "loss": 0.1696,
      "step": 48000
    },
    {
      "epoch": 1.9287361807046848,
      "grad_norm": 1.0864589214324951,
      "learning_rate": 0.00013582533029129305,
      "loss": 0.1832,
      "step": 48500
    },
    {
      "epoch": 1.9486200588562794,
      "grad_norm": 1.9979732036590576,
      "learning_rate": 0.00013516209476309226,
      "loss": 0.1758,
      "step": 49000
    },
    {
      "epoch": 1.968503937007874,
      "grad_norm": 2.2656006813049316,
      "learning_rate": 0.00013450018570594792,
      "loss": 0.1728,
      "step": 49500
    },
    {
      "epoch": 1.9883878151594687,
      "grad_norm": 2.077143669128418,
      "learning_rate": 0.00013383695017774714,
      "loss": 0.1772,
      "step": 50000
    },
    {
      "epoch": 2.0003181420504257,
      "eval_loss": 0.1767009198665619,
      "eval_runtime": 32.0718,
      "eval_samples_per_second": 15.59,
      "eval_steps_per_second": 7.795,
      "step": 50300
    },
    {
      "epoch": 2.0082716933110634,
      "grad_norm": 1.5150219202041626,
      "learning_rate": 0.00013317371464954635,
      "loss": 0.164,
      "step": 50500
    },
    {
      "epoch": 2.028155571462658,
      "grad_norm": 1.2225052118301392,
      "learning_rate": 0.0001325104791213456,
      "loss": 0.1534,
      "step": 51000
    },
    {
      "epoch": 2.0480394496142527,
      "grad_norm": 1.9311732053756714,
      "learning_rate": 0.0001318472435931448,
      "loss": 0.1584,
      "step": 51500
    },
    {
      "epoch": 2.0679233277658473,
      "grad_norm": 2.399226188659668,
      "learning_rate": 0.000131184008064944,
      "loss": 0.153,
      "step": 52000
    },
    {
      "epoch": 2.087807205917442,
      "grad_norm": 2.9771511554718018,
      "learning_rate": 0.00013052077253674325,
      "loss": 0.1528,
      "step": 52500
    },
    {
      "epoch": 2.107691084069037,
      "grad_norm": 1.5730229616165161,
      "learning_rate": 0.0001298588634795989,
      "loss": 0.154,
      "step": 53000
    },
    {
      "epoch": 2.1275749622206317,
      "grad_norm": 1.2603771686553955,
      "learning_rate": 0.0001291956279513981,
      "loss": 0.1575,
      "step": 53500
    },
    {
      "epoch": 2.1474588403722263,
      "grad_norm": 2.315129280090332,
      "learning_rate": 0.00012853371889425373,
      "loss": 0.1603,
      "step": 54000
    },
    {
      "epoch": 2.167342718523821,
      "grad_norm": 1.5500929355621338,
      "learning_rate": 0.00012787048336605297,
      "loss": 0.1544,
      "step": 54500
    },
    {
      "epoch": 2.1872265966754156,
      "grad_norm": 1.2831624746322632,
      "learning_rate": 0.0001272072478378522,
      "loss": 0.143,
      "step": 55000
    },
    {
      "epoch": 2.2003499562554683,
      "eval_loss": 0.1747935712337494,
      "eval_runtime": 31.3745,
      "eval_samples_per_second": 15.936,
      "eval_steps_per_second": 7.968,
      "step": 55330
    },
    {
      "epoch": 2.2071104748270103,
      "grad_norm": 1.4339005947113037,
      "learning_rate": 0.0001265440123096514,
      "loss": 0.1587,
      "step": 55500
    },
    {
      "epoch": 2.226994352978605,
      "grad_norm": 2.5719878673553467,
      "learning_rate": 0.00012588077678145064,
      "loss": 0.1527,
      "step": 56000
    },
    {
      "epoch": 2.2468782311301996,
      "grad_norm": 2.1503918170928955,
      "learning_rate": 0.00012521754125324988,
      "loss": 0.1585,
      "step": 56500
    },
    {
      "epoch": 2.2667621092817942,
      "grad_norm": 3.443103551864624,
      "learning_rate": 0.0001245543057250491,
      "loss": 0.1593,
      "step": 57000
    },
    {
      "epoch": 2.286645987433389,
      "grad_norm": 1.2123481035232544,
      "learning_rate": 0.0001238910701968483,
      "loss": 0.1624,
      "step": 57500
    },
    {
      "epoch": 2.3065298655849835,
      "grad_norm": 1.7718604803085327,
      "learning_rate": 0.00012322783466864754,
      "loss": 0.1569,
      "step": 58000
    },
    {
      "epoch": 2.3264137437365786,
      "grad_norm": 4.025502681732178,
      "learning_rate": 0.00012256592561150315,
      "loss": 0.156,
      "step": 58500
    },
    {
      "epoch": 2.3462976218881733,
      "grad_norm": 2.876533269882202,
      "learning_rate": 0.00012190401655435879,
      "loss": 0.1616,
      "step": 59000
    },
    {
      "epoch": 2.366181500039768,
      "grad_norm": 1.680629849433899,
      "learning_rate": 0.00012124078102615801,
      "loss": 0.1543,
      "step": 59500
    },
    {
      "epoch": 2.3860653781913626,
      "grad_norm": 1.0101946592330933,
      "learning_rate": 0.00012057754549795724,
      "loss": 0.1465,
      "step": 60000
    },
    {
      "epoch": 2.400381770460511,
      "eval_loss": 0.16907282173633575,
      "eval_runtime": 31.0907,
      "eval_samples_per_second": 16.082,
      "eval_steps_per_second": 8.041,
      "step": 60360
    },
    {
      "epoch": 2.405949256342957,
      "grad_norm": 1.9501550197601318,
      "learning_rate": 0.00011991430996975645,
      "loss": 0.1535,
      "step": 60500
    },
    {
      "epoch": 2.425833134494552,
      "grad_norm": 3.678605556488037,
      "learning_rate": 0.0001192510744415557,
      "loss": 0.1453,
      "step": 61000
    },
    {
      "epoch": 2.4457170126461465,
      "grad_norm": 4.075408458709717,
      "learning_rate": 0.00011858783891335492,
      "loss": 0.1494,
      "step": 61500
    },
    {
      "epoch": 2.465600890797741,
      "grad_norm": 1.3122295141220093,
      "learning_rate": 0.00011792460338515414,
      "loss": 0.1593,
      "step": 62000
    },
    {
      "epoch": 2.485484768949336,
      "grad_norm": 2.0857038497924805,
      "learning_rate": 0.00011726136785695337,
      "loss": 0.1514,
      "step": 62500
    },
    {
      "epoch": 2.5053686471009304,
      "grad_norm": 1.6657792329788208,
      "learning_rate": 0.000116599458799809,
      "loss": 0.1501,
      "step": 63000
    },
    {
      "epoch": 2.525252525252525,
      "grad_norm": 1.8100333213806152,
      "learning_rate": 0.00011593622327160823,
      "loss": 0.1519,
      "step": 63500
    },
    {
      "epoch": 2.54513640340412,
      "grad_norm": 2.504148006439209,
      "learning_rate": 0.00011527298774340744,
      "loss": 0.1496,
      "step": 64000
    },
    {
      "epoch": 2.5650202815557144,
      "grad_norm": 1.1805058717727661,
      "learning_rate": 0.00011461107868626306,
      "loss": 0.144,
      "step": 64500
    },
    {
      "epoch": 2.5849041597073095,
      "grad_norm": 3.109494686126709,
      "learning_rate": 0.0001139478431580623,
      "loss": 0.1517,
      "step": 65000
    },
    {
      "epoch": 2.600413584665553,
      "eval_loss": 0.17017363011837006,
      "eval_runtime": 31.1173,
      "eval_samples_per_second": 16.068,
      "eval_steps_per_second": 8.034,
      "step": 65390
    },
    {
      "epoch": 2.604788037858904,
      "grad_norm": 2.1431307792663574,
      "learning_rate": 0.00011328460762986153,
      "loss": 0.1561,
      "step": 65500
    },
    {
      "epoch": 2.6246719160104988,
      "grad_norm": 1.5804929733276367,
      "learning_rate": 0.00011262137210166075,
      "loss": 0.1536,
      "step": 66000
    },
    {
      "epoch": 2.6445557941620934,
      "grad_norm": 3.0483312606811523,
      "learning_rate": 0.00011195813657345997,
      "loss": 0.1497,
      "step": 66500
    },
    {
      "epoch": 2.664439672313688,
      "grad_norm": 2.7714364528656006,
      "learning_rate": 0.00011129490104525919,
      "loss": 0.1493,
      "step": 67000
    },
    {
      "epoch": 2.6843235504652827,
      "grad_norm": 1.1538747549057007,
      "learning_rate": 0.00011063299198811483,
      "loss": 0.1586,
      "step": 67500
    },
    {
      "epoch": 2.7042074286168774,
      "grad_norm": 1.946218729019165,
      "learning_rate": 0.00010996975645991405,
      "loss": 0.1484,
      "step": 68000
    },
    {
      "epoch": 2.724091306768472,
      "grad_norm": 1.453487753868103,
      "learning_rate": 0.00010930652093171328,
      "loss": 0.1545,
      "step": 68500
    },
    {
      "epoch": 2.7439751849200666,
      "grad_norm": 1.9838638305664062,
      "learning_rate": 0.00010864328540351249,
      "loss": 0.1443,
      "step": 69000
    },
    {
      "epoch": 2.7638590630716617,
      "grad_norm": 0.83545982837677,
      "learning_rate": 0.00010798004987531172,
      "loss": 0.148,
      "step": 69500
    },
    {
      "epoch": 2.783742941223256,
      "grad_norm": 2.1687493324279785,
      "learning_rate": 0.00010731681434711096,
      "loss": 0.1533,
      "step": 70000
    },
    {
      "epoch": 2.800445398870596,
      "eval_loss": 0.16508887708187103,
      "eval_runtime": 31.1958,
      "eval_samples_per_second": 16.028,
      "eval_steps_per_second": 8.014,
      "step": 70420
    },
    {
      "epoch": 2.803626819374851,
      "grad_norm": 1.3648791313171387,
      "learning_rate": 0.00010665357881891018,
      "loss": 0.1419,
      "step": 70500
    },
    {
      "epoch": 2.8235106975264457,
      "grad_norm": 1.7863802909851074,
      "learning_rate": 0.00010599034329070941,
      "loss": 0.1482,
      "step": 71000
    },
    {
      "epoch": 2.8433945756780403,
      "grad_norm": 3.2346866130828857,
      "learning_rate": 0.00010532710776250862,
      "loss": 0.1425,
      "step": 71500
    },
    {
      "epoch": 2.863278453829635,
      "grad_norm": 0.9211582541465759,
      "learning_rate": 0.00010466519870536427,
      "loss": 0.1501,
      "step": 72000
    },
    {
      "epoch": 2.8831623319812296,
      "grad_norm": 0.1923113614320755,
      "learning_rate": 0.00010400196317716348,
      "loss": 0.1469,
      "step": 72500
    },
    {
      "epoch": 2.9030462101328243,
      "grad_norm": 4.506950855255127,
      "learning_rate": 0.00010333872764896271,
      "loss": 0.1465,
      "step": 73000
    },
    {
      "epoch": 2.922930088284419,
      "grad_norm": 3.244807004928589,
      "learning_rate": 0.00010267549212076194,
      "loss": 0.1425,
      "step": 73500
    },
    {
      "epoch": 2.9428139664360136,
      "grad_norm": 1.8501616716384888,
      "learning_rate": 0.00010201225659256115,
      "loss": 0.1446,
      "step": 74000
    },
    {
      "epoch": 2.962697844587608,
      "grad_norm": 3.9013025760650635,
      "learning_rate": 0.0001013503475354168,
      "loss": 0.1508,
      "step": 74500
    },
    {
      "epoch": 2.9825817227392033,
      "grad_norm": 3.1799283027648926,
      "learning_rate": 0.00010068711200721601,
      "loss": 0.1412,
      "step": 75000
    },
    {
      "epoch": 3.000477213075638,
      "eval_loss": 0.1569422334432602,
      "eval_runtime": 33.8027,
      "eval_samples_per_second": 14.792,
      "eval_steps_per_second": 7.396,
      "step": 75450
    },
    {
      "epoch": 3.002465600890798,
      "grad_norm": 1.1885572671890259,
      "learning_rate": 0.00010002520295007163,
      "loss": 0.1334,
      "step": 75500
    },
    {
      "epoch": 3.0223494790423926,
      "grad_norm": 0.8424841165542603,
      "learning_rate": 9.936196742187086e-05,
      "loss": 0.1261,
      "step": 76000
    },
    {
      "epoch": 3.0422333571939872,
      "grad_norm": 1.2642099857330322,
      "learning_rate": 9.869873189367008e-05,
      "loss": 0.1233,
      "step": 76500
    },
    {
      "epoch": 3.062117235345582,
      "grad_norm": 3.7216506004333496,
      "learning_rate": 9.803549636546931e-05,
      "loss": 0.1257,
      "step": 77000
    },
    {
      "epoch": 3.0820011134971765,
      "grad_norm": 1.576157808303833,
      "learning_rate": 9.737358730832493e-05,
      "loss": 0.1257,
      "step": 77500
    },
    {
      "epoch": 3.101884991648771,
      "grad_norm": 3.2462401390075684,
      "learning_rate": 9.671035178012417e-05,
      "loss": 0.1387,
      "step": 78000
    },
    {
      "epoch": 3.121768869800366,
      "grad_norm": 0.38061007857322693,
      "learning_rate": 9.604711625192338e-05,
      "loss": 0.1288,
      "step": 78500
    },
    {
      "epoch": 3.1416527479519605,
      "grad_norm": 1.4168367385864258,
      "learning_rate": 9.538388072372262e-05,
      "loss": 0.125,
      "step": 79000
    },
    {
      "epoch": 3.161536626103555,
      "grad_norm": 1.8440510034561157,
      "learning_rate": 9.472064519552183e-05,
      "loss": 0.1203,
      "step": 79500
    },
    {
      "epoch": 3.1814205042551498,
      "grad_norm": 0.012771493755280972,
      "learning_rate": 9.405873613837747e-05,
      "loss": 0.1237,
      "step": 80000
    },
    {
      "epoch": 3.2005090272806807,
      "eval_loss": 0.15806268155574799,
      "eval_runtime": 31.0713,
      "eval_samples_per_second": 16.092,
      "eval_steps_per_second": 8.046,
      "step": 80480
    },
    {
      "epoch": 3.2013043824067444,
      "grad_norm": 2.36724853515625,
      "learning_rate": 9.33955006101767e-05,
      "loss": 0.1253,
      "step": 80500
    },
    {
      "epoch": 3.221188260558339,
      "grad_norm": 1.2032071352005005,
      "learning_rate": 9.273226508197591e-05,
      "loss": 0.1321,
      "step": 81000
    },
    {
      "epoch": 3.241072138709934,
      "grad_norm": 1.6413310766220093,
      "learning_rate": 9.206902955377515e-05,
      "loss": 0.1256,
      "step": 81500
    },
    {
      "epoch": 3.260956016861529,
      "grad_norm": 1.263128638267517,
      "learning_rate": 9.140579402557436e-05,
      "loss": 0.1281,
      "step": 82000
    },
    {
      "epoch": 3.2808398950131235,
      "grad_norm": 2.12233304977417,
      "learning_rate": 9.074255849737359e-05,
      "loss": 0.1299,
      "step": 82500
    },
    {
      "epoch": 3.300723773164718,
      "grad_norm": 1.265188217163086,
      "learning_rate": 9.007932296917282e-05,
      "loss": 0.1226,
      "step": 83000
    },
    {
      "epoch": 3.3206076513163127,
      "grad_norm": 1.391507625579834,
      "learning_rate": 8.941608744097204e-05,
      "loss": 0.1269,
      "step": 83500
    },
    {
      "epoch": 3.3404915294679074,
      "grad_norm": 1.941706657409668,
      "learning_rate": 8.875285191277126e-05,
      "loss": 0.1306,
      "step": 84000
    },
    {
      "epoch": 3.360375407619502,
      "grad_norm": 1.1882116794586182,
      "learning_rate": 8.809094285562688e-05,
      "loss": 0.1287,
      "step": 84500
    },
    {
      "epoch": 3.3802592857710967,
      "grad_norm": 2.2689340114593506,
      "learning_rate": 8.742903379848252e-05,
      "loss": 0.1281,
      "step": 85000
    },
    {
      "epoch": 3.4001431639226913,
      "grad_norm": 2.9482505321502686,
      "learning_rate": 8.676579827028175e-05,
      "loss": 0.131,
      "step": 85500
    },
    {
      "epoch": 3.400540841485723,
      "eval_loss": 0.15611666440963745,
      "eval_runtime": 31.2043,
      "eval_samples_per_second": 16.023,
      "eval_steps_per_second": 8.012,
      "step": 85510
    },
    {
      "epoch": 3.420027042074286,
      "grad_norm": 1.8892121315002441,
      "learning_rate": 8.610256274208097e-05,
      "loss": 0.1261,
      "step": 86000
    },
    {
      "epoch": 3.4399109202258806,
      "grad_norm": 2.923980236053467,
      "learning_rate": 8.54393272138802e-05,
      "loss": 0.1304,
      "step": 86500
    },
    {
      "epoch": 3.4597947983774757,
      "grad_norm": 0.46090424060821533,
      "learning_rate": 8.477609168567942e-05,
      "loss": 0.1263,
      "step": 87000
    },
    {
      "epoch": 3.4796786765290704,
      "grad_norm": 1.307573676109314,
      "learning_rate": 8.411285615747865e-05,
      "loss": 0.1217,
      "step": 87500
    },
    {
      "epoch": 3.499562554680665,
      "grad_norm": 1.9948647022247314,
      "learning_rate": 8.345094710033427e-05,
      "loss": 0.1298,
      "step": 88000
    },
    {
      "epoch": 3.5194464328322597,
      "grad_norm": 1.2264705896377563,
      "learning_rate": 8.27877115721335e-05,
      "loss": 0.1254,
      "step": 88500
    },
    {
      "epoch": 3.5393303109838543,
      "grad_norm": 1.249614953994751,
      "learning_rate": 8.212447604393274e-05,
      "loss": 0.1217,
      "step": 89000
    },
    {
      "epoch": 3.559214189135449,
      "grad_norm": 1.5197207927703857,
      "learning_rate": 8.146124051573195e-05,
      "loss": 0.1276,
      "step": 89500
    },
    {
      "epoch": 3.5790980672870436,
      "grad_norm": 0.8200188875198364,
      "learning_rate": 8.079800498753118e-05,
      "loss": 0.1283,
      "step": 90000
    },
    {
      "epoch": 3.5989819454386383,
      "grad_norm": 1.3526813983917236,
      "learning_rate": 8.01347694593304e-05,
      "loss": 0.1221,
      "step": 90500
    },
    {
      "epoch": 3.600572655690766,
      "eval_loss": 0.14859890937805176,
      "eval_runtime": 32.6446,
      "eval_samples_per_second": 15.316,
      "eval_steps_per_second": 7.658,
      "step": 90540
    },
    {
      "epoch": 3.618865823590233,
      "grad_norm": 1.9000824689865112,
      "learning_rate": 7.947153393112963e-05,
      "loss": 0.1215,
      "step": 91000
    },
    {
      "epoch": 3.638749701741828,
      "grad_norm": 1.2409000396728516,
      "learning_rate": 7.880829840292885e-05,
      "loss": 0.1259,
      "step": 91500
    },
    {
      "epoch": 3.658633579893422,
      "grad_norm": 1.7296925783157349,
      "learning_rate": 7.814506287472808e-05,
      "loss": 0.1245,
      "step": 92000
    },
    {
      "epoch": 3.6785174580450173,
      "grad_norm": 1.3685667514801025,
      "learning_rate": 7.748315381758371e-05,
      "loss": 0.127,
      "step": 92500
    },
    {
      "epoch": 3.698401336196612,
      "grad_norm": 2.338923692703247,
      "learning_rate": 7.681991828938293e-05,
      "loss": 0.119,
      "step": 93000
    },
    {
      "epoch": 3.7182852143482066,
      "grad_norm": 2.3641035556793213,
      "learning_rate": 7.615668276118215e-05,
      "loss": 0.1256,
      "step": 93500
    },
    {
      "epoch": 3.7381690924998012,
      "grad_norm": 1.444541335105896,
      "learning_rate": 7.549477370403777e-05,
      "loss": 0.1218,
      "step": 94000
    },
    {
      "epoch": 3.758052970651396,
      "grad_norm": 2.2175920009613037,
      "learning_rate": 7.4831538175837e-05,
      "loss": 0.1181,
      "step": 94500
    },
    {
      "epoch": 3.7779368488029905,
      "grad_norm": 2.533275604248047,
      "learning_rate": 7.416830264763624e-05,
      "loss": 0.1203,
      "step": 95000
    },
    {
      "epoch": 3.797820726954585,
      "grad_norm": 2.5117480754852295,
      "learning_rate": 7.350506711943545e-05,
      "loss": 0.1161,
      "step": 95500
    },
    {
      "epoch": 3.8006044698958084,
      "eval_loss": 0.14638979732990265,
      "eval_runtime": 33.193,
      "eval_samples_per_second": 15.063,
      "eval_steps_per_second": 7.532,
      "step": 95570
    },
    {
      "epoch": 3.81770460510618,
      "grad_norm": 0.9872569441795349,
      "learning_rate": 7.284183159123469e-05,
      "loss": 0.1257,
      "step": 96000
    },
    {
      "epoch": 3.8375884832577745,
      "grad_norm": 2.7399699687957764,
      "learning_rate": 7.21785960630339e-05,
      "loss": 0.1309,
      "step": 96500
    },
    {
      "epoch": 3.8574723614093696,
      "grad_norm": 1.9438740015029907,
      "learning_rate": 7.151536053483313e-05,
      "loss": 0.1196,
      "step": 97000
    },
    {
      "epoch": 3.8773562395609638,
      "grad_norm": 3.683006763458252,
      "learning_rate": 7.085212500663237e-05,
      "loss": 0.1207,
      "step": 97500
    },
    {
      "epoch": 3.897240117712559,
      "grad_norm": 2.3406124114990234,
      "learning_rate": 7.019154242054439e-05,
      "loss": 0.1202,
      "step": 98000
    },
    {
      "epoch": 3.9171239958641535,
      "grad_norm": 1.3052023649215698,
      "learning_rate": 6.952830689234361e-05,
      "loss": 0.1233,
      "step": 98500
    },
    {
      "epoch": 3.937007874015748,
      "grad_norm": 1.8666225671768188,
      "learning_rate": 6.886507136414284e-05,
      "loss": 0.1217,
      "step": 99000
    },
    {
      "epoch": 3.956891752167343,
      "grad_norm": 1.7979810237884521,
      "learning_rate": 6.820183583594206e-05,
      "loss": 0.1194,
      "step": 99500
    },
    {
      "epoch": 3.9767756303189374,
      "grad_norm": 0.9724803566932678,
      "learning_rate": 6.753860030774129e-05,
      "loss": 0.1276,
      "step": 100000
    },
    {
      "epoch": 3.996659508470532,
      "grad_norm": 3.251120090484619,
      "learning_rate": 6.687536477954052e-05,
      "loss": 0.1165,
      "step": 100500
    },
    {
      "epoch": 4.000636284100851,
      "eval_loss": 0.1412682831287384,
      "eval_runtime": 31.0761,
      "eval_samples_per_second": 16.09,
      "eval_steps_per_second": 8.045,
      "step": 100600
    },
    {
      "epoch": 4.016543386622127,
      "grad_norm": 3.6826343536376953,
      "learning_rate": 6.621212925133974e-05,
      "loss": 0.1046,
      "step": 101000
    },
    {
      "epoch": 4.036427264773722,
      "grad_norm": 5.360718250274658,
      "learning_rate": 6.554889372313896e-05,
      "loss": 0.1032,
      "step": 101500
    },
    {
      "epoch": 4.056311142925316,
      "grad_norm": 1.780490756034851,
      "learning_rate": 6.488698466599459e-05,
      "loss": 0.1014,
      "step": 102000
    },
    {
      "epoch": 4.076195021076911,
      "grad_norm": 2.984694480895996,
      "learning_rate": 6.422374913779382e-05,
      "loss": 0.1044,
      "step": 102500
    },
    {
      "epoch": 4.096078899228505,
      "grad_norm": 0.6237806677818298,
      "learning_rate": 6.356184008064945e-05,
      "loss": 0.1009,
      "step": 103000
    },
    {
      "epoch": 4.1159627773801,
      "grad_norm": 3.3006651401519775,
      "learning_rate": 6.289860455244868e-05,
      "loss": 0.1017,
      "step": 103500
    },
    {
      "epoch": 4.135846655531695,
      "grad_norm": 1.9449480772018433,
      "learning_rate": 6.223536902424789e-05,
      "loss": 0.1043,
      "step": 104000
    },
    {
      "epoch": 4.15573053368329,
      "grad_norm": 2.495291233062744,
      "learning_rate": 6.157213349604713e-05,
      "loss": 0.1024,
      "step": 104500
    },
    {
      "epoch": 4.175614411834884,
      "grad_norm": 0.23489686846733093,
      "learning_rate": 6.090889796784635e-05,
      "loss": 0.1031,
      "step": 105000
    },
    {
      "epoch": 4.195498289986479,
      "grad_norm": 3.0352611541748047,
      "learning_rate": 6.024566243964557e-05,
      "loss": 0.0994,
      "step": 105500
    },
    {
      "epoch": 4.2006680983058935,
      "eval_loss": 0.14581650495529175,
      "eval_runtime": 35.7264,
      "eval_samples_per_second": 13.995,
      "eval_steps_per_second": 6.998,
      "step": 105630
    },
    {
      "epoch": 4.215382168138074,
      "grad_norm": 1.1670618057250977,
      "learning_rate": 5.95824269114448e-05,
      "loss": 0.1045,
      "step": 106000
    },
    {
      "epoch": 4.235266046289668,
      "grad_norm": 2.0033414363861084,
      "learning_rate": 5.891919138324402e-05,
      "loss": 0.1016,
      "step": 106500
    },
    {
      "epoch": 4.255149924441263,
      "grad_norm": 1.4979524612426758,
      "learning_rate": 5.825728232609965e-05,
      "loss": 0.1079,
      "step": 107000
    },
    {
      "epoch": 4.275033802592858,
      "grad_norm": 1.7079240083694458,
      "learning_rate": 5.7594046797898874e-05,
      "loss": 0.1016,
      "step": 107500
    },
    {
      "epoch": 4.294917680744453,
      "grad_norm": 1.1605221033096313,
      "learning_rate": 5.6930811269698107e-05,
      "loss": 0.1008,
      "step": 108000
    },
    {
      "epoch": 4.314801558896047,
      "grad_norm": 1.4162037372589111,
      "learning_rate": 5.6267575741497326e-05,
      "loss": 0.1083,
      "step": 108500
    },
    {
      "epoch": 4.334685437047642,
      "grad_norm": 2.373020887374878,
      "learning_rate": 5.5604340213296545e-05,
      "loss": 0.1073,
      "step": 109000
    },
    {
      "epoch": 4.354569315199236,
      "grad_norm": 2.836911916732788,
      "learning_rate": 5.494375762720858e-05,
      "loss": 0.1019,
      "step": 109500
    },
    {
      "epoch": 4.374453193350831,
      "grad_norm": 3.7747974395751953,
      "learning_rate": 5.42805220990078e-05,
      "loss": 0.1038,
      "step": 110000
    },
    {
      "epoch": 4.3943370715024255,
      "grad_norm": 1.4646281003952026,
      "learning_rate": 5.361728657080703e-05,
      "loss": 0.1113,
      "step": 110500
    },
    {
      "epoch": 4.4006999125109365,
      "eval_loss": 0.14834672212600708,
      "eval_runtime": 37.228,
      "eval_samples_per_second": 13.431,
      "eval_steps_per_second": 6.715,
      "step": 110660
    },
    {
      "epoch": 4.414220949654021,
      "grad_norm": 2.5483474731445312,
      "learning_rate": 5.295405104260626e-05,
      "loss": 0.1006,
      "step": 111000
    },
    {
      "epoch": 4.434104827805616,
      "grad_norm": 2.5747101306915283,
      "learning_rate": 5.229081551440548e-05,
      "loss": 0.1082,
      "step": 111500
    },
    {
      "epoch": 4.45398870595721,
      "grad_norm": 0.23354358971118927,
      "learning_rate": 5.16275799862047e-05,
      "loss": 0.1032,
      "step": 112000
    },
    {
      "epoch": 4.473872584108805,
      "grad_norm": 2.4420604705810547,
      "learning_rate": 5.096434445800393e-05,
      "loss": 0.106,
      "step": 112500
    },
    {
      "epoch": 4.493756462260399,
      "grad_norm": 1.2000492811203003,
      "learning_rate": 5.030110892980315e-05,
      "loss": 0.0961,
      "step": 113000
    },
    {
      "epoch": 4.513640340411994,
      "grad_norm": 1.8864790201187134,
      "learning_rate": 4.963919987265878e-05,
      "loss": 0.0978,
      "step": 113500
    },
    {
      "epoch": 4.5335242185635884,
      "grad_norm": 0.45711418986320496,
      "learning_rate": 4.897596434445801e-05,
      "loss": 0.094,
      "step": 114000
    },
    {
      "epoch": 4.5534080967151835,
      "grad_norm": 7.053537845611572,
      "learning_rate": 4.831272881625723e-05,
      "loss": 0.1,
      "step": 114500
    },
    {
      "epoch": 4.573291974866778,
      "grad_norm": 3.276334047317505,
      "learning_rate": 4.765081975911286e-05,
      "loss": 0.0996,
      "step": 115000
    },
    {
      "epoch": 4.593175853018373,
      "grad_norm": 1.4245741367340088,
      "learning_rate": 4.6987584230912085e-05,
      "loss": 0.0996,
      "step": 115500
    },
    {
      "epoch": 4.600731726715979,
      "eval_loss": 0.14262481033802032,
      "eval_runtime": 32.3953,
      "eval_samples_per_second": 15.434,
      "eval_steps_per_second": 7.717,
      "step": 115690
    },
    {
      "epoch": 4.613059731169967,
      "grad_norm": 2.3254101276397705,
      "learning_rate": 4.6324348702711304e-05,
      "loss": 0.1017,
      "step": 116000
    },
    {
      "epoch": 4.632943609321562,
      "grad_norm": 1.167358636856079,
      "learning_rate": 4.566111317451054e-05,
      "loss": 0.0981,
      "step": 116500
    },
    {
      "epoch": 4.652827487473157,
      "grad_norm": 4.5853376388549805,
      "learning_rate": 4.499787764630976e-05,
      "loss": 0.1016,
      "step": 117000
    },
    {
      "epoch": 4.672711365624751,
      "grad_norm": 1.4917422533035278,
      "learning_rate": 4.433464211810899e-05,
      "loss": 0.1018,
      "step": 117500
    },
    {
      "epoch": 4.6925952437763465,
      "grad_norm": 0.012467560358345509,
      "learning_rate": 4.367273306096461e-05,
      "loss": 0.0975,
      "step": 118000
    },
    {
      "epoch": 4.712479121927941,
      "grad_norm": 0.8470689058303833,
      "learning_rate": 4.3009497532763836e-05,
      "loss": 0.0965,
      "step": 118500
    },
    {
      "epoch": 4.732363000079536,
      "grad_norm": 0.43946540355682373,
      "learning_rate": 4.234626200456306e-05,
      "loss": 0.1014,
      "step": 119000
    },
    {
      "epoch": 4.75224687823113,
      "grad_norm": 2.230210304260254,
      "learning_rate": 4.168302647636229e-05,
      "loss": 0.0996,
      "step": 119500
    },
    {
      "epoch": 4.772130756382725,
      "grad_norm": 3.4537596702575684,
      "learning_rate": 4.1019790948161514e-05,
      "loss": 0.0966,
      "step": 120000
    },
    {
      "epoch": 4.792014634534319,
      "grad_norm": 1.7490534782409668,
      "learning_rate": 4.035655541996074e-05,
      "loss": 0.1116,
      "step": 120500
    },
    {
      "epoch": 4.800763540921022,
      "eval_loss": 0.1392485797405243,
      "eval_runtime": 34.3723,
      "eval_samples_per_second": 14.547,
      "eval_steps_per_second": 7.273,
      "step": 120720
    },
    {
      "epoch": 4.811898512685914,
      "grad_norm": 0.9216163158416748,
      "learning_rate": 3.969331989175996e-05,
      "loss": 0.0965,
      "step": 121000
    },
    {
      "epoch": 4.831782390837509,
      "grad_norm": 4.56651496887207,
      "learning_rate": 3.9030084363559186e-05,
      "loss": 0.0965,
      "step": 121500
    },
    {
      "epoch": 4.851666268989104,
      "grad_norm": 1.2061914205551147,
      "learning_rate": 3.8368175306414814e-05,
      "loss": 0.1059,
      "step": 122000
    },
    {
      "epoch": 4.871550147140699,
      "grad_norm": 1.8895795345306396,
      "learning_rate": 3.770493977821405e-05,
      "loss": 0.0988,
      "step": 122500
    },
    {
      "epoch": 4.891434025292293,
      "grad_norm": 2.730050563812256,
      "learning_rate": 3.7041704250013266e-05,
      "loss": 0.1022,
      "step": 123000
    },
    {
      "epoch": 4.911317903443888,
      "grad_norm": 2.6674962043762207,
      "learning_rate": 3.637846872181249e-05,
      "loss": 0.0966,
      "step": 123500
    },
    {
      "epoch": 4.931201781595482,
      "grad_norm": 5.602296352386475,
      "learning_rate": 3.571655966466812e-05,
      "loss": 0.0944,
      "step": 124000
    },
    {
      "epoch": 4.951085659747077,
      "grad_norm": 2.389179229736328,
      "learning_rate": 3.505332413646734e-05,
      "loss": 0.0975,
      "step": 124500
    },
    {
      "epoch": 4.970969537898672,
      "grad_norm": 3.1410694122314453,
      "learning_rate": 3.4390088608266565e-05,
      "loss": 0.0923,
      "step": 125000
    },
    {
      "epoch": 4.990853416050267,
      "grad_norm": 0.3184365928173065,
      "learning_rate": 3.37268530800658e-05,
      "loss": 0.1004,
      "step": 125500
    },
    {
      "epoch": 5.000795355126064,
      "eval_loss": 0.13804183900356293,
      "eval_runtime": 34.1527,
      "eval_samples_per_second": 14.64,
      "eval_steps_per_second": 7.32,
      "step": 125750
    },
    {
      "epoch": 5.010737294201861,
      "grad_norm": 1.9156638383865356,
      "learning_rate": 3.3063617551865024e-05,
      "loss": 0.0899,
      "step": 126000
    },
    {
      "epoch": 5.030621172353456,
      "grad_norm": 1.1794829368591309,
      "learning_rate": 3.2401708494720645e-05,
      "loss": 0.0762,
      "step": 126500
    },
    {
      "epoch": 5.05050505050505,
      "grad_norm": 2.2873692512512207,
      "learning_rate": 3.173847296651987e-05,
      "loss": 0.0823,
      "step": 127000
    },
    {
      "epoch": 5.070388928656645,
      "grad_norm": 2.4998435974121094,
      "learning_rate": 3.10752374383191e-05,
      "loss": 0.0747,
      "step": 127500
    },
    {
      "epoch": 5.0902728068082395,
      "grad_norm": 3.9110782146453857,
      "learning_rate": 3.041200191011832e-05,
      "loss": 0.0853,
      "step": 128000
    },
    {
      "epoch": 5.1101566849598345,
      "grad_norm": 3.3292062282562256,
      "learning_rate": 2.9748766381917546e-05,
      "loss": 0.0839,
      "step": 128500
    },
    {
      "epoch": 5.13004056311143,
      "grad_norm": 4.433184623718262,
      "learning_rate": 2.9085530853716776e-05,
      "loss": 0.0802,
      "step": 129000
    },
    {
      "epoch": 5.149924441263024,
      "grad_norm": 3.4676051139831543,
      "learning_rate": 2.8422295325515995e-05,
      "loss": 0.0838,
      "step": 129500
    },
    {
      "epoch": 5.169808319414619,
      "grad_norm": 1.5169734954833984,
      "learning_rate": 2.7759059797315225e-05,
      "loss": 0.0783,
      "step": 130000
    },
    {
      "epoch": 5.189692197566213,
      "grad_norm": 1.637904405593872,
      "learning_rate": 2.709582426911445e-05,
      "loss": 0.0814,
      "step": 130500
    },
    {
      "epoch": 5.200827169331106,
      "eval_loss": 0.141804039478302,
      "eval_runtime": 32.6973,
      "eval_samples_per_second": 15.292,
      "eval_steps_per_second": 7.646,
      "step": 130780
    },
    {
      "epoch": 5.209576075717808,
      "grad_norm": 2.086423873901367,
      "learning_rate": 2.643391521197008e-05,
      "loss": 0.0831,
      "step": 131000
    },
    {
      "epoch": 5.229459953869402,
      "grad_norm": 0.6672555804252625,
      "learning_rate": 2.57720061548257e-05,
      "loss": 0.0862,
      "step": 131500
    },
    {
      "epoch": 5.2493438320209975,
      "grad_norm": 0.04646410793066025,
      "learning_rate": 2.510877062662493e-05,
      "loss": 0.0792,
      "step": 132000
    },
    {
      "epoch": 5.269227710172592,
      "grad_norm": 4.182920932769775,
      "learning_rate": 2.4445535098424155e-05,
      "loss": 0.0802,
      "step": 132500
    },
    {
      "epoch": 5.289111588324187,
      "grad_norm": 3.0408833026885986,
      "learning_rate": 2.3782299570223378e-05,
      "loss": 0.0836,
      "step": 133000
    },
    {
      "epoch": 5.308995466475781,
      "grad_norm": 1.9850901365280151,
      "learning_rate": 2.3120390513079006e-05,
      "loss": 0.0779,
      "step": 133500
    },
    {
      "epoch": 5.328879344627376,
      "grad_norm": 2.1532881259918213,
      "learning_rate": 2.2457154984878232e-05,
      "loss": 0.0813,
      "step": 134000
    },
    {
      "epoch": 5.348763222778971,
      "grad_norm": 2.2527785301208496,
      "learning_rate": 2.1793919456677455e-05,
      "loss": 0.0761,
      "step": 134500
    },
    {
      "epoch": 5.368647100930565,
      "grad_norm": 1.829942226409912,
      "learning_rate": 2.1130683928476684e-05,
      "loss": 0.0813,
      "step": 135000
    },
    {
      "epoch": 5.3885309790821605,
      "grad_norm": 2.5190794467926025,
      "learning_rate": 2.046877487133231e-05,
      "loss": 0.086,
      "step": 135500
    },
    {
      "epoch": 5.400858983536149,
      "eval_loss": 0.1405518800020218,
      "eval_runtime": 32.4801,
      "eval_samples_per_second": 15.394,
      "eval_steps_per_second": 7.697,
      "step": 135810
    },
    {
      "epoch": 5.408414857233755,
      "grad_norm": 0.9381058812141418,
      "learning_rate": 1.9805539343131535e-05,
      "loss": 0.0769,
      "step": 136000
    },
    {
      "epoch": 5.42829873538535,
      "grad_norm": 3.986593246459961,
      "learning_rate": 1.914230381493076e-05,
      "loss": 0.08,
      "step": 136500
    },
    {
      "epoch": 5.448182613536944,
      "grad_norm": 1.4495145082473755,
      "learning_rate": 1.848039475778639e-05,
      "loss": 0.0787,
      "step": 137000
    },
    {
      "epoch": 5.468066491688539,
      "grad_norm": 2.496209144592285,
      "learning_rate": 1.781715922958561e-05,
      "loss": 0.0808,
      "step": 137500
    },
    {
      "epoch": 5.487950369840133,
      "grad_norm": 1.4230667352676392,
      "learning_rate": 1.7153923701384837e-05,
      "loss": 0.0818,
      "step": 138000
    },
    {
      "epoch": 5.507834247991728,
      "grad_norm": 0.07668805867433548,
      "learning_rate": 1.6490688173184063e-05,
      "loss": 0.0786,
      "step": 138500
    },
    {
      "epoch": 5.5277181261433235,
      "grad_norm": 2.7259907722473145,
      "learning_rate": 1.5827452644983286e-05,
      "loss": 0.0839,
      "step": 139000
    },
    {
      "epoch": 5.547602004294918,
      "grad_norm": 1.5834600925445557,
      "learning_rate": 1.5164217116782512e-05,
      "loss": 0.0807,
      "step": 139500
    },
    {
      "epoch": 5.567485882446512,
      "grad_norm": 1.825166940689087,
      "learning_rate": 1.4500981588581738e-05,
      "loss": 0.0786,
      "step": 140000
    },
    {
      "epoch": 5.587369760598107,
      "grad_norm": 0.33820098638534546,
      "learning_rate": 1.3837746060380963e-05,
      "loss": 0.0731,
      "step": 140500
    },
    {
      "epoch": 5.600890797741192,
      "eval_loss": 0.1415776163339615,
      "eval_runtime": 32.1699,
      "eval_samples_per_second": 15.542,
      "eval_steps_per_second": 7.771,
      "step": 140840
    },
    {
      "epoch": 5.607253638749702,
      "grad_norm": 0.2709919214248657,
      "learning_rate": 1.317583700323659e-05,
      "loss": 0.0755,
      "step": 141000
    },
    {
      "epoch": 5.627137516901296,
      "grad_norm": 2.084345817565918,
      "learning_rate": 1.2512601475035815e-05,
      "loss": 0.0784,
      "step": 141500
    },
    {
      "epoch": 5.647021395052891,
      "grad_norm": 1.5787373781204224,
      "learning_rate": 1.1849365946835041e-05,
      "loss": 0.0724,
      "step": 142000
    },
    {
      "epoch": 5.666905273204486,
      "grad_norm": 2.0165510177612305,
      "learning_rate": 1.1186130418634267e-05,
      "loss": 0.0814,
      "step": 142500
    },
    {
      "epoch": 5.686789151356081,
      "grad_norm": 1.0443742275238037,
      "learning_rate": 1.052289489043349e-05,
      "loss": 0.0749,
      "step": 143000
    },
    {
      "epoch": 5.706673029507675,
      "grad_norm": 0.3988170921802521,
      "learning_rate": 9.859659362232716e-06,
      "loss": 0.0805,
      "step": 143500
    },
    {
      "epoch": 5.72655690765927,
      "grad_norm": 1.6842561960220337,
      "learning_rate": 9.196423834031942e-06,
      "loss": 0.0859,
      "step": 144000
    },
    {
      "epoch": 5.746440785810865,
      "grad_norm": 3.893829584121704,
      "learning_rate": 8.533188305831168e-06,
      "loss": 0.0834,
      "step": 144500
    },
    {
      "epoch": 5.766324663962459,
      "grad_norm": 3.017742156982422,
      "learning_rate": 7.871279248686794e-06,
      "loss": 0.0767,
      "step": 145000
    },
    {
      "epoch": 5.786208542114053,
      "grad_norm": 1.0755605697631836,
      "learning_rate": 7.2080437204860195e-06,
      "loss": 0.0807,
      "step": 145500
    },
    {
      "epoch": 5.800922611946234,
      "eval_loss": 0.14066436886787415,
      "eval_runtime": 37.1151,
      "eval_samples_per_second": 13.472,
      "eval_steps_per_second": 6.736,
      "step": 145870
    },
    {
      "epoch": 5.8060924202656485,
      "grad_norm": 2.315115451812744,
      "learning_rate": 6.544808192285244e-06,
      "loss": 0.0737,
      "step": 146000
    },
    {
      "epoch": 5.825976298417244,
      "grad_norm": 1.7753525972366333,
      "learning_rate": 5.88157266408447e-06,
      "loss": 0.0781,
      "step": 146500
    },
    {
      "epoch": 5.845860176568838,
      "grad_norm": 0.942420244216919,
      "learning_rate": 5.218337135883695e-06,
      "loss": 0.0785,
      "step": 147000
    },
    {
      "epoch": 5.865744054720433,
      "grad_norm": 1.6908499002456665,
      "learning_rate": 4.5551016076829205e-06,
      "loss": 0.0773,
      "step": 147500
    },
    {
      "epoch": 5.885627932872027,
      "grad_norm": 3.541743755340576,
      "learning_rate": 3.891866079482146e-06,
      "loss": 0.082,
      "step": 148000
    },
    {
      "epoch": 5.905511811023622,
      "grad_norm": 1.41182541847229,
      "learning_rate": 3.228630551281371e-06,
      "loss": 0.0755,
      "step": 148500
    },
    {
      "epoch": 5.925395689175216,
      "grad_norm": 1.457543134689331,
      "learning_rate": 2.5653950230805963e-06,
      "loss": 0.0721,
      "step": 149000
    },
    {
      "epoch": 5.9452795673268115,
      "grad_norm": 1.3256011009216309,
      "learning_rate": 1.9034859659362233e-06,
      "loss": 0.0756,
      "step": 149500
    },
    {
      "epoch": 5.965163445478407,
      "grad_norm": 4.112858295440674,
      "learning_rate": 1.2402504377354487e-06,
      "loss": 0.0792,
      "step": 150000
    },
    {
      "epoch": 5.985047323630001,
      "grad_norm": 9.184978485107422,
      "learning_rate": 5.77014909534674e-07,
      "loss": 0.0763,
      "step": 150500
    },
    {
      "epoch": 6.0,
      "step": 150876,
      "total_flos": 9.584515856152166e+17,
      "train_loss": 0.1514410440879876,
      "train_runtime": 79770.8244,
      "train_samples_per_second": 3.783,
      "train_steps_per_second": 1.891
    }
  ],
  "logging_steps": 500,
  "max_steps": 150876,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.584515856152166e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}