{
  "best_global_step": 1500,
  "best_metric": 0.03645886108279228,
  "best_model_checkpoint": "/kaggle/working/models/checkpoints/v4/sql2nosql/checkpoint-1500",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.1209821447730064,
      "epoch": 0.020005001250312578,
      "grad_norm": 1.7302765846252441,
      "learning_rate": 7.2e-06,
      "loss": 1.3475940704345704,
      "mean_token_accuracy": 0.6775479473173618,
      "num_tokens": 65756.0,
      "step": 10
    },
    {
      "entropy": 1.0718491852283478,
      "epoch": 0.040010002500625155,
      "grad_norm": 1.598680853843689,
      "learning_rate": 1.52e-05,
      "loss": 1.1782958984375,
      "mean_token_accuracy": 0.6978455670177937,
      "num_tokens": 127350.0,
      "step": 20
    },
    {
      "entropy": 0.9551009342074395,
      "epoch": 0.06001500375093773,
      "grad_norm": 1.7650020122528076,
      "learning_rate": 2.32e-05,
      "loss": 0.9067621231079102,
      "mean_token_accuracy": 0.7337856531143189,
      "num_tokens": 191970.0,
      "step": 30
    },
    {
      "entropy": 0.7395371329039335,
      "epoch": 0.08002000500125031,
      "grad_norm": 1.2754874229431152,
      "learning_rate": 3.12e-05,
      "loss": 0.6462621212005615,
      "mean_token_accuracy": 0.815752174705267,
      "num_tokens": 257775.0,
      "step": 40
    },
    {
      "entropy": 0.5132245156913996,
      "epoch": 0.1000250062515629,
      "grad_norm": 1.6131094694137573,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 0.4761983871459961,
      "mean_token_accuracy": 0.8625759541988373,
      "num_tokens": 319507.0,
      "step": 50
    },
    {
      "entropy": 0.35333055444061756,
      "epoch": 0.12003000750187547,
      "grad_norm": 1.9517418146133423,
      "learning_rate": 4.72e-05,
      "loss": 0.34327900409698486,
      "mean_token_accuracy": 0.9010166764259339,
      "num_tokens": 380952.0,
      "step": 60
    },
    {
      "entropy": 0.2581108913756907,
      "epoch": 0.14003500875218805,
      "grad_norm": 1.8101425170898438,
      "learning_rate": 5.520000000000001e-05,
      "loss": 0.25612337589263917,
      "mean_token_accuracy": 0.9237508125603199,
      "num_tokens": 443849.0,
      "step": 70
    },
    {
      "entropy": 0.23392133321613073,
      "epoch": 0.16004001000250062,
      "grad_norm": 1.4003955125808716,
      "learning_rate": 6.32e-05,
      "loss": 0.20029537677764891,
      "mean_token_accuracy": 0.940458069741726,
      "num_tokens": 505868.0,
      "step": 80
    },
    {
      "entropy": 0.1773814239539206,
      "epoch": 0.1800450112528132,
      "grad_norm": 1.2149077653884888,
      "learning_rate": 7.12e-05,
      "loss": 0.1597532272338867,
      "mean_token_accuracy": 0.9488845877349377,
      "num_tokens": 570792.0,
      "step": 90
    },
    {
      "entropy": 0.16433264631778002,
      "epoch": 0.2000500125031258,
      "grad_norm": 1.4650107622146606,
      "learning_rate": 7.920000000000001e-05,
      "loss": 0.15449292659759523,
      "mean_token_accuracy": 0.9529499627649785,
      "num_tokens": 634476.0,
      "step": 100
    },
    {
      "entropy": 0.1567419718950987,
      "epoch": 0.22005501375343836,
      "grad_norm": 1.2412768602371216,
      "learning_rate": 8.72e-05,
      "loss": 0.14979339838027955,
      "mean_token_accuracy": 0.9547351159155368,
      "num_tokens": 697928.0,
      "step": 110
    },
    {
      "entropy": 0.14790078573860227,
      "epoch": 0.24006001500375093,
      "grad_norm": 1.0941258668899536,
      "learning_rate": 9.52e-05,
      "loss": 0.15537588596343993,
      "mean_token_accuracy": 0.9548502139747143,
      "num_tokens": 761970.0,
      "step": 120
    },
    {
      "entropy": 0.14068418610841035,
      "epoch": 0.26006501625406353,
      "grad_norm": 1.3730136156082153,
      "learning_rate": 0.0001032,
      "loss": 0.12376174926757813,
      "mean_token_accuracy": 0.9643007285892964,
      "num_tokens": 827125.0,
      "step": 130
    },
    {
      "entropy": 0.10366909941658378,
      "epoch": 0.2800700175043761,
      "grad_norm": 0.859259843826294,
      "learning_rate": 0.00011120000000000002,
      "loss": 0.10419689416885376,
      "mean_token_accuracy": 0.9708971530199051,
      "num_tokens": 890387.0,
      "step": 140
    },
    {
      "entropy": 0.10762261427007616,
      "epoch": 0.30007501875468867,
      "grad_norm": 0.9888283610343933,
      "learning_rate": 0.0001192,
      "loss": 0.09227448105812072,
      "mean_token_accuracy": 0.9713370814919472,
      "num_tokens": 956032.0,
      "step": 150
    },
    {
      "entropy": 0.1130900933407247,
      "epoch": 0.32008002000500124,
      "grad_norm": 0.9545535445213318,
      "learning_rate": 0.0001272,
      "loss": 0.1083062767982483,
      "mean_token_accuracy": 0.9707557298243046,
      "num_tokens": 1020072.0,
      "step": 160
    },
    {
      "entropy": 0.0895556318340823,
      "epoch": 0.3400850212553138,
      "grad_norm": 0.9866153597831726,
      "learning_rate": 0.0001352,
      "loss": 0.07606152892112732,
      "mean_token_accuracy": 0.973467419296503,
      "num_tokens": 1083781.0,
      "step": 170
    },
    {
      "entropy": 0.09170240506064146,
      "epoch": 0.3600900225056264,
      "grad_norm": 1.0180600881576538,
      "learning_rate": 0.0001432,
      "loss": 0.10022411346435547,
      "mean_token_accuracy": 0.9705780848860741,
      "num_tokens": 1145466.0,
      "step": 180
    },
    {
      "entropy": 0.11426927694119512,
      "epoch": 0.380095023755939,
      "grad_norm": 1.129612684249878,
      "learning_rate": 0.00015120000000000002,
      "loss": 0.10711104869842529,
      "mean_token_accuracy": 0.9698869682848453,
      "num_tokens": 1208095.0,
      "step": 190
    },
    {
      "entropy": 0.09287996906787158,
      "epoch": 0.4001000250062516,
      "grad_norm": 0.6177782416343689,
      "learning_rate": 0.00015920000000000002,
      "loss": 0.08630093932151794,
      "mean_token_accuracy": 0.977255067974329,
      "num_tokens": 1273403.0,
      "step": 200
    },
    {
      "entropy": 0.0744264661334455,
      "epoch": 0.42010502625656415,
      "grad_norm": 0.766137957572937,
      "learning_rate": 0.0001672,
      "loss": 0.07393635511398315,
      "mean_token_accuracy": 0.9792576834559441,
      "num_tokens": 1334848.0,
      "step": 210
    },
    {
      "entropy": 0.10690853425767273,
      "epoch": 0.4401100275068767,
      "grad_norm": 0.6585826277732849,
      "learning_rate": 0.0001752,
      "loss": 0.09994233846664428,
      "mean_token_accuracy": 0.972873219102621,
      "num_tokens": 1400976.0,
      "step": 220
    },
    {
      "entropy": 0.06925062141381204,
      "epoch": 0.4601150287571893,
      "grad_norm": 0.7318869233131409,
      "learning_rate": 0.0001832,
      "loss": 0.06631548404693603,
      "mean_token_accuracy": 0.9832681395113468,
      "num_tokens": 1462530.0,
      "step": 230
    },
    {
      "entropy": 0.08352572850417346,
      "epoch": 0.48012003000750186,
      "grad_norm": 0.46499672532081604,
      "learning_rate": 0.0001912,
      "loss": 0.09161096215248107,
      "mean_token_accuracy": 0.9757984310388566,
      "num_tokens": 1523663.0,
      "step": 240
    },
    {
      "entropy": 0.06921031260862946,
      "epoch": 0.5001250312578145,
      "grad_norm": 0.6253766417503357,
      "learning_rate": 0.00019920000000000002,
      "loss": 0.05958831310272217,
      "mean_token_accuracy": 0.982395163923502,
      "num_tokens": 1588657.0,
      "step": 250
    },
    {
      "entropy": 0.06543620774755254,
      "epoch": 0.5201300325081271,
      "grad_norm": 0.6349820494651794,
      "learning_rate": 0.00019999822839757118,
      "loss": 0.06900651454925537,
      "mean_token_accuracy": 0.9801992796361446,
      "num_tokens": 1653260.0,
      "step": 260
    },
    {
      "entropy": 0.06953847317490727,
      "epoch": 0.5401350337584396,
      "grad_norm": 0.6046717166900635,
      "learning_rate": 0.00019999210442038162,
      "loss": 0.06709518432617187,
      "mean_token_accuracy": 0.9797540627419948,
      "num_tokens": 1718687.0,
      "step": 270
    },
    {
      "entropy": 0.08993639135733247,
      "epoch": 0.5601400350087522,
      "grad_norm": 0.7929577231407166,
      "learning_rate": 0.00019998160646461522,
      "loss": 0.09603813290596008,
      "mean_token_accuracy": 0.9745012931525707,
      "num_tokens": 1779338.0,
      "step": 280
    },
    {
      "entropy": 0.06307904429268092,
      "epoch": 0.5801450362590648,
      "grad_norm": 0.3916383683681488,
      "learning_rate": 0.00019996673498948658,
      "loss": 0.06412749290466309,
      "mean_token_accuracy": 0.9828441753983498,
      "num_tokens": 1842872.0,
      "step": 290
    },
    {
      "entropy": 0.05729433842934668,
      "epoch": 0.6001500375093773,
      "grad_norm": 0.38883084058761597,
      "learning_rate": 0.00019994749064552214,
      "loss": 0.051730161905288695,
      "mean_token_accuracy": 0.9857619300484657,
      "num_tokens": 1904154.0,
      "step": 300
    },
    {
      "entropy": 0.03964498438872397,
      "epoch": 0.6201550387596899,
      "grad_norm": 0.6382091045379639,
      "learning_rate": 0.0001999238742745319,
      "loss": 0.042625024914741516,
      "mean_token_accuracy": 0.9874393172562123,
      "num_tokens": 1966060.0,
      "step": 310
    },
    {
      "entropy": 0.05733265266753733,
      "epoch": 0.6401600400100025,
      "grad_norm": 0.6905948519706726,
      "learning_rate": 0.00019989588690957249,
      "loss": 0.06732727885246277,
      "mean_token_accuracy": 0.982555440813303,
      "num_tokens": 2031939.0,
      "step": 320
    },
    {
      "entropy": 0.07143733124248683,
      "epoch": 0.660165041260315,
      "grad_norm": 0.4337286949157715,
      "learning_rate": 0.0001998635297749018,
      "loss": 0.058036553859710696,
      "mean_token_accuracy": 0.9844090364873409,
      "num_tokens": 2095271.0,
      "step": 330
    },
    {
      "entropy": 0.04944599290611222,
      "epoch": 0.6801700425106276,
      "grad_norm": 0.3526521921157837,
      "learning_rate": 0.00019982680428592584,
      "loss": 0.053340816497802736,
      "mean_token_accuracy": 0.9839612312614918,
      "num_tokens": 2158321.0,
      "step": 340
    },
    {
      "entropy": 0.06322509178426117,
      "epoch": 0.7001750437609402,
      "grad_norm": 0.5422689318656921,
      "learning_rate": 0.00019978571204913638,
      "loss": 0.051858377456665036,
      "mean_token_accuracy": 0.9867838315665722,
      "num_tokens": 2225062.0,
      "step": 350
    },
    {
      "entropy": 0.046866965352091935,
      "epoch": 0.7201800450112528,
      "grad_norm": 0.7101793885231018,
      "learning_rate": 0.000199740254862041,
      "loss": 0.05608519911766052,
      "mean_token_accuracy": 0.9851541951298713,
      "num_tokens": 2291157.0,
      "step": 360
    },
    {
      "entropy": 0.06618997184559702,
      "epoch": 0.7401850462615653,
      "grad_norm": 0.6186705827713013,
      "learning_rate": 0.00019969043471308436,
      "loss": 0.056741136312484744,
      "mean_token_accuracy": 0.9844918318092823,
      "num_tokens": 2353621.0,
      "step": 370
    },
    {
      "entropy": 0.052203354140510784,
      "epoch": 0.760190047511878,
      "grad_norm": 0.9665183424949646,
      "learning_rate": 0.00019963625378156114,
      "loss": 0.05278732180595398,
      "mean_token_accuracy": 0.9863728702068328,
      "num_tokens": 2415518.0,
      "step": 380
    },
    {
      "entropy": 0.04799637275282294,
      "epoch": 0.7801950487621906,
      "grad_norm": 0.45738333463668823,
      "learning_rate": 0.00019957771443752083,
      "loss": 0.044900187849998476,
      "mean_token_accuracy": 0.9872696734964848,
      "num_tokens": 2477716.0,
      "step": 390
    },
    {
      "entropy": 0.045062902639620005,
      "epoch": 0.8002000500125032,
      "grad_norm": 0.4040285646915436,
      "learning_rate": 0.00019951481924166396,
      "loss": 0.043438228964805606,
      "mean_token_accuracy": 0.9878960207104683,
      "num_tokens": 2540807.0,
      "step": 400
    },
    {
      "entropy": 0.036265855759847906,
      "epoch": 0.8202050512628157,
      "grad_norm": 0.4411618411540985,
      "learning_rate": 0.0001994475709452301,
      "loss": 0.04615793526172638,
      "mean_token_accuracy": 0.9870041370391845,
      "num_tokens": 2603019.0,
      "step": 410
    },
    {
      "entropy": 0.05851077587576583,
      "epoch": 0.8402100525131283,
      "grad_norm": 0.31862467527389526,
      "learning_rate": 0.0001993759724898777,
      "loss": 0.045778676867485046,
      "mean_token_accuracy": 0.9873291261494159,
      "num_tokens": 2667216.0,
      "step": 420
    },
    {
      "entropy": 0.03860169492545538,
      "epoch": 0.8602150537634409,
      "grad_norm": 0.29344919323921204,
      "learning_rate": 0.00019930002700755507,
      "loss": 0.03828286230564117,
      "mean_token_accuracy": 0.988411296159029,
      "num_tokens": 2729805.0,
      "step": 430
    },
    {
      "entropy": 0.03951184251927771,
      "epoch": 0.8802200550137534,
      "grad_norm": 0.4283672571182251,
      "learning_rate": 0.00019921973782036366,
      "loss": 0.038545310497283936,
      "mean_token_accuracy": 0.9878928653895855,
      "num_tokens": 2793292.0,
      "step": 440
    },
    {
      "entropy": 0.028071055363398045,
      "epoch": 0.900225056264066,
      "grad_norm": 0.7173567414283752,
      "learning_rate": 0.0001991351084404126,
      "loss": 0.03233465254306793,
      "mean_token_accuracy": 0.9912850938737392,
      "num_tokens": 2855909.0,
      "step": 450
    },
    {
      "entropy": 0.04781279567687306,
      "epoch": 0.9202300575143786,
      "grad_norm": 0.3380762040615082,
      "learning_rate": 0.00019904614256966512,
      "loss": 0.040875044465065,
      "mean_token_accuracy": 0.9889427930116653,
      "num_tokens": 2916597.0,
      "step": 460
    },
    {
      "entropy": 0.045654052757890896,
      "epoch": 0.9402350587646912,
      "grad_norm": 0.21747978031635284,
      "learning_rate": 0.0001989528440997767,
      "loss": 0.04227911233901978,
      "mean_token_accuracy": 0.987830163538456,
      "num_tokens": 2977012.0,
      "step": 470
    },
    {
      "entropy": 0.039897680119611326,
      "epoch": 0.9602400600150037,
      "grad_norm": 0.3250534236431122,
      "learning_rate": 0.00019885521711192453,
      "loss": 0.03866271674633026,
      "mean_token_accuracy": 0.9888145305216313,
      "num_tokens": 3039866.0,
      "step": 480
    },
    {
      "entropy": 0.0344677664746996,
      "epoch": 0.9802450612653163,
      "grad_norm": 0.5867555141448975,
      "learning_rate": 0.00019875326587662941,
      "loss": 0.030780380964279173,
      "mean_token_accuracy": 0.9910528786480427,
      "num_tokens": 3102450.0,
      "step": 490
    },
    {
      "entropy": 0.03601513006665473,
      "epoch": 1.0,
      "grad_norm": 0.19166764616966248,
      "learning_rate": 0.00019864699485356866,
      "loss": 0.04189955592155457,
      "mean_token_accuracy": 0.989976388744161,
      "num_tokens": 3162838.0,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_entropy": 0.051686967685315256,
      "eval_loss": 0.04872545599937439,
      "eval_mean_token_accuracy": 0.9849747346865164,
      "eval_num_tokens": 3162838.0,
      "eval_runtime": 69.4136,
      "eval_samples_per_second": 14.911,
      "eval_steps_per_second": 7.463,
      "step": 500
    },
    {
      "entropy": 0.04407569046597928,
      "epoch": 1.0200050012503126,
      "grad_norm": 0.2645007371902466,
      "learning_rate": 0.00019853640869138103,
      "loss": 0.03468368649482727,
      "mean_token_accuracy": 0.990016209334135,
      "num_tokens": 3228763.0,
      "step": 510
    },
    {
      "entropy": 0.035535094334045426,
      "epoch": 1.0400100025006251,
      "grad_norm": 0.40270644426345825,
      "learning_rate": 0.00019842151222746357,
      "loss": 0.03389493525028229,
      "mean_token_accuracy": 0.9887583516538143,
      "num_tokens": 3293010.0,
      "step": 520
    },
    {
      "entropy": 0.028601143200648948,
      "epoch": 1.0600150037509377,
      "grad_norm": 0.2681713104248047,
      "learning_rate": 0.00019830231048775977,
      "loss": 0.030898410081863403,
      "mean_token_accuracy": 0.9903686709702015,
      "num_tokens": 3354919.0,
      "step": 530
    },
    {
      "entropy": 0.034907517378451304,
      "epoch": 1.0800200050012503,
      "grad_norm": 0.23451267182826996,
      "learning_rate": 0.00019817880868653993,
      "loss": 0.03407395482063293,
      "mean_token_accuracy": 0.9890573389828206,
      "num_tokens": 3415784.0,
      "step": 540
    },
    {
      "entropy": 0.033313815778819846,
      "epoch": 1.1000250062515629,
      "grad_norm": 0.23950988054275513,
      "learning_rate": 0.0001980510122261729,
      "loss": 0.029421544075012206,
      "mean_token_accuracy": 0.9916689246892929,
      "num_tokens": 3480364.0,
      "step": 550
    },
    {
      "entropy": 0.02486751726246439,
      "epoch": 1.1200300075018754,
      "grad_norm": 0.4551165997982025,
      "learning_rate": 0.00019791892669688988,
      "loss": 0.020369285345077516,
      "mean_token_accuracy": 0.9940513521432877,
      "num_tokens": 3542678.0,
      "step": 560
    },
    {
      "entropy": 0.029150180192664264,
      "epoch": 1.140035008752188,
      "grad_norm": 0.20443572103977203,
      "learning_rate": 0.00019778255787653978,
      "loss": 0.02770337164402008,
      "mean_token_accuracy": 0.9897698886692524,
      "num_tokens": 3609086.0,
      "step": 570
    },
    {
      "entropy": 0.031121585314394906,
      "epoch": 1.1600400100025006,
      "grad_norm": 0.3653818368911743,
      "learning_rate": 0.00019764191173033663,
      "loss": 0.025546741485595704,
      "mean_token_accuracy": 0.9925875566899777,
      "num_tokens": 3670310.0,
      "step": 580
    },
    {
      "entropy": 0.02902457951568067,
      "epoch": 1.1800450112528131,
      "grad_norm": 0.4811317026615143,
      "learning_rate": 0.00019749699441059843,
      "loss": 0.027299800515174867,
      "mean_token_accuracy": 0.9906462356448174,
      "num_tokens": 3735307.0,
      "step": 590
    },
    {
      "entropy": 0.030135014103143475,
      "epoch": 1.2000500125031257,
      "grad_norm": 0.21709105372428894,
      "learning_rate": 0.00019734781225647828,
      "loss": 0.02665548324584961,
      "mean_token_accuracy": 0.9932463668286801,
      "num_tokens": 3797213.0,
      "step": 600
    },
    {
      "entropy": 0.03371675145754125,
      "epoch": 1.2200550137534383,
      "grad_norm": 0.631148636341095,
      "learning_rate": 0.00019719437179368688,
      "loss": 0.03535972833633423,
      "mean_token_accuracy": 0.9895499177277088,
      "num_tokens": 3859400.0,
      "step": 610
    },
    {
      "entropy": 0.028748418507166206,
      "epoch": 1.2400600150037508,
      "grad_norm": 0.42123743891716003,
      "learning_rate": 0.00019703667973420706,
      "loss": 0.027658408880233763,
      "mean_token_accuracy": 0.9910015679895878,
      "num_tokens": 3920848.0,
      "step": 620
    },
    {
      "entropy": 0.029886699473718182,
      "epoch": 1.2600650162540634,
      "grad_norm": 0.23473748564720154,
      "learning_rate": 0.00019687474297600045,
      "loss": 0.028230640292167663,
      "mean_token_accuracy": 0.9922301307320595,
      "num_tokens": 3984529.0,
      "step": 630
    },
    {
      "entropy": 0.03151440276997164,
      "epoch": 1.280070017504376,
      "grad_norm": 0.2611916661262512,
      "learning_rate": 0.00019670856860270542,
      "loss": 0.027012214064598083,
      "mean_token_accuracy": 0.9917375601828098,
      "num_tokens": 4048411.0,
      "step": 640
    },
    {
      "entropy": 0.027149295064737088,
      "epoch": 1.3000750187546886,
      "grad_norm": 0.12378375232219696,
      "learning_rate": 0.0001965381638833274,
      "loss": 0.023584455251693726,
      "mean_token_accuracy": 0.9936468034982682,
      "num_tokens": 4112116.0,
      "step": 650
    },
    {
      "entropy": 0.028390987019520253,
      "epoch": 1.3200800200050011,
      "grad_norm": 0.18841037154197693,
      "learning_rate": 0.00019636353627192082,
      "loss": 0.025832393765449525,
      "mean_token_accuracy": 0.9933209784328938,
      "num_tokens": 4175458.0,
      "step": 660
    },
    {
      "entropy": 0.032682666774780954,
      "epoch": 1.3400850212553137,
      "grad_norm": 0.292369544506073,
      "learning_rate": 0.00019618469340726319,
      "loss": 0.032191649079322815,
      "mean_token_accuracy": 0.9895716637372971,
      "num_tokens": 4236466.0,
      "step": 670
    },
    {
      "entropy": 0.032574003856279884,
      "epoch": 1.3600900225056263,
      "grad_norm": 0.2205415517091751,
      "learning_rate": 0.00019600164311252068,
      "loss": 0.02866535782814026,
      "mean_token_accuracy": 0.991417796164751,
      "num_tokens": 4299775.0,
      "step": 680
    },
    {
      "entropy": 0.04169052811048459,
      "epoch": 1.380095023755939,
      "grad_norm": 0.3276292681694031,
      "learning_rate": 0.00019581439339490624,
      "loss": 0.03524776101112366,
      "mean_token_accuracy": 0.9886757604777813,
      "num_tokens": 4365609.0,
      "step": 690
    },
    {
      "entropy": 0.03188371795695275,
      "epoch": 1.4001000250062516,
      "grad_norm": 0.4103780686855316,
      "learning_rate": 0.0001956229524453291,
      "loss": 0.035996857285499576,
      "mean_token_accuracy": 0.9892666183412075,
      "num_tokens": 4432330.0,
      "step": 700
    },
    {
      "entropy": 0.02871296225930564,
      "epoch": 1.4201050262565642,
      "grad_norm": 0.24775762856006622,
      "learning_rate": 0.00019542732863803662,
      "loss": 0.02328706532716751,
      "mean_token_accuracy": 0.9937438026070595,
      "num_tokens": 4496465.0,
      "step": 710
    },
    {
      "entropy": 0.03388670613931026,
      "epoch": 1.4401100275068768,
      "grad_norm": 0.25881657004356384,
      "learning_rate": 0.00019522753053024787,
      "loss": 0.03300818204879761,
      "mean_token_accuracy": 0.9890970505774022,
      "num_tokens": 4560755.0,
      "step": 720
    },
    {
      "entropy": 0.03670836841047276,
      "epoch": 1.4601150287571893,
      "grad_norm": 0.23787066340446472,
      "learning_rate": 0.00019502356686177926,
      "loss": 0.03214167952537537,
      "mean_token_accuracy": 0.9903383336961269,
      "num_tokens": 4625202.0,
      "step": 730
    },
    {
      "entropy": 0.03299383492558263,
      "epoch": 1.480120030007502,
      "grad_norm": 0.31929901242256165,
      "learning_rate": 0.00019481544655466245,
      "loss": 0.031187814474105836,
      "mean_token_accuracy": 0.9905215993523597,
      "num_tokens": 4692174.0,
      "step": 740
    },
    {
      "entropy": 0.023855643330898603,
      "epoch": 1.5001250312578145,
      "grad_norm": 0.24006661772727966,
      "learning_rate": 0.00019460317871275394,
      "loss": 0.025655516982078554,
      "mean_token_accuracy": 0.9913376875221729,
      "num_tokens": 4753192.0,
      "step": 750
    },
    {
      "entropy": 0.033190094074234365,
      "epoch": 1.520130032508127,
      "grad_norm": 0.2834339737892151,
      "learning_rate": 0.00019438677262133668,
      "loss": 0.03126271665096283,
      "mean_token_accuracy": 0.9907064586877823,
      "num_tokens": 4817884.0,
      "step": 760
    },
    {
      "entropy": 0.03176074127841275,
      "epoch": 1.5401350337584396,
      "grad_norm": 0.4165857136249542,
      "learning_rate": 0.00019416623774671425,
      "loss": 0.033395078778266904,
      "mean_token_accuracy": 0.9902952447533607,
      "num_tokens": 4879601.0,
      "step": 770
    },
    {
      "entropy": 0.03441936054150574,
      "epoch": 1.5601400350087522,
      "grad_norm": 0.5297831296920776,
      "learning_rate": 0.00019394158373579645,
      "loss": 0.02917470932006836,
      "mean_token_accuracy": 0.9917018190026283,
      "num_tokens": 4946026.0,
      "step": 780
    },
    {
      "entropy": 0.029212182367336935,
      "epoch": 1.5801450362590648,
      "grad_norm": 0.3761133849620819,
      "learning_rate": 0.00019371282041567752,
      "loss": 0.02672044336795807,
      "mean_token_accuracy": 0.9930847369134426,
      "num_tokens": 5009097.0,
      "step": 790
    },
    {
      "entropy": 0.03556556548574008,
      "epoch": 1.6001500375093773,
      "grad_norm": 0.2545328438282013,
      "learning_rate": 0.0001934799577932062,
      "loss": 0.0338908702135086,
      "mean_token_accuracy": 0.990411739051342,
      "num_tokens": 5076314.0,
      "step": 800
    },
    {
      "entropy": 0.02515874128730502,
      "epoch": 1.62015503875969,
      "grad_norm": 0.3428667187690735,
      "learning_rate": 0.0001932430060545479,
      "loss": 0.022386419773101806,
      "mean_token_accuracy": 0.9932228110730648,
      "num_tokens": 5135522.0,
      "step": 810
    },
    {
      "entropy": 0.021622967024450192,
      "epoch": 1.6401600400100025,
      "grad_norm": 0.19730041921138763,
      "learning_rate": 0.00019300197556473936,
      "loss": 0.02211230844259262,
      "mean_token_accuracy": 0.993052315711975,
      "num_tokens": 5198651.0,
      "step": 820
    },
    {
      "entropy": 0.022755468662944624,
      "epoch": 1.660165041260315,
      "grad_norm": 0.24043497443199158,
      "learning_rate": 0.00019275687686723497,
      "loss": 0.021593029797077178,
      "mean_token_accuracy": 0.9927247293293476,
      "num_tokens": 5258441.0,
      "step": 830
    },
    {
      "entropy": 0.02988760783628095,
      "epoch": 1.6801700425106276,
      "grad_norm": 0.22126108407974243,
      "learning_rate": 0.0001925077206834458,
      "loss": 0.02743455469608307,
      "mean_token_accuracy": 0.9924947433173656,
      "num_tokens": 5323635.0,
      "step": 840
    },
    {
      "entropy": 0.022633779112948105,
      "epoch": 1.7001750437609402,
      "grad_norm": 0.2528051435947418,
      "learning_rate": 0.00019225451791227052,
      "loss": 0.01915370374917984,
      "mean_token_accuracy": 0.9935295671224594,
      "num_tokens": 5388098.0,
      "step": 850
    },
    {
      "entropy": 0.024347139010205864,
      "epoch": 1.7201800450112528,
      "grad_norm": 0.35426992177963257,
      "learning_rate": 0.00019199727962961852,
      "loss": 0.025216898322105406,
      "mean_token_accuracy": 0.9927972495555878,
      "num_tokens": 5450303.0,
      "step": 860
    },
    {
      "entropy": 0.02343553317186888,
      "epoch": 1.7401850462615653,
      "grad_norm": 0.2603664994239807,
      "learning_rate": 0.00019173601708792566,
      "loss": 0.022424712777137756,
      "mean_token_accuracy": 0.9926920354366302,
      "num_tokens": 5513723.0,
      "step": 870
    },
    {
      "entropy": 0.03952418522676453,
      "epoch": 1.7601900475118781,
      "grad_norm": 0.21528302133083344,
      "learning_rate": 0.0001914707417156619,
      "loss": 0.03777352273464203,
      "mean_token_accuracy": 0.9895162962377071,
      "num_tokens": 5575942.0,
      "step": 880
    },
    {
      "entropy": 0.037056630512233825,
      "epoch": 1.7801950487621907,
      "grad_norm": 0.39761167764663696,
      "learning_rate": 0.00019120146511683142,
      "loss": 0.02916957139968872,
      "mean_token_accuracy": 0.9906649015843868,
      "num_tokens": 5640243.0,
      "step": 890
    },
    {
      "entropy": 0.022086267481790857,
      "epoch": 1.8002000500125033,
      "grad_norm": 0.43694376945495605,
      "learning_rate": 0.00019092819907046493,
      "loss": 0.023631574213504793,
      "mean_token_accuracy": 0.9936031945049763,
      "num_tokens": 5702162.0,
      "step": 900
    },
    {
      "entropy": 0.028572715594782493,
      "epoch": 1.8202050512628158,
      "grad_norm": 0.4214474558830261,
      "learning_rate": 0.00019065095553010458,
      "loss": 0.025940075516700745,
      "mean_token_accuracy": 0.9905624501407146,
      "num_tokens": 5764127.0,
      "step": 910
    },
    {
      "entropy": 0.028239472245331854,
      "epoch": 1.8402100525131284,
      "grad_norm": 0.40316465497016907,
      "learning_rate": 0.00019036974662328096,
      "loss": 0.027613845467567445,
      "mean_token_accuracy": 0.991669587045908,
      "num_tokens": 5827235.0,
      "step": 920
    },
    {
      "entropy": 0.028241146955406294,
      "epoch": 1.860215053763441,
      "grad_norm": 0.524972677230835,
      "learning_rate": 0.0001900845846509827,
      "loss": 0.03177001476287842,
      "mean_token_accuracy": 0.9903169378638268,
      "num_tokens": 5890040.0,
      "step": 930
    },
    {
      "entropy": 0.03260187199921347,
      "epoch": 1.8802200550137536,
      "grad_norm": 0.43904051184654236,
      "learning_rate": 0.00018979548208711817,
      "loss": 0.02139996737241745,
      "mean_token_accuracy": 0.9935068063437938,
      "num_tokens": 5953149.0,
      "step": 940
    },
    {
      "entropy": 0.025593279630993494,
      "epoch": 1.9002250562640661,
      "grad_norm": 0.25154390931129456,
      "learning_rate": 0.00018950245157797014,
      "loss": 0.02489333599805832,
      "mean_token_accuracy": 0.9933448024094105,
      "num_tokens": 6014655.0,
      "step": 950
    },
    {
      "entropy": 0.025231685642211232,
      "epoch": 1.9202300575143787,
      "grad_norm": 0.2318635731935501,
      "learning_rate": 0.00018920550594164238,
      "loss": 0.02543329894542694,
      "mean_token_accuracy": 0.9929649449884892,
      "num_tokens": 6078667.0,
      "step": 960
    },
    {
      "entropy": 0.0415392193797743,
      "epoch": 1.9402350587646913,
      "grad_norm": 0.2020518183708191,
      "learning_rate": 0.00018890465816749896,
      "loss": 0.03582499623298645,
      "mean_token_accuracy": 0.9897747471928596,
      "num_tokens": 6139688.0,
      "step": 970
    },
    {
      "entropy": 0.026439224516798275,
      "epoch": 1.9602400600150038,
      "grad_norm": 0.25302180647850037,
      "learning_rate": 0.00018859992141559615,
      "loss": 0.023522551357746124,
      "mean_token_accuracy": 0.9931276544928551,
      "num_tokens": 6203193.0,
      "step": 980
    },
    {
      "entropy": 0.02257391346647637,
      "epoch": 1.9802450612653164,
      "grad_norm": 0.37129050493240356,
      "learning_rate": 0.00018829130901610667,
      "loss": 0.023089873790740966,
      "mean_token_accuracy": 0.9930807471275329,
      "num_tokens": 6265353.0,
      "step": 990
    },
    {
      "entropy": 0.024826381788765894,
      "epoch": 2.0,
      "grad_norm": 0.1416839361190796,
      "learning_rate": 0.00018797883446873662,
      "loss": 0.020879687368869783,
      "mean_token_accuracy": 0.9931328839893583,
      "num_tokens": 6325676.0,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_entropy": 0.028477752043098808,
      "eval_loss": 0.042340315878391266,
      "eval_mean_token_accuracy": 0.988268693211456,
      "eval_num_tokens": 6325676.0,
      "eval_runtime": 69.2562,
      "eval_samples_per_second": 14.945,
      "eval_steps_per_second": 7.479,
      "step": 1000
    },
    {
      "entropy": 0.021088267347658986,
      "epoch": 2.0200050012503126,
      "grad_norm": 0.33506715297698975,
      "learning_rate": 0.00018766251144213504,
      "loss": 0.0183292493224144,
      "mean_token_accuracy": 0.9939773567020893,
      "num_tokens": 6391016.0,
      "step": 1010
    },
    {
      "entropy": 0.021377279089938382,
      "epoch": 2.040010002500625,
      "grad_norm": 0.10510263592004776,
      "learning_rate": 0.00018734235377329582,
      "loss": 0.016541089117527007,
      "mean_token_accuracy": 0.9943198271095752,
      "num_tokens": 6455040.0,
      "step": 1020
    },
    {
      "entropy": 0.01835900279111229,
      "epoch": 2.0600150037509377,
      "grad_norm": 0.25538530945777893,
      "learning_rate": 0.0001870183754669526,
      "loss": 0.017676208913326264,
      "mean_token_accuracy": 0.9943479098379612,
      "num_tokens": 6517983.0,
      "step": 1030
    },
    {
      "entropy": 0.024707998137455434,
      "epoch": 2.0800200050012503,
      "grad_norm": 0.1953907310962677,
      "learning_rate": 0.00018669059069496594,
      "loss": 0.018231543898582458,
      "mean_token_accuracy": 0.9940299488604069,
      "num_tokens": 6582155.0,
      "step": 1040
    },
    {
      "entropy": 0.016339628079731484,
      "epoch": 2.100025006251563,
      "grad_norm": 0.23519866168498993,
      "learning_rate": 0.00018635901379570377,
      "loss": 0.015705856680870055,
      "mean_token_accuracy": 0.9944866336882114,
      "num_tokens": 6644403.0,
      "step": 1050
    },
    {
      "entropy": 0.021186151236179285,
      "epoch": 2.1200300075018754,
      "grad_norm": 0.7782704830169678,
      "learning_rate": 0.00018602365927341375,
      "loss": 0.0239964097738266,
      "mean_token_accuracy": 0.9942199148237705,
      "num_tokens": 6709057.0,
      "step": 1060
    },
    {
      "entropy": 0.02400836138986051,
      "epoch": 2.140035008752188,
      "grad_norm": 0.25117242336273193,
      "learning_rate": 0.0001856845417975891,
      "loss": 0.0214329332113266,
      "mean_token_accuracy": 0.99395372569561,
      "num_tokens": 6772040.0,
      "step": 1070
    },
    {
      "entropy": 0.01854798578133341,
      "epoch": 2.1600400100025006,
      "grad_norm": 0.24927817285060883,
      "learning_rate": 0.0001853416762023268,
      "loss": 0.01707075983285904,
      "mean_token_accuracy": 0.9950113117694854,
      "num_tokens": 6835866.0,
      "step": 1080
    },
    {
      "entropy": 0.019260429358109833,
      "epoch": 2.180045011252813,
      "grad_norm": 0.24353672564029694,
      "learning_rate": 0.00018499507748567873,
      "loss": 0.014785376191139222,
      "mean_token_accuracy": 0.9959283553063869,
      "num_tokens": 6899725.0,
      "step": 1090
    },
    {
      "entropy": 0.02487965851032641,
      "epoch": 2.2000500125031257,
      "grad_norm": 0.288215309381485,
      "learning_rate": 0.00018464476080899559,
      "loss": 0.019350311160087584,
      "mean_token_accuracy": 0.9941258184611798,
      "num_tokens": 6963141.0,
      "step": 1100
    },
    {
      "entropy": 0.01580278711626306,
      "epoch": 2.2200550137534383,
      "grad_norm": 0.0646059513092041,
      "learning_rate": 0.00018429074149626363,
      "loss": 0.017427970468997956,
      "mean_token_accuracy": 0.9952689491212368,
      "num_tokens": 7028325.0,
      "step": 1110
    },
    {
      "entropy": 0.020199327028240077,
      "epoch": 2.240060015003751,
      "grad_norm": 0.160948246717453,
      "learning_rate": 0.0001839330350334345,
      "loss": 0.01726052612066269,
      "mean_token_accuracy": 0.9933249458670617,
      "num_tokens": 7092920.0,
      "step": 1120
    },
    {
      "entropy": 0.021389624777657445,
      "epoch": 2.2600650162540634,
      "grad_norm": 0.2359917163848877,
      "learning_rate": 0.00018357165706774767,
      "loss": 0.01608244627714157,
      "mean_token_accuracy": 0.9940837919712067,
      "num_tokens": 7160997.0,
      "step": 1130
    },
    {
      "entropy": 0.02022790874907514,
      "epoch": 2.280070017504376,
      "grad_norm": 0.07118914276361465,
      "learning_rate": 0.00018320662340704609,
      "loss": 0.020376379787921905,
      "mean_token_accuracy": 0.9942706093192101,
      "num_tokens": 7226007.0,
      "step": 1140
    },
    {
      "entropy": 0.01867675751855131,
      "epoch": 2.3000750187546886,
      "grad_norm": 0.25310513377189636,
      "learning_rate": 0.00018283795001908457,
      "loss": 0.01612715721130371,
      "mean_token_accuracy": 0.9942055746912957,
      "num_tokens": 7287986.0,
      "step": 1150
    },
    {
      "entropy": 0.015577676898101345,
      "epoch": 2.320080020005001,
      "grad_norm": 0.2142266184091568,
      "learning_rate": 0.0001824656530308315,
      "loss": 0.016605789959430694,
      "mean_token_accuracy": 0.9948085315525532,
      "num_tokens": 7349421.0,
      "step": 1160
    },
    {
      "entropy": 0.020357475349737798,
      "epoch": 2.3400850212553137,
      "grad_norm": 0.07947535812854767,
      "learning_rate": 0.00018208974872776322,
      "loss": 0.018196111917495726,
      "mean_token_accuracy": 0.993843074887991,
      "num_tokens": 7411941.0,
      "step": 1170
    },
    {
      "entropy": 0.017023067966511006,
      "epoch": 2.3600900225056263,
      "grad_norm": 0.19202570617198944,
      "learning_rate": 0.00018171025355315164,
      "loss": 0.016091108322143555,
      "mean_token_accuracy": 0.9953217662870883,
      "num_tokens": 7474356.0,
      "step": 1180
    },
    {
      "entropy": 0.01825423450791277,
      "epoch": 2.380095023755939,
      "grad_norm": 0.24460658431053162,
      "learning_rate": 0.00018132718410734515,
      "loss": 0.015425822138786316,
      "mean_token_accuracy": 0.9943965286016464,
      "num_tokens": 7536565.0,
      "step": 1190
    },
    {
      "entropy": 0.015542891589575447,
      "epoch": 2.4001000250062514,
      "grad_norm": 0.24659694731235504,
      "learning_rate": 0.00018094055714704225,
      "loss": 0.013514925539493561,
      "mean_token_accuracy": 0.9943179704248906,
      "num_tokens": 7599751.0,
      "step": 1200
    },
    {
      "entropy": 0.015307287167524919,
      "epoch": 2.420105026256564,
      "grad_norm": 0.2550601065158844,
      "learning_rate": 0.0001805503895845587,
      "loss": 0.012726837396621704,
      "mean_token_accuracy": 0.9962130591273308,
      "num_tokens": 7661520.0,
      "step": 1210
    },
    {
      "entropy": 0.01749844834121177,
      "epoch": 2.4401100275068766,
      "grad_norm": 0.15603283047676086,
      "learning_rate": 0.00018015669848708767,
      "loss": 0.012257835268974305,
      "mean_token_accuracy": 0.9954387851059436,
      "num_tokens": 7726291.0,
      "step": 1220
    },
    {
      "entropy": 0.019737370508664753,
      "epoch": 2.460115028757189,
      "grad_norm": 0.22746174037456512,
      "learning_rate": 0.0001797595010759531,
      "loss": 0.018561355769634247,
      "mean_token_accuracy": 0.9940820440649987,
      "num_tokens": 7788766.0,
      "step": 1230
    },
    {
      "entropy": 0.015929855390277227,
      "epoch": 2.4801200300075017,
      "grad_norm": 0.3152976334095001,
      "learning_rate": 0.0001793588147258565,
      "loss": 0.013849316537380219,
      "mean_token_accuracy": 0.9952766291797162,
      "num_tokens": 7851396.0,
      "step": 1240
    },
    {
      "entropy": 0.02108022033935413,
      "epoch": 2.5001250312578147,
      "grad_norm": 0.2762889862060547,
      "learning_rate": 0.00017895465696411692,
      "loss": 0.019299986958503722,
      "mean_token_accuracy": 0.9940553769469261,
      "num_tokens": 7915421.0,
      "step": 1250
    },
    {
      "entropy": 0.020866505106096157,
      "epoch": 2.520130032508127,
      "grad_norm": 0.24086585640907288,
      "learning_rate": 0.00017854704546990408,
      "loss": 0.020875005424022673,
      "mean_token_accuracy": 0.9940896175801754,
      "num_tokens": 7977133.0,
      "step": 1260
    },
    {
      "entropy": 0.020857046739547514,
      "epoch": 2.54013503375844,
      "grad_norm": 0.23781460523605347,
      "learning_rate": 0.0001781359980734653,
      "loss": 0.018643879890441896,
      "mean_token_accuracy": 0.9929048053920269,
      "num_tokens": 8040191.0,
      "step": 1270
    },
    {
      "entropy": 0.021198420476866885,
      "epoch": 2.560140035008752,
      "grad_norm": 0.2054099142551422,
      "learning_rate": 0.0001777215327553452,
      "loss": 0.017947974801063537,
      "mean_token_accuracy": 0.9934561550617218,
      "num_tokens": 8100152.0,
      "step": 1280
    },
    {
      "entropy": 0.018285114454920405,
      "epoch": 2.580145036259065,
      "grad_norm": 0.469458669424057,
      "learning_rate": 0.00017730366764559955,
      "loss": 0.0151192307472229,
      "mean_token_accuracy": 0.9951836079359054,
      "num_tokens": 8162591.0,
      "step": 1290
    },
    {
      "entropy": 0.018374070005665998,
      "epoch": 2.600150037509377,
      "grad_norm": 0.2882782518863678,
      "learning_rate": 0.00017688242102300192,
      "loss": 0.0184975802898407,
      "mean_token_accuracy": 0.9948060251772404,
      "num_tokens": 8225224.0,
      "step": 1300
    },
    {
      "entropy": 0.022974171601526906,
      "epoch": 2.62015503875969,
      "grad_norm": 0.21924524009227753,
      "learning_rate": 0.00017645781131424423,
      "loss": 0.023296315968036652,
      "mean_token_accuracy": 0.9924322210252285,
      "num_tokens": 8290274.0,
      "step": 1310
    },
    {
      "entropy": 0.02169415617827326,
      "epoch": 2.6401600400100023,
      "grad_norm": 0.354758620262146,
      "learning_rate": 0.00017602985709313073,
      "loss": 0.017917373776435853,
      "mean_token_accuracy": 0.9948078036308289,
      "num_tokens": 8354370.0,
      "step": 1320
    },
    {
      "entropy": 0.016268294051405972,
      "epoch": 2.6601650412603153,
      "grad_norm": 0.4151551127433777,
      "learning_rate": 0.00017559857707976536,
      "loss": 0.012286465615034103,
      "mean_token_accuracy": 0.9961770802736283,
      "num_tokens": 8415022.0,
      "step": 1330
    },
    {
      "entropy": 0.0170176492218161,
      "epoch": 2.6801700425106274,
      "grad_norm": 0.4065402150154114,
      "learning_rate": 0.0001751639901397331,
      "loss": 0.01499750316143036,
      "mean_token_accuracy": 0.9952280893921852,
      "num_tokens": 8478286.0,
      "step": 1340
    },
    {
      "entropy": 0.018367627350380646,
      "epoch": 2.7001750437609404,
      "grad_norm": 0.1335880309343338,
      "learning_rate": 0.0001747261152832746,
      "loss": 0.015018537640571594,
      "mean_token_accuracy": 0.9949840374290944,
      "num_tokens": 8541154.0,
      "step": 1350
    },
    {
      "entropy": 0.019403737914399245,
      "epoch": 2.7201800450112525,
      "grad_norm": 0.20553363859653473,
      "learning_rate": 0.00017428497166445452,
      "loss": 0.019237272441387177,
      "mean_token_accuracy": 0.9930156201124192,
      "num_tokens": 8605361.0,
      "step": 1360
    },
    {
      "entropy": 0.02298831292137038,
      "epoch": 2.7401850462615656,
      "grad_norm": 0.31576329469680786,
      "learning_rate": 0.00017384057858032393,
      "loss": 0.019147740304470064,
      "mean_token_accuracy": 0.9930574476718903,
      "num_tokens": 8669155.0,
      "step": 1370
    },
    {
      "entropy": 0.01718737747578416,
      "epoch": 2.760190047511878,
      "grad_norm": 0.21389739215373993,
      "learning_rate": 0.00017339295547007594,
      "loss": 0.017152118682861327,
      "mean_token_accuracy": 0.995334691554308,
      "num_tokens": 8735019.0,
      "step": 1380
    },
    {
      "entropy": 0.019683032816101332,
      "epoch": 2.7801950487621907,
      "grad_norm": 0.26539650559425354,
      "learning_rate": 0.00017294212191419547,
      "loss": 0.019429606199264527,
      "mean_token_accuracy": 0.9936951123178005,
      "num_tokens": 8800972.0,
      "step": 1390
    },
    {
      "entropy": 0.018636453218641692,
      "epoch": 2.8002000500125033,
      "grad_norm": 0.2084941565990448,
      "learning_rate": 0.00017248809763360277,
      "loss": 0.01909356415271759,
      "mean_token_accuracy": 0.9940625011920929,
      "num_tokens": 8867596.0,
      "step": 1400
    },
    {
      "entropy": 0.023316194582730532,
      "epoch": 2.820205051262816,
      "grad_norm": 0.21403200924396515,
      "learning_rate": 0.0001720309024887907,
      "loss": 0.018046848475933075,
      "mean_token_accuracy": 0.9939217306673527,
      "num_tokens": 8927516.0,
      "step": 1410
    },
    {
      "entropy": 0.01655098560877377,
      "epoch": 2.8402100525131284,
      "grad_norm": 0.33148443698883057,
      "learning_rate": 0.000171570556478956,
      "loss": 0.018755699694156646,
      "mean_token_accuracy": 0.9942230053246022,
      "num_tokens": 8989089.0,
      "step": 1420
    },
    {
      "entropy": 0.02549898542056326,
      "epoch": 2.860215053763441,
      "grad_norm": 0.18521511554718018,
      "learning_rate": 0.00017110707974112447,
      "loss": 0.024141687154769897,
      "mean_token_accuracy": 0.9930150359869003,
      "num_tokens": 9054435.0,
      "step": 1430
    },
    {
      "entropy": 0.01900827525241766,
      "epoch": 2.8802200550137536,
      "grad_norm": 0.23067928850650787,
      "learning_rate": 0.0001706404925492701,
      "loss": 0.020157690346240997,
      "mean_token_accuracy": 0.9943146407604218,
      "num_tokens": 9114274.0,
      "step": 1440
    },
    {
      "entropy": 0.01623811650206335,
      "epoch": 2.900225056264066,
      "grad_norm": 0.5235961079597473,
      "learning_rate": 0.00017017081531342813,
      "loss": 0.014147150516510009,
      "mean_token_accuracy": 0.9944271765649318,
      "num_tokens": 9174841.0,
      "step": 1450
    },
    {
      "entropy": 0.023971330239146483,
      "epoch": 2.9202300575143787,
      "grad_norm": 0.11377973109483719,
      "learning_rate": 0.00016969806857880241,
      "loss": 0.022856634855270386,
      "mean_token_accuracy": 0.9932261377573013,
      "num_tokens": 9239456.0,
      "step": 1460
    },
    {
      "entropy": 0.02552748184389202,
      "epoch": 2.9402350587646913,
      "grad_norm": 0.12835904955863953,
      "learning_rate": 0.00016922227302486667,
      "loss": 0.02486345320940018,
      "mean_token_accuracy": 0.9918816141784191,
      "num_tokens": 9305144.0,
      "step": 1470
    },
    {
      "entropy": 0.019867337332107125,
      "epoch": 2.960240060015004,
      "grad_norm": 0.18033206462860107,
      "learning_rate": 0.00016874344946445974,
      "loss": 0.018220885097980498,
      "mean_token_accuracy": 0.9945706635713577,
      "num_tokens": 9365854.0,
      "step": 1480
    },
    {
      "entropy": 0.01948691344150575,
      "epoch": 2.9802450612653164,
      "grad_norm": 0.2495020180940628,
      "learning_rate": 0.00016826161884287533,
      "loss": 0.01618766337633133,
      "mean_token_accuracy": 0.9956672258675099,
      "num_tokens": 9427287.0,
      "step": 1490
    },
    {
      "entropy": 0.022937397798228586,
      "epoch": 3.0,
      "grad_norm": 0.2727943956851959,
      "learning_rate": 0.00016777680223694575,
      "loss": 0.024902991950511932,
      "mean_token_accuracy": 0.9909723401069641,
      "num_tokens": 9488514.0,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_entropy": 0.02839457441272365,
      "eval_loss": 0.03645886108279228,
      "eval_mean_token_accuracy": 0.9902446437986661,
      "eval_num_tokens": 9488514.0,
      "eval_runtime": 69.2921,
      "eval_samples_per_second": 14.937,
      "eval_steps_per_second": 7.476,
      "step": 1500
    }
  ],
  "logging_steps": 10,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.1295589576933376e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}