{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 2229,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.3228859663009644,
      "epoch": 0.013458950201884253,
      "grad_norm": 1.4003818035125732,
      "learning_rate": 2.4193548387096776e-06,
      "loss": 0.3947859287261963,
      "mean_token_accuracy": 0.8766823470592499,
      "num_tokens": 179351.0,
      "step": 10
    },
    {
      "entropy": 1.3223209857940674,
      "epoch": 0.026917900403768506,
      "grad_norm": 1.4293484687805176,
      "learning_rate": 5.1075268817204305e-06,
      "loss": 0.36992483139038085,
      "mean_token_accuracy": 0.8841759026050567,
      "num_tokens": 358681.0,
      "step": 20
    },
    {
      "entropy": 1.3241522073745728,
      "epoch": 0.040376850605652756,
      "grad_norm": 1.3410050868988037,
      "learning_rate": 7.795698924731183e-06,
      "loss": 0.3491047859191895,
      "mean_token_accuracy": 0.8843624591827393,
      "num_tokens": 538488.0,
      "step": 30
    },
    {
      "entropy": 1.3318131804466247,
      "epoch": 0.05383580080753701,
      "grad_norm": 1.08042573928833,
      "learning_rate": 1.0483870967741936e-05,
      "loss": 0.30254349708557127,
      "mean_token_accuracy": 0.8926964640617371,
      "num_tokens": 718118.0,
      "step": 40
    },
    {
      "entropy": 1.3337590098381042,
      "epoch": 0.06729475100942127,
      "grad_norm": 0.8615964651107788,
      "learning_rate": 1.3172043010752688e-05,
      "loss": 0.26572132110595703,
      "mean_token_accuracy": 0.9041078448295593,
      "num_tokens": 897209.0,
      "step": 50
    },
    {
      "entropy": 1.3126742005348206,
      "epoch": 0.08075370121130551,
      "grad_norm": 0.9317428469657898,
      "learning_rate": 1.586021505376344e-05,
      "loss": 0.23249435424804688,
      "mean_token_accuracy": 0.9135358154773712,
      "num_tokens": 1076448.0,
      "step": 60
    },
    {
      "entropy": 1.2961446046829224,
      "epoch": 0.09421265141318977,
      "grad_norm": 1.1822317838668823,
      "learning_rate": 1.8548387096774193e-05,
      "loss": 0.21359801292419434,
      "mean_token_accuracy": 0.9192376554012298,
      "num_tokens": 1255796.0,
      "step": 70
    },
    {
      "entropy": 1.2832220673561097,
      "epoch": 0.10767160161507403,
      "grad_norm": 1.1605602502822876,
      "learning_rate": 2.1236559139784946e-05,
      "loss": 0.20212924480438232,
      "mean_token_accuracy": 0.9217188417911529,
      "num_tokens": 1435611.0,
      "step": 80
    },
    {
      "entropy": 1.2762320637702942,
      "epoch": 0.12113055181695828,
      "grad_norm": 1.2006449699401855,
      "learning_rate": 2.39247311827957e-05,
      "loss": 0.18676402568817138,
      "mean_token_accuracy": 0.9310197174549103,
      "num_tokens": 1614872.0,
      "step": 90
    },
    {
      "entropy": 1.274475383758545,
      "epoch": 0.13458950201884254,
      "grad_norm": 1.2656220197677612,
      "learning_rate": 2.661290322580645e-05,
      "loss": 0.17775335311889648,
      "mean_token_accuracy": 0.9305342018604279,
      "num_tokens": 1794050.0,
      "step": 100
    },
    {
      "entropy": 1.2607086300849915,
      "epoch": 0.1480484522207268,
      "grad_norm": 1.2415285110473633,
      "learning_rate": 2.9301075268817207e-05,
      "loss": 0.17471678256988527,
      "mean_token_accuracy": 0.9317916095256805,
      "num_tokens": 1973292.0,
      "step": 110
    },
    {
      "entropy": 1.2618489027023316,
      "epoch": 0.16150740242261102,
      "grad_norm": 1.268917202949524,
      "learning_rate": 3.198924731182796e-05,
      "loss": 0.17752587795257568,
      "mean_token_accuracy": 0.9324639737606049,
      "num_tokens": 2152526.0,
      "step": 120
    },
    {
      "entropy": 1.2617292761802674,
      "epoch": 0.17496635262449528,
      "grad_norm": 1.1972988843917847,
      "learning_rate": 3.467741935483872e-05,
      "loss": 0.1649715781211853,
      "mean_token_accuracy": 0.9361909210681916,
      "num_tokens": 2331952.0,
      "step": 130
    },
    {
      "entropy": 1.2492274522781373,
      "epoch": 0.18842530282637954,
      "grad_norm": 1.2891550064086914,
      "learning_rate": 3.736559139784947e-05,
      "loss": 0.15520352125167847,
      "mean_token_accuracy": 0.939128577709198,
      "num_tokens": 2510125.0,
      "step": 140
    },
    {
      "entropy": 1.2332049250602721,
      "epoch": 0.2018842530282638,
      "grad_norm": 1.4108549356460571,
      "learning_rate": 4.005376344086022e-05,
      "loss": 0.16094473600387574,
      "mean_token_accuracy": 0.936911940574646,
      "num_tokens": 2690100.0,
      "step": 150
    },
    {
      "entropy": 1.2368563890457154,
      "epoch": 0.21534320323014805,
      "grad_norm": 1.2720197439193726,
      "learning_rate": 4.2741935483870973e-05,
      "loss": 0.15074102878570556,
      "mean_token_accuracy": 0.9419298887252807,
      "num_tokens": 2869072.0,
      "step": 160
    },
    {
      "entropy": 1.2400941371917724,
      "epoch": 0.2288021534320323,
      "grad_norm": 1.1434999704360962,
      "learning_rate": 4.543010752688172e-05,
      "loss": 0.15364220142364501,
      "mean_token_accuracy": 0.9395892560482025,
      "num_tokens": 3048179.0,
      "step": 170
    },
    {
      "entropy": 1.2370286583900452,
      "epoch": 0.24226110363391656,
      "grad_norm": 1.1543998718261719,
      "learning_rate": 4.811827956989248e-05,
      "loss": 0.1518303394317627,
      "mean_token_accuracy": 0.9403793513774872,
      "num_tokens": 3227648.0,
      "step": 180
    },
    {
      "entropy": 1.2381223678588866,
      "epoch": 0.2557200538358008,
      "grad_norm": 1.2674320936203003,
      "learning_rate": 5.080645161290323e-05,
      "loss": 0.14720194339752196,
      "mean_token_accuracy": 0.9413308918476104,
      "num_tokens": 3406443.0,
      "step": 190
    },
    {
      "entropy": 1.2209740161895752,
      "epoch": 0.2691790040376851,
      "grad_norm": 1.3895965814590454,
      "learning_rate": 5.349462365591398e-05,
      "loss": 0.14467229843139648,
      "mean_token_accuracy": 0.9425995826721192,
      "num_tokens": 3585139.0,
      "step": 200
    },
    {
      "entropy": 1.202421200275421,
      "epoch": 0.28263795423956933,
      "grad_norm": 1.1748970746994019,
      "learning_rate": 5.618279569892473e-05,
      "loss": 0.14418480396270753,
      "mean_token_accuracy": 0.944035142660141,
      "num_tokens": 3764815.0,
      "step": 210
    },
    {
      "entropy": 1.2019111514091492,
      "epoch": 0.2960969044414536,
      "grad_norm": 1.498073935508728,
      "learning_rate": 5.887096774193549e-05,
      "loss": 0.1409894585609436,
      "mean_token_accuracy": 0.9451915919780731,
      "num_tokens": 3944066.0,
      "step": 220
    },
    {
      "entropy": 1.208519995212555,
      "epoch": 0.30955585464333785,
      "grad_norm": 1.3500250577926636,
      "learning_rate": 6.155913978494624e-05,
      "loss": 0.14032076597213744,
      "mean_token_accuracy": 0.9437436699867249,
      "num_tokens": 4122561.0,
      "step": 230
    },
    {
      "entropy": 1.2071591854095458,
      "epoch": 0.32301480484522205,
      "grad_norm": 1.2740904092788696,
      "learning_rate": 6.4247311827957e-05,
      "loss": 0.13652472496032714,
      "mean_token_accuracy": 0.9458230376243592,
      "num_tokens": 4301558.0,
      "step": 240
    },
    {
      "entropy": 1.2087280392646789,
      "epoch": 0.3364737550471063,
      "grad_norm": 1.2498955726623535,
      "learning_rate": 6.693548387096774e-05,
      "loss": 0.137939453125,
      "mean_token_accuracy": 0.9447133004665375,
      "num_tokens": 4480708.0,
      "step": 250
    },
    {
      "entropy": 1.1996983647346497,
      "epoch": 0.34993270524899056,
      "grad_norm": 1.4520384073257446,
      "learning_rate": 6.962365591397851e-05,
      "loss": 0.14351414442062377,
      "mean_token_accuracy": 0.9436001539230346,
      "num_tokens": 4659616.0,
      "step": 260
    },
    {
      "entropy": 1.195546042919159,
      "epoch": 0.3633916554508748,
      "grad_norm": 1.1850366592407227,
      "learning_rate": 7.231182795698926e-05,
      "loss": 0.13713514804840088,
      "mean_token_accuracy": 0.946206146478653,
      "num_tokens": 4838619.0,
      "step": 270
    },
    {
      "entropy": 1.1904785871505736,
      "epoch": 0.3768506056527591,
      "grad_norm": 1.4309216737747192,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.13457468748092652,
      "mean_token_accuracy": 0.9473767280578613,
      "num_tokens": 5017668.0,
      "step": 280
    },
    {
      "entropy": 1.1928183078765868,
      "epoch": 0.39030955585464333,
      "grad_norm": 1.1399856805801392,
      "learning_rate": 7.768817204301076e-05,
      "loss": 0.13286283016204833,
      "mean_token_accuracy": 0.9468790471553803,
      "num_tokens": 5197412.0,
      "step": 290
    },
    {
      "entropy": 1.1921735286712647,
      "epoch": 0.4037685060565276,
      "grad_norm": 1.3263301849365234,
      "learning_rate": 8.037634408602151e-05,
      "loss": 0.1312130570411682,
      "mean_token_accuracy": 0.9481729447841645,
      "num_tokens": 5377026.0,
      "step": 300
    },
    {
      "entropy": 1.1928237080574036,
      "epoch": 0.41722745625841184,
      "grad_norm": 1.6233983039855957,
      "learning_rate": 8.306451612903227e-05,
      "loss": 0.1323886036872864,
      "mean_token_accuracy": 0.9465976357460022,
      "num_tokens": 5556708.0,
      "step": 310
    },
    {
      "entropy": 1.1836108446121216,
      "epoch": 0.4306864064602961,
      "grad_norm": 1.3058593273162842,
      "learning_rate": 8.575268817204302e-05,
      "loss": 0.13815442323684693,
      "mean_token_accuracy": 0.9450344681739807,
      "num_tokens": 5735495.0,
      "step": 320
    },
    {
      "entropy": 1.179207694530487,
      "epoch": 0.44414535666218036,
      "grad_norm": 1.43180513381958,
      "learning_rate": 8.844086021505377e-05,
      "loss": 0.13185629844665528,
      "mean_token_accuracy": 0.9473332643508912,
      "num_tokens": 5914707.0,
      "step": 330
    },
    {
      "entropy": 1.1725443840026855,
      "epoch": 0.4576043068640646,
      "grad_norm": 1.391575813293457,
      "learning_rate": 9.112903225806452e-05,
      "loss": 0.13046096563339232,
      "mean_token_accuracy": 0.9486047685146332,
      "num_tokens": 6093788.0,
      "step": 340
    },
    {
      "entropy": 1.17471581697464,
      "epoch": 0.47106325706594887,
      "grad_norm": 1.5146636962890625,
      "learning_rate": 9.381720430107528e-05,
      "loss": 0.13146305084228516,
      "mean_token_accuracy": 0.9477684259414673,
      "num_tokens": 6272916.0,
      "step": 350
    },
    {
      "entropy": 1.1633504509925843,
      "epoch": 0.4845222072678331,
      "grad_norm": 1.1611223220825195,
      "learning_rate": 9.650537634408603e-05,
      "loss": 0.12906880378723146,
      "mean_token_accuracy": 0.9491180777549744,
      "num_tokens": 6452766.0,
      "step": 360
    },
    {
      "entropy": 1.1688304781913756,
      "epoch": 0.4979811574697174,
      "grad_norm": 1.5618613958358765,
      "learning_rate": 9.919354838709678e-05,
      "loss": 0.1296112060546875,
      "mean_token_accuracy": 0.9494616508483886,
      "num_tokens": 6632087.0,
      "step": 370
    },
    {
      "entropy": 1.1618135571479797,
      "epoch": 0.5114401076716016,
      "grad_norm": 1.259940266609192,
      "learning_rate": 9.999975729865971e-05,
      "loss": 0.12792425155639647,
      "mean_token_accuracy": 0.9487492978572846,
      "num_tokens": 6811480.0,
      "step": 380
    },
    {
      "entropy": 1.1698922395706177,
      "epoch": 0.5248990578734859,
      "grad_norm": 1.4075767993927002,
      "learning_rate": 9.999856856307314e-05,
      "loss": 0.12948644161224365,
      "mean_token_accuracy": 0.9471172571182251,
      "num_tokens": 6990914.0,
      "step": 390
    },
    {
      "entropy": 1.1714325428009034,
      "epoch": 0.5383580080753702,
      "grad_norm": 1.2077672481536865,
      "learning_rate": 9.999638923896533e-05,
      "loss": 0.12964634895324706,
      "mean_token_accuracy": 0.9472773969173431,
      "num_tokens": 7170193.0,
      "step": 400
    },
    {
      "entropy": 1.172813320159912,
      "epoch": 0.5518169582772544,
      "grad_norm": 1.506191611289978,
      "learning_rate": 9.999321936951374e-05,
      "loss": 0.12280762195587158,
      "mean_token_accuracy": 0.9517758071422577,
      "num_tokens": 7349496.0,
      "step": 410
    },
    {
      "entropy": 1.179931104183197,
      "epoch": 0.5652759084791387,
      "grad_norm": 1.2185982465744019,
      "learning_rate": 9.998905901752091e-05,
      "loss": 0.12760610580444337,
      "mean_token_accuracy": 0.948979276418686,
      "num_tokens": 7528648.0,
      "step": 420
    },
    {
      "entropy": 1.1817825198173524,
      "epoch": 0.5787348586810229,
      "grad_norm": 1.285261631011963,
      "learning_rate": 9.998390826541315e-05,
      "loss": 0.13066411018371582,
      "mean_token_accuracy": 0.947555410861969,
      "num_tokens": 7707487.0,
      "step": 430
    },
    {
      "entropy": 1.1945344924926757,
      "epoch": 0.5921938088829072,
      "grad_norm": 1.2778966426849365,
      "learning_rate": 9.997776721523888e-05,
      "loss": 0.13003890514373778,
      "mean_token_accuracy": 0.9476695477962493,
      "num_tokens": 7886452.0,
      "step": 440
    },
    {
      "entropy": 1.192435622215271,
      "epoch": 0.6056527590847914,
      "grad_norm": 1.2676047086715698,
      "learning_rate": 9.99706359886667e-05,
      "loss": 0.13059219121932983,
      "mean_token_accuracy": 0.9467391848564148,
      "num_tokens": 8065093.0,
      "step": 450
    },
    {
      "entropy": 1.1952194094657898,
      "epoch": 0.6191117092866757,
      "grad_norm": 1.1667490005493164,
      "learning_rate": 9.996251472698281e-05,
      "loss": 0.1308892250061035,
      "mean_token_accuracy": 0.9474103152751923,
      "num_tokens": 8245294.0,
      "step": 460
    },
    {
      "entropy": 1.1830523014068604,
      "epoch": 0.6325706594885598,
      "grad_norm": 1.4168891906738281,
      "learning_rate": 9.995340359108844e-05,
      "loss": 0.1230043888092041,
      "mean_token_accuracy": 0.9503998339176178,
      "num_tokens": 8424334.0,
      "step": 470
    },
    {
      "entropy": 1.1928761839866637,
      "epoch": 0.6460296096904441,
      "grad_norm": 1.4041002988815308,
      "learning_rate": 9.994330276149649e-05,
      "loss": 0.12544957399368287,
      "mean_token_accuracy": 0.9496485233306885,
      "num_tokens": 8603758.0,
      "step": 480
    },
    {
      "entropy": 1.180042278766632,
      "epoch": 0.6594885598923284,
      "grad_norm": 1.2816451787948608,
      "learning_rate": 9.993221243832797e-05,
      "loss": 0.1197009801864624,
      "mean_token_accuracy": 0.9527871966361999,
      "num_tokens": 8782936.0,
      "step": 490
    },
    {
      "entropy": 1.1822824954986573,
      "epoch": 0.6729475100942126,
      "grad_norm": 1.0554879903793335,
      "learning_rate": 9.992013284130816e-05,
      "loss": 0.12416183948516846,
      "mean_token_accuracy": 0.9488660097122192,
      "num_tokens": 8962286.0,
      "step": 500
    },
    {
      "entropy": 1.1877296924591065,
      "epoch": 0.6864064602960969,
      "grad_norm": 2.272390842437744,
      "learning_rate": 9.990706420976206e-05,
      "loss": 0.12660024166107178,
      "mean_token_accuracy": 0.9495710134506226,
      "num_tokens": 9141244.0,
      "step": 510
    },
    {
      "entropy": 1.195984995365143,
      "epoch": 0.6998654104979811,
      "grad_norm": 1.106213927268982,
      "learning_rate": 9.989300680260985e-05,
      "loss": 0.12362114191055298,
      "mean_token_accuracy": 0.9512333691120147,
      "num_tokens": 9319796.0,
      "step": 520
    },
    {
      "entropy": 1.1847575902938843,
      "epoch": 0.7133243606998654,
      "grad_norm": 1.2385672330856323,
      "learning_rate": 9.98779608983616e-05,
      "loss": 0.12053300142288208,
      "mean_token_accuracy": 0.9517379641532898,
      "num_tokens": 9498637.0,
      "step": 530
    },
    {
      "entropy": 1.1887083888053893,
      "epoch": 0.7267833109017496,
      "grad_norm": 1.1992591619491577,
      "learning_rate": 9.986192679511189e-05,
      "loss": 0.12146525382995606,
      "mean_token_accuracy": 0.9509431838989257,
      "num_tokens": 9678136.0,
      "step": 540
    },
    {
      "entropy": 1.1956284999847413,
      "epoch": 0.7402422611036339,
      "grad_norm": 1.4346206188201904,
      "learning_rate": 9.984490481053372e-05,
      "loss": 0.12582865953445435,
      "mean_token_accuracy": 0.9495353937149048,
      "num_tokens": 9856943.0,
      "step": 550
    },
    {
      "entropy": 1.183466374874115,
      "epoch": 0.7537012113055181,
      "grad_norm": 1.1207399368286133,
      "learning_rate": 9.982689528187244e-05,
      "loss": 0.11938930749893188,
      "mean_token_accuracy": 0.9524446070194245,
      "num_tokens": 10036305.0,
      "step": 560
    },
    {
      "entropy": 1.1996586084365846,
      "epoch": 0.7671601615074024,
      "grad_norm": 1.0004290342330933,
      "learning_rate": 9.98078985659389e-05,
      "loss": 0.1256342649459839,
      "mean_token_accuracy": 0.9488184452056885,
      "num_tokens": 10215049.0,
      "step": 570
    },
    {
      "entropy": 1.2113700032234191,
      "epoch": 0.7806191117092867,
      "grad_norm": 1.243759036064148,
      "learning_rate": 9.978791503910246e-05,
      "loss": 0.11844713687896728,
      "mean_token_accuracy": 0.9520347356796265,
      "num_tokens": 10393513.0,
      "step": 580
    },
    {
      "entropy": 1.2171649575233459,
      "epoch": 0.7940780619111709,
      "grad_norm": 1.2175841331481934,
      "learning_rate": 9.97669450972835e-05,
      "loss": 0.1155052900314331,
      "mean_token_accuracy": 0.954187548160553,
      "num_tokens": 10572502.0,
      "step": 590
    },
    {
      "entropy": 1.2295325994491577,
      "epoch": 0.8075370121130552,
      "grad_norm": 1.1670854091644287,
      "learning_rate": 9.974498915594557e-05,
      "loss": 0.12255362272262574,
      "mean_token_accuracy": 0.9510588347911835,
      "num_tokens": 10751857.0,
      "step": 600
    },
    {
      "entropy": 1.2220084905624389,
      "epoch": 0.8209959623149394,
      "grad_norm": 1.3236212730407715,
      "learning_rate": 9.97220476500872e-05,
      "loss": 0.1217005968093872,
      "mean_token_accuracy": 0.9508337616920471,
      "num_tokens": 10931362.0,
      "step": 610
    },
    {
      "entropy": 1.20922110080719,
      "epoch": 0.8344549125168237,
      "grad_norm": 1.2529112100601196,
      "learning_rate": 9.969812103423325e-05,
      "loss": 0.11833038330078124,
      "mean_token_accuracy": 0.9529603838920593,
      "num_tokens": 11111075.0,
      "step": 620
    },
    {
      "entropy": 1.2208962082862853,
      "epoch": 0.847913862718708,
      "grad_norm": 1.2380986213684082,
      "learning_rate": 9.967320978242592e-05,
      "loss": 0.12019131183624268,
      "mean_token_accuracy": 0.9517916083335877,
      "num_tokens": 11289952.0,
      "step": 630
    },
    {
      "entropy": 1.206966769695282,
      "epoch": 0.8613728129205922,
      "grad_norm": 1.2476933002471924,
      "learning_rate": 9.964731438821533e-05,
      "loss": 0.11783044338226319,
      "mean_token_accuracy": 0.9523464858531951,
      "num_tokens": 11469661.0,
      "step": 640
    },
    {
      "entropy": 1.2062023997306823,
      "epoch": 0.8748317631224765,
      "grad_norm": 1.4155808687210083,
      "learning_rate": 9.962043536464978e-05,
      "loss": 0.12099127769470215,
      "mean_token_accuracy": 0.9519050180912018,
      "num_tokens": 11648570.0,
      "step": 650
    },
    {
      "entropy": 1.2050026655197144,
      "epoch": 0.8882907133243607,
      "grad_norm": 1.309507966041565,
      "learning_rate": 9.959257324426556e-05,
      "loss": 0.11565302610397339,
      "mean_token_accuracy": 0.9535071849822998,
      "num_tokens": 11827640.0,
      "step": 660
    },
    {
      "entropy": 1.2138132452964783,
      "epoch": 0.901749663526245,
      "grad_norm": 1.150227427482605,
      "learning_rate": 9.95637285790764e-05,
      "loss": 0.11565654277801514,
      "mean_token_accuracy": 0.9536015927791596,
      "num_tokens": 12006419.0,
      "step": 670
    },
    {
      "entropy": 1.2211383819580077,
      "epoch": 0.9152086137281292,
      "grad_norm": 1.3185595273971558,
      "learning_rate": 9.953390194056258e-05,
      "loss": 0.11686277389526367,
      "mean_token_accuracy": 0.9518564403057098,
      "num_tokens": 12184806.0,
      "step": 680
    },
    {
      "entropy": 1.233402180671692,
      "epoch": 0.9286675639300135,
      "grad_norm": 1.160781979560852,
      "learning_rate": 9.950309391965947e-05,
      "loss": 0.11723113059997559,
      "mean_token_accuracy": 0.9525671184062958,
      "num_tokens": 12363767.0,
      "step": 690
    },
    {
      "entropy": 1.2254271149635314,
      "epoch": 0.9421265141318977,
      "grad_norm": 1.0756208896636963,
      "learning_rate": 9.947130512674602e-05,
      "loss": 0.11969656944274902,
      "mean_token_accuracy": 0.9499428868293762,
      "num_tokens": 12542727.0,
      "step": 700
    },
    {
      "entropy": 1.2217535138130189,
      "epoch": 0.955585464333782,
      "grad_norm": 1.131346344947815,
      "learning_rate": 9.943853619163255e-05,
      "loss": 0.11605353355407715,
      "mean_token_accuracy": 0.9536243081092834,
      "num_tokens": 12721825.0,
      "step": 710
    },
    {
      "entropy": 1.2145210385322571,
      "epoch": 0.9690444145356663,
      "grad_norm": 1.0480105876922607,
      "learning_rate": 9.94047877635482e-05,
      "loss": 0.11278635263442993,
      "mean_token_accuracy": 0.9553675949573517,
      "num_tokens": 12902291.0,
      "step": 720
    },
    {
      "entropy": 1.2308586597442628,
      "epoch": 0.9825033647375505,
      "grad_norm": 1.1793105602264404,
      "learning_rate": 9.93700605111283e-05,
      "loss": 0.11050724983215332,
      "mean_token_accuracy": 0.9547911584377289,
      "num_tokens": 13082065.0,
      "step": 730
    },
    {
      "entropy": 1.2493423819541931,
      "epoch": 0.9959623149394348,
      "grad_norm": 1.289297103881836,
      "learning_rate": 9.933435512240084e-05,
      "loss": 0.11567041873931885,
      "mean_token_accuracy": 0.9526383280754089,
      "num_tokens": 13261041.0,
      "step": 740
    },
    {
      "epoch": 1.0,
      "eval_entropy": 1.229653848204643,
      "eval_loss": 0.11524093896150589,
      "eval_mean_token_accuracy": 0.9533016294430775,
      "eval_num_tokens": 13314766.0,
      "eval_runtime": 13.2833,
      "eval_samples_per_second": 376.412,
      "eval_steps_per_second": 11.819,
      "step": 743
    },
    {
      "entropy": 1.233021354675293,
      "epoch": 1.009421265141319,
      "grad_norm": 1.3314229249954224,
      "learning_rate": 9.929767230477305e-05,
      "loss": 0.10805211067199708,
      "mean_token_accuracy": 0.9576423108577728,
      "num_tokens": 13440438.0,
      "step": 750
    },
    {
      "entropy": 1.2277064085006715,
      "epoch": 1.0228802153432033,
      "grad_norm": 1.194751262664795,
      "learning_rate": 9.92600127850173e-05,
      "loss": 0.09916897416114807,
      "mean_token_accuracy": 0.9617054045200348,
      "num_tokens": 13619055.0,
      "step": 760
    },
    {
      "entropy": 1.2193793654441833,
      "epoch": 1.0363391655450875,
      "grad_norm": 1.3674660921096802,
      "learning_rate": 9.922137730925673e-05,
      "loss": 0.09446401596069336,
      "mean_token_accuracy": 0.9620481312274933,
      "num_tokens": 13798753.0,
      "step": 770
    },
    {
      "entropy": 1.2068825483322143,
      "epoch": 1.0497981157469718,
      "grad_norm": 1.175338625907898,
      "learning_rate": 9.918176664295041e-05,
      "loss": 0.09437270164489746,
      "mean_token_accuracy": 0.9626644790172577,
      "num_tokens": 13978445.0,
      "step": 780
    },
    {
      "entropy": 1.2036906003952026,
      "epoch": 1.063257065948856,
      "grad_norm": 1.3463256359100342,
      "learning_rate": 9.914118157087824e-05,
      "loss": 0.09322788715362548,
      "mean_token_accuracy": 0.9640586376190186,
      "num_tokens": 14157400.0,
      "step": 790
    },
    {
      "entropy": 1.195313036441803,
      "epoch": 1.0767160161507403,
      "grad_norm": 1.2407623529434204,
      "learning_rate": 9.909962289712538e-05,
      "loss": 0.10000712871551513,
      "mean_token_accuracy": 0.9604595184326172,
      "num_tokens": 14337209.0,
      "step": 800
    },
    {
      "entropy": 1.1997171878814696,
      "epoch": 1.0901749663526246,
      "grad_norm": 1.4212044477462769,
      "learning_rate": 9.905709144506629e-05,
      "loss": 0.09967402815818786,
      "mean_token_accuracy": 0.9616046726703644,
      "num_tokens": 14516327.0,
      "step": 810
    },
    {
      "entropy": 1.1916023015975952,
      "epoch": 1.1036339165545088,
      "grad_norm": 1.296561598777771,
      "learning_rate": 9.901358805734846e-05,
      "loss": 0.09139133095741273,
      "mean_token_accuracy": 0.9634343802928924,
      "num_tokens": 14695257.0,
      "step": 820
    },
    {
      "entropy": 1.1924808621406555,
      "epoch": 1.117092866756393,
      "grad_norm": 1.3003162145614624,
      "learning_rate": 9.89691135958757e-05,
      "loss": 0.0935364007949829,
      "mean_token_accuracy": 0.9624580383300781,
      "num_tokens": 14874609.0,
      "step": 830
    },
    {
      "entropy": 1.2110714197158814,
      "epoch": 1.1305518169582773,
      "grad_norm": 1.341585636138916,
      "learning_rate": 9.892366894179105e-05,
      "loss": 0.09882450699806214,
      "mean_token_accuracy": 0.961030250787735,
      "num_tokens": 15053971.0,
      "step": 840
    },
    {
      "entropy": 1.2037933468818665,
      "epoch": 1.1440107671601616,
      "grad_norm": 1.4529608488082886,
      "learning_rate": 9.887725499545937e-05,
      "loss": 0.09266124367713928,
      "mean_token_accuracy": 0.9641264617443085,
      "num_tokens": 15233217.0,
      "step": 850
    },
    {
      "entropy": 1.210195577144623,
      "epoch": 1.1574697173620458,
      "grad_norm": 0.9631540179252625,
      "learning_rate": 9.882987267644939e-05,
      "loss": 0.09560335874557495,
      "mean_token_accuracy": 0.9616110920906067,
      "num_tokens": 15412460.0,
      "step": 860
    },
    {
      "entropy": 1.210800564289093,
      "epoch": 1.17092866756393,
      "grad_norm": 1.078429937362671,
      "learning_rate": 9.878152292351563e-05,
      "loss": 0.0967819094657898,
      "mean_token_accuracy": 0.960888934135437,
      "num_tokens": 15590882.0,
      "step": 870
    },
    {
      "entropy": 1.204759907722473,
      "epoch": 1.1843876177658144,
      "grad_norm": 1.132325530052185,
      "learning_rate": 9.873220669457975e-05,
      "loss": 0.09479628801345825,
      "mean_token_accuracy": 0.9629071593284607,
      "num_tokens": 15770658.0,
      "step": 880
    },
    {
      "entropy": 1.2071414232254027,
      "epoch": 1.1978465679676986,
      "grad_norm": 1.0902807712554932,
      "learning_rate": 9.868192496671147e-05,
      "loss": 0.09629296064376831,
      "mean_token_accuracy": 0.9622378885746002,
      "num_tokens": 15950126.0,
      "step": 890
    },
    {
      "entropy": 1.2057334661483765,
      "epoch": 1.2113055181695827,
      "grad_norm": 1.2059770822525024,
      "learning_rate": 9.86306787361094e-05,
      "loss": 0.09764755368232728,
      "mean_token_accuracy": 0.962404465675354,
      "num_tokens": 16129315.0,
      "step": 900
    },
    {
      "entropy": 1.2059934735298157,
      "epoch": 1.224764468371467,
      "grad_norm": 1.5481969118118286,
      "learning_rate": 9.857846901808117e-05,
      "loss": 0.09670655727386475,
      "mean_token_accuracy": 0.9619201004505158,
      "num_tokens": 16307839.0,
      "step": 910
    },
    {
      "entropy": 1.2136071562767028,
      "epoch": 1.2382234185733512,
      "grad_norm": 1.1171293258666992,
      "learning_rate": 9.852529684702329e-05,
      "loss": 0.09502402544021607,
      "mean_token_accuracy": 0.9619876623153687,
      "num_tokens": 16487021.0,
      "step": 920
    },
    {
      "entropy": 1.232442605495453,
      "epoch": 1.2516823687752354,
      "grad_norm": 1.2836637496948242,
      "learning_rate": 9.847116327640082e-05,
      "loss": 0.09930729866027832,
      "mean_token_accuracy": 0.9604007601737976,
      "num_tokens": 16665995.0,
      "step": 930
    },
    {
      "entropy": 1.2321829080581665,
      "epoch": 1.2651413189771197,
      "grad_norm": 1.1843444108963013,
      "learning_rate": 9.841606937872632e-05,
      "loss": 0.10086537599563598,
      "mean_token_accuracy": 0.9602800250053406,
      "num_tokens": 16845090.0,
      "step": 940
    },
    {
      "entropy": 1.233613657951355,
      "epoch": 1.278600269179004,
      "grad_norm": 1.3496166467666626,
      "learning_rate": 9.836001624553869e-05,
      "loss": 0.09795907735824586,
      "mean_token_accuracy": 0.9610718429088593,
      "num_tokens": 17024295.0,
      "step": 950
    },
    {
      "entropy": 1.2220023155212403,
      "epoch": 1.2920592193808882,
      "grad_norm": 1.238175392150879,
      "learning_rate": 9.830300498738152e-05,
      "loss": 0.09709340333938599,
      "mean_token_accuracy": 0.9621975898742676,
      "num_tokens": 17203525.0,
      "step": 960
    },
    {
      "entropy": 1.2186771392822267,
      "epoch": 1.3055181695827724,
      "grad_norm": 1.0820763111114502,
      "learning_rate": 9.824503673378112e-05,
      "loss": 0.09260507822036743,
      "mean_token_accuracy": 0.9632427036762238,
      "num_tokens": 17382269.0,
      "step": 970
    },
    {
      "entropy": 1.212946391105652,
      "epoch": 1.3189771197846567,
      "grad_norm": 1.253194808959961,
      "learning_rate": 9.81861126332241e-05,
      "loss": 0.10012803077697754,
      "mean_token_accuracy": 0.9596389472484589,
      "num_tokens": 17561951.0,
      "step": 980
    },
    {
      "entropy": 1.2092903971672058,
      "epoch": 1.332436069986541,
      "grad_norm": 1.6471713781356812,
      "learning_rate": 9.812623385313461e-05,
      "loss": 0.1032632827758789,
      "mean_token_accuracy": 0.9594815850257874,
      "num_tokens": 17741116.0,
      "step": 990
    },
    {
      "entropy": 1.2158336997032166,
      "epoch": 1.3458950201884252,
      "grad_norm": 1.076393723487854,
      "learning_rate": 9.806540157985131e-05,
      "loss": 0.09857285022735596,
      "mean_token_accuracy": 0.9608540177345276,
      "num_tokens": 17920249.0,
      "step": 1000
    },
    {
      "entropy": 1.2093246698379516,
      "epoch": 1.3593539703903095,
      "grad_norm": 1.1203004121780396,
      "learning_rate": 9.800361701860368e-05,
      "loss": 0.09807900190353394,
      "mean_token_accuracy": 0.9611685931682586,
      "num_tokens": 18099006.0,
      "step": 1010
    },
    {
      "entropy": 1.2070690989494324,
      "epoch": 1.3728129205921937,
      "grad_norm": 1.3285764455795288,
      "learning_rate": 9.794088139348835e-05,
      "loss": 0.10283086299896241,
      "mean_token_accuracy": 0.9585156977176666,
      "num_tokens": 18277971.0,
      "step": 1020
    },
    {
      "entropy": 1.2022451281547546,
      "epoch": 1.386271870794078,
      "grad_norm": 1.0949617624282837,
      "learning_rate": 9.787719594744468e-05,
      "loss": 0.10161725282669068,
      "mean_token_accuracy": 0.9598902583122253,
      "num_tokens": 18457464.0,
      "step": 1030
    },
    {
      "entropy": 1.2045769929885863,
      "epoch": 1.3997308209959622,
      "grad_norm": 1.008150577545166,
      "learning_rate": 9.781256194223023e-05,
      "loss": 0.10038440227508545,
      "mean_token_accuracy": 0.960367614030838,
      "num_tokens": 18636876.0,
      "step": 1040
    },
    {
      "entropy": 1.204549217224121,
      "epoch": 1.4131897711978465,
      "grad_norm": 1.0495935678482056,
      "learning_rate": 9.774698065839577e-05,
      "loss": 0.09564157128334046,
      "mean_token_accuracy": 0.9625212967395782,
      "num_tokens": 18816243.0,
      "step": 1050
    },
    {
      "entropy": 1.2045063614845275,
      "epoch": 1.4266487213997308,
      "grad_norm": 1.2372835874557495,
      "learning_rate": 9.768045339525979e-05,
      "loss": 0.09781360626220703,
      "mean_token_accuracy": 0.9605839848518372,
      "num_tokens": 18995594.0,
      "step": 1060
    },
    {
      "entropy": 1.2258678078651428,
      "epoch": 1.440107671601615,
      "grad_norm": 1.0772687196731567,
      "learning_rate": 9.76129814708829e-05,
      "loss": 0.09291026592254639,
      "mean_token_accuracy": 0.9634248733520507,
      "num_tokens": 19173887.0,
      "step": 1070
    },
    {
      "entropy": 1.2230794191360475,
      "epoch": 1.4535666218034993,
      "grad_norm": 1.2008293867111206,
      "learning_rate": 9.754456622204167e-05,
      "loss": 0.09285001754760742,
      "mean_token_accuracy": 0.9633622407913208,
      "num_tokens": 19352678.0,
      "step": 1080
    },
    {
      "entropy": 1.2313218355178832,
      "epoch": 1.4670255720053835,
      "grad_norm": 1.5826188325881958,
      "learning_rate": 9.747520900420209e-05,
      "loss": 0.1002782940864563,
      "mean_token_accuracy": 0.9600823521614075,
      "num_tokens": 19532077.0,
      "step": 1090
    },
    {
      "entropy": 1.2246542692184448,
      "epoch": 1.4804845222072678,
      "grad_norm": 1.3970143795013428,
      "learning_rate": 9.740491119149277e-05,
      "loss": 0.1005969524383545,
      "mean_token_accuracy": 0.9596368432044983,
      "num_tokens": 19710609.0,
      "step": 1100
    },
    {
      "entropy": 1.207058048248291,
      "epoch": 1.493943472409152,
      "grad_norm": 1.3544780015945435,
      "learning_rate": 9.733367417667773e-05,
      "loss": 0.09367164373397827,
      "mean_token_accuracy": 0.9632523238658905,
      "num_tokens": 19889820.0,
      "step": 1110
    },
    {
      "entropy": 1.2027259588241577,
      "epoch": 1.5074024226110363,
      "grad_norm": 1.2393465042114258,
      "learning_rate": 9.726149937112873e-05,
      "loss": 0.09854428172111511,
      "mean_token_accuracy": 0.9612930059432984,
      "num_tokens": 20069561.0,
      "step": 1120
    },
    {
      "entropy": 1.2199820518493651,
      "epoch": 1.5208613728129206,
      "grad_norm": 1.4061861038208008,
      "learning_rate": 9.718838820479743e-05,
      "loss": 0.09687533378601074,
      "mean_token_accuracy": 0.9612306416034698,
      "num_tokens": 20249088.0,
      "step": 1130
    },
    {
      "entropy": 1.2114709615707397,
      "epoch": 1.5343203230148048,
      "grad_norm": 1.2970331907272339,
      "learning_rate": 9.711434212618691e-05,
      "loss": 0.09762253165245056,
      "mean_token_accuracy": 0.9609376013278961,
      "num_tokens": 20428600.0,
      "step": 1140
    },
    {
      "entropy": 1.1982413172721862,
      "epoch": 1.547779273216689,
      "grad_norm": 1.621308445930481,
      "learning_rate": 9.703936260232308e-05,
      "loss": 0.09679374098777771,
      "mean_token_accuracy": 0.9625207364559174,
      "num_tokens": 20608047.0,
      "step": 1150
    },
    {
      "entropy": 1.195889377593994,
      "epoch": 1.5612382234185733,
      "grad_norm": 1.2940045595169067,
      "learning_rate": 9.696345111872557e-05,
      "loss": 0.09699609279632568,
      "mean_token_accuracy": 0.96190345287323,
      "num_tokens": 20787142.0,
      "step": 1160
    },
    {
      "entropy": 1.1944554448127747,
      "epoch": 1.5746971736204576,
      "grad_norm": 1.3155335187911987,
      "learning_rate": 9.688660917937838e-05,
      "loss": 0.09831242561340332,
      "mean_token_accuracy": 0.9606768429279328,
      "num_tokens": 20966230.0,
      "step": 1170
    },
    {
      "entropy": 1.1957345604896545,
      "epoch": 1.5881561238223418,
      "grad_norm": 1.2948030233383179,
      "learning_rate": 9.68088383066999e-05,
      "loss": 0.09834452867507934,
      "mean_token_accuracy": 0.9612827241420746,
      "num_tokens": 21145768.0,
      "step": 1180
    },
    {
      "entropy": 1.2020023703575133,
      "epoch": 1.601615074024226,
      "grad_norm": 1.0523329973220825,
      "learning_rate": 9.673014004151292e-05,
      "loss": 0.09663949012756348,
      "mean_token_accuracy": 0.9620199799537659,
      "num_tokens": 21324592.0,
      "step": 1190
    },
    {
      "entropy": 1.1892358779907226,
      "epoch": 1.6150740242261103,
      "grad_norm": 1.1584330797195435,
      "learning_rate": 9.665051594301407e-05,
      "loss": 0.09669581055641174,
      "mean_token_accuracy": 0.961614978313446,
      "num_tokens": 21504539.0,
      "step": 1200
    },
    {
      "entropy": 1.190696406364441,
      "epoch": 1.6285329744279946,
      "grad_norm": 1.1194695234298706,
      "learning_rate": 9.656996758874284e-05,
      "loss": 0.09648081660270691,
      "mean_token_accuracy": 0.9612169206142426,
      "num_tokens": 21683905.0,
      "step": 1210
    },
    {
      "entropy": 1.2076977849006654,
      "epoch": 1.6419919246298789,
      "grad_norm": 1.1297376155853271,
      "learning_rate": 9.648849657455044e-05,
      "loss": 0.09605686664581299,
      "mean_token_accuracy": 0.961658376455307,
      "num_tokens": 21862162.0,
      "step": 1220
    },
    {
      "entropy": 1.2096962213516236,
      "epoch": 1.6554508748317631,
      "grad_norm": 1.2401906251907349,
      "learning_rate": 9.640610451456811e-05,
      "loss": 0.09206328392028809,
      "mean_token_accuracy": 0.962989890575409,
      "num_tokens": 22041015.0,
      "step": 1230
    },
    {
      "entropy": 1.216509222984314,
      "epoch": 1.6689098250336474,
      "grad_norm": 1.2637176513671875,
      "learning_rate": 9.632279304117517e-05,
      "loss": 0.09614999294281006,
      "mean_token_accuracy": 0.9613571405410767,
      "num_tokens": 22220655.0,
      "step": 1240
    },
    {
      "entropy": 1.2126171827316283,
      "epoch": 1.6823687752355316,
      "grad_norm": 1.2879180908203125,
      "learning_rate": 9.623856380496664e-05,
      "loss": 0.09818092584609986,
      "mean_token_accuracy": 0.9603166699409484,
      "num_tokens": 22399957.0,
      "step": 1250
    },
    {
      "entropy": 1.182448434829712,
      "epoch": 1.695827725437416,
      "grad_norm": 1.0547544956207275,
      "learning_rate": 9.615341847472059e-05,
      "loss": 0.0945388674736023,
      "mean_token_accuracy": 0.9623521089553833,
      "num_tokens": 22579222.0,
      "step": 1260
    },
    {
      "entropy": 1.1886864185333252,
      "epoch": 1.7092866756393001,
      "grad_norm": 1.4119364023208618,
      "learning_rate": 9.606735873736505e-05,
      "loss": 0.0979494333267212,
      "mean_token_accuracy": 0.9607987105846405,
      "num_tokens": 22758487.0,
      "step": 1270
    },
    {
      "entropy": 1.1918489813804627,
      "epoch": 1.7227456258411844,
      "grad_norm": 1.2551711797714233,
      "learning_rate": 9.598038629794461e-05,
      "loss": 0.09586712718009949,
      "mean_token_accuracy": 0.9615644454956055,
      "num_tokens": 22936708.0,
      "step": 1280
    },
    {
      "entropy": 1.1913212060928344,
      "epoch": 1.7362045760430687,
      "grad_norm": 1.0276069641113281,
      "learning_rate": 9.589250287958657e-05,
      "loss": 0.09535220861434937,
      "mean_token_accuracy": 0.9606883823871613,
      "num_tokens": 23116329.0,
      "step": 1290
    },
    {
      "entropy": 1.203085219860077,
      "epoch": 1.749663526244953,
      "grad_norm": 1.2456278800964355,
      "learning_rate": 9.580371022346693e-05,
      "loss": 0.09598281383514404,
      "mean_token_accuracy": 0.9608144044876099,
      "num_tokens": 23295164.0,
      "step": 1300
    },
    {
      "entropy": 1.1892922878265382,
      "epoch": 1.7631224764468372,
      "grad_norm": 1.1159876585006714,
      "learning_rate": 9.571401008877572e-05,
      "loss": 0.09096106886863708,
      "mean_token_accuracy": 0.9636982321739197,
      "num_tokens": 23474377.0,
      "step": 1310
    },
    {
      "entropy": 1.2096730828285218,
      "epoch": 1.7765814266487214,
      "grad_norm": 1.420886516571045,
      "learning_rate": 9.562340425268233e-05,
      "loss": 0.0925740659236908,
      "mean_token_accuracy": 0.9629011929035187,
      "num_tokens": 23653389.0,
      "step": 1320
    },
    {
      "entropy": 1.2122852802276611,
      "epoch": 1.7900403768506057,
      "grad_norm": 1.1587319374084473,
      "learning_rate": 9.553189451030019e-05,
      "loss": 0.09554123878479004,
      "mean_token_accuracy": 0.9622859060764313,
      "num_tokens": 23832469.0,
      "step": 1330
    },
    {
      "entropy": 1.2176487922668457,
      "epoch": 1.80349932705249,
      "grad_norm": 1.147444248199463,
      "learning_rate": 9.543948267465115e-05,
      "loss": 0.09707238674163818,
      "mean_token_accuracy": 0.9612141191959381,
      "num_tokens": 24011518.0,
      "step": 1340
    },
    {
      "entropy": 1.2231361389160156,
      "epoch": 1.8169582772543742,
      "grad_norm": 1.1775709390640259,
      "learning_rate": 9.534617057662977e-05,
      "loss": 0.09654755592346191,
      "mean_token_accuracy": 0.9617958247661591,
      "num_tokens": 24190267.0,
      "step": 1350
    },
    {
      "entropy": 1.2070120811462401,
      "epoch": 1.8304172274562585,
      "grad_norm": 1.1315947771072388,
      "learning_rate": 9.525196006496679e-05,
      "loss": 0.09382581114768981,
      "mean_token_accuracy": 0.9625270128250122,
      "num_tokens": 24369982.0,
      "step": 1360
    },
    {
      "entropy": 1.2057390093803406,
      "epoch": 1.8438761776581427,
      "grad_norm": 1.1973934173583984,
      "learning_rate": 9.515685300619271e-05,
      "loss": 0.09683746099472046,
      "mean_token_accuracy": 0.9607476830482483,
      "num_tokens": 24549256.0,
      "step": 1370
    },
    {
      "entropy": 1.207427191734314,
      "epoch": 1.857335127860027,
      "grad_norm": 1.3193334341049194,
      "learning_rate": 9.506085128460065e-05,
      "loss": 0.09461041688919067,
      "mean_token_accuracy": 0.9628551185131073,
      "num_tokens": 24727544.0,
      "step": 1380
    },
    {
      "entropy": 1.2011541604995728,
      "epoch": 1.8707940780619112,
      "grad_norm": 1.0681352615356445,
      "learning_rate": 9.496395680220918e-05,
      "loss": 0.0960330069065094,
      "mean_token_accuracy": 0.9622460305690765,
      "num_tokens": 24907721.0,
      "step": 1390
    },
    {
      "entropy": 1.2034499764442443,
      "epoch": 1.8842530282637955,
      "grad_norm": 1.2765101194381714,
      "learning_rate": 9.486617147872446e-05,
      "loss": 0.09376740455627441,
      "mean_token_accuracy": 0.9624415040016174,
      "num_tokens": 25086496.0,
      "step": 1400
    },
    {
      "entropy": 1.1899038195610045,
      "epoch": 1.8977119784656797,
      "grad_norm": 1.1333132982254028,
      "learning_rate": 9.476749725150235e-05,
      "loss": 0.09668049812316895,
      "mean_token_accuracy": 0.9621514558792115,
      "num_tokens": 25266204.0,
      "step": 1410
    },
    {
      "entropy": 1.1895189881324768,
      "epoch": 1.911170928667564,
      "grad_norm": 1.310587763786316,
      "learning_rate": 9.466793607550995e-05,
      "loss": 0.0920013129711151,
      "mean_token_accuracy": 0.963368022441864,
      "num_tokens": 25445604.0,
      "step": 1420
    },
    {
      "entropy": 1.1946001529693604,
      "epoch": 1.9246298788694483,
      "grad_norm": 1.3619959354400635,
      "learning_rate": 9.45674899232869e-05,
      "loss": 0.09906838536262512,
      "mean_token_accuracy": 0.9604476511478424,
      "num_tokens": 25624947.0,
      "step": 1430
    },
    {
      "entropy": 1.2078219771385192,
      "epoch": 1.9380888290713325,
      "grad_norm": 1.152220606803894,
      "learning_rate": 9.446616078490626e-05,
      "loss": 0.09479650259017944,
      "mean_token_accuracy": 0.9627270400524139,
      "num_tokens": 25804643.0,
      "step": 1440
    },
    {
      "entropy": 1.2207356214523315,
      "epoch": 1.9515477792732168,
      "grad_norm": 1.186161994934082,
      "learning_rate": 9.436395066793518e-05,
      "loss": 0.09636704921722412,
      "mean_token_accuracy": 0.9604843854904175,
      "num_tokens": 25984119.0,
      "step": 1450
    },
    {
      "entropy": 1.205849301815033,
      "epoch": 1.965006729475101,
      "grad_norm": 1.409846544265747,
      "learning_rate": 9.426086159739496e-05,
      "loss": 0.09743249416351318,
      "mean_token_accuracy": 0.9608718931674958,
      "num_tokens": 26163483.0,
      "step": 1460
    },
    {
      "entropy": 1.2089114785194397,
      "epoch": 1.9784656796769853,
      "grad_norm": 1.2226805686950684,
      "learning_rate": 9.415689561572107e-05,
      "loss": 0.09131012558937072,
      "mean_token_accuracy": 0.9631811439990997,
      "num_tokens": 26342666.0,
      "step": 1470
    },
    {
      "entropy": 1.1953012466430664,
      "epoch": 1.9919246298788695,
      "grad_norm": 1.0700947046279907,
      "learning_rate": 9.405205478272267e-05,
      "loss": 0.09140577316284179,
      "mean_token_accuracy": 0.9642649590969086,
      "num_tokens": 26521895.0,
      "step": 1480
    },
    {
      "epoch": 2.0,
      "eval_entropy": 1.1869331590688912,
      "eval_loss": 0.10991495102643967,
      "eval_mean_token_accuracy": 0.9554494638351878,
      "eval_num_tokens": 26629596.0,
      "eval_runtime": 12.7631,
      "eval_samples_per_second": 391.753,
      "eval_steps_per_second": 12.301,
      "step": 1486
    },
    {
      "entropy": 1.1832876205444336,
      "epoch": 2.005383580080754,
      "grad_norm": 1.0044941902160645,
      "learning_rate": 9.394634117554173e-05,
      "loss": 0.0840892255306244,
      "mean_token_accuracy": 0.967725521326065,
      "num_tokens": 26701394.0,
      "step": 1490
    },
    {
      "entropy": 1.159238350391388,
      "epoch": 2.018842530282638,
      "grad_norm": 1.4198471307754517,
      "learning_rate": 9.38397568886119e-05,
      "loss": 0.07137876152992248,
      "mean_token_accuracy": 0.9723187386989594,
      "num_tokens": 26880890.0,
      "step": 1500
    },
    {
      "entropy": 1.165151631832123,
      "epoch": 2.0323014804845223,
      "grad_norm": 1.1602118015289307,
      "learning_rate": 9.373230403361712e-05,
      "loss": 0.06463043689727783,
      "mean_token_accuracy": 0.9757274091243744,
      "num_tokens": 27059741.0,
      "step": 1510
    },
    {
      "entropy": 1.1644548654556275,
      "epoch": 2.0457604306864066,
      "grad_norm": 1.3322592973709106,
      "learning_rate": 9.362398473944958e-05,
      "loss": 0.07388677597045898,
      "mean_token_accuracy": 0.971617478132248,
      "num_tokens": 27238125.0,
      "step": 1520
    },
    {
      "entropy": 1.1564043641090394,
      "epoch": 2.059219380888291,
      "grad_norm": 1.1690629720687866,
      "learning_rate": 9.35148011521677e-05,
      "loss": 0.06990204453468322,
      "mean_token_accuracy": 0.9719981133937836,
      "num_tokens": 27417172.0,
      "step": 1530
    },
    {
      "entropy": 1.1576861262321472,
      "epoch": 2.072678331090175,
      "grad_norm": 1.7016727924346924,
      "learning_rate": 9.340475543495364e-05,
      "loss": 0.06850625276565551,
      "mean_token_accuracy": 0.9732699453830719,
      "num_tokens": 27596848.0,
      "step": 1540
    },
    {
      "entropy": 1.1597527265548706,
      "epoch": 2.0861372812920593,
      "grad_norm": 1.1524600982666016,
      "learning_rate": 9.329384976807023e-05,
      "loss": 0.06980778574943543,
      "mean_token_accuracy": 0.9729204118251801,
      "num_tokens": 27775617.0,
      "step": 1550
    },
    {
      "entropy": 1.1575467109680175,
      "epoch": 2.0995962314939436,
      "grad_norm": 1.4498176574707031,
      "learning_rate": 9.318208634881802e-05,
      "loss": 0.07390267252922059,
      "mean_token_accuracy": 0.9713942348957062,
      "num_tokens": 27954133.0,
      "step": 1560
    },
    {
      "entropy": 1.1575489521026612,
      "epoch": 2.113055181695828,
      "grad_norm": 1.243706464767456,
      "learning_rate": 9.306946739149161e-05,
      "loss": 0.06798491477966309,
      "mean_token_accuracy": 0.973270720243454,
      "num_tokens": 28133292.0,
      "step": 1570
    },
    {
      "entropy": 1.1514037609100343,
      "epoch": 2.126514131897712,
      "grad_norm": 1.256933331489563,
      "learning_rate": 9.29559951273358e-05,
      "loss": 0.0749699592590332,
      "mean_token_accuracy": 0.9699350416660308,
      "num_tokens": 28312726.0,
      "step": 1580
    },
    {
      "entropy": 1.1511113524436951,
      "epoch": 2.1399730820995964,
      "grad_norm": 1.1914122104644775,
      "learning_rate": 9.284167180450141e-05,
      "loss": 0.06752681732177734,
      "mean_token_accuracy": 0.9743177771568299,
      "num_tokens": 28492614.0,
      "step": 1590
    },
    {
      "entropy": 1.1382261991500855,
      "epoch": 2.1534320323014806,
      "grad_norm": 1.109575867652893,
      "learning_rate": 9.272649968800069e-05,
      "loss": 0.06449686884880065,
      "mean_token_accuracy": 0.9755833566188812,
      "num_tokens": 28671719.0,
      "step": 1600
    },
    {
      "entropy": 1.1397038459777833,
      "epoch": 2.166890982503365,
      "grad_norm": 1.3151781558990479,
      "learning_rate": 9.26104810596625e-05,
      "loss": 0.07052424550056458,
      "mean_token_accuracy": 0.972499680519104,
      "num_tokens": 28851056.0,
      "step": 1610
    },
    {
      "entropy": 1.136834406852722,
      "epoch": 2.180349932705249,
      "grad_norm": 1.4410197734832764,
      "learning_rate": 9.249361821808708e-05,
      "loss": 0.06850321292877197,
      "mean_token_accuracy": 0.9728572845458985,
      "num_tokens": 29030750.0,
      "step": 1620
    },
    {
      "entropy": 1.140534520149231,
      "epoch": 2.1938088829071334,
      "grad_norm": 1.1493765115737915,
      "learning_rate": 9.237591347860052e-05,
      "loss": 0.06934296488761901,
      "mean_token_accuracy": 0.972960364818573,
      "num_tokens": 29210336.0,
      "step": 1630
    },
    {
      "entropy": 1.138876986503601,
      "epoch": 2.2072678331090176,
      "grad_norm": 1.0383925437927246,
      "learning_rate": 9.225736917320886e-05,
      "loss": 0.06877213716506958,
      "mean_token_accuracy": 0.9730811774730682,
      "num_tokens": 29389788.0,
      "step": 1640
    },
    {
      "entropy": 1.1449824213981628,
      "epoch": 2.220726783310902,
      "grad_norm": 1.374165654182434,
      "learning_rate": 9.213798765055187e-05,
      "loss": 0.07060860991477966,
      "mean_token_accuracy": 0.9721363008022308,
      "num_tokens": 29569111.0,
      "step": 1650
    },
    {
      "entropy": 1.143638014793396,
      "epoch": 2.234185733512786,
      "grad_norm": 1.0196412801742554,
      "learning_rate": 9.20177712758566e-05,
      "loss": 0.07119340896606445,
      "mean_token_accuracy": 0.9731849789619446,
      "num_tokens": 29747986.0,
      "step": 1660
    },
    {
      "entropy": 1.1349515676498414,
      "epoch": 2.2476446837146704,
      "grad_norm": 1.1247919797897339,
      "learning_rate": 9.189672243089046e-05,
      "loss": 0.07071832418441773,
      "mean_token_accuracy": 0.9731756567955017,
      "num_tokens": 29927276.0,
      "step": 1670
    },
    {
      "entropy": 1.1376260280609132,
      "epoch": 2.2611036339165547,
      "grad_norm": 1.4197320938110352,
      "learning_rate": 9.177484351391402e-05,
      "loss": 0.07115572690963745,
      "mean_token_accuracy": 0.9723047018051147,
      "num_tokens": 30106267.0,
      "step": 1680
    },
    {
      "entropy": 1.1282797813415528,
      "epoch": 2.274562584118439,
      "grad_norm": 1.0774035453796387,
      "learning_rate": 9.165213693963355e-05,
      "loss": 0.068689626455307,
      "mean_token_accuracy": 0.9729084491729736,
      "num_tokens": 30285658.0,
      "step": 1690
    },
    {
      "entropy": 1.140774166584015,
      "epoch": 2.288021534320323,
      "grad_norm": 1.5625728368759155,
      "learning_rate": 9.152860513915314e-05,
      "loss": 0.07172787189483643,
      "mean_token_accuracy": 0.9718713641166687,
      "num_tokens": 30464579.0,
      "step": 1700
    },
    {
      "entropy": 1.1416746616363525,
      "epoch": 2.3014804845222074,
      "grad_norm": 1.2159788608551025,
      "learning_rate": 9.140425055992648e-05,
      "loss": 0.07109695672988892,
      "mean_token_accuracy": 0.9723007261753083,
      "num_tokens": 30643566.0,
      "step": 1710
    },
    {
      "entropy": 1.137896478176117,
      "epoch": 2.3149394347240917,
      "grad_norm": 1.1671864986419678,
      "learning_rate": 9.127907566570853e-05,
      "loss": 0.07048168182373046,
      "mean_token_accuracy": 0.9725371599197388,
      "num_tokens": 30822593.0,
      "step": 1720
    },
    {
      "entropy": 1.121722447872162,
      "epoch": 2.328398384925976,
      "grad_norm": 1.3899428844451904,
      "learning_rate": 9.115308293650653e-05,
      "loss": 0.07030471563339233,
      "mean_token_accuracy": 0.972334086894989,
      "num_tokens": 31001872.0,
      "step": 1730
    },
    {
      "entropy": 1.1253960013389588,
      "epoch": 2.34185733512786,
      "grad_norm": 1.3878816366195679,
      "learning_rate": 9.102627486853099e-05,
      "loss": 0.06956568956375123,
      "mean_token_accuracy": 0.9728380262851715,
      "num_tokens": 31181325.0,
      "step": 1740
    },
    {
      "entropy": 1.1276877760887145,
      "epoch": 2.3553162853297445,
      "grad_norm": 1.057102084159851,
      "learning_rate": 9.089865397414614e-05,
      "loss": 0.07267707586288452,
      "mean_token_accuracy": 0.9716822624206543,
      "num_tokens": 31360101.0,
      "step": 1750
    },
    {
      "entropy": 1.1322511553764343,
      "epoch": 2.3687752355316287,
      "grad_norm": 0.9549157023429871,
      "learning_rate": 9.077022278182024e-05,
      "loss": 0.0700565218925476,
      "mean_token_accuracy": 0.9737613677978516,
      "num_tokens": 31539212.0,
      "step": 1760
    },
    {
      "entropy": 1.1198163986206056,
      "epoch": 2.382234185733513,
      "grad_norm": 1.2134599685668945,
      "learning_rate": 9.064098383607545e-05,
      "loss": 0.07131816148757934,
      "mean_token_accuracy": 0.9710030317306518,
      "num_tokens": 31718375.0,
      "step": 1770
    },
    {
      "entropy": 1.1219655275344849,
      "epoch": 2.3956931359353972,
      "grad_norm": 1.3444453477859497,
      "learning_rate": 9.051093969743738e-05,
      "loss": 0.06774230003356933,
      "mean_token_accuracy": 0.9737627685070038,
      "num_tokens": 31897547.0,
      "step": 1780
    },
    {
      "entropy": 1.112294065952301,
      "epoch": 2.409152086137281,
      "grad_norm": 1.5304001569747925,
      "learning_rate": 9.03800929423844e-05,
      "loss": 0.0724343478679657,
      "mean_token_accuracy": 0.9721882760524749,
      "num_tokens": 32076717.0,
      "step": 1790
    },
    {
      "entropy": 1.1336388826370238,
      "epoch": 2.4226110363391653,
      "grad_norm": 1.2315043210983276,
      "learning_rate": 9.024844616329662e-05,
      "loss": 0.07212550640106201,
      "mean_token_accuracy": 0.9726522386074066,
      "num_tokens": 32255927.0,
      "step": 1800
    },
    {
      "entropy": 1.1251046061515808,
      "epoch": 2.4360699865410496,
      "grad_norm": 1.3251651525497437,
      "learning_rate": 9.011600196840447e-05,
      "loss": 0.07009173035621644,
      "mean_token_accuracy": 0.9724473178386688,
      "num_tokens": 32434929.0,
      "step": 1810
    },
    {
      "entropy": 1.1328514456748962,
      "epoch": 2.449528936742934,
      "grad_norm": 1.2601144313812256,
      "learning_rate": 8.998276298173707e-05,
      "loss": 0.0719257116317749,
      "mean_token_accuracy": 0.9720721006393432,
      "num_tokens": 32614063.0,
      "step": 1820
    },
    {
      "entropy": 1.136410367488861,
      "epoch": 2.462987886944818,
      "grad_norm": 1.2086918354034424,
      "learning_rate": 8.984873184307017e-05,
      "loss": 0.07017306089401246,
      "mean_token_accuracy": 0.9722030460834503,
      "num_tokens": 32793068.0,
      "step": 1830
    },
    {
      "entropy": 1.13520849943161,
      "epoch": 2.4764468371467023,
      "grad_norm": 1.453801155090332,
      "learning_rate": 8.971391120787397e-05,
      "loss": 0.07180649638175965,
      "mean_token_accuracy": 0.9726110398769379,
      "num_tokens": 32972445.0,
      "step": 1840
    },
    {
      "entropy": 1.1437011241912842,
      "epoch": 2.4899057873485866,
      "grad_norm": 1.1886014938354492,
      "learning_rate": 8.957830374726042e-05,
      "loss": 0.07153818607330323,
      "mean_token_accuracy": 0.9720338463783265,
      "num_tokens": 33151976.0,
      "step": 1850
    },
    {
      "entropy": 1.1325899600982665,
      "epoch": 2.503364737550471,
      "grad_norm": 1.1960384845733643,
      "learning_rate": 8.944191214793028e-05,
      "loss": 0.06935594081878663,
      "mean_token_accuracy": 0.9729972183704376,
      "num_tokens": 33330611.0,
      "step": 1860
    },
    {
      "entropy": 1.127972149848938,
      "epoch": 2.516823687752355,
      "grad_norm": 1.1048696041107178,
      "learning_rate": 8.930473911212e-05,
      "loss": 0.07217252850532532,
      "mean_token_accuracy": 0.9718475580215454,
      "num_tokens": 33509614.0,
      "step": 1870
    },
    {
      "entropy": 1.1299184799194335,
      "epoch": 2.5302826379542394,
      "grad_norm": 1.6520979404449463,
      "learning_rate": 8.916678735754809e-05,
      "loss": 0.07317680716514588,
      "mean_token_accuracy": 0.971524566411972,
      "num_tokens": 33688724.0,
      "step": 1880
    },
    {
      "entropy": 1.1270474672317505,
      "epoch": 2.5437415881561236,
      "grad_norm": 1.1285676956176758,
      "learning_rate": 8.902805961736123e-05,
      "loss": 0.07085765600204467,
      "mean_token_accuracy": 0.9733552634716034,
      "num_tokens": 33868061.0,
      "step": 1890
    },
    {
      "entropy": 1.1217491984367371,
      "epoch": 2.557200538358008,
      "grad_norm": 1.2642406225204468,
      "learning_rate": 8.88885586400803e-05,
      "loss": 0.06978695392608643,
      "mean_token_accuracy": 0.9726768732070923,
      "num_tokens": 34047387.0,
      "step": 1900
    },
    {
      "entropy": 1.1316059112548829,
      "epoch": 2.570659488559892,
      "grad_norm": 1.3016549348831177,
      "learning_rate": 8.874828718954576e-05,
      "loss": 0.07102057337760925,
      "mean_token_accuracy": 0.9723258554935456,
      "num_tokens": 34227141.0,
      "step": 1910
    },
    {
      "entropy": 1.137197768688202,
      "epoch": 2.5841184387617764,
      "grad_norm": 1.1534605026245117,
      "learning_rate": 8.86072480448629e-05,
      "loss": 0.07511197328567505,
      "mean_token_accuracy": 0.9702324509620667,
      "num_tokens": 34406526.0,
      "step": 1920
    },
    {
      "entropy": 1.1349146008491515,
      "epoch": 2.5975773889636606,
      "grad_norm": 1.3732489347457886,
      "learning_rate": 8.84654440003469e-05,
      "loss": 0.07147140502929687,
      "mean_token_accuracy": 0.9726706743240356,
      "num_tokens": 34586116.0,
      "step": 1930
    },
    {
      "entropy": 1.1294147491455078,
      "epoch": 2.611036339165545,
      "grad_norm": 0.9715967178344727,
      "learning_rate": 8.83228778654674e-05,
      "loss": 0.07225455045700073,
      "mean_token_accuracy": 0.9726594388484955,
      "num_tokens": 34765289.0,
      "step": 1940
    },
    {
      "entropy": 1.130816388130188,
      "epoch": 2.624495289367429,
      "grad_norm": 1.293736219406128,
      "learning_rate": 8.817955246479276e-05,
      "loss": 0.06845389604568482,
      "mean_token_accuracy": 0.9736943006515503,
      "num_tokens": 34944224.0,
      "step": 1950
    },
    {
      "entropy": 1.1200148224830628,
      "epoch": 2.6379542395693134,
      "grad_norm": 1.2962090969085693,
      "learning_rate": 8.803547063793422e-05,
      "loss": 0.07189736366271973,
      "mean_token_accuracy": 0.9717683315277099,
      "num_tokens": 35123825.0,
      "step": 1960
    },
    {
      "entropy": 1.1301296949386597,
      "epoch": 2.6514131897711977,
      "grad_norm": 1.4734028577804565,
      "learning_rate": 8.789063523948958e-05,
      "loss": 0.0702283263206482,
      "mean_token_accuracy": 0.9727118015289307,
      "num_tokens": 35302914.0,
      "step": 1970
    },
    {
      "entropy": 1.1409568905830383,
      "epoch": 2.664872139973082,
      "grad_norm": 1.4132834672927856,
      "learning_rate": 8.774504913898663e-05,
      "loss": 0.07676968574523926,
      "mean_token_accuracy": 0.9695852339267731,
      "num_tokens": 35481783.0,
      "step": 1980
    },
    {
      "entropy": 1.1532965421676635,
      "epoch": 2.678331090174966,
      "grad_norm": 1.1802046298980713,
      "learning_rate": 8.75987152208264e-05,
      "loss": 0.06539074182510377,
      "mean_token_accuracy": 0.9745090186595917,
      "num_tokens": 35660431.0,
      "step": 1990
    },
    {
      "entropy": 1.1450922250747682,
      "epoch": 2.6917900403768504,
      "grad_norm": 1.1922194957733154,
      "learning_rate": 8.745163638422583e-05,
      "loss": 0.07205181121826172,
      "mean_token_accuracy": 0.9712056815624237,
      "num_tokens": 35839308.0,
      "step": 2000
    },
    {
      "entropy": 1.1444142818450929,
      "epoch": 2.7052489905787347,
      "grad_norm": 1.1868208646774292,
      "learning_rate": 8.730381554316051e-05,
      "loss": 0.07235864400863648,
      "mean_token_accuracy": 0.9725943446159363,
      "num_tokens": 36018734.0,
      "step": 2010
    },
    {
      "entropy": 1.133234965801239,
      "epoch": 2.718707940780619,
      "grad_norm": 1.4924181699752808,
      "learning_rate": 8.715525562630687e-05,
      "loss": 0.07137352228164673,
      "mean_token_accuracy": 0.9720249474048615,
      "num_tokens": 36197607.0,
      "step": 2020
    },
    {
      "entropy": 1.150734007358551,
      "epoch": 2.732166890982503,
      "grad_norm": 1.0833524465560913,
      "learning_rate": 8.700595957698411e-05,
      "loss": 0.07287259101867676,
      "mean_token_accuracy": 0.9720607042312622,
      "num_tokens": 36377137.0,
      "step": 2030
    },
    {
      "entropy": 1.15074782371521,
      "epoch": 2.7456258411843875,
      "grad_norm": 1.1532199382781982,
      "learning_rate": 8.685593035309598e-05,
      "loss": 0.07189793586730957,
      "mean_token_accuracy": 0.971609354019165,
      "num_tokens": 36556438.0,
      "step": 2040
    },
    {
      "entropy": 1.1425267219543458,
      "epoch": 2.7590847913862717,
      "grad_norm": 1.65394926071167,
      "learning_rate": 8.670517092707213e-05,
      "loss": 0.07228031158447265,
      "mean_token_accuracy": 0.972437036037445,
      "num_tokens": 36734936.0,
      "step": 2050
    },
    {
      "entropy": 1.1443095088005066,
      "epoch": 2.772543741588156,
      "grad_norm": 1.1756309270858765,
      "learning_rate": 8.655368428580919e-05,
      "loss": 0.07032725811004639,
      "mean_token_accuracy": 0.9716470181941986,
      "num_tokens": 36913871.0,
      "step": 2060
    },
    {
      "entropy": 1.1369405388832092,
      "epoch": 2.7860026917900402,
      "grad_norm": 1.2845005989074707,
      "learning_rate": 8.640147343061165e-05,
      "loss": 0.07300193309783935,
      "mean_token_accuracy": 0.971499103307724,
      "num_tokens": 37093380.0,
      "step": 2070
    },
    {
      "entropy": 1.1441598296165467,
      "epoch": 2.7994616419919245,
      "grad_norm": 1.0858702659606934,
      "learning_rate": 8.624854137713234e-05,
      "loss": 0.07180417776107788,
      "mean_token_accuracy": 0.9721956551074982,
      "num_tokens": 37272902.0,
      "step": 2080
    },
    {
      "entropy": 1.1500964164733887,
      "epoch": 2.8129205921938087,
      "grad_norm": 1.1274408102035522,
      "learning_rate": 8.609489115531278e-05,
      "loss": 0.07155272960662842,
      "mean_token_accuracy": 0.971377295255661,
      "num_tokens": 37451897.0,
      "step": 2090
    },
    {
      "entropy": 1.1386480689048768,
      "epoch": 2.826379542395693,
      "grad_norm": 1.1894359588623047,
      "learning_rate": 8.594052580932301e-05,
      "loss": 0.06719542145729065,
      "mean_token_accuracy": 0.9733208954334259,
      "num_tokens": 37631343.0,
      "step": 2100
    },
    {
      "entropy": 1.1420456409454345,
      "epoch": 2.8398384925975773,
      "grad_norm": 1.266627550125122,
      "learning_rate": 8.578544839750141e-05,
      "loss": 0.06839650273323059,
      "mean_token_accuracy": 0.9735166966915131,
      "num_tokens": 37811111.0,
      "step": 2110
    },
    {
      "entropy": 1.136332881450653,
      "epoch": 2.8532974427994615,
      "grad_norm": 1.6329811811447144,
      "learning_rate": 8.562966199229399e-05,
      "loss": 0.0761029601097107,
      "mean_token_accuracy": 0.9703849673271179,
      "num_tokens": 37991040.0,
      "step": 2120
    },
    {
      "entropy": 1.1302747488021851,
      "epoch": 2.8667563930013458,
      "grad_norm": 1.355210304260254,
      "learning_rate": 8.547316968019363e-05,
      "loss": 0.07443415522575378,
      "mean_token_accuracy": 0.971444720029831,
      "num_tokens": 38169539.0,
      "step": 2130
    },
    {
      "entropy": 1.1407261610031127,
      "epoch": 2.88021534320323,
      "grad_norm": 1.0608882904052734,
      "learning_rate": 8.531597456167885e-05,
      "loss": 0.07463377118110656,
      "mean_token_accuracy": 0.9705821752548218,
      "num_tokens": 38348423.0,
      "step": 2140
    },
    {
      "entropy": 1.121661639213562,
      "epoch": 2.8936742934051143,
      "grad_norm": 1.4264295101165771,
      "learning_rate": 8.515807975115239e-05,
      "loss": 0.06971895098686218,
      "mean_token_accuracy": 0.9723855495452881,
      "num_tokens": 38527736.0,
      "step": 2150
    },
    {
      "entropy": 1.117723774909973,
      "epoch": 2.9071332436069985,
      "grad_norm": 1.114941120147705,
      "learning_rate": 8.499948837687959e-05,
      "loss": 0.07229661345481872,
      "mean_token_accuracy": 0.9710954666137696,
      "num_tokens": 38706893.0,
      "step": 2160
    },
    {
      "entropy": 1.12820805311203,
      "epoch": 2.920592193808883,
      "grad_norm": 1.219247817993164,
      "learning_rate": 8.484020358092625e-05,
      "loss": 0.07078794836997986,
      "mean_token_accuracy": 0.9723483324050903,
      "num_tokens": 38886237.0,
      "step": 2170
    },
    {
      "entropy": 1.1263705372810364,
      "epoch": 2.934051144010767,
      "grad_norm": 1.9336564540863037,
      "learning_rate": 8.468022851909657e-05,
      "loss": 0.07343355417251587,
      "mean_token_accuracy": 0.9712878108024597,
      "num_tokens": 39065800.0,
      "step": 2180
    },
    {
      "entropy": 1.1296117424964904,
      "epoch": 2.9475100942126513,
      "grad_norm": 1.1307294368743896,
      "learning_rate": 8.451956636087046e-05,
      "loss": 0.07211248874664307,
      "mean_token_accuracy": 0.9710810720920563,
      "num_tokens": 39245521.0,
      "step": 2190
    },
    {
      "entropy": 1.1172988891601563,
      "epoch": 2.9609690444145356,
      "grad_norm": 1.1942963600158691,
      "learning_rate": 8.435822028934087e-05,
      "loss": 0.07098879814147949,
      "mean_token_accuracy": 0.9725285410881043,
      "num_tokens": 39424709.0,
      "step": 2200
    },
    {
      "entropy": 1.1155216932296752,
      "epoch": 2.97442799461642,
      "grad_norm": 1.241036057472229,
      "learning_rate": 8.41961935011506e-05,
      "loss": 0.06951723098754883,
      "mean_token_accuracy": 0.9731462299823761,
      "num_tokens": 39603326.0,
      "step": 2210
    },
    {
      "entropy": 1.1296806812286377,
      "epoch": 2.987886944818304,
      "grad_norm": 1.249014139175415,
      "learning_rate": 8.403348920642911e-05,
      "loss": 0.07304394245147705,
      "mean_token_accuracy": 0.9720990836620331,
      "num_tokens": 39782396.0,
      "step": 2220
    },
    {
      "epoch": 3.0,
      "eval_entropy": 1.1091555235492196,
      "eval_loss": 0.10757029801607132,
      "eval_mean_token_accuracy": 0.9568785178433558,
      "eval_num_tokens": 39944284.0,
      "eval_runtime": 12.7473,
      "eval_samples_per_second": 392.239,
      "eval_steps_per_second": 12.316,
      "step": 2229
    }
  ],
  "logging_steps": 10,
  "max_steps": 7430,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.9036814964267418e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}