{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.914164606146238,
  "eval_steps": 3000,
  "global_step": 33000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.639010858535767,
      "epoch": 0.0004415400918403391,
      "grad_norm": 1.046875,
      "learning_rate": 2e-06,
      "loss": 7.4495,
      "mean_token_accuracy": 0.08912035375833512,
      "num_tokens": 9390.0,
      "step": 5
    },
    {
      "entropy": 7.592498016357422,
      "epoch": 0.0008830801836806782,
      "grad_norm": 1.015625,
      "learning_rate": 4.5e-06,
      "loss": 7.5473,
      "mean_token_accuracy": 0.08600941896438599,
      "num_tokens": 18671.0,
      "step": 10
    },
    {
      "entropy": 7.60152039527893,
      "epoch": 0.0013246202755210173,
      "grad_norm": 1.09375,
      "learning_rate": 7e-06,
      "loss": 7.4983,
      "mean_token_accuracy": 0.09218712449073792,
      "num_tokens": 27614.0,
      "step": 15
    },
    {
      "entropy": 7.6614744663238525,
      "epoch": 0.0017661603673613563,
      "grad_norm": 0.97265625,
      "learning_rate": 9.5e-06,
      "loss": 7.6113,
      "mean_token_accuracy": 0.0849214605987072,
      "num_tokens": 37850.0,
      "step": 20
    },
    {
      "entropy": 7.640540361404419,
      "epoch": 0.0022077004592016957,
      "grad_norm": 0.96875,
      "learning_rate": 1.2e-05,
      "loss": 7.4958,
      "mean_token_accuracy": 0.10092130824923515,
      "num_tokens": 47166.0,
      "step": 25
    },
    {
      "entropy": 7.689523506164551,
      "epoch": 0.0026492405510420347,
      "grad_norm": 1.0,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 7.485,
      "mean_token_accuracy": 0.09342324361205101,
      "num_tokens": 55500.0,
      "step": 30
    },
    {
      "entropy": 7.649916553497315,
      "epoch": 0.0030907806428823736,
      "grad_norm": 1.0078125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 7.4904,
      "mean_token_accuracy": 0.08857546746730804,
      "num_tokens": 63851.0,
      "step": 35
    },
    {
      "entropy": 7.70954270362854,
      "epoch": 0.0035323207347227126,
      "grad_norm": 0.99609375,
      "learning_rate": 1.95e-05,
      "loss": 7.5264,
      "mean_token_accuracy": 0.09447742104530335,
      "num_tokens": 73697.0,
      "step": 40
    },
    {
      "entropy": 7.706955003738403,
      "epoch": 0.003973860826563052,
      "grad_norm": 1.203125,
      "learning_rate": 2.2e-05,
      "loss": 7.4835,
      "mean_token_accuracy": 0.09832259118556977,
      "num_tokens": 83000.0,
      "step": 45
    },
    {
      "entropy": 7.693207168579102,
      "epoch": 0.004415400918403391,
      "grad_norm": 0.87890625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 7.4891,
      "mean_token_accuracy": 0.09248929545283317,
      "num_tokens": 92982.0,
      "step": 50
    },
    {
      "entropy": 7.701184225082398,
      "epoch": 0.00485694101024373,
      "grad_norm": 1.0859375,
      "learning_rate": 2.7e-05,
      "loss": 7.4475,
      "mean_token_accuracy": 0.0948328971862793,
      "num_tokens": 101455.0,
      "step": 55
    },
    {
      "entropy": 7.600388383865356,
      "epoch": 0.005298481102084069,
      "grad_norm": 0.92578125,
      "learning_rate": 2.95e-05,
      "loss": 7.4784,
      "mean_token_accuracy": 0.09275375679135323,
      "num_tokens": 110782.0,
      "step": 60
    },
    {
      "entropy": 7.643292331695557,
      "epoch": 0.005740021193924408,
      "grad_norm": 0.953125,
      "learning_rate": 3.2e-05,
      "loss": 7.3593,
      "mean_token_accuracy": 0.09755741134285927,
      "num_tokens": 119241.0,
      "step": 65
    },
    {
      "entropy": 7.646019697189331,
      "epoch": 0.006181561285764747,
      "grad_norm": 1.0390625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 7.4621,
      "mean_token_accuracy": 0.08904825896024704,
      "num_tokens": 127903.0,
      "step": 70
    },
    {
      "entropy": 7.611093521118164,
      "epoch": 0.006623101377605086,
      "grad_norm": 0.98828125,
      "learning_rate": 3.7e-05,
      "loss": 7.4256,
      "mean_token_accuracy": 0.09207391366362572,
      "num_tokens": 137370.0,
      "step": 75
    },
    {
      "entropy": 7.66001524925232,
      "epoch": 0.007064641469445425,
      "grad_norm": 0.94140625,
      "learning_rate": 3.95e-05,
      "loss": 7.3699,
      "mean_token_accuracy": 0.09808079078793526,
      "num_tokens": 146582.0,
      "step": 80
    },
    {
      "entropy": 7.637924289703369,
      "epoch": 0.007506181561285765,
      "grad_norm": 1.0234375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 7.3648,
      "mean_token_accuracy": 0.09920610412955284,
      "num_tokens": 154933.0,
      "step": 85
    },
    {
      "entropy": 7.600069618225097,
      "epoch": 0.007947721653126103,
      "grad_norm": 0.97265625,
      "learning_rate": 4.45e-05,
      "loss": 7.4711,
      "mean_token_accuracy": 0.0958680585026741,
      "num_tokens": 165157.0,
      "step": 90
    },
    {
      "entropy": 7.64748182296753,
      "epoch": 0.008389261744966443,
      "grad_norm": 0.93359375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.379,
      "mean_token_accuracy": 0.09376930147409439,
      "num_tokens": 174958.0,
      "step": 95
    },
    {
      "entropy": 7.598965072631836,
      "epoch": 0.008830801836806783,
      "grad_norm": 1.0390625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 7.4361,
      "mean_token_accuracy": 0.0984335981309414,
      "num_tokens": 184256.0,
      "step": 100
    },
    {
      "entropy": 7.684039211273193,
      "epoch": 0.009272341928647121,
      "grad_norm": 1.015625,
      "learning_rate": 5.2e-05,
      "loss": 7.367,
      "mean_token_accuracy": 0.10001294463872909,
      "num_tokens": 192894.0,
      "step": 105
    },
    {
      "entropy": 7.646014642715454,
      "epoch": 0.00971388202048746,
      "grad_norm": 1.0234375,
      "learning_rate": 5.45e-05,
      "loss": 7.4273,
      "mean_token_accuracy": 0.09812369421124459,
      "num_tokens": 202675.0,
      "step": 110
    },
    {
      "entropy": 7.599092102050781,
      "epoch": 0.010155422112327799,
      "grad_norm": 0.97265625,
      "learning_rate": 5.7e-05,
      "loss": 7.4314,
      "mean_token_accuracy": 0.09605375379323959,
      "num_tokens": 212261.0,
      "step": 115
    },
    {
      "entropy": 7.727504396438599,
      "epoch": 0.010596962204168139,
      "grad_norm": 1.046875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 7.5105,
      "mean_token_accuracy": 0.09431948810815811,
      "num_tokens": 222329.0,
      "step": 120
    },
    {
      "entropy": 7.697684812545776,
      "epoch": 0.011038502296008477,
      "grad_norm": 1.125,
      "learning_rate": 6.2e-05,
      "loss": 7.4354,
      "mean_token_accuracy": 0.09025316089391708,
      "num_tokens": 231247.0,
      "step": 125
    },
    {
      "entropy": 7.657842016220092,
      "epoch": 0.011480042387848817,
      "grad_norm": 1.21875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.379,
      "mean_token_accuracy": 0.1007770724594593,
      "num_tokens": 239978.0,
      "step": 130
    },
    {
      "entropy": 7.621602201461792,
      "epoch": 0.011921582479689156,
      "grad_norm": 0.98828125,
      "learning_rate": 6.7e-05,
      "loss": 7.4038,
      "mean_token_accuracy": 0.09848067536950111,
      "num_tokens": 249735.0,
      "step": 135
    },
    {
      "entropy": 7.6923524856567385,
      "epoch": 0.012363122571529495,
      "grad_norm": 1.234375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.4418,
      "mean_token_accuracy": 0.09509150311350822,
      "num_tokens": 259369.0,
      "step": 140
    },
    {
      "entropy": 7.714404106140137,
      "epoch": 0.012804662663369834,
      "grad_norm": 0.97265625,
      "learning_rate": 7.2e-05,
      "loss": 7.4701,
      "mean_token_accuracy": 0.09397952705621719,
      "num_tokens": 268645.0,
      "step": 145
    },
    {
      "entropy": 7.6344895362854,
      "epoch": 0.013246202755210172,
      "grad_norm": 1.078125,
      "learning_rate": 7.45e-05,
      "loss": 7.3506,
      "mean_token_accuracy": 0.10026702061295509,
      "num_tokens": 276667.0,
      "step": 150
    },
    {
      "entropy": 7.631509923934937,
      "epoch": 0.013687742847050512,
      "grad_norm": 0.9453125,
      "learning_rate": 7.7e-05,
      "loss": 7.3888,
      "mean_token_accuracy": 0.10270725786685944,
      "num_tokens": 286017.0,
      "step": 155
    },
    {
      "entropy": 7.641970491409301,
      "epoch": 0.01412928293889085,
      "grad_norm": 0.9921875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.4111,
      "mean_token_accuracy": 0.09624978005886078,
      "num_tokens": 295631.0,
      "step": 160
    },
    {
      "entropy": 7.598629140853882,
      "epoch": 0.01457082303073119,
      "grad_norm": 0.9921875,
      "learning_rate": 8.2e-05,
      "loss": 7.4006,
      "mean_token_accuracy": 0.10352006703615188,
      "num_tokens": 304704.0,
      "step": 165
    },
    {
      "entropy": 7.7567332744598385,
      "epoch": 0.01501236312257153,
      "grad_norm": 0.99609375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.4231,
      "mean_token_accuracy": 0.10067867860198021,
      "num_tokens": 314195.0,
      "step": 170
    },
    {
      "entropy": 7.665441656112671,
      "epoch": 0.015453903214411868,
      "grad_norm": 1.2421875,
      "learning_rate": 8.7e-05,
      "loss": 7.4392,
      "mean_token_accuracy": 0.09234830960631371,
      "num_tokens": 323379.0,
      "step": 175
    },
    {
      "entropy": 7.631372451782227,
      "epoch": 0.015895443306252206,
      "grad_norm": 1.0546875,
      "learning_rate": 8.95e-05,
      "loss": 7.3662,
      "mean_token_accuracy": 0.10526679679751397,
      "num_tokens": 332322.0,
      "step": 180
    },
    {
      "entropy": 7.573805236816407,
      "epoch": 0.016336983398092548,
      "grad_norm": 1.140625,
      "learning_rate": 9.2e-05,
      "loss": 7.3379,
      "mean_token_accuracy": 0.10242495015263557,
      "num_tokens": 341735.0,
      "step": 185
    },
    {
      "entropy": 7.7565899848937985,
      "epoch": 0.016778523489932886,
      "grad_norm": 1.0546875,
      "learning_rate": 9.45e-05,
      "loss": 7.4344,
      "mean_token_accuracy": 0.09886153489351272,
      "num_tokens": 351209.0,
      "step": 190
    },
    {
      "entropy": 7.6179423332214355,
      "epoch": 0.017220063581773224,
      "grad_norm": 1.09375,
      "learning_rate": 9.7e-05,
      "loss": 7.3605,
      "mean_token_accuracy": 0.10053950771689416,
      "num_tokens": 360467.0,
      "step": 195
    },
    {
      "entropy": 7.6365471363067625,
      "epoch": 0.017661603673613566,
      "grad_norm": 1.0859375,
      "learning_rate": 9.95e-05,
      "loss": 7.4033,
      "mean_token_accuracy": 0.10138513222336769,
      "num_tokens": 370361.0,
      "step": 200
    },
    {
      "entropy": 7.720054578781128,
      "epoch": 0.018103143765453904,
      "grad_norm": 1.328125,
      "learning_rate": 0.000102,
      "loss": 7.4181,
      "mean_token_accuracy": 0.09296029135584831,
      "num_tokens": 380366.0,
      "step": 205
    },
    {
      "entropy": 7.615629529953003,
      "epoch": 0.018544683857294242,
      "grad_norm": 1.109375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.3449,
      "mean_token_accuracy": 0.10289899110794068,
      "num_tokens": 390690.0,
      "step": 210
    },
    {
      "entropy": 7.594337797164917,
      "epoch": 0.01898622394913458,
      "grad_norm": 1.03125,
      "learning_rate": 0.000107,
      "loss": 7.4248,
      "mean_token_accuracy": 0.09637770131230354,
      "num_tokens": 400722.0,
      "step": 215
    },
    {
      "entropy": 7.605887174606323,
      "epoch": 0.01942776404097492,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001095,
      "loss": 7.3072,
      "mean_token_accuracy": 0.09810378924012184,
      "num_tokens": 410223.0,
      "step": 220
    },
    {
      "entropy": 7.63714017868042,
      "epoch": 0.01986930413281526,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000112,
      "loss": 7.3788,
      "mean_token_accuracy": 0.10300084874033928,
      "num_tokens": 420214.0,
      "step": 225
    },
    {
      "entropy": 7.656309795379639,
      "epoch": 0.020310844224655598,
      "grad_norm": 0.984375,
      "learning_rate": 0.0001145,
      "loss": 7.3416,
      "mean_token_accuracy": 0.09932817742228509,
      "num_tokens": 429407.0,
      "step": 230
    },
    {
      "entropy": 7.613865327835083,
      "epoch": 0.02075238431649594,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.2821,
      "mean_token_accuracy": 0.0999271385371685,
      "num_tokens": 438403.0,
      "step": 235
    },
    {
      "entropy": 7.571551370620727,
      "epoch": 0.021193924408336277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.3115,
      "mean_token_accuracy": 0.10332573652267456,
      "num_tokens": 447466.0,
      "step": 240
    },
    {
      "entropy": 7.66993556022644,
      "epoch": 0.021635464500176615,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000122,
      "loss": 7.4665,
      "mean_token_accuracy": 0.09346566423773765,
      "num_tokens": 457141.0,
      "step": 245
    },
    {
      "entropy": 7.626456832885742,
      "epoch": 0.022077004592016954,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001245,
      "loss": 7.3079,
      "mean_token_accuracy": 0.09943239539861679,
      "num_tokens": 465708.0,
      "step": 250
    },
    {
      "entropy": 7.6611328125,
      "epoch": 0.022518544683857295,
      "grad_norm": 1.203125,
      "learning_rate": 0.000127,
      "loss": 7.3345,
      "mean_token_accuracy": 0.09984651207923889,
      "num_tokens": 475369.0,
      "step": 255
    },
    {
      "entropy": 7.552011871337891,
      "epoch": 0.022960084775697633,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001295,
      "loss": 7.2893,
      "mean_token_accuracy": 0.10191124230623246,
      "num_tokens": 484249.0,
      "step": 260
    },
    {
      "entropy": 7.669333648681641,
      "epoch": 0.02340162486753797,
      "grad_norm": 1.203125,
      "learning_rate": 0.000132,
      "loss": 7.4184,
      "mean_token_accuracy": 0.09950976893305778,
      "num_tokens": 493303.0,
      "step": 265
    },
    {
      "entropy": 7.627326536178589,
      "epoch": 0.023843164959378313,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.3782,
      "mean_token_accuracy": 0.10428153648972512,
      "num_tokens": 501503.0,
      "step": 270
    },
    {
      "entropy": 7.684601449966431,
      "epoch": 0.02428470505121865,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.2836,
      "mean_token_accuracy": 0.10307212695479392,
      "num_tokens": 509661.0,
      "step": 275
    },
    {
      "entropy": 7.573146200180053,
      "epoch": 0.02472624514305899,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001395,
      "loss": 7.2937,
      "mean_token_accuracy": 0.0982547052204609,
      "num_tokens": 519464.0,
      "step": 280
    },
    {
      "entropy": 7.555501937866211,
      "epoch": 0.025167785234899327,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.2655,
      "mean_token_accuracy": 0.1019679419696331,
      "num_tokens": 527968.0,
      "step": 285
    },
    {
      "entropy": 7.598510599136352,
      "epoch": 0.02560932532673967,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001445,
      "loss": 7.3321,
      "mean_token_accuracy": 0.10574981644749641,
      "num_tokens": 537234.0,
      "step": 290
    },
    {
      "entropy": 7.564266014099121,
      "epoch": 0.026050865418580007,
      "grad_norm": 1.171875,
      "learning_rate": 0.000147,
      "loss": 7.2971,
      "mean_token_accuracy": 0.1034348577260971,
      "num_tokens": 546398.0,
      "step": 295
    },
    {
      "entropy": 7.582965517044068,
      "epoch": 0.026492405510420345,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001495,
      "loss": 7.3228,
      "mean_token_accuracy": 0.10241033658385276,
      "num_tokens": 555362.0,
      "step": 300
    },
    {
      "entropy": 7.702879476547241,
      "epoch": 0.026933945602260687,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000152,
      "loss": 7.3457,
      "mean_token_accuracy": 0.09650135710835457,
      "num_tokens": 564575.0,
      "step": 305
    },
    {
      "entropy": 7.648530769348144,
      "epoch": 0.027375485694101025,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.3769,
      "mean_token_accuracy": 0.10071438252925873,
      "num_tokens": 573915.0,
      "step": 310
    },
    {
      "entropy": 7.583766126632691,
      "epoch": 0.027817025785941363,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000157,
      "loss": 7.2641,
      "mean_token_accuracy": 0.1006499283015728,
      "num_tokens": 583216.0,
      "step": 315
    },
    {
      "entropy": 7.5586583614349365,
      "epoch": 0.0282585658777817,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001595,
      "loss": 7.322,
      "mean_token_accuracy": 0.10538614764809609,
      "num_tokens": 591955.0,
      "step": 320
    },
    {
      "entropy": 7.593068265914917,
      "epoch": 0.028700105969622042,
      "grad_norm": 1.09375,
      "learning_rate": 0.000162,
      "loss": 7.3037,
      "mean_token_accuracy": 0.10241075083613396,
      "num_tokens": 600999.0,
      "step": 325
    },
    {
      "entropy": 7.519618797302246,
      "epoch": 0.02914164606146238,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.2217,
      "mean_token_accuracy": 0.1135806605219841,
      "num_tokens": 609478.0,
      "step": 330
    },
    {
      "entropy": 7.557135772705078,
      "epoch": 0.02958318615330272,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.3182,
      "mean_token_accuracy": 0.1028580516576767,
      "num_tokens": 618348.0,
      "step": 335
    },
    {
      "entropy": 7.581590604782105,
      "epoch": 0.03002472624514306,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.3462,
      "mean_token_accuracy": 0.10086702331900596,
      "num_tokens": 628548.0,
      "step": 340
    },
    {
      "entropy": 7.617287588119507,
      "epoch": 0.0304662663369834,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.2931,
      "mean_token_accuracy": 0.10549068599939346,
      "num_tokens": 637489.0,
      "step": 345
    },
    {
      "entropy": 7.5078435897827145,
      "epoch": 0.030907806428823736,
      "grad_norm": 1.078125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.2827,
      "mean_token_accuracy": 0.10264097377657891,
      "num_tokens": 646715.0,
      "step": 350
    },
    {
      "entropy": 7.540816068649292,
      "epoch": 0.031349346520664075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000177,
      "loss": 7.2795,
      "mean_token_accuracy": 0.10515663996338845,
      "num_tokens": 656858.0,
      "step": 355
    },
    {
      "entropy": 7.596448993682861,
      "epoch": 0.03179088661250441,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001795,
      "loss": 7.2682,
      "mean_token_accuracy": 0.10302142426371574,
      "num_tokens": 665968.0,
      "step": 360
    },
    {
      "entropy": 7.53389573097229,
      "epoch": 0.03223242670434476,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000182,
      "loss": 7.2449,
      "mean_token_accuracy": 0.10066767483949661,
      "num_tokens": 674295.0,
      "step": 365
    },
    {
      "entropy": 7.472217559814453,
      "epoch": 0.032673966796185096,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001845,
      "loss": 7.2024,
      "mean_token_accuracy": 0.1055987648665905,
      "num_tokens": 683559.0,
      "step": 370
    },
    {
      "entropy": 7.468035459518433,
      "epoch": 0.033115506888025434,
      "grad_norm": 1.4375,
      "learning_rate": 0.000187,
      "loss": 7.2347,
      "mean_token_accuracy": 0.104932302236557,
      "num_tokens": 692402.0,
      "step": 375
    },
    {
      "entropy": 7.603985595703125,
      "epoch": 0.03355704697986577,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001895,
      "loss": 7.3376,
      "mean_token_accuracy": 0.09718681424856186,
      "num_tokens": 702052.0,
      "step": 380
    },
    {
      "entropy": 7.571860313415527,
      "epoch": 0.03399858707170611,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000192,
      "loss": 7.2579,
      "mean_token_accuracy": 0.103890460729599,
      "num_tokens": 711926.0,
      "step": 385
    },
    {
      "entropy": 7.518250036239624,
      "epoch": 0.03444012716354645,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001945,
      "loss": 7.2055,
      "mean_token_accuracy": 0.10119849517941475,
      "num_tokens": 720948.0,
      "step": 390
    },
    {
      "entropy": 7.473824882507325,
      "epoch": 0.034881667255386786,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.2572,
      "mean_token_accuracy": 0.1049205407500267,
      "num_tokens": 729611.0,
      "step": 395
    },
    {
      "entropy": 7.686066150665283,
      "epoch": 0.03532320734722713,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.2316,
      "mean_token_accuracy": 0.0993150919675827,
      "num_tokens": 738433.0,
      "step": 400
    },
    {
      "entropy": 7.416238927841187,
      "epoch": 0.03576474743906747,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000202,
      "loss": 7.2808,
      "mean_token_accuracy": 0.10098983719944954,
      "num_tokens": 747310.0,
      "step": 405
    },
    {
      "entropy": 7.518594074249267,
      "epoch": 0.03620628753090781,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.2464,
      "mean_token_accuracy": 0.10577662512660027,
      "num_tokens": 756362.0,
      "step": 410
    },
    {
      "entropy": 7.609608840942383,
      "epoch": 0.036647827622748146,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000207,
      "loss": 7.2509,
      "mean_token_accuracy": 0.10020640343427659,
      "num_tokens": 764978.0,
      "step": 415
    },
    {
      "entropy": 7.396434926986695,
      "epoch": 0.037089367714588484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002095,
      "loss": 7.2338,
      "mean_token_accuracy": 0.10793569758534431,
      "num_tokens": 774058.0,
      "step": 420
    },
    {
      "entropy": 7.5231493473052975,
      "epoch": 0.03753090780642882,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000212,
      "loss": 7.2202,
      "mean_token_accuracy": 0.10111990198493004,
      "num_tokens": 783332.0,
      "step": 425
    },
    {
      "entropy": 7.536509370803833,
      "epoch": 0.03797244789826916,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002145,
      "loss": 7.2874,
      "mean_token_accuracy": 0.09968122690916062,
      "num_tokens": 792481.0,
      "step": 430
    },
    {
      "entropy": 7.3964588165283205,
      "epoch": 0.038413987990109505,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.1932,
      "mean_token_accuracy": 0.10685456469655037,
      "num_tokens": 801396.0,
      "step": 435
    },
    {
      "entropy": 7.5557657241821286,
      "epoch": 0.03885552808194984,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002195,
      "loss": 7.2548,
      "mean_token_accuracy": 0.09790834933519363,
      "num_tokens": 809903.0,
      "step": 440
    },
    {
      "entropy": 7.494475698471069,
      "epoch": 0.03929706817379018,
      "grad_norm": 1.109375,
      "learning_rate": 0.000222,
      "loss": 7.2027,
      "mean_token_accuracy": 0.10747286528348923,
      "num_tokens": 819144.0,
      "step": 445
    },
    {
      "entropy": 7.3877339363098145,
      "epoch": 0.03973860826563052,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002245,
      "loss": 7.2456,
      "mean_token_accuracy": 0.1021496333181858,
      "num_tokens": 828881.0,
      "step": 450
    },
    {
      "entropy": 7.51954607963562,
      "epoch": 0.04018014835747086,
      "grad_norm": 1.078125,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.1614,
      "mean_token_accuracy": 0.09784932136535644,
      "num_tokens": 837588.0,
      "step": 455
    },
    {
      "entropy": 7.3755847930908205,
      "epoch": 0.040621688449311195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.1635,
      "mean_token_accuracy": 0.1048378475010395,
      "num_tokens": 847002.0,
      "step": 460
    },
    {
      "entropy": 7.388737916946411,
      "epoch": 0.041063228541151534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.1795,
      "mean_token_accuracy": 0.10455363765358924,
      "num_tokens": 855791.0,
      "step": 465
    },
    {
      "entropy": 7.4601222515106205,
      "epoch": 0.04150476863299188,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.2059,
      "mean_token_accuracy": 0.11087375283241271,
      "num_tokens": 865392.0,
      "step": 470
    },
    {
      "entropy": 7.434926748275757,
      "epoch": 0.04194630872483222,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000237,
      "loss": 7.2927,
      "mean_token_accuracy": 0.10048188641667366,
      "num_tokens": 874807.0,
      "step": 475
    },
    {
      "entropy": 7.5444385528564455,
      "epoch": 0.042387848816672555,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002395,
      "loss": 7.25,
      "mean_token_accuracy": 0.1069977454841137,
      "num_tokens": 883928.0,
      "step": 480
    },
    {
      "entropy": 7.370575952529907,
      "epoch": 0.04282938890851289,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000242,
      "loss": 7.1509,
      "mean_token_accuracy": 0.10749021843075753,
      "num_tokens": 893047.0,
      "step": 485
    },
    {
      "entropy": 7.429861927032471,
      "epoch": 0.04327092900035323,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002445,
      "loss": 7.1449,
      "mean_token_accuracy": 0.10997916385531425,
      "num_tokens": 901645.0,
      "step": 490
    },
    {
      "entropy": 7.455361604690552,
      "epoch": 0.04371246909219357,
      "grad_norm": 1.296875,
      "learning_rate": 0.000247,
      "loss": 7.1695,
      "mean_token_accuracy": 0.1042562760412693,
      "num_tokens": 911169.0,
      "step": 495
    },
    {
      "entropy": 7.393462133407593,
      "epoch": 0.04415400918403391,
      "grad_norm": 1.125,
      "learning_rate": 0.0002495,
      "loss": 7.2024,
      "mean_token_accuracy": 0.10118649303913116,
      "num_tokens": 921382.0,
      "step": 500
    },
    {
      "entropy": 7.3356465816497805,
      "epoch": 0.04459554927587425,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000252,
      "loss": 7.1024,
      "mean_token_accuracy": 0.11079763695597648,
      "num_tokens": 930409.0,
      "step": 505
    },
    {
      "entropy": 7.3716787815094,
      "epoch": 0.04503708936771459,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002545,
      "loss": 7.077,
      "mean_token_accuracy": 0.11271328553557396,
      "num_tokens": 939049.0,
      "step": 510
    },
    {
      "entropy": 7.437454271316528,
      "epoch": 0.04547862945955493,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000257,
      "loss": 7.1052,
      "mean_token_accuracy": 0.10970233231782914,
      "num_tokens": 947575.0,
      "step": 515
    },
    {
      "entropy": 7.417500114440918,
      "epoch": 0.045920169551395267,
      "grad_norm": 1.125,
      "learning_rate": 0.0002595,
      "loss": 7.2637,
      "mean_token_accuracy": 0.10168002024292946,
      "num_tokens": 957848.0,
      "step": 520
    },
    {
      "entropy": 7.374029064178467,
      "epoch": 0.046361709643235605,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000262,
      "loss": 7.1028,
      "mean_token_accuracy": 0.10596310794353485,
      "num_tokens": 966521.0,
      "step": 525
    },
    {
      "entropy": 7.3952889919281,
      "epoch": 0.04680324973507594,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.1719,
      "mean_token_accuracy": 0.110343337059021,
      "num_tokens": 975827.0,
      "step": 530
    },
    {
      "entropy": 7.375133419036866,
      "epoch": 0.04724478982691628,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.1173,
      "mean_token_accuracy": 0.11087020114064217,
      "num_tokens": 985292.0,
      "step": 535
    },
    {
      "entropy": 7.2821883201599125,
      "epoch": 0.047686329918756626,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.1083,
      "mean_token_accuracy": 0.11026296317577362,
      "num_tokens": 994791.0,
      "step": 540
    },
    {
      "entropy": 7.3696636199951175,
      "epoch": 0.048127870010596964,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.0853,
      "mean_token_accuracy": 0.11716266572475434,
      "num_tokens": 1003700.0,
      "step": 545
    },
    {
      "entropy": 7.283825445175171,
      "epoch": 0.0485694101024373,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002745,
      "loss": 7.1168,
      "mean_token_accuracy": 0.10780237540602684,
      "num_tokens": 1012682.0,
      "step": 550
    },
    {
      "entropy": 7.344115352630615,
      "epoch": 0.04901095019427764,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000277,
      "loss": 7.1338,
      "mean_token_accuracy": 0.10582566857337952,
      "num_tokens": 1021018.0,
      "step": 555
    },
    {
      "entropy": 7.34993805885315,
      "epoch": 0.04945249028611798,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002795,
      "loss": 7.0103,
      "mean_token_accuracy": 0.11074965745210648,
      "num_tokens": 1029744.0,
      "step": 560
    },
    {
      "entropy": 7.1853831768035885,
      "epoch": 0.049894030377958316,
      "grad_norm": 1.25,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.098,
      "mean_token_accuracy": 0.11121385991573333,
      "num_tokens": 1038805.0,
      "step": 565
    },
    {
      "entropy": 7.3726812362670895,
      "epoch": 0.050335570469798654,
      "grad_norm": 1.25,
      "learning_rate": 0.0002845,
      "loss": 7.1603,
      "mean_token_accuracy": 0.10785874202847481,
      "num_tokens": 1047656.0,
      "step": 570
    },
    {
      "entropy": 7.3822227954864506,
      "epoch": 0.050777110561639,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000287,
      "loss": 7.1509,
      "mean_token_accuracy": 0.10956593379378318,
      "num_tokens": 1056598.0,
      "step": 575
    },
    {
      "entropy": 7.472517013549805,
      "epoch": 0.05121865065347934,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002895,
      "loss": 7.1581,
      "mean_token_accuracy": 0.10991680398583412,
      "num_tokens": 1065226.0,
      "step": 580
    },
    {
      "entropy": 7.192884206771851,
      "epoch": 0.051660190745319676,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000292,
      "loss": 7.0768,
      "mean_token_accuracy": 0.11151133328676224,
      "num_tokens": 1074661.0,
      "step": 585
    },
    {
      "entropy": 7.38781328201294,
      "epoch": 0.052101730837160014,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002945,
      "loss": 7.1923,
      "mean_token_accuracy": 0.10381263568997383,
      "num_tokens": 1083921.0,
      "step": 590
    },
    {
      "entropy": 7.323068380355835,
      "epoch": 0.05254327092900035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000297,
      "loss": 7.0614,
      "mean_token_accuracy": 0.11602075025439262,
      "num_tokens": 1093399.0,
      "step": 595
    },
    {
      "entropy": 7.208582735061645,
      "epoch": 0.05298481102084069,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002995,
      "loss": 7.1418,
      "mean_token_accuracy": 0.10923573076725006,
      "num_tokens": 1104065.0,
      "step": 600
    },
    {
      "entropy": 7.303856897354126,
      "epoch": 0.05342635111268103,
      "grad_norm": 1.15625,
      "learning_rate": 0.000302,
      "loss": 7.039,
      "mean_token_accuracy": 0.11398651152849197,
      "num_tokens": 1112995.0,
      "step": 605
    },
    {
      "entropy": 7.268756914138794,
      "epoch": 0.05386789120452137,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003045,
      "loss": 7.0618,
      "mean_token_accuracy": 0.10772402063012124,
      "num_tokens": 1121637.0,
      "step": 610
    },
    {
      "entropy": 7.217564678192138,
      "epoch": 0.05430943129636171,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000307,
      "loss": 7.0115,
      "mean_token_accuracy": 0.11113513186573983,
      "num_tokens": 1131166.0,
      "step": 615
    },
    {
      "entropy": 7.258164548873902,
      "epoch": 0.05475097138820205,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003095,
      "loss": 7.0109,
      "mean_token_accuracy": 0.11285025104880334,
      "num_tokens": 1140888.0,
      "step": 620
    },
    {
      "entropy": 7.336677837371826,
      "epoch": 0.05519251148004239,
      "grad_norm": 1.15625,
      "learning_rate": 0.000312,
      "loss": 7.1858,
      "mean_token_accuracy": 0.10437602549791336,
      "num_tokens": 1150278.0,
      "step": 625
    },
    {
      "entropy": 7.263944149017334,
      "epoch": 0.055634051571882726,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003145,
      "loss": 7.1197,
      "mean_token_accuracy": 0.1070900171995163,
      "num_tokens": 1159348.0,
      "step": 630
    },
    {
      "entropy": 7.358008766174317,
      "epoch": 0.056075591663723064,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000317,
      "loss": 7.1418,
      "mean_token_accuracy": 0.10531127899885177,
      "num_tokens": 1168883.0,
      "step": 635
    },
    {
      "entropy": 7.3235701560974125,
      "epoch": 0.0565171317555634,
      "grad_norm": 1.25,
      "learning_rate": 0.0003195,
      "loss": 7.0476,
      "mean_token_accuracy": 0.11230480372905731,
      "num_tokens": 1178572.0,
      "step": 640
    },
    {
      "entropy": 7.20401668548584,
      "epoch": 0.05695867184740375,
      "grad_norm": 1.0625,
      "learning_rate": 0.000322,
      "loss": 7.1383,
      "mean_token_accuracy": 0.10713300257921218,
      "num_tokens": 1188909.0,
      "step": 645
    },
    {
      "entropy": 7.316440725326538,
      "epoch": 0.057400211939244085,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.0315,
      "mean_token_accuracy": 0.10825848579406738,
      "num_tokens": 1197705.0,
      "step": 650
    },
    {
      "entropy": 7.163294839859009,
      "epoch": 0.05784175203108442,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.8835,
      "mean_token_accuracy": 0.11359791234135627,
      "num_tokens": 1206351.0,
      "step": 655
    },
    {
      "entropy": 7.221329212188721,
      "epoch": 0.05828329212292476,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.9963,
      "mean_token_accuracy": 0.10536118149757386,
      "num_tokens": 1214984.0,
      "step": 660
    },
    {
      "entropy": 7.231037664413452,
      "epoch": 0.0587248322147651,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.0925,
      "mean_token_accuracy": 0.10655909925699233,
      "num_tokens": 1224485.0,
      "step": 665
    },
    {
      "entropy": 7.283737277984619,
      "epoch": 0.05916637230660544,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.099,
      "mean_token_accuracy": 0.10594365671277046,
      "num_tokens": 1233560.0,
      "step": 670
    },
    {
      "entropy": 7.223735809326172,
      "epoch": 0.05960791239844578,
      "grad_norm": 1.34375,
      "learning_rate": 0.000337,
      "loss": 7.0297,
      "mean_token_accuracy": 0.1108290746808052,
      "num_tokens": 1242628.0,
      "step": 675
    },
    {
      "entropy": 7.167131948471069,
      "epoch": 0.06004945249028612,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003395,
      "loss": 6.9188,
      "mean_token_accuracy": 0.11332646831870079,
      "num_tokens": 1251004.0,
      "step": 680
    },
    {
      "entropy": 7.260364675521851,
      "epoch": 0.06049099258212646,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000342,
      "loss": 7.1323,
      "mean_token_accuracy": 0.1082038976252079,
      "num_tokens": 1260344.0,
      "step": 685
    },
    {
      "entropy": 7.175093746185302,
      "epoch": 0.0609325326739668,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.0298,
      "mean_token_accuracy": 0.11300270706415176,
      "num_tokens": 1269988.0,
      "step": 690
    },
    {
      "entropy": 7.1856037139892575,
      "epoch": 0.061374072765807135,
      "grad_norm": 1.21875,
      "learning_rate": 0.000347,
      "loss": 7.004,
      "mean_token_accuracy": 0.11544976383447647,
      "num_tokens": 1280912.0,
      "step": 695
    },
    {
      "entropy": 7.1747633934021,
      "epoch": 0.06181561285764747,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003495,
      "loss": 7.0176,
      "mean_token_accuracy": 0.1107965774834156,
      "num_tokens": 1289684.0,
      "step": 700
    },
    {
      "entropy": 7.233529663085937,
      "epoch": 0.06225715294948781,
      "grad_norm": 1.109375,
      "learning_rate": 0.000352,
      "loss": 7.0457,
      "mean_token_accuracy": 0.10817870572209358,
      "num_tokens": 1298853.0,
      "step": 705
    },
    {
      "entropy": 7.151789426803589,
      "epoch": 0.06269869304132815,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003545,
      "loss": 6.9907,
      "mean_token_accuracy": 0.11804186925292015,
      "num_tokens": 1309112.0,
      "step": 710
    },
    {
      "entropy": 7.147295379638672,
      "epoch": 0.0631402331331685,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000357,
      "loss": 6.9855,
      "mean_token_accuracy": 0.11006599962711335,
      "num_tokens": 1319064.0,
      "step": 715
    },
    {
      "entropy": 7.0800573348999025,
      "epoch": 0.06358177322500883,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003595,
      "loss": 6.9575,
      "mean_token_accuracy": 0.12022612541913986,
      "num_tokens": 1327889.0,
      "step": 720
    },
    {
      "entropy": 7.178079271316529,
      "epoch": 0.06402331331684917,
      "grad_norm": 1.1875,
      "learning_rate": 0.000362,
      "loss": 6.9869,
      "mean_token_accuracy": 0.10886182188987732,
      "num_tokens": 1337241.0,
      "step": 725
    },
    {
      "entropy": 7.1545867919921875,
      "epoch": 0.06446485340868952,
      "grad_norm": 1.125,
      "learning_rate": 0.0003645,
      "loss": 7.0356,
      "mean_token_accuracy": 0.11060537174344062,
      "num_tokens": 1346527.0,
      "step": 730
    },
    {
      "entropy": 7.176617908477783,
      "epoch": 0.06490639350052985,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000367,
      "loss": 7.036,
      "mean_token_accuracy": 0.11137905716896057,
      "num_tokens": 1355677.0,
      "step": 735
    },
    {
      "entropy": 7.172519874572754,
      "epoch": 0.06534793359237019,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003695,
      "loss": 7.0174,
      "mean_token_accuracy": 0.11485954448580742,
      "num_tokens": 1364874.0,
      "step": 740
    },
    {
      "entropy": 7.176022386550903,
      "epoch": 0.06578947368421052,
      "grad_norm": 1.265625,
      "learning_rate": 0.000372,
      "loss": 6.9532,
      "mean_token_accuracy": 0.11629517674446106,
      "num_tokens": 1373717.0,
      "step": 745
    },
    {
      "entropy": 7.189621257781982,
      "epoch": 0.06623101377605087,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003745,
      "loss": 6.9034,
      "mean_token_accuracy": 0.11833730638027191,
      "num_tokens": 1382767.0,
      "step": 750
    },
    {
      "entropy": 7.006224584579468,
      "epoch": 0.0666725538678912,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000377,
      "loss": 6.96,
      "mean_token_accuracy": 0.11204482167959214,
      "num_tokens": 1391190.0,
      "step": 755
    },
    {
      "entropy": 7.079714679718018,
      "epoch": 0.06711409395973154,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003795,
      "loss": 6.9625,
      "mean_token_accuracy": 0.11262108534574508,
      "num_tokens": 1400722.0,
      "step": 760
    },
    {
      "entropy": 7.229566478729248,
      "epoch": 0.06755563405157189,
      "grad_norm": 1.171875,
      "learning_rate": 0.000382,
      "loss": 6.9853,
      "mean_token_accuracy": 0.1128837876021862,
      "num_tokens": 1409328.0,
      "step": 765
    },
    {
      "entropy": 6.942206907272339,
      "epoch": 0.06799717414341222,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003845,
      "loss": 6.8585,
      "mean_token_accuracy": 0.12095727175474166,
      "num_tokens": 1418504.0,
      "step": 770
    },
    {
      "entropy": 7.204732608795166,
      "epoch": 0.06843871423525257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.01,
      "mean_token_accuracy": 0.10958118066191673,
      "num_tokens": 1427690.0,
      "step": 775
    },
    {
      "entropy": 7.102281284332276,
      "epoch": 0.0688802543270929,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.9896,
      "mean_token_accuracy": 0.11304903626441956,
      "num_tokens": 1436798.0,
      "step": 780
    },
    {
      "entropy": 7.056014490127564,
      "epoch": 0.06932179441893324,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.9927,
      "mean_token_accuracy": 0.11028492376208306,
      "num_tokens": 1446357.0,
      "step": 785
    },
    {
      "entropy": 7.063942003250122,
      "epoch": 0.06976333451077357,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.9007,
      "mean_token_accuracy": 0.11615959778428078,
      "num_tokens": 1455998.0,
      "step": 790
    },
    {
      "entropy": 7.0978930473327635,
      "epoch": 0.07020487460261392,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.9697,
      "mean_token_accuracy": 0.10868826881051064,
      "num_tokens": 1465237.0,
      "step": 795
    },
    {
      "entropy": 7.081945991516113,
      "epoch": 0.07064641469445426,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003995,
      "loss": 6.9174,
      "mean_token_accuracy": 0.12127232849597931,
      "num_tokens": 1474363.0,
      "step": 800
    },
    {
      "entropy": 7.076627969741821,
      "epoch": 0.0710879547862946,
      "grad_norm": 1.390625,
      "learning_rate": 0.000402,
      "loss": 6.9709,
      "mean_token_accuracy": 0.11975041329860688,
      "num_tokens": 1483379.0,
      "step": 805
    },
    {
      "entropy": 7.171514463424683,
      "epoch": 0.07152949487813494,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004045,
      "loss": 6.972,
      "mean_token_accuracy": 0.11165242791175842,
      "num_tokens": 1492507.0,
      "step": 810
    },
    {
      "entropy": 6.919444894790649,
      "epoch": 0.07197103496997527,
      "grad_norm": 1.25,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.8347,
      "mean_token_accuracy": 0.12047068104147911,
      "num_tokens": 1500888.0,
      "step": 815
    },
    {
      "entropy": 7.032200956344605,
      "epoch": 0.07241257506181561,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004095,
      "loss": 6.9006,
      "mean_token_accuracy": 0.11496629267930984,
      "num_tokens": 1510310.0,
      "step": 820
    },
    {
      "entropy": 7.029713153839111,
      "epoch": 0.07285411515365595,
      "grad_norm": 1.140625,
      "learning_rate": 0.000412,
      "loss": 6.8871,
      "mean_token_accuracy": 0.1200887382030487,
      "num_tokens": 1519427.0,
      "step": 825
    },
    {
      "entropy": 7.0805525302886965,
      "epoch": 0.07329565524549629,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004145,
      "loss": 6.9008,
      "mean_token_accuracy": 0.11543082296848298,
      "num_tokens": 1529456.0,
      "step": 830
    },
    {
      "entropy": 6.988912868499756,
      "epoch": 0.07373719533733664,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000417,
      "loss": 6.8479,
      "mean_token_accuracy": 0.12111616134643555,
      "num_tokens": 1537695.0,
      "step": 835
    },
    {
      "entropy": 6.990428161621094,
      "epoch": 0.07417873542917697,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004195,
      "loss": 6.9813,
      "mean_token_accuracy": 0.1161614142358303,
      "num_tokens": 1547511.0,
      "step": 840
    },
    {
      "entropy": 7.147986602783203,
      "epoch": 0.07462027552101731,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000422,
      "loss": 6.9878,
      "mean_token_accuracy": 0.11540340334177017,
      "num_tokens": 1557035.0,
      "step": 845
    },
    {
      "entropy": 6.997569465637207,
      "epoch": 0.07506181561285764,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004245,
      "loss": 6.9226,
      "mean_token_accuracy": 0.12383709400892258,
      "num_tokens": 1566773.0,
      "step": 850
    },
    {
      "entropy": 6.967204141616821,
      "epoch": 0.07550335570469799,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000427,
      "loss": 6.9218,
      "mean_token_accuracy": 0.11128390282392502,
      "num_tokens": 1576873.0,
      "step": 855
    },
    {
      "entropy": 7.097699403762817,
      "epoch": 0.07594489579653832,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004295,
      "loss": 6.9533,
      "mean_token_accuracy": 0.10889460295438766,
      "num_tokens": 1586170.0,
      "step": 860
    },
    {
      "entropy": 7.129969358444214,
      "epoch": 0.07638643588837866,
      "grad_norm": 1.21875,
      "learning_rate": 0.000432,
      "loss": 6.9899,
      "mean_token_accuracy": 0.11445967480540276,
      "num_tokens": 1596054.0,
      "step": 865
    },
    {
      "entropy": 6.944463539123535,
      "epoch": 0.07682797598021901,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004345,
      "loss": 6.8683,
      "mean_token_accuracy": 0.1222694993019104,
      "num_tokens": 1604544.0,
      "step": 870
    },
    {
      "entropy": 7.091968202590943,
      "epoch": 0.07726951607205934,
      "grad_norm": 1.21875,
      "learning_rate": 0.000437,
      "loss": 6.8804,
      "mean_token_accuracy": 0.12263753786683082,
      "num_tokens": 1614580.0,
      "step": 875
    },
    {
      "entropy": 6.969458436965942,
      "epoch": 0.07771105616389969,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004395,
      "loss": 6.9714,
      "mean_token_accuracy": 0.1178787350654602,
      "num_tokens": 1624701.0,
      "step": 880
    },
    {
      "entropy": 6.956935214996338,
      "epoch": 0.07815259625574002,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000442,
      "loss": 6.9033,
      "mean_token_accuracy": 0.11575518846511841,
      "num_tokens": 1634085.0,
      "step": 885
    },
    {
      "entropy": 6.968543863296508,
      "epoch": 0.07859413634758036,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004445,
      "loss": 6.8836,
      "mean_token_accuracy": 0.12341264188289643,
      "num_tokens": 1643190.0,
      "step": 890
    },
    {
      "entropy": 6.951219654083252,
      "epoch": 0.0790356764394207,
      "grad_norm": 1.078125,
      "learning_rate": 0.000447,
      "loss": 6.9128,
      "mean_token_accuracy": 0.11554477885365486,
      "num_tokens": 1652705.0,
      "step": 895
    },
    {
      "entropy": 7.081302642822266,
      "epoch": 0.07947721653126104,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.9005,
      "mean_token_accuracy": 0.10998637154698372,
      "num_tokens": 1662210.0,
      "step": 900
    },
    {
      "entropy": 6.992254829406738,
      "epoch": 0.07991875662310138,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.9455,
      "mean_token_accuracy": 0.11279257908463478,
      "num_tokens": 1671893.0,
      "step": 905
    },
    {
      "entropy": 7.003918933868408,
      "epoch": 0.08036029671494171,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.9,
      "mean_token_accuracy": 0.11731264442205429,
      "num_tokens": 1681217.0,
      "step": 910
    },
    {
      "entropy": 6.996629953384399,
      "epoch": 0.08080183680678206,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.9166,
      "mean_token_accuracy": 0.1132784716784954,
      "num_tokens": 1690447.0,
      "step": 915
    },
    {
      "entropy": 6.97693395614624,
      "epoch": 0.08124337689862239,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.001,
      "mean_token_accuracy": 0.10703598409891128,
      "num_tokens": 1700355.0,
      "step": 920
    },
    {
      "entropy": 7.006402778625488,
      "epoch": 0.08168491699046274,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000462,
      "loss": 6.8873,
      "mean_token_accuracy": 0.11095858663320542,
      "num_tokens": 1709449.0,
      "step": 925
    },
    {
      "entropy": 6.987693881988525,
      "epoch": 0.08212645708230307,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004645,
      "loss": 6.9375,
      "mean_token_accuracy": 0.1106826864182949,
      "num_tokens": 1718838.0,
      "step": 930
    },
    {
      "entropy": 6.954984951019287,
      "epoch": 0.08256799717414341,
      "grad_norm": 1.109375,
      "learning_rate": 0.000467,
      "loss": 6.8892,
      "mean_token_accuracy": 0.11695531234145165,
      "num_tokens": 1728594.0,
      "step": 935
    },
    {
      "entropy": 7.014099025726319,
      "epoch": 0.08300953726598376,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004695,
      "loss": 6.9301,
      "mean_token_accuracy": 0.11335264146327972,
      "num_tokens": 1738814.0,
      "step": 940
    },
    {
      "entropy": 6.966581773757935,
      "epoch": 0.08345107735782409,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000472,
      "loss": 6.8557,
      "mean_token_accuracy": 0.12260387167334556,
      "num_tokens": 1747644.0,
      "step": 945
    },
    {
      "entropy": 6.951725816726684,
      "epoch": 0.08389261744966443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004745,
      "loss": 6.8981,
      "mean_token_accuracy": 0.11759259030222893,
      "num_tokens": 1757489.0,
      "step": 950
    },
    {
      "entropy": 6.9648946762084964,
      "epoch": 0.08433415754150476,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000477,
      "loss": 6.8207,
      "mean_token_accuracy": 0.1227583721280098,
      "num_tokens": 1767546.0,
      "step": 955
    },
    {
      "entropy": 6.863551807403565,
      "epoch": 0.08477569763334511,
      "grad_norm": 1.5,
      "learning_rate": 0.0004795,
      "loss": 6.8835,
      "mean_token_accuracy": 0.11026808023452758,
      "num_tokens": 1776035.0,
      "step": 960
    },
    {
      "entropy": 7.006232595443725,
      "epoch": 0.08521723772518544,
      "grad_norm": 0.90625,
      "learning_rate": 0.000482,
      "loss": 6.9522,
      "mean_token_accuracy": 0.11547720953822135,
      "num_tokens": 1786161.0,
      "step": 965
    },
    {
      "entropy": 7.080778980255127,
      "epoch": 0.08565877781702579,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004845,
      "loss": 7.0038,
      "mean_token_accuracy": 0.11200195625424385,
      "num_tokens": 1796093.0,
      "step": 970
    },
    {
      "entropy": 6.917213439941406,
      "epoch": 0.08610031790886613,
      "grad_norm": 1.03125,
      "learning_rate": 0.000487,
      "loss": 6.8549,
      "mean_token_accuracy": 0.11064035892486572,
      "num_tokens": 1805574.0,
      "step": 975
    },
    {
      "entropy": 6.922301959991455,
      "epoch": 0.08654185800070646,
      "grad_norm": 1.125,
      "learning_rate": 0.0004895,
      "loss": 6.8877,
      "mean_token_accuracy": 0.11633643060922623,
      "num_tokens": 1815175.0,
      "step": 980
    },
    {
      "entropy": 6.897391080856323,
      "epoch": 0.08698339809254681,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000492,
      "loss": 6.8116,
      "mean_token_accuracy": 0.11961451843380928,
      "num_tokens": 1824683.0,
      "step": 985
    },
    {
      "entropy": 6.917155456542969,
      "epoch": 0.08742493818438714,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004945,
      "loss": 6.7871,
      "mean_token_accuracy": 0.1191464401781559,
      "num_tokens": 1833174.0,
      "step": 990
    },
    {
      "entropy": 6.89791784286499,
      "epoch": 0.08786647827622748,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000497,
      "loss": 6.9069,
      "mean_token_accuracy": 0.1123793788254261,
      "num_tokens": 1842409.0,
      "step": 995
    },
    {
      "entropy": 6.737509679794312,
      "epoch": 0.08830801836806781,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995,
      "loss": 6.8022,
      "mean_token_accuracy": 0.11974574252963066,
      "num_tokens": 1851441.0,
      "step": 1000
    },
    {
      "entropy": 7.128527450561523,
      "epoch": 0.08874955845990816,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499999998589561,
      "loss": 6.9312,
      "mean_token_accuracy": 0.1117840752005577,
      "num_tokens": 1861188.0,
      "step": 1005
    },
    {
      "entropy": 6.797428703308105,
      "epoch": 0.0891910985517485,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999999928596523,
      "loss": 6.8707,
      "mean_token_accuracy": 0.11862308084964752,
      "num_tokens": 1870284.0,
      "step": 1010
    },
    {
      "entropy": 6.825656938552856,
      "epoch": 0.08963263864358884,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999999827221219,
      "loss": 6.8224,
      "mean_token_accuracy": 0.12426056489348411,
      "num_tokens": 1879744.0,
      "step": 1015
    },
    {
      "entropy": 6.9824902534484865,
      "epoch": 0.09007417873542918,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999999681769696,
      "loss": 6.8388,
      "mean_token_accuracy": 0.11695261672139168,
      "num_tokens": 1889241.0,
      "step": 1020
    },
    {
      "entropy": 6.868342161178589,
      "epoch": 0.09051571882726951,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499999949224196,
      "loss": 6.8452,
      "mean_token_accuracy": 0.11889938712120056,
      "num_tokens": 1898247.0,
      "step": 1025
    },
    {
      "entropy": 6.884699106216431,
      "epoch": 0.09095725891910986,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999999258638013,
      "loss": 6.7941,
      "mean_token_accuracy": 0.11864137351512909,
      "num_tokens": 1907559.0,
      "step": 1030
    },
    {
      "entropy": 6.788152074813842,
      "epoch": 0.09139879901095019,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999998980957861,
      "loss": 6.8008,
      "mean_token_accuracy": 0.12519625276327134,
      "num_tokens": 1917072.0,
      "step": 1035
    },
    {
      "entropy": 6.877279567718506,
      "epoch": 0.09184033910279053,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999998659201508,
      "loss": 6.7593,
      "mean_token_accuracy": 0.12310572415590286,
      "num_tokens": 1926597.0,
      "step": 1040
    },
    {
      "entropy": 6.864975309371948,
      "epoch": 0.09228187919463088,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999998293368961,
      "loss": 6.8101,
      "mean_token_accuracy": 0.11514781787991524,
      "num_tokens": 1935978.0,
      "step": 1045
    },
    {
      "entropy": 6.792648220062256,
      "epoch": 0.09272341928647121,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999997883460227,
      "loss": 6.7612,
      "mean_token_accuracy": 0.12067044004797936,
      "num_tokens": 1944424.0,
      "step": 1050
    },
    {
      "entropy": 6.8677427768707275,
      "epoch": 0.09316495937831155,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999997429475314,
      "loss": 6.7275,
      "mean_token_accuracy": 0.12832702696323395,
      "num_tokens": 1953844.0,
      "step": 1055
    },
    {
      "entropy": 6.81104793548584,
      "epoch": 0.09360649947015189,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999996931414232,
      "loss": 6.8238,
      "mean_token_accuracy": 0.12011751234531402,
      "num_tokens": 1963974.0,
      "step": 1060
    },
    {
      "entropy": 6.822019290924072,
      "epoch": 0.09404803956199223,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999996389276988,
      "loss": 6.8032,
      "mean_token_accuracy": 0.1252795398235321,
      "num_tokens": 1973466.0,
      "step": 1065
    },
    {
      "entropy": 6.78282470703125,
      "epoch": 0.09448957965383256,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999995803063596,
      "loss": 6.8039,
      "mean_token_accuracy": 0.12010984346270562,
      "num_tokens": 1983478.0,
      "step": 1070
    },
    {
      "entropy": 6.908577823638916,
      "epoch": 0.0949311197456729,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999995172774065,
      "loss": 6.8817,
      "mean_token_accuracy": 0.11865456476807594,
      "num_tokens": 1992775.0,
      "step": 1075
    },
    {
      "entropy": 6.911471366882324,
      "epoch": 0.09537265983751325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999994498408408,
      "loss": 6.8435,
      "mean_token_accuracy": 0.11468391343951226,
      "num_tokens": 2002526.0,
      "step": 1080
    },
    {
      "entropy": 6.848548507690429,
      "epoch": 0.09581419992935358,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999993779966639,
      "loss": 6.8681,
      "mean_token_accuracy": 0.11680729538202286,
      "num_tokens": 2012476.0,
      "step": 1085
    },
    {
      "entropy": 6.714018678665161,
      "epoch": 0.09625574002119393,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999993017448771,
      "loss": 6.707,
      "mean_token_accuracy": 0.13380512073636056,
      "num_tokens": 2021252.0,
      "step": 1090
    },
    {
      "entropy": 6.966219282150268,
      "epoch": 0.09669728011303426,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999992210854821,
      "loss": 6.8432,
      "mean_token_accuracy": 0.1207717202603817,
      "num_tokens": 2031438.0,
      "step": 1095
    },
    {
      "entropy": 6.852664613723755,
      "epoch": 0.0971388202048746,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999991360184801,
      "loss": 6.9024,
      "mean_token_accuracy": 0.11384126916527748,
      "num_tokens": 2041319.0,
      "step": 1100
    },
    {
      "entropy": 6.858384132385254,
      "epoch": 0.09758036029671494,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999990465438731,
      "loss": 6.8825,
      "mean_token_accuracy": 0.11868280768394471,
      "num_tokens": 2052060.0,
      "step": 1105
    },
    {
      "entropy": 6.8333351612091064,
      "epoch": 0.09802190038855528,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999989526616628,
      "loss": 6.7546,
      "mean_token_accuracy": 0.12631451785564424,
      "num_tokens": 2061331.0,
      "step": 1110
    },
    {
      "entropy": 6.765726852416992,
      "epoch": 0.09846344048039563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999988543718509,
      "loss": 6.7792,
      "mean_token_accuracy": 0.11744962036609649,
      "num_tokens": 2070006.0,
      "step": 1115
    },
    {
      "entropy": 6.895434951782226,
      "epoch": 0.09890498057223596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999987516744394,
      "loss": 6.7292,
      "mean_token_accuracy": 0.12839764803647996,
      "num_tokens": 2079089.0,
      "step": 1120
    },
    {
      "entropy": 6.732869672775268,
      "epoch": 0.0993465206640763,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999986445694303,
      "loss": 6.7255,
      "mean_token_accuracy": 0.12237351760268211,
      "num_tokens": 2087237.0,
      "step": 1125
    },
    {
      "entropy": 6.71789960861206,
      "epoch": 0.09978806075591663,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999985330568258,
      "loss": 6.6987,
      "mean_token_accuracy": 0.1242702692747116,
      "num_tokens": 2096598.0,
      "step": 1130
    },
    {
      "entropy": 6.881210708618164,
      "epoch": 0.10022960084775698,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999984171366278,
      "loss": 6.7841,
      "mean_token_accuracy": 0.1141000434756279,
      "num_tokens": 2106106.0,
      "step": 1135
    },
    {
      "entropy": 6.750670385360718,
      "epoch": 0.10067114093959731,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499998296808839,
      "loss": 6.7573,
      "mean_token_accuracy": 0.12421151250600815,
      "num_tokens": 2115478.0,
      "step": 1140
    },
    {
      "entropy": 6.846238422393799,
      "epoch": 0.10111268103143765,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004999981720734615,
      "loss": 6.7956,
      "mean_token_accuracy": 0.12258475199341774,
      "num_tokens": 2124439.0,
      "step": 1145
    },
    {
      "entropy": 6.845457935333252,
      "epoch": 0.101554221123278,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999980429304977,
      "loss": 6.8113,
      "mean_token_accuracy": 0.1178690455853939,
      "num_tokens": 2133718.0,
      "step": 1150
    },
    {
      "entropy": 6.688090801239014,
      "epoch": 0.10199576121511833,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999979093799502,
      "loss": 6.7085,
      "mean_token_accuracy": 0.12130860090255738,
      "num_tokens": 2142978.0,
      "step": 1155
    },
    {
      "entropy": 6.7933430671691895,
      "epoch": 0.10243730130695868,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999977714218217,
      "loss": 6.7791,
      "mean_token_accuracy": 0.11885071843862534,
      "num_tokens": 2152250.0,
      "step": 1160
    },
    {
      "entropy": 6.853244686126709,
      "epoch": 0.102878841398799,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999976290561147,
      "loss": 6.7725,
      "mean_token_accuracy": 0.1168168194591999,
      "num_tokens": 2161620.0,
      "step": 1165
    },
    {
      "entropy": 6.816542768478394,
      "epoch": 0.10332038149063935,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999974822828322,
      "loss": 6.7835,
      "mean_token_accuracy": 0.12128284797072411,
      "num_tokens": 2170856.0,
      "step": 1170
    },
    {
      "entropy": 6.761290454864502,
      "epoch": 0.10376192158247968,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499997331101977,
      "loss": 6.7743,
      "mean_token_accuracy": 0.11757255122065544,
      "num_tokens": 2180926.0,
      "step": 1175
    },
    {
      "entropy": 6.82963809967041,
      "epoch": 0.10420346167432003,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499997175513552,
      "loss": 6.7584,
      "mean_token_accuracy": 0.11870861351490021,
      "num_tokens": 2190248.0,
      "step": 1180
    },
    {
      "entropy": 6.8007221698760985,
      "epoch": 0.10464500176616037,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999970155175603,
      "loss": 6.7745,
      "mean_token_accuracy": 0.11943883895874023,
      "num_tokens": 2199833.0,
      "step": 1185
    },
    {
      "entropy": 6.701777648925781,
      "epoch": 0.1050865418580007,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499996851114005,
      "loss": 6.7307,
      "mean_token_accuracy": 0.12810362204909326,
      "num_tokens": 2208240.0,
      "step": 1190
    },
    {
      "entropy": 6.828345394134521,
      "epoch": 0.10552808194984105,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999966823028894,
      "loss": 6.7824,
      "mean_token_accuracy": 0.11475062593817711,
      "num_tokens": 2218758.0,
      "step": 1195
    },
    {
      "entropy": 6.8126058101654055,
      "epoch": 0.10596962204168138,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999965090842168,
      "loss": 6.7898,
      "mean_token_accuracy": 0.12589357271790505,
      "num_tokens": 2228443.0,
      "step": 1200
    },
    {
      "entropy": 6.724305534362793,
      "epoch": 0.10641116213352173,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999963314579905,
      "loss": 6.6237,
      "mean_token_accuracy": 0.13539545610547066,
      "num_tokens": 2236787.0,
      "step": 1205
    },
    {
      "entropy": 6.805136013031006,
      "epoch": 0.10685270222536206,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999961494242139,
      "loss": 6.8017,
      "mean_token_accuracy": 0.11990576833486558,
      "num_tokens": 2247089.0,
      "step": 1210
    },
    {
      "entropy": 6.714448690414429,
      "epoch": 0.1072942423172024,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999959629828908,
      "loss": 6.7348,
      "mean_token_accuracy": 0.11736459657549858,
      "num_tokens": 2256045.0,
      "step": 1215
    },
    {
      "entropy": 6.793351697921753,
      "epoch": 0.10773578240904275,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999957721340248,
      "loss": 6.7883,
      "mean_token_accuracy": 0.12527842447161674,
      "num_tokens": 2265835.0,
      "step": 1220
    },
    {
      "entropy": 6.671676206588745,
      "epoch": 0.10817732250088308,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999955768776195,
      "loss": 6.7241,
      "mean_token_accuracy": 0.12204784303903579,
      "num_tokens": 2275318.0,
      "step": 1225
    },
    {
      "entropy": 6.7978355884552,
      "epoch": 0.10861886259272342,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999953772136788,
      "loss": 6.7242,
      "mean_token_accuracy": 0.12499865740537644,
      "num_tokens": 2284821.0,
      "step": 1230
    },
    {
      "entropy": 6.821506834030151,
      "epoch": 0.10906040268456375,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999951731422068,
      "loss": 6.6958,
      "mean_token_accuracy": 0.12446941509842872,
      "num_tokens": 2294013.0,
      "step": 1235
    },
    {
      "entropy": 6.718976879119873,
      "epoch": 0.1095019427764041,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999949646632072,
      "loss": 6.6939,
      "mean_token_accuracy": 0.12443441674113273,
      "num_tokens": 2302727.0,
      "step": 1240
    },
    {
      "entropy": 6.7134024143219,
      "epoch": 0.10994348286824443,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999947517766842,
      "loss": 6.7145,
      "mean_token_accuracy": 0.12263145819306373,
      "num_tokens": 2312032.0,
      "step": 1245
    },
    {
      "entropy": 6.909180307388306,
      "epoch": 0.11038502296008477,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499994534482642,
      "loss": 6.8067,
      "mean_token_accuracy": 0.11781661435961724,
      "num_tokens": 2321839.0,
      "step": 1250
    },
    {
      "entropy": 6.693206644058227,
      "epoch": 0.11082656305192512,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999943127810847,
      "loss": 6.7725,
      "mean_token_accuracy": 0.11574066355824471,
      "num_tokens": 2331255.0,
      "step": 1255
    },
    {
      "entropy": 6.676486921310425,
      "epoch": 0.11126810314376545,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999940866720169,
      "loss": 6.598,
      "mean_token_accuracy": 0.133713548630476,
      "num_tokens": 2340038.0,
      "step": 1260
    },
    {
      "entropy": 6.7074402332305905,
      "epoch": 0.1117096432356058,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999938561554429,
      "loss": 6.7001,
      "mean_token_accuracy": 0.12409981712698936,
      "num_tokens": 2348901.0,
      "step": 1265
    },
    {
      "entropy": 6.729129076004028,
      "epoch": 0.11215118332744613,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999936212313672,
      "loss": 6.7966,
      "mean_token_accuracy": 0.11647377163171768,
      "num_tokens": 2358842.0,
      "step": 1270
    },
    {
      "entropy": 6.7503026008605955,
      "epoch": 0.11259272341928647,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999933818997943,
      "loss": 6.6867,
      "mean_token_accuracy": 0.12584296390414237,
      "num_tokens": 2368650.0,
      "step": 1275
    },
    {
      "entropy": 6.747146368026733,
      "epoch": 0.1130342635111268,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999931381607292,
      "loss": 6.7419,
      "mean_token_accuracy": 0.12453764751553535,
      "num_tokens": 2377916.0,
      "step": 1280
    },
    {
      "entropy": 6.746145820617675,
      "epoch": 0.11347580360296715,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999928900141764,
      "loss": 6.6865,
      "mean_token_accuracy": 0.12603561356663703,
      "num_tokens": 2387507.0,
      "step": 1285
    },
    {
      "entropy": 6.6909308433532715,
      "epoch": 0.1139173436948075,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499992637460141,
      "loss": 6.7423,
      "mean_token_accuracy": 0.1275649629533291,
      "num_tokens": 2396148.0,
      "step": 1290
    },
    {
      "entropy": 6.7133759498596195,
      "epoch": 0.11435888378664782,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999923804986275,
      "loss": 6.6046,
      "mean_token_accuracy": 0.12199260294437408,
      "num_tokens": 2404891.0,
      "step": 1295
    },
    {
      "entropy": 6.716338968276977,
      "epoch": 0.11480042387848817,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999921191296415,
      "loss": 6.6426,
      "mean_token_accuracy": 0.12343374639749527,
      "num_tokens": 2414406.0,
      "step": 1300
    },
    {
      "entropy": 6.697209453582763,
      "epoch": 0.1152419639703285,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999918533531877,
      "loss": 6.7299,
      "mean_token_accuracy": 0.127056036144495,
      "num_tokens": 2424363.0,
      "step": 1305
    },
    {
      "entropy": 6.757014656066895,
      "epoch": 0.11568350406216885,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999915831692714,
      "loss": 6.6775,
      "mean_token_accuracy": 0.12560488805174827,
      "num_tokens": 2433753.0,
      "step": 1310
    },
    {
      "entropy": 6.593924283981323,
      "epoch": 0.11612504415400918,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999913085778981,
      "loss": 6.6861,
      "mean_token_accuracy": 0.12244237959384918,
      "num_tokens": 2443275.0,
      "step": 1315
    },
    {
      "entropy": 6.822639799118042,
      "epoch": 0.11656658424584952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999910295790729,
      "loss": 6.7075,
      "mean_token_accuracy": 0.12074168473482132,
      "num_tokens": 2452510.0,
      "step": 1320
    },
    {
      "entropy": 6.731257343292237,
      "epoch": 0.11700812433768987,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999907461728014,
      "loss": 6.8117,
      "mean_token_accuracy": 0.11980628594756126,
      "num_tokens": 2462742.0,
      "step": 1325
    },
    {
      "entropy": 6.71476149559021,
      "epoch": 0.1174496644295302,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999904583590893,
      "loss": 6.6781,
      "mean_token_accuracy": 0.12219401150941848,
      "num_tokens": 2471409.0,
      "step": 1330
    },
    {
      "entropy": 6.749363136291504,
      "epoch": 0.11789120452137054,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999901661379418,
      "loss": 6.6228,
      "mean_token_accuracy": 0.13008193373680116,
      "num_tokens": 2481011.0,
      "step": 1335
    },
    {
      "entropy": 6.63396897315979,
      "epoch": 0.11833274461321087,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999898695093652,
      "loss": 6.7195,
      "mean_token_accuracy": 0.12151179760694504,
      "num_tokens": 2490664.0,
      "step": 1340
    },
    {
      "entropy": 6.728178644180298,
      "epoch": 0.11877428470505122,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999895684733648,
      "loss": 6.6462,
      "mean_token_accuracy": 0.1303670547902584,
      "num_tokens": 2499799.0,
      "step": 1345
    },
    {
      "entropy": 6.716060447692871,
      "epoch": 0.11921582479689156,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004999892630299467,
      "loss": 6.6483,
      "mean_token_accuracy": 0.12420448064804077,
      "num_tokens": 2508780.0,
      "step": 1350
    },
    {
      "entropy": 6.657323360443115,
      "epoch": 0.1196573648887319,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999889531791171,
      "loss": 6.6462,
      "mean_token_accuracy": 0.12205975502729416,
      "num_tokens": 2517741.0,
      "step": 1355
    },
    {
      "entropy": 6.71348876953125,
      "epoch": 0.12009890498057224,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999886389208817,
      "loss": 6.7302,
      "mean_token_accuracy": 0.12077968642115593,
      "num_tokens": 2528742.0,
      "step": 1360
    },
    {
      "entropy": 6.5724108695983885,
      "epoch": 0.12054044507241257,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999883202552468,
      "loss": 6.6741,
      "mean_token_accuracy": 0.1292316846549511,
      "num_tokens": 2538609.0,
      "step": 1365
    },
    {
      "entropy": 6.831192445755005,
      "epoch": 0.12098198516425292,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999879971822189,
      "loss": 6.6579,
      "mean_token_accuracy": 0.12047442719340325,
      "num_tokens": 2547772.0,
      "step": 1370
    },
    {
      "entropy": 6.59716944694519,
      "epoch": 0.12142352525609325,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999876697018038,
      "loss": 6.6253,
      "mean_token_accuracy": 0.12871280014514924,
      "num_tokens": 2556114.0,
      "step": 1375
    },
    {
      "entropy": 6.76294059753418,
      "epoch": 0.1218650653479336,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999873378140085,
      "loss": 6.6682,
      "mean_token_accuracy": 0.1281101770699024,
      "num_tokens": 2566814.0,
      "step": 1380
    },
    {
      "entropy": 6.600197601318359,
      "epoch": 0.12230660543977394,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999870015188389,
      "loss": 6.6317,
      "mean_token_accuracy": 0.12663180232048035,
      "num_tokens": 2576030.0,
      "step": 1385
    },
    {
      "entropy": 6.6567483901977536,
      "epoch": 0.12274814553161427,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999866608163021,
      "loss": 6.6546,
      "mean_token_accuracy": 0.12772715091705322,
      "num_tokens": 2585756.0,
      "step": 1390
    },
    {
      "entropy": 6.739302158355713,
      "epoch": 0.12318968562345461,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999863157064045,
      "loss": 6.6928,
      "mean_token_accuracy": 0.12649391070008278,
      "num_tokens": 2595676.0,
      "step": 1395
    },
    {
      "entropy": 6.681546878814697,
      "epoch": 0.12363122571529495,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999859661891529,
      "loss": 6.7124,
      "mean_token_accuracy": 0.12418720349669457,
      "num_tokens": 2606197.0,
      "step": 1400
    },
    {
      "entropy": 6.692468786239624,
      "epoch": 0.12407276580713529,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999856122645543,
      "loss": 6.6308,
      "mean_token_accuracy": 0.12420927360653877,
      "num_tokens": 2615311.0,
      "step": 1405
    },
    {
      "entropy": 6.625854730606079,
      "epoch": 0.12451430589897562,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999852539326154,
      "loss": 6.551,
      "mean_token_accuracy": 0.1275542564690113,
      "num_tokens": 2624074.0,
      "step": 1410
    },
    {
      "entropy": 6.581278514862061,
      "epoch": 0.12495584599081597,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999848911933434,
      "loss": 6.6894,
      "mean_token_accuracy": 0.12939536869525908,
      "num_tokens": 2633877.0,
      "step": 1415
    },
    {
      "entropy": 6.65780930519104,
      "epoch": 0.1253973860826563,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999845240467453,
      "loss": 6.6164,
      "mean_token_accuracy": 0.12634376361966132,
      "num_tokens": 2643330.0,
      "step": 1420
    },
    {
      "entropy": 6.688624095916748,
      "epoch": 0.12583892617449666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999841524928282,
      "loss": 6.5957,
      "mean_token_accuracy": 0.130678478628397,
      "num_tokens": 2652070.0,
      "step": 1425
    },
    {
      "entropy": 6.651975011825561,
      "epoch": 0.126280466266337,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999837765315997,
      "loss": 6.6467,
      "mean_token_accuracy": 0.13028737679123878,
      "num_tokens": 2660546.0,
      "step": 1430
    },
    {
      "entropy": 6.5479024887084964,
      "epoch": 0.12672200635817732,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999833961630669,
      "loss": 6.5377,
      "mean_token_accuracy": 0.12899447232484818,
      "num_tokens": 2669938.0,
      "step": 1435
    },
    {
      "entropy": 6.7292591571807865,
      "epoch": 0.12716354645001765,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999830113872374,
      "loss": 6.6675,
      "mean_token_accuracy": 0.12595592364668845,
      "num_tokens": 2679814.0,
      "step": 1440
    },
    {
      "entropy": 6.554772853851318,
      "epoch": 0.127605086541858,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999826222041186,
      "loss": 6.571,
      "mean_token_accuracy": 0.13299084976315498,
      "num_tokens": 2688733.0,
      "step": 1445
    },
    {
      "entropy": 6.636107873916626,
      "epoch": 0.12804662663369834,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999822286137182,
      "loss": 6.5779,
      "mean_token_accuracy": 0.12820412516593932,
      "num_tokens": 2697744.0,
      "step": 1450
    },
    {
      "entropy": 6.6388514041900635,
      "epoch": 0.12848816672553867,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999818306160439,
      "loss": 6.5266,
      "mean_token_accuracy": 0.1340452216565609,
      "num_tokens": 2707037.0,
      "step": 1455
    },
    {
      "entropy": 6.645256471633911,
      "epoch": 0.12892970681737903,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999814282111034,
      "loss": 6.6874,
      "mean_token_accuracy": 0.12385595068335534,
      "num_tokens": 2717345.0,
      "step": 1460
    },
    {
      "entropy": 6.667556524276733,
      "epoch": 0.12937124690921936,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999810213989047,
      "loss": 6.638,
      "mean_token_accuracy": 0.11816325932741165,
      "num_tokens": 2726892.0,
      "step": 1465
    },
    {
      "entropy": 6.621801376342773,
      "epoch": 0.1298127870010597,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999806101794558,
      "loss": 6.6041,
      "mean_token_accuracy": 0.1266481436789036,
      "num_tokens": 2736479.0,
      "step": 1470
    },
    {
      "entropy": 6.701800394058227,
      "epoch": 0.13025432709290002,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999801945527648,
      "loss": 6.6545,
      "mean_token_accuracy": 0.12567580863833427,
      "num_tokens": 2745998.0,
      "step": 1475
    },
    {
      "entropy": 6.620102596282959,
      "epoch": 0.13069586718474038,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999797745188395,
      "loss": 6.641,
      "mean_token_accuracy": 0.1308413989841938,
      "num_tokens": 2754346.0,
      "step": 1480
    },
    {
      "entropy": 6.6592470645904545,
      "epoch": 0.13113740727658071,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999793500776886,
      "loss": 6.5835,
      "mean_token_accuracy": 0.13052424117922784,
      "num_tokens": 2763391.0,
      "step": 1485
    },
    {
      "entropy": 6.62844033241272,
      "epoch": 0.13157894736842105,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999789212293201,
      "loss": 6.6395,
      "mean_token_accuracy": 0.12945614531636238,
      "num_tokens": 2772764.0,
      "step": 1490
    },
    {
      "entropy": 6.712308549880982,
      "epoch": 0.1320204874602614,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999784879737423,
      "loss": 6.7289,
      "mean_token_accuracy": 0.12319694384932518,
      "num_tokens": 2782312.0,
      "step": 1495
    },
    {
      "entropy": 6.650492525100708,
      "epoch": 0.13246202755210174,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999780503109642,
      "loss": 6.6299,
      "mean_token_accuracy": 0.12477036267518997,
      "num_tokens": 2791159.0,
      "step": 1500
    },
    {
      "entropy": 6.561451959609985,
      "epoch": 0.13290356764394207,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999776082409939,
      "loss": 6.4526,
      "mean_token_accuracy": 0.13791982010006903,
      "num_tokens": 2799319.0,
      "step": 1505
    },
    {
      "entropy": 6.534754419326783,
      "epoch": 0.1333451077357824,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999771617638401,
      "loss": 6.582,
      "mean_token_accuracy": 0.13077868148684502,
      "num_tokens": 2807401.0,
      "step": 1510
    },
    {
      "entropy": 6.649140977859497,
      "epoch": 0.13378664782762276,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999767108795118,
      "loss": 6.642,
      "mean_token_accuracy": 0.12535738274455072,
      "num_tokens": 2817734.0,
      "step": 1515
    },
    {
      "entropy": 6.605966520309448,
      "epoch": 0.1342281879194631,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999762555880176,
      "loss": 6.6236,
      "mean_token_accuracy": 0.1270806223154068,
      "num_tokens": 2828235.0,
      "step": 1520
    },
    {
      "entropy": 6.663729429244995,
      "epoch": 0.13466972801130342,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999757958893666,
      "loss": 6.651,
      "mean_token_accuracy": 0.1310298413038254,
      "num_tokens": 2837453.0,
      "step": 1525
    },
    {
      "entropy": 6.549145174026489,
      "epoch": 0.13511126810314378,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999753317835677,
      "loss": 6.6279,
      "mean_token_accuracy": 0.12039558663964271,
      "num_tokens": 2847055.0,
      "step": 1530
    },
    {
      "entropy": 6.707547044754028,
      "epoch": 0.1355528081949841,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999748632706299,
      "loss": 6.611,
      "mean_token_accuracy": 0.13395099639892577,
      "num_tokens": 2857101.0,
      "step": 1535
    },
    {
      "entropy": 6.556164789199829,
      "epoch": 0.13599434828682444,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999743903505626,
      "loss": 6.515,
      "mean_token_accuracy": 0.13099609762430192,
      "num_tokens": 2866685.0,
      "step": 1540
    },
    {
      "entropy": 6.649285745620728,
      "epoch": 0.13643588837866477,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999739130233749,
      "loss": 6.6349,
      "mean_token_accuracy": 0.12779910936951638,
      "num_tokens": 2876022.0,
      "step": 1545
    },
    {
      "entropy": 6.624913215637207,
      "epoch": 0.13687742847050513,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999734312890761,
      "loss": 6.5416,
      "mean_token_accuracy": 0.12898361757397653,
      "num_tokens": 2885560.0,
      "step": 1550
    },
    {
      "entropy": 6.541625452041626,
      "epoch": 0.13731896856234546,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999729451476757,
      "loss": 6.5757,
      "mean_token_accuracy": 0.1289501816034317,
      "num_tokens": 2894686.0,
      "step": 1555
    },
    {
      "entropy": 6.620435476303101,
      "epoch": 0.1377605086541858,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999724545991835,
      "loss": 6.5948,
      "mean_token_accuracy": 0.13569336235523224,
      "num_tokens": 2904390.0,
      "step": 1560
    },
    {
      "entropy": 6.616004848480225,
      "epoch": 0.13820204874602615,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999719596436086,
      "loss": 6.6554,
      "mean_token_accuracy": 0.1249268040060997,
      "num_tokens": 2913311.0,
      "step": 1565
    },
    {
      "entropy": 6.633422946929931,
      "epoch": 0.13864358883786648,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999714602809611,
      "loss": 6.5368,
      "mean_token_accuracy": 0.13287945315241814,
      "num_tokens": 2923196.0,
      "step": 1570
    },
    {
      "entropy": 6.549054431915283,
      "epoch": 0.1390851289297068,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999709565112506,
      "loss": 6.6,
      "mean_token_accuracy": 0.12382271289825439,
      "num_tokens": 2932813.0,
      "step": 1575
    },
    {
      "entropy": 6.6186319351196286,
      "epoch": 0.13952666902154715,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499970448334487,
      "loss": 6.5139,
      "mean_token_accuracy": 0.12139623984694481,
      "num_tokens": 2942694.0,
      "step": 1580
    },
    {
      "entropy": 6.508273315429688,
      "epoch": 0.1399682091133875,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999699357506803,
      "loss": 6.4663,
      "mean_token_accuracy": 0.13404878079891205,
      "num_tokens": 2950932.0,
      "step": 1585
    },
    {
      "entropy": 6.620526647567749,
      "epoch": 0.14040974920522784,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999694187598406,
      "loss": 6.5628,
      "mean_token_accuracy": 0.13192685768008233,
      "num_tokens": 2960102.0,
      "step": 1590
    },
    {
      "entropy": 6.592943286895752,
      "epoch": 0.14085128929706817,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004999688973619777,
      "loss": 6.5609,
      "mean_token_accuracy": 0.1262449197471142,
      "num_tokens": 2969968.0,
      "step": 1595
    },
    {
      "entropy": 6.520618915557861,
      "epoch": 0.14129282938890853,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999683715571022,
      "loss": 6.5212,
      "mean_token_accuracy": 0.13312431499361993,
      "num_tokens": 2978880.0,
      "step": 1600
    },
    {
      "entropy": 6.610848045349121,
      "epoch": 0.14173436948074886,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999678413452242,
      "loss": 6.554,
      "mean_token_accuracy": 0.1260514296591282,
      "num_tokens": 2988369.0,
      "step": 1605
    },
    {
      "entropy": 6.534283065795899,
      "epoch": 0.1421759095725892,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999673067263542,
      "loss": 6.5949,
      "mean_token_accuracy": 0.12671327590942383,
      "num_tokens": 2997070.0,
      "step": 1610
    },
    {
      "entropy": 6.656943464279175,
      "epoch": 0.14261744966442952,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999667677005026,
      "loss": 6.5329,
      "mean_token_accuracy": 0.13323748856782913,
      "num_tokens": 3006547.0,
      "step": 1615
    },
    {
      "entropy": 6.516927480697632,
      "epoch": 0.14305898975626988,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999662242676799,
      "loss": 6.5522,
      "mean_token_accuracy": 0.13190415725111962,
      "num_tokens": 3015821.0,
      "step": 1620
    },
    {
      "entropy": 6.561876678466797,
      "epoch": 0.1435005298481102,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999656764278968,
      "loss": 6.5403,
      "mean_token_accuracy": 0.1307166889309883,
      "num_tokens": 3024750.0,
      "step": 1625
    },
    {
      "entropy": 6.569908905029297,
      "epoch": 0.14394206993995054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999651241811642,
      "loss": 6.4929,
      "mean_token_accuracy": 0.13378000110387803,
      "num_tokens": 3033345.0,
      "step": 1630
    },
    {
      "entropy": 6.47226619720459,
      "epoch": 0.1443836100317909,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999645675274925,
      "loss": 6.4813,
      "mean_token_accuracy": 0.1357632078230381,
      "num_tokens": 3042060.0,
      "step": 1635
    },
    {
      "entropy": 6.586079454421997,
      "epoch": 0.14482515012363123,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999640064668931,
      "loss": 6.624,
      "mean_token_accuracy": 0.1263129860162735,
      "num_tokens": 3052490.0,
      "step": 1640
    },
    {
      "entropy": 6.671514511108398,
      "epoch": 0.14526669021547156,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999634409993766,
      "loss": 6.5901,
      "mean_token_accuracy": 0.12612482234835626,
      "num_tokens": 3061934.0,
      "step": 1645
    },
    {
      "entropy": 6.503233861923218,
      "epoch": 0.1457082303073119,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999628711249544,
      "loss": 6.4934,
      "mean_token_accuracy": 0.13857336044311525,
      "num_tokens": 3070890.0,
      "step": 1650
    },
    {
      "entropy": 6.627433586120605,
      "epoch": 0.14614977039915225,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999622968436373,
      "loss": 6.5164,
      "mean_token_accuracy": 0.13164258152246475,
      "num_tokens": 3079933.0,
      "step": 1655
    },
    {
      "entropy": 6.464597463607788,
      "epoch": 0.14659131049099258,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999617181554369,
      "loss": 6.5699,
      "mean_token_accuracy": 0.13076355457305908,
      "num_tokens": 3089910.0,
      "step": 1660
    },
    {
      "entropy": 6.61089825630188,
      "epoch": 0.1470328505828329,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004999611350603643,
      "loss": 6.5471,
      "mean_token_accuracy": 0.13240002393722533,
      "num_tokens": 3098676.0,
      "step": 1665
    },
    {
      "entropy": 6.586600494384766,
      "epoch": 0.14747439067467327,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499960547558431,
      "loss": 6.6647,
      "mean_token_accuracy": 0.1285034753382206,
      "num_tokens": 3109055.0,
      "step": 1670
    },
    {
      "entropy": 6.58641209602356,
      "epoch": 0.1479159307665136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999599556496486,
      "loss": 6.5113,
      "mean_token_accuracy": 0.13572623655200006,
      "num_tokens": 3117517.0,
      "step": 1675
    },
    {
      "entropy": 6.508319330215454,
      "epoch": 0.14835747085835393,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999593593340286,
      "loss": 6.5022,
      "mean_token_accuracy": 0.13704506903886796,
      "num_tokens": 3126606.0,
      "step": 1680
    },
    {
      "entropy": 6.596501302719116,
      "epoch": 0.14879901095019427,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999587586115826,
      "loss": 6.5136,
      "mean_token_accuracy": 0.13149842694401742,
      "num_tokens": 3135444.0,
      "step": 1685
    },
    {
      "entropy": 6.51367220878601,
      "epoch": 0.14924055104203462,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999581534823226,
      "loss": 6.4727,
      "mean_token_accuracy": 0.13636851012706758,
      "num_tokens": 3144967.0,
      "step": 1690
    },
    {
      "entropy": 6.575021123886108,
      "epoch": 0.14968209113387496,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999575439462601,
      "loss": 6.4868,
      "mean_token_accuracy": 0.1300053507089615,
      "num_tokens": 3153898.0,
      "step": 1695
    },
    {
      "entropy": 6.453029918670654,
      "epoch": 0.1501236312257153,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999569300034075,
      "loss": 6.555,
      "mean_token_accuracy": 0.12720515877008437,
      "num_tokens": 3162341.0,
      "step": 1700
    },
    {
      "entropy": 6.498763990402222,
      "epoch": 0.15056517131755565,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999563116537764,
      "loss": 6.3365,
      "mean_token_accuracy": 0.14578044191002845,
      "num_tokens": 3171420.0,
      "step": 1705
    },
    {
      "entropy": 6.4892902851104735,
      "epoch": 0.15100671140939598,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999556888973792,
      "loss": 6.3966,
      "mean_token_accuracy": 0.1321087971329689,
      "num_tokens": 3180221.0,
      "step": 1710
    },
    {
      "entropy": 6.587397718429566,
      "epoch": 0.1514482515012363,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999550617342279,
      "loss": 6.5511,
      "mean_token_accuracy": 0.12823838815093042,
      "num_tokens": 3189156.0,
      "step": 1715
    },
    {
      "entropy": 6.466203594207764,
      "epoch": 0.15188979159307664,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499954430164335,
      "loss": 6.4738,
      "mean_token_accuracy": 0.13134835958480834,
      "num_tokens": 3199870.0,
      "step": 1720
    },
    {
      "entropy": 6.586092329025268,
      "epoch": 0.152331331684917,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999537941877127,
      "loss": 6.4556,
      "mean_token_accuracy": 0.13151518255472183,
      "num_tokens": 3208815.0,
      "step": 1725
    },
    {
      "entropy": 6.4404736995697025,
      "epoch": 0.15277287177675733,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999531538043735,
      "loss": 6.5381,
      "mean_token_accuracy": 0.13203879222273826,
      "num_tokens": 3218692.0,
      "step": 1730
    },
    {
      "entropy": 6.55628137588501,
      "epoch": 0.15321441186859766,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999525090143298,
      "loss": 6.5023,
      "mean_token_accuracy": 0.14053747728466986,
      "num_tokens": 3227604.0,
      "step": 1735
    },
    {
      "entropy": 6.555537223815918,
      "epoch": 0.15365595196043802,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999518598175946,
      "loss": 6.5109,
      "mean_token_accuracy": 0.13781933188438417,
      "num_tokens": 3237912.0,
      "step": 1740
    },
    {
      "entropy": 6.438277339935302,
      "epoch": 0.15409749205227835,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999512062141805,
      "loss": 6.3744,
      "mean_token_accuracy": 0.13822528719902039,
      "num_tokens": 3246003.0,
      "step": 1745
    },
    {
      "entropy": 6.541148376464844,
      "epoch": 0.15453903214411868,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999505482040999,
      "loss": 6.5237,
      "mean_token_accuracy": 0.13249014243483542,
      "num_tokens": 3256363.0,
      "step": 1750
    },
    {
      "entropy": 6.529036283493042,
      "epoch": 0.154980572235959,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999498857873662,
      "loss": 6.5163,
      "mean_token_accuracy": 0.13468330428004266,
      "num_tokens": 3265822.0,
      "step": 1755
    },
    {
      "entropy": 6.361684417724609,
      "epoch": 0.15542211232779937,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999492189639921,
      "loss": 6.4164,
      "mean_token_accuracy": 0.13653916344046593,
      "num_tokens": 3274614.0,
      "step": 1760
    },
    {
      "entropy": 6.6026630878448485,
      "epoch": 0.1558636524196397,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999485477339907,
      "loss": 6.4873,
      "mean_token_accuracy": 0.13441504836082457,
      "num_tokens": 3283800.0,
      "step": 1765
    },
    {
      "entropy": 6.481634330749512,
      "epoch": 0.15630519251148003,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999478720973753,
      "loss": 6.4147,
      "mean_token_accuracy": 0.1378687009215355,
      "num_tokens": 3293221.0,
      "step": 1770
    },
    {
      "entropy": 6.518844079971314,
      "epoch": 0.1567467326033204,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499947192054159,
      "loss": 6.5621,
      "mean_token_accuracy": 0.1299850597977638,
      "num_tokens": 3302852.0,
      "step": 1775
    },
    {
      "entropy": 6.573839330673218,
      "epoch": 0.15718827269516072,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499946507604355,
      "loss": 6.4284,
      "mean_token_accuracy": 0.13676528111100197,
      "num_tokens": 3311752.0,
      "step": 1780
    },
    {
      "entropy": 6.403056526184082,
      "epoch": 0.15762981278700106,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499945818747977,
      "loss": 6.5135,
      "mean_token_accuracy": 0.1319083459675312,
      "num_tokens": 3321339.0,
      "step": 1785
    },
    {
      "entropy": 6.6045581817626955,
      "epoch": 0.1580713528788414,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999451254850383,
      "loss": 6.5346,
      "mean_token_accuracy": 0.1300017662346363,
      "num_tokens": 3330269.0,
      "step": 1790
    },
    {
      "entropy": 6.480741930007935,
      "epoch": 0.15851289297068175,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999444278155525,
      "loss": 6.4478,
      "mean_token_accuracy": 0.13558795750141145,
      "num_tokens": 3340770.0,
      "step": 1795
    },
    {
      "entropy": 6.4864849090576175,
      "epoch": 0.15895443306252208,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999437257395333,
      "loss": 6.5388,
      "mean_token_accuracy": 0.13693441525101663,
      "num_tokens": 3349976.0,
      "step": 1800
    },
    {
      "entropy": 6.576467275619507,
      "epoch": 0.1593959731543624,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999430192569944,
      "loss": 6.5822,
      "mean_token_accuracy": 0.1267798624932766,
      "num_tokens": 3359764.0,
      "step": 1805
    },
    {
      "entropy": 6.459140777587891,
      "epoch": 0.15983751324620277,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999423083679498,
      "loss": 6.4624,
      "mean_token_accuracy": 0.13131050616502762,
      "num_tokens": 3369939.0,
      "step": 1810
    },
    {
      "entropy": 6.611289548873901,
      "epoch": 0.1602790533380431,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999415930724133,
      "loss": 6.604,
      "mean_token_accuracy": 0.12783848196268083,
      "num_tokens": 3381326.0,
      "step": 1815
    },
    {
      "entropy": 6.5344682216644285,
      "epoch": 0.16072059342988343,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999408733703988,
      "loss": 6.5567,
      "mean_token_accuracy": 0.12708184048533439,
      "num_tokens": 3391016.0,
      "step": 1820
    },
    {
      "entropy": 6.508758783340454,
      "epoch": 0.16116213352172376,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999401492619207,
      "loss": 6.3051,
      "mean_token_accuracy": 0.1379851721227169,
      "num_tokens": 3400294.0,
      "step": 1825
    },
    {
      "entropy": 6.467001914978027,
      "epoch": 0.16160367361356412,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999394207469928,
      "loss": 6.523,
      "mean_token_accuracy": 0.1270366355776787,
      "num_tokens": 3409685.0,
      "step": 1830
    },
    {
      "entropy": 6.615599775314331,
      "epoch": 0.16204521370540445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999386878256297,
      "loss": 6.4286,
      "mean_token_accuracy": 0.13395086154341698,
      "num_tokens": 3418946.0,
      "step": 1835
    },
    {
      "entropy": 6.307619047164917,
      "epoch": 0.16248675379724478,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999379504978457,
      "loss": 6.4423,
      "mean_token_accuracy": 0.1329544149339199,
      "num_tokens": 3428245.0,
      "step": 1840
    },
    {
      "entropy": 6.515369462966919,
      "epoch": 0.16292829388908514,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499937208763655,
      "loss": 6.4496,
      "mean_token_accuracy": 0.13294086754322051,
      "num_tokens": 3437580.0,
      "step": 1845
    },
    {
      "entropy": 6.529568195343018,
      "epoch": 0.16336983398092547,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999364626230724,
      "loss": 6.4279,
      "mean_token_accuracy": 0.13660631999373435,
      "num_tokens": 3446393.0,
      "step": 1850
    },
    {
      "entropy": 6.442787027359008,
      "epoch": 0.1638113740727658,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999357120761124,
      "loss": 6.5321,
      "mean_token_accuracy": 0.13097543716430665,
      "num_tokens": 3455956.0,
      "step": 1855
    },
    {
      "entropy": 6.608609199523926,
      "epoch": 0.16425291416460613,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999349571227898,
      "loss": 6.5277,
      "mean_token_accuracy": 0.12765606567263604,
      "num_tokens": 3465722.0,
      "step": 1860
    },
    {
      "entropy": 6.513270521163941,
      "epoch": 0.1646944542564465,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999341977631193,
      "loss": 6.4959,
      "mean_token_accuracy": 0.13204836025834082,
      "num_tokens": 3475467.0,
      "step": 1865
    },
    {
      "entropy": 6.463746118545532,
      "epoch": 0.16513599434828682,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999334339971157,
      "loss": 6.3874,
      "mean_token_accuracy": 0.13608791455626487,
      "num_tokens": 3484931.0,
      "step": 1870
    },
    {
      "entropy": 6.5012932300567625,
      "epoch": 0.16557753444012716,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999326658247942,
      "loss": 6.5008,
      "mean_token_accuracy": 0.1320129007101059,
      "num_tokens": 3494001.0,
      "step": 1875
    },
    {
      "entropy": 6.521371078491211,
      "epoch": 0.16601907453196751,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999318932461696,
      "loss": 6.424,
      "mean_token_accuracy": 0.1406968742609024,
      "num_tokens": 3503021.0,
      "step": 1880
    },
    {
      "entropy": 6.379944896697998,
      "epoch": 0.16646061462380785,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999311162612571,
      "loss": 6.4661,
      "mean_token_accuracy": 0.1362703375518322,
      "num_tokens": 3513005.0,
      "step": 1885
    },
    {
      "entropy": 6.564940690994263,
      "epoch": 0.16690215471564818,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499930334870072,
      "loss": 6.5404,
      "mean_token_accuracy": 0.13439082428812982,
      "num_tokens": 3523219.0,
      "step": 1890
    },
    {
      "entropy": 6.511162185668946,
      "epoch": 0.1673436948074885,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999295490726296,
      "loss": 6.4859,
      "mean_token_accuracy": 0.13242177516222,
      "num_tokens": 3532917.0,
      "step": 1895
    },
    {
      "entropy": 6.521901321411133,
      "epoch": 0.16778523489932887,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999287588689453,
      "loss": 6.4915,
      "mean_token_accuracy": 0.136155141890049,
      "num_tokens": 3542845.0,
      "step": 1900
    },
    {
      "entropy": 6.521121215820313,
      "epoch": 0.1682267749911692,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999279642590344,
      "loss": 6.5074,
      "mean_token_accuracy": 0.13129648566246033,
      "num_tokens": 3552126.0,
      "step": 1905
    },
    {
      "entropy": 6.5425803661346436,
      "epoch": 0.16866831508300953,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999271652429127,
      "loss": 6.5284,
      "mean_token_accuracy": 0.12366664186120033,
      "num_tokens": 3561254.0,
      "step": 1910
    },
    {
      "entropy": 6.458356142044067,
      "epoch": 0.1691098551748499,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999263618205958,
      "loss": 6.366,
      "mean_token_accuracy": 0.1399925693869591,
      "num_tokens": 3569781.0,
      "step": 1915
    },
    {
      "entropy": 6.48570237159729,
      "epoch": 0.16955139526669022,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999255539920993,
      "loss": 6.4831,
      "mean_token_accuracy": 0.1311805158853531,
      "num_tokens": 3579664.0,
      "step": 1920
    },
    {
      "entropy": 6.526191329956054,
      "epoch": 0.16999293535853055,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999247417574391,
      "loss": 6.5194,
      "mean_token_accuracy": 0.13419721275568008,
      "num_tokens": 3588671.0,
      "step": 1925
    },
    {
      "entropy": 6.495071268081665,
      "epoch": 0.17043447545037088,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999239251166312,
      "loss": 6.3976,
      "mean_token_accuracy": 0.13579057157039642,
      "num_tokens": 3597656.0,
      "step": 1930
    },
    {
      "entropy": 6.4537488460540775,
      "epoch": 0.17087601554221124,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999231040696914,
      "loss": 6.4361,
      "mean_token_accuracy": 0.13407086580991745,
      "num_tokens": 3608017.0,
      "step": 1935
    },
    {
      "entropy": 6.464614009857177,
      "epoch": 0.17131755563405157,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999222786166361,
      "loss": 6.5024,
      "mean_token_accuracy": 0.13418009057641028,
      "num_tokens": 3618189.0,
      "step": 1940
    },
    {
      "entropy": 6.500387525558471,
      "epoch": 0.1717590957258919,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999214487574812,
      "loss": 6.4369,
      "mean_token_accuracy": 0.13329068347811698,
      "num_tokens": 3627211.0,
      "step": 1945
    },
    {
      "entropy": 6.505857801437378,
      "epoch": 0.17220063581773226,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999206144922431,
      "loss": 6.3857,
      "mean_token_accuracy": 0.132901331782341,
      "num_tokens": 3636781.0,
      "step": 1950
    },
    {
      "entropy": 6.490985107421875,
      "epoch": 0.1726421759095726,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499919775820938,
      "loss": 6.4826,
      "mean_token_accuracy": 0.13853701204061508,
      "num_tokens": 3644891.0,
      "step": 1955
    },
    {
      "entropy": 6.4972686767578125,
      "epoch": 0.17308371600141292,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999189327435825,
      "loss": 6.505,
      "mean_token_accuracy": 0.1336333230137825,
      "num_tokens": 3655477.0,
      "step": 1960
    },
    {
      "entropy": 6.5157805442810055,
      "epoch": 0.17352525609325326,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999180852601929,
      "loss": 6.5269,
      "mean_token_accuracy": 0.13449802100658417,
      "num_tokens": 3664542.0,
      "step": 1965
    },
    {
      "entropy": 6.446292495727539,
      "epoch": 0.17396679618509361,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000499917233370786,
      "loss": 6.4281,
      "mean_token_accuracy": 0.1332317002117634,
      "num_tokens": 3673806.0,
      "step": 1970
    },
    {
      "entropy": 6.496902084350586,
      "epoch": 0.17440833627693395,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999163770753784,
      "loss": 6.4363,
      "mean_token_accuracy": 0.13492944985628127,
      "num_tokens": 3683238.0,
      "step": 1975
    },
    {
      "entropy": 6.456824207305909,
      "epoch": 0.17484987636877428,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004999155163739869,
      "loss": 6.4036,
      "mean_token_accuracy": 0.13336892798542976,
      "num_tokens": 3692161.0,
      "step": 1980
    },
    {
      "entropy": 6.478030490875244,
      "epoch": 0.17529141646061464,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999146512666284,
      "loss": 6.4329,
      "mean_token_accuracy": 0.13266484811902046,
      "num_tokens": 3701431.0,
      "step": 1985
    },
    {
      "entropy": 6.466426849365234,
      "epoch": 0.17573295655245497,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999137817533197,
      "loss": 6.4202,
      "mean_token_accuracy": 0.14008525460958482,
      "num_tokens": 3710963.0,
      "step": 1990
    },
    {
      "entropy": 6.430937433242798,
      "epoch": 0.1761744966442953,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999129078340779,
      "loss": 6.4103,
      "mean_token_accuracy": 0.1410228542983532,
      "num_tokens": 3720177.0,
      "step": 1995
    },
    {
      "entropy": 6.454744625091553,
      "epoch": 0.17661603673613563,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999120295089202,
      "loss": 6.3475,
      "mean_token_accuracy": 0.1424284115433693,
      "num_tokens": 3728123.0,
      "step": 2000
    },
    {
      "entropy": 6.492793893814087,
      "epoch": 0.177057576827976,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999111467778639,
      "loss": 6.5131,
      "mean_token_accuracy": 0.134304940700531,
      "num_tokens": 3736869.0,
      "step": 2005
    },
    {
      "entropy": 6.4386314868927,
      "epoch": 0.17749911691981632,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499910259640926,
      "loss": 6.3728,
      "mean_token_accuracy": 0.13866495788097383,
      "num_tokens": 3745830.0,
      "step": 2010
    },
    {
      "entropy": 6.426714324951172,
      "epoch": 0.17794065701165665,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499909368098124,
      "loss": 6.4177,
      "mean_token_accuracy": 0.13295648768544197,
      "num_tokens": 3755019.0,
      "step": 2015
    },
    {
      "entropy": 6.44696569442749,
      "epoch": 0.178382197103497,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999084721494754,
      "loss": 6.3988,
      "mean_token_accuracy": 0.13579348400235175,
      "num_tokens": 3764814.0,
      "step": 2020
    },
    {
      "entropy": 6.390467166900635,
      "epoch": 0.17882373719533734,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999075717949978,
      "loss": 6.3543,
      "mean_token_accuracy": 0.14030898809432985,
      "num_tokens": 3774258.0,
      "step": 2025
    },
    {
      "entropy": 6.3934080600738525,
      "epoch": 0.17926527728717767,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999066670347089,
      "loss": 6.406,
      "mean_token_accuracy": 0.13901298716664315,
      "num_tokens": 3783961.0,
      "step": 2030
    },
    {
      "entropy": 6.370501375198364,
      "epoch": 0.179706817379018,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999057578686261,
      "loss": 6.3714,
      "mean_token_accuracy": 0.13467539772391318,
      "num_tokens": 3792417.0,
      "step": 2035
    },
    {
      "entropy": 6.449629449844361,
      "epoch": 0.18014835747085836,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999048442967675,
      "loss": 6.3437,
      "mean_token_accuracy": 0.13966423645615578,
      "num_tokens": 3801479.0,
      "step": 2040
    },
    {
      "entropy": 6.386091041564941,
      "epoch": 0.1805898975626987,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999039263191508,
      "loss": 6.4155,
      "mean_token_accuracy": 0.1346621833741665,
      "num_tokens": 3810799.0,
      "step": 2045
    },
    {
      "entropy": 6.469657516479492,
      "epoch": 0.18103143765453902,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999030039357943,
      "loss": 6.4601,
      "mean_token_accuracy": 0.134067452698946,
      "num_tokens": 3820966.0,
      "step": 2050
    },
    {
      "entropy": 6.439054012298584,
      "epoch": 0.18147297774637938,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999020771467158,
      "loss": 6.4266,
      "mean_token_accuracy": 0.12985620722174646,
      "num_tokens": 3829247.0,
      "step": 2055
    },
    {
      "entropy": 6.5768702030181885,
      "epoch": 0.1819145178382197,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999011459519335,
      "loss": 6.3752,
      "mean_token_accuracy": 0.14039287865161895,
      "num_tokens": 3838114.0,
      "step": 2060
    },
    {
      "entropy": 6.355933761596679,
      "epoch": 0.18235605793006004,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999002103514655,
      "loss": 6.4479,
      "mean_token_accuracy": 0.13630082532763482,
      "num_tokens": 3848075.0,
      "step": 2065
    },
    {
      "entropy": 6.573451566696167,
      "epoch": 0.18279759802190038,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998992703453304,
      "loss": 6.4793,
      "mean_token_accuracy": 0.13254386261105539,
      "num_tokens": 3857934.0,
      "step": 2070
    },
    {
      "entropy": 6.425156974792481,
      "epoch": 0.18323913811374073,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998983259335466,
      "loss": 6.3334,
      "mean_token_accuracy": 0.14134026542305947,
      "num_tokens": 3866707.0,
      "step": 2075
    },
    {
      "entropy": 6.496882581710816,
      "epoch": 0.18368067820558107,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998973771161324,
      "loss": 6.4122,
      "mean_token_accuracy": 0.13621948882937432,
      "num_tokens": 3875233.0,
      "step": 2080
    },
    {
      "entropy": 6.395424509048462,
      "epoch": 0.1841222182974214,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998964238931065,
      "loss": 6.3969,
      "mean_token_accuracy": 0.14365216121077537,
      "num_tokens": 3885173.0,
      "step": 2085
    },
    {
      "entropy": 6.43842887878418,
      "epoch": 0.18456375838926176,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998954662644876,
      "loss": 6.3698,
      "mean_token_accuracy": 0.141506065428257,
      "num_tokens": 3894198.0,
      "step": 2090
    },
    {
      "entropy": 6.405653476715088,
      "epoch": 0.1850052984811021,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998945042302943,
      "loss": 6.362,
      "mean_token_accuracy": 0.13782091364264487,
      "num_tokens": 3904076.0,
      "step": 2095
    },
    {
      "entropy": 6.422827577590942,
      "epoch": 0.18544683857294242,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998935377905457,
      "loss": 6.4827,
      "mean_token_accuracy": 0.13269177302718163,
      "num_tokens": 3913204.0,
      "step": 2100
    },
    {
      "entropy": 6.533098840713501,
      "epoch": 0.18588837866478275,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998925669452605,
      "loss": 6.4243,
      "mean_token_accuracy": 0.13815055713057517,
      "num_tokens": 3922148.0,
      "step": 2105
    },
    {
      "entropy": 6.385688829421997,
      "epoch": 0.1863299187566231,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998915916944579,
      "loss": 6.402,
      "mean_token_accuracy": 0.13988569304347037,
      "num_tokens": 3931333.0,
      "step": 2110
    },
    {
      "entropy": 6.45911545753479,
      "epoch": 0.18677145884846344,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998906120381568,
      "loss": 6.3701,
      "mean_token_accuracy": 0.14163922518491745,
      "num_tokens": 3941061.0,
      "step": 2115
    },
    {
      "entropy": 6.4747344017028805,
      "epoch": 0.18721299894030377,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998896279763766,
      "loss": 6.4645,
      "mean_token_accuracy": 0.13232587426900863,
      "num_tokens": 3950075.0,
      "step": 2120
    },
    {
      "entropy": 6.432454586029053,
      "epoch": 0.18765453903214413,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998886395091365,
      "loss": 6.3139,
      "mean_token_accuracy": 0.1374908208847046,
      "num_tokens": 3958885.0,
      "step": 2125
    },
    {
      "entropy": 6.344161605834961,
      "epoch": 0.18809607912398446,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998876466364559,
      "loss": 6.4104,
      "mean_token_accuracy": 0.13860961571335792,
      "num_tokens": 3968218.0,
      "step": 2130
    },
    {
      "entropy": 6.387483358383179,
      "epoch": 0.1885376192158248,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998866493583541,
      "loss": 6.3306,
      "mean_token_accuracy": 0.13824384585022925,
      "num_tokens": 3977435.0,
      "step": 2135
    },
    {
      "entropy": 6.347888994216919,
      "epoch": 0.18897915930766512,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998856476748509,
      "loss": 6.3315,
      "mean_token_accuracy": 0.14486501812934877,
      "num_tokens": 3986608.0,
      "step": 2140
    },
    {
      "entropy": 6.364703226089477,
      "epoch": 0.18942069939950548,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998846415859656,
      "loss": 6.3417,
      "mean_token_accuracy": 0.14277436435222626,
      "num_tokens": 3996087.0,
      "step": 2145
    },
    {
      "entropy": 6.452169370651245,
      "epoch": 0.1898622394913458,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998836310917182,
      "loss": 6.3897,
      "mean_token_accuracy": 0.13653551563620567,
      "num_tokens": 4006257.0,
      "step": 2150
    },
    {
      "entropy": 6.4531981468200685,
      "epoch": 0.19030377958318614,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998826161921282,
      "loss": 6.4047,
      "mean_token_accuracy": 0.14247702434659004,
      "num_tokens": 4015904.0,
      "step": 2155
    },
    {
      "entropy": 6.370388221740723,
      "epoch": 0.1907453196750265,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998815968872157,
      "loss": 6.3914,
      "mean_token_accuracy": 0.13624223321676254,
      "num_tokens": 4025417.0,
      "step": 2160
    },
    {
      "entropy": 6.354130029678345,
      "epoch": 0.19118685976686683,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998805731770007,
      "loss": 6.2668,
      "mean_token_accuracy": 0.15688675343990327,
      "num_tokens": 4035181.0,
      "step": 2165
    },
    {
      "entropy": 6.40911750793457,
      "epoch": 0.19162839985870717,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499879545061503,
      "loss": 6.4414,
      "mean_token_accuracy": 0.13603056818246842,
      "num_tokens": 4045112.0,
      "step": 2170
    },
    {
      "entropy": 6.498990821838379,
      "epoch": 0.1920699399505475,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998785125407432,
      "loss": 6.5464,
      "mean_token_accuracy": 0.12719694674015045,
      "num_tokens": 4054566.0,
      "step": 2175
    },
    {
      "entropy": 6.520226097106933,
      "epoch": 0.19251148004238786,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499877475614741,
      "loss": 6.3688,
      "mean_token_accuracy": 0.13931149691343309,
      "num_tokens": 4063960.0,
      "step": 2180
    },
    {
      "entropy": 6.299141502380371,
      "epoch": 0.1929530201342282,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998764342835169,
      "loss": 6.3007,
      "mean_token_accuracy": 0.14462162256240846,
      "num_tokens": 4072620.0,
      "step": 2185
    },
    {
      "entropy": 6.3178962707519535,
      "epoch": 0.19339456022606852,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998753885470915,
      "loss": 6.2638,
      "mean_token_accuracy": 0.142801184207201,
      "num_tokens": 4081590.0,
      "step": 2190
    },
    {
      "entropy": 6.386137342453003,
      "epoch": 0.19383610031790888,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998743384054851,
      "loss": 6.4061,
      "mean_token_accuracy": 0.13850954100489615,
      "num_tokens": 4090758.0,
      "step": 2195
    },
    {
      "entropy": 6.348768806457519,
      "epoch": 0.1942776404097492,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998732838587183,
      "loss": 6.2249,
      "mean_token_accuracy": 0.14862391501665115,
      "num_tokens": 4099281.0,
      "step": 2200
    },
    {
      "entropy": 6.402739572525024,
      "epoch": 0.19471918050158954,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998722249068118,
      "loss": 6.3303,
      "mean_token_accuracy": 0.14005536511540412,
      "num_tokens": 4108953.0,
      "step": 2205
    },
    {
      "entropy": 6.359877395629883,
      "epoch": 0.19516072059342987,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998711615497863,
      "loss": 6.4579,
      "mean_token_accuracy": 0.13361181691288948,
      "num_tokens": 4118799.0,
      "step": 2210
    },
    {
      "entropy": 6.544803190231323,
      "epoch": 0.19560226068527023,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998700937876626,
      "loss": 6.4205,
      "mean_token_accuracy": 0.13423945084214212,
      "num_tokens": 4127862.0,
      "step": 2215
    },
    {
      "entropy": 6.3850291728973385,
      "epoch": 0.19604380077711056,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998690216204615,
      "loss": 6.495,
      "mean_token_accuracy": 0.1222709484398365,
      "num_tokens": 4139029.0,
      "step": 2220
    },
    {
      "entropy": 6.464286041259766,
      "epoch": 0.1964853408689509,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998679450482043,
      "loss": 6.2958,
      "mean_token_accuracy": 0.14723266661167145,
      "num_tokens": 4148257.0,
      "step": 2225
    },
    {
      "entropy": 6.36017689704895,
      "epoch": 0.19692688096079125,
      "grad_norm": 1.25,
      "learning_rate": 0.000499866864070912,
      "loss": 6.2594,
      "mean_token_accuracy": 0.13788010850548743,
      "num_tokens": 4157626.0,
      "step": 2230
    },
    {
      "entropy": 6.3908546447753904,
      "epoch": 0.19736842105263158,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998657786886056,
      "loss": 6.4709,
      "mean_token_accuracy": 0.13577495142817497,
      "num_tokens": 4166804.0,
      "step": 2235
    },
    {
      "entropy": 6.486610317230225,
      "epoch": 0.1978099611444719,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998646889013066,
      "loss": 6.3604,
      "mean_token_accuracy": 0.13759335353970528,
      "num_tokens": 4175701.0,
      "step": 2240
    },
    {
      "entropy": 6.4614592552185055,
      "epoch": 0.19825150123631224,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998635947090362,
      "loss": 6.4333,
      "mean_token_accuracy": 0.14182267859578132,
      "num_tokens": 4184711.0,
      "step": 2245
    },
    {
      "entropy": 6.411881828308106,
      "epoch": 0.1986930413281526,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004998624961118158,
      "loss": 6.4015,
      "mean_token_accuracy": 0.14113648906350135,
      "num_tokens": 4193931.0,
      "step": 2250
    },
    {
      "entropy": 6.377731609344482,
      "epoch": 0.19913458141999293,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499861393109667,
      "loss": 6.2849,
      "mean_token_accuracy": 0.14618429020047188,
      "num_tokens": 4203000.0,
      "step": 2255
    },
    {
      "entropy": 6.252184343338013,
      "epoch": 0.19957612151183327,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998602857026114,
      "loss": 6.2994,
      "mean_token_accuracy": 0.1432739280164242,
      "num_tokens": 4211977.0,
      "step": 2260
    },
    {
      "entropy": 6.46929440498352,
      "epoch": 0.20001766160367362,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998591738906708,
      "loss": 6.3915,
      "mean_token_accuracy": 0.14305510893464088,
      "num_tokens": 4220375.0,
      "step": 2265
    },
    {
      "entropy": 6.426153373718262,
      "epoch": 0.20045920169551396,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998580576738668,
      "loss": 6.3733,
      "mean_token_accuracy": 0.1404948763549328,
      "num_tokens": 4230506.0,
      "step": 2270
    },
    {
      "entropy": 6.3839240074157715,
      "epoch": 0.2009007417873543,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998569370522213,
      "loss": 6.3415,
      "mean_token_accuracy": 0.1391644887626171,
      "num_tokens": 4240235.0,
      "step": 2275
    },
    {
      "entropy": 6.3975495338439945,
      "epoch": 0.20134228187919462,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998558120257563,
      "loss": 6.3999,
      "mean_token_accuracy": 0.13654018118977546,
      "num_tokens": 4249775.0,
      "step": 2280
    },
    {
      "entropy": 6.365513944625855,
      "epoch": 0.20178382197103498,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004998546825944938,
      "loss": 6.3373,
      "mean_token_accuracy": 0.14069335907697678,
      "num_tokens": 4258523.0,
      "step": 2285
    },
    {
      "entropy": 6.384157276153564,
      "epoch": 0.2022253620628753,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000499853548758456,
      "loss": 6.3589,
      "mean_token_accuracy": 0.13782192021608353,
      "num_tokens": 4267683.0,
      "step": 2290
    },
    {
      "entropy": 6.489559173583984,
      "epoch": 0.20266690215471564,
      "grad_norm": 3.359375,
      "learning_rate": 0.000499852410517665,
      "loss": 6.4409,
      "mean_token_accuracy": 0.13151715248823165,
      "num_tokens": 4276903.0,
      "step": 2295
    },
    {
      "entropy": 6.439543437957764,
      "epoch": 0.203108442246556,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004998512678721431,
      "loss": 6.3807,
      "mean_token_accuracy": 0.13557614907622337,
      "num_tokens": 4287257.0,
      "step": 2300
    },
    {
      "entropy": 6.444064092636109,
      "epoch": 0.20354998233839633,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499850120821913,
      "loss": 6.4005,
      "mean_token_accuracy": 0.1363210029900074,
      "num_tokens": 4297345.0,
      "step": 2305
    },
    {
      "entropy": 6.469133949279785,
      "epoch": 0.20399152243023666,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004998489693669967,
      "loss": 6.3436,
      "mean_token_accuracy": 0.135118205845356,
      "num_tokens": 4306533.0,
      "step": 2310
    },
    {
      "entropy": 6.341489553451538,
      "epoch": 0.204433062522077,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499847813507417,
      "loss": 6.4077,
      "mean_token_accuracy": 0.13763028532266616,
      "num_tokens": 4316338.0,
      "step": 2315
    },
    {
      "entropy": 6.4129331588745115,
      "epoch": 0.20487460261391735,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998466532431966,
      "loss": 6.3616,
      "mean_token_accuracy": 0.13560457453131675,
      "num_tokens": 4326585.0,
      "step": 2320
    },
    {
      "entropy": 6.5451561450958256,
      "epoch": 0.20531614270575768,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998454885743581,
      "loss": 6.4974,
      "mean_token_accuracy": 0.131976468116045,
      "num_tokens": 4336490.0,
      "step": 2325
    },
    {
      "entropy": 6.331731176376342,
      "epoch": 0.205757682797598,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998443195009242,
      "loss": 6.3524,
      "mean_token_accuracy": 0.1353374347090721,
      "num_tokens": 4346264.0,
      "step": 2330
    },
    {
      "entropy": 6.46915283203125,
      "epoch": 0.20619922288943837,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998431460229182,
      "loss": 6.3484,
      "mean_token_accuracy": 0.14326995760202407,
      "num_tokens": 4355102.0,
      "step": 2335
    },
    {
      "entropy": 6.449292993545532,
      "epoch": 0.2066407629812787,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004998419681403627,
      "loss": 6.5183,
      "mean_token_accuracy": 0.13123422861099243,
      "num_tokens": 4365569.0,
      "step": 2340
    },
    {
      "entropy": 6.420946168899536,
      "epoch": 0.20708230307311903,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998407858532809,
      "loss": 6.3442,
      "mean_token_accuracy": 0.14129015430808067,
      "num_tokens": 4375437.0,
      "step": 2345
    },
    {
      "entropy": 6.4041321754455565,
      "epoch": 0.20752384316495937,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499839599161696,
      "loss": 6.3431,
      "mean_token_accuracy": 0.1439853347837925,
      "num_tokens": 4384420.0,
      "step": 2350
    },
    {
      "entropy": 6.3517293453216555,
      "epoch": 0.20796538325679972,
      "grad_norm": 1.25,
      "learning_rate": 0.0004998384080656314,
      "loss": 6.3093,
      "mean_token_accuracy": 0.13698634803295134,
      "num_tokens": 4393730.0,
      "step": 2355
    },
    {
      "entropy": 6.3013396739959715,
      "epoch": 0.20840692334864006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00049983721256511,
      "loss": 6.2912,
      "mean_token_accuracy": 0.14068662524223327,
      "num_tokens": 4402731.0,
      "step": 2360
    },
    {
      "entropy": 6.36918888092041,
      "epoch": 0.2088484634404804,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004998360126601556,
      "loss": 6.3489,
      "mean_token_accuracy": 0.14043711125850677,
      "num_tokens": 4411606.0,
      "step": 2365
    },
    {
      "entropy": 6.387992668151855,
      "epoch": 0.20929000353232075,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004998348083507916,
      "loss": 6.4201,
      "mean_token_accuracy": 0.13762734830379486,
      "num_tokens": 4421685.0,
      "step": 2370
    },
    {
      "entropy": 6.478447484970093,
      "epoch": 0.20973154362416108,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004998335996370416,
      "loss": 6.3708,
      "mean_token_accuracy": 0.14082587361335755,
      "num_tokens": 4431765.0,
      "step": 2375
    },
    {
      "entropy": 6.374440622329712,
      "epoch": 0.2101730837160014,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998323865189291,
      "loss": 6.343,
      "mean_token_accuracy": 0.13578818812966348,
      "num_tokens": 4441191.0,
      "step": 2380
    },
    {
      "entropy": 6.384360408782959,
      "epoch": 0.21061462380784174,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998311689964781,
      "loss": 6.4306,
      "mean_token_accuracy": 0.13119565919041634,
      "num_tokens": 4450156.0,
      "step": 2385
    },
    {
      "entropy": 6.4847077369689945,
      "epoch": 0.2110561638996821,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004998299470697125,
      "loss": 6.4317,
      "mean_token_accuracy": 0.1381568767130375,
      "num_tokens": 4459466.0,
      "step": 2390
    },
    {
      "entropy": 6.431920385360717,
      "epoch": 0.21149770399152243,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004998287207386559,
      "loss": 6.4087,
      "mean_token_accuracy": 0.13905522376298904,
      "num_tokens": 4468539.0,
      "step": 2395
    },
    {
      "entropy": 6.486487150192261,
      "epoch": 0.21193924408336276,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998274900033326,
      "loss": 6.3309,
      "mean_token_accuracy": 0.14359693378210067,
      "num_tokens": 4477579.0,
      "step": 2400
    },
    {
      "entropy": 6.27841157913208,
      "epoch": 0.21238078417520312,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998262548637667,
      "loss": 6.2944,
      "mean_token_accuracy": 0.14606404006481172,
      "num_tokens": 4486800.0,
      "step": 2405
    },
    {
      "entropy": 6.3170582294464115,
      "epoch": 0.21282232426704345,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004998250153199822,
      "loss": 6.2406,
      "mean_token_accuracy": 0.1487855888903141,
      "num_tokens": 4495985.0,
      "step": 2410
    },
    {
      "entropy": 6.4847211837768555,
      "epoch": 0.21326386435888378,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998237713720036,
      "loss": 6.4147,
      "mean_token_accuracy": 0.13529004976153375,
      "num_tokens": 4504944.0,
      "step": 2415
    },
    {
      "entropy": 6.306296968460083,
      "epoch": 0.2137054044507241,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998225230198552,
      "loss": 6.3031,
      "mean_token_accuracy": 0.14425254613161087,
      "num_tokens": 4515402.0,
      "step": 2420
    },
    {
      "entropy": 6.379396724700928,
      "epoch": 0.21414694454256447,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998212702635614,
      "loss": 6.374,
      "mean_token_accuracy": 0.13573372215032578,
      "num_tokens": 4525009.0,
      "step": 2425
    },
    {
      "entropy": 6.499070882797241,
      "epoch": 0.2145884846344048,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004998200131031469,
      "loss": 6.4094,
      "mean_token_accuracy": 0.13427471220493317,
      "num_tokens": 4534460.0,
      "step": 2430
    },
    {
      "entropy": 6.376511430740356,
      "epoch": 0.21503002472624513,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998187515386361,
      "loss": 6.2374,
      "mean_token_accuracy": 0.14696551114320755,
      "num_tokens": 4543748.0,
      "step": 2435
    },
    {
      "entropy": 6.321905422210693,
      "epoch": 0.2154715648180855,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998174855700538,
      "loss": 6.3617,
      "mean_token_accuracy": 0.14429828226566316,
      "num_tokens": 4552722.0,
      "step": 2440
    },
    {
      "entropy": 6.440483331680298,
      "epoch": 0.21591310490992582,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004998162151974248,
      "loss": 6.2207,
      "mean_token_accuracy": 0.14380936026573182,
      "num_tokens": 4561607.0,
      "step": 2445
    },
    {
      "entropy": 6.3644684791564945,
      "epoch": 0.21635464500176615,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499814940420774,
      "loss": 6.5024,
      "mean_token_accuracy": 0.1272777035832405,
      "num_tokens": 4572524.0,
      "step": 2450
    },
    {
      "entropy": 6.4338010311126705,
      "epoch": 0.21679618509360649,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998136612401266,
      "loss": 6.3239,
      "mean_token_accuracy": 0.14616527855396272,
      "num_tokens": 4581601.0,
      "step": 2455
    },
    {
      "entropy": 6.408474111557007,
      "epoch": 0.21723772518544684,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998123776555071,
      "loss": 6.3758,
      "mean_token_accuracy": 0.1353377342224121,
      "num_tokens": 4591795.0,
      "step": 2460
    },
    {
      "entropy": 6.354497385025025,
      "epoch": 0.21767926527728718,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998110896669412,
      "loss": 6.3251,
      "mean_token_accuracy": 0.13686212673783302,
      "num_tokens": 4600745.0,
      "step": 2465
    },
    {
      "entropy": 6.451937007904053,
      "epoch": 0.2181208053691275,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998097972744539,
      "loss": 6.3817,
      "mean_token_accuracy": 0.13510738238692283,
      "num_tokens": 4610490.0,
      "step": 2470
    },
    {
      "entropy": 6.394930648803711,
      "epoch": 0.21856234546096787,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998085004780705,
      "loss": 6.3089,
      "mean_token_accuracy": 0.14037252739071845,
      "num_tokens": 4619511.0,
      "step": 2475
    },
    {
      "entropy": 6.379614734649659,
      "epoch": 0.2190038855528082,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998071992778164,
      "loss": 6.3428,
      "mean_token_accuracy": 0.1402582421898842,
      "num_tokens": 4628186.0,
      "step": 2480
    },
    {
      "entropy": 6.451162719726563,
      "epoch": 0.21944542564464853,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499805893673717,
      "loss": 6.378,
      "mean_token_accuracy": 0.14208944439888,
      "num_tokens": 4637431.0,
      "step": 2485
    },
    {
      "entropy": 6.312719869613647,
      "epoch": 0.21988696573648886,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004998045836657982,
      "loss": 6.2251,
      "mean_token_accuracy": 0.14331654608249664,
      "num_tokens": 4646627.0,
      "step": 2490
    },
    {
      "entropy": 6.346300935745239,
      "epoch": 0.22032850582832922,
      "grad_norm": 1.25,
      "learning_rate": 0.0004998032692540853,
      "loss": 6.324,
      "mean_token_accuracy": 0.13909852951765062,
      "num_tokens": 4656095.0,
      "step": 2495
    },
    {
      "entropy": 6.4727942943573,
      "epoch": 0.22077004592016955,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998019504386044,
      "loss": 6.4001,
      "mean_token_accuracy": 0.13648373931646346,
      "num_tokens": 4665807.0,
      "step": 2500
    },
    {
      "entropy": 6.379480361938477,
      "epoch": 0.22121158601200988,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004998006272193809,
      "loss": 6.3575,
      "mean_token_accuracy": 0.13852972015738488,
      "num_tokens": 4674459.0,
      "step": 2505
    },
    {
      "entropy": 6.410311126708985,
      "epoch": 0.22165312610385024,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997992995964412,
      "loss": 6.4505,
      "mean_token_accuracy": 0.13554090186953544,
      "num_tokens": 4684063.0,
      "step": 2510
    },
    {
      "entropy": 6.3212048530578615,
      "epoch": 0.22209466619569057,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004997979675698109,
      "loss": 6.3377,
      "mean_token_accuracy": 0.14127737581729888,
      "num_tokens": 4692807.0,
      "step": 2515
    },
    {
      "entropy": 6.487754249572754,
      "epoch": 0.2225362062875309,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997966311395164,
      "loss": 6.2628,
      "mean_token_accuracy": 0.14333414733409883,
      "num_tokens": 4701100.0,
      "step": 2520
    },
    {
      "entropy": 6.231193733215332,
      "epoch": 0.22297774637937123,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004997952903055836,
      "loss": 6.3098,
      "mean_token_accuracy": 0.1349740982055664,
      "num_tokens": 4710697.0,
      "step": 2525
    },
    {
      "entropy": 6.37754340171814,
      "epoch": 0.2234192864712116,
      "grad_norm": 1.4375,
      "learning_rate": 0.000499793945068039,
      "loss": 6.2937,
      "mean_token_accuracy": 0.13749002516269684,
      "num_tokens": 4718745.0,
      "step": 2530
    },
    {
      "entropy": 6.354239130020142,
      "epoch": 0.22386082656305192,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997925954269088,
      "loss": 6.2648,
      "mean_token_accuracy": 0.14231488481163979,
      "num_tokens": 4728056.0,
      "step": 2535
    },
    {
      "entropy": 6.382005739212036,
      "epoch": 0.22430236665489225,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997912413822196,
      "loss": 6.391,
      "mean_token_accuracy": 0.1377481795847416,
      "num_tokens": 4737605.0,
      "step": 2540
    },
    {
      "entropy": 6.34613995552063,
      "epoch": 0.2247439067467326,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997898829339979,
      "loss": 6.2206,
      "mean_token_accuracy": 0.14988200217485428,
      "num_tokens": 4746168.0,
      "step": 2545
    },
    {
      "entropy": 6.318027210235596,
      "epoch": 0.22518544683857294,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00049978852008227,
      "loss": 6.29,
      "mean_token_accuracy": 0.142624107003212,
      "num_tokens": 4755072.0,
      "step": 2550
    },
    {
      "entropy": 6.386806201934815,
      "epoch": 0.22562698693041328,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499787152827063,
      "loss": 6.3083,
      "mean_token_accuracy": 0.1480213187634945,
      "num_tokens": 4764580.0,
      "step": 2555
    },
    {
      "entropy": 6.287759494781494,
      "epoch": 0.2260685270222536,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997857811684035,
      "loss": 6.3079,
      "mean_token_accuracy": 0.1506627842783928,
      "num_tokens": 4774135.0,
      "step": 2560
    },
    {
      "entropy": 6.41878023147583,
      "epoch": 0.22651006711409397,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004997844051063183,
      "loss": 6.4043,
      "mean_token_accuracy": 0.13596237674355507,
      "num_tokens": 4784733.0,
      "step": 2565
    },
    {
      "entropy": 6.420801305770874,
      "epoch": 0.2269516072059343,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997830246408346,
      "loss": 6.3239,
      "mean_token_accuracy": 0.14653001427650453,
      "num_tokens": 4795327.0,
      "step": 2570
    },
    {
      "entropy": 6.26131682395935,
      "epoch": 0.22739314729777463,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004997816397719791,
      "loss": 6.3033,
      "mean_token_accuracy": 0.1419690243899822,
      "num_tokens": 4804314.0,
      "step": 2575
    },
    {
      "entropy": 6.447992181777954,
      "epoch": 0.227834687389615,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997802504997792,
      "loss": 6.3864,
      "mean_token_accuracy": 0.1353736087679863,
      "num_tokens": 4813637.0,
      "step": 2580
    },
    {
      "entropy": 6.362198352813721,
      "epoch": 0.22827622748145532,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997788568242621,
      "loss": 6.2733,
      "mean_token_accuracy": 0.14258976951241492,
      "num_tokens": 4823094.0,
      "step": 2585
    },
    {
      "entropy": 6.313487911224366,
      "epoch": 0.22871776757329565,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499777458745455,
      "loss": 6.2101,
      "mean_token_accuracy": 0.14641790390014647,
      "num_tokens": 4833199.0,
      "step": 2590
    },
    {
      "entropy": 6.383407735824585,
      "epoch": 0.22915930766513598,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004997760562633853,
      "loss": 6.3168,
      "mean_token_accuracy": 0.1408907800912857,
      "num_tokens": 4842970.0,
      "step": 2595
    },
    {
      "entropy": 6.429384851455689,
      "epoch": 0.22960084775697634,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997746493780804,
      "loss": 6.4179,
      "mean_token_accuracy": 0.1343531347811222,
      "num_tokens": 4852043.0,
      "step": 2600
    },
    {
      "entropy": 6.356952238082886,
      "epoch": 0.23004238784881667,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499773238089568,
      "loss": 6.2974,
      "mean_token_accuracy": 0.13784752711653708,
      "num_tokens": 4862232.0,
      "step": 2605
    },
    {
      "entropy": 6.3309001445770265,
      "epoch": 0.230483927940657,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004997718223978758,
      "loss": 6.2265,
      "mean_token_accuracy": 0.14307404607534407,
      "num_tokens": 4871186.0,
      "step": 2610
    },
    {
      "entropy": 6.393591213226318,
      "epoch": 0.23092546803249736,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997704023030315,
      "loss": 6.3176,
      "mean_token_accuracy": 0.14954473525285722,
      "num_tokens": 4879974.0,
      "step": 2615
    },
    {
      "entropy": 6.334340620040893,
      "epoch": 0.2313670081243377,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997689778050627,
      "loss": 6.3876,
      "mean_token_accuracy": 0.138017987459898,
      "num_tokens": 4890300.0,
      "step": 2620
    },
    {
      "entropy": 6.35406002998352,
      "epoch": 0.23180854821617802,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997675489039975,
      "loss": 6.3527,
      "mean_token_accuracy": 0.13665685877203942,
      "num_tokens": 4900428.0,
      "step": 2625
    },
    {
      "entropy": 6.393978548049927,
      "epoch": 0.23225008830801835,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004997661155998638,
      "loss": 6.3461,
      "mean_token_accuracy": 0.14447346180677414,
      "num_tokens": 4910092.0,
      "step": 2630
    },
    {
      "entropy": 6.395442962646484,
      "epoch": 0.2326916283998587,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004997646778926898,
      "loss": 6.3456,
      "mean_token_accuracy": 0.14186301603913307,
      "num_tokens": 4919593.0,
      "step": 2635
    },
    {
      "entropy": 6.36322603225708,
      "epoch": 0.23313316849169904,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004997632357825035,
      "loss": 6.3172,
      "mean_token_accuracy": 0.13682103008031846,
      "num_tokens": 4929098.0,
      "step": 2640
    },
    {
      "entropy": 6.4403660774230955,
      "epoch": 0.23357470858353938,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004997617892693333,
      "loss": 6.3653,
      "mean_token_accuracy": 0.13785530328750611,
      "num_tokens": 4938265.0,
      "step": 2645
    },
    {
      "entropy": 6.4328492164611815,
      "epoch": 0.23401624867537973,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004997603383532075,
      "loss": 6.3177,
      "mean_token_accuracy": 0.14054129794239997,
      "num_tokens": 4946694.0,
      "step": 2650
    },
    {
      "entropy": 6.297847509384155,
      "epoch": 0.23445778876722007,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004997588830341545,
      "loss": 6.3275,
      "mean_token_accuracy": 0.14105295538902282,
      "num_tokens": 4955296.0,
      "step": 2655
    },
    {
      "entropy": 6.317044591903686,
      "epoch": 0.2348993288590604,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004997574233122028,
      "loss": 6.3017,
      "mean_token_accuracy": 0.14537747204303741,
      "num_tokens": 4964409.0,
      "step": 2660
    },
    {
      "entropy": 6.3555162906646725,
      "epoch": 0.23534086895090076,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004997559591873809,
      "loss": 6.3035,
      "mean_token_accuracy": 0.14873416274785994,
      "num_tokens": 4973449.0,
      "step": 2665
    },
    {
      "entropy": 6.351046800613403,
      "epoch": 0.2357824090427411,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004997544906597178,
      "loss": 6.3045,
      "mean_token_accuracy": 0.14574557095766066,
      "num_tokens": 4983057.0,
      "step": 2670
    },
    {
      "entropy": 6.3079445362091064,
      "epoch": 0.23622394913458142,
      "grad_norm": 2.0,
      "learning_rate": 0.0004997530177292418,
      "loss": 6.3526,
      "mean_token_accuracy": 0.1371928870677948,
      "num_tokens": 4991950.0,
      "step": 2675
    },
    {
      "entropy": 6.4892011165618895,
      "epoch": 0.23666548922642175,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004997515403959823,
      "loss": 6.3031,
      "mean_token_accuracy": 0.14117597341537474,
      "num_tokens": 5001042.0,
      "step": 2680
    },
    {
      "entropy": 6.349958276748657,
      "epoch": 0.2371070293182621,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004997500586599677,
      "loss": 6.2512,
      "mean_token_accuracy": 0.14559754729270935,
      "num_tokens": 5009827.0,
      "step": 2685
    },
    {
      "entropy": 6.331127786636353,
      "epoch": 0.23754856941010244,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997485725212274,
      "loss": 6.2914,
      "mean_token_accuracy": 0.13917716667056085,
      "num_tokens": 5018708.0,
      "step": 2690
    },
    {
      "entropy": 6.290257930755615,
      "epoch": 0.23799010950194277,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004997470819797903,
      "loss": 6.1657,
      "mean_token_accuracy": 0.15060664638876914,
      "num_tokens": 5027522.0,
      "step": 2695
    },
    {
      "entropy": 6.301162242889404,
      "epoch": 0.23843164959378313,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997455870356857,
      "loss": 6.3071,
      "mean_token_accuracy": 0.14116470664739608,
      "num_tokens": 5035755.0,
      "step": 2700
    },
    {
      "entropy": 6.376118612289429,
      "epoch": 0.23887318968562346,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004997440876889429,
      "loss": 6.2543,
      "mean_token_accuracy": 0.14213244915008544,
      "num_tokens": 5045289.0,
      "step": 2705
    },
    {
      "entropy": 6.249507427215576,
      "epoch": 0.2393147297774638,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004997425839395913,
      "loss": 6.2774,
      "mean_token_accuracy": 0.14211364090442657,
      "num_tokens": 5053774.0,
      "step": 2710
    },
    {
      "entropy": 6.390804290771484,
      "epoch": 0.23975626986930412,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997410757876602,
      "loss": 6.3637,
      "mean_token_accuracy": 0.13094362914562224,
      "num_tokens": 5062911.0,
      "step": 2715
    },
    {
      "entropy": 6.374802780151367,
      "epoch": 0.24019780996114448,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997395632331793,
      "loss": 6.231,
      "mean_token_accuracy": 0.14750899523496627,
      "num_tokens": 5072107.0,
      "step": 2720
    },
    {
      "entropy": 6.239001655578614,
      "epoch": 0.2406393500529848,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004997380462761781,
      "loss": 6.189,
      "mean_token_accuracy": 0.14651150107383729,
      "num_tokens": 5080588.0,
      "step": 2725
    },
    {
      "entropy": 6.313574028015137,
      "epoch": 0.24108089014482514,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004997365249166864,
      "loss": 6.3843,
      "mean_token_accuracy": 0.14336365312337876,
      "num_tokens": 5090262.0,
      "step": 2730
    },
    {
      "entropy": 6.42013373374939,
      "epoch": 0.2415224302366655,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997349991547342,
      "loss": 6.2793,
      "mean_token_accuracy": 0.15069962069392204,
      "num_tokens": 5099285.0,
      "step": 2735
    },
    {
      "entropy": 6.363212537765503,
      "epoch": 0.24196397032850583,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997334689903509,
      "loss": 6.337,
      "mean_token_accuracy": 0.146900225430727,
      "num_tokens": 5109115.0,
      "step": 2740
    },
    {
      "entropy": 6.39803113937378,
      "epoch": 0.24240551042034617,
      "grad_norm": 1.375,
      "learning_rate": 0.0004997319344235668,
      "loss": 6.3625,
      "mean_token_accuracy": 0.14113707169890405,
      "num_tokens": 5117977.0,
      "step": 2745
    },
    {
      "entropy": 6.412462043762207,
      "epoch": 0.2428470505121865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000499730395454412,
      "loss": 6.3246,
      "mean_token_accuracy": 0.14121332690119742,
      "num_tokens": 5127457.0,
      "step": 2750
    },
    {
      "entropy": 6.294469976425171,
      "epoch": 0.24328859060402686,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997288520829166,
      "loss": 6.3746,
      "mean_token_accuracy": 0.13444311991333963,
      "num_tokens": 5137310.0,
      "step": 2755
    },
    {
      "entropy": 6.404769420623779,
      "epoch": 0.2437301306958672,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004997273043091107,
      "loss": 6.3024,
      "mean_token_accuracy": 0.13945599272847176,
      "num_tokens": 5146963.0,
      "step": 2760
    },
    {
      "entropy": 6.3221978664398195,
      "epoch": 0.24417167078770752,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004997257521330248,
      "loss": 6.285,
      "mean_token_accuracy": 0.13930008858442305,
      "num_tokens": 5155521.0,
      "step": 2765
    },
    {
      "entropy": 6.304354858398438,
      "epoch": 0.24461321087954788,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004997241955546892,
      "loss": 6.2707,
      "mean_token_accuracy": 0.14256517514586448,
      "num_tokens": 5165182.0,
      "step": 2770
    },
    {
      "entropy": 6.362063407897949,
      "epoch": 0.2450547509713882,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997226345741343,
      "loss": 6.2908,
      "mean_token_accuracy": 0.13857515007257462,
      "num_tokens": 5175511.0,
      "step": 2775
    },
    {
      "entropy": 6.305971288681031,
      "epoch": 0.24549629106322854,
      "grad_norm": 1.3125,
      "learning_rate": 0.000499721069191391,
      "loss": 6.2295,
      "mean_token_accuracy": 0.14581285938620567,
      "num_tokens": 5184772.0,
      "step": 2780
    },
    {
      "entropy": 6.251179599761963,
      "epoch": 0.24593783115506887,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997194994064896,
      "loss": 6.3205,
      "mean_token_accuracy": 0.14159369021654128,
      "num_tokens": 5195136.0,
      "step": 2785
    },
    {
      "entropy": 6.401678276062012,
      "epoch": 0.24637937124690923,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000499717925219461,
      "loss": 6.2658,
      "mean_token_accuracy": 0.15039479061961175,
      "num_tokens": 5203163.0,
      "step": 2790
    },
    {
      "entropy": 6.363914871215821,
      "epoch": 0.24682091133874956,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004997163466303362,
      "loss": 6.342,
      "mean_token_accuracy": 0.1374757319688797,
      "num_tokens": 5213233.0,
      "step": 2795
    },
    {
      "entropy": 6.326372671127319,
      "epoch": 0.2472624514305899,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499714763639146,
      "loss": 6.2748,
      "mean_token_accuracy": 0.13834284916520118,
      "num_tokens": 5222940.0,
      "step": 2800
    },
    {
      "entropy": 6.359001636505127,
      "epoch": 0.24770399152243025,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004997131762459211,
      "loss": 6.2926,
      "mean_token_accuracy": 0.1355063237249851,
      "num_tokens": 5232263.0,
      "step": 2805
    },
    {
      "entropy": 6.353963899612427,
      "epoch": 0.24814553161427058,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004997115844506932,
      "loss": 6.2777,
      "mean_token_accuracy": 0.13894380629062653,
      "num_tokens": 5241536.0,
      "step": 2810
    },
    {
      "entropy": 6.347019195556641,
      "epoch": 0.2485870717061109,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004997099882534929,
      "loss": 6.3014,
      "mean_token_accuracy": 0.13601062670350075,
      "num_tokens": 5250702.0,
      "step": 2815
    },
    {
      "entropy": 6.308573961257935,
      "epoch": 0.24902861179795124,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997083876543519,
      "loss": 6.3152,
      "mean_token_accuracy": 0.13795805871486663,
      "num_tokens": 5259811.0,
      "step": 2820
    },
    {
      "entropy": 6.473266696929931,
      "epoch": 0.2494701518897916,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997067826533014,
      "loss": 6.3819,
      "mean_token_accuracy": 0.1361972101032734,
      "num_tokens": 5270518.0,
      "step": 2825
    },
    {
      "entropy": 6.389288711547851,
      "epoch": 0.24991169198163193,
      "grad_norm": 1.625,
      "learning_rate": 0.0004997051732503726,
      "loss": 6.2551,
      "mean_token_accuracy": 0.13983014822006226,
      "num_tokens": 5279538.0,
      "step": 2830
    },
    {
      "entropy": 6.26657543182373,
      "epoch": 0.25035323207347226,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004997035594455975,
      "loss": 6.3026,
      "mean_token_accuracy": 0.13869107514619827,
      "num_tokens": 5289633.0,
      "step": 2835
    },
    {
      "entropy": 6.390535449981689,
      "epoch": 0.2507947721653126,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997019412390074,
      "loss": 6.42,
      "mean_token_accuracy": 0.13744667321443557,
      "num_tokens": 5299148.0,
      "step": 2840
    },
    {
      "entropy": 6.3600469589233395,
      "epoch": 0.2512363122571529,
      "grad_norm": 1.359375,
      "learning_rate": 0.000499700318630634,
      "loss": 6.2602,
      "mean_token_accuracy": 0.14515333026647567,
      "num_tokens": 5309090.0,
      "step": 2845
    },
    {
      "entropy": 6.343839168548584,
      "epoch": 0.2516778523489933,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004996986916205092,
      "loss": 6.3523,
      "mean_token_accuracy": 0.13558987081050872,
      "num_tokens": 5318798.0,
      "step": 2850
    },
    {
      "entropy": 6.27102484703064,
      "epoch": 0.25211939244083365,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004996970602086648,
      "loss": 6.2258,
      "mean_token_accuracy": 0.1445697531104088,
      "num_tokens": 5327915.0,
      "step": 2855
    },
    {
      "entropy": 6.295015048980713,
      "epoch": 0.252560932532674,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004996954243951327,
      "loss": 6.2654,
      "mean_token_accuracy": 0.14920969903469086,
      "num_tokens": 5336970.0,
      "step": 2860
    },
    {
      "entropy": 6.284700727462768,
      "epoch": 0.2530024726245143,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004996937841799451,
      "loss": 6.2196,
      "mean_token_accuracy": 0.1482589215040207,
      "num_tokens": 5345167.0,
      "step": 2865
    },
    {
      "entropy": 6.241301822662353,
      "epoch": 0.25344401271635464,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004996921395631342,
      "loss": 6.2041,
      "mean_token_accuracy": 0.14476508051156997,
      "num_tokens": 5353399.0,
      "step": 2870
    },
    {
      "entropy": 6.302233171463013,
      "epoch": 0.25388555280819497,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000499690490544732,
      "loss": 6.3187,
      "mean_token_accuracy": 0.1422203630208969,
      "num_tokens": 5363224.0,
      "step": 2875
    },
    {
      "entropy": 6.388623285293579,
      "epoch": 0.2543270929000353,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004996888371247707,
      "loss": 6.3351,
      "mean_token_accuracy": 0.1384458288550377,
      "num_tokens": 5372274.0,
      "step": 2880
    },
    {
      "entropy": 6.4166820526123045,
      "epoch": 0.2547686329918757,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000499687179303283,
      "loss": 6.308,
      "mean_token_accuracy": 0.14665373042225838,
      "num_tokens": 5380240.0,
      "step": 2885
    },
    {
      "entropy": 6.302547883987427,
      "epoch": 0.255210173083716,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004996855170803012,
      "loss": 6.1651,
      "mean_token_accuracy": 0.14814262390136718,
      "num_tokens": 5389390.0,
      "step": 2890
    },
    {
      "entropy": 6.2304370403289795,
      "epoch": 0.25565171317555635,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004996838504558581,
      "loss": 6.332,
      "mean_token_accuracy": 0.1434281162917614,
      "num_tokens": 5399425.0,
      "step": 2895
    },
    {
      "entropy": 6.390418291091919,
      "epoch": 0.2560932532673967,
      "grad_norm": 1.40625,
      "learning_rate": 0.000499682179429986,
      "loss": 6.3432,
      "mean_token_accuracy": 0.14031529799103737,
      "num_tokens": 5408717.0,
      "step": 2900
    },
    {
      "entropy": 6.369663047790527,
      "epoch": 0.256534793359237,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004996805040027178,
      "loss": 6.2789,
      "mean_token_accuracy": 0.13554130122065544,
      "num_tokens": 5418475.0,
      "step": 2905
    },
    {
      "entropy": 6.373133182525635,
      "epoch": 0.25697633345107734,
      "grad_norm": 1.75,
      "learning_rate": 0.0004996788241740863,
      "loss": 6.3139,
      "mean_token_accuracy": 0.14075277075171472,
      "num_tokens": 5428403.0,
      "step": 2910
    },
    {
      "entropy": 6.349458980560303,
      "epoch": 0.2574178735429177,
      "grad_norm": 1.375,
      "learning_rate": 0.0004996771399441243,
      "loss": 6.372,
      "mean_token_accuracy": 0.13378326818346978,
      "num_tokens": 5437347.0,
      "step": 2915
    },
    {
      "entropy": 6.4478528022766115,
      "epoch": 0.25785941363475806,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004996754513128652,
      "loss": 6.2655,
      "mean_token_accuracy": 0.14531978368759155,
      "num_tokens": 5446804.0,
      "step": 2920
    },
    {
      "entropy": 6.286684846878051,
      "epoch": 0.2583009537265984,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004996737582803416,
      "loss": 6.2076,
      "mean_token_accuracy": 0.14176183342933654,
      "num_tokens": 5455888.0,
      "step": 2925
    },
    {
      "entropy": 6.301067447662353,
      "epoch": 0.2587424938184387,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004996720608465868,
      "loss": 6.2077,
      "mean_token_accuracy": 0.14722970575094224,
      "num_tokens": 5463977.0,
      "step": 2930
    },
    {
      "entropy": 6.3211047649383545,
      "epoch": 0.25918403391027905,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004996703590116342,
      "loss": 6.327,
      "mean_token_accuracy": 0.13881488144397736,
      "num_tokens": 5473780.0,
      "step": 2935
    },
    {
      "entropy": 6.309579610824585,
      "epoch": 0.2596255740021194,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004996686527755171,
      "loss": 6.1882,
      "mean_token_accuracy": 0.14595898613333702,
      "num_tokens": 5482151.0,
      "step": 2940
    },
    {
      "entropy": 6.318369293212891,
      "epoch": 0.2600671140939597,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004996669421382687,
      "loss": 6.2299,
      "mean_token_accuracy": 0.1491067260503769,
      "num_tokens": 5491103.0,
      "step": 2945
    },
    {
      "entropy": 6.297426509857178,
      "epoch": 0.26050865418580005,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004996652270999228,
      "loss": 6.2482,
      "mean_token_accuracy": 0.14331730753183364,
      "num_tokens": 5500367.0,
      "step": 2950
    },
    {
      "entropy": 6.3517374992370605,
      "epoch": 0.26095019427764043,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996635076605128,
      "loss": 6.2829,
      "mean_token_accuracy": 0.14498205482959747,
      "num_tokens": 5509631.0,
      "step": 2955
    },
    {
      "entropy": 6.456418180465699,
      "epoch": 0.26139173436948077,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004996617838200725,
      "loss": 6.2799,
      "mean_token_accuracy": 0.14098481461405754,
      "num_tokens": 5518635.0,
      "step": 2960
    },
    {
      "entropy": 6.273197174072266,
      "epoch": 0.2618332744613211,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004996600555786357,
      "loss": 6.2366,
      "mean_token_accuracy": 0.14240477085113526,
      "num_tokens": 5527696.0,
      "step": 2965
    },
    {
      "entropy": 6.333998394012451,
      "epoch": 0.26227481455316143,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004996583229362362,
      "loss": 6.216,
      "mean_token_accuracy": 0.14474834203720094,
      "num_tokens": 5536632.0,
      "step": 2970
    },
    {
      "entropy": 6.4111855030059814,
      "epoch": 0.26271635464500176,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004996565858929078,
      "loss": 6.2995,
      "mean_token_accuracy": 0.14190069884061812,
      "num_tokens": 5545825.0,
      "step": 2975
    },
    {
      "entropy": 6.240679502487183,
      "epoch": 0.2631578947368421,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004996548444486847,
      "loss": 6.1819,
      "mean_token_accuracy": 0.14833824932575226,
      "num_tokens": 5555158.0,
      "step": 2980
    },
    {
      "entropy": 6.252728462219238,
      "epoch": 0.2635994348286824,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004996530986036008,
      "loss": 6.126,
      "mean_token_accuracy": 0.14824561923742294,
      "num_tokens": 5564218.0,
      "step": 2985
    },
    {
      "entropy": 6.236001491546631,
      "epoch": 0.2640409749205228,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004996513483576907,
      "loss": 6.2712,
      "mean_token_accuracy": 0.14679816737771034,
      "num_tokens": 5572760.0,
      "step": 2990
    },
    {
      "entropy": 6.370929002761841,
      "epoch": 0.26448251501236314,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004996495937109884,
      "loss": 6.3147,
      "mean_token_accuracy": 0.14080894514918327,
      "num_tokens": 5581660.0,
      "step": 2995
    },
    {
      "entropy": 6.323193359375,
      "epoch": 0.26492405510420347,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004996478346635283,
      "loss": 6.229,
      "mean_token_accuracy": 0.14318683445453645,
      "num_tokens": 5590664.0,
      "step": 3000
    },
    {
      "epoch": 0.26492405510420347,
      "eval_entropy": 6.108612703278003,
      "eval_loss": 6.312756538391113,
      "eval_mean_token_accuracy": 0.14662449901759,
      "eval_num_tokens": 5590664.0,
      "eval_runtime": 26.935,
      "eval_samples_per_second": 1311.12,
      "eval_steps_per_second": 163.913,
      "step": 3000
    },
    {
      "entropy": 6.254260587692261,
      "epoch": 0.2653655951960438,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004996460712153448,
      "loss": 6.1098,
      "mean_token_accuracy": 0.16003013849258424,
      "num_tokens": 5598727.0,
      "step": 3005
    },
    {
      "entropy": 6.19307279586792,
      "epoch": 0.26580713528788413,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004996443033664726,
      "loss": 6.3173,
      "mean_token_accuracy": 0.1360806442797184,
      "num_tokens": 5608549.0,
      "step": 3010
    },
    {
      "entropy": 6.317127704620361,
      "epoch": 0.26624867537972446,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004996425311169463,
      "loss": 6.2454,
      "mean_token_accuracy": 0.14779156744480132,
      "num_tokens": 5617628.0,
      "step": 3015
    },
    {
      "entropy": 6.371944999694824,
      "epoch": 0.2666902154715648,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004996407544668005,
      "loss": 6.303,
      "mean_token_accuracy": 0.1368146575987339,
      "num_tokens": 5627182.0,
      "step": 3020
    },
    {
      "entropy": 6.376071310043335,
      "epoch": 0.2671317555634052,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004996389734160701,
      "loss": 6.323,
      "mean_token_accuracy": 0.13930497989058493,
      "num_tokens": 5636413.0,
      "step": 3025
    },
    {
      "entropy": 6.400035238265991,
      "epoch": 0.2675732956552455,
      "grad_norm": 1.328125,
      "learning_rate": 0.00049963718796479,
      "loss": 6.2654,
      "mean_token_accuracy": 0.14247873574495315,
      "num_tokens": 5646025.0,
      "step": 3030
    },
    {
      "entropy": 6.326315689086914,
      "epoch": 0.26801483574708584,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996353981129952,
      "loss": 6.2818,
      "mean_token_accuracy": 0.1449798859655857,
      "num_tokens": 5656577.0,
      "step": 3035
    },
    {
      "entropy": 6.399411296844482,
      "epoch": 0.2684563758389262,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004996336038607206,
      "loss": 6.2581,
      "mean_token_accuracy": 0.14687425941228865,
      "num_tokens": 5665163.0,
      "step": 3040
    },
    {
      "entropy": 6.332171678543091,
      "epoch": 0.2688979159307665,
      "grad_norm": 1.375,
      "learning_rate": 0.0004996318052080015,
      "loss": 6.2768,
      "mean_token_accuracy": 0.14574099630117415,
      "num_tokens": 5675196.0,
      "step": 3045
    },
    {
      "entropy": 6.325894498825074,
      "epoch": 0.26933945602260684,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996300021548731,
      "loss": 6.2028,
      "mean_token_accuracy": 0.14763407558202743,
      "num_tokens": 5684361.0,
      "step": 3050
    },
    {
      "entropy": 6.266258382797242,
      "epoch": 0.26978099611444717,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004996281947013707,
      "loss": 6.1977,
      "mean_token_accuracy": 0.14618466496467591,
      "num_tokens": 5694187.0,
      "step": 3055
    },
    {
      "entropy": 6.300919008255005,
      "epoch": 0.27022253620628756,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004996263828475298,
      "loss": 6.2712,
      "mean_token_accuracy": 0.13954370617866516,
      "num_tokens": 5702858.0,
      "step": 3060
    },
    {
      "entropy": 6.261748504638672,
      "epoch": 0.2706640762981279,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004996245665933857,
      "loss": 6.1632,
      "mean_token_accuracy": 0.14856553450226784,
      "num_tokens": 5712727.0,
      "step": 3065
    },
    {
      "entropy": 6.303600072860718,
      "epoch": 0.2711056163899682,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004996227459389741,
      "loss": 6.2886,
      "mean_token_accuracy": 0.14338717311620713,
      "num_tokens": 5722904.0,
      "step": 3070
    },
    {
      "entropy": 6.386627197265625,
      "epoch": 0.27154715648180855,
      "grad_norm": 1.375,
      "learning_rate": 0.0004996209208843307,
      "loss": 6.3874,
      "mean_token_accuracy": 0.14143950417637824,
      "num_tokens": 5732228.0,
      "step": 3075
    },
    {
      "entropy": 6.375475549697876,
      "epoch": 0.2719886965736489,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996190914294912,
      "loss": 6.3109,
      "mean_token_accuracy": 0.14234673753380775,
      "num_tokens": 5743548.0,
      "step": 3080
    },
    {
      "entropy": 6.288537931442261,
      "epoch": 0.2724302366654892,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004996172575744914,
      "loss": 6.3338,
      "mean_token_accuracy": 0.13253869488835335,
      "num_tokens": 5754112.0,
      "step": 3085
    },
    {
      "entropy": 6.302958822250366,
      "epoch": 0.27287177675732954,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004996154193193673,
      "loss": 6.1664,
      "mean_token_accuracy": 0.14899687618017196,
      "num_tokens": 5763377.0,
      "step": 3090
    },
    {
      "entropy": 6.1695150375366214,
      "epoch": 0.27331331684916993,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004996135766641549,
      "loss": 6.1159,
      "mean_token_accuracy": 0.15076552629470824,
      "num_tokens": 5772000.0,
      "step": 3095
    },
    {
      "entropy": 6.356336355209351,
      "epoch": 0.27375485694101026,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004996117296088903,
      "loss": 6.2563,
      "mean_token_accuracy": 0.14282927811145782,
      "num_tokens": 5782192.0,
      "step": 3100
    },
    {
      "entropy": 6.394344520568848,
      "epoch": 0.2741963970328506,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004996098781536095,
      "loss": 6.2999,
      "mean_token_accuracy": 0.14321804642677308,
      "num_tokens": 5791163.0,
      "step": 3105
    },
    {
      "entropy": 6.244565391540528,
      "epoch": 0.2746379371246909,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004996080222983492,
      "loss": 6.3145,
      "mean_token_accuracy": 0.13703466728329658,
      "num_tokens": 5801815.0,
      "step": 3110
    },
    {
      "entropy": 6.4173187732696535,
      "epoch": 0.27507947721653125,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004996061620431453,
      "loss": 6.3207,
      "mean_token_accuracy": 0.1415737859904766,
      "num_tokens": 5811545.0,
      "step": 3115
    },
    {
      "entropy": 6.3702785015106205,
      "epoch": 0.2755210173083716,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004996042973880344,
      "loss": 6.3307,
      "mean_token_accuracy": 0.1431344673037529,
      "num_tokens": 5820285.0,
      "step": 3120
    },
    {
      "entropy": 6.2595337390899655,
      "epoch": 0.2759625574002119,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004996024283330532,
      "loss": 6.1762,
      "mean_token_accuracy": 0.14116988331079483,
      "num_tokens": 5829284.0,
      "step": 3125
    },
    {
      "entropy": 6.305311965942383,
      "epoch": 0.2764040974920523,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499600554878238,
      "loss": 6.3521,
      "mean_token_accuracy": 0.1364928737282753,
      "num_tokens": 5838973.0,
      "step": 3130
    },
    {
      "entropy": 6.342598533630371,
      "epoch": 0.27684563758389263,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004995986770236258,
      "loss": 6.2177,
      "mean_token_accuracy": 0.14314799010753632,
      "num_tokens": 5847434.0,
      "step": 3135
    },
    {
      "entropy": 6.290801858901977,
      "epoch": 0.27728717767573297,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004995967947692533,
      "loss": 6.2193,
      "mean_token_accuracy": 0.14620081037282945,
      "num_tokens": 5856464.0,
      "step": 3140
    },
    {
      "entropy": 6.327192163467407,
      "epoch": 0.2777287177675733,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004995949081151571,
      "loss": 6.2846,
      "mean_token_accuracy": 0.14579862579703332,
      "num_tokens": 5865892.0,
      "step": 3145
    },
    {
      "entropy": 6.339615058898926,
      "epoch": 0.2781702578594136,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995930170613746,
      "loss": 6.2622,
      "mean_token_accuracy": 0.14785820320248605,
      "num_tokens": 5874606.0,
      "step": 3150
    },
    {
      "entropy": 6.3258908748626705,
      "epoch": 0.27861179795125396,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004995911216079425,
      "loss": 6.2597,
      "mean_token_accuracy": 0.1452340230345726,
      "num_tokens": 5883853.0,
      "step": 3155
    },
    {
      "entropy": 6.370532512664795,
      "epoch": 0.2790533380430943,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004995892217548981,
      "loss": 6.2883,
      "mean_token_accuracy": 0.1421412155032158,
      "num_tokens": 5892841.0,
      "step": 3160
    },
    {
      "entropy": 6.340239238739014,
      "epoch": 0.2794948781349347,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004995873175022786,
      "loss": 6.259,
      "mean_token_accuracy": 0.14697980433702468,
      "num_tokens": 5901966.0,
      "step": 3165
    },
    {
      "entropy": 6.274954986572266,
      "epoch": 0.279936418226775,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004995854088501213,
      "loss": 6.1552,
      "mean_token_accuracy": 0.14191541820764542,
      "num_tokens": 5910003.0,
      "step": 3170
    },
    {
      "entropy": 6.269457006454468,
      "epoch": 0.28037795831861534,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004995834957984634,
      "loss": 6.3159,
      "mean_token_accuracy": 0.1394389145076275,
      "num_tokens": 5920307.0,
      "step": 3175
    },
    {
      "entropy": 6.303321027755738,
      "epoch": 0.28081949841045567,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004995815783473428,
      "loss": 6.1647,
      "mean_token_accuracy": 0.15029042363166809,
      "num_tokens": 5929875.0,
      "step": 3180
    },
    {
      "entropy": 6.301052713394165,
      "epoch": 0.281261038502296,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004995796564967967,
      "loss": 6.127,
      "mean_token_accuracy": 0.1431355059146881,
      "num_tokens": 5938570.0,
      "step": 3185
    },
    {
      "entropy": 6.2796125411987305,
      "epoch": 0.28170257859413633,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004995777302468628,
      "loss": 6.2874,
      "mean_token_accuracy": 0.1460157573223114,
      "num_tokens": 5947693.0,
      "step": 3190
    },
    {
      "entropy": 6.33156566619873,
      "epoch": 0.28214411868597666,
      "grad_norm": 2.0,
      "learning_rate": 0.0004995757995975789,
      "loss": 6.3641,
      "mean_token_accuracy": 0.14592357650399207,
      "num_tokens": 5957377.0,
      "step": 3195
    },
    {
      "entropy": 6.409634447097778,
      "epoch": 0.28258565877781705,
      "grad_norm": 1.5,
      "learning_rate": 0.0004995738645489828,
      "loss": 6.2884,
      "mean_token_accuracy": 0.14403491765260695,
      "num_tokens": 5966443.0,
      "step": 3200
    },
    {
      "entropy": 6.332985877990723,
      "epoch": 0.2830271988696574,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004995719251011124,
      "loss": 6.2072,
      "mean_token_accuracy": 0.14735419005155564,
      "num_tokens": 5975027.0,
      "step": 3205
    },
    {
      "entropy": 6.324332141876221,
      "epoch": 0.2834687389614977,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004995699812540058,
      "loss": 6.2752,
      "mean_token_accuracy": 0.14320788607001306,
      "num_tokens": 5983722.0,
      "step": 3210
    },
    {
      "entropy": 6.271279239654541,
      "epoch": 0.28391027905333804,
      "grad_norm": 1.5,
      "learning_rate": 0.000499568033007701,
      "loss": 6.2811,
      "mean_token_accuracy": 0.13799498602747917,
      "num_tokens": 5993358.0,
      "step": 3215
    },
    {
      "entropy": 6.32847261428833,
      "epoch": 0.2843518191451784,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004995660803622361,
      "loss": 6.2185,
      "mean_token_accuracy": 0.15176873803138732,
      "num_tokens": 6002743.0,
      "step": 3220
    },
    {
      "entropy": 6.428246402740479,
      "epoch": 0.2847933592370187,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004995641233176494,
      "loss": 6.3779,
      "mean_token_accuracy": 0.13898470029234886,
      "num_tokens": 6013727.0,
      "step": 3225
    },
    {
      "entropy": 6.417748928070068,
      "epoch": 0.28523489932885904,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004995621618739792,
      "loss": 6.3063,
      "mean_token_accuracy": 0.13999874591827394,
      "num_tokens": 6023397.0,
      "step": 3230
    },
    {
      "entropy": 6.362044048309326,
      "epoch": 0.2856764394206994,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000499560196031264,
      "loss": 6.2001,
      "mean_token_accuracy": 0.15036201179027558,
      "num_tokens": 6032453.0,
      "step": 3235
    },
    {
      "entropy": 6.2208168506622314,
      "epoch": 0.28611797951253976,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004995582257895423,
      "loss": 6.1826,
      "mean_token_accuracy": 0.13672572746872902,
      "num_tokens": 6042441.0,
      "step": 3240
    },
    {
      "entropy": 6.303688001632691,
      "epoch": 0.2865595196043801,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004995562511488528,
      "loss": 6.1648,
      "mean_token_accuracy": 0.1514339789748192,
      "num_tokens": 6050881.0,
      "step": 3245
    },
    {
      "entropy": 6.293939399719238,
      "epoch": 0.2870010596962204,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004995542721092337,
      "loss": 6.2304,
      "mean_token_accuracy": 0.14860680252313613,
      "num_tokens": 6060804.0,
      "step": 3250
    },
    {
      "entropy": 6.365213298797608,
      "epoch": 0.28744259978806075,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004995522886707244,
      "loss": 6.3369,
      "mean_token_accuracy": 0.13872155621647836,
      "num_tokens": 6070774.0,
      "step": 3255
    },
    {
      "entropy": 6.407730340957642,
      "epoch": 0.2878841398799011,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004995503008333634,
      "loss": 6.3082,
      "mean_token_accuracy": 0.1409797862172127,
      "num_tokens": 6080158.0,
      "step": 3260
    },
    {
      "entropy": 6.329580640792846,
      "epoch": 0.2883256799717414,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995483085971897,
      "loss": 6.2127,
      "mean_token_accuracy": 0.15062418729066848,
      "num_tokens": 6089183.0,
      "step": 3265
    },
    {
      "entropy": 6.3140839576721195,
      "epoch": 0.2887672200635818,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004995463119622424,
      "loss": 6.2869,
      "mean_token_accuracy": 0.1393837496638298,
      "num_tokens": 6098536.0,
      "step": 3270
    },
    {
      "entropy": 6.201459646224976,
      "epoch": 0.28920876015542213,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004995443109285604,
      "loss": 6.0704,
      "mean_token_accuracy": 0.15581393390893936,
      "num_tokens": 6107745.0,
      "step": 3275
    },
    {
      "entropy": 6.263937854766846,
      "epoch": 0.28965030024726246,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004995423054961832,
      "loss": 6.2581,
      "mean_token_accuracy": 0.14954922050237657,
      "num_tokens": 6117512.0,
      "step": 3280
    },
    {
      "entropy": 6.305516576766967,
      "epoch": 0.2900918403391028,
      "grad_norm": 1.359375,
      "learning_rate": 0.00049954029566515,
      "loss": 6.2557,
      "mean_token_accuracy": 0.14468607753515245,
      "num_tokens": 6126030.0,
      "step": 3285
    },
    {
      "entropy": 6.4128021717071535,
      "epoch": 0.2905333804309431,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004995382814355,
      "loss": 6.3,
      "mean_token_accuracy": 0.13850856572389603,
      "num_tokens": 6134888.0,
      "step": 3290
    },
    {
      "entropy": 6.3890117645263675,
      "epoch": 0.29097492052278345,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004995362628072728,
      "loss": 6.2668,
      "mean_token_accuracy": 0.14920098334550858,
      "num_tokens": 6144274.0,
      "step": 3295
    },
    {
      "entropy": 6.165565633773804,
      "epoch": 0.2914164606146238,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004995342397805078,
      "loss": 6.2744,
      "mean_token_accuracy": 0.14215826392173767,
      "num_tokens": 6153406.0,
      "step": 3300
    },
    {
      "entropy": 6.289848852157593,
      "epoch": 0.29185800070646417,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004995322123552448,
      "loss": 6.1934,
      "mean_token_accuracy": 0.14832887351512908,
      "num_tokens": 6162743.0,
      "step": 3305
    },
    {
      "entropy": 6.275933074951172,
      "epoch": 0.2922995407983045,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004995301805315235,
      "loss": 6.1805,
      "mean_token_accuracy": 0.1520318605005741,
      "num_tokens": 6171997.0,
      "step": 3310
    },
    {
      "entropy": 6.268936443328857,
      "epoch": 0.29274108089014483,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004995281443093837,
      "loss": 6.2017,
      "mean_token_accuracy": 0.1493746504187584,
      "num_tokens": 6181275.0,
      "step": 3315
    },
    {
      "entropy": 6.2795538902282715,
      "epoch": 0.29318262098198516,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004995261036888652,
      "loss": 6.2847,
      "mean_token_accuracy": 0.14096189811825752,
      "num_tokens": 6191640.0,
      "step": 3320
    },
    {
      "entropy": 6.350857257843018,
      "epoch": 0.2936241610738255,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004995240586700081,
      "loss": 6.2653,
      "mean_token_accuracy": 0.14238337501883508,
      "num_tokens": 6201508.0,
      "step": 3325
    },
    {
      "entropy": 6.261187982559204,
      "epoch": 0.2940657011656658,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004995220092528522,
      "loss": 6.1515,
      "mean_token_accuracy": 0.15054052993655204,
      "num_tokens": 6209902.0,
      "step": 3330
    },
    {
      "entropy": 6.294927644729614,
      "epoch": 0.29450724125750616,
      "grad_norm": 1.421875,
      "learning_rate": 0.000499519955437438,
      "loss": 6.308,
      "mean_token_accuracy": 0.14006401896476744,
      "num_tokens": 6219760.0,
      "step": 3335
    },
    {
      "entropy": 6.316092014312744,
      "epoch": 0.29494878134934654,
      "grad_norm": 1.375,
      "learning_rate": 0.0004995178972238054,
      "loss": 6.2726,
      "mean_token_accuracy": 0.14333543106913565,
      "num_tokens": 6228721.0,
      "step": 3340
    },
    {
      "entropy": 6.320749425888062,
      "epoch": 0.2953903214411869,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000499515834611995,
      "loss": 6.1222,
      "mean_token_accuracy": 0.15580927431583405,
      "num_tokens": 6237070.0,
      "step": 3345
    },
    {
      "entropy": 6.313472318649292,
      "epoch": 0.2958318615330272,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004995137676020472,
      "loss": 6.2513,
      "mean_token_accuracy": 0.13897364661097528,
      "num_tokens": 6245659.0,
      "step": 3350
    },
    {
      "entropy": 6.307816123962402,
      "epoch": 0.29627340162486754,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004995116961940023,
      "loss": 6.2262,
      "mean_token_accuracy": 0.14813175648450852,
      "num_tokens": 6255175.0,
      "step": 3355
    },
    {
      "entropy": 6.2639384269714355,
      "epoch": 0.29671494171670787,
      "grad_norm": 1.25,
      "learning_rate": 0.0004995096203879009,
      "loss": 6.2404,
      "mean_token_accuracy": 0.13731699511408807,
      "num_tokens": 6264962.0,
      "step": 3360
    },
    {
      "entropy": 6.2340796947479244,
      "epoch": 0.2971564818085482,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004995075401837837,
      "loss": 6.1452,
      "mean_token_accuracy": 0.15479466691613197,
      "num_tokens": 6273411.0,
      "step": 3365
    },
    {
      "entropy": 6.206551790237427,
      "epoch": 0.29759802190038853,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004995054555816915,
      "loss": 6.1274,
      "mean_token_accuracy": 0.15238229930400848,
      "num_tokens": 6282618.0,
      "step": 3370
    },
    {
      "entropy": 6.228631782531738,
      "epoch": 0.2980395619922289,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004995033665816651,
      "loss": 6.1255,
      "mean_token_accuracy": 0.15363436192274094,
      "num_tokens": 6292008.0,
      "step": 3375
    },
    {
      "entropy": 6.169542646408081,
      "epoch": 0.29848110208406925,
      "grad_norm": 1.375,
      "learning_rate": 0.0004995012731837454,
      "loss": 6.1025,
      "mean_token_accuracy": 0.15055510327219962,
      "num_tokens": 6301079.0,
      "step": 3380
    },
    {
      "entropy": 6.324261808395386,
      "epoch": 0.2989226421759096,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004994991753879736,
      "loss": 6.2718,
      "mean_token_accuracy": 0.14163144826889038,
      "num_tokens": 6310543.0,
      "step": 3385
    },
    {
      "entropy": 6.359718656539917,
      "epoch": 0.2993641822677499,
      "grad_norm": 1.875,
      "learning_rate": 0.0004994970731943904,
      "loss": 6.2171,
      "mean_token_accuracy": 0.14785383343696595,
      "num_tokens": 6320037.0,
      "step": 3390
    },
    {
      "entropy": 6.246334791183472,
      "epoch": 0.29980572235959024,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004994949666030374,
      "loss": 6.2478,
      "mean_token_accuracy": 0.14385821670293808,
      "num_tokens": 6328948.0,
      "step": 3395
    },
    {
      "entropy": 6.221922874450684,
      "epoch": 0.3002472624514306,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004994928556139557,
      "loss": 6.1036,
      "mean_token_accuracy": 0.15099234580993653,
      "num_tokens": 6337449.0,
      "step": 3400
    },
    {
      "entropy": 6.229247903823852,
      "epoch": 0.3006888025432709,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004994907402271865,
      "loss": 6.1396,
      "mean_token_accuracy": 0.15253903567790986,
      "num_tokens": 6346130.0,
      "step": 3405
    },
    {
      "entropy": 6.190823078155518,
      "epoch": 0.3011303426351113,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004994886204427715,
      "loss": 6.1814,
      "mean_token_accuracy": 0.15078601539134978,
      "num_tokens": 6354973.0,
      "step": 3410
    },
    {
      "entropy": 6.341151475906372,
      "epoch": 0.3015718827269516,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004994864962607519,
      "loss": 6.2574,
      "mean_token_accuracy": 0.14179314076900482,
      "num_tokens": 6364107.0,
      "step": 3415
    },
    {
      "entropy": 6.328457355499268,
      "epoch": 0.30201342281879195,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004994843676811697,
      "loss": 6.2059,
      "mean_token_accuracy": 0.1432513862848282,
      "num_tokens": 6372859.0,
      "step": 3420
    },
    {
      "entropy": 6.23663182258606,
      "epoch": 0.3024549629106323,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004994822347040664,
      "loss": 6.092,
      "mean_token_accuracy": 0.15835918337106705,
      "num_tokens": 6381818.0,
      "step": 3425
    },
    {
      "entropy": 6.300030136108399,
      "epoch": 0.3028965030024726,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004994800973294837,
      "loss": 6.2498,
      "mean_token_accuracy": 0.14488281309604645,
      "num_tokens": 6391460.0,
      "step": 3430
    },
    {
      "entropy": 6.328920793533325,
      "epoch": 0.30333804309431295,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004994779555574636,
      "loss": 6.2547,
      "mean_token_accuracy": 0.14561234712600707,
      "num_tokens": 6401461.0,
      "step": 3435
    },
    {
      "entropy": 6.293675470352173,
      "epoch": 0.3037795831861533,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000499475809388048,
      "loss": 6.1428,
      "mean_token_accuracy": 0.1438477359712124,
      "num_tokens": 6410435.0,
      "step": 3440
    },
    {
      "entropy": 6.337829685211181,
      "epoch": 0.30422112327799367,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000499473658821279,
      "loss": 6.3612,
      "mean_token_accuracy": 0.1367957629263401,
      "num_tokens": 6420710.0,
      "step": 3445
    },
    {
      "entropy": 6.429217433929443,
      "epoch": 0.304662663369834,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004994715038571986,
      "loss": 6.2124,
      "mean_token_accuracy": 0.1464938424527645,
      "num_tokens": 6429882.0,
      "step": 3450
    },
    {
      "entropy": 6.211455345153809,
      "epoch": 0.30510420346167433,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004994693444958493,
      "loss": 6.1677,
      "mean_token_accuracy": 0.15407330840826033,
      "num_tokens": 6439183.0,
      "step": 3455
    },
    {
      "entropy": 6.292701482772827,
      "epoch": 0.30554574355351466,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000499467180737273,
      "loss": 6.3449,
      "mean_token_accuracy": 0.1406208023428917,
      "num_tokens": 6448460.0,
      "step": 3460
    },
    {
      "entropy": 6.349296760559082,
      "epoch": 0.305987283645355,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004994650125815124,
      "loss": 6.2131,
      "mean_token_accuracy": 0.14687474071979523,
      "num_tokens": 6457687.0,
      "step": 3465
    },
    {
      "entropy": 6.28594217300415,
      "epoch": 0.3064288237371953,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004994628400286097,
      "loss": 6.1717,
      "mean_token_accuracy": 0.148153455555439,
      "num_tokens": 6467057.0,
      "step": 3470
    },
    {
      "entropy": 6.252876710891724,
      "epoch": 0.30687036382903565,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004994606630786078,
      "loss": 6.2493,
      "mean_token_accuracy": 0.1461620330810547,
      "num_tokens": 6477484.0,
      "step": 3475
    },
    {
      "entropy": 6.321864604949951,
      "epoch": 0.30731190392087604,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004994584817315492,
      "loss": 6.3043,
      "mean_token_accuracy": 0.1443194717168808,
      "num_tokens": 6488381.0,
      "step": 3480
    },
    {
      "entropy": 6.296575021743775,
      "epoch": 0.30775344401271637,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004994562959874765,
      "loss": 6.196,
      "mean_token_accuracy": 0.14554316326975822,
      "num_tokens": 6498279.0,
      "step": 3485
    },
    {
      "entropy": 6.269347381591797,
      "epoch": 0.3081949841045567,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004994541058464326,
      "loss": 6.2109,
      "mean_token_accuracy": 0.14428319782018661,
      "num_tokens": 6508008.0,
      "step": 3490
    },
    {
      "entropy": 6.34035325050354,
      "epoch": 0.30863652419639703,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004994519113084605,
      "loss": 6.2474,
      "mean_token_accuracy": 0.14133610650897027,
      "num_tokens": 6517263.0,
      "step": 3495
    },
    {
      "entropy": 6.349197673797607,
      "epoch": 0.30907806428823736,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004994497123736029,
      "loss": 6.3315,
      "mean_token_accuracy": 0.13482841178774835,
      "num_tokens": 6527682.0,
      "step": 3500
    },
    {
      "entropy": 6.293285512924195,
      "epoch": 0.3095196043800777,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004994475090419034,
      "loss": 6.1941,
      "mean_token_accuracy": 0.14769349247217178,
      "num_tokens": 6537143.0,
      "step": 3505
    },
    {
      "entropy": 6.316049528121948,
      "epoch": 0.309961144471918,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004994453013134047,
      "loss": 6.2011,
      "mean_token_accuracy": 0.14517299979925155,
      "num_tokens": 6546561.0,
      "step": 3510
    },
    {
      "entropy": 6.277545070648193,
      "epoch": 0.3104026845637584,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004994430891881502,
      "loss": 6.2015,
      "mean_token_accuracy": 0.14352679401636123,
      "num_tokens": 6555806.0,
      "step": 3515
    },
    {
      "entropy": 6.268626976013183,
      "epoch": 0.31084422465559874,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004994408726661832,
      "loss": 6.231,
      "mean_token_accuracy": 0.1447139360010624,
      "num_tokens": 6564836.0,
      "step": 3520
    },
    {
      "entropy": 6.2405225276947025,
      "epoch": 0.3112857647474391,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004994386517475472,
      "loss": 6.2323,
      "mean_token_accuracy": 0.13839124515652657,
      "num_tokens": 6574997.0,
      "step": 3525
    },
    {
      "entropy": 6.326260805130005,
      "epoch": 0.3117273048392794,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004994364264322856,
      "loss": 6.2421,
      "mean_token_accuracy": 0.14486346989870072,
      "num_tokens": 6584589.0,
      "step": 3530
    },
    {
      "entropy": 6.287415504455566,
      "epoch": 0.31216884493111974,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004994341967204421,
      "loss": 6.186,
      "mean_token_accuracy": 0.15003243535757066,
      "num_tokens": 6592898.0,
      "step": 3535
    },
    {
      "entropy": 6.322659063339233,
      "epoch": 0.31261038502296007,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004994319626120603,
      "loss": 6.1982,
      "mean_token_accuracy": 0.15189156383275987,
      "num_tokens": 6602454.0,
      "step": 3540
    },
    {
      "entropy": 6.218482780456543,
      "epoch": 0.3130519251148004,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004994297241071841,
      "loss": 6.2155,
      "mean_token_accuracy": 0.1479742333292961,
      "num_tokens": 6612454.0,
      "step": 3545
    },
    {
      "entropy": 6.29142951965332,
      "epoch": 0.3134934652066408,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000499427481205857,
      "loss": 6.1926,
      "mean_token_accuracy": 0.15044232308864594,
      "num_tokens": 6621286.0,
      "step": 3550
    },
    {
      "entropy": 6.223646450042724,
      "epoch": 0.3139350052984811,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004994252339081234,
      "loss": 6.1325,
      "mean_token_accuracy": 0.15221345126628877,
      "num_tokens": 6629684.0,
      "step": 3555
    },
    {
      "entropy": 6.260682678222656,
      "epoch": 0.31437654539032145,
      "grad_norm": 1.234375,
      "learning_rate": 0.000499422982214027,
      "loss": 6.2336,
      "mean_token_accuracy": 0.15286535322666167,
      "num_tokens": 6638836.0,
      "step": 3560
    },
    {
      "entropy": 6.343380880355835,
      "epoch": 0.3148180854821618,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004994207261236121,
      "loss": 6.1827,
      "mean_token_accuracy": 0.14693615287542344,
      "num_tokens": 6647337.0,
      "step": 3565
    },
    {
      "entropy": 6.189584636688233,
      "epoch": 0.3152596255740021,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004994184656369227,
      "loss": 6.1817,
      "mean_token_accuracy": 0.14338988438248634,
      "num_tokens": 6657246.0,
      "step": 3570
    },
    {
      "entropy": 6.359672164916992,
      "epoch": 0.31570116566584244,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004994162007540033,
      "loss": 6.3109,
      "mean_token_accuracy": 0.13514265939593315,
      "num_tokens": 6667981.0,
      "step": 3575
    },
    {
      "entropy": 6.32397427558899,
      "epoch": 0.3161427057576828,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004994139314748981,
      "loss": 6.16,
      "mean_token_accuracy": 0.15298724174499512,
      "num_tokens": 6677462.0,
      "step": 3580
    },
    {
      "entropy": 6.176218175888062,
      "epoch": 0.31658424584952316,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004994116577996517,
      "loss": 6.1994,
      "mean_token_accuracy": 0.14874908104538917,
      "num_tokens": 6687225.0,
      "step": 3585
    },
    {
      "entropy": 6.252984666824341,
      "epoch": 0.3170257859413635,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004994093797283084,
      "loss": 6.2788,
      "mean_token_accuracy": 0.14002985283732414,
      "num_tokens": 6696345.0,
      "step": 3590
    },
    {
      "entropy": 6.361448812484741,
      "epoch": 0.3174673260332038,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004994070972609132,
      "loss": 6.2544,
      "mean_token_accuracy": 0.14232457354664801,
      "num_tokens": 6706207.0,
      "step": 3595
    },
    {
      "entropy": 6.315224981307983,
      "epoch": 0.31790886612504415,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004994048103975103,
      "loss": 6.1593,
      "mean_token_accuracy": 0.14630399346351625,
      "num_tokens": 6714430.0,
      "step": 3600
    },
    {
      "entropy": 6.237560892105103,
      "epoch": 0.3183504062168845,
      "grad_norm": 1.5,
      "learning_rate": 0.000499402519138145,
      "loss": 6.1883,
      "mean_token_accuracy": 0.1458396703004837,
      "num_tokens": 6724139.0,
      "step": 3605
    },
    {
      "entropy": 6.241400671005249,
      "epoch": 0.3187919463087248,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004994002234828619,
      "loss": 6.2346,
      "mean_token_accuracy": 0.1450707234442234,
      "num_tokens": 6733525.0,
      "step": 3610
    },
    {
      "entropy": 6.343650865554809,
      "epoch": 0.31923348640056515,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000499397923431706,
      "loss": 6.3247,
      "mean_token_accuracy": 0.13921364694833754,
      "num_tokens": 6743738.0,
      "step": 3615
    },
    {
      "entropy": 6.342299556732177,
      "epoch": 0.31967502649240553,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004993956189847226,
      "loss": 6.1986,
      "mean_token_accuracy": 0.1489830307662487,
      "num_tokens": 6752900.0,
      "step": 3620
    },
    {
      "entropy": 6.32194037437439,
      "epoch": 0.32011656658424587,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004993933101419565,
      "loss": 6.158,
      "mean_token_accuracy": 0.1525537133216858,
      "num_tokens": 6761776.0,
      "step": 3625
    },
    {
      "entropy": 6.306015205383301,
      "epoch": 0.3205581066760862,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004993909969034531,
      "loss": 6.3116,
      "mean_token_accuracy": 0.14113244116306306,
      "num_tokens": 6771543.0,
      "step": 3630
    },
    {
      "entropy": 6.283383131027222,
      "epoch": 0.3209996467679265,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004993886792692576,
      "loss": 6.1941,
      "mean_token_accuracy": 0.1407588765025139,
      "num_tokens": 6780010.0,
      "step": 3635
    },
    {
      "entropy": 6.345666599273682,
      "epoch": 0.32144118685976686,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004993863572394156,
      "loss": 6.2455,
      "mean_token_accuracy": 0.1481287196278572,
      "num_tokens": 6788830.0,
      "step": 3640
    },
    {
      "entropy": 6.319739246368409,
      "epoch": 0.3218827269516072,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004993840308139724,
      "loss": 6.2361,
      "mean_token_accuracy": 0.14790267944335939,
      "num_tokens": 6799121.0,
      "step": 3645
    },
    {
      "entropy": 6.272012615203858,
      "epoch": 0.3223242670434475,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004993816999929738,
      "loss": 6.1873,
      "mean_token_accuracy": 0.1459096811711788,
      "num_tokens": 6808519.0,
      "step": 3650
    },
    {
      "entropy": 6.246146488189697,
      "epoch": 0.3227658071352879,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004993793647764651,
      "loss": 6.2032,
      "mean_token_accuracy": 0.15324788242578508,
      "num_tokens": 6817583.0,
      "step": 3655
    },
    {
      "entropy": 6.358306646347046,
      "epoch": 0.32320734722712824,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004993770251644923,
      "loss": 6.2018,
      "mean_token_accuracy": 0.14584973454475403,
      "num_tokens": 6827732.0,
      "step": 3660
    },
    {
      "entropy": 6.219063949584961,
      "epoch": 0.32364888731896857,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004993746811571013,
      "loss": 6.1804,
      "mean_token_accuracy": 0.14989982694387435,
      "num_tokens": 6836903.0,
      "step": 3665
    },
    {
      "entropy": 6.157657623291016,
      "epoch": 0.3240904274108089,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004993723327543379,
      "loss": 6.184,
      "mean_token_accuracy": 0.15377924591302872,
      "num_tokens": 6845684.0,
      "step": 3670
    },
    {
      "entropy": 6.22325701713562,
      "epoch": 0.32453196750264923,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000499369979956248,
      "loss": 6.163,
      "mean_token_accuracy": 0.14639684483408927,
      "num_tokens": 6854650.0,
      "step": 3675
    },
    {
      "entropy": 6.313974905014038,
      "epoch": 0.32497350759448956,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004993676227628779,
      "loss": 6.182,
      "mean_token_accuracy": 0.14079167023301126,
      "num_tokens": 6864340.0,
      "step": 3680
    },
    {
      "entropy": 6.354143953323364,
      "epoch": 0.3254150476863299,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004993652611742736,
      "loss": 6.2106,
      "mean_token_accuracy": 0.14864241406321527,
      "num_tokens": 6873244.0,
      "step": 3685
    },
    {
      "entropy": 6.2044641971588135,
      "epoch": 0.3258565877781703,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004993628951904815,
      "loss": 6.0951,
      "mean_token_accuracy": 0.15563804358243943,
      "num_tokens": 6882392.0,
      "step": 3690
    },
    {
      "entropy": 6.283623170852661,
      "epoch": 0.3262981278700106,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004993605248115479,
      "loss": 6.3166,
      "mean_token_accuracy": 0.1375298321247101,
      "num_tokens": 6891515.0,
      "step": 3695
    },
    {
      "entropy": 6.39955735206604,
      "epoch": 0.32673966796185094,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004993581500375191,
      "loss": 6.2063,
      "mean_token_accuracy": 0.14319797828793526,
      "num_tokens": 6900104.0,
      "step": 3700
    },
    {
      "entropy": 6.301495885848999,
      "epoch": 0.3271812080536913,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004993557708684417,
      "loss": 6.3156,
      "mean_token_accuracy": 0.1383724741637707,
      "num_tokens": 6910443.0,
      "step": 3705
    },
    {
      "entropy": 6.2295175075531,
      "epoch": 0.3276227481455316,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004993533873043625,
      "loss": 6.1773,
      "mean_token_accuracy": 0.1472519099712372,
      "num_tokens": 6919882.0,
      "step": 3710
    },
    {
      "entropy": 6.3492731094360355,
      "epoch": 0.32806428823737194,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000499350999345328,
      "loss": 6.3019,
      "mean_token_accuracy": 0.1388927437365055,
      "num_tokens": 6929757.0,
      "step": 3715
    },
    {
      "entropy": 6.29475884437561,
      "epoch": 0.32850582832921227,
      "grad_norm": 1.265625,
      "learning_rate": 0.000499348606991385,
      "loss": 6.2597,
      "mean_token_accuracy": 0.13804953098297118,
      "num_tokens": 6938578.0,
      "step": 3720
    },
    {
      "entropy": 6.323968458175659,
      "epoch": 0.32894736842105265,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004993462102425805,
      "loss": 6.1738,
      "mean_token_accuracy": 0.1445832960307598,
      "num_tokens": 6947920.0,
      "step": 3725
    },
    {
      "entropy": 6.224078845977783,
      "epoch": 0.329388908512893,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004993438090989612,
      "loss": 6.1218,
      "mean_token_accuracy": 0.1494213819503784,
      "num_tokens": 6957291.0,
      "step": 3730
    },
    {
      "entropy": 6.270564126968384,
      "epoch": 0.3298304486047333,
      "grad_norm": 1.5,
      "learning_rate": 0.0004993414035605743,
      "loss": 6.1795,
      "mean_token_accuracy": 0.15251390635967255,
      "num_tokens": 6966504.0,
      "step": 3735
    },
    {
      "entropy": 6.331764554977417,
      "epoch": 0.33027198869657365,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004993389936274669,
      "loss": 6.213,
      "mean_token_accuracy": 0.14998254030942917,
      "num_tokens": 6976234.0,
      "step": 3740
    },
    {
      "entropy": 6.225750160217285,
      "epoch": 0.330713528788414,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004993365792996862,
      "loss": 6.2674,
      "mean_token_accuracy": 0.1404741570353508,
      "num_tokens": 6986372.0,
      "step": 3745
    },
    {
      "entropy": 6.320698499679565,
      "epoch": 0.3311550688802543,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004993341605772795,
      "loss": 6.1535,
      "mean_token_accuracy": 0.1522038422524929,
      "num_tokens": 6995553.0,
      "step": 3750
    },
    {
      "entropy": 6.248960161209107,
      "epoch": 0.33159660897209464,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993317374602941,
      "loss": 6.0784,
      "mean_token_accuracy": 0.1559259220957756,
      "num_tokens": 7005675.0,
      "step": 3755
    },
    {
      "entropy": 6.173632669448852,
      "epoch": 0.33203814906393503,
      "grad_norm": 1.25,
      "learning_rate": 0.0004993293099487777,
      "loss": 6.2065,
      "mean_token_accuracy": 0.14199020266532897,
      "num_tokens": 7014953.0,
      "step": 3760
    },
    {
      "entropy": 6.267611455917359,
      "epoch": 0.33247968915577536,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004993268780427776,
      "loss": 6.2142,
      "mean_token_accuracy": 0.15134309455752373,
      "num_tokens": 7025075.0,
      "step": 3765
    },
    {
      "entropy": 6.312801742553711,
      "epoch": 0.3329212292476157,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004993244417423416,
      "loss": 6.2068,
      "mean_token_accuracy": 0.15050857216119767,
      "num_tokens": 7034286.0,
      "step": 3770
    },
    {
      "entropy": 6.307934761047363,
      "epoch": 0.333362769339456,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004993220010475174,
      "loss": 6.2453,
      "mean_token_accuracy": 0.14248408898711204,
      "num_tokens": 7043957.0,
      "step": 3775
    },
    {
      "entropy": 6.289632654190063,
      "epoch": 0.33380430943129635,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004993195559583526,
      "loss": 6.1572,
      "mean_token_accuracy": 0.1487005241215229,
      "num_tokens": 7053170.0,
      "step": 3780
    },
    {
      "entropy": 6.1731133460998535,
      "epoch": 0.3342458495231367,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004993171064748954,
      "loss": 6.1612,
      "mean_token_accuracy": 0.14813224822282792,
      "num_tokens": 7062469.0,
      "step": 3785
    },
    {
      "entropy": 6.255552530288696,
      "epoch": 0.334687389614977,
      "grad_norm": 2.125,
      "learning_rate": 0.0004993146525971937,
      "loss": 6.1493,
      "mean_token_accuracy": 0.15376724749803544,
      "num_tokens": 7071463.0,
      "step": 3790
    },
    {
      "entropy": 6.3743010520935055,
      "epoch": 0.3351289297068174,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004993121943252955,
      "loss": 6.2374,
      "mean_token_accuracy": 0.13975025340914726,
      "num_tokens": 7081574.0,
      "step": 3795
    },
    {
      "entropy": 6.262507772445678,
      "epoch": 0.33557046979865773,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004993097316592489,
      "loss": 6.1487,
      "mean_token_accuracy": 0.15553452372550963,
      "num_tokens": 7090835.0,
      "step": 3800
    },
    {
      "entropy": 6.15237603187561,
      "epoch": 0.33601200989049806,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004993072645991023,
      "loss": 6.0825,
      "mean_token_accuracy": 0.1396365225315094,
      "num_tokens": 7100521.0,
      "step": 3805
    },
    {
      "entropy": 6.21912112236023,
      "epoch": 0.3364535499823384,
      "grad_norm": 1.359375,
      "learning_rate": 0.000499304793144904,
      "loss": 6.1712,
      "mean_token_accuracy": 0.1433882623910904,
      "num_tokens": 7109011.0,
      "step": 3810
    },
    {
      "entropy": 6.306987667083741,
      "epoch": 0.3368950900741787,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004993023172967022,
      "loss": 6.1968,
      "mean_token_accuracy": 0.1458859533071518,
      "num_tokens": 7119128.0,
      "step": 3815
    },
    {
      "entropy": 6.302616548538208,
      "epoch": 0.33733663016601906,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004992998370545458,
      "loss": 6.2696,
      "mean_token_accuracy": 0.13797667175531386,
      "num_tokens": 7128313.0,
      "step": 3820
    },
    {
      "entropy": 6.243565607070923,
      "epoch": 0.3377781702578594,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004992973524184831,
      "loss": 6.2332,
      "mean_token_accuracy": 0.145349483191967,
      "num_tokens": 7137567.0,
      "step": 3825
    },
    {
      "entropy": 6.308232927322388,
      "epoch": 0.3382197103496998,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004992948633885627,
      "loss": 6.2107,
      "mean_token_accuracy": 0.147455532848835,
      "num_tokens": 7147254.0,
      "step": 3830
    },
    {
      "entropy": 6.2316985607147215,
      "epoch": 0.3386612504415401,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004992923699648335,
      "loss": 6.1269,
      "mean_token_accuracy": 0.15241755098104476,
      "num_tokens": 7156215.0,
      "step": 3835
    },
    {
      "entropy": 6.3128533363342285,
      "epoch": 0.33910279053338044,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004992898721473445,
      "loss": 6.2145,
      "mean_token_accuracy": 0.1414943240582943,
      "num_tokens": 7165653.0,
      "step": 3840
    },
    {
      "entropy": 6.208751821517945,
      "epoch": 0.33954433062522077,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004992873699361444,
      "loss": 6.1056,
      "mean_token_accuracy": 0.14766833633184434,
      "num_tokens": 7174797.0,
      "step": 3845
    },
    {
      "entropy": 6.29413628578186,
      "epoch": 0.3399858707170611,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004992848633312822,
      "loss": 6.1408,
      "mean_token_accuracy": 0.14888592660427094,
      "num_tokens": 7184040.0,
      "step": 3850
    },
    {
      "entropy": 6.273240566253662,
      "epoch": 0.34042741080890143,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004992823523328071,
      "loss": 6.097,
      "mean_token_accuracy": 0.15565756857395172,
      "num_tokens": 7193533.0,
      "step": 3855
    },
    {
      "entropy": 6.19666600227356,
      "epoch": 0.34086895090074176,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004992798369407684,
      "loss": 6.1902,
      "mean_token_accuracy": 0.14624166190624238,
      "num_tokens": 7203355.0,
      "step": 3860
    },
    {
      "entropy": 6.274708843231201,
      "epoch": 0.34131049099258215,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004992773171552152,
      "loss": 6.1325,
      "mean_token_accuracy": 0.14895608872175217,
      "num_tokens": 7212569.0,
      "step": 3865
    },
    {
      "entropy": 6.172758293151856,
      "epoch": 0.3417520310844225,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004992747929761968,
      "loss": 6.1614,
      "mean_token_accuracy": 0.15170100927352906,
      "num_tokens": 7221886.0,
      "step": 3870
    },
    {
      "entropy": 6.239678812026978,
      "epoch": 0.3421935711762628,
      "grad_norm": 1.5,
      "learning_rate": 0.0004992722644037628,
      "loss": 6.2203,
      "mean_token_accuracy": 0.14662540405988694,
      "num_tokens": 7232065.0,
      "step": 3875
    },
    {
      "entropy": 6.307174253463745,
      "epoch": 0.34263511126810314,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004992697314379628,
      "loss": 6.1509,
      "mean_token_accuracy": 0.1447463095188141,
      "num_tokens": 7242287.0,
      "step": 3880
    },
    {
      "entropy": 6.272569704055786,
      "epoch": 0.3430766513599435,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004992671940788462,
      "loss": 6.1405,
      "mean_token_accuracy": 0.14477290213108063,
      "num_tokens": 7250638.0,
      "step": 3885
    },
    {
      "entropy": 6.187071228027344,
      "epoch": 0.3435181914517838,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004992646523264628,
      "loss": 6.1539,
      "mean_token_accuracy": 0.15122336894273758,
      "num_tokens": 7259526.0,
      "step": 3890
    },
    {
      "entropy": 6.216576433181762,
      "epoch": 0.34395973154362414,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004992621061808625,
      "loss": 6.1455,
      "mean_token_accuracy": 0.1504262939095497,
      "num_tokens": 7269260.0,
      "step": 3895
    },
    {
      "entropy": 6.225019454956055,
      "epoch": 0.3444012716354645,
      "grad_norm": 2.21875,
      "learning_rate": 0.000499259555642095,
      "loss": 6.167,
      "mean_token_accuracy": 0.143419449031353,
      "num_tokens": 7279061.0,
      "step": 3900
    },
    {
      "entropy": 6.263490152359009,
      "epoch": 0.34484281172730485,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004992570007102104,
      "loss": 6.15,
      "mean_token_accuracy": 0.15289402306079863,
      "num_tokens": 7289721.0,
      "step": 3905
    },
    {
      "entropy": 6.392903089523315,
      "epoch": 0.3452843518191452,
      "grad_norm": 1.625,
      "learning_rate": 0.0004992544413852587,
      "loss": 6.2987,
      "mean_token_accuracy": 0.14017098546028137,
      "num_tokens": 7299836.0,
      "step": 3910
    },
    {
      "entropy": 6.341606092453003,
      "epoch": 0.3457258919109855,
      "grad_norm": 1.4375,
      "learning_rate": 0.00049925187766729,
      "loss": 6.3001,
      "mean_token_accuracy": 0.1420486383140087,
      "num_tokens": 7309306.0,
      "step": 3915
    },
    {
      "entropy": 6.297093105316162,
      "epoch": 0.34616743200282585,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004992493095563545,
      "loss": 6.1999,
      "mean_token_accuracy": 0.15256548225879668,
      "num_tokens": 7318486.0,
      "step": 3920
    },
    {
      "entropy": 6.253756761550903,
      "epoch": 0.3466089720946662,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004992467370525026,
      "loss": 6.1648,
      "mean_token_accuracy": 0.1465355783700943,
      "num_tokens": 7327764.0,
      "step": 3925
    },
    {
      "entropy": 6.29128246307373,
      "epoch": 0.3470505121865065,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004992441601557848,
      "loss": 6.2269,
      "mean_token_accuracy": 0.15001798868179322,
      "num_tokens": 7337118.0,
      "step": 3930
    },
    {
      "entropy": 6.215613794326782,
      "epoch": 0.3474920522783469,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004992415788662514,
      "loss": 6.1704,
      "mean_token_accuracy": 0.14473639652132989,
      "num_tokens": 7345973.0,
      "step": 3935
    },
    {
      "entropy": 6.328490686416626,
      "epoch": 0.34793359237018723,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004992389931839529,
      "loss": 6.1743,
      "mean_token_accuracy": 0.14794324338436127,
      "num_tokens": 7355776.0,
      "step": 3940
    },
    {
      "entropy": 6.2720846176147464,
      "epoch": 0.34837513246202756,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004992364031089401,
      "loss": 6.1858,
      "mean_token_accuracy": 0.1513002872467041,
      "num_tokens": 7364820.0,
      "step": 3945
    },
    {
      "entropy": 6.277421283721924,
      "epoch": 0.3488166725538679,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004992338086412636,
      "loss": 6.2091,
      "mean_token_accuracy": 0.14662266448140143,
      "num_tokens": 7373881.0,
      "step": 3950
    },
    {
      "entropy": 6.260824489593506,
      "epoch": 0.3492582126457082,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004992312097809744,
      "loss": 6.1732,
      "mean_token_accuracy": 0.15294966846704483,
      "num_tokens": 7383095.0,
      "step": 3955
    },
    {
      "entropy": 6.289446401596069,
      "epoch": 0.34969975273754855,
      "grad_norm": 1.5,
      "learning_rate": 0.0004992286065281234,
      "loss": 6.1876,
      "mean_token_accuracy": 0.14803554639220237,
      "num_tokens": 7392702.0,
      "step": 3960
    },
    {
      "entropy": 6.307240962982178,
      "epoch": 0.3501412928293889,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004992259988827614,
      "loss": 6.3518,
      "mean_token_accuracy": 0.13984315916895868,
      "num_tokens": 7402560.0,
      "step": 3965
    },
    {
      "entropy": 6.256998872756958,
      "epoch": 0.35058283292122927,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004992233868449397,
      "loss": 6.1293,
      "mean_token_accuracy": 0.15598249137401582,
      "num_tokens": 7412375.0,
      "step": 3970
    },
    {
      "entropy": 6.326172494888306,
      "epoch": 0.3510243730130696,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004992207704147093,
      "loss": 6.1884,
      "mean_token_accuracy": 0.14315188080072402,
      "num_tokens": 7420734.0,
      "step": 3975
    },
    {
      "entropy": 6.247087144851685,
      "epoch": 0.35146591310490993,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004992181495921216,
      "loss": 6.1647,
      "mean_token_accuracy": 0.14719347208738326,
      "num_tokens": 7430156.0,
      "step": 3980
    },
    {
      "entropy": 6.216844654083252,
      "epoch": 0.35190745319675026,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004992155243772277,
      "loss": 6.1851,
      "mean_token_accuracy": 0.1513088494539261,
      "num_tokens": 7439189.0,
      "step": 3985
    },
    {
      "entropy": 6.258322238922119,
      "epoch": 0.3523489932885906,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004992128947700795,
      "loss": 6.1879,
      "mean_token_accuracy": 0.14627946466207503,
      "num_tokens": 7448768.0,
      "step": 3990
    },
    {
      "entropy": 6.253720235824585,
      "epoch": 0.3527905333804309,
      "grad_norm": 1.703125,
      "learning_rate": 0.000499210260770728,
      "loss": 6.1574,
      "mean_token_accuracy": 0.14627916291356086,
      "num_tokens": 7458468.0,
      "step": 3995
    },
    {
      "entropy": 6.266324186325074,
      "epoch": 0.35323207347227126,
      "grad_norm": 1.9296875,
      "learning_rate": 0.000499207622379225,
      "loss": 6.2229,
      "mean_token_accuracy": 0.1418443389236927,
      "num_tokens": 7467351.0,
      "step": 4000
    },
    {
      "entropy": 6.345872783660889,
      "epoch": 0.35367361356411164,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004992049795956222,
      "loss": 6.19,
      "mean_token_accuracy": 0.1431963935494423,
      "num_tokens": 7477116.0,
      "step": 4005
    },
    {
      "entropy": 6.257971954345703,
      "epoch": 0.354115153655952,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004992023324199715,
      "loss": 6.1043,
      "mean_token_accuracy": 0.15152290165424348,
      "num_tokens": 7485216.0,
      "step": 4010
    },
    {
      "entropy": 6.25113000869751,
      "epoch": 0.3545566937477923,
      "grad_norm": 1.625,
      "learning_rate": 0.0004991996808523245,
      "loss": 6.2547,
      "mean_token_accuracy": 0.14295927956700324,
      "num_tokens": 7495084.0,
      "step": 4015
    },
    {
      "entropy": 6.294170665740967,
      "epoch": 0.35499823383963264,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004991970248927332,
      "loss": 6.2849,
      "mean_token_accuracy": 0.14390699565410614,
      "num_tokens": 7503219.0,
      "step": 4020
    },
    {
      "entropy": 6.271115255355835,
      "epoch": 0.35543977393147297,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004991943645412498,
      "loss": 6.0751,
      "mean_token_accuracy": 0.15546453595161439,
      "num_tokens": 7511961.0,
      "step": 4025
    },
    {
      "entropy": 6.16305022239685,
      "epoch": 0.3558813140233133,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004991916997979263,
      "loss": 6.1843,
      "mean_token_accuracy": 0.1484257772564888,
      "num_tokens": 7521927.0,
      "step": 4030
    },
    {
      "entropy": 6.162647819519043,
      "epoch": 0.35632285411515363,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004991890306628149,
      "loss": 6.1326,
      "mean_token_accuracy": 0.151800037920475,
      "num_tokens": 7531093.0,
      "step": 4035
    },
    {
      "entropy": 6.230826377868652,
      "epoch": 0.356764394206994,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004991863571359678,
      "loss": 6.094,
      "mean_token_accuracy": 0.14804090559482574,
      "num_tokens": 7539813.0,
      "step": 4040
    },
    {
      "entropy": 6.34772834777832,
      "epoch": 0.35720593429883435,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004991836792174376,
      "loss": 6.2649,
      "mean_token_accuracy": 0.1414028562605381,
      "num_tokens": 7548954.0,
      "step": 4045
    },
    {
      "entropy": 6.269259119033814,
      "epoch": 0.3576474743906747,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004991809969072765,
      "loss": 6.1202,
      "mean_token_accuracy": 0.14746632128953935,
      "num_tokens": 7558275.0,
      "step": 4050
    },
    {
      "entropy": 6.181514644622803,
      "epoch": 0.358089014482515,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004991783102055371,
      "loss": 6.1689,
      "mean_token_accuracy": 0.15268539562821387,
      "num_tokens": 7567870.0,
      "step": 4055
    },
    {
      "entropy": 6.279887056350708,
      "epoch": 0.35853055457435534,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004991756191122723,
      "loss": 6.1269,
      "mean_token_accuracy": 0.15328674614429474,
      "num_tokens": 7577485.0,
      "step": 4060
    },
    {
      "entropy": 6.298182344436645,
      "epoch": 0.3589720946661957,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004991729236275346,
      "loss": 6.1762,
      "mean_token_accuracy": 0.14742286801338195,
      "num_tokens": 7587398.0,
      "step": 4065
    },
    {
      "entropy": 6.24832558631897,
      "epoch": 0.359413634758036,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004991702237513768,
      "loss": 6.1763,
      "mean_token_accuracy": 0.1459968388080597,
      "num_tokens": 7596310.0,
      "step": 4070
    },
    {
      "entropy": 6.258885526657105,
      "epoch": 0.3598551748498764,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004991675194838517,
      "loss": 6.1955,
      "mean_token_accuracy": 0.138567902892828,
      "num_tokens": 7606971.0,
      "step": 4075
    },
    {
      "entropy": 6.2136951923370365,
      "epoch": 0.3602967149417167,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004991648108250125,
      "loss": 6.175,
      "mean_token_accuracy": 0.14450421258807183,
      "num_tokens": 7616636.0,
      "step": 4080
    },
    {
      "entropy": 6.337588310241699,
      "epoch": 0.36073825503355705,
      "grad_norm": 1.703125,
      "learning_rate": 0.000499162097774912,
      "loss": 6.214,
      "mean_token_accuracy": 0.14829875528812408,
      "num_tokens": 7625128.0,
      "step": 4085
    },
    {
      "entropy": 6.297661161422729,
      "epoch": 0.3611797951253974,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004991593803336037,
      "loss": 6.1301,
      "mean_token_accuracy": 0.1504852756857872,
      "num_tokens": 7634224.0,
      "step": 4090
    },
    {
      "entropy": 6.2248612403869625,
      "epoch": 0.3616213352172377,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004991566585011405,
      "loss": 6.1054,
      "mean_token_accuracy": 0.1508861094713211,
      "num_tokens": 7643392.0,
      "step": 4095
    },
    {
      "entropy": 6.141217947006226,
      "epoch": 0.36206287530907805,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004991539322775758,
      "loss": 6.142,
      "mean_token_accuracy": 0.1469010017812252,
      "num_tokens": 7652491.0,
      "step": 4100
    },
    {
      "entropy": 6.402379751205444,
      "epoch": 0.3625044154009184,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004991512016629632,
      "loss": 6.3451,
      "mean_token_accuracy": 0.13396419063210488,
      "num_tokens": 7663192.0,
      "step": 4105
    },
    {
      "entropy": 6.351581001281739,
      "epoch": 0.36294595549275877,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004991484666573558,
      "loss": 6.1872,
      "mean_token_accuracy": 0.14387465864419938,
      "num_tokens": 7672801.0,
      "step": 4110
    },
    {
      "entropy": 6.290612125396729,
      "epoch": 0.3633874955845991,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004991457272608077,
      "loss": 6.2029,
      "mean_token_accuracy": 0.142944598197937,
      "num_tokens": 7683013.0,
      "step": 4115
    },
    {
      "entropy": 6.285777425765991,
      "epoch": 0.3638290356764394,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004991429834733721,
      "loss": 6.2166,
      "mean_token_accuracy": 0.14354896545410156,
      "num_tokens": 7692760.0,
      "step": 4120
    },
    {
      "entropy": 6.293794918060303,
      "epoch": 0.36427057576827976,
      "grad_norm": 2.890625,
      "learning_rate": 0.000499140235295103,
      "loss": 6.1902,
      "mean_token_accuracy": 0.14732723981142043,
      "num_tokens": 7702005.0,
      "step": 4125
    },
    {
      "entropy": 6.299161052703857,
      "epoch": 0.3647121158601201,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004991374827260542,
      "loss": 6.2069,
      "mean_token_accuracy": 0.14357001036405564,
      "num_tokens": 7711604.0,
      "step": 4130
    },
    {
      "entropy": 6.247972822189331,
      "epoch": 0.3651536559519604,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004991347257662795,
      "loss": 6.1049,
      "mean_token_accuracy": 0.14599167481064795,
      "num_tokens": 7720434.0,
      "step": 4135
    },
    {
      "entropy": 6.294910478591919,
      "epoch": 0.36559519604380075,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000499131964415833,
      "loss": 6.166,
      "mean_token_accuracy": 0.14839906394481658,
      "num_tokens": 7730733.0,
      "step": 4140
    },
    {
      "entropy": 6.198695325851441,
      "epoch": 0.36603673613564114,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004991291986747689,
      "loss": 6.235,
      "mean_token_accuracy": 0.14558938667178153,
      "num_tokens": 7740394.0,
      "step": 4145
    },
    {
      "entropy": 6.325007724761963,
      "epoch": 0.36647827622748147,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004991264285431412,
      "loss": 6.1828,
      "mean_token_accuracy": 0.150047105550766,
      "num_tokens": 7748919.0,
      "step": 4150
    },
    {
      "entropy": 6.360399150848389,
      "epoch": 0.3669198163193218,
      "grad_norm": 1.875,
      "learning_rate": 0.0004991236540210041,
      "loss": 6.2051,
      "mean_token_accuracy": 0.1464771181344986,
      "num_tokens": 7758591.0,
      "step": 4155
    },
    {
      "entropy": 6.263759708404541,
      "epoch": 0.36736135641116213,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004991208751084122,
      "loss": 6.2409,
      "mean_token_accuracy": 0.13913936614990235,
      "num_tokens": 7768269.0,
      "step": 4160
    },
    {
      "entropy": 6.311656665802002,
      "epoch": 0.36780289650300246,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004991180918054199,
      "loss": 6.257,
      "mean_token_accuracy": 0.14464529231190681,
      "num_tokens": 7776980.0,
      "step": 4165
    },
    {
      "entropy": 6.337720155715942,
      "epoch": 0.3682444365948428,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004991153041120815,
      "loss": 6.1695,
      "mean_token_accuracy": 0.1493624746799469,
      "num_tokens": 7786664.0,
      "step": 4170
    },
    {
      "entropy": 6.2292993545532225,
      "epoch": 0.3686859766866831,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004991125120284519,
      "loss": 6.0267,
      "mean_token_accuracy": 0.15732113569974898,
      "num_tokens": 7794981.0,
      "step": 4175
    },
    {
      "entropy": 6.180199718475341,
      "epoch": 0.3691275167785235,
      "grad_norm": 6.78125,
      "learning_rate": 0.0004991097155545856,
      "loss": 6.1299,
      "mean_token_accuracy": 0.15222777128219606,
      "num_tokens": 7804850.0,
      "step": 4180
    },
    {
      "entropy": 6.182018852233886,
      "epoch": 0.36956905687036384,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004991069146905374,
      "loss": 6.1419,
      "mean_token_accuracy": 0.15296721160411836,
      "num_tokens": 7814117.0,
      "step": 4185
    },
    {
      "entropy": 6.265089225769043,
      "epoch": 0.3700105969622042,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004991041094363621,
      "loss": 6.1782,
      "mean_token_accuracy": 0.15054446160793306,
      "num_tokens": 7823384.0,
      "step": 4190
    },
    {
      "entropy": 6.2933553695678714,
      "epoch": 0.3704521370540445,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004991012997921149,
      "loss": 6.1724,
      "mean_token_accuracy": 0.15083738043904305,
      "num_tokens": 7834072.0,
      "step": 4195
    },
    {
      "entropy": 6.280925703048706,
      "epoch": 0.37089367714588484,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004990984857578506,
      "loss": 6.1744,
      "mean_token_accuracy": 0.1427485764026642,
      "num_tokens": 7843334.0,
      "step": 4200
    },
    {
      "entropy": 6.127236223220825,
      "epoch": 0.37133521723772517,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004990956673336245,
      "loss": 6.0968,
      "mean_token_accuracy": 0.1535086452960968,
      "num_tokens": 7853187.0,
      "step": 4205
    },
    {
      "entropy": 6.232405376434326,
      "epoch": 0.3717767573295655,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004990928445194917,
      "loss": 6.1824,
      "mean_token_accuracy": 0.14675197005271912,
      "num_tokens": 7862262.0,
      "step": 4210
    },
    {
      "entropy": 6.2625322341918945,
      "epoch": 0.3722182974214059,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004990900173155074,
      "loss": 6.1323,
      "mean_token_accuracy": 0.15006745010614395,
      "num_tokens": 7870725.0,
      "step": 4215
    },
    {
      "entropy": 6.220727872848511,
      "epoch": 0.3726598375132462,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004990871857217273,
      "loss": 6.0515,
      "mean_token_accuracy": 0.15157599598169327,
      "num_tokens": 7879492.0,
      "step": 4220
    },
    {
      "entropy": 6.237200546264648,
      "epoch": 0.37310137760508655,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004990843497382066,
      "loss": 6.1537,
      "mean_token_accuracy": 0.14744771271944046,
      "num_tokens": 7888710.0,
      "step": 4225
    },
    {
      "entropy": 6.243387746810913,
      "epoch": 0.3735429176969269,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004990815093650009,
      "loss": 6.1786,
      "mean_token_accuracy": 0.1538126990199089,
      "num_tokens": 7898313.0,
      "step": 4230
    },
    {
      "entropy": 6.194939756393433,
      "epoch": 0.3739844577887672,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004990786646021659,
      "loss": 6.1903,
      "mean_token_accuracy": 0.14287389069795609,
      "num_tokens": 7906726.0,
      "step": 4235
    },
    {
      "entropy": 6.255891227722168,
      "epoch": 0.37442599788060754,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004990758154497573,
      "loss": 6.1098,
      "mean_token_accuracy": 0.15329550951719284,
      "num_tokens": 7915554.0,
      "step": 4240
    },
    {
      "entropy": 6.313123893737793,
      "epoch": 0.3748675379724479,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004990729619078309,
      "loss": 6.2365,
      "mean_token_accuracy": 0.14729351848363875,
      "num_tokens": 7924732.0,
      "step": 4245
    },
    {
      "entropy": 6.307825803756714,
      "epoch": 0.37530907806428826,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004990701039764427,
      "loss": 6.1502,
      "mean_token_accuracy": 0.1527288243174553,
      "num_tokens": 7933808.0,
      "step": 4250
    },
    {
      "entropy": 6.24105749130249,
      "epoch": 0.3757506181561286,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004990672416556485,
      "loss": 6.1186,
      "mean_token_accuracy": 0.15374407172203064,
      "num_tokens": 7942558.0,
      "step": 4255
    },
    {
      "entropy": 6.174380731582642,
      "epoch": 0.3761921582479689,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004990643749455045,
      "loss": 6.2051,
      "mean_token_accuracy": 0.14767509549856186,
      "num_tokens": 7951951.0,
      "step": 4260
    },
    {
      "entropy": 6.311297607421875,
      "epoch": 0.37663369833980925,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004990615038460667,
      "loss": 6.1174,
      "mean_token_accuracy": 0.15629696547985078,
      "num_tokens": 7960543.0,
      "step": 4265
    },
    {
      "entropy": 6.212013912200928,
      "epoch": 0.3770752384316496,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004990586283573916,
      "loss": 6.1004,
      "mean_token_accuracy": 0.15100028067827226,
      "num_tokens": 7970273.0,
      "step": 4270
    },
    {
      "entropy": 6.24426040649414,
      "epoch": 0.3775167785234899,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004990557484795355,
      "loss": 6.1821,
      "mean_token_accuracy": 0.15400006845593453,
      "num_tokens": 7978888.0,
      "step": 4275
    },
    {
      "entropy": 6.259699487686158,
      "epoch": 0.37795831861533025,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004990528642125545,
      "loss": 6.1865,
      "mean_token_accuracy": 0.1407089427113533,
      "num_tokens": 7988843.0,
      "step": 4280
    },
    {
      "entropy": 6.279107713699341,
      "epoch": 0.37839985870717063,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004990499755565055,
      "loss": 6.1638,
      "mean_token_accuracy": 0.151794371008873,
      "num_tokens": 7998403.0,
      "step": 4285
    },
    {
      "entropy": 6.285480642318726,
      "epoch": 0.37884139879901096,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004990470825114448,
      "loss": 6.1885,
      "mean_token_accuracy": 0.15127992779016494,
      "num_tokens": 8007493.0,
      "step": 4290
    },
    {
      "entropy": 6.1836748123168945,
      "epoch": 0.3792829388908513,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004990441850774292,
      "loss": 6.1603,
      "mean_token_accuracy": 0.15351675748825072,
      "num_tokens": 8016399.0,
      "step": 4295
    },
    {
      "entropy": 6.262436246871948,
      "epoch": 0.3797244789826916,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004990412832545155,
      "loss": 6.1635,
      "mean_token_accuracy": 0.14691063463687898,
      "num_tokens": 8025891.0,
      "step": 4300
    },
    {
      "entropy": 6.238006687164306,
      "epoch": 0.38016601907453196,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004990383770427603,
      "loss": 6.0859,
      "mean_token_accuracy": 0.1509350799024105,
      "num_tokens": 8034733.0,
      "step": 4305
    },
    {
      "entropy": 6.230528879165649,
      "epoch": 0.3806075591663723,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004990354664422209,
      "loss": 6.1483,
      "mean_token_accuracy": 0.1521308660507202,
      "num_tokens": 8044443.0,
      "step": 4310
    },
    {
      "entropy": 6.138578844070435,
      "epoch": 0.3810490992582126,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004990325514529541,
      "loss": 6.1303,
      "mean_token_accuracy": 0.16342166811227798,
      "num_tokens": 8053920.0,
      "step": 4315
    },
    {
      "entropy": 6.3028443336486815,
      "epoch": 0.381490639350053,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004990296320750169,
      "loss": 6.1227,
      "mean_token_accuracy": 0.15118969082832337,
      "num_tokens": 8063455.0,
      "step": 4320
    },
    {
      "entropy": 6.22663893699646,
      "epoch": 0.38193217944189334,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004990267083084667,
      "loss": 6.1995,
      "mean_token_accuracy": 0.1500270701944828,
      "num_tokens": 8073353.0,
      "step": 4325
    },
    {
      "entropy": 6.255289125442505,
      "epoch": 0.38237371953373367,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004990237801533607,
      "loss": 6.2035,
      "mean_token_accuracy": 0.15115233659744262,
      "num_tokens": 8082164.0,
      "step": 4330
    },
    {
      "entropy": 6.303492021560669,
      "epoch": 0.382815259625574,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004990208476097562,
      "loss": 6.2161,
      "mean_token_accuracy": 0.1453220874071121,
      "num_tokens": 8090790.0,
      "step": 4335
    },
    {
      "entropy": 6.417842245101928,
      "epoch": 0.38325679971741433,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004990179106777109,
      "loss": 6.2292,
      "mean_token_accuracy": 0.14599840193986893,
      "num_tokens": 8100412.0,
      "step": 4340
    },
    {
      "entropy": 6.189248132705688,
      "epoch": 0.38369833980925466,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004990149693572819,
      "loss": 6.2178,
      "mean_token_accuracy": 0.1462481401860714,
      "num_tokens": 8110002.0,
      "step": 4345
    },
    {
      "entropy": 6.205133438110352,
      "epoch": 0.384139879901095,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004990120236485271,
      "loss": 6.1141,
      "mean_token_accuracy": 0.1545722119510174,
      "num_tokens": 8118898.0,
      "step": 4350
    },
    {
      "entropy": 6.2361159324646,
      "epoch": 0.3845814199929354,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004990090735515043,
      "loss": 6.1703,
      "mean_token_accuracy": 0.1435256503522396,
      "num_tokens": 8128207.0,
      "step": 4355
    },
    {
      "entropy": 6.186733150482178,
      "epoch": 0.3850229600847757,
      "grad_norm": 1.640625,
      "learning_rate": 0.000499006119066271,
      "loss": 6.1558,
      "mean_token_accuracy": 0.14698365032672883,
      "num_tokens": 8137317.0,
      "step": 4360
    },
    {
      "entropy": 6.266343832015991,
      "epoch": 0.38546450017661604,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004990031601928854,
      "loss": 6.1825,
      "mean_token_accuracy": 0.1447221465408802,
      "num_tokens": 8146519.0,
      "step": 4365
    },
    {
      "entropy": 6.168535375595093,
      "epoch": 0.3859060402684564,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004990001969314051,
      "loss": 6.2218,
      "mean_token_accuracy": 0.145457149296999,
      "num_tokens": 8155916.0,
      "step": 4370
    },
    {
      "entropy": 6.3184326171875,
      "epoch": 0.3863475803602967,
      "grad_norm": 1.5,
      "learning_rate": 0.0004989972292818884,
      "loss": 6.231,
      "mean_token_accuracy": 0.14837146773934365,
      "num_tokens": 8165251.0,
      "step": 4375
    },
    {
      "entropy": 6.2823505878448485,
      "epoch": 0.38678912045213704,
      "grad_norm": 5.40625,
      "learning_rate": 0.0004989942572443934,
      "loss": 6.2042,
      "mean_token_accuracy": 0.14634532183408738,
      "num_tokens": 8174761.0,
      "step": 4380
    },
    {
      "entropy": 6.271600246429443,
      "epoch": 0.38723066054397737,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004989912808189784,
      "loss": 6.1563,
      "mean_token_accuracy": 0.15100886076688766,
      "num_tokens": 8182980.0,
      "step": 4385
    },
    {
      "entropy": 6.3505312442779545,
      "epoch": 0.38767220063581775,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004989883000057013,
      "loss": 6.2051,
      "mean_token_accuracy": 0.14624489918351175,
      "num_tokens": 8193306.0,
      "step": 4390
    },
    {
      "entropy": 6.2322680950164795,
      "epoch": 0.3881137407276581,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000498985314804621,
      "loss": 6.0946,
      "mean_token_accuracy": 0.15022510960698127,
      "num_tokens": 8203446.0,
      "step": 4395
    },
    {
      "entropy": 6.238686370849609,
      "epoch": 0.3885552808194984,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004989823252157958,
      "loss": 6.1412,
      "mean_token_accuracy": 0.1490713208913803,
      "num_tokens": 8212414.0,
      "step": 4400
    },
    {
      "entropy": 6.13543381690979,
      "epoch": 0.38899682091133875,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004989793312392841,
      "loss": 6.0629,
      "mean_token_accuracy": 0.15065995454788209,
      "num_tokens": 8221509.0,
      "step": 4405
    },
    {
      "entropy": 6.232558631896973,
      "epoch": 0.3894383610031791,
      "grad_norm": 1.625,
      "learning_rate": 0.0004989763328751448,
      "loss": 6.1626,
      "mean_token_accuracy": 0.14932333379983903,
      "num_tokens": 8230017.0,
      "step": 4410
    },
    {
      "entropy": 6.253946590423584,
      "epoch": 0.3898799010950194,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004989733301234365,
      "loss": 6.1223,
      "mean_token_accuracy": 0.15184631198644638,
      "num_tokens": 8238853.0,
      "step": 4415
    },
    {
      "entropy": 6.283807992935181,
      "epoch": 0.39032144118685974,
      "grad_norm": 1.40625,
      "learning_rate": 0.000498970322984218,
      "loss": 6.2002,
      "mean_token_accuracy": 0.14423604309558868,
      "num_tokens": 8248112.0,
      "step": 4420
    },
    {
      "entropy": 6.273688220977784,
      "epoch": 0.39076298127870013,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004989673114575483,
      "loss": 6.1516,
      "mean_token_accuracy": 0.14697378426790236,
      "num_tokens": 8257166.0,
      "step": 4425
    },
    {
      "entropy": 6.248018789291382,
      "epoch": 0.39120452137054046,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004989642955434863,
      "loss": 6.1751,
      "mean_token_accuracy": 0.14618921428918838,
      "num_tokens": 8266229.0,
      "step": 4430
    },
    {
      "entropy": 6.263542175292969,
      "epoch": 0.3916460614623808,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004989612752420912,
      "loss": 6.1318,
      "mean_token_accuracy": 0.147259022295475,
      "num_tokens": 8275102.0,
      "step": 4435
    },
    {
      "entropy": 6.21520299911499,
      "epoch": 0.3920876015542211,
      "grad_norm": 1.5625,
      "learning_rate": 0.000498958250553422,
      "loss": 6.0509,
      "mean_token_accuracy": 0.15448229461908342,
      "num_tokens": 8283847.0,
      "step": 4440
    },
    {
      "entropy": 6.29284291267395,
      "epoch": 0.39252914164606145,
      "grad_norm": 1.5,
      "learning_rate": 0.0004989552214775381,
      "loss": 6.1753,
      "mean_token_accuracy": 0.1464612901210785,
      "num_tokens": 8292622.0,
      "step": 4445
    },
    {
      "entropy": 6.291398048400879,
      "epoch": 0.3929706817379018,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004989521880144988,
      "loss": 6.1034,
      "mean_token_accuracy": 0.16104813367128373,
      "num_tokens": 8301026.0,
      "step": 4450
    },
    {
      "entropy": 6.259631872177124,
      "epoch": 0.3934122218297421,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004989491501643635,
      "loss": 6.3562,
      "mean_token_accuracy": 0.13695246875286102,
      "num_tokens": 8310977.0,
      "step": 4455
    },
    {
      "entropy": 6.219717025756836,
      "epoch": 0.3938537619215825,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004989461079271916,
      "loss": 6.1221,
      "mean_token_accuracy": 0.1505828931927681,
      "num_tokens": 8319391.0,
      "step": 4460
    },
    {
      "entropy": 6.210220909118652,
      "epoch": 0.39429530201342283,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004989430613030429,
      "loss": 6.0801,
      "mean_token_accuracy": 0.1500351071357727,
      "num_tokens": 8328639.0,
      "step": 4465
    },
    {
      "entropy": 6.262903022766113,
      "epoch": 0.39473684210526316,
      "grad_norm": 2.421875,
      "learning_rate": 0.000498940010291977,
      "loss": 6.1916,
      "mean_token_accuracy": 0.1418144792318344,
      "num_tokens": 8338190.0,
      "step": 4470
    },
    {
      "entropy": 6.208887529373169,
      "epoch": 0.3951783821971035,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004989369548940536,
      "loss": 6.1396,
      "mean_token_accuracy": 0.14933243542909622,
      "num_tokens": 8346874.0,
      "step": 4475
    },
    {
      "entropy": 6.2679730415344235,
      "epoch": 0.3956199222889438,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004989338951093327,
      "loss": 6.1072,
      "mean_token_accuracy": 0.1559783160686493,
      "num_tokens": 8356446.0,
      "step": 4480
    },
    {
      "entropy": 6.254531574249268,
      "epoch": 0.39606146238078416,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004989308309378741,
      "loss": 6.1142,
      "mean_token_accuracy": 0.15238502025604247,
      "num_tokens": 8365694.0,
      "step": 4485
    },
    {
      "entropy": 6.228513717651367,
      "epoch": 0.3965030024726245,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004989277623797379,
      "loss": 6.1785,
      "mean_token_accuracy": 0.1451736718416214,
      "num_tokens": 8374282.0,
      "step": 4490
    },
    {
      "entropy": 6.223094749450683,
      "epoch": 0.3969445425644649,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004989246894349841,
      "loss": 6.1571,
      "mean_token_accuracy": 0.14543229341506958,
      "num_tokens": 8383315.0,
      "step": 4495
    },
    {
      "entropy": 6.248504543304444,
      "epoch": 0.3973860826563052,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004989216121036732,
      "loss": 6.1193,
      "mean_token_accuracy": 0.14976042211055757,
      "num_tokens": 8392263.0,
      "step": 4500
    },
    {
      "entropy": 6.268833160400391,
      "epoch": 0.39782762274814554,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004989185303858651,
      "loss": 6.1598,
      "mean_token_accuracy": 0.14398288130760192,
      "num_tokens": 8400734.0,
      "step": 4505
    },
    {
      "entropy": 6.275659704208374,
      "epoch": 0.39826916283998587,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004989154442816203,
      "loss": 6.1967,
      "mean_token_accuracy": 0.14542601853609086,
      "num_tokens": 8410635.0,
      "step": 4510
    },
    {
      "entropy": 6.305443334579468,
      "epoch": 0.3987107029318262,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004989123537909994,
      "loss": 6.165,
      "mean_token_accuracy": 0.14762203097343446,
      "num_tokens": 8420111.0,
      "step": 4515
    },
    {
      "entropy": 6.190962457656861,
      "epoch": 0.39915224302366653,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004989092589140629,
      "loss": 6.1101,
      "mean_token_accuracy": 0.14062744081020356,
      "num_tokens": 8429459.0,
      "step": 4520
    },
    {
      "entropy": 6.277060890197754,
      "epoch": 0.39959378311550686,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004989061596508712,
      "loss": 6.1288,
      "mean_token_accuracy": 0.14999366477131842,
      "num_tokens": 8438083.0,
      "step": 4525
    },
    {
      "entropy": 6.259339284896851,
      "epoch": 0.40003532320734725,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004989030560014853,
      "loss": 6.1662,
      "mean_token_accuracy": 0.15049690380692482,
      "num_tokens": 8447713.0,
      "step": 4530
    },
    {
      "entropy": 6.255221700668335,
      "epoch": 0.4004768632991876,
      "grad_norm": 5.75,
      "learning_rate": 0.0004988999479659657,
      "loss": 6.1189,
      "mean_token_accuracy": 0.14965896531939507,
      "num_tokens": 8457394.0,
      "step": 4535
    },
    {
      "entropy": 6.226453971862793,
      "epoch": 0.4009184033910279,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004988968355443737,
      "loss": 6.0989,
      "mean_token_accuracy": 0.15315678864717483,
      "num_tokens": 8467333.0,
      "step": 4540
    },
    {
      "entropy": 6.249733924865723,
      "epoch": 0.40135994348286824,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004988937187367699,
      "loss": 6.1927,
      "mean_token_accuracy": 0.14078941121697425,
      "num_tokens": 8477530.0,
      "step": 4545
    },
    {
      "entropy": 6.2826484680175785,
      "epoch": 0.4018014835747086,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004988905975432154,
      "loss": 6.2625,
      "mean_token_accuracy": 0.14484824985265732,
      "num_tokens": 8486861.0,
      "step": 4550
    },
    {
      "entropy": 6.243760204315185,
      "epoch": 0.4022430236665489,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004988874719637715,
      "loss": 6.1034,
      "mean_token_accuracy": 0.15113344937562942,
      "num_tokens": 8496541.0,
      "step": 4555
    },
    {
      "entropy": 6.19863486289978,
      "epoch": 0.40268456375838924,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004988843419984994,
      "loss": 6.1531,
      "mean_token_accuracy": 0.1470022514462471,
      "num_tokens": 8505667.0,
      "step": 4560
    },
    {
      "entropy": 6.358606481552124,
      "epoch": 0.4031261038502296,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004988812076474604,
      "loss": 6.1486,
      "mean_token_accuracy": 0.14644313231110573,
      "num_tokens": 8515133.0,
      "step": 4565
    },
    {
      "entropy": 6.305288457870484,
      "epoch": 0.40356764394206995,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004988780689107158,
      "loss": 6.1064,
      "mean_token_accuracy": 0.15373924225568772,
      "num_tokens": 8524797.0,
      "step": 4570
    },
    {
      "entropy": 6.173681545257568,
      "epoch": 0.4040091840339103,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004988749257883271,
      "loss": 6.1367,
      "mean_token_accuracy": 0.14695328921079637,
      "num_tokens": 8534667.0,
      "step": 4575
    },
    {
      "entropy": 6.227071952819824,
      "epoch": 0.4044507241257506,
      "grad_norm": 1.765625,
      "learning_rate": 0.000498871778280356,
      "loss": 6.1302,
      "mean_token_accuracy": 0.15038311779499053,
      "num_tokens": 8543874.0,
      "step": 4580
    },
    {
      "entropy": 6.192816591262817,
      "epoch": 0.40489226421759095,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004988686263868641,
      "loss": 6.1467,
      "mean_token_accuracy": 0.14648784697055817,
      "num_tokens": 8553620.0,
      "step": 4585
    },
    {
      "entropy": 6.25673189163208,
      "epoch": 0.4053338043094313,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004988654701079131,
      "loss": 6.2143,
      "mean_token_accuracy": 0.14080783054232598,
      "num_tokens": 8563857.0,
      "step": 4590
    },
    {
      "entropy": 6.334765005111694,
      "epoch": 0.4057753444012716,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004988623094435649,
      "loss": 6.2139,
      "mean_token_accuracy": 0.14208807647228242,
      "num_tokens": 8572677.0,
      "step": 4595
    },
    {
      "entropy": 6.229012393951416,
      "epoch": 0.406216884493112,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004988591443938813,
      "loss": 6.1269,
      "mean_token_accuracy": 0.14795654267072678,
      "num_tokens": 8581907.0,
      "step": 4600
    },
    {
      "entropy": 6.262136697769165,
      "epoch": 0.4066584245849523,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004988559749589244,
      "loss": 6.1629,
      "mean_token_accuracy": 0.14598316252231597,
      "num_tokens": 8591701.0,
      "step": 4605
    },
    {
      "entropy": 6.257154321670532,
      "epoch": 0.40709996467679266,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004988528011387563,
      "loss": 6.1899,
      "mean_token_accuracy": 0.15008021965622903,
      "num_tokens": 8601054.0,
      "step": 4610
    },
    {
      "entropy": 6.234865713119507,
      "epoch": 0.407541504768633,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004988496229334392,
      "loss": 6.1083,
      "mean_token_accuracy": 0.15296214148402215,
      "num_tokens": 8610346.0,
      "step": 4615
    },
    {
      "entropy": 6.223312854766846,
      "epoch": 0.4079830448604733,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004988464403430352,
      "loss": 6.1006,
      "mean_token_accuracy": 0.1481364607810974,
      "num_tokens": 8620823.0,
      "step": 4620
    },
    {
      "entropy": 6.2206339836120605,
      "epoch": 0.40842458495231365,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004988432533676067,
      "loss": 6.2553,
      "mean_token_accuracy": 0.14241225197911261,
      "num_tokens": 8630184.0,
      "step": 4625
    },
    {
      "entropy": 6.259246778488159,
      "epoch": 0.408866125044154,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004988400620072163,
      "loss": 6.2588,
      "mean_token_accuracy": 0.1362864449620247,
      "num_tokens": 8640064.0,
      "step": 4630
    },
    {
      "entropy": 6.3511247634887695,
      "epoch": 0.40930766513599437,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004988368662619263,
      "loss": 6.1484,
      "mean_token_accuracy": 0.14784810915589333,
      "num_tokens": 8650503.0,
      "step": 4635
    },
    {
      "entropy": 6.333515739440918,
      "epoch": 0.4097492052278347,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004988336661317994,
      "loss": 6.1629,
      "mean_token_accuracy": 0.14861233532428741,
      "num_tokens": 8659125.0,
      "step": 4640
    },
    {
      "entropy": 6.213559198379516,
      "epoch": 0.41019074531967503,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004988304616168984,
      "loss": 6.1769,
      "mean_token_accuracy": 0.14591192007064818,
      "num_tokens": 8668193.0,
      "step": 4645
    },
    {
      "entropy": 6.266979026794433,
      "epoch": 0.41063228541151536,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004988272527172858,
      "loss": 6.1308,
      "mean_token_accuracy": 0.14889822229743005,
      "num_tokens": 8677515.0,
      "step": 4650
    },
    {
      "entropy": 6.287114191055298,
      "epoch": 0.4110738255033557,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004988240394330246,
      "loss": 6.1746,
      "mean_token_accuracy": 0.14187708795070647,
      "num_tokens": 8687549.0,
      "step": 4655
    },
    {
      "entropy": 6.256554746627808,
      "epoch": 0.411515365595196,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004988208217641778,
      "loss": 6.1761,
      "mean_token_accuracy": 0.14865077808499336,
      "num_tokens": 8697371.0,
      "step": 4660
    },
    {
      "entropy": 6.30476655960083,
      "epoch": 0.41195690568703636,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004988175997108086,
      "loss": 6.1793,
      "mean_token_accuracy": 0.1483299732208252,
      "num_tokens": 8707193.0,
      "step": 4665
    },
    {
      "entropy": 6.268969869613647,
      "epoch": 0.41239844577887674,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004988143732729797,
      "loss": 6.1778,
      "mean_token_accuracy": 0.1447226196527481,
      "num_tokens": 8716052.0,
      "step": 4670
    },
    {
      "entropy": 6.170227193832398,
      "epoch": 0.4128399858707171,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004988111424507546,
      "loss": 6.1174,
      "mean_token_accuracy": 0.15311618894338608,
      "num_tokens": 8726140.0,
      "step": 4675
    },
    {
      "entropy": 6.254574155807495,
      "epoch": 0.4132815259625574,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004988079072441964,
      "loss": 6.1107,
      "mean_token_accuracy": 0.15668862909078599,
      "num_tokens": 8735299.0,
      "step": 4680
    },
    {
      "entropy": 6.258963775634766,
      "epoch": 0.41372306605439774,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004988046676533687,
      "loss": 6.179,
      "mean_token_accuracy": 0.1478872299194336,
      "num_tokens": 8744686.0,
      "step": 4685
    },
    {
      "entropy": 6.226867914199829,
      "epoch": 0.41416460614623807,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004988014236783347,
      "loss": 6.171,
      "mean_token_accuracy": 0.15285128355026245,
      "num_tokens": 8754942.0,
      "step": 4690
    },
    {
      "entropy": 6.317941951751709,
      "epoch": 0.4146061462380784,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004987981753191582,
      "loss": 6.1695,
      "mean_token_accuracy": 0.14410044327378274,
      "num_tokens": 8764054.0,
      "step": 4695
    },
    {
      "entropy": 6.133100175857544,
      "epoch": 0.41504768632991873,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004987949225759027,
      "loss": 6.0421,
      "mean_token_accuracy": 0.15263513773679732,
      "num_tokens": 8773050.0,
      "step": 4700
    },
    {
      "entropy": 6.275085163116455,
      "epoch": 0.4154892264217591,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004987916654486321,
      "loss": 6.1478,
      "mean_token_accuracy": 0.1481424480676651,
      "num_tokens": 8782476.0,
      "step": 4705
    },
    {
      "entropy": 6.360612583160401,
      "epoch": 0.41593076651359945,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004987884039374099,
      "loss": 6.215,
      "mean_token_accuracy": 0.1436464659869671,
      "num_tokens": 8791147.0,
      "step": 4710
    },
    {
      "entropy": 6.36751766204834,
      "epoch": 0.4163723066054398,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004987851380423001,
      "loss": 6.2116,
      "mean_token_accuracy": 0.14959510266780854,
      "num_tokens": 8801151.0,
      "step": 4715
    },
    {
      "entropy": 6.216678428649902,
      "epoch": 0.4168138466972801,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004987818677633668,
      "loss": 6.157,
      "mean_token_accuracy": 0.14351932033896447,
      "num_tokens": 8809587.0,
      "step": 4720
    },
    {
      "entropy": 6.13917121887207,
      "epoch": 0.41725538678912044,
      "grad_norm": 2.171875,
      "learning_rate": 0.000498778593100674,
      "loss": 6.054,
      "mean_token_accuracy": 0.1474153071641922,
      "num_tokens": 8818168.0,
      "step": 4725
    },
    {
      "entropy": 6.221515893936157,
      "epoch": 0.4176969268809608,
      "grad_norm": 2.375,
      "learning_rate": 0.0004987753140542857,
      "loss": 6.1075,
      "mean_token_accuracy": 0.15322076976299287,
      "num_tokens": 8827477.0,
      "step": 4730
    },
    {
      "entropy": 6.230283880233765,
      "epoch": 0.4181384669728011,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004987720306242664,
      "loss": 6.0908,
      "mean_token_accuracy": 0.15419567972421647,
      "num_tokens": 8837067.0,
      "step": 4735
    },
    {
      "entropy": 6.259425640106201,
      "epoch": 0.4185800070646415,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004987687428106803,
      "loss": 6.1961,
      "mean_token_accuracy": 0.15190616473555565,
      "num_tokens": 8845790.0,
      "step": 4740
    },
    {
      "entropy": 6.233168458938598,
      "epoch": 0.4190215471564818,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004987654506135917,
      "loss": 6.1166,
      "mean_token_accuracy": 0.15246631503105162,
      "num_tokens": 8855242.0,
      "step": 4745
    },
    {
      "entropy": 6.357547903060913,
      "epoch": 0.41946308724832215,
      "grad_norm": 1.625,
      "learning_rate": 0.0004987621540330652,
      "loss": 6.1807,
      "mean_token_accuracy": 0.15104667320847512,
      "num_tokens": 8864459.0,
      "step": 4750
    },
    {
      "entropy": 6.257883977890015,
      "epoch": 0.4199046273401625,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004987588530691653,
      "loss": 6.2325,
      "mean_token_accuracy": 0.14518377780914307,
      "num_tokens": 8875028.0,
      "step": 4755
    },
    {
      "entropy": 6.165983200073242,
      "epoch": 0.4203461674320028,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004987555477219569,
      "loss": 6.0833,
      "mean_token_accuracy": 0.15509303510189057,
      "num_tokens": 8883857.0,
      "step": 4760
    },
    {
      "entropy": 6.305104160308838,
      "epoch": 0.42078770752384315,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004987522379915045,
      "loss": 6.0764,
      "mean_token_accuracy": 0.15581088662147521,
      "num_tokens": 8893499.0,
      "step": 4765
    },
    {
      "entropy": 6.269616603851318,
      "epoch": 0.4212292476156835,
      "grad_norm": 1.609375,
      "learning_rate": 0.000498748923877873,
      "loss": 6.1035,
      "mean_token_accuracy": 0.15332883447408677,
      "num_tokens": 8903368.0,
      "step": 4770
    },
    {
      "entropy": 6.1651966094970705,
      "epoch": 0.42167078770752386,
      "grad_norm": 2.125,
      "learning_rate": 0.0004987456053811273,
      "loss": 6.1849,
      "mean_token_accuracy": 0.1498742736876011,
      "num_tokens": 8912701.0,
      "step": 4775
    },
    {
      "entropy": 6.17396912574768,
      "epoch": 0.4221123277993642,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004987422825013325,
      "loss": 6.1473,
      "mean_token_accuracy": 0.14589054882526398,
      "num_tokens": 8921962.0,
      "step": 4780
    },
    {
      "entropy": 6.349455690383911,
      "epoch": 0.4225538678912045,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004987389552385536,
      "loss": 6.1536,
      "mean_token_accuracy": 0.14763372913002967,
      "num_tokens": 8931923.0,
      "step": 4785
    },
    {
      "entropy": 6.243163013458252,
      "epoch": 0.42299540798304486,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004987356235928558,
      "loss": 6.1744,
      "mean_token_accuracy": 0.14302204996347428,
      "num_tokens": 8940403.0,
      "step": 4790
    },
    {
      "entropy": 6.256076335906982,
      "epoch": 0.4234369480748852,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004987322875643044,
      "loss": 6.0935,
      "mean_token_accuracy": 0.1572992831468582,
      "num_tokens": 8949377.0,
      "step": 4795
    },
    {
      "entropy": 6.306877994537354,
      "epoch": 0.4238784881667255,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004987289471529647,
      "loss": 6.2359,
      "mean_token_accuracy": 0.145481688529253,
      "num_tokens": 8958719.0,
      "step": 4800
    },
    {
      "entropy": 6.2402099609375,
      "epoch": 0.42432002825856585,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004987256023589022,
      "loss": 6.1936,
      "mean_token_accuracy": 0.14106479585170745,
      "num_tokens": 8968226.0,
      "step": 4805
    },
    {
      "entropy": 6.20019416809082,
      "epoch": 0.42476156835040624,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004987222531821824,
      "loss": 6.1456,
      "mean_token_accuracy": 0.14612633734941483,
      "num_tokens": 8976670.0,
      "step": 4810
    },
    {
      "entropy": 6.179481172561646,
      "epoch": 0.42520310844224657,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004987188996228709,
      "loss": 6.0789,
      "mean_token_accuracy": 0.15591740757226943,
      "num_tokens": 8986185.0,
      "step": 4815
    },
    {
      "entropy": 6.212154388427734,
      "epoch": 0.4256446485340869,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004987155416810334,
      "loss": 6.0864,
      "mean_token_accuracy": 0.1552813544869423,
      "num_tokens": 8995124.0,
      "step": 4820
    },
    {
      "entropy": 6.21709885597229,
      "epoch": 0.42608618862592723,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004987121793567356,
      "loss": 6.1716,
      "mean_token_accuracy": 0.1470434084534645,
      "num_tokens": 9004380.0,
      "step": 4825
    },
    {
      "entropy": 6.242340612411499,
      "epoch": 0.42652772871776756,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004987088126500436,
      "loss": 6.0855,
      "mean_token_accuracy": 0.15288129597902297,
      "num_tokens": 9013791.0,
      "step": 4830
    },
    {
      "entropy": 6.20204725265503,
      "epoch": 0.4269692688096079,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000498705441561023,
      "loss": 6.1038,
      "mean_token_accuracy": 0.16311327815055848,
      "num_tokens": 9023076.0,
      "step": 4835
    },
    {
      "entropy": 6.332134342193603,
      "epoch": 0.4274108089014482,
      "grad_norm": 1.625,
      "learning_rate": 0.0004987020660897401,
      "loss": 6.2081,
      "mean_token_accuracy": 0.14697263911366462,
      "num_tokens": 9032720.0,
      "step": 4840
    },
    {
      "entropy": 6.2755121231079105,
      "epoch": 0.4278523489932886,
      "grad_norm": 2.421875,
      "learning_rate": 0.000498698686236261,
      "loss": 6.1613,
      "mean_token_accuracy": 0.146524578332901,
      "num_tokens": 9042652.0,
      "step": 4845
    },
    {
      "entropy": 6.2472004890441895,
      "epoch": 0.42829388908512894,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004986953020006519,
      "loss": 6.2262,
      "mean_token_accuracy": 0.14156229123473169,
      "num_tokens": 9052172.0,
      "step": 4850
    },
    {
      "entropy": 6.226924276351928,
      "epoch": 0.4287354291769693,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004986919133829788,
      "loss": 6.1085,
      "mean_token_accuracy": 0.15182769894599915,
      "num_tokens": 9061798.0,
      "step": 4855
    },
    {
      "entropy": 6.158733892440796,
      "epoch": 0.4291769692688096,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004986885203833086,
      "loss": 6.04,
      "mean_token_accuracy": 0.15590932518243789,
      "num_tokens": 9070429.0,
      "step": 4860
    },
    {
      "entropy": 6.165717315673828,
      "epoch": 0.42961850936064994,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004986851230017075,
      "loss": 6.1517,
      "mean_token_accuracy": 0.15533166229724885,
      "num_tokens": 9080200.0,
      "step": 4865
    },
    {
      "entropy": 6.206667804718018,
      "epoch": 0.43006004945249027,
      "grad_norm": 1.75,
      "learning_rate": 0.0004986817212382419,
      "loss": 6.1441,
      "mean_token_accuracy": 0.14960070550441742,
      "num_tokens": 9089555.0,
      "step": 4870
    },
    {
      "entropy": 6.226404857635498,
      "epoch": 0.4305015895443306,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004986783150929786,
      "loss": 6.1278,
      "mean_token_accuracy": 0.14620326682925225,
      "num_tokens": 9099091.0,
      "step": 4875
    },
    {
      "entropy": 6.274065351486206,
      "epoch": 0.430943129636171,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004986749045659845,
      "loss": 6.1929,
      "mean_token_accuracy": 0.1484577551484108,
      "num_tokens": 9109008.0,
      "step": 4880
    },
    {
      "entropy": 6.215702390670776,
      "epoch": 0.4313846697280113,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004986714896573261,
      "loss": 6.0642,
      "mean_token_accuracy": 0.15413544476032257,
      "num_tokens": 9117882.0,
      "step": 4885
    },
    {
      "entropy": 6.18838906288147,
      "epoch": 0.43182620981985165,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004986680703670704,
      "loss": 6.2156,
      "mean_token_accuracy": 0.14986138939857482,
      "num_tokens": 9126860.0,
      "step": 4890
    },
    {
      "entropy": 6.218597459793091,
      "epoch": 0.432267749911692,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004986646466952845,
      "loss": 6.1027,
      "mean_token_accuracy": 0.15009103044867517,
      "num_tokens": 9135819.0,
      "step": 4895
    },
    {
      "entropy": 6.20029296875,
      "epoch": 0.4327092900035323,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004986612186420353,
      "loss": 6.0957,
      "mean_token_accuracy": 0.15253715515136718,
      "num_tokens": 9145302.0,
      "step": 4900
    },
    {
      "entropy": 6.156060600280762,
      "epoch": 0.43315083009537264,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004986577862073901,
      "loss": 6.1599,
      "mean_token_accuracy": 0.14464066177606583,
      "num_tokens": 9154667.0,
      "step": 4905
    },
    {
      "entropy": 6.355712270736694,
      "epoch": 0.43359237018721297,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004986543493914159,
      "loss": 6.2065,
      "mean_token_accuracy": 0.1452227145433426,
      "num_tokens": 9164562.0,
      "step": 4910
    },
    {
      "entropy": 6.344915914535522,
      "epoch": 0.43403391027905336,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004986509081941805,
      "loss": 6.2122,
      "mean_token_accuracy": 0.14627591297030448,
      "num_tokens": 9174872.0,
      "step": 4915
    },
    {
      "entropy": 6.154640483856201,
      "epoch": 0.4344754503708937,
      "grad_norm": 1.75,
      "learning_rate": 0.0004986474626157507,
      "loss": 5.982,
      "mean_token_accuracy": 0.16471818685531617,
      "num_tokens": 9184322.0,
      "step": 4920
    },
    {
      "entropy": 6.106767034530639,
      "epoch": 0.434916990462734,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004986440126561945,
      "loss": 6.118,
      "mean_token_accuracy": 0.14820999652147293,
      "num_tokens": 9194450.0,
      "step": 4925
    },
    {
      "entropy": 6.2701087474823,
      "epoch": 0.43535853055457435,
      "grad_norm": 1.75,
      "learning_rate": 0.0004986405583155792,
      "loss": 6.1522,
      "mean_token_accuracy": 0.1471060834825039,
      "num_tokens": 9203658.0,
      "step": 4930
    },
    {
      "entropy": 6.234064769744873,
      "epoch": 0.4358000706464147,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004986370995939725,
      "loss": 6.1212,
      "mean_token_accuracy": 0.15071560591459274,
      "num_tokens": 9213609.0,
      "step": 4935
    },
    {
      "entropy": 6.175269746780396,
      "epoch": 0.436241610738255,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004986336364914423,
      "loss": 6.1706,
      "mean_token_accuracy": 0.1434772104024887,
      "num_tokens": 9222704.0,
      "step": 4940
    },
    {
      "entropy": 6.222776937484741,
      "epoch": 0.43668315083009535,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004986301690080564,
      "loss": 6.1661,
      "mean_token_accuracy": 0.14416092187166213,
      "num_tokens": 9231599.0,
      "step": 4945
    },
    {
      "entropy": 6.2662248611450195,
      "epoch": 0.43712469092193573,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004986266971438826,
      "loss": 6.1764,
      "mean_token_accuracy": 0.14897412061691284,
      "num_tokens": 9241886.0,
      "step": 4950
    },
    {
      "entropy": 6.277956199645996,
      "epoch": 0.43756623101377606,
      "grad_norm": 1.796875,
      "learning_rate": 0.000498623220898989,
      "loss": 6.1638,
      "mean_token_accuracy": 0.1432771749794483,
      "num_tokens": 9251085.0,
      "step": 4955
    },
    {
      "entropy": 6.278365135192871,
      "epoch": 0.4380077711056164,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004986197402734436,
      "loss": 6.1487,
      "mean_token_accuracy": 0.1480664610862732,
      "num_tokens": 9259601.0,
      "step": 4960
    },
    {
      "entropy": 6.258785533905029,
      "epoch": 0.4384493111974567,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004986162552673148,
      "loss": 6.1799,
      "mean_token_accuracy": 0.15204193890094758,
      "num_tokens": 9268935.0,
      "step": 4965
    },
    {
      "entropy": 6.276970481872558,
      "epoch": 0.43889085128929706,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004986127658806706,
      "loss": 6.1838,
      "mean_token_accuracy": 0.14441031441092492,
      "num_tokens": 9277647.0,
      "step": 4970
    },
    {
      "entropy": 6.257901382446289,
      "epoch": 0.4393323913811374,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004986092721135796,
      "loss": 6.1254,
      "mean_token_accuracy": 0.1464274510741234,
      "num_tokens": 9286610.0,
      "step": 4975
    },
    {
      "entropy": 6.224040937423706,
      "epoch": 0.4397739314729777,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004986057739661101,
      "loss": 6.2149,
      "mean_token_accuracy": 0.14213823676109313,
      "num_tokens": 9295946.0,
      "step": 4980
    },
    {
      "entropy": 6.199743700027466,
      "epoch": 0.4402154715648181,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004986022714383307,
      "loss": 6.1125,
      "mean_token_accuracy": 0.1442783683538437,
      "num_tokens": 9304903.0,
      "step": 4985
    },
    {
      "entropy": 6.1956946849823,
      "epoch": 0.44065701165665844,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004985987645303099,
      "loss": 6.0175,
      "mean_token_accuracy": 0.15016455501317977,
      "num_tokens": 9313606.0,
      "step": 4990
    },
    {
      "entropy": 6.185058498382569,
      "epoch": 0.44109855174849877,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004985952532421164,
      "loss": 6.1373,
      "mean_token_accuracy": 0.14618215709924698,
      "num_tokens": 9322815.0,
      "step": 4995
    },
    {
      "entropy": 6.249926280975342,
      "epoch": 0.4415400918403391,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004985917375738193,
      "loss": 6.0576,
      "mean_token_accuracy": 0.15385938584804534,
      "num_tokens": 9332630.0,
      "step": 5000
    },
    {
      "entropy": 6.112989044189453,
      "epoch": 0.44198163193217943,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004985882175254871,
      "loss": 6.1087,
      "mean_token_accuracy": 0.14714636355638505,
      "num_tokens": 9342216.0,
      "step": 5005
    },
    {
      "entropy": 6.237312364578247,
      "epoch": 0.44242317202401976,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004985846930971887,
      "loss": 6.147,
      "mean_token_accuracy": 0.14364741295576094,
      "num_tokens": 9352295.0,
      "step": 5010
    },
    {
      "entropy": 6.195395755767822,
      "epoch": 0.4428647121158601,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004985811642889937,
      "loss": 6.145,
      "mean_token_accuracy": 0.14929724484682083,
      "num_tokens": 9361462.0,
      "step": 5015
    },
    {
      "entropy": 6.209699726104736,
      "epoch": 0.4433062522077005,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004985776311009705,
      "loss": 6.0618,
      "mean_token_accuracy": 0.14563873261213303,
      "num_tokens": 9370379.0,
      "step": 5020
    },
    {
      "entropy": 6.244296598434448,
      "epoch": 0.4437477922995408,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004985740935331888,
      "loss": 6.1192,
      "mean_token_accuracy": 0.1475251004099846,
      "num_tokens": 9379255.0,
      "step": 5025
    },
    {
      "entropy": 6.226190853118896,
      "epoch": 0.44418933239138114,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004985705515857177,
      "loss": 6.0889,
      "mean_token_accuracy": 0.15009573549032212,
      "num_tokens": 9389313.0,
      "step": 5030
    },
    {
      "entropy": 6.096516418457031,
      "epoch": 0.4446308724832215,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004985670052586268,
      "loss": 5.8827,
      "mean_token_accuracy": 0.16914348155260087,
      "num_tokens": 9397778.0,
      "step": 5035
    },
    {
      "entropy": 6.196593523025513,
      "epoch": 0.4450724125750618,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004985634545519853,
      "loss": 6.1952,
      "mean_token_accuracy": 0.1398940645158291,
      "num_tokens": 9407831.0,
      "step": 5040
    },
    {
      "entropy": 6.167946147918701,
      "epoch": 0.44551395266690214,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004985598994658629,
      "loss": 6.0666,
      "mean_token_accuracy": 0.1540227361023426,
      "num_tokens": 9418458.0,
      "step": 5045
    },
    {
      "entropy": 6.179390811920166,
      "epoch": 0.44595549275874247,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004985563400003291,
      "loss": 5.9951,
      "mean_token_accuracy": 0.16488620936870574,
      "num_tokens": 9426911.0,
      "step": 5050
    },
    {
      "entropy": 6.27180323600769,
      "epoch": 0.44639703285058285,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004985527761554539,
      "loss": 6.1607,
      "mean_token_accuracy": 0.14800687357783318,
      "num_tokens": 9435896.0,
      "step": 5055
    },
    {
      "entropy": 6.298493957519531,
      "epoch": 0.4468385729424232,
      "grad_norm": 2.015625,
      "learning_rate": 0.000498549207931307,
      "loss": 6.181,
      "mean_token_accuracy": 0.14514534771442414,
      "num_tokens": 9445272.0,
      "step": 5060
    },
    {
      "entropy": 6.22353572845459,
      "epoch": 0.4472801130342635,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004985456353279581,
      "loss": 6.132,
      "mean_token_accuracy": 0.14213817343115806,
      "num_tokens": 9455377.0,
      "step": 5065
    },
    {
      "entropy": 6.2302265644073485,
      "epoch": 0.44772165312610385,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004985420583454774,
      "loss": 6.154,
      "mean_token_accuracy": 0.15052020102739333,
      "num_tokens": 9464918.0,
      "step": 5070
    },
    {
      "entropy": 6.2570312976837155,
      "epoch": 0.4481631932179442,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004985384769839349,
      "loss": 6.1742,
      "mean_token_accuracy": 0.14404505118727684,
      "num_tokens": 9473322.0,
      "step": 5075
    },
    {
      "entropy": 6.266055107116699,
      "epoch": 0.4486047333097845,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004985348912434008,
      "loss": 6.0661,
      "mean_token_accuracy": 0.15727601945400238,
      "num_tokens": 9482255.0,
      "step": 5080
    },
    {
      "entropy": 6.288808965682984,
      "epoch": 0.44904627340162484,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004985313011239452,
      "loss": 6.1165,
      "mean_token_accuracy": 0.1513061538338661,
      "num_tokens": 9491709.0,
      "step": 5085
    },
    {
      "entropy": 6.160472440719604,
      "epoch": 0.4494878134934652,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004985277066256388,
      "loss": 6.1041,
      "mean_token_accuracy": 0.15101252347230912,
      "num_tokens": 9500594.0,
      "step": 5090
    },
    {
      "entropy": 6.163977861404419,
      "epoch": 0.44992935358530556,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004985241077485515,
      "loss": 6.0951,
      "mean_token_accuracy": 0.15175875127315522,
      "num_tokens": 9509088.0,
      "step": 5095
    },
    {
      "entropy": 6.2438524723052975,
      "epoch": 0.4503708936771459,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004985205044927541,
      "loss": 6.0798,
      "mean_token_accuracy": 0.15726079791784286,
      "num_tokens": 9517776.0,
      "step": 5100
    },
    {
      "entropy": 6.192723655700684,
      "epoch": 0.4508124337689862,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004985168968583173,
      "loss": 6.146,
      "mean_token_accuracy": 0.15224774926900864,
      "num_tokens": 9527080.0,
      "step": 5105
    },
    {
      "entropy": 6.150369167327881,
      "epoch": 0.45125397386082655,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004985132848453114,
      "loss": 5.9983,
      "mean_token_accuracy": 0.16127897500991822,
      "num_tokens": 9536358.0,
      "step": 5110
    },
    {
      "entropy": 6.167308330535889,
      "epoch": 0.4516955139526669,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004985096684538075,
      "loss": 6.1038,
      "mean_token_accuracy": 0.14964734613895417,
      "num_tokens": 9545528.0,
      "step": 5115
    },
    {
      "entropy": 6.190183210372925,
      "epoch": 0.4521370540445072,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004985060476838763,
      "loss": 6.1315,
      "mean_token_accuracy": 0.14921002686023713,
      "num_tokens": 9554433.0,
      "step": 5120
    },
    {
      "entropy": 6.230302476882935,
      "epoch": 0.4525785941363476,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004985024225355887,
      "loss": 6.0682,
      "mean_token_accuracy": 0.14985187724232674,
      "num_tokens": 9563932.0,
      "step": 5125
    },
    {
      "entropy": 6.27160873413086,
      "epoch": 0.45302013422818793,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004984987930090158,
      "loss": 6.0615,
      "mean_token_accuracy": 0.15488119423389435,
      "num_tokens": 9572829.0,
      "step": 5130
    },
    {
      "entropy": 6.151999282836914,
      "epoch": 0.45346167432002826,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004984951591042285,
      "loss": 6.1005,
      "mean_token_accuracy": 0.15698079913854598,
      "num_tokens": 9583597.0,
      "step": 5135
    },
    {
      "entropy": 6.26218466758728,
      "epoch": 0.4539032144118686,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004984915208212983,
      "loss": 6.1823,
      "mean_token_accuracy": 0.14492006301879884,
      "num_tokens": 9593114.0,
      "step": 5140
    },
    {
      "entropy": 6.235624980926514,
      "epoch": 0.4543447545037089,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004984878781602964,
      "loss": 6.0829,
      "mean_token_accuracy": 0.14600640833377837,
      "num_tokens": 9601454.0,
      "step": 5145
    },
    {
      "entropy": 6.186273860931396,
      "epoch": 0.45478629459554926,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004984842311212939,
      "loss": 6.1448,
      "mean_token_accuracy": 0.15090979114174843,
      "num_tokens": 9611994.0,
      "step": 5150
    },
    {
      "entropy": 6.242185878753662,
      "epoch": 0.4552278346873896,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004984805797043625,
      "loss": 6.0986,
      "mean_token_accuracy": 0.14918104112148284,
      "num_tokens": 9621318.0,
      "step": 5155
    },
    {
      "entropy": 6.221137046813965,
      "epoch": 0.45566937477923,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004984769239095736,
      "loss": 6.0973,
      "mean_token_accuracy": 0.15199169963598252,
      "num_tokens": 9630270.0,
      "step": 5160
    },
    {
      "entropy": 6.247491025924683,
      "epoch": 0.4561109148710703,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004984732637369989,
      "loss": 6.1263,
      "mean_token_accuracy": 0.1468114122748375,
      "num_tokens": 9640391.0,
      "step": 5165
    },
    {
      "entropy": 6.236334943771363,
      "epoch": 0.45655245496291064,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004984695991867099,
      "loss": 6.1433,
      "mean_token_accuracy": 0.1472519591450691,
      "num_tokens": 9648827.0,
      "step": 5170
    },
    {
      "entropy": 6.168450736999512,
      "epoch": 0.45699399505475097,
      "grad_norm": 2.125,
      "learning_rate": 0.0004984659302587788,
      "loss": 6.0673,
      "mean_token_accuracy": 0.14995986744761466,
      "num_tokens": 9657940.0,
      "step": 5175
    },
    {
      "entropy": 6.161103534698486,
      "epoch": 0.4574355351465913,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000498462256953277,
      "loss": 6.1943,
      "mean_token_accuracy": 0.14195918291807175,
      "num_tokens": 9668296.0,
      "step": 5180
    },
    {
      "entropy": 6.272454166412354,
      "epoch": 0.45787707523843163,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004984585792702767,
      "loss": 6.1321,
      "mean_token_accuracy": 0.1457594834268093,
      "num_tokens": 9677914.0,
      "step": 5185
    },
    {
      "entropy": 6.276125431060791,
      "epoch": 0.45831861533027196,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004984548972098501,
      "loss": 6.2682,
      "mean_token_accuracy": 0.13744014501571655,
      "num_tokens": 9687339.0,
      "step": 5190
    },
    {
      "entropy": 6.200534772872925,
      "epoch": 0.45876015542211235,
      "grad_norm": 1.96875,
      "learning_rate": 0.000498451210772069,
      "loss": 6.0492,
      "mean_token_accuracy": 0.15557005926966666,
      "num_tokens": 9696889.0,
      "step": 5195
    },
    {
      "entropy": 6.138857126235962,
      "epoch": 0.4592016955139527,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004984475199570058,
      "loss": 6.0119,
      "mean_token_accuracy": 0.1522158980369568,
      "num_tokens": 9705498.0,
      "step": 5200
    },
    {
      "entropy": 6.187509822845459,
      "epoch": 0.459643235605793,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004984438247647329,
      "loss": 6.0819,
      "mean_token_accuracy": 0.14996571764349936,
      "num_tokens": 9713936.0,
      "step": 5205
    },
    {
      "entropy": 6.235674858093262,
      "epoch": 0.46008477569763334,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004984401251953223,
      "loss": 6.139,
      "mean_token_accuracy": 0.14663990586996078,
      "num_tokens": 9723924.0,
      "step": 5210
    },
    {
      "entropy": 6.2642858028411865,
      "epoch": 0.4605263157894737,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004984364212488469,
      "loss": 6.1785,
      "mean_token_accuracy": 0.14377184957265854,
      "num_tokens": 9733427.0,
      "step": 5215
    },
    {
      "entropy": 6.205567359924316,
      "epoch": 0.460967855881314,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004984327129253789,
      "loss": 5.9885,
      "mean_token_accuracy": 0.15839618146419526,
      "num_tokens": 9742054.0,
      "step": 5220
    },
    {
      "entropy": 6.19922513961792,
      "epoch": 0.46140939597315433,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004984290002249914,
      "loss": 6.1593,
      "mean_token_accuracy": 0.1428479090332985,
      "num_tokens": 9752447.0,
      "step": 5225
    },
    {
      "entropy": 6.184007072448731,
      "epoch": 0.4618509360649947,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004984252831477567,
      "loss": 6.0063,
      "mean_token_accuracy": 0.15819374471902847,
      "num_tokens": 9760878.0,
      "step": 5230
    },
    {
      "entropy": 6.133184146881104,
      "epoch": 0.46229247615683505,
      "grad_norm": 2.25,
      "learning_rate": 0.0004984215616937477,
      "loss": 6.0676,
      "mean_token_accuracy": 0.15267323404550553,
      "num_tokens": 9770200.0,
      "step": 5235
    },
    {
      "entropy": 6.275788307189941,
      "epoch": 0.4627340162486754,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004984178358630374,
      "loss": 6.1659,
      "mean_token_accuracy": 0.14913044422864913,
      "num_tokens": 9780303.0,
      "step": 5240
    },
    {
      "entropy": 6.21355676651001,
      "epoch": 0.4631755563405157,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004984141056556989,
      "loss": 6.062,
      "mean_token_accuracy": 0.15438947528600694,
      "num_tokens": 9790248.0,
      "step": 5245
    },
    {
      "entropy": 6.2150005340576175,
      "epoch": 0.46361709643235605,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004984103710718051,
      "loss": 6.1757,
      "mean_token_accuracy": 0.14253960698843002,
      "num_tokens": 9799345.0,
      "step": 5250
    },
    {
      "entropy": 6.234236717224121,
      "epoch": 0.4640586365241964,
      "grad_norm": 1.84375,
      "learning_rate": 0.000498406632111429,
      "loss": 6.0315,
      "mean_token_accuracy": 0.1504400633275509,
      "num_tokens": 9808561.0,
      "step": 5255
    },
    {
      "entropy": 6.295517539978027,
      "epoch": 0.4645001766160367,
      "grad_norm": 2.125,
      "learning_rate": 0.0004984028887746443,
      "loss": 6.091,
      "mean_token_accuracy": 0.15317275375127792,
      "num_tokens": 9818324.0,
      "step": 5260
    },
    {
      "entropy": 6.0996545314788815,
      "epoch": 0.4649417167078771,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004983991410615239,
      "loss": 6.0616,
      "mean_token_accuracy": 0.16138627082109452,
      "num_tokens": 9827935.0,
      "step": 5265
    },
    {
      "entropy": 6.135337018966675,
      "epoch": 0.4653832567997174,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004983953889721414,
      "loss": 6.0966,
      "mean_token_accuracy": 0.15333109125494956,
      "num_tokens": 9837118.0,
      "step": 5270
    },
    {
      "entropy": 6.33495831489563,
      "epoch": 0.46582479689155776,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004983916325065703,
      "loss": 6.1469,
      "mean_token_accuracy": 0.14710031673312188,
      "num_tokens": 9846197.0,
      "step": 5275
    },
    {
      "entropy": 6.320817136764527,
      "epoch": 0.4662663369833981,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004983878716648842,
      "loss": 6.2274,
      "mean_token_accuracy": 0.1372292719781399,
      "num_tokens": 9856120.0,
      "step": 5280
    },
    {
      "entropy": 6.224937200546265,
      "epoch": 0.4667078770752384,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004983841064471567,
      "loss": 6.1598,
      "mean_token_accuracy": 0.14868675619363786,
      "num_tokens": 9865599.0,
      "step": 5285
    },
    {
      "entropy": 6.239090204238892,
      "epoch": 0.46714941716707875,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004983803368534617,
      "loss": 6.0918,
      "mean_token_accuracy": 0.14730772823095323,
      "num_tokens": 9876471.0,
      "step": 5290
    },
    {
      "entropy": 6.2774077415466305,
      "epoch": 0.4675909572589191,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004983765628838728,
      "loss": 6.1372,
      "mean_token_accuracy": 0.14997045919299126,
      "num_tokens": 9887680.0,
      "step": 5295
    },
    {
      "entropy": 6.305510187149048,
      "epoch": 0.46803249735075947,
      "grad_norm": 2.5,
      "learning_rate": 0.0004983727845384641,
      "loss": 6.2229,
      "mean_token_accuracy": 0.13749768882989882,
      "num_tokens": 9897366.0,
      "step": 5300
    },
    {
      "entropy": 6.251697826385498,
      "epoch": 0.4684740374425998,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004983690018173096,
      "loss": 6.17,
      "mean_token_accuracy": 0.14041578024625778,
      "num_tokens": 9907437.0,
      "step": 5305
    },
    {
      "entropy": 6.193480968475342,
      "epoch": 0.46891557753444013,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004983652147204834,
      "loss": 5.9589,
      "mean_token_accuracy": 0.16042340472340583,
      "num_tokens": 9915766.0,
      "step": 5310
    },
    {
      "entropy": 6.205143594741822,
      "epoch": 0.46935711762628046,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004983614232480598,
      "loss": 6.1733,
      "mean_token_accuracy": 0.14470496103167535,
      "num_tokens": 9925389.0,
      "step": 5315
    },
    {
      "entropy": 6.167950963973999,
      "epoch": 0.4697986577181208,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004983576274001127,
      "loss": 6.2229,
      "mean_token_accuracy": 0.14586539566516876,
      "num_tokens": 9935798.0,
      "step": 5320
    },
    {
      "entropy": 6.2063414573669435,
      "epoch": 0.4702401978099611,
      "grad_norm": 2.15625,
      "learning_rate": 0.000498353827176717,
      "loss": 6.0719,
      "mean_token_accuracy": 0.15585483759641647,
      "num_tokens": 9945579.0,
      "step": 5325
    },
    {
      "entropy": 6.267688894271851,
      "epoch": 0.4706817379018015,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004983500225779466,
      "loss": 6.0985,
      "mean_token_accuracy": 0.1498646095395088,
      "num_tokens": 9955248.0,
      "step": 5330
    },
    {
      "entropy": 6.230977582931518,
      "epoch": 0.47112327799364184,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004983462136038764,
      "loss": 6.0943,
      "mean_token_accuracy": 0.1582035481929779,
      "num_tokens": 9965078.0,
      "step": 5335
    },
    {
      "entropy": 6.220054006576538,
      "epoch": 0.4715648180854822,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004983424002545809,
      "loss": 6.0784,
      "mean_token_accuracy": 0.1562138557434082,
      "num_tokens": 9973932.0,
      "step": 5340
    },
    {
      "entropy": 6.264338159561158,
      "epoch": 0.4720063581773225,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004983385825301348,
      "loss": 6.1322,
      "mean_token_accuracy": 0.14699085503816606,
      "num_tokens": 9983315.0,
      "step": 5345
    },
    {
      "entropy": 6.2415910243988035,
      "epoch": 0.47244789826916284,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004983347604306129,
      "loss": 6.0708,
      "mean_token_accuracy": 0.14864065274596214,
      "num_tokens": 9992987.0,
      "step": 5350
    },
    {
      "entropy": 6.094457197189331,
      "epoch": 0.47288943836100317,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004983309339560899,
      "loss": 6.1439,
      "mean_token_accuracy": 0.14318862557411194,
      "num_tokens": 10002268.0,
      "step": 5355
    },
    {
      "entropy": 6.2612992286682125,
      "epoch": 0.4733309784528435,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004983271031066412,
      "loss": 6.1586,
      "mean_token_accuracy": 0.14363493770360947,
      "num_tokens": 10011772.0,
      "step": 5360
    },
    {
      "entropy": 6.305984926223755,
      "epoch": 0.4737725185446839,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004983232678823414,
      "loss": 6.141,
      "mean_token_accuracy": 0.15256721675395965,
      "num_tokens": 10021069.0,
      "step": 5365
    },
    {
      "entropy": 6.123571062088013,
      "epoch": 0.4742140586365242,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004983194282832657,
      "loss": 6.0592,
      "mean_token_accuracy": 0.15314415395259856,
      "num_tokens": 10029706.0,
      "step": 5370
    },
    {
      "entropy": 6.0487165451049805,
      "epoch": 0.47465559872836455,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004983155843094895,
      "loss": 6.0282,
      "mean_token_accuracy": 0.1549852669239044,
      "num_tokens": 10039633.0,
      "step": 5375
    },
    {
      "entropy": 6.2324432849884035,
      "epoch": 0.4750971388202049,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004983117359610881,
      "loss": 6.0862,
      "mean_token_accuracy": 0.15131546556949615,
      "num_tokens": 10048675.0,
      "step": 5380
    },
    {
      "entropy": 6.257778930664062,
      "epoch": 0.4755386789120452,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004983078832381367,
      "loss": 6.0631,
      "mean_token_accuracy": 0.15427327901124954,
      "num_tokens": 10057447.0,
      "step": 5385
    },
    {
      "entropy": 6.1880207538604735,
      "epoch": 0.47598021900388554,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004983040261407109,
      "loss": 6.1598,
      "mean_token_accuracy": 0.14713627547025682,
      "num_tokens": 10067020.0,
      "step": 5390
    },
    {
      "entropy": 6.261177015304566,
      "epoch": 0.47642175909572587,
      "grad_norm": 1.875,
      "learning_rate": 0.0004983001646688863,
      "loss": 6.1696,
      "mean_token_accuracy": 0.14345274344086648,
      "num_tokens": 10076466.0,
      "step": 5395
    },
    {
      "entropy": 6.160317230224609,
      "epoch": 0.47686329918756626,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004982962988227383,
      "loss": 6.0432,
      "mean_token_accuracy": 0.15397232472896577,
      "num_tokens": 10085384.0,
      "step": 5400
    },
    {
      "entropy": 6.20677080154419,
      "epoch": 0.4773048392794066,
      "grad_norm": 1.765625,
      "learning_rate": 0.000498292428602343,
      "loss": 6.1742,
      "mean_token_accuracy": 0.1478663094341755,
      "num_tokens": 10094015.0,
      "step": 5405
    },
    {
      "entropy": 6.2046802043914795,
      "epoch": 0.4777463793712469,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004982885540077758,
      "loss": 6.0629,
      "mean_token_accuracy": 0.15281429886817932,
      "num_tokens": 10103107.0,
      "step": 5410
    },
    {
      "entropy": 6.079674625396729,
      "epoch": 0.47818791946308725,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004982846750391129,
      "loss": 5.9917,
      "mean_token_accuracy": 0.15909103006124498,
      "num_tokens": 10112408.0,
      "step": 5415
    },
    {
      "entropy": 6.1567846775054935,
      "epoch": 0.4786294595549276,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004982807916964303,
      "loss": 6.0707,
      "mean_token_accuracy": 0.14905139431357384,
      "num_tokens": 10121605.0,
      "step": 5420
    },
    {
      "entropy": 6.115311241149902,
      "epoch": 0.4790709996467679,
      "grad_norm": 2.484375,
      "learning_rate": 0.000498276903979804,
      "loss": 6.0543,
      "mean_token_accuracy": 0.1517949789762497,
      "num_tokens": 10130484.0,
      "step": 5425
    },
    {
      "entropy": 6.133319520950318,
      "epoch": 0.47951253973860825,
      "grad_norm": 2.21875,
      "learning_rate": 0.00049827301188931,
      "loss": 6.0355,
      "mean_token_accuracy": 0.15710298717021942,
      "num_tokens": 10140002.0,
      "step": 5430
    },
    {
      "entropy": 6.278661727905273,
      "epoch": 0.47995407983044863,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004982691154250247,
      "loss": 6.1998,
      "mean_token_accuracy": 0.15036016926169396,
      "num_tokens": 10150287.0,
      "step": 5435
    },
    {
      "entropy": 6.18853178024292,
      "epoch": 0.48039561992228896,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004982652145870245,
      "loss": 6.0384,
      "mean_token_accuracy": 0.15560515448451043,
      "num_tokens": 10160615.0,
      "step": 5440
    },
    {
      "entropy": 6.165748548507691,
      "epoch": 0.4808371600141293,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004982613093753856,
      "loss": 6.0834,
      "mean_token_accuracy": 0.1541260801255703,
      "num_tokens": 10170271.0,
      "step": 5445
    },
    {
      "entropy": 6.203157663345337,
      "epoch": 0.4812787001059696,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004982573997901847,
      "loss": 6.0805,
      "mean_token_accuracy": 0.15566618293523787,
      "num_tokens": 10179663.0,
      "step": 5450
    },
    {
      "entropy": 6.230932712554932,
      "epoch": 0.48172024019780996,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004982534858314982,
      "loss": 6.1355,
      "mean_token_accuracy": 0.14900892227888107,
      "num_tokens": 10188608.0,
      "step": 5455
    },
    {
      "entropy": 6.262550163269043,
      "epoch": 0.4821617802896503,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004982495674994031,
      "loss": 6.0865,
      "mean_token_accuracy": 0.14918412119150162,
      "num_tokens": 10198462.0,
      "step": 5460
    },
    {
      "entropy": 6.1284778118133545,
      "epoch": 0.4826033203814906,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004982456447939758,
      "loss": 6.0629,
      "mean_token_accuracy": 0.15871936827898026,
      "num_tokens": 10208362.0,
      "step": 5465
    },
    {
      "entropy": 6.189786338806153,
      "epoch": 0.483044860473331,
      "grad_norm": 1.875,
      "learning_rate": 0.0004982417177152933,
      "loss": 6.0889,
      "mean_token_accuracy": 0.15344267934560776,
      "num_tokens": 10217915.0,
      "step": 5470
    },
    {
      "entropy": 6.235037326812744,
      "epoch": 0.48348640056517134,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004982377862634325,
      "loss": 6.131,
      "mean_token_accuracy": 0.15101420134305954,
      "num_tokens": 10226913.0,
      "step": 5475
    },
    {
      "entropy": 6.189169836044312,
      "epoch": 0.48392794065701167,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004982338504384705,
      "loss": 6.0879,
      "mean_token_accuracy": 0.15291856229305267,
      "num_tokens": 10236516.0,
      "step": 5480
    },
    {
      "entropy": 6.292510843276977,
      "epoch": 0.484369480748852,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004982299102404843,
      "loss": 6.1347,
      "mean_token_accuracy": 0.15164220482110977,
      "num_tokens": 10245492.0,
      "step": 5485
    },
    {
      "entropy": 6.15688533782959,
      "epoch": 0.48481102084069233,
      "grad_norm": 1.953125,
      "learning_rate": 0.000498225965669551,
      "loss": 5.9841,
      "mean_token_accuracy": 0.16536709070205688,
      "num_tokens": 10254094.0,
      "step": 5490
    },
    {
      "entropy": 6.089360761642456,
      "epoch": 0.48525256093253266,
      "grad_norm": 2.125,
      "learning_rate": 0.0004982220167257482,
      "loss": 6.0383,
      "mean_token_accuracy": 0.15727589279413223,
      "num_tokens": 10263645.0,
      "step": 5495
    },
    {
      "entropy": 6.084077024459839,
      "epoch": 0.485694101024373,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004982180634091529,
      "loss": 6.0128,
      "mean_token_accuracy": 0.1580739438533783,
      "num_tokens": 10273416.0,
      "step": 5500
    },
    {
      "entropy": 6.23244342803955,
      "epoch": 0.4861356411162134,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004982141057198427,
      "loss": 6.1204,
      "mean_token_accuracy": 0.15273454710841178,
      "num_tokens": 10283194.0,
      "step": 5505
    },
    {
      "entropy": 6.240085029602051,
      "epoch": 0.4865771812080537,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004982101436578952,
      "loss": 6.1013,
      "mean_token_accuracy": 0.14988823384046554,
      "num_tokens": 10292188.0,
      "step": 5510
    },
    {
      "entropy": 6.161333322525024,
      "epoch": 0.48701872129989404,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004982061772233878,
      "loss": 6.0099,
      "mean_token_accuracy": 0.16448492854833602,
      "num_tokens": 10301539.0,
      "step": 5515
    },
    {
      "entropy": 6.117511987686157,
      "epoch": 0.4874602613917344,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004982022064163984,
      "loss": 6.0548,
      "mean_token_accuracy": 0.15493491739034654,
      "num_tokens": 10310379.0,
      "step": 5520
    },
    {
      "entropy": 6.296630859375,
      "epoch": 0.4879018014835747,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004981982312370047,
      "loss": 6.242,
      "mean_token_accuracy": 0.13949622288346292,
      "num_tokens": 10320622.0,
      "step": 5525
    },
    {
      "entropy": 6.197968816757202,
      "epoch": 0.48834334157541504,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004981942516852847,
      "loss": 6.0672,
      "mean_token_accuracy": 0.1509483724832535,
      "num_tokens": 10330525.0,
      "step": 5530
    },
    {
      "entropy": 6.176139831542969,
      "epoch": 0.48878488166725537,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004981902677613161,
      "loss": 6.0748,
      "mean_token_accuracy": 0.15572689101099968,
      "num_tokens": 10340288.0,
      "step": 5535
    },
    {
      "entropy": 6.263591957092285,
      "epoch": 0.48922642175909575,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004981862794651771,
      "loss": 6.1882,
      "mean_token_accuracy": 0.14224747344851493,
      "num_tokens": 10349836.0,
      "step": 5540
    },
    {
      "entropy": 6.236993360519409,
      "epoch": 0.4896679618509361,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004981822867969459,
      "loss": 6.1135,
      "mean_token_accuracy": 0.14976053535938263,
      "num_tokens": 10359481.0,
      "step": 5545
    },
    {
      "entropy": 6.228906965255737,
      "epoch": 0.4901095019427764,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004981782897567006,
      "loss": 6.068,
      "mean_token_accuracy": 0.1566969633102417,
      "num_tokens": 10369837.0,
      "step": 5550
    },
    {
      "entropy": 6.180895662307739,
      "epoch": 0.49055104203461675,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004981742883445195,
      "loss": 6.0438,
      "mean_token_accuracy": 0.15206340700387955,
      "num_tokens": 10379276.0,
      "step": 5555
    },
    {
      "entropy": 6.1696826934814455,
      "epoch": 0.4909925821264571,
      "grad_norm": 2.375,
      "learning_rate": 0.000498170282560481,
      "loss": 6.1867,
      "mean_token_accuracy": 0.14743926301598548,
      "num_tokens": 10388878.0,
      "step": 5560
    },
    {
      "entropy": 6.312894344329834,
      "epoch": 0.4914341222182974,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004981662724046637,
      "loss": 6.1512,
      "mean_token_accuracy": 0.14519662261009217,
      "num_tokens": 10399148.0,
      "step": 5565
    },
    {
      "entropy": 6.235131168365479,
      "epoch": 0.49187566231013774,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000498162257877146,
      "loss": 6.0919,
      "mean_token_accuracy": 0.15135070234537124,
      "num_tokens": 10409002.0,
      "step": 5570
    },
    {
      "entropy": 6.129899597167968,
      "epoch": 0.4923172024019781,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004981582389780065,
      "loss": 6.1344,
      "mean_token_accuracy": 0.1505287192761898,
      "num_tokens": 10418083.0,
      "step": 5575
    },
    {
      "entropy": 6.293564653396606,
      "epoch": 0.49275874249381846,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004981542157073241,
      "loss": 6.1486,
      "mean_token_accuracy": 0.14746079742908477,
      "num_tokens": 10428134.0,
      "step": 5580
    },
    {
      "entropy": 6.274387216567993,
      "epoch": 0.4932002825856588,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004981501880651775,
      "loss": 6.132,
      "mean_token_accuracy": 0.14803507775068284,
      "num_tokens": 10437436.0,
      "step": 5585
    },
    {
      "entropy": 6.2339592456817625,
      "epoch": 0.4936418226774991,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004981461560516457,
      "loss": 6.0992,
      "mean_token_accuracy": 0.14704242050647737,
      "num_tokens": 10446176.0,
      "step": 5590
    },
    {
      "entropy": 6.229759883880615,
      "epoch": 0.49408336276933945,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004981421196668075,
      "loss": 6.243,
      "mean_token_accuracy": 0.14259758442640305,
      "num_tokens": 10455953.0,
      "step": 5595
    },
    {
      "entropy": 6.307909250259399,
      "epoch": 0.4945249028611798,
      "grad_norm": 2.0,
      "learning_rate": 0.0004981380789107422,
      "loss": 6.0736,
      "mean_token_accuracy": 0.15389894247055053,
      "num_tokens": 10465129.0,
      "step": 5600
    },
    {
      "entropy": 6.122000789642334,
      "epoch": 0.4949664429530201,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004981340337835287,
      "loss": 6.0793,
      "mean_token_accuracy": 0.1542934447526932,
      "num_tokens": 10474376.0,
      "step": 5605
    },
    {
      "entropy": 6.166776704788208,
      "epoch": 0.4954079830448605,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004981299842852464,
      "loss": 6.1663,
      "mean_token_accuracy": 0.1488489493727684,
      "num_tokens": 10484849.0,
      "step": 5610
    },
    {
      "entropy": 6.202208805084228,
      "epoch": 0.49584952313670083,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004981259304159747,
      "loss": 6.0273,
      "mean_token_accuracy": 0.15956118404865266,
      "num_tokens": 10493950.0,
      "step": 5615
    },
    {
      "entropy": 6.168651723861695,
      "epoch": 0.49629106322854116,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004981218721757929,
      "loss": 5.9799,
      "mean_token_accuracy": 0.1569823607802391,
      "num_tokens": 10502890.0,
      "step": 5620
    },
    {
      "entropy": 6.119230270385742,
      "epoch": 0.4967326033203815,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004981178095647805,
      "loss": 5.9866,
      "mean_token_accuracy": 0.15033361464738845,
      "num_tokens": 10511702.0,
      "step": 5625
    },
    {
      "entropy": 6.212582731246949,
      "epoch": 0.4971741434122218,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004981137425830171,
      "loss": 6.0833,
      "mean_token_accuracy": 0.146825347840786,
      "num_tokens": 10520212.0,
      "step": 5630
    },
    {
      "entropy": 6.213112783432007,
      "epoch": 0.49761568350406216,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004981096712305825,
      "loss": 6.0966,
      "mean_token_accuracy": 0.1567651391029358,
      "num_tokens": 10529652.0,
      "step": 5635
    },
    {
      "entropy": 6.214666318893433,
      "epoch": 0.4980572235959025,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004981055955075561,
      "loss": 6.1651,
      "mean_token_accuracy": 0.14563403353095056,
      "num_tokens": 10537980.0,
      "step": 5640
    },
    {
      "entropy": 6.205415534973144,
      "epoch": 0.4984987636877429,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004981015154140181,
      "loss": 6.0204,
      "mean_token_accuracy": 0.15574650317430497,
      "num_tokens": 10546854.0,
      "step": 5645
    },
    {
      "entropy": 6.154613637924195,
      "epoch": 0.4989403037795832,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004980974309500483,
      "loss": 6.0621,
      "mean_token_accuracy": 0.15574722364544868,
      "num_tokens": 10555691.0,
      "step": 5650
    },
    {
      "entropy": 6.17102541923523,
      "epoch": 0.49938184387142354,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004980933421157267,
      "loss": 6.1159,
      "mean_token_accuracy": 0.14486125558614732,
      "num_tokens": 10564750.0,
      "step": 5655
    },
    {
      "entropy": 6.234163522720337,
      "epoch": 0.49982338396326387,
      "grad_norm": 1.625,
      "learning_rate": 0.0004980892489111334,
      "loss": 6.152,
      "mean_token_accuracy": 0.14877569377422334,
      "num_tokens": 10573515.0,
      "step": 5660
    },
    {
      "entropy": 6.125738573074341,
      "epoch": 0.5002649240551043,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004980851513363486,
      "loss": 6.0233,
      "mean_token_accuracy": 0.15632108598947525,
      "num_tokens": 10582730.0,
      "step": 5665
    },
    {
      "entropy": 6.051721811294556,
      "epoch": 0.5007064641469445,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004980810493914526,
      "loss": 5.9253,
      "mean_token_accuracy": 0.1711601883172989,
      "num_tokens": 10591267.0,
      "step": 5670
    },
    {
      "entropy": 6.154175090789795,
      "epoch": 0.5011480042387849,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004980769430765256,
      "loss": 6.04,
      "mean_token_accuracy": 0.1528927817940712,
      "num_tokens": 10599776.0,
      "step": 5675
    },
    {
      "entropy": 6.192328596115113,
      "epoch": 0.5015895443306252,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004980728323916484,
      "loss": 6.033,
      "mean_token_accuracy": 0.1478010207414627,
      "num_tokens": 10609670.0,
      "step": 5680
    },
    {
      "entropy": 6.142802715301514,
      "epoch": 0.5020310844224656,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004980687173369009,
      "loss": 5.9971,
      "mean_token_accuracy": 0.15713515877723694,
      "num_tokens": 10619032.0,
      "step": 5685
    },
    {
      "entropy": 6.176899385452271,
      "epoch": 0.5024726245143059,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004980645979123644,
      "loss": 6.0839,
      "mean_token_accuracy": 0.14918567463755608,
      "num_tokens": 10628851.0,
      "step": 5690
    },
    {
      "entropy": 6.220816135406494,
      "epoch": 0.5029141646061462,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004980604741181192,
      "loss": 6.0664,
      "mean_token_accuracy": 0.1565829783678055,
      "num_tokens": 10638335.0,
      "step": 5695
    },
    {
      "entropy": 6.225756311416626,
      "epoch": 0.5033557046979866,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004980563459542461,
      "loss": 6.1504,
      "mean_token_accuracy": 0.15053488165140153,
      "num_tokens": 10648450.0,
      "step": 5700
    },
    {
      "entropy": 6.272598075866699,
      "epoch": 0.5037972447898269,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004980522134208261,
      "loss": 6.168,
      "mean_token_accuracy": 0.14580240920186044,
      "num_tokens": 10658700.0,
      "step": 5705
    },
    {
      "entropy": 6.281301689147949,
      "epoch": 0.5042387848816673,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004980480765179401,
      "loss": 6.183,
      "mean_token_accuracy": 0.14910355508327483,
      "num_tokens": 10667952.0,
      "step": 5710
    },
    {
      "entropy": 6.137789487838745,
      "epoch": 0.5046803249735076,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004980439352456692,
      "loss": 6.0103,
      "mean_token_accuracy": 0.1529277339577675,
      "num_tokens": 10677228.0,
      "step": 5715
    },
    {
      "entropy": 6.194245338439941,
      "epoch": 0.505121865065348,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004980397896040944,
      "loss": 6.0869,
      "mean_token_accuracy": 0.15276115983724595,
      "num_tokens": 10686183.0,
      "step": 5720
    },
    {
      "entropy": 6.287981605529785,
      "epoch": 0.5055634051571882,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004980356395932969,
      "loss": 6.1249,
      "mean_token_accuracy": 0.15401637256145478,
      "num_tokens": 10695293.0,
      "step": 5725
    },
    {
      "entropy": 6.208388042449951,
      "epoch": 0.5060049452490286,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004980314852133581,
      "loss": 6.0737,
      "mean_token_accuracy": 0.155954746901989,
      "num_tokens": 10704853.0,
      "step": 5730
    },
    {
      "entropy": 6.1891162395477295,
      "epoch": 0.506446485340869,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004980273264643594,
      "loss": 6.0797,
      "mean_token_accuracy": 0.14783507138490676,
      "num_tokens": 10714307.0,
      "step": 5735
    },
    {
      "entropy": 6.068199825286865,
      "epoch": 0.5068880254327093,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004980231633463822,
      "loss": 5.9478,
      "mean_token_accuracy": 0.14717553332448005,
      "num_tokens": 10723513.0,
      "step": 5740
    },
    {
      "entropy": 6.131747579574585,
      "epoch": 0.5073295655245497,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004980189958595081,
      "loss": 6.1132,
      "mean_token_accuracy": 0.15128831416368485,
      "num_tokens": 10732809.0,
      "step": 5745
    },
    {
      "entropy": 6.229391098022461,
      "epoch": 0.5077711056163899,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004980148240038186,
      "loss": 6.1388,
      "mean_token_accuracy": 0.15277309194207192,
      "num_tokens": 10742960.0,
      "step": 5750
    },
    {
      "entropy": 6.311017084121704,
      "epoch": 0.5082126457082303,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004980106477793957,
      "loss": 6.2002,
      "mean_token_accuracy": 0.15205241590738297,
      "num_tokens": 10752835.0,
      "step": 5755
    },
    {
      "entropy": 6.192711448669433,
      "epoch": 0.5086541858000706,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004980064671863209,
      "loss": 6.0765,
      "mean_token_accuracy": 0.1547958582639694,
      "num_tokens": 10762139.0,
      "step": 5760
    },
    {
      "entropy": 6.211809253692627,
      "epoch": 0.509095725891911,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004980022822246763,
      "loss": 6.1501,
      "mean_token_accuracy": 0.14922941774129866,
      "num_tokens": 10771961.0,
      "step": 5765
    },
    {
      "entropy": 6.349077844619751,
      "epoch": 0.5095372659837514,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004979980928945439,
      "loss": 6.1847,
      "mean_token_accuracy": 0.1427362911403179,
      "num_tokens": 10780163.0,
      "step": 5770
    },
    {
      "entropy": 6.252031850814819,
      "epoch": 0.5099788060755917,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004979938991960056,
      "loss": 6.1567,
      "mean_token_accuracy": 0.1454293832182884,
      "num_tokens": 10788339.0,
      "step": 5775
    },
    {
      "entropy": 6.193761348724365,
      "epoch": 0.510420346167432,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004979897011291436,
      "loss": 6.1412,
      "mean_token_accuracy": 0.1439156413078308,
      "num_tokens": 10797315.0,
      "step": 5780
    },
    {
      "entropy": 6.285819005966187,
      "epoch": 0.5108618862592723,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004979854986940402,
      "loss": 6.0717,
      "mean_token_accuracy": 0.14682874381542205,
      "num_tokens": 10806864.0,
      "step": 5785
    },
    {
      "entropy": 6.234613418579102,
      "epoch": 0.5113034263511127,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004979812918907777,
      "loss": 6.0984,
      "mean_token_accuracy": 0.15000496208667755,
      "num_tokens": 10816417.0,
      "step": 5790
    },
    {
      "entropy": 6.24019980430603,
      "epoch": 0.511744966442953,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004979770807194385,
      "loss": 6.1482,
      "mean_token_accuracy": 0.1514828845858574,
      "num_tokens": 10825692.0,
      "step": 5795
    },
    {
      "entropy": 6.236111831665039,
      "epoch": 0.5121865065347934,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004979728651801051,
      "loss": 6.1403,
      "mean_token_accuracy": 0.1545632392168045,
      "num_tokens": 10834652.0,
      "step": 5800
    },
    {
      "entropy": 6.167957305908203,
      "epoch": 0.5126280466266337,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004979686452728602,
      "loss": 6.1464,
      "mean_token_accuracy": 0.14476050958037376,
      "num_tokens": 10844491.0,
      "step": 5805
    },
    {
      "entropy": 6.158088731765747,
      "epoch": 0.513069586718474,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004979644209977863,
      "loss": 6.0836,
      "mean_token_accuracy": 0.1580489918589592,
      "num_tokens": 10853948.0,
      "step": 5810
    },
    {
      "entropy": 6.193817520141602,
      "epoch": 0.5135111268103144,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004979601923549661,
      "loss": 6.0705,
      "mean_token_accuracy": 0.15333878844976426,
      "num_tokens": 10861745.0,
      "step": 5815
    },
    {
      "entropy": 6.259048175811768,
      "epoch": 0.5139526669021547,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004979559593444826,
      "loss": 6.1645,
      "mean_token_accuracy": 0.1452197715640068,
      "num_tokens": 10871239.0,
      "step": 5820
    },
    {
      "entropy": 6.205961465835571,
      "epoch": 0.5143942069939951,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004979517219664187,
      "loss": 6.0248,
      "mean_token_accuracy": 0.15704307109117507,
      "num_tokens": 10880883.0,
      "step": 5825
    },
    {
      "entropy": 6.200279664993286,
      "epoch": 0.5148357470858353,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004979474802208572,
      "loss": 6.0787,
      "mean_token_accuracy": 0.15447696894407273,
      "num_tokens": 10890002.0,
      "step": 5830
    },
    {
      "entropy": 6.135254621505737,
      "epoch": 0.5152772871776757,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004979432341078816,
      "loss": 6.0187,
      "mean_token_accuracy": 0.14993426129221915,
      "num_tokens": 10898299.0,
      "step": 5835
    },
    {
      "entropy": 6.215742015838623,
      "epoch": 0.5157188272695161,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004979389836275746,
      "loss": 6.2195,
      "mean_token_accuracy": 0.14315202385187148,
      "num_tokens": 10907918.0,
      "step": 5840
    },
    {
      "entropy": 6.2449322700500485,
      "epoch": 0.5161603673613564,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004979347287800198,
      "loss": 6.089,
      "mean_token_accuracy": 0.15006405860185623,
      "num_tokens": 10917318.0,
      "step": 5845
    },
    {
      "entropy": 6.182945251464844,
      "epoch": 0.5166019074531968,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004979304695653005,
      "loss": 6.0404,
      "mean_token_accuracy": 0.1500822015106678,
      "num_tokens": 10926338.0,
      "step": 5850
    },
    {
      "entropy": 6.164806938171386,
      "epoch": 0.5170434475450371,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004979262059835001,
      "loss": 6.0823,
      "mean_token_accuracy": 0.15123574808239937,
      "num_tokens": 10935244.0,
      "step": 5855
    },
    {
      "entropy": 6.243174409866333,
      "epoch": 0.5174849876368774,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004979219380347021,
      "loss": 6.0969,
      "mean_token_accuracy": 0.15276483595371246,
      "num_tokens": 10943764.0,
      "step": 5860
    },
    {
      "entropy": 6.290911149978638,
      "epoch": 0.5179265277287177,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004979176657189901,
      "loss": 6.0856,
      "mean_token_accuracy": 0.1533655822277069,
      "num_tokens": 10952580.0,
      "step": 5865
    },
    {
      "entropy": 6.23861780166626,
      "epoch": 0.5183680678205581,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004979133890364477,
      "loss": 6.1112,
      "mean_token_accuracy": 0.15561173781752585,
      "num_tokens": 10961745.0,
      "step": 5870
    },
    {
      "entropy": 6.081136846542359,
      "epoch": 0.5188096079123985,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000497909107987159,
      "loss": 6.0203,
      "mean_token_accuracy": 0.1529902383685112,
      "num_tokens": 10971018.0,
      "step": 5875
    },
    {
      "entropy": 6.202692270278931,
      "epoch": 0.5192511480042388,
      "grad_norm": 1.75,
      "learning_rate": 0.0004979048225712074,
      "loss": 6.0744,
      "mean_token_accuracy": 0.1534593366086483,
      "num_tokens": 10980070.0,
      "step": 5880
    },
    {
      "entropy": 6.22016716003418,
      "epoch": 0.5196926880960792,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004979005327886772,
      "loss": 6.0273,
      "mean_token_accuracy": 0.1515096217393875,
      "num_tokens": 10989090.0,
      "step": 5885
    },
    {
      "entropy": 6.23254246711731,
      "epoch": 0.5201342281879194,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004978962386396525,
      "loss": 6.0883,
      "mean_token_accuracy": 0.15314121246337892,
      "num_tokens": 10998652.0,
      "step": 5890
    },
    {
      "entropy": 6.148999738693237,
      "epoch": 0.5205757682797598,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004978919401242171,
      "loss": 5.9882,
      "mean_token_accuracy": 0.15903434306383132,
      "num_tokens": 11006323.0,
      "step": 5895
    },
    {
      "entropy": 6.012803792953491,
      "epoch": 0.5210173083716001,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004978876372424553,
      "loss": 5.9238,
      "mean_token_accuracy": 0.164156411588192,
      "num_tokens": 11015054.0,
      "step": 5900
    },
    {
      "entropy": 6.143123817443848,
      "epoch": 0.5214588484634405,
      "grad_norm": 5.1875,
      "learning_rate": 0.0004978833299944515,
      "loss": 6.1265,
      "mean_token_accuracy": 0.14407195448875426,
      "num_tokens": 11025787.0,
      "step": 5905
    },
    {
      "entropy": 6.360718059539795,
      "epoch": 0.5219003885552809,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004978790183802901,
      "loss": 6.2388,
      "mean_token_accuracy": 0.1436489887535572,
      "num_tokens": 11036091.0,
      "step": 5910
    },
    {
      "entropy": 6.328134918212891,
      "epoch": 0.5223419286471211,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004978747024000554,
      "loss": 6.1939,
      "mean_token_accuracy": 0.1392632395029068,
      "num_tokens": 11046442.0,
      "step": 5915
    },
    {
      "entropy": 6.268065500259399,
      "epoch": 0.5227834687389615,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004978703820538321,
      "loss": 6.2112,
      "mean_token_accuracy": 0.14451470226049423,
      "num_tokens": 11056907.0,
      "step": 5920
    },
    {
      "entropy": 6.211714887619019,
      "epoch": 0.5232250088308018,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004978660573417048,
      "loss": 6.0584,
      "mean_token_accuracy": 0.15640886425971984,
      "num_tokens": 11065298.0,
      "step": 5925
    },
    {
      "entropy": 6.202786159515381,
      "epoch": 0.5236665489226422,
      "grad_norm": 3.265625,
      "learning_rate": 0.000497861728263758,
      "loss": 6.0566,
      "mean_token_accuracy": 0.15214993953704833,
      "num_tokens": 11074509.0,
      "step": 5930
    },
    {
      "entropy": 6.189145088195801,
      "epoch": 0.5241080890144825,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004978573948200769,
      "loss": 6.0983,
      "mean_token_accuracy": 0.15263655185699462,
      "num_tokens": 11082433.0,
      "step": 5935
    },
    {
      "entropy": 6.122281789779663,
      "epoch": 0.5245496291063229,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004978530570107461,
      "loss": 6.0632,
      "mean_token_accuracy": 0.15261141806840897,
      "num_tokens": 11090838.0,
      "step": 5940
    },
    {
      "entropy": 6.231547832489014,
      "epoch": 0.5249911691981632,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004978487148358506,
      "loss": 6.0673,
      "mean_token_accuracy": 0.1546216756105423,
      "num_tokens": 11100048.0,
      "step": 5945
    },
    {
      "entropy": 6.150852632522583,
      "epoch": 0.5254327092900035,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004978443682954756,
      "loss": 6.0743,
      "mean_token_accuracy": 0.1524834528565407,
      "num_tokens": 11109871.0,
      "step": 5950
    },
    {
      "entropy": 6.17802734375,
      "epoch": 0.5258742493818439,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004978400173897061,
      "loss": 6.0653,
      "mean_token_accuracy": 0.14837429970502852,
      "num_tokens": 11119450.0,
      "step": 5955
    },
    {
      "entropy": 6.265403985977173,
      "epoch": 0.5263157894736842,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004978356621186275,
      "loss": 6.007,
      "mean_token_accuracy": 0.16124734580516814,
      "num_tokens": 11128321.0,
      "step": 5960
    },
    {
      "entropy": 6.202954912185669,
      "epoch": 0.5267573295655246,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004978313024823249,
      "loss": 6.0737,
      "mean_token_accuracy": 0.15226287469267846,
      "num_tokens": 11138657.0,
      "step": 5965
    },
    {
      "entropy": 6.148827409744262,
      "epoch": 0.5271988696573648,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004978269384808839,
      "loss": 6.0597,
      "mean_token_accuracy": 0.1582752376794815,
      "num_tokens": 11147704.0,
      "step": 5970
    },
    {
      "entropy": 6.209830904006958,
      "epoch": 0.5276404097492052,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004978225701143898,
      "loss": 6.1387,
      "mean_token_accuracy": 0.14596268758177758,
      "num_tokens": 11156979.0,
      "step": 5975
    },
    {
      "entropy": 6.260457229614258,
      "epoch": 0.5280819498410456,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004978181973829284,
      "loss": 6.0266,
      "mean_token_accuracy": 0.15336811989545823,
      "num_tokens": 11165539.0,
      "step": 5980
    },
    {
      "entropy": 6.3307692527771,
      "epoch": 0.5285234899328859,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004978138202865851,
      "loss": 6.1168,
      "mean_token_accuracy": 0.14098541140556337,
      "num_tokens": 11175911.0,
      "step": 5985
    },
    {
      "entropy": 6.204238653182983,
      "epoch": 0.5289650300247263,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004978094388254459,
      "loss": 6.1287,
      "mean_token_accuracy": 0.14792783409357071,
      "num_tokens": 11185365.0,
      "step": 5990
    },
    {
      "entropy": 6.221013116836548,
      "epoch": 0.5294065701165666,
      "grad_norm": 5.125,
      "learning_rate": 0.0004978050529995965,
      "loss": 6.0366,
      "mean_token_accuracy": 0.1510949432849884,
      "num_tokens": 11195318.0,
      "step": 5995
    },
    {
      "entropy": 6.264184522628784,
      "epoch": 0.5298481102084069,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004978006628091228,
      "loss": 6.1571,
      "mean_token_accuracy": 0.14806697890162468,
      "num_tokens": 11205246.0,
      "step": 6000
    },
    {
      "epoch": 0.5298481102084069,
      "eval_entropy": 6.074740511752741,
      "eval_loss": 6.123631954193115,
      "eval_mean_token_accuracy": 0.1567293423705229,
      "eval_num_tokens": 11205246.0,
      "eval_runtime": 26.285,
      "eval_samples_per_second": 1343.541,
      "eval_steps_per_second": 167.966,
      "step": 6000
    },
    {
      "entropy": 6.261495876312256,
      "epoch": 0.5302896503002472,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004977962682541108,
      "loss": 6.1245,
      "mean_token_accuracy": 0.14799828678369523,
      "num_tokens": 11215183.0,
      "step": 6005
    },
    {
      "entropy": 6.234362173080444,
      "epoch": 0.5307311903920876,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004977918693346467,
      "loss": 6.0681,
      "mean_token_accuracy": 0.14836506694555282,
      "num_tokens": 11224127.0,
      "step": 6010
    },
    {
      "entropy": 6.2427223205566404,
      "epoch": 0.531172730483928,
      "grad_norm": 2.375,
      "learning_rate": 0.0004977874660508165,
      "loss": 6.1051,
      "mean_token_accuracy": 0.15863558948040007,
      "num_tokens": 11233014.0,
      "step": 6015
    },
    {
      "entropy": 6.267404651641845,
      "epoch": 0.5316142705757683,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004977830584027065,
      "loss": 6.1116,
      "mean_token_accuracy": 0.15553241968154907,
      "num_tokens": 11242251.0,
      "step": 6020
    },
    {
      "entropy": 6.252231168746948,
      "epoch": 0.5320558106676087,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004977786463904033,
      "loss": 6.1432,
      "mean_token_accuracy": 0.14830528125166892,
      "num_tokens": 11251947.0,
      "step": 6025
    },
    {
      "entropy": 6.1382335186004635,
      "epoch": 0.5324973507594489,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004977742300139929,
      "loss": 6.1673,
      "mean_token_accuracy": 0.148697080463171,
      "num_tokens": 11261125.0,
      "step": 6030
    },
    {
      "entropy": 6.280923986434937,
      "epoch": 0.5329388908512893,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004977698092735621,
      "loss": 6.068,
      "mean_token_accuracy": 0.15308852344751359,
      "num_tokens": 11270334.0,
      "step": 6035
    },
    {
      "entropy": 6.235616731643677,
      "epoch": 0.5333804309431296,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004977653841691975,
      "loss": 6.06,
      "mean_token_accuracy": 0.15150801390409468,
      "num_tokens": 11280497.0,
      "step": 6040
    },
    {
      "entropy": 6.170658874511719,
      "epoch": 0.53382197103497,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004977609547009857,
      "loss": 5.9952,
      "mean_token_accuracy": 0.15689463764429093,
      "num_tokens": 11288636.0,
      "step": 6045
    },
    {
      "entropy": 6.238555002212524,
      "epoch": 0.5342635111268104,
      "grad_norm": 2.125,
      "learning_rate": 0.0004977565208690134,
      "loss": 6.1211,
      "mean_token_accuracy": 0.15063835978507994,
      "num_tokens": 11297519.0,
      "step": 6050
    },
    {
      "entropy": 6.268319177627563,
      "epoch": 0.5347050512186506,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004977520826733677,
      "loss": 6.0851,
      "mean_token_accuracy": 0.15677888840436935,
      "num_tokens": 11306280.0,
      "step": 6055
    },
    {
      "entropy": 6.24368577003479,
      "epoch": 0.535146591310491,
      "grad_norm": 2.375,
      "learning_rate": 0.0004977476401141353,
      "loss": 6.071,
      "mean_token_accuracy": 0.14682250916957856,
      "num_tokens": 11315366.0,
      "step": 6060
    },
    {
      "entropy": 6.185540246963501,
      "epoch": 0.5355881314023313,
      "grad_norm": 2.75,
      "learning_rate": 0.0004977431931914032,
      "loss": 6.0787,
      "mean_token_accuracy": 0.14786505848169326,
      "num_tokens": 11324664.0,
      "step": 6065
    },
    {
      "entropy": 6.199446153640747,
      "epoch": 0.5360296714941717,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004977387419052586,
      "loss": 6.0797,
      "mean_token_accuracy": 0.1473124697804451,
      "num_tokens": 11334118.0,
      "step": 6070
    },
    {
      "entropy": 6.347864866256714,
      "epoch": 0.536471211586012,
      "grad_norm": 3.03125,
      "learning_rate": 0.000497734286255789,
      "loss": 6.175,
      "mean_token_accuracy": 0.14238944202661513,
      "num_tokens": 11342684.0,
      "step": 6075
    },
    {
      "entropy": 6.262286758422851,
      "epoch": 0.5369127516778524,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004977298262430813,
      "loss": 6.1101,
      "mean_token_accuracy": 0.1559641495347023,
      "num_tokens": 11351298.0,
      "step": 6080
    },
    {
      "entropy": 6.177248239517212,
      "epoch": 0.5373542917696927,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004977253618672229,
      "loss": 6.048,
      "mean_token_accuracy": 0.15297479629516603,
      "num_tokens": 11360264.0,
      "step": 6085
    },
    {
      "entropy": 6.191631174087524,
      "epoch": 0.537795831861533,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004977208931283013,
      "loss": 6.0225,
      "mean_token_accuracy": 0.15780718624591827,
      "num_tokens": 11369815.0,
      "step": 6090
    },
    {
      "entropy": 6.261722660064697,
      "epoch": 0.5382373719533734,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004977164200264042,
      "loss": 6.1168,
      "mean_token_accuracy": 0.14988852441310882,
      "num_tokens": 11379131.0,
      "step": 6095
    },
    {
      "entropy": 6.200402641296387,
      "epoch": 0.5386789120452137,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004977119425616191,
      "loss": 5.9978,
      "mean_token_accuracy": 0.15366772264242173,
      "num_tokens": 11387692.0,
      "step": 6100
    },
    {
      "entropy": 6.184014987945557,
      "epoch": 0.5391204521370541,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004977074607340336,
      "loss": 6.0627,
      "mean_token_accuracy": 0.15367623716592788,
      "num_tokens": 11396278.0,
      "step": 6105
    },
    {
      "entropy": 6.098484373092651,
      "epoch": 0.5395619922288943,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004977029745437357,
      "loss": 6.0075,
      "mean_token_accuracy": 0.16174166202545165,
      "num_tokens": 11404790.0,
      "step": 6110
    },
    {
      "entropy": 6.2133594989776615,
      "epoch": 0.5400035323207347,
      "grad_norm": 2.21875,
      "learning_rate": 0.000497698483990813,
      "loss": 6.0825,
      "mean_token_accuracy": 0.14521845802664757,
      "num_tokens": 11414614.0,
      "step": 6115
    },
    {
      "entropy": 6.2395154476165775,
      "epoch": 0.5404450724125751,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004976939890753539,
      "loss": 6.1112,
      "mean_token_accuracy": 0.1490165799856186,
      "num_tokens": 11423696.0,
      "step": 6120
    },
    {
      "entropy": 6.1709897994995115,
      "epoch": 0.5408866125044154,
      "grad_norm": 2.90625,
      "learning_rate": 0.000497689489797446,
      "loss": 6.0606,
      "mean_token_accuracy": 0.1530643433332443,
      "num_tokens": 11432669.0,
      "step": 6125
    },
    {
      "entropy": 6.204414415359497,
      "epoch": 0.5413281525962558,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004976849861571778,
      "loss": 6.1017,
      "mean_token_accuracy": 0.15045553892850877,
      "num_tokens": 11442132.0,
      "step": 6130
    },
    {
      "entropy": 6.218031549453736,
      "epoch": 0.541769692688096,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004976804781546373,
      "loss": 6.0728,
      "mean_token_accuracy": 0.14818092808127403,
      "num_tokens": 11451821.0,
      "step": 6135
    },
    {
      "entropy": 6.208802270889282,
      "epoch": 0.5422112327799364,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004976759657899129,
      "loss": 6.0682,
      "mean_token_accuracy": 0.15299722403287888,
      "num_tokens": 11460242.0,
      "step": 6140
    },
    {
      "entropy": 6.111813926696778,
      "epoch": 0.5426527728717767,
      "grad_norm": 3.765625,
      "learning_rate": 0.000497671449063093,
      "loss": 6.08,
      "mean_token_accuracy": 0.1515306919813156,
      "num_tokens": 11469588.0,
      "step": 6145
    },
    {
      "entropy": 6.246411895751953,
      "epoch": 0.5430943129636171,
      "grad_norm": 2.265625,
      "learning_rate": 0.000497666927974266,
      "loss": 6.1593,
      "mean_token_accuracy": 0.14249540939927102,
      "num_tokens": 11480401.0,
      "step": 6150
    },
    {
      "entropy": 6.300544309616089,
      "epoch": 0.5435358530554575,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004976624025235205,
      "loss": 6.0559,
      "mean_token_accuracy": 0.14820022881031036,
      "num_tokens": 11489117.0,
      "step": 6155
    },
    {
      "entropy": 6.183306360244751,
      "epoch": 0.5439773931472978,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004976578727109453,
      "loss": 6.0201,
      "mean_token_accuracy": 0.15444311127066612,
      "num_tokens": 11497612.0,
      "step": 6160
    },
    {
      "entropy": 6.119126510620117,
      "epoch": 0.5444189332391381,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004976533385366289,
      "loss": 6.1239,
      "mean_token_accuracy": 0.14944978952407836,
      "num_tokens": 11507024.0,
      "step": 6165
    },
    {
      "entropy": 6.167967653274536,
      "epoch": 0.5448604733309784,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004976488000006602,
      "loss": 6.0558,
      "mean_token_accuracy": 0.15538153201341628,
      "num_tokens": 11516873.0,
      "step": 6170
    },
    {
      "entropy": 6.2573628425598145,
      "epoch": 0.5453020134228188,
      "grad_norm": 2.125,
      "learning_rate": 0.0004976442571031282,
      "loss": 6.1195,
      "mean_token_accuracy": 0.1560243234038353,
      "num_tokens": 11526128.0,
      "step": 6175
    },
    {
      "entropy": 6.223786067962647,
      "epoch": 0.5457435535146591,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004976397098441218,
      "loss": 6.1373,
      "mean_token_accuracy": 0.15402478873729705,
      "num_tokens": 11535030.0,
      "step": 6180
    },
    {
      "entropy": 6.111001682281494,
      "epoch": 0.5461850936064995,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004976351582237301,
      "loss": 6.0279,
      "mean_token_accuracy": 0.1489868089556694,
      "num_tokens": 11543113.0,
      "step": 6185
    },
    {
      "entropy": 6.244951248168945,
      "epoch": 0.5466266336983399,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004976306022420423,
      "loss": 6.0814,
      "mean_token_accuracy": 0.14897802099585533,
      "num_tokens": 11552744.0,
      "step": 6190
    },
    {
      "entropy": 6.2333760261535645,
      "epoch": 0.5470681737901801,
      "grad_norm": 2.125,
      "learning_rate": 0.0004976260418991476,
      "loss": 6.0815,
      "mean_token_accuracy": 0.15112528651952745,
      "num_tokens": 11561274.0,
      "step": 6195
    },
    {
      "entropy": 6.188093137741089,
      "epoch": 0.5475097138820205,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004976214771951354,
      "loss": 6.1228,
      "mean_token_accuracy": 0.15548378378152847,
      "num_tokens": 11570485.0,
      "step": 6200
    },
    {
      "entropy": 6.247988700866699,
      "epoch": 0.5479512539738608,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004976169081300949,
      "loss": 6.1145,
      "mean_token_accuracy": 0.14788752794265747,
      "num_tokens": 11579380.0,
      "step": 6205
    },
    {
      "entropy": 6.229219913482666,
      "epoch": 0.5483927940657012,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004976123347041159,
      "loss": 6.0947,
      "mean_token_accuracy": 0.1510343536734581,
      "num_tokens": 11588686.0,
      "step": 6210
    },
    {
      "entropy": 6.136067247390747,
      "epoch": 0.5488343341575415,
      "grad_norm": 7.28125,
      "learning_rate": 0.0004976077569172879,
      "loss": 6.0207,
      "mean_token_accuracy": 0.15724413692951203,
      "num_tokens": 11598547.0,
      "step": 6215
    },
    {
      "entropy": 6.204976606369018,
      "epoch": 0.5492758742493818,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004976031747697006,
      "loss": 6.0959,
      "mean_token_accuracy": 0.15085032731294631,
      "num_tokens": 11608407.0,
      "step": 6220
    },
    {
      "entropy": 6.1893705368042,
      "epoch": 0.5497174143412222,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004975985882614435,
      "loss": 6.0526,
      "mean_token_accuracy": 0.15429275631904601,
      "num_tokens": 11618641.0,
      "step": 6225
    },
    {
      "entropy": 6.2130521774292,
      "epoch": 0.5501589544330625,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004975939973926068,
      "loss": 6.1113,
      "mean_token_accuracy": 0.14406439512968064,
      "num_tokens": 11627082.0,
      "step": 6230
    },
    {
      "entropy": 6.180857801437378,
      "epoch": 0.5506004945249029,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004975894021632803,
      "loss": 6.0049,
      "mean_token_accuracy": 0.1561313599348068,
      "num_tokens": 11636045.0,
      "step": 6235
    },
    {
      "entropy": 6.171938514709472,
      "epoch": 0.5510420346167432,
      "grad_norm": 2.21875,
      "learning_rate": 0.000497584802573554,
      "loss": 6.09,
      "mean_token_accuracy": 0.1516232445836067,
      "num_tokens": 11646516.0,
      "step": 6240
    },
    {
      "entropy": 6.1169013500213625,
      "epoch": 0.5514835747085836,
      "grad_norm": 2.796875,
      "learning_rate": 0.000497580198623518,
      "loss": 6.0152,
      "mean_token_accuracy": 0.15931856781244277,
      "num_tokens": 11655512.0,
      "step": 6245
    },
    {
      "entropy": 6.179685831069946,
      "epoch": 0.5519251148004238,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004975755903132623,
      "loss": 6.1158,
      "mean_token_accuracy": 0.15324108451604843,
      "num_tokens": 11665641.0,
      "step": 6250
    },
    {
      "entropy": 6.242779970169067,
      "epoch": 0.5523666548922642,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004975709776428776,
      "loss": 6.0864,
      "mean_token_accuracy": 0.14270863980054854,
      "num_tokens": 11674585.0,
      "step": 6255
    },
    {
      "entropy": 6.1917078495025635,
      "epoch": 0.5528081949841046,
      "grad_norm": 2.5,
      "learning_rate": 0.000497566360612454,
      "loss": 5.9814,
      "mean_token_accuracy": 0.16099566966295242,
      "num_tokens": 11683498.0,
      "step": 6260
    },
    {
      "entropy": 6.150371551513672,
      "epoch": 0.5532497350759449,
      "grad_norm": 2.40625,
      "learning_rate": 0.000497561739222082,
      "loss": 6.0133,
      "mean_token_accuracy": 0.15729362815618514,
      "num_tokens": 11692663.0,
      "step": 6265
    },
    {
      "entropy": 6.237178993225098,
      "epoch": 0.5536912751677853,
      "grad_norm": 2.125,
      "learning_rate": 0.0004975571134718521,
      "loss": 6.1624,
      "mean_token_accuracy": 0.14940599352121353,
      "num_tokens": 11701906.0,
      "step": 6270
    },
    {
      "entropy": 6.188458633422852,
      "epoch": 0.5541328152596255,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004975524833618548,
      "loss": 6.0612,
      "mean_token_accuracy": 0.15533912926912308,
      "num_tokens": 11710932.0,
      "step": 6275
    },
    {
      "entropy": 6.136988687515259,
      "epoch": 0.5545743553514659,
      "grad_norm": 2.4375,
      "learning_rate": 0.000497547848892181,
      "loss": 6.0925,
      "mean_token_accuracy": 0.15054061263799667,
      "num_tokens": 11719953.0,
      "step": 6280
    },
    {
      "entropy": 6.216779804229736,
      "epoch": 0.5550158954433062,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004975432100629215,
      "loss": 5.9751,
      "mean_token_accuracy": 0.15986845195293425,
      "num_tokens": 11728706.0,
      "step": 6285
    },
    {
      "entropy": 6.262577915191651,
      "epoch": 0.5554574355351466,
      "grad_norm": 2.375,
      "learning_rate": 0.000497538566874167,
      "loss": 6.1512,
      "mean_token_accuracy": 0.14726722687482835,
      "num_tokens": 11738537.0,
      "step": 6290
    },
    {
      "entropy": 6.1831403255462645,
      "epoch": 0.555898975626987,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004975339193260084,
      "loss": 5.9988,
      "mean_token_accuracy": 0.15619935542345048,
      "num_tokens": 11747503.0,
      "step": 6295
    },
    {
      "entropy": 6.209629535675049,
      "epoch": 0.5563405157188273,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004975292674185371,
      "loss": 6.0719,
      "mean_token_accuracy": 0.15479969680309297,
      "num_tokens": 11756501.0,
      "step": 6300
    },
    {
      "entropy": 6.174160957336426,
      "epoch": 0.5567820558106676,
      "grad_norm": 1.75,
      "learning_rate": 0.0004975246111518439,
      "loss": 6.0483,
      "mean_token_accuracy": 0.1491239570081234,
      "num_tokens": 11765483.0,
      "step": 6305
    },
    {
      "entropy": 6.230826044082642,
      "epoch": 0.5572235959025079,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004975199505260202,
      "loss": 6.1862,
      "mean_token_accuracy": 0.14385148659348487,
      "num_tokens": 11774941.0,
      "step": 6310
    },
    {
      "entropy": 6.284946060180664,
      "epoch": 0.5576651359943483,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004975152855411571,
      "loss": 6.1552,
      "mean_token_accuracy": 0.1457800880074501,
      "num_tokens": 11785080.0,
      "step": 6315
    },
    {
      "entropy": 6.319910860061645,
      "epoch": 0.5581066760861886,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004975106161973462,
      "loss": 6.0255,
      "mean_token_accuracy": 0.15467681735754013,
      "num_tokens": 11793933.0,
      "step": 6320
    },
    {
      "entropy": 6.182535457611084,
      "epoch": 0.558548216178029,
      "grad_norm": 2.125,
      "learning_rate": 0.000497505942494679,
      "loss": 6.0675,
      "mean_token_accuracy": 0.14724954590201378,
      "num_tokens": 11803592.0,
      "step": 6325
    },
    {
      "entropy": 6.128745031356812,
      "epoch": 0.5589897562698694,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004975012644332467,
      "loss": 5.9718,
      "mean_token_accuracy": 0.15875890329480172,
      "num_tokens": 11812397.0,
      "step": 6330
    },
    {
      "entropy": 6.2402550220489506,
      "epoch": 0.5594312963617096,
      "grad_norm": 2.375,
      "learning_rate": 0.0004974965820131413,
      "loss": 6.1241,
      "mean_token_accuracy": 0.1507368728518486,
      "num_tokens": 11821734.0,
      "step": 6335
    },
    {
      "entropy": 6.170193338394165,
      "epoch": 0.55987283645355,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004974918952344544,
      "loss": 6.0176,
      "mean_token_accuracy": 0.1553915947675705,
      "num_tokens": 11832107.0,
      "step": 6340
    },
    {
      "entropy": 6.16438479423523,
      "epoch": 0.5603143765453903,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004974872040972777,
      "loss": 6.0291,
      "mean_token_accuracy": 0.15436061918735505,
      "num_tokens": 11841983.0,
      "step": 6345
    },
    {
      "entropy": 6.111740016937256,
      "epoch": 0.5607559166372307,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004974825086017033,
      "loss": 6.0285,
      "mean_token_accuracy": 0.15739574581384658,
      "num_tokens": 11851836.0,
      "step": 6350
    },
    {
      "entropy": 6.1818804264068605,
      "epoch": 0.561197456729071,
      "grad_norm": 2.78125,
      "learning_rate": 0.000497477808747823,
      "loss": 6.0737,
      "mean_token_accuracy": 0.15442853420972824,
      "num_tokens": 11861109.0,
      "step": 6355
    },
    {
      "entropy": 6.214493465423584,
      "epoch": 0.5616389968209113,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004974731045357292,
      "loss": 6.1666,
      "mean_token_accuracy": 0.1480507642030716,
      "num_tokens": 11871015.0,
      "step": 6360
    },
    {
      "entropy": 6.224303102493286,
      "epoch": 0.5620805369127517,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004974683959655136,
      "loss": 6.0559,
      "mean_token_accuracy": 0.15506404340267183,
      "num_tokens": 11881306.0,
      "step": 6365
    },
    {
      "entropy": 6.2358441829681395,
      "epoch": 0.562522077004592,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004974636830372686,
      "loss": 5.9684,
      "mean_token_accuracy": 0.1575288876891136,
      "num_tokens": 11890251.0,
      "step": 6370
    },
    {
      "entropy": 6.170713281631469,
      "epoch": 0.5629636170964324,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004974589657510866,
      "loss": 6.0966,
      "mean_token_accuracy": 0.1509570896625519,
      "num_tokens": 11899662.0,
      "step": 6375
    },
    {
      "entropy": 6.143237447738647,
      "epoch": 0.5634051571882727,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004974542441070601,
      "loss": 6.0628,
      "mean_token_accuracy": 0.15524649620056152,
      "num_tokens": 11908699.0,
      "step": 6380
    },
    {
      "entropy": 6.159485769271851,
      "epoch": 0.563846697280113,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004974495181052814,
      "loss": 6.0416,
      "mean_token_accuracy": 0.15841482132673262,
      "num_tokens": 11916637.0,
      "step": 6385
    },
    {
      "entropy": 6.245145463943482,
      "epoch": 0.5642882373719533,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004974447877458433,
      "loss": 6.1138,
      "mean_token_accuracy": 0.15061435103416443,
      "num_tokens": 11926063.0,
      "step": 6390
    },
    {
      "entropy": 6.221379137039184,
      "epoch": 0.5647297774637937,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004974400530288381,
      "loss": 5.9968,
      "mean_token_accuracy": 0.15577640980482102,
      "num_tokens": 11934861.0,
      "step": 6395
    },
    {
      "entropy": 6.233156442642212,
      "epoch": 0.5651713175556341,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004974353139543589,
      "loss": 6.0984,
      "mean_token_accuracy": 0.15241503417491914,
      "num_tokens": 11944403.0,
      "step": 6400
    },
    {
      "entropy": 6.242249774932861,
      "epoch": 0.5656128576474744,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004974305705224983,
      "loss": 6.192,
      "mean_token_accuracy": 0.15282924622297286,
      "num_tokens": 11953306.0,
      "step": 6405
    },
    {
      "entropy": 6.240070533752442,
      "epoch": 0.5660543977393148,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004974258227333493,
      "loss": 6.0747,
      "mean_token_accuracy": 0.1487444445490837,
      "num_tokens": 11962727.0,
      "step": 6410
    },
    {
      "entropy": 6.250161600112915,
      "epoch": 0.566495937831155,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004974210705870052,
      "loss": 6.2021,
      "mean_token_accuracy": 0.14626676738262176,
      "num_tokens": 11972375.0,
      "step": 6415
    },
    {
      "entropy": 6.143079137802124,
      "epoch": 0.5669374779229954,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004974163140835586,
      "loss": 5.9903,
      "mean_token_accuracy": 0.15524720996618271,
      "num_tokens": 11981989.0,
      "step": 6420
    },
    {
      "entropy": 6.185615587234497,
      "epoch": 0.5673790180148357,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004974115532231029,
      "loss": 6.0347,
      "mean_token_accuracy": 0.1525355249643326,
      "num_tokens": 11990865.0,
      "step": 6425
    },
    {
      "entropy": 6.183894205093384,
      "epoch": 0.5678205581066761,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004974067880057315,
      "loss": 6.0403,
      "mean_token_accuracy": 0.15843118131160736,
      "num_tokens": 12000697.0,
      "step": 6430
    },
    {
      "entropy": 6.191749143600464,
      "epoch": 0.5682620981985165,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004974020184315375,
      "loss": 6.112,
      "mean_token_accuracy": 0.15010613352060317,
      "num_tokens": 12010560.0,
      "step": 6435
    },
    {
      "entropy": 6.25274658203125,
      "epoch": 0.5687036382903567,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004973972445006144,
      "loss": 6.1426,
      "mean_token_accuracy": 0.15021998137235643,
      "num_tokens": 12020760.0,
      "step": 6440
    },
    {
      "entropy": 6.1001549243927,
      "epoch": 0.5691451783821971,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004973924662130559,
      "loss": 5.9806,
      "mean_token_accuracy": 0.15531480312347412,
      "num_tokens": 12030467.0,
      "step": 6445
    },
    {
      "entropy": 6.1398577213287355,
      "epoch": 0.5695867184740374,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004973876835689554,
      "loss": 6.0788,
      "mean_token_accuracy": 0.1474594548344612,
      "num_tokens": 12039897.0,
      "step": 6450
    },
    {
      "entropy": 6.215720844268799,
      "epoch": 0.5700282585658778,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004973828965684066,
      "loss": 6.102,
      "mean_token_accuracy": 0.14926670640707015,
      "num_tokens": 12048997.0,
      "step": 6455
    },
    {
      "entropy": 6.244099140167236,
      "epoch": 0.5704697986577181,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004973781052115035,
      "loss": 6.0028,
      "mean_token_accuracy": 0.15570057332515716,
      "num_tokens": 12057493.0,
      "step": 6460
    },
    {
      "entropy": 6.127003002166748,
      "epoch": 0.5709113387495585,
      "grad_norm": 2.375,
      "learning_rate": 0.0004973733094983396,
      "loss": 6.0991,
      "mean_token_accuracy": 0.1567182257771492,
      "num_tokens": 12066329.0,
      "step": 6465
    },
    {
      "entropy": 6.141180372238159,
      "epoch": 0.5713528788413988,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004973685094290092,
      "loss": 6.102,
      "mean_token_accuracy": 0.15247246474027634,
      "num_tokens": 12076022.0,
      "step": 6470
    },
    {
      "entropy": 6.237234592437744,
      "epoch": 0.5717944189332391,
      "grad_norm": 1.90625,
      "learning_rate": 0.000497363705003606,
      "loss": 6.0739,
      "mean_token_accuracy": 0.14747185185551642,
      "num_tokens": 12085750.0,
      "step": 6475
    },
    {
      "entropy": 6.275226354598999,
      "epoch": 0.5722359590250795,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004973588962222244,
      "loss": 6.0831,
      "mean_token_accuracy": 0.1538494572043419,
      "num_tokens": 12094935.0,
      "step": 6480
    },
    {
      "entropy": 6.150537633895874,
      "epoch": 0.5726774991169198,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004973540830849585,
      "loss": 6.022,
      "mean_token_accuracy": 0.15123365074396133,
      "num_tokens": 12104970.0,
      "step": 6485
    },
    {
      "entropy": 6.1925688743591305,
      "epoch": 0.5731190392087602,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004973492655919025,
      "loss": 6.1351,
      "mean_token_accuracy": 0.14966658353805543,
      "num_tokens": 12116195.0,
      "step": 6490
    },
    {
      "entropy": 6.265226793289185,
      "epoch": 0.5735605793006004,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004973444437431508,
      "loss": 6.0662,
      "mean_token_accuracy": 0.15317095220088958,
      "num_tokens": 12125485.0,
      "step": 6495
    },
    {
      "entropy": 6.144006729125977,
      "epoch": 0.5740021193924408,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004973396175387981,
      "loss": 6.0788,
      "mean_token_accuracy": 0.15357084721326827,
      "num_tokens": 12135474.0,
      "step": 6500
    },
    {
      "entropy": 6.165819597244263,
      "epoch": 0.5744436594842812,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004973347869789385,
      "loss": 6.1009,
      "mean_token_accuracy": 0.1501133732497692,
      "num_tokens": 12144945.0,
      "step": 6505
    },
    {
      "entropy": 6.256242799758911,
      "epoch": 0.5748851995761215,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004973299520636671,
      "loss": 6.0569,
      "mean_token_accuracy": 0.15410331934690474,
      "num_tokens": 12154893.0,
      "step": 6510
    },
    {
      "entropy": 6.225384521484375,
      "epoch": 0.5753267396679619,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004973251127930782,
      "loss": 6.1315,
      "mean_token_accuracy": 0.14763701856136321,
      "num_tokens": 12164047.0,
      "step": 6515
    },
    {
      "entropy": 6.24685001373291,
      "epoch": 0.5757682797598022,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004973202691672669,
      "loss": 6.1279,
      "mean_token_accuracy": 0.1465580239892006,
      "num_tokens": 12172965.0,
      "step": 6520
    },
    {
      "entropy": 6.203349494934082,
      "epoch": 0.5762098198516425,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004973154211863279,
      "loss": 6.1462,
      "mean_token_accuracy": 0.1492430865764618,
      "num_tokens": 12183534.0,
      "step": 6525
    },
    {
      "entropy": 6.190635442733765,
      "epoch": 0.5766513599434828,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004973105688503563,
      "loss": 6.0962,
      "mean_token_accuracy": 0.15006413906812668,
      "num_tokens": 12194370.0,
      "step": 6530
    },
    {
      "entropy": 6.201846075057984,
      "epoch": 0.5770929000353232,
      "grad_norm": 1.875,
      "learning_rate": 0.000497305712159447,
      "loss": 6.0899,
      "mean_token_accuracy": 0.15036289542913436,
      "num_tokens": 12203765.0,
      "step": 6535
    },
    {
      "entropy": 6.2407386779785154,
      "epoch": 0.5775344401271636,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004973008511136954,
      "loss": 6.0436,
      "mean_token_accuracy": 0.15642048344016074,
      "num_tokens": 12212889.0,
      "step": 6540
    },
    {
      "entropy": 6.228720664978027,
      "epoch": 0.5779759802190039,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004972959857131963,
      "loss": 6.0991,
      "mean_token_accuracy": 0.1515391692519188,
      "num_tokens": 12222055.0,
      "step": 6545
    },
    {
      "entropy": 6.237844610214234,
      "epoch": 0.5784175203108443,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004972911159580454,
      "loss": 6.1281,
      "mean_token_accuracy": 0.15056926906108856,
      "num_tokens": 12232092.0,
      "step": 6550
    },
    {
      "entropy": 6.21602201461792,
      "epoch": 0.5788590604026845,
      "grad_norm": 1.640625,
      "learning_rate": 0.000497286241848338,
      "loss": 6.1419,
      "mean_token_accuracy": 0.14570123925805092,
      "num_tokens": 12241411.0,
      "step": 6555
    },
    {
      "entropy": 6.213965797424317,
      "epoch": 0.5793006004945249,
      "grad_norm": 2.125,
      "learning_rate": 0.0004972813633841696,
      "loss": 6.0186,
      "mean_token_accuracy": 0.15196094810962676,
      "num_tokens": 12249735.0,
      "step": 6560
    },
    {
      "entropy": 6.210372972488403,
      "epoch": 0.5797421405863652,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004972764805656355,
      "loss": 6.0338,
      "mean_token_accuracy": 0.15679688304662703,
      "num_tokens": 12259132.0,
      "step": 6565
    },
    {
      "entropy": 6.1579570293426515,
      "epoch": 0.5801836806782056,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004972715933928318,
      "loss": 6.1017,
      "mean_token_accuracy": 0.1559670940041542,
      "num_tokens": 12269401.0,
      "step": 6570
    },
    {
      "entropy": 6.211545133590699,
      "epoch": 0.580625220770046,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004972667018658539,
      "loss": 6.1179,
      "mean_token_accuracy": 0.14414386153221131,
      "num_tokens": 12279163.0,
      "step": 6575
    },
    {
      "entropy": 6.173906469345093,
      "epoch": 0.5810667608618862,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004972618059847976,
      "loss": 5.9978,
      "mean_token_accuracy": 0.15999294966459274,
      "num_tokens": 12287942.0,
      "step": 6580
    },
    {
      "entropy": 6.2115003108978275,
      "epoch": 0.5815083009537266,
      "grad_norm": 2.375,
      "learning_rate": 0.0004972569057497591,
      "loss": 6.0068,
      "mean_token_accuracy": 0.15378794223070144,
      "num_tokens": 12296717.0,
      "step": 6585
    },
    {
      "entropy": 6.207294654846192,
      "epoch": 0.5819498410455669,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004972520011608341,
      "loss": 5.9907,
      "mean_token_accuracy": 0.16442489475011826,
      "num_tokens": 12305983.0,
      "step": 6590
    },
    {
      "entropy": 6.11067795753479,
      "epoch": 0.5823913811374073,
      "grad_norm": 3.625,
      "learning_rate": 0.0004972470922181189,
      "loss": 6.064,
      "mean_token_accuracy": 0.1454851284623146,
      "num_tokens": 12315338.0,
      "step": 6595
    },
    {
      "entropy": 6.159762525558472,
      "epoch": 0.5828329212292476,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004972421789217094,
      "loss": 6.1611,
      "mean_token_accuracy": 0.14217343926429749,
      "num_tokens": 12324828.0,
      "step": 6600
    },
    {
      "entropy": 6.256775808334351,
      "epoch": 0.583274461321088,
      "grad_norm": 3.625,
      "learning_rate": 0.0004972372612717022,
      "loss": 6.025,
      "mean_token_accuracy": 0.1642207235097885,
      "num_tokens": 12332844.0,
      "step": 6605
    },
    {
      "entropy": 6.231286430358887,
      "epoch": 0.5837160014129283,
      "grad_norm": 4.9375,
      "learning_rate": 0.0004972323392681934,
      "loss": 6.0346,
      "mean_token_accuracy": 0.15260550677776336,
      "num_tokens": 12342062.0,
      "step": 6610
    },
    {
      "entropy": 6.233716201782227,
      "epoch": 0.5841575415047686,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004972274129112794,
      "loss": 6.128,
      "mean_token_accuracy": 0.14080557674169542,
      "num_tokens": 12352071.0,
      "step": 6615
    },
    {
      "entropy": 6.3159643650054935,
      "epoch": 0.584599081596609,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004972224822010567,
      "loss": 6.2045,
      "mean_token_accuracy": 0.14211712777614594,
      "num_tokens": 12362025.0,
      "step": 6620
    },
    {
      "entropy": 6.222633790969849,
      "epoch": 0.5850406216884493,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004972175471376222,
      "loss": 6.2029,
      "mean_token_accuracy": 0.1442934051156044,
      "num_tokens": 12371547.0,
      "step": 6625
    },
    {
      "entropy": 6.262058734893799,
      "epoch": 0.5854821617802897,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004972126077210722,
      "loss": 6.0817,
      "mean_token_accuracy": 0.15001257061958312,
      "num_tokens": 12381385.0,
      "step": 6630
    },
    {
      "entropy": 6.322567796707153,
      "epoch": 0.5859237018721299,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004972076639515035,
      "loss": 6.1214,
      "mean_token_accuracy": 0.15077711939811705,
      "num_tokens": 12391302.0,
      "step": 6635
    },
    {
      "entropy": 6.136204528808594,
      "epoch": 0.5863652419639703,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004972027158290131,
      "loss": 6.038,
      "mean_token_accuracy": 0.15844113975763321,
      "num_tokens": 12400388.0,
      "step": 6640
    },
    {
      "entropy": 6.101259660720825,
      "epoch": 0.5868067820558107,
      "grad_norm": 2.5625,
      "learning_rate": 0.000497197763353698,
      "loss": 6.1047,
      "mean_token_accuracy": 0.1440892532467842,
      "num_tokens": 12409695.0,
      "step": 6645
    },
    {
      "entropy": 6.210428762435913,
      "epoch": 0.587248322147651,
      "grad_norm": 2.5,
      "learning_rate": 0.000497192806525655,
      "loss": 5.956,
      "mean_token_accuracy": 0.1615053281188011,
      "num_tokens": 12418728.0,
      "step": 6650
    },
    {
      "entropy": 6.184083366394043,
      "epoch": 0.5876898622394914,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004971878453449811,
      "loss": 6.1403,
      "mean_token_accuracy": 0.15096160173416137,
      "num_tokens": 12428072.0,
      "step": 6655
    },
    {
      "entropy": 6.221541213989258,
      "epoch": 0.5881314023313317,
      "grad_norm": 2.421875,
      "learning_rate": 0.000497182879811774,
      "loss": 6.0572,
      "mean_token_accuracy": 0.159123557806015,
      "num_tokens": 12436624.0,
      "step": 6660
    },
    {
      "entropy": 6.16471152305603,
      "epoch": 0.588572942423172,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004971779099261304,
      "loss": 6.0573,
      "mean_token_accuracy": 0.14437438026070595,
      "num_tokens": 12446370.0,
      "step": 6665
    },
    {
      "entropy": 6.233976030349732,
      "epoch": 0.5890144825150123,
      "grad_norm": 2.6875,
      "learning_rate": 0.000497172935688148,
      "loss": 6.114,
      "mean_token_accuracy": 0.15265168845653534,
      "num_tokens": 12456705.0,
      "step": 6670
    },
    {
      "entropy": 6.233513212203979,
      "epoch": 0.5894560226068527,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004971679570979242,
      "loss": 6.1202,
      "mean_token_accuracy": 0.15143783241510392,
      "num_tokens": 12466150.0,
      "step": 6675
    },
    {
      "entropy": 6.15090160369873,
      "epoch": 0.5898975626986931,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004971629741555565,
      "loss": 5.9865,
      "mean_token_accuracy": 0.1553258329629898,
      "num_tokens": 12475051.0,
      "step": 6680
    },
    {
      "entropy": 6.15589075088501,
      "epoch": 0.5903391027905334,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004971579868611423,
      "loss": 6.0754,
      "mean_token_accuracy": 0.1528732568025589,
      "num_tokens": 12485427.0,
      "step": 6685
    },
    {
      "entropy": 6.150468969345093,
      "epoch": 0.5907806428823738,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004971529952147797,
      "loss": 6.0171,
      "mean_token_accuracy": 0.15561191588640214,
      "num_tokens": 12494489.0,
      "step": 6690
    },
    {
      "entropy": 6.251090431213379,
      "epoch": 0.591222182974214,
      "grad_norm": 2.359375,
      "learning_rate": 0.000497147999216566,
      "loss": 6.0594,
      "mean_token_accuracy": 0.15239985883235932,
      "num_tokens": 12503656.0,
      "step": 6695
    },
    {
      "entropy": 6.278953886032104,
      "epoch": 0.5916637230660544,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004971429988665996,
      "loss": 6.0904,
      "mean_token_accuracy": 0.14635108262300492,
      "num_tokens": 12512940.0,
      "step": 6700
    },
    {
      "entropy": 6.216693925857544,
      "epoch": 0.5921052631578947,
      "grad_norm": 2.078125,
      "learning_rate": 0.000497137994164978,
      "loss": 6.049,
      "mean_token_accuracy": 0.15235408991575242,
      "num_tokens": 12523186.0,
      "step": 6705
    },
    {
      "entropy": 6.208224725723267,
      "epoch": 0.5925468032497351,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004971329851117996,
      "loss": 6.0792,
      "mean_token_accuracy": 0.14888448268175125,
      "num_tokens": 12533018.0,
      "step": 6710
    },
    {
      "entropy": 6.291949510574341,
      "epoch": 0.5929883433415755,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004971279717071622,
      "loss": 6.1424,
      "mean_token_accuracy": 0.15083410516381263,
      "num_tokens": 12542889.0,
      "step": 6715
    },
    {
      "entropy": 6.280454683303833,
      "epoch": 0.5934298834334157,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004971229539511643,
      "loss": 6.1197,
      "mean_token_accuracy": 0.14412906095385553,
      "num_tokens": 12553625.0,
      "step": 6720
    },
    {
      "entropy": 6.154693746566773,
      "epoch": 0.5938714235252561,
      "grad_norm": 2.46875,
      "learning_rate": 0.000497117931843904,
      "loss": 6.0614,
      "mean_token_accuracy": 0.1494505375623703,
      "num_tokens": 12563012.0,
      "step": 6725
    },
    {
      "entropy": 6.299040460586548,
      "epoch": 0.5943129636170964,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004971129053854797,
      "loss": 6.19,
      "mean_token_accuracy": 0.14032588005065919,
      "num_tokens": 12572892.0,
      "step": 6730
    },
    {
      "entropy": 6.226144170761108,
      "epoch": 0.5947545037089368,
      "grad_norm": 2.25,
      "learning_rate": 0.00049710787457599,
      "loss": 6.005,
      "mean_token_accuracy": 0.1601989150047302,
      "num_tokens": 12581550.0,
      "step": 6735
    },
    {
      "entropy": 6.205187177658081,
      "epoch": 0.5951960438007771,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004971028394155332,
      "loss": 6.1132,
      "mean_token_accuracy": 0.15160066187381743,
      "num_tokens": 12590599.0,
      "step": 6740
    },
    {
      "entropy": 6.233553647994995,
      "epoch": 0.5956375838926175,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004970977999042082,
      "loss": 6.0533,
      "mean_token_accuracy": 0.15343946889042853,
      "num_tokens": 12599497.0,
      "step": 6745
    },
    {
      "entropy": 6.162289619445801,
      "epoch": 0.5960791239844578,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004970927560421137,
      "loss": 6.1035,
      "mean_token_accuracy": 0.1528333231806755,
      "num_tokens": 12608273.0,
      "step": 6750
    },
    {
      "entropy": 6.151650285720825,
      "epoch": 0.5965206640762981,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004970877078293483,
      "loss": 5.9908,
      "mean_token_accuracy": 0.1540530242025852,
      "num_tokens": 12617287.0,
      "step": 6755
    },
    {
      "entropy": 6.2019072532653805,
      "epoch": 0.5969622041681385,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004970826552660111,
      "loss": 6.0679,
      "mean_token_accuracy": 0.15437413454055787,
      "num_tokens": 12626874.0,
      "step": 6760
    },
    {
      "entropy": 6.175334405899048,
      "epoch": 0.5974037442599788,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004970775983522009,
      "loss": 6.1008,
      "mean_token_accuracy": 0.15122495591640472,
      "num_tokens": 12635218.0,
      "step": 6765
    },
    {
      "entropy": 6.212725591659546,
      "epoch": 0.5978452843518192,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004970725370880168,
      "loss": 6.0673,
      "mean_token_accuracy": 0.15267443880438805,
      "num_tokens": 12644395.0,
      "step": 6770
    },
    {
      "entropy": 6.247377157211304,
      "epoch": 0.5982868244436594,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004970674714735581,
      "loss": 6.1117,
      "mean_token_accuracy": 0.15235370993614197,
      "num_tokens": 12654540.0,
      "step": 6775
    },
    {
      "entropy": 6.217602729797363,
      "epoch": 0.5987283645354998,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004970624015089239,
      "loss": 5.965,
      "mean_token_accuracy": 0.1598532095551491,
      "num_tokens": 12663546.0,
      "step": 6780
    },
    {
      "entropy": 6.069162797927857,
      "epoch": 0.5991699046273402,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004970573271942135,
      "loss": 6.0505,
      "mean_token_accuracy": 0.15797914415597916,
      "num_tokens": 12672353.0,
      "step": 6785
    },
    {
      "entropy": 6.119115257263184,
      "epoch": 0.5996114447191805,
      "grad_norm": 2.625,
      "learning_rate": 0.0004970522485295265,
      "loss": 5.9604,
      "mean_token_accuracy": 0.1626288414001465,
      "num_tokens": 12680833.0,
      "step": 6790
    },
    {
      "entropy": 6.2099771976470945,
      "epoch": 0.6000529848110209,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004970471655149621,
      "loss": 5.931,
      "mean_token_accuracy": 0.15862324982881545,
      "num_tokens": 12690033.0,
      "step": 6795
    },
    {
      "entropy": 6.035942649841308,
      "epoch": 0.6004945249028611,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00049704207815062,
      "loss": 6.0148,
      "mean_token_accuracy": 0.16100204288959502,
      "num_tokens": 12700127.0,
      "step": 6800
    },
    {
      "entropy": 6.246385383605957,
      "epoch": 0.6009360649947015,
      "grad_norm": 2.375,
      "learning_rate": 0.0004970369864366,
      "loss": 6.0568,
      "mean_token_accuracy": 0.1493418201804161,
      "num_tokens": 12709803.0,
      "step": 6805
    },
    {
      "entropy": 6.23317322731018,
      "epoch": 0.6013776050865418,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004970318903730017,
      "loss": 6.0184,
      "mean_token_accuracy": 0.15744948983192444,
      "num_tokens": 12718876.0,
      "step": 6810
    },
    {
      "entropy": 6.226846408843994,
      "epoch": 0.6018191451783822,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004970267899599249,
      "loss": 6.1384,
      "mean_token_accuracy": 0.15181976407766343,
      "num_tokens": 12728745.0,
      "step": 6815
    },
    {
      "entropy": 6.207247638702393,
      "epoch": 0.6022606852702226,
      "grad_norm": 2.25,
      "learning_rate": 0.0004970216851974695,
      "loss": 6.1304,
      "mean_token_accuracy": 0.1460263289511204,
      "num_tokens": 12738764.0,
      "step": 6820
    },
    {
      "entropy": 6.148588848114014,
      "epoch": 0.6027022253620629,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004970165760857356,
      "loss": 6.0685,
      "mean_token_accuracy": 0.14464377611875534,
      "num_tokens": 12747888.0,
      "step": 6825
    },
    {
      "entropy": 6.315575790405274,
      "epoch": 0.6031437654539032,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004970114626248232,
      "loss": 6.0977,
      "mean_token_accuracy": 0.14810532927513123,
      "num_tokens": 12757763.0,
      "step": 6830
    },
    {
      "entropy": 6.293575620651245,
      "epoch": 0.6035853055457435,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004970063448148326,
      "loss": 6.1372,
      "mean_token_accuracy": 0.14629552736878396,
      "num_tokens": 12767924.0,
      "step": 6835
    },
    {
      "entropy": 6.260718154907226,
      "epoch": 0.6040268456375839,
      "grad_norm": 5.90625,
      "learning_rate": 0.0004970012226558639,
      "loss": 6.1513,
      "mean_token_accuracy": 0.14904167130589485,
      "num_tokens": 12777140.0,
      "step": 6840
    },
    {
      "entropy": 6.217737531661987,
      "epoch": 0.6044683857294242,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004969960961480176,
      "loss": 6.1159,
      "mean_token_accuracy": 0.15016963630914687,
      "num_tokens": 12786731.0,
      "step": 6845
    },
    {
      "entropy": 6.2195882320404055,
      "epoch": 0.6049099258212646,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004969909652913939,
      "loss": 6.0566,
      "mean_token_accuracy": 0.15309870839118958,
      "num_tokens": 12796839.0,
      "step": 6850
    },
    {
      "entropy": 6.253087329864502,
      "epoch": 0.605351465913105,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004969858300860935,
      "loss": 6.1138,
      "mean_token_accuracy": 0.15200683176517488,
      "num_tokens": 12805944.0,
      "step": 6855
    },
    {
      "entropy": 6.224487781524658,
      "epoch": 0.6057930060049452,
      "grad_norm": 3.875,
      "learning_rate": 0.0004969806905322169,
      "loss": 6.0726,
      "mean_token_accuracy": 0.15439479649066926,
      "num_tokens": 12814692.0,
      "step": 6860
    },
    {
      "entropy": 6.1771105289459225,
      "epoch": 0.6062345460967856,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004969755466298649,
      "loss": 6.0233,
      "mean_token_accuracy": 0.1491582840681076,
      "num_tokens": 12824506.0,
      "step": 6865
    },
    {
      "entropy": 6.228961563110351,
      "epoch": 0.6066760861886259,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004969703983791382,
      "loss": 6.0964,
      "mean_token_accuracy": 0.15290898382663726,
      "num_tokens": 12835157.0,
      "step": 6870
    },
    {
      "entropy": 6.189265441894531,
      "epoch": 0.6071176262804663,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004969652457801376,
      "loss": 6.0108,
      "mean_token_accuracy": 0.1485551714897156,
      "num_tokens": 12844281.0,
      "step": 6875
    },
    {
      "entropy": 6.2684972286224365,
      "epoch": 0.6075591663723066,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004969600888329641,
      "loss": 6.0707,
      "mean_token_accuracy": 0.1507108047604561,
      "num_tokens": 12854486.0,
      "step": 6880
    },
    {
      "entropy": 6.215403413772583,
      "epoch": 0.608000706464147,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004969549275377186,
      "loss": 6.0526,
      "mean_token_accuracy": 0.15018294453620912,
      "num_tokens": 12863574.0,
      "step": 6885
    },
    {
      "entropy": 6.19040732383728,
      "epoch": 0.6084422465559873,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004969497618945024,
      "loss": 6.0325,
      "mean_token_accuracy": 0.15197444781661035,
      "num_tokens": 12872450.0,
      "step": 6890
    },
    {
      "entropy": 6.17571005821228,
      "epoch": 0.6088837866478276,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004969445919034165,
      "loss": 5.9714,
      "mean_token_accuracy": 0.16512659937143326,
      "num_tokens": 12881299.0,
      "step": 6895
    },
    {
      "entropy": 6.0891955375671385,
      "epoch": 0.609325326739668,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004969394175645623,
      "loss": 5.9488,
      "mean_token_accuracy": 0.15983871519565582,
      "num_tokens": 12889943.0,
      "step": 6900
    },
    {
      "entropy": 6.0837867736816404,
      "epoch": 0.6097668668315083,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004969342388780411,
      "loss": 6.0627,
      "mean_token_accuracy": 0.15418387949466705,
      "num_tokens": 12899497.0,
      "step": 6905
    },
    {
      "entropy": 6.157145881652832,
      "epoch": 0.6102084069233487,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004969290558439546,
      "loss": 6.0392,
      "mean_token_accuracy": 0.15607356578111647,
      "num_tokens": 12908677.0,
      "step": 6910
    },
    {
      "entropy": 6.222147655487061,
      "epoch": 0.6106499470151889,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004969238684624038,
      "loss": 6.155,
      "mean_token_accuracy": 0.14163556098937988,
      "num_tokens": 12918468.0,
      "step": 6915
    },
    {
      "entropy": 6.208994007110595,
      "epoch": 0.6110914871070293,
      "grad_norm": 4.75,
      "learning_rate": 0.0004969186767334908,
      "loss": 6.0333,
      "mean_token_accuracy": 0.15937585681676864,
      "num_tokens": 12928527.0,
      "step": 6920
    },
    {
      "entropy": 6.227367925643921,
      "epoch": 0.6115330271988697,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004969134806573172,
      "loss": 5.9931,
      "mean_token_accuracy": 0.1560715138912201,
      "num_tokens": 12937021.0,
      "step": 6925
    },
    {
      "entropy": 6.0965118408203125,
      "epoch": 0.61197456729071,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004969082802339846,
      "loss": 5.9683,
      "mean_token_accuracy": 0.1567614756524563,
      "num_tokens": 12945570.0,
      "step": 6930
    },
    {
      "entropy": 6.151629972457886,
      "epoch": 0.6124161073825504,
      "grad_norm": 3.984375,
      "learning_rate": 0.000496903075463595,
      "loss": 6.0941,
      "mean_token_accuracy": 0.14968045204877853,
      "num_tokens": 12954594.0,
      "step": 6935
    },
    {
      "entropy": 6.1665760517120365,
      "epoch": 0.6128576474743906,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004968978663462504,
      "loss": 6.0703,
      "mean_token_accuracy": 0.15526689887046813,
      "num_tokens": 12963328.0,
      "step": 6940
    },
    {
      "entropy": 6.171751880645752,
      "epoch": 0.613299187566231,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004968926528820529,
      "loss": 6.0662,
      "mean_token_accuracy": 0.15428778827190398,
      "num_tokens": 12972806.0,
      "step": 6945
    },
    {
      "entropy": 6.232146310806274,
      "epoch": 0.6137407276580713,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004968874350711044,
      "loss": 6.0649,
      "mean_token_accuracy": 0.1464855320751667,
      "num_tokens": 12981940.0,
      "step": 6950
    },
    {
      "entropy": 6.231241750717163,
      "epoch": 0.6141822677499117,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004968822129135072,
      "loss": 6.0182,
      "mean_token_accuracy": 0.15274877846240997,
      "num_tokens": 12990501.0,
      "step": 6955
    },
    {
      "entropy": 6.256838703155518,
      "epoch": 0.6146238078417521,
      "grad_norm": 4.25,
      "learning_rate": 0.0004968769864093637,
      "loss": 6.0867,
      "mean_token_accuracy": 0.14663174301385878,
      "num_tokens": 12999734.0,
      "step": 6960
    },
    {
      "entropy": 6.219386291503906,
      "epoch": 0.6150653479335924,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004968717555587762,
      "loss": 6.1369,
      "mean_token_accuracy": 0.14457945972681047,
      "num_tokens": 13009323.0,
      "step": 6965
    },
    {
      "entropy": 6.275314235687256,
      "epoch": 0.6155068880254327,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004968665203618472,
      "loss": 6.1536,
      "mean_token_accuracy": 0.15292704552412034,
      "num_tokens": 13018136.0,
      "step": 6970
    },
    {
      "entropy": 6.254905414581299,
      "epoch": 0.615948428117273,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004968612808186793,
      "loss": 6.1091,
      "mean_token_accuracy": 0.15201055482029915,
      "num_tokens": 13026494.0,
      "step": 6975
    },
    {
      "entropy": 6.192426538467407,
      "epoch": 0.6163899682091134,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004968560369293751,
      "loss": 6.1093,
      "mean_token_accuracy": 0.15276225358247758,
      "num_tokens": 13036548.0,
      "step": 6980
    },
    {
      "entropy": 6.232642126083374,
      "epoch": 0.6168315083009537,
      "grad_norm": 4.125,
      "learning_rate": 0.0004968507886940372,
      "loss": 6.1121,
      "mean_token_accuracy": 0.15091473534703254,
      "num_tokens": 13045653.0,
      "step": 6985
    },
    {
      "entropy": 6.289934301376343,
      "epoch": 0.6172730483927941,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004968455361127686,
      "loss": 6.1739,
      "mean_token_accuracy": 0.14828065782785416,
      "num_tokens": 13056204.0,
      "step": 6990
    },
    {
      "entropy": 6.243044137954712,
      "epoch": 0.6177145884846345,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004968402791856721,
      "loss": 6.1285,
      "mean_token_accuracy": 0.15255769044160844,
      "num_tokens": 13067215.0,
      "step": 6995
    },
    {
      "entropy": 6.24944486618042,
      "epoch": 0.6181561285764747,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004968350179128507,
      "loss": 6.0627,
      "mean_token_accuracy": 0.15274736136198044,
      "num_tokens": 13076824.0,
      "step": 7000
    },
    {
      "entropy": 6.172572994232178,
      "epoch": 0.6185976686683151,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004968297522944075,
      "loss": 6.0354,
      "mean_token_accuracy": 0.15531302243471146,
      "num_tokens": 13086205.0,
      "step": 7005
    },
    {
      "entropy": 6.105914640426636,
      "epoch": 0.6190392087601554,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004968244823304455,
      "loss": 6.0018,
      "mean_token_accuracy": 0.15155387222766875,
      "num_tokens": 13096381.0,
      "step": 7010
    },
    {
      "entropy": 6.20639476776123,
      "epoch": 0.6194807488519958,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004968192080210683,
      "loss": 6.0391,
      "mean_token_accuracy": 0.1583210527896881,
      "num_tokens": 13106533.0,
      "step": 7015
    },
    {
      "entropy": 6.117431354522705,
      "epoch": 0.619922288943836,
      "grad_norm": 4.125,
      "learning_rate": 0.0004968139293663787,
      "loss": 6.0547,
      "mean_token_accuracy": 0.15728403776884078,
      "num_tokens": 13116605.0,
      "step": 7020
    },
    {
      "entropy": 6.140177392959595,
      "epoch": 0.6203638290356764,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004968086463664806,
      "loss": 6.0207,
      "mean_token_accuracy": 0.15957427769899368,
      "num_tokens": 13125316.0,
      "step": 7025
    },
    {
      "entropy": 6.286570930480957,
      "epoch": 0.6208053691275168,
      "grad_norm": 3.421875,
      "learning_rate": 0.000496803359021477,
      "loss": 6.1795,
      "mean_token_accuracy": 0.15137581676244735,
      "num_tokens": 13135595.0,
      "step": 7030
    },
    {
      "entropy": 6.3050120830535885,
      "epoch": 0.6212469092193571,
      "grad_norm": 2.0,
      "learning_rate": 0.0004967980673314719,
      "loss": 6.1225,
      "mean_token_accuracy": 0.15151166468858718,
      "num_tokens": 13144395.0,
      "step": 7035
    },
    {
      "entropy": 6.1149821281433105,
      "epoch": 0.6216884493111975,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004967927712965688,
      "loss": 6.0612,
      "mean_token_accuracy": 0.15384516268968582,
      "num_tokens": 13153906.0,
      "step": 7040
    },
    {
      "entropy": 6.1296850681304935,
      "epoch": 0.6221299894030378,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004967874709168714,
      "loss": 6.0122,
      "mean_token_accuracy": 0.1552685961127281,
      "num_tokens": 13163840.0,
      "step": 7045
    },
    {
      "entropy": 6.168993759155273,
      "epoch": 0.6225715294948782,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004967821661924837,
      "loss": 6.038,
      "mean_token_accuracy": 0.15430732667446137,
      "num_tokens": 13173224.0,
      "step": 7050
    },
    {
      "entropy": 6.19411449432373,
      "epoch": 0.6230130695867184,
      "grad_norm": 6.875,
      "learning_rate": 0.0004967768571235094,
      "loss": 6.0777,
      "mean_token_accuracy": 0.1492239847779274,
      "num_tokens": 13182664.0,
      "step": 7055
    },
    {
      "entropy": 6.154684066772461,
      "epoch": 0.6234546096785588,
      "grad_norm": 2.5,
      "learning_rate": 0.0004967715437100526,
      "loss": 5.9867,
      "mean_token_accuracy": 0.1553950473666191,
      "num_tokens": 13191076.0,
      "step": 7060
    },
    {
      "entropy": 6.296986627578735,
      "epoch": 0.6238961497703992,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004967662259522174,
      "loss": 6.1738,
      "mean_token_accuracy": 0.14851178973913193,
      "num_tokens": 13201075.0,
      "step": 7065
    },
    {
      "entropy": 6.217172002792358,
      "epoch": 0.6243376898622395,
      "grad_norm": 2.0625,
      "learning_rate": 0.000496760903850108,
      "loss": 6.1084,
      "mean_token_accuracy": 0.15260327011346816,
      "num_tokens": 13210602.0,
      "step": 7070
    },
    {
      "entropy": 6.235305547714233,
      "epoch": 0.6247792299540799,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004967555774038286,
      "loss": 6.0738,
      "mean_token_accuracy": 0.15017012655735015,
      "num_tokens": 13219601.0,
      "step": 7075
    },
    {
      "entropy": 6.2360265254974365,
      "epoch": 0.6252207700459201,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004967502466134836,
      "loss": 6.1494,
      "mean_token_accuracy": 0.15250493586063385,
      "num_tokens": 13230675.0,
      "step": 7080
    },
    {
      "entropy": 6.27086296081543,
      "epoch": 0.6256623101377605,
      "grad_norm": 2.625,
      "learning_rate": 0.0004967449114791773,
      "loss": 6.0931,
      "mean_token_accuracy": 0.15382839292287825,
      "num_tokens": 13240349.0,
      "step": 7085
    },
    {
      "entropy": 6.238660764694214,
      "epoch": 0.6261038502296008,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004967395720010143,
      "loss": 6.1546,
      "mean_token_accuracy": 0.14545532166957856,
      "num_tokens": 13249867.0,
      "step": 7090
    },
    {
      "entropy": 6.212581157684326,
      "epoch": 0.6265453903214412,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004967342281790992,
      "loss": 6.0925,
      "mean_token_accuracy": 0.15382098108530046,
      "num_tokens": 13259143.0,
      "step": 7095
    },
    {
      "entropy": 6.256375122070312,
      "epoch": 0.6269869304132816,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004967288800135368,
      "loss": 6.0149,
      "mean_token_accuracy": 0.149423186480999,
      "num_tokens": 13268673.0,
      "step": 7100
    },
    {
      "entropy": 6.232032728195191,
      "epoch": 0.6274284705051218,
      "grad_norm": 2.625,
      "learning_rate": 0.0004967235275044315,
      "loss": 6.0383,
      "mean_token_accuracy": 0.1562743902206421,
      "num_tokens": 13278035.0,
      "step": 7105
    },
    {
      "entropy": 6.261171579360962,
      "epoch": 0.6278700105969622,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004967181706518886,
      "loss": 6.1578,
      "mean_token_accuracy": 0.14218074977397918,
      "num_tokens": 13286950.0,
      "step": 7110
    },
    {
      "entropy": 6.111796236038208,
      "epoch": 0.6283115506888025,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004967128094560128,
      "loss": 6.0202,
      "mean_token_accuracy": 0.16301421150565149,
      "num_tokens": 13296425.0,
      "step": 7115
    },
    {
      "entropy": 6.250481128692627,
      "epoch": 0.6287530907806429,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004967074439169092,
      "loss": 6.1432,
      "mean_token_accuracy": 0.14082750529050828,
      "num_tokens": 13306094.0,
      "step": 7120
    },
    {
      "entropy": 6.209989309310913,
      "epoch": 0.6291946308724832,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004967020740346827,
      "loss": 6.0726,
      "mean_token_accuracy": 0.1573947176337242,
      "num_tokens": 13315181.0,
      "step": 7125
    },
    {
      "entropy": 6.201513481140137,
      "epoch": 0.6296361709643236,
      "grad_norm": 2.375,
      "learning_rate": 0.0004966966998094387,
      "loss": 5.9838,
      "mean_token_accuracy": 0.15726577639579772,
      "num_tokens": 13324108.0,
      "step": 7130
    },
    {
      "entropy": 6.2309064865112305,
      "epoch": 0.630077711056164,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004966913212412825,
      "loss": 6.0686,
      "mean_token_accuracy": 0.15295007675886155,
      "num_tokens": 13333101.0,
      "step": 7135
    },
    {
      "entropy": 6.212774181365967,
      "epoch": 0.6305192511480042,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004966859383303194,
      "loss": 6.0015,
      "mean_token_accuracy": 0.1559315010905266,
      "num_tokens": 13341804.0,
      "step": 7140
    },
    {
      "entropy": 6.160013818740845,
      "epoch": 0.6309607912398446,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004966805510766549,
      "loss": 6.0336,
      "mean_token_accuracy": 0.15144553557038307,
      "num_tokens": 13350911.0,
      "step": 7145
    },
    {
      "entropy": 6.2372581481933596,
      "epoch": 0.6314023313316849,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004966751594803944,
      "loss": 6.0261,
      "mean_token_accuracy": 0.15334241688251496,
      "num_tokens": 13360657.0,
      "step": 7150
    },
    {
      "entropy": 6.166117763519287,
      "epoch": 0.6318438714235253,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004966697635416437,
      "loss": 5.9945,
      "mean_token_accuracy": 0.1584376350045204,
      "num_tokens": 13370462.0,
      "step": 7155
    },
    {
      "entropy": 6.263478851318359,
      "epoch": 0.6322854115153655,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004966643632605082,
      "loss": 6.1045,
      "mean_token_accuracy": 0.14802244305610657,
      "num_tokens": 13379446.0,
      "step": 7160
    },
    {
      "entropy": 6.201040315628052,
      "epoch": 0.6327269516072059,
      "grad_norm": 2.328125,
      "learning_rate": 0.000496658958637094,
      "loss": 6.1054,
      "mean_token_accuracy": 0.1525284618139267,
      "num_tokens": 13389238.0,
      "step": 7165
    },
    {
      "entropy": 6.1420855045318605,
      "epoch": 0.6331684916990463,
      "grad_norm": 2.703125,
      "learning_rate": 0.000496653549671507,
      "loss": 6.1154,
      "mean_token_accuracy": 0.14715484976768495,
      "num_tokens": 13398613.0,
      "step": 7170
    },
    {
      "entropy": 6.264933013916016,
      "epoch": 0.6336100317908866,
      "grad_norm": 2.625,
      "learning_rate": 0.0004966481363638529,
      "loss": 6.0982,
      "mean_token_accuracy": 0.15162092745304107,
      "num_tokens": 13407600.0,
      "step": 7175
    },
    {
      "entropy": 6.142405271530151,
      "epoch": 0.634051571882727,
      "grad_norm": 2.25,
      "learning_rate": 0.0004966427187142378,
      "loss": 5.9838,
      "mean_token_accuracy": 0.1597929358482361,
      "num_tokens": 13415974.0,
      "step": 7180
    },
    {
      "entropy": 6.245382308959961,
      "epoch": 0.6344931119745673,
      "grad_norm": 2.125,
      "learning_rate": 0.0004966372967227681,
      "loss": 6.0112,
      "mean_token_accuracy": 0.15867528319358826,
      "num_tokens": 13425628.0,
      "step": 7185
    },
    {
      "entropy": 6.155143213272095,
      "epoch": 0.6349346520664076,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004966318703895496,
      "loss": 6.0202,
      "mean_token_accuracy": 0.16059588491916657,
      "num_tokens": 13435573.0,
      "step": 7190
    },
    {
      "entropy": 6.180757522583008,
      "epoch": 0.6353761921582479,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004966264397146888,
      "loss": 5.9861,
      "mean_token_accuracy": 0.155886735022068,
      "num_tokens": 13443969.0,
      "step": 7195
    },
    {
      "entropy": 6.26367564201355,
      "epoch": 0.6358177322500883,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004966210046982921,
      "loss": 6.0873,
      "mean_token_accuracy": 0.14780050665140151,
      "num_tokens": 13452823.0,
      "step": 7200
    },
    {
      "entropy": 6.246481275558471,
      "epoch": 0.6362592723419287,
      "grad_norm": 6.5625,
      "learning_rate": 0.0004966155653404661,
      "loss": 6.0847,
      "mean_token_accuracy": 0.15006960928440094,
      "num_tokens": 13461090.0,
      "step": 7205
    },
    {
      "entropy": 6.266584396362305,
      "epoch": 0.636700812433769,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004966101216413172,
      "loss": 6.0998,
      "mean_token_accuracy": 0.143781378865242,
      "num_tokens": 13471425.0,
      "step": 7210
    },
    {
      "entropy": 6.295030927658081,
      "epoch": 0.6371423525256094,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004966046736009519,
      "loss": 6.1334,
      "mean_token_accuracy": 0.1519760623574257,
      "num_tokens": 13479755.0,
      "step": 7215
    },
    {
      "entropy": 6.20279541015625,
      "epoch": 0.6375838926174496,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004965992212194772,
      "loss": 6.0216,
      "mean_token_accuracy": 0.15963679254055024,
      "num_tokens": 13488854.0,
      "step": 7220
    },
    {
      "entropy": 6.139471006393433,
      "epoch": 0.63802543270929,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004965937644969998,
      "loss": 6.0855,
      "mean_token_accuracy": 0.1470048598945141,
      "num_tokens": 13497987.0,
      "step": 7225
    },
    {
      "entropy": 6.185364294052124,
      "epoch": 0.6384669728011303,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004965883034336266,
      "loss": 6.1126,
      "mean_token_accuracy": 0.15082333609461784,
      "num_tokens": 13506960.0,
      "step": 7230
    },
    {
      "entropy": 6.267209434509278,
      "epoch": 0.6389085128929707,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004965828380294645,
      "loss": 6.0478,
      "mean_token_accuracy": 0.16055953800678252,
      "num_tokens": 13515907.0,
      "step": 7235
    },
    {
      "entropy": 6.232587671279907,
      "epoch": 0.6393500529848111,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004965773682846207,
      "loss": 6.1523,
      "mean_token_accuracy": 0.15573379397392273,
      "num_tokens": 13524876.0,
      "step": 7240
    },
    {
      "entropy": 6.185802936553955,
      "epoch": 0.6397915930766513,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004965718941992021,
      "loss": 6.0312,
      "mean_token_accuracy": 0.15648015439510346,
      "num_tokens": 13532891.0,
      "step": 7245
    },
    {
      "entropy": 6.307486963272095,
      "epoch": 0.6402331331684917,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004965664157733162,
      "loss": 6.1176,
      "mean_token_accuracy": 0.15312286764383315,
      "num_tokens": 13542409.0,
      "step": 7250
    },
    {
      "entropy": 6.226913166046143,
      "epoch": 0.640674673260332,
      "grad_norm": 2.625,
      "learning_rate": 0.0004965609330070703,
      "loss": 6.0814,
      "mean_token_accuracy": 0.1505043089389801,
      "num_tokens": 13551947.0,
      "step": 7255
    },
    {
      "entropy": 6.246300411224365,
      "epoch": 0.6411162133521724,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004965554459005717,
      "loss": 6.1256,
      "mean_token_accuracy": 0.15032473504543303,
      "num_tokens": 13560812.0,
      "step": 7260
    },
    {
      "entropy": 6.283781480789185,
      "epoch": 0.6415577534440127,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004965499544539278,
      "loss": 6.1123,
      "mean_token_accuracy": 0.14852680265903473,
      "num_tokens": 13570197.0,
      "step": 7265
    },
    {
      "entropy": 6.1979292869567875,
      "epoch": 0.641999293535853,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004965444586672463,
      "loss": 6.0288,
      "mean_token_accuracy": 0.15019979029893876,
      "num_tokens": 13579614.0,
      "step": 7270
    },
    {
      "entropy": 6.203632497787476,
      "epoch": 0.6424408336276934,
      "grad_norm": 4.0,
      "learning_rate": 0.000496538958540635,
      "loss": 6.0889,
      "mean_token_accuracy": 0.15143698453903198,
      "num_tokens": 13589511.0,
      "step": 7275
    },
    {
      "entropy": 6.26300163269043,
      "epoch": 0.6428823737195337,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004965334540742015,
      "loss": 6.0463,
      "mean_token_accuracy": 0.1591418817639351,
      "num_tokens": 13597718.0,
      "step": 7280
    },
    {
      "entropy": 6.192359066009521,
      "epoch": 0.6433239138113741,
      "grad_norm": 2.375,
      "learning_rate": 0.0004965279452680535,
      "loss": 6.1226,
      "mean_token_accuracy": 0.1504811927676201,
      "num_tokens": 13607599.0,
      "step": 7285
    },
    {
      "entropy": 6.262105798721313,
      "epoch": 0.6437654539032144,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004965224321222991,
      "loss": 6.1584,
      "mean_token_accuracy": 0.1500078409910202,
      "num_tokens": 13616513.0,
      "step": 7290
    },
    {
      "entropy": 6.173467588424683,
      "epoch": 0.6442069939950548,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004965169146370462,
      "loss": 6.0219,
      "mean_token_accuracy": 0.15918111354112624,
      "num_tokens": 13625505.0,
      "step": 7295
    },
    {
      "entropy": 6.249496698379517,
      "epoch": 0.644648534086895,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004965113928124031,
      "loss": 6.1037,
      "mean_token_accuracy": 0.15476782619953156,
      "num_tokens": 13635340.0,
      "step": 7300
    },
    {
      "entropy": 6.183323574066162,
      "epoch": 0.6450900741787354,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004965058666484775,
      "loss": 6.0321,
      "mean_token_accuracy": 0.1503544792532921,
      "num_tokens": 13644886.0,
      "step": 7305
    },
    {
      "entropy": 6.183138990402222,
      "epoch": 0.6455316142705758,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004965003361453782,
      "loss": 6.0759,
      "mean_token_accuracy": 0.1546678811311722,
      "num_tokens": 13654093.0,
      "step": 7310
    },
    {
      "entropy": 6.262458944320679,
      "epoch": 0.6459731543624161,
      "grad_norm": 2.375,
      "learning_rate": 0.0004964948013032132,
      "loss": 6.1199,
      "mean_token_accuracy": 0.14781120419502258,
      "num_tokens": 13663095.0,
      "step": 7315
    },
    {
      "entropy": 6.282963609695434,
      "epoch": 0.6464146944542565,
      "grad_norm": 2.125,
      "learning_rate": 0.000496489262122091,
      "loss": 6.1553,
      "mean_token_accuracy": 0.14864987134933472,
      "num_tokens": 13673624.0,
      "step": 7320
    },
    {
      "entropy": 6.259463548660278,
      "epoch": 0.6468562345460968,
      "grad_norm": 7.4375,
      "learning_rate": 0.0004964837186021201,
      "loss": 6.0348,
      "mean_token_accuracy": 0.15352552831172944,
      "num_tokens": 13683145.0,
      "step": 7325
    },
    {
      "entropy": 6.175587797164917,
      "epoch": 0.6472977746379371,
      "grad_norm": 3.75,
      "learning_rate": 0.0004964781707434093,
      "loss": 6.0329,
      "mean_token_accuracy": 0.15152520984411239,
      "num_tokens": 13691963.0,
      "step": 7330
    },
    {
      "entropy": 6.037362384796142,
      "epoch": 0.6477393147297774,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004964726185460669,
      "loss": 5.9579,
      "mean_token_accuracy": 0.1579597070813179,
      "num_tokens": 13700926.0,
      "step": 7335
    },
    {
      "entropy": 6.1612313747406,
      "epoch": 0.6481808548216178,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004964670620102019,
      "loss": 6.1494,
      "mean_token_accuracy": 0.14145292192697526,
      "num_tokens": 13709824.0,
      "step": 7340
    },
    {
      "entropy": 6.25114688873291,
      "epoch": 0.6486223949134582,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004964615011359231,
      "loss": 6.0579,
      "mean_token_accuracy": 0.151515481621027,
      "num_tokens": 13718789.0,
      "step": 7345
    },
    {
      "entropy": 6.252131652832031,
      "epoch": 0.6490639350052985,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004964559359233395,
      "loss": 6.0618,
      "mean_token_accuracy": 0.15075841844081878,
      "num_tokens": 13728180.0,
      "step": 7350
    },
    {
      "entropy": 6.1781408309936525,
      "epoch": 0.6495054750971389,
      "grad_norm": 2.375,
      "learning_rate": 0.0004964503663725601,
      "loss": 6.167,
      "mean_token_accuracy": 0.14491574019193648,
      "num_tokens": 13737819.0,
      "step": 7355
    },
    {
      "entropy": 6.216016721725464,
      "epoch": 0.6499470151889791,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004964447924836938,
      "loss": 6.1369,
      "mean_token_accuracy": 0.149229533970356,
      "num_tokens": 13747519.0,
      "step": 7360
    },
    {
      "entropy": 6.343541479110717,
      "epoch": 0.6503885552808195,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004964392142568501,
      "loss": 6.1307,
      "mean_token_accuracy": 0.14465169459581376,
      "num_tokens": 13757821.0,
      "step": 7365
    },
    {
      "entropy": 6.229691362380981,
      "epoch": 0.6508300953726598,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004964336316921381,
      "loss": 5.9414,
      "mean_token_accuracy": 0.15893535315990448,
      "num_tokens": 13766560.0,
      "step": 7370
    },
    {
      "entropy": 6.181404256820679,
      "epoch": 0.6512716354645002,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004964280447896672,
      "loss": 5.9699,
      "mean_token_accuracy": 0.1556891031563282,
      "num_tokens": 13775560.0,
      "step": 7375
    },
    {
      "entropy": 6.093255043029785,
      "epoch": 0.6517131755563406,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004964224535495468,
      "loss": 6.0828,
      "mean_token_accuracy": 0.14826790988445282,
      "num_tokens": 13784391.0,
      "step": 7380
    },
    {
      "entropy": 6.163384532928466,
      "epoch": 0.6521547156481808,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004964168579718864,
      "loss": 6.047,
      "mean_token_accuracy": 0.15087808892130852,
      "num_tokens": 13794008.0,
      "step": 7385
    },
    {
      "entropy": 6.277848768234253,
      "epoch": 0.6525962557400212,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004964112580567959,
      "loss": 6.0799,
      "mean_token_accuracy": 0.14790103286504747,
      "num_tokens": 13804321.0,
      "step": 7390
    },
    {
      "entropy": 6.251108884811401,
      "epoch": 0.6530377958318615,
      "grad_norm": 10.625,
      "learning_rate": 0.0004964056538043845,
      "loss": 6.0947,
      "mean_token_accuracy": 0.1505176082253456,
      "num_tokens": 13813492.0,
      "step": 7395
    },
    {
      "entropy": 6.119813537597656,
      "epoch": 0.6534793359237019,
      "grad_norm": 2.125,
      "learning_rate": 0.0004964000452147624,
      "loss": 6.07,
      "mean_token_accuracy": 0.1533577561378479,
      "num_tokens": 13822471.0,
      "step": 7400
    },
    {
      "entropy": 6.178601408004761,
      "epoch": 0.6539208760155422,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004963944322880394,
      "loss": 6.1343,
      "mean_token_accuracy": 0.14370774999260902,
      "num_tokens": 13832789.0,
      "step": 7405
    },
    {
      "entropy": 6.242937612533569,
      "epoch": 0.6543624161073825,
      "grad_norm": 2.125,
      "learning_rate": 0.0004963888150243253,
      "loss": 6.1002,
      "mean_token_accuracy": 0.14843902736902237,
      "num_tokens": 13842230.0,
      "step": 7410
    },
    {
      "entropy": 6.277557611465454,
      "epoch": 0.6548039561992229,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004963831934237302,
      "loss": 6.0289,
      "mean_token_accuracy": 0.15670600831508635,
      "num_tokens": 13851542.0,
      "step": 7415
    },
    {
      "entropy": 6.187560272216797,
      "epoch": 0.6552454962910632,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004963775674863643,
      "loss": 6.0288,
      "mean_token_accuracy": 0.1488627851009369,
      "num_tokens": 13860753.0,
      "step": 7420
    },
    {
      "entropy": 6.172817325592041,
      "epoch": 0.6556870363829036,
      "grad_norm": 2.75,
      "learning_rate": 0.0004963719372123378,
      "loss": 6.0395,
      "mean_token_accuracy": 0.15519573241472245,
      "num_tokens": 13870133.0,
      "step": 7425
    },
    {
      "entropy": 6.218771266937256,
      "epoch": 0.6561285764747439,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004963663026017609,
      "loss": 6.1058,
      "mean_token_accuracy": 0.15374775975942612,
      "num_tokens": 13879086.0,
      "step": 7430
    },
    {
      "entropy": 6.224050903320313,
      "epoch": 0.6565701165665843,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004963606636547441,
      "loss": 6.0514,
      "mean_token_accuracy": 0.1545367494225502,
      "num_tokens": 13888706.0,
      "step": 7435
    },
    {
      "entropy": 6.183874034881592,
      "epoch": 0.6570116566584245,
      "grad_norm": 1.875,
      "learning_rate": 0.0004963550203713976,
      "loss": 6.0948,
      "mean_token_accuracy": 0.15013214200735092,
      "num_tokens": 13897455.0,
      "step": 7440
    },
    {
      "entropy": 6.207463502883911,
      "epoch": 0.6574531967502649,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004963493727518323,
      "loss": 6.0951,
      "mean_token_accuracy": 0.15092161446809768,
      "num_tokens": 13907697.0,
      "step": 7445
    },
    {
      "entropy": 6.19521541595459,
      "epoch": 0.6578947368421053,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004963437207961585,
      "loss": 6.1067,
      "mean_token_accuracy": 0.15172283574938775,
      "num_tokens": 13917124.0,
      "step": 7450
    },
    {
      "entropy": 6.222240877151489,
      "epoch": 0.6583362769339456,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004963380645044874,
      "loss": 6.0407,
      "mean_token_accuracy": 0.15824299454689025,
      "num_tokens": 13926434.0,
      "step": 7455
    },
    {
      "entropy": 6.138532924652099,
      "epoch": 0.658777817025786,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004963324038769292,
      "loss": 6.0144,
      "mean_token_accuracy": 0.1565467596054077,
      "num_tokens": 13934883.0,
      "step": 7460
    },
    {
      "entropy": 6.204624605178833,
      "epoch": 0.6592193571176262,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004963267389135953,
      "loss": 6.0827,
      "mean_token_accuracy": 0.15051436573266982,
      "num_tokens": 13944514.0,
      "step": 7465
    },
    {
      "entropy": 6.168144702911377,
      "epoch": 0.6596608972094666,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004963210696145964,
      "loss": 6.0078,
      "mean_token_accuracy": 0.1553253263235092,
      "num_tokens": 13953434.0,
      "step": 7470
    },
    {
      "entropy": 6.130171871185302,
      "epoch": 0.6601024373013069,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004963153959800436,
      "loss": 6.0286,
      "mean_token_accuracy": 0.16009896844625474,
      "num_tokens": 13962493.0,
      "step": 7475
    },
    {
      "entropy": 6.192154836654663,
      "epoch": 0.6605439773931473,
      "grad_norm": 2.125,
      "learning_rate": 0.000496309718010048,
      "loss": 6.0678,
      "mean_token_accuracy": 0.15374763309955597,
      "num_tokens": 13971193.0,
      "step": 7480
    },
    {
      "entropy": 6.266092729568482,
      "epoch": 0.6609855174849877,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004963040357047209,
      "loss": 6.1029,
      "mean_token_accuracy": 0.1511322468519211,
      "num_tokens": 13981829.0,
      "step": 7485
    },
    {
      "entropy": 6.223195028305054,
      "epoch": 0.661427057576828,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004962983490641737,
      "loss": 6.035,
      "mean_token_accuracy": 0.15042887330055238,
      "num_tokens": 13990262.0,
      "step": 7490
    },
    {
      "entropy": 6.1686547756195065,
      "epoch": 0.6618685976686683,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004962926580885177,
      "loss": 6.0737,
      "mean_token_accuracy": 0.1577219158411026,
      "num_tokens": 14000100.0,
      "step": 7495
    },
    {
      "entropy": 6.214493942260742,
      "epoch": 0.6623101377605086,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004962869627778642,
      "loss": 6.1151,
      "mean_token_accuracy": 0.15706808865070343,
      "num_tokens": 14009330.0,
      "step": 7500
    },
    {
      "entropy": 6.23420147895813,
      "epoch": 0.662751677852349,
      "grad_norm": 2.125,
      "learning_rate": 0.0004962812631323252,
      "loss": 6.0684,
      "mean_token_accuracy": 0.15241270512342453,
      "num_tokens": 14018109.0,
      "step": 7505
    },
    {
      "entropy": 6.213022470474243,
      "epoch": 0.6631932179441893,
      "grad_norm": 2.65625,
      "learning_rate": 0.000496275559152012,
      "loss": 5.9906,
      "mean_token_accuracy": 0.15551604926586152,
      "num_tokens": 14027318.0,
      "step": 7510
    },
    {
      "entropy": 6.2161946296691895,
      "epoch": 0.6636347580360297,
      "grad_norm": 2.375,
      "learning_rate": 0.0004962698508370365,
      "loss": 6.1051,
      "mean_token_accuracy": 0.15343084782361985,
      "num_tokens": 14037061.0,
      "step": 7515
    },
    {
      "entropy": 6.179489421844482,
      "epoch": 0.6640762981278701,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004962641381875103,
      "loss": 6.0699,
      "mean_token_accuracy": 0.14859853237867354,
      "num_tokens": 14046008.0,
      "step": 7520
    },
    {
      "entropy": 6.203066730499268,
      "epoch": 0.6645178382197103,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004962584212035456,
      "loss": 6.0769,
      "mean_token_accuracy": 0.14658914059400557,
      "num_tokens": 14054535.0,
      "step": 7525
    },
    {
      "entropy": 6.205629682540893,
      "epoch": 0.6649593783115507,
      "grad_norm": 2.625,
      "learning_rate": 0.0004962526998852543,
      "loss": 6.0521,
      "mean_token_accuracy": 0.1541662707924843,
      "num_tokens": 14063680.0,
      "step": 7530
    },
    {
      "entropy": 6.24108362197876,
      "epoch": 0.665400918403391,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004962469742327483,
      "loss": 6.0845,
      "mean_token_accuracy": 0.1557173639535904,
      "num_tokens": 14072206.0,
      "step": 7535
    },
    {
      "entropy": 6.241954326629639,
      "epoch": 0.6658424584952314,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00049624124424614,
      "loss": 6.0976,
      "mean_token_accuracy": 0.15491850823163986,
      "num_tokens": 14081943.0,
      "step": 7540
    },
    {
      "entropy": 6.207191371917725,
      "epoch": 0.6662839985870717,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004962355099255415,
      "loss": 6.0566,
      "mean_token_accuracy": 0.1543470561504364,
      "num_tokens": 14091695.0,
      "step": 7545
    },
    {
      "entropy": 6.092154788970947,
      "epoch": 0.666725538678912,
      "grad_norm": 3.125,
      "learning_rate": 0.0004962297712710653,
      "loss": 5.9816,
      "mean_token_accuracy": 0.15961627513170243,
      "num_tokens": 14100811.0,
      "step": 7550
    },
    {
      "entropy": 6.1884294033050535,
      "epoch": 0.6671670787707524,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004962240282828236,
      "loss": 6.1548,
      "mean_token_accuracy": 0.14150830507278442,
      "num_tokens": 14109818.0,
      "step": 7555
    },
    {
      "entropy": 6.2431878566741945,
      "epoch": 0.6676086188625927,
      "grad_norm": 2.46875,
      "learning_rate": 0.000496218280960929,
      "loss": 6.0784,
      "mean_token_accuracy": 0.15114578008651733,
      "num_tokens": 14119811.0,
      "step": 7560
    },
    {
      "entropy": 6.316728115081787,
      "epoch": 0.6680501589544331,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004962125293054941,
      "loss": 6.0431,
      "mean_token_accuracy": 0.1584528237581253,
      "num_tokens": 14128963.0,
      "step": 7565
    },
    {
      "entropy": 6.191557216644287,
      "epoch": 0.6684916990462734,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004962067733166317,
      "loss": 6.0948,
      "mean_token_accuracy": 0.14102213308215142,
      "num_tokens": 14138532.0,
      "step": 7570
    },
    {
      "entropy": 6.112533760070801,
      "epoch": 0.6689332391381138,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004962010129944542,
      "loss": 6.0604,
      "mean_token_accuracy": 0.15434442609548568,
      "num_tokens": 14148107.0,
      "step": 7575
    },
    {
      "entropy": 6.277549076080322,
      "epoch": 0.669374779229954,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004961952483390747,
      "loss": 6.1156,
      "mean_token_accuracy": 0.1456269219517708,
      "num_tokens": 14158823.0,
      "step": 7580
    },
    {
      "entropy": 6.2724374294281,
      "epoch": 0.6698163193217944,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004961894793506062,
      "loss": 6.0665,
      "mean_token_accuracy": 0.15882535129785538,
      "num_tokens": 14168147.0,
      "step": 7585
    },
    {
      "entropy": 6.207061910629273,
      "epoch": 0.6702578594136348,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004961837060291615,
      "loss": 6.0889,
      "mean_token_accuracy": 0.15001036822795868,
      "num_tokens": 14177123.0,
      "step": 7590
    },
    {
      "entropy": 6.213112258911133,
      "epoch": 0.6706993995054751,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004961779283748538,
      "loss": 6.0762,
      "mean_token_accuracy": 0.14930718392133713,
      "num_tokens": 14186614.0,
      "step": 7595
    },
    {
      "entropy": 6.258604288101196,
      "epoch": 0.6711409395973155,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004961721463877963,
      "loss": 6.0879,
      "mean_token_accuracy": 0.147329892963171,
      "num_tokens": 14195310.0,
      "step": 7600
    },
    {
      "entropy": 6.242795372009278,
      "epoch": 0.6715824796891557,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004961663600681023,
      "loss": 6.0477,
      "mean_token_accuracy": 0.14992639869451524,
      "num_tokens": 14204321.0,
      "step": 7605
    },
    {
      "entropy": 6.1641522407531735,
      "epoch": 0.6720240197809961,
      "grad_norm": 1.921875,
      "learning_rate": 0.000496160569415885,
      "loss": 6.0743,
      "mean_token_accuracy": 0.14891463667154312,
      "num_tokens": 14213615.0,
      "step": 7610
    },
    {
      "entropy": 6.224883890151977,
      "epoch": 0.6724655598728364,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004961547744312579,
      "loss": 5.9574,
      "mean_token_accuracy": 0.15733491480350495,
      "num_tokens": 14222715.0,
      "step": 7615
    },
    {
      "entropy": 6.1939027309417725,
      "epoch": 0.6729070999646768,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004961489751143346,
      "loss": 6.0857,
      "mean_token_accuracy": 0.1541801318526268,
      "num_tokens": 14232117.0,
      "step": 7620
    },
    {
      "entropy": 6.13014850616455,
      "epoch": 0.6733486400565172,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004961431714652287,
      "loss": 5.9731,
      "mean_token_accuracy": 0.16042798459529878,
      "num_tokens": 14240606.0,
      "step": 7625
    },
    {
      "entropy": 6.220015954971314,
      "epoch": 0.6737901801483575,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004961373634840538,
      "loss": 6.073,
      "mean_token_accuracy": 0.14794691205024718,
      "num_tokens": 14249844.0,
      "step": 7630
    },
    {
      "entropy": 6.200179195404052,
      "epoch": 0.6742317202401978,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004961315511709237,
      "loss": 6.0879,
      "mean_token_accuracy": 0.15268362164497376,
      "num_tokens": 14259231.0,
      "step": 7635
    },
    {
      "entropy": 6.250634336471558,
      "epoch": 0.6746732603320381,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004961257345259524,
      "loss": 6.1228,
      "mean_token_accuracy": 0.15392789170145987,
      "num_tokens": 14268508.0,
      "step": 7640
    },
    {
      "entropy": 6.212841749191284,
      "epoch": 0.6751148004238785,
      "grad_norm": 3.75,
      "learning_rate": 0.0004961199135492535,
      "loss": 6.0158,
      "mean_token_accuracy": 0.1586117058992386,
      "num_tokens": 14278057.0,
      "step": 7645
    },
    {
      "entropy": 6.137183141708374,
      "epoch": 0.6755563405157188,
      "grad_norm": 2.25,
      "learning_rate": 0.0004961140882409414,
      "loss": 6.1053,
      "mean_token_accuracy": 0.14983683377504348,
      "num_tokens": 14287225.0,
      "step": 7650
    },
    {
      "entropy": 6.176516056060791,
      "epoch": 0.6759978806075592,
      "grad_norm": 2.15625,
      "learning_rate": 0.00049610825860113,
      "loss": 6.0602,
      "mean_token_accuracy": 0.15462789237499236,
      "num_tokens": 14296529.0,
      "step": 7655
    },
    {
      "entropy": 6.218318700790405,
      "epoch": 0.6764394206993996,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004961024246299336,
      "loss": 5.9889,
      "mean_token_accuracy": 0.15721210241317748,
      "num_tokens": 14305675.0,
      "step": 7660
    },
    {
      "entropy": 6.232691097259521,
      "epoch": 0.6768809607912398,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004960965863274665,
      "loss": 5.9634,
      "mean_token_accuracy": 0.15535827577114106,
      "num_tokens": 14313632.0,
      "step": 7665
    },
    {
      "entropy": 6.118216848373413,
      "epoch": 0.6773225008830802,
      "grad_norm": 3.203125,
      "learning_rate": 0.000496090743693843,
      "loss": 6.0886,
      "mean_token_accuracy": 0.15240846276283265,
      "num_tokens": 14323353.0,
      "step": 7670
    },
    {
      "entropy": 6.241943311691284,
      "epoch": 0.6777640409749205,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004960848967291775,
      "loss": 6.064,
      "mean_token_accuracy": 0.1510762870311737,
      "num_tokens": 14331953.0,
      "step": 7675
    },
    {
      "entropy": 6.291945219039917,
      "epoch": 0.6782055810667609,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004960790454335846,
      "loss": 6.1253,
      "mean_token_accuracy": 0.1440570555627346,
      "num_tokens": 14342456.0,
      "step": 7680
    },
    {
      "entropy": 6.166342258453369,
      "epoch": 0.6786471211586012,
      "grad_norm": 2.28125,
      "learning_rate": 0.000496073189807179,
      "loss": 5.9019,
      "mean_token_accuracy": 0.1624265655875206,
      "num_tokens": 14351013.0,
      "step": 7685
    },
    {
      "entropy": 6.187205171585083,
      "epoch": 0.6790886612504415,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004960673298500753,
      "loss": 6.18,
      "mean_token_accuracy": 0.14918676763772964,
      "num_tokens": 14360887.0,
      "step": 7690
    },
    {
      "entropy": 6.226143932342529,
      "epoch": 0.6795302013422819,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004960614655623883,
      "loss": 6.0989,
      "mean_token_accuracy": 0.148998162150383,
      "num_tokens": 14370510.0,
      "step": 7695
    },
    {
      "entropy": 6.229071617126465,
      "epoch": 0.6799717414341222,
      "grad_norm": 3.03125,
      "learning_rate": 0.000496055596944233,
      "loss": 6.0895,
      "mean_token_accuracy": 0.1563186064362526,
      "num_tokens": 14379315.0,
      "step": 7700
    },
    {
      "entropy": 6.21696834564209,
      "epoch": 0.6804132815259626,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004960497239957243,
      "loss": 6.0804,
      "mean_token_accuracy": 0.14839715361595154,
      "num_tokens": 14388537.0,
      "step": 7705
    },
    {
      "entropy": 6.247610473632813,
      "epoch": 0.6808548216178029,
      "grad_norm": 2.125,
      "learning_rate": 0.0004960438467169772,
      "loss": 6.1075,
      "mean_token_accuracy": 0.14529118984937667,
      "num_tokens": 14398347.0,
      "step": 7710
    },
    {
      "entropy": 6.288898611068726,
      "epoch": 0.6812963617096433,
      "grad_norm": 2.375,
      "learning_rate": 0.0004960379651081068,
      "loss": 6.0266,
      "mean_token_accuracy": 0.1562356874346733,
      "num_tokens": 14407304.0,
      "step": 7715
    },
    {
      "entropy": 6.2475076675415036,
      "epoch": 0.6817379018014835,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004960320791692283,
      "loss": 6.0411,
      "mean_token_accuracy": 0.15162470787763596,
      "num_tokens": 14416641.0,
      "step": 7720
    },
    {
      "entropy": 6.248139762878418,
      "epoch": 0.6821794418933239,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004960261889004572,
      "loss": 6.1653,
      "mean_token_accuracy": 0.14672458916902542,
      "num_tokens": 14426095.0,
      "step": 7725
    },
    {
      "entropy": 6.227323007583618,
      "epoch": 0.6826209819851643,
      "grad_norm": 1.75,
      "learning_rate": 0.0004960202943019088,
      "loss": 6.0558,
      "mean_token_accuracy": 0.15345034152269363,
      "num_tokens": 14434563.0,
      "step": 7730
    },
    {
      "entropy": 6.118875980377197,
      "epoch": 0.6830625220770046,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004960143953736984,
      "loss": 5.9894,
      "mean_token_accuracy": 0.15708500742912293,
      "num_tokens": 14444319.0,
      "step": 7735
    },
    {
      "entropy": 6.229998683929443,
      "epoch": 0.683504062168845,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004960084921159419,
      "loss": 6.0232,
      "mean_token_accuracy": 0.15335571765899658,
      "num_tokens": 14453381.0,
      "step": 7740
    },
    {
      "entropy": 6.130149507522583,
      "epoch": 0.6839456022606852,
      "grad_norm": 1.875,
      "learning_rate": 0.0004960025845287546,
      "loss": 6.0223,
      "mean_token_accuracy": 0.15564830154180526,
      "num_tokens": 14462660.0,
      "step": 7745
    },
    {
      "entropy": 6.125526428222656,
      "epoch": 0.6843871423525256,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004959966726122523,
      "loss": 5.9577,
      "mean_token_accuracy": 0.16211079359054564,
      "num_tokens": 14471820.0,
      "step": 7750
    },
    {
      "entropy": 6.218053960800171,
      "epoch": 0.6848286824443659,
      "grad_norm": 2.53125,
      "learning_rate": 0.000495990756366551,
      "loss": 6.1168,
      "mean_token_accuracy": 0.1515028163790703,
      "num_tokens": 14481663.0,
      "step": 7755
    },
    {
      "entropy": 6.257606124877929,
      "epoch": 0.6852702225362063,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004959848357917664,
      "loss": 6.1271,
      "mean_token_accuracy": 0.147647662460804,
      "num_tokens": 14491515.0,
      "step": 7760
    },
    {
      "entropy": 6.142368125915527,
      "epoch": 0.6857117626280467,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004959789108880145,
      "loss": 5.9671,
      "mean_token_accuracy": 0.14912231415510177,
      "num_tokens": 14500852.0,
      "step": 7765
    },
    {
      "entropy": 6.136193132400512,
      "epoch": 0.686153302719887,
      "grad_norm": 2.75,
      "learning_rate": 0.0004959729816554115,
      "loss": 6.0328,
      "mean_token_accuracy": 0.15739069730043412,
      "num_tokens": 14509771.0,
      "step": 7770
    },
    {
      "entropy": 6.211749124526977,
      "epoch": 0.6865948428117273,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004959670480940734,
      "loss": 6.0121,
      "mean_token_accuracy": 0.16098477467894554,
      "num_tokens": 14519377.0,
      "step": 7775
    },
    {
      "entropy": 6.21663122177124,
      "epoch": 0.6870363829035676,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004959611102041165,
      "loss": 6.0525,
      "mean_token_accuracy": 0.15086856484413147,
      "num_tokens": 14528998.0,
      "step": 7780
    },
    {
      "entropy": 6.172790050506592,
      "epoch": 0.687477922995408,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004959551679856571,
      "loss": 6.0792,
      "mean_token_accuracy": 0.15509206503629686,
      "num_tokens": 14538153.0,
      "step": 7785
    },
    {
      "entropy": 6.196965932846069,
      "epoch": 0.6879194630872483,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004959492214388117,
      "loss": 6.0166,
      "mean_token_accuracy": 0.1575303018093109,
      "num_tokens": 14547631.0,
      "step": 7790
    },
    {
      "entropy": 6.256254625320435,
      "epoch": 0.6883610031790887,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004959432705636968,
      "loss": 5.9797,
      "mean_token_accuracy": 0.15786290168762207,
      "num_tokens": 14557685.0,
      "step": 7795
    },
    {
      "entropy": 6.231467437744141,
      "epoch": 0.688802543270929,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004959373153604287,
      "loss": 6.0993,
      "mean_token_accuracy": 0.15257465690374375,
      "num_tokens": 14567360.0,
      "step": 7800
    },
    {
      "entropy": 6.125985431671142,
      "epoch": 0.6892440833627693,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004959313558291243,
      "loss": 6.0254,
      "mean_token_accuracy": 0.15141796320676804,
      "num_tokens": 14576916.0,
      "step": 7805
    },
    {
      "entropy": 6.101062536239624,
      "epoch": 0.6896856234546097,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004959253919699004,
      "loss": 5.9499,
      "mean_token_accuracy": 0.16243618130683898,
      "num_tokens": 14586716.0,
      "step": 7810
    },
    {
      "entropy": 6.2262571334838865,
      "epoch": 0.69012716354645,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004959194237828735,
      "loss": 5.9998,
      "mean_token_accuracy": 0.153834930062294,
      "num_tokens": 14595690.0,
      "step": 7815
    },
    {
      "entropy": 6.181614828109741,
      "epoch": 0.6905687036382904,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004959134512681609,
      "loss": 6.0546,
      "mean_token_accuracy": 0.14944540560245514,
      "num_tokens": 14605481.0,
      "step": 7820
    },
    {
      "entropy": 6.180599164962769,
      "epoch": 0.6910102437301306,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004959074744258794,
      "loss": 6.116,
      "mean_token_accuracy": 0.15085441544651984,
      "num_tokens": 14615242.0,
      "step": 7825
    },
    {
      "entropy": 6.201753234863281,
      "epoch": 0.691451783821971,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004959014932561461,
      "loss": 5.9454,
      "mean_token_accuracy": 0.16299830228090287,
      "num_tokens": 14624343.0,
      "step": 7830
    },
    {
      "entropy": 6.127270126342774,
      "epoch": 0.6918933239138114,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004958955077590782,
      "loss": 5.9414,
      "mean_token_accuracy": 0.16462224572896958,
      "num_tokens": 14633014.0,
      "step": 7835
    },
    {
      "entropy": 6.168978500366211,
      "epoch": 0.6923348640056517,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004958895179347929,
      "loss": 6.0763,
      "mean_token_accuracy": 0.1522584706544876,
      "num_tokens": 14642796.0,
      "step": 7840
    },
    {
      "entropy": 6.25012173652649,
      "epoch": 0.6927764040974921,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004958835237834075,
      "loss": 6.1059,
      "mean_token_accuracy": 0.14263201132416725,
      "num_tokens": 14652388.0,
      "step": 7845
    },
    {
      "entropy": 6.271516036987305,
      "epoch": 0.6932179441893324,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004958775253050396,
      "loss": 6.0596,
      "mean_token_accuracy": 0.15313490480184555,
      "num_tokens": 14661219.0,
      "step": 7850
    },
    {
      "entropy": 6.182176208496093,
      "epoch": 0.6936594842811727,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004958715224998066,
      "loss": 6.0727,
      "mean_token_accuracy": 0.15574069023132325,
      "num_tokens": 14670878.0,
      "step": 7855
    },
    {
      "entropy": 6.181418514251709,
      "epoch": 0.694101024373013,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004958655153678261,
      "loss": 6.0899,
      "mean_token_accuracy": 0.1497741535305977,
      "num_tokens": 14680893.0,
      "step": 7860
    },
    {
      "entropy": 6.235378074645996,
      "epoch": 0.6945425644648534,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004958595039092156,
      "loss": 6.0022,
      "mean_token_accuracy": 0.15413426607847214,
      "num_tokens": 14689249.0,
      "step": 7865
    },
    {
      "entropy": 6.23575348854065,
      "epoch": 0.6949841045566938,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004958534881240932,
      "loss": 6.0297,
      "mean_token_accuracy": 0.15611442476511,
      "num_tokens": 14698724.0,
      "step": 7870
    },
    {
      "entropy": 6.180769920349121,
      "epoch": 0.6954256446485341,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004958474680125766,
      "loss": 5.9864,
      "mean_token_accuracy": 0.15255807638168334,
      "num_tokens": 14707756.0,
      "step": 7875
    },
    {
      "entropy": 6.145475435256958,
      "epoch": 0.6958671847403745,
      "grad_norm": 5.15625,
      "learning_rate": 0.0004958414435747837,
      "loss": 6.027,
      "mean_token_accuracy": 0.1529964953660965,
      "num_tokens": 14717416.0,
      "step": 7880
    },
    {
      "entropy": 6.201198482513428,
      "epoch": 0.6963087248322147,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004958354148108324,
      "loss": 6.0707,
      "mean_token_accuracy": 0.16048150211572648,
      "num_tokens": 14727321.0,
      "step": 7885
    },
    {
      "entropy": 6.18621597290039,
      "epoch": 0.6967502649240551,
      "grad_norm": 3.390625,
      "learning_rate": 0.000495829381720841,
      "loss": 5.9491,
      "mean_token_accuracy": 0.166436567902565,
      "num_tokens": 14736298.0,
      "step": 7890
    },
    {
      "entropy": 6.208941507339477,
      "epoch": 0.6971918050158954,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004958233443049275,
      "loss": 6.0093,
      "mean_token_accuracy": 0.14525908082723618,
      "num_tokens": 14745893.0,
      "step": 7895
    },
    {
      "entropy": 6.208337497711182,
      "epoch": 0.6976333451077358,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004958173025632103,
      "loss": 6.0529,
      "mean_token_accuracy": 0.14887382239103317,
      "num_tokens": 14755277.0,
      "step": 7900
    },
    {
      "entropy": 6.2192895889282225,
      "epoch": 0.6980748851995762,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004958112564958079,
      "loss": 6.1697,
      "mean_token_accuracy": 0.14462502300739288,
      "num_tokens": 14764369.0,
      "step": 7905
    },
    {
      "entropy": 6.16321964263916,
      "epoch": 0.6985164252914164,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004958052061028385,
      "loss": 6.0086,
      "mean_token_accuracy": 0.16280398070812224,
      "num_tokens": 14772970.0,
      "step": 7910
    },
    {
      "entropy": 6.204917478561401,
      "epoch": 0.6989579653832568,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004957991513844205,
      "loss": 6.0686,
      "mean_token_accuracy": 0.15557245910167694,
      "num_tokens": 14782862.0,
      "step": 7915
    },
    {
      "entropy": 6.263249015808105,
      "epoch": 0.6993995054750971,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004957930923406729,
      "loss": 6.0712,
      "mean_token_accuracy": 0.15803339332342148,
      "num_tokens": 14792042.0,
      "step": 7920
    },
    {
      "entropy": 6.223450803756714,
      "epoch": 0.6998410455669375,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004957870289717142,
      "loss": 6.0063,
      "mean_token_accuracy": 0.1534290134906769,
      "num_tokens": 14801032.0,
      "step": 7925
    },
    {
      "entropy": 6.15586667060852,
      "epoch": 0.7002825856587778,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004957809612776631,
      "loss": 6.0196,
      "mean_token_accuracy": 0.1556229293346405,
      "num_tokens": 14810594.0,
      "step": 7930
    },
    {
      "entropy": 6.19131817817688,
      "epoch": 0.7007241257506182,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004957748892586384,
      "loss": 6.1236,
      "mean_token_accuracy": 0.14985551685094833,
      "num_tokens": 14820570.0,
      "step": 7935
    },
    {
      "entropy": 6.108098936080933,
      "epoch": 0.7011656658424585,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004957688129147594,
      "loss": 5.9444,
      "mean_token_accuracy": 0.16571446657180786,
      "num_tokens": 14830078.0,
      "step": 7940
    },
    {
      "entropy": 6.1342144966125485,
      "epoch": 0.7016072059342988,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004957627322461448,
      "loss": 6.0137,
      "mean_token_accuracy": 0.15698135048151016,
      "num_tokens": 14839974.0,
      "step": 7945
    },
    {
      "entropy": 6.232937955856324,
      "epoch": 0.7020487460261392,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004957566472529139,
      "loss": 6.1025,
      "mean_token_accuracy": 0.14428324848413468,
      "num_tokens": 14849551.0,
      "step": 7950
    },
    {
      "entropy": 6.2377479553222654,
      "epoch": 0.7024902861179795,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004957505579351858,
      "loss": 6.0336,
      "mean_token_accuracy": 0.1549524500966072,
      "num_tokens": 14859011.0,
      "step": 7955
    },
    {
      "entropy": 6.175362968444825,
      "epoch": 0.7029318262098199,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004957444642930798,
      "loss": 5.9942,
      "mean_token_accuracy": 0.15593235790729523,
      "num_tokens": 14867948.0,
      "step": 7960
    },
    {
      "entropy": 6.191630506515503,
      "epoch": 0.7033733663016601,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004957383663267152,
      "loss": 6.051,
      "mean_token_accuracy": 0.15497976690530776,
      "num_tokens": 14877233.0,
      "step": 7965
    },
    {
      "entropy": 6.204564094543457,
      "epoch": 0.7038149063935005,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004957322640362118,
      "loss": 6.069,
      "mean_token_accuracy": 0.14801982194185256,
      "num_tokens": 14887439.0,
      "step": 7970
    },
    {
      "entropy": 6.322771501541138,
      "epoch": 0.7042564464853409,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004957261574216887,
      "loss": 6.1465,
      "mean_token_accuracy": 0.14907680749893187,
      "num_tokens": 14897064.0,
      "step": 7975
    },
    {
      "entropy": 6.244523668289185,
      "epoch": 0.7046979865771812,
      "grad_norm": 1.875,
      "learning_rate": 0.0004957200464832656,
      "loss": 6.1031,
      "mean_token_accuracy": 0.14721770584583282,
      "num_tokens": 14907054.0,
      "step": 7980
    },
    {
      "entropy": 6.220310115814209,
      "epoch": 0.7051395266690216,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004957139312210626,
      "loss": 6.0558,
      "mean_token_accuracy": 0.15061642229557037,
      "num_tokens": 14916969.0,
      "step": 7985
    },
    {
      "entropy": 6.189560461044311,
      "epoch": 0.7055810667608619,
      "grad_norm": 2.34375,
      "learning_rate": 0.000495707811635199,
      "loss": 6.0849,
      "mean_token_accuracy": 0.1534519761800766,
      "num_tokens": 14926404.0,
      "step": 7990
    },
    {
      "entropy": 6.195728206634522,
      "epoch": 0.7060226068527022,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004957016877257949,
      "loss": 6.0637,
      "mean_token_accuracy": 0.14685604795813562,
      "num_tokens": 14935693.0,
      "step": 7995
    },
    {
      "entropy": 6.2307899475097654,
      "epoch": 0.7064641469445425,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004956955594929704,
      "loss": 6.01,
      "mean_token_accuracy": 0.15188159197568893,
      "num_tokens": 14945147.0,
      "step": 8000
    },
    {
      "entropy": 6.178026580810547,
      "epoch": 0.7069056870363829,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004956894269368454,
      "loss": 6.0336,
      "mean_token_accuracy": 0.15496408343315124,
      "num_tokens": 14954022.0,
      "step": 8005
    },
    {
      "entropy": 6.117526531219482,
      "epoch": 0.7073472271282233,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004956832900575399,
      "loss": 6.0382,
      "mean_token_accuracy": 0.15371241867542268,
      "num_tokens": 14962977.0,
      "step": 8010
    },
    {
      "entropy": 6.195847368240356,
      "epoch": 0.7077887672200636,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004956771488551742,
      "loss": 6.0909,
      "mean_token_accuracy": 0.14825759455561638,
      "num_tokens": 14973420.0,
      "step": 8015
    },
    {
      "entropy": 6.267194128036499,
      "epoch": 0.708230307311904,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004956710033298689,
      "loss": 6.035,
      "mean_token_accuracy": 0.15323473662137985,
      "num_tokens": 14981762.0,
      "step": 8020
    },
    {
      "entropy": 6.320011281967163,
      "epoch": 0.7086718474037442,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000495664853481744,
      "loss": 6.1254,
      "mean_token_accuracy": 0.1446717455983162,
      "num_tokens": 14991449.0,
      "step": 8025
    },
    {
      "entropy": 6.248271751403808,
      "epoch": 0.7091133874955846,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004956586993109201,
      "loss": 6.0381,
      "mean_token_accuracy": 0.15330882221460343,
      "num_tokens": 15000802.0,
      "step": 8030
    },
    {
      "entropy": 6.082692623138428,
      "epoch": 0.7095549275874249,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004956525408175179,
      "loss": 6.0586,
      "mean_token_accuracy": 0.15180643945932387,
      "num_tokens": 15010580.0,
      "step": 8035
    },
    {
      "entropy": 6.142378139495849,
      "epoch": 0.7099964676792653,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004956463780016578,
      "loss": 6.0869,
      "mean_token_accuracy": 0.14902941286563873,
      "num_tokens": 15020059.0,
      "step": 8040
    },
    {
      "entropy": 6.221126222610474,
      "epoch": 0.7104380077711057,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004956402108634607,
      "loss": 6.0477,
      "mean_token_accuracy": 0.152384951710701,
      "num_tokens": 15030268.0,
      "step": 8045
    },
    {
      "entropy": 6.179953908920288,
      "epoch": 0.7108795478629459,
      "grad_norm": 2.625,
      "learning_rate": 0.0004956340394030473,
      "loss": 6.0078,
      "mean_token_accuracy": 0.1581128403544426,
      "num_tokens": 15038865.0,
      "step": 8050
    },
    {
      "entropy": 6.272174596786499,
      "epoch": 0.7113210879547863,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004956278636205386,
      "loss": 6.1555,
      "mean_token_accuracy": 0.14396643489599228,
      "num_tokens": 15048427.0,
      "step": 8055
    },
    {
      "entropy": 6.2368580341339115,
      "epoch": 0.7117626280466266,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004956216835160556,
      "loss": 6.0152,
      "mean_token_accuracy": 0.16587988734245301,
      "num_tokens": 15057366.0,
      "step": 8060
    },
    {
      "entropy": 6.218137741088867,
      "epoch": 0.712204168138467,
      "grad_norm": 3.5,
      "learning_rate": 0.0004956154990897191,
      "loss": 6.0775,
      "mean_token_accuracy": 0.14875958263874053,
      "num_tokens": 15067098.0,
      "step": 8065
    },
    {
      "entropy": 6.218372488021851,
      "epoch": 0.7126457082303073,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004956093103416505,
      "loss": 6.062,
      "mean_token_accuracy": 0.15535494238138198,
      "num_tokens": 15075734.0,
      "step": 8070
    },
    {
      "entropy": 6.20798077583313,
      "epoch": 0.7130872483221476,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004956031172719711,
      "loss": 6.1019,
      "mean_token_accuracy": 0.15078829303383828,
      "num_tokens": 15084625.0,
      "step": 8075
    },
    {
      "entropy": 6.084878635406494,
      "epoch": 0.713528788413988,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004955969198808021,
      "loss": 5.9352,
      "mean_token_accuracy": 0.16838667690753936,
      "num_tokens": 15093256.0,
      "step": 8080
    },
    {
      "entropy": 6.141441059112549,
      "epoch": 0.7139703285058283,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004955907181682648,
      "loss": 5.9937,
      "mean_token_accuracy": 0.15782455503940582,
      "num_tokens": 15103175.0,
      "step": 8085
    },
    {
      "entropy": 6.172451734542847,
      "epoch": 0.7144118685976687,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004955845121344809,
      "loss": 6.0842,
      "mean_token_accuracy": 0.154188871383667,
      "num_tokens": 15112789.0,
      "step": 8090
    },
    {
      "entropy": 6.190401649475097,
      "epoch": 0.714853408689509,
      "grad_norm": 2.125,
      "learning_rate": 0.0004955783017795717,
      "loss": 6.1132,
      "mean_token_accuracy": 0.15388823449611663,
      "num_tokens": 15121869.0,
      "step": 8095
    },
    {
      "entropy": 6.15615177154541,
      "epoch": 0.7152949487813494,
      "grad_norm": 2.125,
      "learning_rate": 0.000495572087103659,
      "loss": 5.9973,
      "mean_token_accuracy": 0.15638259947299957,
      "num_tokens": 15131441.0,
      "step": 8100
    },
    {
      "entropy": 6.16906967163086,
      "epoch": 0.7157364888731896,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004955658681068647,
      "loss": 6.0608,
      "mean_token_accuracy": 0.1528923600912094,
      "num_tokens": 15140641.0,
      "step": 8105
    },
    {
      "entropy": 6.263287353515625,
      "epoch": 0.71617802896503,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004955596447893105,
      "loss": 6.0468,
      "mean_token_accuracy": 0.1551143765449524,
      "num_tokens": 15150073.0,
      "step": 8110
    },
    {
      "entropy": 6.11682391166687,
      "epoch": 0.7166195690568704,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004955534171511181,
      "loss": 5.8063,
      "mean_token_accuracy": 0.17062758654356003,
      "num_tokens": 15159879.0,
      "step": 8115
    },
    {
      "entropy": 6.120118474960327,
      "epoch": 0.7170611091487107,
      "grad_norm": 2.015625,
      "learning_rate": 0.00049554718519241,
      "loss": 6.0203,
      "mean_token_accuracy": 0.15730814039707183,
      "num_tokens": 15168316.0,
      "step": 8120
    },
    {
      "entropy": 6.1357043266296385,
      "epoch": 0.7175026492405511,
      "grad_norm": 5.46875,
      "learning_rate": 0.0004955409489133078,
      "loss": 6.0343,
      "mean_token_accuracy": 0.1505425527691841,
      "num_tokens": 15176700.0,
      "step": 8125
    },
    {
      "entropy": 6.227250671386718,
      "epoch": 0.7179441893323913,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004955347083139338,
      "loss": 6.0417,
      "mean_token_accuracy": 0.1585338830947876,
      "num_tokens": 15185997.0,
      "step": 8130
    },
    {
      "entropy": 6.14000449180603,
      "epoch": 0.7183857294242317,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004955284633944104,
      "loss": 6.0411,
      "mean_token_accuracy": 0.15232947319746018,
      "num_tokens": 15195341.0,
      "step": 8135
    },
    {
      "entropy": 6.185124254226684,
      "epoch": 0.718827269516072,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004955222141548597,
      "loss": 6.0902,
      "mean_token_accuracy": 0.15339289903640746,
      "num_tokens": 15205501.0,
      "step": 8140
    },
    {
      "entropy": 6.161205530166626,
      "epoch": 0.7192688096079124,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004955159605954043,
      "loss": 6.0393,
      "mean_token_accuracy": 0.15205711349844933,
      "num_tokens": 15214778.0,
      "step": 8145
    },
    {
      "entropy": 6.311834096908569,
      "epoch": 0.7197103496997528,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004955097027161666,
      "loss": 6.0639,
      "mean_token_accuracy": 0.15708200111985207,
      "num_tokens": 15224201.0,
      "step": 8150
    },
    {
      "entropy": 6.216530847549438,
      "epoch": 0.7201518897915931,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004955034405172693,
      "loss": 6.0027,
      "mean_token_accuracy": 0.16143639236688614,
      "num_tokens": 15233472.0,
      "step": 8155
    },
    {
      "entropy": 6.148335742950439,
      "epoch": 0.7205934298834334,
      "grad_norm": 2.765625,
      "learning_rate": 0.000495497173998835,
      "loss": 6.0696,
      "mean_token_accuracy": 0.15243514031171798,
      "num_tokens": 15242323.0,
      "step": 8160
    },
    {
      "entropy": 6.167079496383667,
      "epoch": 0.7210349699752737,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004954909031609864,
      "loss": 6.0843,
      "mean_token_accuracy": 0.14983757585287094,
      "num_tokens": 15252952.0,
      "step": 8165
    },
    {
      "entropy": 6.204553461074829,
      "epoch": 0.7214765100671141,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004954846280038465,
      "loss": 6.0603,
      "mean_token_accuracy": 0.1506972149014473,
      "num_tokens": 15261815.0,
      "step": 8170
    },
    {
      "entropy": 6.255751132965088,
      "epoch": 0.7219180501589544,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004954783485275381,
      "loss": 6.1082,
      "mean_token_accuracy": 0.15254272818565368,
      "num_tokens": 15271026.0,
      "step": 8175
    },
    {
      "entropy": 6.19788007736206,
      "epoch": 0.7223595902507948,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004954720647321843,
      "loss": 5.9651,
      "mean_token_accuracy": 0.1562256060540676,
      "num_tokens": 15279795.0,
      "step": 8180
    },
    {
      "entropy": 6.155232620239258,
      "epoch": 0.7228011303426352,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004954657766179081,
      "loss": 6.0821,
      "mean_token_accuracy": 0.15056323632597923,
      "num_tokens": 15289113.0,
      "step": 8185
    },
    {
      "entropy": 6.194340467453003,
      "epoch": 0.7232426704344754,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004954594841848327,
      "loss": 5.9395,
      "mean_token_accuracy": 0.16234055683016776,
      "num_tokens": 15298271.0,
      "step": 8190
    },
    {
      "entropy": 6.262794113159179,
      "epoch": 0.7236842105263158,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004954531874330814,
      "loss": 6.0269,
      "mean_token_accuracy": 0.15222135111689566,
      "num_tokens": 15307677.0,
      "step": 8195
    },
    {
      "entropy": 6.061808633804321,
      "epoch": 0.7241257506181561,
      "grad_norm": 2.25,
      "learning_rate": 0.0004954468863627774,
      "loss": 6.0352,
      "mean_token_accuracy": 0.1541186660528183,
      "num_tokens": 15316637.0,
      "step": 8200
    },
    {
      "entropy": 6.157130289077759,
      "epoch": 0.7245672907099965,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004954405809740445,
      "loss": 6.0196,
      "mean_token_accuracy": 0.15551941692829133,
      "num_tokens": 15325221.0,
      "step": 8205
    },
    {
      "entropy": 6.266123628616333,
      "epoch": 0.7250088308018368,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004954342712670058,
      "loss": 6.0362,
      "mean_token_accuracy": 0.1526273787021637,
      "num_tokens": 15335877.0,
      "step": 8210
    },
    {
      "entropy": 6.191796398162841,
      "epoch": 0.7254503708936771,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004954279572417852,
      "loss": 6.0759,
      "mean_token_accuracy": 0.1485827647149563,
      "num_tokens": 15345096.0,
      "step": 8215
    },
    {
      "entropy": 6.133485746383667,
      "epoch": 0.7258919109855175,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004954216388985063,
      "loss": 5.9009,
      "mean_token_accuracy": 0.16367556452751159,
      "num_tokens": 15353785.0,
      "step": 8220
    },
    {
      "entropy": 6.198287105560302,
      "epoch": 0.7263334510773578,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004954153162372928,
      "loss": 5.9914,
      "mean_token_accuracy": 0.15659495443105698,
      "num_tokens": 15362761.0,
      "step": 8225
    },
    {
      "entropy": 6.200320911407471,
      "epoch": 0.7267749911691982,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004954089892582686,
      "loss": 6.0033,
      "mean_token_accuracy": 0.14984651207923888,
      "num_tokens": 15371951.0,
      "step": 8230
    },
    {
      "entropy": 6.1897303581237795,
      "epoch": 0.7272165312610385,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004954026579615577,
      "loss": 6.0632,
      "mean_token_accuracy": 0.1566293716430664,
      "num_tokens": 15380277.0,
      "step": 8235
    },
    {
      "entropy": 6.193405818939209,
      "epoch": 0.7276580713528789,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004953963223472841,
      "loss": 6.062,
      "mean_token_accuracy": 0.15038451701402664,
      "num_tokens": 15390951.0,
      "step": 8240
    },
    {
      "entropy": 6.28805890083313,
      "epoch": 0.7280996114447191,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004953899824155718,
      "loss": 6.0329,
      "mean_token_accuracy": 0.15698810070753097,
      "num_tokens": 15399394.0,
      "step": 8245
    },
    {
      "entropy": 6.1586767673492435,
      "epoch": 0.7285411515365595,
      "grad_norm": 2.5,
      "learning_rate": 0.0004953836381665452,
      "loss": 5.9735,
      "mean_token_accuracy": 0.1548301711678505,
      "num_tokens": 15408644.0,
      "step": 8250
    },
    {
      "entropy": 6.174379491806031,
      "epoch": 0.7289826916283999,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004953772896003284,
      "loss": 6.0737,
      "mean_token_accuracy": 0.15424104779958725,
      "num_tokens": 15417398.0,
      "step": 8255
    },
    {
      "entropy": 6.245291805267334,
      "epoch": 0.7294242317202402,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004953709367170458,
      "loss": 6.1137,
      "mean_token_accuracy": 0.1443786732852459,
      "num_tokens": 15426532.0,
      "step": 8260
    },
    {
      "entropy": 6.244936513900757,
      "epoch": 0.7298657718120806,
      "grad_norm": 2.953125,
      "learning_rate": 0.000495364579516822,
      "loss": 6.095,
      "mean_token_accuracy": 0.14691477119922638,
      "num_tokens": 15436371.0,
      "step": 8265
    },
    {
      "entropy": 6.3124534606933596,
      "epoch": 0.7303073119039208,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004953582179997813,
      "loss": 6.1494,
      "mean_token_accuracy": 0.14667709469795226,
      "num_tokens": 15446600.0,
      "step": 8270
    },
    {
      "entropy": 6.254750108718872,
      "epoch": 0.7307488519957612,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004953518521660484,
      "loss": 5.9877,
      "mean_token_accuracy": 0.1541195407509804,
      "num_tokens": 15455547.0,
      "step": 8275
    },
    {
      "entropy": 6.181308889389038,
      "epoch": 0.7311903920876015,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004953454820157481,
      "loss": 6.0464,
      "mean_token_accuracy": 0.15462103337049485,
      "num_tokens": 15464543.0,
      "step": 8280
    },
    {
      "entropy": 6.208321475982666,
      "epoch": 0.7316319321794419,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004953391075490051,
      "loss": 6.1598,
      "mean_token_accuracy": 0.1457721471786499,
      "num_tokens": 15473626.0,
      "step": 8285
    },
    {
      "entropy": 6.221294593811035,
      "epoch": 0.7320734722712823,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004953327287659444,
      "loss": 6.0933,
      "mean_token_accuracy": 0.1486469939351082,
      "num_tokens": 15484164.0,
      "step": 8290
    },
    {
      "entropy": 6.314439582824707,
      "epoch": 0.7325150123631226,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004953263456666907,
      "loss": 6.102,
      "mean_token_accuracy": 0.14443486854434012,
      "num_tokens": 15493718.0,
      "step": 8295
    },
    {
      "entropy": 6.220274448394775,
      "epoch": 0.7329565524549629,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004953199582513693,
      "loss": 6.0242,
      "mean_token_accuracy": 0.15442363172769547,
      "num_tokens": 15503339.0,
      "step": 8300
    },
    {
      "entropy": 6.215155076980591,
      "epoch": 0.7333980925468032,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004953135665201052,
      "loss": 6.006,
      "mean_token_accuracy": 0.1588119901716709,
      "num_tokens": 15513314.0,
      "step": 8305
    },
    {
      "entropy": 6.217810535430909,
      "epoch": 0.7338396326386436,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004953071704730237,
      "loss": 6.0986,
      "mean_token_accuracy": 0.14826320856809616,
      "num_tokens": 15522943.0,
      "step": 8310
    },
    {
      "entropy": 6.1630918979644775,
      "epoch": 0.7342811727304839,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004953007701102499,
      "loss": 6.0575,
      "mean_token_accuracy": 0.1495142787694931,
      "num_tokens": 15531975.0,
      "step": 8315
    },
    {
      "entropy": 6.1869665622711185,
      "epoch": 0.7347227128223243,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004952943654319094,
      "loss": 5.9788,
      "mean_token_accuracy": 0.15272131115198134,
      "num_tokens": 15541139.0,
      "step": 8320
    },
    {
      "entropy": 6.212423419952392,
      "epoch": 0.7351642529141647,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004952879564381276,
      "loss": 6.0591,
      "mean_token_accuracy": 0.15568502247333527,
      "num_tokens": 15550604.0,
      "step": 8325
    },
    {
      "entropy": 6.161788749694824,
      "epoch": 0.7356057930060049,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004952815431290298,
      "loss": 5.9785,
      "mean_token_accuracy": 0.15971437096595764,
      "num_tokens": 15558995.0,
      "step": 8330
    },
    {
      "entropy": 6.235270690917969,
      "epoch": 0.7360473330978453,
      "grad_norm": 2.921875,
      "learning_rate": 0.000495275125504742,
      "loss": 6.0522,
      "mean_token_accuracy": 0.15424182116985322,
      "num_tokens": 15568229.0,
      "step": 8335
    },
    {
      "entropy": 6.141129970550537,
      "epoch": 0.7364888731896856,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004952687035653899,
      "loss": 5.9708,
      "mean_token_accuracy": 0.15418651700019836,
      "num_tokens": 15577042.0,
      "step": 8340
    },
    {
      "entropy": 6.1271415710449215,
      "epoch": 0.736930413281526,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004952622773110989,
      "loss": 5.9139,
      "mean_token_accuracy": 0.16094250828027726,
      "num_tokens": 15585825.0,
      "step": 8345
    },
    {
      "entropy": 6.182744836807251,
      "epoch": 0.7373719533733663,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004952558467419953,
      "loss": 5.9343,
      "mean_token_accuracy": 0.15798150151968002,
      "num_tokens": 15595486.0,
      "step": 8350
    },
    {
      "entropy": 6.092209577560425,
      "epoch": 0.7378134934652066,
      "grad_norm": 1.8671875,
      "learning_rate": 0.000495249411858205,
      "loss": 5.9585,
      "mean_token_accuracy": 0.14684529975056648,
      "num_tokens": 15605312.0,
      "step": 8355
    },
    {
      "entropy": 6.1702048778533936,
      "epoch": 0.738255033557047,
      "grad_norm": 3.625,
      "learning_rate": 0.0004952429726598538,
      "loss": 5.9953,
      "mean_token_accuracy": 0.15861203223466874,
      "num_tokens": 15615185.0,
      "step": 8360
    },
    {
      "entropy": 6.253500270843506,
      "epoch": 0.7386965736488873,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004952365291470682,
      "loss": 6.0975,
      "mean_token_accuracy": 0.153415647149086,
      "num_tokens": 15625259.0,
      "step": 8365
    },
    {
      "entropy": 6.236084794998169,
      "epoch": 0.7391381137407277,
      "grad_norm": 1.875,
      "learning_rate": 0.0004952300813199742,
      "loss": 6.0742,
      "mean_token_accuracy": 0.15204451978206635,
      "num_tokens": 15635040.0,
      "step": 8370
    },
    {
      "entropy": 6.262276887893677,
      "epoch": 0.739579653832568,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004952236291786981,
      "loss": 6.0712,
      "mean_token_accuracy": 0.15011000037193298,
      "num_tokens": 15644996.0,
      "step": 8375
    },
    {
      "entropy": 6.096345233917236,
      "epoch": 0.7400211939244083,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004952171727233665,
      "loss": 6.0005,
      "mean_token_accuracy": 0.1511855512857437,
      "num_tokens": 15654598.0,
      "step": 8380
    },
    {
      "entropy": 6.156092548370362,
      "epoch": 0.7404627340162486,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004952107119541055,
      "loss": 6.0067,
      "mean_token_accuracy": 0.1565088838338852,
      "num_tokens": 15664213.0,
      "step": 8385
    },
    {
      "entropy": 6.1887593269348145,
      "epoch": 0.740904274108089,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004952042468710421,
      "loss": 6.0605,
      "mean_token_accuracy": 0.1537364087998867,
      "num_tokens": 15673595.0,
      "step": 8390
    },
    {
      "entropy": 6.240141248703003,
      "epoch": 0.7413458141999294,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004951977774743027,
      "loss": 6.1115,
      "mean_token_accuracy": 0.14618029668927193,
      "num_tokens": 15682753.0,
      "step": 8395
    },
    {
      "entropy": 6.133825016021729,
      "epoch": 0.7417873542917697,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004951913037640139,
      "loss": 5.9389,
      "mean_token_accuracy": 0.16434774845838546,
      "num_tokens": 15691508.0,
      "step": 8400
    },
    {
      "entropy": 6.120059823989868,
      "epoch": 0.7422288943836101,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004951848257403029,
      "loss": 5.9389,
      "mean_token_accuracy": 0.15547937601804734,
      "num_tokens": 15701110.0,
      "step": 8405
    },
    {
      "entropy": 6.063517713546753,
      "epoch": 0.7426704344754503,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004951783434032963,
      "loss": 5.8822,
      "mean_token_accuracy": 0.16648623794317247,
      "num_tokens": 15710565.0,
      "step": 8410
    },
    {
      "entropy": 6.129230642318726,
      "epoch": 0.7431119745672907,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004951718567531211,
      "loss": 6.0531,
      "mean_token_accuracy": 0.14975019171833992,
      "num_tokens": 15718364.0,
      "step": 8415
    },
    {
      "entropy": 6.169689178466797,
      "epoch": 0.743553514659131,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004951653657899045,
      "loss": 6.1328,
      "mean_token_accuracy": 0.14661180898547171,
      "num_tokens": 15729127.0,
      "step": 8420
    },
    {
      "entropy": 6.192731428146362,
      "epoch": 0.7439950547509714,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004951588705137737,
      "loss": 6.1017,
      "mean_token_accuracy": 0.1482056498527527,
      "num_tokens": 15738491.0,
      "step": 8425
    },
    {
      "entropy": 6.24168210029602,
      "epoch": 0.7444365948428118,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004951523709248557,
      "loss": 5.9737,
      "mean_token_accuracy": 0.15786256790161132,
      "num_tokens": 15748346.0,
      "step": 8430
    },
    {
      "entropy": 6.125617933273316,
      "epoch": 0.744878134934652,
      "grad_norm": 2.078125,
      "learning_rate": 0.000495145867023278,
      "loss": 6.0018,
      "mean_token_accuracy": 0.1626955583691597,
      "num_tokens": 15757242.0,
      "step": 8435
    },
    {
      "entropy": 6.2744465351104735,
      "epoch": 0.7453196750264924,
      "grad_norm": 2.640625,
      "learning_rate": 0.000495139358809168,
      "loss": 6.0348,
      "mean_token_accuracy": 0.15362615138292313,
      "num_tokens": 15767236.0,
      "step": 8440
    },
    {
      "entropy": 6.171437835693359,
      "epoch": 0.7457612151183327,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004951328462826532,
      "loss": 6.0764,
      "mean_token_accuracy": 0.14767611175775527,
      "num_tokens": 15776977.0,
      "step": 8445
    },
    {
      "entropy": 6.1203724384307865,
      "epoch": 0.7462027552101731,
      "grad_norm": 13.5625,
      "learning_rate": 0.0004951263294438611,
      "loss": 5.9988,
      "mean_token_accuracy": 0.1552292600274086,
      "num_tokens": 15787333.0,
      "step": 8450
    },
    {
      "entropy": 6.117383146286011,
      "epoch": 0.7466442953020134,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004951198082929194,
      "loss": 5.8902,
      "mean_token_accuracy": 0.1694972649216652,
      "num_tokens": 15794773.0,
      "step": 8455
    },
    {
      "entropy": 6.115877676010132,
      "epoch": 0.7470858353938538,
      "grad_norm": 2.375,
      "learning_rate": 0.0004951132828299558,
      "loss": 5.994,
      "mean_token_accuracy": 0.16377333998680116,
      "num_tokens": 15803803.0,
      "step": 8460
    },
    {
      "entropy": 6.182858276367187,
      "epoch": 0.7475273754856941,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004951067530550982,
      "loss": 6.0366,
      "mean_token_accuracy": 0.15338488817214965,
      "num_tokens": 15814021.0,
      "step": 8465
    },
    {
      "entropy": 6.186029148101807,
      "epoch": 0.7479689155775344,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004951002189684745,
      "loss": 6.0211,
      "mean_token_accuracy": 0.14898501634597777,
      "num_tokens": 15822974.0,
      "step": 8470
    },
    {
      "entropy": 6.1186785221099855,
      "epoch": 0.7484104556693748,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004950936805702127,
      "loss": 5.9307,
      "mean_token_accuracy": 0.16209083497524263,
      "num_tokens": 15832202.0,
      "step": 8475
    },
    {
      "entropy": 6.187338018417359,
      "epoch": 0.7488519957612151,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004950871378604409,
      "loss": 6.033,
      "mean_token_accuracy": 0.15920102149248122,
      "num_tokens": 15841223.0,
      "step": 8480
    },
    {
      "entropy": 6.17196831703186,
      "epoch": 0.7492935358530555,
      "grad_norm": 2.25,
      "learning_rate": 0.0004950805908392872,
      "loss": 5.9752,
      "mean_token_accuracy": 0.1617097333073616,
      "num_tokens": 15850260.0,
      "step": 8485
    },
    {
      "entropy": 6.116999387741089,
      "epoch": 0.7497350759448957,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004950740395068799,
      "loss": 6.0575,
      "mean_token_accuracy": 0.15411865711212158,
      "num_tokens": 15859780.0,
      "step": 8490
    },
    {
      "entropy": 6.143398094177246,
      "epoch": 0.7501766160367361,
      "grad_norm": 2.25,
      "learning_rate": 0.0004950674838633473,
      "loss": 5.9971,
      "mean_token_accuracy": 0.15415159314870835,
      "num_tokens": 15868331.0,
      "step": 8495
    },
    {
      "entropy": 6.169262361526489,
      "epoch": 0.7506181561285765,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004950609239088178,
      "loss": 6.0497,
      "mean_token_accuracy": 0.14991126954555511,
      "num_tokens": 15877359.0,
      "step": 8500
    },
    {
      "entropy": 6.193778324127197,
      "epoch": 0.7510596962204168,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004950543596434201,
      "loss": 5.9772,
      "mean_token_accuracy": 0.16402651369571686,
      "num_tokens": 15886976.0,
      "step": 8505
    },
    {
      "entropy": 6.131670570373535,
      "epoch": 0.7515012363122572,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004950477910672825,
      "loss": 6.0348,
      "mean_token_accuracy": 0.15271446779370307,
      "num_tokens": 15896139.0,
      "step": 8510
    },
    {
      "entropy": 6.180541944503784,
      "epoch": 0.7519427764040975,
      "grad_norm": 2.203125,
      "learning_rate": 0.000495041218180534,
      "loss": 6.0177,
      "mean_token_accuracy": 0.16259529143571855,
      "num_tokens": 15904672.0,
      "step": 8515
    },
    {
      "entropy": 6.059027624130249,
      "epoch": 0.7523843164959378,
      "grad_norm": 2.15625,
      "learning_rate": 0.000495034640983303,
      "loss": 5.8731,
      "mean_token_accuracy": 0.16886622160673143,
      "num_tokens": 15912716.0,
      "step": 8520
    },
    {
      "entropy": 6.12007884979248,
      "epoch": 0.7528258565877781,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004950280594757186,
      "loss": 5.9946,
      "mean_token_accuracy": 0.15706919729709626,
      "num_tokens": 15921974.0,
      "step": 8525
    },
    {
      "entropy": 6.191877841949463,
      "epoch": 0.7532673966796185,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004950214736579097,
      "loss": 5.9961,
      "mean_token_accuracy": 0.1594957433640957,
      "num_tokens": 15930859.0,
      "step": 8530
    },
    {
      "entropy": 6.105285406112671,
      "epoch": 0.7537089367714589,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004950148835300051,
      "loss": 5.933,
      "mean_token_accuracy": 0.15683687925338746,
      "num_tokens": 15940052.0,
      "step": 8535
    },
    {
      "entropy": 6.158630323410034,
      "epoch": 0.7541504768632992,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004950082890921341,
      "loss": 5.998,
      "mean_token_accuracy": 0.15748081728816032,
      "num_tokens": 15948422.0,
      "step": 8540
    },
    {
      "entropy": 6.182331562042236,
      "epoch": 0.7545920169551396,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004950016903444261,
      "loss": 6.0974,
      "mean_token_accuracy": 0.15164448618888854,
      "num_tokens": 15957433.0,
      "step": 8545
    },
    {
      "entropy": 6.238806486129761,
      "epoch": 0.7550335570469798,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004949950872870099,
      "loss": 6.0159,
      "mean_token_accuracy": 0.15430637300014496,
      "num_tokens": 15966542.0,
      "step": 8550
    },
    {
      "entropy": 6.178037452697754,
      "epoch": 0.7554750971388202,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000494988479920015,
      "loss": 6.1505,
      "mean_token_accuracy": 0.1452166497707367,
      "num_tokens": 15977067.0,
      "step": 8555
    },
    {
      "entropy": 6.255996131896973,
      "epoch": 0.7559166372306605,
      "grad_norm": 2.515625,
      "learning_rate": 0.000494981868243571,
      "loss": 6.0632,
      "mean_token_accuracy": 0.15499598830938338,
      "num_tokens": 15986502.0,
      "step": 8560
    },
    {
      "entropy": 6.271188640594483,
      "epoch": 0.7563581773225009,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004949752522578073,
      "loss": 6.1111,
      "mean_token_accuracy": 0.15028804168105125,
      "num_tokens": 15996788.0,
      "step": 8565
    },
    {
      "entropy": 6.170350408554077,
      "epoch": 0.7567997174143413,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004949686319628535,
      "loss": 6.0201,
      "mean_token_accuracy": 0.15122852921485902,
      "num_tokens": 16007212.0,
      "step": 8570
    },
    {
      "entropy": 6.107031393051147,
      "epoch": 0.7572412575061815,
      "grad_norm": 2.0,
      "learning_rate": 0.0004949620073588394,
      "loss": 6.0547,
      "mean_token_accuracy": 0.15020963698625564,
      "num_tokens": 16017000.0,
      "step": 8575
    },
    {
      "entropy": 6.2585564136505125,
      "epoch": 0.7576827975980219,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004949553784458945,
      "loss": 6.1396,
      "mean_token_accuracy": 0.14561762511730195,
      "num_tokens": 16026647.0,
      "step": 8580
    },
    {
      "entropy": 6.2295958518981935,
      "epoch": 0.7581243376898622,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004949487452241489,
      "loss": 5.8828,
      "mean_token_accuracy": 0.16303810477256775,
      "num_tokens": 16034773.0,
      "step": 8585
    },
    {
      "entropy": 6.237812662124634,
      "epoch": 0.7585658777817026,
      "grad_norm": 2.625,
      "learning_rate": 0.0004949421076937325,
      "loss": 6.0022,
      "mean_token_accuracy": 0.15621815025806426,
      "num_tokens": 16043787.0,
      "step": 8590
    },
    {
      "entropy": 6.059094953536987,
      "epoch": 0.7590074178735429,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004949354658547753,
      "loss": 5.9761,
      "mean_token_accuracy": 0.16451401114463807,
      "num_tokens": 16053130.0,
      "step": 8595
    },
    {
      "entropy": 6.142942190170288,
      "epoch": 0.7594489579653833,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004949288197074074,
      "loss": 6.1217,
      "mean_token_accuracy": 0.1474703349173069,
      "num_tokens": 16062529.0,
      "step": 8600
    },
    {
      "entropy": 6.1555428981781,
      "epoch": 0.7598904980572236,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004949221692517589,
      "loss": 5.9824,
      "mean_token_accuracy": 0.14887360334396363,
      "num_tokens": 16071932.0,
      "step": 8605
    },
    {
      "entropy": 6.282268905639649,
      "epoch": 0.7603320381490639,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004949155144879602,
      "loss": 5.9612,
      "mean_token_accuracy": 0.15913195461034774,
      "num_tokens": 16081077.0,
      "step": 8610
    },
    {
      "entropy": 6.123275375366211,
      "epoch": 0.7607735782409043,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004949088554161417,
      "loss": 5.8976,
      "mean_token_accuracy": 0.16462857276201248,
      "num_tokens": 16089725.0,
      "step": 8615
    },
    {
      "entropy": 6.1253915309906,
      "epoch": 0.7612151183327446,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004949021920364339,
      "loss": 6.0818,
      "mean_token_accuracy": 0.1539457008242607,
      "num_tokens": 16099362.0,
      "step": 8620
    },
    {
      "entropy": 6.263493871688842,
      "epoch": 0.761656658424585,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004948955243489671,
      "loss": 6.1044,
      "mean_token_accuracy": 0.15207269489765168,
      "num_tokens": 16108746.0,
      "step": 8625
    },
    {
      "entropy": 6.254348278045654,
      "epoch": 0.7620981985164252,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004948888523538721,
      "loss": 6.2109,
      "mean_token_accuracy": 0.1440314069390297,
      "num_tokens": 16118902.0,
      "step": 8630
    },
    {
      "entropy": 6.151745843887329,
      "epoch": 0.7625397386082656,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004948821760512795,
      "loss": 5.9669,
      "mean_token_accuracy": 0.1618361622095108,
      "num_tokens": 16127781.0,
      "step": 8635
    },
    {
      "entropy": 6.161589670181274,
      "epoch": 0.762981278700106,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004948754954413201,
      "loss": 5.9428,
      "mean_token_accuracy": 0.1669818416237831,
      "num_tokens": 16136004.0,
      "step": 8640
    },
    {
      "entropy": 6.114984655380249,
      "epoch": 0.7634228187919463,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004948688105241248,
      "loss": 5.9832,
      "mean_token_accuracy": 0.16015140414237977,
      "num_tokens": 16145023.0,
      "step": 8645
    },
    {
      "entropy": 6.139699029922485,
      "epoch": 0.7638643588837867,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004948621212998246,
      "loss": 6.0611,
      "mean_token_accuracy": 0.15055262818932533,
      "num_tokens": 16154242.0,
      "step": 8650
    },
    {
      "entropy": 6.218677186965943,
      "epoch": 0.764305898975627,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004948554277685505,
      "loss": 6.0142,
      "mean_token_accuracy": 0.14951191693544388,
      "num_tokens": 16163252.0,
      "step": 8655
    },
    {
      "entropy": 6.244760608673095,
      "epoch": 0.7647474390674673,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004948487299304335,
      "loss": 5.9431,
      "mean_token_accuracy": 0.15852085798978804,
      "num_tokens": 16171363.0,
      "step": 8660
    },
    {
      "entropy": 6.172284317016602,
      "epoch": 0.7651889791593076,
      "grad_norm": 2.421875,
      "learning_rate": 0.000494842027785605,
      "loss": 6.1235,
      "mean_token_accuracy": 0.15257852524518967,
      "num_tokens": 16181234.0,
      "step": 8665
    },
    {
      "entropy": 6.188773202896118,
      "epoch": 0.765630519251148,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004948353213341961,
      "loss": 6.0368,
      "mean_token_accuracy": 0.1514037013053894,
      "num_tokens": 16189420.0,
      "step": 8670
    },
    {
      "entropy": 6.278617191314697,
      "epoch": 0.7660720593429884,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004948286105763385,
      "loss": 6.2021,
      "mean_token_accuracy": 0.14024263024330139,
      "num_tokens": 16199202.0,
      "step": 8675
    },
    {
      "entropy": 6.261602783203125,
      "epoch": 0.7665135994348287,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004948218955121632,
      "loss": 6.033,
      "mean_token_accuracy": 0.15275065898895263,
      "num_tokens": 16208481.0,
      "step": 8680
    },
    {
      "entropy": 6.184469747543335,
      "epoch": 0.766955139526669,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004948151761418021,
      "loss": 6.0913,
      "mean_token_accuracy": 0.15930637568235398,
      "num_tokens": 16217282.0,
      "step": 8685
    },
    {
      "entropy": 6.13504114151001,
      "epoch": 0.7673966796185093,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004948084524653868,
      "loss": 5.9651,
      "mean_token_accuracy": 0.15824726969003677,
      "num_tokens": 16226567.0,
      "step": 8690
    },
    {
      "entropy": 6.193227529525757,
      "epoch": 0.7678382197103497,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004948017244830489,
      "loss": 6.1761,
      "mean_token_accuracy": 0.14736137688159942,
      "num_tokens": 16236255.0,
      "step": 8695
    },
    {
      "entropy": 6.180487108230591,
      "epoch": 0.76827975980219,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004947949921949201,
      "loss": 5.9276,
      "mean_token_accuracy": 0.16612430810928344,
      "num_tokens": 16244838.0,
      "step": 8700
    },
    {
      "entropy": 6.048501920700073,
      "epoch": 0.7687212998940304,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004947882556011325,
      "loss": 5.864,
      "mean_token_accuracy": 0.16519371271133423,
      "num_tokens": 16254101.0,
      "step": 8705
    },
    {
      "entropy": 6.045390367507935,
      "epoch": 0.7691628399858708,
      "grad_norm": 1.734375,
      "learning_rate": 0.000494781514701818,
      "loss": 5.9404,
      "mean_token_accuracy": 0.1638119101524353,
      "num_tokens": 16263496.0,
      "step": 8710
    },
    {
      "entropy": 6.197959566116333,
      "epoch": 0.769604380077711,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004947747694971085,
      "loss": 6.018,
      "mean_token_accuracy": 0.15746343284845352,
      "num_tokens": 16271884.0,
      "step": 8715
    },
    {
      "entropy": 6.186513757705688,
      "epoch": 0.7700459201695514,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004947680199871363,
      "loss": 5.9421,
      "mean_token_accuracy": 0.16937040835618972,
      "num_tokens": 16281197.0,
      "step": 8720
    },
    {
      "entropy": 6.087084770202637,
      "epoch": 0.7704874602613917,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004947612661720336,
      "loss": 6.0195,
      "mean_token_accuracy": 0.1552564576268196,
      "num_tokens": 16289908.0,
      "step": 8725
    },
    {
      "entropy": 6.107022476196289,
      "epoch": 0.7709290003532321,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004947545080519327,
      "loss": 6.0016,
      "mean_token_accuracy": 0.15504290387034417,
      "num_tokens": 16299237.0,
      "step": 8730
    },
    {
      "entropy": 6.244728994369507,
      "epoch": 0.7713705404450724,
      "grad_norm": 1.9609375,
      "learning_rate": 0.000494747745626966,
      "loss": 6.0161,
      "mean_token_accuracy": 0.1612411007285118,
      "num_tokens": 16308175.0,
      "step": 8735
    },
    {
      "entropy": 6.150093460083008,
      "epoch": 0.7718120805369127,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004947409788972659,
      "loss": 5.9777,
      "mean_token_accuracy": 0.1611553907394409,
      "num_tokens": 16317121.0,
      "step": 8740
    },
    {
      "entropy": 6.105035400390625,
      "epoch": 0.7722536206287531,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004947342078629649,
      "loss": 5.9399,
      "mean_token_accuracy": 0.15897390842437745,
      "num_tokens": 16325764.0,
      "step": 8745
    },
    {
      "entropy": 6.109813976287842,
      "epoch": 0.7726951607205934,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004947274325241959,
      "loss": 5.9563,
      "mean_token_accuracy": 0.15956080257892608,
      "num_tokens": 16333823.0,
      "step": 8750
    },
    {
      "entropy": 6.125399732589722,
      "epoch": 0.7731367008124338,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004947206528810913,
      "loss": 5.9579,
      "mean_token_accuracy": 0.16569003760814666,
      "num_tokens": 16341918.0,
      "step": 8755
    },
    {
      "entropy": 6.074509716033935,
      "epoch": 0.7735782409042741,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004947138689337843,
      "loss": 6.0194,
      "mean_token_accuracy": 0.15725950747728348,
      "num_tokens": 16351904.0,
      "step": 8760
    },
    {
      "entropy": 6.157044219970703,
      "epoch": 0.7740197809961145,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004947070806824074,
      "loss": 6.0175,
      "mean_token_accuracy": 0.15132663100957872,
      "num_tokens": 16360921.0,
      "step": 8765
    },
    {
      "entropy": 6.202144718170166,
      "epoch": 0.7744613210879547,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004947002881270937,
      "loss": 6.0581,
      "mean_token_accuracy": 0.14726337194442748,
      "num_tokens": 16370929.0,
      "step": 8770
    },
    {
      "entropy": 6.24249849319458,
      "epoch": 0.7749028611797951,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004946934912679764,
      "loss": 6.0456,
      "mean_token_accuracy": 0.1561698243021965,
      "num_tokens": 16380099.0,
      "step": 8775
    },
    {
      "entropy": 6.2365148067474365,
      "epoch": 0.7753444012716355,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004946866901051885,
      "loss": 6.0564,
      "mean_token_accuracy": 0.15523979514837266,
      "num_tokens": 16389109.0,
      "step": 8780
    },
    {
      "entropy": 6.14391942024231,
      "epoch": 0.7757859413634758,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004946798846388634,
      "loss": 6.1027,
      "mean_token_accuracy": 0.15341169089078904,
      "num_tokens": 16398601.0,
      "step": 8785
    },
    {
      "entropy": 6.095554780960083,
      "epoch": 0.7762274814553162,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004946730748691342,
      "loss": 5.9552,
      "mean_token_accuracy": 0.16528091281652452,
      "num_tokens": 16406950.0,
      "step": 8790
    },
    {
      "entropy": 6.209650373458862,
      "epoch": 0.7766690215471564,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004946662607961344,
      "loss": 5.9834,
      "mean_token_accuracy": 0.1604272246360779,
      "num_tokens": 16414680.0,
      "step": 8795
    },
    {
      "entropy": 6.168493175506592,
      "epoch": 0.7771105616389968,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004946594424199976,
      "loss": 6.0461,
      "mean_token_accuracy": 0.1582956314086914,
      "num_tokens": 16424060.0,
      "step": 8800
    },
    {
      "entropy": 6.140843391418457,
      "epoch": 0.7775521017308371,
      "grad_norm": 1.875,
      "learning_rate": 0.0004946526197408571,
      "loss": 6.0426,
      "mean_token_accuracy": 0.15482948273420333,
      "num_tokens": 16433691.0,
      "step": 8805
    },
    {
      "entropy": 6.209283351898193,
      "epoch": 0.7779936418226775,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004946457927588467,
      "loss": 6.0597,
      "mean_token_accuracy": 0.15235007852315902,
      "num_tokens": 16443234.0,
      "step": 8810
    },
    {
      "entropy": 6.144260358810425,
      "epoch": 0.7784351819145179,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004946389614741002,
      "loss": 6.0082,
      "mean_token_accuracy": 0.16013663113117219,
      "num_tokens": 16452084.0,
      "step": 8815
    },
    {
      "entropy": 6.145121288299561,
      "epoch": 0.7788767220063582,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004946321258867513,
      "loss": 5.9754,
      "mean_token_accuracy": 0.1602022811770439,
      "num_tokens": 16460135.0,
      "step": 8820
    },
    {
      "entropy": 6.144701051712036,
      "epoch": 0.7793182620981985,
      "grad_norm": 2.015625,
      "learning_rate": 0.000494625285996934,
      "loss": 6.1058,
      "mean_token_accuracy": 0.15144651457667352,
      "num_tokens": 16469919.0,
      "step": 8825
    },
    {
      "entropy": 6.140231657028198,
      "epoch": 0.7797598021900388,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004946184418047823,
      "loss": 6.022,
      "mean_token_accuracy": 0.15065136328339576,
      "num_tokens": 16478473.0,
      "step": 8830
    },
    {
      "entropy": 6.235374116897583,
      "epoch": 0.7802013422818792,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004946115933104301,
      "loss": 6.0279,
      "mean_token_accuracy": 0.1558874547481537,
      "num_tokens": 16486878.0,
      "step": 8835
    },
    {
      "entropy": 6.199049854278565,
      "epoch": 0.7806428823737195,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004946047405140119,
      "loss": 6.0759,
      "mean_token_accuracy": 0.1492841623723507,
      "num_tokens": 16496798.0,
      "step": 8840
    },
    {
      "entropy": 6.202003431320191,
      "epoch": 0.7810844224655599,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004945978834156616,
      "loss": 5.994,
      "mean_token_accuracy": 0.16024575084447862,
      "num_tokens": 16506508.0,
      "step": 8845
    },
    {
      "entropy": 6.0723021030426025,
      "epoch": 0.7815259625574003,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004945910220155136,
      "loss": 5.9722,
      "mean_token_accuracy": 0.1620054602622986,
      "num_tokens": 16514979.0,
      "step": 8850
    },
    {
      "entropy": 6.188809442520141,
      "epoch": 0.7819675026492405,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004945841563137025,
      "loss": 6.0436,
      "mean_token_accuracy": 0.15457658618688583,
      "num_tokens": 16524654.0,
      "step": 8855
    },
    {
      "entropy": 6.233149766921997,
      "epoch": 0.7824090427410809,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004945772863103626,
      "loss": 6.1173,
      "mean_token_accuracy": 0.1509366050362587,
      "num_tokens": 16534486.0,
      "step": 8860
    },
    {
      "entropy": 6.245261812210083,
      "epoch": 0.7828505828329212,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004945704120056286,
      "loss": 6.0401,
      "mean_token_accuracy": 0.16267454922199248,
      "num_tokens": 16543718.0,
      "step": 8865
    },
    {
      "entropy": 6.09795355796814,
      "epoch": 0.7832921229247616,
      "grad_norm": 2.21875,
      "learning_rate": 0.000494563533399635,
      "loss": 5.9766,
      "mean_token_accuracy": 0.1589898332953453,
      "num_tokens": 16553159.0,
      "step": 8870
    },
    {
      "entropy": 6.214016628265381,
      "epoch": 0.7837336630166019,
      "grad_norm": 3.5,
      "learning_rate": 0.0004945566504925167,
      "loss": 6.0601,
      "mean_token_accuracy": 0.15150932148098945,
      "num_tokens": 16562398.0,
      "step": 8875
    },
    {
      "entropy": 6.241987991333008,
      "epoch": 0.7841752031084422,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004945497632844084,
      "loss": 5.9994,
      "mean_token_accuracy": 0.1608886316418648,
      "num_tokens": 16571530.0,
      "step": 8880
    },
    {
      "entropy": 6.15855860710144,
      "epoch": 0.7846167432002826,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004945428717754453,
      "loss": 6.0128,
      "mean_token_accuracy": 0.1525505930185318,
      "num_tokens": 16581074.0,
      "step": 8885
    },
    {
      "entropy": 6.088472747802735,
      "epoch": 0.7850582832921229,
      "grad_norm": 1.75,
      "learning_rate": 0.000494535975965762,
      "loss": 5.8857,
      "mean_token_accuracy": 0.16745850592851638,
      "num_tokens": 16589574.0,
      "step": 8890
    },
    {
      "entropy": 6.060295724868775,
      "epoch": 0.7854998233839633,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004945290758554939,
      "loss": 5.967,
      "mean_token_accuracy": 0.16401208192110062,
      "num_tokens": 16599623.0,
      "step": 8895
    },
    {
      "entropy": 6.118352174758911,
      "epoch": 0.7859413634758036,
      "grad_norm": 3.390625,
      "learning_rate": 0.000494522171444776,
      "loss": 6.0456,
      "mean_token_accuracy": 0.14564495235681535,
      "num_tokens": 16608559.0,
      "step": 8900
    },
    {
      "entropy": 6.165759563446045,
      "epoch": 0.786382903567644,
      "grad_norm": 2.75,
      "learning_rate": 0.0004945152627337437,
      "loss": 5.9585,
      "mean_token_accuracy": 0.16287342160940171,
      "num_tokens": 16617148.0,
      "step": 8905
    },
    {
      "entropy": 6.087662029266357,
      "epoch": 0.7868244436594842,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004945083497225321,
      "loss": 5.9767,
      "mean_token_accuracy": 0.16042562276124955,
      "num_tokens": 16626405.0,
      "step": 8910
    },
    {
      "entropy": 6.153976058959961,
      "epoch": 0.7872659837513246,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004945014324112769,
      "loss": 6.0194,
      "mean_token_accuracy": 0.15582138448953628,
      "num_tokens": 16635116.0,
      "step": 8915
    },
    {
      "entropy": 6.114471054077148,
      "epoch": 0.787707523843165,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004944945108001134,
      "loss": 5.9926,
      "mean_token_accuracy": 0.1544615477323532,
      "num_tokens": 16644381.0,
      "step": 8920
    },
    {
      "entropy": 6.18104887008667,
      "epoch": 0.7881490639350053,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004944875848891773,
      "loss": 6.0524,
      "mean_token_accuracy": 0.15410684943199157,
      "num_tokens": 16654275.0,
      "step": 8925
    },
    {
      "entropy": 6.228330326080322,
      "epoch": 0.7885906040268457,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004944806546786042,
      "loss": 6.0999,
      "mean_token_accuracy": 0.14967502653598785,
      "num_tokens": 16663484.0,
      "step": 8930
    },
    {
      "entropy": 6.175157356262207,
      "epoch": 0.7890321441186859,
      "grad_norm": 2.15625,
      "learning_rate": 0.00049447372016853,
      "loss": 6.0366,
      "mean_token_accuracy": 0.15706061720848083,
      "num_tokens": 16672490.0,
      "step": 8935
    },
    {
      "entropy": 6.1683669090271,
      "epoch": 0.7894736842105263,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004944667813590904,
      "loss": 6.0157,
      "mean_token_accuracy": 0.15035513937473297,
      "num_tokens": 16681242.0,
      "step": 8940
    },
    {
      "entropy": 6.104328966140747,
      "epoch": 0.7899152243023666,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004944598382504214,
      "loss": 5.9353,
      "mean_token_accuracy": 0.15881823003292084,
      "num_tokens": 16690359.0,
      "step": 8945
    },
    {
      "entropy": 6.171369123458862,
      "epoch": 0.790356764394207,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004944528908426589,
      "loss": 6.0449,
      "mean_token_accuracy": 0.16055190712213516,
      "num_tokens": 16700774.0,
      "step": 8950
    },
    {
      "entropy": 6.287743616104126,
      "epoch": 0.7907983044860474,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004944459391359391,
      "loss": 6.1122,
      "mean_token_accuracy": 0.15050581097602844,
      "num_tokens": 16710513.0,
      "step": 8955
    },
    {
      "entropy": 6.176932334899902,
      "epoch": 0.7912398445778877,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004944389831303982,
      "loss": 6.0271,
      "mean_token_accuracy": 0.1555835410952568,
      "num_tokens": 16720785.0,
      "step": 8960
    },
    {
      "entropy": 6.202974414825439,
      "epoch": 0.791681384669728,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004944320228261723,
      "loss": 6.0091,
      "mean_token_accuracy": 0.1591140389442444,
      "num_tokens": 16729083.0,
      "step": 8965
    },
    {
      "entropy": 6.150328063964844,
      "epoch": 0.7921229247615683,
      "grad_norm": 2.296875,
      "learning_rate": 0.000494425058223398,
      "loss": 6.0207,
      "mean_token_accuracy": 0.1523827463388443,
      "num_tokens": 16738313.0,
      "step": 8970
    },
    {
      "entropy": 6.1048441410064695,
      "epoch": 0.7925644648534087,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004944180893222117,
      "loss": 5.9611,
      "mean_token_accuracy": 0.16035664975643157,
      "num_tokens": 16747191.0,
      "step": 8975
    },
    {
      "entropy": 6.144352149963379,
      "epoch": 0.793006004945249,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004944111161227496,
      "loss": 6.0032,
      "mean_token_accuracy": 0.15728989839553834,
      "num_tokens": 16755921.0,
      "step": 8980
    },
    {
      "entropy": 6.203505754470825,
      "epoch": 0.7934475450370894,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004944041386251486,
      "loss": 6.144,
      "mean_token_accuracy": 0.1521465003490448,
      "num_tokens": 16765394.0,
      "step": 8985
    },
    {
      "entropy": 6.20770001411438,
      "epoch": 0.7938890851289298,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004943971568295454,
      "loss": 6.0234,
      "mean_token_accuracy": 0.16135077625513078,
      "num_tokens": 16774824.0,
      "step": 8990
    },
    {
      "entropy": 6.168958568572998,
      "epoch": 0.79433062522077,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004943901707360765,
      "loss": 6.1627,
      "mean_token_accuracy": 0.14651539474725722,
      "num_tokens": 16783945.0,
      "step": 8995
    },
    {
      "entropy": 6.158242082595825,
      "epoch": 0.7947721653126104,
      "grad_norm": 1.796875,
      "learning_rate": 0.000494383180344879,
      "loss": 5.9415,
      "mean_token_accuracy": 0.1593190088868141,
      "num_tokens": 16792618.0,
      "step": 9000
    },
    {
      "epoch": 0.7947721653126104,
      "eval_entropy": 5.975076497208627,
      "eval_loss": 6.048604488372803,
      "eval_mean_token_accuracy": 0.16088079484720164,
      "eval_num_tokens": 16792618.0,
      "eval_runtime": 26.2413,
      "eval_samples_per_second": 1345.781,
      "eval_steps_per_second": 168.246,
      "step": 9000
    },
    {
      "entropy": 6.16330132484436,
      "epoch": 0.7952137054044507,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004943761856560898,
      "loss": 6.0054,
      "mean_token_accuracy": 0.16190404444932938,
      "num_tokens": 16801210.0,
      "step": 9005
    },
    {
      "entropy": 6.19029974937439,
      "epoch": 0.7956552454962911,
      "grad_norm": 2.171875,
      "learning_rate": 0.000494369186669846,
      "loss": 6.0443,
      "mean_token_accuracy": 0.16017620116472245,
      "num_tokens": 16810371.0,
      "step": 9010
    },
    {
      "entropy": 6.174519681930542,
      "epoch": 0.7960967855881314,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004943621833862844,
      "loss": 5.9585,
      "mean_token_accuracy": 0.15855027139186859,
      "num_tokens": 16818886.0,
      "step": 9015
    },
    {
      "entropy": 6.178792524337768,
      "epoch": 0.7965383256799717,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004943551758055425,
      "loss": 6.0172,
      "mean_token_accuracy": 0.1579152137041092,
      "num_tokens": 16827705.0,
      "step": 9020
    },
    {
      "entropy": 6.2442505836486815,
      "epoch": 0.7969798657718121,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004943481639277574,
      "loss": 6.0142,
      "mean_token_accuracy": 0.15518439263105394,
      "num_tokens": 16836313.0,
      "step": 9025
    },
    {
      "entropy": 6.213355827331543,
      "epoch": 0.7974214058636524,
      "grad_norm": 4.125,
      "learning_rate": 0.0004943411477530665,
      "loss": 6.0734,
      "mean_token_accuracy": 0.14613794684410095,
      "num_tokens": 16845016.0,
      "step": 9030
    },
    {
      "entropy": 6.18611102104187,
      "epoch": 0.7978629459554928,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004943341272816073,
      "loss": 5.9978,
      "mean_token_accuracy": 0.1506240576505661,
      "num_tokens": 16854025.0,
      "step": 9035
    },
    {
      "entropy": 6.2279809474945065,
      "epoch": 0.7983044860473331,
      "grad_norm": 1.75,
      "learning_rate": 0.0004943271025135173,
      "loss": 5.9687,
      "mean_token_accuracy": 0.16130097955465317,
      "num_tokens": 16862859.0,
      "step": 9040
    },
    {
      "entropy": 6.198914575576782,
      "epoch": 0.7987460261391734,
      "grad_norm": 2.25,
      "learning_rate": 0.0004943200734489341,
      "loss": 6.0707,
      "mean_token_accuracy": 0.15602338537573815,
      "num_tokens": 16872226.0,
      "step": 9045
    },
    {
      "entropy": 6.120191383361816,
      "epoch": 0.7991875662310137,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004943130400879953,
      "loss": 5.949,
      "mean_token_accuracy": 0.16058978736400603,
      "num_tokens": 16881977.0,
      "step": 9050
    },
    {
      "entropy": 6.167643356323242,
      "epoch": 0.7996291063228541,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004943060024308389,
      "loss": 6.0691,
      "mean_token_accuracy": 0.14636371284723282,
      "num_tokens": 16891656.0,
      "step": 9055
    },
    {
      "entropy": 6.198838472366333,
      "epoch": 0.8000706464146945,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004942989604776026,
      "loss": 5.9387,
      "mean_token_accuracy": 0.1607286438345909,
      "num_tokens": 16900944.0,
      "step": 9060
    },
    {
      "entropy": 6.1139894962310795,
      "epoch": 0.8005121865065348,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004942919142284244,
      "loss": 5.9587,
      "mean_token_accuracy": 0.1589059628546238,
      "num_tokens": 16910154.0,
      "step": 9065
    },
    {
      "entropy": 6.167115831375122,
      "epoch": 0.8009537265983752,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004942848636834423,
      "loss": 6.088,
      "mean_token_accuracy": 0.15294271409511567,
      "num_tokens": 16919694.0,
      "step": 9070
    },
    {
      "entropy": 6.1631608486175535,
      "epoch": 0.8013952666902154,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004942778088427944,
      "loss": 6.0452,
      "mean_token_accuracy": 0.15437108725309373,
      "num_tokens": 16928530.0,
      "step": 9075
    },
    {
      "entropy": 6.217332363128662,
      "epoch": 0.8018368067820558,
      "grad_norm": 2.28125,
      "learning_rate": 0.000494270749706619,
      "loss": 6.0569,
      "mean_token_accuracy": 0.1532064601778984,
      "num_tokens": 16938614.0,
      "step": 9080
    },
    {
      "entropy": 6.273269748687744,
      "epoch": 0.8022783468738961,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004942636862750544,
      "loss": 6.0024,
      "mean_token_accuracy": 0.1548839457333088,
      "num_tokens": 16947655.0,
      "step": 9085
    },
    {
      "entropy": 6.263075590133667,
      "epoch": 0.8027198869657365,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004942566185482387,
      "loss": 6.1003,
      "mean_token_accuracy": 0.15037893801927565,
      "num_tokens": 16957101.0,
      "step": 9090
    },
    {
      "entropy": 6.146580791473388,
      "epoch": 0.8031614270575769,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004942495465263106,
      "loss": 6.109,
      "mean_token_accuracy": 0.14742673188447952,
      "num_tokens": 16966421.0,
      "step": 9095
    },
    {
      "entropy": 6.202206945419311,
      "epoch": 0.8036029671494171,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004942424702094085,
      "loss": 6.0191,
      "mean_token_accuracy": 0.15906717628240585,
      "num_tokens": 16975656.0,
      "step": 9100
    },
    {
      "entropy": 6.178380012512207,
      "epoch": 0.8040445072412575,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004942353895976713,
      "loss": 6.0599,
      "mean_token_accuracy": 0.15076268017292022,
      "num_tokens": 16984409.0,
      "step": 9105
    },
    {
      "entropy": 6.16157898902893,
      "epoch": 0.8044860473330978,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004942283046912374,
      "loss": 5.9762,
      "mean_token_accuracy": 0.14847689867019653,
      "num_tokens": 16992728.0,
      "step": 9110
    },
    {
      "entropy": 6.166329860687256,
      "epoch": 0.8049275874249382,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004942212154902456,
      "loss": 5.8957,
      "mean_token_accuracy": 0.1641251653432846,
      "num_tokens": 17001616.0,
      "step": 9115
    },
    {
      "entropy": 6.1823100566864015,
      "epoch": 0.8053691275167785,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000494214121994835,
      "loss": 6.049,
      "mean_token_accuracy": 0.15651159435510636,
      "num_tokens": 17010655.0,
      "step": 9120
    },
    {
      "entropy": 6.135058641433716,
      "epoch": 0.8058106676086189,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004942070242051444,
      "loss": 6.0237,
      "mean_token_accuracy": 0.1511933371424675,
      "num_tokens": 17019819.0,
      "step": 9125
    },
    {
      "entropy": 6.190236616134643,
      "epoch": 0.8062522077004592,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004941999221213129,
      "loss": 6.0379,
      "mean_token_accuracy": 0.15519107580184938,
      "num_tokens": 17029218.0,
      "step": 9130
    },
    {
      "entropy": 6.196532678604126,
      "epoch": 0.8066937477922995,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004941928157434796,
      "loss": 5.9583,
      "mean_token_accuracy": 0.15509601384401323,
      "num_tokens": 17037880.0,
      "step": 9135
    },
    {
      "entropy": 6.095403337478638,
      "epoch": 0.8071352878841399,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004941857050717836,
      "loss": 6.0598,
      "mean_token_accuracy": 0.15429162681102754,
      "num_tokens": 17047934.0,
      "step": 9140
    },
    {
      "entropy": 6.279002666473389,
      "epoch": 0.8075768279759802,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004941785901063642,
      "loss": 6.1084,
      "mean_token_accuracy": 0.14977463185787201,
      "num_tokens": 17057258.0,
      "step": 9145
    },
    {
      "entropy": 6.214845800399781,
      "epoch": 0.8080183680678206,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004941714708473611,
      "loss": 6.0519,
      "mean_token_accuracy": 0.15372628569602967,
      "num_tokens": 17066387.0,
      "step": 9150
    },
    {
      "entropy": 6.08232626914978,
      "epoch": 0.8084599081596608,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004941643472949134,
      "loss": 5.9797,
      "mean_token_accuracy": 0.15843379348516465,
      "num_tokens": 17075952.0,
      "step": 9155
    },
    {
      "entropy": 6.0682563304901125,
      "epoch": 0.8089014482515012,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004941572194491608,
      "loss": 5.8706,
      "mean_token_accuracy": 0.16113847345113755,
      "num_tokens": 17084209.0,
      "step": 9160
    },
    {
      "entropy": 6.09498815536499,
      "epoch": 0.8093429883433416,
      "grad_norm": 2.90625,
      "learning_rate": 0.000494150087310243,
      "loss": 6.0974,
      "mean_token_accuracy": 0.15568703263998032,
      "num_tokens": 17093395.0,
      "step": 9165
    },
    {
      "entropy": 6.211379718780518,
      "epoch": 0.8097845284351819,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004941429508782995,
      "loss": 5.9763,
      "mean_token_accuracy": 0.15933869704604148,
      "num_tokens": 17102646.0,
      "step": 9170
    },
    {
      "entropy": 6.223939275741577,
      "epoch": 0.8102260685270223,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004941358101534703,
      "loss": 5.9914,
      "mean_token_accuracy": 0.1557866096496582,
      "num_tokens": 17111530.0,
      "step": 9175
    },
    {
      "entropy": 6.101489305496216,
      "epoch": 0.8106676086188626,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004941286651358952,
      "loss": 5.9333,
      "mean_token_accuracy": 0.1631128028035164,
      "num_tokens": 17119659.0,
      "step": 9180
    },
    {
      "entropy": 5.9786371231079105,
      "epoch": 0.8111091487107029,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004941215158257142,
      "loss": 5.9632,
      "mean_token_accuracy": 0.1578883334994316,
      "num_tokens": 17127931.0,
      "step": 9185
    },
    {
      "entropy": 6.224986839294433,
      "epoch": 0.8115506888025432,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004941143622230672,
      "loss": 6.0128,
      "mean_token_accuracy": 0.15638066306710244,
      "num_tokens": 17137374.0,
      "step": 9190
    },
    {
      "entropy": 6.306176853179932,
      "epoch": 0.8119922288943836,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004941072043280946,
      "loss": 6.0675,
      "mean_token_accuracy": 0.14437773153185846,
      "num_tokens": 17147257.0,
      "step": 9195
    },
    {
      "entropy": 6.164927577972412,
      "epoch": 0.812433768986224,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004941000421409365,
      "loss": 5.9507,
      "mean_token_accuracy": 0.15665391683578492,
      "num_tokens": 17156233.0,
      "step": 9200
    },
    {
      "entropy": 6.225679111480713,
      "epoch": 0.8128753090780643,
      "grad_norm": 2.375,
      "learning_rate": 0.0004940928756617331,
      "loss": 6.1359,
      "mean_token_accuracy": 0.14476195946335793,
      "num_tokens": 17166364.0,
      "step": 9205
    },
    {
      "entropy": 6.20369143486023,
      "epoch": 0.8133168491699047,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004940857048906248,
      "loss": 6.0208,
      "mean_token_accuracy": 0.15950246453285216,
      "num_tokens": 17175853.0,
      "step": 9210
    },
    {
      "entropy": 6.222797155380249,
      "epoch": 0.8137583892617449,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004940785298277523,
      "loss": 6.1214,
      "mean_token_accuracy": 0.14765160977840425,
      "num_tokens": 17187849.0,
      "step": 9215
    },
    {
      "entropy": 6.20315260887146,
      "epoch": 0.8141999293535853,
      "grad_norm": 2.53125,
      "learning_rate": 0.000494071350473256,
      "loss": 5.986,
      "mean_token_accuracy": 0.15655021518468856,
      "num_tokens": 17196646.0,
      "step": 9220
    },
    {
      "entropy": 6.183213424682617,
      "epoch": 0.8146414694454256,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004940641668272765,
      "loss": 6.1141,
      "mean_token_accuracy": 0.15241051316261292,
      "num_tokens": 17205919.0,
      "step": 9225
    },
    {
      "entropy": 6.197967433929444,
      "epoch": 0.815083009537266,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004940569788899544,
      "loss": 6.0825,
      "mean_token_accuracy": 0.15468350201845169,
      "num_tokens": 17215085.0,
      "step": 9230
    },
    {
      "entropy": 6.15929069519043,
      "epoch": 0.8155245496291064,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004940497866614309,
      "loss": 6.056,
      "mean_token_accuracy": 0.14778956323862075,
      "num_tokens": 17223633.0,
      "step": 9235
    },
    {
      "entropy": 6.147996044158935,
      "epoch": 0.8159660897209466,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004940425901418465,
      "loss": 5.9745,
      "mean_token_accuracy": 0.15986088961362838,
      "num_tokens": 17232915.0,
      "step": 9240
    },
    {
      "entropy": 6.181144189834595,
      "epoch": 0.816407629812787,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004940353893313423,
      "loss": 6.1006,
      "mean_token_accuracy": 0.1512872740626335,
      "num_tokens": 17242392.0,
      "step": 9245
    },
    {
      "entropy": 6.270976161956787,
      "epoch": 0.8168491699046273,
      "grad_norm": 5.71875,
      "learning_rate": 0.0004940281842300596,
      "loss": 6.1767,
      "mean_token_accuracy": 0.14558177962899208,
      "num_tokens": 17252259.0,
      "step": 9250
    },
    {
      "entropy": 6.290340375900269,
      "epoch": 0.8172907099964677,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004940209748381392,
      "loss": 6.0376,
      "mean_token_accuracy": 0.14993744641542434,
      "num_tokens": 17261584.0,
      "step": 9255
    },
    {
      "entropy": 6.1955766677856445,
      "epoch": 0.817732250088308,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004940137611557225,
      "loss": 5.9871,
      "mean_token_accuracy": 0.1564241111278534,
      "num_tokens": 17270901.0,
      "step": 9260
    },
    {
      "entropy": 6.110502481460571,
      "epoch": 0.8181737901801484,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004940065431829508,
      "loss": 6.0165,
      "mean_token_accuracy": 0.1528618276119232,
      "num_tokens": 17280024.0,
      "step": 9265
    },
    {
      "entropy": 6.171491527557373,
      "epoch": 0.8186153302719887,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004939993209199654,
      "loss": 6.0735,
      "mean_token_accuracy": 0.15109170973300934,
      "num_tokens": 17290131.0,
      "step": 9270
    },
    {
      "entropy": 6.316831159591675,
      "epoch": 0.819056870363829,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004939920943669079,
      "loss": 6.0763,
      "mean_token_accuracy": 0.149622143805027,
      "num_tokens": 17299453.0,
      "step": 9275
    },
    {
      "entropy": 6.217446279525757,
      "epoch": 0.8194984104556694,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004939848635239198,
      "loss": 6.0406,
      "mean_token_accuracy": 0.1519092798233032,
      "num_tokens": 17309344.0,
      "step": 9280
    },
    {
      "entropy": 6.164099025726318,
      "epoch": 0.8199399505475097,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004939776283911429,
      "loss": 5.9662,
      "mean_token_accuracy": 0.1645274356007576,
      "num_tokens": 17318636.0,
      "step": 9285
    },
    {
      "entropy": 6.105147838592529,
      "epoch": 0.8203814906393501,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004939703889687186,
      "loss": 6.0213,
      "mean_token_accuracy": 0.15480365753173828,
      "num_tokens": 17328146.0,
      "step": 9290
    },
    {
      "entropy": 6.14805326461792,
      "epoch": 0.8208230307311903,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004939631452567891,
      "loss": 6.0843,
      "mean_token_accuracy": 0.15634123831987382,
      "num_tokens": 17338542.0,
      "step": 9295
    },
    {
      "entropy": 6.275281000137329,
      "epoch": 0.8212645708230307,
      "grad_norm": 2.734375,
      "learning_rate": 0.000493955897255496,
      "loss": 6.04,
      "mean_token_accuracy": 0.15544226467609407,
      "num_tokens": 17347577.0,
      "step": 9300
    },
    {
      "entropy": 6.2426682472229,
      "epoch": 0.8217061109148711,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004939486449649814,
      "loss": 6.0423,
      "mean_token_accuracy": 0.15123710185289382,
      "num_tokens": 17356554.0,
      "step": 9305
    },
    {
      "entropy": 6.186700057983399,
      "epoch": 0.8221476510067114,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004939413883853873,
      "loss": 6.0493,
      "mean_token_accuracy": 0.15451455116271973,
      "num_tokens": 17366432.0,
      "step": 9310
    },
    {
      "entropy": 6.191228437423706,
      "epoch": 0.8225891910985518,
      "grad_norm": 2.5,
      "learning_rate": 0.000493934127516856,
      "loss": 6.018,
      "mean_token_accuracy": 0.15782462805509567,
      "num_tokens": 17376556.0,
      "step": 9315
    },
    {
      "entropy": 6.176034069061279,
      "epoch": 0.823030731190392,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004939268623595297,
      "loss": 6.0829,
      "mean_token_accuracy": 0.15094019025564193,
      "num_tokens": 17385382.0,
      "step": 9320
    },
    {
      "entropy": 6.195343017578125,
      "epoch": 0.8234722712822324,
      "grad_norm": 2.875,
      "learning_rate": 0.0004939195929135506,
      "loss": 5.9479,
      "mean_token_accuracy": 0.1533026024699211,
      "num_tokens": 17393229.0,
      "step": 9325
    },
    {
      "entropy": 6.169883394241333,
      "epoch": 0.8239138113740727,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004939123191790612,
      "loss": 5.9319,
      "mean_token_accuracy": 0.14991182833909988,
      "num_tokens": 17402439.0,
      "step": 9330
    },
    {
      "entropy": 6.150929021835327,
      "epoch": 0.8243553514659131,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004939050411562041,
      "loss": 6.0917,
      "mean_token_accuracy": 0.14788064062595369,
      "num_tokens": 17412349.0,
      "step": 9335
    },
    {
      "entropy": 6.054622459411621,
      "epoch": 0.8247968915577535,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004938977588451216,
      "loss": 5.8675,
      "mean_token_accuracy": 0.1703634589910507,
      "num_tokens": 17420842.0,
      "step": 9340
    },
    {
      "entropy": 6.145426797866821,
      "epoch": 0.8252384316495938,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004938904722459565,
      "loss": 5.94,
      "mean_token_accuracy": 0.1631227657198906,
      "num_tokens": 17429877.0,
      "step": 9345
    },
    {
      "entropy": 6.153570747375488,
      "epoch": 0.8256799717414341,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004938831813588515,
      "loss": 6.0604,
      "mean_token_accuracy": 0.15108452886343002,
      "num_tokens": 17439508.0,
      "step": 9350
    },
    {
      "entropy": 6.1563780307769775,
      "epoch": 0.8261215118332744,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004938758861839495,
      "loss": 5.9997,
      "mean_token_accuracy": 0.15649935901165007,
      "num_tokens": 17449004.0,
      "step": 9355
    },
    {
      "entropy": 6.141077184677124,
      "epoch": 0.8265630519251148,
      "grad_norm": 4.25,
      "learning_rate": 0.0004938685867213934,
      "loss": 6.0038,
      "mean_token_accuracy": 0.1581357717514038,
      "num_tokens": 17458462.0,
      "step": 9360
    },
    {
      "entropy": 6.223224925994873,
      "epoch": 0.8270045920169551,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004938612829713262,
      "loss": 6.0288,
      "mean_token_accuracy": 0.15867939889431,
      "num_tokens": 17466808.0,
      "step": 9365
    },
    {
      "entropy": 6.172831058502197,
      "epoch": 0.8274461321087955,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004938539749338909,
      "loss": 6.0742,
      "mean_token_accuracy": 0.1540021702647209,
      "num_tokens": 17476441.0,
      "step": 9370
    },
    {
      "entropy": 6.1683203220367435,
      "epoch": 0.8278876722006359,
      "grad_norm": 6.1875,
      "learning_rate": 0.0004938466626092308,
      "loss": 6.0027,
      "mean_token_accuracy": 0.16268929094076157,
      "num_tokens": 17485393.0,
      "step": 9375
    },
    {
      "entropy": 6.125380802154541,
      "epoch": 0.8283292122924761,
      "grad_norm": 1.703125,
      "learning_rate": 0.000493839345997489,
      "loss": 6.0305,
      "mean_token_accuracy": 0.15758230686187744,
      "num_tokens": 17494904.0,
      "step": 9380
    },
    {
      "entropy": 6.257401657104492,
      "epoch": 0.8287707523843165,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004938320250988087,
      "loss": 6.0235,
      "mean_token_accuracy": 0.15207177996635438,
      "num_tokens": 17505137.0,
      "step": 9385
    },
    {
      "entropy": 6.20851697921753,
      "epoch": 0.8292122924761568,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004938246999133337,
      "loss": 6.0834,
      "mean_token_accuracy": 0.155420882999897,
      "num_tokens": 17514039.0,
      "step": 9390
    },
    {
      "entropy": 6.104730176925659,
      "epoch": 0.8296538325679972,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004938173704412071,
      "loss": 5.9529,
      "mean_token_accuracy": 0.163827283680439,
      "num_tokens": 17522423.0,
      "step": 9395
    },
    {
      "entropy": 6.13248643875122,
      "epoch": 0.8300953726598375,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004938100366825728,
      "loss": 5.9137,
      "mean_token_accuracy": 0.16646003127098083,
      "num_tokens": 17531958.0,
      "step": 9400
    },
    {
      "entropy": 6.148837900161743,
      "epoch": 0.8305369127516778,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004938026986375742,
      "loss": 6.0062,
      "mean_token_accuracy": 0.16234309375286102,
      "num_tokens": 17540905.0,
      "step": 9405
    },
    {
      "entropy": 6.1506743907928465,
      "epoch": 0.8309784528435182,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004937953563063553,
      "loss": 6.0534,
      "mean_token_accuracy": 0.1501004308462143,
      "num_tokens": 17551007.0,
      "step": 9410
    },
    {
      "entropy": 6.153183650970459,
      "epoch": 0.8314199929353585,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004937880096890598,
      "loss": 5.8739,
      "mean_token_accuracy": 0.16787810176610946,
      "num_tokens": 17559403.0,
      "step": 9415
    },
    {
      "entropy": 6.184926891326905,
      "epoch": 0.8318615330271989,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004937806587858315,
      "loss": 6.0475,
      "mean_token_accuracy": 0.15986477732658386,
      "num_tokens": 17569191.0,
      "step": 9420
    },
    {
      "entropy": 6.258034801483154,
      "epoch": 0.8323030731190392,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004937733035968147,
      "loss": 6.0953,
      "mean_token_accuracy": 0.14976078867912293,
      "num_tokens": 17578538.0,
      "step": 9425
    },
    {
      "entropy": 6.188538408279419,
      "epoch": 0.8327446132108796,
      "grad_norm": 2.375,
      "learning_rate": 0.0004937659441221531,
      "loss": 5.9661,
      "mean_token_accuracy": 0.15306929647922515,
      "num_tokens": 17587344.0,
      "step": 9430
    },
    {
      "entropy": 6.130722427368164,
      "epoch": 0.8331861533027198,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004937585803619912,
      "loss": 5.9615,
      "mean_token_accuracy": 0.1581699386239052,
      "num_tokens": 17596033.0,
      "step": 9435
    },
    {
      "entropy": 6.176023197174072,
      "epoch": 0.8336276933945602,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004937512123164731,
      "loss": 5.9247,
      "mean_token_accuracy": 0.1648713991045952,
      "num_tokens": 17605860.0,
      "step": 9440
    },
    {
      "entropy": 6.164800596237183,
      "epoch": 0.8340692334864006,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004937438399857431,
      "loss": 6.0333,
      "mean_token_accuracy": 0.15173615366220475,
      "num_tokens": 17615819.0,
      "step": 9445
    },
    {
      "entropy": 6.1563958644866945,
      "epoch": 0.8345107735782409,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004937364633699459,
      "loss": 6.0285,
      "mean_token_accuracy": 0.15544223189353942,
      "num_tokens": 17625478.0,
      "step": 9450
    },
    {
      "entropy": 6.206226634979248,
      "epoch": 0.8349523136700813,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004937290824692255,
      "loss": 5.9959,
      "mean_token_accuracy": 0.1604430839419365,
      "num_tokens": 17635329.0,
      "step": 9455
    },
    {
      "entropy": 6.0999308109283445,
      "epoch": 0.8353938537619215,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004937216972837269,
      "loss": 5.9751,
      "mean_token_accuracy": 0.15586129426956177,
      "num_tokens": 17644129.0,
      "step": 9460
    },
    {
      "entropy": 6.07632360458374,
      "epoch": 0.8358353938537619,
      "grad_norm": 2.25,
      "learning_rate": 0.0004937143078135946,
      "loss": 5.94,
      "mean_token_accuracy": 0.16034181118011476,
      "num_tokens": 17652836.0,
      "step": 9465
    },
    {
      "entropy": 6.203134632110595,
      "epoch": 0.8362769339456022,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004937069140589735,
      "loss": 5.969,
      "mean_token_accuracy": 0.15560818165540696,
      "num_tokens": 17661222.0,
      "step": 9470
    },
    {
      "entropy": 6.170781993865967,
      "epoch": 0.8367184740374426,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004936995160200083,
      "loss": 5.9338,
      "mean_token_accuracy": 0.1625818833708763,
      "num_tokens": 17669637.0,
      "step": 9475
    },
    {
      "entropy": 6.180658292770386,
      "epoch": 0.837160014129283,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004936921136968439,
      "loss": 6.1231,
      "mean_token_accuracy": 0.1459084391593933,
      "num_tokens": 17678589.0,
      "step": 9480
    },
    {
      "entropy": 6.209010219573974,
      "epoch": 0.8376015542211233,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004936847070896254,
      "loss": 6.0875,
      "mean_token_accuracy": 0.15022720098495485,
      "num_tokens": 17689129.0,
      "step": 9485
    },
    {
      "entropy": 6.264245510101318,
      "epoch": 0.8380430943129636,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004936772961984979,
      "loss": 6.0895,
      "mean_token_accuracy": 0.15328721404075624,
      "num_tokens": 17698864.0,
      "step": 9490
    },
    {
      "entropy": 6.217590379714966,
      "epoch": 0.8384846344048039,
      "grad_norm": 2.75,
      "learning_rate": 0.0004936698810236065,
      "loss": 6.1524,
      "mean_token_accuracy": 0.14972497373819352,
      "num_tokens": 17707900.0,
      "step": 9495
    },
    {
      "entropy": 6.180559921264648,
      "epoch": 0.8389261744966443,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004936624615650964,
      "loss": 5.985,
      "mean_token_accuracy": 0.1569998010993004,
      "num_tokens": 17716983.0,
      "step": 9500
    },
    {
      "entropy": 6.2115813255310055,
      "epoch": 0.8393677145884846,
      "grad_norm": 2.25,
      "learning_rate": 0.0004936550378231133,
      "loss": 5.9936,
      "mean_token_accuracy": 0.1554318845272064,
      "num_tokens": 17726503.0,
      "step": 9505
    },
    {
      "entropy": 6.199985313415527,
      "epoch": 0.839809254680325,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004936476097978023,
      "loss": 6.0589,
      "mean_token_accuracy": 0.15271791815757751,
      "num_tokens": 17735708.0,
      "step": 9510
    },
    {
      "entropy": 6.238025188446045,
      "epoch": 0.8402507947721654,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004936401774893088,
      "loss": 6.0111,
      "mean_token_accuracy": 0.15702128112316133,
      "num_tokens": 17745169.0,
      "step": 9515
    },
    {
      "entropy": 6.28457989692688,
      "epoch": 0.8406923348640056,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004936327408977786,
      "loss": 6.1275,
      "mean_token_accuracy": 0.14965770095586778,
      "num_tokens": 17755630.0,
      "step": 9520
    },
    {
      "entropy": 6.213114309310913,
      "epoch": 0.841133874955846,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004936253000233574,
      "loss": 6.0052,
      "mean_token_accuracy": 0.1566065639257431,
      "num_tokens": 17764469.0,
      "step": 9525
    },
    {
      "entropy": 6.128476524353028,
      "epoch": 0.8415754150476863,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004936178548661909,
      "loss": 6.0905,
      "mean_token_accuracy": 0.14764439612627028,
      "num_tokens": 17774977.0,
      "step": 9530
    },
    {
      "entropy": 6.143621635437012,
      "epoch": 0.8420169551395267,
      "grad_norm": 1.984375,
      "learning_rate": 0.000493610405426425,
      "loss": 6.0011,
      "mean_token_accuracy": 0.1499892368912697,
      "num_tokens": 17784250.0,
      "step": 9535
    },
    {
      "entropy": 6.20766806602478,
      "epoch": 0.842458495231367,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004936029517042056,
      "loss": 6.0459,
      "mean_token_accuracy": 0.15633412301540375,
      "num_tokens": 17793858.0,
      "step": 9540
    },
    {
      "entropy": 6.276909065246582,
      "epoch": 0.8429000353232073,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004935954936996787,
      "loss": 6.1303,
      "mean_token_accuracy": 0.14823518842458724,
      "num_tokens": 17804330.0,
      "step": 9545
    },
    {
      "entropy": 6.184249591827393,
      "epoch": 0.8433415754150477,
      "grad_norm": 2.75,
      "learning_rate": 0.0004935880314129903,
      "loss": 6.0214,
      "mean_token_accuracy": 0.15660575330257415,
      "num_tokens": 17813750.0,
      "step": 9550
    },
    {
      "entropy": 6.1108362674713135,
      "epoch": 0.843783115506888,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004935805648442868,
      "loss": 5.981,
      "mean_token_accuracy": 0.15835636854171753,
      "num_tokens": 17823537.0,
      "step": 9555
    },
    {
      "entropy": 6.182886791229248,
      "epoch": 0.8442246555987284,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004935730939937143,
      "loss": 5.9993,
      "mean_token_accuracy": 0.15982471406459808,
      "num_tokens": 17833298.0,
      "step": 9560
    },
    {
      "entropy": 6.243503379821777,
      "epoch": 0.8446661956905687,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004935656188614192,
      "loss": 5.924,
      "mean_token_accuracy": 0.16516816467046738,
      "num_tokens": 17842339.0,
      "step": 9565
    },
    {
      "entropy": 6.193902778625488,
      "epoch": 0.845107735782409,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004935581394475479,
      "loss": 6.1125,
      "mean_token_accuracy": 0.1564135506749153,
      "num_tokens": 17851827.0,
      "step": 9570
    },
    {
      "entropy": 6.080324697494507,
      "epoch": 0.8455492758742493,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004935506557522469,
      "loss": 5.9882,
      "mean_token_accuracy": 0.15881171822547913,
      "num_tokens": 17860788.0,
      "step": 9575
    },
    {
      "entropy": 6.200874280929566,
      "epoch": 0.8459908159660897,
      "grad_norm": 2.296875,
      "learning_rate": 0.000493543167775663,
      "loss": 5.9783,
      "mean_token_accuracy": 0.15974099338054656,
      "num_tokens": 17869849.0,
      "step": 9580
    },
    {
      "entropy": 6.23571743965149,
      "epoch": 0.8464323560579301,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004935356755179426,
      "loss": 5.993,
      "mean_token_accuracy": 0.15421488881111145,
      "num_tokens": 17879431.0,
      "step": 9585
    },
    {
      "entropy": 6.252764511108398,
      "epoch": 0.8468738961497704,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004935281789792326,
      "loss": 6.1044,
      "mean_token_accuracy": 0.15210022479295732,
      "num_tokens": 17888599.0,
      "step": 9590
    },
    {
      "entropy": 6.190015029907227,
      "epoch": 0.8473154362416108,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004935206781596799,
      "loss": 5.9982,
      "mean_token_accuracy": 0.1624806120991707,
      "num_tokens": 17898662.0,
      "step": 9595
    },
    {
      "entropy": 6.2146673679351805,
      "epoch": 0.847756976333451,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004935131730594315,
      "loss": 6.0349,
      "mean_token_accuracy": 0.15563121289014817,
      "num_tokens": 17907810.0,
      "step": 9600
    },
    {
      "entropy": 6.178008794784546,
      "epoch": 0.8481985164252914,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004935056636786342,
      "loss": 6.0042,
      "mean_token_accuracy": 0.15478068590164185,
      "num_tokens": 17917611.0,
      "step": 9605
    },
    {
      "entropy": 6.188114881515503,
      "epoch": 0.8486400565171317,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004934981500174353,
      "loss": 6.0329,
      "mean_token_accuracy": 0.15326831638813018,
      "num_tokens": 17926076.0,
      "step": 9610
    },
    {
      "entropy": 6.1936304569244385,
      "epoch": 0.8490815966089721,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004934906320759818,
      "loss": 6.0451,
      "mean_token_accuracy": 0.1524820163846016,
      "num_tokens": 17935521.0,
      "step": 9615
    },
    {
      "entropy": 6.2309952735900875,
      "epoch": 0.8495231367008125,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004934831098544211,
      "loss": 6.0237,
      "mean_token_accuracy": 0.15770833045244217,
      "num_tokens": 17944849.0,
      "step": 9620
    },
    {
      "entropy": 6.235608005523682,
      "epoch": 0.8499646767926528,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004934755833529007,
      "loss": 5.9992,
      "mean_token_accuracy": 0.1565394029021263,
      "num_tokens": 17953988.0,
      "step": 9625
    },
    {
      "entropy": 6.154301691055298,
      "epoch": 0.8504062168844931,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004934680525715677,
      "loss": 5.9896,
      "mean_token_accuracy": 0.15951733887195588,
      "num_tokens": 17963319.0,
      "step": 9630
    },
    {
      "entropy": 6.129526615142822,
      "epoch": 0.8508477569763334,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004934605175105698,
      "loss": 5.9604,
      "mean_token_accuracy": 0.16619348227977754,
      "num_tokens": 17972054.0,
      "step": 9635
    },
    {
      "entropy": 6.212171268463135,
      "epoch": 0.8512892970681738,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004934529781700546,
      "loss": 6.0995,
      "mean_token_accuracy": 0.15235573276877404,
      "num_tokens": 17981630.0,
      "step": 9640
    },
    {
      "entropy": 6.247295379638672,
      "epoch": 0.8517308371600141,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004934454345501698,
      "loss": 6.045,
      "mean_token_accuracy": 0.1502332419157028,
      "num_tokens": 17990987.0,
      "step": 9645
    },
    {
      "entropy": 6.117348384857178,
      "epoch": 0.8521723772518545,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004934378866510633,
      "loss": 5.9442,
      "mean_token_accuracy": 0.1672571614384651,
      "num_tokens": 17999284.0,
      "step": 9650
    },
    {
      "entropy": 6.1755430698394775,
      "epoch": 0.8526139173436948,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004934303344728828,
      "loss": 5.9889,
      "mean_token_accuracy": 0.15463931411504744,
      "num_tokens": 18008670.0,
      "step": 9655
    },
    {
      "entropy": 6.129124164581299,
      "epoch": 0.8530554574355351,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004934227780157761,
      "loss": 5.9668,
      "mean_token_accuracy": 0.15953436493873596,
      "num_tokens": 18018374.0,
      "step": 9660
    },
    {
      "entropy": 6.236544036865235,
      "epoch": 0.8534969975273755,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004934152172798916,
      "loss": 6.0639,
      "mean_token_accuracy": 0.15600237548351287,
      "num_tokens": 18026943.0,
      "step": 9665
    },
    {
      "entropy": 6.235983610153198,
      "epoch": 0.8539385376192158,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004934076522653772,
      "loss": 6.0029,
      "mean_token_accuracy": 0.15417421311140062,
      "num_tokens": 18036270.0,
      "step": 9670
    },
    {
      "entropy": 6.143337106704712,
      "epoch": 0.8543800777110562,
      "grad_norm": 2.328125,
      "learning_rate": 0.000493400082972381,
      "loss": 5.9936,
      "mean_token_accuracy": 0.1586397185921669,
      "num_tokens": 18045611.0,
      "step": 9675
    },
    {
      "entropy": 6.150987148284912,
      "epoch": 0.8548216178028964,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004933925094010516,
      "loss": 6.0247,
      "mean_token_accuracy": 0.15174849480390548,
      "num_tokens": 18055152.0,
      "step": 9680
    },
    {
      "entropy": 6.204892015457153,
      "epoch": 0.8552631578947368,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004933849315515369,
      "loss": 6.0152,
      "mean_token_accuracy": 0.14948640912771224,
      "num_tokens": 18064376.0,
      "step": 9685
    },
    {
      "entropy": 6.090399599075317,
      "epoch": 0.8557046979865772,
      "grad_norm": 4.25,
      "learning_rate": 0.0004933773494239857,
      "loss": 5.94,
      "mean_token_accuracy": 0.16883918195962905,
      "num_tokens": 18073569.0,
      "step": 9690
    },
    {
      "entropy": 6.110611486434936,
      "epoch": 0.8561462380784175,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004933697630185464,
      "loss": 5.9969,
      "mean_token_accuracy": 0.1602720320224762,
      "num_tokens": 18083264.0,
      "step": 9695
    },
    {
      "entropy": 6.22624568939209,
      "epoch": 0.8565877781702579,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004933621723353677,
      "loss": 5.9994,
      "mean_token_accuracy": 0.15230705291032792,
      "num_tokens": 18093151.0,
      "step": 9700
    },
    {
      "entropy": 6.208186960220337,
      "epoch": 0.8570293182620982,
      "grad_norm": 2.703125,
      "learning_rate": 0.000493354577374598,
      "loss": 5.9289,
      "mean_token_accuracy": 0.16737866401672363,
      "num_tokens": 18102087.0,
      "step": 9705
    },
    {
      "entropy": 6.187351655960083,
      "epoch": 0.8574708583539385,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004933469781363865,
      "loss": 6.0346,
      "mean_token_accuracy": 0.15789082199335097,
      "num_tokens": 18111854.0,
      "step": 9710
    },
    {
      "entropy": 6.136016178131103,
      "epoch": 0.8579123984457788,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004933393746208819,
      "loss": 6.0629,
      "mean_token_accuracy": 0.15245209634304047,
      "num_tokens": 18120520.0,
      "step": 9715
    },
    {
      "entropy": 6.211365509033203,
      "epoch": 0.8583539385376192,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004933317668282331,
      "loss": 6.1116,
      "mean_token_accuracy": 0.1523909091949463,
      "num_tokens": 18130557.0,
      "step": 9720
    },
    {
      "entropy": 6.193790864944458,
      "epoch": 0.8587954786294596,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004933241547585891,
      "loss": 5.9211,
      "mean_token_accuracy": 0.16193769872188568,
      "num_tokens": 18139199.0,
      "step": 9725
    },
    {
      "entropy": 6.155925941467285,
      "epoch": 0.8592370187212999,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004933165384120991,
      "loss": 5.9702,
      "mean_token_accuracy": 0.16056935489177704,
      "num_tokens": 18150039.0,
      "step": 9730
    },
    {
      "entropy": 6.1194170951843265,
      "epoch": 0.8596785588131403,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004933089177889124,
      "loss": 5.9771,
      "mean_token_accuracy": 0.15308154970407487,
      "num_tokens": 18159451.0,
      "step": 9735
    },
    {
      "entropy": 6.193133115768433,
      "epoch": 0.8601200989049805,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004933012928891781,
      "loss": 6.073,
      "mean_token_accuracy": 0.1550351247191429,
      "num_tokens": 18169896.0,
      "step": 9740
    },
    {
      "entropy": 6.174573802947998,
      "epoch": 0.8605616389968209,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004932936637130456,
      "loss": 5.9386,
      "mean_token_accuracy": 0.16603296995162964,
      "num_tokens": 18177764.0,
      "step": 9745
    },
    {
      "entropy": 6.169020938873291,
      "epoch": 0.8610031790886612,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004932860302606644,
      "loss": 6.0167,
      "mean_token_accuracy": 0.15849282890558242,
      "num_tokens": 18186940.0,
      "step": 9750
    },
    {
      "entropy": 6.194835758209228,
      "epoch": 0.8614447191805016,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004932783925321838,
      "loss": 5.8995,
      "mean_token_accuracy": 0.16262447088956833,
      "num_tokens": 18195374.0,
      "step": 9755
    },
    {
      "entropy": 6.117531776428223,
      "epoch": 0.861886259272342,
      "grad_norm": 2.09375,
      "learning_rate": 0.000493270750527754,
      "loss": 6.0212,
      "mean_token_accuracy": 0.15546985268592833,
      "num_tokens": 18204779.0,
      "step": 9760
    },
    {
      "entropy": 6.091239595413208,
      "epoch": 0.8623277993641822,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004932631042475241,
      "loss": 5.9414,
      "mean_token_accuracy": 0.16219218373298644,
      "num_tokens": 18214099.0,
      "step": 9765
    },
    {
      "entropy": 6.16319899559021,
      "epoch": 0.8627693394560226,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004932554536916441,
      "loss": 6.0175,
      "mean_token_accuracy": 0.15754653364419938,
      "num_tokens": 18223190.0,
      "step": 9770
    },
    {
      "entropy": 6.178204345703125,
      "epoch": 0.8632108795478629,
      "grad_norm": 3.171875,
      "learning_rate": 0.000493247798860264,
      "loss": 5.9915,
      "mean_token_accuracy": 0.1578979194164276,
      "num_tokens": 18232380.0,
      "step": 9775
    },
    {
      "entropy": 6.176240873336792,
      "epoch": 0.8636524196397033,
      "grad_norm": 4.125,
      "learning_rate": 0.0004932401397535337,
      "loss": 6.0937,
      "mean_token_accuracy": 0.15338415801525115,
      "num_tokens": 18242621.0,
      "step": 9780
    },
    {
      "entropy": 6.2324787139892575,
      "epoch": 0.8640939597315436,
      "grad_norm": 2.5625,
      "learning_rate": 0.000493232476371603,
      "loss": 6.1318,
      "mean_token_accuracy": 0.15454535335302352,
      "num_tokens": 18253183.0,
      "step": 9785
    },
    {
      "entropy": 6.208017826080322,
      "epoch": 0.864535499823384,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004932248087146223,
      "loss": 5.9138,
      "mean_token_accuracy": 0.15757145136594772,
      "num_tokens": 18261967.0,
      "step": 9790
    },
    {
      "entropy": 6.152089548110962,
      "epoch": 0.8649770399152243,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004932171367827417,
      "loss": 6.0224,
      "mean_token_accuracy": 0.15886688679456712,
      "num_tokens": 18270902.0,
      "step": 9795
    },
    {
      "entropy": 6.119718074798584,
      "epoch": 0.8654185800070646,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004932094605761115,
      "loss": 6.0116,
      "mean_token_accuracy": 0.15925500839948653,
      "num_tokens": 18279387.0,
      "step": 9800
    },
    {
      "entropy": 6.106378841400146,
      "epoch": 0.865860120098905,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004932017800948821,
      "loss": 5.8453,
      "mean_token_accuracy": 0.16939669251441955,
      "num_tokens": 18288363.0,
      "step": 9805
    },
    {
      "entropy": 6.334258270263672,
      "epoch": 0.8663016601907453,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004931940953392038,
      "loss": 6.1468,
      "mean_token_accuracy": 0.14759631007909774,
      "num_tokens": 18297398.0,
      "step": 9810
    },
    {
      "entropy": 6.31011815071106,
      "epoch": 0.8667432002825857,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004931864063092274,
      "loss": 6.0036,
      "mean_token_accuracy": 0.1502399615943432,
      "num_tokens": 18306230.0,
      "step": 9815
    },
    {
      "entropy": 6.187512826919556,
      "epoch": 0.8671847403744259,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004931787130051034,
      "loss": 6.1044,
      "mean_token_accuracy": 0.16098907440900803,
      "num_tokens": 18316395.0,
      "step": 9820
    },
    {
      "entropy": 6.212162637710572,
      "epoch": 0.8676262804662663,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004931710154269824,
      "loss": 6.0132,
      "mean_token_accuracy": 0.16348406225442885,
      "num_tokens": 18326020.0,
      "step": 9825
    },
    {
      "entropy": 6.163374042510986,
      "epoch": 0.8680678205581067,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004931633135750154,
      "loss": 6.0028,
      "mean_token_accuracy": 0.15662624835968017,
      "num_tokens": 18335320.0,
      "step": 9830
    },
    {
      "entropy": 6.153064966201782,
      "epoch": 0.868509360649947,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004931556074493531,
      "loss": 6.0605,
      "mean_token_accuracy": 0.15074578672647476,
      "num_tokens": 18345137.0,
      "step": 9835
    },
    {
      "entropy": 6.156153726577759,
      "epoch": 0.8689509007417874,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004931478970501465,
      "loss": 6.0596,
      "mean_token_accuracy": 0.15259919315576553,
      "num_tokens": 18354289.0,
      "step": 9840
    },
    {
      "entropy": 6.29138593673706,
      "epoch": 0.8693924408336277,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004931401823775466,
      "loss": 6.135,
      "mean_token_accuracy": 0.14294005930423737,
      "num_tokens": 18363701.0,
      "step": 9845
    },
    {
      "entropy": 6.255833435058594,
      "epoch": 0.869833980925468,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004931324634317047,
      "loss": 6.0139,
      "mean_token_accuracy": 0.15377718731760978,
      "num_tokens": 18373336.0,
      "step": 9850
    },
    {
      "entropy": 6.233364200592041,
      "epoch": 0.8702755210173083,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004931247402127719,
      "loss": 6.0086,
      "mean_token_accuracy": 0.1552053853869438,
      "num_tokens": 18383233.0,
      "step": 9855
    },
    {
      "entropy": 6.228805875778198,
      "epoch": 0.8707170611091487,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004931170127208995,
      "loss": 6.0661,
      "mean_token_accuracy": 0.15398486107587814,
      "num_tokens": 18393214.0,
      "step": 9860
    },
    {
      "entropy": 6.146592283248902,
      "epoch": 0.8711586012009891,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004931092809562388,
      "loss": 5.946,
      "mean_token_accuracy": 0.1565675213932991,
      "num_tokens": 18401773.0,
      "step": 9865
    },
    {
      "entropy": 6.112184190750122,
      "epoch": 0.8716001412928294,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004931015449189414,
      "loss": 5.9951,
      "mean_token_accuracy": 0.15512312427163125,
      "num_tokens": 18411748.0,
      "step": 9870
    },
    {
      "entropy": 6.177978467941284,
      "epoch": 0.8720416813846698,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004930938046091587,
      "loss": 6.0476,
      "mean_token_accuracy": 0.15289223715662956,
      "num_tokens": 18421399.0,
      "step": 9875
    },
    {
      "entropy": 6.1912150382995605,
      "epoch": 0.87248322147651,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004930860600270425,
      "loss": 6.048,
      "mean_token_accuracy": 0.15060875192284584,
      "num_tokens": 18430052.0,
      "step": 9880
    },
    {
      "entropy": 6.186047840118408,
      "epoch": 0.8729247615683504,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004930783111727443,
      "loss": 5.8845,
      "mean_token_accuracy": 0.16470766216516494,
      "num_tokens": 18438268.0,
      "step": 9885
    },
    {
      "entropy": 6.079912853240967,
      "epoch": 0.8733663016601907,
      "grad_norm": 2.3125,
      "learning_rate": 0.000493070558046416,
      "loss": 6.0036,
      "mean_token_accuracy": 0.15994871109724046,
      "num_tokens": 18446964.0,
      "step": 9890
    },
    {
      "entropy": 6.154115200042725,
      "epoch": 0.8738078417520311,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004930628006482097,
      "loss": 5.9974,
      "mean_token_accuracy": 0.15065375491976737,
      "num_tokens": 18455701.0,
      "step": 9895
    },
    {
      "entropy": 6.197438716888428,
      "epoch": 0.8742493818438715,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004930550389782769,
      "loss": 5.9724,
      "mean_token_accuracy": 0.15478905737400056,
      "num_tokens": 18465635.0,
      "step": 9900
    },
    {
      "entropy": 6.170751237869263,
      "epoch": 0.8746909219357117,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004930472730367701,
      "loss": 5.9129,
      "mean_token_accuracy": 0.15688419491052627,
      "num_tokens": 18474833.0,
      "step": 9905
    },
    {
      "entropy": 6.145848369598388,
      "epoch": 0.8751324620275521,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004930395028238412,
      "loss": 5.9868,
      "mean_token_accuracy": 0.15899734795093537,
      "num_tokens": 18482987.0,
      "step": 9910
    },
    {
      "entropy": 6.132942628860474,
      "epoch": 0.8755740021193924,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004930317283396423,
      "loss": 5.9045,
      "mean_token_accuracy": 0.1640846237540245,
      "num_tokens": 18491274.0,
      "step": 9915
    },
    {
      "entropy": 6.167538642883301,
      "epoch": 0.8760155422112328,
      "grad_norm": 3.46875,
      "learning_rate": 0.000493023949584326,
      "loss": 6.0527,
      "mean_token_accuracy": 0.15335492864251138,
      "num_tokens": 18500811.0,
      "step": 9920
    },
    {
      "entropy": 6.064701843261719,
      "epoch": 0.8764570823030731,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004930161665580445,
      "loss": 5.8384,
      "mean_token_accuracy": 0.17185902297496797,
      "num_tokens": 18509475.0,
      "step": 9925
    },
    {
      "entropy": 6.170683908462524,
      "epoch": 0.8768986223949135,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004930083792609502,
      "loss": 5.9832,
      "mean_token_accuracy": 0.1548695906996727,
      "num_tokens": 18519410.0,
      "step": 9930
    },
    {
      "entropy": 5.974826192855835,
      "epoch": 0.8773401624867538,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004930005876931958,
      "loss": 5.8249,
      "mean_token_accuracy": 0.17380041480064393,
      "num_tokens": 18529194.0,
      "step": 9935
    },
    {
      "entropy": 6.168081188201905,
      "epoch": 0.8777817025785941,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004929927918549339,
      "loss": 6.0463,
      "mean_token_accuracy": 0.14996799752116202,
      "num_tokens": 18539166.0,
      "step": 9940
    },
    {
      "entropy": 6.2125404357910154,
      "epoch": 0.8782232426704345,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004929849917463171,
      "loss": 5.9887,
      "mean_token_accuracy": 0.16084574833512305,
      "num_tokens": 18548166.0,
      "step": 9945
    },
    {
      "entropy": 6.147589111328125,
      "epoch": 0.8786647827622748,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004929771873674984,
      "loss": 5.985,
      "mean_token_accuracy": 0.1554001420736313,
      "num_tokens": 18557041.0,
      "step": 9950
    },
    {
      "entropy": 6.150298070907593,
      "epoch": 0.8791063228541152,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004929693787186305,
      "loss": 6.0116,
      "mean_token_accuracy": 0.15107946917414666,
      "num_tokens": 18566461.0,
      "step": 9955
    },
    {
      "entropy": 6.1634551048278805,
      "epoch": 0.8795478629459554,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004929615657998664,
      "loss": 5.9419,
      "mean_token_accuracy": 0.1572975307703018,
      "num_tokens": 18576334.0,
      "step": 9960
    },
    {
      "entropy": 6.205463933944702,
      "epoch": 0.8799894030377958,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004929537486113593,
      "loss": 5.9623,
      "mean_token_accuracy": 0.1574566476047039,
      "num_tokens": 18584882.0,
      "step": 9965
    },
    {
      "entropy": 6.226665925979614,
      "epoch": 0.8804309431296362,
      "grad_norm": 2.5,
      "learning_rate": 0.0004929459271532621,
      "loss": 6.0103,
      "mean_token_accuracy": 0.1574738770723343,
      "num_tokens": 18593669.0,
      "step": 9970
    },
    {
      "entropy": 6.148843240737915,
      "epoch": 0.8808724832214765,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004929381014257282,
      "loss": 6.0501,
      "mean_token_accuracy": 0.1535251572728157,
      "num_tokens": 18603357.0,
      "step": 9975
    },
    {
      "entropy": 6.149325132369995,
      "epoch": 0.8813140233133169,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004929302714289107,
      "loss": 5.9397,
      "mean_token_accuracy": 0.1659372180700302,
      "num_tokens": 18611856.0,
      "step": 9980
    },
    {
      "entropy": 6.14805383682251,
      "epoch": 0.8817555634051572,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004929224371629634,
      "loss": 6.0145,
      "mean_token_accuracy": 0.1571170061826706,
      "num_tokens": 18620500.0,
      "step": 9985
    },
    {
      "entropy": 6.111777973175049,
      "epoch": 0.8821971034969975,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004929145986280394,
      "loss": 5.9444,
      "mean_token_accuracy": 0.16665309220552443,
      "num_tokens": 18628836.0,
      "step": 9990
    },
    {
      "entropy": 6.169977283477783,
      "epoch": 0.8826386435888378,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004929067558242922,
      "loss": 5.8904,
      "mean_token_accuracy": 0.16612378358840943,
      "num_tokens": 18636831.0,
      "step": 9995
    },
    {
      "entropy": 6.173154354095459,
      "epoch": 0.8830801836806782,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004928989087518758,
      "loss": 6.0369,
      "mean_token_accuracy": 0.1609070673584938,
      "num_tokens": 18646327.0,
      "step": 10000
    },
    {
      "entropy": 6.114136409759522,
      "epoch": 0.8835217237725186,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004928910574109435,
      "loss": 6.0673,
      "mean_token_accuracy": 0.15682611614465714,
      "num_tokens": 18656125.0,
      "step": 10005
    },
    {
      "entropy": 6.20459623336792,
      "epoch": 0.8839632638643589,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004928832018016495,
      "loss": 6.0328,
      "mean_token_accuracy": 0.15780654102563857,
      "num_tokens": 18665853.0,
      "step": 10010
    },
    {
      "entropy": 6.258661460876465,
      "epoch": 0.8844048039561992,
      "grad_norm": 2.125,
      "learning_rate": 0.0004928753419241472,
      "loss": 6.0983,
      "mean_token_accuracy": 0.15185749977827073,
      "num_tokens": 18676674.0,
      "step": 10015
    },
    {
      "entropy": 6.1970985412597654,
      "epoch": 0.8848463440480395,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004928674777785912,
      "loss": 5.9439,
      "mean_token_accuracy": 0.16666018813848496,
      "num_tokens": 18685535.0,
      "step": 10020
    },
    {
      "entropy": 6.1253903865814205,
      "epoch": 0.8852878841398799,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004928596093651351,
      "loss": 5.9636,
      "mean_token_accuracy": 0.15550997406244277,
      "num_tokens": 18694186.0,
      "step": 10025
    },
    {
      "entropy": 6.115320158004761,
      "epoch": 0.8857294242317202,
      "grad_norm": 4.875,
      "learning_rate": 0.000492851736683933,
      "loss": 5.9543,
      "mean_token_accuracy": 0.1645532190799713,
      "num_tokens": 18703061.0,
      "step": 10030
    },
    {
      "entropy": 6.151181602478028,
      "epoch": 0.8861709643235606,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004928438597351395,
      "loss": 5.9118,
      "mean_token_accuracy": 0.16242409944534303,
      "num_tokens": 18712466.0,
      "step": 10035
    },
    {
      "entropy": 6.182442283630371,
      "epoch": 0.886612504415401,
      "grad_norm": 2.25,
      "learning_rate": 0.0004928359785189086,
      "loss": 6.0178,
      "mean_token_accuracy": 0.15392802581191062,
      "num_tokens": 18721522.0,
      "step": 10040
    },
    {
      "entropy": 6.12618727684021,
      "epoch": 0.8870540445072412,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004928280930353948,
      "loss": 5.9416,
      "mean_token_accuracy": 0.15530440360307693,
      "num_tokens": 18730488.0,
      "step": 10045
    },
    {
      "entropy": 6.158182525634766,
      "epoch": 0.8874955845990816,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004928202032847525,
      "loss": 5.8864,
      "mean_token_accuracy": 0.16544022411108017,
      "num_tokens": 18740318.0,
      "step": 10050
    },
    {
      "entropy": 6.146950149536133,
      "epoch": 0.8879371246909219,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004928123092671362,
      "loss": 5.9917,
      "mean_token_accuracy": 0.15361434519290923,
      "num_tokens": 18748731.0,
      "step": 10055
    },
    {
      "entropy": 6.050906038284301,
      "epoch": 0.8883786647827623,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004928044109827008,
      "loss": 5.9194,
      "mean_token_accuracy": 0.16573346555233,
      "num_tokens": 18757847.0,
      "step": 10060
    },
    {
      "entropy": 6.267030811309814,
      "epoch": 0.8888202048746026,
      "grad_norm": 5.59375,
      "learning_rate": 0.0004927965084316007,
      "loss": 6.0377,
      "mean_token_accuracy": 0.1447622910141945,
      "num_tokens": 18767429.0,
      "step": 10065
    },
    {
      "entropy": 6.183168506622314,
      "epoch": 0.889261744966443,
      "grad_norm": 1.875,
      "learning_rate": 0.000492788601613991,
      "loss": 5.9443,
      "mean_token_accuracy": 0.1628161758184433,
      "num_tokens": 18776652.0,
      "step": 10070
    },
    {
      "entropy": 6.131083583831787,
      "epoch": 0.8897032850582833,
      "grad_norm": 3.5,
      "learning_rate": 0.0004927806905300265,
      "loss": 5.9772,
      "mean_token_accuracy": 0.15684217661619188,
      "num_tokens": 18785687.0,
      "step": 10075
    },
    {
      "entropy": 6.125252103805542,
      "epoch": 0.8901448251501236,
      "grad_norm": 2.203125,
      "learning_rate": 0.000492772775179862,
      "loss": 6.0162,
      "mean_token_accuracy": 0.15839558988809585,
      "num_tokens": 18794924.0,
      "step": 10080
    },
    {
      "entropy": 6.181124544143676,
      "epoch": 0.890586365241964,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004927648555636528,
      "loss": 5.8787,
      "mean_token_accuracy": 0.1720139965415001,
      "num_tokens": 18803359.0,
      "step": 10085
    },
    {
      "entropy": 6.148335695266724,
      "epoch": 0.8910279053338043,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004927569316815539,
      "loss": 6.1219,
      "mean_token_accuracy": 0.14714574366807937,
      "num_tokens": 18813496.0,
      "step": 10090
    },
    {
      "entropy": 6.149403619766235,
      "epoch": 0.8914694454256447,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004927490035337205,
      "loss": 5.9629,
      "mean_token_accuracy": 0.16067009270191193,
      "num_tokens": 18823274.0,
      "step": 10095
    },
    {
      "entropy": 6.1795487880706785,
      "epoch": 0.8919109855174849,
      "grad_norm": 1.8828125,
      "learning_rate": 0.000492741071120308,
      "loss": 5.9607,
      "mean_token_accuracy": 0.16393790245056153,
      "num_tokens": 18832436.0,
      "step": 10100
    },
    {
      "entropy": 6.214590406417846,
      "epoch": 0.8923525256093253,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004927331344414717,
      "loss": 6.0233,
      "mean_token_accuracy": 0.15794878751039504,
      "num_tokens": 18841923.0,
      "step": 10105
    },
    {
      "entropy": 6.201234197616577,
      "epoch": 0.8927940657011657,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004927251934973672,
      "loss": 6.0736,
      "mean_token_accuracy": 0.15925167202949525,
      "num_tokens": 18850957.0,
      "step": 10110
    },
    {
      "entropy": 6.264265060424805,
      "epoch": 0.893235605793006,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004927172482881499,
      "loss": 6.0676,
      "mean_token_accuracy": 0.15099077001214029,
      "num_tokens": 18859622.0,
      "step": 10115
    },
    {
      "entropy": 6.223895263671875,
      "epoch": 0.8936771458848464,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004927092988139755,
      "loss": 6.0428,
      "mean_token_accuracy": 0.15183982253074646,
      "num_tokens": 18868555.0,
      "step": 10120
    },
    {
      "entropy": 6.211693000793457,
      "epoch": 0.8941186859766866,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004927013450749999,
      "loss": 6.019,
      "mean_token_accuracy": 0.15297604501247405,
      "num_tokens": 18877445.0,
      "step": 10125
    },
    {
      "entropy": 6.171515369415284,
      "epoch": 0.894560226068527,
      "grad_norm": 7.59375,
      "learning_rate": 0.0004926933870713787,
      "loss": 5.9209,
      "mean_token_accuracy": 0.15988947749137877,
      "num_tokens": 18886386.0,
      "step": 10130
    },
    {
      "entropy": 6.149491453170777,
      "epoch": 0.8950017661603673,
      "grad_norm": 2.125,
      "learning_rate": 0.0004926854248032678,
      "loss": 6.0348,
      "mean_token_accuracy": 0.15640042424201966,
      "num_tokens": 18895016.0,
      "step": 10135
    },
    {
      "entropy": 6.085942983627319,
      "epoch": 0.8954433062522077,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004926774582708234,
      "loss": 5.8979,
      "mean_token_accuracy": 0.16461617052555083,
      "num_tokens": 18904126.0,
      "step": 10140
    },
    {
      "entropy": 6.114335441589356,
      "epoch": 0.8958848463440481,
      "grad_norm": 2.0,
      "learning_rate": 0.0004926694874742012,
      "loss": 5.9586,
      "mean_token_accuracy": 0.1573253482580185,
      "num_tokens": 18913064.0,
      "step": 10145
    },
    {
      "entropy": 6.165401983261108,
      "epoch": 0.8963263864358884,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004926615124135577,
      "loss": 6.0607,
      "mean_token_accuracy": 0.1534641534090042,
      "num_tokens": 18921189.0,
      "step": 10150
    },
    {
      "entropy": 6.199530220031738,
      "epoch": 0.8967679265277287,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004926535330890488,
      "loss": 5.974,
      "mean_token_accuracy": 0.159789377450943,
      "num_tokens": 18930818.0,
      "step": 10155
    },
    {
      "entropy": 6.257216835021973,
      "epoch": 0.897209466619569,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004926455495008311,
      "loss": 5.9908,
      "mean_token_accuracy": 0.15595454275608062,
      "num_tokens": 18940376.0,
      "step": 10160
    },
    {
      "entropy": 6.172543859481811,
      "epoch": 0.8976510067114094,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004926375616490608,
      "loss": 6.0041,
      "mean_token_accuracy": 0.16691479980945587,
      "num_tokens": 18949418.0,
      "step": 10165
    },
    {
      "entropy": 6.140330648422241,
      "epoch": 0.8980925468032497,
      "grad_norm": 7.15625,
      "learning_rate": 0.0004926295695338945,
      "loss": 5.9703,
      "mean_token_accuracy": 0.1607614979147911,
      "num_tokens": 18959315.0,
      "step": 10170
    },
    {
      "entropy": 6.1289197444915775,
      "epoch": 0.8985340868950901,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004926215731554887,
      "loss": 5.8859,
      "mean_token_accuracy": 0.16685865074396133,
      "num_tokens": 18968336.0,
      "step": 10175
    },
    {
      "entropy": 6.141733074188233,
      "epoch": 0.8989756269869305,
      "grad_norm": 1.6875,
      "learning_rate": 0.000492613572514,
      "loss": 5.958,
      "mean_token_accuracy": 0.1616265818476677,
      "num_tokens": 18977123.0,
      "step": 10180
    },
    {
      "entropy": 6.176858758926391,
      "epoch": 0.8994171670787707,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004926055676095851,
      "loss": 6.0825,
      "mean_token_accuracy": 0.1572777435183525,
      "num_tokens": 18986150.0,
      "step": 10185
    },
    {
      "entropy": 6.252068185806275,
      "epoch": 0.8998587071706111,
      "grad_norm": 1.625,
      "learning_rate": 0.0004925975584424012,
      "loss": 6.0836,
      "mean_token_accuracy": 0.15544864535331726,
      "num_tokens": 18996010.0,
      "step": 10190
    },
    {
      "entropy": 6.203353500366211,
      "epoch": 0.9003002472624514,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004925895450126046,
      "loss": 6.0569,
      "mean_token_accuracy": 0.14781340956687927,
      "num_tokens": 19005100.0,
      "step": 10195
    },
    {
      "entropy": 6.158134269714355,
      "epoch": 0.9007417873542918,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004925815273203526,
      "loss": 5.9553,
      "mean_token_accuracy": 0.16117661744356154,
      "num_tokens": 19013844.0,
      "step": 10200
    },
    {
      "entropy": 6.237428331375122,
      "epoch": 0.901183327446132,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004925735053658022,
      "loss": 6.101,
      "mean_token_accuracy": 0.1538894847035408,
      "num_tokens": 19024218.0,
      "step": 10205
    },
    {
      "entropy": 6.201617336273193,
      "epoch": 0.9016248675379724,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004925654791491106,
      "loss": 6.0243,
      "mean_token_accuracy": 0.1535754010081291,
      "num_tokens": 19034725.0,
      "step": 10210
    },
    {
      "entropy": 6.197865295410156,
      "epoch": 0.9020664076298128,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004925574486704351,
      "loss": 6.0435,
      "mean_token_accuracy": 0.1501529857516289,
      "num_tokens": 19044597.0,
      "step": 10215
    },
    {
      "entropy": 6.275617885589599,
      "epoch": 0.9025079477216531,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004925494139299329,
      "loss": 6.176,
      "mean_token_accuracy": 0.14526568949222565,
      "num_tokens": 19055124.0,
      "step": 10220
    },
    {
      "entropy": 6.176630258560181,
      "epoch": 0.9029494878134935,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004925413749277613,
      "loss": 5.9432,
      "mean_token_accuracy": 0.15719038546085357,
      "num_tokens": 19064516.0,
      "step": 10225
    },
    {
      "entropy": 6.128312587738037,
      "epoch": 0.9033910279053338,
      "grad_norm": 2.375,
      "learning_rate": 0.0004925333316640779,
      "loss": 5.9976,
      "mean_token_accuracy": 0.161532860994339,
      "num_tokens": 19072926.0,
      "step": 10230
    },
    {
      "entropy": 6.2058759212493895,
      "epoch": 0.9038325679971742,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004925252841390403,
      "loss": 6.0408,
      "mean_token_accuracy": 0.1520572617650032,
      "num_tokens": 19082313.0,
      "step": 10235
    },
    {
      "entropy": 6.148319673538208,
      "epoch": 0.9042741080890144,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004925172323528061,
      "loss": 5.9824,
      "mean_token_accuracy": 0.1626274913549423,
      "num_tokens": 19091891.0,
      "step": 10240
    },
    {
      "entropy": 6.185559797286987,
      "epoch": 0.9047156481808548,
      "grad_norm": 1.9375,
      "learning_rate": 0.000492509176305533,
      "loss": 5.9893,
      "mean_token_accuracy": 0.15698708444833756,
      "num_tokens": 19100311.0,
      "step": 10245
    },
    {
      "entropy": 6.193223094940185,
      "epoch": 0.9051571882726952,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004925011159973788,
      "loss": 6.0128,
      "mean_token_accuracy": 0.1560434177517891,
      "num_tokens": 19110463.0,
      "step": 10250
    },
    {
      "entropy": 6.235753107070923,
      "epoch": 0.9055987283645355,
      "grad_norm": 5.3125,
      "learning_rate": 0.0004924930514285015,
      "loss": 5.9917,
      "mean_token_accuracy": 0.1641075938940048,
      "num_tokens": 19120723.0,
      "step": 10255
    },
    {
      "entropy": 6.142373132705688,
      "epoch": 0.9060402684563759,
      "grad_norm": 2.21875,
      "learning_rate": 0.000492484982599059,
      "loss": 6.007,
      "mean_token_accuracy": 0.15472819954156875,
      "num_tokens": 19130259.0,
      "step": 10260
    },
    {
      "entropy": 6.232355165481567,
      "epoch": 0.9064818085482161,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004924769095092094,
      "loss": 6.0484,
      "mean_token_accuracy": 0.15291386246681213,
      "num_tokens": 19140256.0,
      "step": 10265
    },
    {
      "entropy": 6.187994813919067,
      "epoch": 0.9069233486400565,
      "grad_norm": 4.75,
      "learning_rate": 0.0004924688321591109,
      "loss": 6.0521,
      "mean_token_accuracy": 0.15677061378955842,
      "num_tokens": 19149165.0,
      "step": 10270
    },
    {
      "entropy": 6.072564315795899,
      "epoch": 0.9073648887318968,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004924607505489215,
      "loss": 5.9736,
      "mean_token_accuracy": 0.1617050975561142,
      "num_tokens": 19159254.0,
      "step": 10275
    },
    {
      "entropy": 6.197022867202759,
      "epoch": 0.9078064288237372,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004924526646787998,
      "loss": 6.0264,
      "mean_token_accuracy": 0.15401589646935462,
      "num_tokens": 19168166.0,
      "step": 10280
    },
    {
      "entropy": 6.284684085845948,
      "epoch": 0.9082479689155776,
      "grad_norm": 2.796875,
      "learning_rate": 0.000492444574548904,
      "loss": 5.9976,
      "mean_token_accuracy": 0.15589857399463652,
      "num_tokens": 19176887.0,
      "step": 10285
    },
    {
      "entropy": 6.12396273612976,
      "epoch": 0.9086895090074179,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004924364801593927,
      "loss": 5.9254,
      "mean_token_accuracy": 0.1621394783258438,
      "num_tokens": 19185510.0,
      "step": 10290
    },
    {
      "entropy": 6.0319455623626705,
      "epoch": 0.9091310490992582,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004924283815104243,
      "loss": 5.9667,
      "mean_token_accuracy": 0.16688053756952287,
      "num_tokens": 19195459.0,
      "step": 10295
    },
    {
      "entropy": 6.176657485961914,
      "epoch": 0.9095725891910985,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004924202786021578,
      "loss": 5.989,
      "mean_token_accuracy": 0.15756237059831618,
      "num_tokens": 19204590.0,
      "step": 10300
    },
    {
      "entropy": 6.190572357177734,
      "epoch": 0.9100141292829389,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004924121714347515,
      "loss": 5.9534,
      "mean_token_accuracy": 0.15884590595960618,
      "num_tokens": 19213302.0,
      "step": 10305
    },
    {
      "entropy": 6.197195100784302,
      "epoch": 0.9104556693747792,
      "grad_norm": 2.25,
      "learning_rate": 0.0004924040600083644,
      "loss": 6.0684,
      "mean_token_accuracy": 0.154947692155838,
      "num_tokens": 19223172.0,
      "step": 10310
    },
    {
      "entropy": 6.18108868598938,
      "epoch": 0.9108972094666196,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004923959443231554,
      "loss": 5.9656,
      "mean_token_accuracy": 0.15735325068235398,
      "num_tokens": 19232234.0,
      "step": 10315
    },
    {
      "entropy": 6.096418237686157,
      "epoch": 0.91133874955846,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004923878243792836,
      "loss": 5.9418,
      "mean_token_accuracy": 0.1593674212694168,
      "num_tokens": 19241070.0,
      "step": 10320
    },
    {
      "entropy": 6.177568054199218,
      "epoch": 0.9117802896503002,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004923797001769079,
      "loss": 6.0312,
      "mean_token_accuracy": 0.16176247596740723,
      "num_tokens": 19251094.0,
      "step": 10325
    },
    {
      "entropy": 6.284003639221192,
      "epoch": 0.9122218297421406,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004923715717161875,
      "loss": 6.1177,
      "mean_token_accuracy": 0.1512012779712677,
      "num_tokens": 19260588.0,
      "step": 10330
    },
    {
      "entropy": 6.246473407745361,
      "epoch": 0.9126633698339809,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004923634389972817,
      "loss": 6.1614,
      "mean_token_accuracy": 0.14666686132550238,
      "num_tokens": 19270687.0,
      "step": 10335
    },
    {
      "entropy": 6.129995346069336,
      "epoch": 0.9131049099258213,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004923553020203497,
      "loss": 5.9474,
      "mean_token_accuracy": 0.16146068274974823,
      "num_tokens": 19278739.0,
      "step": 10340
    },
    {
      "entropy": 6.315760231018066,
      "epoch": 0.9135464500176615,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004923471607855511,
      "loss": 6.1081,
      "mean_token_accuracy": 0.14812798649072648,
      "num_tokens": 19289184.0,
      "step": 10345
    },
    {
      "entropy": 6.219142246246338,
      "epoch": 0.9139879901095019,
      "grad_norm": 2.96875,
      "learning_rate": 0.000492339015293045,
      "loss": 5.9659,
      "mean_token_accuracy": 0.1682812362909317,
      "num_tokens": 19298390.0,
      "step": 10350
    },
    {
      "entropy": 6.11060905456543,
      "epoch": 0.9144295302013423,
      "grad_norm": 6.90625,
      "learning_rate": 0.0004923308655429913,
      "loss": 5.9021,
      "mean_token_accuracy": 0.16550161987543105,
      "num_tokens": 19307227.0,
      "step": 10355
    },
    {
      "entropy": 6.1038957118988035,
      "epoch": 0.9148710702931826,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004923227115355495,
      "loss": 5.9588,
      "mean_token_accuracy": 0.15407850295305253,
      "num_tokens": 19317701.0,
      "step": 10360
    },
    {
      "entropy": 6.190516138076783,
      "epoch": 0.915312610385023,
      "grad_norm": 2.375,
      "learning_rate": 0.0004923145532708794,
      "loss": 6.0538,
      "mean_token_accuracy": 0.15957052856683732,
      "num_tokens": 19326983.0,
      "step": 10365
    },
    {
      "entropy": 6.193438386917114,
      "epoch": 0.9157541504768633,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004923063907491408,
      "loss": 6.0553,
      "mean_token_accuracy": 0.15762412548065186,
      "num_tokens": 19337157.0,
      "step": 10370
    },
    {
      "entropy": 6.217849206924439,
      "epoch": 0.9161956905687036,
      "grad_norm": 4.375,
      "learning_rate": 0.0004922982239704936,
      "loss": 6.1498,
      "mean_token_accuracy": 0.15656513199210167,
      "num_tokens": 19346845.0,
      "step": 10375
    },
    {
      "entropy": 6.305586433410644,
      "epoch": 0.9166372306605439,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004922900529350978,
      "loss": 6.027,
      "mean_token_accuracy": 0.15455891340970992,
      "num_tokens": 19356905.0,
      "step": 10380
    },
    {
      "entropy": 6.249670124053955,
      "epoch": 0.9170787707523843,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004922818776431134,
      "loss": 6.012,
      "mean_token_accuracy": 0.1489650472998619,
      "num_tokens": 19366595.0,
      "step": 10385
    },
    {
      "entropy": 6.233332109451294,
      "epoch": 0.9175203108442247,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004922736980947005,
      "loss": 6.0644,
      "mean_token_accuracy": 0.15061762630939485,
      "num_tokens": 19377213.0,
      "step": 10390
    },
    {
      "entropy": 6.136075878143311,
      "epoch": 0.917961850936065,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004922655142900196,
      "loss": 5.9352,
      "mean_token_accuracy": 0.1552181825041771,
      "num_tokens": 19385510.0,
      "step": 10395
    },
    {
      "entropy": 6.085175561904907,
      "epoch": 0.9184033910279054,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004922573262292308,
      "loss": 5.9464,
      "mean_token_accuracy": 0.1548845037817955,
      "num_tokens": 19394505.0,
      "step": 10400
    },
    {
      "entropy": 6.119608592987061,
      "epoch": 0.9188449311197456,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004922491339124945,
      "loss": 5.8446,
      "mean_token_accuracy": 0.16599608659744264,
      "num_tokens": 19403140.0,
      "step": 10405
    },
    {
      "entropy": 6.160563516616821,
      "epoch": 0.919286471211586,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004922409373399713,
      "loss": 6.0634,
      "mean_token_accuracy": 0.15320063084363938,
      "num_tokens": 19412709.0,
      "step": 10410
    },
    {
      "entropy": 6.123647928237915,
      "epoch": 0.9197280113034263,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004922327365118216,
      "loss": 5.8959,
      "mean_token_accuracy": 0.1659088060259819,
      "num_tokens": 19422442.0,
      "step": 10415
    },
    {
      "entropy": 6.180219507217407,
      "epoch": 0.9201695513952667,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004922245314282062,
      "loss": 6.0409,
      "mean_token_accuracy": 0.15570426285266875,
      "num_tokens": 19432153.0,
      "step": 10420
    },
    {
      "entropy": 6.168195962905884,
      "epoch": 0.9206110914871071,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004922163220892858,
      "loss": 5.8333,
      "mean_token_accuracy": 0.17357273697853087,
      "num_tokens": 19440811.0,
      "step": 10425
    },
    {
      "entropy": 6.117291784286499,
      "epoch": 0.9210526315789473,
      "grad_norm": 2.859375,
      "learning_rate": 0.000492208108495221,
      "loss": 6.0344,
      "mean_token_accuracy": 0.15521479845046998,
      "num_tokens": 19451330.0,
      "step": 10430
    },
    {
      "entropy": 6.094224691390991,
      "epoch": 0.9214941716707877,
      "grad_norm": 1.9453125,
      "learning_rate": 0.000492199890646173,
      "loss": 5.8815,
      "mean_token_accuracy": 0.16824816912412643,
      "num_tokens": 19459689.0,
      "step": 10435
    },
    {
      "entropy": 6.163274574279785,
      "epoch": 0.921935711762628,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004921916685423028,
      "loss": 5.943,
      "mean_token_accuracy": 0.16071141958236695,
      "num_tokens": 19468110.0,
      "step": 10440
    },
    {
      "entropy": 6.167320251464844,
      "epoch": 0.9223772518544684,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004921834421837711,
      "loss": 5.9947,
      "mean_token_accuracy": 0.15582612305879592,
      "num_tokens": 19477190.0,
      "step": 10445
    },
    {
      "entropy": 6.2047117233276365,
      "epoch": 0.9228187919463087,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004921752115707394,
      "loss": 6.0946,
      "mean_token_accuracy": 0.15232176333665848,
      "num_tokens": 19485757.0,
      "step": 10450
    },
    {
      "entropy": 6.197108173370362,
      "epoch": 0.9232603320381491,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004921669767033688,
      "loss": 6.0333,
      "mean_token_accuracy": 0.1551002785563469,
      "num_tokens": 19494846.0,
      "step": 10455
    },
    {
      "entropy": 6.180310535430908,
      "epoch": 0.9237018721299894,
      "grad_norm": 9.0625,
      "learning_rate": 0.0004921587375818207,
      "loss": 6.0293,
      "mean_token_accuracy": 0.15100041329860686,
      "num_tokens": 19504307.0,
      "step": 10460
    },
    {
      "entropy": 6.214234972000122,
      "epoch": 0.9241434122218297,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004921504942062564,
      "loss": 5.9805,
      "mean_token_accuracy": 0.15465851426124572,
      "num_tokens": 19512741.0,
      "step": 10465
    },
    {
      "entropy": 6.100347805023193,
      "epoch": 0.9245849523136701,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004921422465768374,
      "loss": 5.8617,
      "mean_token_accuracy": 0.16154198795557023,
      "num_tokens": 19521823.0,
      "step": 10470
    },
    {
      "entropy": 6.15104398727417,
      "epoch": 0.9250264924055104,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004921339946937253,
      "loss": 6.0634,
      "mean_token_accuracy": 0.15580296963453294,
      "num_tokens": 19531048.0,
      "step": 10475
    },
    {
      "entropy": 6.199530696868896,
      "epoch": 0.9254680324973508,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004921257385570818,
      "loss": 6.0152,
      "mean_token_accuracy": 0.15597717314958573,
      "num_tokens": 19539543.0,
      "step": 10480
    },
    {
      "entropy": 6.238590097427368,
      "epoch": 0.925909572589191,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004921174781670685,
      "loss": 6.1305,
      "mean_token_accuracy": 0.15179005190730094,
      "num_tokens": 19549242.0,
      "step": 10485
    },
    {
      "entropy": 6.275363397598267,
      "epoch": 0.9263511126810314,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004921092135238473,
      "loss": 6.0273,
      "mean_token_accuracy": 0.14577364027500153,
      "num_tokens": 19558259.0,
      "step": 10490
    },
    {
      "entropy": 6.22588005065918,
      "epoch": 0.9267926527728718,
      "grad_norm": 2.390625,
      "learning_rate": 0.00049210094462758,
      "loss": 6.0251,
      "mean_token_accuracy": 0.16097075939178468,
      "num_tokens": 19567327.0,
      "step": 10495
    },
    {
      "entropy": 6.193928194046021,
      "epoch": 0.9272341928647121,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004920926714784288,
      "loss": 6.0363,
      "mean_token_accuracy": 0.15664020031690598,
      "num_tokens": 19576696.0,
      "step": 10500
    },
    {
      "entropy": 6.151621580123901,
      "epoch": 0.9276757329565525,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004920843940765555,
      "loss": 5.9212,
      "mean_token_accuracy": 0.16299362182617189,
      "num_tokens": 19585708.0,
      "step": 10505
    },
    {
      "entropy": 6.2330797672271725,
      "epoch": 0.9281172730483928,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004920761124221224,
      "loss": 6.0814,
      "mean_token_accuracy": 0.16604722887277604,
      "num_tokens": 19595199.0,
      "step": 10510
    },
    {
      "entropy": 6.218397569656372,
      "epoch": 0.9285588131402331,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004920678265152919,
      "loss": 6.0204,
      "mean_token_accuracy": 0.14686674624681473,
      "num_tokens": 19604781.0,
      "step": 10515
    },
    {
      "entropy": 6.154498529434204,
      "epoch": 0.9290003532320734,
      "grad_norm": 2.0625,
      "learning_rate": 0.000492059536356226,
      "loss": 5.9744,
      "mean_token_accuracy": 0.16245217323303224,
      "num_tokens": 19614203.0,
      "step": 10520
    },
    {
      "entropy": 6.171329021453857,
      "epoch": 0.9294418933239138,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004920512419450872,
      "loss": 5.9991,
      "mean_token_accuracy": 0.15724168121814727,
      "num_tokens": 19623664.0,
      "step": 10525
    },
    {
      "entropy": 6.220635652542114,
      "epoch": 0.9298834334157542,
      "grad_norm": 2.265625,
      "learning_rate": 0.000492042943282038,
      "loss": 5.9778,
      "mean_token_accuracy": 0.15525819957256318,
      "num_tokens": 19632960.0,
      "step": 10530
    },
    {
      "entropy": 6.1476593017578125,
      "epoch": 0.9303249735075945,
      "grad_norm": 2.28125,
      "learning_rate": 0.000492034640367241,
      "loss": 6.0964,
      "mean_token_accuracy": 0.1484966441988945,
      "num_tokens": 19643087.0,
      "step": 10535
    },
    {
      "entropy": 6.23501410484314,
      "epoch": 0.9307665135994349,
      "grad_norm": 2.8125,
      "learning_rate": 0.000492026333200859,
      "loss": 6.0565,
      "mean_token_accuracy": 0.15510230362415314,
      "num_tokens": 19653015.0,
      "step": 10540
    },
    {
      "entropy": 6.26780948638916,
      "epoch": 0.9312080536912751,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004920180217830543,
      "loss": 6.0616,
      "mean_token_accuracy": 0.1520492762327194,
      "num_tokens": 19661569.0,
      "step": 10545
    },
    {
      "entropy": 6.125591278076172,
      "epoch": 0.9316495937831155,
      "grad_norm": 2.453125,
      "learning_rate": 0.00049200970611399,
      "loss": 5.8859,
      "mean_token_accuracy": 0.16664395481348038,
      "num_tokens": 19669559.0,
      "step": 10550
    },
    {
      "entropy": 6.075623226165772,
      "epoch": 0.9320911338749558,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004920013861938291,
      "loss": 5.9133,
      "mean_token_accuracy": 0.167872416973114,
      "num_tokens": 19678766.0,
      "step": 10555
    },
    {
      "entropy": 6.12970929145813,
      "epoch": 0.9325326739667962,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004919930620227345,
      "loss": 6.0406,
      "mean_token_accuracy": 0.1478048160672188,
      "num_tokens": 19688422.0,
      "step": 10560
    },
    {
      "entropy": 6.293255186080932,
      "epoch": 0.9329742140586366,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004919847336008692,
      "loss": 6.078,
      "mean_token_accuracy": 0.152406769990921,
      "num_tokens": 19697206.0,
      "step": 10565
    },
    {
      "entropy": 6.224672317504883,
      "epoch": 0.9334157541504768,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004919764009283963,
      "loss": 6.0471,
      "mean_token_accuracy": 0.14844200015068054,
      "num_tokens": 19706065.0,
      "step": 10570
    },
    {
      "entropy": 6.173615884780884,
      "epoch": 0.9338572942423172,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004919680640054792,
      "loss": 5.9629,
      "mean_token_accuracy": 0.15830276012420655,
      "num_tokens": 19715402.0,
      "step": 10575
    },
    {
      "entropy": 6.136301708221436,
      "epoch": 0.9342988343341575,
      "grad_norm": 2.46875,
      "learning_rate": 0.000491959722832281,
      "loss": 5.9186,
      "mean_token_accuracy": 0.16106877475976944,
      "num_tokens": 19724328.0,
      "step": 10580
    },
    {
      "entropy": 6.099761629104615,
      "epoch": 0.9347403744259979,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004919513774089653,
      "loss": 6.0097,
      "mean_token_accuracy": 0.16225920021533966,
      "num_tokens": 19734087.0,
      "step": 10585
    },
    {
      "entropy": 6.261999607086182,
      "epoch": 0.9351819145178382,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004919430277356955,
      "loss": 5.988,
      "mean_token_accuracy": 0.16102648824453353,
      "num_tokens": 19743010.0,
      "step": 10590
    },
    {
      "entropy": 6.181056118011474,
      "epoch": 0.9356234546096786,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004919346738126351,
      "loss": 5.9494,
      "mean_token_accuracy": 0.16381199955940245,
      "num_tokens": 19752212.0,
      "step": 10595
    },
    {
      "entropy": 6.174293947219849,
      "epoch": 0.9360649947015189,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004919263156399478,
      "loss": 6.015,
      "mean_token_accuracy": 0.15577296018600464,
      "num_tokens": 19762496.0,
      "step": 10600
    },
    {
      "entropy": 6.186666774749756,
      "epoch": 0.9365065347933592,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004919179532177976,
      "loss": 6.0245,
      "mean_token_accuracy": 0.15240861624479293,
      "num_tokens": 19772665.0,
      "step": 10605
    },
    {
      "entropy": 6.137942838668823,
      "epoch": 0.9369480748851996,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004919095865463479,
      "loss": 5.8469,
      "mean_token_accuracy": 0.16629263162612914,
      "num_tokens": 19781639.0,
      "step": 10610
    },
    {
      "entropy": 6.17881965637207,
      "epoch": 0.9373896149770399,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004919012156257627,
      "loss": 5.9596,
      "mean_token_accuracy": 0.1548631563782692,
      "num_tokens": 19791475.0,
      "step": 10615
    },
    {
      "entropy": 6.12343921661377,
      "epoch": 0.9378311550688803,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004918928404562061,
      "loss": 5.9502,
      "mean_token_accuracy": 0.15676655918359755,
      "num_tokens": 19800967.0,
      "step": 10620
    },
    {
      "entropy": 6.125112295150757,
      "epoch": 0.9382726951607206,
      "grad_norm": 2.625,
      "learning_rate": 0.0004918844610378421,
      "loss": 6.0685,
      "mean_token_accuracy": 0.15577329844236373,
      "num_tokens": 19810303.0,
      "step": 10625
    },
    {
      "entropy": 6.278899049758911,
      "epoch": 0.9387142352525609,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004918760773708348,
      "loss": 5.9912,
      "mean_token_accuracy": 0.15593978464603425,
      "num_tokens": 19819049.0,
      "step": 10630
    },
    {
      "entropy": 6.300095891952514,
      "epoch": 0.9391557753444013,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004918676894553485,
      "loss": 6.0213,
      "mean_token_accuracy": 0.15708844661712645,
      "num_tokens": 19828253.0,
      "step": 10635
    },
    {
      "entropy": 6.178868627548217,
      "epoch": 0.9395973154362416,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004918592972915476,
      "loss": 5.9873,
      "mean_token_accuracy": 0.16300770342350007,
      "num_tokens": 19837673.0,
      "step": 10640
    },
    {
      "entropy": 6.079726028442383,
      "epoch": 0.940038855528082,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004918509008795963,
      "loss": 6.0048,
      "mean_token_accuracy": 0.15207261741161346,
      "num_tokens": 19846289.0,
      "step": 10645
    },
    {
      "entropy": 6.096834373474121,
      "epoch": 0.9404803956199222,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004918425002196594,
      "loss": 5.9845,
      "mean_token_accuracy": 0.1529189318418503,
      "num_tokens": 19855384.0,
      "step": 10650
    },
    {
      "entropy": 6.148316192626953,
      "epoch": 0.9409219357117626,
      "grad_norm": 15.9375,
      "learning_rate": 0.000491834095311901,
      "loss": 5.9311,
      "mean_token_accuracy": 0.16442809849977494,
      "num_tokens": 19864688.0,
      "step": 10655
    },
    {
      "entropy": 6.205455541610718,
      "epoch": 0.941363475803603,
      "grad_norm": 2.375,
      "learning_rate": 0.000491825686156486,
      "loss": 5.9377,
      "mean_token_accuracy": 0.16152185797691346,
      "num_tokens": 19873692.0,
      "step": 10660
    },
    {
      "entropy": 6.235633182525635,
      "epoch": 0.9418050158954433,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004918172727535792,
      "loss": 6.0262,
      "mean_token_accuracy": 0.15295447111129762,
      "num_tokens": 19882063.0,
      "step": 10665
    },
    {
      "entropy": 6.193088674545288,
      "epoch": 0.9422465559872837,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004918088551033454,
      "loss": 5.9956,
      "mean_token_accuracy": 0.15469479113817214,
      "num_tokens": 19891727.0,
      "step": 10670
    },
    {
      "entropy": 6.111576414108276,
      "epoch": 0.942688096079124,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004918004332059493,
      "loss": 6.0266,
      "mean_token_accuracy": 0.16044130325317382,
      "num_tokens": 19901851.0,
      "step": 10675
    },
    {
      "entropy": 6.191736364364624,
      "epoch": 0.9431296361709643,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004917920070615561,
      "loss": 5.9773,
      "mean_token_accuracy": 0.1507607340812683,
      "num_tokens": 19911092.0,
      "step": 10680
    },
    {
      "entropy": 6.124292087554932,
      "epoch": 0.9435711762628046,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004917835766703307,
      "loss": 5.9583,
      "mean_token_accuracy": 0.16241815090179443,
      "num_tokens": 19920842.0,
      "step": 10685
    },
    {
      "entropy": 6.205943632125854,
      "epoch": 0.944012716354645,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004917751420324384,
      "loss": 6.0482,
      "mean_token_accuracy": 0.15033227056264878,
      "num_tokens": 19930833.0,
      "step": 10690
    },
    {
      "entropy": 6.245380783081055,
      "epoch": 0.9444542564464854,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004917667031480444,
      "loss": 6.0621,
      "mean_token_accuracy": 0.15881533920764923,
      "num_tokens": 19940514.0,
      "step": 10695
    },
    {
      "entropy": 6.185396575927735,
      "epoch": 0.9448957965383257,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004917582600173138,
      "loss": 5.8882,
      "mean_token_accuracy": 0.16307059526443482,
      "num_tokens": 19948949.0,
      "step": 10700
    },
    {
      "entropy": 6.0753607749938965,
      "epoch": 0.9453373366301661,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004917498126404123,
      "loss": 5.8861,
      "mean_token_accuracy": 0.16055179089307786,
      "num_tokens": 19958410.0,
      "step": 10705
    },
    {
      "entropy": 6.1283721923828125,
      "epoch": 0.9457788767220063,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004917413610175052,
      "loss": 5.9033,
      "mean_token_accuracy": 0.16776129007339477,
      "num_tokens": 19968160.0,
      "step": 10710
    },
    {
      "entropy": 6.223766326904297,
      "epoch": 0.9462204168138467,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004917329051487582,
      "loss": 6.0449,
      "mean_token_accuracy": 0.1501573257148266,
      "num_tokens": 19976811.0,
      "step": 10715
    },
    {
      "entropy": 6.282238340377807,
      "epoch": 0.946661956905687,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004917244450343368,
      "loss": 6.1145,
      "mean_token_accuracy": 0.14723313003778457,
      "num_tokens": 19986326.0,
      "step": 10720
    },
    {
      "entropy": 6.090417909622192,
      "epoch": 0.9471034969975274,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004917159806744067,
      "loss": 5.8476,
      "mean_token_accuracy": 0.15740818083286284,
      "num_tokens": 19996255.0,
      "step": 10725
    },
    {
      "entropy": 6.161284685134888,
      "epoch": 0.9475450370893678,
      "grad_norm": 6.375,
      "learning_rate": 0.0004917075120691338,
      "loss": 6.0485,
      "mean_token_accuracy": 0.1610925927758217,
      "num_tokens": 20004805.0,
      "step": 10730
    },
    {
      "entropy": 6.204595232009888,
      "epoch": 0.947986577181208,
      "grad_norm": 2.125,
      "learning_rate": 0.0004916990392186841,
      "loss": 5.949,
      "mean_token_accuracy": 0.1597094565629959,
      "num_tokens": 20014189.0,
      "step": 10735
    },
    {
      "entropy": 6.234174203872681,
      "epoch": 0.9484281172730484,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004916905621232233,
      "loss": 5.9548,
      "mean_token_accuracy": 0.16005349457263945,
      "num_tokens": 20023261.0,
      "step": 10740
    },
    {
      "entropy": 6.092266082763672,
      "epoch": 0.9488696573648887,
      "grad_norm": 5.9375,
      "learning_rate": 0.0004916820807829177,
      "loss": 5.9201,
      "mean_token_accuracy": 0.16162907779216767,
      "num_tokens": 20032433.0,
      "step": 10745
    },
    {
      "entropy": 6.102322673797607,
      "epoch": 0.9493111974567291,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004916735951979335,
      "loss": 6.039,
      "mean_token_accuracy": 0.15769112557172776,
      "num_tokens": 20041455.0,
      "step": 10750
    },
    {
      "entropy": 6.188523387908935,
      "epoch": 0.9497527375485694,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004916651053684366,
      "loss": 6.0707,
      "mean_token_accuracy": 0.15521103143692017,
      "num_tokens": 20050088.0,
      "step": 10755
    },
    {
      "entropy": 6.220887088775635,
      "epoch": 0.9501942776404098,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004916566112945936,
      "loss": 6.015,
      "mean_token_accuracy": 0.15838380306959152,
      "num_tokens": 20059114.0,
      "step": 10760
    },
    {
      "entropy": 6.190140247344971,
      "epoch": 0.9506358177322501,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004916481129765708,
      "loss": 6.0082,
      "mean_token_accuracy": 0.15650569051504135,
      "num_tokens": 20068692.0,
      "step": 10765
    },
    {
      "entropy": 6.13848032951355,
      "epoch": 0.9510773578240904,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004916396104145347,
      "loss": 6.0573,
      "mean_token_accuracy": 0.1557210013270378,
      "num_tokens": 20078380.0,
      "step": 10770
    },
    {
      "entropy": 6.095428800582885,
      "epoch": 0.9515188979159308,
      "grad_norm": 2.0,
      "learning_rate": 0.0004916311036086518,
      "loss": 5.9188,
      "mean_token_accuracy": 0.17131393253803254,
      "num_tokens": 20087849.0,
      "step": 10775
    },
    {
      "entropy": 6.14254937171936,
      "epoch": 0.9519604380077711,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004916225925590889,
      "loss": 5.8453,
      "mean_token_accuracy": 0.1693214550614357,
      "num_tokens": 20097253.0,
      "step": 10780
    },
    {
      "entropy": 6.199139165878296,
      "epoch": 0.9524019780996115,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004916140772660124,
      "loss": 6.0141,
      "mean_token_accuracy": 0.15481267273426055,
      "num_tokens": 20105931.0,
      "step": 10785
    },
    {
      "entropy": 6.219844341278076,
      "epoch": 0.9528435181914517,
      "grad_norm": 7.75,
      "learning_rate": 0.0004916055577295895,
      "loss": 5.9411,
      "mean_token_accuracy": 0.16423480063676835,
      "num_tokens": 20115546.0,
      "step": 10790
    },
    {
      "entropy": 6.245632743835449,
      "epoch": 0.9532850582832921,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004915970339499868,
      "loss": 5.9491,
      "mean_token_accuracy": 0.16496830433607101,
      "num_tokens": 20125009.0,
      "step": 10795
    },
    {
      "entropy": 6.0463440895080565,
      "epoch": 0.9537265983751325,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004915885059273714,
      "loss": 5.9165,
      "mean_token_accuracy": 0.16282762736082076,
      "num_tokens": 20134007.0,
      "step": 10800
    },
    {
      "entropy": 6.1316979885101315,
      "epoch": 0.9541681384669728,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004915799736619104,
      "loss": 5.9346,
      "mean_token_accuracy": 0.16109953820705414,
      "num_tokens": 20142617.0,
      "step": 10805
    },
    {
      "entropy": 6.1474696636199955,
      "epoch": 0.9546096785588132,
      "grad_norm": 3.03125,
      "learning_rate": 0.000491571437153771,
      "loss": 5.9399,
      "mean_token_accuracy": 0.16747409105300903,
      "num_tokens": 20151732.0,
      "step": 10810
    },
    {
      "entropy": 6.218894624710083,
      "epoch": 0.9550512186506535,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004915628964031202,
      "loss": 6.081,
      "mean_token_accuracy": 0.15158325284719468,
      "num_tokens": 20161733.0,
      "step": 10815
    },
    {
      "entropy": 6.227372789382935,
      "epoch": 0.9554927587424938,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004915543514101253,
      "loss": 5.9769,
      "mean_token_accuracy": 0.15670328438282013,
      "num_tokens": 20171022.0,
      "step": 10820
    },
    {
      "entropy": 6.172809410095215,
      "epoch": 0.9559342988343341,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004915458021749541,
      "loss": 6.0131,
      "mean_token_accuracy": 0.15471852123737334,
      "num_tokens": 20180809.0,
      "step": 10825
    },
    {
      "entropy": 6.175866937637329,
      "epoch": 0.9563758389261745,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004915372486977736,
      "loss": 5.9278,
      "mean_token_accuracy": 0.16139951944351197,
      "num_tokens": 20189614.0,
      "step": 10830
    },
    {
      "entropy": 6.228444051742554,
      "epoch": 0.9568173790180149,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004915286909787516,
      "loss": 6.0879,
      "mean_token_accuracy": 0.1520192638039589,
      "num_tokens": 20199136.0,
      "step": 10835
    },
    {
      "entropy": 6.1951131343841555,
      "epoch": 0.9572589191098552,
      "grad_norm": 4.5,
      "learning_rate": 0.0004915201290180556,
      "loss": 5.9582,
      "mean_token_accuracy": 0.16031280308961868,
      "num_tokens": 20208768.0,
      "step": 10840
    },
    {
      "entropy": 6.137907934188843,
      "epoch": 0.9577004592016956,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004915115628158535,
      "loss": 5.9426,
      "mean_token_accuracy": 0.15742172300815582,
      "num_tokens": 20218109.0,
      "step": 10845
    },
    {
      "entropy": 6.167422342300415,
      "epoch": 0.9581419992935358,
      "grad_norm": 3.984375,
      "learning_rate": 0.000491502992372313,
      "loss": 5.9709,
      "mean_token_accuracy": 0.159721340239048,
      "num_tokens": 20226768.0,
      "step": 10850
    },
    {
      "entropy": 6.164302730560303,
      "epoch": 0.9585835393853762,
      "grad_norm": 2.71875,
      "learning_rate": 0.000491494417687602,
      "loss": 5.931,
      "mean_token_accuracy": 0.16306218802928923,
      "num_tokens": 20236519.0,
      "step": 10855
    },
    {
      "entropy": 6.12832612991333,
      "epoch": 0.9590250794772165,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004914858387618886,
      "loss": 5.9336,
      "mean_token_accuracy": 0.16003130823373796,
      "num_tokens": 20245611.0,
      "step": 10860
    },
    {
      "entropy": 6.034099531173706,
      "epoch": 0.9594666195690569,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004914772555953406,
      "loss": 5.8619,
      "mean_token_accuracy": 0.16994949132204057,
      "num_tokens": 20255089.0,
      "step": 10865
    },
    {
      "entropy": 6.136065864562989,
      "epoch": 0.9599081596608973,
      "grad_norm": 2.75,
      "learning_rate": 0.0004914686681881264,
      "loss": 6.0136,
      "mean_token_accuracy": 0.1583478830754757,
      "num_tokens": 20264794.0,
      "step": 10870
    },
    {
      "entropy": 6.162253999710083,
      "epoch": 0.9603496997527375,
      "grad_norm": 3.78125,
      "learning_rate": 0.000491460076540414,
      "loss": 5.8834,
      "mean_token_accuracy": 0.17484040260314943,
      "num_tokens": 20274088.0,
      "step": 10875
    },
    {
      "entropy": 6.193916702270508,
      "epoch": 0.9607912398445779,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004914514806523718,
      "loss": 6.0727,
      "mean_token_accuracy": 0.15178250670433044,
      "num_tokens": 20283837.0,
      "step": 10880
    },
    {
      "entropy": 6.193194770812989,
      "epoch": 0.9612327799364182,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004914428805241683,
      "loss": 5.9377,
      "mean_token_accuracy": 0.15788443386554718,
      "num_tokens": 20293287.0,
      "step": 10885
    },
    {
      "entropy": 6.091219902038574,
      "epoch": 0.9616743200282586,
      "grad_norm": 2.125,
      "learning_rate": 0.0004914342761559718,
      "loss": 5.9446,
      "mean_token_accuracy": 0.15868934839963914,
      "num_tokens": 20302262.0,
      "step": 10890
    },
    {
      "entropy": 6.24610071182251,
      "epoch": 0.9621158601200989,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004914256675479509,
      "loss": 6.0211,
      "mean_token_accuracy": 0.15657830536365508,
      "num_tokens": 20312475.0,
      "step": 10895
    },
    {
      "entropy": 6.239964199066162,
      "epoch": 0.9625574002119393,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004914170547002742,
      "loss": 6.0375,
      "mean_token_accuracy": 0.15694003701210021,
      "num_tokens": 20322243.0,
      "step": 10900
    },
    {
      "entropy": 6.197062206268311,
      "epoch": 0.9629989403037796,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004914084376131106,
      "loss": 5.9302,
      "mean_token_accuracy": 0.15753385573625564,
      "num_tokens": 20331299.0,
      "step": 10905
    },
    {
      "entropy": 6.076598596572876,
      "epoch": 0.9634404803956199,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004913998162866288,
      "loss": 5.8866,
      "mean_token_accuracy": 0.16386901140213012,
      "num_tokens": 20340800.0,
      "step": 10910
    },
    {
      "entropy": 6.118331813812256,
      "epoch": 0.9638820204874603,
      "grad_norm": 5.0,
      "learning_rate": 0.0004913911907209976,
      "loss": 5.9563,
      "mean_token_accuracy": 0.16420629024505615,
      "num_tokens": 20350292.0,
      "step": 10915
    },
    {
      "entropy": 6.187256240844727,
      "epoch": 0.9643235605793006,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004913825609163861,
      "loss": 6.0399,
      "mean_token_accuracy": 0.15299595892429352,
      "num_tokens": 20360109.0,
      "step": 10920
    },
    {
      "entropy": 6.1300853252410885,
      "epoch": 0.964765100671141,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004913739268729633,
      "loss": 5.8661,
      "mean_token_accuracy": 0.16548800468444824,
      "num_tokens": 20368917.0,
      "step": 10925
    },
    {
      "entropy": 6.2461127758026125,
      "epoch": 0.9652066407629812,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004913652885908983,
      "loss": 6.0697,
      "mean_token_accuracy": 0.1535951465368271,
      "num_tokens": 20378580.0,
      "step": 10930
    },
    {
      "entropy": 6.219760227203369,
      "epoch": 0.9656481808548216,
      "grad_norm": 2.875,
      "learning_rate": 0.0004913566460703603,
      "loss": 5.9033,
      "mean_token_accuracy": 0.1624210759997368,
      "num_tokens": 20387497.0,
      "step": 10935
    },
    {
      "entropy": 6.162945938110352,
      "epoch": 0.966089720946662,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004913479993115187,
      "loss": 6.0476,
      "mean_token_accuracy": 0.15562895089387893,
      "num_tokens": 20396586.0,
      "step": 10940
    },
    {
      "entropy": 6.174739599227905,
      "epoch": 0.9665312610385023,
      "grad_norm": 9.5625,
      "learning_rate": 0.0004913393483145428,
      "loss": 6.1271,
      "mean_token_accuracy": 0.14758048504590987,
      "num_tokens": 20406183.0,
      "step": 10945
    },
    {
      "entropy": 6.298451709747314,
      "epoch": 0.9669728011303427,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004913306930796022,
      "loss": 6.0835,
      "mean_token_accuracy": 0.14926106110215187,
      "num_tokens": 20415716.0,
      "step": 10950
    },
    {
      "entropy": 6.254632377624512,
      "epoch": 0.967414341222183,
      "grad_norm": 2.375,
      "learning_rate": 0.0004913220336068662,
      "loss": 5.9834,
      "mean_token_accuracy": 0.15914528518915178,
      "num_tokens": 20424715.0,
      "step": 10955
    },
    {
      "entropy": 6.145887804031372,
      "epoch": 0.9678558813140233,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004913133698965048,
      "loss": 5.9686,
      "mean_token_accuracy": 0.16440486609935762,
      "num_tokens": 20434686.0,
      "step": 10960
    },
    {
      "entropy": 6.157244634628296,
      "epoch": 0.9682974214058636,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004913047019486874,
      "loss": 5.974,
      "mean_token_accuracy": 0.16015694290399551,
      "num_tokens": 20443619.0,
      "step": 10965
    },
    {
      "entropy": 6.139456415176392,
      "epoch": 0.968738961497704,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004912960297635839,
      "loss": 5.9023,
      "mean_token_accuracy": 0.16531845480203627,
      "num_tokens": 20453540.0,
      "step": 10970
    },
    {
      "entropy": 6.184057998657226,
      "epoch": 0.9691805015895444,
      "grad_norm": 4.125,
      "learning_rate": 0.0004912873533413641,
      "loss": 6.0138,
      "mean_token_accuracy": 0.1574922278523445,
      "num_tokens": 20463395.0,
      "step": 10975
    },
    {
      "entropy": 6.218158340454101,
      "epoch": 0.9696220416813847,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004912786726821982,
      "loss": 5.9655,
      "mean_token_accuracy": 0.15509623140096665,
      "num_tokens": 20472302.0,
      "step": 10980
    },
    {
      "entropy": 6.126210069656372,
      "epoch": 0.970063581773225,
      "grad_norm": 2.609375,
      "learning_rate": 0.000491269987786256,
      "loss": 5.9908,
      "mean_token_accuracy": 0.1569495067000389,
      "num_tokens": 20482077.0,
      "step": 10985
    },
    {
      "entropy": 6.19876651763916,
      "epoch": 0.9705051218650653,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004912612986537078,
      "loss": 5.9797,
      "mean_token_accuracy": 0.15929123461246492,
      "num_tokens": 20492177.0,
      "step": 10990
    },
    {
      "entropy": 6.199370384216309,
      "epoch": 0.9709466619569057,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004912526052847238,
      "loss": 6.0219,
      "mean_token_accuracy": 0.1591154769062996,
      "num_tokens": 20502063.0,
      "step": 10995
    },
    {
      "entropy": 6.135434913635254,
      "epoch": 0.971388202048746,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004912439076794742,
      "loss": 5.9438,
      "mean_token_accuracy": 0.15248878076672553,
      "num_tokens": 20510631.0,
      "step": 11000
    },
    {
      "entropy": 6.1080974578857425,
      "epoch": 0.9718297421405864,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004912352058381293,
      "loss": 5.9895,
      "mean_token_accuracy": 0.15816915333271026,
      "num_tokens": 20519978.0,
      "step": 11005
    },
    {
      "entropy": 6.161657285690308,
      "epoch": 0.9722712822324268,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004912264997608599,
      "loss": 5.9769,
      "mean_token_accuracy": 0.15872621834278106,
      "num_tokens": 20529346.0,
      "step": 11010
    },
    {
      "entropy": 6.2405179977417,
      "epoch": 0.972712822324267,
      "grad_norm": 5.375,
      "learning_rate": 0.0004912177894478364,
      "loss": 5.9468,
      "mean_token_accuracy": 0.1512127101421356,
      "num_tokens": 20537619.0,
      "step": 11015
    },
    {
      "entropy": 6.218065452575684,
      "epoch": 0.9731543624161074,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004912090748992291,
      "loss": 5.9873,
      "mean_token_accuracy": 0.1572415143251419,
      "num_tokens": 20547011.0,
      "step": 11020
    },
    {
      "entropy": 6.196943759918213,
      "epoch": 0.9735959025079477,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004912003561152092,
      "loss": 5.931,
      "mean_token_accuracy": 0.15849291533231735,
      "num_tokens": 20556425.0,
      "step": 11025
    },
    {
      "entropy": 6.156350564956665,
      "epoch": 0.9740374425997881,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004911916330959472,
      "loss": 5.9256,
      "mean_token_accuracy": 0.16159031242132188,
      "num_tokens": 20565330.0,
      "step": 11030
    },
    {
      "entropy": 6.072202873229981,
      "epoch": 0.9744789826916284,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004911829058416141,
      "loss": 6.0361,
      "mean_token_accuracy": 0.16514350175857545,
      "num_tokens": 20575320.0,
      "step": 11035
    },
    {
      "entropy": 6.181253576278687,
      "epoch": 0.9749205227834687,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004911741743523808,
      "loss": 5.9265,
      "mean_token_accuracy": 0.16285687908530236,
      "num_tokens": 20585105.0,
      "step": 11040
    },
    {
      "entropy": 6.172753810882568,
      "epoch": 0.9753620628753091,
      "grad_norm": 2.5,
      "learning_rate": 0.0004911654386284184,
      "loss": 5.9814,
      "mean_token_accuracy": 0.15713930875062943,
      "num_tokens": 20593889.0,
      "step": 11045
    },
    {
      "entropy": 6.154466199874878,
      "epoch": 0.9758036029671494,
      "grad_norm": 2.734375,
      "learning_rate": 0.000491156698669898,
      "loss": 6.0155,
      "mean_token_accuracy": 0.1601693570613861,
      "num_tokens": 20602322.0,
      "step": 11050
    },
    {
      "entropy": 6.21231894493103,
      "epoch": 0.9762451430589898,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004911479544769908,
      "loss": 6.1091,
      "mean_token_accuracy": 0.14768424034118652,
      "num_tokens": 20612077.0,
      "step": 11055
    },
    {
      "entropy": 6.289440536499024,
      "epoch": 0.9766866831508301,
      "grad_norm": 3.171875,
      "learning_rate": 0.000491139206049868,
      "loss": 6.0462,
      "mean_token_accuracy": 0.15645182877779007,
      "num_tokens": 20620683.0,
      "step": 11060
    },
    {
      "entropy": 6.279990577697754,
      "epoch": 0.9771282232426705,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004911304533887011,
      "loss": 6.0929,
      "mean_token_accuracy": 0.1489789068698883,
      "num_tokens": 20631124.0,
      "step": 11065
    },
    {
      "entropy": 6.194825553894043,
      "epoch": 0.9775697633345107,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004911216964936616,
      "loss": 5.9338,
      "mean_token_accuracy": 0.15422664284706117,
      "num_tokens": 20640515.0,
      "step": 11070
    },
    {
      "entropy": 6.151896047592163,
      "epoch": 0.9780113034263511,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004911129353649211,
      "loss": 5.9407,
      "mean_token_accuracy": 0.16204164773225785,
      "num_tokens": 20649819.0,
      "step": 11075
    },
    {
      "entropy": 6.195270729064942,
      "epoch": 0.9784528435181915,
      "grad_norm": 2.34375,
      "learning_rate": 0.000491104170002651,
      "loss": 5.9594,
      "mean_token_accuracy": 0.15603119134902954,
      "num_tokens": 20657831.0,
      "step": 11080
    },
    {
      "entropy": 6.219133758544922,
      "epoch": 0.9788943836100318,
      "grad_norm": 2.484375,
      "learning_rate": 0.000491095400407023,
      "loss": 6.0647,
      "mean_token_accuracy": 0.1545814350247383,
      "num_tokens": 20666807.0,
      "step": 11085
    },
    {
      "entropy": 6.188013410568237,
      "epoch": 0.9793359237018722,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004910866265782091,
      "loss": 5.9122,
      "mean_token_accuracy": 0.1632992908358574,
      "num_tokens": 20674859.0,
      "step": 11090
    },
    {
      "entropy": 6.225031471252441,
      "epoch": 0.9797774637937124,
      "grad_norm": 2.875,
      "learning_rate": 0.0004910778485163811,
      "loss": 5.9995,
      "mean_token_accuracy": 0.15673470944166185,
      "num_tokens": 20684060.0,
      "step": 11095
    },
    {
      "entropy": 6.113849401473999,
      "epoch": 0.9802190038855528,
      "grad_norm": 3.828125,
      "learning_rate": 0.000491069066221711,
      "loss": 5.9228,
      "mean_token_accuracy": 0.16676190495491028,
      "num_tokens": 20693446.0,
      "step": 11100
    },
    {
      "entropy": 6.178444194793701,
      "epoch": 0.9806605439773931,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004910602796943707,
      "loss": 6.0555,
      "mean_token_accuracy": 0.15749317556619644,
      "num_tokens": 20702635.0,
      "step": 11105
    },
    {
      "entropy": 6.259019804000855,
      "epoch": 0.9811020840692335,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004910514889345323,
      "loss": 6.0168,
      "mean_token_accuracy": 0.1519618071615696,
      "num_tokens": 20712256.0,
      "step": 11110
    },
    {
      "entropy": 6.156629848480224,
      "epoch": 0.9815436241610739,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004910426939423682,
      "loss": 5.9555,
      "mean_token_accuracy": 0.1574217200279236,
      "num_tokens": 20720175.0,
      "step": 11115
    },
    {
      "entropy": 6.180657577514649,
      "epoch": 0.9819851642529142,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004910338947180506,
      "loss": 5.9952,
      "mean_token_accuracy": 0.16044948101043702,
      "num_tokens": 20729849.0,
      "step": 11120
    },
    {
      "entropy": 6.180977296829224,
      "epoch": 0.9824267043447545,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004910250912617519,
      "loss": 5.9287,
      "mean_token_accuracy": 0.16036087423563003,
      "num_tokens": 20739129.0,
      "step": 11125
    },
    {
      "entropy": 6.130973815917969,
      "epoch": 0.9828682444365948,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004910162835736445,
      "loss": 5.8996,
      "mean_token_accuracy": 0.16647030115127565,
      "num_tokens": 20748632.0,
      "step": 11130
    },
    {
      "entropy": 6.19669394493103,
      "epoch": 0.9833097845284352,
      "grad_norm": 3.84375,
      "learning_rate": 0.000491007471653901,
      "loss": 5.9654,
      "mean_token_accuracy": 0.1556306153535843,
      "num_tokens": 20758535.0,
      "step": 11135
    },
    {
      "entropy": 6.193940210342407,
      "epoch": 0.9837513246202755,
      "grad_norm": 2.0,
      "learning_rate": 0.0004909986555026939,
      "loss": 5.9795,
      "mean_token_accuracy": 0.15689342916011811,
      "num_tokens": 20768982.0,
      "step": 11140
    },
    {
      "entropy": 6.205114030838013,
      "epoch": 0.9841928647121159,
      "grad_norm": 2.421875,
      "learning_rate": 0.000490989835120196,
      "loss": 6.0411,
      "mean_token_accuracy": 0.1563404068350792,
      "num_tokens": 20778936.0,
      "step": 11145
    },
    {
      "entropy": 6.167696332931518,
      "epoch": 0.9846344048039563,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004909810105065801,
      "loss": 6.0345,
      "mean_token_accuracy": 0.15165605247020722,
      "num_tokens": 20788618.0,
      "step": 11150
    },
    {
      "entropy": 6.2161524295806885,
      "epoch": 0.9850759448957965,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004909721816620191,
      "loss": 6.0159,
      "mean_token_accuracy": 0.1615686759352684,
      "num_tokens": 20797721.0,
      "step": 11155
    },
    {
      "entropy": 6.186040878295898,
      "epoch": 0.9855174849876369,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004909633485866857,
      "loss": 5.9761,
      "mean_token_accuracy": 0.16043892204761506,
      "num_tokens": 20807305.0,
      "step": 11160
    },
    {
      "entropy": 6.07858567237854,
      "epoch": 0.9859590250794772,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004909545112807533,
      "loss": 5.9331,
      "mean_token_accuracy": 0.1650419846177101,
      "num_tokens": 20816984.0,
      "step": 11165
    },
    {
      "entropy": 6.1921858310699465,
      "epoch": 0.9864005651713176,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004909456697443948,
      "loss": 6.0483,
      "mean_token_accuracy": 0.152926018089056,
      "num_tokens": 20825738.0,
      "step": 11170
    },
    {
      "entropy": 6.22014422416687,
      "epoch": 0.9868421052631579,
      "grad_norm": 4.125,
      "learning_rate": 0.0004909368239777835,
      "loss": 5.9935,
      "mean_token_accuracy": 0.1584202915430069,
      "num_tokens": 20834135.0,
      "step": 11175
    },
    {
      "entropy": 6.304402923583984,
      "epoch": 0.9872836453549982,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004909279739810926,
      "loss": 6.0888,
      "mean_token_accuracy": 0.15151295438408852,
      "num_tokens": 20843689.0,
      "step": 11180
    },
    {
      "entropy": 6.219657325744629,
      "epoch": 0.9877251854468386,
      "grad_norm": 2.75,
      "learning_rate": 0.0004909191197544954,
      "loss": 5.9551,
      "mean_token_accuracy": 0.1585159108042717,
      "num_tokens": 20851724.0,
      "step": 11185
    },
    {
      "entropy": 6.251550006866455,
      "epoch": 0.9881667255386789,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004909102612981655,
      "loss": 6.1485,
      "mean_token_accuracy": 0.14939290881156922,
      "num_tokens": 20862325.0,
      "step": 11190
    },
    {
      "entropy": 6.225513887405396,
      "epoch": 0.9886082656305193,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004909013986122763,
      "loss": 5.9396,
      "mean_token_accuracy": 0.16449217200279237,
      "num_tokens": 20871046.0,
      "step": 11195
    },
    {
      "entropy": 6.2365467071533205,
      "epoch": 0.9890498057223596,
      "grad_norm": 3.375,
      "learning_rate": 0.0004908925316970016,
      "loss": 5.9695,
      "mean_token_accuracy": 0.15662687718868257,
      "num_tokens": 20881605.0,
      "step": 11200
    },
    {
      "entropy": 6.089207506179809,
      "epoch": 0.9894913458142,
      "grad_norm": 4.21875,
      "learning_rate": 0.000490883660552515,
      "loss": 5.9293,
      "mean_token_accuracy": 0.15434980243444443,
      "num_tokens": 20891170.0,
      "step": 11205
    },
    {
      "entropy": 6.140697956085205,
      "epoch": 0.9899328859060402,
      "grad_norm": 2.75,
      "learning_rate": 0.0004908747851789902,
      "loss": 5.9338,
      "mean_token_accuracy": 0.14825039803981782,
      "num_tokens": 20900307.0,
      "step": 11210
    },
    {
      "entropy": 6.206425142288208,
      "epoch": 0.9903744259978806,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004908659055766011,
      "loss": 6.0787,
      "mean_token_accuracy": 0.1500147297978401,
      "num_tokens": 20910623.0,
      "step": 11215
    },
    {
      "entropy": 6.315029430389404,
      "epoch": 0.990815966089721,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004908570217455218,
      "loss": 5.9929,
      "mean_token_accuracy": 0.16046351343393325,
      "num_tokens": 20919744.0,
      "step": 11220
    },
    {
      "entropy": 6.202160263061524,
      "epoch": 0.9912575061815613,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004908481336859262,
      "loss": 6.0605,
      "mean_token_accuracy": 0.15378574877977372,
      "num_tokens": 20928701.0,
      "step": 11225
    },
    {
      "entropy": 6.174495458602905,
      "epoch": 0.9916990462734017,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004908392413979884,
      "loss": 6.0277,
      "mean_token_accuracy": 0.14975389391183852,
      "num_tokens": 20937786.0,
      "step": 11230
    },
    {
      "entropy": 6.186076211929321,
      "epoch": 0.9921405863652419,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004908303448818825,
      "loss": 5.9562,
      "mean_token_accuracy": 0.16092230379581451,
      "num_tokens": 20947214.0,
      "step": 11235
    },
    {
      "entropy": 6.109399318695068,
      "epoch": 0.9925821264570823,
      "grad_norm": 2.765625,
      "learning_rate": 0.000490821444137783,
      "loss": 6.0237,
      "mean_token_accuracy": 0.1595834255218506,
      "num_tokens": 20956041.0,
      "step": 11240
    },
    {
      "entropy": 6.1674988746643065,
      "epoch": 0.9930236665489226,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004908125391658641,
      "loss": 6.0311,
      "mean_token_accuracy": 0.15522620379924773,
      "num_tokens": 20966162.0,
      "step": 11245
    },
    {
      "entropy": 6.203303289413452,
      "epoch": 0.993465206640763,
      "grad_norm": 3.125,
      "learning_rate": 0.0004908036299663003,
      "loss": 6.0081,
      "mean_token_accuracy": 0.1580747187137604,
      "num_tokens": 20976025.0,
      "step": 11250
    },
    {
      "entropy": 6.190470504760742,
      "epoch": 0.9939067467326034,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004907947165392662,
      "loss": 6.0132,
      "mean_token_accuracy": 0.1540180116891861,
      "num_tokens": 20985606.0,
      "step": 11255
    },
    {
      "entropy": 6.189827251434326,
      "epoch": 0.9943482868244437,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004907857988849363,
      "loss": 6.0122,
      "mean_token_accuracy": 0.1569234013557434,
      "num_tokens": 20994308.0,
      "step": 11260
    },
    {
      "entropy": 6.114636325836182,
      "epoch": 0.994789826916284,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004907768770034853,
      "loss": 5.9451,
      "mean_token_accuracy": 0.16427878886461258,
      "num_tokens": 21003783.0,
      "step": 11265
    },
    {
      "entropy": 6.157221126556396,
      "epoch": 0.9952313670081243,
      "grad_norm": 2.5,
      "learning_rate": 0.000490767950895088,
      "loss": 5.9323,
      "mean_token_accuracy": 0.15766710937023162,
      "num_tokens": 21012870.0,
      "step": 11270
    },
    {
      "entropy": 6.245476579666137,
      "epoch": 0.9956729070999647,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004907590205599192,
      "loss": 6.0218,
      "mean_token_accuracy": 0.1500089094042778,
      "num_tokens": 21021987.0,
      "step": 11275
    },
    {
      "entropy": 6.177654600143432,
      "epoch": 0.996114447191805,
      "grad_norm": 5.6875,
      "learning_rate": 0.000490750085998154,
      "loss": 6.01,
      "mean_token_accuracy": 0.1551889404654503,
      "num_tokens": 21030977.0,
      "step": 11280
    },
    {
      "entropy": 6.102250242233277,
      "epoch": 0.9965559872836454,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004907411472099672,
      "loss": 5.974,
      "mean_token_accuracy": 0.15588120222091675,
      "num_tokens": 21039259.0,
      "step": 11285
    },
    {
      "entropy": 6.218702745437622,
      "epoch": 0.9969975273754857,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004907322041955341,
      "loss": 6.0466,
      "mean_token_accuracy": 0.1561511978507042,
      "num_tokens": 21049321.0,
      "step": 11290
    },
    {
      "entropy": 6.286624622344971,
      "epoch": 0.997439067467326,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004907232569550298,
      "loss": 6.0963,
      "mean_token_accuracy": 0.15165919065475464,
      "num_tokens": 21059469.0,
      "step": 11295
    },
    {
      "entropy": 6.217846250534057,
      "epoch": 0.9978806075591664,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004907143054886297,
      "loss": 6.0264,
      "mean_token_accuracy": 0.15524942427873611,
      "num_tokens": 21068694.0,
      "step": 11300
    },
    {
      "entropy": 6.136050844192505,
      "epoch": 0.9983221476510067,
      "grad_norm": 2.296875,
      "learning_rate": 0.000490705349796509,
      "loss": 5.9753,
      "mean_token_accuracy": 0.15849747359752656,
      "num_tokens": 21077232.0,
      "step": 11305
    },
    {
      "entropy": 6.226991510391235,
      "epoch": 0.9987636877428471,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004906963898788431,
      "loss": 5.9824,
      "mean_token_accuracy": 0.154877470433712,
      "num_tokens": 21086517.0,
      "step": 11310
    },
    {
      "entropy": 6.236574411392212,
      "epoch": 0.9992052278346873,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004906874257358075,
      "loss": 6.0508,
      "mean_token_accuracy": 0.16054894030094147,
      "num_tokens": 21096969.0,
      "step": 11315
    },
    {
      "entropy": 6.190933656692505,
      "epoch": 0.9996467679265277,
      "grad_norm": 1.8671875,
      "learning_rate": 0.000490678457367578,
      "loss": 5.9732,
      "mean_token_accuracy": 0.15344223529100418,
      "num_tokens": 21106554.0,
      "step": 11320
    },
    {
      "entropy": 6.206276178359985,
      "epoch": 1.000088308018368,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004906694847743302,
      "loss": 5.9893,
      "mean_token_accuracy": 0.15358800143003465,
      "num_tokens": 21114829.0,
      "step": 11325
    },
    {
      "entropy": 6.214368391036987,
      "epoch": 1.0005298481102085,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004906605079562399,
      "loss": 5.9471,
      "mean_token_accuracy": 0.16381950825452804,
      "num_tokens": 21123568.0,
      "step": 11330
    },
    {
      "entropy": 6.154477500915528,
      "epoch": 1.0009713882020488,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004906515269134827,
      "loss": 5.9903,
      "mean_token_accuracy": 0.15432658270001412,
      "num_tokens": 21132811.0,
      "step": 11335
    },
    {
      "entropy": 6.174828577041626,
      "epoch": 1.001412928293889,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004906425416462349,
      "loss": 6.0133,
      "mean_token_accuracy": 0.15606007725000381,
      "num_tokens": 21141549.0,
      "step": 11340
    },
    {
      "entropy": 6.1920520782470705,
      "epoch": 1.0018544683857293,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004906335521546723,
      "loss": 5.9161,
      "mean_token_accuracy": 0.1637442409992218,
      "num_tokens": 21150603.0,
      "step": 11345
    },
    {
      "entropy": 6.032572221755982,
      "epoch": 1.0022960084775698,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004906245584389712,
      "loss": 5.7609,
      "mean_token_accuracy": 0.17506401091814042,
      "num_tokens": 21159842.0,
      "step": 11350
    },
    {
      "entropy": 6.12627854347229,
      "epoch": 1.00273754856941,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004906155604993075,
      "loss": 5.8985,
      "mean_token_accuracy": 0.1639697715640068,
      "num_tokens": 21169139.0,
      "step": 11355
    },
    {
      "entropy": 6.089358377456665,
      "epoch": 1.0031790886612504,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004906065583358576,
      "loss": 5.8647,
      "mean_token_accuracy": 0.16264504343271255,
      "num_tokens": 21178059.0,
      "step": 11360
    },
    {
      "entropy": 6.115440225601196,
      "epoch": 1.0036206287530909,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004905975519487978,
      "loss": 5.8943,
      "mean_token_accuracy": 0.1575951173901558,
      "num_tokens": 21187966.0,
      "step": 11365
    },
    {
      "entropy": 6.09847412109375,
      "epoch": 1.0040621688449312,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004905885413383046,
      "loss": 5.9365,
      "mean_token_accuracy": 0.15951112359762193,
      "num_tokens": 21197460.0,
      "step": 11370
    },
    {
      "entropy": 6.166648912429809,
      "epoch": 1.0045037089367714,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004905795265045546,
      "loss": 5.9483,
      "mean_token_accuracy": 0.1568487100303173,
      "num_tokens": 21206977.0,
      "step": 11375
    },
    {
      "entropy": 6.168020343780517,
      "epoch": 1.0049452490286117,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004905705074477242,
      "loss": 5.9052,
      "mean_token_accuracy": 0.15654271543025972,
      "num_tokens": 21216046.0,
      "step": 11380
    },
    {
      "entropy": 6.181176233291626,
      "epoch": 1.0053867891204522,
      "grad_norm": 2.5,
      "learning_rate": 0.0004905614841679901,
      "loss": 5.9385,
      "mean_token_accuracy": 0.15977095663547516,
      "num_tokens": 21225612.0,
      "step": 11385
    },
    {
      "entropy": 6.1381330490112305,
      "epoch": 1.0058283292122925,
      "grad_norm": 2.75,
      "learning_rate": 0.0004905524566655292,
      "loss": 5.8602,
      "mean_token_accuracy": 0.16055045053362846,
      "num_tokens": 21233641.0,
      "step": 11390
    },
    {
      "entropy": 6.090469884872436,
      "epoch": 1.0062698693041328,
      "grad_norm": 5.96875,
      "learning_rate": 0.0004905434249405183,
      "loss": 5.9051,
      "mean_token_accuracy": 0.15970632433891296,
      "num_tokens": 21243444.0,
      "step": 11395
    },
    {
      "entropy": 6.095047521591186,
      "epoch": 1.0067114093959733,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004905343889931342,
      "loss": 5.8751,
      "mean_token_accuracy": 0.16488119661808015,
      "num_tokens": 21253473.0,
      "step": 11400
    },
    {
      "entropy": 6.152161026000977,
      "epoch": 1.0071529494878135,
      "grad_norm": 2.59375,
      "learning_rate": 0.000490525348823554,
      "loss": 5.844,
      "mean_token_accuracy": 0.17127221822738647,
      "num_tokens": 21262049.0,
      "step": 11405
    },
    {
      "entropy": 6.125896072387695,
      "epoch": 1.0075944895796538,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004905163044319549,
      "loss": 5.8781,
      "mean_token_accuracy": 0.15909457951784134,
      "num_tokens": 21270909.0,
      "step": 11410
    },
    {
      "entropy": 6.091321897506714,
      "epoch": 1.008036029671494,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004905072558185139,
      "loss": 5.9928,
      "mean_token_accuracy": 0.1581793926656246,
      "num_tokens": 21280503.0,
      "step": 11415
    },
    {
      "entropy": 6.166311883926392,
      "epoch": 1.0084775697633346,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004904982029834083,
      "loss": 5.951,
      "mean_token_accuracy": 0.15546180456876754,
      "num_tokens": 21289957.0,
      "step": 11420
    },
    {
      "entropy": 6.182868528366089,
      "epoch": 1.0089191098551749,
      "grad_norm": 8.375,
      "learning_rate": 0.0004904891459268154,
      "loss": 5.9537,
      "mean_token_accuracy": 0.15713533461093904,
      "num_tokens": 21298799.0,
      "step": 11425
    },
    {
      "entropy": 6.142367506027222,
      "epoch": 1.0093606499470151,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004904800846489128,
      "loss": 5.9532,
      "mean_token_accuracy": 0.156003238260746,
      "num_tokens": 21308251.0,
      "step": 11430
    },
    {
      "entropy": 6.201836109161377,
      "epoch": 1.0098021900388556,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004904710191498779,
      "loss": 5.9665,
      "mean_token_accuracy": 0.1572624810039997,
      "num_tokens": 21317705.0,
      "step": 11435
    },
    {
      "entropy": 6.180490493774414,
      "epoch": 1.010243730130696,
      "grad_norm": 3.375,
      "learning_rate": 0.0004904619494298882,
      "loss": 5.929,
      "mean_token_accuracy": 0.15343269407749177,
      "num_tokens": 21326867.0,
      "step": 11440
    },
    {
      "entropy": 6.190150737762451,
      "epoch": 1.0106852702225362,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004904528754891215,
      "loss": 5.9551,
      "mean_token_accuracy": 0.16153139024972915,
      "num_tokens": 21336054.0,
      "step": 11445
    },
    {
      "entropy": 6.165220642089844,
      "epoch": 1.0111268103143765,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004904437973277555,
      "loss": 5.8814,
      "mean_token_accuracy": 0.16096238791942596,
      "num_tokens": 21344457.0,
      "step": 11450
    },
    {
      "entropy": 6.2367006778717045,
      "epoch": 1.011568350406217,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004904347149459681,
      "loss": 5.9824,
      "mean_token_accuracy": 0.1612090528011322,
      "num_tokens": 21353639.0,
      "step": 11455
    },
    {
      "entropy": 6.103877258300781,
      "epoch": 1.0120098904980572,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004904256283439371,
      "loss": 5.7885,
      "mean_token_accuracy": 0.16498116105794908,
      "num_tokens": 21362005.0,
      "step": 11460
    },
    {
      "entropy": 6.1524499416351315,
      "epoch": 1.0124514305898975,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004904165375218405,
      "loss": 5.9533,
      "mean_token_accuracy": 0.15746854841709138,
      "num_tokens": 21371917.0,
      "step": 11465
    },
    {
      "entropy": 6.16218113899231,
      "epoch": 1.012892970681738,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004904074424798565,
      "loss": 5.9735,
      "mean_token_accuracy": 0.15811279118061067,
      "num_tokens": 21381500.0,
      "step": 11470
    },
    {
      "entropy": 6.175776100158691,
      "epoch": 1.0133345107735783,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004903983432181632,
      "loss": 5.8288,
      "mean_token_accuracy": 0.16764068305492402,
      "num_tokens": 21391280.0,
      "step": 11475
    },
    {
      "entropy": 6.2048516273498535,
      "epoch": 1.0137760508654186,
      "grad_norm": 2.75,
      "learning_rate": 0.0004903892397369388,
      "loss": 5.9268,
      "mean_token_accuracy": 0.15447692275047303,
      "num_tokens": 21400371.0,
      "step": 11480
    },
    {
      "entropy": 6.035159730911255,
      "epoch": 1.0142175909572588,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004903801320363618,
      "loss": 5.9501,
      "mean_token_accuracy": 0.16072324067354202,
      "num_tokens": 21410493.0,
      "step": 11485
    },
    {
      "entropy": 6.150381088256836,
      "epoch": 1.0146591310490993,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004903710201166105,
      "loss": 6.0369,
      "mean_token_accuracy": 0.14643384665250778,
      "num_tokens": 21420516.0,
      "step": 11490
    },
    {
      "entropy": 6.187597990036011,
      "epoch": 1.0151006711409396,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004903619039778634,
      "loss": 5.9575,
      "mean_token_accuracy": 0.15411232709884642,
      "num_tokens": 21429614.0,
      "step": 11495
    },
    {
      "entropy": 6.219703292846679,
      "epoch": 1.0155422112327799,
      "grad_norm": 4.09375,
      "learning_rate": 0.000490352783620299,
      "loss": 5.9102,
      "mean_token_accuracy": 0.1611112102866173,
      "num_tokens": 21438595.0,
      "step": 11500
    },
    {
      "entropy": 6.2136434555053714,
      "epoch": 1.0159837513246204,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004903436590440961,
      "loss": 5.9671,
      "mean_token_accuracy": 0.15861888378858566,
      "num_tokens": 21447708.0,
      "step": 11505
    },
    {
      "entropy": 6.180592584609985,
      "epoch": 1.0164252914164607,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004903345302494334,
      "loss": 5.9205,
      "mean_token_accuracy": 0.1601620078086853,
      "num_tokens": 21457210.0,
      "step": 11510
    },
    {
      "entropy": 6.191863584518432,
      "epoch": 1.016866831508301,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004903253972364897,
      "loss": 5.9389,
      "mean_token_accuracy": 0.15268738567829132,
      "num_tokens": 21466247.0,
      "step": 11515
    },
    {
      "entropy": 6.164012002944946,
      "epoch": 1.0173083716001412,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004903162600054439,
      "loss": 5.8735,
      "mean_token_accuracy": 0.1608436480164528,
      "num_tokens": 21475634.0,
      "step": 11520
    },
    {
      "entropy": 6.195175266265869,
      "epoch": 1.0177499116919817,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004903071185564751,
      "loss": 6.0045,
      "mean_token_accuracy": 0.15861720144748687,
      "num_tokens": 21485523.0,
      "step": 11525
    },
    {
      "entropy": 6.1413788318634035,
      "epoch": 1.018191451783822,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004902979728897623,
      "loss": 5.9403,
      "mean_token_accuracy": 0.1597276136279106,
      "num_tokens": 21494909.0,
      "step": 11530
    },
    {
      "entropy": 6.155840253829956,
      "epoch": 1.0186329918756623,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004902888230054846,
      "loss": 5.9702,
      "mean_token_accuracy": 0.15895597785711288,
      "num_tokens": 21504413.0,
      "step": 11535
    },
    {
      "entropy": 6.122620058059693,
      "epoch": 1.0190745319675028,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004902796689038213,
      "loss": 5.8599,
      "mean_token_accuracy": 0.16754451021552086,
      "num_tokens": 21514068.0,
      "step": 11540
    },
    {
      "entropy": 6.154145002365112,
      "epoch": 1.019516072059343,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004902705105849517,
      "loss": 5.8788,
      "mean_token_accuracy": 0.16204680800437926,
      "num_tokens": 21523536.0,
      "step": 11545
    },
    {
      "entropy": 6.113324308395386,
      "epoch": 1.0199576121511833,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004902613480490553,
      "loss": 5.965,
      "mean_token_accuracy": 0.15359012633562089,
      "num_tokens": 21532554.0,
      "step": 11550
    },
    {
      "entropy": 6.049493074417114,
      "epoch": 1.0203991522430236,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004902521812963116,
      "loss": 5.8511,
      "mean_token_accuracy": 0.16149965971708297,
      "num_tokens": 21542508.0,
      "step": 11555
    },
    {
      "entropy": 6.200096559524536,
      "epoch": 1.020840692334864,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004902430103268998,
      "loss": 5.9368,
      "mean_token_accuracy": 0.14979047179222107,
      "num_tokens": 21551954.0,
      "step": 11560
    },
    {
      "entropy": 6.244258451461792,
      "epoch": 1.0212822324267044,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004902338351410002,
      "loss": 5.8989,
      "mean_token_accuracy": 0.16174011081457138,
      "num_tokens": 21561676.0,
      "step": 11565
    },
    {
      "entropy": 6.194683694839478,
      "epoch": 1.0217237725185446,
      "grad_norm": 3.4375,
      "learning_rate": 0.000490224655738792,
      "loss": 5.9995,
      "mean_token_accuracy": 0.1473441541194916,
      "num_tokens": 21571916.0,
      "step": 11570
    },
    {
      "entropy": 6.119585227966309,
      "epoch": 1.0221653126103851,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004902154721204552,
      "loss": 5.9213,
      "mean_token_accuracy": 0.16167986690998076,
      "num_tokens": 21581892.0,
      "step": 11575
    },
    {
      "entropy": 6.128266525268555,
      "epoch": 1.0226068527022254,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004902062842861696,
      "loss": 5.8725,
      "mean_token_accuracy": 0.16149003356695174,
      "num_tokens": 21591390.0,
      "step": 11580
    },
    {
      "entropy": 6.196033906936646,
      "epoch": 1.0230483927940657,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004901970922361154,
      "loss": 5.9845,
      "mean_token_accuracy": 0.15903691053390503,
      "num_tokens": 21600766.0,
      "step": 11585
    },
    {
      "entropy": 6.170470571517944,
      "epoch": 1.023489932885906,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004901878959704726,
      "loss": 5.8975,
      "mean_token_accuracy": 0.16713234633207322,
      "num_tokens": 21609264.0,
      "step": 11590
    },
    {
      "entropy": 6.205210113525391,
      "epoch": 1.0239314729777464,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004901786954894212,
      "loss": 5.9855,
      "mean_token_accuracy": 0.1524491161108017,
      "num_tokens": 21618409.0,
      "step": 11595
    },
    {
      "entropy": 6.130446910858154,
      "epoch": 1.0243730130695867,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004901694907931416,
      "loss": 5.8477,
      "mean_token_accuracy": 0.15585287660360336,
      "num_tokens": 21628158.0,
      "step": 11600
    },
    {
      "entropy": 6.152302265167236,
      "epoch": 1.024814553161427,
      "grad_norm": 2.890625,
      "learning_rate": 0.000490160281881814,
      "loss": 5.9875,
      "mean_token_accuracy": 0.16069798320531845,
      "num_tokens": 21637859.0,
      "step": 11605
    },
    {
      "entropy": 6.184852123260498,
      "epoch": 1.0252560932532675,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004901510687556188,
      "loss": 5.9927,
      "mean_token_accuracy": 0.15484195500612258,
      "num_tokens": 21647389.0,
      "step": 11610
    },
    {
      "entropy": 6.150085973739624,
      "epoch": 1.0256976333451078,
      "grad_norm": 8.125,
      "learning_rate": 0.0004901418514147367,
      "loss": 5.9445,
      "mean_token_accuracy": 0.15903750360012053,
      "num_tokens": 21655857.0,
      "step": 11615
    },
    {
      "entropy": 6.150744438171387,
      "epoch": 1.026139173436948,
      "grad_norm": 2.546875,
      "learning_rate": 0.000490132629859348,
      "loss": 5.8962,
      "mean_token_accuracy": 0.16650709062814711,
      "num_tokens": 21666150.0,
      "step": 11620
    },
    {
      "entropy": 6.154983329772949,
      "epoch": 1.0265807135287883,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004901234040896334,
      "loss": 5.9416,
      "mean_token_accuracy": 0.15803727358579636,
      "num_tokens": 21675662.0,
      "step": 11625
    },
    {
      "entropy": 6.196630907058716,
      "epoch": 1.0270222536206288,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004901141741057737,
      "loss": 5.9894,
      "mean_token_accuracy": 0.15326767265796662,
      "num_tokens": 21685721.0,
      "step": 11630
    },
    {
      "entropy": 6.1874120235443115,
      "epoch": 1.027463793712469,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004901049399079495,
      "loss": 5.947,
      "mean_token_accuracy": 0.16126787662506104,
      "num_tokens": 21694507.0,
      "step": 11635
    },
    {
      "entropy": 6.195686292648316,
      "epoch": 1.0279053338043094,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004900957014963421,
      "loss": 5.9507,
      "mean_token_accuracy": 0.1644703760743141,
      "num_tokens": 21704092.0,
      "step": 11640
    },
    {
      "entropy": 6.100425577163696,
      "epoch": 1.0283468738961499,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004900864588711321,
      "loss": 5.8766,
      "mean_token_accuracy": 0.16685758382081986,
      "num_tokens": 21713313.0,
      "step": 11645
    },
    {
      "entropy": 6.11095404624939,
      "epoch": 1.0287884139879901,
      "grad_norm": 2.375,
      "learning_rate": 0.0004900772120325008,
      "loss": 5.9207,
      "mean_token_accuracy": 0.16339877992868423,
      "num_tokens": 21721622.0,
      "step": 11650
    },
    {
      "entropy": 6.114428520202637,
      "epoch": 1.0292299540798304,
      "grad_norm": 2.75,
      "learning_rate": 0.0004900679609806291,
      "loss": 5.8843,
      "mean_token_accuracy": 0.1624878764152527,
      "num_tokens": 21730914.0,
      "step": 11655
    },
    {
      "entropy": 6.149040985107422,
      "epoch": 1.0296714941716707,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004900587057156985,
      "loss": 5.8667,
      "mean_token_accuracy": 0.163518089056015,
      "num_tokens": 21740124.0,
      "step": 11660
    },
    {
      "entropy": 6.176793336868286,
      "epoch": 1.0301130342635112,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004900494462378902,
      "loss": 5.9652,
      "mean_token_accuracy": 0.1554120734333992,
      "num_tokens": 21750013.0,
      "step": 11665
    },
    {
      "entropy": 6.1557331562042235,
      "epoch": 1.0305545743553515,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004900401825473854,
      "loss": 5.9452,
      "mean_token_accuracy": 0.158529195189476,
      "num_tokens": 21758842.0,
      "step": 11670
    },
    {
      "entropy": 6.116462087631225,
      "epoch": 1.0309961144471917,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004900309146443658,
      "loss": 5.8492,
      "mean_token_accuracy": 0.1661784291267395,
      "num_tokens": 21767002.0,
      "step": 11675
    },
    {
      "entropy": 6.13418664932251,
      "epoch": 1.0314376545390322,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004900216425290128,
      "loss": 5.946,
      "mean_token_accuracy": 0.1634654611349106,
      "num_tokens": 21775372.0,
      "step": 11680
    },
    {
      "entropy": 6.157120037078857,
      "epoch": 1.0318791946308725,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004900123662015082,
      "loss": 5.9946,
      "mean_token_accuracy": 0.1534305691719055,
      "num_tokens": 21784541.0,
      "step": 11685
    },
    {
      "entropy": 6.193939828872681,
      "epoch": 1.0323207347227128,
      "grad_norm": 2.75,
      "learning_rate": 0.0004900030856620336,
      "loss": 5.8802,
      "mean_token_accuracy": 0.16111927032470702,
      "num_tokens": 21794525.0,
      "step": 11690
    },
    {
      "entropy": 6.254008054733276,
      "epoch": 1.032762274814553,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004899938009107708,
      "loss": 5.9328,
      "mean_token_accuracy": 0.15929633378982544,
      "num_tokens": 21804469.0,
      "step": 11695
    },
    {
      "entropy": 6.096216106414795,
      "epoch": 1.0332038149063936,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004899845119479019,
      "loss": 5.9296,
      "mean_token_accuracy": 0.16315037310123442,
      "num_tokens": 21813986.0,
      "step": 11700
    },
    {
      "entropy": 6.149116039276123,
      "epoch": 1.0336453549982338,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004899752187736085,
      "loss": 5.9257,
      "mean_token_accuracy": 0.16401932388544083,
      "num_tokens": 21824083.0,
      "step": 11705
    },
    {
      "entropy": 6.121707201004028,
      "epoch": 1.0340868950900741,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004899659213880728,
      "loss": 5.8642,
      "mean_token_accuracy": 0.16949550211429595,
      "num_tokens": 21833526.0,
      "step": 11710
    },
    {
      "entropy": 6.1797881603240965,
      "epoch": 1.0345284351819146,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004899566197914771,
      "loss": 5.9265,
      "mean_token_accuracy": 0.16507390141487122,
      "num_tokens": 21842537.0,
      "step": 11715
    },
    {
      "entropy": 6.175706672668457,
      "epoch": 1.034969975273755,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004899473139840033,
      "loss": 5.9196,
      "mean_token_accuracy": 0.15725167691707612,
      "num_tokens": 21851604.0,
      "step": 11720
    },
    {
      "entropy": 6.090625,
      "epoch": 1.0354115153655952,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004899380039658339,
      "loss": 5.9913,
      "mean_token_accuracy": 0.15647540837526322,
      "num_tokens": 21861480.0,
      "step": 11725
    },
    {
      "entropy": 6.141532373428345,
      "epoch": 1.0358530554574354,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004899286897371513,
      "loss": 5.9551,
      "mean_token_accuracy": 0.1585635870695114,
      "num_tokens": 21870901.0,
      "step": 11730
    },
    {
      "entropy": 6.136163282394409,
      "epoch": 1.036294595549276,
      "grad_norm": 2.703125,
      "learning_rate": 0.000489919371298138,
      "loss": 5.8344,
      "mean_token_accuracy": 0.17009264677762986,
      "num_tokens": 21880491.0,
      "step": 11735
    },
    {
      "entropy": 6.09901065826416,
      "epoch": 1.0367361356411162,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004899100486489763,
      "loss": 5.937,
      "mean_token_accuracy": 0.15906979888677597,
      "num_tokens": 21889833.0,
      "step": 11740
    },
    {
      "entropy": 6.076237916946411,
      "epoch": 1.0371776757329565,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004899007217898491,
      "loss": 5.945,
      "mean_token_accuracy": 0.15788724273443222,
      "num_tokens": 21899222.0,
      "step": 11745
    },
    {
      "entropy": 6.227149915695191,
      "epoch": 1.037619215824797,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004898913907209389,
      "loss": 5.9578,
      "mean_token_accuracy": 0.15345034524798393,
      "num_tokens": 21908848.0,
      "step": 11750
    },
    {
      "entropy": 6.198890542984008,
      "epoch": 1.0380607559166373,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004898820554424285,
      "loss": 5.8843,
      "mean_token_accuracy": 0.15585824102163315,
      "num_tokens": 21917867.0,
      "step": 11755
    },
    {
      "entropy": 6.2142982006073,
      "epoch": 1.0385022960084775,
      "grad_norm": 2.359375,
      "learning_rate": 0.000489872715954501,
      "loss": 5.9687,
      "mean_token_accuracy": 0.152188478410244,
      "num_tokens": 21927476.0,
      "step": 11760
    },
    {
      "entropy": 6.214330387115479,
      "epoch": 1.0389438361003178,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004898633722573391,
      "loss": 6.0303,
      "mean_token_accuracy": 0.15347810685634614,
      "num_tokens": 21937068.0,
      "step": 11765
    },
    {
      "entropy": 6.129259490966797,
      "epoch": 1.0393853761921583,
      "grad_norm": 2.90625,
      "learning_rate": 0.000489854024351126,
      "loss": 5.8966,
      "mean_token_accuracy": 0.16052083373069764,
      "num_tokens": 21946144.0,
      "step": 11770
    },
    {
      "entropy": 6.185786390304566,
      "epoch": 1.0398269162839986,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004898446722360446,
      "loss": 5.8318,
      "mean_token_accuracy": 0.16848260313272476,
      "num_tokens": 21955346.0,
      "step": 11775
    },
    {
      "entropy": 6.1774369239807125,
      "epoch": 1.0402684563758389,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004898353159122784,
      "loss": 5.9246,
      "mean_token_accuracy": 0.15950198024511336,
      "num_tokens": 21965034.0,
      "step": 11780
    },
    {
      "entropy": 6.155341625213623,
      "epoch": 1.0407099964676794,
      "grad_norm": 5.71875,
      "learning_rate": 0.0004898259553800106,
      "loss": 5.9058,
      "mean_token_accuracy": 0.15801970213651656,
      "num_tokens": 21975167.0,
      "step": 11785
    },
    {
      "entropy": 6.178949689865112,
      "epoch": 1.0411515365595196,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004898165906394244,
      "loss": 5.8888,
      "mean_token_accuracy": 0.1605347737669945,
      "num_tokens": 21984372.0,
      "step": 11790
    },
    {
      "entropy": 6.113554239273071,
      "epoch": 1.04159307665136,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004898072216907033,
      "loss": 5.8976,
      "mean_token_accuracy": 0.15484814941883088,
      "num_tokens": 21993513.0,
      "step": 11795
    },
    {
      "entropy": 6.145312404632568,
      "epoch": 1.0420346167432002,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004897978485340311,
      "loss": 5.9649,
      "mean_token_accuracy": 0.15129612386226654,
      "num_tokens": 22002593.0,
      "step": 11800
    },
    {
      "entropy": 6.229594421386719,
      "epoch": 1.0424761568350407,
      "grad_norm": 2.703125,
      "learning_rate": 0.000489788471169591,
      "loss": 5.9708,
      "mean_token_accuracy": 0.15879204273223876,
      "num_tokens": 22012296.0,
      "step": 11805
    },
    {
      "entropy": 6.222621250152588,
      "epoch": 1.042917696926881,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004897790895975671,
      "loss": 6.0244,
      "mean_token_accuracy": 0.15057723820209504,
      "num_tokens": 22021267.0,
      "step": 11810
    },
    {
      "entropy": 6.146600437164307,
      "epoch": 1.0433592370187212,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004897697038181428,
      "loss": 5.9483,
      "mean_token_accuracy": 0.16155195981264114,
      "num_tokens": 22030878.0,
      "step": 11815
    },
    {
      "entropy": 6.079737377166748,
      "epoch": 1.0438007771105617,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004897603138315022,
      "loss": 5.9586,
      "mean_token_accuracy": 0.16057838797569274,
      "num_tokens": 22040245.0,
      "step": 11820
    },
    {
      "entropy": 6.184117650985717,
      "epoch": 1.044242317202402,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004897509196378293,
      "loss": 5.9458,
      "mean_token_accuracy": 0.16077676713466643,
      "num_tokens": 22050169.0,
      "step": 11825
    },
    {
      "entropy": 6.250207996368408,
      "epoch": 1.0446838572942423,
      "grad_norm": 3.296875,
      "learning_rate": 0.000489741521237308,
      "loss": 5.9736,
      "mean_token_accuracy": 0.15720196366310119,
      "num_tokens": 22059168.0,
      "step": 11830
    },
    {
      "entropy": 6.138293361663818,
      "epoch": 1.0451253973860826,
      "grad_norm": 6.5,
      "learning_rate": 0.0004897321186301223,
      "loss": 5.9075,
      "mean_token_accuracy": 0.1591096416115761,
      "num_tokens": 22068326.0,
      "step": 11835
    },
    {
      "entropy": 6.160605764389038,
      "epoch": 1.045566937477923,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004897227118164566,
      "loss": 5.9326,
      "mean_token_accuracy": 0.16288507133722305,
      "num_tokens": 22078026.0,
      "step": 11840
    },
    {
      "entropy": 6.159685182571411,
      "epoch": 1.0460084775697633,
      "grad_norm": 2.671875,
      "learning_rate": 0.000489713300796495,
      "loss": 5.9839,
      "mean_token_accuracy": 0.15039085000753402,
      "num_tokens": 22088230.0,
      "step": 11845
    },
    {
      "entropy": 6.077940797805786,
      "epoch": 1.0464500176616036,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004897038855704221,
      "loss": 5.795,
      "mean_token_accuracy": 0.17047708332538605,
      "num_tokens": 22096667.0,
      "step": 11850
    },
    {
      "entropy": 6.1295373916625975,
      "epoch": 1.0468915577534441,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004896944661384222,
      "loss": 5.9218,
      "mean_token_accuracy": 0.16085873395204545,
      "num_tokens": 22106871.0,
      "step": 11855
    },
    {
      "entropy": 6.202170372009277,
      "epoch": 1.0473330978452844,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004896850425006797,
      "loss": 5.8754,
      "mean_token_accuracy": 0.1568978562951088,
      "num_tokens": 22114954.0,
      "step": 11860
    },
    {
      "entropy": 6.123904180526734,
      "epoch": 1.0477746379371247,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004896756146573794,
      "loss": 5.9086,
      "mean_token_accuracy": 0.15844995230436326,
      "num_tokens": 22125293.0,
      "step": 11865
    },
    {
      "entropy": 6.214665269851684,
      "epoch": 1.048216178028965,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004896661826087059,
      "loss": 5.9516,
      "mean_token_accuracy": 0.16156904697418212,
      "num_tokens": 22133731.0,
      "step": 11870
    },
    {
      "entropy": 6.090568351745605,
      "epoch": 1.0486577181208054,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004896567463548439,
      "loss": 5.9287,
      "mean_token_accuracy": 0.15373207181692122,
      "num_tokens": 22143007.0,
      "step": 11875
    },
    {
      "entropy": 6.097213315963745,
      "epoch": 1.0490992582126457,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004896473058959783,
      "loss": 5.9702,
      "mean_token_accuracy": 0.16133466511964797,
      "num_tokens": 22152044.0,
      "step": 11880
    },
    {
      "entropy": 6.24034276008606,
      "epoch": 1.049540798304486,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004896378612322942,
      "loss": 6.006,
      "mean_token_accuracy": 0.15323544591665267,
      "num_tokens": 22161473.0,
      "step": 11885
    },
    {
      "entropy": 6.170629644393921,
      "epoch": 1.0499823383963265,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004896284123639763,
      "loss": 5.9185,
      "mean_token_accuracy": 0.1587260454893112,
      "num_tokens": 22171704.0,
      "step": 11890
    },
    {
      "entropy": 6.1624367237091064,
      "epoch": 1.0504238784881668,
      "grad_norm": 2.90625,
      "learning_rate": 0.00048961895929121,
      "loss": 5.9177,
      "mean_token_accuracy": 0.15345404297113419,
      "num_tokens": 22180593.0,
      "step": 11895
    },
    {
      "entropy": 6.148451614379883,
      "epoch": 1.050865418580007,
      "grad_norm": 2.25,
      "learning_rate": 0.0004896095020141802,
      "loss": 5.8895,
      "mean_token_accuracy": 0.16307324469089507,
      "num_tokens": 22189552.0,
      "step": 11900
    },
    {
      "entropy": 6.156101417541504,
      "epoch": 1.0513069586718473,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004896000405330724,
      "loss": 5.9152,
      "mean_token_accuracy": 0.15955967903137208,
      "num_tokens": 22199314.0,
      "step": 11905
    },
    {
      "entropy": 6.04547085762024,
      "epoch": 1.0517484987636878,
      "grad_norm": 2.15625,
      "learning_rate": 0.000489590574848072,
      "loss": 5.8943,
      "mean_token_accuracy": 0.16911342293024062,
      "num_tokens": 22208128.0,
      "step": 11910
    },
    {
      "entropy": 6.062590312957764,
      "epoch": 1.052190038855528,
      "grad_norm": 3.4375,
      "learning_rate": 0.000489581104959364,
      "loss": 5.8633,
      "mean_token_accuracy": 0.1628858581185341,
      "num_tokens": 22217700.0,
      "step": 11915
    },
    {
      "entropy": 6.165154886245728,
      "epoch": 1.0526315789473684,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004895716308671343,
      "loss": 5.9322,
      "mean_token_accuracy": 0.1612838864326477,
      "num_tokens": 22225941.0,
      "step": 11920
    },
    {
      "entropy": 6.1347064018249515,
      "epoch": 1.0530731190392089,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004895621525715685,
      "loss": 5.8912,
      "mean_token_accuracy": 0.16051552891731263,
      "num_tokens": 22234935.0,
      "step": 11925
    },
    {
      "entropy": 6.132886457443237,
      "epoch": 1.0535146591310491,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004895526700728521,
      "loss": 5.9747,
      "mean_token_accuracy": 0.16112820357084273,
      "num_tokens": 22244087.0,
      "step": 11930
    },
    {
      "entropy": 6.146267557144165,
      "epoch": 1.0539561992228894,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004895431833711708,
      "loss": 5.9258,
      "mean_token_accuracy": 0.15657443851232528,
      "num_tokens": 22254277.0,
      "step": 11935
    },
    {
      "entropy": 6.17157735824585,
      "epoch": 1.0543977393147297,
      "grad_norm": 3.5,
      "learning_rate": 0.0004895336924667107,
      "loss": 5.8815,
      "mean_token_accuracy": 0.16508007943630218,
      "num_tokens": 22263150.0,
      "step": 11940
    },
    {
      "entropy": 6.173756694793701,
      "epoch": 1.0548392794065702,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004895241973596576,
      "loss": 5.9115,
      "mean_token_accuracy": 0.16058234125375748,
      "num_tokens": 22271973.0,
      "step": 11945
    },
    {
      "entropy": 6.107911205291748,
      "epoch": 1.0552808194984105,
      "grad_norm": 2.625,
      "learning_rate": 0.0004895146980501973,
      "loss": 5.8854,
      "mean_token_accuracy": 0.1664498120546341,
      "num_tokens": 22280637.0,
      "step": 11950
    },
    {
      "entropy": 6.124010848999023,
      "epoch": 1.0557223595902507,
      "grad_norm": 2.125,
      "learning_rate": 0.0004895051945385163,
      "loss": 5.9039,
      "mean_token_accuracy": 0.16170118302106856,
      "num_tokens": 22290601.0,
      "step": 11955
    },
    {
      "entropy": 6.182231855392456,
      "epoch": 1.0561638996820912,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004894956868248004,
      "loss": 5.9424,
      "mean_token_accuracy": 0.1595696061849594,
      "num_tokens": 22300212.0,
      "step": 11960
    },
    {
      "entropy": 6.214186763763427,
      "epoch": 1.0566054397739315,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004894861749092359,
      "loss": 5.9153,
      "mean_token_accuracy": 0.1592661112546921,
      "num_tokens": 22309171.0,
      "step": 11965
    },
    {
      "entropy": 6.099295663833618,
      "epoch": 1.0570469798657718,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004894766587920094,
      "loss": 5.9269,
      "mean_token_accuracy": 0.16791395246982574,
      "num_tokens": 22318457.0,
      "step": 11970
    },
    {
      "entropy": 6.208569288253784,
      "epoch": 1.057488519957612,
      "grad_norm": 2.578125,
      "learning_rate": 0.000489467138473307,
      "loss": 5.9277,
      "mean_token_accuracy": 0.1591622695326805,
      "num_tokens": 22327077.0,
      "step": 11975
    },
    {
      "entropy": 6.142443895339966,
      "epoch": 1.0579300600494526,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004894576139533154,
      "loss": 5.9294,
      "mean_token_accuracy": 0.1621837630867958,
      "num_tokens": 22336259.0,
      "step": 11980
    },
    {
      "entropy": 6.147655057907104,
      "epoch": 1.0583716001412928,
      "grad_norm": 3.9375,
      "learning_rate": 0.000489448085232221,
      "loss": 5.8967,
      "mean_token_accuracy": 0.17437706738710404,
      "num_tokens": 22344978.0,
      "step": 11985
    },
    {
      "entropy": 6.140650081634521,
      "epoch": 1.058813140233133,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004894385523102107,
      "loss": 5.9268,
      "mean_token_accuracy": 0.157879801094532,
      "num_tokens": 22353950.0,
      "step": 11990
    },
    {
      "entropy": 6.189157724380493,
      "epoch": 1.0592546803249736,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004894290151874711,
      "loss": 5.8813,
      "mean_token_accuracy": 0.15687581300735473,
      "num_tokens": 22363100.0,
      "step": 11995
    },
    {
      "entropy": 6.090399932861328,
      "epoch": 1.0596962204168139,
      "grad_norm": 2.59375,
      "learning_rate": 0.000489419473864189,
      "loss": 5.8611,
      "mean_token_accuracy": 0.16944732517004013,
      "num_tokens": 22371979.0,
      "step": 12000
    },
    {
      "epoch": 1.0596962204168139,
      "eval_entropy": 5.936019211852267,
      "eval_loss": 6.008758068084717,
      "eval_mean_token_accuracy": 0.16387403600808076,
      "eval_num_tokens": 22371979.0,
      "eval_runtime": 26.0247,
      "eval_samples_per_second": 1356.981,
      "eval_steps_per_second": 169.647,
      "step": 12000
    },
    {
      "entropy": 6.114017915725708,
      "epoch": 1.0601377605086542,
      "grad_norm": 4.5,
      "learning_rate": 0.0004894099283405515,
      "loss": 5.8816,
      "mean_token_accuracy": 0.16182024776935577,
      "num_tokens": 22380987.0,
      "step": 12005
    },
    {
      "entropy": 6.134852933883667,
      "epoch": 1.0605793006004944,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004894003786167452,
      "loss": 5.9509,
      "mean_token_accuracy": 0.1527145341038704,
      "num_tokens": 22389367.0,
      "step": 12010
    },
    {
      "entropy": 6.195053482055664,
      "epoch": 1.061020840692335,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004893908246929576,
      "loss": 6.0065,
      "mean_token_accuracy": 0.1539388418197632,
      "num_tokens": 22399158.0,
      "step": 12015
    },
    {
      "entropy": 6.239836120605469,
      "epoch": 1.0614623807841752,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004893812665693755,
      "loss": 5.9707,
      "mean_token_accuracy": 0.15816060453653336,
      "num_tokens": 22408245.0,
      "step": 12020
    },
    {
      "entropy": 6.172027349472046,
      "epoch": 1.0619039208760155,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004893717042461865,
      "loss": 5.9485,
      "mean_token_accuracy": 0.16011442244052887,
      "num_tokens": 22417014.0,
      "step": 12025
    },
    {
      "entropy": 6.192820358276367,
      "epoch": 1.062345460967856,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004893621377235777,
      "loss": 5.8893,
      "mean_token_accuracy": 0.16173512786626815,
      "num_tokens": 22426504.0,
      "step": 12030
    },
    {
      "entropy": 6.2348034381866455,
      "epoch": 1.0627870010596963,
      "grad_norm": 3.875,
      "learning_rate": 0.0004893525670017364,
      "loss": 5.9944,
      "mean_token_accuracy": 0.15601857602596284,
      "num_tokens": 22435859.0,
      "step": 12035
    },
    {
      "entropy": 6.217994928359985,
      "epoch": 1.0632285411515365,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004893429920808503,
      "loss": 5.9204,
      "mean_token_accuracy": 0.16233511120080948,
      "num_tokens": 22445112.0,
      "step": 12040
    },
    {
      "entropy": 6.179212856292724,
      "epoch": 1.0636700812433768,
      "grad_norm": 3.609375,
      "learning_rate": 0.000489333412961107,
      "loss": 5.9679,
      "mean_token_accuracy": 0.15648476928472518,
      "num_tokens": 22454602.0,
      "step": 12045
    },
    {
      "entropy": 6.229143381118774,
      "epoch": 1.0641116213352173,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004893238296426939,
      "loss": 5.9923,
      "mean_token_accuracy": 0.1561363086104393,
      "num_tokens": 22463918.0,
      "step": 12050
    },
    {
      "entropy": 6.149765300750732,
      "epoch": 1.0645531614270576,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004893142421257989,
      "loss": 5.8841,
      "mean_token_accuracy": 0.15609944760799407,
      "num_tokens": 22473059.0,
      "step": 12055
    },
    {
      "entropy": 6.110332059860229,
      "epoch": 1.0649947015188979,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004893046504106099,
      "loss": 5.9448,
      "mean_token_accuracy": 0.15709808468818665,
      "num_tokens": 22482615.0,
      "step": 12060
    },
    {
      "entropy": 6.111856079101562,
      "epoch": 1.0654362416107384,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004892950544973145,
      "loss": 5.8673,
      "mean_token_accuracy": 0.16662584245204926,
      "num_tokens": 22492141.0,
      "step": 12065
    },
    {
      "entropy": 6.180486536026001,
      "epoch": 1.0658777817025786,
      "grad_norm": 2.25,
      "learning_rate": 0.0004892854543861009,
      "loss": 5.8295,
      "mean_token_accuracy": 0.16754715293645858,
      "num_tokens": 22502084.0,
      "step": 12070
    },
    {
      "entropy": 6.156108236312866,
      "epoch": 1.066319321794419,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004892758500771572,
      "loss": 5.8538,
      "mean_token_accuracy": 0.1638255760073662,
      "num_tokens": 22510565.0,
      "step": 12075
    },
    {
      "entropy": 6.177877473831177,
      "epoch": 1.0667608618862592,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004892662415706714,
      "loss": 5.9665,
      "mean_token_accuracy": 0.15753857493400575,
      "num_tokens": 22520089.0,
      "step": 12080
    },
    {
      "entropy": 6.118342351913452,
      "epoch": 1.0672024019780997,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004892566288668319,
      "loss": 5.8933,
      "mean_token_accuracy": 0.1633586660027504,
      "num_tokens": 22529008.0,
      "step": 12085
    },
    {
      "entropy": 6.140356922149659,
      "epoch": 1.06764394206994,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004892470119658267,
      "loss": 5.8654,
      "mean_token_accuracy": 0.164863882958889,
      "num_tokens": 22538040.0,
      "step": 12090
    },
    {
      "entropy": 6.191899681091309,
      "epoch": 1.0680854821617802,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004892373908678445,
      "loss": 5.9458,
      "mean_token_accuracy": 0.15595253705978393,
      "num_tokens": 22548786.0,
      "step": 12095
    },
    {
      "entropy": 6.234998321533203,
      "epoch": 1.0685270222536207,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004892277655730736,
      "loss": 5.9289,
      "mean_token_accuracy": 0.162250292301178,
      "num_tokens": 22559841.0,
      "step": 12100
    },
    {
      "entropy": 6.207377529144287,
      "epoch": 1.068968562345461,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004892181360817027,
      "loss": 5.867,
      "mean_token_accuracy": 0.1664956107735634,
      "num_tokens": 22569338.0,
      "step": 12105
    },
    {
      "entropy": 6.111291885375977,
      "epoch": 1.0694101024373013,
      "grad_norm": 5.125,
      "learning_rate": 0.0004892085023939204,
      "loss": 5.9291,
      "mean_token_accuracy": 0.155917289853096,
      "num_tokens": 22578184.0,
      "step": 12110
    },
    {
      "entropy": 6.176279306411743,
      "epoch": 1.0698516425291416,
      "grad_norm": 4.375,
      "learning_rate": 0.0004891988645099151,
      "loss": 6.0379,
      "mean_token_accuracy": 0.1512170098721981,
      "num_tokens": 22587007.0,
      "step": 12115
    },
    {
      "entropy": 6.1934202194213865,
      "epoch": 1.070293182620982,
      "grad_norm": 20.0,
      "learning_rate": 0.0004891892224298761,
      "loss": 5.958,
      "mean_token_accuracy": 0.15347342044115067,
      "num_tokens": 22596898.0,
      "step": 12120
    },
    {
      "entropy": 6.175861072540283,
      "epoch": 1.0707347227128223,
      "grad_norm": 2.9375,
      "learning_rate": 0.000489179576153992,
      "loss": 5.8603,
      "mean_token_accuracy": 0.17192627489566803,
      "num_tokens": 22605627.0,
      "step": 12125
    },
    {
      "entropy": 6.128075122833252,
      "epoch": 1.0711762628046626,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004891699256824518,
      "loss": 5.8313,
      "mean_token_accuracy": 0.16555165946483613,
      "num_tokens": 22615269.0,
      "step": 12130
    },
    {
      "entropy": 6.102174139022827,
      "epoch": 1.071617802896503,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004891602710154446,
      "loss": 5.8827,
      "mean_token_accuracy": 0.1575715348124504,
      "num_tokens": 22625048.0,
      "step": 12135
    },
    {
      "entropy": 6.094519901275635,
      "epoch": 1.0720593429883434,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004891506121531593,
      "loss": 5.8761,
      "mean_token_accuracy": 0.1634797066450119,
      "num_tokens": 22634146.0,
      "step": 12140
    },
    {
      "entropy": 6.146231031417846,
      "epoch": 1.0725008830801837,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004891409490957855,
      "loss": 5.9562,
      "mean_token_accuracy": 0.159510750323534,
      "num_tokens": 22642915.0,
      "step": 12145
    },
    {
      "entropy": 6.2338377952575685,
      "epoch": 1.072942423172024,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004891312818435122,
      "loss": 5.8788,
      "mean_token_accuracy": 0.15790450870990752,
      "num_tokens": 22651546.0,
      "step": 12150
    },
    {
      "entropy": 6.223159980773926,
      "epoch": 1.0733839632638644,
      "grad_norm": 3.390625,
      "learning_rate": 0.000489121610396529,
      "loss": 5.9171,
      "mean_token_accuracy": 0.1610031321644783,
      "num_tokens": 22660712.0,
      "step": 12155
    },
    {
      "entropy": 6.140164184570312,
      "epoch": 1.0738255033557047,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004891119347550251,
      "loss": 5.8801,
      "mean_token_accuracy": 0.16884622126817703,
      "num_tokens": 22669428.0,
      "step": 12160
    },
    {
      "entropy": 6.134841394424439,
      "epoch": 1.074267043447545,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004891022549191902,
      "loss": 5.9763,
      "mean_token_accuracy": 0.15740178823471068,
      "num_tokens": 22679126.0,
      "step": 12165
    },
    {
      "entropy": 6.046198177337646,
      "epoch": 1.0747085835393855,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004890925708892138,
      "loss": 5.8643,
      "mean_token_accuracy": 0.16726138740777968,
      "num_tokens": 22688640.0,
      "step": 12170
    },
    {
      "entropy": 6.173946762084961,
      "epoch": 1.0751501236312258,
      "grad_norm": 3.625,
      "learning_rate": 0.000489082882665286,
      "loss": 5.8406,
      "mean_token_accuracy": 0.16655810922384262,
      "num_tokens": 22698158.0,
      "step": 12175
    },
    {
      "entropy": 6.106610918045044,
      "epoch": 1.075591663723066,
      "grad_norm": 3.453125,
      "learning_rate": 0.000489073190247596,
      "loss": 5.7853,
      "mean_token_accuracy": 0.17605790197849275,
      "num_tokens": 22707649.0,
      "step": 12180
    },
    {
      "entropy": 6.20052547454834,
      "epoch": 1.0760332038149063,
      "grad_norm": 3.859375,
      "learning_rate": 0.000489063493636334,
      "loss": 5.9799,
      "mean_token_accuracy": 0.1654104471206665,
      "num_tokens": 22717123.0,
      "step": 12185
    },
    {
      "entropy": 6.157053995132446,
      "epoch": 1.0764747439067468,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004890537928316901,
      "loss": 5.9086,
      "mean_token_accuracy": 0.16282155960798264,
      "num_tokens": 22725921.0,
      "step": 12190
    },
    {
      "entropy": 6.175580406188965,
      "epoch": 1.076916283998587,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004890440878338538,
      "loss": 6.0119,
      "mean_token_accuracy": 0.15466470122337342,
      "num_tokens": 22735798.0,
      "step": 12195
    },
    {
      "entropy": 6.1869574069976805,
      "epoch": 1.0773578240904274,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004890343786430158,
      "loss": 5.9862,
      "mean_token_accuracy": 0.1566761016845703,
      "num_tokens": 22745583.0,
      "step": 12200
    },
    {
      "entropy": 6.1388157367706295,
      "epoch": 1.0777993641822678,
      "grad_norm": 3.140625,
      "learning_rate": 0.000489024665259366,
      "loss": 5.8229,
      "mean_token_accuracy": 0.1679653689265251,
      "num_tokens": 22754693.0,
      "step": 12205
    },
    {
      "entropy": 6.05749831199646,
      "epoch": 1.0782409042741081,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004890149476830947,
      "loss": 5.8255,
      "mean_token_accuracy": 0.16756767332553862,
      "num_tokens": 22763416.0,
      "step": 12210
    },
    {
      "entropy": 6.107419824600219,
      "epoch": 1.0786824443659484,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004890052259143924,
      "loss": 5.8998,
      "mean_token_accuracy": 0.1657376393675804,
      "num_tokens": 22772411.0,
      "step": 12215
    },
    {
      "entropy": 6.068491840362549,
      "epoch": 1.0791239844577887,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004889954999534493,
      "loss": 5.7567,
      "mean_token_accuracy": 0.1732191562652588,
      "num_tokens": 22780148.0,
      "step": 12220
    },
    {
      "entropy": 6.0916108131408695,
      "epoch": 1.0795655245496292,
      "grad_norm": 4.625,
      "learning_rate": 0.0004889857698004561,
      "loss": 5.869,
      "mean_token_accuracy": 0.16153077334165572,
      "num_tokens": 22789453.0,
      "step": 12225
    },
    {
      "entropy": 6.190332221984863,
      "epoch": 1.0800070646414694,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004889760354556033,
      "loss": 5.9167,
      "mean_token_accuracy": 0.16472706943750381,
      "num_tokens": 22798800.0,
      "step": 12230
    },
    {
      "entropy": 6.125522041320801,
      "epoch": 1.0804486047333097,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004889662969190817,
      "loss": 5.9726,
      "mean_token_accuracy": 0.1637173995375633,
      "num_tokens": 22807766.0,
      "step": 12235
    },
    {
      "entropy": 6.102245473861695,
      "epoch": 1.0808901448251502,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004889565541910822,
      "loss": 5.8814,
      "mean_token_accuracy": 0.16091295182704926,
      "num_tokens": 22817316.0,
      "step": 12240
    },
    {
      "entropy": 6.132245779037476,
      "epoch": 1.0813316849169905,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004889468072717953,
      "loss": 5.9222,
      "mean_token_accuracy": 0.16387611627578735,
      "num_tokens": 22827082.0,
      "step": 12245
    },
    {
      "entropy": 6.159386014938354,
      "epoch": 1.0817732250088308,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004889370561614122,
      "loss": 5.9044,
      "mean_token_accuracy": 0.16351822763681412,
      "num_tokens": 22836114.0,
      "step": 12250
    },
    {
      "entropy": 6.191832399368286,
      "epoch": 1.082214765100671,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004889273008601238,
      "loss": 6.0257,
      "mean_token_accuracy": 0.15433973893523217,
      "num_tokens": 22845653.0,
      "step": 12255
    },
    {
      "entropy": 6.22887487411499,
      "epoch": 1.0826563051925115,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004889175413681213,
      "loss": 5.9441,
      "mean_token_accuracy": 0.16206600964069368,
      "num_tokens": 22855472.0,
      "step": 12260
    },
    {
      "entropy": 6.143603515625,
      "epoch": 1.0830978452843518,
      "grad_norm": 2.125,
      "learning_rate": 0.0004889077776855957,
      "loss": 5.8751,
      "mean_token_accuracy": 0.16495681256055833,
      "num_tokens": 22864223.0,
      "step": 12265
    },
    {
      "entropy": 6.133127117156983,
      "epoch": 1.083539385376192,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004888980098127385,
      "loss": 5.9604,
      "mean_token_accuracy": 0.15613909363746642,
      "num_tokens": 22872868.0,
      "step": 12270
    },
    {
      "entropy": 6.190570926666259,
      "epoch": 1.0839809254680326,
      "grad_norm": 2.875,
      "learning_rate": 0.0004888882377497409,
      "loss": 5.9698,
      "mean_token_accuracy": 0.15288522988557815,
      "num_tokens": 22881867.0,
      "step": 12275
    },
    {
      "entropy": 6.202322006225586,
      "epoch": 1.0844224655598729,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004888784614967943,
      "loss": 5.8926,
      "mean_token_accuracy": 0.15369175523519515,
      "num_tokens": 22890904.0,
      "step": 12280
    },
    {
      "entropy": 6.130778217315674,
      "epoch": 1.0848640056517131,
      "grad_norm": 3.125,
      "learning_rate": 0.0004888686810540905,
      "loss": 5.8424,
      "mean_token_accuracy": 0.1671067714691162,
      "num_tokens": 22900128.0,
      "step": 12285
    },
    {
      "entropy": 6.162302684783936,
      "epoch": 1.0853055457435534,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004888588964218208,
      "loss": 5.9475,
      "mean_token_accuracy": 0.15772829204797745,
      "num_tokens": 22910101.0,
      "step": 12290
    },
    {
      "entropy": 6.16146240234375,
      "epoch": 1.085747085835394,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004888491076001768,
      "loss": 5.8911,
      "mean_token_accuracy": 0.1635946586728096,
      "num_tokens": 22919372.0,
      "step": 12295
    },
    {
      "entropy": 6.167607736587525,
      "epoch": 1.0861886259272342,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004888393145893505,
      "loss": 5.9287,
      "mean_token_accuracy": 0.15791655331850052,
      "num_tokens": 22929053.0,
      "step": 12300
    },
    {
      "entropy": 6.098089218139648,
      "epoch": 1.0866301660190745,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004888295173895335,
      "loss": 5.8579,
      "mean_token_accuracy": 0.16462911069393157,
      "num_tokens": 22938173.0,
      "step": 12305
    },
    {
      "entropy": 6.102741050720215,
      "epoch": 1.087071706110915,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004888197160009181,
      "loss": 5.9862,
      "mean_token_accuracy": 0.15943123251199723,
      "num_tokens": 22947076.0,
      "step": 12310
    },
    {
      "entropy": 6.113621377944947,
      "epoch": 1.0875132462027552,
      "grad_norm": 2.53125,
      "learning_rate": 0.000488809910423696,
      "loss": 5.9082,
      "mean_token_accuracy": 0.16993424743413926,
      "num_tokens": 22956267.0,
      "step": 12315
    },
    {
      "entropy": 6.262680530548096,
      "epoch": 1.0879547862945955,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004888001006580592,
      "loss": 5.9979,
      "mean_token_accuracy": 0.1547674521803856,
      "num_tokens": 22966751.0,
      "step": 12320
    },
    {
      "entropy": 6.077016067504883,
      "epoch": 1.0883963263864358,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004887902867042002,
      "loss": 5.9194,
      "mean_token_accuracy": 0.16017358005046844,
      "num_tokens": 22975862.0,
      "step": 12325
    },
    {
      "entropy": 6.117411947250366,
      "epoch": 1.0888378664782763,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004887804685623111,
      "loss": 5.8654,
      "mean_token_accuracy": 0.16339120864868165,
      "num_tokens": 22985085.0,
      "step": 12330
    },
    {
      "entropy": 6.240997791290283,
      "epoch": 1.0892794065701166,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004887706462325841,
      "loss": 6.0032,
      "mean_token_accuracy": 0.15804048776626586,
      "num_tokens": 22993718.0,
      "step": 12335
    },
    {
      "entropy": 6.155885887145996,
      "epoch": 1.0897209466619568,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004887608197152117,
      "loss": 5.8497,
      "mean_token_accuracy": 0.17430692464113234,
      "num_tokens": 23002611.0,
      "step": 12340
    },
    {
      "entropy": 6.117863702774048,
      "epoch": 1.0901624867537973,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004887509890103864,
      "loss": 6.0135,
      "mean_token_accuracy": 0.15509911328554155,
      "num_tokens": 23012333.0,
      "step": 12345
    },
    {
      "entropy": 6.161442995071411,
      "epoch": 1.0906040268456376,
      "grad_norm": 3.375,
      "learning_rate": 0.0004887411541183009,
      "loss": 5.9338,
      "mean_token_accuracy": 0.1575099855661392,
      "num_tokens": 23020998.0,
      "step": 12350
    },
    {
      "entropy": 6.215446281433105,
      "epoch": 1.091045566937478,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004887313150391477,
      "loss": 5.9363,
      "mean_token_accuracy": 0.15863421857357024,
      "num_tokens": 23029520.0,
      "step": 12355
    },
    {
      "entropy": 6.200163316726685,
      "epoch": 1.0914871070293182,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004887214717731196,
      "loss": 5.9947,
      "mean_token_accuracy": 0.15386056303977966,
      "num_tokens": 23039224.0,
      "step": 12360
    },
    {
      "entropy": 6.1819822788238525,
      "epoch": 1.0919286471211587,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004887116243204094,
      "loss": 5.8827,
      "mean_token_accuracy": 0.16292284280061722,
      "num_tokens": 23047851.0,
      "step": 12365
    },
    {
      "entropy": 6.038466930389404,
      "epoch": 1.092370187212999,
      "grad_norm": 3.1875,
      "learning_rate": 0.00048870177268121,
      "loss": 5.9087,
      "mean_token_accuracy": 0.1561818964779377,
      "num_tokens": 23057242.0,
      "step": 12370
    },
    {
      "entropy": 6.126602554321289,
      "epoch": 1.0928117273048392,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004886919168557145,
      "loss": 6.0226,
      "mean_token_accuracy": 0.15501100122928618,
      "num_tokens": 23066873.0,
      "step": 12375
    },
    {
      "entropy": 6.198664236068725,
      "epoch": 1.0932532673966797,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004886820568441158,
      "loss": 6.0039,
      "mean_token_accuracy": 0.1533634513616562,
      "num_tokens": 23076448.0,
      "step": 12380
    },
    {
      "entropy": 6.210957050323486,
      "epoch": 1.09369480748852,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004886721926466072,
      "loss": 5.9044,
      "mean_token_accuracy": 0.15767279267311096,
      "num_tokens": 23084759.0,
      "step": 12385
    },
    {
      "entropy": 6.116765308380127,
      "epoch": 1.0941363475803603,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004886623242633817,
      "loss": 5.8428,
      "mean_token_accuracy": 0.16857688128948212,
      "num_tokens": 23093236.0,
      "step": 12390
    },
    {
      "entropy": 6.1179845333099365,
      "epoch": 1.0945778876722005,
      "grad_norm": 2.375,
      "learning_rate": 0.000488652451694633,
      "loss": 5.8985,
      "mean_token_accuracy": 0.16083543747663498,
      "num_tokens": 23101621.0,
      "step": 12395
    },
    {
      "entropy": 6.1819006443023685,
      "epoch": 1.095019427764041,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004886425749405542,
      "loss": 5.9501,
      "mean_token_accuracy": 0.15517825186252593,
      "num_tokens": 23110882.0,
      "step": 12400
    },
    {
      "entropy": 6.236010360717773,
      "epoch": 1.0954609678558813,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004886326940013388,
      "loss": 5.9398,
      "mean_token_accuracy": 0.1623895585536957,
      "num_tokens": 23120669.0,
      "step": 12405
    },
    {
      "entropy": 6.144532537460327,
      "epoch": 1.0959025079477216,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004886228088771805,
      "loss": 5.8506,
      "mean_token_accuracy": 0.161375030875206,
      "num_tokens": 23129524.0,
      "step": 12410
    },
    {
      "entropy": 6.0762065887451175,
      "epoch": 1.096344048039562,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004886129195682729,
      "loss": 5.9163,
      "mean_token_accuracy": 0.15916515290737152,
      "num_tokens": 23139144.0,
      "step": 12415
    },
    {
      "entropy": 6.101714324951172,
      "epoch": 1.0967855881314024,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004886030260748096,
      "loss": 5.8357,
      "mean_token_accuracy": 0.16195028424263,
      "num_tokens": 23148600.0,
      "step": 12420
    },
    {
      "entropy": 6.122948122024536,
      "epoch": 1.0972271282232426,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004885931283969845,
      "loss": 5.8164,
      "mean_token_accuracy": 0.16671625077724456,
      "num_tokens": 23156831.0,
      "step": 12425
    },
    {
      "entropy": 6.231880044937133,
      "epoch": 1.097668668315083,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004885832265349915,
      "loss": 5.993,
      "mean_token_accuracy": 0.15441508293151857,
      "num_tokens": 23166453.0,
      "step": 12430
    },
    {
      "entropy": 6.151757001876831,
      "epoch": 1.0981102084069234,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004885733204890246,
      "loss": 5.8598,
      "mean_token_accuracy": 0.15992325395345688,
      "num_tokens": 23175832.0,
      "step": 12435
    },
    {
      "entropy": 6.1306596279144285,
      "epoch": 1.0985517484987637,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004885634102592778,
      "loss": 5.887,
      "mean_token_accuracy": 0.16349006593227386,
      "num_tokens": 23185349.0,
      "step": 12440
    },
    {
      "entropy": 6.144681596755982,
      "epoch": 1.098993288590604,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004885534958459454,
      "loss": 6.0362,
      "mean_token_accuracy": 0.15734102576971054,
      "num_tokens": 23195391.0,
      "step": 12445
    },
    {
      "entropy": 6.112054252624512,
      "epoch": 1.0994348286824445,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004885435772492213,
      "loss": 5.9729,
      "mean_token_accuracy": 0.15919270664453505,
      "num_tokens": 23204689.0,
      "step": 12450
    },
    {
      "entropy": 6.134461545944214,
      "epoch": 1.0998763687742847,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004885336544693,
      "loss": 5.9124,
      "mean_token_accuracy": 0.166460819542408,
      "num_tokens": 23213182.0,
      "step": 12455
    },
    {
      "entropy": 6.083830499649048,
      "epoch": 1.100317908866125,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004885237275063758,
      "loss": 5.913,
      "mean_token_accuracy": 0.1541294515132904,
      "num_tokens": 23222058.0,
      "step": 12460
    },
    {
      "entropy": 6.193151569366455,
      "epoch": 1.1007594489579653,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004885137963606433,
      "loss": 5.9547,
      "mean_token_accuracy": 0.16158088445663452,
      "num_tokens": 23231747.0,
      "step": 12465
    },
    {
      "entropy": 6.175371170043945,
      "epoch": 1.1012009890498058,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004885038610322969,
      "loss": 5.8995,
      "mean_token_accuracy": 0.16137738078832625,
      "num_tokens": 23240402.0,
      "step": 12470
    },
    {
      "entropy": 6.202411127090454,
      "epoch": 1.101642529141646,
      "grad_norm": 2.625,
      "learning_rate": 0.0004884939215215314,
      "loss": 5.9631,
      "mean_token_accuracy": 0.16475873440504074,
      "num_tokens": 23249344.0,
      "step": 12475
    },
    {
      "entropy": 6.184190273284912,
      "epoch": 1.1020840692334863,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004884839778285413,
      "loss": 5.9529,
      "mean_token_accuracy": 0.16150661706924438,
      "num_tokens": 23258157.0,
      "step": 12480
    },
    {
      "entropy": 6.092105579376221,
      "epoch": 1.1025256093253268,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004884740299535216,
      "loss": 5.932,
      "mean_token_accuracy": 0.14843451157212256,
      "num_tokens": 23267798.0,
      "step": 12485
    },
    {
      "entropy": 6.161636066436768,
      "epoch": 1.1029671494171671,
      "grad_norm": 2.25,
      "learning_rate": 0.000488464077896667,
      "loss": 5.9454,
      "mean_token_accuracy": 0.1618686556816101,
      "num_tokens": 23276429.0,
      "step": 12490
    },
    {
      "entropy": 6.1453760147094725,
      "epoch": 1.1034086895090074,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004884541216581724,
      "loss": 5.8573,
      "mean_token_accuracy": 0.15965035557746887,
      "num_tokens": 23285878.0,
      "step": 12495
    },
    {
      "entropy": 6.116464996337891,
      "epoch": 1.1038502296008477,
      "grad_norm": 3.078125,
      "learning_rate": 0.000488444161238233,
      "loss": 5.8777,
      "mean_token_accuracy": 0.16719256043434144,
      "num_tokens": 23295683.0,
      "step": 12500
    },
    {
      "entropy": 6.148839902877808,
      "epoch": 1.1042917696926882,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004884341966370439,
      "loss": 5.9604,
      "mean_token_accuracy": 0.15567479133605958,
      "num_tokens": 23305539.0,
      "step": 12505
    },
    {
      "entropy": 6.147060394287109,
      "epoch": 1.1047333097845284,
      "grad_norm": 3.125,
      "learning_rate": 0.0004884242278548003,
      "loss": 5.9269,
      "mean_token_accuracy": 0.16066510677337648,
      "num_tokens": 23314889.0,
      "step": 12510
    },
    {
      "entropy": 6.170310401916504,
      "epoch": 1.1051748498763687,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004884142548916975,
      "loss": 5.9554,
      "mean_token_accuracy": 0.15886897593736649,
      "num_tokens": 23324425.0,
      "step": 12515
    },
    {
      "entropy": 6.123000907897949,
      "epoch": 1.1056163899682092,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004884042777479308,
      "loss": 5.856,
      "mean_token_accuracy": 0.1647001251578331,
      "num_tokens": 23332994.0,
      "step": 12520
    },
    {
      "entropy": 6.141663312911987,
      "epoch": 1.1060579300600495,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004883942964236957,
      "loss": 5.8317,
      "mean_token_accuracy": 0.16078163534402848,
      "num_tokens": 23342108.0,
      "step": 12525
    },
    {
      "entropy": 6.086635971069336,
      "epoch": 1.1064994701518898,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004883843109191877,
      "loss": 5.8503,
      "mean_token_accuracy": 0.16816559433937073,
      "num_tokens": 23351439.0,
      "step": 12530
    },
    {
      "entropy": 6.159261703491211,
      "epoch": 1.10694101024373,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004883743212346021,
      "loss": 5.9539,
      "mean_token_accuracy": 0.16020542681217192,
      "num_tokens": 23360591.0,
      "step": 12535
    },
    {
      "entropy": 6.167931842803955,
      "epoch": 1.1073825503355705,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004883643273701353,
      "loss": 5.9077,
      "mean_token_accuracy": 0.163707834482193,
      "num_tokens": 23369854.0,
      "step": 12540
    },
    {
      "entropy": 6.166998195648193,
      "epoch": 1.1078240904274108,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004883543293259826,
      "loss": 5.9789,
      "mean_token_accuracy": 0.1555547297000885,
      "num_tokens": 23378071.0,
      "step": 12545
    },
    {
      "entropy": 6.157186269760132,
      "epoch": 1.108265630519251,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004883443271023397,
      "loss": 5.9746,
      "mean_token_accuracy": 0.15041225999593735,
      "num_tokens": 23387083.0,
      "step": 12550
    },
    {
      "entropy": 6.142925882339478,
      "epoch": 1.1087071706110916,
      "grad_norm": 7.125,
      "learning_rate": 0.000488334320699403,
      "loss": 5.7993,
      "mean_token_accuracy": 0.16230424791574477,
      "num_tokens": 23395008.0,
      "step": 12555
    },
    {
      "entropy": 6.18157548904419,
      "epoch": 1.1091487107029319,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004883243101173681,
      "loss": 5.9053,
      "mean_token_accuracy": 0.16364568918943406,
      "num_tokens": 23404193.0,
      "step": 12560
    },
    {
      "entropy": 6.128438568115234,
      "epoch": 1.1095902507947721,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004883142953564315,
      "loss": 5.9004,
      "mean_token_accuracy": 0.16206439658999444,
      "num_tokens": 23413353.0,
      "step": 12565
    },
    {
      "entropy": 6.139471435546875,
      "epoch": 1.1100317908866124,
      "grad_norm": 3.078125,
      "learning_rate": 0.000488304276416789,
      "loss": 5.9472,
      "mean_token_accuracy": 0.16016733050346374,
      "num_tokens": 23422797.0,
      "step": 12570
    },
    {
      "entropy": 6.216736316680908,
      "epoch": 1.110473330978453,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004882942532986372,
      "loss": 6.025,
      "mean_token_accuracy": 0.16250147074460983,
      "num_tokens": 23432705.0,
      "step": 12575
    },
    {
      "entropy": 6.159796333312988,
      "epoch": 1.1109148710702932,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004882842260021721,
      "loss": 5.8848,
      "mean_token_accuracy": 0.1626063108444214,
      "num_tokens": 23441613.0,
      "step": 12580
    },
    {
      "entropy": 6.101108741760254,
      "epoch": 1.1113564111621335,
      "grad_norm": 4.8125,
      "learning_rate": 0.00048827419452759045,
      "loss": 5.9605,
      "mean_token_accuracy": 0.15925377756357192,
      "num_tokens": 23450975.0,
      "step": 12585
    },
    {
      "entropy": 6.157285881042481,
      "epoch": 1.111797951253974,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048826415887508857,
      "loss": 5.9438,
      "mean_token_accuracy": 0.1590849280357361,
      "num_tokens": 23460646.0,
      "step": 12590
    },
    {
      "entropy": 6.33017578125,
      "epoch": 1.1122394913458142,
      "grad_norm": 2.859375,
      "learning_rate": 0.00048825411904486303,
      "loss": 6.0191,
      "mean_token_accuracy": 0.14977509528398514,
      "num_tokens": 23470454.0,
      "step": 12595
    },
    {
      "entropy": 6.1781083106994625,
      "epoch": 1.1126810314376545,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004882440750371108,
      "loss": 5.9155,
      "mean_token_accuracy": 0.16248905658721924,
      "num_tokens": 23480013.0,
      "step": 12600
    },
    {
      "entropy": 6.124111938476562,
      "epoch": 1.1131225715294948,
      "grad_norm": 3.703125,
      "learning_rate": 0.00048823402685202824,
      "loss": 5.8974,
      "mean_token_accuracy": 0.1607903614640236,
      "num_tokens": 23489485.0,
      "step": 12605
    },
    {
      "entropy": 6.042632579803467,
      "epoch": 1.1135641116213353,
      "grad_norm": 2.75,
      "learning_rate": 0.0004882239744898124,
      "loss": 5.8813,
      "mean_token_accuracy": 0.1654872253537178,
      "num_tokens": 23498147.0,
      "step": 12610
    },
    {
      "entropy": 6.011892986297608,
      "epoch": 1.1140056517131756,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004882139179506602,
      "loss": 5.7956,
      "mean_token_accuracy": 0.1666394755244255,
      "num_tokens": 23506614.0,
      "step": 12615
    },
    {
      "entropy": 6.153559684753418,
      "epoch": 1.1144471918050158,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004882038572347685,
      "loss": 6.0483,
      "mean_token_accuracy": 0.1509348526597023,
      "num_tokens": 23514927.0,
      "step": 12620
    },
    {
      "entropy": 6.217209100723267,
      "epoch": 1.1148887318968563,
      "grad_norm": 3.09375,
      "learning_rate": 0.00048819379234233464,
      "loss": 6.0366,
      "mean_token_accuracy": 0.15386761873960494,
      "num_tokens": 23525298.0,
      "step": 12625
    },
    {
      "entropy": 6.3298015117645265,
      "epoch": 1.1153302719886966,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004881837232735556,
      "loss": 5.9483,
      "mean_token_accuracy": 0.15949916541576387,
      "num_tokens": 23533995.0,
      "step": 12630
    },
    {
      "entropy": 6.083431768417358,
      "epoch": 1.1157718120805369,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004881736500286286,
      "loss": 5.9297,
      "mean_token_accuracy": 0.15916224122047423,
      "num_tokens": 23543900.0,
      "step": 12635
    },
    {
      "entropy": 6.156477165222168,
      "epoch": 1.1162133521723772,
      "grad_norm": 3.203125,
      "learning_rate": 0.00048816357260775115,
      "loss": 5.9899,
      "mean_token_accuracy": 0.15882350504398346,
      "num_tokens": 23553790.0,
      "step": 12640
    },
    {
      "entropy": 6.1945747375488285,
      "epoch": 1.1166548922642177,
      "grad_norm": 3.75,
      "learning_rate": 0.0004881534910111204,
      "loss": 5.9494,
      "mean_token_accuracy": 0.16102498918771743,
      "num_tokens": 23563473.0,
      "step": 12645
    },
    {
      "entropy": 6.224649858474732,
      "epoch": 1.117096432356058,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048814340523893406,
      "loss": 5.8307,
      "mean_token_accuracy": 0.1708875447511673,
      "num_tokens": 23572495.0,
      "step": 12650
    },
    {
      "entropy": 6.1610143184661865,
      "epoch": 1.1175379724478982,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048813331529138965,
      "loss": 6.007,
      "mean_token_accuracy": 0.14832639172673226,
      "num_tokens": 23582945.0,
      "step": 12655
    },
    {
      "entropy": 6.175566339492798,
      "epoch": 1.1179795125397387,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004881232211686848,
      "loss": 6.038,
      "mean_token_accuracy": 0.14907224103808403,
      "num_tokens": 23592370.0,
      "step": 12660
    },
    {
      "entropy": 6.194454860687256,
      "epoch": 1.118421052631579,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004881131228710172,
      "loss": 5.9674,
      "mean_token_accuracy": 0.15318025052547454,
      "num_tokens": 23601589.0,
      "step": 12665
    },
    {
      "entropy": 6.215276527404785,
      "epoch": 1.1188625927234193,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004881030203985848,
      "loss": 5.9012,
      "mean_token_accuracy": 0.1597913920879364,
      "num_tokens": 23611456.0,
      "step": 12670
    },
    {
      "entropy": 6.1739644527435305,
      "epoch": 1.1193041328152595,
      "grad_norm": 5.78125,
      "learning_rate": 0.00048809291375158544,
      "loss": 5.9085,
      "mean_token_accuracy": 0.15762727260589598,
      "num_tokens": 23620362.0,
      "step": 12675
    },
    {
      "entropy": 6.175387573242188,
      "epoch": 1.1197456729071,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048808280293021703,
      "loss": 5.9482,
      "mean_token_accuracy": 0.16135822236537933,
      "num_tokens": 23629647.0,
      "step": 12680
    },
    {
      "entropy": 6.173170709609986,
      "epoch": 1.1201872129989403,
      "grad_norm": 3.84375,
      "learning_rate": 0.00048807268793467774,
      "loss": 5.8904,
      "mean_token_accuracy": 0.16001069843769072,
      "num_tokens": 23639201.0,
      "step": 12685
    },
    {
      "entropy": 6.141100263595581,
      "epoch": 1.1206287530907806,
      "grad_norm": 3.328125,
      "learning_rate": 0.00048806256876516565,
      "loss": 5.875,
      "mean_token_accuracy": 0.1625053733587265,
      "num_tokens": 23648166.0,
      "step": 12690
    },
    {
      "entropy": 6.107354784011841,
      "epoch": 1.121070293182621,
      "grad_norm": 4.3125,
      "learning_rate": 0.00048805244542187906,
      "loss": 6.0437,
      "mean_token_accuracy": 0.15732126981019973,
      "num_tokens": 23658478.0,
      "step": 12695
    },
    {
      "entropy": 6.123895168304443,
      "epoch": 1.1215118332744614,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004880423179050161,
      "loss": 5.8395,
      "mean_token_accuracy": 0.1617825448513031,
      "num_tokens": 23667934.0,
      "step": 12700
    },
    {
      "entropy": 6.099366903305054,
      "epoch": 1.1219533733663016,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004880321862147755,
      "loss": 5.8988,
      "mean_token_accuracy": 0.1578386642038822,
      "num_tokens": 23676164.0,
      "step": 12705
    },
    {
      "entropy": 6.167014312744141,
      "epoch": 1.122394913458142,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048802205035135537,
      "loss": 5.9692,
      "mean_token_accuracy": 0.15662605315446854,
      "num_tokens": 23686223.0,
      "step": 12710
    },
    {
      "entropy": 6.179031658172607,
      "epoch": 1.1228364535499824,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048801191031495444,
      "loss": 5.9846,
      "mean_token_accuracy": 0.1548076316714287,
      "num_tokens": 23695043.0,
      "step": 12715
    },
    {
      "entropy": 6.259233999252319,
      "epoch": 1.1232779936418227,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004880017661057714,
      "loss": 5.9884,
      "mean_token_accuracy": 0.15324808657169342,
      "num_tokens": 23704248.0,
      "step": 12720
    },
    {
      "entropy": 6.214432859420777,
      "epoch": 1.123719533733663,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004879916177240049,
      "loss": 5.8863,
      "mean_token_accuracy": 0.15993714332580566,
      "num_tokens": 23713724.0,
      "step": 12725
    },
    {
      "entropy": 6.30100383758545,
      "epoch": 1.1241610738255035,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004879814651698537,
      "loss": 6.0732,
      "mean_token_accuracy": 0.1507812835276127,
      "num_tokens": 23723039.0,
      "step": 12730
    },
    {
      "entropy": 6.16924638748169,
      "epoch": 1.1246026139173437,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048797130844351673,
      "loss": 5.8503,
      "mean_token_accuracy": 0.15733642876148224,
      "num_tokens": 23732024.0,
      "step": 12735
    },
    {
      "entropy": 6.12670316696167,
      "epoch": 1.125044154009184,
      "grad_norm": 4.40625,
      "learning_rate": 0.00048796114754519294,
      "loss": 5.9876,
      "mean_token_accuracy": 0.15579846501350403,
      "num_tokens": 23740739.0,
      "step": 12740
    },
    {
      "entropy": 6.108487462997436,
      "epoch": 1.1254856941010243,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048795098247508145,
      "loss": 5.918,
      "mean_token_accuracy": 0.15626540333032607,
      "num_tokens": 23750405.0,
      "step": 12745
    },
    {
      "entropy": 6.16664080619812,
      "epoch": 1.1259272341928648,
      "grad_norm": 6.8125,
      "learning_rate": 0.0004879408132333813,
      "loss": 5.9109,
      "mean_token_accuracy": 0.16076355725526809,
      "num_tokens": 23759318.0,
      "step": 12750
    },
    {
      "entropy": 6.214537191390991,
      "epoch": 1.126368774284705,
      "grad_norm": 5.4375,
      "learning_rate": 0.00048793063982029166,
      "loss": 6.0286,
      "mean_token_accuracy": 0.15366530269384385,
      "num_tokens": 23769202.0,
      "step": 12755
    },
    {
      "entropy": 6.163922071456909,
      "epoch": 1.1268103143765453,
      "grad_norm": 2.0625,
      "learning_rate": 0.000487920462236012,
      "loss": 5.8883,
      "mean_token_accuracy": 0.1640685483813286,
      "num_tokens": 23778478.0,
      "step": 12760
    },
    {
      "entropy": 6.1662345886230465,
      "epoch": 1.1272518544683856,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004879102804807415,
      "loss": 5.9802,
      "mean_token_accuracy": 0.15960298031568526,
      "num_tokens": 23788044.0,
      "step": 12765
    },
    {
      "entropy": 6.157622671127319,
      "epoch": 1.127693394560226,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048790009455467967,
      "loss": 5.9573,
      "mean_token_accuracy": 0.1509365901350975,
      "num_tokens": 23796351.0,
      "step": 12770
    },
    {
      "entropy": 6.168710517883301,
      "epoch": 1.1281349346520664,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004878899044580261,
      "loss": 5.9644,
      "mean_token_accuracy": 0.1563441962003708,
      "num_tokens": 23805553.0,
      "step": 12775
    },
    {
      "entropy": 6.151319456100464,
      "epoch": 1.1285764747439067,
      "grad_norm": 4.1875,
      "learning_rate": 0.00048787971019098044,
      "loss": 5.8118,
      "mean_token_accuracy": 0.1696744054555893,
      "num_tokens": 23813742.0,
      "step": 12780
    },
    {
      "entropy": 6.161671113967896,
      "epoch": 1.1290180148357472,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004878695117537423,
      "loss": 5.9251,
      "mean_token_accuracy": 0.16025556176900863,
      "num_tokens": 23822135.0,
      "step": 12785
    },
    {
      "entropy": 6.222481203079224,
      "epoch": 1.1294595549275874,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004878593091465114,
      "loss": 5.9833,
      "mean_token_accuracy": 0.16105886101722716,
      "num_tokens": 23832021.0,
      "step": 12790
    },
    {
      "entropy": 6.202785634994507,
      "epoch": 1.1299010950194277,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004878491023694877,
      "loss": 5.9417,
      "mean_token_accuracy": 0.15718482732772826,
      "num_tokens": 23842432.0,
      "step": 12795
    },
    {
      "entropy": 6.217890501022339,
      "epoch": 1.1303426351112682,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004878388914228713,
      "loss": 5.9836,
      "mean_token_accuracy": 0.15357162952423095,
      "num_tokens": 23852172.0,
      "step": 12800
    },
    {
      "entropy": 6.137939214706421,
      "epoch": 1.1307841752031085,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004878286763068619,
      "loss": 5.8557,
      "mean_token_accuracy": 0.16467740535736083,
      "num_tokens": 23861178.0,
      "step": 12805
    },
    {
      "entropy": 6.180500316619873,
      "epoch": 1.1312257152949488,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004878184570216598,
      "loss": 5.9759,
      "mean_token_accuracy": 0.15340608656406401,
      "num_tokens": 23869998.0,
      "step": 12810
    },
    {
      "entropy": 6.15168833732605,
      "epoch": 1.131667255386789,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048780823356746523,
      "loss": 5.9132,
      "mean_token_accuracy": 0.1560283973813057,
      "num_tokens": 23880404.0,
      "step": 12815
    },
    {
      "entropy": 6.174927854537964,
      "epoch": 1.1321087954786295,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004877980059444784,
      "loss": 5.8418,
      "mean_token_accuracy": 0.15855402797460555,
      "num_tokens": 23889643.0,
      "step": 12820
    },
    {
      "entropy": 6.198969888687134,
      "epoch": 1.1325503355704698,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004877877741528996,
      "loss": 5.9169,
      "mean_token_accuracy": 0.16071036607027053,
      "num_tokens": 23899264.0,
      "step": 12825
    },
    {
      "entropy": 6.180214929580688,
      "epoch": 1.13299187566231,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004877775381929294,
      "loss": 5.9635,
      "mean_token_accuracy": 0.15552644431591034,
      "num_tokens": 23908685.0,
      "step": 12830
    },
    {
      "entropy": 6.152384996414185,
      "epoch": 1.1334334157541504,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004877672980647682,
      "loss": 5.9497,
      "mean_token_accuracy": 0.15492501258850097,
      "num_tokens": 23917838.0,
      "step": 12835
    },
    {
      "entropy": 6.194139194488526,
      "epoch": 1.1338749558459909,
      "grad_norm": 3.0625,
      "learning_rate": 0.00048775705376861663,
      "loss": 5.9523,
      "mean_token_accuracy": 0.1608953669667244,
      "num_tokens": 23927498.0,
      "step": 12840
    },
    {
      "entropy": 6.2101281642913815,
      "epoch": 1.1343164959378311,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004877468053046753,
      "loss": 5.9693,
      "mean_token_accuracy": 0.16637535393238068,
      "num_tokens": 23936205.0,
      "step": 12845
    },
    {
      "entropy": 6.113933992385864,
      "epoch": 1.1347580360296714,
      "grad_norm": 2.859375,
      "learning_rate": 0.00048773655267314516,
      "loss": 5.8724,
      "mean_token_accuracy": 0.16375424861907958,
      "num_tokens": 23945897.0,
      "step": 12850
    },
    {
      "entropy": 6.2041229724884035,
      "epoch": 1.135199576121512,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004877262958742268,
      "loss": 5.903,
      "mean_token_accuracy": 0.15940342396497725,
      "num_tokens": 23954873.0,
      "step": 12855
    },
    {
      "entropy": 6.126399993896484,
      "epoch": 1.1356411162133522,
      "grad_norm": 4.71875,
      "learning_rate": 0.00048771603490812144,
      "loss": 6.0093,
      "mean_token_accuracy": 0.15629909709095954,
      "num_tokens": 23964631.0,
      "step": 12860
    },
    {
      "entropy": 6.16305980682373,
      "epoch": 1.1360826563051925,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004877057697750299,
      "loss": 5.8912,
      "mean_token_accuracy": 0.1583813928067684,
      "num_tokens": 23973935.0,
      "step": 12865
    },
    {
      "entropy": 6.210225296020508,
      "epoch": 1.136524196397033,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004876955004751533,
      "loss": 6.0135,
      "mean_token_accuracy": 0.15347768515348434,
      "num_tokens": 23984340.0,
      "step": 12870
    },
    {
      "entropy": 6.267869281768799,
      "epoch": 1.1369657364888732,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048768522700869284,
      "loss": 6.0876,
      "mean_token_accuracy": 0.1525693953037262,
      "num_tokens": 23994387.0,
      "step": 12875
    },
    {
      "entropy": 6.127750015258789,
      "epoch": 1.1374072765807135,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048767494937584964,
      "loss": 5.8996,
      "mean_token_accuracy": 0.1604439526796341,
      "num_tokens": 24003391.0,
      "step": 12880
    },
    {
      "entropy": 6.143089723587036,
      "epoch": 1.1378488166725538,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048766466757682527,
      "loss": 5.8791,
      "mean_token_accuracy": 0.1647971972823143,
      "num_tokens": 24013501.0,
      "step": 12885
    },
    {
      "entropy": 6.130701732635498,
      "epoch": 1.1382903567643943,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004876543816118209,
      "loss": 5.8413,
      "mean_token_accuracy": 0.16390649527311324,
      "num_tokens": 24022675.0,
      "step": 12890
    },
    {
      "entropy": 6.2308478355407715,
      "epoch": 1.1387318968562345,
      "grad_norm": 2.5,
      "learning_rate": 0.00048764409148103815,
      "loss": 6.0174,
      "mean_token_accuracy": 0.15676412284374236,
      "num_tokens": 24032542.0,
      "step": 12895
    },
    {
      "entropy": 6.176291275024414,
      "epoch": 1.1391734369480748,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004876337971846787,
      "loss": 5.8946,
      "mean_token_accuracy": 0.16045114845037461,
      "num_tokens": 24041342.0,
      "step": 12900
    },
    {
      "entropy": 6.171636152267456,
      "epoch": 1.139614977039915,
      "grad_norm": 3.171875,
      "learning_rate": 0.000487623498722944,
      "loss": 5.9422,
      "mean_token_accuracy": 0.16535146534442902,
      "num_tokens": 24051429.0,
      "step": 12905
    },
    {
      "entropy": 6.186244201660156,
      "epoch": 1.1400565171317556,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004876131960960358,
      "loss": 5.917,
      "mean_token_accuracy": 0.1657955154776573,
      "num_tokens": 24061074.0,
      "step": 12910
    },
    {
      "entropy": 6.168979454040527,
      "epoch": 1.1404980572235959,
      "grad_norm": 3.734375,
      "learning_rate": 0.00048760288930415615,
      "loss": 5.9204,
      "mean_token_accuracy": 0.16312191486358643,
      "num_tokens": 24069622.0,
      "step": 12915
    },
    {
      "entropy": 6.14166259765625,
      "epoch": 1.1409395973154361,
      "grad_norm": 2.625,
      "learning_rate": 0.00048759257834750673,
      "loss": 5.9566,
      "mean_token_accuracy": 0.1613275185227394,
      "num_tokens": 24078393.0,
      "step": 12920
    },
    {
      "entropy": 6.1301041603088375,
      "epoch": 1.1413811374072766,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004875822632262897,
      "loss": 5.9338,
      "mean_token_accuracy": 0.1595478229224682,
      "num_tokens": 24089157.0,
      "step": 12925
    },
    {
      "entropy": 6.162030458450317,
      "epoch": 1.141822677499117,
      "grad_norm": 2.5,
      "learning_rate": 0.00048757194394070697,
      "loss": 5.897,
      "mean_token_accuracy": 0.1616758957505226,
      "num_tokens": 24098022.0,
      "step": 12930
    },
    {
      "entropy": 6.127862882614136,
      "epoch": 1.1422642175909572,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048756162049096077,
      "loss": 5.8588,
      "mean_token_accuracy": 0.1631949618458748,
      "num_tokens": 24107629.0,
      "step": 12935
    },
    {
      "entropy": 6.1903746128082275,
      "epoch": 1.1427057576827977,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048755129287725334,
      "loss": 5.9263,
      "mean_token_accuracy": 0.16060164272785188,
      "num_tokens": 24116903.0,
      "step": 12940
    },
    {
      "entropy": 6.098018407821655,
      "epoch": 1.143147297774638,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048754096109978684,
      "loss": 5.8663,
      "mean_token_accuracy": 0.15904103517532348,
      "num_tokens": 24126433.0,
      "step": 12945
    },
    {
      "entropy": 6.237703943252564,
      "epoch": 1.1435888378664782,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004875306251587639,
      "loss": 6.0814,
      "mean_token_accuracy": 0.1499664954841137,
      "num_tokens": 24136603.0,
      "step": 12950
    },
    {
      "entropy": 6.198658990859985,
      "epoch": 1.1440303779583185,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004875202850543869,
      "loss": 5.9421,
      "mean_token_accuracy": 0.1543191336095333,
      "num_tokens": 24145656.0,
      "step": 12955
    },
    {
      "entropy": 6.168276786804199,
      "epoch": 1.144471918050159,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048750994078685837,
      "loss": 5.9868,
      "mean_token_accuracy": 0.1520389199256897,
      "num_tokens": 24155044.0,
      "step": 12960
    },
    {
      "entropy": 6.168300819396973,
      "epoch": 1.1449134581419993,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004874995923563809,
      "loss": 5.9176,
      "mean_token_accuracy": 0.16050723642110826,
      "num_tokens": 24163987.0,
      "step": 12965
    },
    {
      "entropy": 6.210289192199707,
      "epoch": 1.1453549982338396,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004874892397631573,
      "loss": 5.8959,
      "mean_token_accuracy": 0.15463410168886185,
      "num_tokens": 24173542.0,
      "step": 12970
    },
    {
      "entropy": 6.0940813541412355,
      "epoch": 1.1457965383256798,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004874788830073904,
      "loss": 5.8971,
      "mean_token_accuracy": 0.15726284980773925,
      "num_tokens": 24182785.0,
      "step": 12975
    },
    {
      "entropy": 6.151674413681031,
      "epoch": 1.1462380784175203,
      "grad_norm": 2.96875,
      "learning_rate": 0.000487468522089283,
      "loss": 5.9204,
      "mean_token_accuracy": 0.15491933822631837,
      "num_tokens": 24193477.0,
      "step": 12980
    },
    {
      "entropy": 6.147600507736206,
      "epoch": 1.1466796185093606,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048745815700903806,
      "loss": 5.9047,
      "mean_token_accuracy": 0.1584320306777954,
      "num_tokens": 24202289.0,
      "step": 12985
    },
    {
      "entropy": 6.048496198654175,
      "epoch": 1.147121158601201,
      "grad_norm": 7.4375,
      "learning_rate": 0.00048744778776685865,
      "loss": 5.8398,
      "mean_token_accuracy": 0.16988199353218078,
      "num_tokens": 24212199.0,
      "step": 12990
    },
    {
      "entropy": 6.139628458023071,
      "epoch": 1.1475626986930414,
      "grad_norm": 3.28125,
      "learning_rate": 0.00048743741436294794,
      "loss": 5.9561,
      "mean_token_accuracy": 0.16567686945199966,
      "num_tokens": 24220984.0,
      "step": 12995
    },
    {
      "entropy": 6.161849308013916,
      "epoch": 1.1480042387848817,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004874270367975091,
      "loss": 5.9217,
      "mean_token_accuracy": 0.1658036917448044,
      "num_tokens": 24229599.0,
      "step": 13000
    },
    {
      "entropy": 6.172030305862426,
      "epoch": 1.148445778876722,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004874166550707454,
      "loss": 5.9468,
      "mean_token_accuracy": 0.15842569917440413,
      "num_tokens": 24238721.0,
      "step": 13005
    },
    {
      "entropy": 6.188222360610962,
      "epoch": 1.1488873189685624,
      "grad_norm": 2.5,
      "learning_rate": 0.00048740626918286026,
      "loss": 5.9527,
      "mean_token_accuracy": 0.15307596549391747,
      "num_tokens": 24248431.0,
      "step": 13010
    },
    {
      "entropy": 6.1717675685882565,
      "epoch": 1.1493288590604027,
      "grad_norm": 3.609375,
      "learning_rate": 0.00048739587913405706,
      "loss": 5.9031,
      "mean_token_accuracy": 0.16315290853381156,
      "num_tokens": 24257770.0,
      "step": 13015
    },
    {
      "entropy": 6.195352554321289,
      "epoch": 1.149770399152243,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048738548492453944,
      "loss": 5.9111,
      "mean_token_accuracy": 0.15852010399103164,
      "num_tokens": 24266402.0,
      "step": 13020
    },
    {
      "entropy": 6.182618618011475,
      "epoch": 1.1502119392440833,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004873750865545109,
      "loss": 5.9073,
      "mean_token_accuracy": 0.15980733186006546,
      "num_tokens": 24275164.0,
      "step": 13025
    },
    {
      "entropy": 6.1739013671875,
      "epoch": 1.1506534793359238,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048736468402417525,
      "loss": 5.9877,
      "mean_token_accuracy": 0.15904164016246797,
      "num_tokens": 24285650.0,
      "step": 13030
    },
    {
      "entropy": 6.1579667091369625,
      "epoch": 1.151095019427764,
      "grad_norm": 2.71875,
      "learning_rate": 0.00048735427733373617,
      "loss": 5.9313,
      "mean_token_accuracy": 0.1569092810153961,
      "num_tokens": 24294745.0,
      "step": 13035
    },
    {
      "entropy": 6.212623643875122,
      "epoch": 1.1515365595196043,
      "grad_norm": 2.5,
      "learning_rate": 0.00048734386648339767,
      "loss": 5.9126,
      "mean_token_accuracy": 0.1645822435617447,
      "num_tokens": 24303837.0,
      "step": 13040
    },
    {
      "entropy": 6.121044731140136,
      "epoch": 1.1519780996114446,
      "grad_norm": 2.625,
      "learning_rate": 0.0004873334514733634,
      "loss": 5.9815,
      "mean_token_accuracy": 0.15832275301218032,
      "num_tokens": 24313417.0,
      "step": 13045
    },
    {
      "entropy": 6.228082180023193,
      "epoch": 1.152419639703285,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004873230323038378,
      "loss": 5.8974,
      "mean_token_accuracy": 0.16411705017089845,
      "num_tokens": 24322439.0,
      "step": 13050
    },
    {
      "entropy": 6.249577903747559,
      "epoch": 1.1528611797951254,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004873126089750246,
      "loss": 5.9566,
      "mean_token_accuracy": 0.16161733716726304,
      "num_tokens": 24332509.0,
      "step": 13055
    },
    {
      "entropy": 6.156072044372559,
      "epoch": 1.1533027198869656,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004873021814871282,
      "loss": 5.9777,
      "mean_token_accuracy": 0.1539487287402153,
      "num_tokens": 24342252.0,
      "step": 13060
    },
    {
      "entropy": 6.118894624710083,
      "epoch": 1.1537442599788061,
      "grad_norm": 9.1875,
      "learning_rate": 0.0004872917498403528,
      "loss": 5.9761,
      "mean_token_accuracy": 0.1629844516515732,
      "num_tokens": 24352459.0,
      "step": 13065
    },
    {
      "entropy": 6.198663759231567,
      "epoch": 1.1541858000706464,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004872813140349028,
      "loss": 5.8607,
      "mean_token_accuracy": 0.16332223117351533,
      "num_tokens": 24361788.0,
      "step": 13070
    },
    {
      "entropy": 6.217107486724854,
      "epoch": 1.1546273401624867,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004872708740709825,
      "loss": 5.9483,
      "mean_token_accuracy": 0.15652839541435243,
      "num_tokens": 24372117.0,
      "step": 13075
    },
    {
      "entropy": 6.111956453323364,
      "epoch": 1.1550688802543272,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048726042994879656,
      "loss": 5.8875,
      "mean_token_accuracy": 0.16594249457120896,
      "num_tokens": 24380833.0,
      "step": 13080
    },
    {
      "entropy": 6.110033416748047,
      "epoch": 1.1555104203461675,
      "grad_norm": 3.09375,
      "learning_rate": 0.00048724998166854944,
      "loss": 5.9825,
      "mean_token_accuracy": 0.16378928273916243,
      "num_tokens": 24390258.0,
      "step": 13085
    },
    {
      "entropy": 6.196791553497315,
      "epoch": 1.1559519604380077,
      "grad_norm": 1.96875,
      "learning_rate": 0.000487239529230446,
      "loss": 5.9789,
      "mean_token_accuracy": 0.15774571001529694,
      "num_tokens": 24399359.0,
      "step": 13090
    },
    {
      "entropy": 6.209712028503418,
      "epoch": 1.156393500529848,
      "grad_norm": 15.875,
      "learning_rate": 0.00048722907263469084,
      "loss": 5.9585,
      "mean_token_accuracy": 0.16555930078029632,
      "num_tokens": 24408012.0,
      "step": 13095
    },
    {
      "entropy": 6.235305643081665,
      "epoch": 1.1568350406216885,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004872186118814889,
      "loss": 5.9484,
      "mean_token_accuracy": 0.15444328933954238,
      "num_tokens": 24416804.0,
      "step": 13100
    },
    {
      "entropy": 6.080541563034058,
      "epoch": 1.1572765807135288,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048720814697104497,
      "loss": 5.7509,
      "mean_token_accuracy": 0.1696800172328949,
      "num_tokens": 24425401.0,
      "step": 13105
    },
    {
      "entropy": 6.037707424163818,
      "epoch": 1.157718120805369,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048719767790356415,
      "loss": 5.826,
      "mean_token_accuracy": 0.16340615302324296,
      "num_tokens": 24434473.0,
      "step": 13110
    },
    {
      "entropy": 6.227966976165772,
      "epoch": 1.1581596608972093,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048718720467925156,
      "loss": 6.043,
      "mean_token_accuracy": 0.14917874932289124,
      "num_tokens": 24443154.0,
      "step": 13115
    },
    {
      "entropy": 6.215178680419922,
      "epoch": 1.1586012009890498,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004871767272983122,
      "loss": 5.8801,
      "mean_token_accuracy": 0.1701144054532051,
      "num_tokens": 24451308.0,
      "step": 13120
    },
    {
      "entropy": 6.184889650344848,
      "epoch": 1.1590427410808901,
      "grad_norm": 6.21875,
      "learning_rate": 0.0004871662457609515,
      "loss": 5.8974,
      "mean_token_accuracy": 0.16406641006469727,
      "num_tokens": 24459997.0,
      "step": 13125
    },
    {
      "entropy": 6.141274261474609,
      "epoch": 1.1594842811727304,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004871557600673747,
      "loss": 5.9277,
      "mean_token_accuracy": 0.16226707100868226,
      "num_tokens": 24469448.0,
      "step": 13130
    },
    {
      "entropy": 6.129578638076782,
      "epoch": 1.1599258212645709,
      "grad_norm": 4.21875,
      "learning_rate": 0.00048714527021778714,
      "loss": 5.9128,
      "mean_token_accuracy": 0.160499732196331,
      "num_tokens": 24478304.0,
      "step": 13135
    },
    {
      "entropy": 6.115665864944458,
      "epoch": 1.1603673613564112,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004871347762123945,
      "loss": 5.8106,
      "mean_token_accuracy": 0.1724348247051239,
      "num_tokens": 24487300.0,
      "step": 13140
    },
    {
      "entropy": 6.113453578948975,
      "epoch": 1.1608089014482514,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048712427805140224,
      "loss": 5.901,
      "mean_token_accuracy": 0.16196317970752716,
      "num_tokens": 24495893.0,
      "step": 13145
    },
    {
      "entropy": 6.1374163150787355,
      "epoch": 1.161250441540092,
      "grad_norm": 2.5,
      "learning_rate": 0.0004871137757350159,
      "loss": 5.8662,
      "mean_token_accuracy": 0.16316550970077515,
      "num_tokens": 24504603.0,
      "step": 13150
    },
    {
      "entropy": 6.122546243667602,
      "epoch": 1.1616919816319322,
      "grad_norm": 4.28125,
      "learning_rate": 0.00048710326926344153,
      "loss": 5.8635,
      "mean_token_accuracy": 0.16920806989073753,
      "num_tokens": 24513148.0,
      "step": 13155
    },
    {
      "entropy": 6.23073000907898,
      "epoch": 1.1621335217237725,
      "grad_norm": 3.203125,
      "learning_rate": 0.00048709275863688455,
      "loss": 6.0072,
      "mean_token_accuracy": 0.15268441587686538,
      "num_tokens": 24524006.0,
      "step": 13160
    },
    {
      "entropy": 6.177923154830933,
      "epoch": 1.1625750618156128,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048708224385555116,
      "loss": 5.8387,
      "mean_token_accuracy": 0.17196518182754517,
      "num_tokens": 24532468.0,
      "step": 13165
    },
    {
      "entropy": 6.164284515380859,
      "epoch": 1.1630166019074533,
      "grad_norm": 2.25,
      "learning_rate": 0.0004870717249196472,
      "loss": 5.9401,
      "mean_token_accuracy": 0.15457817912101746,
      "num_tokens": 24540887.0,
      "step": 13170
    },
    {
      "entropy": 6.185343647003174,
      "epoch": 1.1634581419992935,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004870612018293788,
      "loss": 5.879,
      "mean_token_accuracy": 0.1544166013598442,
      "num_tokens": 24550565.0,
      "step": 13175
    },
    {
      "entropy": 6.209562587738037,
      "epoch": 1.1638996820911338,
      "grad_norm": 2.953125,
      "learning_rate": 0.000487050674584952,
      "loss": 5.8445,
      "mean_token_accuracy": 0.16708905547857283,
      "num_tokens": 24560283.0,
      "step": 13180
    },
    {
      "entropy": 6.11170654296875,
      "epoch": 1.164341222182974,
      "grad_norm": 3.203125,
      "learning_rate": 0.000487040143186573,
      "loss": 5.8311,
      "mean_token_accuracy": 0.16198592782020568,
      "num_tokens": 24569527.0,
      "step": 13185
    },
    {
      "entropy": 6.0781200408935545,
      "epoch": 1.1647827622748146,
      "grad_norm": 2.375,
      "learning_rate": 0.00048702960763444833,
      "loss": 5.9204,
      "mean_token_accuracy": 0.1569120854139328,
      "num_tokens": 24578702.0,
      "step": 13190
    },
    {
      "entropy": 6.204309177398682,
      "epoch": 1.1652243023666549,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004870190679287842,
      "loss": 5.8865,
      "mean_token_accuracy": 0.16081749945878981,
      "num_tokens": 24588252.0,
      "step": 13195
    },
    {
      "entropy": 6.1760461807250975,
      "epoch": 1.1656658424584951,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004870085240697871,
      "loss": 5.8617,
      "mean_token_accuracy": 0.1601746678352356,
      "num_tokens": 24596989.0,
      "step": 13200
    },
    {
      "entropy": 6.150515747070313,
      "epoch": 1.1661073825503356,
      "grad_norm": 3.875,
      "learning_rate": 0.00048699797605766365,
      "loss": 5.9995,
      "mean_token_accuracy": 0.15753665268421174,
      "num_tokens": 24607051.0,
      "step": 13205
    },
    {
      "entropy": 6.2017967224121096,
      "epoch": 1.166548922642176,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048698742389262036,
      "loss": 5.9867,
      "mean_token_accuracy": 0.15080763548612594,
      "num_tokens": 24616561.0,
      "step": 13210
    },
    {
      "entropy": 6.182310962677002,
      "epoch": 1.1669904627340162,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048697686757486397,
      "loss": 5.9579,
      "mean_token_accuracy": 0.15699755102396012,
      "num_tokens": 24625942.0,
      "step": 13215
    },
    {
      "entropy": 6.223864126205444,
      "epoch": 1.1674320028258567,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004869663071046013,
      "loss": 5.9254,
      "mean_token_accuracy": 0.15231055542826652,
      "num_tokens": 24635646.0,
      "step": 13220
    },
    {
      "entropy": 6.083964204788208,
      "epoch": 1.167873542917697,
      "grad_norm": 3.75,
      "learning_rate": 0.00048695574248203935,
      "loss": 5.9577,
      "mean_token_accuracy": 0.15992813110351561,
      "num_tokens": 24645147.0,
      "step": 13225
    },
    {
      "entropy": 6.091044330596924,
      "epoch": 1.1683150830095372,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004869451737073848,
      "loss": 5.9298,
      "mean_token_accuracy": 0.15731989741325378,
      "num_tokens": 24654796.0,
      "step": 13230
    },
    {
      "entropy": 6.1002613544464115,
      "epoch": 1.1687566231013775,
      "grad_norm": 2.375,
      "learning_rate": 0.0004869346007808449,
      "loss": 5.8666,
      "mean_token_accuracy": 0.16188845038414001,
      "num_tokens": 24663765.0,
      "step": 13235
    },
    {
      "entropy": 6.262855482101441,
      "epoch": 1.169198163193218,
      "grad_norm": 3.765625,
      "learning_rate": 0.00048692402370262674,
      "loss": 5.9849,
      "mean_token_accuracy": 0.1545178234577179,
      "num_tokens": 24673758.0,
      "step": 13240
    },
    {
      "entropy": 6.221972465515137,
      "epoch": 1.1696397032850583,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048691344247293747,
      "loss": 5.8467,
      "mean_token_accuracy": 0.16410696357488633,
      "num_tokens": 24682751.0,
      "step": 13245
    },
    {
      "entropy": 6.158407592773438,
      "epoch": 1.1700812433768986,
      "grad_norm": 2.25,
      "learning_rate": 0.00048690285709198435,
      "loss": 5.9597,
      "mean_token_accuracy": 0.158477720618248,
      "num_tokens": 24691375.0,
      "step": 13250
    },
    {
      "entropy": 6.144349384307861,
      "epoch": 1.1705227834687388,
      "grad_norm": 3.25,
      "learning_rate": 0.0004868922675599747,
      "loss": 5.9008,
      "mean_token_accuracy": 0.16543617844581604,
      "num_tokens": 24699852.0,
      "step": 13255
    },
    {
      "entropy": 6.151238584518433,
      "epoch": 1.1709643235605793,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004868816738771162,
      "loss": 5.8205,
      "mean_token_accuracy": 0.1628445565700531,
      "num_tokens": 24708857.0,
      "step": 13260
    },
    {
      "entropy": 6.1484630584716795,
      "epoch": 1.1714058636524196,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004868710760436161,
      "loss": 5.978,
      "mean_token_accuracy": 0.15893639922142028,
      "num_tokens": 24718262.0,
      "step": 13265
    },
    {
      "entropy": 6.137684535980225,
      "epoch": 1.1718474037442599,
      "grad_norm": 3.390625,
      "learning_rate": 0.00048686047405968217,
      "loss": 5.9552,
      "mean_token_accuracy": 0.16079557538032532,
      "num_tokens": 24727458.0,
      "step": 13270
    },
    {
      "entropy": 6.207680511474609,
      "epoch": 1.1722889438361004,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048684986792552206,
      "loss": 6.0053,
      "mean_token_accuracy": 0.15345407724380494,
      "num_tokens": 24737297.0,
      "step": 13275
    },
    {
      "entropy": 6.25681734085083,
      "epoch": 1.1727304839279407,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048683925764134346,
      "loss": 5.8994,
      "mean_token_accuracy": 0.16747342348098754,
      "num_tokens": 24746939.0,
      "step": 13280
    },
    {
      "entropy": 6.186113262176514,
      "epoch": 1.173172024019781,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004868286432073543,
      "loss": 6.0444,
      "mean_token_accuracy": 0.1579015001654625,
      "num_tokens": 24756907.0,
      "step": 13285
    },
    {
      "entropy": 6.110030841827393,
      "epoch": 1.1736135641116214,
      "grad_norm": 5.5,
      "learning_rate": 0.00048681802462376253,
      "loss": 5.846,
      "mean_token_accuracy": 0.16282231509685516,
      "num_tokens": 24765997.0,
      "step": 13290
    },
    {
      "entropy": 6.228509187698364,
      "epoch": 1.1740551042034617,
      "grad_norm": 4.0625,
      "learning_rate": 0.00048680740189077605,
      "loss": 5.9229,
      "mean_token_accuracy": 0.15141807049512862,
      "num_tokens": 24775090.0,
      "step": 13295
    },
    {
      "entropy": 6.174625825881958,
      "epoch": 1.174496644295302,
      "grad_norm": 2.625,
      "learning_rate": 0.00048679677500860314,
      "loss": 5.9026,
      "mean_token_accuracy": 0.16119746267795562,
      "num_tokens": 24784709.0,
      "step": 13300
    },
    {
      "entropy": 6.20410304069519,
      "epoch": 1.1749381843871423,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004867861439774519,
      "loss": 5.9244,
      "mean_token_accuracy": 0.1608407825231552,
      "num_tokens": 24794473.0,
      "step": 13305
    },
    {
      "entropy": 6.1915966987609865,
      "epoch": 1.1753797244789828,
      "grad_norm": 2.5,
      "learning_rate": 0.00048677550879753037,
      "loss": 6.0279,
      "mean_token_accuracy": 0.1547059342265129,
      "num_tokens": 24804718.0,
      "step": 13310
    },
    {
      "entropy": 6.1425261974334715,
      "epoch": 1.175821264570823,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048676486946904715,
      "loss": 5.9228,
      "mean_token_accuracy": 0.15841422826051713,
      "num_tokens": 24813632.0,
      "step": 13315
    },
    {
      "entropy": 6.083115434646606,
      "epoch": 1.1762628046626633,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004867542259922105,
      "loss": 5.8547,
      "mean_token_accuracy": 0.1671317309141159,
      "num_tokens": 24823395.0,
      "step": 13320
    },
    {
      "entropy": 6.2250324249267575,
      "epoch": 1.1767043447545038,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004867435783672291,
      "loss": 5.9713,
      "mean_token_accuracy": 0.15184925198554994,
      "num_tokens": 24832869.0,
      "step": 13325
    },
    {
      "entropy": 6.231447219848633,
      "epoch": 1.177145884846344,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004867329265943114,
      "loss": 5.8908,
      "mean_token_accuracy": 0.1585858106613159,
      "num_tokens": 24841706.0,
      "step": 13330
    },
    {
      "entropy": 6.206498670578003,
      "epoch": 1.1775874249381844,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004867222706736661,
      "loss": 5.9315,
      "mean_token_accuracy": 0.1613654911518097,
      "num_tokens": 24851149.0,
      "step": 13335
    },
    {
      "entropy": 6.197022247314453,
      "epoch": 1.1780289650300246,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004867116106055019,
      "loss": 5.8945,
      "mean_token_accuracy": 0.16297131925821304,
      "num_tokens": 24860002.0,
      "step": 13340
    },
    {
      "entropy": 6.128525114059448,
      "epoch": 1.1784705051218651,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048670094639002766,
      "loss": 5.986,
      "mean_token_accuracy": 0.15572923123836518,
      "num_tokens": 24869457.0,
      "step": 13345
    },
    {
      "entropy": 6.04019341468811,
      "epoch": 1.1789120452137054,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004866902780274523,
      "loss": 5.9126,
      "mean_token_accuracy": 0.1607941582798958,
      "num_tokens": 24879527.0,
      "step": 13350
    },
    {
      "entropy": 6.146878337860107,
      "epoch": 1.1793535853055457,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004866796055179847,
      "loss": 5.933,
      "mean_token_accuracy": 0.16120410561561585,
      "num_tokens": 24889401.0,
      "step": 13355
    },
    {
      "entropy": 6.216059970855713,
      "epoch": 1.1797951253973862,
      "grad_norm": 2.625,
      "learning_rate": 0.0004866689288618342,
      "loss": 5.9333,
      "mean_token_accuracy": 0.15970335453748702,
      "num_tokens": 24897911.0,
      "step": 13360
    },
    {
      "entropy": 6.1216864585876465,
      "epoch": 1.1802366654892265,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004866582480592096,
      "loss": 5.8886,
      "mean_token_accuracy": 0.1663259521126747,
      "num_tokens": 24906969.0,
      "step": 13365
    },
    {
      "entropy": 6.1119410514831545,
      "epoch": 1.1806782055810667,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004866475631103203,
      "loss": 5.9225,
      "mean_token_accuracy": 0.16083779484033583,
      "num_tokens": 24915713.0,
      "step": 13370
    },
    {
      "entropy": 6.141808748245239,
      "epoch": 1.181119745672907,
      "grad_norm": 3.4375,
      "learning_rate": 0.00048663687401537573,
      "loss": 5.9279,
      "mean_token_accuracy": 0.15738131999969482,
      "num_tokens": 24925484.0,
      "step": 13375
    },
    {
      "entropy": 6.169105577468872,
      "epoch": 1.1815612857647475,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048662618077458507,
      "loss": 5.9477,
      "mean_token_accuracy": 0.1669459268450737,
      "num_tokens": 24934751.0,
      "step": 13380
    },
    {
      "entropy": 6.197234058380127,
      "epoch": 1.1820028258565878,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004866154833881579,
      "loss": 5.9893,
      "mean_token_accuracy": 0.15621086210012436,
      "num_tokens": 24945059.0,
      "step": 13385
    },
    {
      "entropy": 6.202260875701905,
      "epoch": 1.182444365948428,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004866047818563038,
      "loss": 5.9611,
      "mean_token_accuracy": 0.15845003873109817,
      "num_tokens": 24954678.0,
      "step": 13390
    },
    {
      "entropy": 6.127795648574829,
      "epoch": 1.1828859060402686,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004865940761792323,
      "loss": 5.9429,
      "mean_token_accuracy": 0.15084182769060134,
      "num_tokens": 24964093.0,
      "step": 13395
    },
    {
      "entropy": 6.235414981842041,
      "epoch": 1.1833274461321088,
      "grad_norm": 2.5,
      "learning_rate": 0.0004865833663571533,
      "loss": 5.8759,
      "mean_token_accuracy": 0.16595400273799896,
      "num_tokens": 24973903.0,
      "step": 13400
    },
    {
      "entropy": 6.107465744018555,
      "epoch": 1.183768986223949,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048657265239027635,
      "loss": 5.8899,
      "mean_token_accuracy": 0.160612852871418,
      "num_tokens": 24982458.0,
      "step": 13405
    },
    {
      "entropy": 6.080062198638916,
      "epoch": 1.1842105263157894,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004865619342788116,
      "loss": 5.944,
      "mean_token_accuracy": 0.157385890185833,
      "num_tokens": 24992233.0,
      "step": 13410
    },
    {
      "entropy": 6.184826898574829,
      "epoch": 1.1846520664076299,
      "grad_norm": 3.5625,
      "learning_rate": 0.00048655121202296873,
      "loss": 5.8856,
      "mean_token_accuracy": 0.15693662613630294,
      "num_tokens": 25001855.0,
      "step": 13415
    },
    {
      "entropy": 6.161380243301392,
      "epoch": 1.1850936064994702,
      "grad_norm": 1.890625,
      "learning_rate": 0.00048654048562295807,
      "loss": 5.7895,
      "mean_token_accuracy": 0.16418375819921494,
      "num_tokens": 25010950.0,
      "step": 13420
    },
    {
      "entropy": 6.184538125991821,
      "epoch": 1.1855351465913104,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004865297550789895,
      "loss": 5.979,
      "mean_token_accuracy": 0.15356025993824005,
      "num_tokens": 25020816.0,
      "step": 13425
    },
    {
      "entropy": 6.228384256362915,
      "epoch": 1.185976686683151,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004865190203912734,
      "loss": 5.9976,
      "mean_token_accuracy": 0.15983387678861619,
      "num_tokens": 25030236.0,
      "step": 13430
    },
    {
      "entropy": 6.137870597839355,
      "epoch": 1.1864182267749912,
      "grad_norm": 2.484375,
      "learning_rate": 0.00048650828156001995,
      "loss": 5.8861,
      "mean_token_accuracy": 0.16400271505117417,
      "num_tokens": 25040236.0,
      "step": 13435
    },
    {
      "entropy": 6.173695373535156,
      "epoch": 1.1868597668668315,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004864975385854396,
      "loss": 6.007,
      "mean_token_accuracy": 0.15775451213121414,
      "num_tokens": 25049835.0,
      "step": 13440
    },
    {
      "entropy": 6.154194355010986,
      "epoch": 1.1873013069586718,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048648679146774275,
      "loss": 5.9501,
      "mean_token_accuracy": 0.1562986582517624,
      "num_tokens": 25058947.0,
      "step": 13445
    },
    {
      "entropy": 6.193396091461182,
      "epoch": 1.1877428470505123,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004864760402071399,
      "loss": 5.9294,
      "mean_token_accuracy": 0.1595063403248787,
      "num_tokens": 25069128.0,
      "step": 13450
    },
    {
      "entropy": 6.186780643463135,
      "epoch": 1.1881843871423525,
      "grad_norm": 2.859375,
      "learning_rate": 0.00048646528480384175,
      "loss": 5.9879,
      "mean_token_accuracy": 0.1523025467991829,
      "num_tokens": 25079717.0,
      "step": 13455
    },
    {
      "entropy": 6.118858861923218,
      "epoch": 1.1886259272341928,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004864545252580588,
      "loss": 5.9238,
      "mean_token_accuracy": 0.15855868086218833,
      "num_tokens": 25088656.0,
      "step": 13460
    },
    {
      "entropy": 6.136055564880371,
      "epoch": 1.1890674673260333,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048644376157000213,
      "loss": 5.926,
      "mean_token_accuracy": 0.15793594121932983,
      "num_tokens": 25098066.0,
      "step": 13465
    },
    {
      "entropy": 6.149987506866455,
      "epoch": 1.1895090074178736,
      "grad_norm": 3.0,
      "learning_rate": 0.0004864329937398823,
      "loss": 5.9553,
      "mean_token_accuracy": 0.15770915821194648,
      "num_tokens": 25107530.0,
      "step": 13470
    },
    {
      "entropy": 6.171546888351441,
      "epoch": 1.1899505475097139,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048642222176791043,
      "loss": 5.8863,
      "mean_token_accuracy": 0.15822558850049973,
      "num_tokens": 25116226.0,
      "step": 13475
    },
    {
      "entropy": 6.150636959075928,
      "epoch": 1.1903920876015541,
      "grad_norm": 2.25,
      "learning_rate": 0.0004864114456542975,
      "loss": 6.0269,
      "mean_token_accuracy": 0.14910721331834792,
      "num_tokens": 25125619.0,
      "step": 13480
    },
    {
      "entropy": 6.1670585632324215,
      "epoch": 1.1908336276933946,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004864006653992546,
      "loss": 5.8844,
      "mean_token_accuracy": 0.1635897010564804,
      "num_tokens": 25134782.0,
      "step": 13485
    },
    {
      "entropy": 6.239220666885376,
      "epoch": 1.191275167785235,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048638988100299283,
      "loss": 5.9466,
      "mean_token_accuracy": 0.1643083930015564,
      "num_tokens": 25144709.0,
      "step": 13490
    },
    {
      "entropy": 6.202182865142822,
      "epoch": 1.1917167078770752,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004863790924657236,
      "loss": 5.8333,
      "mean_token_accuracy": 0.1616229996085167,
      "num_tokens": 25153771.0,
      "step": 13495
    },
    {
      "entropy": 6.113958311080933,
      "epoch": 1.1921582479689157,
      "grad_norm": 3.046875,
      "learning_rate": 0.000486368299787658,
      "loss": 5.962,
      "mean_token_accuracy": 0.16108824014663697,
      "num_tokens": 25163217.0,
      "step": 13500
    },
    {
      "entropy": 6.113112735748291,
      "epoch": 1.192599788060756,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048635750296900784,
      "loss": 5.9725,
      "mean_token_accuracy": 0.15988246351480484,
      "num_tokens": 25172762.0,
      "step": 13505
    },
    {
      "entropy": 6.173691082000732,
      "epoch": 1.1930413281525962,
      "grad_norm": 2.125,
      "learning_rate": 0.0004863467020099842,
      "loss": 5.8777,
      "mean_token_accuracy": 0.16410644203424454,
      "num_tokens": 25182349.0,
      "step": 13510
    },
    {
      "entropy": 6.234494829177857,
      "epoch": 1.1934828682444365,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048633589691079895,
      "loss": 5.9839,
      "mean_token_accuracy": 0.15102794617414475,
      "num_tokens": 25191971.0,
      "step": 13515
    },
    {
      "entropy": 6.113103628158569,
      "epoch": 1.193924408336277,
      "grad_norm": 2.125,
      "learning_rate": 0.0004863250876716637,
      "loss": 5.7549,
      "mean_token_accuracy": 0.16959406286478043,
      "num_tokens": 25200476.0,
      "step": 13520
    },
    {
      "entropy": 6.056025171279908,
      "epoch": 1.1943659484281173,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004863142742927902,
      "loss": 5.8709,
      "mean_token_accuracy": 0.16417582482099533,
      "num_tokens": 25209748.0,
      "step": 13525
    },
    {
      "entropy": 6.177553796768189,
      "epoch": 1.1948074885199576,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004863034567743901,
      "loss": 5.9467,
      "mean_token_accuracy": 0.15654509365558625,
      "num_tokens": 25220140.0,
      "step": 13530
    },
    {
      "entropy": 6.11725435256958,
      "epoch": 1.195249028611798,
      "grad_norm": 2.0,
      "learning_rate": 0.0004862926351166756,
      "loss": 5.8271,
      "mean_token_accuracy": 0.1715938687324524,
      "num_tokens": 25228424.0,
      "step": 13535
    },
    {
      "entropy": 6.212284183502197,
      "epoch": 1.1956905687036383,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004862818093198585,
      "loss": 5.95,
      "mean_token_accuracy": 0.1611993908882141,
      "num_tokens": 25237448.0,
      "step": 13540
    },
    {
      "entropy": 6.115144205093384,
      "epoch": 1.1961321087954786,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004862709793841509,
      "loss": 5.8446,
      "mean_token_accuracy": 0.1682332143187523,
      "num_tokens": 25247124.0,
      "step": 13545
    },
    {
      "entropy": 6.005498456954956,
      "epoch": 1.1965736488873189,
      "grad_norm": 2.375,
      "learning_rate": 0.00048626014530976495,
      "loss": 5.8935,
      "mean_token_accuracy": 0.16573696881532668,
      "num_tokens": 25257363.0,
      "step": 13550
    },
    {
      "entropy": 6.1136373519897464,
      "epoch": 1.1970151889791594,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048624930709691296,
      "loss": 5.9502,
      "mean_token_accuracy": 0.15379948765039445,
      "num_tokens": 25266837.0,
      "step": 13555
    },
    {
      "entropy": 6.195064306259155,
      "epoch": 1.1974567290709996,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048623846474580716,
      "loss": 6.0006,
      "mean_token_accuracy": 0.15028007179498673,
      "num_tokens": 25275767.0,
      "step": 13560
    },
    {
      "entropy": 6.231287050247192,
      "epoch": 1.19789826916284,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048622761825665994,
      "loss": 5.9334,
      "mean_token_accuracy": 0.15911966562271118,
      "num_tokens": 25285212.0,
      "step": 13565
    },
    {
      "entropy": 6.095887994766235,
      "epoch": 1.1983398092546804,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004862167676296839,
      "loss": 5.8442,
      "mean_token_accuracy": 0.16009492799639702,
      "num_tokens": 25294698.0,
      "step": 13570
    },
    {
      "entropy": 6.059245014190674,
      "epoch": 1.1987813493465207,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004862059128650915,
      "loss": 5.9115,
      "mean_token_accuracy": 0.16499557793140412,
      "num_tokens": 25304343.0,
      "step": 13575
    },
    {
      "entropy": 6.1906756401062015,
      "epoch": 1.199222889438361,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048619505396309525,
      "loss": 5.8242,
      "mean_token_accuracy": 0.15956140011548997,
      "num_tokens": 25312542.0,
      "step": 13580
    },
    {
      "entropy": 6.173790645599365,
      "epoch": 1.1996644295302012,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004861841909239081,
      "loss": 5.9449,
      "mean_token_accuracy": 0.16140115708112718,
      "num_tokens": 25322665.0,
      "step": 13585
    },
    {
      "entropy": 6.129951047897339,
      "epoch": 1.2001059696220417,
      "grad_norm": 5.84375,
      "learning_rate": 0.00048617332374774276,
      "loss": 5.9005,
      "mean_token_accuracy": 0.16430671364068986,
      "num_tokens": 25332234.0,
      "step": 13590
    },
    {
      "entropy": 6.182890844345093,
      "epoch": 1.200547509713882,
      "grad_norm": 2.75,
      "learning_rate": 0.00048616245243481216,
      "loss": 5.9934,
      "mean_token_accuracy": 0.15685328394174575,
      "num_tokens": 25341462.0,
      "step": 13595
    },
    {
      "entropy": 6.131249046325683,
      "epoch": 1.2009890498057223,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048615157698532913,
      "loss": 5.9169,
      "mean_token_accuracy": 0.16424605399370193,
      "num_tokens": 25351771.0,
      "step": 13600
    },
    {
      "entropy": 6.136184453964233,
      "epoch": 1.2014305898975628,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048614069739950683,
      "loss": 5.8721,
      "mean_token_accuracy": 0.16261929720640184,
      "num_tokens": 25360108.0,
      "step": 13605
    },
    {
      "entropy": 6.131735229492188,
      "epoch": 1.201872129989403,
      "grad_norm": 2.625,
      "learning_rate": 0.0004861298136775583,
      "loss": 5.8891,
      "mean_token_accuracy": 0.1624024197459221,
      "num_tokens": 25369635.0,
      "step": 13610
    },
    {
      "entropy": 6.100260639190674,
      "epoch": 1.2023136700812433,
      "grad_norm": 2.625,
      "learning_rate": 0.0004861189258196969,
      "loss": 5.8581,
      "mean_token_accuracy": 0.16749583333730697,
      "num_tokens": 25379623.0,
      "step": 13615
    },
    {
      "entropy": 6.163019418716431,
      "epoch": 1.2027552101730836,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004861080338261357,
      "loss": 5.9406,
      "mean_token_accuracy": 0.15702980607748032,
      "num_tokens": 25388591.0,
      "step": 13620
    },
    {
      "entropy": 6.17725977897644,
      "epoch": 1.2031967502649241,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004860971376970882,
      "loss": 5.8435,
      "mean_token_accuracy": 0.16299673914909363,
      "num_tokens": 25397865.0,
      "step": 13625
    },
    {
      "entropy": 6.075814199447632,
      "epoch": 1.2036382903567644,
      "grad_norm": 3.0,
      "learning_rate": 0.00048608623743276794,
      "loss": 5.7859,
      "mean_token_accuracy": 0.16911190748214722,
      "num_tokens": 25407143.0,
      "step": 13630
    },
    {
      "entropy": 6.1407585620880125,
      "epoch": 1.2040798304486047,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048607533303338823,
      "loss": 5.9284,
      "mean_token_accuracy": 0.16223665624856948,
      "num_tokens": 25415721.0,
      "step": 13635
    },
    {
      "entropy": 6.14788932800293,
      "epoch": 1.2045213705404452,
      "grad_norm": 6.875,
      "learning_rate": 0.00048606442449916284,
      "loss": 5.9796,
      "mean_token_accuracy": 0.1607537016272545,
      "num_tokens": 25425319.0,
      "step": 13640
    },
    {
      "entropy": 6.1613319396972654,
      "epoch": 1.2049629106322854,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048605351183030537,
      "loss": 5.9161,
      "mean_token_accuracy": 0.16481170803308487,
      "num_tokens": 25435477.0,
      "step": 13645
    },
    {
      "entropy": 6.137243270874023,
      "epoch": 1.2054044507241257,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004860425950270296,
      "loss": 5.9081,
      "mean_token_accuracy": 0.16187706589698792,
      "num_tokens": 25445633.0,
      "step": 13650
    },
    {
      "entropy": 6.131955146789551,
      "epoch": 1.205845990815966,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048603167408954953,
      "loss": 5.8764,
      "mean_token_accuracy": 0.15950068980455398,
      "num_tokens": 25455228.0,
      "step": 13655
    },
    {
      "entropy": 6.148524236679077,
      "epoch": 1.2062875309078065,
      "grad_norm": 5.40625,
      "learning_rate": 0.00048602074901807894,
      "loss": 5.8819,
      "mean_token_accuracy": 0.16517825722694396,
      "num_tokens": 25463992.0,
      "step": 13660
    },
    {
      "entropy": 6.175719308853149,
      "epoch": 1.2067290709996468,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048600981981283185,
      "loss": 5.8634,
      "mean_token_accuracy": 0.16691384613513946,
      "num_tokens": 25472985.0,
      "step": 13665
    },
    {
      "entropy": 6.140690469741822,
      "epoch": 1.207170611091487,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004859988864740225,
      "loss": 5.8818,
      "mean_token_accuracy": 0.16197107136249542,
      "num_tokens": 25482165.0,
      "step": 13670
    },
    {
      "entropy": 6.056623315811157,
      "epoch": 1.2076121511833275,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004859879490018649,
      "loss": 5.9793,
      "mean_token_accuracy": 0.16203480511903762,
      "num_tokens": 25490643.0,
      "step": 13675
    },
    {
      "entropy": 6.120284461975098,
      "epoch": 1.2080536912751678,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004859770073965733,
      "loss": 5.8664,
      "mean_token_accuracy": 0.1650485798716545,
      "num_tokens": 25499954.0,
      "step": 13680
    },
    {
      "entropy": 6.209693431854248,
      "epoch": 1.208495231367008,
      "grad_norm": 2.484375,
      "learning_rate": 0.00048596606165836223,
      "loss": 5.9249,
      "mean_token_accuracy": 0.15819624364376067,
      "num_tokens": 25509512.0,
      "step": 13685
    },
    {
      "entropy": 6.172435998916626,
      "epoch": 1.2089367714588484,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00048595511178744594,
      "loss": 5.8412,
      "mean_token_accuracy": 0.16026907190680503,
      "num_tokens": 25518909.0,
      "step": 13690
    },
    {
      "entropy": 6.0585425853729244,
      "epoch": 1.2093783115506889,
      "grad_norm": 2.28125,
      "learning_rate": 0.000485944157784039,
      "loss": 5.9203,
      "mean_token_accuracy": 0.15861292630434037,
      "num_tokens": 25528221.0,
      "step": 13695
    },
    {
      "entropy": 6.139790153503418,
      "epoch": 1.2098198516425291,
      "grad_norm": 2.96875,
      "learning_rate": 0.000485933199648356,
      "loss": 5.9234,
      "mean_token_accuracy": 0.16150010973215104,
      "num_tokens": 25537428.0,
      "step": 13700
    },
    {
      "entropy": 6.163102626800537,
      "epoch": 1.2102613917343694,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048592223738061155,
      "loss": 5.9159,
      "mean_token_accuracy": 0.1621326193213463,
      "num_tokens": 25547608.0,
      "step": 13705
    },
    {
      "entropy": 6.2079113006591795,
      "epoch": 1.21070293182621,
      "grad_norm": 2.625,
      "learning_rate": 0.00048591127098102045,
      "loss": 5.9825,
      "mean_token_accuracy": 0.1559573456645012,
      "num_tokens": 25557482.0,
      "step": 13710
    },
    {
      "entropy": 6.12672758102417,
      "epoch": 1.2111444719180502,
      "grad_norm": 2.125,
      "learning_rate": 0.0004859003004497974,
      "loss": 5.8604,
      "mean_token_accuracy": 0.1620791271328926,
      "num_tokens": 25566967.0,
      "step": 13715
    },
    {
      "entropy": 6.152239274978638,
      "epoch": 1.2115860120098905,
      "grad_norm": 3.671875,
      "learning_rate": 0.00048588932578715754,
      "loss": 5.8578,
      "mean_token_accuracy": 0.16489053815603255,
      "num_tokens": 25576114.0,
      "step": 13720
    },
    {
      "entropy": 6.210437774658203,
      "epoch": 1.2120275521017307,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048587834699331567,
      "loss": 6.0647,
      "mean_token_accuracy": 0.14975523203611374,
      "num_tokens": 25585953.0,
      "step": 13725
    },
    {
      "entropy": 6.2288818359375,
      "epoch": 1.2124690921935712,
      "grad_norm": 2.421875,
      "learning_rate": 0.000485867364068487,
      "loss": 6.0321,
      "mean_token_accuracy": 0.15451286286115645,
      "num_tokens": 25595445.0,
      "step": 13730
    },
    {
      "entropy": 6.230136966705322,
      "epoch": 1.2129106322854115,
      "grad_norm": 5.4375,
      "learning_rate": 0.00048585637701288646,
      "loss": 5.9031,
      "mean_token_accuracy": 0.16195821315050124,
      "num_tokens": 25605883.0,
      "step": 13735
    },
    {
      "entropy": 6.194358062744141,
      "epoch": 1.2133521723772518,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004858453858267295,
      "loss": 5.9484,
      "mean_token_accuracy": 0.1579759567975998,
      "num_tokens": 25615653.0,
      "step": 13740
    },
    {
      "entropy": 6.095244264602661,
      "epoch": 1.2137937124690923,
      "grad_norm": 5.46875,
      "learning_rate": 0.0004858343905102314,
      "loss": 5.871,
      "mean_token_accuracy": 0.16299858391284944,
      "num_tokens": 25625043.0,
      "step": 13745
    },
    {
      "entropy": 6.165210485458374,
      "epoch": 1.2142352525609326,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048582339106360736,
      "loss": 5.8855,
      "mean_token_accuracy": 0.16728933155536652,
      "num_tokens": 25633642.0,
      "step": 13750
    },
    {
      "entropy": 6.2057489395141605,
      "epoch": 1.2146767926527728,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048581238748707304,
      "loss": 5.9392,
      "mean_token_accuracy": 0.15919444113969802,
      "num_tokens": 25642154.0,
      "step": 13755
    },
    {
      "entropy": 6.133049726486206,
      "epoch": 1.2151183327446131,
      "grad_norm": 2.453125,
      "learning_rate": 0.000485801379780844,
      "loss": 5.8307,
      "mean_token_accuracy": 0.16360081136226653,
      "num_tokens": 25650250.0,
      "step": 13760
    },
    {
      "entropy": 6.115797424316407,
      "epoch": 1.2155598728364536,
      "grad_norm": 2.828125,
      "learning_rate": 0.00048579036794513584,
      "loss": 5.8827,
      "mean_token_accuracy": 0.16627621203660964,
      "num_tokens": 25659789.0,
      "step": 13765
    },
    {
      "entropy": 6.120511722564697,
      "epoch": 1.216001412928294,
      "grad_norm": 1.734375,
      "learning_rate": 0.00048577935198016413,
      "loss": 5.9403,
      "mean_token_accuracy": 0.15867630243301392,
      "num_tokens": 25668658.0,
      "step": 13770
    },
    {
      "entropy": 6.160347652435303,
      "epoch": 1.2164429530201342,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004857683318861449,
      "loss": 5.9365,
      "mean_token_accuracy": 0.15854731649160386,
      "num_tokens": 25679082.0,
      "step": 13775
    },
    {
      "entropy": 6.202041149139404,
      "epoch": 1.2168844931119747,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048575730766329384,
      "loss": 5.9092,
      "mean_token_accuracy": 0.15646228790283204,
      "num_tokens": 25688408.0,
      "step": 13780
    },
    {
      "entropy": 6.1509631156921385,
      "epoch": 1.217326033203815,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048574627931182704,
      "loss": 5.8894,
      "mean_token_accuracy": 0.16455817371606826,
      "num_tokens": 25697415.0,
      "step": 13785
    },
    {
      "entropy": 6.15203971862793,
      "epoch": 1.2177675732956552,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048573524683196047,
      "loss": 5.9599,
      "mean_token_accuracy": 0.15235176384449006,
      "num_tokens": 25706330.0,
      "step": 13790
    },
    {
      "entropy": 6.21326093673706,
      "epoch": 1.2182091133874955,
      "grad_norm": 2.0,
      "learning_rate": 0.0004857242102239102,
      "loss": 5.9326,
      "mean_token_accuracy": 0.15663277059793473,
      "num_tokens": 25716198.0,
      "step": 13795
    },
    {
      "entropy": 6.211286830902099,
      "epoch": 1.218650653479336,
      "grad_norm": 11.5625,
      "learning_rate": 0.0004857131694878926,
      "loss": 5.9214,
      "mean_token_accuracy": 0.15139016807079314,
      "num_tokens": 25726486.0,
      "step": 13800
    },
    {
      "entropy": 6.147220516204834,
      "epoch": 1.2190921935711763,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048570212462412376,
      "loss": 5.8401,
      "mean_token_accuracy": 0.16819383352994918,
      "num_tokens": 25735559.0,
      "step": 13805
    },
    {
      "entropy": 6.137135219573975,
      "epoch": 1.2195337336630165,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004856910756328202,
      "loss": 5.922,
      "mean_token_accuracy": 0.15584810078144073,
      "num_tokens": 25744781.0,
      "step": 13810
    },
    {
      "entropy": 6.105981111526489,
      "epoch": 1.219975273754857,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048568002251419835,
      "loss": 5.8809,
      "mean_token_accuracy": 0.16684502214193345,
      "num_tokens": 25753984.0,
      "step": 13815
    },
    {
      "entropy": 6.161437606811523,
      "epoch": 1.2204168138466973,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004856689652684746,
      "loss": 5.9624,
      "mean_token_accuracy": 0.15715058296918868,
      "num_tokens": 25763378.0,
      "step": 13820
    },
    {
      "entropy": 6.21608190536499,
      "epoch": 1.2208583539385376,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004856579038958656,
      "loss": 5.876,
      "mean_token_accuracy": 0.16399963051080704,
      "num_tokens": 25771907.0,
      "step": 13825
    },
    {
      "entropy": 6.160857629776001,
      "epoch": 1.2212998940303779,
      "grad_norm": 7.46875,
      "learning_rate": 0.00048564683839658805,
      "loss": 5.9255,
      "mean_token_accuracy": 0.16224211305379868,
      "num_tokens": 25780464.0,
      "step": 13830
    },
    {
      "entropy": 6.106551647186279,
      "epoch": 1.2217414341222184,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048563576877085887,
      "loss": 5.9417,
      "mean_token_accuracy": 0.16728258579969407,
      "num_tokens": 25790300.0,
      "step": 13835
    },
    {
      "entropy": 6.125736951828003,
      "epoch": 1.2221829742140586,
      "grad_norm": 2.75,
      "learning_rate": 0.0004856246950188947,
      "loss": 5.848,
      "mean_token_accuracy": 0.1637830078601837,
      "num_tokens": 25799103.0,
      "step": 13840
    },
    {
      "entropy": 6.183549404144287,
      "epoch": 1.222624514305899,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048561361714091257,
      "loss": 5.9488,
      "mean_token_accuracy": 0.15795571282505988,
      "num_tokens": 25808506.0,
      "step": 13845
    },
    {
      "entropy": 6.169549608230591,
      "epoch": 1.2230660543977394,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048560253513712945,
      "loss": 5.897,
      "mean_token_accuracy": 0.15889372080564498,
      "num_tokens": 25817965.0,
      "step": 13850
    },
    {
      "entropy": 6.093393850326538,
      "epoch": 1.2235075944895797,
      "grad_norm": 1.875,
      "learning_rate": 0.0004855914490077624,
      "loss": 5.8957,
      "mean_token_accuracy": 0.1627109855413437,
      "num_tokens": 25827874.0,
      "step": 13855
    },
    {
      "entropy": 6.239730882644653,
      "epoch": 1.22394913458142,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048558035875302864,
      "loss": 5.9811,
      "mean_token_accuracy": 0.16338664591312407,
      "num_tokens": 25836840.0,
      "step": 13860
    },
    {
      "entropy": 6.169374942779541,
      "epoch": 1.2243906746732602,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004855692643731455,
      "loss": 5.895,
      "mean_token_accuracy": 0.16576620787382126,
      "num_tokens": 25846344.0,
      "step": 13865
    },
    {
      "entropy": 6.13448429107666,
      "epoch": 1.2248322147651007,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048555816586833006,
      "loss": 5.8724,
      "mean_token_accuracy": 0.16877156645059585,
      "num_tokens": 25855336.0,
      "step": 13870
    },
    {
      "entropy": 6.169673919677734,
      "epoch": 1.225273754856941,
      "grad_norm": 2.5,
      "learning_rate": 0.00048554706323880007,
      "loss": 5.9226,
      "mean_token_accuracy": 0.16086411774158477,
      "num_tokens": 25864402.0,
      "step": 13875
    },
    {
      "entropy": 6.134567975997925,
      "epoch": 1.2257152949487813,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004855359564847728,
      "loss": 5.9041,
      "mean_token_accuracy": 0.15963491350412368,
      "num_tokens": 25873326.0,
      "step": 13880
    },
    {
      "entropy": 6.154598522186279,
      "epoch": 1.2261568350406218,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048552484560646585,
      "loss": 6.0001,
      "mean_token_accuracy": 0.1461065262556076,
      "num_tokens": 25882521.0,
      "step": 13885
    },
    {
      "entropy": 6.144202804565429,
      "epoch": 1.226598375132462,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048551373060409694,
      "loss": 5.8178,
      "mean_token_accuracy": 0.16841204911470414,
      "num_tokens": 25891087.0,
      "step": 13890
    },
    {
      "entropy": 6.154122352600098,
      "epoch": 1.2270399152243023,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048550261147788375,
      "loss": 5.88,
      "mean_token_accuracy": 0.15948003083467482,
      "num_tokens": 25900757.0,
      "step": 13895
    },
    {
      "entropy": 6.139804744720459,
      "epoch": 1.2274814553161426,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048549148822804415,
      "loss": 5.8697,
      "mean_token_accuracy": 0.1640913486480713,
      "num_tokens": 25909801.0,
      "step": 13900
    },
    {
      "entropy": 6.183643770217896,
      "epoch": 1.227922995407983,
      "grad_norm": 2.109375,
      "learning_rate": 0.000485480360854796,
      "loss": 5.9958,
      "mean_token_accuracy": 0.14888071939349173,
      "num_tokens": 25919142.0,
      "step": 13905
    },
    {
      "entropy": 6.179129362106323,
      "epoch": 1.2283645354998234,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004854692293583573,
      "loss": 5.8762,
      "mean_token_accuracy": 0.16878688782453538,
      "num_tokens": 25927560.0,
      "step": 13910
    },
    {
      "entropy": 6.1683567523956295,
      "epoch": 1.2288060755916637,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048545809373894605,
      "loss": 5.9942,
      "mean_token_accuracy": 0.15973826497793198,
      "num_tokens": 25937184.0,
      "step": 13915
    },
    {
      "entropy": 6.271181583404541,
      "epoch": 1.2292476156835042,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048544695399678044,
      "loss": 6.0021,
      "mean_token_accuracy": 0.1467764362692833,
      "num_tokens": 25946641.0,
      "step": 13920
    },
    {
      "entropy": 6.237927389144898,
      "epoch": 1.2296891557753444,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004854358101320787,
      "loss": 6.049,
      "mean_token_accuracy": 0.16022181808948516,
      "num_tokens": 25955910.0,
      "step": 13925
    },
    {
      "entropy": 6.144511890411377,
      "epoch": 1.2301306958671847,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048542466214505914,
      "loss": 5.9394,
      "mean_token_accuracy": 0.16163881719112397,
      "num_tokens": 25966580.0,
      "step": 13930
    },
    {
      "entropy": 6.1884512424469,
      "epoch": 1.230572235959025,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00048541351003594015,
      "loss": 5.8176,
      "mean_token_accuracy": 0.16138723194599153,
      "num_tokens": 25975815.0,
      "step": 13935
    },
    {
      "entropy": 6.10997896194458,
      "epoch": 1.2310137760508655,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004854023538049401,
      "loss": 5.9121,
      "mean_token_accuracy": 0.16547756046056747,
      "num_tokens": 25984784.0,
      "step": 13940
    },
    {
      "entropy": 6.098403596878052,
      "epoch": 1.2314553161427058,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048539119345227774,
      "loss": 5.8799,
      "mean_token_accuracy": 0.1653710812330246,
      "num_tokens": 25994125.0,
      "step": 13945
    },
    {
      "entropy": 6.143418645858764,
      "epoch": 1.231896856234546,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004853800289781714,
      "loss": 5.8921,
      "mean_token_accuracy": 0.16527310907840728,
      "num_tokens": 26002876.0,
      "step": 13950
    },
    {
      "entropy": 6.166758871078491,
      "epoch": 1.2323383963263865,
      "grad_norm": 1.96875,
      "learning_rate": 0.00048536886038284,
      "loss": 5.9019,
      "mean_token_accuracy": 0.16711250096559524,
      "num_tokens": 26012591.0,
      "step": 13955
    },
    {
      "entropy": 6.178992080688476,
      "epoch": 1.2327799364182268,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048535768766650225,
      "loss": 6.003,
      "mean_token_accuracy": 0.1515873596072197,
      "num_tokens": 26022107.0,
      "step": 13960
    },
    {
      "entropy": 6.178382539749146,
      "epoch": 1.233221476510067,
      "grad_norm": 2.796875,
      "learning_rate": 0.000485346510829377,
      "loss": 5.9639,
      "mean_token_accuracy": 0.16601064056158066,
      "num_tokens": 26031035.0,
      "step": 13965
    },
    {
      "entropy": 6.125119733810425,
      "epoch": 1.2336630166019074,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048533532987168334,
      "loss": 5.826,
      "mean_token_accuracy": 0.16693026423454285,
      "num_tokens": 26039980.0,
      "step": 13970
    },
    {
      "entropy": 6.188949108123779,
      "epoch": 1.2341045566937479,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048532414479364016,
      "loss": 6.0172,
      "mean_token_accuracy": 0.1519097715616226,
      "num_tokens": 26049389.0,
      "step": 13975
    },
    {
      "entropy": 6.151670598983765,
      "epoch": 1.2345460967855881,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004853129555954666,
      "loss": 5.8962,
      "mean_token_accuracy": 0.16317246705293656,
      "num_tokens": 26058868.0,
      "step": 13980
    },
    {
      "entropy": 6.136835813522339,
      "epoch": 1.2349876368774284,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004853017622773818,
      "loss": 5.9526,
      "mean_token_accuracy": 0.15672052949666976,
      "num_tokens": 26068025.0,
      "step": 13985
    },
    {
      "entropy": 6.173934078216552,
      "epoch": 1.235429176969269,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004852905648396051,
      "loss": 5.9081,
      "mean_token_accuracy": 0.1544721782207489,
      "num_tokens": 26076410.0,
      "step": 13990
    },
    {
      "entropy": 6.128337287902832,
      "epoch": 1.2358707170611092,
      "grad_norm": 2.0,
      "learning_rate": 0.00048527936328235583,
      "loss": 5.8651,
      "mean_token_accuracy": 0.16472303867340088,
      "num_tokens": 26085624.0,
      "step": 13995
    },
    {
      "entropy": 6.173004293441773,
      "epoch": 1.2363122571529495,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004852681576058534,
      "loss": 5.9532,
      "mean_token_accuracy": 0.15905184745788575,
      "num_tokens": 26094898.0,
      "step": 14000
    },
    {
      "entropy": 6.140864658355713,
      "epoch": 1.2367537972447897,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004852569478103174,
      "loss": 5.9093,
      "mean_token_accuracy": 0.1570506900548935,
      "num_tokens": 26105041.0,
      "step": 14005
    },
    {
      "entropy": 6.144279956817627,
      "epoch": 1.2371953373366302,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048524573389596736,
      "loss": 5.8933,
      "mean_token_accuracy": 0.16431440263986588,
      "num_tokens": 26114140.0,
      "step": 14010
    },
    {
      "entropy": 6.07152738571167,
      "epoch": 1.2376368774284705,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048523451586302295,
      "loss": 5.7642,
      "mean_token_accuracy": 0.16753955632448198,
      "num_tokens": 26122611.0,
      "step": 14015
    },
    {
      "entropy": 6.05628514289856,
      "epoch": 1.2380784175203108,
      "grad_norm": 2.109375,
      "learning_rate": 0.000485223293711704,
      "loss": 5.8383,
      "mean_token_accuracy": 0.1654874250292778,
      "num_tokens": 26131548.0,
      "step": 14020
    },
    {
      "entropy": 6.102871990203857,
      "epoch": 1.2385199576121513,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048521206744223024,
      "loss": 5.8317,
      "mean_token_accuracy": 0.15633988827466966,
      "num_tokens": 26139806.0,
      "step": 14025
    },
    {
      "entropy": 6.184936237335205,
      "epoch": 1.2389614977039916,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048520083705482165,
      "loss": 5.9427,
      "mean_token_accuracy": 0.1581635892391205,
      "num_tokens": 26148960.0,
      "step": 14030
    },
    {
      "entropy": 6.18954291343689,
      "epoch": 1.2394030377958318,
      "grad_norm": 2.5,
      "learning_rate": 0.0004851896025496982,
      "loss": 5.9559,
      "mean_token_accuracy": 0.1615045428276062,
      "num_tokens": 26158498.0,
      "step": 14035
    },
    {
      "entropy": 6.075376701354981,
      "epoch": 1.239844577887672,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048517836392707994,
      "loss": 5.8065,
      "mean_token_accuracy": 0.1683821365237236,
      "num_tokens": 26167319.0,
      "step": 14040
    },
    {
      "entropy": 6.022907018661499,
      "epoch": 1.2402861179795126,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048516712118718713,
      "loss": 5.8174,
      "mean_token_accuracy": 0.16261593699455262,
      "num_tokens": 26176059.0,
      "step": 14045
    },
    {
      "entropy": 6.102324914932251,
      "epoch": 1.2407276580713529,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004851558743302399,
      "loss": 5.8302,
      "mean_token_accuracy": 0.1683965727686882,
      "num_tokens": 26185102.0,
      "step": 14050
    },
    {
      "entropy": 6.158196020126343,
      "epoch": 1.2411691981631932,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004851446233564586,
      "loss": 5.8742,
      "mean_token_accuracy": 0.1645967409014702,
      "num_tokens": 26194873.0,
      "step": 14055
    },
    {
      "entropy": 6.1414063453674315,
      "epoch": 1.2416107382550337,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004851333682660637,
      "loss": 5.9206,
      "mean_token_accuracy": 0.15780261307954788,
      "num_tokens": 26205362.0,
      "step": 14060
    },
    {
      "entropy": 6.062950515747071,
      "epoch": 1.242052278346874,
      "grad_norm": 3.125,
      "learning_rate": 0.00048512210905927555,
      "loss": 5.8143,
      "mean_token_accuracy": 0.16553541123867035,
      "num_tokens": 26214266.0,
      "step": 14065
    },
    {
      "entropy": 6.242000102996826,
      "epoch": 1.2424938184387142,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004851108457363149,
      "loss": 5.9881,
      "mean_token_accuracy": 0.16291094869375228,
      "num_tokens": 26223069.0,
      "step": 14070
    },
    {
      "entropy": 6.16287899017334,
      "epoch": 1.2429353585305545,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004850995782974022,
      "loss": 5.9039,
      "mean_token_accuracy": 0.16355999261140824,
      "num_tokens": 26232547.0,
      "step": 14075
    },
    {
      "entropy": 6.16363377571106,
      "epoch": 1.243376898622395,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004850883067427582,
      "loss": 5.8891,
      "mean_token_accuracy": 0.15883788019418715,
      "num_tokens": 26241014.0,
      "step": 14080
    },
    {
      "entropy": 6.083933115005493,
      "epoch": 1.2438184387142353,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048507703107260384,
      "loss": 5.9343,
      "mean_token_accuracy": 0.1633957013487816,
      "num_tokens": 26250784.0,
      "step": 14085
    },
    {
      "entropy": 6.136890459060669,
      "epoch": 1.2442599788060755,
      "grad_norm": 2.375,
      "learning_rate": 0.00048506575128715984,
      "loss": 5.8564,
      "mean_token_accuracy": 0.16451779007911682,
      "num_tokens": 26259769.0,
      "step": 14090
    },
    {
      "entropy": 6.17093334197998,
      "epoch": 1.244701518897916,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004850544673866474,
      "loss": 5.8908,
      "mean_token_accuracy": 0.15885442569851876,
      "num_tokens": 26269038.0,
      "step": 14095
    },
    {
      "entropy": 6.182549715042114,
      "epoch": 1.2451430589897563,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004850431793712873,
      "loss": 5.8784,
      "mean_token_accuracy": 0.15211578011512755,
      "num_tokens": 26278657.0,
      "step": 14100
    },
    {
      "entropy": 6.090300464630127,
      "epoch": 1.2455845990815966,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048503188724130075,
      "loss": 5.8797,
      "mean_token_accuracy": 0.16596818417310716,
      "num_tokens": 26287458.0,
      "step": 14105
    },
    {
      "entropy": 6.125815773010254,
      "epoch": 1.2460261391734369,
      "grad_norm": 1.9375,
      "learning_rate": 0.000485020590996909,
      "loss": 5.9272,
      "mean_token_accuracy": 0.15662363916635513,
      "num_tokens": 26296324.0,
      "step": 14110
    },
    {
      "entropy": 6.149014949798584,
      "epoch": 1.2464676792652774,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00048500929063833334,
      "loss": 5.8424,
      "mean_token_accuracy": 0.16634784489870072,
      "num_tokens": 26304799.0,
      "step": 14115
    },
    {
      "entropy": 6.131788158416748,
      "epoch": 1.2469092193571176,
      "grad_norm": 1.921875,
      "learning_rate": 0.00048499798616579505,
      "loss": 5.9649,
      "mean_token_accuracy": 0.1626861184835434,
      "num_tokens": 26314223.0,
      "step": 14120
    },
    {
      "entropy": 6.180202770233154,
      "epoch": 1.247350759448958,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048498667757951574,
      "loss": 5.9632,
      "mean_token_accuracy": 0.1531592920422554,
      "num_tokens": 26323599.0,
      "step": 14125
    },
    {
      "entropy": 6.151395988464356,
      "epoch": 1.2477922995407984,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004849753648797168,
      "loss": 5.9923,
      "mean_token_accuracy": 0.15974041521549226,
      "num_tokens": 26331929.0,
      "step": 14130
    },
    {
      "entropy": 6.127089214324951,
      "epoch": 1.2482338396326387,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004849640480666199,
      "loss": 5.8848,
      "mean_token_accuracy": 0.16590515971183778,
      "num_tokens": 26340929.0,
      "step": 14135
    },
    {
      "entropy": 6.214146089553833,
      "epoch": 1.248675379724479,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004849527271404467,
      "loss": 5.8569,
      "mean_token_accuracy": 0.15572652220726013,
      "num_tokens": 26350393.0,
      "step": 14140
    },
    {
      "entropy": 6.212586307525635,
      "epoch": 1.2491169198163192,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048494140210141897,
      "loss": 6.0766,
      "mean_token_accuracy": 0.15760573968291283,
      "num_tokens": 26360709.0,
      "step": 14145
    },
    {
      "entropy": 6.143700647354126,
      "epoch": 1.2495584599081597,
      "grad_norm": 1.875,
      "learning_rate": 0.00048493007294975855,
      "loss": 5.8686,
      "mean_token_accuracy": 0.16504170149564742,
      "num_tokens": 26369447.0,
      "step": 14150
    },
    {
      "entropy": 6.146756267547607,
      "epoch": 1.25,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00048491873968568746,
      "loss": 5.855,
      "mean_token_accuracy": 0.16688266023993492,
      "num_tokens": 26379281.0,
      "step": 14155
    },
    {
      "entropy": 5.988578414916992,
      "epoch": 1.2504415400918403,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004849074023094276,
      "loss": 5.7976,
      "mean_token_accuracy": 0.17447122633457185,
      "num_tokens": 26387664.0,
      "step": 14160
    },
    {
      "entropy": 5.986051416397094,
      "epoch": 1.2508830801836806,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048489606082120106,
      "loss": 5.8213,
      "mean_token_accuracy": 0.15860421657562257,
      "num_tokens": 26397485.0,
      "step": 14165
    },
    {
      "entropy": 6.107970714569092,
      "epoch": 1.251324620275521,
      "grad_norm": 2.25,
      "learning_rate": 0.0004848847152212301,
      "loss": 5.7964,
      "mean_token_accuracy": 0.1685743436217308,
      "num_tokens": 26406307.0,
      "step": 14170
    },
    {
      "entropy": 6.193047094345093,
      "epoch": 1.2517661603673613,
      "grad_norm": 2.234375,
      "learning_rate": 0.000484873365509737,
      "loss": 6.0017,
      "mean_token_accuracy": 0.15746570080518724,
      "num_tokens": 26416155.0,
      "step": 14175
    },
    {
      "entropy": 6.133742761611939,
      "epoch": 1.2522077004592016,
      "grad_norm": 4.65625,
      "learning_rate": 0.00048486201168694397,
      "loss": 5.8793,
      "mean_token_accuracy": 0.16526299417018891,
      "num_tokens": 26424147.0,
      "step": 14180
    },
    {
      "entropy": 6.1153538703918455,
      "epoch": 1.252649240551042,
      "grad_norm": 3.609375,
      "learning_rate": 0.00048485065375307353,
      "loss": 5.8863,
      "mean_token_accuracy": 0.16146370023488998,
      "num_tokens": 26433249.0,
      "step": 14185
    },
    {
      "entropy": 6.137222003936768,
      "epoch": 1.2530907806428824,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048483929170834805,
      "loss": 5.9433,
      "mean_token_accuracy": 0.16384687423706054,
      "num_tokens": 26442346.0,
      "step": 14190
    },
    {
      "entropy": 6.146828317642212,
      "epoch": 1.2535323207347226,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048482792555299027,
      "loss": 5.8913,
      "mean_token_accuracy": 0.16533481031656266,
      "num_tokens": 26450885.0,
      "step": 14195
    },
    {
      "entropy": 6.143717241287232,
      "epoch": 1.2539738608265631,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004848165552872227,
      "loss": 5.9226,
      "mean_token_accuracy": 0.16853769421577453,
      "num_tokens": 26459827.0,
      "step": 14200
    },
    {
      "entropy": 6.089367198944092,
      "epoch": 1.2544154009184034,
      "grad_norm": 2.25,
      "learning_rate": 0.0004848051809112682,
      "loss": 5.9073,
      "mean_token_accuracy": 0.1613314315676689,
      "num_tokens": 26468357.0,
      "step": 14205
    },
    {
      "entropy": 6.1293073177337645,
      "epoch": 1.2548569410102437,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048479380242534955,
      "loss": 5.9997,
      "mean_token_accuracy": 0.1588381201028824,
      "num_tokens": 26477396.0,
      "step": 14210
    },
    {
      "entropy": 6.179404830932617,
      "epoch": 1.2552984811020842,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004847824198296896,
      "loss": 5.948,
      "mean_token_accuracy": 0.1606784790754318,
      "num_tokens": 26486468.0,
      "step": 14215
    },
    {
      "entropy": 6.213735294342041,
      "epoch": 1.2557400211939245,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004847710331245113,
      "loss": 5.8993,
      "mean_token_accuracy": 0.1633673846721649,
      "num_tokens": 26496157.0,
      "step": 14220
    },
    {
      "entropy": 6.061906909942627,
      "epoch": 1.2561815612857647,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048475964231003776,
      "loss": 5.7761,
      "mean_token_accuracy": 0.17620088160037994,
      "num_tokens": 26504573.0,
      "step": 14225
    },
    {
      "entropy": 6.01952805519104,
      "epoch": 1.256623101377605,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004847482473864922,
      "loss": 5.751,
      "mean_token_accuracy": 0.17447698712348939,
      "num_tokens": 26513441.0,
      "step": 14230
    },
    {
      "entropy": 6.1289002895355225,
      "epoch": 1.2570646414694453,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004847368483540977,
      "loss": 5.9697,
      "mean_token_accuracy": 0.15450180172920228,
      "num_tokens": 26522911.0,
      "step": 14235
    },
    {
      "entropy": 6.088858222961425,
      "epoch": 1.2575061815612858,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048472544521307765,
      "loss": 5.8259,
      "mean_token_accuracy": 0.168184857070446,
      "num_tokens": 26531388.0,
      "step": 14240
    },
    {
      "entropy": 6.148261499404907,
      "epoch": 1.257947721653126,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004847140379636554,
      "loss": 5.8277,
      "mean_token_accuracy": 0.1657315105199814,
      "num_tokens": 26540500.0,
      "step": 14245
    },
    {
      "entropy": 6.1033295631408695,
      "epoch": 1.2583892617449663,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004847026266060545,
      "loss": 5.8336,
      "mean_token_accuracy": 0.16738886535167694,
      "num_tokens": 26550041.0,
      "step": 14250
    },
    {
      "entropy": 6.08073844909668,
      "epoch": 1.2588308018368068,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004846912111404983,
      "loss": 5.8132,
      "mean_token_accuracy": 0.16374696642160416,
      "num_tokens": 26559594.0,
      "step": 14255
    },
    {
      "entropy": 6.134618186950684,
      "epoch": 1.2592723419286471,
      "grad_norm": 1.9375,
      "learning_rate": 0.00048467979156721055,
      "loss": 5.8958,
      "mean_token_accuracy": 0.16100245267152785,
      "num_tokens": 26569143.0,
      "step": 14260
    },
    {
      "entropy": 6.075584793090821,
      "epoch": 1.2597138820204874,
      "grad_norm": 2.0,
      "learning_rate": 0.00048466836788641496,
      "loss": 5.8709,
      "mean_token_accuracy": 0.16191661655902861,
      "num_tokens": 26578040.0,
      "step": 14265
    },
    {
      "entropy": 6.110110807418823,
      "epoch": 1.260155422112328,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048465694009833525,
      "loss": 5.9016,
      "mean_token_accuracy": 0.16245102733373643,
      "num_tokens": 26588283.0,
      "step": 14270
    },
    {
      "entropy": 6.169938564300537,
      "epoch": 1.2605969622041682,
      "grad_norm": 2.375,
      "learning_rate": 0.00048464550820319537,
      "loss": 5.9417,
      "mean_token_accuracy": 0.15852103233337403,
      "num_tokens": 26597815.0,
      "step": 14275
    },
    {
      "entropy": 6.100296974182129,
      "epoch": 1.2610385022960084,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004846340722012192,
      "loss": 5.9146,
      "mean_token_accuracy": 0.1597320780158043,
      "num_tokens": 26606752.0,
      "step": 14280
    },
    {
      "entropy": 6.158532190322876,
      "epoch": 1.261480042387849,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004846226320926308,
      "loss": 5.8915,
      "mean_token_accuracy": 0.16063199043273926,
      "num_tokens": 26615312.0,
      "step": 14285
    },
    {
      "entropy": 6.184799385070801,
      "epoch": 1.2619215824796892,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004846111878776542,
      "loss": 5.8876,
      "mean_token_accuracy": 0.16116779446601867,
      "num_tokens": 26624642.0,
      "step": 14290
    },
    {
      "entropy": 6.052973508834839,
      "epoch": 1.2623631225715295,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004845997395565137,
      "loss": 5.7603,
      "mean_token_accuracy": 0.17240442037582399,
      "num_tokens": 26633250.0,
      "step": 14295
    },
    {
      "entropy": 6.061658334732056,
      "epoch": 1.2628046626633698,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004845882871294335,
      "loss": 5.92,
      "mean_token_accuracy": 0.16076537370681762,
      "num_tokens": 26642422.0,
      "step": 14300
    },
    {
      "entropy": 6.150954103469848,
      "epoch": 1.26324620275521,
      "grad_norm": 2.5,
      "learning_rate": 0.0004845768305966379,
      "loss": 5.9583,
      "mean_token_accuracy": 0.15743603110313414,
      "num_tokens": 26651978.0,
      "step": 14305
    },
    {
      "entropy": 6.1525288105010985,
      "epoch": 1.2636877428470505,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004845653699583515,
      "loss": 5.9573,
      "mean_token_accuracy": 0.1546889066696167,
      "num_tokens": 26662075.0,
      "step": 14310
    },
    {
      "entropy": 6.153447103500366,
      "epoch": 1.2641292829388908,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048455390521479855,
      "loss": 5.8636,
      "mean_token_accuracy": 0.16401289254426957,
      "num_tokens": 26670813.0,
      "step": 14315
    },
    {
      "entropy": 6.144858694076538,
      "epoch": 1.264570823030731,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048454243636620383,
      "loss": 5.8906,
      "mean_token_accuracy": 0.16578105837106705,
      "num_tokens": 26679867.0,
      "step": 14320
    },
    {
      "entropy": 6.084038591384887,
      "epoch": 1.2650123631225716,
      "grad_norm": 1.8203125,
      "learning_rate": 0.000484530963412792,
      "loss": 5.9121,
      "mean_token_accuracy": 0.15640238672494888,
      "num_tokens": 26688523.0,
      "step": 14325
    },
    {
      "entropy": 6.114808654785156,
      "epoch": 1.2654539032144119,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048451948635478765,
      "loss": 5.9695,
      "mean_token_accuracy": 0.16364617198705672,
      "num_tokens": 26697922.0,
      "step": 14330
    },
    {
      "entropy": 6.158629322052002,
      "epoch": 1.2658954433062521,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004845080051924158,
      "loss": 5.9157,
      "mean_token_accuracy": 0.15545468032360077,
      "num_tokens": 26707058.0,
      "step": 14335
    },
    {
      "entropy": 6.158748531341553,
      "epoch": 1.2663369833980926,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004844965199259013,
      "loss": 5.9035,
      "mean_token_accuracy": 0.16495700180530548,
      "num_tokens": 26715873.0,
      "step": 14340
    },
    {
      "entropy": 6.170320844650268,
      "epoch": 1.266778523489933,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004844850305554692,
      "loss": 5.9055,
      "mean_token_accuracy": 0.16377891302108766,
      "num_tokens": 26726253.0,
      "step": 14345
    },
    {
      "entropy": 6.143414831161499,
      "epoch": 1.2672200635817732,
      "grad_norm": 2.625,
      "learning_rate": 0.00048447353708134426,
      "loss": 5.9287,
      "mean_token_accuracy": 0.16029662042856216,
      "num_tokens": 26737077.0,
      "step": 14350
    },
    {
      "entropy": 6.134037160873413,
      "epoch": 1.2676616036736137,
      "grad_norm": 1.96875,
      "learning_rate": 0.000484462039503752,
      "loss": 5.9326,
      "mean_token_accuracy": 0.15892160981893538,
      "num_tokens": 26746603.0,
      "step": 14355
    },
    {
      "entropy": 6.08798565864563,
      "epoch": 1.268103143765454,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048445053782291753,
      "loss": 5.8972,
      "mean_token_accuracy": 0.16114922016859054,
      "num_tokens": 26756174.0,
      "step": 14360
    },
    {
      "entropy": 6.116337156295776,
      "epoch": 1.2685446838572942,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048443903203906614,
      "loss": 5.8928,
      "mean_token_accuracy": 0.16467928588390351,
      "num_tokens": 26765124.0,
      "step": 14365
    },
    {
      "entropy": 6.109856700897216,
      "epoch": 1.2689862239491345,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004844275221524232,
      "loss": 5.8811,
      "mean_token_accuracy": 0.17070176005363463,
      "num_tokens": 26774438.0,
      "step": 14370
    },
    {
      "entropy": 6.210801219940185,
      "epoch": 1.2694277640409748,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048441600816321424,
      "loss": 5.9172,
      "mean_token_accuracy": 0.1579091191291809,
      "num_tokens": 26783613.0,
      "step": 14375
    },
    {
      "entropy": 6.105120325088501,
      "epoch": 1.2698693041328153,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004844044900716647,
      "loss": 5.7676,
      "mean_token_accuracy": 0.16843381524085999,
      "num_tokens": 26792136.0,
      "step": 14380
    },
    {
      "entropy": 6.0391148090362545,
      "epoch": 1.2703108442246556,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004843929678780004,
      "loss": 5.7884,
      "mean_token_accuracy": 0.17705477327108382,
      "num_tokens": 26801481.0,
      "step": 14385
    },
    {
      "entropy": 6.069174575805664,
      "epoch": 1.2707523843164958,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004843814415824469,
      "loss": 5.8658,
      "mean_token_accuracy": 0.16673449277877808,
      "num_tokens": 26810503.0,
      "step": 14390
    },
    {
      "entropy": 6.198712348937988,
      "epoch": 1.2711939244083363,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048436991118523003,
      "loss": 5.9038,
      "mean_token_accuracy": 0.16482168734073638,
      "num_tokens": 26820007.0,
      "step": 14395
    },
    {
      "entropy": 6.135466194152832,
      "epoch": 1.2716354645001766,
      "grad_norm": 3.609375,
      "learning_rate": 0.00048435837668657565,
      "loss": 5.8228,
      "mean_token_accuracy": 0.1683567062020302,
      "num_tokens": 26829088.0,
      "step": 14400
    },
    {
      "entropy": 6.088929128646851,
      "epoch": 1.272077004592017,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004843468380867098,
      "loss": 5.9164,
      "mean_token_accuracy": 0.16359835118055344,
      "num_tokens": 26838619.0,
      "step": 14405
    },
    {
      "entropy": 6.108640670776367,
      "epoch": 1.2725185446838574,
      "grad_norm": 2.75,
      "learning_rate": 0.0004843352953858585,
      "loss": 5.9344,
      "mean_token_accuracy": 0.1615082025527954,
      "num_tokens": 26848728.0,
      "step": 14410
    },
    {
      "entropy": 6.143708229064941,
      "epoch": 1.2729600847756977,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004843237485842477,
      "loss": 5.8505,
      "mean_token_accuracy": 0.17027464807033538,
      "num_tokens": 26859293.0,
      "step": 14415
    },
    {
      "entropy": 6.12141752243042,
      "epoch": 1.273401624867538,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048431219768210374,
      "loss": 5.9081,
      "mean_token_accuracy": 0.1629863888025284,
      "num_tokens": 26868818.0,
      "step": 14420
    },
    {
      "entropy": 6.207045698165894,
      "epoch": 1.2738431649593784,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004843006426796529,
      "loss": 5.991,
      "mean_token_accuracy": 0.1566545620560646,
      "num_tokens": 26879245.0,
      "step": 14425
    },
    {
      "entropy": 6.158621215820313,
      "epoch": 1.2742847050512187,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004842890835771215,
      "loss": 5.8771,
      "mean_token_accuracy": 0.1620942696928978,
      "num_tokens": 26888074.0,
      "step": 14430
    },
    {
      "entropy": 6.132410717010498,
      "epoch": 1.274726245143059,
      "grad_norm": 1.96875,
      "learning_rate": 0.000484277520374736,
      "loss": 5.9138,
      "mean_token_accuracy": 0.15504443496465684,
      "num_tokens": 26897324.0,
      "step": 14435
    },
    {
      "entropy": 6.2166553974151615,
      "epoch": 1.2751677852348993,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048426595307272285,
      "loss": 6.0221,
      "mean_token_accuracy": 0.15333397835493087,
      "num_tokens": 26906797.0,
      "step": 14440
    },
    {
      "entropy": 6.209755325317383,
      "epoch": 1.2756093253267395,
      "grad_norm": 2.0,
      "learning_rate": 0.00048425438167130875,
      "loss": 5.9388,
      "mean_token_accuracy": 0.1663684904575348,
      "num_tokens": 26915737.0,
      "step": 14445
    },
    {
      "entropy": 6.081834125518799,
      "epoch": 1.27605086541858,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048424280617072026,
      "loss": 5.8803,
      "mean_token_accuracy": 0.16682081371545793,
      "num_tokens": 26925364.0,
      "step": 14450
    },
    {
      "entropy": 6.108405208587646,
      "epoch": 1.2764924055104203,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048423122657118424,
      "loss": 5.8691,
      "mean_token_accuracy": 0.1697370171546936,
      "num_tokens": 26935188.0,
      "step": 14455
    },
    {
      "entropy": 6.109171676635742,
      "epoch": 1.2769339456022606,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048421964287292745,
      "loss": 5.8597,
      "mean_token_accuracy": 0.16356942057609558,
      "num_tokens": 26944606.0,
      "step": 14460
    },
    {
      "entropy": 6.183118057250977,
      "epoch": 1.277375485694101,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004842080550761769,
      "loss": 5.9759,
      "mean_token_accuracy": 0.15626660883426666,
      "num_tokens": 26955147.0,
      "step": 14465
    },
    {
      "entropy": 6.133596706390381,
      "epoch": 1.2778170257859414,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004841964631811595,
      "loss": 5.9184,
      "mean_token_accuracy": 0.1682221621274948,
      "num_tokens": 26964243.0,
      "step": 14470
    },
    {
      "entropy": 6.190114974975586,
      "epoch": 1.2782585658777816,
      "grad_norm": 1.890625,
      "learning_rate": 0.00048418486718810244,
      "loss": 5.9028,
      "mean_token_accuracy": 0.16112356930971145,
      "num_tokens": 26973284.0,
      "step": 14475
    },
    {
      "entropy": 6.18262734413147,
      "epoch": 1.2787001059696221,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004841732670972327,
      "loss": 5.9903,
      "mean_token_accuracy": 0.1584943115711212,
      "num_tokens": 26982757.0,
      "step": 14480
    },
    {
      "entropy": 6.164980840682984,
      "epoch": 1.2791416460614624,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004841616629087776,
      "loss": 5.9234,
      "mean_token_accuracy": 0.15870108753442763,
      "num_tokens": 26991784.0,
      "step": 14485
    },
    {
      "entropy": 6.195461940765381,
      "epoch": 1.2795831861533027,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048415005462296455,
      "loss": 5.925,
      "mean_token_accuracy": 0.15015334486961365,
      "num_tokens": 27000700.0,
      "step": 14490
    },
    {
      "entropy": 6.165881538391114,
      "epoch": 1.2800247262451432,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004841384422400209,
      "loss": 5.881,
      "mean_token_accuracy": 0.17133558839559554,
      "num_tokens": 27010089.0,
      "step": 14495
    },
    {
      "entropy": 6.090375089645386,
      "epoch": 1.2804662663369835,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004841268257601741,
      "loss": 5.885,
      "mean_token_accuracy": 0.16691755950450898,
      "num_tokens": 27018961.0,
      "step": 14500
    },
    {
      "entropy": 6.148070478439331,
      "epoch": 1.2809078064288237,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004841152051836517,
      "loss": 5.9356,
      "mean_token_accuracy": 0.16089101135730743,
      "num_tokens": 27027905.0,
      "step": 14505
    },
    {
      "entropy": 6.140289926528931,
      "epoch": 1.281349346520664,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048410358051068135,
      "loss": 5.859,
      "mean_token_accuracy": 0.16071921437978745,
      "num_tokens": 27036969.0,
      "step": 14510
    },
    {
      "entropy": 6.054096698760986,
      "epoch": 1.2817908866125043,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004840919517414908,
      "loss": 5.8292,
      "mean_token_accuracy": 0.17177568078041078,
      "num_tokens": 27046654.0,
      "step": 14515
    },
    {
      "entropy": 6.160114002227783,
      "epoch": 1.2822324267043448,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00048408031887630775,
      "loss": 5.9421,
      "mean_token_accuracy": 0.16171510368585587,
      "num_tokens": 27058492.0,
      "step": 14520
    },
    {
      "entropy": 6.185419082641602,
      "epoch": 1.282673966796185,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004840686819153604,
      "loss": 5.9313,
      "mean_token_accuracy": 0.16024121046066284,
      "num_tokens": 27068863.0,
      "step": 14525
    },
    {
      "entropy": 6.132033395767212,
      "epoch": 1.2831155068880253,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004840570408588762,
      "loss": 5.8992,
      "mean_token_accuracy": 0.16224589496850966,
      "num_tokens": 27078793.0,
      "step": 14530
    },
    {
      "entropy": 6.066058921813965,
      "epoch": 1.2835570469798658,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004840453957070836,
      "loss": 5.7058,
      "mean_token_accuracy": 0.17713941931724547,
      "num_tokens": 27087093.0,
      "step": 14535
    },
    {
      "entropy": 6.119988012313843,
      "epoch": 1.283998587071706,
      "grad_norm": 2.5,
      "learning_rate": 0.00048403374646021057,
      "loss": 5.8232,
      "mean_token_accuracy": 0.16361542344093322,
      "num_tokens": 27096601.0,
      "step": 14540
    },
    {
      "entropy": 6.100772953033447,
      "epoch": 1.2844401271635464,
      "grad_norm": 3.0625,
      "learning_rate": 0.00048402209311848533,
      "loss": 5.8515,
      "mean_token_accuracy": 0.16600432693958284,
      "num_tokens": 27105041.0,
      "step": 14545
    },
    {
      "entropy": 6.101114559173584,
      "epoch": 1.2848816672553869,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048401043568213614,
      "loss": 5.927,
      "mean_token_accuracy": 0.16112814247608184,
      "num_tokens": 27114252.0,
      "step": 14550
    },
    {
      "entropy": 6.143374300003051,
      "epoch": 1.2853232073472272,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004839987741513914,
      "loss": 5.851,
      "mean_token_accuracy": 0.16172419786453246,
      "num_tokens": 27122480.0,
      "step": 14555
    },
    {
      "entropy": 6.114062023162842,
      "epoch": 1.2857647474390674,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00048398710852647954,
      "loss": 5.9046,
      "mean_token_accuracy": 0.159881991147995,
      "num_tokens": 27132322.0,
      "step": 14560
    },
    {
      "entropy": 6.0440285205841064,
      "epoch": 1.286206287530908,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048397543880762905,
      "loss": 5.7701,
      "mean_token_accuracy": 0.17634997218847276,
      "num_tokens": 27140782.0,
      "step": 14565
    },
    {
      "entropy": 6.061186504364014,
      "epoch": 1.2866478276227482,
      "grad_norm": 3.140625,
      "learning_rate": 0.00048396376499506866,
      "loss": 5.8139,
      "mean_token_accuracy": 0.16377085894346238,
      "num_tokens": 27150058.0,
      "step": 14570
    },
    {
      "entropy": 6.089512825012207,
      "epoch": 1.2870893677145885,
      "grad_norm": 1.625,
      "learning_rate": 0.00048395208708902686,
      "loss": 5.8486,
      "mean_token_accuracy": 0.16396327018737794,
      "num_tokens": 27159574.0,
      "step": 14575
    },
    {
      "entropy": 6.119163084030151,
      "epoch": 1.2875309078064288,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00048394040508973257,
      "loss": 5.8675,
      "mean_token_accuracy": 0.1665537938475609,
      "num_tokens": 27168025.0,
      "step": 14580
    },
    {
      "entropy": 6.033315896987915,
      "epoch": 1.287972447898269,
      "grad_norm": 2.25,
      "learning_rate": 0.0004839287189974145,
      "loss": 5.8759,
      "mean_token_accuracy": 0.16489376351237298,
      "num_tokens": 27177108.0,
      "step": 14585
    },
    {
      "entropy": 6.097931241989135,
      "epoch": 1.2884139879901095,
      "grad_norm": 1.6875,
      "learning_rate": 0.00048391702881230177,
      "loss": 5.9287,
      "mean_token_accuracy": 0.1560143455862999,
      "num_tokens": 27187486.0,
      "step": 14590
    },
    {
      "entropy": 6.214470434188843,
      "epoch": 1.2888555280819498,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004839053345346232,
      "loss": 5.9183,
      "mean_token_accuracy": 0.15378315448760987,
      "num_tokens": 27195962.0,
      "step": 14595
    },
    {
      "entropy": 6.122103357315064,
      "epoch": 1.28929706817379,
      "grad_norm": 1.875,
      "learning_rate": 0.0004838936361646079,
      "loss": 5.8991,
      "mean_token_accuracy": 0.16736670732498168,
      "num_tokens": 27206354.0,
      "step": 14600
    },
    {
      "entropy": 6.027134132385254,
      "epoch": 1.2897386082656306,
      "grad_norm": 5.5625,
      "learning_rate": 0.0004838819337024852,
      "loss": 5.7738,
      "mean_token_accuracy": 0.1755576863884926,
      "num_tokens": 27215326.0,
      "step": 14605
    },
    {
      "entropy": 6.133112668991089,
      "epoch": 1.2901801483574709,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048387022714848406,
      "loss": 5.8949,
      "mean_token_accuracy": 0.1606071799993515,
      "num_tokens": 27224253.0,
      "step": 14610
    },
    {
      "entropy": 6.138223695755005,
      "epoch": 1.2906216884493111,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000483858516502834,
      "loss": 5.8747,
      "mean_token_accuracy": 0.16328553929924966,
      "num_tokens": 27233007.0,
      "step": 14615
    },
    {
      "entropy": 6.104684782028198,
      "epoch": 1.2910632285411516,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004838468017657644,
      "loss": 5.8134,
      "mean_token_accuracy": 0.16842241287231446,
      "num_tokens": 27242127.0,
      "step": 14620
    },
    {
      "entropy": 6.124634170532227,
      "epoch": 1.291504768632992,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048383508293750474,
      "loss": 5.974,
      "mean_token_accuracy": 0.15246982127428055,
      "num_tokens": 27252753.0,
      "step": 14625
    },
    {
      "entropy": 6.151149320602417,
      "epoch": 1.2919463087248322,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004838233600182846,
      "loss": 5.9559,
      "mean_token_accuracy": 0.15627553910017014,
      "num_tokens": 27262498.0,
      "step": 14630
    },
    {
      "entropy": 6.214147233963013,
      "epoch": 1.2923878488166727,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004838116330083336,
      "loss": 5.9429,
      "mean_token_accuracy": 0.15430187582969665,
      "num_tokens": 27271524.0,
      "step": 14635
    },
    {
      "entropy": 6.033311462402343,
      "epoch": 1.292829388908513,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004837999019078814,
      "loss": 5.7891,
      "mean_token_accuracy": 0.17727081030607222,
      "num_tokens": 27280649.0,
      "step": 14640
    },
    {
      "entropy": 6.132535409927368,
      "epoch": 1.2932709290003532,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004837881667171579,
      "loss": 5.8911,
      "mean_token_accuracy": 0.15832727998495102,
      "num_tokens": 27289593.0,
      "step": 14645
    },
    {
      "entropy": 6.154211759567261,
      "epoch": 1.2937124690921935,
      "grad_norm": 5.09375,
      "learning_rate": 0.00048377642743639297,
      "loss": 5.8838,
      "mean_token_accuracy": 0.15846479535102845,
      "num_tokens": 27298070.0,
      "step": 14650
    },
    {
      "entropy": 6.0713883399963375,
      "epoch": 1.2941540091840338,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004837646840658166,
      "loss": 5.8367,
      "mean_token_accuracy": 0.16703768968582153,
      "num_tokens": 27306738.0,
      "step": 14655
    },
    {
      "entropy": 6.057452917098999,
      "epoch": 1.2945955492758743,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048375293660565875,
      "loss": 5.806,
      "mean_token_accuracy": 0.1690780997276306,
      "num_tokens": 27315243.0,
      "step": 14660
    },
    {
      "entropy": 6.063826990127564,
      "epoch": 1.2950370893677146,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004837411850561496,
      "loss": 5.827,
      "mean_token_accuracy": 0.16601891815662384,
      "num_tokens": 27325327.0,
      "step": 14665
    },
    {
      "entropy": 6.135980796813965,
      "epoch": 1.2954786294595548,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048372942941751935,
      "loss": 5.8388,
      "mean_token_accuracy": 0.16478551179170609,
      "num_tokens": 27333436.0,
      "step": 14670
    },
    {
      "entropy": 6.142597389221192,
      "epoch": 1.2959201695513953,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004837176696899984,
      "loss": 5.9235,
      "mean_token_accuracy": 0.15309327989816665,
      "num_tokens": 27342689.0,
      "step": 14675
    },
    {
      "entropy": 6.159892749786377,
      "epoch": 1.2963617096432356,
      "grad_norm": 1.921875,
      "learning_rate": 0.00048370590587381684,
      "loss": 5.9153,
      "mean_token_accuracy": 0.16493655294179915,
      "num_tokens": 27352347.0,
      "step": 14680
    },
    {
      "entropy": 6.1565186977386475,
      "epoch": 1.2968032497350759,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048369413796920535,
      "loss": 5.9271,
      "mean_token_accuracy": 0.15912128239870071,
      "num_tokens": 27362153.0,
      "step": 14685
    },
    {
      "entropy": 6.053378105163574,
      "epoch": 1.2972447898269164,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004836823659763944,
      "loss": 5.8673,
      "mean_token_accuracy": 0.1583820253610611,
      "num_tokens": 27371126.0,
      "step": 14690
    },
    {
      "entropy": 6.070938777923584,
      "epoch": 1.2976863299187567,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048367058989561457,
      "loss": 5.8012,
      "mean_token_accuracy": 0.1651498258113861,
      "num_tokens": 27380727.0,
      "step": 14695
    },
    {
      "entropy": 6.123284435272216,
      "epoch": 1.298127870010597,
      "grad_norm": 8.75,
      "learning_rate": 0.0004836588097270966,
      "loss": 5.8562,
      "mean_token_accuracy": 0.16518203765153885,
      "num_tokens": 27389970.0,
      "step": 14700
    },
    {
      "entropy": 6.07294864654541,
      "epoch": 1.2985694101024374,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048364702547107113,
      "loss": 5.7924,
      "mean_token_accuracy": 0.16545184776186944,
      "num_tokens": 27399177.0,
      "step": 14705
    },
    {
      "entropy": 6.131297159194946,
      "epoch": 1.2990109501942777,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004836352371277692,
      "loss": 5.9119,
      "mean_token_accuracy": 0.16712456047534943,
      "num_tokens": 27408776.0,
      "step": 14710
    },
    {
      "entropy": 6.1579790115356445,
      "epoch": 1.299452490286118,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004836234446974216,
      "loss": 5.9211,
      "mean_token_accuracy": 0.15698197185993196,
      "num_tokens": 27418708.0,
      "step": 14715
    },
    {
      "entropy": 6.161049127578735,
      "epoch": 1.2998940303779583,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004836116481802593,
      "loss": 5.8676,
      "mean_token_accuracy": 0.16068847924470903,
      "num_tokens": 27428027.0,
      "step": 14720
    },
    {
      "entropy": 6.059080743789673,
      "epoch": 1.3003355704697985,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048359984757651356,
      "loss": 5.8477,
      "mean_token_accuracy": 0.167088121175766,
      "num_tokens": 27438214.0,
      "step": 14725
    },
    {
      "entropy": 6.089139747619629,
      "epoch": 1.300777110561639,
      "grad_norm": 1.8125,
      "learning_rate": 0.00048358804288641536,
      "loss": 5.8079,
      "mean_token_accuracy": 0.1652594432234764,
      "num_tokens": 27447413.0,
      "step": 14730
    },
    {
      "entropy": 6.043606090545654,
      "epoch": 1.3012186506534793,
      "grad_norm": 2.0,
      "learning_rate": 0.00048357623411019607,
      "loss": 5.8367,
      "mean_token_accuracy": 0.16637206077575684,
      "num_tokens": 27456316.0,
      "step": 14735
    },
    {
      "entropy": 6.027954626083374,
      "epoch": 1.3016601907453196,
      "grad_norm": 2.21875,
      "learning_rate": 0.000483564421248087,
      "loss": 5.8795,
      "mean_token_accuracy": 0.16721034198999404,
      "num_tokens": 27465346.0,
      "step": 14740
    },
    {
      "entropy": 6.0546328067779545,
      "epoch": 1.30210173083716,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048355260430031947,
      "loss": 5.734,
      "mean_token_accuracy": 0.1723693937063217,
      "num_tokens": 27473367.0,
      "step": 14745
    },
    {
      "entropy": 6.189047908782959,
      "epoch": 1.3025432709290004,
      "grad_norm": 2.203125,
      "learning_rate": 0.000483540783267125,
      "loss": 5.9092,
      "mean_token_accuracy": 0.15992856174707412,
      "num_tokens": 27482187.0,
      "step": 14750
    },
    {
      "entropy": 6.041848230361938,
      "epoch": 1.3029848110208406,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048352895814873526,
      "loss": 5.8628,
      "mean_token_accuracy": 0.16338201314210893,
      "num_tokens": 27492406.0,
      "step": 14755
    },
    {
      "entropy": 6.137863254547119,
      "epoch": 1.3034263511126811,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004835171289453818,
      "loss": 5.9298,
      "mean_token_accuracy": 0.166038478910923,
      "num_tokens": 27501929.0,
      "step": 14760
    },
    {
      "entropy": 6.13617787361145,
      "epoch": 1.3038678912045214,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004835052956572964,
      "loss": 5.7808,
      "mean_token_accuracy": 0.1754772201180458,
      "num_tokens": 27510657.0,
      "step": 14765
    },
    {
      "entropy": 6.070120143890381,
      "epoch": 1.3043094312963617,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048349345828471084,
      "loss": 5.9116,
      "mean_token_accuracy": 0.1547280415892601,
      "num_tokens": 27519726.0,
      "step": 14770
    },
    {
      "entropy": 6.0759790420532225,
      "epoch": 1.3047509713882022,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048348161682785704,
      "loss": 5.8346,
      "mean_token_accuracy": 0.16799575984477996,
      "num_tokens": 27528876.0,
      "step": 14775
    },
    {
      "entropy": 6.152000617980957,
      "epoch": 1.3051925114800424,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004834697712869669,
      "loss": 5.8871,
      "mean_token_accuracy": 0.16133493781089783,
      "num_tokens": 27537782.0,
      "step": 14780
    },
    {
      "entropy": 6.131542634963989,
      "epoch": 1.3056340515718827,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004834579216622725,
      "loss": 5.91,
      "mean_token_accuracy": 0.15888519287109376,
      "num_tokens": 27547322.0,
      "step": 14785
    },
    {
      "entropy": 6.066386461257935,
      "epoch": 1.306075591663723,
      "grad_norm": 1.734375,
      "learning_rate": 0.000483446067954006,
      "loss": 5.8444,
      "mean_token_accuracy": 0.16652489006519317,
      "num_tokens": 27556070.0,
      "step": 14790
    },
    {
      "entropy": 6.074157190322876,
      "epoch": 1.3065171317555633,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00048343421016239945,
      "loss": 5.7794,
      "mean_token_accuracy": 0.17245520651340485,
      "num_tokens": 27565951.0,
      "step": 14795
    },
    {
      "entropy": 6.137390661239624,
      "epoch": 1.3069586718474038,
      "grad_norm": 2.25,
      "learning_rate": 0.0004834223482876854,
      "loss": 5.8563,
      "mean_token_accuracy": 0.16035415083169938,
      "num_tokens": 27575102.0,
      "step": 14800
    },
    {
      "entropy": 6.1102954864501955,
      "epoch": 1.307400211939244,
      "grad_norm": 2.375,
      "learning_rate": 0.0004834104823300961,
      "loss": 5.9327,
      "mean_token_accuracy": 0.15517421066761017,
      "num_tokens": 27584854.0,
      "step": 14805
    },
    {
      "entropy": 6.087147808074951,
      "epoch": 1.3078417520310843,
      "grad_norm": 1.8125,
      "learning_rate": 0.00048339861228986395,
      "loss": 5.8791,
      "mean_token_accuracy": 0.16814199835062027,
      "num_tokens": 27593403.0,
      "step": 14810
    },
    {
      "entropy": 6.150009822845459,
      "epoch": 1.3082832921229248,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048338673816722145,
      "loss": 5.9597,
      "mean_token_accuracy": 0.16064391434192657,
      "num_tokens": 27602162.0,
      "step": 14815
    },
    {
      "entropy": 6.095381069183349,
      "epoch": 1.308724832214765,
      "grad_norm": 3.234375,
      "learning_rate": 0.00048337485996240127,
      "loss": 5.9436,
      "mean_token_accuracy": 0.15897045135498047,
      "num_tokens": 27611433.0,
      "step": 14820
    },
    {
      "entropy": 6.117966175079346,
      "epoch": 1.3091663723066054,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004833629776756362,
      "loss": 5.7991,
      "mean_token_accuracy": 0.16362854093313217,
      "num_tokens": 27620443.0,
      "step": 14825
    },
    {
      "entropy": 6.164895009994507,
      "epoch": 1.3096079123984459,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048335109130715875,
      "loss": 5.8608,
      "mean_token_accuracy": 0.16003518104553222,
      "num_tokens": 27629507.0,
      "step": 14830
    },
    {
      "entropy": 6.107555246353149,
      "epoch": 1.3100494524902861,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004833392008572021,
      "loss": 5.8665,
      "mean_token_accuracy": 0.16597609370946884,
      "num_tokens": 27638355.0,
      "step": 14835
    },
    {
      "entropy": 6.1141993522644045,
      "epoch": 1.3104909925821264,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048332730632599883,
      "loss": 5.9061,
      "mean_token_accuracy": 0.16002965420484544,
      "num_tokens": 27649074.0,
      "step": 14840
    },
    {
      "entropy": 6.130582714080811,
      "epoch": 1.310932532673967,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004833154077137821,
      "loss": 5.9186,
      "mean_token_accuracy": 0.16170259714126586,
      "num_tokens": 27658564.0,
      "step": 14845
    },
    {
      "entropy": 6.117284917831421,
      "epoch": 1.3113740727658072,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048330350502078515,
      "loss": 5.8453,
      "mean_token_accuracy": 0.1653812810778618,
      "num_tokens": 27667963.0,
      "step": 14850
    },
    {
      "entropy": 6.132467746734619,
      "epoch": 1.3118156128576475,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048329159824724087,
      "loss": 5.9281,
      "mean_token_accuracy": 0.1623508483171463,
      "num_tokens": 27677528.0,
      "step": 14855
    },
    {
      "entropy": 6.1240575313568115,
      "epoch": 1.3122571529494877,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048327968739338275,
      "loss": 5.8323,
      "mean_token_accuracy": 0.17114934474229812,
      "num_tokens": 27687659.0,
      "step": 14860
    },
    {
      "entropy": 6.027389669418335,
      "epoch": 1.312698693041328,
      "grad_norm": 3.546875,
      "learning_rate": 0.00048326777245944396,
      "loss": 5.7765,
      "mean_token_accuracy": 0.1733567014336586,
      "num_tokens": 27697222.0,
      "step": 14865
    },
    {
      "entropy": 6.070391416549683,
      "epoch": 1.3131402331331685,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000483255853445658,
      "loss": 5.8606,
      "mean_token_accuracy": 0.16391192525625228,
      "num_tokens": 27705664.0,
      "step": 14870
    },
    {
      "entropy": 6.034143924713135,
      "epoch": 1.3135817732250088,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048324393035225824,
      "loss": 5.7994,
      "mean_token_accuracy": 0.1712170198559761,
      "num_tokens": 27714879.0,
      "step": 14875
    },
    {
      "entropy": 6.087355995178223,
      "epoch": 1.314023313316849,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004832320031794783,
      "loss": 5.9291,
      "mean_token_accuracy": 0.15606244653463364,
      "num_tokens": 27724749.0,
      "step": 14880
    },
    {
      "entropy": 6.1193657398223875,
      "epoch": 1.3144648534086896,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004832200719275519,
      "loss": 5.883,
      "mean_token_accuracy": 0.1629638969898224,
      "num_tokens": 27733109.0,
      "step": 14885
    },
    {
      "entropy": 5.940599203109741,
      "epoch": 1.3149063935005298,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004832081365967128,
      "loss": 5.6814,
      "mean_token_accuracy": 0.17373916655778884,
      "num_tokens": 27742079.0,
      "step": 14890
    },
    {
      "entropy": 6.133149671554565,
      "epoch": 1.3153479335923701,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048319619718719455,
      "loss": 5.9351,
      "mean_token_accuracy": 0.15382995456457138,
      "num_tokens": 27750900.0,
      "step": 14895
    },
    {
      "entropy": 6.117257022857666,
      "epoch": 1.3157894736842106,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00048318425369923133,
      "loss": 5.8121,
      "mean_token_accuracy": 0.17089105993509293,
      "num_tokens": 27760392.0,
      "step": 14900
    },
    {
      "entropy": 6.077043914794922,
      "epoch": 1.316231013776051,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004831723061330569,
      "loss": 5.8621,
      "mean_token_accuracy": 0.16401863396167754,
      "num_tokens": 27768805.0,
      "step": 14905
    },
    {
      "entropy": 6.14566478729248,
      "epoch": 1.3166725538678912,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00048316035448890546,
      "loss": 5.9572,
      "mean_token_accuracy": 0.15272333323955536,
      "num_tokens": 27777165.0,
      "step": 14910
    },
    {
      "entropy": 6.142856931686401,
      "epoch": 1.3171140939597317,
      "grad_norm": 3.125,
      "learning_rate": 0.000483148398767011,
      "loss": 5.8941,
      "mean_token_accuracy": 0.1675186425447464,
      "num_tokens": 27786709.0,
      "step": 14915
    },
    {
      "entropy": 6.020650815963745,
      "epoch": 1.317555634051572,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048313643896760776,
      "loss": 5.8282,
      "mean_token_accuracy": 0.1644199937582016,
      "num_tokens": 27795977.0,
      "step": 14920
    },
    {
      "entropy": 6.069112348556518,
      "epoch": 1.3179971741434122,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048312447509093017,
      "loss": 5.8497,
      "mean_token_accuracy": 0.1610893115401268,
      "num_tokens": 27804838.0,
      "step": 14925
    },
    {
      "entropy": 6.1101641178131105,
      "epoch": 1.3184387142352525,
      "grad_norm": 2.5,
      "learning_rate": 0.0004831125071372124,
      "loss": 5.8168,
      "mean_token_accuracy": 0.17083584368228913,
      "num_tokens": 27813245.0,
      "step": 14930
    },
    {
      "entropy": 6.143509101867676,
      "epoch": 1.3188802543270928,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004831005351066889,
      "loss": 5.9462,
      "mean_token_accuracy": 0.1577487975358963,
      "num_tokens": 27822509.0,
      "step": 14935
    },
    {
      "entropy": 6.057284212112426,
      "epoch": 1.3193217944189333,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004830885589995944,
      "loss": 5.739,
      "mean_token_accuracy": 0.17275231033563615,
      "num_tokens": 27831140.0,
      "step": 14940
    },
    {
      "entropy": 6.113650321960449,
      "epoch": 1.3197633345107735,
      "grad_norm": 2.953125,
      "learning_rate": 0.00048307657881616333,
      "loss": 5.8685,
      "mean_token_accuracy": 0.16350204795598983,
      "num_tokens": 27839537.0,
      "step": 14945
    },
    {
      "entropy": 6.093905973434448,
      "epoch": 1.3202048746026138,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048306459455663045,
      "loss": 5.8371,
      "mean_token_accuracy": 0.16539138406515122,
      "num_tokens": 27848888.0,
      "step": 14950
    },
    {
      "entropy": 6.192476367950439,
      "epoch": 1.3206464146944543,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004830526062212305,
      "loss": 5.9064,
      "mean_token_accuracy": 0.16160476952791214,
      "num_tokens": 27858910.0,
      "step": 14955
    },
    {
      "entropy": 6.081178760528564,
      "epoch": 1.3210879547862946,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048304061381019826,
      "loss": 5.7951,
      "mean_token_accuracy": 0.16869389712810517,
      "num_tokens": 27867967.0,
      "step": 14960
    },
    {
      "entropy": 6.000652122497558,
      "epoch": 1.3215294948781349,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048302861732376875,
      "loss": 5.8202,
      "mean_token_accuracy": 0.1637197270989418,
      "num_tokens": 27877317.0,
      "step": 14965
    },
    {
      "entropy": 6.110893774032593,
      "epoch": 1.3219710349699754,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048301661676217693,
      "loss": 5.8824,
      "mean_token_accuracy": 0.16272113025188445,
      "num_tokens": 27886400.0,
      "step": 14970
    },
    {
      "entropy": 6.110931062698365,
      "epoch": 1.3224125750618156,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004830046121256579,
      "loss": 5.8643,
      "mean_token_accuracy": 0.16234778612852097,
      "num_tokens": 27895816.0,
      "step": 14975
    },
    {
      "entropy": 6.176324319839478,
      "epoch": 1.322854115153656,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004829926034144468,
      "loss": 5.9098,
      "mean_token_accuracy": 0.16054308861494065,
      "num_tokens": 27905614.0,
      "step": 14980
    },
    {
      "entropy": 6.206574010848999,
      "epoch": 1.3232956552454964,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004829805906287789,
      "loss": 5.9158,
      "mean_token_accuracy": 0.1620567411184311,
      "num_tokens": 27914714.0,
      "step": 14985
    },
    {
      "entropy": 6.07873158454895,
      "epoch": 1.3237371953373367,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004829685737688895,
      "loss": 5.8619,
      "mean_token_accuracy": 0.1676024705171585,
      "num_tokens": 27924821.0,
      "step": 14990
    },
    {
      "entropy": 6.125785398483276,
      "epoch": 1.324178735429177,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048295655283501403,
      "loss": 5.9145,
      "mean_token_accuracy": 0.1634213760495186,
      "num_tokens": 27934028.0,
      "step": 14995
    },
    {
      "entropy": 6.137412214279175,
      "epoch": 1.3246202755210172,
      "grad_norm": 1.875,
      "learning_rate": 0.000482944527827388,
      "loss": 5.8734,
      "mean_token_accuracy": 0.1577913299202919,
      "num_tokens": 27943550.0,
      "step": 15000
    },
    {
      "epoch": 1.3246202755210172,
      "eval_entropy": 5.9305301974070845,
      "eval_loss": 5.950167179107666,
      "eval_mean_token_accuracy": 0.16813784923514147,
      "eval_num_tokens": 27943550.0,
      "eval_runtime": 26.3542,
      "eval_samples_per_second": 1340.013,
      "eval_steps_per_second": 167.525,
      "step": 15000
    },
    {
      "entropy": 6.136676836013794,
      "epoch": 1.3250618156128575,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004829324987462469,
      "loss": 5.8568,
      "mean_token_accuracy": 0.1614982783794403,
      "num_tokens": 27952752.0,
      "step": 15005
    },
    {
      "entropy": 6.096404361724853,
      "epoch": 1.325503355704698,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004829204655918264,
      "loss": 5.9058,
      "mean_token_accuracy": 0.16487836539745332,
      "num_tokens": 27961535.0,
      "step": 15010
    },
    {
      "entropy": 6.122967195510864,
      "epoch": 1.3259448957965383,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048290842836436226,
      "loss": 5.8462,
      "mean_token_accuracy": 0.16730811297893525,
      "num_tokens": 27969644.0,
      "step": 15015
    },
    {
      "entropy": 6.147615957260132,
      "epoch": 1.3263864358883786,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004828963870640902,
      "loss": 5.9832,
      "mean_token_accuracy": 0.1559483677148819,
      "num_tokens": 27978859.0,
      "step": 15020
    },
    {
      "entropy": 6.185392999649048,
      "epoch": 1.326827975980219,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048288434169124625,
      "loss": 5.9069,
      "mean_token_accuracy": 0.16501372754573823,
      "num_tokens": 27988693.0,
      "step": 15025
    },
    {
      "entropy": 6.1454404354095455,
      "epoch": 1.3272695160720593,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004828722922460662,
      "loss": 5.838,
      "mean_token_accuracy": 0.17279563695192338,
      "num_tokens": 27998196.0,
      "step": 15030
    },
    {
      "entropy": 6.1566626071929935,
      "epoch": 1.3277110561638996,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048286023872878614,
      "loss": 5.9807,
      "mean_token_accuracy": 0.15159865319728852,
      "num_tokens": 28008743.0,
      "step": 15035
    },
    {
      "entropy": 6.172740125656128,
      "epoch": 1.3281525962557401,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004828481811396423,
      "loss": 5.955,
      "mean_token_accuracy": 0.15462292432785035,
      "num_tokens": 28017684.0,
      "step": 15040
    },
    {
      "entropy": 6.06938910484314,
      "epoch": 1.3285941363475804,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004828361194788708,
      "loss": 5.9038,
      "mean_token_accuracy": 0.16645537465810775,
      "num_tokens": 28027992.0,
      "step": 15045
    },
    {
      "entropy": 6.2779663562774655,
      "epoch": 1.3290356764394207,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048282405374670785,
      "loss": 5.9774,
      "mean_token_accuracy": 0.15799216330051422,
      "num_tokens": 28037457.0,
      "step": 15050
    },
    {
      "entropy": 6.156169271469116,
      "epoch": 1.3294772165312612,
      "grad_norm": 3.953125,
      "learning_rate": 0.00048281198394338986,
      "loss": 5.9466,
      "mean_token_accuracy": 0.16258110255002975,
      "num_tokens": 28046881.0,
      "step": 15055
    },
    {
      "entropy": 6.104385280609131,
      "epoch": 1.3299187566231014,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00048279991006915335,
      "loss": 5.7887,
      "mean_token_accuracy": 0.16446478366851808,
      "num_tokens": 28056932.0,
      "step": 15060
    },
    {
      "entropy": 6.137291526794433,
      "epoch": 1.3303602967149417,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048278783212423484,
      "loss": 5.9742,
      "mean_token_accuracy": 0.16084498018026352,
      "num_tokens": 28066843.0,
      "step": 15065
    },
    {
      "entropy": 6.148873281478882,
      "epoch": 1.330801836806782,
      "grad_norm": 3.875,
      "learning_rate": 0.0004827757501088708,
      "loss": 5.8659,
      "mean_token_accuracy": 0.1645237997174263,
      "num_tokens": 28076329.0,
      "step": 15070
    },
    {
      "entropy": 6.1862977027893065,
      "epoch": 1.3312433768986223,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048276366402329803,
      "loss": 5.9458,
      "mean_token_accuracy": 0.1656655862927437,
      "num_tokens": 28086402.0,
      "step": 15075
    },
    {
      "entropy": 6.170673608779907,
      "epoch": 1.3316849169904628,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004827515738677532,
      "loss": 5.8724,
      "mean_token_accuracy": 0.16392226815223693,
      "num_tokens": 28095030.0,
      "step": 15080
    },
    {
      "entropy": 6.0416021823883055,
      "epoch": 1.332126457082303,
      "grad_norm": 1.75,
      "learning_rate": 0.0004827394796424732,
      "loss": 5.8632,
      "mean_token_accuracy": 0.16345753967761995,
      "num_tokens": 28105138.0,
      "step": 15085
    },
    {
      "entropy": 6.124411916732788,
      "epoch": 1.3325679971741433,
      "grad_norm": 2.25,
      "learning_rate": 0.00048272738134769503,
      "loss": 5.8838,
      "mean_token_accuracy": 0.1591957166790962,
      "num_tokens": 28114359.0,
      "step": 15090
    },
    {
      "entropy": 6.171829414367676,
      "epoch": 1.3330095372659838,
      "grad_norm": 1.796875,
      "learning_rate": 0.00048271527898365545,
      "loss": 5.868,
      "mean_token_accuracy": 0.1697617530822754,
      "num_tokens": 28124045.0,
      "step": 15095
    },
    {
      "entropy": 6.070909881591797,
      "epoch": 1.333451077357824,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004827031725505917,
      "loss": 5.8856,
      "mean_token_accuracy": 0.16883034259080887,
      "num_tokens": 28132990.0,
      "step": 15100
    },
    {
      "entropy": 6.069250011444092,
      "epoch": 1.3338926174496644,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004826910620487411,
      "loss": 5.8463,
      "mean_token_accuracy": 0.1676555335521698,
      "num_tokens": 28142504.0,
      "step": 15105
    },
    {
      "entropy": 6.150550031661988,
      "epoch": 1.3343341575415049,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004826789474783405,
      "loss": 5.9499,
      "mean_token_accuracy": 0.16239131093025208,
      "num_tokens": 28151227.0,
      "step": 15110
    },
    {
      "entropy": 6.161597537994385,
      "epoch": 1.3347756976333451,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004826668288396276,
      "loss": 5.8522,
      "mean_token_accuracy": 0.16688032746315,
      "num_tokens": 28160244.0,
      "step": 15115
    },
    {
      "entropy": 6.074572610855102,
      "epoch": 1.3352172377251854,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00048265470613283954,
      "loss": 5.8626,
      "mean_token_accuracy": 0.16348717510700225,
      "num_tokens": 28169295.0,
      "step": 15120
    },
    {
      "entropy": 6.1810746669769285,
      "epoch": 1.335658777817026,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048264257935821394,
      "loss": 5.9231,
      "mean_token_accuracy": 0.16096550822257996,
      "num_tokens": 28179199.0,
      "step": 15125
    },
    {
      "entropy": 6.158259868621826,
      "epoch": 1.3361003179088662,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004826304485159882,
      "loss": 5.9349,
      "mean_token_accuracy": 0.16522632390260697,
      "num_tokens": 28188315.0,
      "step": 15130
    },
    {
      "entropy": 6.107345390319824,
      "epoch": 1.3365418580007065,
      "grad_norm": 2.125,
      "learning_rate": 0.00048261831360640016,
      "loss": 5.9248,
      "mean_token_accuracy": 0.16162197589874266,
      "num_tokens": 28197776.0,
      "step": 15135
    },
    {
      "entropy": 6.112831926345825,
      "epoch": 1.3369833980925467,
      "grad_norm": 2.0,
      "learning_rate": 0.00048260617462968735,
      "loss": 5.9132,
      "mean_token_accuracy": 0.16529201567173005,
      "num_tokens": 28207985.0,
      "step": 15140
    },
    {
      "entropy": 6.127112913131714,
      "epoch": 1.337424938184387,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048259403158608773,
      "loss": 5.7263,
      "mean_token_accuracy": 0.17778486013412476,
      "num_tokens": 28216321.0,
      "step": 15145
    },
    {
      "entropy": 6.167921924591065,
      "epoch": 1.3378664782762275,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004825818844758391,
      "loss": 5.9407,
      "mean_token_accuracy": 0.15713327676057814,
      "num_tokens": 28226763.0,
      "step": 15150
    },
    {
      "entropy": 6.11635947227478,
      "epoch": 1.3383080183680678,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048256973329917943,
      "loss": 5.8527,
      "mean_token_accuracy": 0.16685983389616013,
      "num_tokens": 28236123.0,
      "step": 15155
    },
    {
      "entropy": 6.156151962280274,
      "epoch": 1.338749558459908,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048255757805634665,
      "loss": 5.9557,
      "mean_token_accuracy": 0.15697105675935746,
      "num_tokens": 28245848.0,
      "step": 15160
    },
    {
      "entropy": 6.04344539642334,
      "epoch": 1.3391910985517486,
      "grad_norm": 2.828125,
      "learning_rate": 0.00048254541874757895,
      "loss": 5.9458,
      "mean_token_accuracy": 0.16697599589824677,
      "num_tokens": 28255414.0,
      "step": 15165
    },
    {
      "entropy": 6.128244924545288,
      "epoch": 1.3396326386435888,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048253325537311463,
      "loss": 5.7836,
      "mean_token_accuracy": 0.16900660395622252,
      "num_tokens": 28264162.0,
      "step": 15170
    },
    {
      "entropy": 6.113693952560425,
      "epoch": 1.3400741787354291,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048252108793319184,
      "loss": 5.8482,
      "mean_token_accuracy": 0.1691725730895996,
      "num_tokens": 28273146.0,
      "step": 15175
    },
    {
      "entropy": 6.063716888427734,
      "epoch": 1.3405157188272696,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004825089164280489,
      "loss": 5.8095,
      "mean_token_accuracy": 0.16298898607492446,
      "num_tokens": 28281865.0,
      "step": 15180
    },
    {
      "entropy": 6.079371595382691,
      "epoch": 1.3409572589191099,
      "grad_norm": 2.984375,
      "learning_rate": 0.00048249674085792433,
      "loss": 5.7647,
      "mean_token_accuracy": 0.16839976012706756,
      "num_tokens": 28291197.0,
      "step": 15185
    },
    {
      "entropy": 6.104543781280517,
      "epoch": 1.3413987990109502,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048248456122305667,
      "loss": 5.9622,
      "mean_token_accuracy": 0.15247435718774796,
      "num_tokens": 28300283.0,
      "step": 15190
    },
    {
      "entropy": 6.136643743515014,
      "epoch": 1.3418403391027907,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004824723775236844,
      "loss": 6.0144,
      "mean_token_accuracy": 0.15439607352018356,
      "num_tokens": 28309121.0,
      "step": 15195
    },
    {
      "entropy": 6.106071519851684,
      "epoch": 1.342281879194631,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004824601897600463,
      "loss": 5.8175,
      "mean_token_accuracy": 0.16596537083387375,
      "num_tokens": 28318377.0,
      "step": 15200
    },
    {
      "entropy": 6.150601053237915,
      "epoch": 1.3427234192864712,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048244799793238104,
      "loss": 5.8115,
      "mean_token_accuracy": 0.16415223330259324,
      "num_tokens": 28328341.0,
      "step": 15205
    },
    {
      "entropy": 6.202287197113037,
      "epoch": 1.3431649593783115,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004824358020409275,
      "loss": 5.995,
      "mean_token_accuracy": 0.16071592420339584,
      "num_tokens": 28337340.0,
      "step": 15210
    },
    {
      "entropy": 6.119330549240113,
      "epoch": 1.3436064994701518,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004824236020859246,
      "loss": 5.9218,
      "mean_token_accuracy": 0.16256964206695557,
      "num_tokens": 28347653.0,
      "step": 15215
    },
    {
      "entropy": 6.221112442016602,
      "epoch": 1.3440480395619923,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048241139806761124,
      "loss": 5.9578,
      "mean_token_accuracy": 0.1585196889936924,
      "num_tokens": 28356596.0,
      "step": 15220
    },
    {
      "entropy": 6.120379877090454,
      "epoch": 1.3444895796538325,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048239918998622657,
      "loss": 5.938,
      "mean_token_accuracy": 0.16215635538101197,
      "num_tokens": 28366253.0,
      "step": 15225
    },
    {
      "entropy": 6.0570769786834715,
      "epoch": 1.3449311197456728,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004823869778420097,
      "loss": 5.788,
      "mean_token_accuracy": 0.17162175476551056,
      "num_tokens": 28374815.0,
      "step": 15230
    },
    {
      "entropy": 6.193132829666138,
      "epoch": 1.3453726598375133,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048237476163519994,
      "loss": 5.8964,
      "mean_token_accuracy": 0.1565113767981529,
      "num_tokens": 28384221.0,
      "step": 15235
    },
    {
      "entropy": 6.143459272384644,
      "epoch": 1.3458141999293536,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004823625413660365,
      "loss": 5.865,
      "mean_token_accuracy": 0.16818929463624954,
      "num_tokens": 28393453.0,
      "step": 15240
    },
    {
      "entropy": 6.0584112167358395,
      "epoch": 1.3462557400211939,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048235031703475884,
      "loss": 5.789,
      "mean_token_accuracy": 0.17384424209594726,
      "num_tokens": 28402343.0,
      "step": 15245
    },
    {
      "entropy": 6.037799596786499,
      "epoch": 1.3466972801130344,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004823380886416064,
      "loss": 5.8431,
      "mean_token_accuracy": 0.16947484016418457,
      "num_tokens": 28411527.0,
      "step": 15250
    },
    {
      "entropy": 6.120686340332031,
      "epoch": 1.3471388202048746,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004823258561868187,
      "loss": 5.9594,
      "mean_token_accuracy": 0.1578349232673645,
      "num_tokens": 28422518.0,
      "step": 15255
    },
    {
      "entropy": 6.1936675071716305,
      "epoch": 1.347580360296715,
      "grad_norm": 4.34375,
      "learning_rate": 0.00048231361967063534,
      "loss": 5.8766,
      "mean_token_accuracy": 0.1585938587784767,
      "num_tokens": 28431692.0,
      "step": 15260
    },
    {
      "entropy": 6.142662429809571,
      "epoch": 1.3480219003885554,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004823013790932961,
      "loss": 5.8965,
      "mean_token_accuracy": 0.16168369501829147,
      "num_tokens": 28440304.0,
      "step": 15265
    },
    {
      "entropy": 6.169588088989258,
      "epoch": 1.3484634404803957,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004822891344550408,
      "loss": 5.9636,
      "mean_token_accuracy": 0.15273201912641526,
      "num_tokens": 28450265.0,
      "step": 15270
    },
    {
      "entropy": 6.083650541305542,
      "epoch": 1.348904980572236,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048227688575610915,
      "loss": 5.8518,
      "mean_token_accuracy": 0.16304405629634858,
      "num_tokens": 28459699.0,
      "step": 15275
    },
    {
      "entropy": 6.128487491607666,
      "epoch": 1.3493465206640762,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004822646329967413,
      "loss": 5.8776,
      "mean_token_accuracy": 0.1602812945842743,
      "num_tokens": 28468523.0,
      "step": 15280
    },
    {
      "entropy": 6.232688283920288,
      "epoch": 1.3497880607559165,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004822523761771771,
      "loss": 5.9432,
      "mean_token_accuracy": 0.15543479919433595,
      "num_tokens": 28477629.0,
      "step": 15285
    },
    {
      "entropy": 6.1279926776885985,
      "epoch": 1.350229600847757,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004822401152976568,
      "loss": 5.8125,
      "mean_token_accuracy": 0.1725485473871231,
      "num_tokens": 28486517.0,
      "step": 15290
    },
    {
      "entropy": 6.127835369110107,
      "epoch": 1.3506711409395973,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048222785035842045,
      "loss": 5.9435,
      "mean_token_accuracy": 0.1604157194495201,
      "num_tokens": 28496103.0,
      "step": 15295
    },
    {
      "entropy": 6.114990139007569,
      "epoch": 1.3511126810314376,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048221558135970834,
      "loss": 5.9022,
      "mean_token_accuracy": 0.1671994984149933,
      "num_tokens": 28505538.0,
      "step": 15300
    },
    {
      "entropy": 6.014170551300049,
      "epoch": 1.351554221123278,
      "grad_norm": 1.75,
      "learning_rate": 0.00048220330830176086,
      "loss": 5.7758,
      "mean_token_accuracy": 0.16750406473875046,
      "num_tokens": 28514322.0,
      "step": 15305
    },
    {
      "entropy": 6.1670534133911135,
      "epoch": 1.3519957612151183,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004821910311848184,
      "loss": 5.8724,
      "mean_token_accuracy": 0.16740487962961198,
      "num_tokens": 28524298.0,
      "step": 15310
    },
    {
      "entropy": 6.176371288299561,
      "epoch": 1.3524373013069586,
      "grad_norm": 1.875,
      "learning_rate": 0.0004821787500091215,
      "loss": 5.8966,
      "mean_token_accuracy": 0.16443534642457963,
      "num_tokens": 28533567.0,
      "step": 15315
    },
    {
      "entropy": 6.062994003295898,
      "epoch": 1.352878841398799,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048216646477491074,
      "loss": 5.8673,
      "mean_token_accuracy": 0.1581305205821991,
      "num_tokens": 28541746.0,
      "step": 15320
    },
    {
      "entropy": 6.107037019729614,
      "epoch": 1.3533203814906394,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004821541754824267,
      "loss": 5.9307,
      "mean_token_accuracy": 0.15478355139493943,
      "num_tokens": 28551685.0,
      "step": 15325
    },
    {
      "entropy": 6.183929395675659,
      "epoch": 1.3537619215824797,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004821418821319102,
      "loss": 5.9363,
      "mean_token_accuracy": 0.16133300215005875,
      "num_tokens": 28560935.0,
      "step": 15330
    },
    {
      "entropy": 6.146428537368775,
      "epoch": 1.3542034616743202,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004821295847236021,
      "loss": 5.7901,
      "mean_token_accuracy": 0.17096901834011077,
      "num_tokens": 28569506.0,
      "step": 15335
    },
    {
      "entropy": 6.10943636894226,
      "epoch": 1.3546450017661604,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004821172832577431,
      "loss": 5.8616,
      "mean_token_accuracy": 0.16612724661827089,
      "num_tokens": 28579792.0,
      "step": 15340
    },
    {
      "entropy": 6.146883344650268,
      "epoch": 1.3550865418580007,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004821049777345744,
      "loss": 5.9496,
      "mean_token_accuracy": 0.16473130583763124,
      "num_tokens": 28589756.0,
      "step": 15345
    },
    {
      "entropy": 6.079454660415649,
      "epoch": 1.355528081949841,
      "grad_norm": 2.078125,
      "learning_rate": 0.000482092668154337,
      "loss": 5.8572,
      "mean_token_accuracy": 0.16899795979261398,
      "num_tokens": 28599853.0,
      "step": 15350
    },
    {
      "entropy": 6.0947753429412845,
      "epoch": 1.3559696220416813,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048208035451727195,
      "loss": 5.8872,
      "mean_token_accuracy": 0.16147305965423583,
      "num_tokens": 28608576.0,
      "step": 15355
    },
    {
      "entropy": 6.184477090835571,
      "epoch": 1.3564111621335218,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004820680368236206,
      "loss": 5.9353,
      "mean_token_accuracy": 0.15532443299889565,
      "num_tokens": 28618820.0,
      "step": 15360
    },
    {
      "entropy": 6.145905685424805,
      "epoch": 1.356852702225362,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004820557150736241,
      "loss": 5.8527,
      "mean_token_accuracy": 0.1681053727865219,
      "num_tokens": 28628092.0,
      "step": 15365
    },
    {
      "entropy": 6.1453804016113285,
      "epoch": 1.3572942423172023,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048204338926752396,
      "loss": 5.9441,
      "mean_token_accuracy": 0.15953940153121948,
      "num_tokens": 28637849.0,
      "step": 15370
    },
    {
      "entropy": 6.052766561508179,
      "epoch": 1.3577357824090428,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004820310594055616,
      "loss": 5.8335,
      "mean_token_accuracy": 0.1650378569960594,
      "num_tokens": 28646957.0,
      "step": 15375
    },
    {
      "entropy": 6.150359869003296,
      "epoch": 1.358177322500883,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004820187254879784,
      "loss": 5.9161,
      "mean_token_accuracy": 0.16387245804071426,
      "num_tokens": 28655768.0,
      "step": 15380
    },
    {
      "entropy": 6.167734956741333,
      "epoch": 1.3586188625927234,
      "grad_norm": 1.859375,
      "learning_rate": 0.00048200638751501616,
      "loss": 5.903,
      "mean_token_accuracy": 0.1627979800105095,
      "num_tokens": 28664719.0,
      "step": 15385
    },
    {
      "entropy": 6.112593650817871,
      "epoch": 1.3590604026845639,
      "grad_norm": 2.125,
      "learning_rate": 0.0004819940454869166,
      "loss": 5.8764,
      "mean_token_accuracy": 0.1607330784201622,
      "num_tokens": 28673568.0,
      "step": 15390
    },
    {
      "entropy": 6.055149078369141,
      "epoch": 1.3595019427764041,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004819816994039213,
      "loss": 5.851,
      "mean_token_accuracy": 0.16435939371585845,
      "num_tokens": 28682558.0,
      "step": 15395
    },
    {
      "entropy": 6.169664573669434,
      "epoch": 1.3599434828682444,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048196934926627227,
      "loss": 5.9203,
      "mean_token_accuracy": 0.16732561886310576,
      "num_tokens": 28691965.0,
      "step": 15400
    },
    {
      "entropy": 6.110429286956787,
      "epoch": 1.360385022960085,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004819569950742114,
      "loss": 5.8172,
      "mean_token_accuracy": 0.16938089728355407,
      "num_tokens": 28700805.0,
      "step": 15405
    },
    {
      "entropy": 6.076837396621704,
      "epoch": 1.3608265630519252,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004819446368279808,
      "loss": 5.8199,
      "mean_token_accuracy": 0.1727551817893982,
      "num_tokens": 28710169.0,
      "step": 15410
    },
    {
      "entropy": 6.174251508712769,
      "epoch": 1.3612681031437655,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00048193227452782237,
      "loss": 5.9944,
      "mean_token_accuracy": 0.1586844727396965,
      "num_tokens": 28718656.0,
      "step": 15415
    },
    {
      "entropy": 6.184745740890503,
      "epoch": 1.3617096432356057,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004819199081739783,
      "loss": 5.9244,
      "mean_token_accuracy": 0.16398369818925856,
      "num_tokens": 28728546.0,
      "step": 15420
    },
    {
      "entropy": 6.247153902053833,
      "epoch": 1.362151183327446,
      "grad_norm": 2.890625,
      "learning_rate": 0.000481907537766691,
      "loss": 5.8272,
      "mean_token_accuracy": 0.16780071705579758,
      "num_tokens": 28737267.0,
      "step": 15425
    },
    {
      "entropy": 6.135275363922119,
      "epoch": 1.3625927234192865,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004818951633062027,
      "loss": 5.9032,
      "mean_token_accuracy": 0.16297065764665603,
      "num_tokens": 28746721.0,
      "step": 15430
    },
    {
      "entropy": 6.1039337635040285,
      "epoch": 1.3630342635111268,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048188278479275584,
      "loss": 5.9492,
      "mean_token_accuracy": 0.1639252483844757,
      "num_tokens": 28756099.0,
      "step": 15435
    },
    {
      "entropy": 6.100718641281128,
      "epoch": 1.363475803602967,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004818704022265928,
      "loss": 5.8669,
      "mean_token_accuracy": 0.16386017203330994,
      "num_tokens": 28765463.0,
      "step": 15440
    },
    {
      "entropy": 6.087878799438476,
      "epoch": 1.3639173436948075,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048185801560795627,
      "loss": 5.851,
      "mean_token_accuracy": 0.16973695158958435,
      "num_tokens": 28775284.0,
      "step": 15445
    },
    {
      "entropy": 6.219679975509644,
      "epoch": 1.3643588837866478,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048184562493708895,
      "loss": 5.8971,
      "mean_token_accuracy": 0.17263949066400527,
      "num_tokens": 28784524.0,
      "step": 15450
    },
    {
      "entropy": 6.146704530715942,
      "epoch": 1.364800423878488,
      "grad_norm": 3.703125,
      "learning_rate": 0.00048183323021423334,
      "loss": 5.853,
      "mean_token_accuracy": 0.16110181361436843,
      "num_tokens": 28792942.0,
      "step": 15455
    },
    {
      "entropy": 6.101429510116577,
      "epoch": 1.3652419639703286,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048182083143963247,
      "loss": 5.8413,
      "mean_token_accuracy": 0.16819911748170852,
      "num_tokens": 28801746.0,
      "step": 15460
    },
    {
      "entropy": 6.083650064468384,
      "epoch": 1.3656835040621689,
      "grad_norm": 2.25,
      "learning_rate": 0.0004818084286135291,
      "loss": 5.8796,
      "mean_token_accuracy": 0.16120848655700684,
      "num_tokens": 28810276.0,
      "step": 15465
    },
    {
      "entropy": 6.152009201049805,
      "epoch": 1.3661250441540091,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004817960217361663,
      "loss": 5.9182,
      "mean_token_accuracy": 0.16191227585077286,
      "num_tokens": 28819545.0,
      "step": 15470
    },
    {
      "entropy": 6.149415922164917,
      "epoch": 1.3665665842458496,
      "grad_norm": 3.78125,
      "learning_rate": 0.000481783610807787,
      "loss": 5.7815,
      "mean_token_accuracy": 0.16721295565366745,
      "num_tokens": 28828163.0,
      "step": 15475
    },
    {
      "entropy": 6.099937772750854,
      "epoch": 1.36700812433769,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004817711958286343,
      "loss": 5.9088,
      "mean_token_accuracy": 0.1608690768480301,
      "num_tokens": 28837952.0,
      "step": 15480
    },
    {
      "entropy": 6.149483394622803,
      "epoch": 1.3674496644295302,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004817587767989516,
      "loss": 5.9532,
      "mean_token_accuracy": 0.15384592562913896,
      "num_tokens": 28847702.0,
      "step": 15485
    },
    {
      "entropy": 6.207688617706299,
      "epoch": 1.3678912045213705,
      "grad_norm": 2.375,
      "learning_rate": 0.00048174635371898197,
      "loss": 5.9072,
      "mean_token_accuracy": 0.15858109593391417,
      "num_tokens": 28856936.0,
      "step": 15490
    },
    {
      "entropy": 6.125268077850341,
      "epoch": 1.3683327446132107,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004817339265889689,
      "loss": 5.8348,
      "mean_token_accuracy": 0.16983504444360734,
      "num_tokens": 28867072.0,
      "step": 15495
    },
    {
      "entropy": 6.016877937316894,
      "epoch": 1.3687742847050512,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004817214954091557,
      "loss": 5.7741,
      "mean_token_accuracy": 0.16902895718812944,
      "num_tokens": 28877546.0,
      "step": 15500
    },
    {
      "entropy": 6.136899995803833,
      "epoch": 1.3692158247968915,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00048170906017978605,
      "loss": 5.8626,
      "mean_token_accuracy": 0.16621137261390687,
      "num_tokens": 28886804.0,
      "step": 15505
    },
    {
      "entropy": 6.126304960250854,
      "epoch": 1.3696573648887318,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004816966209011034,
      "loss": 5.9334,
      "mean_token_accuracy": 0.15568396598100662,
      "num_tokens": 28896476.0,
      "step": 15510
    },
    {
      "entropy": 6.164620733261108,
      "epoch": 1.3700989049805723,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048168417757335155,
      "loss": 5.9145,
      "mean_token_accuracy": 0.16056760847568513,
      "num_tokens": 28906269.0,
      "step": 15515
    },
    {
      "entropy": 6.093124723434448,
      "epoch": 1.3705404450724126,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004816717301967743,
      "loss": 5.8443,
      "mean_token_accuracy": 0.1644322782754898,
      "num_tokens": 28915792.0,
      "step": 15520
    },
    {
      "entropy": 6.099399518966675,
      "epoch": 1.3709819851642528,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004816592787716152,
      "loss": 5.8848,
      "mean_token_accuracy": 0.15966594815254212,
      "num_tokens": 28925065.0,
      "step": 15525
    },
    {
      "entropy": 6.111478185653686,
      "epoch": 1.3714235252560933,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048164682329811855,
      "loss": 5.8849,
      "mean_token_accuracy": 0.15538350790739058,
      "num_tokens": 28933879.0,
      "step": 15530
    },
    {
      "entropy": 6.017357444763183,
      "epoch": 1.3718650653479336,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048163436377652804,
      "loss": 5.7378,
      "mean_token_accuracy": 0.17379885613918306,
      "num_tokens": 28943176.0,
      "step": 15535
    },
    {
      "entropy": 6.168413305282593,
      "epoch": 1.372306605439774,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004816219002070879,
      "loss": 5.9189,
      "mean_token_accuracy": 0.15961471498012542,
      "num_tokens": 28952770.0,
      "step": 15540
    },
    {
      "entropy": 6.040417671203613,
      "epoch": 1.3727481455316144,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004816094325900422,
      "loss": 5.7396,
      "mean_token_accuracy": 0.17644925862550737,
      "num_tokens": 28961186.0,
      "step": 15545
    },
    {
      "entropy": 6.135634469985962,
      "epoch": 1.3731896856234547,
      "grad_norm": 2.125,
      "learning_rate": 0.00048159696092563533,
      "loss": 5.8891,
      "mean_token_accuracy": 0.16268136352300644,
      "num_tokens": 28971787.0,
      "step": 15550
    },
    {
      "entropy": 6.145185136795044,
      "epoch": 1.373631225715295,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004815844852141114,
      "loss": 5.9735,
      "mean_token_accuracy": 0.1635147973895073,
      "num_tokens": 28982095.0,
      "step": 15555
    },
    {
      "entropy": 6.083566904067993,
      "epoch": 1.3740727658071352,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048157200545571497,
      "loss": 5.8733,
      "mean_token_accuracy": 0.16704943776130676,
      "num_tokens": 28991084.0,
      "step": 15560
    },
    {
      "entropy": 6.14418249130249,
      "epoch": 1.3745143058989755,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004815595216506903,
      "loss": 5.8837,
      "mean_token_accuracy": 0.15926523953676225,
      "num_tokens": 29000544.0,
      "step": 15565
    },
    {
      "entropy": 6.1139379978179935,
      "epoch": 1.374955845990816,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004815470337992822,
      "loss": 5.7799,
      "mean_token_accuracy": 0.16660087704658508,
      "num_tokens": 29009366.0,
      "step": 15570
    },
    {
      "entropy": 6.089523553848267,
      "epoch": 1.3753973860826563,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00048153454190173517,
      "loss": 5.9387,
      "mean_token_accuracy": 0.1598697856068611,
      "num_tokens": 29018730.0,
      "step": 15575
    },
    {
      "entropy": 6.19139142036438,
      "epoch": 1.3758389261744965,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00048152204595829394,
      "loss": 5.9826,
      "mean_token_accuracy": 0.15205013900995254,
      "num_tokens": 29030255.0,
      "step": 15580
    },
    {
      "entropy": 6.180130243301392,
      "epoch": 1.376280466266337,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004815095459692032,
      "loss": 5.8089,
      "mean_token_accuracy": 0.16379355639219284,
      "num_tokens": 29039030.0,
      "step": 15585
    },
    {
      "entropy": 6.137860822677612,
      "epoch": 1.3767220063581773,
      "grad_norm": 2.0625,
      "learning_rate": 0.000481497041934708,
      "loss": 5.8469,
      "mean_token_accuracy": 0.16842200309038163,
      "num_tokens": 29047504.0,
      "step": 15590
    },
    {
      "entropy": 6.1818489074707035,
      "epoch": 1.3771635464500176,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048148453385505317,
      "loss": 5.9584,
      "mean_token_accuracy": 0.1618385136127472,
      "num_tokens": 29056488.0,
      "step": 15595
    },
    {
      "entropy": 6.126311349868774,
      "epoch": 1.377605086541858,
      "grad_norm": 2.828125,
      "learning_rate": 0.00048147202173048366,
      "loss": 5.8963,
      "mean_token_accuracy": 0.17358059734106063,
      "num_tokens": 29066067.0,
      "step": 15600
    },
    {
      "entropy": 6.190014839172363,
      "epoch": 1.3780466266336984,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004814595055612447,
      "loss": 5.8801,
      "mean_token_accuracy": 0.16412414759397506,
      "num_tokens": 29075136.0,
      "step": 15605
    },
    {
      "entropy": 6.142491006851197,
      "epoch": 1.3784881667255386,
      "grad_norm": 2.484375,
      "learning_rate": 0.00048144698534758153,
      "loss": 5.8593,
      "mean_token_accuracy": 0.16961864978075028,
      "num_tokens": 29084970.0,
      "step": 15610
    },
    {
      "entropy": 6.049011945724487,
      "epoch": 1.3789297068173791,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048143446108973924,
      "loss": 5.7912,
      "mean_token_accuracy": 0.16672810167074203,
      "num_tokens": 29093234.0,
      "step": 15615
    },
    {
      "entropy": 6.117179298400879,
      "epoch": 1.3793712469092194,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00048142193278796335,
      "loss": 5.8387,
      "mean_token_accuracy": 0.16513116359710694,
      "num_tokens": 29102115.0,
      "step": 15620
    },
    {
      "entropy": 6.1502272605896,
      "epoch": 1.3798127870010597,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004814094004424992,
      "loss": 5.8581,
      "mean_token_accuracy": 0.17045588046312332,
      "num_tokens": 29111065.0,
      "step": 15625
    },
    {
      "entropy": 6.079172277450562,
      "epoch": 1.3802543270929,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004813968640535922,
      "loss": 5.8835,
      "mean_token_accuracy": 0.16478243172168733,
      "num_tokens": 29120238.0,
      "step": 15630
    },
    {
      "entropy": 6.102296543121338,
      "epoch": 1.3806958671847402,
      "grad_norm": 2.359375,
      "learning_rate": 0.000481384323621488,
      "loss": 5.8132,
      "mean_token_accuracy": 0.16345491707324983,
      "num_tokens": 29129062.0,
      "step": 15635
    },
    {
      "entropy": 6.158713245391846,
      "epoch": 1.3811374072765807,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004813717791464324,
      "loss": 5.9323,
      "mean_token_accuracy": 0.15704514384269713,
      "num_tokens": 29138990.0,
      "step": 15640
    },
    {
      "entropy": 6.127796983718872,
      "epoch": 1.381578947368421,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048135923062867094,
      "loss": 5.8188,
      "mean_token_accuracy": 0.16983270049095153,
      "num_tokens": 29147507.0,
      "step": 15645
    },
    {
      "entropy": 6.082017660140991,
      "epoch": 1.3820204874602613,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004813466780684495,
      "loss": 5.9282,
      "mean_token_accuracy": 0.16170967668294906,
      "num_tokens": 29156277.0,
      "step": 15650
    },
    {
      "entropy": 6.123262357711792,
      "epoch": 1.3824620275521018,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004813341214660141,
      "loss": 6.0291,
      "mean_token_accuracy": 0.15477319806814194,
      "num_tokens": 29165925.0,
      "step": 15655
    },
    {
      "entropy": 6.257296180725097,
      "epoch": 1.382903567643942,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004813215608216105,
      "loss": 5.9302,
      "mean_token_accuracy": 0.16036395877599716,
      "num_tokens": 29175811.0,
      "step": 15660
    },
    {
      "entropy": 6.1712291717529295,
      "epoch": 1.3833451077357823,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048130899613548487,
      "loss": 5.927,
      "mean_token_accuracy": 0.16482851207256316,
      "num_tokens": 29185100.0,
      "step": 15665
    },
    {
      "entropy": 6.145036268234253,
      "epoch": 1.3837866478276228,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048129642740788337,
      "loss": 5.9026,
      "mean_token_accuracy": 0.1651301220059395,
      "num_tokens": 29194430.0,
      "step": 15670
    },
    {
      "entropy": 6.123716020584107,
      "epoch": 1.3842281879194631,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004812838546390521,
      "loss": 5.8559,
      "mean_token_accuracy": 0.16822090446949006,
      "num_tokens": 29203897.0,
      "step": 15675
    },
    {
      "entropy": 6.1042036533355715,
      "epoch": 1.3846697280113034,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004812712778292375,
      "loss": 5.9788,
      "mean_token_accuracy": 0.16073347479104996,
      "num_tokens": 29213474.0,
      "step": 15680
    },
    {
      "entropy": 6.160219812393189,
      "epoch": 1.3851112681031439,
      "grad_norm": 2.6875,
      "learning_rate": 0.000481258696978686,
      "loss": 5.9278,
      "mean_token_accuracy": 0.16495075523853303,
      "num_tokens": 29222176.0,
      "step": 15685
    },
    {
      "entropy": 6.090711545944214,
      "epoch": 1.3855528081949842,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048124611208764375,
      "loss": 5.8691,
      "mean_token_accuracy": 0.15829688012599946,
      "num_tokens": 29231885.0,
      "step": 15690
    },
    {
      "entropy": 6.103654909133911,
      "epoch": 1.3859943482868244,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004812335231563576,
      "loss": 5.8659,
      "mean_token_accuracy": 0.1671243578195572,
      "num_tokens": 29241921.0,
      "step": 15695
    },
    {
      "entropy": 6.104442024230957,
      "epoch": 1.3864358883786647,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048122093018507396,
      "loss": 5.9392,
      "mean_token_accuracy": 0.16232561618089675,
      "num_tokens": 29251308.0,
      "step": 15700
    },
    {
      "entropy": 6.141607093811035,
      "epoch": 1.386877428470505,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004812083331740396,
      "loss": 5.8688,
      "mean_token_accuracy": 0.1718305304646492,
      "num_tokens": 29260628.0,
      "step": 15705
    },
    {
      "entropy": 6.209476518630981,
      "epoch": 1.3873189685623455,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048119573212350127,
      "loss": 5.883,
      "mean_token_accuracy": 0.15705794990062713,
      "num_tokens": 29269530.0,
      "step": 15710
    },
    {
      "entropy": 6.101268911361695,
      "epoch": 1.3877605086541858,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048118312703370576,
      "loss": 5.9092,
      "mean_token_accuracy": 0.15910112708806992,
      "num_tokens": 29279846.0,
      "step": 15715
    },
    {
      "entropy": 6.127383899688721,
      "epoch": 1.388202048746026,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048117051790490007,
      "loss": 5.8928,
      "mean_token_accuracy": 0.1618381842970848,
      "num_tokens": 29289822.0,
      "step": 15720
    },
    {
      "entropy": 6.120386552810669,
      "epoch": 1.3886435888378665,
      "grad_norm": 5.03125,
      "learning_rate": 0.00048115790473733124,
      "loss": 5.8872,
      "mean_token_accuracy": 0.16319378167390824,
      "num_tokens": 29299454.0,
      "step": 15725
    },
    {
      "entropy": 6.139145469665527,
      "epoch": 1.3890851289297068,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048114528753124625,
      "loss": 5.9303,
      "mean_token_accuracy": 0.15999872386455535,
      "num_tokens": 29308398.0,
      "step": 15730
    },
    {
      "entropy": 6.129378890991211,
      "epoch": 1.389526669021547,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048113266628689236,
      "loss": 5.8909,
      "mean_token_accuracy": 0.1709734857082367,
      "num_tokens": 29318968.0,
      "step": 15735
    },
    {
      "entropy": 6.196834373474121,
      "epoch": 1.3899682091133876,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004811200410045167,
      "loss": 5.9586,
      "mean_token_accuracy": 0.15296978428959845,
      "num_tokens": 29329745.0,
      "step": 15740
    },
    {
      "entropy": 6.132255268096924,
      "epoch": 1.3904097492052279,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004811074116843667,
      "loss": 5.7726,
      "mean_token_accuracy": 0.171654511988163,
      "num_tokens": 29338438.0,
      "step": 15745
    },
    {
      "entropy": 6.1217388153076175,
      "epoch": 1.3908512892970681,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048109477832668977,
      "loss": 5.8522,
      "mean_token_accuracy": 0.16359959542751312,
      "num_tokens": 29347783.0,
      "step": 15750
    },
    {
      "entropy": 6.079904079437256,
      "epoch": 1.3912928293889086,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004810821409317334,
      "loss": 5.9077,
      "mean_token_accuracy": 0.16538021564483643,
      "num_tokens": 29357062.0,
      "step": 15755
    },
    {
      "entropy": 6.092018079757691,
      "epoch": 1.391734369480749,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048106949949974494,
      "loss": 5.8927,
      "mean_token_accuracy": 0.16877734959125518,
      "num_tokens": 29366361.0,
      "step": 15760
    },
    {
      "entropy": 6.074779939651489,
      "epoch": 1.3921759095725892,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004810568540309723,
      "loss": 5.9176,
      "mean_token_accuracy": 0.1550411954522133,
      "num_tokens": 29375992.0,
      "step": 15765
    },
    {
      "entropy": 6.151559066772461,
      "epoch": 1.3926174496644295,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048104420452566303,
      "loss": 5.9655,
      "mean_token_accuracy": 0.15881908386945726,
      "num_tokens": 29386408.0,
      "step": 15770
    },
    {
      "entropy": 6.168401908874512,
      "epoch": 1.3930589897562697,
      "grad_norm": 2.125,
      "learning_rate": 0.00048103155098406505,
      "loss": 5.9577,
      "mean_token_accuracy": 0.15898243337869644,
      "num_tokens": 29395881.0,
      "step": 15775
    },
    {
      "entropy": 6.107079029083252,
      "epoch": 1.3935005298481102,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048101889340642604,
      "loss": 5.7752,
      "mean_token_accuracy": 0.16928510665893554,
      "num_tokens": 29404397.0,
      "step": 15780
    },
    {
      "entropy": 6.030300712585449,
      "epoch": 1.3939420699399505,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004810062317929941,
      "loss": 5.8583,
      "mean_token_accuracy": 0.16855769008398055,
      "num_tokens": 29413219.0,
      "step": 15785
    },
    {
      "entropy": 6.1403223991394045,
      "epoch": 1.3943836100317908,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004809935661440172,
      "loss": 5.939,
      "mean_token_accuracy": 0.15871401280164718,
      "num_tokens": 29422727.0,
      "step": 15790
    },
    {
      "entropy": 6.20472297668457,
      "epoch": 1.3948251501236313,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004809808964597436,
      "loss": 5.9203,
      "mean_token_accuracy": 0.16529241800308228,
      "num_tokens": 29432183.0,
      "step": 15795
    },
    {
      "entropy": 6.104766178131103,
      "epoch": 1.3952666902154716,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048096822274042145,
      "loss": 5.8313,
      "mean_token_accuracy": 0.16376353204250335,
      "num_tokens": 29440821.0,
      "step": 15800
    },
    {
      "entropy": 6.134339094161987,
      "epoch": 1.3957082303073118,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00048095554498629885,
      "loss": 5.964,
      "mean_token_accuracy": 0.15938032567501068,
      "num_tokens": 29450128.0,
      "step": 15805
    },
    {
      "entropy": 6.084202623367309,
      "epoch": 1.3961497703991523,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00048094286319762426,
      "loss": 5.838,
      "mean_token_accuracy": 0.16619508415460588,
      "num_tokens": 29459271.0,
      "step": 15810
    },
    {
      "entropy": 6.106001996994019,
      "epoch": 1.3965913104909926,
      "grad_norm": 7.0625,
      "learning_rate": 0.00048093017737464613,
      "loss": 5.7881,
      "mean_token_accuracy": 0.17325164377689362,
      "num_tokens": 29468517.0,
      "step": 15815
    },
    {
      "entropy": 6.10490403175354,
      "epoch": 1.3970328505828329,
      "grad_norm": 4.3125,
      "learning_rate": 0.00048091748751761295,
      "loss": 5.9612,
      "mean_token_accuracy": 0.1573534831404686,
      "num_tokens": 29478330.0,
      "step": 15820
    },
    {
      "entropy": 6.139644575119019,
      "epoch": 1.3974743906746734,
      "grad_norm": 1.75,
      "learning_rate": 0.0004809047936267732,
      "loss": 5.9082,
      "mean_token_accuracy": 0.15455982983112335,
      "num_tokens": 29487190.0,
      "step": 15825
    },
    {
      "entropy": 6.157718658447266,
      "epoch": 1.3979159307665137,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048089209570237573,
      "loss": 5.9897,
      "mean_token_accuracy": 0.15826574414968492,
      "num_tokens": 29495885.0,
      "step": 15830
    },
    {
      "entropy": 6.195947790145874,
      "epoch": 1.398357470858354,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004808793937446692,
      "loss": 5.8654,
      "mean_token_accuracy": 0.15851812362670897,
      "num_tokens": 29505524.0,
      "step": 15835
    },
    {
      "entropy": 6.11465277671814,
      "epoch": 1.3987990109501942,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004808666877539025,
      "loss": 5.8273,
      "mean_token_accuracy": 0.16863056421279907,
      "num_tokens": 29514003.0,
      "step": 15840
    },
    {
      "entropy": 6.040834665298462,
      "epoch": 1.3992405510420345,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048085397773032433,
      "loss": 5.8677,
      "mean_token_accuracy": 0.16476200222969056,
      "num_tokens": 29523236.0,
      "step": 15845
    },
    {
      "entropy": 6.117907905578614,
      "epoch": 1.399682091133875,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004808412636741839,
      "loss": 5.9107,
      "mean_token_accuracy": 0.16730762124061585,
      "num_tokens": 29532852.0,
      "step": 15850
    },
    {
      "entropy": 6.094393968582153,
      "epoch": 1.4001236312257153,
      "grad_norm": 3.875,
      "learning_rate": 0.00048082854558573017,
      "loss": 5.8581,
      "mean_token_accuracy": 0.1682341679930687,
      "num_tokens": 29542035.0,
      "step": 15855
    },
    {
      "entropy": 6.092104530334472,
      "epoch": 1.4005651713175555,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048081582346521236,
      "loss": 5.8526,
      "mean_token_accuracy": 0.1652159124612808,
      "num_tokens": 29551978.0,
      "step": 15860
    },
    {
      "entropy": 6.223153400421142,
      "epoch": 1.401006711409396,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004808030973128795,
      "loss": 5.9962,
      "mean_token_accuracy": 0.15778125822544098,
      "num_tokens": 29562011.0,
      "step": 15865
    },
    {
      "entropy": 6.125807189941407,
      "epoch": 1.4014482515012363,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048079036712898106,
      "loss": 5.8447,
      "mean_token_accuracy": 0.17101458907127381,
      "num_tokens": 29571498.0,
      "step": 15870
    },
    {
      "entropy": 6.132891702651977,
      "epoch": 1.4018897915930766,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048077763291376637,
      "loss": 5.794,
      "mean_token_accuracy": 0.1728857785463333,
      "num_tokens": 29580704.0,
      "step": 15875
    },
    {
      "entropy": 6.142060375213623,
      "epoch": 1.402331331684917,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004807648946674849,
      "loss": 5.7979,
      "mean_token_accuracy": 0.17174144089221954,
      "num_tokens": 29590066.0,
      "step": 15880
    },
    {
      "entropy": 6.029634094238281,
      "epoch": 1.4027728717767574,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004807521523903862,
      "loss": 5.9031,
      "mean_token_accuracy": 0.1633729785680771,
      "num_tokens": 29599473.0,
      "step": 15885
    },
    {
      "entropy": 6.095937871932984,
      "epoch": 1.4032144118685976,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048073940608271974,
      "loss": 5.9002,
      "mean_token_accuracy": 0.16311468780040742,
      "num_tokens": 29609371.0,
      "step": 15890
    },
    {
      "entropy": 6.1717887878417965,
      "epoch": 1.4036559519604381,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004807266557447354,
      "loss": 5.7791,
      "mean_token_accuracy": 0.1717279762029648,
      "num_tokens": 29618821.0,
      "step": 15895
    },
    {
      "entropy": 6.068723869323731,
      "epoch": 1.4040974920522784,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004807139013766829,
      "loss": 5.8071,
      "mean_token_accuracy": 0.17336508482694626,
      "num_tokens": 29627491.0,
      "step": 15900
    },
    {
      "entropy": 6.059764051437378,
      "epoch": 1.4045390321441187,
      "grad_norm": 2.0,
      "learning_rate": 0.00048070114297881205,
      "loss": 5.8965,
      "mean_token_accuracy": 0.16345272958278656,
      "num_tokens": 29637699.0,
      "step": 15905
    },
    {
      "entropy": 6.12862548828125,
      "epoch": 1.404980572235959,
      "grad_norm": 2.5,
      "learning_rate": 0.0004806883805513728,
      "loss": 5.8685,
      "mean_token_accuracy": 0.16485465168952942,
      "num_tokens": 29646439.0,
      "step": 15910
    },
    {
      "entropy": 6.143293333053589,
      "epoch": 1.4054221123277992,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004806756140946151,
      "loss": 5.9232,
      "mean_token_accuracy": 0.1546403706073761,
      "num_tokens": 29655704.0,
      "step": 15915
    },
    {
      "entropy": 6.178645420074463,
      "epoch": 1.4058636524196397,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048066284360878913,
      "loss": 5.9834,
      "mean_token_accuracy": 0.15822216868400574,
      "num_tokens": 29665163.0,
      "step": 15920
    },
    {
      "entropy": 6.149802589416504,
      "epoch": 1.40630519251148,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048065006909414507,
      "loss": 5.8916,
      "mean_token_accuracy": 0.16327484846115112,
      "num_tokens": 29674803.0,
      "step": 15925
    },
    {
      "entropy": 6.157119035720825,
      "epoch": 1.4067467326033203,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004806372905509331,
      "loss": 5.8794,
      "mean_token_accuracy": 0.16487651392817498,
      "num_tokens": 29685149.0,
      "step": 15930
    },
    {
      "entropy": 6.062006521224975,
      "epoch": 1.4071882726951608,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048062450797940347,
      "loss": 5.7738,
      "mean_token_accuracy": 0.1761588841676712,
      "num_tokens": 29694224.0,
      "step": 15935
    },
    {
      "entropy": 6.101453876495361,
      "epoch": 1.407629812787001,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004806117213798068,
      "loss": 5.8697,
      "mean_token_accuracy": 0.16886991560459136,
      "num_tokens": 29704512.0,
      "step": 15940
    },
    {
      "entropy": 6.089446544647217,
      "epoch": 1.4080713528788413,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048059893075239334,
      "loss": 5.7894,
      "mean_token_accuracy": 0.171515491604805,
      "num_tokens": 29712990.0,
      "step": 15945
    },
    {
      "entropy": 6.169225788116455,
      "epoch": 1.4085128929706818,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048058613609741374,
      "loss": 5.9304,
      "mean_token_accuracy": 0.15995712727308273,
      "num_tokens": 29722844.0,
      "step": 15950
    },
    {
      "entropy": 6.161352491378784,
      "epoch": 1.408954433062522,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004805733374151188,
      "loss": 5.8195,
      "mean_token_accuracy": 0.15880099087953567,
      "num_tokens": 29731593.0,
      "step": 15955
    },
    {
      "entropy": 6.131821632385254,
      "epoch": 1.4093959731543624,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048056053470575894,
      "loss": 5.9064,
      "mean_token_accuracy": 0.15979454666376114,
      "num_tokens": 29741052.0,
      "step": 15960
    },
    {
      "entropy": 6.136772346496582,
      "epoch": 1.4098375132462029,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048054772796958517,
      "loss": 5.8934,
      "mean_token_accuracy": 0.16224837005138398,
      "num_tokens": 29750187.0,
      "step": 15965
    },
    {
      "entropy": 6.081807851791382,
      "epoch": 1.4102790533380432,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048053491720684836,
      "loss": 5.8501,
      "mean_token_accuracy": 0.16276826560497284,
      "num_tokens": 29759574.0,
      "step": 15970
    },
    {
      "entropy": 6.058559274673462,
      "epoch": 1.4107205934298834,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004805221024177994,
      "loss": 5.8792,
      "mean_token_accuracy": 0.16131340712308884,
      "num_tokens": 29769549.0,
      "step": 15975
    },
    {
      "entropy": 6.188282012939453,
      "epoch": 1.4111621335217237,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004805092836026893,
      "loss": 5.8933,
      "mean_token_accuracy": 0.16495360136032106,
      "num_tokens": 29778643.0,
      "step": 15980
    },
    {
      "entropy": 6.140232706069947,
      "epoch": 1.411603673613564,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004804964607617693,
      "loss": 5.8593,
      "mean_token_accuracy": 0.16024386286735534,
      "num_tokens": 29788685.0,
      "step": 15985
    },
    {
      "entropy": 6.044860172271728,
      "epoch": 1.4120452137054045,
      "grad_norm": 2.828125,
      "learning_rate": 0.00048048363389529045,
      "loss": 5.8194,
      "mean_token_accuracy": 0.16884915828704833,
      "num_tokens": 29798648.0,
      "step": 15990
    },
    {
      "entropy": 6.06784462928772,
      "epoch": 1.4124867537972448,
      "grad_norm": 3.09375,
      "learning_rate": 0.000480470803003504,
      "loss": 5.8736,
      "mean_token_accuracy": 0.16122069507837294,
      "num_tokens": 29807582.0,
      "step": 15995
    },
    {
      "entropy": 6.118914222717285,
      "epoch": 1.412928293889085,
      "grad_norm": 4.25,
      "learning_rate": 0.0004804579680866614,
      "loss": 5.887,
      "mean_token_accuracy": 0.1623173400759697,
      "num_tokens": 29817443.0,
      "step": 16000
    },
    {
      "entropy": 6.153106307983398,
      "epoch": 1.4133698339809255,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004804451291450142,
      "loss": 5.8688,
      "mean_token_accuracy": 0.169584022462368,
      "num_tokens": 29825796.0,
      "step": 16005
    },
    {
      "entropy": 6.089987134933471,
      "epoch": 1.4138113740727658,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048043228617881365,
      "loss": 5.8263,
      "mean_token_accuracy": 0.16512692719697952,
      "num_tokens": 29834874.0,
      "step": 16010
    },
    {
      "entropy": 6.086905431747437,
      "epoch": 1.414252914164606,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004804194391883114,
      "loss": 5.883,
      "mean_token_accuracy": 0.15792076885700226,
      "num_tokens": 29843884.0,
      "step": 16015
    },
    {
      "entropy": 6.172228145599365,
      "epoch": 1.4146944542564466,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048040658817375927,
      "loss": 5.989,
      "mean_token_accuracy": 0.15276834815740586,
      "num_tokens": 29853180.0,
      "step": 16020
    },
    {
      "entropy": 6.1056303024292,
      "epoch": 1.4151359943482869,
      "grad_norm": 2.828125,
      "learning_rate": 0.00048039373313540885,
      "loss": 5.834,
      "mean_token_accuracy": 0.16269521266222,
      "num_tokens": 29862926.0,
      "step": 16025
    },
    {
      "entropy": 6.038877868652344,
      "epoch": 1.4155775344401271,
      "grad_norm": 2.015625,
      "learning_rate": 0.000480380874073512,
      "loss": 5.827,
      "mean_token_accuracy": 0.167874313890934,
      "num_tokens": 29871450.0,
      "step": 16030
    },
    {
      "entropy": 6.084260892868042,
      "epoch": 1.4160190745319676,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048036801098832067,
      "loss": 5.8208,
      "mean_token_accuracy": 0.1719295233488083,
      "num_tokens": 29880142.0,
      "step": 16035
    },
    {
      "entropy": 6.069157266616822,
      "epoch": 1.416460614623808,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004803551438800868,
      "loss": 5.8481,
      "mean_token_accuracy": 0.1637227177619934,
      "num_tokens": 29888668.0,
      "step": 16040
    },
    {
      "entropy": 6.119636392593383,
      "epoch": 1.4169021547156482,
      "grad_norm": 3.484375,
      "learning_rate": 0.00048034227274906235,
      "loss": 5.9745,
      "mean_token_accuracy": 0.16190259158611298,
      "num_tokens": 29898727.0,
      "step": 16045
    },
    {
      "entropy": 6.119180059432983,
      "epoch": 1.4173436948074885,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048032939759549964,
      "loss": 5.9034,
      "mean_token_accuracy": 0.1519879288971424,
      "num_tokens": 29909089.0,
      "step": 16050
    },
    {
      "entropy": 6.152333736419678,
      "epoch": 1.4177852348993287,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048031651841965085,
      "loss": 5.8376,
      "mean_token_accuracy": 0.16603053212165833,
      "num_tokens": 29918169.0,
      "step": 16055
    },
    {
      "entropy": 6.082733392715454,
      "epoch": 1.4182267749911692,
      "grad_norm": 2.75,
      "learning_rate": 0.0004803036352217682,
      "loss": 5.9083,
      "mean_token_accuracy": 0.17020384669303895,
      "num_tokens": 29928160.0,
      "step": 16060
    },
    {
      "entropy": 6.063603448867798,
      "epoch": 1.4186683150830095,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048029074800210413,
      "loss": 5.907,
      "mean_token_accuracy": 0.15762772262096406,
      "num_tokens": 29937695.0,
      "step": 16065
    },
    {
      "entropy": 6.173110818862915,
      "epoch": 1.4191098551748498,
      "grad_norm": 2.359375,
      "learning_rate": 0.000480277856760911,
      "loss": 5.9047,
      "mean_token_accuracy": 0.165798057615757,
      "num_tokens": 29947380.0,
      "step": 16070
    },
    {
      "entropy": 6.182863378524781,
      "epoch": 1.4195513952666903,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048026496149844146,
      "loss": 5.9704,
      "mean_token_accuracy": 0.15434986352920532,
      "num_tokens": 29956313.0,
      "step": 16075
    },
    {
      "entropy": 6.070312547683716,
      "epoch": 1.4199929353585306,
      "grad_norm": 3.28125,
      "learning_rate": 0.00048025206221494806,
      "loss": 5.8252,
      "mean_token_accuracy": 0.16633400917053223,
      "num_tokens": 29965502.0,
      "step": 16080
    },
    {
      "entropy": 6.105660915374756,
      "epoch": 1.4204344754503708,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004802391589106835,
      "loss": 5.8046,
      "mean_token_accuracy": 0.17077498584985734,
      "num_tokens": 29974243.0,
      "step": 16085
    },
    {
      "entropy": 6.087031030654908,
      "epoch": 1.4208760155422113,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004802262515859005,
      "loss": 5.7705,
      "mean_token_accuracy": 0.1746451422572136,
      "num_tokens": 29983966.0,
      "step": 16090
    },
    {
      "entropy": 6.040146780014038,
      "epoch": 1.4213175556340516,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004802133402408519,
      "loss": 5.8677,
      "mean_token_accuracy": 0.16261519193649293,
      "num_tokens": 29993365.0,
      "step": 16095
    },
    {
      "entropy": 6.1389075756073,
      "epoch": 1.4217590957258919,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048020042487579076,
      "loss": 5.9019,
      "mean_token_accuracy": 0.16038127318024636,
      "num_tokens": 30001919.0,
      "step": 16100
    },
    {
      "entropy": 6.068770790100098,
      "epoch": 1.4222006358177324,
      "grad_norm": 3.703125,
      "learning_rate": 0.00048018750549097,
      "loss": 5.8438,
      "mean_token_accuracy": 0.16224301904439925,
      "num_tokens": 30011353.0,
      "step": 16105
    },
    {
      "entropy": 6.125130319595337,
      "epoch": 1.4226421759095726,
      "grad_norm": 2.375,
      "learning_rate": 0.00048017458208664265,
      "loss": 5.9274,
      "mean_token_accuracy": 0.1609000474214554,
      "num_tokens": 30019994.0,
      "step": 16110
    },
    {
      "entropy": 6.0759461402893065,
      "epoch": 1.423083716001413,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000480161654663062,
      "loss": 5.7611,
      "mean_token_accuracy": 0.16583640575408937,
      "num_tokens": 30028934.0,
      "step": 16115
    },
    {
      "entropy": 6.171993064880371,
      "epoch": 1.4235252560932532,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004801487232204811,
      "loss": 5.8166,
      "mean_token_accuracy": 0.16902725249528885,
      "num_tokens": 30038069.0,
      "step": 16120
    },
    {
      "entropy": 6.117969512939453,
      "epoch": 1.4239667961850935,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004801357877591535,
      "loss": 5.9286,
      "mean_token_accuracy": 0.1579977884888649,
      "num_tokens": 30047307.0,
      "step": 16125
    },
    {
      "entropy": 6.147633647918701,
      "epoch": 1.424408336276934,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048012284827933245,
      "loss": 5.8875,
      "mean_token_accuracy": 0.163276606798172,
      "num_tokens": 30057429.0,
      "step": 16130
    },
    {
      "entropy": 6.055267286300659,
      "epoch": 1.4248498763687742,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004801099047812715,
      "loss": 5.8298,
      "mean_token_accuracy": 0.16643744856119155,
      "num_tokens": 30067327.0,
      "step": 16135
    },
    {
      "entropy": 6.137884140014648,
      "epoch": 1.4252914164606145,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004800969572652241,
      "loss": 5.939,
      "mean_token_accuracy": 0.15360855609178542,
      "num_tokens": 30077103.0,
      "step": 16140
    },
    {
      "entropy": 6.189132404327393,
      "epoch": 1.425732956552455,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048008400573144394,
      "loss": 5.9865,
      "mean_token_accuracy": 0.15452441275119783,
      "num_tokens": 30087539.0,
      "step": 16145
    },
    {
      "entropy": 6.193690538406372,
      "epoch": 1.4261744966442953,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048007105018018487,
      "loss": 5.8607,
      "mean_token_accuracy": 0.16532308608293533,
      "num_tokens": 30097153.0,
      "step": 16150
    },
    {
      "entropy": 6.110301160812378,
      "epoch": 1.4266160367361356,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048005809061170044,
      "loss": 5.9386,
      "mean_token_accuracy": 0.16120788007974624,
      "num_tokens": 30106291.0,
      "step": 16155
    },
    {
      "entropy": 6.072012138366699,
      "epoch": 1.427057576827976,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004800451270262447,
      "loss": 5.9576,
      "mean_token_accuracy": 0.1584857687354088,
      "num_tokens": 30116441.0,
      "step": 16160
    },
    {
      "entropy": 6.059629201889038,
      "epoch": 1.4274991169198163,
      "grad_norm": 2.5,
      "learning_rate": 0.0004800321594240714,
      "loss": 5.8184,
      "mean_token_accuracy": 0.17985545396804808,
      "num_tokens": 30125040.0,
      "step": 16165
    },
    {
      "entropy": 6.156841659545899,
      "epoch": 1.4279406570116566,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004800191878054349,
      "loss": 5.8119,
      "mean_token_accuracy": 0.16971911638975143,
      "num_tokens": 30134284.0,
      "step": 16170
    },
    {
      "entropy": 6.210902833938599,
      "epoch": 1.4283821971034971,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048000621217058904,
      "loss": 5.7915,
      "mean_token_accuracy": 0.17633806467056273,
      "num_tokens": 30143679.0,
      "step": 16175
    },
    {
      "entropy": 6.041277647018433,
      "epoch": 1.4288237371953374,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047999323251978805,
      "loss": 5.7475,
      "mean_token_accuracy": 0.1762945219874382,
      "num_tokens": 30152011.0,
      "step": 16180
    },
    {
      "entropy": 5.981799173355102,
      "epoch": 1.4292652772871777,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047998024885328625,
      "loss": 5.8556,
      "mean_token_accuracy": 0.161075459420681,
      "num_tokens": 30161949.0,
      "step": 16185
    },
    {
      "entropy": 6.13063154220581,
      "epoch": 1.429706817379018,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047996726117133795,
      "loss": 5.8689,
      "mean_token_accuracy": 0.16735961139202118,
      "num_tokens": 30172003.0,
      "step": 16190
    },
    {
      "entropy": 6.0955297470092775,
      "epoch": 1.4301483574708582,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004799542694741975,
      "loss": 5.8351,
      "mean_token_accuracy": 0.177482707798481,
      "num_tokens": 30181918.0,
      "step": 16195
    },
    {
      "entropy": 6.132780885696411,
      "epoch": 1.4305898975626987,
      "grad_norm": 4.09375,
      "learning_rate": 0.00047994127376211953,
      "loss": 5.8198,
      "mean_token_accuracy": 0.16379327476024627,
      "num_tokens": 30191032.0,
      "step": 16200
    },
    {
      "entropy": 6.074625158309937,
      "epoch": 1.431031437654539,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004799282740353586,
      "loss": 5.8293,
      "mean_token_accuracy": 0.16282529830932618,
      "num_tokens": 30200276.0,
      "step": 16205
    },
    {
      "entropy": 6.1657298564910885,
      "epoch": 1.4314729777463793,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047991527029416924,
      "loss": 5.8847,
      "mean_token_accuracy": 0.16564644128084183,
      "num_tokens": 30208574.0,
      "step": 16210
    },
    {
      "entropy": 6.1583123207092285,
      "epoch": 1.4319145178382198,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004799022625388064,
      "loss": 5.9081,
      "mean_token_accuracy": 0.15934095978736879,
      "num_tokens": 30218912.0,
      "step": 16215
    },
    {
      "entropy": 6.178378963470459,
      "epoch": 1.43235605793006,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004798892507695247,
      "loss": 5.9667,
      "mean_token_accuracy": 0.161487440764904,
      "num_tokens": 30228325.0,
      "step": 16220
    },
    {
      "entropy": 6.113603925704956,
      "epoch": 1.4327975980219003,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004798762349865791,
      "loss": 5.8499,
      "mean_token_accuracy": 0.16566318422555923,
      "num_tokens": 30237169.0,
      "step": 16225
    },
    {
      "entropy": 6.075502252578735,
      "epoch": 1.4332391381137408,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004798632151902246,
      "loss": 5.8239,
      "mean_token_accuracy": 0.16892409324645996,
      "num_tokens": 30246452.0,
      "step": 16230
    },
    {
      "entropy": 6.127521324157715,
      "epoch": 1.433680678205581,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047985019138071616,
      "loss": 5.8753,
      "mean_token_accuracy": 0.16883043497800826,
      "num_tokens": 30256055.0,
      "step": 16235
    },
    {
      "entropy": 6.107683229446411,
      "epoch": 1.4341222182974214,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047983716355830903,
      "loss": 5.7612,
      "mean_token_accuracy": 0.1641386330127716,
      "num_tokens": 30265764.0,
      "step": 16240
    },
    {
      "entropy": 6.145148420333863,
      "epoch": 1.4345637583892619,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00047982413172325845,
      "loss": 5.8995,
      "mean_token_accuracy": 0.1677064321935177,
      "num_tokens": 30274490.0,
      "step": 16245
    },
    {
      "entropy": 6.069685363769532,
      "epoch": 1.4350052984811021,
      "grad_norm": 1.890625,
      "learning_rate": 0.00047981109587581945,
      "loss": 5.8244,
      "mean_token_accuracy": 0.17229797691106796,
      "num_tokens": 30283330.0,
      "step": 16250
    },
    {
      "entropy": 6.137584543228149,
      "epoch": 1.4354468385729424,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004797980560162477,
      "loss": 6.0319,
      "mean_token_accuracy": 0.15261051654815674,
      "num_tokens": 30293542.0,
      "step": 16255
    },
    {
      "entropy": 6.140758609771728,
      "epoch": 1.4358883786647827,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004797850121447984,
      "loss": 5.9052,
      "mean_token_accuracy": 0.1695887640118599,
      "num_tokens": 30302122.0,
      "step": 16260
    },
    {
      "entropy": 6.1034870624542235,
      "epoch": 1.436329918756623,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047977196426172725,
      "loss": 5.9441,
      "mean_token_accuracy": 0.16791289299726486,
      "num_tokens": 30311490.0,
      "step": 16265
    },
    {
      "entropy": 6.049110698699951,
      "epoch": 1.4367714588484635,
      "grad_norm": 2.0,
      "learning_rate": 0.00047975891236728973,
      "loss": 5.8266,
      "mean_token_accuracy": 0.16991001665592192,
      "num_tokens": 30321199.0,
      "step": 16270
    },
    {
      "entropy": 6.202510929107666,
      "epoch": 1.4372129989403037,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047974585646174165,
      "loss": 5.8737,
      "mean_token_accuracy": 0.16286528259515762,
      "num_tokens": 30330114.0,
      "step": 16275
    },
    {
      "entropy": 6.1361555576324465,
      "epoch": 1.437654539032144,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047973279654533866,
      "loss": 5.8739,
      "mean_token_accuracy": 0.16734967082738877,
      "num_tokens": 30339290.0,
      "step": 16280
    },
    {
      "entropy": 6.0517942905426025,
      "epoch": 1.4380960791239845,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004797197326183365,
      "loss": 5.8312,
      "mean_token_accuracy": 0.16903034448623658,
      "num_tokens": 30349595.0,
      "step": 16285
    },
    {
      "entropy": 6.107483720779419,
      "epoch": 1.4385376192158248,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004797066646809914,
      "loss": 5.8897,
      "mean_token_accuracy": 0.16099362969398498,
      "num_tokens": 30358010.0,
      "step": 16290
    },
    {
      "entropy": 6.165457391738892,
      "epoch": 1.438979159307665,
      "grad_norm": 2.9375,
      "learning_rate": 0.00047969359273355907,
      "loss": 5.9627,
      "mean_token_accuracy": 0.16302094161510466,
      "num_tokens": 30367408.0,
      "step": 16295
    },
    {
      "entropy": 6.115442895889283,
      "epoch": 1.4394206993995056,
      "grad_norm": 1.9375,
      "learning_rate": 0.00047968051677629565,
      "loss": 5.9431,
      "mean_token_accuracy": 0.16147204637527465,
      "num_tokens": 30376475.0,
      "step": 16300
    },
    {
      "entropy": 6.145323991775513,
      "epoch": 1.4398622394913458,
      "grad_norm": 2.375,
      "learning_rate": 0.00047966743680945734,
      "loss": 5.8646,
      "mean_token_accuracy": 0.16147609949111938,
      "num_tokens": 30385361.0,
      "step": 16305
    },
    {
      "entropy": 6.148641633987427,
      "epoch": 1.4403037795831861,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004796543528333004,
      "loss": 5.891,
      "mean_token_accuracy": 0.1635327234864235,
      "num_tokens": 30393594.0,
      "step": 16310
    },
    {
      "entropy": 6.074825763702393,
      "epoch": 1.4407453196750266,
      "grad_norm": 2.53125,
      "learning_rate": 0.000479641264848081,
      "loss": 5.8718,
      "mean_token_accuracy": 0.16489749699831008,
      "num_tokens": 30403444.0,
      "step": 16315
    },
    {
      "entropy": 6.020571231842041,
      "epoch": 1.441186859766867,
      "grad_norm": 2.25,
      "learning_rate": 0.0004796281728540556,
      "loss": 5.7725,
      "mean_token_accuracy": 0.16638872027397156,
      "num_tokens": 30412485.0,
      "step": 16320
    },
    {
      "entropy": 6.136004114151001,
      "epoch": 1.4416283998587072,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00047961507685148077,
      "loss": 5.818,
      "mean_token_accuracy": 0.16860155612230301,
      "num_tokens": 30421104.0,
      "step": 16325
    },
    {
      "entropy": 6.076471138000488,
      "epoch": 1.4420699399505474,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047960197684061287,
      "loss": 5.844,
      "mean_token_accuracy": 0.16615869700908661,
      "num_tokens": 30430061.0,
      "step": 16330
    },
    {
      "entropy": 6.088811635971069,
      "epoch": 1.4425114800423877,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004795888728217086,
      "loss": 5.7529,
      "mean_token_accuracy": 0.16537652760744095,
      "num_tokens": 30437645.0,
      "step": 16335
    },
    {
      "entropy": 5.995330095291138,
      "epoch": 1.4429530201342282,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00047957576479502466,
      "loss": 5.7425,
      "mean_token_accuracy": 0.17875225841999054,
      "num_tokens": 30447805.0,
      "step": 16340
    },
    {
      "entropy": 6.138200807571411,
      "epoch": 1.4433945602260685,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004795626527608179,
      "loss": 5.891,
      "mean_token_accuracy": 0.1682120956480503,
      "num_tokens": 30457524.0,
      "step": 16345
    },
    {
      "entropy": 6.190278577804565,
      "epoch": 1.4438361003179088,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004795495367193451,
      "loss": 5.8979,
      "mean_token_accuracy": 0.16679812967777252,
      "num_tokens": 30467082.0,
      "step": 16350
    },
    {
      "entropy": 6.133280611038208,
      "epoch": 1.4442776404097493,
      "grad_norm": 1.859375,
      "learning_rate": 0.00047953641667086315,
      "loss": 5.922,
      "mean_token_accuracy": 0.1620195671916008,
      "num_tokens": 30476251.0,
      "step": 16355
    },
    {
      "entropy": 6.003503704071045,
      "epoch": 1.4447191805015895,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047952329261562914,
      "loss": 5.8213,
      "mean_token_accuracy": 0.16314259469509124,
      "num_tokens": 30485997.0,
      "step": 16360
    },
    {
      "entropy": 6.117480516433716,
      "epoch": 1.4451607205934298,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047951016455390023,
      "loss": 5.9321,
      "mean_token_accuracy": 0.16438680440187453,
      "num_tokens": 30494886.0,
      "step": 16365
    },
    {
      "entropy": 6.1632829189300535,
      "epoch": 1.4456022606852703,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004794970324859334,
      "loss": 5.9366,
      "mean_token_accuracy": 0.1525437518954277,
      "num_tokens": 30504194.0,
      "step": 16370
    },
    {
      "entropy": 6.091844701766968,
      "epoch": 1.4460438007771106,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047948389641198615,
      "loss": 5.8095,
      "mean_token_accuracy": 0.16608143895864486,
      "num_tokens": 30513077.0,
      "step": 16375
    },
    {
      "entropy": 6.060039901733399,
      "epoch": 1.4464853408689509,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047947075633231553,
      "loss": 5.8778,
      "mean_token_accuracy": 0.16263142675161363,
      "num_tokens": 30521999.0,
      "step": 16380
    },
    {
      "entropy": 6.177615213394165,
      "epoch": 1.4469268809607914,
      "grad_norm": 1.75,
      "learning_rate": 0.0004794576122471791,
      "loss": 5.9486,
      "mean_token_accuracy": 0.1609589010477066,
      "num_tokens": 30530138.0,
      "step": 16385
    },
    {
      "entropy": 6.082210540771484,
      "epoch": 1.4473684210526316,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004794444641568344,
      "loss": 5.7825,
      "mean_token_accuracy": 0.17482944577932358,
      "num_tokens": 30538850.0,
      "step": 16390
    },
    {
      "entropy": 6.086614894866943,
      "epoch": 1.447809961144472,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004794313120615388,
      "loss": 5.8719,
      "mean_token_accuracy": 0.16175234615802764,
      "num_tokens": 30547593.0,
      "step": 16395
    },
    {
      "entropy": 6.071878099441529,
      "epoch": 1.4482515012363122,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004794181559615502,
      "loss": 5.8423,
      "mean_token_accuracy": 0.1658116564154625,
      "num_tokens": 30557154.0,
      "step": 16400
    },
    {
      "entropy": 6.157874202728271,
      "epoch": 1.4486930413281525,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047940499585712615,
      "loss": 5.8526,
      "mean_token_accuracy": 0.16343866437673568,
      "num_tokens": 30566379.0,
      "step": 16405
    },
    {
      "entropy": 6.043774604797363,
      "epoch": 1.449134581419993,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047939183174852446,
      "loss": 5.7969,
      "mean_token_accuracy": 0.16887260228395462,
      "num_tokens": 30575146.0,
      "step": 16410
    },
    {
      "entropy": 6.091286993026733,
      "epoch": 1.4495761215118332,
      "grad_norm": 2.125,
      "learning_rate": 0.00047937866363600295,
      "loss": 5.8683,
      "mean_token_accuracy": 0.1682022273540497,
      "num_tokens": 30584444.0,
      "step": 16415
    },
    {
      "entropy": 6.144282388687134,
      "epoch": 1.4500176616036735,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004793654915198198,
      "loss": 5.856,
      "mean_token_accuracy": 0.1659972608089447,
      "num_tokens": 30593779.0,
      "step": 16420
    },
    {
      "entropy": 6.125273752212524,
      "epoch": 1.450459201695514,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004793523154002328,
      "loss": 5.8931,
      "mean_token_accuracy": 0.16509217768907547,
      "num_tokens": 30603762.0,
      "step": 16425
    },
    {
      "entropy": 6.166903162002564,
      "epoch": 1.4509007417873543,
      "grad_norm": 1.765625,
      "learning_rate": 0.00047933913527750015,
      "loss": 5.8486,
      "mean_token_accuracy": 0.16470662802457808,
      "num_tokens": 30612907.0,
      "step": 16430
    },
    {
      "entropy": 6.044273376464844,
      "epoch": 1.4513422818791946,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004793259511518801,
      "loss": 5.837,
      "mean_token_accuracy": 0.17739309519529342,
      "num_tokens": 30621537.0,
      "step": 16435
    },
    {
      "entropy": 6.029165458679199,
      "epoch": 1.451783821971035,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004793127630236309,
      "loss": 5.9286,
      "mean_token_accuracy": 0.15234357342123986,
      "num_tokens": 30632143.0,
      "step": 16440
    },
    {
      "entropy": 6.191774177551269,
      "epoch": 1.4522253620628753,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047929957089301085,
      "loss": 5.8696,
      "mean_token_accuracy": 0.15817511677742005,
      "num_tokens": 30641145.0,
      "step": 16445
    },
    {
      "entropy": 6.1487610816955565,
      "epoch": 1.4526669021547156,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047928637476027836,
      "loss": 5.893,
      "mean_token_accuracy": 0.16433389782905578,
      "num_tokens": 30650036.0,
      "step": 16450
    },
    {
      "entropy": 6.1043312549591064,
      "epoch": 1.453108442246556,
      "grad_norm": 1.984375,
      "learning_rate": 0.00047927317462569206,
      "loss": 5.8885,
      "mean_token_accuracy": 0.17362781316041948,
      "num_tokens": 30659552.0,
      "step": 16455
    },
    {
      "entropy": 6.116583061218262,
      "epoch": 1.4535499823383964,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004792599704895103,
      "loss": 5.8209,
      "mean_token_accuracy": 0.16752706319093705,
      "num_tokens": 30668652.0,
      "step": 16460
    },
    {
      "entropy": 6.113993215560913,
      "epoch": 1.4539915224302367,
      "grad_norm": 1.734375,
      "learning_rate": 0.000479246762351992,
      "loss": 5.8924,
      "mean_token_accuracy": 0.16233946681022643,
      "num_tokens": 30678613.0,
      "step": 16465
    },
    {
      "entropy": 6.111586618423462,
      "epoch": 1.454433062522077,
      "grad_norm": 2.34375,
      "learning_rate": 0.00047923355021339576,
      "loss": 5.8504,
      "mean_token_accuracy": 0.15980245172977448,
      "num_tokens": 30688259.0,
      "step": 16470
    },
    {
      "entropy": 6.085546827316284,
      "epoch": 1.4548746026139172,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004792203340739805,
      "loss": 5.9046,
      "mean_token_accuracy": 0.16249998956918715,
      "num_tokens": 30697309.0,
      "step": 16475
    },
    {
      "entropy": 6.200652170181274,
      "epoch": 1.4553161427057577,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004792071139340049,
      "loss": 5.9483,
      "mean_token_accuracy": 0.16065697520971298,
      "num_tokens": 30707278.0,
      "step": 16480
    },
    {
      "entropy": 6.147679138183594,
      "epoch": 1.455757682797598,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004791938897937281,
      "loss": 5.8527,
      "mean_token_accuracy": 0.16501333862543105,
      "num_tokens": 30716840.0,
      "step": 16485
    },
    {
      "entropy": 6.090538549423218,
      "epoch": 1.4561992228894383,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047918066165340926,
      "loss": 5.9188,
      "mean_token_accuracy": 0.15906209051609038,
      "num_tokens": 30727120.0,
      "step": 16490
    },
    {
      "entropy": 6.010635757446289,
      "epoch": 1.4566407629812788,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004791674295133072,
      "loss": 5.8204,
      "mean_token_accuracy": 0.1674734428524971,
      "num_tokens": 30735543.0,
      "step": 16495
    },
    {
      "entropy": 6.1078931331634525,
      "epoch": 1.457082303073119,
      "grad_norm": 2.5,
      "learning_rate": 0.00047915419337368146,
      "loss": 5.8923,
      "mean_token_accuracy": 0.1585860386490822,
      "num_tokens": 30744971.0,
      "step": 16500
    },
    {
      "entropy": 6.097987270355224,
      "epoch": 1.4575238431649593,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00047914095323479115,
      "loss": 5.7907,
      "mean_token_accuracy": 0.17311002761125566,
      "num_tokens": 30753444.0,
      "step": 16505
    },
    {
      "entropy": 6.122252082824707,
      "epoch": 1.4579653832567998,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004791277090968956,
      "loss": 5.8817,
      "mean_token_accuracy": 0.1644444525241852,
      "num_tokens": 30761469.0,
      "step": 16510
    },
    {
      "entropy": 6.050733280181885,
      "epoch": 1.45840692334864,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004791144609602544,
      "loss": 5.786,
      "mean_token_accuracy": 0.17038584053516387,
      "num_tokens": 30770773.0,
      "step": 16515
    },
    {
      "entropy": 6.078069257736206,
      "epoch": 1.4588484634404804,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047910120882512696,
      "loss": 5.9115,
      "mean_token_accuracy": 0.16382840424776077,
      "num_tokens": 30779624.0,
      "step": 16520
    },
    {
      "entropy": 6.0982013702392575,
      "epoch": 1.4592900035323209,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004790879526917729,
      "loss": 5.8253,
      "mean_token_accuracy": 0.16577726900577544,
      "num_tokens": 30788671.0,
      "step": 16525
    },
    {
      "entropy": 6.210943269729614,
      "epoch": 1.4597315436241611,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004790746925604519,
      "loss": 5.9024,
      "mean_token_accuracy": 0.1667320489883423,
      "num_tokens": 30797470.0,
      "step": 16530
    },
    {
      "entropy": 6.112218761444092,
      "epoch": 1.4601730837160014,
      "grad_norm": 2.125,
      "learning_rate": 0.00047906142843142383,
      "loss": 5.8865,
      "mean_token_accuracy": 0.1677073732018471,
      "num_tokens": 30807209.0,
      "step": 16535
    },
    {
      "entropy": 6.163278818130493,
      "epoch": 1.4606146238078417,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047904816030494834,
      "loss": 5.9137,
      "mean_token_accuracy": 0.1586761325597763,
      "num_tokens": 30816156.0,
      "step": 16540
    },
    {
      "entropy": 6.120845937728882,
      "epoch": 1.461056163899682,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004790348881812855,
      "loss": 5.7906,
      "mean_token_accuracy": 0.17404769062995912,
      "num_tokens": 30825439.0,
      "step": 16545
    },
    {
      "entropy": 6.099930715560913,
      "epoch": 1.4614977039915225,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047902161206069514,
      "loss": 5.8385,
      "mean_token_accuracy": 0.16998015940189362,
      "num_tokens": 30834953.0,
      "step": 16550
    },
    {
      "entropy": 6.124677610397339,
      "epoch": 1.4619392440833627,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047900833194343764,
      "loss": 5.8782,
      "mean_token_accuracy": 0.1615422785282135,
      "num_tokens": 30843534.0,
      "step": 16555
    },
    {
      "entropy": 6.016527700424194,
      "epoch": 1.462380784175203,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004789950478297727,
      "loss": 5.7865,
      "mean_token_accuracy": 0.1739972323179245,
      "num_tokens": 30851869.0,
      "step": 16560
    },
    {
      "entropy": 6.060797452926636,
      "epoch": 1.4628223242670435,
      "grad_norm": 2.0,
      "learning_rate": 0.00047898175971996095,
      "loss": 5.9364,
      "mean_token_accuracy": 0.1589815929532051,
      "num_tokens": 30862262.0,
      "step": 16565
    },
    {
      "entropy": 6.206104707717896,
      "epoch": 1.4632638643588838,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047896846761426255,
      "loss": 5.9277,
      "mean_token_accuracy": 0.15740045309066772,
      "num_tokens": 30872223.0,
      "step": 16570
    },
    {
      "entropy": 6.129883241653443,
      "epoch": 1.463705404450724,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004789551715129378,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16485995948314666,
      "num_tokens": 30881834.0,
      "step": 16575
    },
    {
      "entropy": 6.127930593490601,
      "epoch": 1.4641469445425646,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004789418714162472,
      "loss": 5.9299,
      "mean_token_accuracy": 0.1612404927611351,
      "num_tokens": 30893414.0,
      "step": 16580
    },
    {
      "entropy": 6.159403133392334,
      "epoch": 1.4645884846344048,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004789285673244514,
      "loss": 5.9721,
      "mean_token_accuracy": 0.15337469279766083,
      "num_tokens": 30902345.0,
      "step": 16585
    },
    {
      "entropy": 6.055581331253052,
      "epoch": 1.465030024726245,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004789152592378109,
      "loss": 5.6716,
      "mean_token_accuracy": 0.18701363056898118,
      "num_tokens": 30911619.0,
      "step": 16590
    },
    {
      "entropy": 6.153064584732055,
      "epoch": 1.4654715648180856,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047890194715658645,
      "loss": 5.9088,
      "mean_token_accuracy": 0.15721338242292404,
      "num_tokens": 30921515.0,
      "step": 16595
    },
    {
      "entropy": 6.17427248954773,
      "epoch": 1.4659131049099259,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004788886310810389,
      "loss": 5.8345,
      "mean_token_accuracy": 0.17332156598567963,
      "num_tokens": 30930807.0,
      "step": 16600
    },
    {
      "entropy": 6.154349040985108,
      "epoch": 1.4663546450017662,
      "grad_norm": 3.21875,
      "learning_rate": 0.00047887531101142893,
      "loss": 5.932,
      "mean_token_accuracy": 0.156033456325531,
      "num_tokens": 30939266.0,
      "step": 16605
    },
    {
      "entropy": 6.1407880783081055,
      "epoch": 1.4667961850936064,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047886198694801764,
      "loss": 5.8207,
      "mean_token_accuracy": 0.16336972564458846,
      "num_tokens": 30948119.0,
      "step": 16610
    },
    {
      "entropy": 6.085747385025025,
      "epoch": 1.4672377251854467,
      "grad_norm": 3.171875,
      "learning_rate": 0.00047884865889106593,
      "loss": 5.7728,
      "mean_token_accuracy": 0.17088960558176042,
      "num_tokens": 30957005.0,
      "step": 16615
    },
    {
      "entropy": 5.979638385772705,
      "epoch": 1.4676792652772872,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047883532684083493,
      "loss": 5.8021,
      "mean_token_accuracy": 0.17437647581100463,
      "num_tokens": 30967144.0,
      "step": 16620
    },
    {
      "entropy": 6.065267276763916,
      "epoch": 1.4681208053691275,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004788219907975858,
      "loss": 5.7771,
      "mean_token_accuracy": 0.1695861831307411,
      "num_tokens": 30975315.0,
      "step": 16625
    },
    {
      "entropy": 6.204185962677002,
      "epoch": 1.4685623454609678,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004788086507615798,
      "loss": 5.9518,
      "mean_token_accuracy": 0.160202094912529,
      "num_tokens": 30984705.0,
      "step": 16630
    },
    {
      "entropy": 6.1908375263214115,
      "epoch": 1.4690038855528083,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004787953067330782,
      "loss": 5.9418,
      "mean_token_accuracy": 0.15874718129634857,
      "num_tokens": 30994936.0,
      "step": 16635
    },
    {
      "entropy": 6.0459671974182125,
      "epoch": 1.4694454256446485,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004787819587123425,
      "loss": 5.8405,
      "mean_token_accuracy": 0.16103943437337875,
      "num_tokens": 31004011.0,
      "step": 16640
    },
    {
      "entropy": 6.131539964675904,
      "epoch": 1.4698869657364888,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004787686066996341,
      "loss": 5.889,
      "mean_token_accuracy": 0.1587212160229683,
      "num_tokens": 31013852.0,
      "step": 16645
    },
    {
      "entropy": 6.094918489456177,
      "epoch": 1.4703285058283293,
      "grad_norm": 2.765625,
      "learning_rate": 0.00047875525069521457,
      "loss": 5.8183,
      "mean_token_accuracy": 0.16892816722393036,
      "num_tokens": 31023456.0,
      "step": 16650
    },
    {
      "entropy": 6.11794638633728,
      "epoch": 1.4707700459201696,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004787418906993456,
      "loss": 5.9176,
      "mean_token_accuracy": 0.1613260582089424,
      "num_tokens": 31032828.0,
      "step": 16655
    },
    {
      "entropy": 6.089264011383056,
      "epoch": 1.4712115860120099,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047872852671228884,
      "loss": 5.8307,
      "mean_token_accuracy": 0.16881431490182877,
      "num_tokens": 31043017.0,
      "step": 16660
    },
    {
      "entropy": 6.041978979110718,
      "epoch": 1.4716531261038504,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00047871515873430614,
      "loss": 5.7225,
      "mean_token_accuracy": 0.17817163616418838,
      "num_tokens": 31051491.0,
      "step": 16665
    },
    {
      "entropy": 6.051099252700806,
      "epoch": 1.4720946661956906,
      "grad_norm": 2.375,
      "learning_rate": 0.0004787017867656593,
      "loss": 5.7735,
      "mean_token_accuracy": 0.17896751314401627,
      "num_tokens": 31060375.0,
      "step": 16670
    },
    {
      "entropy": 6.1610908031463625,
      "epoch": 1.472536206287531,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004786884108066103,
      "loss": 5.9757,
      "mean_token_accuracy": 0.15622382909059523,
      "num_tokens": 31071410.0,
      "step": 16675
    },
    {
      "entropy": 6.11457314491272,
      "epoch": 1.4729777463793712,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004786750308574212,
      "loss": 5.7101,
      "mean_token_accuracy": 0.18190546780824662,
      "num_tokens": 31080123.0,
      "step": 16680
    },
    {
      "entropy": 6.052502727508545,
      "epoch": 1.4734192864712115,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004786616469183541,
      "loss": 5.8748,
      "mean_token_accuracy": 0.16587188392877578,
      "num_tokens": 31089870.0,
      "step": 16685
    },
    {
      "entropy": 6.111912870407105,
      "epoch": 1.473860826563052,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004786482589896711,
      "loss": 5.8834,
      "mean_token_accuracy": 0.165750715136528,
      "num_tokens": 31098682.0,
      "step": 16690
    },
    {
      "entropy": 6.074671983718872,
      "epoch": 1.4743023666548922,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004786348670716346,
      "loss": 5.9062,
      "mean_token_accuracy": 0.1592576324939728,
      "num_tokens": 31107785.0,
      "step": 16695
    },
    {
      "entropy": 6.021370220184326,
      "epoch": 1.4747439067467325,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004786214711645068,
      "loss": 5.7318,
      "mean_token_accuracy": 0.17026282250881195,
      "num_tokens": 31116853.0,
      "step": 16700
    },
    {
      "entropy": 6.10271258354187,
      "epoch": 1.475185446838573,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004786080712685503,
      "loss": 5.835,
      "mean_token_accuracy": 0.16705643832683564,
      "num_tokens": 31126260.0,
      "step": 16705
    },
    {
      "entropy": 6.138799381256104,
      "epoch": 1.4756269869304133,
      "grad_norm": 1.859375,
      "learning_rate": 0.00047859466738402743,
      "loss": 5.9094,
      "mean_token_accuracy": 0.16703267246484757,
      "num_tokens": 31135047.0,
      "step": 16710
    },
    {
      "entropy": 6.16530556678772,
      "epoch": 1.4760685270222536,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047858125951120073,
      "loss": 5.8513,
      "mean_token_accuracy": 0.16246282756328584,
      "num_tokens": 31144177.0,
      "step": 16715
    },
    {
      "entropy": 6.136130475997925,
      "epoch": 1.476510067114094,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000478567847650333,
      "loss": 5.9032,
      "mean_token_accuracy": 0.15859589576721192,
      "num_tokens": 31153586.0,
      "step": 16720
    },
    {
      "entropy": 6.156058883666992,
      "epoch": 1.4769516072059343,
      "grad_norm": 2.359375,
      "learning_rate": 0.000478554431801687,
      "loss": 6.0316,
      "mean_token_accuracy": 0.16775701195001602,
      "num_tokens": 31163252.0,
      "step": 16725
    },
    {
      "entropy": 6.096574211120606,
      "epoch": 1.4773931472977746,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004785410119655253,
      "loss": 5.7972,
      "mean_token_accuracy": 0.1694275349378586,
      "num_tokens": 31171947.0,
      "step": 16730
    },
    {
      "entropy": 6.0835874557495115,
      "epoch": 1.477834687389615,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00047852758814211104,
      "loss": 5.8364,
      "mean_token_accuracy": 0.1691820852458477,
      "num_tokens": 31180502.0,
      "step": 16735
    },
    {
      "entropy": 6.154789638519287,
      "epoch": 1.4782762274814554,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047851416033170704,
      "loss": 5.9209,
      "mean_token_accuracy": 0.15854550153017044,
      "num_tokens": 31190744.0,
      "step": 16740
    },
    {
      "entropy": 6.149501371383667,
      "epoch": 1.4787177675732956,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004785007285345764,
      "loss": 5.9584,
      "mean_token_accuracy": 0.1622500717639923,
      "num_tokens": 31200869.0,
      "step": 16745
    },
    {
      "entropy": 6.1348827362060545,
      "epoch": 1.479159307665136,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004784872927509822,
      "loss": 5.8625,
      "mean_token_accuracy": 0.17138958126306533,
      "num_tokens": 31210046.0,
      "step": 16750
    },
    {
      "entropy": 6.0879851341247555,
      "epoch": 1.4796008477569762,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004784738529811877,
      "loss": 5.7898,
      "mean_token_accuracy": 0.16822390407323837,
      "num_tokens": 31219140.0,
      "step": 16755
    },
    {
      "entropy": 6.062300062179565,
      "epoch": 1.4800423878488167,
      "grad_norm": 1.703125,
      "learning_rate": 0.00047846040922545604,
      "loss": 5.7942,
      "mean_token_accuracy": 0.1735643595457077,
      "num_tokens": 31227993.0,
      "step": 16760
    },
    {
      "entropy": 6.054637813568116,
      "epoch": 1.480483927940657,
      "grad_norm": 1.953125,
      "learning_rate": 0.00047844696148405074,
      "loss": 5.8271,
      "mean_token_accuracy": 0.16599794179201127,
      "num_tokens": 31236959.0,
      "step": 16765
    },
    {
      "entropy": 6.115944814682007,
      "epoch": 1.4809254680324972,
      "grad_norm": 1.96875,
      "learning_rate": 0.00047843350975723523,
      "loss": 5.7515,
      "mean_token_accuracy": 0.1751209408044815,
      "num_tokens": 31246373.0,
      "step": 16770
    },
    {
      "entropy": 6.062704753875733,
      "epoch": 1.4813670081243377,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004784200540452729,
      "loss": 5.8762,
      "mean_token_accuracy": 0.16236329227685928,
      "num_tokens": 31255556.0,
      "step": 16775
    },
    {
      "entropy": 6.042606496810913,
      "epoch": 1.481808548216178,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004784065943484274,
      "loss": 5.7943,
      "mean_token_accuracy": 0.1733111932873726,
      "num_tokens": 31264824.0,
      "step": 16780
    },
    {
      "entropy": 6.050362300872803,
      "epoch": 1.4822500883080183,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047839313066696246,
      "loss": 5.7636,
      "mean_token_accuracy": 0.17941183000802993,
      "num_tokens": 31273946.0,
      "step": 16785
    },
    {
      "entropy": 6.087690353393555,
      "epoch": 1.4826916283998588,
      "grad_norm": 1.984375,
      "learning_rate": 0.00047837966300114164,
      "loss": 5.8284,
      "mean_token_accuracy": 0.16756643801927568,
      "num_tokens": 31283492.0,
      "step": 16790
    },
    {
      "entropy": 6.136430740356445,
      "epoch": 1.483133168491699,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047836619135122906,
      "loss": 5.8756,
      "mean_token_accuracy": 0.16756242662668228,
      "num_tokens": 31292250.0,
      "step": 16795
    },
    {
      "entropy": 6.129665231704712,
      "epoch": 1.4835747085835393,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047835271571748834,
      "loss": 6.0026,
      "mean_token_accuracy": 0.15289610624313354,
      "num_tokens": 31302868.0,
      "step": 16800
    },
    {
      "entropy": 6.125290250778198,
      "epoch": 1.4840162486753798,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004783392361001836,
      "loss": 5.7688,
      "mean_token_accuracy": 0.17371761798858643,
      "num_tokens": 31311918.0,
      "step": 16805
    },
    {
      "entropy": 6.0620016098022464,
      "epoch": 1.4844577887672201,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004783257524995788,
      "loss": 5.7853,
      "mean_token_accuracy": 0.17009174078702927,
      "num_tokens": 31321339.0,
      "step": 16810
    },
    {
      "entropy": 6.11047592163086,
      "epoch": 1.4848993288590604,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004783122649159383,
      "loss": 5.7706,
      "mean_token_accuracy": 0.1641666367650032,
      "num_tokens": 31330498.0,
      "step": 16815
    },
    {
      "entropy": 6.0488334655761715,
      "epoch": 1.4853408689509007,
      "grad_norm": 2.203125,
      "learning_rate": 0.000478298773349526,
      "loss": 5.8017,
      "mean_token_accuracy": 0.1696183294057846,
      "num_tokens": 31339676.0,
      "step": 16820
    },
    {
      "entropy": 6.161963748931885,
      "epoch": 1.485782409042741,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004782852778006064,
      "loss": 5.9708,
      "mean_token_accuracy": 0.15382129326462746,
      "num_tokens": 31350072.0,
      "step": 16825
    },
    {
      "entropy": 6.1765745162963865,
      "epoch": 1.4862239491345814,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047827177826944396,
      "loss": 5.8364,
      "mean_token_accuracy": 0.17558974027633667,
      "num_tokens": 31359992.0,
      "step": 16830
    },
    {
      "entropy": 6.142035341262817,
      "epoch": 1.4866654892264217,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047825827475630285,
      "loss": 5.8468,
      "mean_token_accuracy": 0.17307675033807754,
      "num_tokens": 31369434.0,
      "step": 16835
    },
    {
      "entropy": 6.095015954971314,
      "epoch": 1.487107029318262,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00047824476726144775,
      "loss": 5.8548,
      "mean_token_accuracy": 0.16512990444898606,
      "num_tokens": 31378773.0,
      "step": 16840
    },
    {
      "entropy": 6.167878341674805,
      "epoch": 1.4875485694101025,
      "grad_norm": 2.125,
      "learning_rate": 0.00047823125578514325,
      "loss": 5.8869,
      "mean_token_accuracy": 0.16091246381402016,
      "num_tokens": 31388219.0,
      "step": 16845
    },
    {
      "entropy": 6.01156096458435,
      "epoch": 1.4879901095019428,
      "grad_norm": 2.890625,
      "learning_rate": 0.00047821774032765407,
      "loss": 5.8014,
      "mean_token_accuracy": 0.1722875252366066,
      "num_tokens": 31397216.0,
      "step": 16850
    },
    {
      "entropy": 5.973207426071167,
      "epoch": 1.488431649593783,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004782042208892449,
      "loss": 5.7873,
      "mean_token_accuracy": 0.17524673119187356,
      "num_tokens": 31406861.0,
      "step": 16855
    },
    {
      "entropy": 6.088246011734009,
      "epoch": 1.4888731896856235,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047819069747018065,
      "loss": 5.8625,
      "mean_token_accuracy": 0.16829375475645064,
      "num_tokens": 31416387.0,
      "step": 16860
    },
    {
      "entropy": 6.1217546463012695,
      "epoch": 1.4893147297774638,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004781771700707262,
      "loss": 5.7749,
      "mean_token_accuracy": 0.1686438351869583,
      "num_tokens": 31425056.0,
      "step": 16865
    },
    {
      "entropy": 6.155984830856323,
      "epoch": 1.489756269869304,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004781636386911465,
      "loss": 5.9689,
      "mean_token_accuracy": 0.1603533536195755,
      "num_tokens": 31435746.0,
      "step": 16870
    },
    {
      "entropy": 6.1432360172271725,
      "epoch": 1.4901978099611446,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004781501033317067,
      "loss": 5.9863,
      "mean_token_accuracy": 0.15970577597618102,
      "num_tokens": 31444808.0,
      "step": 16875
    },
    {
      "entropy": 6.152387857437134,
      "epoch": 1.4906393500529849,
      "grad_norm": 2.375,
      "learning_rate": 0.0004781365639926718,
      "loss": 5.8766,
      "mean_token_accuracy": 0.16602828353643417,
      "num_tokens": 31454205.0,
      "step": 16880
    },
    {
      "entropy": 6.169824695587158,
      "epoch": 1.4910808901448251,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004781230206743073,
      "loss": 5.9091,
      "mean_token_accuracy": 0.1637486547231674,
      "num_tokens": 31463446.0,
      "step": 16885
    },
    {
      "entropy": 6.119858264923096,
      "epoch": 1.4915224302366654,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004781094733768783,
      "loss": 5.8511,
      "mean_token_accuracy": 0.17152807116508484,
      "num_tokens": 31473168.0,
      "step": 16890
    },
    {
      "entropy": 6.066344404220581,
      "epoch": 1.4919639703285057,
      "grad_norm": 3.75,
      "learning_rate": 0.00047809592210065017,
      "loss": 5.7617,
      "mean_token_accuracy": 0.17701584994792938,
      "num_tokens": 31482899.0,
      "step": 16895
    },
    {
      "entropy": 6.100191593170166,
      "epoch": 1.4924055104203462,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047808236684588847,
      "loss": 5.7715,
      "mean_token_accuracy": 0.16601161956787108,
      "num_tokens": 31492503.0,
      "step": 16900
    },
    {
      "entropy": 5.997424793243408,
      "epoch": 1.4928470505121865,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004780688076128587,
      "loss": 5.7227,
      "mean_token_accuracy": 0.1828492045402527,
      "num_tokens": 31501520.0,
      "step": 16905
    },
    {
      "entropy": 6.089253711700439,
      "epoch": 1.4932885906040267,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047805524440182654,
      "loss": 5.8666,
      "mean_token_accuracy": 0.17120354026556014,
      "num_tokens": 31510821.0,
      "step": 16910
    },
    {
      "entropy": 6.116029739379883,
      "epoch": 1.4937301306958672,
      "grad_norm": 2.5,
      "learning_rate": 0.00047804167721305756,
      "loss": 5.8547,
      "mean_token_accuracy": 0.17002387046813966,
      "num_tokens": 31520472.0,
      "step": 16915
    },
    {
      "entropy": 6.063102293014526,
      "epoch": 1.4941716707877075,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047802810604681757,
      "loss": 5.8744,
      "mean_token_accuracy": 0.17188760638237,
      "num_tokens": 31529129.0,
      "step": 16920
    },
    {
      "entropy": 6.035191345214844,
      "epoch": 1.4946132108795478,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047801453090337256,
      "loss": 5.8123,
      "mean_token_accuracy": 0.1693618878722191,
      "num_tokens": 31537897.0,
      "step": 16925
    },
    {
      "entropy": 6.140631103515625,
      "epoch": 1.4950547509713883,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004780009517829882,
      "loss": 5.8746,
      "mean_token_accuracy": 0.16316870152950286,
      "num_tokens": 31547656.0,
      "step": 16930
    },
    {
      "entropy": 6.136087322235108,
      "epoch": 1.4954962910632286,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004779873686859308,
      "loss": 5.8169,
      "mean_token_accuracy": 0.16767246127128602,
      "num_tokens": 31557303.0,
      "step": 16935
    },
    {
      "entropy": 6.143079710006714,
      "epoch": 1.4959378311550688,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047797378161246613,
      "loss": 5.8559,
      "mean_token_accuracy": 0.16396174877882003,
      "num_tokens": 31565749.0,
      "step": 16940
    },
    {
      "entropy": 6.087357425689698,
      "epoch": 1.4963793712469093,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047796019056286067,
      "loss": 5.872,
      "mean_token_accuracy": 0.16910456120967865,
      "num_tokens": 31575286.0,
      "step": 16945
    },
    {
      "entropy": 6.169096612930298,
      "epoch": 1.4968209113387496,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004779465955373804,
      "loss": 5.8178,
      "mean_token_accuracy": 0.17204560786485673,
      "num_tokens": 31583723.0,
      "step": 16950
    },
    {
      "entropy": 6.00992522239685,
      "epoch": 1.49726245143059,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004779329965362918,
      "loss": 5.6867,
      "mean_token_accuracy": 0.18534846156835555,
      "num_tokens": 31593171.0,
      "step": 16955
    },
    {
      "entropy": 5.995798349380493,
      "epoch": 1.4977039915224302,
      "grad_norm": 2.25,
      "learning_rate": 0.00047791939355986123,
      "loss": 5.8609,
      "mean_token_accuracy": 0.1654835671186447,
      "num_tokens": 31603958.0,
      "step": 16960
    },
    {
      "entropy": 6.031079435348511,
      "epoch": 1.4981455316142704,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004779057866083551,
      "loss": 5.8612,
      "mean_token_accuracy": 0.164565047621727,
      "num_tokens": 31612460.0,
      "step": 16965
    },
    {
      "entropy": 6.203945589065552,
      "epoch": 1.498587071706111,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047789217568204004,
      "loss": 5.9136,
      "mean_token_accuracy": 0.16604970395565033,
      "num_tokens": 31622114.0,
      "step": 16970
    },
    {
      "entropy": 6.188137149810791,
      "epoch": 1.4990286117979512,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047787856078118266,
      "loss": 5.856,
      "mean_token_accuracy": 0.1649842992424965,
      "num_tokens": 31632272.0,
      "step": 16975
    },
    {
      "entropy": 6.155510759353637,
      "epoch": 1.4994701518897915,
      "grad_norm": 2.375,
      "learning_rate": 0.0004778649419060496,
      "loss": 5.8049,
      "mean_token_accuracy": 0.1681971177458763,
      "num_tokens": 31642235.0,
      "step": 16980
    },
    {
      "entropy": 6.102774381637573,
      "epoch": 1.499911691981632,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004778513190569077,
      "loss": 5.9096,
      "mean_token_accuracy": 0.15545578226447104,
      "num_tokens": 31652773.0,
      "step": 16985
    },
    {
      "entropy": 6.197693157196045,
      "epoch": 1.5003532320734723,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004778376922340239,
      "loss": 5.8941,
      "mean_token_accuracy": 0.16292841583490372,
      "num_tokens": 31662681.0,
      "step": 16990
    },
    {
      "entropy": 6.101391553878784,
      "epoch": 1.5007947721653125,
      "grad_norm": 1.8515625,
      "learning_rate": 0.000477824061437665,
      "loss": 5.7838,
      "mean_token_accuracy": 0.16881538182497025,
      "num_tokens": 31671576.0,
      "step": 16995
    },
    {
      "entropy": 6.147274971008301,
      "epoch": 1.501236312257153,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047781042666809816,
      "loss": 6.0051,
      "mean_token_accuracy": 0.15614731311798097,
      "num_tokens": 31681149.0,
      "step": 17000
    },
    {
      "entropy": 6.075467300415039,
      "epoch": 1.5016778523489933,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004777967879255904,
      "loss": 5.8246,
      "mean_token_accuracy": 0.16849614530801774,
      "num_tokens": 31690007.0,
      "step": 17005
    },
    {
      "entropy": 6.221605205535889,
      "epoch": 1.5021193924408336,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004777831452104088,
      "loss": 5.9665,
      "mean_token_accuracy": 0.15869542509317397,
      "num_tokens": 31700055.0,
      "step": 17010
    },
    {
      "entropy": 6.1071125030517575,
      "epoch": 1.502560932532674,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004777694985228208,
      "loss": 5.8401,
      "mean_token_accuracy": 0.16893051266670228,
      "num_tokens": 31708255.0,
      "step": 17015
    },
    {
      "entropy": 6.031527900695801,
      "epoch": 1.5030024726245141,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047775584786309365,
      "loss": 5.8535,
      "mean_token_accuracy": 0.1649830862879753,
      "num_tokens": 31716855.0,
      "step": 17020
    },
    {
      "entropy": 6.082430028915406,
      "epoch": 1.5034440127163546,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047774219323149474,
      "loss": 5.8206,
      "mean_token_accuracy": 0.1744278833270073,
      "num_tokens": 31726153.0,
      "step": 17025
    },
    {
      "entropy": 6.062655448913574,
      "epoch": 1.5038855528081951,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004777285346282916,
      "loss": 5.8429,
      "mean_token_accuracy": 0.17389040142297746,
      "num_tokens": 31736935.0,
      "step": 17030
    },
    {
      "entropy": 6.139467716217041,
      "epoch": 1.5043270929000352,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004777148720537517,
      "loss": 5.9051,
      "mean_token_accuracy": 0.16433268189430236,
      "num_tokens": 31746402.0,
      "step": 17035
    },
    {
      "entropy": 6.101178598403931,
      "epoch": 1.5047686329918757,
      "grad_norm": 2.78125,
      "learning_rate": 0.00047770120550814276,
      "loss": 5.7584,
      "mean_token_accuracy": 0.1637473538517952,
      "num_tokens": 31755615.0,
      "step": 17040
    },
    {
      "entropy": 6.065146207809448,
      "epoch": 1.505210173083716,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004776875349917325,
      "loss": 5.8881,
      "mean_token_accuracy": 0.16716915667057036,
      "num_tokens": 31765701.0,
      "step": 17045
    },
    {
      "entropy": 6.089950370788574,
      "epoch": 1.5056517131755562,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004776738605047887,
      "loss": 5.8762,
      "mean_token_accuracy": 0.1676294907927513,
      "num_tokens": 31774674.0,
      "step": 17050
    },
    {
      "entropy": 6.107439756393433,
      "epoch": 1.5060932532673967,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004776601820475793,
      "loss": 5.8146,
      "mean_token_accuracy": 0.16173949539661409,
      "num_tokens": 31784871.0,
      "step": 17055
    },
    {
      "entropy": 6.105894756317139,
      "epoch": 1.506534793359237,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004776464996203721,
      "loss": 5.7787,
      "mean_token_accuracy": 0.17518564015626908,
      "num_tokens": 31793274.0,
      "step": 17060
    },
    {
      "entropy": 6.072642993927002,
      "epoch": 1.5069763334510773,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004776328132234353,
      "loss": 5.9528,
      "mean_token_accuracy": 0.15331857576966285,
      "num_tokens": 31803543.0,
      "step": 17065
    },
    {
      "entropy": 6.184981966018677,
      "epoch": 1.5074178735429178,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004776191228570368,
      "loss": 5.9342,
      "mean_token_accuracy": 0.1568526431918144,
      "num_tokens": 31814529.0,
      "step": 17070
    },
    {
      "entropy": 6.228585290908813,
      "epoch": 1.507859413634758,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004776054285214451,
      "loss": 5.8894,
      "mean_token_accuracy": 0.16705289036035537,
      "num_tokens": 31824064.0,
      "step": 17075
    },
    {
      "entropy": 6.1461437225341795,
      "epoch": 1.5083009537265983,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047759173021692813,
      "loss": 5.8953,
      "mean_token_accuracy": 0.16069892346858977,
      "num_tokens": 31833881.0,
      "step": 17080
    },
    {
      "entropy": 6.0819196701049805,
      "epoch": 1.5087424938184388,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004775780279437545,
      "loss": 5.8461,
      "mean_token_accuracy": 0.16622475236654283,
      "num_tokens": 31842987.0,
      "step": 17085
    },
    {
      "entropy": 6.132349491119385,
      "epoch": 1.5091840339102789,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047756432170219243,
      "loss": 5.7933,
      "mean_token_accuracy": 0.16437864005565644,
      "num_tokens": 31851806.0,
      "step": 17090
    },
    {
      "entropy": 6.205032777786255,
      "epoch": 1.5096255740021194,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004775506114925106,
      "loss": 5.9494,
      "mean_token_accuracy": 0.1621339276432991,
      "num_tokens": 31861857.0,
      "step": 17095
    },
    {
      "entropy": 6.159710550308228,
      "epoch": 1.5100671140939599,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047753689731497734,
      "loss": 5.8475,
      "mean_token_accuracy": 0.16587883979082108,
      "num_tokens": 31871252.0,
      "step": 17100
    },
    {
      "entropy": 6.150771999359131,
      "epoch": 1.5105086541858,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004775231791698616,
      "loss": 5.9324,
      "mean_token_accuracy": 0.15629052370786667,
      "num_tokens": 31880734.0,
      "step": 17105
    },
    {
      "entropy": 6.116977405548096,
      "epoch": 1.5109501942776404,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047750945705743187,
      "loss": 5.8261,
      "mean_token_accuracy": 0.1616917446255684,
      "num_tokens": 31889304.0,
      "step": 17110
    },
    {
      "entropy": 6.127993774414063,
      "epoch": 1.5113917343694807,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004774957309779571,
      "loss": 5.8348,
      "mean_token_accuracy": 0.1676137402653694,
      "num_tokens": 31898586.0,
      "step": 17115
    },
    {
      "entropy": 6.166815662384034,
      "epoch": 1.511833274461321,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004774820009317061,
      "loss": 5.9133,
      "mean_token_accuracy": 0.16282063573598862,
      "num_tokens": 31908153.0,
      "step": 17120
    },
    {
      "entropy": 6.016722106933594,
      "epoch": 1.5122748145531615,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047746826691894787,
      "loss": 5.6796,
      "mean_token_accuracy": 0.18392256498336793,
      "num_tokens": 31916467.0,
      "step": 17125
    },
    {
      "entropy": 6.019726514816284,
      "epoch": 1.5127163546450018,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047745452893995145,
      "loss": 5.8405,
      "mean_token_accuracy": 0.16480594873428345,
      "num_tokens": 31925469.0,
      "step": 17130
    },
    {
      "entropy": 6.119672203063965,
      "epoch": 1.513157894736842,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047744078699498587,
      "loss": 5.8911,
      "mean_token_accuracy": 0.15898177027702332,
      "num_tokens": 31934747.0,
      "step": 17135
    },
    {
      "entropy": 6.178226804733276,
      "epoch": 1.5135994348286825,
      "grad_norm": 4.96875,
      "learning_rate": 0.00047742704108432046,
      "loss": 5.7878,
      "mean_token_accuracy": 0.16157239079475402,
      "num_tokens": 31943773.0,
      "step": 17140
    },
    {
      "entropy": 6.066216850280762,
      "epoch": 1.5140409749205228,
      "grad_norm": 2.953125,
      "learning_rate": 0.00047741329120822446,
      "loss": 5.8796,
      "mean_token_accuracy": 0.16855469942092896,
      "num_tokens": 31952699.0,
      "step": 17145
    },
    {
      "entropy": 6.041545152664185,
      "epoch": 1.514482515012363,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004773995373669672,
      "loss": 5.8559,
      "mean_token_accuracy": 0.15691616982221604,
      "num_tokens": 31962845.0,
      "step": 17150
    },
    {
      "entropy": 6.1267486095428465,
      "epoch": 1.5149240551042036,
      "grad_norm": 5.96875,
      "learning_rate": 0.0004773857795608181,
      "loss": 5.7526,
      "mean_token_accuracy": 0.17602582424879074,
      "num_tokens": 31972437.0,
      "step": 17155
    },
    {
      "entropy": 6.173786067962647,
      "epoch": 1.5153655951960436,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047737201779004667,
      "loss": 5.9301,
      "mean_token_accuracy": 0.166225403547287,
      "num_tokens": 31981652.0,
      "step": 17160
    },
    {
      "entropy": 6.108345413208008,
      "epoch": 1.5158071352878841,
      "grad_norm": 16.25,
      "learning_rate": 0.0004773582520549226,
      "loss": 5.8596,
      "mean_token_accuracy": 0.16720700412988662,
      "num_tokens": 31990088.0,
      "step": 17165
    },
    {
      "entropy": 6.11037392616272,
      "epoch": 1.5162486753797246,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004773444823557153,
      "loss": 5.8807,
      "mean_token_accuracy": 0.16836710572242736,
      "num_tokens": 31999386.0,
      "step": 17170
    },
    {
      "entropy": 6.143088006973267,
      "epoch": 1.5166902154715647,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004773307086926948,
      "loss": 5.7883,
      "mean_token_accuracy": 0.17056186944246293,
      "num_tokens": 32007695.0,
      "step": 17175
    },
    {
      "entropy": 6.154992246627808,
      "epoch": 1.5171317555634052,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047731693106613075,
      "loss": 5.95,
      "mean_token_accuracy": 0.16174111664295196,
      "num_tokens": 32018300.0,
      "step": 17180
    },
    {
      "entropy": 6.139096355438232,
      "epoch": 1.5175732956552455,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004773031494762931,
      "loss": 5.9091,
      "mean_token_accuracy": 0.16173823475837706,
      "num_tokens": 32028386.0,
      "step": 17185
    },
    {
      "entropy": 6.182955026626587,
      "epoch": 1.5180148357470857,
      "grad_norm": 2.375,
      "learning_rate": 0.0004772893639234518,
      "loss": 5.9226,
      "mean_token_accuracy": 0.163718094676733,
      "num_tokens": 32037260.0,
      "step": 17190
    },
    {
      "entropy": 6.15783166885376,
      "epoch": 1.5184563758389262,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004772755744078769,
      "loss": 5.9211,
      "mean_token_accuracy": 0.1667024776339531,
      "num_tokens": 32047654.0,
      "step": 17195
    },
    {
      "entropy": 6.108058738708496,
      "epoch": 1.5188979159307665,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004772617809298385,
      "loss": 5.8505,
      "mean_token_accuracy": 0.16301622092723847,
      "num_tokens": 32056791.0,
      "step": 17200
    },
    {
      "entropy": 6.143093776702881,
      "epoch": 1.5193394560226068,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004772479834896069,
      "loss": 5.8461,
      "mean_token_accuracy": 0.16170306801795958,
      "num_tokens": 32066561.0,
      "step": 17205
    },
    {
      "entropy": 6.228625535964966,
      "epoch": 1.5197809961144473,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047723418208745226,
      "loss": 5.8786,
      "mean_token_accuracy": 0.16210633367300034,
      "num_tokens": 32075801.0,
      "step": 17210
    },
    {
      "entropy": 6.062961530685425,
      "epoch": 1.5202225362062876,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004772203767236451,
      "loss": 5.808,
      "mean_token_accuracy": 0.17412050068378448,
      "num_tokens": 32085274.0,
      "step": 17215
    },
    {
      "entropy": 6.079689311981201,
      "epoch": 1.5206640762981278,
      "grad_norm": 1.625,
      "learning_rate": 0.00047720656739845573,
      "loss": 5.8492,
      "mean_token_accuracy": 0.16835268437862397,
      "num_tokens": 32094481.0,
      "step": 17220
    },
    {
      "entropy": 6.048946237564087,
      "epoch": 1.5211056163899683,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004771927541121547,
      "loss": 5.8163,
      "mean_token_accuracy": 0.17228095531463622,
      "num_tokens": 32103094.0,
      "step": 17225
    },
    {
      "entropy": 6.070489120483399,
      "epoch": 1.5215471564818084,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004771789368650126,
      "loss": 5.8186,
      "mean_token_accuracy": 0.16577275097370148,
      "num_tokens": 32112230.0,
      "step": 17230
    },
    {
      "entropy": 6.104430198669434,
      "epoch": 1.5219886965736489,
      "grad_norm": 1.875,
      "learning_rate": 0.00047716511565730013,
      "loss": 5.8549,
      "mean_token_accuracy": 0.1713423475623131,
      "num_tokens": 32121472.0,
      "step": 17235
    },
    {
      "entropy": 6.165273761749267,
      "epoch": 1.5224302366654894,
      "grad_norm": 2.59375,
      "learning_rate": 0.000477151290489288,
      "loss": 5.8496,
      "mean_token_accuracy": 0.169678170979023,
      "num_tokens": 32130849.0,
      "step": 17240
    },
    {
      "entropy": 6.086892604827881,
      "epoch": 1.5228717767573294,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004771374613612471,
      "loss": 5.802,
      "mean_token_accuracy": 0.16939686238765717,
      "num_tokens": 32140874.0,
      "step": 17245
    },
    {
      "entropy": 6.084832620620728,
      "epoch": 1.52331331684917,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004771236282734482,
      "loss": 5.7801,
      "mean_token_accuracy": 0.16830243915319443,
      "num_tokens": 32150774.0,
      "step": 17250
    },
    {
      "entropy": 6.0836982250213625,
      "epoch": 1.5237548569410102,
      "grad_norm": 1.625,
      "learning_rate": 0.00047710979122616247,
      "loss": 5.8067,
      "mean_token_accuracy": 0.16750760227441788,
      "num_tokens": 32159825.0,
      "step": 17255
    },
    {
      "entropy": 6.095811414718628,
      "epoch": 1.5241963970328505,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004770959502196609,
      "loss": 5.8337,
      "mean_token_accuracy": 0.17115994840860366,
      "num_tokens": 32168443.0,
      "step": 17260
    },
    {
      "entropy": 6.08999662399292,
      "epoch": 1.524637937124691,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047708210525421445,
      "loss": 5.817,
      "mean_token_accuracy": 0.16795896589756013,
      "num_tokens": 32178065.0,
      "step": 17265
    },
    {
      "entropy": 6.072080183029175,
      "epoch": 1.5250794772165313,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047706825633009466,
      "loss": 5.7223,
      "mean_token_accuracy": 0.17456792593002318,
      "num_tokens": 32187274.0,
      "step": 17270
    },
    {
      "entropy": 6.064247560501099,
      "epoch": 1.5255210173083715,
      "grad_norm": 5.96875,
      "learning_rate": 0.00047705440344757264,
      "loss": 5.6994,
      "mean_token_accuracy": 0.1779107540845871,
      "num_tokens": 32195811.0,
      "step": 17275
    },
    {
      "entropy": 5.92962532043457,
      "epoch": 1.525962557400212,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047704054660691964,
      "loss": 5.7991,
      "mean_token_accuracy": 0.171577525138855,
      "num_tokens": 32204958.0,
      "step": 17280
    },
    {
      "entropy": 6.05693998336792,
      "epoch": 1.5264040974920523,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004770266858084074,
      "loss": 5.8174,
      "mean_token_accuracy": 0.1735185131430626,
      "num_tokens": 32214299.0,
      "step": 17285
    },
    {
      "entropy": 6.183889293670655,
      "epoch": 1.5268456375838926,
      "grad_norm": 1.953125,
      "learning_rate": 0.00047701282105230716,
      "loss": 5.8798,
      "mean_token_accuracy": 0.16397885531187056,
      "num_tokens": 32223353.0,
      "step": 17290
    },
    {
      "entropy": 6.130427122116089,
      "epoch": 1.527287177675733,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004769989523388906,
      "loss": 5.863,
      "mean_token_accuracy": 0.1585610568523407,
      "num_tokens": 32232641.0,
      "step": 17295
    },
    {
      "entropy": 6.159122800827026,
      "epoch": 1.5277287177675731,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004769850796684296,
      "loss": 5.9211,
      "mean_token_accuracy": 0.1594293847680092,
      "num_tokens": 32243028.0,
      "step": 17300
    },
    {
      "entropy": 6.162701797485352,
      "epoch": 1.5281702578594136,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004769712030411957,
      "loss": 5.8569,
      "mean_token_accuracy": 0.17022625505924224,
      "num_tokens": 32253660.0,
      "step": 17305
    },
    {
      "entropy": 6.151645946502685,
      "epoch": 1.5286117979512541,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047695732245746076,
      "loss": 5.8579,
      "mean_token_accuracy": 0.16790457665920258,
      "num_tokens": 32263267.0,
      "step": 17310
    },
    {
      "entropy": 6.1199180603027346,
      "epoch": 1.5290533380430942,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004769434379174967,
      "loss": 5.8129,
      "mean_token_accuracy": 0.17420368641614914,
      "num_tokens": 32273929.0,
      "step": 17315
    },
    {
      "entropy": 6.091553211212158,
      "epoch": 1.5294948781349347,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004769295494215757,
      "loss": 5.8247,
      "mean_token_accuracy": 0.1727677434682846,
      "num_tokens": 32282717.0,
      "step": 17320
    },
    {
      "entropy": 6.15964093208313,
      "epoch": 1.529936418226775,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00047691565696996954,
      "loss": 5.8948,
      "mean_token_accuracy": 0.16669653058052064,
      "num_tokens": 32293225.0,
      "step": 17325
    },
    {
      "entropy": 6.179988241195678,
      "epoch": 1.5303779583186152,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004769017605629505,
      "loss": 5.8149,
      "mean_token_accuracy": 0.1702434316277504,
      "num_tokens": 32301582.0,
      "step": 17330
    },
    {
      "entropy": 6.05822787284851,
      "epoch": 1.5308194984104557,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004768878602007909,
      "loss": 5.7758,
      "mean_token_accuracy": 0.1659368544816971,
      "num_tokens": 32310131.0,
      "step": 17335
    },
    {
      "entropy": 6.087152576446533,
      "epoch": 1.531261038502296,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004768739558837628,
      "loss": 5.8519,
      "mean_token_accuracy": 0.16661548465490342,
      "num_tokens": 32320038.0,
      "step": 17340
    },
    {
      "entropy": 6.1244645595550535,
      "epoch": 1.5317025785941363,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004768600476121388,
      "loss": 5.8899,
      "mean_token_accuracy": 0.1629193052649498,
      "num_tokens": 32329568.0,
      "step": 17345
    },
    {
      "entropy": 6.1587567806243895,
      "epoch": 1.5321441186859768,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047684613538619125,
      "loss": 5.8434,
      "mean_token_accuracy": 0.16886794567108154,
      "num_tokens": 32338940.0,
      "step": 17350
    },
    {
      "entropy": 6.151344299316406,
      "epoch": 1.532585658777817,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004768322192061927,
      "loss": 5.8019,
      "mean_token_accuracy": 0.17227451652288436,
      "num_tokens": 32347757.0,
      "step": 17355
    },
    {
      "entropy": 6.011085128784179,
      "epoch": 1.5330271988696573,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004768182990724158,
      "loss": 5.7508,
      "mean_token_accuracy": 0.1769898071885109,
      "num_tokens": 32357099.0,
      "step": 17360
    },
    {
      "entropy": 6.092490482330322,
      "epoch": 1.5334687389614978,
      "grad_norm": 3.625,
      "learning_rate": 0.00047680437498513317,
      "loss": 5.8927,
      "mean_token_accuracy": 0.16415278166532515,
      "num_tokens": 32366548.0,
      "step": 17365
    },
    {
      "entropy": 6.204706001281738,
      "epoch": 1.5339102790533379,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047679044694461765,
      "loss": 5.9907,
      "mean_token_accuracy": 0.15458527952432632,
      "num_tokens": 32377435.0,
      "step": 17370
    },
    {
      "entropy": 6.219340085983276,
      "epoch": 1.5343518191451784,
      "grad_norm": 2.125,
      "learning_rate": 0.000476776514951142,
      "loss": 5.8682,
      "mean_token_accuracy": 0.15746877044439317,
      "num_tokens": 32386922.0,
      "step": 17375
    },
    {
      "entropy": 6.06431097984314,
      "epoch": 1.5347933592370189,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047676257900497915,
      "loss": 5.9066,
      "mean_token_accuracy": 0.16283479183912278,
      "num_tokens": 32396350.0,
      "step": 17380
    },
    {
      "entropy": 5.9955932140350345,
      "epoch": 1.535234899328859,
      "grad_norm": 1.953125,
      "learning_rate": 0.00047674863910640217,
      "loss": 5.6973,
      "mean_token_accuracy": 0.1767934262752533,
      "num_tokens": 32405678.0,
      "step": 17385
    },
    {
      "entropy": 6.145180654525757,
      "epoch": 1.5356764394206994,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047673469525568406,
      "loss": 5.8247,
      "mean_token_accuracy": 0.17113334834575653,
      "num_tokens": 32415793.0,
      "step": 17390
    },
    {
      "entropy": 6.073387956619262,
      "epoch": 1.5361179795125397,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047672074745309804,
      "loss": 5.7301,
      "mean_token_accuracy": 0.17700450569391252,
      "num_tokens": 32424735.0,
      "step": 17395
    },
    {
      "entropy": 6.026446866989136,
      "epoch": 1.53655951960438,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004767067956989172,
      "loss": 5.8697,
      "mean_token_accuracy": 0.1691689595580101,
      "num_tokens": 32435213.0,
      "step": 17400
    },
    {
      "entropy": 6.071824169158935,
      "epoch": 1.5370010596962205,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004766928399934151,
      "loss": 5.751,
      "mean_token_accuracy": 0.17125995382666587,
      "num_tokens": 32443932.0,
      "step": 17405
    },
    {
      "entropy": 6.108694267272949,
      "epoch": 1.5374425997880607,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004766788803368649,
      "loss": 5.7346,
      "mean_token_accuracy": 0.1653388410806656,
      "num_tokens": 32453851.0,
      "step": 17410
    },
    {
      "entropy": 6.102496385574341,
      "epoch": 1.537884139879901,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004766649167295401,
      "loss": 5.8303,
      "mean_token_accuracy": 0.167599917948246,
      "num_tokens": 32462906.0,
      "step": 17415
    },
    {
      "entropy": 6.027633476257324,
      "epoch": 1.5383256799717415,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004766509491717144,
      "loss": 5.7781,
      "mean_token_accuracy": 0.17101458162069322,
      "num_tokens": 32471840.0,
      "step": 17420
    },
    {
      "entropy": 6.027692699432373,
      "epoch": 1.5387672200635818,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047663697766366114,
      "loss": 5.7471,
      "mean_token_accuracy": 0.1702307164669037,
      "num_tokens": 32481111.0,
      "step": 17425
    },
    {
      "entropy": 6.0944441795349125,
      "epoch": 1.539208760155422,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004766230022056542,
      "loss": 5.8549,
      "mean_token_accuracy": 0.1633177012205124,
      "num_tokens": 32489827.0,
      "step": 17430
    },
    {
      "entropy": 6.125396633148194,
      "epoch": 1.5396503002472626,
      "grad_norm": 2.9375,
      "learning_rate": 0.00047660902279796736,
      "loss": 5.7777,
      "mean_token_accuracy": 0.17241585105657578,
      "num_tokens": 32499106.0,
      "step": 17435
    },
    {
      "entropy": 6.247645664215088,
      "epoch": 1.5400918403391026,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004765950394408744,
      "loss": 6.0156,
      "mean_token_accuracy": 0.14962735325098037,
      "num_tokens": 32507858.0,
      "step": 17440
    },
    {
      "entropy": 6.110102272033691,
      "epoch": 1.5405333804309431,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004765810521346493,
      "loss": 5.8097,
      "mean_token_accuracy": 0.17039534598588943,
      "num_tokens": 32516761.0,
      "step": 17445
    },
    {
      "entropy": 6.0619882106781,
      "epoch": 1.5409749205227836,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047656706087956604,
      "loss": 5.8278,
      "mean_token_accuracy": 0.1719517230987549,
      "num_tokens": 32526585.0,
      "step": 17450
    },
    {
      "entropy": 6.115849494934082,
      "epoch": 1.5414164606146237,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047655306567589864,
      "loss": 5.8317,
      "mean_token_accuracy": 0.1625863641500473,
      "num_tokens": 32535607.0,
      "step": 17455
    },
    {
      "entropy": 6.049368047714234,
      "epoch": 1.5418580007064642,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047653906652392135,
      "loss": 5.7567,
      "mean_token_accuracy": 0.17191615998744963,
      "num_tokens": 32545401.0,
      "step": 17460
    },
    {
      "entropy": 6.130995178222657,
      "epoch": 1.5422995407983044,
      "grad_norm": 2.375,
      "learning_rate": 0.0004765250634239083,
      "loss": 5.823,
      "mean_token_accuracy": 0.17306451201438905,
      "num_tokens": 32554421.0,
      "step": 17465
    },
    {
      "entropy": 6.144204759597779,
      "epoch": 1.5427410808901447,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047651105637613395,
      "loss": 5.8963,
      "mean_token_accuracy": 0.161521315574646,
      "num_tokens": 32563652.0,
      "step": 17470
    },
    {
      "entropy": 6.106625175476074,
      "epoch": 1.5431826209819852,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004764970453808726,
      "loss": 5.8811,
      "mean_token_accuracy": 0.17197612226009368,
      "num_tokens": 32572812.0,
      "step": 17475
    },
    {
      "entropy": 6.049667024612427,
      "epoch": 1.5436241610738255,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047648303043839875,
      "loss": 5.8094,
      "mean_token_accuracy": 0.17532625943422317,
      "num_tokens": 32581746.0,
      "step": 17480
    },
    {
      "entropy": 6.15779070854187,
      "epoch": 1.5440657011656658,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004764690115489869,
      "loss": 5.8388,
      "mean_token_accuracy": 0.16599915325641632,
      "num_tokens": 32591596.0,
      "step": 17485
    },
    {
      "entropy": 6.06755838394165,
      "epoch": 1.5445072412575063,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047645498871291163,
      "loss": 5.826,
      "mean_token_accuracy": 0.16419705003499985,
      "num_tokens": 32600610.0,
      "step": 17490
    },
    {
      "entropy": 6.139260482788086,
      "epoch": 1.5449487813493465,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004764409619304478,
      "loss": 5.8582,
      "mean_token_accuracy": 0.16197474896907807,
      "num_tokens": 32609827.0,
      "step": 17495
    },
    {
      "entropy": 6.10739278793335,
      "epoch": 1.5453903214411868,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047642693120187007,
      "loss": 5.8807,
      "mean_token_accuracy": 0.1677936613559723,
      "num_tokens": 32618258.0,
      "step": 17500
    },
    {
      "entropy": 6.110782670974731,
      "epoch": 1.5458318615330273,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004764128965274533,
      "loss": 5.8579,
      "mean_token_accuracy": 0.16667877733707429,
      "num_tokens": 32628155.0,
      "step": 17505
    },
    {
      "entropy": 6.155669116973877,
      "epoch": 1.5462734016248674,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047639885790747244,
      "loss": 5.9236,
      "mean_token_accuracy": 0.16263431161642075,
      "num_tokens": 32638347.0,
      "step": 17510
    },
    {
      "entropy": 6.144860649108887,
      "epoch": 1.5467149417167079,
      "grad_norm": 3.25,
      "learning_rate": 0.00047638481534220244,
      "loss": 5.7967,
      "mean_token_accuracy": 0.16820589154958726,
      "num_tokens": 32648636.0,
      "step": 17515
    },
    {
      "entropy": 6.114885425567627,
      "epoch": 1.5471564818085484,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004763707688319185,
      "loss": 5.8337,
      "mean_token_accuracy": 0.167985238134861,
      "num_tokens": 32657762.0,
      "step": 17520
    },
    {
      "entropy": 6.081134510040283,
      "epoch": 1.5475980219003884,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004763567183768958,
      "loss": 5.8374,
      "mean_token_accuracy": 0.1608771026134491,
      "num_tokens": 32668265.0,
      "step": 17525
    },
    {
      "entropy": 6.065871334075927,
      "epoch": 1.548039561992229,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047634266397740934,
      "loss": 5.7515,
      "mean_token_accuracy": 0.1711367830634117,
      "num_tokens": 32678051.0,
      "step": 17530
    },
    {
      "entropy": 6.186676073074341,
      "epoch": 1.5484811020840692,
      "grad_norm": 2.25,
      "learning_rate": 0.00047632860563373466,
      "loss": 5.9376,
      "mean_token_accuracy": 0.16790214478969573,
      "num_tokens": 32687186.0,
      "step": 17535
    },
    {
      "entropy": 6.105017375946045,
      "epoch": 1.5489226421759095,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004763145433461472,
      "loss": 5.8347,
      "mean_token_accuracy": 0.16769083142280578,
      "num_tokens": 32697082.0,
      "step": 17540
    },
    {
      "entropy": 6.111582279205322,
      "epoch": 1.54936418226775,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004763004771149222,
      "loss": 5.8437,
      "mean_token_accuracy": 0.1625613585114479,
      "num_tokens": 32706502.0,
      "step": 17545
    },
    {
      "entropy": 6.153810501098633,
      "epoch": 1.5498057223595902,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004762864069403354,
      "loss": 5.8021,
      "mean_token_accuracy": 0.1680271714925766,
      "num_tokens": 32715035.0,
      "step": 17550
    },
    {
      "entropy": 6.092133283615112,
      "epoch": 1.5502472624514305,
      "grad_norm": 2.0,
      "learning_rate": 0.00047627233282266236,
      "loss": 5.8375,
      "mean_token_accuracy": 0.16458076983690262,
      "num_tokens": 32724713.0,
      "step": 17555
    },
    {
      "entropy": 6.060888576507568,
      "epoch": 1.550688802543271,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004762582547621788,
      "loss": 5.8492,
      "mean_token_accuracy": 0.17185103744268418,
      "num_tokens": 32734584.0,
      "step": 17560
    },
    {
      "entropy": 6.1040277004241945,
      "epoch": 1.5511303426351113,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047624417275916045,
      "loss": 5.8113,
      "mean_token_accuracy": 0.1722095087170601,
      "num_tokens": 32742744.0,
      "step": 17565
    },
    {
      "entropy": 6.050622272491455,
      "epoch": 1.5515718827269516,
      "grad_norm": 2.421875,
      "learning_rate": 0.00047623008681388324,
      "loss": 5.7467,
      "mean_token_accuracy": 0.17936375588178635,
      "num_tokens": 32752051.0,
      "step": 17570
    },
    {
      "entropy": 6.125229740142823,
      "epoch": 1.552013422818792,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047621599692662306,
      "loss": 5.8303,
      "mean_token_accuracy": 0.1644992396235466,
      "num_tokens": 32761740.0,
      "step": 17575
    },
    {
      "entropy": 6.0934717655181885,
      "epoch": 1.5524549629106321,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047620190309765595,
      "loss": 5.7771,
      "mean_token_accuracy": 0.1796782582998276,
      "num_tokens": 32770809.0,
      "step": 17580
    },
    {
      "entropy": 6.136604452133179,
      "epoch": 1.5528965030024726,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047618780532725805,
      "loss": 5.8869,
      "mean_token_accuracy": 0.15972867459058762,
      "num_tokens": 32780614.0,
      "step": 17585
    },
    {
      "entropy": 6.118178272247315,
      "epoch": 1.5533380430943131,
      "grad_norm": 2.5,
      "learning_rate": 0.00047617370361570534,
      "loss": 5.8175,
      "mean_token_accuracy": 0.1706744596362114,
      "num_tokens": 32790336.0,
      "step": 17590
    },
    {
      "entropy": 6.054000186920166,
      "epoch": 1.5537795831861532,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004761595979632743,
      "loss": 5.8011,
      "mean_token_accuracy": 0.17311388552188872,
      "num_tokens": 32798991.0,
      "step": 17595
    },
    {
      "entropy": 6.139200735092163,
      "epoch": 1.5542211232779937,
      "grad_norm": 2.5,
      "learning_rate": 0.0004761454883702411,
      "loss": 5.8591,
      "mean_token_accuracy": 0.16659257411956788,
      "num_tokens": 32807616.0,
      "step": 17600
    },
    {
      "entropy": 6.107923603057861,
      "epoch": 1.554662663369834,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004761313748368821,
      "loss": 5.8853,
      "mean_token_accuracy": 0.16499450355768203,
      "num_tokens": 32817107.0,
      "step": 17605
    },
    {
      "entropy": 6.108745765686035,
      "epoch": 1.5551042034616742,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004761172573634739,
      "loss": 5.863,
      "mean_token_accuracy": 0.1676663115620613,
      "num_tokens": 32827205.0,
      "step": 17610
    },
    {
      "entropy": 6.121722555160522,
      "epoch": 1.5555457435535147,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004761031359502931,
      "loss": 5.7407,
      "mean_token_accuracy": 0.17420295476913453,
      "num_tokens": 32836102.0,
      "step": 17615
    },
    {
      "entropy": 6.113816022872925,
      "epoch": 1.555987283645355,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047608901059761617,
      "loss": 5.8854,
      "mean_token_accuracy": 0.16251321583986283,
      "num_tokens": 32845389.0,
      "step": 17620
    },
    {
      "entropy": 6.041683626174927,
      "epoch": 1.5564288237371953,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047607488130571997,
      "loss": 5.7607,
      "mean_token_accuracy": 0.17484191954135894,
      "num_tokens": 32854186.0,
      "step": 17625
    },
    {
      "entropy": 6.01494722366333,
      "epoch": 1.5568703638290358,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004760607480748811,
      "loss": 5.7723,
      "mean_token_accuracy": 0.1677704244852066,
      "num_tokens": 32863498.0,
      "step": 17630
    },
    {
      "entropy": 6.039358711242675,
      "epoch": 1.557311903920876,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047604661090537657,
      "loss": 5.8852,
      "mean_token_accuracy": 0.16316410005092621,
      "num_tokens": 32872483.0,
      "step": 17635
    },
    {
      "entropy": 6.165162420272827,
      "epoch": 1.5577534440127163,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004760324697974832,
      "loss": 5.8888,
      "mean_token_accuracy": 0.16196616739034653,
      "num_tokens": 32882739.0,
      "step": 17640
    },
    {
      "entropy": 6.227752923965454,
      "epoch": 1.5581949841045568,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004760183247514782,
      "loss": 5.9618,
      "mean_token_accuracy": 0.1599683418869972,
      "num_tokens": 32892707.0,
      "step": 17645
    },
    {
      "entropy": 6.168838024139404,
      "epoch": 1.5586365241963969,
      "grad_norm": 2.34375,
      "learning_rate": 0.00047600417576763843,
      "loss": 5.8625,
      "mean_token_accuracy": 0.16548029631376265,
      "num_tokens": 32901916.0,
      "step": 17650
    },
    {
      "entropy": 6.092421007156372,
      "epoch": 1.5590780642882374,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004759900228462412,
      "loss": 5.8251,
      "mean_token_accuracy": 0.16327809989452363,
      "num_tokens": 32910656.0,
      "step": 17655
    },
    {
      "entropy": 6.181734704971314,
      "epoch": 1.5595196043800779,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004759758659875638,
      "loss": 5.9127,
      "mean_token_accuracy": 0.1692071095108986,
      "num_tokens": 32919579.0,
      "step": 17660
    },
    {
      "entropy": 6.078861951828003,
      "epoch": 1.559961144471918,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047596170519188335,
      "loss": 5.8544,
      "mean_token_accuracy": 0.1651238664984703,
      "num_tokens": 32928271.0,
      "step": 17665
    },
    {
      "entropy": 6.085068655014038,
      "epoch": 1.5604026845637584,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004759475404594776,
      "loss": 5.7633,
      "mean_token_accuracy": 0.17588693350553514,
      "num_tokens": 32937671.0,
      "step": 17670
    },
    {
      "entropy": 6.132420301437378,
      "epoch": 1.5608442246555987,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004759333717906236,
      "loss": 5.9197,
      "mean_token_accuracy": 0.16314699798822402,
      "num_tokens": 32946896.0,
      "step": 17675
    },
    {
      "entropy": 6.114155292510986,
      "epoch": 1.561285764747439,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047591919918559927,
      "loss": 5.9092,
      "mean_token_accuracy": 0.16083289980888366,
      "num_tokens": 32956706.0,
      "step": 17680
    },
    {
      "entropy": 6.1080348014831545,
      "epoch": 1.5617273048392795,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004759050226446821,
      "loss": 5.7592,
      "mean_token_accuracy": 0.17548486292362214,
      "num_tokens": 32965990.0,
      "step": 17685
    },
    {
      "entropy": 6.098467493057251,
      "epoch": 1.5621688449311197,
      "grad_norm": 2.75,
      "learning_rate": 0.0004758908421681497,
      "loss": 5.7529,
      "mean_token_accuracy": 0.17205052673816681,
      "num_tokens": 32975311.0,
      "step": 17690
    },
    {
      "entropy": 6.128306484222412,
      "epoch": 1.56261038502296,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047587665775627997,
      "loss": 5.8585,
      "mean_token_accuracy": 0.1653894618153572,
      "num_tokens": 32984641.0,
      "step": 17695
    },
    {
      "entropy": 6.07186107635498,
      "epoch": 1.5630519251148005,
      "grad_norm": 4.5625,
      "learning_rate": 0.00047586246940935087,
      "loss": 5.8797,
      "mean_token_accuracy": 0.1645336151123047,
      "num_tokens": 32994696.0,
      "step": 17700
    },
    {
      "entropy": 6.110556793212891,
      "epoch": 1.5634934652066408,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047584827712764013,
      "loss": 5.7845,
      "mean_token_accuracy": 0.16879519820213318,
      "num_tokens": 33003392.0,
      "step": 17705
    },
    {
      "entropy": 6.075239944458008,
      "epoch": 1.563935005298481,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004758340809114259,
      "loss": 5.8489,
      "mean_token_accuracy": 0.16600215882062913,
      "num_tokens": 33012896.0,
      "step": 17710
    },
    {
      "entropy": 6.149149322509766,
      "epoch": 1.5643765453903216,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004758198807609862,
      "loss": 5.8874,
      "mean_token_accuracy": 0.16498418003320695,
      "num_tokens": 33022710.0,
      "step": 17715
    },
    {
      "entropy": 6.134741306304932,
      "epoch": 1.5648180854821618,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004758056766765993,
      "loss": 5.8271,
      "mean_token_accuracy": 0.16969270408153533,
      "num_tokens": 33031982.0,
      "step": 17720
    },
    {
      "entropy": 6.115036916732788,
      "epoch": 1.5652596255740021,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047579146865854333,
      "loss": 5.9232,
      "mean_token_accuracy": 0.1653538629412651,
      "num_tokens": 33041496.0,
      "step": 17725
    },
    {
      "entropy": 6.118243408203125,
      "epoch": 1.5657011656658426,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047577725670709675,
      "loss": 5.827,
      "mean_token_accuracy": 0.1665562257170677,
      "num_tokens": 33050764.0,
      "step": 17730
    },
    {
      "entropy": 6.101021480560303,
      "epoch": 1.5661427057576827,
      "grad_norm": 6.46875,
      "learning_rate": 0.000475763040822538,
      "loss": 5.8844,
      "mean_token_accuracy": 0.16952865570783615,
      "num_tokens": 33060809.0,
      "step": 17735
    },
    {
      "entropy": 6.029382371902466,
      "epoch": 1.5665842458495232,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004757488210051453,
      "loss": 5.7267,
      "mean_token_accuracy": 0.17556411921977996,
      "num_tokens": 33070155.0,
      "step": 17740
    },
    {
      "entropy": 6.090768432617187,
      "epoch": 1.5670257859413634,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004757345972551974,
      "loss": 5.836,
      "mean_token_accuracy": 0.17216530740261077,
      "num_tokens": 33079354.0,
      "step": 17745
    },
    {
      "entropy": 6.078714227676391,
      "epoch": 1.5674673260332037,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004757203695729729,
      "loss": 5.7997,
      "mean_token_accuracy": 0.17082267701625825,
      "num_tokens": 33087877.0,
      "step": 17750
    },
    {
      "entropy": 6.1016439437866214,
      "epoch": 1.5679088661250442,
      "grad_norm": 2.75,
      "learning_rate": 0.00047570613795875055,
      "loss": 5.8766,
      "mean_token_accuracy": 0.16317350566387176,
      "num_tokens": 33097375.0,
      "step": 17755
    },
    {
      "entropy": 6.130765533447265,
      "epoch": 1.5683504062168845,
      "grad_norm": 2.125,
      "learning_rate": 0.0004756919024128091,
      "loss": 5.8446,
      "mean_token_accuracy": 0.16809494346380233,
      "num_tokens": 33106336.0,
      "step": 17760
    },
    {
      "entropy": 5.99042820930481,
      "epoch": 1.5687919463087248,
      "grad_norm": 2.125,
      "learning_rate": 0.0004756776629354274,
      "loss": 5.6873,
      "mean_token_accuracy": 0.18224185705184937,
      "num_tokens": 33115753.0,
      "step": 17765
    },
    {
      "entropy": 6.103987979888916,
      "epoch": 1.5692334864005653,
      "grad_norm": 3.5,
      "learning_rate": 0.0004756634195268844,
      "loss": 5.855,
      "mean_token_accuracy": 0.15965788811445236,
      "num_tokens": 33125205.0,
      "step": 17770
    },
    {
      "entropy": 6.085108232498169,
      "epoch": 1.5696750264924055,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004756491721874592,
      "loss": 5.8026,
      "mean_token_accuracy": 0.172939333319664,
      "num_tokens": 33134470.0,
      "step": 17775
    },
    {
      "entropy": 6.041581821441651,
      "epoch": 1.5701165665842458,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047563492091743075,
      "loss": 5.7605,
      "mean_token_accuracy": 0.17684805691242217,
      "num_tokens": 33143346.0,
      "step": 17780
    },
    {
      "entropy": 6.03066930770874,
      "epoch": 1.5705581066760863,
      "grad_norm": 2.6875,
      "learning_rate": 0.00047562066571707836,
      "loss": 5.7801,
      "mean_token_accuracy": 0.1738138735294342,
      "num_tokens": 33153291.0,
      "step": 17785
    },
    {
      "entropy": 6.177571201324463,
      "epoch": 1.5709996467679266,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047560640658668126,
      "loss": 5.9569,
      "mean_token_accuracy": 0.16048300862312317,
      "num_tokens": 33163642.0,
      "step": 17790
    },
    {
      "entropy": 6.171758699417114,
      "epoch": 1.5714411868597669,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047559214352651863,
      "loss": 5.788,
      "mean_token_accuracy": 0.172685644030571,
      "num_tokens": 33173529.0,
      "step": 17795
    },
    {
      "entropy": 6.133573913574219,
      "epoch": 1.5718827269516074,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047557787653687014,
      "loss": 5.7876,
      "mean_token_accuracy": 0.16866244226694108,
      "num_tokens": 33180877.0,
      "step": 17800
    },
    {
      "entropy": 6.070457649230957,
      "epoch": 1.5723242670434474,
      "grad_norm": 2.25,
      "learning_rate": 0.000475563605618015,
      "loss": 5.8679,
      "mean_token_accuracy": 0.16552094370126724,
      "num_tokens": 33190831.0,
      "step": 17805
    },
    {
      "entropy": 6.116821527481079,
      "epoch": 1.572765807135288,
      "grad_norm": 3.46875,
      "learning_rate": 0.000475549330770233,
      "loss": 5.8965,
      "mean_token_accuracy": 0.1590697318315506,
      "num_tokens": 33200313.0,
      "step": 17810
    },
    {
      "entropy": 6.102489566802978,
      "epoch": 1.5732073472271282,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004755350519938036,
      "loss": 5.8374,
      "mean_token_accuracy": 0.1678464502096176,
      "num_tokens": 33209272.0,
      "step": 17815
    },
    {
      "entropy": 6.070160675048828,
      "epoch": 1.5736488873189685,
      "grad_norm": 2.5,
      "learning_rate": 0.00047552076928900663,
      "loss": 5.8107,
      "mean_token_accuracy": 0.1785772755742073,
      "num_tokens": 33218712.0,
      "step": 17820
    },
    {
      "entropy": 6.051478862762451,
      "epoch": 1.574090427410809,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004755064826561217,
      "loss": 5.8966,
      "mean_token_accuracy": 0.16399329900741577,
      "num_tokens": 33227480.0,
      "step": 17825
    },
    {
      "entropy": 6.063170099258423,
      "epoch": 1.5745319675026492,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004754921920954289,
      "loss": 5.801,
      "mean_token_accuracy": 0.16158377677202224,
      "num_tokens": 33236044.0,
      "step": 17830
    },
    {
      "entropy": 6.114393711090088,
      "epoch": 1.5749735075944895,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004754778976072081,
      "loss": 5.8426,
      "mean_token_accuracy": 0.16717129200696945,
      "num_tokens": 33244770.0,
      "step": 17835
    },
    {
      "entropy": 6.098041343688965,
      "epoch": 1.57541504768633,
      "grad_norm": 4.0,
      "learning_rate": 0.00047546359919173933,
      "loss": 5.7673,
      "mean_token_accuracy": 0.171917362511158,
      "num_tokens": 33254178.0,
      "step": 17840
    },
    {
      "entropy": 6.020664024353027,
      "epoch": 1.5758565877781703,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004754492968493026,
      "loss": 5.8312,
      "mean_token_accuracy": 0.172230064868927,
      "num_tokens": 33264210.0,
      "step": 17845
    },
    {
      "entropy": 6.013358020782471,
      "epoch": 1.5762981278700106,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004754349905801782,
      "loss": 5.7489,
      "mean_token_accuracy": 0.17490144968032836,
      "num_tokens": 33273348.0,
      "step": 17850
    },
    {
      "entropy": 6.129894971847534,
      "epoch": 1.576739667961851,
      "grad_norm": 2.5,
      "learning_rate": 0.00047542068038464625,
      "loss": 5.7857,
      "mean_token_accuracy": 0.17762259691953658,
      "num_tokens": 33281906.0,
      "step": 17855
    },
    {
      "entropy": 6.177788782119751,
      "epoch": 1.5771812080536913,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004754063662629872,
      "loss": 5.9101,
      "mean_token_accuracy": 0.16242618411779403,
      "num_tokens": 33291363.0,
      "step": 17860
    },
    {
      "entropy": 6.0520867824554445,
      "epoch": 1.5776227481455316,
      "grad_norm": 2.421875,
      "learning_rate": 0.00047539204821548146,
      "loss": 5.8211,
      "mean_token_accuracy": 0.17256740778684615,
      "num_tokens": 33299785.0,
      "step": 17865
    },
    {
      "entropy": 6.202607727050781,
      "epoch": 1.578064288237372,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004753777262424094,
      "loss": 5.9217,
      "mean_token_accuracy": 0.16302428692579268,
      "num_tokens": 33309655.0,
      "step": 17870
    },
    {
      "entropy": 6.127539014816284,
      "epoch": 1.5785058283292122,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047536340034405166,
      "loss": 5.8888,
      "mean_token_accuracy": 0.16510090380907058,
      "num_tokens": 33320158.0,
      "step": 17875
    },
    {
      "entropy": 6.032534980773926,
      "epoch": 1.5789473684210527,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047534907052068886,
      "loss": 5.7832,
      "mean_token_accuracy": 0.17676863819360733,
      "num_tokens": 33328959.0,
      "step": 17880
    },
    {
      "entropy": 6.141665029525757,
      "epoch": 1.579388908512893,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004753347367726017,
      "loss": 5.894,
      "mean_token_accuracy": 0.16338335424661637,
      "num_tokens": 33338624.0,
      "step": 17885
    },
    {
      "entropy": 6.139796876907349,
      "epoch": 1.5798304486047332,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047532039910007097,
      "loss": 5.8484,
      "mean_token_accuracy": 0.16992269307374955,
      "num_tokens": 33348083.0,
      "step": 17890
    },
    {
      "entropy": 6.106912422180176,
      "epoch": 1.5802719886965737,
      "grad_norm": 3.671875,
      "learning_rate": 0.00047530605750337765,
      "loss": 5.7847,
      "mean_token_accuracy": 0.16629447638988495,
      "num_tokens": 33358226.0,
      "step": 17895
    },
    {
      "entropy": 6.113605403900147,
      "epoch": 1.580713528788414,
      "grad_norm": 3.75,
      "learning_rate": 0.00047529171198280244,
      "loss": 5.9236,
      "mean_token_accuracy": 0.16566363275051116,
      "num_tokens": 33367997.0,
      "step": 17900
    },
    {
      "entropy": 6.128067445755005,
      "epoch": 1.5811550688802543,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047527736253862664,
      "loss": 5.8783,
      "mean_token_accuracy": 0.1596900627017021,
      "num_tokens": 33377574.0,
      "step": 17905
    },
    {
      "entropy": 6.109926366806031,
      "epoch": 1.5815966089720948,
      "grad_norm": 1.9375,
      "learning_rate": 0.00047526300917113116,
      "loss": 5.9044,
      "mean_token_accuracy": 0.16743322014808654,
      "num_tokens": 33387016.0,
      "step": 17910
    },
    {
      "entropy": 6.133172607421875,
      "epoch": 1.582038149063935,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004752486518805972,
      "loss": 5.8346,
      "mean_token_accuracy": 0.15947234481573105,
      "num_tokens": 33396239.0,
      "step": 17915
    },
    {
      "entropy": 6.031153726577759,
      "epoch": 1.5824796891557753,
      "grad_norm": 3.421875,
      "learning_rate": 0.00047523429066730603,
      "loss": 5.7164,
      "mean_token_accuracy": 0.1769706651568413,
      "num_tokens": 33404738.0,
      "step": 17920
    },
    {
      "entropy": 5.965107870101929,
      "epoch": 1.5829212292476158,
      "grad_norm": 3.6875,
      "learning_rate": 0.00047521992553153903,
      "loss": 5.7229,
      "mean_token_accuracy": 0.18078763782978058,
      "num_tokens": 33413670.0,
      "step": 17925
    },
    {
      "entropy": 6.05218358039856,
      "epoch": 1.583362769339456,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004752055564735775,
      "loss": 5.8256,
      "mean_token_accuracy": 0.15891026109457015,
      "num_tokens": 33422841.0,
      "step": 17930
    },
    {
      "entropy": 6.194892168045044,
      "epoch": 1.5838043094312964,
      "grad_norm": 2.5,
      "learning_rate": 0.00047519118349370306,
      "loss": 5.8712,
      "mean_token_accuracy": 0.16661487966775895,
      "num_tokens": 33432023.0,
      "step": 17935
    },
    {
      "entropy": 6.143595218658447,
      "epoch": 1.5842458495231369,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004751768065921971,
      "loss": 5.8816,
      "mean_token_accuracy": 0.15850434601306915,
      "num_tokens": 33442181.0,
      "step": 17940
    },
    {
      "entropy": 6.148207426071167,
      "epoch": 1.584687389614977,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047516242576934144,
      "loss": 5.8991,
      "mean_token_accuracy": 0.15497517734766006,
      "num_tokens": 33451847.0,
      "step": 17945
    },
    {
      "entropy": 6.147199296951294,
      "epoch": 1.5851289297068174,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047514804102541765,
      "loss": 5.8798,
      "mean_token_accuracy": 0.16751822978258132,
      "num_tokens": 33461587.0,
      "step": 17950
    },
    {
      "entropy": 6.201934480667115,
      "epoch": 1.5855704697986577,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047513365236070753,
      "loss": 5.907,
      "mean_token_accuracy": 0.16351648420095444,
      "num_tokens": 33471553.0,
      "step": 17955
    },
    {
      "entropy": 6.028347635269165,
      "epoch": 1.586012009890498,
      "grad_norm": 3.109375,
      "learning_rate": 0.00047511925977549303,
      "loss": 5.7427,
      "mean_token_accuracy": 0.1773436188697815,
      "num_tokens": 33480984.0,
      "step": 17960
    },
    {
      "entropy": 6.066892385482788,
      "epoch": 1.5864535499823385,
      "grad_norm": 4.5,
      "learning_rate": 0.00047510486327005604,
      "loss": 5.8767,
      "mean_token_accuracy": 0.16242087185382842,
      "num_tokens": 33491196.0,
      "step": 17965
    },
    {
      "entropy": 5.952466154098511,
      "epoch": 1.5868950900741787,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047509046284467855,
      "loss": 5.736,
      "mean_token_accuracy": 0.17176616936922073,
      "num_tokens": 33500404.0,
      "step": 17970
    },
    {
      "entropy": 6.184411287307739,
      "epoch": 1.587336630166019,
      "grad_norm": 2.734375,
      "learning_rate": 0.00047507605849964274,
      "loss": 5.9267,
      "mean_token_accuracy": 0.15994629859924317,
      "num_tokens": 33509281.0,
      "step": 17975
    },
    {
      "entropy": 6.183355045318604,
      "epoch": 1.5877781702578595,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004750616502352307,
      "loss": 5.8268,
      "mean_token_accuracy": 0.15847884267568588,
      "num_tokens": 33518556.0,
      "step": 17980
    },
    {
      "entropy": 6.207890319824219,
      "epoch": 1.5882197103496998,
      "grad_norm": 3.421875,
      "learning_rate": 0.00047504723805172474,
      "loss": 6.0002,
      "mean_token_accuracy": 0.15450143069028854,
      "num_tokens": 33529266.0,
      "step": 17985
    },
    {
      "entropy": 6.0935853004455565,
      "epoch": 1.58866125044154,
      "grad_norm": 3.328125,
      "learning_rate": 0.00047503282194940715,
      "loss": 5.8715,
      "mean_token_accuracy": 0.1618405908346176,
      "num_tokens": 33538972.0,
      "step": 17990
    },
    {
      "entropy": 6.095176935195923,
      "epoch": 1.5891027905333805,
      "grad_norm": 5.71875,
      "learning_rate": 0.0004750184019285603,
      "loss": 5.8104,
      "mean_token_accuracy": 0.16553851664066316,
      "num_tokens": 33549528.0,
      "step": 17995
    },
    {
      "entropy": 6.063725090026855,
      "epoch": 1.5895443306252208,
      "grad_norm": 2.5,
      "learning_rate": 0.0004750039779894668,
      "loss": 5.7225,
      "mean_token_accuracy": 0.17327024042606354,
      "num_tokens": 33557828.0,
      "step": 18000
    },
    {
      "epoch": 1.5895443306252208,
      "eval_entropy": 5.861491500102785,
      "eval_loss": 5.894139766693115,
      "eval_mean_token_accuracy": 0.17310694614894653,
      "eval_num_tokens": 33557828.0,
      "eval_runtime": 26.3984,
      "eval_samples_per_second": 1337.772,
      "eval_steps_per_second": 167.245,
      "step": 18000
    },
    {
      "entropy": 6.12619194984436,
      "epoch": 1.589985870717061,
      "grad_norm": 3.796875,
      "learning_rate": 0.00047498955013240907,
      "loss": 5.886,
      "mean_token_accuracy": 0.16243494302034378,
      "num_tokens": 33567087.0,
      "step": 18005
    },
    {
      "entropy": 6.119371700286865,
      "epoch": 1.5904274108089016,
      "grad_norm": 2.984375,
      "learning_rate": 0.00047497511835766984,
      "loss": 5.866,
      "mean_token_accuracy": 0.16054177582263945,
      "num_tokens": 33577442.0,
      "step": 18010
    },
    {
      "entropy": 6.162020540237426,
      "epoch": 1.5908689509007417,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047496068266553165,
      "loss": 5.8702,
      "mean_token_accuracy": 0.16661471575498582,
      "num_tokens": 33585655.0,
      "step": 18015
    },
    {
      "entropy": 6.11450777053833,
      "epoch": 1.5913104909925821,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047494624305627754,
      "loss": 5.8554,
      "mean_token_accuracy": 0.16912199109792708,
      "num_tokens": 33594808.0,
      "step": 18020
    },
    {
      "entropy": 6.097761726379394,
      "epoch": 1.5917520310844224,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004749317995301902,
      "loss": 5.8909,
      "mean_token_accuracy": 0.16618283614516258,
      "num_tokens": 33603924.0,
      "step": 18025
    },
    {
      "entropy": 6.0765392780303955,
      "epoch": 1.5921935711762627,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004749173520875525,
      "loss": 5.7887,
      "mean_token_accuracy": 0.17191557884216307,
      "num_tokens": 33613870.0,
      "step": 18030
    },
    {
      "entropy": 6.108762788772583,
      "epoch": 1.5926351112681032,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047490290072864773,
      "loss": 5.8407,
      "mean_token_accuracy": 0.17372720092535018,
      "num_tokens": 33622938.0,
      "step": 18035
    },
    {
      "entropy": 6.1648698329925535,
      "epoch": 1.5930766513599435,
      "grad_norm": 2.625,
      "learning_rate": 0.00047488844545375886,
      "loss": 5.8916,
      "mean_token_accuracy": 0.1655658647418022,
      "num_tokens": 33631520.0,
      "step": 18040
    },
    {
      "entropy": 6.190776252746582,
      "epoch": 1.5935181914517837,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004748739862631689,
      "loss": 5.8821,
      "mean_token_accuracy": 0.1694650560617447,
      "num_tokens": 33641365.0,
      "step": 18045
    },
    {
      "entropy": 6.100015687942505,
      "epoch": 1.5939597315436242,
      "grad_norm": 3.75,
      "learning_rate": 0.00047485952315716126,
      "loss": 5.6827,
      "mean_token_accuracy": 0.18006293773651122,
      "num_tokens": 33650425.0,
      "step": 18050
    },
    {
      "entropy": 6.1313316822052,
      "epoch": 1.5944012716354645,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047484505613601917,
      "loss": 5.8894,
      "mean_token_accuracy": 0.16520665287971498,
      "num_tokens": 33659928.0,
      "step": 18055
    },
    {
      "entropy": 6.044728851318359,
      "epoch": 1.5948428117273048,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004748305852000262,
      "loss": 5.8698,
      "mean_token_accuracy": 0.17164038121700287,
      "num_tokens": 33668860.0,
      "step": 18060
    },
    {
      "entropy": 6.143427610397339,
      "epoch": 1.5952843518191453,
      "grad_norm": 3.328125,
      "learning_rate": 0.00047481611034946565,
      "loss": 5.868,
      "mean_token_accuracy": 0.16181062906980515,
      "num_tokens": 33679343.0,
      "step": 18065
    },
    {
      "entropy": 6.084942483901978,
      "epoch": 1.5957258919109856,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004748016315846212,
      "loss": 5.7902,
      "mean_token_accuracy": 0.17311810702085495,
      "num_tokens": 33688749.0,
      "step": 18070
    },
    {
      "entropy": 6.13870325088501,
      "epoch": 1.5961674320028258,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00047478714890577633,
      "loss": 5.8786,
      "mean_token_accuracy": 0.16342048048973085,
      "num_tokens": 33698359.0,
      "step": 18075
    },
    {
      "entropy": 6.129868030548096,
      "epoch": 1.5966089720946663,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047477266231321484,
      "loss": 5.8746,
      "mean_token_accuracy": 0.1600186124444008,
      "num_tokens": 33707402.0,
      "step": 18080
    },
    {
      "entropy": 6.086358118057251,
      "epoch": 1.5970505121865064,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047475817180722056,
      "loss": 5.7863,
      "mean_token_accuracy": 0.17048639357089995,
      "num_tokens": 33717043.0,
      "step": 18085
    },
    {
      "entropy": 6.046060609817505,
      "epoch": 1.597492052278347,
      "grad_norm": 2.875,
      "learning_rate": 0.00047474367738807734,
      "loss": 5.7894,
      "mean_token_accuracy": 0.17396627962589264,
      "num_tokens": 33726527.0,
      "step": 18090
    },
    {
      "entropy": 6.043662214279175,
      "epoch": 1.5979335923701872,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047472917905606903,
      "loss": 5.7806,
      "mean_token_accuracy": 0.1720337063074112,
      "num_tokens": 33735357.0,
      "step": 18095
    },
    {
      "entropy": 5.977050065994263,
      "epoch": 1.5983751324620274,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047471467681147974,
      "loss": 5.6831,
      "mean_token_accuracy": 0.18261580765247346,
      "num_tokens": 33743660.0,
      "step": 18100
    },
    {
      "entropy": 6.096145153045654,
      "epoch": 1.598816672553868,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004747001706545935,
      "loss": 5.8652,
      "mean_token_accuracy": 0.1608905479311943,
      "num_tokens": 33752899.0,
      "step": 18105
    },
    {
      "entropy": 6.1640928268432615,
      "epoch": 1.5992582126457082,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004746856605856944,
      "loss": 5.844,
      "mean_token_accuracy": 0.16726600825786592,
      "num_tokens": 33761333.0,
      "step": 18110
    },
    {
      "entropy": 6.0593029975891115,
      "epoch": 1.5996997527375485,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047467114660506687,
      "loss": 5.8899,
      "mean_token_accuracy": 0.16225569248199462,
      "num_tokens": 33770507.0,
      "step": 18115
    },
    {
      "entropy": 6.034752941131591,
      "epoch": 1.600141292829389,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004746566287129951,
      "loss": 5.8228,
      "mean_token_accuracy": 0.1644844338297844,
      "num_tokens": 33781658.0,
      "step": 18120
    },
    {
      "entropy": 6.142946243286133,
      "epoch": 1.6005828329212293,
      "grad_norm": 2.0,
      "learning_rate": 0.0004746421069097636,
      "loss": 5.8972,
      "mean_token_accuracy": 0.16087266504764558,
      "num_tokens": 33792229.0,
      "step": 18125
    },
    {
      "entropy": 6.199428510665894,
      "epoch": 1.6010243730130695,
      "grad_norm": 2.0,
      "learning_rate": 0.00047462758119565663,
      "loss": 5.8831,
      "mean_token_accuracy": 0.16761776506900788,
      "num_tokens": 33801674.0,
      "step": 18130
    },
    {
      "entropy": 6.157083797454834,
      "epoch": 1.60146591310491,
      "grad_norm": 4.21875,
      "learning_rate": 0.000474613051570959,
      "loss": 5.7913,
      "mean_token_accuracy": 0.17747995406389236,
      "num_tokens": 33812066.0,
      "step": 18135
    },
    {
      "entropy": 6.11303744316101,
      "epoch": 1.6019074531967503,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047459851803595524,
      "loss": 5.923,
      "mean_token_accuracy": 0.15796318352222444,
      "num_tokens": 33822206.0,
      "step": 18140
    },
    {
      "entropy": 6.110971879959107,
      "epoch": 1.6023489932885906,
      "grad_norm": 1.953125,
      "learning_rate": 0.00047458398059092995,
      "loss": 5.9263,
      "mean_token_accuracy": 0.16437760591506959,
      "num_tokens": 33832075.0,
      "step": 18145
    },
    {
      "entropy": 6.099058389663696,
      "epoch": 1.602790533380431,
      "grad_norm": 2.859375,
      "learning_rate": 0.00047456943923616793,
      "loss": 5.736,
      "mean_token_accuracy": 0.1736042767763138,
      "num_tokens": 33841523.0,
      "step": 18150
    },
    {
      "entropy": 6.13116602897644,
      "epoch": 1.6032320734722711,
      "grad_norm": 1.953125,
      "learning_rate": 0.00047455489397195416,
      "loss": 5.8074,
      "mean_token_accuracy": 0.16780513823032378,
      "num_tokens": 33850756.0,
      "step": 18155
    },
    {
      "entropy": 6.1135077476501465,
      "epoch": 1.6036736135641116,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047454034479857356,
      "loss": 5.8531,
      "mean_token_accuracy": 0.16833048611879348,
      "num_tokens": 33859611.0,
      "step": 18160
    },
    {
      "entropy": 6.041662359237671,
      "epoch": 1.604115153655952,
      "grad_norm": 1.78125,
      "learning_rate": 0.000474525791716311,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16119749695062638,
      "num_tokens": 33867993.0,
      "step": 18165
    },
    {
      "entropy": 6.143669319152832,
      "epoch": 1.6045566937477922,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047451123472545175,
      "loss": 5.9253,
      "mean_token_accuracy": 0.16710306853055953,
      "num_tokens": 33876785.0,
      "step": 18170
    },
    {
      "entropy": 6.199612951278686,
      "epoch": 1.6049982338396327,
      "grad_norm": 2.765625,
      "learning_rate": 0.00047449667382628083,
      "loss": 5.7851,
      "mean_token_accuracy": 0.17136625200510025,
      "num_tokens": 33885729.0,
      "step": 18175
    },
    {
      "entropy": 6.163266134262085,
      "epoch": 1.605439773931473,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004744821090190835,
      "loss": 5.8281,
      "mean_token_accuracy": 0.17416220307350158,
      "num_tokens": 33894622.0,
      "step": 18180
    },
    {
      "entropy": 6.039811277389527,
      "epoch": 1.6058813140233132,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047446754030414513,
      "loss": 5.902,
      "mean_token_accuracy": 0.16813846826553344,
      "num_tokens": 33904187.0,
      "step": 18185
    },
    {
      "entropy": 6.136535596847534,
      "epoch": 1.6063228541151537,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004744529676817511,
      "loss": 5.9027,
      "mean_token_accuracy": 0.16646450012922287,
      "num_tokens": 33914327.0,
      "step": 18190
    },
    {
      "entropy": 6.183498191833496,
      "epoch": 1.606764394206994,
      "grad_norm": 2.125,
      "learning_rate": 0.00047443839115218685,
      "loss": 5.7439,
      "mean_token_accuracy": 0.1703719362616539,
      "num_tokens": 33923424.0,
      "step": 18195
    },
    {
      "entropy": 6.0624064922332765,
      "epoch": 1.6072059342988343,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000474423810715738,
      "loss": 5.8012,
      "mean_token_accuracy": 0.17058018445968628,
      "num_tokens": 33932145.0,
      "step": 18200
    },
    {
      "entropy": 6.0381495475769045,
      "epoch": 1.6076474743906748,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004744092263726901,
      "loss": 5.8175,
      "mean_token_accuracy": 0.1711421087384224,
      "num_tokens": 33942505.0,
      "step": 18205
    },
    {
      "entropy": 6.1049176216125485,
      "epoch": 1.608089014482515,
      "grad_norm": 2.25,
      "learning_rate": 0.0004743946381233288,
      "loss": 5.8902,
      "mean_token_accuracy": 0.1634010061621666,
      "num_tokens": 33952156.0,
      "step": 18210
    },
    {
      "entropy": 6.12799301147461,
      "epoch": 1.6085305545743553,
      "grad_norm": 1.890625,
      "learning_rate": 0.00047438004596794,
      "loss": 5.872,
      "mean_token_accuracy": 0.1673312082886696,
      "num_tokens": 33962478.0,
      "step": 18215
    },
    {
      "entropy": 6.122446155548095,
      "epoch": 1.6089720946661958,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004743654499068095,
      "loss": 5.811,
      "mean_token_accuracy": 0.16983745247125626,
      "num_tokens": 33971581.0,
      "step": 18220
    },
    {
      "entropy": 6.128451728820801,
      "epoch": 1.609413634758036,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004743508499402232,
      "loss": 5.8079,
      "mean_token_accuracy": 0.1693253517150879,
      "num_tokens": 33981262.0,
      "step": 18225
    },
    {
      "entropy": 6.135485696792602,
      "epoch": 1.6098551748498764,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004743362460684671,
      "loss": 5.8486,
      "mean_token_accuracy": 0.16261979937553406,
      "num_tokens": 33989865.0,
      "step": 18230
    },
    {
      "entropy": 6.118477010726929,
      "epoch": 1.6102967149417167,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047432163829182737,
      "loss": 5.8473,
      "mean_token_accuracy": 0.16548852771520614,
      "num_tokens": 33999261.0,
      "step": 18235
    },
    {
      "entropy": 6.182919025421143,
      "epoch": 1.610738255033557,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047430702661059007,
      "loss": 5.8468,
      "mean_token_accuracy": 0.1630049780011177,
      "num_tokens": 34008471.0,
      "step": 18240
    },
    {
      "entropy": 6.028147602081299,
      "epoch": 1.6111797951253974,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047429241102504144,
      "loss": 5.734,
      "mean_token_accuracy": 0.1753491446375847,
      "num_tokens": 34018305.0,
      "step": 18245
    },
    {
      "entropy": 6.110628509521485,
      "epoch": 1.6116213352172377,
      "grad_norm": 3.640625,
      "learning_rate": 0.00047427779153546787,
      "loss": 5.8611,
      "mean_token_accuracy": 0.16029200553894044,
      "num_tokens": 34027393.0,
      "step": 18250
    },
    {
      "entropy": 6.138890266418457,
      "epoch": 1.612062875309078,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047426316814215563,
      "loss": 5.8054,
      "mean_token_accuracy": 0.17123512774705887,
      "num_tokens": 34035835.0,
      "step": 18255
    },
    {
      "entropy": 6.042346096038818,
      "epoch": 1.6125044154009185,
      "grad_norm": 2.125,
      "learning_rate": 0.00047424854084539134,
      "loss": 5.8801,
      "mean_token_accuracy": 0.16252967566251755,
      "num_tokens": 34045946.0,
      "step": 18260
    },
    {
      "entropy": 6.158521032333374,
      "epoch": 1.6129459554927588,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047423390964546145,
      "loss": 5.9403,
      "mean_token_accuracy": 0.1619979575276375,
      "num_tokens": 34055356.0,
      "step": 18265
    },
    {
      "entropy": 6.268468189239502,
      "epoch": 1.613387495584599,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004742192745426525,
      "loss": 5.9865,
      "mean_token_accuracy": 0.15419042855501175,
      "num_tokens": 34065862.0,
      "step": 18270
    },
    {
      "entropy": 6.213652038574219,
      "epoch": 1.6138290356764395,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004742046355372513,
      "loss": 5.8563,
      "mean_token_accuracy": 0.16763030588626862,
      "num_tokens": 34074980.0,
      "step": 18275
    },
    {
      "entropy": 6.018419075012207,
      "epoch": 1.6142705757682798,
      "grad_norm": 5.09375,
      "learning_rate": 0.00047418999262954466,
      "loss": 5.8662,
      "mean_token_accuracy": 0.16070133298635483,
      "num_tokens": 34085501.0,
      "step": 18280
    },
    {
      "entropy": 6.078172922134399,
      "epoch": 1.61471211586012,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004741753458198192,
      "loss": 5.853,
      "mean_token_accuracy": 0.16296856999397277,
      "num_tokens": 34095564.0,
      "step": 18285
    },
    {
      "entropy": 6.139164543151855,
      "epoch": 1.6151536559519606,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00047416069510836215,
      "loss": 5.8702,
      "mean_token_accuracy": 0.15560472905635833,
      "num_tokens": 34104571.0,
      "step": 18290
    },
    {
      "entropy": 6.147561740875244,
      "epoch": 1.6155951960438006,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004741460404954602,
      "loss": 5.8949,
      "mean_token_accuracy": 0.15459970384836197,
      "num_tokens": 34114272.0,
      "step": 18295
    },
    {
      "entropy": 6.059701538085937,
      "epoch": 1.6160367361356411,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004741313819814007,
      "loss": 5.8206,
      "mean_token_accuracy": 0.17209326326847077,
      "num_tokens": 34125521.0,
      "step": 18300
    },
    {
      "entropy": 6.10121111869812,
      "epoch": 1.6164782762274814,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004741167195664706,
      "loss": 5.8748,
      "mean_token_accuracy": 0.16273993402719497,
      "num_tokens": 34134447.0,
      "step": 18305
    },
    {
      "entropy": 6.072175550460815,
      "epoch": 1.6169198163193217,
      "grad_norm": 2.125,
      "learning_rate": 0.00047410205325095725,
      "loss": 5.7988,
      "mean_token_accuracy": 0.17279000729322433,
      "num_tokens": 34143207.0,
      "step": 18310
    },
    {
      "entropy": 6.089821767807007,
      "epoch": 1.6173613564111622,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004740873830351479,
      "loss": 5.814,
      "mean_token_accuracy": 0.16931044012308122,
      "num_tokens": 34153025.0,
      "step": 18315
    },
    {
      "entropy": 6.188735818862915,
      "epoch": 1.6178028965030025,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047407270891932995,
      "loss": 5.831,
      "mean_token_accuracy": 0.16417529433965683,
      "num_tokens": 34162793.0,
      "step": 18320
    },
    {
      "entropy": 6.101986646652222,
      "epoch": 1.6182444365948427,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047405803090379083,
      "loss": 5.8768,
      "mean_token_accuracy": 0.16687648892402648,
      "num_tokens": 34171941.0,
      "step": 18325
    },
    {
      "entropy": 6.14830002784729,
      "epoch": 1.6186859766866832,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004740433489888182,
      "loss": 5.8312,
      "mean_token_accuracy": 0.1696966201066971,
      "num_tokens": 34180993.0,
      "step": 18330
    },
    {
      "entropy": 6.147417116165161,
      "epoch": 1.6191275167785235,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047402866317469944,
      "loss": 5.772,
      "mean_token_accuracy": 0.1734404131770134,
      "num_tokens": 34190078.0,
      "step": 18335
    },
    {
      "entropy": 5.997342824935913,
      "epoch": 1.6195690568703638,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004740139734617224,
      "loss": 5.7625,
      "mean_token_accuracy": 0.1712697669863701,
      "num_tokens": 34199091.0,
      "step": 18340
    },
    {
      "entropy": 6.1161487102508545,
      "epoch": 1.6200105969622043,
      "grad_norm": 2.125,
      "learning_rate": 0.0004739992798501748,
      "loss": 5.8791,
      "mean_token_accuracy": 0.16644636243581773,
      "num_tokens": 34209042.0,
      "step": 18345
    },
    {
      "entropy": 6.1418968677520756,
      "epoch": 1.6204521370540446,
      "grad_norm": 1.625,
      "learning_rate": 0.00047398458234034445,
      "loss": 5.8466,
      "mean_token_accuracy": 0.16400086134672165,
      "num_tokens": 34217311.0,
      "step": 18350
    },
    {
      "entropy": 6.006878614425659,
      "epoch": 1.6208936771458848,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047396988093251934,
      "loss": 5.7994,
      "mean_token_accuracy": 0.17102440893650056,
      "num_tokens": 34227001.0,
      "step": 18355
    },
    {
      "entropy": 6.084216213226318,
      "epoch": 1.6213352172377253,
      "grad_norm": 2.0,
      "learning_rate": 0.0004739551756269875,
      "loss": 5.8528,
      "mean_token_accuracy": 0.16675664782524108,
      "num_tokens": 34236214.0,
      "step": 18360
    },
    {
      "entropy": 6.163750123977661,
      "epoch": 1.6217767573295654,
      "grad_norm": 2.125,
      "learning_rate": 0.0004739404664240368,
      "loss": 5.7818,
      "mean_token_accuracy": 0.17071945518255233,
      "num_tokens": 34244808.0,
      "step": 18365
    },
    {
      "entropy": 6.082055473327637,
      "epoch": 1.6222182974214059,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047392575332395556,
      "loss": 5.8003,
      "mean_token_accuracy": 0.17396869659423828,
      "num_tokens": 34253355.0,
      "step": 18370
    },
    {
      "entropy": 6.180101728439331,
      "epoch": 1.6226598375132462,
      "grad_norm": 1.875,
      "learning_rate": 0.0004739110363270319,
      "loss": 5.9167,
      "mean_token_accuracy": 0.15865453630685805,
      "num_tokens": 34262235.0,
      "step": 18375
    },
    {
      "entropy": 6.146726369857788,
      "epoch": 1.6231013776050864,
      "grad_norm": 1.984375,
      "learning_rate": 0.00047389631543355424,
      "loss": 5.7882,
      "mean_token_accuracy": 0.16891453117132188,
      "num_tokens": 34272174.0,
      "step": 18380
    },
    {
      "entropy": 6.080994510650635,
      "epoch": 1.623542917696927,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004738815906438108,
      "loss": 5.7872,
      "mean_token_accuracy": 0.17077571153640747,
      "num_tokens": 34281153.0,
      "step": 18385
    },
    {
      "entropy": 6.045297002792358,
      "epoch": 1.6239844577887672,
      "grad_norm": 4.625,
      "learning_rate": 0.00047386686195809017,
      "loss": 5.8297,
      "mean_token_accuracy": 0.17055407762527466,
      "num_tokens": 34291045.0,
      "step": 18390
    },
    {
      "entropy": 6.087706756591797,
      "epoch": 1.6244259978806075,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047385212937668084,
      "loss": 5.7754,
      "mean_token_accuracy": 0.17028800249099732,
      "num_tokens": 34299234.0,
      "step": 18395
    },
    {
      "entropy": 6.086493253707886,
      "epoch": 1.624867537972448,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047383739289987134,
      "loss": 5.867,
      "mean_token_accuracy": 0.16537597477436067,
      "num_tokens": 34309084.0,
      "step": 18400
    },
    {
      "entropy": 6.1628927230834964,
      "epoch": 1.6253090780642883,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004738226525279504,
      "loss": 5.9282,
      "mean_token_accuracy": 0.15648255199193956,
      "num_tokens": 34320039.0,
      "step": 18405
    },
    {
      "entropy": 6.092647647857666,
      "epoch": 1.6257506181561285,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004738079082612068,
      "loss": 5.7526,
      "mean_token_accuracy": 0.1702187478542328,
      "num_tokens": 34329020.0,
      "step": 18410
    },
    {
      "entropy": 6.104157161712647,
      "epoch": 1.626192158247969,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047379316009992927,
      "loss": 5.8557,
      "mean_token_accuracy": 0.1603140115737915,
      "num_tokens": 34337953.0,
      "step": 18415
    },
    {
      "entropy": 6.0291413307189945,
      "epoch": 1.6266336983398093,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047377840804440686,
      "loss": 5.7347,
      "mean_token_accuracy": 0.1758974850177765,
      "num_tokens": 34346564.0,
      "step": 18420
    },
    {
      "entropy": 6.123154020309448,
      "epoch": 1.6270752384316496,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004737636520949285,
      "loss": 5.968,
      "mean_token_accuracy": 0.16519565731287003,
      "num_tokens": 34357163.0,
      "step": 18425
    },
    {
      "entropy": 6.148588705062866,
      "epoch": 1.62751677852349,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004737488922517832,
      "loss": 5.9562,
      "mean_token_accuracy": 0.16161729842424394,
      "num_tokens": 34366731.0,
      "step": 18430
    },
    {
      "entropy": 6.132932710647583,
      "epoch": 1.6279583186153301,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047373412851526013,
      "loss": 5.8099,
      "mean_token_accuracy": 0.17021574527025224,
      "num_tokens": 34375845.0,
      "step": 18435
    },
    {
      "entropy": 6.129991865158081,
      "epoch": 1.6283998587071706,
      "grad_norm": 1.75,
      "learning_rate": 0.00047371936088564856,
      "loss": 5.8054,
      "mean_token_accuracy": 0.16528016626834868,
      "num_tokens": 34385335.0,
      "step": 18440
    },
    {
      "entropy": 6.143348455429077,
      "epoch": 1.628841398799011,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004737045893632377,
      "loss": 5.8826,
      "mean_token_accuracy": 0.16935659795999528,
      "num_tokens": 34394356.0,
      "step": 18445
    },
    {
      "entropy": 6.099414300918579,
      "epoch": 1.6292829388908512,
      "grad_norm": 2.5,
      "learning_rate": 0.00047368981394831696,
      "loss": 5.8453,
      "mean_token_accuracy": 0.16848586052656173,
      "num_tokens": 34403727.0,
      "step": 18450
    },
    {
      "entropy": 6.213435888290405,
      "epoch": 1.6297244789826917,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047367503464117573,
      "loss": 5.8809,
      "mean_token_accuracy": 0.16537810862064362,
      "num_tokens": 34413253.0,
      "step": 18455
    },
    {
      "entropy": 6.052283239364624,
      "epoch": 1.630166019074532,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047366025144210356,
      "loss": 5.8047,
      "mean_token_accuracy": 0.17044736444950104,
      "num_tokens": 34422430.0,
      "step": 18460
    },
    {
      "entropy": 6.08810601234436,
      "epoch": 1.6306075591663722,
      "grad_norm": 2.53125,
      "learning_rate": 0.00047364546435139004,
      "loss": 5.8303,
      "mean_token_accuracy": 0.1690466284751892,
      "num_tokens": 34431111.0,
      "step": 18465
    },
    {
      "entropy": 6.084423971176148,
      "epoch": 1.6310490992582127,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047363067336932493,
      "loss": 5.8274,
      "mean_token_accuracy": 0.17423311471939087,
      "num_tokens": 34440044.0,
      "step": 18470
    },
    {
      "entropy": 6.081137800216675,
      "epoch": 1.631490639350053,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004736158784961978,
      "loss": 5.8407,
      "mean_token_accuracy": 0.16822053790092467,
      "num_tokens": 34450315.0,
      "step": 18475
    },
    {
      "entropy": 6.07003436088562,
      "epoch": 1.6319321794418933,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00047360107973229865,
      "loss": 5.7517,
      "mean_token_accuracy": 0.18084010630846023,
      "num_tokens": 34459380.0,
      "step": 18480
    },
    {
      "entropy": 6.143307638168335,
      "epoch": 1.6323737195337338,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004735862770779172,
      "loss": 5.8683,
      "mean_token_accuracy": 0.16610395908355713,
      "num_tokens": 34468224.0,
      "step": 18485
    },
    {
      "entropy": 6.0548951625823975,
      "epoch": 1.632815259625574,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004735714705333436,
      "loss": 5.7347,
      "mean_token_accuracy": 0.16951258033514022,
      "num_tokens": 34476917.0,
      "step": 18490
    },
    {
      "entropy": 6.045917940139771,
      "epoch": 1.6332567997174143,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047355666009886785,
      "loss": 5.858,
      "mean_token_accuracy": 0.16378099620342254,
      "num_tokens": 34487192.0,
      "step": 18495
    },
    {
      "entropy": 6.129749441146851,
      "epoch": 1.6336983398092548,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00047354184577478,
      "loss": 5.808,
      "mean_token_accuracy": 0.173916058242321,
      "num_tokens": 34496658.0,
      "step": 18500
    },
    {
      "entropy": 6.074455118179321,
      "epoch": 1.6341398799010949,
      "grad_norm": 3.09375,
      "learning_rate": 0.00047352702756137035,
      "loss": 5.719,
      "mean_token_accuracy": 0.17727081924676896,
      "num_tokens": 34505126.0,
      "step": 18505
    },
    {
      "entropy": 6.0856328964233395,
      "epoch": 1.6345814199929354,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004735122054589291,
      "loss": 5.7827,
      "mean_token_accuracy": 0.1696959137916565,
      "num_tokens": 34514236.0,
      "step": 18510
    },
    {
      "entropy": 5.976208686828613,
      "epoch": 1.6350229600847757,
      "grad_norm": 2.0,
      "learning_rate": 0.0004734973794677467,
      "loss": 5.7158,
      "mean_token_accuracy": 0.1738231897354126,
      "num_tokens": 34522609.0,
      "step": 18515
    },
    {
      "entropy": 6.040709066390991,
      "epoch": 1.635464500176616,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047348254958811347,
      "loss": 5.8539,
      "mean_token_accuracy": 0.16867918372154236,
      "num_tokens": 34531742.0,
      "step": 18520
    },
    {
      "entropy": 6.08082537651062,
      "epoch": 1.6359060402684564,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047346771582031994,
      "loss": 5.7779,
      "mean_token_accuracy": 0.17049534916877745,
      "num_tokens": 34540821.0,
      "step": 18525
    },
    {
      "entropy": 6.153197240829468,
      "epoch": 1.6363475803602967,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004734528781646568,
      "loss": 5.8552,
      "mean_token_accuracy": 0.16659105271101,
      "num_tokens": 34549795.0,
      "step": 18530
    },
    {
      "entropy": 6.113088321685791,
      "epoch": 1.636789120452137,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004734380366214146,
      "loss": 5.8798,
      "mean_token_accuracy": 0.16161377876996993,
      "num_tokens": 34559455.0,
      "step": 18535
    },
    {
      "entropy": 6.100339078903199,
      "epoch": 1.6372306605439775,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004734231911908842,
      "loss": 5.8485,
      "mean_token_accuracy": 0.16398931443691253,
      "num_tokens": 34568832.0,
      "step": 18540
    },
    {
      "entropy": 6.087686491012573,
      "epoch": 1.6376722006358178,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047340834187335626,
      "loss": 5.8243,
      "mean_token_accuracy": 0.17221225798130035,
      "num_tokens": 34578630.0,
      "step": 18545
    },
    {
      "entropy": 6.152389860153198,
      "epoch": 1.638113740727658,
      "grad_norm": 1.953125,
      "learning_rate": 0.00047339348866912173,
      "loss": 5.8666,
      "mean_token_accuracy": 0.16704230159521102,
      "num_tokens": 34587782.0,
      "step": 18550
    },
    {
      "entropy": 6.061284828186035,
      "epoch": 1.6385552808194985,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004733786315784716,
      "loss": 5.7856,
      "mean_token_accuracy": 0.1663614481687546,
      "num_tokens": 34597133.0,
      "step": 18555
    },
    {
      "entropy": 6.053738021850586,
      "epoch": 1.6389968209113388,
      "grad_norm": 5.53125,
      "learning_rate": 0.00047336377060169696,
      "loss": 5.8623,
      "mean_token_accuracy": 0.16309555470943451,
      "num_tokens": 34606697.0,
      "step": 18560
    },
    {
      "entropy": 6.135743570327759,
      "epoch": 1.639438361003179,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004733489057390888,
      "loss": 5.8365,
      "mean_token_accuracy": 0.17261281311511995,
      "num_tokens": 34615806.0,
      "step": 18565
    },
    {
      "entropy": 6.145219373703003,
      "epoch": 1.6398799010950196,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004733340369909384,
      "loss": 5.9789,
      "mean_token_accuracy": 0.16026761382818222,
      "num_tokens": 34626563.0,
      "step": 18570
    },
    {
      "entropy": 6.116544723510742,
      "epoch": 1.6403214411868596,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047331916435753706,
      "loss": 5.8658,
      "mean_token_accuracy": 0.16707602441310881,
      "num_tokens": 34636077.0,
      "step": 18575
    },
    {
      "entropy": 6.117063426971436,
      "epoch": 1.6407629812787001,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047330428783917607,
      "loss": 5.9031,
      "mean_token_accuracy": 0.1625363126397133,
      "num_tokens": 34646164.0,
      "step": 18580
    },
    {
      "entropy": 6.118893766403199,
      "epoch": 1.6412045213705404,
      "grad_norm": 3.90625,
      "learning_rate": 0.00047328940743614683,
      "loss": 5.837,
      "mean_token_accuracy": 0.16394413411617278,
      "num_tokens": 34655577.0,
      "step": 18585
    },
    {
      "entropy": 6.1573773384094235,
      "epoch": 1.6416460614623807,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047327452314874094,
      "loss": 5.8558,
      "mean_token_accuracy": 0.16554490774869918,
      "num_tokens": 34664520.0,
      "step": 18590
    },
    {
      "entropy": 6.058693218231201,
      "epoch": 1.6420876015542212,
      "grad_norm": 2.71875,
      "learning_rate": 0.00047325963497724983,
      "loss": 5.7682,
      "mean_token_accuracy": 0.1714266210794449,
      "num_tokens": 34673528.0,
      "step": 18595
    },
    {
      "entropy": 6.137438344955444,
      "epoch": 1.6425291416460615,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004732447429219652,
      "loss": 5.8739,
      "mean_token_accuracy": 0.17458346635103225,
      "num_tokens": 34682563.0,
      "step": 18600
    },
    {
      "entropy": 6.101055479049682,
      "epoch": 1.6429706817379017,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004732298469831789,
      "loss": 5.7759,
      "mean_token_accuracy": 0.17350144535303116,
      "num_tokens": 34691932.0,
      "step": 18605
    },
    {
      "entropy": 6.111643218994141,
      "epoch": 1.6434122218297422,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047321494716118256,
      "loss": 5.7782,
      "mean_token_accuracy": 0.17158440500497818,
      "num_tokens": 34701376.0,
      "step": 18610
    },
    {
      "entropy": 6.074541473388672,
      "epoch": 1.6438537619215825,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047320004345626816,
      "loss": 5.7906,
      "mean_token_accuracy": 0.1667272225022316,
      "num_tokens": 34711112.0,
      "step": 18615
    },
    {
      "entropy": 6.10776948928833,
      "epoch": 1.6442953020134228,
      "grad_norm": 2.0,
      "learning_rate": 0.00047318513586872766,
      "loss": 5.8179,
      "mean_token_accuracy": 0.17048742473125458,
      "num_tokens": 34720429.0,
      "step": 18620
    },
    {
      "entropy": 6.08955078125,
      "epoch": 1.6447368421052633,
      "grad_norm": 1.9453125,
      "learning_rate": 0.000473170224398853,
      "loss": 5.8117,
      "mean_token_accuracy": 0.1703656241297722,
      "num_tokens": 34729422.0,
      "step": 18625
    },
    {
      "entropy": 6.092669582366943,
      "epoch": 1.6451783821971036,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047315530904693643,
      "loss": 5.8362,
      "mean_token_accuracy": 0.16888661980628966,
      "num_tokens": 34738231.0,
      "step": 18630
    },
    {
      "entropy": 6.143431711196899,
      "epoch": 1.6456199222889438,
      "grad_norm": 3.3125,
      "learning_rate": 0.00047314038981327,
      "loss": 5.8795,
      "mean_token_accuracy": 0.16543508619070052,
      "num_tokens": 34747002.0,
      "step": 18635
    },
    {
      "entropy": 6.017608594894409,
      "epoch": 1.6460614623807843,
      "grad_norm": 1.9609375,
      "learning_rate": 0.000473125466698146,
      "loss": 5.7413,
      "mean_token_accuracy": 0.17355683147907258,
      "num_tokens": 34756093.0,
      "step": 18640
    },
    {
      "entropy": 6.061131715774536,
      "epoch": 1.6465030024726244,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004731105397018569,
      "loss": 5.8544,
      "mean_token_accuracy": 0.16881997138261795,
      "num_tokens": 34766031.0,
      "step": 18645
    },
    {
      "entropy": 6.125955438613891,
      "epoch": 1.6469445425644649,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004730956088246949,
      "loss": 5.9129,
      "mean_token_accuracy": 0.1646955728530884,
      "num_tokens": 34776487.0,
      "step": 18650
    },
    {
      "entropy": 6.153037357330322,
      "epoch": 1.6473860826563052,
      "grad_norm": 1.828125,
      "learning_rate": 0.00047308067406695255,
      "loss": 5.7447,
      "mean_token_accuracy": 0.17313628643751144,
      "num_tokens": 34785372.0,
      "step": 18655
    },
    {
      "entropy": 6.094385862350464,
      "epoch": 1.6478276227481454,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047306573542892257,
      "loss": 5.7611,
      "mean_token_accuracy": 0.17166944295167924,
      "num_tokens": 34795453.0,
      "step": 18660
    },
    {
      "entropy": 6.092348241806031,
      "epoch": 1.648269162839986,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047305079291089735,
      "loss": 5.8724,
      "mean_token_accuracy": 0.16628682911396026,
      "num_tokens": 34804818.0,
      "step": 18665
    },
    {
      "entropy": 6.128774929046631,
      "epoch": 1.6487107029318262,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047303584651316985,
      "loss": 5.9021,
      "mean_token_accuracy": 0.16431626230478286,
      "num_tokens": 34814031.0,
      "step": 18670
    },
    {
      "entropy": 6.127244186401367,
      "epoch": 1.6491522430236665,
      "grad_norm": 2.125,
      "learning_rate": 0.0004730208962360327,
      "loss": 5.7733,
      "mean_token_accuracy": 0.1724126249551773,
      "num_tokens": 34823560.0,
      "step": 18675
    },
    {
      "entropy": 6.15035309791565,
      "epoch": 1.649593783115507,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004730059420797788,
      "loss": 5.8184,
      "mean_token_accuracy": 0.16556016206741334,
      "num_tokens": 34833662.0,
      "step": 18680
    },
    {
      "entropy": 6.149046993255615,
      "epoch": 1.6500353232073472,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004729909840447011,
      "loss": 5.9408,
      "mean_token_accuracy": 0.15505362078547477,
      "num_tokens": 34844264.0,
      "step": 18685
    },
    {
      "entropy": 6.100401830673218,
      "epoch": 1.6504768632991875,
      "grad_norm": 2.421875,
      "learning_rate": 0.00047297602213109265,
      "loss": 5.8111,
      "mean_token_accuracy": 0.17430943101644517,
      "num_tokens": 34853708.0,
      "step": 18690
    },
    {
      "entropy": 6.169701480865479,
      "epoch": 1.650918403391028,
      "grad_norm": 2.75,
      "learning_rate": 0.0004729610563392465,
      "loss": 5.8825,
      "mean_token_accuracy": 0.16530174463987352,
      "num_tokens": 34861926.0,
      "step": 18695
    },
    {
      "entropy": 6.0732227802276615,
      "epoch": 1.6513599434828683,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004729460866694558,
      "loss": 5.8009,
      "mean_token_accuracy": 0.1726557120680809,
      "num_tokens": 34870919.0,
      "step": 18700
    },
    {
      "entropy": 6.089860534667968,
      "epoch": 1.6518014835747086,
      "grad_norm": 2.15625,
      "learning_rate": 0.000472931113122014,
      "loss": 5.8389,
      "mean_token_accuracy": 0.17020328044891359,
      "num_tokens": 34879801.0,
      "step": 18705
    },
    {
      "entropy": 6.064186096191406,
      "epoch": 1.652243023666549,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004729161356972141,
      "loss": 5.8406,
      "mean_token_accuracy": 0.16323086619377136,
      "num_tokens": 34888605.0,
      "step": 18710
    },
    {
      "entropy": 6.141759157180786,
      "epoch": 1.6526845637583891,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047290115439534977,
      "loss": 5.8922,
      "mean_token_accuracy": 0.1569688782095909,
      "num_tokens": 34899294.0,
      "step": 18715
    },
    {
      "entropy": 6.203583097457885,
      "epoch": 1.6531261038502296,
      "grad_norm": 2.25,
      "learning_rate": 0.0004728861692167143,
      "loss": 5.8987,
      "mean_token_accuracy": 0.1664890393614769,
      "num_tokens": 34908624.0,
      "step": 18720
    },
    {
      "entropy": 6.097908449172974,
      "epoch": 1.65356764394207,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004728711801616013,
      "loss": 5.7484,
      "mean_token_accuracy": 0.1751505583524704,
      "num_tokens": 34917390.0,
      "step": 18725
    },
    {
      "entropy": 6.079542064666748,
      "epoch": 1.6540091840339102,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004728561872303045,
      "loss": 5.8656,
      "mean_token_accuracy": 0.16248057335615157,
      "num_tokens": 34926652.0,
      "step": 18730
    },
    {
      "entropy": 6.133965873718262,
      "epoch": 1.6544507241257507,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004728411904231175,
      "loss": 5.9213,
      "mean_token_accuracy": 0.15963236838579178,
      "num_tokens": 34936900.0,
      "step": 18735
    },
    {
      "entropy": 6.121544218063354,
      "epoch": 1.654892264217591,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004728261897403341,
      "loss": 5.8711,
      "mean_token_accuracy": 0.16618083119392396,
      "num_tokens": 34946827.0,
      "step": 18740
    },
    {
      "entropy": 6.141269493103027,
      "epoch": 1.6553338043094312,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047281118518224807,
      "loss": 5.8834,
      "mean_token_accuracy": 0.16597179025411607,
      "num_tokens": 34955987.0,
      "step": 18745
    },
    {
      "entropy": 6.108980464935303,
      "epoch": 1.6557753444012717,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004727961767491535,
      "loss": 5.8063,
      "mean_token_accuracy": 0.16927266716957093,
      "num_tokens": 34965001.0,
      "step": 18750
    },
    {
      "entropy": 6.116049528121948,
      "epoch": 1.656216884493112,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004727811644413443,
      "loss": 5.8778,
      "mean_token_accuracy": 0.16350563764572143,
      "num_tokens": 34974113.0,
      "step": 18755
    },
    {
      "entropy": 6.023114013671875,
      "epoch": 1.6566584245849523,
      "grad_norm": 2.0,
      "learning_rate": 0.0004727661482591146,
      "loss": 5.7255,
      "mean_token_accuracy": 0.17749422192573547,
      "num_tokens": 34982535.0,
      "step": 18760
    },
    {
      "entropy": 6.055050277709961,
      "epoch": 1.6570999646767928,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004727511282027585,
      "loss": 5.8429,
      "mean_token_accuracy": 0.16527510285377503,
      "num_tokens": 34991828.0,
      "step": 18765
    },
    {
      "entropy": 6.101134967803955,
      "epoch": 1.657541504768633,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004727361042725703,
      "loss": 5.7973,
      "mean_token_accuracy": 0.16607616990804672,
      "num_tokens": 35001670.0,
      "step": 18770
    },
    {
      "entropy": 6.069123840332031,
      "epoch": 1.6579830448604733,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004727210764688442,
      "loss": 5.7546,
      "mean_token_accuracy": 0.17217485457658768,
      "num_tokens": 35011407.0,
      "step": 18775
    },
    {
      "entropy": 5.98431510925293,
      "epoch": 1.6584245849523138,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004727060447918748,
      "loss": 5.7149,
      "mean_token_accuracy": 0.17702813744544982,
      "num_tokens": 35020753.0,
      "step": 18780
    },
    {
      "entropy": 6.074127674102783,
      "epoch": 1.6588661250441539,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047269100924195623,
      "loss": 5.8734,
      "mean_token_accuracy": 0.16552927792072297,
      "num_tokens": 35030599.0,
      "step": 18785
    },
    {
      "entropy": 6.122203063964844,
      "epoch": 1.6593076651359944,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047267596981938337,
      "loss": 5.8672,
      "mean_token_accuracy": 0.16295130997896196,
      "num_tokens": 35039098.0,
      "step": 18790
    },
    {
      "entropy": 6.117880344390869,
      "epoch": 1.6597492052278346,
      "grad_norm": 3.265625,
      "learning_rate": 0.00047266092652445066,
      "loss": 5.8376,
      "mean_token_accuracy": 0.16440571397542952,
      "num_tokens": 35048286.0,
      "step": 18795
    },
    {
      "entropy": 6.197451448440551,
      "epoch": 1.660190745319675,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004726458793574528,
      "loss": 5.8906,
      "mean_token_accuracy": 0.16686654090881348,
      "num_tokens": 35057331.0,
      "step": 18800
    },
    {
      "entropy": 6.128736972808838,
      "epoch": 1.6606322854115154,
      "grad_norm": 2.375,
      "learning_rate": 0.0004726308283186846,
      "loss": 5.8403,
      "mean_token_accuracy": 0.16483235359191895,
      "num_tokens": 35066321.0,
      "step": 18805
    },
    {
      "entropy": 6.025134563446045,
      "epoch": 1.6610738255033557,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004726157734084409,
      "loss": 5.7697,
      "mean_token_accuracy": 0.17450788021087646,
      "num_tokens": 35075218.0,
      "step": 18810
    },
    {
      "entropy": 6.070403718948365,
      "epoch": 1.661515365595196,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004726007146270164,
      "loss": 5.8385,
      "mean_token_accuracy": 0.16996488273143767,
      "num_tokens": 35084563.0,
      "step": 18815
    },
    {
      "entropy": 6.005402135848999,
      "epoch": 1.6619569056870365,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004725856519747065,
      "loss": 5.7274,
      "mean_token_accuracy": 0.17311567962169647,
      "num_tokens": 35093346.0,
      "step": 18820
    },
    {
      "entropy": 6.049417400360108,
      "epoch": 1.6623984457788767,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004725705854518059,
      "loss": 5.8223,
      "mean_token_accuracy": 0.16694478839635848,
      "num_tokens": 35102501.0,
      "step": 18825
    },
    {
      "entropy": 6.089984035491943,
      "epoch": 1.662839985870717,
      "grad_norm": 2.625,
      "learning_rate": 0.00047255551505861007,
      "loss": 5.8174,
      "mean_token_accuracy": 0.16846184358000754,
      "num_tokens": 35112083.0,
      "step": 18830
    },
    {
      "entropy": 6.067811775207519,
      "epoch": 1.6632815259625575,
      "grad_norm": 4.09375,
      "learning_rate": 0.00047254044079541387,
      "loss": 5.7541,
      "mean_token_accuracy": 0.17685417085886002,
      "num_tokens": 35121380.0,
      "step": 18835
    },
    {
      "entropy": 6.112482929229737,
      "epoch": 1.6637230660543978,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004725253626625129,
      "loss": 5.8892,
      "mean_token_accuracy": 0.15796207785606384,
      "num_tokens": 35130083.0,
      "step": 18840
    },
    {
      "entropy": 6.0659881114959715,
      "epoch": 1.664164606146238,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047251028066020237,
      "loss": 5.7575,
      "mean_token_accuracy": 0.17269061058759688,
      "num_tokens": 35139001.0,
      "step": 18845
    },
    {
      "entropy": 6.015168380737305,
      "epoch": 1.6646061462380786,
      "grad_norm": 1.984375,
      "learning_rate": 0.00047249519478877774,
      "loss": 5.7212,
      "mean_token_accuracy": 0.17552589476108552,
      "num_tokens": 35147701.0,
      "step": 18850
    },
    {
      "entropy": 6.0029267311096195,
      "epoch": 1.6650476863299186,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004724801050485346,
      "loss": 5.7499,
      "mean_token_accuracy": 0.17299622893333436,
      "num_tokens": 35157334.0,
      "step": 18855
    },
    {
      "entropy": 6.1090747833251955,
      "epoch": 1.6654892264217591,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004724650114397685,
      "loss": 5.7825,
      "mean_token_accuracy": 0.17230241894721984,
      "num_tokens": 35166647.0,
      "step": 18860
    },
    {
      "entropy": 6.119015455245972,
      "epoch": 1.6659307665135994,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047244991396277515,
      "loss": 5.7727,
      "mean_token_accuracy": 0.16403061598539354,
      "num_tokens": 35176345.0,
      "step": 18865
    },
    {
      "entropy": 6.108417749404907,
      "epoch": 1.6663723066054397,
      "grad_norm": 3.0625,
      "learning_rate": 0.00047243481261785027,
      "loss": 5.7851,
      "mean_token_accuracy": 0.17539390176534653,
      "num_tokens": 35185793.0,
      "step": 18870
    },
    {
      "entropy": 6.081078052520752,
      "epoch": 1.6668138466972802,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047241970740528975,
      "loss": 5.8797,
      "mean_token_accuracy": 0.1692331090569496,
      "num_tokens": 35195905.0,
      "step": 18875
    },
    {
      "entropy": 6.059637403488159,
      "epoch": 1.6672553867891204,
      "grad_norm": 2.953125,
      "learning_rate": 0.00047240459832538934,
      "loss": 5.7101,
      "mean_token_accuracy": 0.1732350692152977,
      "num_tokens": 35204340.0,
      "step": 18880
    },
    {
      "entropy": 6.111812400817871,
      "epoch": 1.6676969268809607,
      "grad_norm": 4.125,
      "learning_rate": 0.0004723894853784452,
      "loss": 5.8167,
      "mean_token_accuracy": 0.16568848490715027,
      "num_tokens": 35213270.0,
      "step": 18885
    },
    {
      "entropy": 6.04381628036499,
      "epoch": 1.6681384669728012,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004723743685647533,
      "loss": 5.8018,
      "mean_token_accuracy": 0.1666388675570488,
      "num_tokens": 35222652.0,
      "step": 18890
    },
    {
      "entropy": 6.08791184425354,
      "epoch": 1.6685800070646415,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004723592478846097,
      "loss": 5.8116,
      "mean_token_accuracy": 0.16797169297933578,
      "num_tokens": 35231684.0,
      "step": 18895
    },
    {
      "entropy": 6.107256650924683,
      "epoch": 1.6690215471564818,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047234412333831073,
      "loss": 5.8745,
      "mean_token_accuracy": 0.1700949102640152,
      "num_tokens": 35241589.0,
      "step": 18900
    },
    {
      "entropy": 6.1547369956970215,
      "epoch": 1.6694630872483223,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004723289949261527,
      "loss": 5.9551,
      "mean_token_accuracy": 0.15807587951421737,
      "num_tokens": 35250903.0,
      "step": 18905
    },
    {
      "entropy": 6.120497035980224,
      "epoch": 1.6699046273401625,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004723138626484318,
      "loss": 5.8197,
      "mean_token_accuracy": 0.16882286965847015,
      "num_tokens": 35260556.0,
      "step": 18910
    },
    {
      "entropy": 6.047041511535644,
      "epoch": 1.6703461674320028,
      "grad_norm": 2.765625,
      "learning_rate": 0.00047229872650544446,
      "loss": 5.7228,
      "mean_token_accuracy": 0.17861812859773635,
      "num_tokens": 35270237.0,
      "step": 18915
    },
    {
      "entropy": 6.046302747726441,
      "epoch": 1.6707877075238433,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047228358649748745,
      "loss": 5.7555,
      "mean_token_accuracy": 0.1766030415892601,
      "num_tokens": 35279017.0,
      "step": 18920
    },
    {
      "entropy": 5.9885560989379885,
      "epoch": 1.6712292476156834,
      "grad_norm": 5.4375,
      "learning_rate": 0.0004722684426248571,
      "loss": 5.778,
      "mean_token_accuracy": 0.17301517724990845,
      "num_tokens": 35288150.0,
      "step": 18925
    },
    {
      "entropy": 6.090394783020019,
      "epoch": 1.6716707877075239,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047225329488785015,
      "loss": 5.8626,
      "mean_token_accuracy": 0.17515024095773696,
      "num_tokens": 35297463.0,
      "step": 18930
    },
    {
      "entropy": 6.152465391159057,
      "epoch": 1.6721123277993641,
      "grad_norm": 2.78125,
      "learning_rate": 0.00047223814328676336,
      "loss": 5.8115,
      "mean_token_accuracy": 0.17241683900356292,
      "num_tokens": 35307409.0,
      "step": 18935
    },
    {
      "entropy": 6.007883214950562,
      "epoch": 1.6725538678912044,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004722229878218935,
      "loss": 5.7703,
      "mean_token_accuracy": 0.17405946999788285,
      "num_tokens": 35315613.0,
      "step": 18940
    },
    {
      "entropy": 6.012724542617798,
      "epoch": 1.672995407983045,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047220782849353743,
      "loss": 5.8307,
      "mean_token_accuracy": 0.16756608635187148,
      "num_tokens": 35325297.0,
      "step": 18945
    },
    {
      "entropy": 6.10370397567749,
      "epoch": 1.6734369480748852,
      "grad_norm": 2.375,
      "learning_rate": 0.0004721926653019922,
      "loss": 5.801,
      "mean_token_accuracy": 0.16623207330703735,
      "num_tokens": 35334014.0,
      "step": 18950
    },
    {
      "entropy": 6.1236778736114506,
      "epoch": 1.6738784881667255,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004721774982475548,
      "loss": 5.7992,
      "mean_token_accuracy": 0.1742392048239708,
      "num_tokens": 35342784.0,
      "step": 18955
    },
    {
      "entropy": 6.149184322357177,
      "epoch": 1.674320028258566,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00047216232733052237,
      "loss": 5.9449,
      "mean_token_accuracy": 0.16564358472824098,
      "num_tokens": 35351985.0,
      "step": 18960
    },
    {
      "entropy": 6.124332141876221,
      "epoch": 1.6747615683504062,
      "grad_norm": 2.125,
      "learning_rate": 0.0004721471525511921,
      "loss": 5.7876,
      "mean_token_accuracy": 0.16358005702495576,
      "num_tokens": 35360857.0,
      "step": 18965
    },
    {
      "entropy": 6.115841865539551,
      "epoch": 1.6752031084422465,
      "grad_norm": 15.8125,
      "learning_rate": 0.0004721319739098611,
      "loss": 5.7664,
      "mean_token_accuracy": 0.17723292112350464,
      "num_tokens": 35370200.0,
      "step": 18970
    },
    {
      "entropy": 6.112900447845459,
      "epoch": 1.675644648534087,
      "grad_norm": 6.03125,
      "learning_rate": 0.000472116791406827,
      "loss": 5.8357,
      "mean_token_accuracy": 0.16599339693784715,
      "num_tokens": 35379481.0,
      "step": 18975
    },
    {
      "entropy": 5.999763631820679,
      "epoch": 1.6760861886259273,
      "grad_norm": 2.84375,
      "learning_rate": 0.000472101605042387,
      "loss": 5.7947,
      "mean_token_accuracy": 0.1762950137257576,
      "num_tokens": 35387951.0,
      "step": 18980
    },
    {
      "entropy": 6.127461194992065,
      "epoch": 1.6765277287177676,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004720864148168387,
      "loss": 5.8167,
      "mean_token_accuracy": 0.16511853337287902,
      "num_tokens": 35396200.0,
      "step": 18985
    },
    {
      "entropy": 6.143962001800537,
      "epoch": 1.676969268809608,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004720712207304796,
      "loss": 5.7966,
      "mean_token_accuracy": 0.1654999554157257,
      "num_tokens": 35404474.0,
      "step": 18990
    },
    {
      "entropy": 6.189566612243652,
      "epoch": 1.6774108089014481,
      "grad_norm": 3.875,
      "learning_rate": 0.0004720560227836073,
      "loss": 5.954,
      "mean_token_accuracy": 0.16297169029712677,
      "num_tokens": 35414419.0,
      "step": 18995
    },
    {
      "entropy": 6.067425537109375,
      "epoch": 1.6778523489932886,
      "grad_norm": 12.375,
      "learning_rate": 0.0004720408209765197,
      "loss": 5.7734,
      "mean_token_accuracy": 0.16908938437700272,
      "num_tokens": 35423727.0,
      "step": 19000
    },
    {
      "entropy": 6.021581697463989,
      "epoch": 1.6782938890851289,
      "grad_norm": 3.1875,
      "learning_rate": 0.00047202561530951443,
      "loss": 5.7253,
      "mean_token_accuracy": 0.18238083124160767,
      "num_tokens": 35433364.0,
      "step": 19005
    },
    {
      "entropy": 6.195837211608887,
      "epoch": 1.6787354291769692,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004720104057828894,
      "loss": 5.8693,
      "mean_token_accuracy": 0.1637371599674225,
      "num_tokens": 35442665.0,
      "step": 19010
    },
    {
      "entropy": 6.109767770767212,
      "epoch": 1.6791769692688097,
      "grad_norm": 2.578125,
      "learning_rate": 0.00047199519239694257,
      "loss": 5.7505,
      "mean_token_accuracy": 0.1797059029340744,
      "num_tokens": 35451443.0,
      "step": 19015
    },
    {
      "entropy": 6.038609743118286,
      "epoch": 1.67961850936065,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047197997515197195,
      "loss": 5.7071,
      "mean_token_accuracy": 0.18391988426446915,
      "num_tokens": 35460622.0,
      "step": 19020
    },
    {
      "entropy": 6.003350257873535,
      "epoch": 1.6800600494524902,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047196475404827565,
      "loss": 5.8709,
      "mean_token_accuracy": 0.1679638758301735,
      "num_tokens": 35470345.0,
      "step": 19025
    },
    {
      "entropy": 6.129145622253418,
      "epoch": 1.6805015895443307,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047194952908615183,
      "loss": 5.9504,
      "mean_token_accuracy": 0.1622982919216156,
      "num_tokens": 35480460.0,
      "step": 19030
    },
    {
      "entropy": 6.155210161209107,
      "epoch": 1.680943129636171,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004719343002658987,
      "loss": 5.8734,
      "mean_token_accuracy": 0.16270867884159088,
      "num_tokens": 35489983.0,
      "step": 19035
    },
    {
      "entropy": 6.119449090957642,
      "epoch": 1.6813846697280113,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047191906758781475,
      "loss": 5.7945,
      "mean_token_accuracy": 0.17016828209161758,
      "num_tokens": 35498963.0,
      "step": 19040
    },
    {
      "entropy": 6.109153699874878,
      "epoch": 1.6818262098198518,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004719038310521982,
      "loss": 5.8438,
      "mean_token_accuracy": 0.17538014352321624,
      "num_tokens": 35509054.0,
      "step": 19045
    },
    {
      "entropy": 6.120983695983886,
      "epoch": 1.682267749911692,
      "grad_norm": 2.625,
      "learning_rate": 0.0004718885906593476,
      "loss": 5.9794,
      "mean_token_accuracy": 0.15372612327337265,
      "num_tokens": 35518093.0,
      "step": 19050
    },
    {
      "entropy": 6.187893199920654,
      "epoch": 1.6827092900035323,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004718733464095614,
      "loss": 5.8494,
      "mean_token_accuracy": 0.17016295492649078,
      "num_tokens": 35527330.0,
      "step": 19055
    },
    {
      "entropy": 6.145586395263672,
      "epoch": 1.6831508300953728,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047185809830313844,
      "loss": 5.8409,
      "mean_token_accuracy": 0.161556638777256,
      "num_tokens": 35538147.0,
      "step": 19060
    },
    {
      "entropy": 6.144809865951538,
      "epoch": 1.6835923701872129,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047184284634037727,
      "loss": 5.9107,
      "mean_token_accuracy": 0.1590116649866104,
      "num_tokens": 35547930.0,
      "step": 19065
    },
    {
      "entropy": 6.0906219482421875,
      "epoch": 1.6840339102790534,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047182759052157665,
      "loss": 5.762,
      "mean_token_accuracy": 0.17063236087560654,
      "num_tokens": 35557111.0,
      "step": 19070
    },
    {
      "entropy": 6.105663585662842,
      "epoch": 1.6844754503708936,
      "grad_norm": 2.75,
      "learning_rate": 0.0004718123308470355,
      "loss": 5.8867,
      "mean_token_accuracy": 0.16183037161827088,
      "num_tokens": 35565739.0,
      "step": 19075
    },
    {
      "entropy": 6.106256341934204,
      "epoch": 1.684916990462734,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047179706731705263,
      "loss": 5.7468,
      "mean_token_accuracy": 0.1707444965839386,
      "num_tokens": 35574164.0,
      "step": 19080
    },
    {
      "entropy": 6.090675401687622,
      "epoch": 1.6853585305545744,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047178179993192723,
      "loss": 5.8159,
      "mean_token_accuracy": 0.16410738229751587,
      "num_tokens": 35582813.0,
      "step": 19085
    },
    {
      "entropy": 6.128316402435303,
      "epoch": 1.6858000706464147,
      "grad_norm": 3.015625,
      "learning_rate": 0.00047176652869195825,
      "loss": 5.7901,
      "mean_token_accuracy": 0.16931006908416749,
      "num_tokens": 35591884.0,
      "step": 19090
    },
    {
      "entropy": 6.134412097930908,
      "epoch": 1.686241610738255,
      "grad_norm": 4.6875,
      "learning_rate": 0.00047175125359744487,
      "loss": 5.8225,
      "mean_token_accuracy": 0.16840188205242157,
      "num_tokens": 35601824.0,
      "step": 19095
    },
    {
      "entropy": 6.053271341323852,
      "epoch": 1.6866831508300955,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047173597464868634,
      "loss": 5.7654,
      "mean_token_accuracy": 0.17484045028686523,
      "num_tokens": 35610756.0,
      "step": 19100
    },
    {
      "entropy": 5.9778365135192875,
      "epoch": 1.6871246909219357,
      "grad_norm": 2.609375,
      "learning_rate": 0.000471720691845982,
      "loss": 5.782,
      "mean_token_accuracy": 0.17011386007070542,
      "num_tokens": 35620160.0,
      "step": 19105
    },
    {
      "entropy": 6.1530474662780765,
      "epoch": 1.687566231013776,
      "grad_norm": 2.25,
      "learning_rate": 0.00047170540518963114,
      "loss": 5.918,
      "mean_token_accuracy": 0.15778680592775346,
      "num_tokens": 35630487.0,
      "step": 19110
    },
    {
      "entropy": 6.210044002532959,
      "epoch": 1.6880077711056165,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004716901146799333,
      "loss": 5.979,
      "mean_token_accuracy": 0.15712981522083283,
      "num_tokens": 35641924.0,
      "step": 19115
    },
    {
      "entropy": 6.1275519847869875,
      "epoch": 1.6884493111974568,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004716748203171879,
      "loss": 5.788,
      "mean_token_accuracy": 0.16949477642774582,
      "num_tokens": 35651575.0,
      "step": 19120
    },
    {
      "entropy": 6.124764394760132,
      "epoch": 1.688890851289297,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047165952210169475,
      "loss": 5.764,
      "mean_token_accuracy": 0.1735522210597992,
      "num_tokens": 35661481.0,
      "step": 19125
    },
    {
      "entropy": 6.0302050590515135,
      "epoch": 1.6893323913811376,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004716442200337533,
      "loss": 5.7774,
      "mean_token_accuracy": 0.16735107004642485,
      "num_tokens": 35672049.0,
      "step": 19130
    },
    {
      "entropy": 6.071022844314575,
      "epoch": 1.6897739314729776,
      "grad_norm": 2.34375,
      "learning_rate": 0.00047162891411366343,
      "loss": 5.8228,
      "mean_token_accuracy": 0.16902995407581328,
      "num_tokens": 35681653.0,
      "step": 19135
    },
    {
      "entropy": 6.099254131317139,
      "epoch": 1.690215471564818,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047161360434172506,
      "loss": 5.8631,
      "mean_token_accuracy": 0.17202121317386626,
      "num_tokens": 35690868.0,
      "step": 19140
    },
    {
      "entropy": 6.095858383178711,
      "epoch": 1.6906570116566584,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004715982907182379,
      "loss": 5.7882,
      "mean_token_accuracy": 0.17239004820585252,
      "num_tokens": 35699062.0,
      "step": 19145
    },
    {
      "entropy": 6.06781759262085,
      "epoch": 1.6910985517484987,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004715829732435021,
      "loss": 5.8307,
      "mean_token_accuracy": 0.1685429885983467,
      "num_tokens": 35708411.0,
      "step": 19150
    },
    {
      "entropy": 6.006149005889893,
      "epoch": 1.6915400918403392,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047156765191781763,
      "loss": 5.764,
      "mean_token_accuracy": 0.16876054555177689,
      "num_tokens": 35717937.0,
      "step": 19155
    },
    {
      "entropy": 6.093161106109619,
      "epoch": 1.6919816319321794,
      "grad_norm": 2.90625,
      "learning_rate": 0.00047155232674148465,
      "loss": 5.7721,
      "mean_token_accuracy": 0.17122883945703507,
      "num_tokens": 35727712.0,
      "step": 19160
    },
    {
      "entropy": 6.131403207778931,
      "epoch": 1.6924231720240197,
      "grad_norm": 1.875,
      "learning_rate": 0.00047153699771480335,
      "loss": 5.8768,
      "mean_token_accuracy": 0.1637745603919029,
      "num_tokens": 35737076.0,
      "step": 19165
    },
    {
      "entropy": 6.053154373168946,
      "epoch": 1.6928647121158602,
      "grad_norm": 2.3125,
      "learning_rate": 0.000471521664838074,
      "loss": 5.6837,
      "mean_token_accuracy": 0.17750798612833024,
      "num_tokens": 35746436.0,
      "step": 19170
    },
    {
      "entropy": 6.049314546585083,
      "epoch": 1.6933062522077005,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004715063281115971,
      "loss": 5.8665,
      "mean_token_accuracy": 0.17165696769952773,
      "num_tokens": 35755786.0,
      "step": 19175
    },
    {
      "entropy": 6.074800062179565,
      "epoch": 1.6937477922995408,
      "grad_norm": 2.859375,
      "learning_rate": 0.000471490987535673,
      "loss": 5.8287,
      "mean_token_accuracy": 0.1699073776602745,
      "num_tokens": 35764424.0,
      "step": 19180
    },
    {
      "entropy": 6.154159927368164,
      "epoch": 1.6941893323913813,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047147564311060216,
      "loss": 5.7957,
      "mean_token_accuracy": 0.17205065190792085,
      "num_tokens": 35773792.0,
      "step": 19185
    },
    {
      "entropy": 6.091228437423706,
      "epoch": 1.6946308724832215,
      "grad_norm": 2.375,
      "learning_rate": 0.00047146029483668517,
      "loss": 5.7491,
      "mean_token_accuracy": 0.16847403943538666,
      "num_tokens": 35783285.0,
      "step": 19190
    },
    {
      "entropy": 6.096319675445557,
      "epoch": 1.6950724125750618,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004714449427142228,
      "loss": 5.8944,
      "mean_token_accuracy": 0.1662941873073578,
      "num_tokens": 35792476.0,
      "step": 19195
    },
    {
      "entropy": 6.101670837402343,
      "epoch": 1.6955139526669023,
      "grad_norm": 2.8125,
      "learning_rate": 0.00047142958674351563,
      "loss": 5.8045,
      "mean_token_accuracy": 0.17654617726802826,
      "num_tokens": 35801640.0,
      "step": 19200
    },
    {
      "entropy": 6.10167121887207,
      "epoch": 1.6959554927587424,
      "grad_norm": 2.90625,
      "learning_rate": 0.00047141422692486466,
      "loss": 5.8372,
      "mean_token_accuracy": 0.1688803568482399,
      "num_tokens": 35811073.0,
      "step": 19205
    },
    {
      "entropy": 6.122856092453003,
      "epoch": 1.6963970328505829,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047139886325857074,
      "loss": 5.7825,
      "mean_token_accuracy": 0.17257902920246124,
      "num_tokens": 35819814.0,
      "step": 19210
    },
    {
      "entropy": 6.088796901702881,
      "epoch": 1.6968385729424231,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004713834957449347,
      "loss": 5.7668,
      "mean_token_accuracy": 0.17421850115060805,
      "num_tokens": 35828363.0,
      "step": 19215
    },
    {
      "entropy": 6.097899675369263,
      "epoch": 1.6972801130342634,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004713681243842577,
      "loss": 5.7234,
      "mean_token_accuracy": 0.1752650648355484,
      "num_tokens": 35836868.0,
      "step": 19220
    },
    {
      "entropy": 6.166964101791382,
      "epoch": 1.697721653126104,
      "grad_norm": 3.234375,
      "learning_rate": 0.00047135274917684087,
      "loss": 5.8176,
      "mean_token_accuracy": 0.16683898270130157,
      "num_tokens": 35846030.0,
      "step": 19225
    },
    {
      "entropy": 6.041832113265992,
      "epoch": 1.6981631932179442,
      "grad_norm": 2.25,
      "learning_rate": 0.0004713373701229852,
      "loss": 5.684,
      "mean_token_accuracy": 0.18073196709156036,
      "num_tokens": 35854862.0,
      "step": 19230
    },
    {
      "entropy": 5.998066997528076,
      "epoch": 1.6986047333097845,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004713219872229923,
      "loss": 5.7537,
      "mean_token_accuracy": 0.17589168548583983,
      "num_tokens": 35864238.0,
      "step": 19235
    },
    {
      "entropy": 6.057971811294555,
      "epoch": 1.699046273401625,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004713066004771633,
      "loss": 5.7613,
      "mean_token_accuracy": 0.17018077075481414,
      "num_tokens": 35872542.0,
      "step": 19240
    },
    {
      "entropy": 6.151629638671875,
      "epoch": 1.6994878134934652,
      "grad_norm": 3.21875,
      "learning_rate": 0.00047129120988579953,
      "loss": 5.8506,
      "mean_token_accuracy": 0.16743845790624617,
      "num_tokens": 35881899.0,
      "step": 19245
    },
    {
      "entropy": 6.142848682403565,
      "epoch": 1.6999293535853055,
      "grad_norm": 3.578125,
      "learning_rate": 0.00047127581544920263,
      "loss": 5.8094,
      "mean_token_accuracy": 0.169037164747715,
      "num_tokens": 35891970.0,
      "step": 19250
    },
    {
      "entropy": 6.001580047607422,
      "epoch": 1.700370893677146,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004712604171676742,
      "loss": 5.7235,
      "mean_token_accuracy": 0.1694053515791893,
      "num_tokens": 35901286.0,
      "step": 19255
    },
    {
      "entropy": 6.025413274765015,
      "epoch": 1.7008124337689863,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047124501504151574,
      "loss": 5.9479,
      "mean_token_accuracy": 0.1582539439201355,
      "num_tokens": 35912301.0,
      "step": 19260
    },
    {
      "entropy": 6.104783821105957,
      "epoch": 1.7012539738608266,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047122960907102906,
      "loss": 5.8291,
      "mean_token_accuracy": 0.16690873503684997,
      "num_tokens": 35920767.0,
      "step": 19265
    },
    {
      "entropy": 6.13028130531311,
      "epoch": 1.701695513952667,
      "grad_norm": 2.734375,
      "learning_rate": 0.000471214199256516,
      "loss": 5.8887,
      "mean_token_accuracy": 0.16692220717668532,
      "num_tokens": 35930235.0,
      "step": 19270
    },
    {
      "entropy": 6.1388813018798825,
      "epoch": 1.702137054044507,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047119878559827825,
      "loss": 5.8402,
      "mean_token_accuracy": 0.16159826070070266,
      "num_tokens": 35939213.0,
      "step": 19275
    },
    {
      "entropy": 6.164036893844605,
      "epoch": 1.7025785941363476,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047118336809661793,
      "loss": 5.9037,
      "mean_token_accuracy": 0.15856220573186874,
      "num_tokens": 35948221.0,
      "step": 19280
    },
    {
      "entropy": 6.1849359512329105,
      "epoch": 1.7030201342281879,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004711679467518371,
      "loss": 5.8305,
      "mean_token_accuracy": 0.16352578103542328,
      "num_tokens": 35957961.0,
      "step": 19285
    },
    {
      "entropy": 6.139436388015747,
      "epoch": 1.7034616743200282,
      "grad_norm": 3.609375,
      "learning_rate": 0.00047115252156423767,
      "loss": 5.8517,
      "mean_token_accuracy": 0.16613772958517076,
      "num_tokens": 35966742.0,
      "step": 19290
    },
    {
      "entropy": 6.103049325942993,
      "epoch": 1.7039032144118686,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004711370925341219,
      "loss": 5.8041,
      "mean_token_accuracy": 0.1633719399571419,
      "num_tokens": 35976312.0,
      "step": 19295
    },
    {
      "entropy": 6.037829494476318,
      "epoch": 1.704344754503709,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047112165966179206,
      "loss": 5.9062,
      "mean_token_accuracy": 0.17124451398849488,
      "num_tokens": 35985537.0,
      "step": 19300
    },
    {
      "entropy": 6.1216216564178465,
      "epoch": 1.7047862945955492,
      "grad_norm": 4.09375,
      "learning_rate": 0.00047110622294755045,
      "loss": 5.8026,
      "mean_token_accuracy": 0.1674380585551262,
      "num_tokens": 35995004.0,
      "step": 19305
    },
    {
      "entropy": 6.156541633605957,
      "epoch": 1.7052278346873897,
      "grad_norm": 4.25,
      "learning_rate": 0.00047109078239169946,
      "loss": 5.8241,
      "mean_token_accuracy": 0.16694437861442565,
      "num_tokens": 36004786.0,
      "step": 19310
    },
    {
      "entropy": 6.051384115219117,
      "epoch": 1.70566937477923,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004710753379945415,
      "loss": 5.7827,
      "mean_token_accuracy": 0.17595934718847275,
      "num_tokens": 36014047.0,
      "step": 19315
    },
    {
      "entropy": 6.124526262283325,
      "epoch": 1.7061109148710702,
      "grad_norm": 3.3125,
      "learning_rate": 0.00047105988975637927,
      "loss": 5.79,
      "mean_token_accuracy": 0.17057613730430604,
      "num_tokens": 36022921.0,
      "step": 19320
    },
    {
      "entropy": 6.085334444046021,
      "epoch": 1.7065524549629107,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004710444376775153,
      "loss": 5.7959,
      "mean_token_accuracy": 0.16868038326501847,
      "num_tokens": 36032713.0,
      "step": 19325
    },
    {
      "entropy": 6.113410472869873,
      "epoch": 1.706993995054751,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047102898175825224,
      "loss": 5.7789,
      "mean_token_accuracy": 0.16796768009662627,
      "num_tokens": 36041595.0,
      "step": 19330
    },
    {
      "entropy": 6.079406833648681,
      "epoch": 1.7074355351465913,
      "grad_norm": 1.734375,
      "learning_rate": 0.00047101352199889295,
      "loss": 5.8081,
      "mean_token_accuracy": 0.16420574933290483,
      "num_tokens": 36051993.0,
      "step": 19335
    },
    {
      "entropy": 6.073725891113281,
      "epoch": 1.7078770752384318,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004709980583997402,
      "loss": 5.803,
      "mean_token_accuracy": 0.17007434070110322,
      "num_tokens": 36061806.0,
      "step": 19340
    },
    {
      "entropy": 6.0892116069793705,
      "epoch": 1.7083186153302719,
      "grad_norm": 2.15625,
      "learning_rate": 0.000470982590961097,
      "loss": 5.8601,
      "mean_token_accuracy": 0.16894044280052184,
      "num_tokens": 36071390.0,
      "step": 19345
    },
    {
      "entropy": 6.087711381912231,
      "epoch": 1.7087601554221123,
      "grad_norm": 1.96875,
      "learning_rate": 0.00047096711968326626,
      "loss": 5.8173,
      "mean_token_accuracy": 0.1661966323852539,
      "num_tokens": 36080561.0,
      "step": 19350
    },
    {
      "entropy": 6.127880048751831,
      "epoch": 1.7092016955139526,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004709516445665512,
      "loss": 5.7452,
      "mean_token_accuracy": 0.16936004012823105,
      "num_tokens": 36089568.0,
      "step": 19355
    },
    {
      "entropy": 6.082294273376465,
      "epoch": 1.709643235605793,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004709361656112547,
      "loss": 5.7731,
      "mean_token_accuracy": 0.17450327426195145,
      "num_tokens": 36098881.0,
      "step": 19360
    },
    {
      "entropy": 6.070976209640503,
      "epoch": 1.7100847756976334,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004709206828176804,
      "loss": 5.7782,
      "mean_token_accuracy": 0.17481836229562758,
      "num_tokens": 36108209.0,
      "step": 19365
    },
    {
      "entropy": 6.027729654312134,
      "epoch": 1.7105263157894737,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004709051961861311,
      "loss": 5.8251,
      "mean_token_accuracy": 0.17355323135852813,
      "num_tokens": 36117573.0,
      "step": 19370
    },
    {
      "entropy": 6.105583000183105,
      "epoch": 1.710967855881314,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004708897057169105,
      "loss": 5.8494,
      "mean_token_accuracy": 0.17175644487142563,
      "num_tokens": 36127405.0,
      "step": 19375
    },
    {
      "entropy": 6.175522422790527,
      "epoch": 1.7114093959731544,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004708742114103221,
      "loss": 5.8438,
      "mean_token_accuracy": 0.1696135386824608,
      "num_tokens": 36136148.0,
      "step": 19380
    },
    {
      "entropy": 6.087206745147705,
      "epoch": 1.7118509360649947,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004708587132666692,
      "loss": 5.815,
      "mean_token_accuracy": 0.17109988182783126,
      "num_tokens": 36146164.0,
      "step": 19385
    },
    {
      "entropy": 6.112454223632812,
      "epoch": 1.712292476156835,
      "grad_norm": 3.4375,
      "learning_rate": 0.00047084321128625545,
      "loss": 5.8135,
      "mean_token_accuracy": 0.1616964191198349,
      "num_tokens": 36155313.0,
      "step": 19390
    },
    {
      "entropy": 6.114309930801392,
      "epoch": 1.7127340162486755,
      "grad_norm": 3.21875,
      "learning_rate": 0.00047082770546938457,
      "loss": 5.8113,
      "mean_token_accuracy": 0.17233244627714156,
      "num_tokens": 36164902.0,
      "step": 19395
    },
    {
      "entropy": 6.084652376174927,
      "epoch": 1.7131755563405158,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004708121958163604,
      "loss": 5.8614,
      "mean_token_accuracy": 0.1720091372728348,
      "num_tokens": 36174128.0,
      "step": 19400
    },
    {
      "entropy": 6.158577346801758,
      "epoch": 1.713617096432356,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004707966823274866,
      "loss": 5.8847,
      "mean_token_accuracy": 0.16349819898605347,
      "num_tokens": 36183084.0,
      "step": 19405
    },
    {
      "entropy": 6.109979820251465,
      "epoch": 1.7140586365241965,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047078116500306723,
      "loss": 5.8399,
      "mean_token_accuracy": 0.1679747998714447,
      "num_tokens": 36191650.0,
      "step": 19410
    },
    {
      "entropy": 6.0948741912841795,
      "epoch": 1.7145001766160366,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047076564384340616,
      "loss": 5.8253,
      "mean_token_accuracy": 0.16588324308395386,
      "num_tokens": 36200712.0,
      "step": 19415
    },
    {
      "entropy": 6.0532787322998045,
      "epoch": 1.714941716707877,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004707501188488074,
      "loss": 5.7967,
      "mean_token_accuracy": 0.16974953413009644,
      "num_tokens": 36209347.0,
      "step": 19420
    },
    {
      "entropy": 6.102224588394165,
      "epoch": 1.7153832567997174,
      "grad_norm": 5.0,
      "learning_rate": 0.00047073459001957524,
      "loss": 5.805,
      "mean_token_accuracy": 0.16661521196365356,
      "num_tokens": 36218233.0,
      "step": 19425
    },
    {
      "entropy": 6.153863000869751,
      "epoch": 1.7158247968915576,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004707190573560137,
      "loss": 5.8381,
      "mean_token_accuracy": 0.16955134123563767,
      "num_tokens": 36227627.0,
      "step": 19430
    },
    {
      "entropy": 6.03925313949585,
      "epoch": 1.7162663369833981,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004707035208584272,
      "loss": 5.8079,
      "mean_token_accuracy": 0.1719197452068329,
      "num_tokens": 36236298.0,
      "step": 19435
    },
    {
      "entropy": 6.035017299652099,
      "epoch": 1.7167078770752384,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047068798052711996,
      "loss": 5.8643,
      "mean_token_accuracy": 0.17071401476860046,
      "num_tokens": 36245623.0,
      "step": 19440
    },
    {
      "entropy": 6.055868625640869,
      "epoch": 1.7171494171670787,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004706724363623965,
      "loss": 5.7675,
      "mean_token_accuracy": 0.17812453359365463,
      "num_tokens": 36255117.0,
      "step": 19445
    },
    {
      "entropy": 6.096489763259887,
      "epoch": 1.7175909572589192,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047065688836456134,
      "loss": 5.8013,
      "mean_token_accuracy": 0.17089735269546508,
      "num_tokens": 36264535.0,
      "step": 19450
    },
    {
      "entropy": 6.148611831665039,
      "epoch": 1.7180324973507595,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047064133653391895,
      "loss": 5.8449,
      "mean_token_accuracy": 0.16642197519540786,
      "num_tokens": 36274432.0,
      "step": 19455
    },
    {
      "entropy": 6.051980876922608,
      "epoch": 1.7184740374425997,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004706257808707741,
      "loss": 5.7224,
      "mean_token_accuracy": 0.17745835185050965,
      "num_tokens": 36282644.0,
      "step": 19460
    },
    {
      "entropy": 6.002152919769287,
      "epoch": 1.7189155775344402,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047061022137543144,
      "loss": 5.7786,
      "mean_token_accuracy": 0.17448026537895203,
      "num_tokens": 36290885.0,
      "step": 19465
    },
    {
      "entropy": 6.057797145843506,
      "epoch": 1.7193571176262805,
      "grad_norm": 2.421875,
      "learning_rate": 0.00047059465804819576,
      "loss": 5.8709,
      "mean_token_accuracy": 0.16982895284891128,
      "num_tokens": 36299477.0,
      "step": 19470
    },
    {
      "entropy": 6.052576303482056,
      "epoch": 1.7197986577181208,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004705790908893721,
      "loss": 5.8797,
      "mean_token_accuracy": 0.16778782457113267,
      "num_tokens": 36309638.0,
      "step": 19475
    },
    {
      "entropy": 6.1856413841247555,
      "epoch": 1.7202401978099613,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004705635198992651,
      "loss": 5.8953,
      "mean_token_accuracy": 0.16642554849386215,
      "num_tokens": 36319027.0,
      "step": 19480
    },
    {
      "entropy": 6.116280174255371,
      "epoch": 1.7206817379018013,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047054794507818,
      "loss": 5.7666,
      "mean_token_accuracy": 0.16644914597272872,
      "num_tokens": 36327936.0,
      "step": 19485
    },
    {
      "entropy": 6.113155508041382,
      "epoch": 1.7211232779936418,
      "grad_norm": 2.71875,
      "learning_rate": 0.00047053236642642193,
      "loss": 5.951,
      "mean_token_accuracy": 0.1643791005015373,
      "num_tokens": 36337162.0,
      "step": 19490
    },
    {
      "entropy": 6.108543777465821,
      "epoch": 1.7215648180854821,
      "grad_norm": 2.171875,
      "learning_rate": 0.000470516783944296,
      "loss": 5.7993,
      "mean_token_accuracy": 0.17253859490156173,
      "num_tokens": 36345587.0,
      "step": 19495
    },
    {
      "entropy": 6.094468832015991,
      "epoch": 1.7220063581773224,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047050119763210747,
      "loss": 5.8693,
      "mean_token_accuracy": 0.16964970231056214,
      "num_tokens": 36354536.0,
      "step": 19500
    },
    {
      "entropy": 6.050735425949097,
      "epoch": 1.722447898269163,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047048560749016167,
      "loss": 5.8892,
      "mean_token_accuracy": 0.16194503009319305,
      "num_tokens": 36363691.0,
      "step": 19505
    },
    {
      "entropy": 6.212757158279419,
      "epoch": 1.7228894383610032,
      "grad_norm": 2.8125,
      "learning_rate": 0.00047047001351876403,
      "loss": 5.9508,
      "mean_token_accuracy": 0.15841859728097915,
      "num_tokens": 36373898.0,
      "step": 19510
    },
    {
      "entropy": 6.17266616821289,
      "epoch": 1.7233309784528434,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004704544157182199,
      "loss": 5.9431,
      "mean_token_accuracy": 0.16280186772346497,
      "num_tokens": 36384345.0,
      "step": 19515
    },
    {
      "entropy": 6.102886438369751,
      "epoch": 1.723772518544684,
      "grad_norm": 2.765625,
      "learning_rate": 0.00047043881408883497,
      "loss": 5.7886,
      "mean_token_accuracy": 0.17599017322063445,
      "num_tokens": 36393021.0,
      "step": 19520
    },
    {
      "entropy": 6.10330023765564,
      "epoch": 1.7242140586365242,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004704232086309148,
      "loss": 5.8035,
      "mean_token_accuracy": 0.16605061143636704,
      "num_tokens": 36402729.0,
      "step": 19525
    },
    {
      "entropy": 5.984959888458252,
      "epoch": 1.7246555987283645,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047040759934476517,
      "loss": 5.7352,
      "mean_token_accuracy": 0.172627030313015,
      "num_tokens": 36412360.0,
      "step": 19530
    },
    {
      "entropy": 6.046199035644531,
      "epoch": 1.725097138820205,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004703919862306918,
      "loss": 5.7489,
      "mean_token_accuracy": 0.17245676070451738,
      "num_tokens": 36420245.0,
      "step": 19535
    },
    {
      "entropy": 6.035966062545777,
      "epoch": 1.7255386789120453,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004703763692890006,
      "loss": 5.726,
      "mean_token_accuracy": 0.17858747094869615,
      "num_tokens": 36428758.0,
      "step": 19540
    },
    {
      "entropy": 6.00393443107605,
      "epoch": 1.7259802190038855,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004703607485199973,
      "loss": 5.7872,
      "mean_token_accuracy": 0.1681286007165909,
      "num_tokens": 36437883.0,
      "step": 19545
    },
    {
      "entropy": 6.0812845706939695,
      "epoch": 1.726421759095726,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047034512392398815,
      "loss": 5.819,
      "mean_token_accuracy": 0.16681404709815978,
      "num_tokens": 36447950.0,
      "step": 19550
    },
    {
      "entropy": 6.199726867675781,
      "epoch": 1.726863299187566,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004703294955012791,
      "loss": 5.8509,
      "mean_token_accuracy": 0.16741348952054977,
      "num_tokens": 36456993.0,
      "step": 19555
    },
    {
      "entropy": 6.149022102355957,
      "epoch": 1.7273048392794066,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004703138632521762,
      "loss": 5.9213,
      "mean_token_accuracy": 0.16558149605989456,
      "num_tokens": 36465942.0,
      "step": 19560
    },
    {
      "entropy": 6.06173882484436,
      "epoch": 1.7277463793712469,
      "grad_norm": 3.359375,
      "learning_rate": 0.000470298227176986,
      "loss": 5.8751,
      "mean_token_accuracy": 0.16825119853019715,
      "num_tokens": 36474906.0,
      "step": 19565
    },
    {
      "entropy": 6.1172842502594,
      "epoch": 1.7281879194630871,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004702825872760145,
      "loss": 5.8107,
      "mean_token_accuracy": 0.16862916797399521,
      "num_tokens": 36483579.0,
      "step": 19570
    },
    {
      "entropy": 6.074004697799682,
      "epoch": 1.7286294595549276,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004702669435495681,
      "loss": 5.7894,
      "mean_token_accuracy": 0.1761271372437477,
      "num_tokens": 36492541.0,
      "step": 19575
    },
    {
      "entropy": 6.154872274398803,
      "epoch": 1.729070999646768,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004702512959979535,
      "loss": 5.9563,
      "mean_token_accuracy": 0.16102923154830934,
      "num_tokens": 36502833.0,
      "step": 19580
    },
    {
      "entropy": 6.107817268371582,
      "epoch": 1.7295125397386082,
      "grad_norm": 3.421875,
      "learning_rate": 0.00047023564462147697,
      "loss": 5.8255,
      "mean_token_accuracy": 0.1689919799566269,
      "num_tokens": 36513091.0,
      "step": 19585
    },
    {
      "entropy": 6.100772857666016,
      "epoch": 1.7299540798304487,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047021998942044517,
      "loss": 5.8479,
      "mean_token_accuracy": 0.1649123027920723,
      "num_tokens": 36522911.0,
      "step": 19590
    },
    {
      "entropy": 6.163949012756348,
      "epoch": 1.730395619922289,
      "grad_norm": 3.0625,
      "learning_rate": 0.00047020433039516484,
      "loss": 5.8941,
      "mean_token_accuracy": 0.1696918487548828,
      "num_tokens": 36531934.0,
      "step": 19595
    },
    {
      "entropy": 6.094933557510376,
      "epoch": 1.7308371600141292,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004701886675459427,
      "loss": 5.813,
      "mean_token_accuracy": 0.17123899459838868,
      "num_tokens": 36541823.0,
      "step": 19600
    },
    {
      "entropy": 6.106542778015137,
      "epoch": 1.7312787001059697,
      "grad_norm": 3.265625,
      "learning_rate": 0.00047017300087308557,
      "loss": 5.9149,
      "mean_token_accuracy": 0.16031504273414612,
      "num_tokens": 36550954.0,
      "step": 19605
    },
    {
      "entropy": 6.095903158187866,
      "epoch": 1.73172024019781,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004701573303769004,
      "loss": 5.8091,
      "mean_token_accuracy": 0.17046509832143783,
      "num_tokens": 36560689.0,
      "step": 19610
    },
    {
      "entropy": 6.089792633056641,
      "epoch": 1.7321617802896503,
      "grad_norm": 2.015625,
      "learning_rate": 0.000470141656057694,
      "loss": 5.7945,
      "mean_token_accuracy": 0.16714853197336196,
      "num_tokens": 36569587.0,
      "step": 19615
    },
    {
      "entropy": 6.089764738082886,
      "epoch": 1.7326033203814908,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047012597791577364,
      "loss": 5.8392,
      "mean_token_accuracy": 0.16799080967903138,
      "num_tokens": 36579609.0,
      "step": 19620
    },
    {
      "entropy": 6.080531883239746,
      "epoch": 1.7330448604733308,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004701102959514463,
      "loss": 5.8314,
      "mean_token_accuracy": 0.1664298355579376,
      "num_tokens": 36589685.0,
      "step": 19625
    },
    {
      "entropy": 6.119324541091919,
      "epoch": 1.7334864005651713,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047009461016501926,
      "loss": 5.7965,
      "mean_token_accuracy": 0.16852269619703292,
      "num_tokens": 36598225.0,
      "step": 19630
    },
    {
      "entropy": 5.99651346206665,
      "epoch": 1.7339279406570116,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047007892055679976,
      "loss": 5.7279,
      "mean_token_accuracy": 0.1763758346438408,
      "num_tokens": 36607481.0,
      "step": 19635
    },
    {
      "entropy": 6.089128828048706,
      "epoch": 1.7343694807488519,
      "grad_norm": 1.859375,
      "learning_rate": 0.00047006322712709513,
      "loss": 5.8264,
      "mean_token_accuracy": 0.16984937191009522,
      "num_tokens": 36616707.0,
      "step": 19640
    },
    {
      "entropy": 6.168269109725952,
      "epoch": 1.7348110208406924,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047004752987621286,
      "loss": 5.9453,
      "mean_token_accuracy": 0.16389377117156984,
      "num_tokens": 36626290.0,
      "step": 19645
    },
    {
      "entropy": 6.09074854850769,
      "epoch": 1.7352525609325327,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004700318288044603,
      "loss": 5.8425,
      "mean_token_accuracy": 0.1714595824480057,
      "num_tokens": 36635155.0,
      "step": 19650
    },
    {
      "entropy": 6.060588836669922,
      "epoch": 1.735694101024373,
      "grad_norm": 2.890625,
      "learning_rate": 0.00047001612391214527,
      "loss": 5.8194,
      "mean_token_accuracy": 0.17118046283721924,
      "num_tokens": 36644632.0,
      "step": 19655
    },
    {
      "entropy": 6.100505685806274,
      "epoch": 1.7361356411162134,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004700004151995753,
      "loss": 5.8707,
      "mean_token_accuracy": 0.16182403713464738,
      "num_tokens": 36654818.0,
      "step": 19660
    },
    {
      "entropy": 6.0540351390838625,
      "epoch": 1.7365771812080537,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046998470266705795,
      "loss": 5.7152,
      "mean_token_accuracy": 0.1802017092704773,
      "num_tokens": 36664149.0,
      "step": 19665
    },
    {
      "entropy": 6.0944374084472654,
      "epoch": 1.737018721299894,
      "grad_norm": 3.078125,
      "learning_rate": 0.00046996898631490125,
      "loss": 5.7787,
      "mean_token_accuracy": 0.17128826677799225,
      "num_tokens": 36673447.0,
      "step": 19670
    },
    {
      "entropy": 6.199214220046997,
      "epoch": 1.7374602613917345,
      "grad_norm": 1.96875,
      "learning_rate": 0.000469953266143413,
      "loss": 5.9539,
      "mean_token_accuracy": 0.15941453576087952,
      "num_tokens": 36683426.0,
      "step": 19675
    },
    {
      "entropy": 6.153524684906006,
      "epoch": 1.7379018014835748,
      "grad_norm": 2.34375,
      "learning_rate": 0.00046993754215290116,
      "loss": 5.8331,
      "mean_token_accuracy": 0.16669536083936692,
      "num_tokens": 36692708.0,
      "step": 19680
    },
    {
      "entropy": 5.983818435668946,
      "epoch": 1.738343341575415,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004699218143436737,
      "loss": 5.7324,
      "mean_token_accuracy": 0.17691716253757478,
      "num_tokens": 36701527.0,
      "step": 19685
    },
    {
      "entropy": 6.027654647827148,
      "epoch": 1.7387848816672555,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046990608271603874,
      "loss": 5.9114,
      "mean_token_accuracy": 0.16957876831293106,
      "num_tokens": 36711458.0,
      "step": 19690
    },
    {
      "entropy": 6.037820100784302,
      "epoch": 1.7392264217590956,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004698903472703045,
      "loss": 5.805,
      "mean_token_accuracy": 0.17394464164972306,
      "num_tokens": 36721260.0,
      "step": 19695
    },
    {
      "entropy": 6.138068008422851,
      "epoch": 1.739667961850936,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046987460800677915,
      "loss": 5.8248,
      "mean_token_accuracy": 0.16751631796360017,
      "num_tokens": 36730496.0,
      "step": 19700
    },
    {
      "entropy": 6.05646333694458,
      "epoch": 1.7401095019427764,
      "grad_norm": 1.71875,
      "learning_rate": 0.00046985886492577106,
      "loss": 5.8135,
      "mean_token_accuracy": 0.16532355397939683,
      "num_tokens": 36739016.0,
      "step": 19705
    },
    {
      "entropy": 6.097438144683838,
      "epoch": 1.7405510420346166,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004698431180275887,
      "loss": 5.9062,
      "mean_token_accuracy": 0.15973488241434097,
      "num_tokens": 36748259.0,
      "step": 19710
    },
    {
      "entropy": 6.26218729019165,
      "epoch": 1.7409925821264571,
      "grad_norm": 3.0,
      "learning_rate": 0.00046982736731254044,
      "loss": 5.9507,
      "mean_token_accuracy": 0.1565454736351967,
      "num_tokens": 36758174.0,
      "step": 19715
    },
    {
      "entropy": 6.111567783355713,
      "epoch": 1.7414341222182974,
      "grad_norm": 3.125,
      "learning_rate": 0.00046981161278093487,
      "loss": 5.8945,
      "mean_token_accuracy": 0.16601982116699218,
      "num_tokens": 36767286.0,
      "step": 19720
    },
    {
      "entropy": 6.07186861038208,
      "epoch": 1.7418756623101377,
      "grad_norm": 2.375,
      "learning_rate": 0.00046979585443308056,
      "loss": 5.7467,
      "mean_token_accuracy": 0.17302797436714173,
      "num_tokens": 36775818.0,
      "step": 19725
    },
    {
      "entropy": 6.085107898712158,
      "epoch": 1.7423172024019782,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046978009226928633,
      "loss": 5.829,
      "mean_token_accuracy": 0.1721041664481163,
      "num_tokens": 36785489.0,
      "step": 19730
    },
    {
      "entropy": 6.09384126663208,
      "epoch": 1.7427587424938185,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004697643262898608,
      "loss": 5.7964,
      "mean_token_accuracy": 0.1717776447534561,
      "num_tokens": 36794241.0,
      "step": 19735
    },
    {
      "entropy": 6.077766799926758,
      "epoch": 1.7432002825856587,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004697485564951128,
      "loss": 5.8292,
      "mean_token_accuracy": 0.16362475007772445,
      "num_tokens": 36803477.0,
      "step": 19740
    },
    {
      "entropy": 6.07278265953064,
      "epoch": 1.7436418226774992,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004697327828853515,
      "loss": 5.8019,
      "mean_token_accuracy": 0.16485653519630433,
      "num_tokens": 36811974.0,
      "step": 19745
    },
    {
      "entropy": 6.1780225276947025,
      "epoch": 1.7440833627693395,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004697170054608857,
      "loss": 5.9123,
      "mean_token_accuracy": 0.1634774312376976,
      "num_tokens": 36822049.0,
      "step": 19750
    },
    {
      "entropy": 6.140384531021118,
      "epoch": 1.7445249028611798,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004697012242220245,
      "loss": 5.8269,
      "mean_token_accuracy": 0.16831426024436952,
      "num_tokens": 36831989.0,
      "step": 19755
    },
    {
      "entropy": 6.10304365158081,
      "epoch": 1.7449664429530203,
      "grad_norm": 2.8125,
      "learning_rate": 0.00046968543916907704,
      "loss": 5.8637,
      "mean_token_accuracy": 0.16519104689359665,
      "num_tokens": 36841141.0,
      "step": 19760
    },
    {
      "entropy": 6.027408790588379,
      "epoch": 1.7454079830448603,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046966965030235253,
      "loss": 5.7151,
      "mean_token_accuracy": 0.17647875398397445,
      "num_tokens": 36850101.0,
      "step": 19765
    },
    {
      "entropy": 6.094846868515015,
      "epoch": 1.7458495231367008,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046965385762216024,
      "loss": 5.8311,
      "mean_token_accuracy": 0.17147888392210006,
      "num_tokens": 36859234.0,
      "step": 19770
    },
    {
      "entropy": 6.17104172706604,
      "epoch": 1.746291063228541,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004696380611288096,
      "loss": 5.9088,
      "mean_token_accuracy": 0.16515255123376846,
      "num_tokens": 36868937.0,
      "step": 19775
    },
    {
      "entropy": 6.044730949401855,
      "epoch": 1.7467326033203814,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046962226082261014,
      "loss": 5.7654,
      "mean_token_accuracy": 0.17578153312206268,
      "num_tokens": 36878380.0,
      "step": 19780
    },
    {
      "entropy": 6.036913013458252,
      "epoch": 1.7471741434122219,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004696064567038712,
      "loss": 5.8811,
      "mean_token_accuracy": 0.1676212280988693,
      "num_tokens": 36887702.0,
      "step": 19785
    },
    {
      "entropy": 6.08637113571167,
      "epoch": 1.7476156835040622,
      "grad_norm": 2.34375,
      "learning_rate": 0.00046959064877290246,
      "loss": 5.7464,
      "mean_token_accuracy": 0.18063959777355193,
      "num_tokens": 36896777.0,
      "step": 19790
    },
    {
      "entropy": 6.148621225357056,
      "epoch": 1.7480572235959024,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046957483703001366,
      "loss": 5.7683,
      "mean_token_accuracy": 0.1718536615371704,
      "num_tokens": 36905256.0,
      "step": 19795
    },
    {
      "entropy": 6.047731256484985,
      "epoch": 1.748498763687743,
      "grad_norm": 2.375,
      "learning_rate": 0.0004695590214755143,
      "loss": 5.7819,
      "mean_token_accuracy": 0.17247138172388077,
      "num_tokens": 36913767.0,
      "step": 19800
    },
    {
      "entropy": 5.978545427322388,
      "epoch": 1.7489403037795832,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046954320210971444,
      "loss": 5.8065,
      "mean_token_accuracy": 0.17074050158262252,
      "num_tokens": 36922838.0,
      "step": 19805
    },
    {
      "entropy": 6.037084150314331,
      "epoch": 1.7493818438714235,
      "grad_norm": 2.25,
      "learning_rate": 0.00046952737893292396,
      "loss": 5.7119,
      "mean_token_accuracy": 0.17612875252962112,
      "num_tokens": 36931955.0,
      "step": 19810
    },
    {
      "entropy": 6.0667730331420895,
      "epoch": 1.749823383963264,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00046951155194545265,
      "loss": 5.7734,
      "mean_token_accuracy": 0.1687113270163536,
      "num_tokens": 36941505.0,
      "step": 19815
    },
    {
      "entropy": 6.054841566085815,
      "epoch": 1.7502649240551043,
      "grad_norm": 2.25,
      "learning_rate": 0.00046949572114761076,
      "loss": 5.8552,
      "mean_token_accuracy": 0.16072550565004348,
      "num_tokens": 36951007.0,
      "step": 19820
    },
    {
      "entropy": 6.1119975566864015,
      "epoch": 1.7507064641469445,
      "grad_norm": 1.859375,
      "learning_rate": 0.00046947988653970826,
      "loss": 5.8409,
      "mean_token_accuracy": 0.17078903466463088,
      "num_tokens": 36961111.0,
      "step": 19825
    },
    {
      "entropy": 6.100942707061767,
      "epoch": 1.751148004238785,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004694640481220554,
      "loss": 5.8426,
      "mean_token_accuracy": 0.16860979795455933,
      "num_tokens": 36969605.0,
      "step": 19830
    },
    {
      "entropy": 6.128773736953735,
      "epoch": 1.751589544330625,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004694482058949624,
      "loss": 5.8403,
      "mean_token_accuracy": 0.16082397550344468,
      "num_tokens": 36979051.0,
      "step": 19835
    },
    {
      "entropy": 6.063669013977051,
      "epoch": 1.7520310844224656,
      "grad_norm": 3.90625,
      "learning_rate": 0.00046943235985873975,
      "loss": 5.779,
      "mean_token_accuracy": 0.17177076786756515,
      "num_tokens": 36988807.0,
      "step": 19840
    },
    {
      "entropy": 6.079297637939453,
      "epoch": 1.7524726245143059,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004694165100136976,
      "loss": 5.71,
      "mean_token_accuracy": 0.18016556054353713,
      "num_tokens": 36997813.0,
      "step": 19845
    },
    {
      "entropy": 6.092271947860718,
      "epoch": 1.7529141646061461,
      "grad_norm": 2.46875,
      "learning_rate": 0.00046940065636014665,
      "loss": 5.8083,
      "mean_token_accuracy": 0.16886910200119018,
      "num_tokens": 37007849.0,
      "step": 19850
    },
    {
      "entropy": 6.031528520584106,
      "epoch": 1.7533557046979866,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004693847988983974,
      "loss": 5.7032,
      "mean_token_accuracy": 0.18317572325468062,
      "num_tokens": 37016109.0,
      "step": 19855
    },
    {
      "entropy": 6.089900875091553,
      "epoch": 1.753797244789827,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004693689376287605,
      "loss": 5.7942,
      "mean_token_accuracy": 0.16811697483062743,
      "num_tokens": 37025156.0,
      "step": 19860
    },
    {
      "entropy": 6.141596508026123,
      "epoch": 1.7542387848816672,
      "grad_norm": 1.796875,
      "learning_rate": 0.00046935307255154663,
      "loss": 5.9383,
      "mean_token_accuracy": 0.16267292499542235,
      "num_tokens": 37035361.0,
      "step": 19865
    },
    {
      "entropy": 6.1525774002075195,
      "epoch": 1.7546803249735077,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004693372036670667,
      "loss": 5.8488,
      "mean_token_accuracy": 0.16371381133794785,
      "num_tokens": 37045249.0,
      "step": 19870
    },
    {
      "entropy": 6.102923059463501,
      "epoch": 1.755121865065348,
      "grad_norm": 2.796875,
      "learning_rate": 0.00046932133097563137,
      "loss": 5.8397,
      "mean_token_accuracy": 0.1699132189154625,
      "num_tokens": 37054637.0,
      "step": 19875
    },
    {
      "entropy": 6.089144992828369,
      "epoch": 1.7555634051571882,
      "grad_norm": 3.703125,
      "learning_rate": 0.00046930545447755167,
      "loss": 5.8461,
      "mean_token_accuracy": 0.17304728478193282,
      "num_tokens": 37064557.0,
      "step": 19880
    },
    {
      "entropy": 6.085269832611084,
      "epoch": 1.7560049452490287,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004692895741731387,
      "loss": 5.8649,
      "mean_token_accuracy": 0.1619995877146721,
      "num_tokens": 37073915.0,
      "step": 19885
    },
    {
      "entropy": 6.152730321884155,
      "epoch": 1.756446485340869,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004692736900627034,
      "loss": 5.8317,
      "mean_token_accuracy": 0.16984189599752425,
      "num_tokens": 37083008.0,
      "step": 19890
    },
    {
      "entropy": 6.136660242080689,
      "epoch": 1.7568880254327093,
      "grad_norm": 2.625,
      "learning_rate": 0.00046925780214655713,
      "loss": 5.874,
      "mean_token_accuracy": 0.16377387642860414,
      "num_tokens": 37092742.0,
      "step": 19895
    },
    {
      "entropy": 6.154133319854736,
      "epoch": 1.7573295655245498,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004692419104250108,
      "loss": 5.882,
      "mean_token_accuracy": 0.1570454090833664,
      "num_tokens": 37102242.0,
      "step": 19900
    },
    {
      "entropy": 6.112510108947754,
      "epoch": 1.7577711056163898,
      "grad_norm": 4.875,
      "learning_rate": 0.0004692260148983761,
      "loss": 5.8054,
      "mean_token_accuracy": 0.17799424082040788,
      "num_tokens": 37111304.0,
      "step": 19905
    },
    {
      "entropy": 6.095273447036743,
      "epoch": 1.7582126457082303,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004692101155669641,
      "loss": 5.8307,
      "mean_token_accuracy": 0.16950057744979857,
      "num_tokens": 37121420.0,
      "step": 19910
    },
    {
      "entropy": 6.190530395507812,
      "epoch": 1.7586541858000706,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004691942124310865,
      "loss": 5.8419,
      "mean_token_accuracy": 0.17063725292682647,
      "num_tokens": 37131265.0,
      "step": 19915
    },
    {
      "entropy": 6.0620628833770756,
      "epoch": 1.7590957258919109,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046917830549105465,
      "loss": 5.769,
      "mean_token_accuracy": 0.17321426421403885,
      "num_tokens": 37141097.0,
      "step": 19920
    },
    {
      "entropy": 5.94435863494873,
      "epoch": 1.7595372659837514,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004691623947471803,
      "loss": 5.6592,
      "mean_token_accuracy": 0.18381442725658417,
      "num_tokens": 37149687.0,
      "step": 19925
    },
    {
      "entropy": 6.100812149047852,
      "epoch": 1.7599788060755917,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046914648019977495,
      "loss": 5.8145,
      "mean_token_accuracy": 0.17705095559358597,
      "num_tokens": 37159219.0,
      "step": 19930
    },
    {
      "entropy": 6.027207612991333,
      "epoch": 1.760420346167432,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046913056184915054,
      "loss": 5.8203,
      "mean_token_accuracy": 0.16838347762823105,
      "num_tokens": 37168681.0,
      "step": 19935
    },
    {
      "entropy": 5.941357851028442,
      "epoch": 1.7608618862592724,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004691146396956188,
      "loss": 5.7167,
      "mean_token_accuracy": 0.17878993153572081,
      "num_tokens": 37177932.0,
      "step": 19940
    },
    {
      "entropy": 6.082370233535767,
      "epoch": 1.7613034263511127,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004690987137394917,
      "loss": 5.745,
      "mean_token_accuracy": 0.16805656850337983,
      "num_tokens": 37186783.0,
      "step": 19945
    },
    {
      "entropy": 6.136793756484986,
      "epoch": 1.761744966442953,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004690827839810812,
      "loss": 5.9161,
      "mean_token_accuracy": 0.16153229624032975,
      "num_tokens": 37196290.0,
      "step": 19950
    },
    {
      "entropy": 6.0889892578125,
      "epoch": 1.7621865065347935,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046906685042069927,
      "loss": 5.7614,
      "mean_token_accuracy": 0.17487434446811675,
      "num_tokens": 37205057.0,
      "step": 19955
    },
    {
      "entropy": 6.107542037963867,
      "epoch": 1.7626280466266337,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004690509130586582,
      "loss": 5.827,
      "mean_token_accuracy": 0.16696810275316237,
      "num_tokens": 37214078.0,
      "step": 19960
    },
    {
      "entropy": 6.101221895217895,
      "epoch": 1.763069586718474,
      "grad_norm": 4.03125,
      "learning_rate": 0.00046903497189527006,
      "loss": 5.7814,
      "mean_token_accuracy": 0.17532793432474136,
      "num_tokens": 37223232.0,
      "step": 19965
    },
    {
      "entropy": 6.048960256576538,
      "epoch": 1.7635111268103145,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004690190269308472,
      "loss": 5.7869,
      "mean_token_accuracy": 0.17661932855844498,
      "num_tokens": 37232054.0,
      "step": 19970
    },
    {
      "entropy": 6.057698488235474,
      "epoch": 1.7639526669021546,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004690030781657019,
      "loss": 5.7457,
      "mean_token_accuracy": 0.17482176423072815,
      "num_tokens": 37241451.0,
      "step": 19975
    },
    {
      "entropy": 6.0922942638397215,
      "epoch": 1.764394206993995,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004689871256001467,
      "loss": 5.7972,
      "mean_token_accuracy": 0.17089495807886124,
      "num_tokens": 37251029.0,
      "step": 19980
    },
    {
      "entropy": 6.035389852523804,
      "epoch": 1.7648357470858353,
      "grad_norm": 22.5,
      "learning_rate": 0.00046897116923449403,
      "loss": 5.8095,
      "mean_token_accuracy": 0.1662556454539299,
      "num_tokens": 37259594.0,
      "step": 19985
    },
    {
      "entropy": 6.004026556015015,
      "epoch": 1.7652772871776756,
      "grad_norm": 3.03125,
      "learning_rate": 0.00046895520906905635,
      "loss": 5.7401,
      "mean_token_accuracy": 0.17376499325037004,
      "num_tokens": 37268824.0,
      "step": 19990
    },
    {
      "entropy": 6.05657000541687,
      "epoch": 1.7657188272695161,
      "grad_norm": 2.46875,
      "learning_rate": 0.00046893924510414654,
      "loss": 5.7654,
      "mean_token_accuracy": 0.17436862885951995,
      "num_tokens": 37278494.0,
      "step": 19995
    },
    {
      "entropy": 5.957307291030884,
      "epoch": 1.7661603673613564,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004689232773400772,
      "loss": 5.6039,
      "mean_token_accuracy": 0.19008844196796418,
      "num_tokens": 37287648.0,
      "step": 20000
    },
    {
      "entropy": 6.064302730560303,
      "epoch": 1.7666019074531967,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004689073057771611,
      "loss": 5.768,
      "mean_token_accuracy": 0.16729025691747665,
      "num_tokens": 37295938.0,
      "step": 20005
    },
    {
      "entropy": 6.0653585433959964,
      "epoch": 1.7670434475450372,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046889133041571126,
      "loss": 5.7588,
      "mean_token_accuracy": 0.17362576574087143,
      "num_tokens": 37305097.0,
      "step": 20010
    },
    {
      "entropy": 6.088988590240478,
      "epoch": 1.7674849876368774,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004688753512560405,
      "loss": 5.799,
      "mean_token_accuracy": 0.17262169867753982,
      "num_tokens": 37313672.0,
      "step": 20015
    },
    {
      "entropy": 6.1185140132904055,
      "epoch": 1.7679265277287177,
      "grad_norm": 3.625,
      "learning_rate": 0.0004688593682984619,
      "loss": 5.8708,
      "mean_token_accuracy": 0.16818217039108277,
      "num_tokens": 37323342.0,
      "step": 20020
    },
    {
      "entropy": 6.065874099731445,
      "epoch": 1.7683680678205582,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004688433815432884,
      "loss": 5.8365,
      "mean_token_accuracy": 0.1694142997264862,
      "num_tokens": 37332578.0,
      "step": 20025
    },
    {
      "entropy": 6.139076995849609,
      "epoch": 1.7688096079123985,
      "grad_norm": 2.6875,
      "learning_rate": 0.00046882739099083335,
      "loss": 5.8311,
      "mean_token_accuracy": 0.16556988507509232,
      "num_tokens": 37341769.0,
      "step": 20030
    },
    {
      "entropy": 6.160698938369751,
      "epoch": 1.7692511480042388,
      "grad_norm": 2.953125,
      "learning_rate": 0.00046881139664141006,
      "loss": 5.8652,
      "mean_token_accuracy": 0.16831768900156022,
      "num_tokens": 37351158.0,
      "step": 20035
    },
    {
      "entropy": 6.147327280044555,
      "epoch": 1.7696926880960793,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004687953984953316,
      "loss": 5.8131,
      "mean_token_accuracy": 0.1712550103664398,
      "num_tokens": 37360924.0,
      "step": 20040
    },
    {
      "entropy": 6.049875497817993,
      "epoch": 1.7701342281879193,
      "grad_norm": 4.75,
      "learning_rate": 0.0004687793965529115,
      "loss": 5.7864,
      "mean_token_accuracy": 0.17124420702457427,
      "num_tokens": 37369295.0,
      "step": 20045
    },
    {
      "entropy": 6.04566650390625,
      "epoch": 1.7705757682797598,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004687633908144633,
      "loss": 5.7938,
      "mean_token_accuracy": 0.16798923015594483,
      "num_tokens": 37379135.0,
      "step": 20050
    },
    {
      "entropy": 5.994712543487549,
      "epoch": 1.7710173083716,
      "grad_norm": 3.234375,
      "learning_rate": 0.00046874738128030045,
      "loss": 5.6822,
      "mean_token_accuracy": 0.1756627306342125,
      "num_tokens": 37387927.0,
      "step": 20055
    },
    {
      "entropy": 6.0350676536560055,
      "epoch": 1.7714588484634404,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004687313679507366,
      "loss": 5.7228,
      "mean_token_accuracy": 0.1786044180393219,
      "num_tokens": 37397629.0,
      "step": 20060
    },
    {
      "entropy": 6.056587076187133,
      "epoch": 1.7719003885552809,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046871535082608544,
      "loss": 5.7579,
      "mean_token_accuracy": 0.17354556918144226,
      "num_tokens": 37407688.0,
      "step": 20065
    },
    {
      "entropy": 6.082600021362305,
      "epoch": 1.7723419286471211,
      "grad_norm": 4.625,
      "learning_rate": 0.0004686993299066607,
      "loss": 5.7641,
      "mean_token_accuracy": 0.17267437279224396,
      "num_tokens": 37417019.0,
      "step": 20070
    },
    {
      "entropy": 6.129771614074707,
      "epoch": 1.7727834687389614,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004686833051927763,
      "loss": 5.8277,
      "mean_token_accuracy": 0.1660886064171791,
      "num_tokens": 37427685.0,
      "step": 20075
    },
    {
      "entropy": 6.016252660751343,
      "epoch": 1.773225008830802,
      "grad_norm": 19.375,
      "learning_rate": 0.00046866727668474605,
      "loss": 5.7297,
      "mean_token_accuracy": 0.17796356081962586,
      "num_tokens": 37436844.0,
      "step": 20080
    },
    {
      "entropy": 6.010707521438599,
      "epoch": 1.7736665489226422,
      "grad_norm": 2.890625,
      "learning_rate": 0.000468651244382884,
      "loss": 5.7065,
      "mean_token_accuracy": 0.1779193624854088,
      "num_tokens": 37445755.0,
      "step": 20085
    },
    {
      "entropy": 6.078216314315796,
      "epoch": 1.7741080890144825,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004686352082875043,
      "loss": 5.893,
      "mean_token_accuracy": 0.16180832386016847,
      "num_tokens": 37454176.0,
      "step": 20090
    },
    {
      "entropy": 6.106676483154297,
      "epoch": 1.774549629106323,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004686191683989209,
      "loss": 5.7803,
      "mean_token_accuracy": 0.17976059317588805,
      "num_tokens": 37463696.0,
      "step": 20095
    },
    {
      "entropy": 6.138060235977173,
      "epoch": 1.7749911691981632,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004686031247174482,
      "loss": 5.8059,
      "mean_token_accuracy": 0.16886304318904877,
      "num_tokens": 37473651.0,
      "step": 20100
    },
    {
      "entropy": 6.1480451107025145,
      "epoch": 1.7754327092900035,
      "grad_norm": 9.3125,
      "learning_rate": 0.0004685870772434002,
      "loss": 5.866,
      "mean_token_accuracy": 0.16640237271785735,
      "num_tokens": 37483137.0,
      "step": 20105
    },
    {
      "entropy": 6.142318201065064,
      "epoch": 1.775874249381844,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004685710259770916,
      "loss": 5.873,
      "mean_token_accuracy": 0.1649242363870144,
      "num_tokens": 37492029.0,
      "step": 20110
    },
    {
      "entropy": 6.103555536270141,
      "epoch": 1.776315789473684,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046855497091883664,
      "loss": 5.7872,
      "mean_token_accuracy": 0.18035364747047425,
      "num_tokens": 37500421.0,
      "step": 20115
    },
    {
      "entropy": 6.03936538696289,
      "epoch": 1.7767573295655246,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004685389120689499,
      "loss": 5.7756,
      "mean_token_accuracy": 0.17496757358312606,
      "num_tokens": 37510332.0,
      "step": 20120
    },
    {
      "entropy": 6.023328399658203,
      "epoch": 1.7771988696573648,
      "grad_norm": 3.46875,
      "learning_rate": 0.000468522849427746,
      "loss": 5.7848,
      "mean_token_accuracy": 0.1779075562953949,
      "num_tokens": 37519802.0,
      "step": 20125
    },
    {
      "entropy": 6.186013555526733,
      "epoch": 1.7776404097492051,
      "grad_norm": 2.84375,
      "learning_rate": 0.00046850678299553943,
      "loss": 5.9233,
      "mean_token_accuracy": 0.1626991420984268,
      "num_tokens": 37529622.0,
      "step": 20130
    },
    {
      "entropy": 6.051730585098267,
      "epoch": 1.7780819498410456,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004684907127726451,
      "loss": 5.7567,
      "mean_token_accuracy": 0.17914391160011292,
      "num_tokens": 37539446.0,
      "step": 20135
    },
    {
      "entropy": 6.033084964752197,
      "epoch": 1.778523489932886,
      "grad_norm": 2.0,
      "learning_rate": 0.0004684746387593778,
      "loss": 5.7269,
      "mean_token_accuracy": 0.17872604727745056,
      "num_tokens": 37547877.0,
      "step": 20140
    },
    {
      "entropy": 6.109363603591919,
      "epoch": 1.7789650300247262,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004684585609560523,
      "loss": 5.9308,
      "mean_token_accuracy": 0.16783461570739747,
      "num_tokens": 37557841.0,
      "step": 20145
    },
    {
      "entropy": 6.137045812606812,
      "epoch": 1.7794065701165667,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004684424793629836,
      "loss": 5.8798,
      "mean_token_accuracy": 0.1611030474305153,
      "num_tokens": 37566876.0,
      "step": 20150
    },
    {
      "entropy": 6.143455076217651,
      "epoch": 1.779848110208407,
      "grad_norm": 2.515625,
      "learning_rate": 0.00046842639398048683,
      "loss": 5.8367,
      "mean_token_accuracy": 0.16474115550518037,
      "num_tokens": 37575571.0,
      "step": 20155
    },
    {
      "entropy": 6.17484130859375,
      "epoch": 1.7802896503002472,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004684103048088769,
      "loss": 5.9241,
      "mean_token_accuracy": 0.16889073252677916,
      "num_tokens": 37585972.0,
      "step": 20160
    },
    {
      "entropy": 6.115367126464844,
      "epoch": 1.7807311903920877,
      "grad_norm": 2.125,
      "learning_rate": 0.0004683942118484692,
      "loss": 5.8535,
      "mean_token_accuracy": 0.16657682210206987,
      "num_tokens": 37594849.0,
      "step": 20165
    },
    {
      "entropy": 6.093259716033936,
      "epoch": 1.781172730483928,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046837811509957883,
      "loss": 5.7891,
      "mean_token_accuracy": 0.16928683817386628,
      "num_tokens": 37603634.0,
      "step": 20170
    },
    {
      "entropy": 6.10488920211792,
      "epoch": 1.7816142705757683,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046836201456252116,
      "loss": 5.7416,
      "mean_token_accuracy": 0.17953264117240905,
      "num_tokens": 37612518.0,
      "step": 20175
    },
    {
      "entropy": 6.0905234813690186,
      "epoch": 1.7820558106676088,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004683459102376116,
      "loss": 5.7564,
      "mean_token_accuracy": 0.17812949866056443,
      "num_tokens": 37621553.0,
      "step": 20180
    },
    {
      "entropy": 6.096011161804199,
      "epoch": 1.7824973507594488,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046832980212516557,
      "loss": 5.8459,
      "mean_token_accuracy": 0.16855701059103012,
      "num_tokens": 37631263.0,
      "step": 20185
    },
    {
      "entropy": 6.0844989776611325,
      "epoch": 1.7829388908512893,
      "grad_norm": 3.484375,
      "learning_rate": 0.00046831369022549876,
      "loss": 5.741,
      "mean_token_accuracy": 0.17886846363544465,
      "num_tokens": 37640810.0,
      "step": 20190
    },
    {
      "entropy": 6.089738416671753,
      "epoch": 1.7833804309431296,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004682975745389267,
      "loss": 5.8289,
      "mean_token_accuracy": 0.16985976994037627,
      "num_tokens": 37650848.0,
      "step": 20195
    },
    {
      "entropy": 6.112750053405762,
      "epoch": 1.7838219710349699,
      "grad_norm": 2.9375,
      "learning_rate": 0.00046828145506576504,
      "loss": 5.7955,
      "mean_token_accuracy": 0.17283367067575456,
      "num_tokens": 37660370.0,
      "step": 20200
    },
    {
      "entropy": 6.079094409942627,
      "epoch": 1.7842635111268104,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004682653318063296,
      "loss": 5.7583,
      "mean_token_accuracy": 0.17057965695858002,
      "num_tokens": 37669570.0,
      "step": 20205
    },
    {
      "entropy": 6.0790680885314945,
      "epoch": 1.7847050512186506,
      "grad_norm": 3.171875,
      "learning_rate": 0.00046824920476093624,
      "loss": 5.8157,
      "mean_token_accuracy": 0.1683760702610016,
      "num_tokens": 37679585.0,
      "step": 20210
    },
    {
      "entropy": 6.077490234375,
      "epoch": 1.785146591310491,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046823307392990086,
      "loss": 5.8564,
      "mean_token_accuracy": 0.1694411650300026,
      "num_tokens": 37688736.0,
      "step": 20215
    },
    {
      "entropy": 6.1329395294189455,
      "epoch": 1.7855881314023314,
      "grad_norm": 2.875,
      "learning_rate": 0.00046821693931353947,
      "loss": 5.8811,
      "mean_token_accuracy": 0.15908513516187667,
      "num_tokens": 37699104.0,
      "step": 20220
    },
    {
      "entropy": 6.143720293045044,
      "epoch": 1.7860296714941717,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004682008009121681,
      "loss": 5.8682,
      "mean_token_accuracy": 0.17118003517389296,
      "num_tokens": 37708260.0,
      "step": 20225
    },
    {
      "entropy": 6.107336044311523,
      "epoch": 1.786471211586012,
      "grad_norm": 2.09375,
      "learning_rate": 0.000468184658726103,
      "loss": 5.8226,
      "mean_token_accuracy": 0.1685038611292839,
      "num_tokens": 37717442.0,
      "step": 20230
    },
    {
      "entropy": 6.0616100311279295,
      "epoch": 1.7869127516778525,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004681685127556602,
      "loss": 5.8572,
      "mean_token_accuracy": 0.16283371150493622,
      "num_tokens": 37727063.0,
      "step": 20235
    },
    {
      "entropy": 6.080146026611328,
      "epoch": 1.7873542917696927,
      "grad_norm": 2.671875,
      "learning_rate": 0.00046815236300115615,
      "loss": 5.7608,
      "mean_token_accuracy": 0.16819789707660676,
      "num_tokens": 37736441.0,
      "step": 20240
    },
    {
      "entropy": 6.031875371932983,
      "epoch": 1.787795831861533,
      "grad_norm": 2.75,
      "learning_rate": 0.00046813620946290716,
      "loss": 5.7153,
      "mean_token_accuracy": 0.17982517778873444,
      "num_tokens": 37745290.0,
      "step": 20245
    },
    {
      "entropy": 6.072076749801636,
      "epoch": 1.7882373719533735,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004681200521412296,
      "loss": 5.8303,
      "mean_token_accuracy": 0.17126999646425248,
      "num_tokens": 37754786.0,
      "step": 20250
    },
    {
      "entropy": 6.073743009567261,
      "epoch": 1.7886789120452136,
      "grad_norm": 4.03125,
      "learning_rate": 0.00046810389103644015,
      "loss": 5.8286,
      "mean_token_accuracy": 0.16740425825119018,
      "num_tokens": 37763513.0,
      "step": 20255
    },
    {
      "entropy": 6.075063800811767,
      "epoch": 1.789120452137054,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004680877261488552,
      "loss": 5.8114,
      "mean_token_accuracy": 0.16782309412956237,
      "num_tokens": 37771752.0,
      "step": 20260
    },
    {
      "entropy": 6.09251971244812,
      "epoch": 1.7895619922288943,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004680715574787916,
      "loss": 5.8343,
      "mean_token_accuracy": 0.16130578368902207,
      "num_tokens": 37781531.0,
      "step": 20265
    },
    {
      "entropy": 6.054262781143189,
      "epoch": 1.7900035323207346,
      "grad_norm": 2.484375,
      "learning_rate": 0.00046805538502656596,
      "loss": 5.7435,
      "mean_token_accuracy": 0.17107709348201752,
      "num_tokens": 37791469.0,
      "step": 20270
    },
    {
      "entropy": 6.114316463470459,
      "epoch": 1.7904450724125751,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004680392087924952,
      "loss": 5.8291,
      "mean_token_accuracy": 0.16594353467226028,
      "num_tokens": 37800301.0,
      "step": 20275
    },
    {
      "entropy": 6.071118354797363,
      "epoch": 1.7908866125044154,
      "grad_norm": 3.015625,
      "learning_rate": 0.000468023028776896,
      "loss": 5.6681,
      "mean_token_accuracy": 0.18431031107902526,
      "num_tokens": 37808849.0,
      "step": 20280
    },
    {
      "entropy": 6.142718076705933,
      "epoch": 1.7913281525962557,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004680068449800855,
      "loss": 5.8343,
      "mean_token_accuracy": 0.16702621728181838,
      "num_tokens": 37818632.0,
      "step": 20285
    },
    {
      "entropy": 6.077562236785889,
      "epoch": 1.7917696926880962,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004679906574023807,
      "loss": 5.8345,
      "mean_token_accuracy": 0.16573940366506576,
      "num_tokens": 37828158.0,
      "step": 20290
    },
    {
      "entropy": 6.062243223190308,
      "epoch": 1.7922112327799364,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004679744660440987,
      "loss": 5.7605,
      "mean_token_accuracy": 0.16861721724271775,
      "num_tokens": 37836242.0,
      "step": 20295
    },
    {
      "entropy": 6.07423243522644,
      "epoch": 1.7926527728717767,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046795827090555654,
      "loss": 5.8104,
      "mean_token_accuracy": 0.17325833737850188,
      "num_tokens": 37844286.0,
      "step": 20300
    },
    {
      "entropy": 6.046108865737915,
      "epoch": 1.7930943129636172,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004679420719870717,
      "loss": 5.7621,
      "mean_token_accuracy": 0.1713164433836937,
      "num_tokens": 37853440.0,
      "step": 20305
    },
    {
      "entropy": 6.0701988697052,
      "epoch": 1.7935358530554575,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004679258692889613,
      "loss": 5.6983,
      "mean_token_accuracy": 0.17428845316171646,
      "num_tokens": 37862511.0,
      "step": 20310
    },
    {
      "entropy": 6.155171966552734,
      "epoch": 1.7939773931472978,
      "grad_norm": 2.25,
      "learning_rate": 0.000467909662811543,
      "loss": 5.8992,
      "mean_token_accuracy": 0.16147223263978958,
      "num_tokens": 37872797.0,
      "step": 20315
    },
    {
      "entropy": 6.116675329208374,
      "epoch": 1.7944189332391383,
      "grad_norm": 3.046875,
      "learning_rate": 0.000467893452555134,
      "loss": 5.7568,
      "mean_token_accuracy": 0.17537894397974013,
      "num_tokens": 37881859.0,
      "step": 20320
    },
    {
      "entropy": 6.083496761322022,
      "epoch": 1.7948604733309783,
      "grad_norm": 2.53125,
      "learning_rate": 0.00046787723852005197,
      "loss": 5.7862,
      "mean_token_accuracy": 0.17786880135536193,
      "num_tokens": 37891381.0,
      "step": 20325
    },
    {
      "entropy": 5.956803512573242,
      "epoch": 1.7953020134228188,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004678610207066146,
      "loss": 5.6984,
      "mean_token_accuracy": 0.17156352698802949,
      "num_tokens": 37900554.0,
      "step": 20330
    },
    {
      "entropy": 6.136029529571533,
      "epoch": 1.795743553514659,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004678447991151395,
      "loss": 5.7611,
      "mean_token_accuracy": 0.1727130338549614,
      "num_tokens": 37908965.0,
      "step": 20335
    },
    {
      "entropy": 6.056249952316284,
      "epoch": 1.7961850936064994,
      "grad_norm": 2.765625,
      "learning_rate": 0.00046782857374594447,
      "loss": 5.7284,
      "mean_token_accuracy": 0.17350957840681075,
      "num_tokens": 37917555.0,
      "step": 20340
    },
    {
      "entropy": 5.997202634811401,
      "epoch": 1.7966266336983399,
      "grad_norm": 2.125,
      "learning_rate": 0.0004678123445993473,
      "loss": 5.7927,
      "mean_token_accuracy": 0.17182748764753342,
      "num_tokens": 37926334.0,
      "step": 20345
    },
    {
      "entropy": 6.022460842132569,
      "epoch": 1.7970681737901801,
      "grad_norm": 2.328125,
      "learning_rate": 0.000467796111675666,
      "loss": 5.747,
      "mean_token_accuracy": 0.17274363040924073,
      "num_tokens": 37935912.0,
      "step": 20350
    },
    {
      "entropy": 6.0903702735900875,
      "epoch": 1.7975097138820204,
      "grad_norm": 2.640625,
      "learning_rate": 0.00046777987497521845,
      "loss": 5.7964,
      "mean_token_accuracy": 0.17393017262220384,
      "num_tokens": 37945238.0,
      "step": 20355
    },
    {
      "entropy": 6.080442714691162,
      "epoch": 1.797951253973861,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046776363449832286,
      "loss": 5.7662,
      "mean_token_accuracy": 0.17735556066036223,
      "num_tokens": 37954945.0,
      "step": 20360
    },
    {
      "entropy": 6.045280122756958,
      "epoch": 1.7983927940657012,
      "grad_norm": 4.0625,
      "learning_rate": 0.00046774739024529736,
      "loss": 5.6776,
      "mean_token_accuracy": 0.18106734454631807,
      "num_tokens": 37964114.0,
      "step": 20365
    },
    {
      "entropy": 6.095221853256225,
      "epoch": 1.7988343341575415,
      "grad_norm": 2.359375,
      "learning_rate": 0.00046773114221645995,
      "loss": 5.8622,
      "mean_token_accuracy": 0.15842871218919755,
      "num_tokens": 37973774.0,
      "step": 20370
    },
    {
      "entropy": 6.120967674255371,
      "epoch": 1.799275874249382,
      "grad_norm": 3.0,
      "learning_rate": 0.00046771489041212916,
      "loss": 5.8038,
      "mean_token_accuracy": 0.1683808833360672,
      "num_tokens": 37982763.0,
      "step": 20375
    },
    {
      "entropy": 6.172139310836792,
      "epoch": 1.7997174143412222,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004676986348326233,
      "loss": 5.9112,
      "mean_token_accuracy": 0.16594945937395095,
      "num_tokens": 37992302.0,
      "step": 20380
    },
    {
      "entropy": 6.227605390548706,
      "epoch": 1.8001589544330625,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004676823754782607,
      "loss": 5.9338,
      "mean_token_accuracy": 0.15908068269491196,
      "num_tokens": 38001785.0,
      "step": 20385
    },
    {
      "entropy": 6.0478685855865475,
      "epoch": 1.800600494524903,
      "grad_norm": 3.375,
      "learning_rate": 0.00046766611234936,
      "loss": 5.7271,
      "mean_token_accuracy": 0.17077599763870238,
      "num_tokens": 38010823.0,
      "step": 20390
    },
    {
      "entropy": 6.012527942657471,
      "epoch": 1.801042034616743,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004676498454462398,
      "loss": 5.7494,
      "mean_token_accuracy": 0.18201036304235457,
      "num_tokens": 38019718.0,
      "step": 20395
    },
    {
      "entropy": 6.070411443710327,
      "epoch": 1.8014835747085836,
      "grad_norm": 3.234375,
      "learning_rate": 0.00046763357476921856,
      "loss": 5.7778,
      "mean_token_accuracy": 0.17086032330989837,
      "num_tokens": 38028962.0,
      "step": 20400
    },
    {
      "entropy": 6.092631578445435,
      "epoch": 1.8019251148004238,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004676173003186152,
      "loss": 5.7389,
      "mean_token_accuracy": 0.17887519523501397,
      "num_tokens": 38037824.0,
      "step": 20405
    },
    {
      "entropy": 6.111455965042114,
      "epoch": 1.802366654892264,
      "grad_norm": 2.578125,
      "learning_rate": 0.00046760102209474855,
      "loss": 5.8744,
      "mean_token_accuracy": 0.16635147780179976,
      "num_tokens": 38046764.0,
      "step": 20410
    },
    {
      "entropy": 6.096413898468017,
      "epoch": 1.8028081949841046,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004675847400979373,
      "loss": 5.789,
      "mean_token_accuracy": 0.17202276587486268,
      "num_tokens": 38055363.0,
      "step": 20415
    },
    {
      "entropy": 6.097694253921508,
      "epoch": 1.8032497350759449,
      "grad_norm": 1.96875,
      "learning_rate": 0.00046756845432850057,
      "loss": 5.8534,
      "mean_token_accuracy": 0.1675860106945038,
      "num_tokens": 38065471.0,
      "step": 20420
    },
    {
      "entropy": 6.154904222488403,
      "epoch": 1.8036912751677852,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004675521647867573,
      "loss": 5.9156,
      "mean_token_accuracy": 0.16407944560050963,
      "num_tokens": 38075448.0,
      "step": 20425
    },
    {
      "entropy": 6.065900611877441,
      "epoch": 1.8041328152596257,
      "grad_norm": 2.640625,
      "learning_rate": 0.00046753587147302665,
      "loss": 5.8825,
      "mean_token_accuracy": 0.17118899524211884,
      "num_tokens": 38085657.0,
      "step": 20430
    },
    {
      "entropy": 6.127964925765991,
      "epoch": 1.804574355351466,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004675195743876278,
      "loss": 5.8014,
      "mean_token_accuracy": 0.16822673827409745,
      "num_tokens": 38095580.0,
      "step": 20435
    },
    {
      "entropy": 6.216636276245117,
      "epoch": 1.8050158954433062,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046750327353088,
      "loss": 5.8997,
      "mean_token_accuracy": 0.16335139125585557,
      "num_tokens": 38104969.0,
      "step": 20440
    },
    {
      "entropy": 6.149291133880615,
      "epoch": 1.8054574355351467,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004674869689031026,
      "loss": 5.8733,
      "mean_token_accuracy": 0.16422299593687056,
      "num_tokens": 38114536.0,
      "step": 20445
    },
    {
      "entropy": 6.08988676071167,
      "epoch": 1.805898975626987,
      "grad_norm": 2.828125,
      "learning_rate": 0.00046747066050461486,
      "loss": 5.7665,
      "mean_token_accuracy": 0.17371692806482314,
      "num_tokens": 38122969.0,
      "step": 20450
    },
    {
      "entropy": 6.080980587005615,
      "epoch": 1.8063405157188273,
      "grad_norm": 3.171875,
      "learning_rate": 0.00046745434833573635,
      "loss": 5.8402,
      "mean_token_accuracy": 0.16300772726535798,
      "num_tokens": 38132892.0,
      "step": 20455
    },
    {
      "entropy": 5.934664964675903,
      "epoch": 1.8067820558106678,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004674380323967867,
      "loss": 5.8042,
      "mean_token_accuracy": 0.16829438656568527,
      "num_tokens": 38141950.0,
      "step": 20460
    },
    {
      "entropy": 6.145351696014404,
      "epoch": 1.8072235959025078,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004674217126880853,
      "loss": 5.8878,
      "mean_token_accuracy": 0.16407862007617952,
      "num_tokens": 38151314.0,
      "step": 20465
    },
    {
      "entropy": 6.261961507797241,
      "epoch": 1.8076651359943483,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004674053892099521,
      "loss": 5.8746,
      "mean_token_accuracy": 0.16993481963872908,
      "num_tokens": 38159992.0,
      "step": 20470
    },
    {
      "entropy": 6.109126806259155,
      "epoch": 1.8081066760861886,
      "grad_norm": 4.125,
      "learning_rate": 0.00046738906196270663,
      "loss": 5.7742,
      "mean_token_accuracy": 0.16726333796977996,
      "num_tokens": 38168436.0,
      "step": 20475
    },
    {
      "entropy": 5.991465091705322,
      "epoch": 1.8085482161780289,
      "grad_norm": 2.515625,
      "learning_rate": 0.000467372730946669,
      "loss": 5.7729,
      "mean_token_accuracy": 0.17277428656816482,
      "num_tokens": 38177932.0,
      "step": 20480
    },
    {
      "entropy": 6.1263312816619875,
      "epoch": 1.8089897562698694,
      "grad_norm": 2.734375,
      "learning_rate": 0.00046735639616215884,
      "loss": 5.7898,
      "mean_token_accuracy": 0.17274608612060546,
      "num_tokens": 38187727.0,
      "step": 20485
    },
    {
      "entropy": 6.121396493911743,
      "epoch": 1.8094312963617096,
      "grad_norm": 3.484375,
      "learning_rate": 0.00046734005760949634,
      "loss": 5.8514,
      "mean_token_accuracy": 0.1676426038146019,
      "num_tokens": 38198768.0,
      "step": 20490
    },
    {
      "entropy": 6.094011068344116,
      "epoch": 1.80987283645355,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00046732371528900154,
      "loss": 5.8382,
      "mean_token_accuracy": 0.1743569165468216,
      "num_tokens": 38207538.0,
      "step": 20495
    },
    {
      "entropy": 6.160304546356201,
      "epoch": 1.8103143765453904,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004673073692009945,
      "loss": 5.8766,
      "mean_token_accuracy": 0.16444033533334732,
      "num_tokens": 38216254.0,
      "step": 20500
    },
    {
      "entropy": 6.184422302246094,
      "epoch": 1.8107559166372307,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004672910193457955,
      "loss": 5.8576,
      "mean_token_accuracy": 0.16156531274318695,
      "num_tokens": 38225356.0,
      "step": 20505
    },
    {
      "entropy": 6.0654932975769045,
      "epoch": 1.811197456729071,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004672746657237248,
      "loss": 5.7445,
      "mean_token_accuracy": 0.17631847858428956,
      "num_tokens": 38234561.0,
      "step": 20510
    },
    {
      "entropy": 6.015161085128784,
      "epoch": 1.8116389968209115,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004672583083351027,
      "loss": 5.8028,
      "mean_token_accuracy": 0.16754768490791322,
      "num_tokens": 38244418.0,
      "step": 20515
    },
    {
      "entropy": 6.096727657318115,
      "epoch": 1.8120805369127517,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004672419471802497,
      "loss": 5.8017,
      "mean_token_accuracy": 0.17227547615766525,
      "num_tokens": 38253639.0,
      "step": 20520
    },
    {
      "entropy": 6.1182411193847654,
      "epoch": 1.812522077004592,
      "grad_norm": 5.875,
      "learning_rate": 0.00046722558225948627,
      "loss": 5.706,
      "mean_token_accuracy": 0.18689554184675217,
      "num_tokens": 38262401.0,
      "step": 20525
    },
    {
      "entropy": 6.0850132465362545,
      "epoch": 1.8129636170964325,
      "grad_norm": 2.859375,
      "learning_rate": 0.00046720921357313303,
      "loss": 5.7889,
      "mean_token_accuracy": 0.17243619859218598,
      "num_tokens": 38271070.0,
      "step": 20530
    },
    {
      "entropy": 6.100644493103028,
      "epoch": 1.8134051571882726,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004671928411215106,
      "loss": 5.8102,
      "mean_token_accuracy": 0.17495137751102446,
      "num_tokens": 38280675.0,
      "step": 20535
    },
    {
      "entropy": 6.06340184211731,
      "epoch": 1.813846697280113,
      "grad_norm": 3.15625,
      "learning_rate": 0.00046717646490493977,
      "loss": 5.8488,
      "mean_token_accuracy": 0.16519464552402496,
      "num_tokens": 38290116.0,
      "step": 20540
    },
    {
      "entropy": 6.136339426040649,
      "epoch": 1.8142882373719533,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004671600849237412,
      "loss": 5.8119,
      "mean_token_accuracy": 0.17038616240024568,
      "num_tokens": 38299415.0,
      "step": 20545
    },
    {
      "entropy": 5.99956316947937,
      "epoch": 1.8147297774637936,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004671437011782359,
      "loss": 5.7188,
      "mean_token_accuracy": 0.1787692964076996,
      "num_tokens": 38308761.0,
      "step": 20550
    },
    {
      "entropy": 6.060312366485595,
      "epoch": 1.815171317555634,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004671273136687448,
      "loss": 5.8509,
      "mean_token_accuracy": 0.1694723829627037,
      "num_tokens": 38318212.0,
      "step": 20555
    },
    {
      "entropy": 6.086144208908081,
      "epoch": 1.8156128576474744,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004671109223955889,
      "loss": 5.8082,
      "mean_token_accuracy": 0.1736452355980873,
      "num_tokens": 38327233.0,
      "step": 20560
    },
    {
      "entropy": 6.095857858657837,
      "epoch": 1.8160543977393147,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004670945273590892,
      "loss": 5.6914,
      "mean_token_accuracy": 0.18070479184389115,
      "num_tokens": 38336473.0,
      "step": 20565
    },
    {
      "entropy": 6.100157928466797,
      "epoch": 1.8164959378311551,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004670781285595671,
      "loss": 5.857,
      "mean_token_accuracy": 0.16759061962366104,
      "num_tokens": 38345286.0,
      "step": 20570
    },
    {
      "entropy": 6.13175220489502,
      "epoch": 1.8169374779229954,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004670617259973435,
      "loss": 5.7564,
      "mean_token_accuracy": 0.1714487910270691,
      "num_tokens": 38355191.0,
      "step": 20575
    },
    {
      "entropy": 6.086813735961914,
      "epoch": 1.8173790180148357,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046704531967274007,
      "loss": 5.743,
      "mean_token_accuracy": 0.17049239873886107,
      "num_tokens": 38363902.0,
      "step": 20580
    },
    {
      "entropy": 6.077591848373413,
      "epoch": 1.8178205581066762,
      "grad_norm": 2.34375,
      "learning_rate": 0.00046702890958607804,
      "loss": 5.9076,
      "mean_token_accuracy": 0.16567446440458297,
      "num_tokens": 38374828.0,
      "step": 20585
    },
    {
      "entropy": 6.0379315376281735,
      "epoch": 1.8182620981985165,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004670124957376788,
      "loss": 5.7978,
      "mean_token_accuracy": 0.17197397351264954,
      "num_tokens": 38384918.0,
      "step": 20590
    },
    {
      "entropy": 6.144944763183593,
      "epoch": 1.8187036382903567,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004669960781278641,
      "loss": 5.8354,
      "mean_token_accuracy": 0.16581521630287172,
      "num_tokens": 38394005.0,
      "step": 20595
    },
    {
      "entropy": 6.199813652038574,
      "epoch": 1.8191451783821972,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004669796567569553,
      "loss": 5.8637,
      "mean_token_accuracy": 0.16711565256118774,
      "num_tokens": 38402722.0,
      "step": 20600
    },
    {
      "entropy": 6.048141813278198,
      "epoch": 1.8195867184740373,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004669632316252743,
      "loss": 5.7639,
      "mean_token_accuracy": 0.17448830604553223,
      "num_tokens": 38411726.0,
      "step": 20605
    },
    {
      "entropy": 6.146266269683838,
      "epoch": 1.8200282585658778,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046694680273314275,
      "loss": 5.8574,
      "mean_token_accuracy": 0.16246224641799928,
      "num_tokens": 38421187.0,
      "step": 20610
    },
    {
      "entropy": 6.132893037796021,
      "epoch": 1.820469798657718,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004669303700808824,
      "loss": 5.8603,
      "mean_token_accuracy": 0.16648274660110474,
      "num_tokens": 38430722.0,
      "step": 20615
    },
    {
      "entropy": 6.101836109161377,
      "epoch": 1.8209113387495584,
      "grad_norm": 2.375,
      "learning_rate": 0.0004669139336688154,
      "loss": 5.805,
      "mean_token_accuracy": 0.16964921057224275,
      "num_tokens": 38440763.0,
      "step": 20620
    },
    {
      "entropy": 6.116004753112793,
      "epoch": 1.8213528788413988,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046689749349726353,
      "loss": 5.8078,
      "mean_token_accuracy": 0.1723707839846611,
      "num_tokens": 38449298.0,
      "step": 20625
    },
    {
      "entropy": 6.112743616104126,
      "epoch": 1.8217944189332391,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004668810495665489,
      "loss": 5.795,
      "mean_token_accuracy": 0.17065605968236924,
      "num_tokens": 38458941.0,
      "step": 20630
    },
    {
      "entropy": 6.0263622283935545,
      "epoch": 1.8222359590250794,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004668646018769936,
      "loss": 5.7752,
      "mean_token_accuracy": 0.17536549121141434,
      "num_tokens": 38468381.0,
      "step": 20635
    },
    {
      "entropy": 6.108361005783081,
      "epoch": 1.82267749911692,
      "grad_norm": 3.21875,
      "learning_rate": 0.00046684815042891993,
      "loss": 5.8451,
      "mean_token_accuracy": 0.1661509655416012,
      "num_tokens": 38477767.0,
      "step": 20640
    },
    {
      "entropy": 6.077046966552734,
      "epoch": 1.8231190392087602,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004668316952226501,
      "loss": 5.7341,
      "mean_token_accuracy": 0.1715601474046707,
      "num_tokens": 38486884.0,
      "step": 20645
    },
    {
      "entropy": 6.074830961227417,
      "epoch": 1.8235605793006004,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004668152362585064,
      "loss": 5.8309,
      "mean_token_accuracy": 0.16225812286138536,
      "num_tokens": 38496243.0,
      "step": 20650
    },
    {
      "entropy": 6.015606880187988,
      "epoch": 1.824002119392441,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004667987735368113,
      "loss": 5.7275,
      "mean_token_accuracy": 0.1728453665971756,
      "num_tokens": 38504447.0,
      "step": 20655
    },
    {
      "entropy": 6.068365097045898,
      "epoch": 1.8244436594842812,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004667823070578874,
      "loss": 5.8495,
      "mean_token_accuracy": 0.17222337573766708,
      "num_tokens": 38514423.0,
      "step": 20660
    },
    {
      "entropy": 6.114740657806396,
      "epoch": 1.8248851995761215,
      "grad_norm": 2.5,
      "learning_rate": 0.00046676583682205713,
      "loss": 5.7738,
      "mean_token_accuracy": 0.1706966921687126,
      "num_tokens": 38522765.0,
      "step": 20665
    },
    {
      "entropy": 6.125257253646851,
      "epoch": 1.825326739667962,
      "grad_norm": 3.03125,
      "learning_rate": 0.00046674936282964326,
      "loss": 5.9029,
      "mean_token_accuracy": 0.16381777822971344,
      "num_tokens": 38532074.0,
      "step": 20670
    },
    {
      "entropy": 6.079831981658936,
      "epoch": 1.825768279759802,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004667328850809683,
      "loss": 5.7554,
      "mean_token_accuracy": 0.17308495789766312,
      "num_tokens": 38541423.0,
      "step": 20675
    },
    {
      "entropy": 6.087328100204468,
      "epoch": 1.8262098198516425,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004667164035763552,
      "loss": 5.8591,
      "mean_token_accuracy": 0.1685842603445053,
      "num_tokens": 38552369.0,
      "step": 20680
    },
    {
      "entropy": 6.130248165130615,
      "epoch": 1.8266513599434828,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004666999183161268,
      "loss": 5.8198,
      "mean_token_accuracy": 0.17114411145448685,
      "num_tokens": 38561310.0,
      "step": 20685
    },
    {
      "entropy": 6.1262308120727536,
      "epoch": 1.827092900035323,
      "grad_norm": 2.0,
      "learning_rate": 0.00046668342930060604,
      "loss": 5.8666,
      "mean_token_accuracy": 0.1684773564338684,
      "num_tokens": 38571157.0,
      "step": 20690
    },
    {
      "entropy": 6.1203779697418215,
      "epoch": 1.8275344401271636,
      "grad_norm": 3.546875,
      "learning_rate": 0.00046666693653011596,
      "loss": 5.7774,
      "mean_token_accuracy": 0.16994786411523818,
      "num_tokens": 38579862.0,
      "step": 20695
    },
    {
      "entropy": 6.093237686157226,
      "epoch": 1.8279759802190039,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046665044000497954,
      "loss": 5.854,
      "mean_token_accuracy": 0.17067529857158661,
      "num_tokens": 38589784.0,
      "step": 20700
    },
    {
      "entropy": 6.076339244842529,
      "epoch": 1.8284175203108441,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046663393972551994,
      "loss": 5.7652,
      "mean_token_accuracy": 0.1757814571261406,
      "num_tokens": 38598925.0,
      "step": 20705
    },
    {
      "entropy": 6.113490343093872,
      "epoch": 1.8288590604026846,
      "grad_norm": 3.25,
      "learning_rate": 0.00046661743569206056,
      "loss": 5.839,
      "mean_token_accuracy": 0.16599774956703187,
      "num_tokens": 38608245.0,
      "step": 20710
    },
    {
      "entropy": 6.091380786895752,
      "epoch": 1.829300600494525,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004666009279049245,
      "loss": 5.9046,
      "mean_token_accuracy": 0.16655689775943755,
      "num_tokens": 38617586.0,
      "step": 20715
    },
    {
      "entropy": 6.136275720596314,
      "epoch": 1.8297421405863652,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046658441636443526,
      "loss": 5.8328,
      "mean_token_accuracy": 0.17282410115003585,
      "num_tokens": 38626420.0,
      "step": 20720
    },
    {
      "entropy": 6.18027400970459,
      "epoch": 1.8301836806782057,
      "grad_norm": 4.3125,
      "learning_rate": 0.00046656790107091627,
      "loss": 5.7974,
      "mean_token_accuracy": 0.16962956935167312,
      "num_tokens": 38635585.0,
      "step": 20725
    },
    {
      "entropy": 6.092431974411011,
      "epoch": 1.830625220770046,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004665513820246911,
      "loss": 5.7937,
      "mean_token_accuracy": 0.17145867198705672,
      "num_tokens": 38643998.0,
      "step": 20730
    },
    {
      "entropy": 6.087373304367065,
      "epoch": 1.8310667608618862,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004665348592260833,
      "loss": 5.8177,
      "mean_token_accuracy": 0.17502590715885163,
      "num_tokens": 38653018.0,
      "step": 20735
    },
    {
      "entropy": 6.046820974349975,
      "epoch": 1.8315083009537267,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004665183326754165,
      "loss": 5.7578,
      "mean_token_accuracy": 0.17521387040615083,
      "num_tokens": 38662079.0,
      "step": 20740
    },
    {
      "entropy": 6.096765184402466,
      "epoch": 1.8319498410455668,
      "grad_norm": 1.875,
      "learning_rate": 0.0004665018023730144,
      "loss": 5.7872,
      "mean_token_accuracy": 0.16922145783901216,
      "num_tokens": 38669928.0,
      "step": 20745
    },
    {
      "entropy": 6.039730167388916,
      "epoch": 1.8323913811374073,
      "grad_norm": 2.203125,
      "learning_rate": 0.000466485268319201,
      "loss": 5.7652,
      "mean_token_accuracy": 0.1714009314775467,
      "num_tokens": 38679475.0,
      "step": 20750
    },
    {
      "entropy": 6.124115753173828,
      "epoch": 1.8328329212292476,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004664687305143001,
      "loss": 5.8562,
      "mean_token_accuracy": 0.16976838260889054,
      "num_tokens": 38689057.0,
      "step": 20755
    },
    {
      "entropy": 6.128188705444336,
      "epoch": 1.8332744613210878,
      "grad_norm": 3.84375,
      "learning_rate": 0.00046645218895863564,
      "loss": 5.6902,
      "mean_token_accuracy": 0.1861782670021057,
      "num_tokens": 38698362.0,
      "step": 20760
    },
    {
      "entropy": 5.9882972240448,
      "epoch": 1.8337160014129283,
      "grad_norm": 2.953125,
      "learning_rate": 0.00046643564365253173,
      "loss": 5.7333,
      "mean_token_accuracy": 0.17798884212970734,
      "num_tokens": 38708319.0,
      "step": 20765
    },
    {
      "entropy": 6.088434839248658,
      "epoch": 1.8341575415047686,
      "grad_norm": 2.5,
      "learning_rate": 0.0004664190945963124,
      "loss": 5.8361,
      "mean_token_accuracy": 0.1672394335269928,
      "num_tokens": 38717441.0,
      "step": 20770
    },
    {
      "entropy": 6.147144460678101,
      "epoch": 1.834599081596609,
      "grad_norm": 1.75,
      "learning_rate": 0.00046640254179030184,
      "loss": 5.814,
      "mean_token_accuracy": 0.16730164885520935,
      "num_tokens": 38726903.0,
      "step": 20775
    },
    {
      "entropy": 6.111432933807373,
      "epoch": 1.8350406216884494,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004663859852348244,
      "loss": 5.7813,
      "mean_token_accuracy": 0.16867226958274842,
      "num_tokens": 38735983.0,
      "step": 20780
    },
    {
      "entropy": 6.07939395904541,
      "epoch": 1.8354821617802897,
      "grad_norm": 3.046875,
      "learning_rate": 0.00046636942493020437,
      "loss": 5.8605,
      "mean_token_accuracy": 0.1720335826277733,
      "num_tokens": 38745017.0,
      "step": 20785
    },
    {
      "entropy": 5.994217777252198,
      "epoch": 1.83592370187213,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004663528608767661,
      "loss": 5.7239,
      "mean_token_accuracy": 0.17451438307762146,
      "num_tokens": 38754756.0,
      "step": 20790
    },
    {
      "entropy": 6.1012248516082765,
      "epoch": 1.8363652419639704,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004663362930748341,
      "loss": 5.7904,
      "mean_token_accuracy": 0.17205928266048431,
      "num_tokens": 38763695.0,
      "step": 20795
    },
    {
      "entropy": 6.135064554214478,
      "epoch": 1.8368067820558107,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046631972152473295,
      "loss": 5.8643,
      "mean_token_accuracy": 0.16970751136541368,
      "num_tokens": 38773637.0,
      "step": 20800
    },
    {
      "entropy": 6.019438791275024,
      "epoch": 1.837248322147651,
      "grad_norm": 4.15625,
      "learning_rate": 0.00046630314622678724,
      "loss": 5.6845,
      "mean_token_accuracy": 0.18394582867622375,
      "num_tokens": 38783004.0,
      "step": 20805
    },
    {
      "entropy": 6.018321561813354,
      "epoch": 1.8376898622394915,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004662865671813217,
      "loss": 5.7793,
      "mean_token_accuracy": 0.17583214342594147,
      "num_tokens": 38791342.0,
      "step": 20810
    },
    {
      "entropy": 6.037858247756958,
      "epoch": 1.8381314023313315,
      "grad_norm": 2.65625,
      "learning_rate": 0.00046626998438866114,
      "loss": 5.7695,
      "mean_token_accuracy": 0.17465388625860215,
      "num_tokens": 38801046.0,
      "step": 20815
    },
    {
      "entropy": 6.084261035919189,
      "epoch": 1.838572942423172,
      "grad_norm": 2.90625,
      "learning_rate": 0.00046625339784913035,
      "loss": 5.7901,
      "mean_token_accuracy": 0.17411142736673355,
      "num_tokens": 38810806.0,
      "step": 20820
    },
    {
      "entropy": 6.104199647903442,
      "epoch": 1.8390144825150123,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046623680756305427,
      "loss": 5.8281,
      "mean_token_accuracy": 0.1724385380744934,
      "num_tokens": 38819362.0,
      "step": 20825
    },
    {
      "entropy": 6.173845052719116,
      "epoch": 1.8394560226068526,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004662202135307579,
      "loss": 5.8472,
      "mean_token_accuracy": 0.1651548758149147,
      "num_tokens": 38829337.0,
      "step": 20830
    },
    {
      "entropy": 6.073967218399048,
      "epoch": 1.839897562698693,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004662036157525663,
      "loss": 5.8178,
      "mean_token_accuracy": 0.16689627468585969,
      "num_tokens": 38838351.0,
      "step": 20835
    },
    {
      "entropy": 6.098162889480591,
      "epoch": 1.8403391027905334,
      "grad_norm": 2.703125,
      "learning_rate": 0.00046618701422880456,
      "loss": 5.8472,
      "mean_token_accuracy": 0.16811490058898926,
      "num_tokens": 38847830.0,
      "step": 20840
    },
    {
      "entropy": 6.123013591766357,
      "epoch": 1.8407806428823736,
      "grad_norm": 3.734375,
      "learning_rate": 0.000466170408959798,
      "loss": 5.8424,
      "mean_token_accuracy": 0.15953526496887208,
      "num_tokens": 38856945.0,
      "step": 20845
    },
    {
      "entropy": 6.11520037651062,
      "epoch": 1.8412221829742141,
      "grad_norm": 3.375,
      "learning_rate": 0.0004661537999458718,
      "loss": 5.7916,
      "mean_token_accuracy": 0.17117039412260054,
      "num_tokens": 38866493.0,
      "step": 20850
    },
    {
      "entropy": 6.0629487991333,
      "epoch": 1.8416637230660544,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046613718718735143,
      "loss": 5.7262,
      "mean_token_accuracy": 0.17249458730220796,
      "num_tokens": 38875653.0,
      "step": 20855
    },
    {
      "entropy": 5.990917491912842,
      "epoch": 1.8421052631578947,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046612057068456226,
      "loss": 5.6526,
      "mean_token_accuracy": 0.1834740996360779,
      "num_tokens": 38884740.0,
      "step": 20860
    },
    {
      "entropy": 5.958756303787231,
      "epoch": 1.8425468032497352,
      "grad_norm": 3.421875,
      "learning_rate": 0.00046610395043782983,
      "loss": 5.7721,
      "mean_token_accuracy": 0.17412829548120498,
      "num_tokens": 38893801.0,
      "step": 20865
    },
    {
      "entropy": 6.053528738021851,
      "epoch": 1.8429883433415755,
      "grad_norm": 2.953125,
      "learning_rate": 0.00046608732644747963,
      "loss": 5.7963,
      "mean_token_accuracy": 0.17707792073488235,
      "num_tokens": 38902298.0,
      "step": 20870
    },
    {
      "entropy": 6.100785255432129,
      "epoch": 1.8434298834334157,
      "grad_norm": 2.9375,
      "learning_rate": 0.00046607069871383745,
      "loss": 5.7073,
      "mean_token_accuracy": 0.17980646044015886,
      "num_tokens": 38910904.0,
      "step": 20875
    },
    {
      "entropy": 6.062715005874634,
      "epoch": 1.8438714235252562,
      "grad_norm": 15.3125,
      "learning_rate": 0.0004660540672372289,
      "loss": 5.6948,
      "mean_token_accuracy": 0.1730342835187912,
      "num_tokens": 38920405.0,
      "step": 20880
    },
    {
      "entropy": 6.021054887771607,
      "epoch": 1.8443129636170963,
      "grad_norm": 3.75,
      "learning_rate": 0.0004660374320179798,
      "loss": 5.703,
      "mean_token_accuracy": 0.17356231212615966,
      "num_tokens": 38930118.0,
      "step": 20885
    },
    {
      "entropy": 6.044412136077881,
      "epoch": 1.8447545037089368,
      "grad_norm": 2.375,
      "learning_rate": 0.00046602079305641615,
      "loss": 5.8086,
      "mean_token_accuracy": 0.1706959933042526,
      "num_tokens": 38939388.0,
      "step": 20890
    },
    {
      "entropy": 6.085807132720947,
      "epoch": 1.845196043800777,
      "grad_norm": 1.921875,
      "learning_rate": 0.00046600415035286374,
      "loss": 5.8038,
      "mean_token_accuracy": 0.17296133786439896,
      "num_tokens": 38948632.0,
      "step": 20895
    },
    {
      "entropy": 6.138688707351685,
      "epoch": 1.8456375838926173,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004659875039076487,
      "loss": 5.818,
      "mean_token_accuracy": 0.16964947134256364,
      "num_tokens": 38957504.0,
      "step": 20900
    },
    {
      "entropy": 6.055169296264649,
      "epoch": 1.8460791239844578,
      "grad_norm": 2.625,
      "learning_rate": 0.000465970853721097,
      "loss": 5.7918,
      "mean_token_accuracy": 0.16929807364940644,
      "num_tokens": 38967713.0,
      "step": 20905
    },
    {
      "entropy": 6.082954692840576,
      "epoch": 1.8465206640762981,
      "grad_norm": 3.890625,
      "learning_rate": 0.00046595419979353494,
      "loss": 5.8213,
      "mean_token_accuracy": 0.17495082169771195,
      "num_tokens": 38976982.0,
      "step": 20910
    },
    {
      "entropy": 6.047014284133911,
      "epoch": 1.8469622041681384,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004659375421252887,
      "loss": 5.7705,
      "mean_token_accuracy": 0.17424721419811248,
      "num_tokens": 38985479.0,
      "step": 20915
    },
    {
      "entropy": 6.141044616699219,
      "epoch": 1.8474037442599789,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004659208807166846,
      "loss": 5.8564,
      "mean_token_accuracy": 0.15925310105085372,
      "num_tokens": 38995418.0,
      "step": 20920
    },
    {
      "entropy": 6.088508987426758,
      "epoch": 1.8478452843518192,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004659042155680491,
      "loss": 5.7684,
      "mean_token_accuracy": 0.17865591794252395,
      "num_tokens": 39004961.0,
      "step": 20925
    },
    {
      "entropy": 6.028351306915283,
      "epoch": 1.8482868244436594,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004658875466797085,
      "loss": 5.7763,
      "mean_token_accuracy": 0.17899756878614426,
      "num_tokens": 39014282.0,
      "step": 20930
    },
    {
      "entropy": 6.057820415496826,
      "epoch": 1.8487283645355,
      "grad_norm": 3.03125,
      "learning_rate": 0.00046587087405198943,
      "loss": 5.7935,
      "mean_token_accuracy": 0.16766920685768127,
      "num_tokens": 39024480.0,
      "step": 20935
    },
    {
      "entropy": 6.148885679244995,
      "epoch": 1.8491699046273402,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046585419768521854,
      "loss": 5.8199,
      "mean_token_accuracy": 0.16912710666656494,
      "num_tokens": 39034363.0,
      "step": 20940
    },
    {
      "entropy": 6.1271233558654785,
      "epoch": 1.8496114447191805,
      "grad_norm": 3.296875,
      "learning_rate": 0.00046583751757972244,
      "loss": 5.7356,
      "mean_token_accuracy": 0.17860287874937059,
      "num_tokens": 39043823.0,
      "step": 20945
    },
    {
      "entropy": 6.057376718521118,
      "epoch": 1.850052984811021,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004658208337358279,
      "loss": 5.7513,
      "mean_token_accuracy": 0.17544250339269638,
      "num_tokens": 39052219.0,
      "step": 20950
    },
    {
      "entropy": 6.046796989440918,
      "epoch": 1.850494524902861,
      "grad_norm": 2.875,
      "learning_rate": 0.00046580414615386175,
      "loss": 5.8671,
      "mean_token_accuracy": 0.16520621478557587,
      "num_tokens": 39062974.0,
      "step": 20955
    },
    {
      "entropy": 5.90196533203125,
      "epoch": 1.8509360649947015,
      "grad_norm": 2.125,
      "learning_rate": 0.00046578745483415087,
      "loss": 5.6073,
      "mean_token_accuracy": 0.19023892283439636,
      "num_tokens": 39072178.0,
      "step": 20960
    },
    {
      "entropy": 6.0652039527893065,
      "epoch": 1.8513776050865418,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004657707597770223,
      "loss": 5.8024,
      "mean_token_accuracy": 0.16778215318918227,
      "num_tokens": 39081494.0,
      "step": 20965
    },
    {
      "entropy": 6.066865015029907,
      "epoch": 1.851819145178382,
      "grad_norm": 1.8515625,
      "learning_rate": 0.000465754060982803,
      "loss": 5.7912,
      "mean_token_accuracy": 0.16780581325292587,
      "num_tokens": 39090783.0,
      "step": 20970
    },
    {
      "entropy": 6.092477560043335,
      "epoch": 1.8522606852702226,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004657373584518201,
      "loss": 5.8068,
      "mean_token_accuracy": 0.17250518053770064,
      "num_tokens": 39100150.0,
      "step": 20975
    },
    {
      "entropy": 6.111744070053101,
      "epoch": 1.8527022253620629,
      "grad_norm": 4.0,
      "learning_rate": 0.000465720652184401,
      "loss": 5.8543,
      "mean_token_accuracy": 0.16818137615919113,
      "num_tokens": 39109419.0,
      "step": 20980
    },
    {
      "entropy": 6.020004367828369,
      "epoch": 1.8531437654539031,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004657039421808726,
      "loss": 5.718,
      "mean_token_accuracy": 0.1791211560368538,
      "num_tokens": 39118312.0,
      "step": 20985
    },
    {
      "entropy": 6.1326988697052,
      "epoch": 1.8535853055457436,
      "grad_norm": 3.703125,
      "learning_rate": 0.00046568722844156254,
      "loss": 5.8628,
      "mean_token_accuracy": 0.1638197720050812,
      "num_tokens": 39128169.0,
      "step": 20990
    },
    {
      "entropy": 6.110426378250122,
      "epoch": 1.854026845637584,
      "grad_norm": 3.21875,
      "learning_rate": 0.00046567051096679807,
      "loss": 5.8164,
      "mean_token_accuracy": 0.16905437409877777,
      "num_tokens": 39136790.0,
      "step": 20995
    },
    {
      "entropy": 6.096529531478882,
      "epoch": 1.8544683857294242,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004656537897569068,
      "loss": 5.7338,
      "mean_token_accuracy": 0.17872632294893265,
      "num_tokens": 39146974.0,
      "step": 21000
    },
    {
      "epoch": 1.8544683857294242,
      "eval_entropy": 5.859363390284243,
      "eval_loss": 5.853883266448975,
      "eval_mean_token_accuracy": 0.1763305330626869,
      "eval_num_tokens": 39146974.0,
      "eval_runtime": 26.3705,
      "eval_samples_per_second": 1339.188,
      "eval_steps_per_second": 167.422,
      "step": 21000
    },
    {
      "entropy": 6.173826026916504,
      "epoch": 1.8549099258212647,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004656370648122161,
      "loss": 5.9402,
      "mean_token_accuracy": 0.1656165212392807,
      "num_tokens": 39156910.0,
      "step": 21005
    },
    {
      "entropy": 6.159449815750122,
      "epoch": 1.855351465913105,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046562033613305377,
      "loss": 5.8933,
      "mean_token_accuracy": 0.170497564971447,
      "num_tokens": 39165790.0,
      "step": 21010
    },
    {
      "entropy": 6.065485668182373,
      "epoch": 1.8557930060049452,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004656036037197475,
      "loss": 5.6869,
      "mean_token_accuracy": 0.18348097503185273,
      "num_tokens": 39174811.0,
      "step": 21015
    },
    {
      "entropy": 6.022029399871826,
      "epoch": 1.8562345460967857,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004655868675726251,
      "loss": 5.8169,
      "mean_token_accuracy": 0.1653737887740135,
      "num_tokens": 39184325.0,
      "step": 21020
    },
    {
      "entropy": 6.134298992156983,
      "epoch": 1.8566760861886258,
      "grad_norm": 1.921875,
      "learning_rate": 0.00046557012769201425,
      "loss": 5.7697,
      "mean_token_accuracy": 0.17411702126264572,
      "num_tokens": 39194879.0,
      "step": 21025
    },
    {
      "entropy": 6.133053207397461,
      "epoch": 1.8571176262804663,
      "grad_norm": 2.390625,
      "learning_rate": 0.000465553384078243,
      "loss": 5.7962,
      "mean_token_accuracy": 0.17355335205793382,
      "num_tokens": 39204702.0,
      "step": 21030
    },
    {
      "entropy": 6.101826524734497,
      "epoch": 1.8575591663723066,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004655366367316394,
      "loss": 5.8295,
      "mean_token_accuracy": 0.17261516451835632,
      "num_tokens": 39214561.0,
      "step": 21035
    },
    {
      "entropy": 6.092253065109253,
      "epoch": 1.8580007064641468,
      "grad_norm": 2.9375,
      "learning_rate": 0.00046551988565253136,
      "loss": 5.9101,
      "mean_token_accuracy": 0.16380758881568908,
      "num_tokens": 39224411.0,
      "step": 21040
    },
    {
      "entropy": 6.065380334854126,
      "epoch": 1.8584422465559873,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004655031308412471,
      "loss": 5.8233,
      "mean_token_accuracy": 0.16833099126815795,
      "num_tokens": 39233422.0,
      "step": 21045
    },
    {
      "entropy": 6.146573543548584,
      "epoch": 1.8588837866478276,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046548637229811495,
      "loss": 5.8548,
      "mean_token_accuracy": 0.16477475613355635,
      "num_tokens": 39243919.0,
      "step": 21050
    },
    {
      "entropy": 6.187794780731201,
      "epoch": 1.8593253267396679,
      "grad_norm": 3.078125,
      "learning_rate": 0.00046546961002346304,
      "loss": 5.8917,
      "mean_token_accuracy": 0.1640167400240898,
      "num_tokens": 39253682.0,
      "step": 21055
    },
    {
      "entropy": 6.0901734828948975,
      "epoch": 1.8597668668315084,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004654528440176198,
      "loss": 5.8664,
      "mean_token_accuracy": 0.1674874782562256,
      "num_tokens": 39263855.0,
      "step": 21060
    },
    {
      "entropy": 6.118631887435913,
      "epoch": 1.8602084069233487,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004654360742809137,
      "loss": 5.792,
      "mean_token_accuracy": 0.17478817105293273,
      "num_tokens": 39272292.0,
      "step": 21065
    },
    {
      "entropy": 6.120535755157471,
      "epoch": 1.860649947015189,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004654193008136732,
      "loss": 5.7636,
      "mean_token_accuracy": 0.17719945162534714,
      "num_tokens": 39281321.0,
      "step": 21070
    },
    {
      "entropy": 6.104314756393433,
      "epoch": 1.8610914871070294,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046540252361622683,
      "loss": 5.9113,
      "mean_token_accuracy": 0.1624233528971672,
      "num_tokens": 39292255.0,
      "step": 21075
    },
    {
      "entropy": 6.137627410888672,
      "epoch": 1.8615330271988697,
      "grad_norm": 4.34375,
      "learning_rate": 0.00046538574268890334,
      "loss": 5.8361,
      "mean_token_accuracy": 0.16941694915294647,
      "num_tokens": 39301841.0,
      "step": 21080
    },
    {
      "entropy": 6.128303289413452,
      "epoch": 1.86197456729071,
      "grad_norm": 2.703125,
      "learning_rate": 0.00046536895803203144,
      "loss": 5.808,
      "mean_token_accuracy": 0.1663319632411003,
      "num_tokens": 39311238.0,
      "step": 21085
    },
    {
      "entropy": 6.119568729400635,
      "epoch": 1.8624161073825505,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046535216964593997,
      "loss": 5.8307,
      "mean_token_accuracy": 0.17109025418758392,
      "num_tokens": 39319945.0,
      "step": 21090
    },
    {
      "entropy": 6.094813013076783,
      "epoch": 1.8628576474743905,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004653353775309576,
      "loss": 5.7839,
      "mean_token_accuracy": 0.17427863478660582,
      "num_tokens": 39329436.0,
      "step": 21095
    },
    {
      "entropy": 6.11808032989502,
      "epoch": 1.863299187566231,
      "grad_norm": 2.25,
      "learning_rate": 0.0004653185816874136,
      "loss": 5.8166,
      "mean_token_accuracy": 0.1741361677646637,
      "num_tokens": 39338447.0,
      "step": 21100
    },
    {
      "entropy": 5.993050050735474,
      "epoch": 1.8637407276580713,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004653017821156368,
      "loss": 5.724,
      "mean_token_accuracy": 0.18409451395273208,
      "num_tokens": 39346744.0,
      "step": 21105
    },
    {
      "entropy": 6.066872835159302,
      "epoch": 1.8641822677499116,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004652849788159563,
      "loss": 5.8538,
      "mean_token_accuracy": 0.1621278315782547,
      "num_tokens": 39357402.0,
      "step": 21110
    },
    {
      "entropy": 6.172119903564453,
      "epoch": 1.864623807841752,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046526817178870123,
      "loss": 5.8461,
      "mean_token_accuracy": 0.16537647992372512,
      "num_tokens": 39366902.0,
      "step": 21115
    },
    {
      "entropy": 6.175823497772217,
      "epoch": 1.8650653479335924,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046525136103420093,
      "loss": 5.8207,
      "mean_token_accuracy": 0.1709502562880516,
      "num_tokens": 39376330.0,
      "step": 21120
    },
    {
      "entropy": 6.041230726242065,
      "epoch": 1.8655068880254326,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046523454655278466,
      "loss": 5.7763,
      "mean_token_accuracy": 0.18190991133451462,
      "num_tokens": 39385926.0,
      "step": 21125
    },
    {
      "entropy": 6.115786695480347,
      "epoch": 1.8659484281172731,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004652177283447818,
      "loss": 5.7504,
      "mean_token_accuracy": 0.16965205371379852,
      "num_tokens": 39395558.0,
      "step": 21130
    },
    {
      "entropy": 6.1176833152771,
      "epoch": 1.8663899682091134,
      "grad_norm": 2.640625,
      "learning_rate": 0.00046520090641052187,
      "loss": 5.8296,
      "mean_token_accuracy": 0.17404235750436783,
      "num_tokens": 39405072.0,
      "step": 21135
    },
    {
      "entropy": 6.151421928405762,
      "epoch": 1.8668315083009537,
      "grad_norm": 2.78125,
      "learning_rate": 0.00046518408075033437,
      "loss": 5.929,
      "mean_token_accuracy": 0.16529123038053511,
      "num_tokens": 39414714.0,
      "step": 21140
    },
    {
      "entropy": 6.118004894256591,
      "epoch": 1.8672730483927942,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004651672513645488,
      "loss": 5.7188,
      "mean_token_accuracy": 0.17483951151371002,
      "num_tokens": 39423189.0,
      "step": 21145
    },
    {
      "entropy": 6.096082305908203,
      "epoch": 1.8677145884846345,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046515041825349505,
      "loss": 5.8817,
      "mean_token_accuracy": 0.16405032873153685,
      "num_tokens": 39432791.0,
      "step": 21150
    },
    {
      "entropy": 6.145186424255371,
      "epoch": 1.8681561285764747,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004651335814175027,
      "loss": 5.8141,
      "mean_token_accuracy": 0.17019782215356827,
      "num_tokens": 39442849.0,
      "step": 21155
    },
    {
      "entropy": 6.083635234832764,
      "epoch": 1.8685976686683152,
      "grad_norm": 3.078125,
      "learning_rate": 0.00046511674085690155,
      "loss": 5.8284,
      "mean_token_accuracy": 0.16883940100669861,
      "num_tokens": 39452171.0,
      "step": 21160
    },
    {
      "entropy": 6.089730072021484,
      "epoch": 1.8690392087601553,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004650998965720216,
      "loss": 5.7972,
      "mean_token_accuracy": 0.1659482091665268,
      "num_tokens": 39460351.0,
      "step": 21165
    },
    {
      "entropy": 6.184094953536987,
      "epoch": 1.8694807488519958,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046508304856319283,
      "loss": 5.8773,
      "mean_token_accuracy": 0.1659751370549202,
      "num_tokens": 39470050.0,
      "step": 21170
    },
    {
      "entropy": 6.106051635742188,
      "epoch": 1.869922288943836,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046506619683074526,
      "loss": 5.8191,
      "mean_token_accuracy": 0.17182818353176116,
      "num_tokens": 39480192.0,
      "step": 21175
    },
    {
      "entropy": 6.092202186584473,
      "epoch": 1.8703638290356763,
      "grad_norm": 4.1875,
      "learning_rate": 0.00046504934137500903,
      "loss": 5.8579,
      "mean_token_accuracy": 0.16855140626430512,
      "num_tokens": 39490118.0,
      "step": 21180
    },
    {
      "entropy": 6.124629068374634,
      "epoch": 1.8708053691275168,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046503248219631417,
      "loss": 5.8477,
      "mean_token_accuracy": 0.16532353758811952,
      "num_tokens": 39499173.0,
      "step": 21185
    },
    {
      "entropy": 6.040664339065552,
      "epoch": 1.871246909219357,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046501561929499115,
      "loss": 5.7225,
      "mean_token_accuracy": 0.1788926750421524,
      "num_tokens": 39508595.0,
      "step": 21190
    },
    {
      "entropy": 6.059040117263794,
      "epoch": 1.8716884493111974,
      "grad_norm": 3.4375,
      "learning_rate": 0.00046499875267137016,
      "loss": 5.7963,
      "mean_token_accuracy": 0.17927175611257554,
      "num_tokens": 39519300.0,
      "step": 21195
    },
    {
      "entropy": 6.128713083267212,
      "epoch": 1.8721299894030379,
      "grad_norm": 1.625,
      "learning_rate": 0.0004649818823257817,
      "loss": 5.8352,
      "mean_token_accuracy": 0.16680102795362473,
      "num_tokens": 39527722.0,
      "step": 21200
    },
    {
      "entropy": 6.157823276519776,
      "epoch": 1.8725715294948782,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004649650082585562,
      "loss": 5.8099,
      "mean_token_accuracy": 0.17177072316408157,
      "num_tokens": 39537063.0,
      "step": 21205
    },
    {
      "entropy": 6.120093441009521,
      "epoch": 1.8730130695867184,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046494813047002424,
      "loss": 5.826,
      "mean_token_accuracy": 0.17201895862817765,
      "num_tokens": 39546402.0,
      "step": 21210
    },
    {
      "entropy": 6.140852308273315,
      "epoch": 1.873454609678559,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046493124896051634,
      "loss": 5.8199,
      "mean_token_accuracy": 0.17111910730600358,
      "num_tokens": 39555877.0,
      "step": 21215
    },
    {
      "entropy": 6.062350559234619,
      "epoch": 1.8738961497703992,
      "grad_norm": 2.78125,
      "learning_rate": 0.00046491436373036346,
      "loss": 5.7812,
      "mean_token_accuracy": 0.17037922143936157,
      "num_tokens": 39565153.0,
      "step": 21220
    },
    {
      "entropy": 6.094492197036743,
      "epoch": 1.8743376898622395,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004648974747798961,
      "loss": 5.8007,
      "mean_token_accuracy": 0.16861672699451447,
      "num_tokens": 39573775.0,
      "step": 21225
    },
    {
      "entropy": 6.101448917388916,
      "epoch": 1.87477922995408,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004648805821094452,
      "loss": 5.8802,
      "mean_token_accuracy": 0.16456809043884277,
      "num_tokens": 39583159.0,
      "step": 21230
    },
    {
      "entropy": 6.196863746643066,
      "epoch": 1.87522077004592,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046486368571934175,
      "loss": 5.899,
      "mean_token_accuracy": 0.16256113946437836,
      "num_tokens": 39592545.0,
      "step": 21235
    },
    {
      "entropy": 5.974124956130981,
      "epoch": 1.8756623101377605,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004648467856099167,
      "loss": 5.6366,
      "mean_token_accuracy": 0.18383304625749589,
      "num_tokens": 39602059.0,
      "step": 21240
    },
    {
      "entropy": 6.059417963027954,
      "epoch": 1.8761038502296008,
      "grad_norm": 2.90625,
      "learning_rate": 0.000464829881781501,
      "loss": 5.7918,
      "mean_token_accuracy": 0.16888861656188964,
      "num_tokens": 39610777.0,
      "step": 21245
    },
    {
      "entropy": 6.139316844940185,
      "epoch": 1.876545390321441,
      "grad_norm": 2.546875,
      "learning_rate": 0.00046481297423442593,
      "loss": 5.8688,
      "mean_token_accuracy": 0.17126096040010452,
      "num_tokens": 39620782.0,
      "step": 21250
    },
    {
      "entropy": 6.213293266296387,
      "epoch": 1.8769869304132816,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004647960629690226,
      "loss": 5.9224,
      "mean_token_accuracy": 0.1583952412009239,
      "num_tokens": 39631340.0,
      "step": 21255
    },
    {
      "entropy": 6.012655305862427,
      "epoch": 1.8774284705051218,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004647791479856223,
      "loss": 5.7345,
      "mean_token_accuracy": 0.1777779459953308,
      "num_tokens": 39640361.0,
      "step": 21260
    },
    {
      "entropy": 6.088248300552368,
      "epoch": 1.8778700105969621,
      "grad_norm": 1.75,
      "learning_rate": 0.0004647622292845566,
      "loss": 5.7969,
      "mean_token_accuracy": 0.1724010169506073,
      "num_tokens": 39649726.0,
      "step": 21265
    },
    {
      "entropy": 6.13547191619873,
      "epoch": 1.8783115506888026,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004647453068661566,
      "loss": 5.8673,
      "mean_token_accuracy": 0.16787385791540146,
      "num_tokens": 39660226.0,
      "step": 21270
    },
    {
      "entropy": 6.082841253280639,
      "epoch": 1.878753090780643,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046472838073075397,
      "loss": 5.6429,
      "mean_token_accuracy": 0.17697591930627823,
      "num_tokens": 39668731.0,
      "step": 21275
    },
    {
      "entropy": 6.0610991477966305,
      "epoch": 1.8791946308724832,
      "grad_norm": 1.5625,
      "learning_rate": 0.00046471145087868034,
      "loss": 5.8109,
      "mean_token_accuracy": 0.17789752781391144,
      "num_tokens": 39678465.0,
      "step": 21280
    },
    {
      "entropy": 6.045650959014893,
      "epoch": 1.8796361709643237,
      "grad_norm": 2.125,
      "learning_rate": 0.00046469451731026725,
      "loss": 5.6637,
      "mean_token_accuracy": 0.18099646419286727,
      "num_tokens": 39687057.0,
      "step": 21285
    },
    {
      "entropy": 6.032855367660522,
      "epoch": 1.880077711056164,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004646775800258464,
      "loss": 5.8215,
      "mean_token_accuracy": 0.17431067377328874,
      "num_tokens": 39696387.0,
      "step": 21290
    },
    {
      "entropy": 6.089390754699707,
      "epoch": 1.8805192511480042,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046466063902574973,
      "loss": 5.7764,
      "mean_token_accuracy": 0.17333888709545137,
      "num_tokens": 39705756.0,
      "step": 21295
    },
    {
      "entropy": 6.113208436965943,
      "epoch": 1.8809607912398447,
      "grad_norm": 2.34375,
      "learning_rate": 0.00046464369431030894,
      "loss": 5.7723,
      "mean_token_accuracy": 0.17512379735708236,
      "num_tokens": 39713916.0,
      "step": 21300
    },
    {
      "entropy": 6.053884649276734,
      "epoch": 1.8814023313316848,
      "grad_norm": 1.65625,
      "learning_rate": 0.000464626745879856,
      "loss": 5.8987,
      "mean_token_accuracy": 0.1651058942079544,
      "num_tokens": 39722920.0,
      "step": 21305
    },
    {
      "entropy": 6.1752533435821535,
      "epoch": 1.8818438714235253,
      "grad_norm": 3.46875,
      "learning_rate": 0.000464609793734723,
      "loss": 5.87,
      "mean_token_accuracy": 0.1687443032860756,
      "num_tokens": 39732217.0,
      "step": 21310
    },
    {
      "entropy": 6.090468645095825,
      "epoch": 1.8822854115153655,
      "grad_norm": 1.859375,
      "learning_rate": 0.000464592837875242,
      "loss": 5.7088,
      "mean_token_accuracy": 0.1755327343940735,
      "num_tokens": 39741636.0,
      "step": 21315
    },
    {
      "entropy": 6.0907868385314945,
      "epoch": 1.8827269516072058,
      "grad_norm": 2.5,
      "learning_rate": 0.0004645758783017451,
      "loss": 5.826,
      "mean_token_accuracy": 0.17212406545877457,
      "num_tokens": 39751457.0,
      "step": 21320
    },
    {
      "entropy": 6.1437750339508055,
      "epoch": 1.8831684916990463,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00046455891501456456,
      "loss": 5.7877,
      "mean_token_accuracy": 0.1737821415066719,
      "num_tokens": 39760308.0,
      "step": 21325
    },
    {
      "entropy": 6.005112075805664,
      "epoch": 1.8836100317908866,
      "grad_norm": 4.25,
      "learning_rate": 0.00046454194801403274,
      "loss": 5.7298,
      "mean_token_accuracy": 0.17937238663434982,
      "num_tokens": 39770002.0,
      "step": 21330
    },
    {
      "entropy": 5.968388223648072,
      "epoch": 1.8840515718827269,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004645249773004819,
      "loss": 5.7541,
      "mean_token_accuracy": 0.17092153429985046,
      "num_tokens": 39779043.0,
      "step": 21335
    },
    {
      "entropy": 6.147596168518066,
      "epoch": 1.8844931119745674,
      "grad_norm": 2.484375,
      "learning_rate": 0.00046450800287424455,
      "loss": 5.8119,
      "mean_token_accuracy": 0.1653391033411026,
      "num_tokens": 39788545.0,
      "step": 21340
    },
    {
      "entropy": 6.0730071544647215,
      "epoch": 1.8849346520664076,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004644910247356532,
      "loss": 5.7147,
      "mean_token_accuracy": 0.1835902139544487,
      "num_tokens": 39798063.0,
      "step": 21345
    },
    {
      "entropy": 6.0455574035644535,
      "epoch": 1.885376192158248,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004644740428850404,
      "loss": 5.7422,
      "mean_token_accuracy": 0.17773073464632033,
      "num_tokens": 39808622.0,
      "step": 21350
    },
    {
      "entropy": 6.125068664550781,
      "epoch": 1.8858177322500884,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004644570573227389,
      "loss": 5.8575,
      "mean_token_accuracy": 0.1657984584569931,
      "num_tokens": 39817608.0,
      "step": 21355
    },
    {
      "entropy": 6.081325244903565,
      "epoch": 1.8862592723419287,
      "grad_norm": 1.5625,
      "learning_rate": 0.00046444006804908144,
      "loss": 5.6791,
      "mean_token_accuracy": 0.1855430632829666,
      "num_tokens": 39826055.0,
      "step": 21360
    },
    {
      "entropy": 5.96155309677124,
      "epoch": 1.886700812433769,
      "grad_norm": 1.75,
      "learning_rate": 0.0004644230750644007,
      "loss": 5.6803,
      "mean_token_accuracy": 0.1834298327565193,
      "num_tokens": 39835490.0,
      "step": 21365
    },
    {
      "entropy": 6.103626680374146,
      "epoch": 1.8871423525256095,
      "grad_norm": 2.59375,
      "learning_rate": 0.00046440607836902974,
      "loss": 5.8068,
      "mean_token_accuracy": 0.17698344886302947,
      "num_tokens": 39845613.0,
      "step": 21370
    },
    {
      "entropy": 6.114264488220215,
      "epoch": 1.8875838926174495,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046438907796330145,
      "loss": 5.7866,
      "mean_token_accuracy": 0.17371232360601424,
      "num_tokens": 39854495.0,
      "step": 21375
    },
    {
      "entropy": 6.025073575973511,
      "epoch": 1.88802543270929,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046437207384754874,
      "loss": 5.8205,
      "mean_token_accuracy": 0.17138807624578475,
      "num_tokens": 39864528.0,
      "step": 21380
    },
    {
      "entropy": 6.139890909194946,
      "epoch": 1.8884669728011303,
      "grad_norm": 3.125,
      "learning_rate": 0.0004643550660221048,
      "loss": 5.8684,
      "mean_token_accuracy": 0.16607652306556703,
      "num_tokens": 39873780.0,
      "step": 21385
    },
    {
      "entropy": 6.0725260257720945,
      "epoch": 1.8889085128929706,
      "grad_norm": 2.65625,
      "learning_rate": 0.00046433805448730287,
      "loss": 5.6669,
      "mean_token_accuracy": 0.17712335884571076,
      "num_tokens": 39882529.0,
      "step": 21390
    },
    {
      "entropy": 6.143491172790528,
      "epoch": 1.889350052984811,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004643210392434761,
      "loss": 5.7944,
      "mean_token_accuracy": 0.1723927676677704,
      "num_tokens": 39890688.0,
      "step": 21395
    },
    {
      "entropy": 6.068182134628296,
      "epoch": 1.8897915930766513,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046430402029095784,
      "loss": 5.8538,
      "mean_token_accuracy": 0.16682320535182954,
      "num_tokens": 39900849.0,
      "step": 21400
    },
    {
      "entropy": 6.002923059463501,
      "epoch": 1.8902331331684916,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004642869976300815,
      "loss": 5.8222,
      "mean_token_accuracy": 0.16972097605466843,
      "num_tokens": 39911297.0,
      "step": 21405
    },
    {
      "entropy": 6.055374383926392,
      "epoch": 1.8906746732603321,
      "grad_norm": 2.125,
      "learning_rate": 0.0004642699712611806,
      "loss": 5.767,
      "mean_token_accuracy": 0.17551086246967315,
      "num_tokens": 39920312.0,
      "step": 21410
    },
    {
      "entropy": 6.070961046218872,
      "epoch": 1.8911162133521724,
      "grad_norm": 6.21875,
      "learning_rate": 0.0004642529411845885,
      "loss": 5.835,
      "mean_token_accuracy": 0.1677388593554497,
      "num_tokens": 39930128.0,
      "step": 21415
    },
    {
      "entropy": 6.178530311584472,
      "epoch": 1.8915577534440127,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004642359074006389,
      "loss": 5.7797,
      "mean_token_accuracy": 0.17219416946172714,
      "num_tokens": 39939177.0,
      "step": 21420
    },
    {
      "entropy": 6.161017751693725,
      "epoch": 1.8919992935358532,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004642188699096656,
      "loss": 5.8798,
      "mean_token_accuracy": 0.1635308414697647,
      "num_tokens": 39948537.0,
      "step": 21425
    },
    {
      "entropy": 6.080967140197754,
      "epoch": 1.8924408336276934,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004642018287120022,
      "loss": 5.7876,
      "mean_token_accuracy": 0.16851073056459426,
      "num_tokens": 39958623.0,
      "step": 21430
    },
    {
      "entropy": 6.154923295974731,
      "epoch": 1.8928823737195337,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046418478380798263,
      "loss": 5.8915,
      "mean_token_accuracy": 0.170401693880558,
      "num_tokens": 39968736.0,
      "step": 21435
    },
    {
      "entropy": 6.002540493011475,
      "epoch": 1.8933239138113742,
      "grad_norm": 2.5,
      "learning_rate": 0.0004641677351979407,
      "loss": 5.7215,
      "mean_token_accuracy": 0.1756562501192093,
      "num_tokens": 39977600.0,
      "step": 21440
    },
    {
      "entropy": 6.098312425613403,
      "epoch": 1.8937654539032143,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004641506828822104,
      "loss": 5.8398,
      "mean_token_accuracy": 0.16205616742372514,
      "num_tokens": 39987808.0,
      "step": 21445
    },
    {
      "entropy": 6.167420768737793,
      "epoch": 1.8942069939950548,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004641336268611258,
      "loss": 5.9714,
      "mean_token_accuracy": 0.16621507406234742,
      "num_tokens": 39998553.0,
      "step": 21450
    },
    {
      "entropy": 6.233862209320068,
      "epoch": 1.894648534086895,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00046411656713502106,
      "loss": 5.8681,
      "mean_token_accuracy": 0.16706406325101852,
      "num_tokens": 40007529.0,
      "step": 21455
    },
    {
      "entropy": 6.132805395126343,
      "epoch": 1.8950900741787353,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004640995037042303,
      "loss": 5.8564,
      "mean_token_accuracy": 0.16705780178308488,
      "num_tokens": 40017116.0,
      "step": 21460
    },
    {
      "entropy": 6.12165093421936,
      "epoch": 1.8955316142705758,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004640824365690878,
      "loss": 5.8738,
      "mean_token_accuracy": 0.16973905861377717,
      "num_tokens": 40026601.0,
      "step": 21465
    },
    {
      "entropy": 6.082113647460938,
      "epoch": 1.895973154362416,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046406536572992795,
      "loss": 5.7792,
      "mean_token_accuracy": 0.17967993319034575,
      "num_tokens": 40036029.0,
      "step": 21470
    },
    {
      "entropy": 6.167379474639892,
      "epoch": 1.8964146944542564,
      "grad_norm": 1.8515625,
      "learning_rate": 0.000464048291187085,
      "loss": 5.8221,
      "mean_token_accuracy": 0.16262540370225906,
      "num_tokens": 40045892.0,
      "step": 21475
    },
    {
      "entropy": 6.0997960567474365,
      "epoch": 1.8968562345460969,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004640312129408936,
      "loss": 5.7636,
      "mean_token_accuracy": 0.17702283710241318,
      "num_tokens": 40055411.0,
      "step": 21480
    },
    {
      "entropy": 6.087768459320069,
      "epoch": 1.8972977746379371,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004640141309916883,
      "loss": 5.836,
      "mean_token_accuracy": 0.17093580812215806,
      "num_tokens": 40063988.0,
      "step": 21485
    },
    {
      "entropy": 6.05920033454895,
      "epoch": 1.8977393147297774,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004639970453398037,
      "loss": 5.7833,
      "mean_token_accuracy": 0.1750456288456917,
      "num_tokens": 40072671.0,
      "step": 21490
    },
    {
      "entropy": 6.098271131515503,
      "epoch": 1.898180854821618,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00046397995598557434,
      "loss": 5.883,
      "mean_token_accuracy": 0.17087428122758866,
      "num_tokens": 40082191.0,
      "step": 21495
    },
    {
      "entropy": 6.087645626068115,
      "epoch": 1.8986223949134582,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004639628629293352,
      "loss": 5.8426,
      "mean_token_accuracy": 0.1724378541111946,
      "num_tokens": 40090823.0,
      "step": 21500
    },
    {
      "entropy": 6.13563814163208,
      "epoch": 1.8990639350052985,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000463945766171421,
      "loss": 5.8153,
      "mean_token_accuracy": 0.16313320398330688,
      "num_tokens": 40099825.0,
      "step": 21505
    },
    {
      "entropy": 6.177030372619629,
      "epoch": 1.899505475097139,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004639286657121668,
      "loss": 5.8082,
      "mean_token_accuracy": 0.17420322746038436,
      "num_tokens": 40109589.0,
      "step": 21510
    },
    {
      "entropy": 6.079631328582764,
      "epoch": 1.899947015188979,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004639115615519074,
      "loss": 5.7682,
      "mean_token_accuracy": 0.17216488122940063,
      "num_tokens": 40119194.0,
      "step": 21515
    },
    {
      "entropy": 6.123204088211059,
      "epoch": 1.9003885552808195,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000463894453690978,
      "loss": 5.8273,
      "mean_token_accuracy": 0.16813366562128068,
      "num_tokens": 40128184.0,
      "step": 21520
    },
    {
      "entropy": 6.089981412887573,
      "epoch": 1.9008300953726598,
      "grad_norm": 2.578125,
      "learning_rate": 0.00046387734212971363,
      "loss": 5.8406,
      "mean_token_accuracy": 0.16549814790487288,
      "num_tokens": 40137842.0,
      "step": 21525
    },
    {
      "entropy": 6.092873287200928,
      "epoch": 1.9012716354645,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004638602268684496,
      "loss": 5.8174,
      "mean_token_accuracy": 0.16807781904935837,
      "num_tokens": 40147151.0,
      "step": 21530
    },
    {
      "entropy": 6.120345258712769,
      "epoch": 1.9017131755563406,
      "grad_norm": 3.078125,
      "learning_rate": 0.00046384310790752105,
      "loss": 5.7688,
      "mean_token_accuracy": 0.17126996964216232,
      "num_tokens": 40157170.0,
      "step": 21535
    },
    {
      "entropy": 6.079241466522217,
      "epoch": 1.9021547156481808,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004638259852472635,
      "loss": 5.7855,
      "mean_token_accuracy": 0.1681679055094719,
      "num_tokens": 40166787.0,
      "step": 21540
    },
    {
      "entropy": 6.102788209915161,
      "epoch": 1.9025962557400211,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004638088588880122,
      "loss": 5.756,
      "mean_token_accuracy": 0.18017373085021973,
      "num_tokens": 40176640.0,
      "step": 21545
    },
    {
      "entropy": 6.061911153793335,
      "epoch": 1.9030377958318616,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004637917288301029,
      "loss": 5.8104,
      "mean_token_accuracy": 0.17292430251836777,
      "num_tokens": 40185732.0,
      "step": 21550
    },
    {
      "entropy": 6.089268445968628,
      "epoch": 1.9034793359237019,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004637745950738709,
      "loss": 5.7381,
      "mean_token_accuracy": 0.17483177781105042,
      "num_tokens": 40194798.0,
      "step": 21555
    },
    {
      "entropy": 6.131200695037842,
      "epoch": 1.9039208760155422,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004637574576196519,
      "loss": 5.9048,
      "mean_token_accuracy": 0.16370820254087448,
      "num_tokens": 40204507.0,
      "step": 21560
    },
    {
      "entropy": 6.160027074813843,
      "epoch": 1.9043624161073827,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004637403164677817,
      "loss": 5.8243,
      "mean_token_accuracy": 0.16611558347940444,
      "num_tokens": 40213546.0,
      "step": 21565
    },
    {
      "entropy": 6.083746528625488,
      "epoch": 1.904803956199223,
      "grad_norm": 2.609375,
      "learning_rate": 0.00046372317161859605,
      "loss": 5.77,
      "mean_token_accuracy": 0.16845943331718444,
      "num_tokens": 40222571.0,
      "step": 21570
    },
    {
      "entropy": 6.111548376083374,
      "epoch": 1.9052454962910632,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046370602307243083,
      "loss": 5.8572,
      "mean_token_accuracy": 0.16383468061685563,
      "num_tokens": 40231803.0,
      "step": 21575
    },
    {
      "entropy": 6.124693584442139,
      "epoch": 1.9056870363829037,
      "grad_norm": 2.125,
      "learning_rate": 0.00046368887082962187,
      "loss": 5.8946,
      "mean_token_accuracy": 0.1644723355770111,
      "num_tokens": 40241755.0,
      "step": 21580
    },
    {
      "entropy": 6.108580732345581,
      "epoch": 1.9061285764747438,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004636717148905053,
      "loss": 5.7653,
      "mean_token_accuracy": 0.17084273248910903,
      "num_tokens": 40250963.0,
      "step": 21585
    },
    {
      "entropy": 6.154359245300293,
      "epoch": 1.9065701165665843,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004636545552554171,
      "loss": 5.7996,
      "mean_token_accuracy": 0.1683279663324356,
      "num_tokens": 40260626.0,
      "step": 21590
    },
    {
      "entropy": 6.059853506088257,
      "epoch": 1.9070116566584245,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004636373919246935,
      "loss": 5.7564,
      "mean_token_accuracy": 0.17302207201719283,
      "num_tokens": 40269852.0,
      "step": 21595
    },
    {
      "entropy": 6.031315517425537,
      "epoch": 1.9074531967502648,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004636202248986707,
      "loss": 5.7922,
      "mean_token_accuracy": 0.1703391894698143,
      "num_tokens": 40279499.0,
      "step": 21600
    },
    {
      "entropy": 6.149195194244385,
      "epoch": 1.9078947368421053,
      "grad_norm": 1.828125,
      "learning_rate": 0.00046360305417768484,
      "loss": 5.8732,
      "mean_token_accuracy": 0.1654176890850067,
      "num_tokens": 40288131.0,
      "step": 21605
    },
    {
      "entropy": 6.13309416770935,
      "epoch": 1.9083362769339456,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004635858797620725,
      "loss": 5.8258,
      "mean_token_accuracy": 0.16516976356506347,
      "num_tokens": 40296951.0,
      "step": 21610
    },
    {
      "entropy": 6.118100881576538,
      "epoch": 1.9087778170257859,
      "grad_norm": 1.984375,
      "learning_rate": 0.00046356870165217004,
      "loss": 5.7653,
      "mean_token_accuracy": 0.17338330149650574,
      "num_tokens": 40306250.0,
      "step": 21615
    },
    {
      "entropy": 6.077229261398315,
      "epoch": 1.9092193571176264,
      "grad_norm": 2.625,
      "learning_rate": 0.00046355151984831395,
      "loss": 5.7165,
      "mean_token_accuracy": 0.1791114017367363,
      "num_tokens": 40314561.0,
      "step": 21620
    },
    {
      "entropy": 5.976869535446167,
      "epoch": 1.9096608972094666,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046353433435084083,
      "loss": 5.7184,
      "mean_token_accuracy": 0.17474828362464906,
      "num_tokens": 40323580.0,
      "step": 21625
    },
    {
      "entropy": 5.992854738235474,
      "epoch": 1.910102437301307,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004635171451600872,
      "loss": 5.8032,
      "mean_token_accuracy": 0.17070396542549132,
      "num_tokens": 40333747.0,
      "step": 21630
    },
    {
      "entropy": 6.185562562942505,
      "epoch": 1.9105439773931474,
      "grad_norm": 3.984375,
      "learning_rate": 0.00046349995227639015,
      "loss": 5.8778,
      "mean_token_accuracy": 0.17154461443424224,
      "num_tokens": 40343435.0,
      "step": 21635
    },
    {
      "entropy": 6.207742166519165,
      "epoch": 1.9109855174849877,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004634827557000861,
      "loss": 5.8818,
      "mean_token_accuracy": 0.16733755320310592,
      "num_tokens": 40353282.0,
      "step": 21640
    },
    {
      "entropy": 6.143401288986206,
      "epoch": 1.911427057576828,
      "grad_norm": 3.15625,
      "learning_rate": 0.000463465555431512,
      "loss": 5.8813,
      "mean_token_accuracy": 0.16569918394088745,
      "num_tokens": 40363781.0,
      "step": 21645
    },
    {
      "entropy": 6.09151258468628,
      "epoch": 1.9118685976686685,
      "grad_norm": 2.765625,
      "learning_rate": 0.00046344835147100495,
      "loss": 5.7854,
      "mean_token_accuracy": 0.17065882682800293,
      "num_tokens": 40373240.0,
      "step": 21650
    },
    {
      "entropy": 6.093291282653809,
      "epoch": 1.9123101377605085,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00046343114381890187,
      "loss": 5.87,
      "mean_token_accuracy": 0.16901895403862,
      "num_tokens": 40383088.0,
      "step": 21655
    },
    {
      "entropy": 6.155415010452271,
      "epoch": 1.912751677852349,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004634139324755398,
      "loss": 5.7965,
      "mean_token_accuracy": 0.17586801052093506,
      "num_tokens": 40391654.0,
      "step": 21660
    },
    {
      "entropy": 6.070003843307495,
      "epoch": 1.9131932179441893,
      "grad_norm": 2.421875,
      "learning_rate": 0.000463396717441256,
      "loss": 5.7125,
      "mean_token_accuracy": 0.18071299493312837,
      "num_tokens": 40400616.0,
      "step": 21665
    },
    {
      "entropy": 6.022754859924317,
      "epoch": 1.9136347580360296,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004633794987163876,
      "loss": 5.8156,
      "mean_token_accuracy": 0.16231585294008255,
      "num_tokens": 40411271.0,
      "step": 21670
    },
    {
      "entropy": 6.0972943782806395,
      "epoch": 1.91407629812787,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046336227630127205,
      "loss": 5.7119,
      "mean_token_accuracy": 0.17011647969484328,
      "num_tokens": 40420019.0,
      "step": 21675
    },
    {
      "entropy": 6.11604118347168,
      "epoch": 1.9145178382197103,
      "grad_norm": 2.984375,
      "learning_rate": 0.00046334505019624656,
      "loss": 5.8104,
      "mean_token_accuracy": 0.17163215279579164,
      "num_tokens": 40429858.0,
      "step": 21680
    },
    {
      "entropy": 6.0243720531463625,
      "epoch": 1.9149593783115506,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00046332782040164867,
      "loss": 5.7768,
      "mean_token_accuracy": 0.17416307032108308,
      "num_tokens": 40439566.0,
      "step": 21685
    },
    {
      "entropy": 6.009717273712158,
      "epoch": 1.915400918403391,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004633105869178159,
      "loss": 5.7323,
      "mean_token_accuracy": 0.17883324921131133,
      "num_tokens": 40448527.0,
      "step": 21690
    },
    {
      "entropy": 6.103447723388672,
      "epoch": 1.9158424584952314,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004632933497450858,
      "loss": 5.7567,
      "mean_token_accuracy": 0.17720151394605638,
      "num_tokens": 40458076.0,
      "step": 21695
    },
    {
      "entropy": 6.083323001861572,
      "epoch": 1.9162839985870717,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004632761088837961,
      "loss": 5.6816,
      "mean_token_accuracy": 0.18921045064926148,
      "num_tokens": 40467639.0,
      "step": 21700
    },
    {
      "entropy": 6.076544666290284,
      "epoch": 1.9167255386789122,
      "grad_norm": 2.265625,
      "learning_rate": 0.00046325886433428446,
      "loss": 5.7704,
      "mean_token_accuracy": 0.17501137405633926,
      "num_tokens": 40476243.0,
      "step": 21705
    },
    {
      "entropy": 6.075497341156006,
      "epoch": 1.9171670787707524,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046324161609688876,
      "loss": 5.8256,
      "mean_token_accuracy": 0.17157698422670364,
      "num_tokens": 40485789.0,
      "step": 21710
    },
    {
      "entropy": 6.057292318344116,
      "epoch": 1.9176086188625927,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004632243641719469,
      "loss": 5.7421,
      "mean_token_accuracy": 0.17434582561254502,
      "num_tokens": 40494139.0,
      "step": 21715
    },
    {
      "entropy": 6.094521999359131,
      "epoch": 1.9180501589544332,
      "grad_norm": 2.6875,
      "learning_rate": 0.00046320710855979677,
      "loss": 5.7645,
      "mean_token_accuracy": 0.171839240193367,
      "num_tokens": 40503965.0,
      "step": 21720
    },
    {
      "entropy": 6.1057507514953615,
      "epoch": 1.9184916990462733,
      "grad_norm": 2.484375,
      "learning_rate": 0.00046318984926077637,
      "loss": 5.7516,
      "mean_token_accuracy": 0.17855425029993058,
      "num_tokens": 40513140.0,
      "step": 21725
    },
    {
      "entropy": 6.025338792800904,
      "epoch": 1.9189332391381138,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046317258627522393,
      "loss": 5.7696,
      "mean_token_accuracy": 0.1826067253947258,
      "num_tokens": 40522858.0,
      "step": 21730
    },
    {
      "entropy": 6.037998104095459,
      "epoch": 1.919374779229954,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004631553196034774,
      "loss": 5.7567,
      "mean_token_accuracy": 0.17928770333528518,
      "num_tokens": 40532050.0,
      "step": 21735
    },
    {
      "entropy": 6.10838770866394,
      "epoch": 1.9198163193217943,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004631380492458753,
      "loss": 5.738,
      "mean_token_accuracy": 0.17207290679216386,
      "num_tokens": 40540942.0,
      "step": 21740
    },
    {
      "entropy": 6.06263689994812,
      "epoch": 1.9202578594136348,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004631207752027558,
      "loss": 5.7884,
      "mean_token_accuracy": 0.17493458837270737,
      "num_tokens": 40550158.0,
      "step": 21745
    },
    {
      "entropy": 6.073766088485717,
      "epoch": 1.920699399505475,
      "grad_norm": 2.375,
      "learning_rate": 0.0004631034974744573,
      "loss": 5.9023,
      "mean_token_accuracy": 0.16174623966217042,
      "num_tokens": 40559267.0,
      "step": 21750
    },
    {
      "entropy": 6.125134038925171,
      "epoch": 1.9211409395973154,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004630862160613183,
      "loss": 5.7886,
      "mean_token_accuracy": 0.16743289679288864,
      "num_tokens": 40568503.0,
      "step": 21755
    },
    {
      "entropy": 6.130008554458618,
      "epoch": 1.9215824796891559,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004630689309636772,
      "loss": 5.7611,
      "mean_token_accuracy": 0.16947265863418579,
      "num_tokens": 40578464.0,
      "step": 21760
    },
    {
      "entropy": 6.113556241989135,
      "epoch": 1.9220240197809961,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004630516421818728,
      "loss": 5.9122,
      "mean_token_accuracy": 0.16252827793359756,
      "num_tokens": 40588338.0,
      "step": 21765
    },
    {
      "entropy": 6.0907598495483395,
      "epoch": 1.9224655598728364,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00046303434971624357,
      "loss": 5.7664,
      "mean_token_accuracy": 0.1727721318602562,
      "num_tokens": 40598912.0,
      "step": 21770
    },
    {
      "entropy": 6.047838878631592,
      "epoch": 1.922907099964677,
      "grad_norm": 2.25,
      "learning_rate": 0.00046301705356712846,
      "loss": 5.6769,
      "mean_token_accuracy": 0.17443982809782027,
      "num_tokens": 40607084.0,
      "step": 21775
    },
    {
      "entropy": 6.07603268623352,
      "epoch": 1.9233486400565172,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00046299975373486626,
      "loss": 5.7784,
      "mean_token_accuracy": 0.16572931706905364,
      "num_tokens": 40616422.0,
      "step": 21780
    },
    {
      "entropy": 6.050441789627075,
      "epoch": 1.9237901801483575,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004629824502197956,
      "loss": 5.7432,
      "mean_token_accuracy": 0.17481203824281694,
      "num_tokens": 40626189.0,
      "step": 21785
    },
    {
      "entropy": 5.99709677696228,
      "epoch": 1.924231720240198,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046296514302225585,
      "loss": 5.7242,
      "mean_token_accuracy": 0.17655985653400422,
      "num_tokens": 40636053.0,
      "step": 21790
    },
    {
      "entropy": 6.06900806427002,
      "epoch": 1.924673260332038,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004629478321425858,
      "loss": 5.8123,
      "mean_token_accuracy": 0.17019806951284408,
      "num_tokens": 40645159.0,
      "step": 21795
    },
    {
      "entropy": 6.091820430755615,
      "epoch": 1.9251148004238785,
      "grad_norm": 2.125,
      "learning_rate": 0.00046293051758112457,
      "loss": 5.6872,
      "mean_token_accuracy": 0.18223176002502442,
      "num_tokens": 40653890.0,
      "step": 21800
    },
    {
      "entropy": 6.107747554779053,
      "epoch": 1.9255563405157188,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046291319933821143,
      "loss": 5.8664,
      "mean_token_accuracy": 0.16482017934322357,
      "num_tokens": 40663041.0,
      "step": 21805
    },
    {
      "entropy": 6.01526107788086,
      "epoch": 1.925997880607559,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046289587741418553,
      "loss": 5.6964,
      "mean_token_accuracy": 0.18277244865894318,
      "num_tokens": 40670972.0,
      "step": 21810
    },
    {
      "entropy": 6.048628473281861,
      "epoch": 1.9264394206993996,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046287855180938633,
      "loss": 5.8446,
      "mean_token_accuracy": 0.17509347200393677,
      "num_tokens": 40680503.0,
      "step": 21815
    },
    {
      "entropy": 6.124141216278076,
      "epoch": 1.9268809607912398,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046286122252415303,
      "loss": 5.8405,
      "mean_token_accuracy": 0.1726961091160774,
      "num_tokens": 40691339.0,
      "step": 21820
    },
    {
      "entropy": 6.133550405502319,
      "epoch": 1.92732250088308,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046284388955882526,
      "loss": 5.8111,
      "mean_token_accuracy": 0.16965928822755813,
      "num_tokens": 40699744.0,
      "step": 21825
    },
    {
      "entropy": 6.110125017166138,
      "epoch": 1.9277640409749206,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046282655291374257,
      "loss": 5.8105,
      "mean_token_accuracy": 0.16978985518217088,
      "num_tokens": 40709144.0,
      "step": 21830
    },
    {
      "entropy": 6.087880420684814,
      "epoch": 1.9282055810667609,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046280921258924455,
      "loss": 5.8931,
      "mean_token_accuracy": 0.16667681634426118,
      "num_tokens": 40719513.0,
      "step": 21835
    },
    {
      "entropy": 6.093751621246338,
      "epoch": 1.9286471211586012,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046279186858567083,
      "loss": 5.8116,
      "mean_token_accuracy": 0.16393331438302994,
      "num_tokens": 40728846.0,
      "step": 21840
    },
    {
      "entropy": 6.0487861156463625,
      "epoch": 1.9290886612504416,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004627745209033613,
      "loss": 5.7173,
      "mean_token_accuracy": 0.1785743698477745,
      "num_tokens": 40737904.0,
      "step": 21845
    },
    {
      "entropy": 6.0005817890167235,
      "epoch": 1.929530201342282,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004627571695426556,
      "loss": 5.7319,
      "mean_token_accuracy": 0.18728642910718918,
      "num_tokens": 40747328.0,
      "step": 21850
    },
    {
      "entropy": 6.070573568344116,
      "epoch": 1.9299717414341222,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046273981450389374,
      "loss": 5.7644,
      "mean_token_accuracy": 0.16974205225706102,
      "num_tokens": 40755567.0,
      "step": 21855
    },
    {
      "entropy": 6.017500734329223,
      "epoch": 1.9304132815259627,
      "grad_norm": 2.703125,
      "learning_rate": 0.00046272245578741567,
      "loss": 5.772,
      "mean_token_accuracy": 0.16800532937049867,
      "num_tokens": 40764684.0,
      "step": 21860
    },
    {
      "entropy": 6.077537918090821,
      "epoch": 1.9308548216178028,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004627050933935615,
      "loss": 5.7716,
      "mean_token_accuracy": 0.18046178817749023,
      "num_tokens": 40774050.0,
      "step": 21865
    },
    {
      "entropy": 6.082465124130249,
      "epoch": 1.9312963617096433,
      "grad_norm": 2.34375,
      "learning_rate": 0.00046268772732267126,
      "loss": 5.7483,
      "mean_token_accuracy": 0.17499785274267196,
      "num_tokens": 40782722.0,
      "step": 21870
    },
    {
      "entropy": 6.041574478149414,
      "epoch": 1.9317379018014835,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004626703575750852,
      "loss": 5.8141,
      "mean_token_accuracy": 0.17131524831056594,
      "num_tokens": 40791978.0,
      "step": 21875
    },
    {
      "entropy": 6.046136808395386,
      "epoch": 1.9321794418933238,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004626529841511436,
      "loss": 5.8447,
      "mean_token_accuracy": 0.16501759141683578,
      "num_tokens": 40801855.0,
      "step": 21880
    },
    {
      "entropy": 6.148819065093994,
      "epoch": 1.9326209819851643,
      "grad_norm": 1.75,
      "learning_rate": 0.00046263560705118675,
      "loss": 5.9081,
      "mean_token_accuracy": 0.16544333845376968,
      "num_tokens": 40812025.0,
      "step": 21885
    },
    {
      "entropy": 6.140730953216552,
      "epoch": 1.9330625220770046,
      "grad_norm": 2.125,
      "learning_rate": 0.0004626182262755551,
      "loss": 5.728,
      "mean_token_accuracy": 0.16779044270515442,
      "num_tokens": 40820758.0,
      "step": 21890
    },
    {
      "entropy": 6.055291032791137,
      "epoch": 1.9335040621688449,
      "grad_norm": 2.65625,
      "learning_rate": 0.00046260084182458906,
      "loss": 5.827,
      "mean_token_accuracy": 0.1704641029238701,
      "num_tokens": 40830653.0,
      "step": 21895
    },
    {
      "entropy": 6.105960178375244,
      "epoch": 1.9339456022606853,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046258345369862915,
      "loss": 5.8859,
      "mean_token_accuracy": 0.16519241780042648,
      "num_tokens": 40840370.0,
      "step": 21900
    },
    {
      "entropy": 6.161451816558838,
      "epoch": 1.9343871423525256,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004625660618980162,
      "loss": 5.7997,
      "mean_token_accuracy": 0.16797949820756913,
      "num_tokens": 40849590.0,
      "step": 21905
    },
    {
      "entropy": 6.043071651458741,
      "epoch": 1.934828682444366,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046254866642309067,
      "loss": 5.7845,
      "mean_token_accuracy": 0.1733742654323578,
      "num_tokens": 40859271.0,
      "step": 21910
    },
    {
      "entropy": 6.0452752113342285,
      "epoch": 1.9352702225362064,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004625312672741935,
      "loss": 5.8218,
      "mean_token_accuracy": 0.16800014674663544,
      "num_tokens": 40869272.0,
      "step": 21915
    },
    {
      "entropy": 6.222105312347412,
      "epoch": 1.9357117626280467,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004625138644516655,
      "loss": 5.8878,
      "mean_token_accuracy": 0.16493486911058425,
      "num_tokens": 40880327.0,
      "step": 21920
    },
    {
      "entropy": 6.075521612167359,
      "epoch": 1.936153302719887,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004624964579558474,
      "loss": 5.8273,
      "mean_token_accuracy": 0.16807909160852433,
      "num_tokens": 40890407.0,
      "step": 21925
    },
    {
      "entropy": 6.109927320480347,
      "epoch": 1.9365948428117274,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004624790477870804,
      "loss": 5.7808,
      "mean_token_accuracy": 0.1685138612985611,
      "num_tokens": 40899306.0,
      "step": 21930
    },
    {
      "entropy": 6.181375122070312,
      "epoch": 1.9370363829035675,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004624616339457055,
      "loss": 5.8578,
      "mean_token_accuracy": 0.16582290977239608,
      "num_tokens": 40908908.0,
      "step": 21935
    },
    {
      "entropy": 6.115111780166626,
      "epoch": 1.937477922995408,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004624442164320638,
      "loss": 5.8106,
      "mean_token_accuracy": 0.17320494949817658,
      "num_tokens": 40917960.0,
      "step": 21940
    },
    {
      "entropy": 5.998154592514038,
      "epoch": 1.9379194630872483,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004624267952464965,
      "loss": 5.6198,
      "mean_token_accuracy": 0.18235038965940475,
      "num_tokens": 40926364.0,
      "step": 21945
    },
    {
      "entropy": 6.06477632522583,
      "epoch": 1.9383610031790885,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046240937038934484,
      "loss": 5.8231,
      "mean_token_accuracy": 0.16534019708633424,
      "num_tokens": 40936558.0,
      "step": 21950
    },
    {
      "entropy": 6.055356502532959,
      "epoch": 1.938802543270929,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004623919418609502,
      "loss": 5.7805,
      "mean_token_accuracy": 0.17553754895925522,
      "num_tokens": 40945501.0,
      "step": 21955
    },
    {
      "entropy": 6.014885282516479,
      "epoch": 1.9392440833627693,
      "grad_norm": 2.671875,
      "learning_rate": 0.00046237450966165407,
      "loss": 5.7742,
      "mean_token_accuracy": 0.17473527789115906,
      "num_tokens": 40954740.0,
      "step": 21960
    },
    {
      "entropy": 6.061191082000732,
      "epoch": 1.9396856234546096,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00046235707379179775,
      "loss": 5.8474,
      "mean_token_accuracy": 0.1711583614349365,
      "num_tokens": 40963748.0,
      "step": 21965
    },
    {
      "entropy": 6.123156881332397,
      "epoch": 1.94012716354645,
      "grad_norm": 1.625,
      "learning_rate": 0.00046233963425172306,
      "loss": 5.8127,
      "mean_token_accuracy": 0.16981699615716933,
      "num_tokens": 40973907.0,
      "step": 21970
    },
    {
      "entropy": 6.103388929367066,
      "epoch": 1.9405687036382904,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046232219104177136,
      "loss": 5.8221,
      "mean_token_accuracy": 0.16296774595975877,
      "num_tokens": 40983374.0,
      "step": 21975
    },
    {
      "entropy": 6.1597836971282955,
      "epoch": 1.9410102437301306,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004623047441622845,
      "loss": 5.8568,
      "mean_token_accuracy": 0.1592784121632576,
      "num_tokens": 40993305.0,
      "step": 21980
    },
    {
      "entropy": 6.172373962402344,
      "epoch": 1.9414517838219711,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046228729361360427,
      "loss": 5.7825,
      "mean_token_accuracy": 0.16402530670166016,
      "num_tokens": 41002438.0,
      "step": 21985
    },
    {
      "entropy": 5.970716571807861,
      "epoch": 1.9418933239138114,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046226983939607236,
      "loss": 5.7327,
      "mean_token_accuracy": 0.1748810201883316,
      "num_tokens": 41011491.0,
      "step": 21990
    },
    {
      "entropy": 6.055589056015014,
      "epoch": 1.9423348640056517,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004622523815100309,
      "loss": 5.8461,
      "mean_token_accuracy": 0.16704246401786804,
      "num_tokens": 41021091.0,
      "step": 21995
    },
    {
      "entropy": 6.08888349533081,
      "epoch": 1.9427764040974922,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046223491995582187,
      "loss": 5.792,
      "mean_token_accuracy": 0.16747479140758514,
      "num_tokens": 41030719.0,
      "step": 22000
    },
    {
      "entropy": 6.083539438247681,
      "epoch": 1.9432179441893322,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004622174547337871,
      "loss": 5.8317,
      "mean_token_accuracy": 0.16503254175186158,
      "num_tokens": 41039651.0,
      "step": 22005
    },
    {
      "entropy": 6.088931512832642,
      "epoch": 1.9436594842811727,
      "grad_norm": 3.34375,
      "learning_rate": 0.00046219998584426885,
      "loss": 5.7309,
      "mean_token_accuracy": 0.1723812848329544,
      "num_tokens": 41047914.0,
      "step": 22010
    },
    {
      "entropy": 6.077472591400147,
      "epoch": 1.944101024373013,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004621825132876094,
      "loss": 5.8542,
      "mean_token_accuracy": 0.16681549251079558,
      "num_tokens": 41056779.0,
      "step": 22015
    },
    {
      "entropy": 6.105723285675049,
      "epoch": 1.9445425644648533,
      "grad_norm": 2.6875,
      "learning_rate": 0.00046216503706415096,
      "loss": 5.785,
      "mean_token_accuracy": 0.17238941341638564,
      "num_tokens": 41066356.0,
      "step": 22020
    },
    {
      "entropy": 6.165150117874146,
      "epoch": 1.9449841045566938,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004621475571742359,
      "loss": 5.8305,
      "mean_token_accuracy": 0.17884328663349153,
      "num_tokens": 41075766.0,
      "step": 22025
    },
    {
      "entropy": 6.118260669708252,
      "epoch": 1.945425644648534,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046213007361820663,
      "loss": 5.7883,
      "mean_token_accuracy": 0.1724184885621071,
      "num_tokens": 41085503.0,
      "step": 22030
    },
    {
      "entropy": 6.150070333480835,
      "epoch": 1.9458671847403743,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004621125863964057,
      "loss": 5.9642,
      "mean_token_accuracy": 0.1561279535293579,
      "num_tokens": 41096581.0,
      "step": 22035
    },
    {
      "entropy": 6.0784773349761965,
      "epoch": 1.9463087248322148,
      "grad_norm": 2.953125,
      "learning_rate": 0.00046209509550917553,
      "loss": 5.7686,
      "mean_token_accuracy": 0.17354507744312286,
      "num_tokens": 41105226.0,
      "step": 22040
    },
    {
      "entropy": 6.138539981842041,
      "epoch": 1.9467502649240551,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004620776009568589,
      "loss": 5.7911,
      "mean_token_accuracy": 0.17229045927524567,
      "num_tokens": 41113998.0,
      "step": 22045
    },
    {
      "entropy": 6.060680246353149,
      "epoch": 1.9471918050158954,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004620601027397984,
      "loss": 5.7963,
      "mean_token_accuracy": 0.16104075014591218,
      "num_tokens": 41123147.0,
      "step": 22050
    },
    {
      "entropy": 6.0763993740081785,
      "epoch": 1.947633345107736,
      "grad_norm": 1.671875,
      "learning_rate": 0.00046204260085833695,
      "loss": 5.7818,
      "mean_token_accuracy": 0.17426732331514358,
      "num_tokens": 41132317.0,
      "step": 22055
    },
    {
      "entropy": 6.091706895828247,
      "epoch": 1.9480748851995762,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004620250953128173,
      "loss": 5.8387,
      "mean_token_accuracy": 0.16694470196962358,
      "num_tokens": 41140795.0,
      "step": 22060
    },
    {
      "entropy": 6.089817523956299,
      "epoch": 1.9485164252914164,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004620075861035824,
      "loss": 5.7934,
      "mean_token_accuracy": 0.16738156378269195,
      "num_tokens": 41149959.0,
      "step": 22065
    },
    {
      "entropy": 6.084614992141724,
      "epoch": 1.948957965383257,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004619900732309753,
      "loss": 5.8676,
      "mean_token_accuracy": 0.16183963119983674,
      "num_tokens": 41159388.0,
      "step": 22070
    },
    {
      "entropy": 6.233500957489014,
      "epoch": 1.949399505475097,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046197255669533886,
      "loss": 6.0292,
      "mean_token_accuracy": 0.15898094922304154,
      "num_tokens": 41168875.0,
      "step": 22075
    },
    {
      "entropy": 6.181044673919677,
      "epoch": 1.9498410455669375,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046195503649701646,
      "loss": 5.8488,
      "mean_token_accuracy": 0.16562723368406296,
      "num_tokens": 41177621.0,
      "step": 22080
    },
    {
      "entropy": 6.092544841766357,
      "epoch": 1.9502825856587778,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004619375126363513,
      "loss": 5.8706,
      "mean_token_accuracy": 0.167301182448864,
      "num_tokens": 41185911.0,
      "step": 22085
    },
    {
      "entropy": 6.103871393203735,
      "epoch": 1.950724125750618,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00046191998511368646,
      "loss": 5.7668,
      "mean_token_accuracy": 0.17563872933387756,
      "num_tokens": 41194811.0,
      "step": 22090
    },
    {
      "entropy": 5.998917579650879,
      "epoch": 1.9511656658424585,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00046190245392936554,
      "loss": 5.6941,
      "mean_token_accuracy": 0.18211950957775117,
      "num_tokens": 41203955.0,
      "step": 22095
    },
    {
      "entropy": 6.050065898895264,
      "epoch": 1.9516072059342988,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004618849190837317,
      "loss": 5.7409,
      "mean_token_accuracy": 0.17454166412353517,
      "num_tokens": 41212635.0,
      "step": 22100
    },
    {
      "entropy": 6.098102188110351,
      "epoch": 1.952048746026139,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004618673805771287,
      "loss": 5.7893,
      "mean_token_accuracy": 0.1663086473941803,
      "num_tokens": 41221443.0,
      "step": 22105
    },
    {
      "entropy": 6.0908421039581295,
      "epoch": 1.9524902861179796,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004618498384098999,
      "loss": 5.7768,
      "mean_token_accuracy": 0.17070409059524536,
      "num_tokens": 41230089.0,
      "step": 22110
    },
    {
      "entropy": 6.072627162933349,
      "epoch": 1.9529318262098199,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046183229258238905,
      "loss": 5.8826,
      "mean_token_accuracy": 0.16405201256275176,
      "num_tokens": 41239569.0,
      "step": 22115
    },
    {
      "entropy": 6.131079053878784,
      "epoch": 1.9533733663016601,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004618147430949399,
      "loss": 5.7929,
      "mean_token_accuracy": 0.17486875504255295,
      "num_tokens": 41248394.0,
      "step": 22120
    },
    {
      "entropy": 6.0715662956237795,
      "epoch": 1.9538149063935006,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046179718994789624,
      "loss": 5.8044,
      "mean_token_accuracy": 0.17719150483608245,
      "num_tokens": 41258148.0,
      "step": 22125
    },
    {
      "entropy": 6.121745491027832,
      "epoch": 1.954256446485341,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046177963314160175,
      "loss": 5.7223,
      "mean_token_accuracy": 0.1783413589000702,
      "num_tokens": 41266984.0,
      "step": 22130
    },
    {
      "entropy": 6.099260663986206,
      "epoch": 1.9546979865771812,
      "grad_norm": 2.90625,
      "learning_rate": 0.00046176207267640054,
      "loss": 5.8553,
      "mean_token_accuracy": 0.1691174864768982,
      "num_tokens": 41276119.0,
      "step": 22135
    },
    {
      "entropy": 5.991044998168945,
      "epoch": 1.9551395266690217,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004617445085526365,
      "loss": 5.7102,
      "mean_token_accuracy": 0.17640875577926635,
      "num_tokens": 41284945.0,
      "step": 22140
    },
    {
      "entropy": 6.056941652297974,
      "epoch": 1.9555810667608617,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004617269407706537,
      "loss": 5.7138,
      "mean_token_accuracy": 0.18317076116800307,
      "num_tokens": 41294255.0,
      "step": 22145
    },
    {
      "entropy": 6.076195096969604,
      "epoch": 1.9560226068527022,
      "grad_norm": 5.625,
      "learning_rate": 0.0004617093693307964,
      "loss": 5.8235,
      "mean_token_accuracy": 0.1691629871726036,
      "num_tokens": 41303056.0,
      "step": 22150
    },
    {
      "entropy": 6.092026996612549,
      "epoch": 1.9564641469445425,
      "grad_norm": 3.15625,
      "learning_rate": 0.00046169179423340866,
      "loss": 5.8533,
      "mean_token_accuracy": 0.1646975815296173,
      "num_tokens": 41312444.0,
      "step": 22155
    },
    {
      "entropy": 6.152564811706543,
      "epoch": 1.9569056870363828,
      "grad_norm": 2.390625,
      "learning_rate": 0.000461674215478835,
      "loss": 5.9375,
      "mean_token_accuracy": 0.16892132610082627,
      "num_tokens": 41322507.0,
      "step": 22160
    },
    {
      "entropy": 6.12220983505249,
      "epoch": 1.9573472271282233,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004616566330674195,
      "loss": 5.7939,
      "mean_token_accuracy": 0.1764255777001381,
      "num_tokens": 41332507.0,
      "step": 22165
    },
    {
      "entropy": 6.115534543991089,
      "epoch": 1.9577887672200636,
      "grad_norm": 2.25,
      "learning_rate": 0.0004616390469995068,
      "loss": 5.7853,
      "mean_token_accuracy": 0.17447489947080613,
      "num_tokens": 41342085.0,
      "step": 22170
    },
    {
      "entropy": 6.070411825180054,
      "epoch": 1.9582303073119038,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046162145727544126,
      "loss": 5.7332,
      "mean_token_accuracy": 0.18058770298957824,
      "num_tokens": 41351422.0,
      "step": 22175
    },
    {
      "entropy": 6.077708196640015,
      "epoch": 1.9586718474037443,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004616038638955675,
      "loss": 5.7703,
      "mean_token_accuracy": 0.1756131112575531,
      "num_tokens": 41361337.0,
      "step": 22180
    },
    {
      "entropy": 6.029050254821778,
      "epoch": 1.9591133874955846,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046158626686023014,
      "loss": 5.7462,
      "mean_token_accuracy": 0.17536391019821168,
      "num_tokens": 41370734.0,
      "step": 22185
    },
    {
      "entropy": 6.04237585067749,
      "epoch": 1.9595549275874249,
      "grad_norm": 2.859375,
      "learning_rate": 0.00046156866616977406,
      "loss": 5.7529,
      "mean_token_accuracy": 0.17436618357896805,
      "num_tokens": 41380090.0,
      "step": 22190
    },
    {
      "entropy": 6.12188606262207,
      "epoch": 1.9599964676792654,
      "grad_norm": 1.828125,
      "learning_rate": 0.00046155106182454384,
      "loss": 5.8037,
      "mean_token_accuracy": 0.1772506058216095,
      "num_tokens": 41388967.0,
      "step": 22195
    },
    {
      "entropy": 6.042675638198853,
      "epoch": 1.9604380077711057,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004615334538248844,
      "loss": 5.7749,
      "mean_token_accuracy": 0.1712171882390976,
      "num_tokens": 41397429.0,
      "step": 22200
    },
    {
      "entropy": 6.031103801727295,
      "epoch": 1.960879547862946,
      "grad_norm": 2.75,
      "learning_rate": 0.00046151584217114073,
      "loss": 5.7441,
      "mean_token_accuracy": 0.1719260632991791,
      "num_tokens": 41406420.0,
      "step": 22205
    },
    {
      "entropy": 5.967312049865723,
      "epoch": 1.9613210879547864,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004614982268636577,
      "loss": 5.6792,
      "mean_token_accuracy": 0.17842305153608323,
      "num_tokens": 41415994.0,
      "step": 22210
    },
    {
      "entropy": 6.077283716201782,
      "epoch": 1.9617626280466265,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004614806079027806,
      "loss": 5.7002,
      "mean_token_accuracy": 0.1796986863017082,
      "num_tokens": 41425744.0,
      "step": 22215
    },
    {
      "entropy": 6.110962152481079,
      "epoch": 1.962204168138467,
      "grad_norm": 2.71875,
      "learning_rate": 0.00046146298528885435,
      "loss": 5.8751,
      "mean_token_accuracy": 0.16830646991729736,
      "num_tokens": 41435452.0,
      "step": 22220
    },
    {
      "entropy": 6.098014450073242,
      "epoch": 1.9626457082303073,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004614453590222243,
      "loss": 5.8152,
      "mean_token_accuracy": 0.17671353965997696,
      "num_tokens": 41444853.0,
      "step": 22225
    },
    {
      "entropy": 6.000149869918824,
      "epoch": 1.9630872483221475,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004614277291032357,
      "loss": 5.6923,
      "mean_token_accuracy": 0.19407846182584762,
      "num_tokens": 41454569.0,
      "step": 22230
    },
    {
      "entropy": 6.133536911010742,
      "epoch": 1.963528788413988,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004614100955322339,
      "loss": 5.8532,
      "mean_token_accuracy": 0.1720211997628212,
      "num_tokens": 41464159.0,
      "step": 22235
    },
    {
      "entropy": 6.112495851516724,
      "epoch": 1.9639703285058283,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004613924583095644,
      "loss": 5.8698,
      "mean_token_accuracy": 0.16430338025093078,
      "num_tokens": 41474354.0,
      "step": 22240
    },
    {
      "entropy": 6.051172113418579,
      "epoch": 1.9644118685976686,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004613748174355726,
      "loss": 5.7341,
      "mean_token_accuracy": 0.17458372116088866,
      "num_tokens": 41484053.0,
      "step": 22245
    },
    {
      "entropy": 6.111041116714477,
      "epoch": 1.964853408689509,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046135717291060415,
      "loss": 5.7965,
      "mean_token_accuracy": 0.16547825038433076,
      "num_tokens": 41493377.0,
      "step": 22250
    },
    {
      "entropy": 6.03553876876831,
      "epoch": 1.9652949487813494,
      "grad_norm": 1.90625,
      "learning_rate": 0.00046133952473500464,
      "loss": 5.7377,
      "mean_token_accuracy": 0.17641324251890184,
      "num_tokens": 41502361.0,
      "step": 22255
    },
    {
      "entropy": 6.0039678573608395,
      "epoch": 1.9657364888731896,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004613218729091198,
      "loss": 5.7933,
      "mean_token_accuracy": 0.17322905957698823,
      "num_tokens": 41511083.0,
      "step": 22260
    },
    {
      "entropy": 6.120638799667359,
      "epoch": 1.9661780289650301,
      "grad_norm": 6.90625,
      "learning_rate": 0.0004613042174332955,
      "loss": 5.88,
      "mean_token_accuracy": 0.16631219163537025,
      "num_tokens": 41520264.0,
      "step": 22265
    },
    {
      "entropy": 6.164268970489502,
      "epoch": 1.9666195690568704,
      "grad_norm": 2.75,
      "learning_rate": 0.0004612865583078775,
      "loss": 5.7333,
      "mean_token_accuracy": 0.1748616099357605,
      "num_tokens": 41529805.0,
      "step": 22270
    },
    {
      "entropy": 5.974587726593017,
      "epoch": 1.9670611091487107,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004612688955332118,
      "loss": 5.6814,
      "mean_token_accuracy": 0.17812594771385193,
      "num_tokens": 41538348.0,
      "step": 22275
    },
    {
      "entropy": 6.000253486633301,
      "epoch": 1.9675026492405512,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046125122910964427,
      "loss": 5.8055,
      "mean_token_accuracy": 0.16663441956043243,
      "num_tokens": 41547433.0,
      "step": 22280
    },
    {
      "entropy": 6.099403762817383,
      "epoch": 1.9679441893323912,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004612335590375212,
      "loss": 5.8037,
      "mean_token_accuracy": 0.1670122340321541,
      "num_tokens": 41556162.0,
      "step": 22285
    },
    {
      "entropy": 6.156384897232056,
      "epoch": 1.9683857294242317,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004612158853171886,
      "loss": 5.7791,
      "mean_token_accuracy": 0.17200758308172226,
      "num_tokens": 41565415.0,
      "step": 22290
    },
    {
      "entropy": 6.084620666503906,
      "epoch": 1.968827269516072,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004611982079489928,
      "loss": 5.6928,
      "mean_token_accuracy": 0.1847338169813156,
      "num_tokens": 41574728.0,
      "step": 22295
    },
    {
      "entropy": 6.035195970535279,
      "epoch": 1.9692688096079123,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004611805269332798,
      "loss": 5.8352,
      "mean_token_accuracy": 0.16349859684705734,
      "num_tokens": 41584749.0,
      "step": 22300
    },
    {
      "entropy": 6.043558645248413,
      "epoch": 1.9697103496997528,
      "grad_norm": 3.46875,
      "learning_rate": 0.00046116284227039626,
      "loss": 5.7305,
      "mean_token_accuracy": 0.1782889448106289,
      "num_tokens": 41594658.0,
      "step": 22305
    },
    {
      "entropy": 6.081830263137817,
      "epoch": 1.970151889791593,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004611451539606886,
      "loss": 5.8423,
      "mean_token_accuracy": 0.17075575590133668,
      "num_tokens": 41604548.0,
      "step": 22310
    },
    {
      "entropy": 6.127156734466553,
      "epoch": 1.9705934298834333,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004611274620045031,
      "loss": 5.795,
      "mean_token_accuracy": 0.16849096715450287,
      "num_tokens": 41614002.0,
      "step": 22315
    },
    {
      "entropy": 6.109599018096924,
      "epoch": 1.9710349699752738,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004611097664021866,
      "loss": 5.8108,
      "mean_token_accuracy": 0.16453594118356704,
      "num_tokens": 41623793.0,
      "step": 22320
    },
    {
      "entropy": 6.070938587188721,
      "epoch": 1.971476510067114,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00046109206715408554,
      "loss": 5.741,
      "mean_token_accuracy": 0.16723096519708633,
      "num_tokens": 41632608.0,
      "step": 22325
    },
    {
      "entropy": 6.10354642868042,
      "epoch": 1.9719180501589544,
      "grad_norm": 1.921875,
      "learning_rate": 0.00046107436426054676,
      "loss": 5.8354,
      "mean_token_accuracy": 0.16783374398946763,
      "num_tokens": 41641814.0,
      "step": 22330
    },
    {
      "entropy": 6.119575262069702,
      "epoch": 1.9723595902507949,
      "grad_norm": 2.28125,
      "learning_rate": 0.000461056657721917,
      "loss": 5.8076,
      "mean_token_accuracy": 0.17176466733217238,
      "num_tokens": 41650643.0,
      "step": 22335
    },
    {
      "entropy": 6.09545865058899,
      "epoch": 1.9728011303426352,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046103894753854317,
      "loss": 5.818,
      "mean_token_accuracy": 0.16710563451051713,
      "num_tokens": 41661308.0,
      "step": 22340
    },
    {
      "entropy": 6.159102916717529,
      "epoch": 1.9732426704344754,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004610212337107721,
      "loss": 5.8277,
      "mean_token_accuracy": 0.1674237370491028,
      "num_tokens": 41671090.0,
      "step": 22345
    },
    {
      "entropy": 6.056845378875733,
      "epoch": 1.973684210526316,
      "grad_norm": 2.421875,
      "learning_rate": 0.00046100351623895087,
      "loss": 5.7862,
      "mean_token_accuracy": 0.16926652789115906,
      "num_tokens": 41679584.0,
      "step": 22350
    },
    {
      "entropy": 6.009527111053467,
      "epoch": 1.974125750618156,
      "grad_norm": 5.34375,
      "learning_rate": 0.00046098579512342654,
      "loss": 5.7957,
      "mean_token_accuracy": 0.16196219325065614,
      "num_tokens": 41689772.0,
      "step": 22355
    },
    {
      "entropy": 6.113477945327759,
      "epoch": 1.9745672907099965,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004609680703645463,
      "loss": 5.7766,
      "mean_token_accuracy": 0.17247380018234254,
      "num_tokens": 41699196.0,
      "step": 22360
    },
    {
      "entropy": 6.110484552383423,
      "epoch": 1.9750088308018368,
      "grad_norm": 2.796875,
      "learning_rate": 0.00046095034196265726,
      "loss": 5.7537,
      "mean_token_accuracy": 0.17644185870885848,
      "num_tokens": 41707899.0,
      "step": 22365
    },
    {
      "entropy": 6.02688307762146,
      "epoch": 1.975450370893677,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046093260991810685,
      "loss": 5.7794,
      "mean_token_accuracy": 0.17416183948516845,
      "num_tokens": 41716903.0,
      "step": 22370
    },
    {
      "entropy": 6.083221292495727,
      "epoch": 1.9758919109855175,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00046091487423124234,
      "loss": 5.7331,
      "mean_token_accuracy": 0.16851509511470794,
      "num_tokens": 41725676.0,
      "step": 22375
    },
    {
      "entropy": 6.122511911392212,
      "epoch": 1.9763334510773578,
      "grad_norm": 3.390625,
      "learning_rate": 0.00046089713490241117,
      "loss": 5.703,
      "mean_token_accuracy": 0.17383269518613814,
      "num_tokens": 41734720.0,
      "step": 22380
    },
    {
      "entropy": 6.053651237487793,
      "epoch": 1.976774991169198,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004608793919319608,
      "loss": 5.7907,
      "mean_token_accuracy": 0.17582984417676925,
      "num_tokens": 41743866.0,
      "step": 22385
    },
    {
      "entropy": 6.023544549942017,
      "epoch": 1.9772165312610386,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046086164532023894,
      "loss": 5.7774,
      "mean_token_accuracy": 0.17741502523422242,
      "num_tokens": 41754157.0,
      "step": 22390
    },
    {
      "entropy": 6.05854868888855,
      "epoch": 1.9776580713528789,
      "grad_norm": 2.265625,
      "learning_rate": 0.00046084389506759313,
      "loss": 5.7753,
      "mean_token_accuracy": 0.1713033065199852,
      "num_tokens": 41763105.0,
      "step": 22395
    },
    {
      "entropy": 6.117065000534057,
      "epoch": 1.9780996114447191,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046082614117437103,
      "loss": 5.7044,
      "mean_token_accuracy": 0.1731950595974922,
      "num_tokens": 41772623.0,
      "step": 22400
    },
    {
      "entropy": 6.069474458694458,
      "epoch": 1.9785411515365596,
      "grad_norm": 2.8125,
      "learning_rate": 0.00046080838364092057,
      "loss": 5.7912,
      "mean_token_accuracy": 0.16727855801582336,
      "num_tokens": 41781331.0,
      "step": 22405
    },
    {
      "entropy": 6.070024919509888,
      "epoch": 1.9789826916284,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004607906224675896,
      "loss": 5.9154,
      "mean_token_accuracy": 0.16006624847650527,
      "num_tokens": 41789727.0,
      "step": 22410
    },
    {
      "entropy": 6.123909425735474,
      "epoch": 1.9794242317202402,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004607728576547259,
      "loss": 5.7877,
      "mean_token_accuracy": 0.17119552940130234,
      "num_tokens": 41798531.0,
      "step": 22415
    },
    {
      "entropy": 6.119760656356812,
      "epoch": 1.9798657718120807,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046075508920267766,
      "loss": 5.7986,
      "mean_token_accuracy": 0.16944885700941087,
      "num_tokens": 41808304.0,
      "step": 22420
    },
    {
      "entropy": 6.039729213714599,
      "epoch": 1.9803073119039207,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004607373171117929,
      "loss": 5.6923,
      "mean_token_accuracy": 0.17815064489841462,
      "num_tokens": 41817427.0,
      "step": 22425
    },
    {
      "entropy": 6.019547367095948,
      "epoch": 1.9807488519957612,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004607195413824197,
      "loss": 5.7596,
      "mean_token_accuracy": 0.17684614658355713,
      "num_tokens": 41827202.0,
      "step": 22430
    },
    {
      "entropy": 6.054301023483276,
      "epoch": 1.9811903920876015,
      "grad_norm": 2.375,
      "learning_rate": 0.0004607017620149062,
      "loss": 5.714,
      "mean_token_accuracy": 0.1773855835199356,
      "num_tokens": 41836213.0,
      "step": 22435
    },
    {
      "entropy": 6.048216056823731,
      "epoch": 1.9816319321794418,
      "grad_norm": 3.0,
      "learning_rate": 0.0004606839790096009,
      "loss": 5.8254,
      "mean_token_accuracy": 0.16356285363435746,
      "num_tokens": 41845358.0,
      "step": 22440
    },
    {
      "entropy": 6.0881676197052,
      "epoch": 1.9820734722712823,
      "grad_norm": 5.4375,
      "learning_rate": 0.0004606661923668521,
      "loss": 5.7904,
      "mean_token_accuracy": 0.17645473778247833,
      "num_tokens": 41854572.0,
      "step": 22445
    },
    {
      "entropy": 6.08144326210022,
      "epoch": 1.9825150123631226,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004606484020870081,
      "loss": 5.7671,
      "mean_token_accuracy": 0.17829624712467193,
      "num_tokens": 41863643.0,
      "step": 22450
    },
    {
      "entropy": 6.021071863174439,
      "epoch": 1.9829565524549628,
      "grad_norm": 4.28125,
      "learning_rate": 0.00046063060817041747,
      "loss": 5.6977,
      "mean_token_accuracy": 0.1737956151366234,
      "num_tokens": 41873004.0,
      "step": 22455
    },
    {
      "entropy": 6.083002614974975,
      "epoch": 1.9833980925468033,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004606128106174288,
      "loss": 5.7403,
      "mean_token_accuracy": 0.17781067341566087,
      "num_tokens": 41881530.0,
      "step": 22460
    },
    {
      "entropy": 6.024988031387329,
      "epoch": 1.9838396326386436,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004605950094283908,
      "loss": 5.8065,
      "mean_token_accuracy": 0.16909864842891692,
      "num_tokens": 41890342.0,
      "step": 22465
    },
    {
      "entropy": 6.03479266166687,
      "epoch": 1.9842811727304839,
      "grad_norm": 2.84375,
      "learning_rate": 0.00046057720460365205,
      "loss": 5.7359,
      "mean_token_accuracy": 0.17384000420570372,
      "num_tokens": 41899274.0,
      "step": 22470
    },
    {
      "entropy": 6.13137412071228,
      "epoch": 1.9847227128223244,
      "grad_norm": 2.46875,
      "learning_rate": 0.00046055939614356146,
      "loss": 5.7772,
      "mean_token_accuracy": 0.1719484269618988,
      "num_tokens": 41908835.0,
      "step": 22475
    },
    {
      "entropy": 6.054810810089111,
      "epoch": 1.9851642529141647,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004605415840484678,
      "loss": 5.6972,
      "mean_token_accuracy": 0.176234233379364,
      "num_tokens": 41917900.0,
      "step": 22480
    },
    {
      "entropy": 6.053602981567383,
      "epoch": 1.985605793006005,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046052376831872007,
      "loss": 5.7933,
      "mean_token_accuracy": 0.17869770228862764,
      "num_tokens": 41926964.0,
      "step": 22485
    },
    {
      "entropy": 6.071066999435425,
      "epoch": 1.9860473330978454,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004605059489546672,
      "loss": 5.8443,
      "mean_token_accuracy": 0.17304819226264953,
      "num_tokens": 41936401.0,
      "step": 22490
    },
    {
      "entropy": 6.0321207523345945,
      "epoch": 1.9864888731896855,
      "grad_norm": 3.34375,
      "learning_rate": 0.00046048812595665824,
      "loss": 5.7003,
      "mean_token_accuracy": 0.1725921243429184,
      "num_tokens": 41944665.0,
      "step": 22495
    },
    {
      "entropy": 6.023105955123901,
      "epoch": 1.986930413281526,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004604702993250425,
      "loss": 5.7805,
      "mean_token_accuracy": 0.16897769868373871,
      "num_tokens": 41953868.0,
      "step": 22500
    },
    {
      "entropy": 6.036277961730957,
      "epoch": 1.9873719533733663,
      "grad_norm": 6.34375,
      "learning_rate": 0.00046045246906016904,
      "loss": 5.8126,
      "mean_token_accuracy": 0.17801220268011092,
      "num_tokens": 41963013.0,
      "step": 22505
    },
    {
      "entropy": 6.088152027130127,
      "epoch": 1.9878134934652065,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004604346351623872,
      "loss": 5.812,
      "mean_token_accuracy": 0.16001619026064873,
      "num_tokens": 41972926.0,
      "step": 22510
    },
    {
      "entropy": 6.1506448745727536,
      "epoch": 1.988255033557047,
      "grad_norm": 2.375,
      "learning_rate": 0.00046041679763204625,
      "loss": 5.7789,
      "mean_token_accuracy": 0.17821266651153564,
      "num_tokens": 41981773.0,
      "step": 22515
    },
    {
      "entropy": 6.136757802963257,
      "epoch": 1.9886965736488873,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004603989564694957,
      "loss": 5.7678,
      "mean_token_accuracy": 0.17464101314544678,
      "num_tokens": 41990893.0,
      "step": 22520
    },
    {
      "entropy": 6.004096460342407,
      "epoch": 1.9891381137407276,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004603811116750851,
      "loss": 5.7771,
      "mean_token_accuracy": 0.16848766952753066,
      "num_tokens": 42000842.0,
      "step": 22525
    },
    {
      "entropy": 6.125044965744019,
      "epoch": 1.989579653832568,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004603632632491639,
      "loss": 5.7614,
      "mean_token_accuracy": 0.17467122972011567,
      "num_tokens": 42009430.0,
      "step": 22530
    },
    {
      "entropy": 6.078407955169678,
      "epoch": 1.9900211939244083,
      "grad_norm": 2.53125,
      "learning_rate": 0.00046034541119208177,
      "loss": 5.8965,
      "mean_token_accuracy": 0.17102467268705368,
      "num_tokens": 42018985.0,
      "step": 22535
    },
    {
      "entropy": 6.071476459503174,
      "epoch": 1.9904627340162486,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046032755550418845,
      "loss": 5.7754,
      "mean_token_accuracy": 0.16679833382368087,
      "num_tokens": 42028885.0,
      "step": 22540
    },
    {
      "entropy": 6.1595601558685305,
      "epoch": 1.9909042741080891,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004603096961858338,
      "loss": 5.8783,
      "mean_token_accuracy": 0.16122772693634033,
      "num_tokens": 42037838.0,
      "step": 22545
    },
    {
      "entropy": 6.098144578933716,
      "epoch": 1.9913458141999294,
      "grad_norm": 1.859375,
      "learning_rate": 0.00046029183323736755,
      "loss": 5.7576,
      "mean_token_accuracy": 0.16951230466365813,
      "num_tokens": 42047212.0,
      "step": 22550
    },
    {
      "entropy": 6.092751216888428,
      "epoch": 1.9917873542917697,
      "grad_norm": 3.0,
      "learning_rate": 0.0004602739666591397,
      "loss": 5.8177,
      "mean_token_accuracy": 0.1750197485089302,
      "num_tokens": 42056680.0,
      "step": 22555
    },
    {
      "entropy": 6.032433128356933,
      "epoch": 1.9922288943836102,
      "grad_norm": 3.078125,
      "learning_rate": 0.00046025609645150014,
      "loss": 5.8357,
      "mean_token_accuracy": 0.16915468871593475,
      "num_tokens": 42065604.0,
      "step": 22560
    },
    {
      "entropy": 6.099498891830445,
      "epoch": 1.9926704344754502,
      "grad_norm": 2.28125,
      "learning_rate": 0.000460238222614799,
      "loss": 5.7735,
      "mean_token_accuracy": 0.17018410116434096,
      "num_tokens": 42074672.0,
      "step": 22565
    },
    {
      "entropy": 6.0783899307250975,
      "epoch": 1.9931119745672907,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004602203451493865,
      "loss": 5.7381,
      "mean_token_accuracy": 0.1713005393743515,
      "num_tokens": 42083586.0,
      "step": 22570
    },
    {
      "entropy": 6.044497156143189,
      "epoch": 1.993553514659131,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046020246405561283,
      "loss": 5.7359,
      "mean_token_accuracy": 0.17378671020269393,
      "num_tokens": 42092328.0,
      "step": 22575
    },
    {
      "entropy": 5.990148401260376,
      "epoch": 1.9939950547509713,
      "grad_norm": 2.359375,
      "learning_rate": 0.00046018457933382807,
      "loss": 5.7547,
      "mean_token_accuracy": 0.18044360280036925,
      "num_tokens": 42102154.0,
      "step": 22580
    },
    {
      "entropy": 6.0742066383361815,
      "epoch": 1.9944365948428118,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046016669098438286,
      "loss": 5.8196,
      "mean_token_accuracy": 0.16978509724140167,
      "num_tokens": 42111762.0,
      "step": 22585
    },
    {
      "entropy": 6.070820045471192,
      "epoch": 1.994878134934652,
      "grad_norm": 4.375,
      "learning_rate": 0.0004601487990076274,
      "loss": 5.8005,
      "mean_token_accuracy": 0.16722838282585145,
      "num_tokens": 42120631.0,
      "step": 22590
    },
    {
      "entropy": 6.040423250198364,
      "epoch": 1.9953196750264923,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004601309034039123,
      "loss": 5.6588,
      "mean_token_accuracy": 0.19342879951000214,
      "num_tokens": 42129311.0,
      "step": 22595
    },
    {
      "entropy": 6.121557521820068,
      "epoch": 1.9957612151183328,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004601130041735881,
      "loss": 5.9003,
      "mean_token_accuracy": 0.16315261870622635,
      "num_tokens": 42138735.0,
      "step": 22600
    },
    {
      "entropy": 6.203469181060791,
      "epoch": 1.996202755210173,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004600951013170054,
      "loss": 5.8539,
      "mean_token_accuracy": 0.17185495793819427,
      "num_tokens": 42148416.0,
      "step": 22605
    },
    {
      "entropy": 6.153916311264038,
      "epoch": 1.9966442953020134,
      "grad_norm": 2.484375,
      "learning_rate": 0.00046007719483451493,
      "loss": 5.7676,
      "mean_token_accuracy": 0.175704288482666,
      "num_tokens": 42157191.0,
      "step": 22610
    },
    {
      "entropy": 5.964570617675781,
      "epoch": 1.9970858353938539,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046005928472646747,
      "loss": 5.717,
      "mean_token_accuracy": 0.1875031426548958,
      "num_tokens": 42165798.0,
      "step": 22615
    },
    {
      "entropy": 6.062952756881714,
      "epoch": 1.9975273754856941,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046004137099321386,
      "loss": 5.825,
      "mean_token_accuracy": 0.16738241314888,
      "num_tokens": 42175318.0,
      "step": 22620
    },
    {
      "entropy": 6.154802036285401,
      "epoch": 1.9979689155775344,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046002345363510506,
      "loss": 5.8417,
      "mean_token_accuracy": 0.16775001138448714,
      "num_tokens": 42184800.0,
      "step": 22625
    },
    {
      "entropy": 6.115027809143067,
      "epoch": 1.998410455669375,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004600055326524921,
      "loss": 5.6938,
      "mean_token_accuracy": 0.17703184932470323,
      "num_tokens": 42194172.0,
      "step": 22630
    },
    {
      "entropy": 5.981658506393432,
      "epoch": 1.998851995761215,
      "grad_norm": 2.625,
      "learning_rate": 0.00045998760804572583,
      "loss": 5.7374,
      "mean_token_accuracy": 0.18031188398599624,
      "num_tokens": 42203102.0,
      "step": 22635
    },
    {
      "entropy": 5.9814869403839115,
      "epoch": 1.9992935358530555,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045996967981515763,
      "loss": 5.6889,
      "mean_token_accuracy": 0.17985379844903945,
      "num_tokens": 42212421.0,
      "step": 22640
    },
    {
      "entropy": 6.124047040939331,
      "epoch": 1.9997350759448957,
      "grad_norm": 2.390625,
      "learning_rate": 0.00045995174796113856,
      "loss": 5.8007,
      "mean_token_accuracy": 0.169218248128891,
      "num_tokens": 42221528.0,
      "step": 22645
    },
    {
      "entropy": 6.09119348526001,
      "epoch": 2.000176616036736,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00045993381248402,
      "loss": 5.7476,
      "mean_token_accuracy": 0.17077835351228715,
      "num_tokens": 42230628.0,
      "step": 22650
    },
    {
      "entropy": 6.0533284664154055,
      "epoch": 2.0006181561285765,
      "grad_norm": 1.75,
      "learning_rate": 0.00045991587338415306,
      "loss": 5.7322,
      "mean_token_accuracy": 0.17342145889997482,
      "num_tokens": 42239571.0,
      "step": 22655
    },
    {
      "entropy": 6.073413896560669,
      "epoch": 2.001059696220417,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045989793066188953,
      "loss": 5.7114,
      "mean_token_accuracy": 0.17204705774784088,
      "num_tokens": 42248010.0,
      "step": 22660
    },
    {
      "entropy": 6.018349933624267,
      "epoch": 2.001501236312257,
      "grad_norm": 2.21875,
      "learning_rate": 0.00045987998431758063,
      "loss": 5.683,
      "mean_token_accuracy": 0.1771262362599373,
      "num_tokens": 42257251.0,
      "step": 22665
    },
    {
      "entropy": 6.063322830200195,
      "epoch": 2.0019427764040976,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004598620343515779,
      "loss": 5.8085,
      "mean_token_accuracy": 0.1723163202404976,
      "num_tokens": 42266441.0,
      "step": 22670
    },
    {
      "entropy": 6.061972999572754,
      "epoch": 2.0023843164959376,
      "grad_norm": 3.3125,
      "learning_rate": 0.00045984408076423313,
      "loss": 5.7387,
      "mean_token_accuracy": 0.17668651789426804,
      "num_tokens": 42275995.0,
      "step": 22675
    },
    {
      "entropy": 6.123283433914184,
      "epoch": 2.002825856587778,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004598261235558979,
      "loss": 5.7491,
      "mean_token_accuracy": 0.17586032301187515,
      "num_tokens": 42284065.0,
      "step": 22680
    },
    {
      "entropy": 6.042110252380371,
      "epoch": 2.0032673966796186,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045980816272692403,
      "loss": 5.8238,
      "mean_token_accuracy": 0.1735433578491211,
      "num_tokens": 42295309.0,
      "step": 22685
    },
    {
      "entropy": 6.1149547576904295,
      "epoch": 2.0037089367714587,
      "grad_norm": 1.75,
      "learning_rate": 0.00045979019827766345,
      "loss": 5.7217,
      "mean_token_accuracy": 0.17594785690307618,
      "num_tokens": 42304390.0,
      "step": 22690
    },
    {
      "entropy": 6.086695146560669,
      "epoch": 2.004150476863299,
      "grad_norm": 2.4375,
      "learning_rate": 0.00045977223020846785,
      "loss": 5.7722,
      "mean_token_accuracy": 0.17499068081378938,
      "num_tokens": 42313830.0,
      "step": 22695
    },
    {
      "entropy": 6.069741153717041,
      "epoch": 2.0045920169551397,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004597542585196895,
      "loss": 5.7311,
      "mean_token_accuracy": 0.17245448529720306,
      "num_tokens": 42323827.0,
      "step": 22700
    },
    {
      "entropy": 6.146077966690063,
      "epoch": 2.0050335570469797,
      "grad_norm": 1.953125,
      "learning_rate": 0.00045973628321168016,
      "loss": 5.8588,
      "mean_token_accuracy": 0.1672836422920227,
      "num_tokens": 42333221.0,
      "step": 22705
    },
    {
      "entropy": 6.023550748825073,
      "epoch": 2.00547509713882,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045971830428479215,
      "loss": 5.7279,
      "mean_token_accuracy": 0.17514718025922776,
      "num_tokens": 42343036.0,
      "step": 22710
    },
    {
      "entropy": 6.0486937999725345,
      "epoch": 2.0059166372306607,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00045970032173937766,
      "loss": 5.6984,
      "mean_token_accuracy": 0.18180923759937287,
      "num_tokens": 42351916.0,
      "step": 22715
    },
    {
      "entropy": 6.12478814125061,
      "epoch": 2.0063581773225008,
      "grad_norm": 3.640625,
      "learning_rate": 0.00045968233557578883,
      "loss": 5.8354,
      "mean_token_accuracy": 0.16859125196933747,
      "num_tokens": 42362412.0,
      "step": 22720
    },
    {
      "entropy": 6.065492963790893,
      "epoch": 2.0067997174143413,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004596643457943781,
      "loss": 5.7108,
      "mean_token_accuracy": 0.18018664866685868,
      "num_tokens": 42372773.0,
      "step": 22725
    },
    {
      "entropy": 6.110373640060425,
      "epoch": 2.0072412575061818,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004596463523954979,
      "loss": 5.6979,
      "mean_token_accuracy": 0.17578907459974288,
      "num_tokens": 42382009.0,
      "step": 22730
    },
    {
      "entropy": 6.050855445861816,
      "epoch": 2.007682797598022,
      "grad_norm": 2.6875,
      "learning_rate": 0.00045962835537950063,
      "loss": 5.7147,
      "mean_token_accuracy": 0.1729799136519432,
      "num_tokens": 42390961.0,
      "step": 22735
    },
    {
      "entropy": 5.954950952529908,
      "epoch": 2.0081243376898623,
      "grad_norm": 2.515625,
      "learning_rate": 0.00045961035474673893,
      "loss": 5.712,
      "mean_token_accuracy": 0.17625963687896729,
      "num_tokens": 42400483.0,
      "step": 22740
    },
    {
      "entropy": 5.939000988006592,
      "epoch": 2.0085658777817024,
      "grad_norm": 2.421875,
      "learning_rate": 0.00045959235049756537,
      "loss": 5.5723,
      "mean_token_accuracy": 0.18988077342510223,
      "num_tokens": 42409285.0,
      "step": 22745
    },
    {
      "entropy": 6.078508186340332,
      "epoch": 2.009007417873543,
      "grad_norm": 2.640625,
      "learning_rate": 0.00045957434263233253,
      "loss": 5.7683,
      "mean_token_accuracy": 0.1714036449790001,
      "num_tokens": 42418950.0,
      "step": 22750
    },
    {
      "entropy": 6.08935136795044,
      "epoch": 2.0094489579653834,
      "grad_norm": 3.125,
      "learning_rate": 0.00045955633115139343,
      "loss": 5.832,
      "mean_token_accuracy": 0.16589801162481307,
      "num_tokens": 42428285.0,
      "step": 22755
    },
    {
      "entropy": 6.0794212341308596,
      "epoch": 2.0098904980572234,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004595383160551007,
      "loss": 5.7179,
      "mean_token_accuracy": 0.17384958416223525,
      "num_tokens": 42437832.0,
      "step": 22760
    },
    {
      "entropy": 6.0187225341796875,
      "epoch": 2.010332038149064,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004595202973438074,
      "loss": 5.7289,
      "mean_token_accuracy": 0.1753165304660797,
      "num_tokens": 42447704.0,
      "step": 22765
    },
    {
      "entropy": 6.094425964355469,
      "epoch": 2.0107735782409044,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004595022750178663,
      "loss": 5.7921,
      "mean_token_accuracy": 0.1714656502008438,
      "num_tokens": 42456571.0,
      "step": 22770
    },
    {
      "entropy": 6.105397462844849,
      "epoch": 2.0112151183327445,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004594842490776306,
      "loss": 5.7125,
      "mean_token_accuracy": 0.17677250355482102,
      "num_tokens": 42466522.0,
      "step": 22775
    },
    {
      "entropy": 6.061077547073364,
      "epoch": 2.011656658424585,
      "grad_norm": 2.0,
      "learning_rate": 0.00045946621952345335,
      "loss": 5.7157,
      "mean_token_accuracy": 0.16929714679718016,
      "num_tokens": 42475940.0,
      "step": 22780
    },
    {
      "entropy": 6.138867712020874,
      "epoch": 2.0120981985164255,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004594481863556879,
      "loss": 5.815,
      "mean_token_accuracy": 0.16873244643211366,
      "num_tokens": 42486456.0,
      "step": 22785
    },
    {
      "entropy": 6.104576873779297,
      "epoch": 2.0125397386082655,
      "grad_norm": 2.125,
      "learning_rate": 0.00045943014957468726,
      "loss": 5.7917,
      "mean_token_accuracy": 0.17428668588399887,
      "num_tokens": 42495578.0,
      "step": 22790
    },
    {
      "entropy": 6.0331292152404785,
      "epoch": 2.012981278700106,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004594121091808049,
      "loss": 5.6301,
      "mean_token_accuracy": 0.18230628222227097,
      "num_tokens": 42504532.0,
      "step": 22795
    },
    {
      "entropy": 6.085680389404297,
      "epoch": 2.0134228187919465,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045939406517439427,
      "loss": 5.6892,
      "mean_token_accuracy": 0.1717018410563469,
      "num_tokens": 42513683.0,
      "step": 22800
    },
    {
      "entropy": 6.062415981292725,
      "epoch": 2.0138643588837866,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00045937601755580864,
      "loss": 5.777,
      "mean_token_accuracy": 0.1723837822675705,
      "num_tokens": 42522717.0,
      "step": 22805
    },
    {
      "entropy": 6.0807606220245365,
      "epoch": 2.014305898975627,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004593579663254018,
      "loss": 5.8028,
      "mean_token_accuracy": 0.1724116712808609,
      "num_tokens": 42531847.0,
      "step": 22810
    },
    {
      "entropy": 6.040658092498779,
      "epoch": 2.014747439067467,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004593399114835272,
      "loss": 5.7314,
      "mean_token_accuracy": 0.17662145644426347,
      "num_tokens": 42540869.0,
      "step": 22815
    },
    {
      "entropy": 6.058356809616089,
      "epoch": 2.0151889791593076,
      "grad_norm": 2.25,
      "learning_rate": 0.00045932185303053857,
      "loss": 5.713,
      "mean_token_accuracy": 0.17783579975366592,
      "num_tokens": 42550875.0,
      "step": 22820
    },
    {
      "entropy": 6.089735746383667,
      "epoch": 2.015630519251148,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004593037909667898,
      "loss": 5.6905,
      "mean_token_accuracy": 0.1799245446920395,
      "num_tokens": 42560066.0,
      "step": 22825
    },
    {
      "entropy": 5.960052633285523,
      "epoch": 2.016072059342988,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004592857252926344,
      "loss": 5.7166,
      "mean_token_accuracy": 0.17549924552440643,
      "num_tokens": 42568733.0,
      "step": 22830
    },
    {
      "entropy": 6.034108972549438,
      "epoch": 2.0165135994348287,
      "grad_norm": 2.875,
      "learning_rate": 0.00045926765600842653,
      "loss": 5.7358,
      "mean_token_accuracy": 0.17844226509332656,
      "num_tokens": 42577565.0,
      "step": 22835
    },
    {
      "entropy": 6.078424024581909,
      "epoch": 2.016955139526669,
      "grad_norm": 3.609375,
      "learning_rate": 0.00045924958311452006,
      "loss": 5.7788,
      "mean_token_accuracy": 0.17431116104125977,
      "num_tokens": 42586489.0,
      "step": 22840
    },
    {
      "entropy": 6.0531532764434814,
      "epoch": 2.017396679618509,
      "grad_norm": 2.453125,
      "learning_rate": 0.00045923150661126904,
      "loss": 5.8037,
      "mean_token_accuracy": 0.17136487662792205,
      "num_tokens": 42596263.0,
      "step": 22845
    },
    {
      "entropy": 6.056074380874634,
      "epoch": 2.0178382197103497,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004592134264990276,
      "loss": 5.7266,
      "mean_token_accuracy": 0.17534229457378386,
      "num_tokens": 42605787.0,
      "step": 22850
    },
    {
      "entropy": 6.164589262008667,
      "epoch": 2.01827975980219,
      "grad_norm": 2.4375,
      "learning_rate": 0.00045919534277814996,
      "loss": 5.8871,
      "mean_token_accuracy": 0.1654741570353508,
      "num_tokens": 42616014.0,
      "step": 22855
    },
    {
      "entropy": 6.09309720993042,
      "epoch": 2.0187212998940303,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004591772554489902,
      "loss": 5.7622,
      "mean_token_accuracy": 0.17920698076486588,
      "num_tokens": 42624839.0,
      "step": 22860
    },
    {
      "entropy": 6.109853935241699,
      "epoch": 2.0191628399858708,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004591591645119028,
      "loss": 5.7766,
      "mean_token_accuracy": 0.1702759936451912,
      "num_tokens": 42633811.0,
      "step": 22865
    },
    {
      "entropy": 5.983197021484375,
      "epoch": 2.0196043800777113,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045914106996724217,
      "loss": 5.6739,
      "mean_token_accuracy": 0.1870922029018402,
      "num_tokens": 42643140.0,
      "step": 22870
    },
    {
      "entropy": 6.03023362159729,
      "epoch": 2.0200459201695513,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00045912297181536257,
      "loss": 5.6566,
      "mean_token_accuracy": 0.17909153699874877,
      "num_tokens": 42651123.0,
      "step": 22875
    },
    {
      "entropy": 6.026482582092285,
      "epoch": 2.020487460261392,
      "grad_norm": 1.578125,
      "learning_rate": 0.00045910487005661877,
      "loss": 5.7103,
      "mean_token_accuracy": 0.18610048443078994,
      "num_tokens": 42659880.0,
      "step": 22880
    },
    {
      "entropy": 6.016549587249756,
      "epoch": 2.020929000353232,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004590867646913653,
      "loss": 5.6686,
      "mean_token_accuracy": 0.18456330448389052,
      "num_tokens": 42668689.0,
      "step": 22885
    },
    {
      "entropy": 6.0607297897338865,
      "epoch": 2.0213705404450724,
      "grad_norm": 2.328125,
      "learning_rate": 0.00045906865571995666,
      "loss": 5.7204,
      "mean_token_accuracy": 0.18061990141868592,
      "num_tokens": 42678403.0,
      "step": 22890
    },
    {
      "entropy": 6.022908306121826,
      "epoch": 2.021812080536913,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004590505431427479,
      "loss": 5.7307,
      "mean_token_accuracy": 0.1772811532020569,
      "num_tokens": 42688203.0,
      "step": 22895
    },
    {
      "entropy": 6.107380676269531,
      "epoch": 2.022253620628753,
      "grad_norm": 6.6875,
      "learning_rate": 0.0004590324269600936,
      "loss": 5.7909,
      "mean_token_accuracy": 0.16444470584392548,
      "num_tokens": 42698343.0,
      "step": 22900
    },
    {
      "entropy": 6.037564373016357,
      "epoch": 2.0226951607205934,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045901430717234875,
      "loss": 5.6401,
      "mean_token_accuracy": 0.18040127754211427,
      "num_tokens": 42707483.0,
      "step": 22905
    },
    {
      "entropy": 6.070845222473144,
      "epoch": 2.023136700812434,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004589961837798682,
      "loss": 5.7943,
      "mean_token_accuracy": 0.17378006428480147,
      "num_tokens": 42717820.0,
      "step": 22910
    },
    {
      "entropy": 6.1042928218841555,
      "epoch": 2.023578240904274,
      "grad_norm": 2.390625,
      "learning_rate": 0.00045897805678300714,
      "loss": 5.7627,
      "mean_token_accuracy": 0.17683238983154298,
      "num_tokens": 42726996.0,
      "step": 22915
    },
    {
      "entropy": 6.119480562210083,
      "epoch": 2.0240197809961145,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004589599261821205,
      "loss": 5.7291,
      "mean_token_accuracy": 0.17859240174293517,
      "num_tokens": 42735779.0,
      "step": 22920
    },
    {
      "entropy": 6.09844012260437,
      "epoch": 2.024461321087955,
      "grad_norm": 2.375,
      "learning_rate": 0.00045894179197756364,
      "loss": 5.8371,
      "mean_token_accuracy": 0.1681301176548004,
      "num_tokens": 42745089.0,
      "step": 22925
    },
    {
      "entropy": 6.0906110286712645,
      "epoch": 2.024902861179795,
      "grad_norm": 1.90625,
      "learning_rate": 0.00045892365416969164,
      "loss": 5.7936,
      "mean_token_accuracy": 0.1692209631204605,
      "num_tokens": 42754963.0,
      "step": 22930
    },
    {
      "entropy": 6.188189172744751,
      "epoch": 2.0253444012716355,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045890551275885985,
      "loss": 5.8243,
      "mean_token_accuracy": 0.16632246673107148,
      "num_tokens": 42764856.0,
      "step": 22935
    },
    {
      "entropy": 6.138555574417114,
      "epoch": 2.025785941363476,
      "grad_norm": 3.765625,
      "learning_rate": 0.00045888736774542363,
      "loss": 5.7805,
      "mean_token_accuracy": 0.1727130427956581,
      "num_tokens": 42775028.0,
      "step": 22940
    },
    {
      "entropy": 6.045684003829956,
      "epoch": 2.026227481455316,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004588692191297385,
      "loss": 5.7558,
      "mean_token_accuracy": 0.17669824212789537,
      "num_tokens": 42784288.0,
      "step": 22945
    },
    {
      "entropy": 6.11463189125061,
      "epoch": 2.0266690215471566,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004588510669121599,
      "loss": 5.8134,
      "mean_token_accuracy": 0.16901011765003204,
      "num_tokens": 42793356.0,
      "step": 22950
    },
    {
      "entropy": 6.0672831535339355,
      "epoch": 2.0271105616389966,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045883291109304346,
      "loss": 5.6626,
      "mean_token_accuracy": 0.1785956621170044,
      "num_tokens": 42802287.0,
      "step": 22955
    },
    {
      "entropy": 6.051865530014038,
      "epoch": 2.027552101730837,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004588147516727449,
      "loss": 5.8078,
      "mean_token_accuracy": 0.1705307751893997,
      "num_tokens": 42812206.0,
      "step": 22960
    },
    {
      "entropy": 6.130224943161011,
      "epoch": 2.0279936418226776,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004587965886516199,
      "loss": 5.7911,
      "mean_token_accuracy": 0.16724737286567687,
      "num_tokens": 42822046.0,
      "step": 22965
    },
    {
      "entropy": 6.109975147247314,
      "epoch": 2.0284351819145177,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045877842203002417,
      "loss": 5.7008,
      "mean_token_accuracy": 0.1829884558916092,
      "num_tokens": 42831228.0,
      "step": 22970
    },
    {
      "entropy": 6.061490249633789,
      "epoch": 2.028876722006358,
      "grad_norm": 1.75,
      "learning_rate": 0.00045876025180831373,
      "loss": 5.8321,
      "mean_token_accuracy": 0.16947442069649696,
      "num_tokens": 42840683.0,
      "step": 22975
    },
    {
      "entropy": 6.051962423324585,
      "epoch": 2.0293182620981987,
      "grad_norm": 2.4375,
      "learning_rate": 0.00045874207798684446,
      "loss": 5.7177,
      "mean_token_accuracy": 0.17408789992332457,
      "num_tokens": 42849731.0,
      "step": 22980
    },
    {
      "entropy": 6.075231695175171,
      "epoch": 2.0297598021900387,
      "grad_norm": 1.890625,
      "learning_rate": 0.00045872390056597235,
      "loss": 5.7595,
      "mean_token_accuracy": 0.17312418818473815,
      "num_tokens": 42858028.0,
      "step": 22985
    },
    {
      "entropy": 6.060962867736817,
      "epoch": 2.030201342281879,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004587057195460536,
      "loss": 5.7724,
      "mean_token_accuracy": 0.16680307537317277,
      "num_tokens": 42867340.0,
      "step": 22990
    },
    {
      "entropy": 6.149641513824463,
      "epoch": 2.0306428823737197,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00045868753492744424,
      "loss": 5.778,
      "mean_token_accuracy": 0.1739378109574318,
      "num_tokens": 42877006.0,
      "step": 22995
    },
    {
      "entropy": 6.071325397491455,
      "epoch": 2.0310844224655598,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004586693467105005,
      "loss": 5.6597,
      "mean_token_accuracy": 0.1767528921365738,
      "num_tokens": 42886324.0,
      "step": 23000
    },
    {
      "entropy": 6.088351535797119,
      "epoch": 2.0315259625574003,
      "grad_norm": 4.75,
      "learning_rate": 0.0004586511548955788,
      "loss": 5.756,
      "mean_token_accuracy": 0.1761959061026573,
      "num_tokens": 42896411.0,
      "step": 23005
    },
    {
      "entropy": 5.941689586639404,
      "epoch": 2.0319675026492408,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004586329594830353,
      "loss": 5.6328,
      "mean_token_accuracy": 0.18148183673620225,
      "num_tokens": 42905606.0,
      "step": 23010
    },
    {
      "entropy": 5.936578845977783,
      "epoch": 2.032409042741081,
      "grad_norm": 1.453125,
      "learning_rate": 0.00045861476047322664,
      "loss": 5.6023,
      "mean_token_accuracy": 0.1799500212073326,
      "num_tokens": 42914226.0,
      "step": 23015
    },
    {
      "entropy": 6.0961634635925295,
      "epoch": 2.0328505828329213,
      "grad_norm": 1.984375,
      "learning_rate": 0.00045859655786650925,
      "loss": 5.8176,
      "mean_token_accuracy": 0.16826072186231614,
      "num_tokens": 42923910.0,
      "step": 23020
    },
    {
      "entropy": 6.040944242477417,
      "epoch": 2.0332921229247614,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004585783516632397,
      "loss": 5.659,
      "mean_token_accuracy": 0.18811955004930497,
      "num_tokens": 42933203.0,
      "step": 23025
    },
    {
      "entropy": 6.077639150619507,
      "epoch": 2.033733663016602,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004585601418637747,
      "loss": 5.7308,
      "mean_token_accuracy": 0.17890943139791488,
      "num_tokens": 42942143.0,
      "step": 23030
    },
    {
      "entropy": 6.022352504730224,
      "epoch": 2.0341752031084424,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004585419284684708,
      "loss": 5.7559,
      "mean_token_accuracy": 0.17017291337251664,
      "num_tokens": 42951432.0,
      "step": 23035
    },
    {
      "entropy": 6.067526197433471,
      "epoch": 2.0346167432002824,
      "grad_norm": 1.765625,
      "learning_rate": 0.000458523711477685,
      "loss": 5.6811,
      "mean_token_accuracy": 0.17597204446792603,
      "num_tokens": 42961033.0,
      "step": 23040
    },
    {
      "entropy": 6.092260456085205,
      "epoch": 2.035058283292123,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00045850549089177406,
      "loss": 5.663,
      "mean_token_accuracy": 0.1890003576874733,
      "num_tokens": 42971371.0,
      "step": 23045
    },
    {
      "entropy": 6.081429862976075,
      "epoch": 2.0354998233839634,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045848726671109493,
      "loss": 5.7554,
      "mean_token_accuracy": 0.16306892037391663,
      "num_tokens": 42980445.0,
      "step": 23050
    },
    {
      "entropy": 5.9818830490112305,
      "epoch": 2.0359413634758035,
      "grad_norm": 3.046875,
      "learning_rate": 0.00045846903893600457,
      "loss": 5.7359,
      "mean_token_accuracy": 0.17906024158000947,
      "num_tokens": 42989923.0,
      "step": 23055
    },
    {
      "entropy": 5.965462970733642,
      "epoch": 2.036382903567644,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00045845080756686007,
      "loss": 5.6296,
      "mean_token_accuracy": 0.17902407497167588,
      "num_tokens": 42998370.0,
      "step": 23060
    },
    {
      "entropy": 6.1027031421661375,
      "epoch": 2.0368244436594845,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00045843257260401855,
      "loss": 5.7178,
      "mean_token_accuracy": 0.17873544096946717,
      "num_tokens": 43007641.0,
      "step": 23065
    },
    {
      "entropy": 6.015437650680542,
      "epoch": 2.0372659837513245,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045841433404783733,
      "loss": 5.6856,
      "mean_token_accuracy": 0.17976224422454834,
      "num_tokens": 43016271.0,
      "step": 23070
    },
    {
      "entropy": 5.9930487155914305,
      "epoch": 2.037707523843165,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004583960918986736,
      "loss": 5.6191,
      "mean_token_accuracy": 0.18528620451688765,
      "num_tokens": 43024525.0,
      "step": 23075
    },
    {
      "entropy": 6.020774936676025,
      "epoch": 2.0381490639350055,
      "grad_norm": 1.875,
      "learning_rate": 0.00045837784615688473,
      "loss": 5.7384,
      "mean_token_accuracy": 0.17716163843870164,
      "num_tokens": 43033248.0,
      "step": 23080
    },
    {
      "entropy": 6.152990388870239,
      "epoch": 2.0385906040268456,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004583595968228281,
      "loss": 5.9003,
      "mean_token_accuracy": 0.1673475667834282,
      "num_tokens": 43042171.0,
      "step": 23085
    },
    {
      "entropy": 6.118916654586792,
      "epoch": 2.039032144118686,
      "grad_norm": 3.046875,
      "learning_rate": 0.00045834134389686137,
      "loss": 5.7064,
      "mean_token_accuracy": 0.18097981363534926,
      "num_tokens": 43051447.0,
      "step": 23090
    },
    {
      "entropy": 6.077790117263794,
      "epoch": 2.039473684210526,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004583230873793419,
      "loss": 5.6696,
      "mean_token_accuracy": 0.171328504383564,
      "num_tokens": 43059869.0,
      "step": 23095
    },
    {
      "entropy": 5.905318927764893,
      "epoch": 2.0399152243023666,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00045830482727062736,
      "loss": 5.621,
      "mean_token_accuracy": 0.18191699534654618,
      "num_tokens": 43068412.0,
      "step": 23100
    },
    {
      "entropy": 6.061247253417969,
      "epoch": 2.040356764394207,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045828656357107557,
      "loss": 5.7605,
      "mean_token_accuracy": 0.1759234994649887,
      "num_tokens": 43077130.0,
      "step": 23105
    },
    {
      "entropy": 6.183452606201172,
      "epoch": 2.040798304486047,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004582682962810442,
      "loss": 5.7781,
      "mean_token_accuracy": 0.1705264613032341,
      "num_tokens": 43086304.0,
      "step": 23110
    },
    {
      "entropy": 6.106404161453247,
      "epoch": 2.0412398445778877,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004582500254008912,
      "loss": 5.7366,
      "mean_token_accuracy": 0.1828863188624382,
      "num_tokens": 43095734.0,
      "step": 23115
    },
    {
      "entropy": 6.087597322463989,
      "epoch": 2.041681384669728,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00045823175093097444,
      "loss": 5.7174,
      "mean_token_accuracy": 0.17634183019399643,
      "num_tokens": 43104761.0,
      "step": 23120
    },
    {
      "entropy": 6.061728286743164,
      "epoch": 2.042122924761568,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004582134728716518,
      "loss": 5.7552,
      "mean_token_accuracy": 0.17494938820600509,
      "num_tokens": 43114182.0,
      "step": 23125
    },
    {
      "entropy": 5.991589117050171,
      "epoch": 2.0425644648534087,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004581951912232814,
      "loss": 5.7206,
      "mean_token_accuracy": 0.17611633241176605,
      "num_tokens": 43124893.0,
      "step": 23130
    },
    {
      "entropy": 6.103698253631592,
      "epoch": 2.043006004945249,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004581769059862215,
      "loss": 5.7336,
      "mean_token_accuracy": 0.17771054357290267,
      "num_tokens": 43133514.0,
      "step": 23135
    },
    {
      "entropy": 6.104642534255982,
      "epoch": 2.0434475450370893,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004581586171608301,
      "loss": 5.7908,
      "mean_token_accuracy": 0.17167143970727922,
      "num_tokens": 43142082.0,
      "step": 23140
    },
    {
      "entropy": 6.034381914138794,
      "epoch": 2.0438890851289298,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004581403247474657,
      "loss": 5.7592,
      "mean_token_accuracy": 0.16672576516866683,
      "num_tokens": 43151466.0,
      "step": 23145
    },
    {
      "entropy": 6.079020357131958,
      "epoch": 2.0443306252207702,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004581220287464864,
      "loss": 5.8361,
      "mean_token_accuracy": 0.1654016926884651,
      "num_tokens": 43160631.0,
      "step": 23150
    },
    {
      "entropy": 6.093832969665527,
      "epoch": 2.0447721653126103,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004581037291582507,
      "loss": 5.812,
      "mean_token_accuracy": 0.17410966455936433,
      "num_tokens": 43170968.0,
      "step": 23155
    },
    {
      "entropy": 6.084896516799927,
      "epoch": 2.045213705404451,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004580854259831171,
      "loss": 5.7312,
      "mean_token_accuracy": 0.17373128831386567,
      "num_tokens": 43179661.0,
      "step": 23160
    },
    {
      "entropy": 6.020966196060181,
      "epoch": 2.045655245496291,
      "grad_norm": 2.375,
      "learning_rate": 0.0004580671192214441,
      "loss": 5.5834,
      "mean_token_accuracy": 0.18548309355974196,
      "num_tokens": 43188994.0,
      "step": 23165
    },
    {
      "entropy": 6.1175065517425535,
      "epoch": 2.0460967855881314,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00045804880887359036,
      "loss": 5.8535,
      "mean_token_accuracy": 0.16565000712871553,
      "num_tokens": 43199001.0,
      "step": 23170
    },
    {
      "entropy": 6.081899785995484,
      "epoch": 2.046538325679972,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004580304949399146,
      "loss": 5.7069,
      "mean_token_accuracy": 0.17866305708885194,
      "num_tokens": 43208113.0,
      "step": 23175
    },
    {
      "entropy": 6.022565269470215,
      "epoch": 2.046979865771812,
      "grad_norm": 2.21875,
      "learning_rate": 0.00045801217742077544,
      "loss": 5.7181,
      "mean_token_accuracy": 0.17901778370141982,
      "num_tokens": 43216983.0,
      "step": 23180
    },
    {
      "entropy": 6.004829788208008,
      "epoch": 2.0474214058636524,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004579938563165319,
      "loss": 5.6893,
      "mean_token_accuracy": 0.18421900123357773,
      "num_tokens": 43225965.0,
      "step": 23185
    },
    {
      "entropy": 6.078584241867065,
      "epoch": 2.047862945955493,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004579755316275427,
      "loss": 5.6356,
      "mean_token_accuracy": 0.183344866335392,
      "num_tokens": 43235078.0,
      "step": 23190
    },
    {
      "entropy": 6.015656423568726,
      "epoch": 2.048304486047333,
      "grad_norm": 1.765625,
      "learning_rate": 0.000457957203354167,
      "loss": 5.762,
      "mean_token_accuracy": 0.1789546236395836,
      "num_tokens": 43244398.0,
      "step": 23195
    },
    {
      "entropy": 6.08344202041626,
      "epoch": 2.0487460261391734,
      "grad_norm": 3.96875,
      "learning_rate": 0.00045793887149676366,
      "loss": 5.8025,
      "mean_token_accuracy": 0.17441286742687226,
      "num_tokens": 43254638.0,
      "step": 23200
    },
    {
      "entropy": 6.055268049240112,
      "epoch": 2.049187566231014,
      "grad_norm": 1.90625,
      "learning_rate": 0.000457920536055692,
      "loss": 5.6885,
      "mean_token_accuracy": 0.1793655961751938,
      "num_tokens": 43263287.0,
      "step": 23205
    },
    {
      "entropy": 6.047469758987427,
      "epoch": 2.049629106322854,
      "grad_norm": 2.1875,
      "learning_rate": 0.00045790219703131097,
      "loss": 5.6568,
      "mean_token_accuracy": 0.18701719045639037,
      "num_tokens": 43272921.0,
      "step": 23210
    },
    {
      "entropy": 6.070576524734497,
      "epoch": 2.0500706464146945,
      "grad_norm": 2.4375,
      "learning_rate": 0.00045788385442397994,
      "loss": 5.7386,
      "mean_token_accuracy": 0.17656859010457993,
      "num_tokens": 43283423.0,
      "step": 23215
    },
    {
      "entropy": 6.142502307891846,
      "epoch": 2.050512186506535,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004578655082340582,
      "loss": 5.7936,
      "mean_token_accuracy": 0.17540356665849685,
      "num_tokens": 43293031.0,
      "step": 23220
    },
    {
      "entropy": 6.081293249130249,
      "epoch": 2.050953726598375,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004578471584619051,
      "loss": 5.7598,
      "mean_token_accuracy": 0.17727106362581252,
      "num_tokens": 43302775.0,
      "step": 23225
    },
    {
      "entropy": 5.974957799911499,
      "epoch": 2.0513952666902155,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004578288051078803,
      "loss": 5.6879,
      "mean_token_accuracy": 0.1815493643283844,
      "num_tokens": 43312341.0,
      "step": 23230
    },
    {
      "entropy": 6.093283700942993,
      "epoch": 2.0518368067820556,
      "grad_norm": 1.9375,
      "learning_rate": 0.00045781044817234305,
      "loss": 5.7468,
      "mean_token_accuracy": 0.1726027548313141,
      "num_tokens": 43321358.0,
      "step": 23235
    },
    {
      "entropy": 6.0529505729675295,
      "epoch": 2.052278346873896,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045779208765565324,
      "loss": 5.7256,
      "mean_token_accuracy": 0.175204998254776,
      "num_tokens": 43331839.0,
      "step": 23240
    },
    {
      "entropy": 6.048745632171631,
      "epoch": 2.0527198869657366,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004577737235581702,
      "loss": 5.8097,
      "mean_token_accuracy": 0.1690050408244133,
      "num_tokens": 43340384.0,
      "step": 23245
    },
    {
      "entropy": 6.154810905456543,
      "epoch": 2.0531614270575766,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00045775535588025404,
      "loss": 5.761,
      "mean_token_accuracy": 0.17184665352106093,
      "num_tokens": 43349264.0,
      "step": 23250
    },
    {
      "entropy": 6.079110765457154,
      "epoch": 2.053602967149417,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004577369846222643,
      "loss": 5.7521,
      "mean_token_accuracy": 0.1744764819741249,
      "num_tokens": 43358598.0,
      "step": 23255
    },
    {
      "entropy": 6.025182580947876,
      "epoch": 2.0540445072412576,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004577186097845609,
      "loss": 5.7184,
      "mean_token_accuracy": 0.17291803807020187,
      "num_tokens": 43368357.0,
      "step": 23260
    },
    {
      "entropy": 6.152990245819092,
      "epoch": 2.0544860473330977,
      "grad_norm": 1.765625,
      "learning_rate": 0.00045770023136750405,
      "loss": 5.8176,
      "mean_token_accuracy": 0.1752984717488289,
      "num_tokens": 43378143.0,
      "step": 23265
    },
    {
      "entropy": 6.064139032363892,
      "epoch": 2.054927587424938,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004576818493714534,
      "loss": 5.6914,
      "mean_token_accuracy": 0.17971962243318557,
      "num_tokens": 43387010.0,
      "step": 23270
    },
    {
      "entropy": 5.925927734375,
      "epoch": 2.0553691275167787,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004576634637967693,
      "loss": 5.5908,
      "mean_token_accuracy": 0.18118612915277482,
      "num_tokens": 43395864.0,
      "step": 23275
    },
    {
      "entropy": 6.0475969314575195,
      "epoch": 2.0558106676086187,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004576450746438118,
      "loss": 5.6757,
      "mean_token_accuracy": 0.17749825417995452,
      "num_tokens": 43404662.0,
      "step": 23280
    },
    {
      "entropy": 6.095426368713379,
      "epoch": 2.0562522077004592,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004576266819129412,
      "loss": 5.747,
      "mean_token_accuracy": 0.17489632815122605,
      "num_tokens": 43413927.0,
      "step": 23285
    },
    {
      "entropy": 6.045530605316162,
      "epoch": 2.0566937477922997,
      "grad_norm": 1.75,
      "learning_rate": 0.0004576082856045177,
      "loss": 5.8735,
      "mean_token_accuracy": 0.16151500195264817,
      "num_tokens": 43424197.0,
      "step": 23290
    },
    {
      "entropy": 6.136016225814819,
      "epoch": 2.05713528788414,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00045758988571890176,
      "loss": 5.7541,
      "mean_token_accuracy": 0.1698651686310768,
      "num_tokens": 43432630.0,
      "step": 23295
    },
    {
      "entropy": 6.036664819717407,
      "epoch": 2.0575768279759803,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00045757148225645383,
      "loss": 5.7341,
      "mean_token_accuracy": 0.1699366107583046,
      "num_tokens": 43442186.0,
      "step": 23300
    },
    {
      "entropy": 6.015429925918579,
      "epoch": 2.0580183680678203,
      "grad_norm": 2.5625,
      "learning_rate": 0.00045755307521753443,
      "loss": 5.698,
      "mean_token_accuracy": 0.18458699733018874,
      "num_tokens": 43451927.0,
      "step": 23305
    },
    {
      "entropy": 6.109522819519043,
      "epoch": 2.058459908159661,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00045753466460250405,
      "loss": 5.7139,
      "mean_token_accuracy": 0.17335239052772522,
      "num_tokens": 43460879.0,
      "step": 23310
    },
    {
      "entropy": 6.0430684089660645,
      "epoch": 2.0589014482515013,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004575162504117234,
      "loss": 5.8151,
      "mean_token_accuracy": 0.16433346271514893,
      "num_tokens": 43469918.0,
      "step": 23315
    },
    {
      "entropy": 6.055506086349487,
      "epoch": 2.0593429883433414,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004574978326455532,
      "loss": 5.7209,
      "mean_token_accuracy": 0.17833346873521805,
      "num_tokens": 43478779.0,
      "step": 23320
    },
    {
      "entropy": 6.100334739685058,
      "epoch": 2.059784528435182,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004574794113043543,
      "loss": 5.8,
      "mean_token_accuracy": 0.17284142076969147,
      "num_tokens": 43488128.0,
      "step": 23325
    },
    {
      "entropy": 6.00214376449585,
      "epoch": 2.0602260685270224,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004574609863884875,
      "loss": 5.7109,
      "mean_token_accuracy": 0.17653937339782716,
      "num_tokens": 43497567.0,
      "step": 23330
    },
    {
      "entropy": 6.068548011779785,
      "epoch": 2.0606676086188624,
      "grad_norm": 3.640625,
      "learning_rate": 0.00045744255789831377,
      "loss": 5.7619,
      "mean_token_accuracy": 0.16977661103010178,
      "num_tokens": 43507360.0,
      "step": 23335
    },
    {
      "entropy": 6.107602453231811,
      "epoch": 2.061109148710703,
      "grad_norm": 2.125,
      "learning_rate": 0.00045742412583419403,
      "loss": 5.7521,
      "mean_token_accuracy": 0.1776425302028656,
      "num_tokens": 43517601.0,
      "step": 23340
    },
    {
      "entropy": 6.0288983345031735,
      "epoch": 2.0615506888025434,
      "grad_norm": 1.703125,
      "learning_rate": 0.00045740569019648945,
      "loss": 5.6776,
      "mean_token_accuracy": 0.18459907919168472,
      "num_tokens": 43526348.0,
      "step": 23345
    },
    {
      "entropy": 6.032813978195191,
      "epoch": 2.0619922288943835,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004573872509855612,
      "loss": 5.6802,
      "mean_token_accuracy": 0.1812094658613205,
      "num_tokens": 43536106.0,
      "step": 23350
    },
    {
      "entropy": 6.110356140136719,
      "epoch": 2.062433768986224,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004573688082017703,
      "loss": 5.8124,
      "mean_token_accuracy": 0.1702081710100174,
      "num_tokens": 43545110.0,
      "step": 23355
    },
    {
      "entropy": 6.053924703598023,
      "epoch": 2.0628753090780645,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045735036184547825,
      "loss": 5.7242,
      "mean_token_accuracy": 0.17509303241968155,
      "num_tokens": 43554226.0,
      "step": 23360
    },
    {
      "entropy": 6.055877065658569,
      "epoch": 2.0633168491699045,
      "grad_norm": 1.953125,
      "learning_rate": 0.00045733191191704633,
      "loss": 5.7448,
      "mean_token_accuracy": 0.18178366720676423,
      "num_tokens": 43563184.0,
      "step": 23365
    },
    {
      "entropy": 6.0847385883331295,
      "epoch": 2.063758389261745,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004573134584168359,
      "loss": 5.7151,
      "mean_token_accuracy": 0.1863383635878563,
      "num_tokens": 43572595.0,
      "step": 23370
    },
    {
      "entropy": 6.062539958953858,
      "epoch": 2.064199929353585,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004572950013452086,
      "loss": 5.8124,
      "mean_token_accuracy": 0.16977320462465287,
      "num_tokens": 43582211.0,
      "step": 23375
    },
    {
      "entropy": 6.036958646774292,
      "epoch": 2.0646414694454256,
      "grad_norm": 1.75,
      "learning_rate": 0.0004572765407025257,
      "loss": 5.6769,
      "mean_token_accuracy": 0.17824050933122634,
      "num_tokens": 43591060.0,
      "step": 23380
    },
    {
      "entropy": 6.1496381759643555,
      "epoch": 2.065083009537266,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004572580764891493,
      "loss": 5.866,
      "mean_token_accuracy": 0.16508686989545823,
      "num_tokens": 43600057.0,
      "step": 23385
    },
    {
      "entropy": 6.10386176109314,
      "epoch": 2.065524549629106,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004572396087054406,
      "loss": 5.6562,
      "mean_token_accuracy": 0.18082741796970367,
      "num_tokens": 43608609.0,
      "step": 23390
    },
    {
      "entropy": 6.015077590942383,
      "epoch": 2.0659660897209466,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004572211373517618,
      "loss": 5.7803,
      "mean_token_accuracy": 0.16902307122945787,
      "num_tokens": 43618751.0,
      "step": 23395
    },
    {
      "entropy": 6.127391672134399,
      "epoch": 2.066407629812787,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00045720266242847446,
      "loss": 5.8545,
      "mean_token_accuracy": 0.17057742923498154,
      "num_tokens": 43628934.0,
      "step": 23400
    },
    {
      "entropy": 6.140016794204712,
      "epoch": 2.066849169904627,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00045718418393594055,
      "loss": 5.7526,
      "mean_token_accuracy": 0.17612434476613997,
      "num_tokens": 43638512.0,
      "step": 23405
    },
    {
      "entropy": 5.97312741279602,
      "epoch": 2.0672907099964677,
      "grad_norm": 2.359375,
      "learning_rate": 0.00045716570187452214,
      "loss": 5.6907,
      "mean_token_accuracy": 0.1731865406036377,
      "num_tokens": 43647708.0,
      "step": 23410
    },
    {
      "entropy": 6.036051845550537,
      "epoch": 2.067732250088308,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004571472162445813,
      "loss": 5.7674,
      "mean_token_accuracy": 0.17517612725496293,
      "num_tokens": 43657096.0,
      "step": 23415
    },
    {
      "entropy": 6.07252140045166,
      "epoch": 2.0681737901801482,
      "grad_norm": 2.484375,
      "learning_rate": 0.00045712872704647994,
      "loss": 5.776,
      "mean_token_accuracy": 0.1708046928048134,
      "num_tokens": 43667410.0,
      "step": 23420
    },
    {
      "entropy": 6.073540925979614,
      "epoch": 2.0686153302719887,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004571102342805805,
      "loss": 5.7339,
      "mean_token_accuracy": 0.1755052164196968,
      "num_tokens": 43675800.0,
      "step": 23425
    },
    {
      "entropy": 6.0428118228912355,
      "epoch": 2.0690568703638292,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004570917379472451,
      "loss": 5.7328,
      "mean_token_accuracy": 0.1837793245911598,
      "num_tokens": 43684802.0,
      "step": 23430
    },
    {
      "entropy": 5.993852949142456,
      "epoch": 2.0694984104556693,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045707323804683616,
      "loss": 5.7286,
      "mean_token_accuracy": 0.18102988749742507,
      "num_tokens": 43693968.0,
      "step": 23435
    },
    {
      "entropy": 6.153824329376221,
      "epoch": 2.06993995054751,
      "grad_norm": 2.078125,
      "learning_rate": 0.000457054734579716,
      "loss": 5.7604,
      "mean_token_accuracy": 0.16545585691928863,
      "num_tokens": 43703177.0,
      "step": 23440
    },
    {
      "entropy": 6.054857921600342,
      "epoch": 2.07038149063935,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00045703622754624716,
      "loss": 5.7578,
      "mean_token_accuracy": 0.17659079730510713,
      "num_tokens": 43712894.0,
      "step": 23445
    },
    {
      "entropy": 6.054760217666626,
      "epoch": 2.0708230307311903,
      "grad_norm": 1.765625,
      "learning_rate": 0.00045701771694679213,
      "loss": 5.6749,
      "mean_token_accuracy": 0.17841540426015853,
      "num_tokens": 43722591.0,
      "step": 23450
    },
    {
      "entropy": 5.968215227127075,
      "epoch": 2.071264570823031,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045699920278171355,
      "loss": 5.6576,
      "mean_token_accuracy": 0.18652978986501695,
      "num_tokens": 43732251.0,
      "step": 23455
    },
    {
      "entropy": 6.136153364181519,
      "epoch": 2.071706110914871,
      "grad_norm": 1.875,
      "learning_rate": 0.0004569806850513741,
      "loss": 5.8283,
      "mean_token_accuracy": 0.1806676596403122,
      "num_tokens": 43741061.0,
      "step": 23460
    },
    {
      "entropy": 6.105810499191284,
      "epoch": 2.0721476510067114,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00045696216375613647,
      "loss": 5.7823,
      "mean_token_accuracy": 0.17174470126628877,
      "num_tokens": 43750147.0,
      "step": 23465
    },
    {
      "entropy": 6.066600561141968,
      "epoch": 2.072589191098552,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004569436388963636,
      "loss": 5.8383,
      "mean_token_accuracy": 0.16454419791698455,
      "num_tokens": 43760960.0,
      "step": 23470
    },
    {
      "entropy": 6.021958494186402,
      "epoch": 2.073030731190392,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004569251104724184,
      "loss": 5.7519,
      "mean_token_accuracy": 0.17310821861028672,
      "num_tokens": 43771306.0,
      "step": 23475
    },
    {
      "entropy": 6.009669923782349,
      "epoch": 2.0734722712822324,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004569065784846636,
      "loss": 5.6371,
      "mean_token_accuracy": 0.1787103831768036,
      "num_tokens": 43780520.0,
      "step": 23480
    },
    {
      "entropy": 6.100091600418091,
      "epoch": 2.073913811374073,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004568880429334624,
      "loss": 5.776,
      "mean_token_accuracy": 0.17368683069944382,
      "num_tokens": 43789472.0,
      "step": 23485
    },
    {
      "entropy": 6.113467121124268,
      "epoch": 2.074355351465913,
      "grad_norm": 2.390625,
      "learning_rate": 0.00045686950381917805,
      "loss": 5.7935,
      "mean_token_accuracy": 0.16746529936790466,
      "num_tokens": 43798728.0,
      "step": 23490
    },
    {
      "entropy": 6.062205648422241,
      "epoch": 2.0747968915577535,
      "grad_norm": 2.140625,
      "learning_rate": 0.00045685096114217336,
      "loss": 5.7087,
      "mean_token_accuracy": 0.17999023497104644,
      "num_tokens": 43807959.0,
      "step": 23495
    },
    {
      "entropy": 6.108411121368408,
      "epoch": 2.075238431649594,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004568324149028119,
      "loss": 5.7246,
      "mean_token_accuracy": 0.17597604393959046,
      "num_tokens": 43816922.0,
      "step": 23500
    },
    {
      "entropy": 6.017356300354004,
      "epoch": 2.075679971741434,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004568138651014567,
      "loss": 5.7196,
      "mean_token_accuracy": 0.17795283794403077,
      "num_tokens": 43827195.0,
      "step": 23505
    },
    {
      "entropy": 5.975253915786743,
      "epoch": 2.0761215118332745,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004567953117384714,
      "loss": 5.6809,
      "mean_token_accuracy": 0.17806514948606492,
      "num_tokens": 43836289.0,
      "step": 23510
    },
    {
      "entropy": 6.087928819656372,
      "epoch": 2.0765630519251146,
      "grad_norm": 2.0,
      "learning_rate": 0.0004567767548142193,
      "loss": 5.7814,
      "mean_token_accuracy": 0.17479943335056305,
      "num_tokens": 43845918.0,
      "step": 23515
    },
    {
      "entropy": 6.127353286743164,
      "epoch": 2.077004592016955,
      "grad_norm": 1.625,
      "learning_rate": 0.00045675819432906394,
      "loss": 5.7803,
      "mean_token_accuracy": 0.17071448713541032,
      "num_tokens": 43853907.0,
      "step": 23520
    },
    {
      "entropy": 6.02207612991333,
      "epoch": 2.0774461321087956,
      "grad_norm": 3.671875,
      "learning_rate": 0.00045673963028336896,
      "loss": 5.7337,
      "mean_token_accuracy": 0.1771627113223076,
      "num_tokens": 43862992.0,
      "step": 23525
    },
    {
      "entropy": 6.087219095230102,
      "epoch": 2.0778876722006356,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00045672106267749795,
      "loss": 5.7441,
      "mean_token_accuracy": 0.17353140115737914,
      "num_tokens": 43871686.0,
      "step": 23530
    },
    {
      "entropy": 6.003948926925659,
      "epoch": 2.078329212292476,
      "grad_norm": 2.25,
      "learning_rate": 0.0004567024915118148,
      "loss": 5.7381,
      "mean_token_accuracy": 0.1760423183441162,
      "num_tokens": 43881302.0,
      "step": 23535
    },
    {
      "entropy": 6.121306753158569,
      "epoch": 2.0787707523843166,
      "grad_norm": 1.75,
      "learning_rate": 0.00045668391678668296,
      "loss": 5.7983,
      "mean_token_accuracy": 0.1666475936770439,
      "num_tokens": 43889982.0,
      "step": 23540
    },
    {
      "entropy": 6.07760968208313,
      "epoch": 2.0792122924761567,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00045666533850246665,
      "loss": 5.7785,
      "mean_token_accuracy": 0.17647273987531661,
      "num_tokens": 43898385.0,
      "step": 23545
    },
    {
      "entropy": 6.041455411911011,
      "epoch": 2.079653832567997,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004566467566595297,
      "loss": 5.7349,
      "mean_token_accuracy": 0.16386200934648515,
      "num_tokens": 43907882.0,
      "step": 23550
    },
    {
      "entropy": 6.148154544830322,
      "epoch": 2.0800953726598377,
      "grad_norm": 2.140625,
      "learning_rate": 0.00045662817125823605,
      "loss": 5.7613,
      "mean_token_accuracy": 0.17677324414253234,
      "num_tokens": 43916222.0,
      "step": 23555
    },
    {
      "entropy": 6.043903970718384,
      "epoch": 2.0805369127516777,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00045660958229894985,
      "loss": 5.6854,
      "mean_token_accuracy": 0.17748757004737853,
      "num_tokens": 43925181.0,
      "step": 23560
    },
    {
      "entropy": 5.992525339126587,
      "epoch": 2.0809784528435182,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00045659098978203524,
      "loss": 5.7333,
      "mean_token_accuracy": 0.1730632558465004,
      "num_tokens": 43933973.0,
      "step": 23565
    },
    {
      "entropy": 5.98916482925415,
      "epoch": 2.0814199929353587,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004565723937078563,
      "loss": 5.6957,
      "mean_token_accuracy": 0.17935002893209456,
      "num_tokens": 43943297.0,
      "step": 23570
    },
    {
      "entropy": 6.1017866134643555,
      "epoch": 2.081861533027199,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00045655379407677754,
      "loss": 5.7716,
      "mean_token_accuracy": 0.1766517475247383,
      "num_tokens": 43952679.0,
      "step": 23575
    },
    {
      "entropy": 6.123733758926392,
      "epoch": 2.0823030731190393,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004565351908891632,
      "loss": 5.7216,
      "mean_token_accuracy": 0.17303809076547622,
      "num_tokens": 43961629.0,
      "step": 23580
    },
    {
      "entropy": 6.0128623962402346,
      "epoch": 2.0827446132108793,
      "grad_norm": 4.5625,
      "learning_rate": 0.00045651658414537765,
      "loss": 5.7436,
      "mean_token_accuracy": 0.17360084503889084,
      "num_tokens": 43970880.0,
      "step": 23585
    },
    {
      "entropy": 6.030864572525024,
      "epoch": 2.08318615330272,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004564979738457855,
      "loss": 5.7059,
      "mean_token_accuracy": 0.1836896315217018,
      "num_tokens": 43980449.0,
      "step": 23590
    },
    {
      "entropy": 6.01722354888916,
      "epoch": 2.0836276933945603,
      "grad_norm": 1.953125,
      "learning_rate": 0.00045647935999075127,
      "loss": 5.6992,
      "mean_token_accuracy": 0.1726984977722168,
      "num_tokens": 43988716.0,
      "step": 23595
    },
    {
      "entropy": 6.118170499801636,
      "epoch": 2.0840692334864004,
      "grad_norm": 2.59375,
      "learning_rate": 0.00045646074258063956,
      "loss": 5.782,
      "mean_token_accuracy": 0.177912200987339,
      "num_tokens": 43998464.0,
      "step": 23600
    },
    {
      "entropy": 6.088991737365722,
      "epoch": 2.084510773578241,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004564421216158152,
      "loss": 5.7187,
      "mean_token_accuracy": 0.17990431934595108,
      "num_tokens": 44006993.0,
      "step": 23605
    },
    {
      "entropy": 6.059772777557373,
      "epoch": 2.0849523136700814,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004564234970966428,
      "loss": 5.7885,
      "mean_token_accuracy": 0.16979264318943024,
      "num_tokens": 44015801.0,
      "step": 23610
    },
    {
      "entropy": 5.951881170272827,
      "epoch": 2.0853938537619214,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004564048690234872,
      "loss": 5.7129,
      "mean_token_accuracy": 0.171370992064476,
      "num_tokens": 44025069.0,
      "step": 23615
    },
    {
      "entropy": 6.155179071426391,
      "epoch": 2.085835393853762,
      "grad_norm": 2.578125,
      "learning_rate": 0.00045638623739671343,
      "loss": 5.8679,
      "mean_token_accuracy": 0.15780416429042815,
      "num_tokens": 44034700.0,
      "step": 23620
    },
    {
      "entropy": 6.09529972076416,
      "epoch": 2.0862769339456024,
      "grad_norm": 4.0625,
      "learning_rate": 0.00045636760221668646,
      "loss": 5.6795,
      "mean_token_accuracy": 0.1765646979212761,
      "num_tokens": 44044417.0,
      "step": 23625
    },
    {
      "entropy": 6.082326936721802,
      "epoch": 2.0867184740374425,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004563489634837713,
      "loss": 5.7279,
      "mean_token_accuracy": 0.1755108967423439,
      "num_tokens": 44053617.0,
      "step": 23630
    },
    {
      "entropy": 6.033968925476074,
      "epoch": 2.087160014129283,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00045633032119833314,
      "loss": 5.7134,
      "mean_token_accuracy": 0.17140089869499206,
      "num_tokens": 44062738.0,
      "step": 23635
    },
    {
      "entropy": 6.069627141952514,
      "epoch": 2.0876015542211235,
      "grad_norm": 2.4375,
      "learning_rate": 0.000456311675360737,
      "loss": 5.7785,
      "mean_token_accuracy": 0.16690519005060195,
      "num_tokens": 44072154.0,
      "step": 23640
    },
    {
      "entropy": 6.1760039806365965,
      "epoch": 2.0880430943129635,
      "grad_norm": 2.125,
      "learning_rate": 0.0004562930259713483,
      "loss": 5.8201,
      "mean_token_accuracy": 0.16758950054645538,
      "num_tokens": 44081239.0,
      "step": 23645
    },
    {
      "entropy": 6.085016965866089,
      "epoch": 2.088484634404804,
      "grad_norm": 2.234375,
      "learning_rate": 0.00045627437303053236,
      "loss": 5.7127,
      "mean_token_accuracy": 0.1745936319231987,
      "num_tokens": 44090866.0,
      "step": 23650
    },
    {
      "entropy": 6.013209629058838,
      "epoch": 2.088926174496644,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004562557165386546,
      "loss": 5.8171,
      "mean_token_accuracy": 0.16897691637277604,
      "num_tokens": 44100225.0,
      "step": 23655
    },
    {
      "entropy": 6.094102287292481,
      "epoch": 2.0893677145884846,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004562370564960804,
      "loss": 5.7565,
      "mean_token_accuracy": 0.1729419097304344,
      "num_tokens": 44110139.0,
      "step": 23660
    },
    {
      "entropy": 6.111436080932617,
      "epoch": 2.089809254680325,
      "grad_norm": 1.546875,
      "learning_rate": 0.00045621839290317526,
      "loss": 5.719,
      "mean_token_accuracy": 0.17938904464244843,
      "num_tokens": 44118687.0,
      "step": 23665
    },
    {
      "entropy": 6.039404487609863,
      "epoch": 2.090250794772165,
      "grad_norm": 3.078125,
      "learning_rate": 0.00045619972576030495,
      "loss": 5.7359,
      "mean_token_accuracy": 0.1711748018860817,
      "num_tokens": 44127377.0,
      "step": 23670
    },
    {
      "entropy": 6.003701114654541,
      "epoch": 2.0906923348640056,
      "grad_norm": 2.46875,
      "learning_rate": 0.000456181055067835,
      "loss": 5.7496,
      "mean_token_accuracy": 0.17740200459957123,
      "num_tokens": 44136583.0,
      "step": 23675
    },
    {
      "entropy": 6.021934986114502,
      "epoch": 2.091133874955846,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004561623808261313,
      "loss": 5.7324,
      "mean_token_accuracy": 0.17165962755680084,
      "num_tokens": 44145467.0,
      "step": 23680
    },
    {
      "entropy": 6.039149141311645,
      "epoch": 2.091575415047686,
      "grad_norm": 7.875,
      "learning_rate": 0.0004561437030355595,
      "loss": 5.7174,
      "mean_token_accuracy": 0.18085480034351348,
      "num_tokens": 44154932.0,
      "step": 23685
    },
    {
      "entropy": 5.979896545410156,
      "epoch": 2.0920169551395267,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004561250216964857,
      "loss": 5.632,
      "mean_token_accuracy": 0.18613940328359604,
      "num_tokens": 44162791.0,
      "step": 23690
    },
    {
      "entropy": 6.057424020767212,
      "epoch": 2.092458495231367,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004561063368092757,
      "loss": 5.781,
      "mean_token_accuracy": 0.17305790334939958,
      "num_tokens": 44171993.0,
      "step": 23695
    },
    {
      "entropy": 6.091082048416138,
      "epoch": 2.0929000353232072,
      "grad_norm": 1.9375,
      "learning_rate": 0.00045608764837429543,
      "loss": 5.7926,
      "mean_token_accuracy": 0.1736200675368309,
      "num_tokens": 44180807.0,
      "step": 23700
    },
    {
      "entropy": 6.13680510520935,
      "epoch": 2.0933415754150477,
      "grad_norm": 2.296875,
      "learning_rate": 0.00045606895639191126,
      "loss": 5.8303,
      "mean_token_accuracy": 0.17249124348163605,
      "num_tokens": 44191764.0,
      "step": 23705
    },
    {
      "entropy": 6.0706311702728275,
      "epoch": 2.0937831155068882,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045605026086248916,
      "loss": 5.7889,
      "mean_token_accuracy": 0.17373232543468475,
      "num_tokens": 44201165.0,
      "step": 23710
    },
    {
      "entropy": 6.122869634628296,
      "epoch": 2.0942246555987283,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004560315617863954,
      "loss": 5.8039,
      "mean_token_accuracy": 0.1718558430671692,
      "num_tokens": 44210709.0,
      "step": 23715
    },
    {
      "entropy": 6.10222635269165,
      "epoch": 2.0946661956905688,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004560128591639964,
      "loss": 5.7648,
      "mean_token_accuracy": 0.17185956686735154,
      "num_tokens": 44219827.0,
      "step": 23720
    },
    {
      "entropy": 6.132001447677612,
      "epoch": 2.095107735782409,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004559941529956584,
      "loss": 5.8146,
      "mean_token_accuracy": 0.16624186635017396,
      "num_tokens": 44229498.0,
      "step": 23725
    },
    {
      "entropy": 6.0440747261047365,
      "epoch": 2.0955492758742493,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004559754432817479,
      "loss": 5.7322,
      "mean_token_accuracy": 0.17843539714813234,
      "num_tokens": 44238844.0,
      "step": 23730
    },
    {
      "entropy": 6.047912549972534,
      "epoch": 2.09599081596609,
      "grad_norm": 2.296875,
      "learning_rate": 0.00045595673002263125,
      "loss": 5.7308,
      "mean_token_accuracy": 0.1752448186278343,
      "num_tokens": 44247512.0,
      "step": 23735
    },
    {
      "entropy": 6.0723954200744625,
      "epoch": 2.09643235605793,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004559380132186753,
      "loss": 5.6785,
      "mean_token_accuracy": 0.1817456379532814,
      "num_tokens": 44256485.0,
      "step": 23740
    },
    {
      "entropy": 5.944156646728516,
      "epoch": 2.0968738961497704,
      "grad_norm": 2.546875,
      "learning_rate": 0.00045591929287024655,
      "loss": 5.5801,
      "mean_token_accuracy": 0.18451371192932128,
      "num_tokens": 44265095.0,
      "step": 23745
    },
    {
      "entropy": 6.019408416748047,
      "epoch": 2.097315436241611,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004559005689777117,
      "loss": 5.7586,
      "mean_token_accuracy": 0.1693573772907257,
      "num_tokens": 44274582.0,
      "step": 23750
    },
    {
      "entropy": 6.130525732040406,
      "epoch": 2.097756976333451,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004558818415414377,
      "loss": 5.8131,
      "mean_token_accuracy": 0.16966109871864318,
      "num_tokens": 44285752.0,
      "step": 23755
    },
    {
      "entropy": 6.143424654006958,
      "epoch": 2.0981985164252914,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004558631105617912,
      "loss": 5.7682,
      "mean_token_accuracy": 0.16804327815771103,
      "num_tokens": 44295315.0,
      "step": 23760
    },
    {
      "entropy": 6.011955499649048,
      "epoch": 2.098640056517132,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045584437603913923,
      "loss": 5.8469,
      "mean_token_accuracy": 0.16663959622383118,
      "num_tokens": 44305151.0,
      "step": 23765
    },
    {
      "entropy": 6.058171558380127,
      "epoch": 2.099081596608972,
      "grad_norm": 1.703125,
      "learning_rate": 0.00045582563797384886,
      "loss": 5.8031,
      "mean_token_accuracy": 0.16820082515478135,
      "num_tokens": 44314462.0,
      "step": 23770
    },
    {
      "entropy": 6.052392292022705,
      "epoch": 2.0995231367008125,
      "grad_norm": 1.59375,
      "learning_rate": 0.000455806896366287,
      "loss": 5.6219,
      "mean_token_accuracy": 0.19755849391222,
      "num_tokens": 44322782.0,
      "step": 23775
    },
    {
      "entropy": 6.10023775100708,
      "epoch": 2.099964676792653,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004557881512168209,
      "loss": 5.6916,
      "mean_token_accuracy": 0.18137233704328537,
      "num_tokens": 44331588.0,
      "step": 23780
    },
    {
      "entropy": 6.048156881332398,
      "epoch": 2.100406216884493,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004557694025258177,
      "loss": 5.7258,
      "mean_token_accuracy": 0.18416826277971268,
      "num_tokens": 44340997.0,
      "step": 23785
    },
    {
      "entropy": 6.056554508209229,
      "epoch": 2.1008477569763335,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004557506502936448,
      "loss": 5.7562,
      "mean_token_accuracy": 0.17048679441213607,
      "num_tokens": 44350679.0,
      "step": 23790
    },
    {
      "entropy": 6.040418386459351,
      "epoch": 2.1012892970681736,
      "grad_norm": 2.359375,
      "learning_rate": 0.00045573189452066936,
      "loss": 5.6816,
      "mean_token_accuracy": 0.17749429643154144,
      "num_tokens": 44359934.0,
      "step": 23795
    },
    {
      "entropy": 6.015652894973755,
      "epoch": 2.101730837160014,
      "grad_norm": 12.25,
      "learning_rate": 0.00045571313520725897,
      "loss": 5.7471,
      "mean_token_accuracy": 0.17009244710206986,
      "num_tokens": 44369104.0,
      "step": 23800
    },
    {
      "entropy": 6.06161150932312,
      "epoch": 2.1021723772518546,
      "grad_norm": 3.3125,
      "learning_rate": 0.00045569437235378103,
      "loss": 5.7389,
      "mean_token_accuracy": 0.173904250562191,
      "num_tokens": 44379645.0,
      "step": 23805
    },
    {
      "entropy": 6.111919689178467,
      "epoch": 2.1026139173436946,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004556756059606031,
      "loss": 5.7573,
      "mean_token_accuracy": 0.17693840116262435,
      "num_tokens": 44389276.0,
      "step": 23810
    },
    {
      "entropy": 6.048802089691162,
      "epoch": 2.103055457435535,
      "grad_norm": 6.25,
      "learning_rate": 0.0004556568360280928,
      "loss": 5.738,
      "mean_token_accuracy": 0.17227324843406677,
      "num_tokens": 44399114.0,
      "step": 23815
    },
    {
      "entropy": 5.89023790359497,
      "epoch": 2.1034969975273756,
      "grad_norm": 2.5625,
      "learning_rate": 0.00045563806255661784,
      "loss": 5.4766,
      "mean_token_accuracy": 0.2094142258167267,
      "num_tokens": 44406917.0,
      "step": 23820
    },
    {
      "entropy": 6.102201080322265,
      "epoch": 2.1039385376192157,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004556192855465459,
      "loss": 5.7625,
      "mean_token_accuracy": 0.16426923125982285,
      "num_tokens": 44416678.0,
      "step": 23825
    },
    {
      "entropy": 5.970564937591552,
      "epoch": 2.104380077711056,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004556005049982449,
      "loss": 5.6372,
      "mean_token_accuracy": 0.18183527141809464,
      "num_tokens": 44425427.0,
      "step": 23830
    },
    {
      "entropy": 6.049872732162475,
      "epoch": 2.1048216178028967,
      "grad_norm": 2.40625,
      "learning_rate": 0.00045558172091208284,
      "loss": 5.7166,
      "mean_token_accuracy": 0.17256854772567748,
      "num_tokens": 44434267.0,
      "step": 23835
    },
    {
      "entropy": 6.073463726043701,
      "epoch": 2.1052631578947367,
      "grad_norm": 1.96875,
      "learning_rate": 0.00045556293328842746,
      "loss": 5.7209,
      "mean_token_accuracy": 0.18110367506742478,
      "num_tokens": 44444037.0,
      "step": 23840
    },
    {
      "entropy": 6.053682422637939,
      "epoch": 2.1057046979865772,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004555441421276469,
      "loss": 5.7073,
      "mean_token_accuracy": 0.17837031483650206,
      "num_tokens": 44453707.0,
      "step": 23845
    },
    {
      "entropy": 6.063917446136474,
      "epoch": 2.1061462380784177,
      "grad_norm": 5.59375,
      "learning_rate": 0.00045552534743010934,
      "loss": 5.712,
      "mean_token_accuracy": 0.17191759049892424,
      "num_tokens": 44464357.0,
      "step": 23850
    },
    {
      "entropy": 6.1204918384552,
      "epoch": 2.1065877781702578,
      "grad_norm": 1.84375,
      "learning_rate": 0.00045550654919618283,
      "loss": 5.8173,
      "mean_token_accuracy": 0.17030971199274064,
      "num_tokens": 44473803.0,
      "step": 23855
    },
    {
      "entropy": 6.077618074417114,
      "epoch": 2.1070293182620983,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004554877474262357,
      "loss": 5.7607,
      "mean_token_accuracy": 0.17263709604740143,
      "num_tokens": 44482830.0,
      "step": 23860
    },
    {
      "entropy": 6.037305068969727,
      "epoch": 2.1074708583539383,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004554689421206362,
      "loss": 5.7328,
      "mean_token_accuracy": 0.17612526565790176,
      "num_tokens": 44493045.0,
      "step": 23865
    },
    {
      "entropy": 6.041487312316894,
      "epoch": 2.107912398445779,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004554501332797529,
      "loss": 5.7026,
      "mean_token_accuracy": 0.17910281717777252,
      "num_tokens": 44503304.0,
      "step": 23870
    },
    {
      "entropy": 6.085078477859497,
      "epoch": 2.1083539385376193,
      "grad_norm": 2.578125,
      "learning_rate": 0.00045543132090395403,
      "loss": 5.7882,
      "mean_token_accuracy": 0.1733369469642639,
      "num_tokens": 44512337.0,
      "step": 23875
    },
    {
      "entropy": 6.094937753677368,
      "epoch": 2.1087954786294594,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004554125049936082,
      "loss": 5.7275,
      "mean_token_accuracy": 0.17947376370429993,
      "num_tokens": 44521881.0,
      "step": 23880
    },
    {
      "entropy": 6.053909349441528,
      "epoch": 2.1092370187213,
      "grad_norm": 1.78125,
      "learning_rate": 0.000455393685549084,
      "loss": 5.7518,
      "mean_token_accuracy": 0.1737428367137909,
      "num_tokens": 44531351.0,
      "step": 23885
    },
    {
      "entropy": 6.079000616073609,
      "epoch": 2.1096785588131404,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004553748625707501,
      "loss": 5.7773,
      "mean_token_accuracy": 0.17754469364881514,
      "num_tokens": 44540592.0,
      "step": 23890
    },
    {
      "entropy": 6.038525104522705,
      "epoch": 2.1101200989049804,
      "grad_norm": 3.25,
      "learning_rate": 0.00045535603605897516,
      "loss": 5.6957,
      "mean_token_accuracy": 0.18500210642814635,
      "num_tokens": 44550019.0,
      "step": 23895
    },
    {
      "entropy": 6.032449531555176,
      "epoch": 2.110561638996821,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004553372060141282,
      "loss": 5.7339,
      "mean_token_accuracy": 0.17283661812543868,
      "num_tokens": 44559423.0,
      "step": 23900
    },
    {
      "entropy": 5.959686803817749,
      "epoch": 2.1110031790886614,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004553183724365778,
      "loss": 5.7004,
      "mean_token_accuracy": 0.18014152497053146,
      "num_tokens": 44568792.0,
      "step": 23905
    },
    {
      "entropy": 6.038976621627808,
      "epoch": 2.1114447191805015,
      "grad_norm": 2.25,
      "learning_rate": 0.0004552995353266931,
      "loss": 5.7441,
      "mean_token_accuracy": 0.17441533952951432,
      "num_tokens": 44578191.0,
      "step": 23910
    },
    {
      "entropy": 6.08317551612854,
      "epoch": 2.111886259272342,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004552806946848431,
      "loss": 5.7018,
      "mean_token_accuracy": 0.17960938215255737,
      "num_tokens": 44587402.0,
      "step": 23915
    },
    {
      "entropy": 6.102370023727417,
      "epoch": 2.1123277993641825,
      "grad_norm": 2.109375,
      "learning_rate": 0.00045526185051139676,
      "loss": 5.8412,
      "mean_token_accuracy": 0.16092807799577713,
      "num_tokens": 44596241.0,
      "step": 23920
    },
    {
      "entropy": 6.020882892608642,
      "epoch": 2.1127693394560225,
      "grad_norm": 2.234375,
      "learning_rate": 0.00045524300280672326,
      "loss": 5.6426,
      "mean_token_accuracy": 0.18139759302139283,
      "num_tokens": 44604796.0,
      "step": 23925
    },
    {
      "entropy": 6.065481901168823,
      "epoch": 2.113210879547863,
      "grad_norm": 1.5625,
      "learning_rate": 0.00045522415157119185,
      "loss": 5.798,
      "mean_token_accuracy": 0.16563344448804856,
      "num_tokens": 44615208.0,
      "step": 23930
    },
    {
      "entropy": 6.011007022857666,
      "epoch": 2.113652419639703,
      "grad_norm": 2.0,
      "learning_rate": 0.0004552052968051719,
      "loss": 5.7362,
      "mean_token_accuracy": 0.18108452409505843,
      "num_tokens": 44623937.0,
      "step": 23935
    },
    {
      "entropy": 6.0833213329315186,
      "epoch": 2.1140939597315436,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004551864385090326,
      "loss": 5.7192,
      "mean_token_accuracy": 0.1737353175878525,
      "num_tokens": 44634116.0,
      "step": 23940
    },
    {
      "entropy": 6.142103147506714,
      "epoch": 2.114535499823384,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004551675766831434,
      "loss": 5.7673,
      "mean_token_accuracy": 0.17710809409618378,
      "num_tokens": 44643436.0,
      "step": 23945
    },
    {
      "entropy": 6.06249680519104,
      "epoch": 2.114977039915224,
      "grad_norm": 2.125,
      "learning_rate": 0.00045514871132787394,
      "loss": 5.7868,
      "mean_token_accuracy": 0.17562808841466904,
      "num_tokens": 44654077.0,
      "step": 23950
    },
    {
      "entropy": 6.073651361465454,
      "epoch": 2.1154185800070646,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00045512984244359364,
      "loss": 5.7412,
      "mean_token_accuracy": 0.1748410075902939,
      "num_tokens": 44663785.0,
      "step": 23955
    },
    {
      "entropy": 5.945398139953613,
      "epoch": 2.115860120098905,
      "grad_norm": 2.359375,
      "learning_rate": 0.00045511097003067216,
      "loss": 5.6051,
      "mean_token_accuracy": 0.18745265305042266,
      "num_tokens": 44672750.0,
      "step": 23960
    },
    {
      "entropy": 5.995215320587159,
      "epoch": 2.116301660190745,
      "grad_norm": 1.609375,
      "learning_rate": 0.00045509209408947914,
      "loss": 5.6932,
      "mean_token_accuracy": 0.1842673659324646,
      "num_tokens": 44681814.0,
      "step": 23965
    },
    {
      "entropy": 5.995731592178345,
      "epoch": 2.1167432002825857,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004550732146203845,
      "loss": 5.7041,
      "mean_token_accuracy": 0.17657448649406432,
      "num_tokens": 44690926.0,
      "step": 23970
    },
    {
      "entropy": 6.148920392990112,
      "epoch": 2.117184740374426,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00045505433162375806,
      "loss": 5.8692,
      "mean_token_accuracy": 0.1731095492839813,
      "num_tokens": 44700625.0,
      "step": 23975
    },
    {
      "entropy": 6.101783895492554,
      "epoch": 2.117626280466266,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004550354450999696,
      "loss": 5.7649,
      "mean_token_accuracy": 0.1754211813211441,
      "num_tokens": 44710703.0,
      "step": 23980
    },
    {
      "entropy": 6.039858341217041,
      "epoch": 2.1180678205581067,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004550165550493891,
      "loss": 5.6908,
      "mean_token_accuracy": 0.18105313926935196,
      "num_tokens": 44719412.0,
      "step": 23985
    },
    {
      "entropy": 6.066136693954467,
      "epoch": 2.118509360649947,
      "grad_norm": 2.4375,
      "learning_rate": 0.00045499766147238675,
      "loss": 5.8107,
      "mean_token_accuracy": 0.17512366771697999,
      "num_tokens": 44729161.0,
      "step": 23990
    },
    {
      "entropy": 6.14470682144165,
      "epoch": 2.1189509007417873,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00045497876436933257,
      "loss": 5.806,
      "mean_token_accuracy": 0.16657124012708663,
      "num_tokens": 44738985.0,
      "step": 23995
    },
    {
      "entropy": 6.124337816238404,
      "epoch": 2.1193924408336278,
      "grad_norm": 1.671875,
      "learning_rate": 0.00045495986374059674,
      "loss": 5.7164,
      "mean_token_accuracy": 0.18006630837917328,
      "num_tokens": 44748070.0,
      "step": 24000
    },
    {
      "epoch": 2.1193924408336278,
      "eval_entropy": 5.8434180657698915,
      "eval_loss": 5.815832138061523,
      "eval_mean_token_accuracy": 0.18016821731236318,
      "eval_num_tokens": 44748070.0,
      "eval_runtime": 26.2661,
      "eval_samples_per_second": 1344.511,
      "eval_steps_per_second": 168.088,
      "step": 24000
    },
    {
      "entropy": 5.9801126480102536,
      "epoch": 2.119833980925468,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004549409595865496,
      "loss": 5.6554,
      "mean_token_accuracy": 0.180684594810009,
      "num_tokens": 44756607.0,
      "step": 24005
    },
    {
      "entropy": 5.99581208229065,
      "epoch": 2.1202755210173083,
      "grad_norm": 1.765625,
      "learning_rate": 0.00045492205190756133,
      "loss": 5.7708,
      "mean_token_accuracy": 0.17556989639997483,
      "num_tokens": 44765435.0,
      "step": 24010
    },
    {
      "entropy": 6.043754005432129,
      "epoch": 2.120717061109149,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004549031407040024,
      "loss": 5.7346,
      "mean_token_accuracy": 0.1729056566953659,
      "num_tokens": 44775589.0,
      "step": 24015
    },
    {
      "entropy": 6.117159271240235,
      "epoch": 2.121158601200989,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004548842259762432,
      "loss": 5.762,
      "mean_token_accuracy": 0.17266097366809846,
      "num_tokens": 44785692.0,
      "step": 24020
    },
    {
      "entropy": 6.013881874084473,
      "epoch": 2.1216001412928294,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004548653077246544,
      "loss": 5.6472,
      "mean_token_accuracy": 0.18281587660312654,
      "num_tokens": 44794074.0,
      "step": 24025
    },
    {
      "entropy": 6.018519735336303,
      "epoch": 2.12204168138467,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004548463859496065,
      "loss": 5.7624,
      "mean_token_accuracy": 0.17261515259742738,
      "num_tokens": 44803959.0,
      "step": 24030
    },
    {
      "entropy": 5.983622264862061,
      "epoch": 2.12248322147651,
      "grad_norm": 2.25,
      "learning_rate": 0.00045482746065147014,
      "loss": 5.6941,
      "mean_token_accuracy": 0.17479044646024705,
      "num_tokens": 44813294.0,
      "step": 24035
    },
    {
      "entropy": 6.066862916946411,
      "epoch": 2.1229247615683504,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004548085318306162,
      "loss": 5.7033,
      "mean_token_accuracy": 0.17637211978435516,
      "num_tokens": 44822788.0,
      "step": 24040
    },
    {
      "entropy": 6.01959376335144,
      "epoch": 2.123366301660191,
      "grad_norm": 2.28125,
      "learning_rate": 0.00045478959948741526,
      "loss": 5.6563,
      "mean_token_accuracy": 0.18011990785598755,
      "num_tokens": 44832236.0,
      "step": 24045
    },
    {
      "entropy": 6.106993627548218,
      "epoch": 2.123807841752031,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004547706636222384,
      "loss": 5.7312,
      "mean_token_accuracy": 0.17797573208808898,
      "num_tokens": 44841258.0,
      "step": 24050
    },
    {
      "entropy": 6.05029616355896,
      "epoch": 2.1242493818438715,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004547517242354565,
      "loss": 5.7216,
      "mean_token_accuracy": 0.17079202234745025,
      "num_tokens": 44850365.0,
      "step": 24055
    },
    {
      "entropy": 6.036679267883301,
      "epoch": 2.124690921935712,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004547327813274405,
      "loss": 5.764,
      "mean_token_accuracy": 0.17281196266412735,
      "num_tokens": 44860451.0,
      "step": 24060
    },
    {
      "entropy": 6.103848934173584,
      "epoch": 2.125132462027552,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004547138348985615,
      "loss": 5.6956,
      "mean_token_accuracy": 0.17822781652212144,
      "num_tokens": 44868815.0,
      "step": 24065
    },
    {
      "entropy": 6.066459894180298,
      "epoch": 2.1255740021193925,
      "grad_norm": 1.96875,
      "learning_rate": 0.00045469488494919076,
      "loss": 5.7901,
      "mean_token_accuracy": 0.17657415568828583,
      "num_tokens": 44878046.0,
      "step": 24070
    },
    {
      "entropy": 6.033400630950927,
      "epoch": 2.126015542211233,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004546759314796994,
      "loss": 5.6612,
      "mean_token_accuracy": 0.18332472890615464,
      "num_tokens": 44886915.0,
      "step": 24075
    },
    {
      "entropy": 6.046270799636841,
      "epoch": 2.126457082303073,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004546569744904586,
      "loss": 5.7361,
      "mean_token_accuracy": 0.17417203783988952,
      "num_tokens": 44895629.0,
      "step": 24080
    },
    {
      "entropy": 6.099378824234009,
      "epoch": 2.1268986223949136,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00045463801398184004,
      "loss": 5.8701,
      "mean_token_accuracy": 0.16949711740016937,
      "num_tokens": 44905913.0,
      "step": 24085
    },
    {
      "entropy": 6.149720621109009,
      "epoch": 2.1273401624867536,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004546190499542149,
      "loss": 5.7769,
      "mean_token_accuracy": 0.16766871362924576,
      "num_tokens": 44915329.0,
      "step": 24090
    },
    {
      "entropy": 6.0992296695709225,
      "epoch": 2.127781702578594,
      "grad_norm": 1.875,
      "learning_rate": 0.0004546000824079547,
      "loss": 5.7242,
      "mean_token_accuracy": 0.1766329139471054,
      "num_tokens": 44923810.0,
      "step": 24095
    },
    {
      "entropy": 6.031911706924438,
      "epoch": 2.1282232426704346,
      "grad_norm": 2.390625,
      "learning_rate": 0.000454581111343431,
      "loss": 5.7667,
      "mean_token_accuracy": 0.17630589455366136,
      "num_tokens": 44932375.0,
      "step": 24100
    },
    {
      "entropy": 6.081184816360474,
      "epoch": 2.1286647827622747,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004545621367610155,
      "loss": 5.7543,
      "mean_token_accuracy": 0.17274098992347717,
      "num_tokens": 44941341.0,
      "step": 24105
    },
    {
      "entropy": 6.108899068832398,
      "epoch": 2.129106322854115,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004545431586610799,
      "loss": 5.7703,
      "mean_token_accuracy": 0.1775650292634964,
      "num_tokens": 44951221.0,
      "step": 24110
    },
    {
      "entropy": 5.993212366104126,
      "epoch": 2.1295478629459557,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004545241770439958,
      "loss": 5.6693,
      "mean_token_accuracy": 0.18601171523332596,
      "num_tokens": 44960254.0,
      "step": 24115
    },
    {
      "entropy": 6.115652704238892,
      "epoch": 2.1299894030377957,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004545051919101353,
      "loss": 5.8393,
      "mean_token_accuracy": 0.1631707638502121,
      "num_tokens": 44969023.0,
      "step": 24120
    },
    {
      "entropy": 5.947831439971924,
      "epoch": 2.130430943129636,
      "grad_norm": 1.71875,
      "learning_rate": 0.00045448620325987013,
      "loss": 5.7142,
      "mean_token_accuracy": 0.1780805066227913,
      "num_tokens": 44978942.0,
      "step": 24125
    },
    {
      "entropy": 6.029355764389038,
      "epoch": 2.1308724832214767,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00045446721109357243,
      "loss": 5.6757,
      "mean_token_accuracy": 0.1740359768271446,
      "num_tokens": 44987794.0,
      "step": 24130
    },
    {
      "entropy": 6.115400171279907,
      "epoch": 2.1313140233133168,
      "grad_norm": 1.671875,
      "learning_rate": 0.00045444821541161404,
      "loss": 5.7946,
      "mean_token_accuracy": 0.1688767120242119,
      "num_tokens": 44995781.0,
      "step": 24135
    },
    {
      "entropy": 6.050333881378174,
      "epoch": 2.1317555634051573,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004544292162143671,
      "loss": 5.7294,
      "mean_token_accuracy": 0.17321905642747878,
      "num_tokens": 45004731.0,
      "step": 24140
    },
    {
      "entropy": 6.0841662883758545,
      "epoch": 2.1321971034969973,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004544102135022039,
      "loss": 5.7735,
      "mean_token_accuracy": 0.16835902035236358,
      "num_tokens": 45014312.0,
      "step": 24145
    },
    {
      "entropy": 6.134602785110474,
      "epoch": 2.132638643588838,
      "grad_norm": 1.75,
      "learning_rate": 0.0004543912072754967,
      "loss": 5.8339,
      "mean_token_accuracy": 0.1681660920381546,
      "num_tokens": 45023903.0,
      "step": 24150
    },
    {
      "entropy": 6.118334770202637,
      "epoch": 2.1330801836806783,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004543721975346177,
      "loss": 5.7618,
      "mean_token_accuracy": 0.1679817706346512,
      "num_tokens": 45033330.0,
      "step": 24155
    },
    {
      "entropy": 5.93076720237732,
      "epoch": 2.1335217237725184,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00045435318427993946,
      "loss": 5.6068,
      "mean_token_accuracy": 0.18273475617170334,
      "num_tokens": 45042862.0,
      "step": 24160
    },
    {
      "entropy": 6.080313014984131,
      "epoch": 2.133963263864359,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004543341675118343,
      "loss": 5.7048,
      "mean_token_accuracy": 0.1794311821460724,
      "num_tokens": 45052198.0,
      "step": 24165
    },
    {
      "entropy": 6.14905195236206,
      "epoch": 2.1344048039561994,
      "grad_norm": 1.59375,
      "learning_rate": 0.00045431514723067487,
      "loss": 5.7567,
      "mean_token_accuracy": 0.17908408790826796,
      "num_tokens": 45060625.0,
      "step": 24170
    },
    {
      "entropy": 5.990757608413697,
      "epoch": 2.1348463440480394,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00045429612343683354,
      "loss": 5.6856,
      "mean_token_accuracy": 0.18024740368127823,
      "num_tokens": 45069557.0,
      "step": 24175
    },
    {
      "entropy": 6.005916261672974,
      "epoch": 2.13528788413988,
      "grad_norm": 1.671875,
      "learning_rate": 0.00045427709613068317,
      "loss": 5.7792,
      "mean_token_accuracy": 0.17593892365694047,
      "num_tokens": 45079281.0,
      "step": 24180
    },
    {
      "entropy": 6.129705905914307,
      "epoch": 2.1357294242317204,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004542580653125965,
      "loss": 5.7843,
      "mean_token_accuracy": 0.17222469598054885,
      "num_tokens": 45089363.0,
      "step": 24185
    },
    {
      "entropy": 6.060103702545166,
      "epoch": 2.1361709643235605,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00045423903098294624,
      "loss": 5.7035,
      "mean_token_accuracy": 0.1652398243546486,
      "num_tokens": 45098735.0,
      "step": 24190
    },
    {
      "entropy": 6.098920297622681,
      "epoch": 2.136612504415401,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045421999314210535,
      "loss": 5.8618,
      "mean_token_accuracy": 0.1660047024488449,
      "num_tokens": 45109173.0,
      "step": 24195
    },
    {
      "entropy": 6.077979946136475,
      "epoch": 2.1370540445072415,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004542009517904466,
      "loss": 5.8496,
      "mean_token_accuracy": 0.16836312264204026,
      "num_tokens": 45119153.0,
      "step": 24200
    },
    {
      "entropy": 6.051446104049683,
      "epoch": 2.1374955845990815,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004541819069283432,
      "loss": 5.7342,
      "mean_token_accuracy": 0.17674380242824556,
      "num_tokens": 45128219.0,
      "step": 24205
    },
    {
      "entropy": 6.093689632415772,
      "epoch": 2.137937124690922,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004541628585561682,
      "loss": 5.78,
      "mean_token_accuracy": 0.17441326826810838,
      "num_tokens": 45137777.0,
      "step": 24210
    },
    {
      "entropy": 6.092958450317383,
      "epoch": 2.1383786647827625,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004541438066742946,
      "loss": 5.7197,
      "mean_token_accuracy": 0.1814879760146141,
      "num_tokens": 45147553.0,
      "step": 24215
    },
    {
      "entropy": 6.100968694686889,
      "epoch": 2.1388202048746026,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004541247512830958,
      "loss": 5.7798,
      "mean_token_accuracy": 0.16399392783641814,
      "num_tokens": 45157443.0,
      "step": 24220
    },
    {
      "entropy": 6.065377140045166,
      "epoch": 2.139261744966443,
      "grad_norm": 1.9375,
      "learning_rate": 0.000454105692382945,
      "loss": 5.8066,
      "mean_token_accuracy": 0.17214321941137314,
      "num_tokens": 45167279.0,
      "step": 24225
    },
    {
      "entropy": 6.141276693344116,
      "epoch": 2.139703285058283,
      "grad_norm": 2.34375,
      "learning_rate": 0.00045408662997421546,
      "loss": 5.8394,
      "mean_token_accuracy": 0.17419105023145676,
      "num_tokens": 45176515.0,
      "step": 24230
    },
    {
      "entropy": 6.1541801452636715,
      "epoch": 2.1401448251501236,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004540675640572807,
      "loss": 5.7535,
      "mean_token_accuracy": 0.1739774614572525,
      "num_tokens": 45185501.0,
      "step": 24235
    },
    {
      "entropy": 6.113752746582032,
      "epoch": 2.140586365241964,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004540484946325143,
      "loss": 5.8239,
      "mean_token_accuracy": 0.16978019177913667,
      "num_tokens": 45194881.0,
      "step": 24240
    },
    {
      "entropy": 6.0723412990570065,
      "epoch": 2.141027905333804,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045402942170028965,
      "loss": 5.7509,
      "mean_token_accuracy": 0.17636565268039703,
      "num_tokens": 45204463.0,
      "step": 24245
    },
    {
      "entropy": 6.059255838394165,
      "epoch": 2.1414694454256447,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00045401034526098053,
      "loss": 5.7698,
      "mean_token_accuracy": 0.17043178528547287,
      "num_tokens": 45214302.0,
      "step": 24250
    },
    {
      "entropy": 6.082703161239624,
      "epoch": 2.141910985517485,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004539912653149604,
      "loss": 5.7733,
      "mean_token_accuracy": 0.1696614682674408,
      "num_tokens": 45224565.0,
      "step": 24255
    },
    {
      "entropy": 5.9812963008880615,
      "epoch": 2.142352525609325,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004539721818626033,
      "loss": 5.6565,
      "mean_token_accuracy": 0.18399664312601088,
      "num_tokens": 45234574.0,
      "step": 24260
    },
    {
      "entropy": 5.989774942398071,
      "epoch": 2.1427940657011657,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004539530949042829,
      "loss": 5.6459,
      "mean_token_accuracy": 0.1789031982421875,
      "num_tokens": 45242707.0,
      "step": 24265
    },
    {
      "entropy": 6.1041992664337155,
      "epoch": 2.143235605793006,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004539340044403731,
      "loss": 5.8952,
      "mean_token_accuracy": 0.16298793405294418,
      "num_tokens": 45252873.0,
      "step": 24270
    },
    {
      "entropy": 6.094801139831543,
      "epoch": 2.1436771458848463,
      "grad_norm": 2.421875,
      "learning_rate": 0.000453914910471248,
      "loss": 5.7438,
      "mean_token_accuracy": 0.17693345248699188,
      "num_tokens": 45262833.0,
      "step": 24275
    },
    {
      "entropy": 6.029384803771973,
      "epoch": 2.1441186859766868,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004538958129972815,
      "loss": 5.7513,
      "mean_token_accuracy": 0.17815324515104294,
      "num_tokens": 45271798.0,
      "step": 24280
    },
    {
      "entropy": 6.082404613494873,
      "epoch": 2.144560226068527,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004538767120188479,
      "loss": 5.68,
      "mean_token_accuracy": 0.1789723441004753,
      "num_tokens": 45281470.0,
      "step": 24285
    },
    {
      "entropy": 6.123211288452149,
      "epoch": 2.1450017661603673,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004538576075363211,
      "loss": 5.7351,
      "mean_token_accuracy": 0.17237321883440018,
      "num_tokens": 45290685.0,
      "step": 24290
    },
    {
      "entropy": 6.054110813140869,
      "epoch": 2.145443306252208,
      "grad_norm": 3.53125,
      "learning_rate": 0.00045383849955007554,
      "loss": 5.7117,
      "mean_token_accuracy": 0.17471117675304412,
      "num_tokens": 45299631.0,
      "step": 24295
    },
    {
      "entropy": 6.086636257171631,
      "epoch": 2.145884846344048,
      "grad_norm": 5.71875,
      "learning_rate": 0.00045381938806048544,
      "loss": 5.8373,
      "mean_token_accuracy": 0.17218175530433655,
      "num_tokens": 45309871.0,
      "step": 24300
    },
    {
      "entropy": 6.108902502059936,
      "epoch": 2.1463263864358884,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004538002730679253,
      "loss": 5.6149,
      "mean_token_accuracy": 0.18412218242883682,
      "num_tokens": 45319233.0,
      "step": 24305
    },
    {
      "entropy": 6.003004741668701,
      "epoch": 2.146767926527729,
      "grad_norm": 1.859375,
      "learning_rate": 0.00045378115457276946,
      "loss": 5.7111,
      "mean_token_accuracy": 0.17469239830970765,
      "num_tokens": 45329467.0,
      "step": 24310
    },
    {
      "entropy": 6.026943159103394,
      "epoch": 2.147209466619569,
      "grad_norm": 2.484375,
      "learning_rate": 0.00045376203257539244,
      "loss": 5.7174,
      "mean_token_accuracy": 0.17661588042974471,
      "num_tokens": 45339124.0,
      "step": 24315
    },
    {
      "entropy": 6.080534601211548,
      "epoch": 2.1476510067114094,
      "grad_norm": 2.328125,
      "learning_rate": 0.00045374290707616893,
      "loss": 5.8434,
      "mean_token_accuracy": 0.1682404860854149,
      "num_tokens": 45348973.0,
      "step": 24320
    },
    {
      "entropy": 6.134699153900146,
      "epoch": 2.14809254680325,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004537237780754735,
      "loss": 5.7841,
      "mean_token_accuracy": 0.1717480778694153,
      "num_tokens": 45359161.0,
      "step": 24325
    },
    {
      "entropy": 6.100856637954712,
      "epoch": 2.14853408689509,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004537046455736809,
      "loss": 5.773,
      "mean_token_accuracy": 0.17893183827400208,
      "num_tokens": 45369019.0,
      "step": 24330
    },
    {
      "entropy": 6.031752443313598,
      "epoch": 2.1489756269869305,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004536855095711659,
      "loss": 5.6852,
      "mean_token_accuracy": 0.18304708302021028,
      "num_tokens": 45378491.0,
      "step": 24335
    },
    {
      "entropy": 6.055395078659058,
      "epoch": 2.149417167078771,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004536663700683035,
      "loss": 5.7744,
      "mean_token_accuracy": 0.17196986824274063,
      "num_tokens": 45387651.0,
      "step": 24340
    },
    {
      "entropy": 6.150043964385986,
      "epoch": 2.149858707170611,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004536472270654685,
      "loss": 5.7866,
      "mean_token_accuracy": 0.16701131612062453,
      "num_tokens": 45396524.0,
      "step": 24345
    },
    {
      "entropy": 6.075572538375854,
      "epoch": 2.1503002472624515,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004536280805630358,
      "loss": 5.7808,
      "mean_token_accuracy": 0.1756866380572319,
      "num_tokens": 45404849.0,
      "step": 24350
    },
    {
      "entropy": 6.075936317443848,
      "epoch": 2.150741787354292,
      "grad_norm": 1.578125,
      "learning_rate": 0.00045360893056138076,
      "loss": 5.6981,
      "mean_token_accuracy": 0.17938045412302017,
      "num_tokens": 45413628.0,
      "step": 24355
    },
    {
      "entropy": 6.049666976928711,
      "epoch": 2.151183327446132,
      "grad_norm": 2.296875,
      "learning_rate": 0.00045358977706087827,
      "loss": 5.7224,
      "mean_token_accuracy": 0.18417874127626419,
      "num_tokens": 45423190.0,
      "step": 24360
    },
    {
      "entropy": 5.990799760818481,
      "epoch": 2.1516248675379726,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004535706200619036,
      "loss": 5.6865,
      "mean_token_accuracy": 0.17843838632106782,
      "num_tokens": 45432075.0,
      "step": 24365
    },
    {
      "entropy": 6.055817413330078,
      "epoch": 2.1520664076298126,
      "grad_norm": 2.0,
      "learning_rate": 0.00045355145956483205,
      "loss": 5.845,
      "mean_token_accuracy": 0.16700876504182816,
      "num_tokens": 45442627.0,
      "step": 24370
    },
    {
      "entropy": 6.084476613998413,
      "epoch": 2.152507947721653,
      "grad_norm": 2.265625,
      "learning_rate": 0.00045353229557003896,
      "loss": 5.7252,
      "mean_token_accuracy": 0.1783043399453163,
      "num_tokens": 45452818.0,
      "step": 24375
    },
    {
      "entropy": 6.040212345123291,
      "epoch": 2.1529494878134936,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004535131280778997,
      "loss": 5.6995,
      "mean_token_accuracy": 0.18211660236120225,
      "num_tokens": 45461964.0,
      "step": 24380
    },
    {
      "entropy": 6.036298131942749,
      "epoch": 2.1533910279053337,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004534939570887898,
      "loss": 5.7129,
      "mean_token_accuracy": 0.17944927811622619,
      "num_tokens": 45471700.0,
      "step": 24385
    },
    {
      "entropy": 6.05089545249939,
      "epoch": 2.153832567997174,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004534747826030848,
      "loss": 5.6914,
      "mean_token_accuracy": 0.17261157929897308,
      "num_tokens": 45481291.0,
      "step": 24390
    },
    {
      "entropy": 6.035101413726807,
      "epoch": 2.1542741080890146,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004534556046211604,
      "loss": 5.725,
      "mean_token_accuracy": 0.1729696810245514,
      "num_tokens": 45490317.0,
      "step": 24395
    },
    {
      "entropy": 6.080319929122925,
      "epoch": 2.1547156481808547,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00045343642314339206,
      "loss": 5.79,
      "mean_token_accuracy": 0.17096616476774215,
      "num_tokens": 45499688.0,
      "step": 24400
    },
    {
      "entropy": 6.0606780529022215,
      "epoch": 2.155157188272695,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004534172381701558,
      "loss": 5.712,
      "mean_token_accuracy": 0.18266336172819136,
      "num_tokens": 45508951.0,
      "step": 24405
    },
    {
      "entropy": 6.126812267303467,
      "epoch": 2.1555987283645357,
      "grad_norm": 1.75,
      "learning_rate": 0.0004533980497018272,
      "loss": 5.8017,
      "mean_token_accuracy": 0.1770400285720825,
      "num_tokens": 45518955.0,
      "step": 24410
    },
    {
      "entropy": 6.1009923934936525,
      "epoch": 2.1560402684563758,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004533788577387824,
      "loss": 5.7591,
      "mean_token_accuracy": 0.16928358525037765,
      "num_tokens": 45527869.0,
      "step": 24415
    },
    {
      "entropy": 6.008814334869385,
      "epoch": 2.1564818085482163,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004533596622813971,
      "loss": 5.6753,
      "mean_token_accuracy": 0.18052051067352295,
      "num_tokens": 45537250.0,
      "step": 24420
    },
    {
      "entropy": 5.990625190734863,
      "epoch": 2.1569233486400563,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004533404633300475,
      "loss": 5.7296,
      "mean_token_accuracy": 0.1704665392637253,
      "num_tokens": 45547204.0,
      "step": 24425
    },
    {
      "entropy": 5.981370449066162,
      "epoch": 2.157364888731897,
      "grad_norm": 1.5625,
      "learning_rate": 0.00045332126088510965,
      "loss": 5.7153,
      "mean_token_accuracy": 0.17207056432962417,
      "num_tokens": 45555464.0,
      "step": 24430
    },
    {
      "entropy": 6.094027757644653,
      "epoch": 2.1578064288237373,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004533020549469597,
      "loss": 5.6645,
      "mean_token_accuracy": 0.1825187847018242,
      "num_tokens": 45564797.0,
      "step": 24435
    },
    {
      "entropy": 6.064813470840454,
      "epoch": 2.1582479689155774,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004532828455159739,
      "loss": 5.801,
      "mean_token_accuracy": 0.16723596900701523,
      "num_tokens": 45573317.0,
      "step": 24440
    },
    {
      "entropy": 5.984539747238159,
      "epoch": 2.158689509007418,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004532636325925286,
      "loss": 5.7087,
      "mean_token_accuracy": 0.1704849436879158,
      "num_tokens": 45582838.0,
      "step": 24445
    },
    {
      "entropy": 6.060503625869751,
      "epoch": 2.1591310490992583,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045324441617700014,
      "loss": 5.6103,
      "mean_token_accuracy": 0.18069412112236022,
      "num_tokens": 45592031.0,
      "step": 24450
    },
    {
      "entropy": 6.037649488449096,
      "epoch": 2.1595725891910984,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004532251962697649,
      "loss": 5.8098,
      "mean_token_accuracy": 0.17883156687021257,
      "num_tokens": 45601353.0,
      "step": 24455
    },
    {
      "entropy": 5.994347190856933,
      "epoch": 2.160014129282939,
      "grad_norm": 1.96875,
      "learning_rate": 0.00045320597287119943,
      "loss": 5.7249,
      "mean_token_accuracy": 0.17485858350992203,
      "num_tokens": 45611064.0,
      "step": 24460
    },
    {
      "entropy": 6.126097106933594,
      "epoch": 2.1604556693747794,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004531867459816803,
      "loss": 5.7391,
      "mean_token_accuracy": 0.17616110891103745,
      "num_tokens": 45620704.0,
      "step": 24465
    },
    {
      "entropy": 6.155863666534424,
      "epoch": 2.1608972094666195,
      "grad_norm": 1.71875,
      "learning_rate": 0.00045316751560158414,
      "loss": 5.7984,
      "mean_token_accuracy": 0.16882864534854888,
      "num_tokens": 45629967.0,
      "step": 24470
    },
    {
      "entropy": 6.117015409469604,
      "epoch": 2.16133874955846,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004531482817312877,
      "loss": 5.8084,
      "mean_token_accuracy": 0.16963985115289687,
      "num_tokens": 45639104.0,
      "step": 24475
    },
    {
      "entropy": 6.06401629447937,
      "epoch": 2.1617802896503004,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00045312904437116774,
      "loss": 5.871,
      "mean_token_accuracy": 0.16282245963811875,
      "num_tokens": 45649217.0,
      "step": 24480
    },
    {
      "entropy": 6.121313571929932,
      "epoch": 2.1622218297421405,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004531098035216011,
      "loss": 5.8399,
      "mean_token_accuracy": 0.1657326936721802,
      "num_tokens": 45658969.0,
      "step": 24485
    },
    {
      "entropy": 6.023886823654175,
      "epoch": 2.162663369833981,
      "grad_norm": 2.15625,
      "learning_rate": 0.00045309055918296465,
      "loss": 5.6368,
      "mean_token_accuracy": 0.1858410954475403,
      "num_tokens": 45668427.0,
      "step": 24490
    },
    {
      "entropy": 5.957619905471802,
      "epoch": 2.1631049099258215,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004530713113556355,
      "loss": 5.5908,
      "mean_token_accuracy": 0.18484682589769363,
      "num_tokens": 45676675.0,
      "step": 24495
    },
    {
      "entropy": 5.9724523544311525,
      "epoch": 2.1635464500176615,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00045305206003999057,
      "loss": 5.6233,
      "mean_token_accuracy": 0.18563942313194276,
      "num_tokens": 45685756.0,
      "step": 24500
    },
    {
      "entropy": 6.106640434265136,
      "epoch": 2.163987990109502,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004530328052364071,
      "loss": 5.7403,
      "mean_token_accuracy": 0.17958677411079407,
      "num_tokens": 45694770.0,
      "step": 24505
    },
    {
      "entropy": 6.0374328136444095,
      "epoch": 2.164429530201342,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004530135469452622,
      "loss": 5.6435,
      "mean_token_accuracy": 0.18216670751571656,
      "num_tokens": 45703940.0,
      "step": 24510
    },
    {
      "entropy": 6.031857204437256,
      "epoch": 2.1648710702931826,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004529942851669333,
      "loss": 5.7919,
      "mean_token_accuracy": 0.1722160503268242,
      "num_tokens": 45713875.0,
      "step": 24515
    },
    {
      "entropy": 6.073873090744018,
      "epoch": 2.165312610385023,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004529750199017974,
      "loss": 5.7972,
      "mean_token_accuracy": 0.16576025784015655,
      "num_tokens": 45723870.0,
      "step": 24520
    },
    {
      "entropy": 5.995216608047485,
      "epoch": 2.165754150476863,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004529557511502322,
      "loss": 5.6988,
      "mean_token_accuracy": 0.17947032451629638,
      "num_tokens": 45733269.0,
      "step": 24525
    },
    {
      "entropy": 6.075911808013916,
      "epoch": 2.1661956905687036,
      "grad_norm": 2.0,
      "learning_rate": 0.00045293647891261494,
      "loss": 5.7749,
      "mean_token_accuracy": 0.16606410145759581,
      "num_tokens": 45743208.0,
      "step": 24530
    },
    {
      "entropy": 6.177947282791138,
      "epoch": 2.166637230660544,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004529172031893233,
      "loss": 5.6981,
      "mean_token_accuracy": 0.18266727924346923,
      "num_tokens": 45752411.0,
      "step": 24535
    },
    {
      "entropy": 6.009743499755859,
      "epoch": 2.167078770752384,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00045289792398073487,
      "loss": 5.7134,
      "mean_token_accuracy": 0.179906165599823,
      "num_tokens": 45761770.0,
      "step": 24540
    },
    {
      "entropy": 6.0136627674102785,
      "epoch": 2.1675203108442247,
      "grad_norm": 1.515625,
      "learning_rate": 0.00045287864128722724,
      "loss": 5.7028,
      "mean_token_accuracy": 0.17485223710536957,
      "num_tokens": 45771227.0,
      "step": 24545
    },
    {
      "entropy": 5.996882772445678,
      "epoch": 2.167961850936065,
      "grad_norm": 2.625,
      "learning_rate": 0.0004528593551091782,
      "loss": 5.6479,
      "mean_token_accuracy": 0.1816339448094368,
      "num_tokens": 45780558.0,
      "step": 24550
    },
    {
      "entropy": 6.009071350097656,
      "epoch": 2.1684033910279052,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045284006544696554,
      "loss": 5.7636,
      "mean_token_accuracy": 0.17727895975112914,
      "num_tokens": 45790078.0,
      "step": 24555
    },
    {
      "entropy": 6.055781412124634,
      "epoch": 2.1688449311197457,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045282077230096715,
      "loss": 5.7275,
      "mean_token_accuracy": 0.17577870637178422,
      "num_tokens": 45799962.0,
      "step": 24560
    },
    {
      "entropy": 6.011180877685547,
      "epoch": 2.169286471211586,
      "grad_norm": 1.53125,
      "learning_rate": 0.00045280147567156097,
      "loss": 5.614,
      "mean_token_accuracy": 0.19487434923648833,
      "num_tokens": 45809599.0,
      "step": 24565
    },
    {
      "entropy": 5.937513828277588,
      "epoch": 2.1697280113034263,
      "grad_norm": 1.765625,
      "learning_rate": 0.000452782175559125,
      "loss": 5.6683,
      "mean_token_accuracy": 0.18337731212377548,
      "num_tokens": 45818318.0,
      "step": 24570
    },
    {
      "entropy": 6.0836492538452145,
      "epoch": 2.170169551395267,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004527628719640374,
      "loss": 5.7781,
      "mean_token_accuracy": 0.17641180008649826,
      "num_tokens": 45828083.0,
      "step": 24575
    },
    {
      "entropy": 6.081655979156494,
      "epoch": 2.170611091487107,
      "grad_norm": 2.171875,
      "learning_rate": 0.00045274356488667614,
      "loss": 5.7402,
      "mean_token_accuracy": 0.17199944257736205,
      "num_tokens": 45838441.0,
      "step": 24580
    },
    {
      "entropy": 6.039579010009765,
      "epoch": 2.1710526315789473,
      "grad_norm": 1.90625,
      "learning_rate": 0.00045272425432741953,
      "loss": 5.7389,
      "mean_token_accuracy": 0.17408603727817534,
      "num_tokens": 45848047.0,
      "step": 24585
    },
    {
      "entropy": 6.075168418884277,
      "epoch": 2.171494171670788,
      "grad_norm": 1.546875,
      "learning_rate": 0.00045270494028664586,
      "loss": 5.7449,
      "mean_token_accuracy": 0.17228063941001892,
      "num_tokens": 45857675.0,
      "step": 24590
    },
    {
      "entropy": 6.009345769882202,
      "epoch": 2.171935711762628,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004526856227647335,
      "loss": 5.6769,
      "mean_token_accuracy": 0.18456998765468596,
      "num_tokens": 45866506.0,
      "step": 24595
    },
    {
      "entropy": 6.132862758636475,
      "epoch": 2.1723772518544684,
      "grad_norm": 1.75,
      "learning_rate": 0.0004526663017620609,
      "loss": 5.7685,
      "mean_token_accuracy": 0.17352699786424636,
      "num_tokens": 45875955.0,
      "step": 24600
    },
    {
      "entropy": 6.080857038497925,
      "epoch": 2.172818791946309,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004526469772790063,
      "loss": 5.7117,
      "mean_token_accuracy": 0.1772049441933632,
      "num_tokens": 45884518.0,
      "step": 24605
    },
    {
      "entropy": 5.953561496734619,
      "epoch": 2.173260332038149,
      "grad_norm": 2.125,
      "learning_rate": 0.0004526276493159487,
      "loss": 5.6651,
      "mean_token_accuracy": 0.18071816265583038,
      "num_tokens": 45894006.0,
      "step": 24610
    },
    {
      "entropy": 5.970248222351074,
      "epoch": 2.1737018721299894,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004526083178732663,
      "loss": 5.6438,
      "mean_token_accuracy": 0.18505288660526276,
      "num_tokens": 45902739.0,
      "step": 24615
    },
    {
      "entropy": 5.936252784729004,
      "epoch": 2.17414341222183,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045258898295133797,
      "loss": 5.5694,
      "mean_token_accuracy": 0.18849026113748552,
      "num_tokens": 45911418.0,
      "step": 24620
    },
    {
      "entropy": 6.026677036285401,
      "epoch": 2.17458495231367,
      "grad_norm": 3.5625,
      "learning_rate": 0.00045256964455054255,
      "loss": 5.7262,
      "mean_token_accuracy": 0.18204082548618317,
      "num_tokens": 45921135.0,
      "step": 24625
    },
    {
      "entropy": 6.062057304382324,
      "epoch": 2.1750264924055105,
      "grad_norm": 2.421875,
      "learning_rate": 0.00045255030267125874,
      "loss": 5.6697,
      "mean_token_accuracy": 0.17753322720527648,
      "num_tokens": 45930276.0,
      "step": 24630
    },
    {
      "entropy": 6.07206358909607,
      "epoch": 2.175468032497351,
      "grad_norm": 2.0,
      "learning_rate": 0.00045253095731386553,
      "loss": 5.7079,
      "mean_token_accuracy": 0.17191194891929626,
      "num_tokens": 45939466.0,
      "step": 24635
    },
    {
      "entropy": 6.049936771392822,
      "epoch": 2.175909572589191,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045251160847874174,
      "loss": 5.6565,
      "mean_token_accuracy": 0.17622894048690796,
      "num_tokens": 45948158.0,
      "step": 24640
    },
    {
      "entropy": 6.055775880813599,
      "epoch": 2.1763511126810315,
      "grad_norm": 2.296875,
      "learning_rate": 0.00045249225616626656,
      "loss": 5.8166,
      "mean_token_accuracy": 0.17108219861984253,
      "num_tokens": 45957759.0,
      "step": 24645
    },
    {
      "entropy": 6.036308002471924,
      "epoch": 2.1767926527728716,
      "grad_norm": 2.171875,
      "learning_rate": 0.000452472900376819,
      "loss": 5.7137,
      "mean_token_accuracy": 0.1749040186405182,
      "num_tokens": 45966689.0,
      "step": 24650
    },
    {
      "entropy": 6.111075687408447,
      "epoch": 2.177234192864712,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00045245354111077827,
      "loss": 5.8053,
      "mean_token_accuracy": 0.17096136957407,
      "num_tokens": 45975618.0,
      "step": 24655
    },
    {
      "entropy": 6.126403093338013,
      "epoch": 2.1776757329565526,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00045243417836852355,
      "loss": 5.71,
      "mean_token_accuracy": 0.17604216784238816,
      "num_tokens": 45984500.0,
      "step": 24660
    },
    {
      "entropy": 6.06546049118042,
      "epoch": 2.1781172730483926,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004524148121504343,
      "loss": 5.7356,
      "mean_token_accuracy": 0.17040737718343735,
      "num_tokens": 45994471.0,
      "step": 24665
    },
    {
      "entropy": 6.12561182975769,
      "epoch": 2.178558813140233,
      "grad_norm": 1.734375,
      "learning_rate": 0.00045239544245688967,
      "loss": 5.8067,
      "mean_token_accuracy": 0.169516359269619,
      "num_tokens": 46003757.0,
      "step": 24670
    },
    {
      "entropy": 6.051066160202026,
      "epoch": 2.1790003532320736,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004523760692882693,
      "loss": 5.6531,
      "mean_token_accuracy": 0.18267683386802674,
      "num_tokens": 46013512.0,
      "step": 24675
    },
    {
      "entropy": 6.070747756958008,
      "epoch": 2.1794418933239137,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004523566926449526,
      "loss": 5.7172,
      "mean_token_accuracy": 0.1808073028922081,
      "num_tokens": 46022867.0,
      "step": 24680
    },
    {
      "entropy": 6.015013408660889,
      "epoch": 2.179883433415754,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045233731252731905,
      "loss": 5.7415,
      "mean_token_accuracy": 0.17586106359958648,
      "num_tokens": 46031799.0,
      "step": 24685
    },
    {
      "entropy": 6.028363752365112,
      "epoch": 2.1803249735075947,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00045231792893574846,
      "loss": 5.777,
      "mean_token_accuracy": 0.16805811673402787,
      "num_tokens": 46041917.0,
      "step": 24690
    },
    {
      "entropy": 6.104132795333863,
      "epoch": 2.1807665135994347,
      "grad_norm": 3.625,
      "learning_rate": 0.00045229854187062054,
      "loss": 5.7444,
      "mean_token_accuracy": 0.17390409708023072,
      "num_tokens": 46051629.0,
      "step": 24695
    },
    {
      "entropy": 6.003335428237915,
      "epoch": 2.1812080536912752,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004522791513323149,
      "loss": 5.663,
      "mean_token_accuracy": 0.18825505971908568,
      "num_tokens": 46060840.0,
      "step": 24700
    },
    {
      "entropy": 5.932952737808227,
      "epoch": 2.1816495937831153,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00045225975732121165,
      "loss": 5.6783,
      "mean_token_accuracy": 0.1804504543542862,
      "num_tokens": 46069774.0,
      "step": 24705
    },
    {
      "entropy": 6.078145742416382,
      "epoch": 2.182091133874956,
      "grad_norm": 1.875,
      "learning_rate": 0.00045224035983769054,
      "loss": 5.805,
      "mean_token_accuracy": 0.1669962465763092,
      "num_tokens": 46079377.0,
      "step": 24710
    },
    {
      "entropy": 6.010259628295898,
      "epoch": 2.1825326739667963,
      "grad_norm": 2.21875,
      "learning_rate": 0.00045222095888213155,
      "loss": 5.6878,
      "mean_token_accuracy": 0.18172694742679596,
      "num_tokens": 46088190.0,
      "step": 24715
    },
    {
      "entropy": 6.1052031993865965,
      "epoch": 2.1829742140586363,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00045220155445491465,
      "loss": 5.7354,
      "mean_token_accuracy": 0.18364651799201964,
      "num_tokens": 46097487.0,
      "step": 24720
    },
    {
      "entropy": 6.174369287490845,
      "epoch": 2.183415754150477,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004521821465564203,
      "loss": 5.805,
      "mean_token_accuracy": 0.17522414177656173,
      "num_tokens": 46106554.0,
      "step": 24725
    },
    {
      "entropy": 6.0853431224823,
      "epoch": 2.1838572942423173,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004521627351870283,
      "loss": 5.81,
      "mean_token_accuracy": 0.16841111779212953,
      "num_tokens": 46115665.0,
      "step": 24730
    },
    {
      "entropy": 6.076347208023071,
      "epoch": 2.1842988343341574,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004521433203471192,
      "loss": 5.6877,
      "mean_token_accuracy": 0.1818760961294174,
      "num_tokens": 46124863.0,
      "step": 24735
    },
    {
      "entropy": 6.138796472549439,
      "epoch": 2.184740374425998,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00045212390203707315,
      "loss": 5.7131,
      "mean_token_accuracy": 0.17471237033605574,
      "num_tokens": 46133986.0,
      "step": 24740
    },
    {
      "entropy": 5.991206312179566,
      "epoch": 2.1851819145178384,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004521044802572705,
      "loss": 5.6449,
      "mean_token_accuracy": 0.18672870695590973,
      "num_tokens": 46142234.0,
      "step": 24745
    },
    {
      "entropy": 5.977221155166626,
      "epoch": 2.1856234546096784,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00045208505500809196,
      "loss": 5.7379,
      "mean_token_accuracy": 0.18073842227458953,
      "num_tokens": 46150976.0,
      "step": 24750
    },
    {
      "entropy": 6.0430260181427,
      "epoch": 2.186064994701519,
      "grad_norm": 2.984375,
      "learning_rate": 0.00045206562628991785,
      "loss": 5.7493,
      "mean_token_accuracy": 0.17779294550418853,
      "num_tokens": 46160541.0,
      "step": 24755
    },
    {
      "entropy": 6.110772562026978,
      "epoch": 2.1865065347933594,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045204619410312875,
      "loss": 5.7928,
      "mean_token_accuracy": 0.1720835417509079,
      "num_tokens": 46170958.0,
      "step": 24760
    },
    {
      "entropy": 6.098491859436035,
      "epoch": 2.1869480748851995,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004520267584481055,
      "loss": 5.7297,
      "mean_token_accuracy": 0.17830943763256074,
      "num_tokens": 46180293.0,
      "step": 24765
    },
    {
      "entropy": 6.026238107681275,
      "epoch": 2.18738961497704,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00045200731932522873,
      "loss": 5.6907,
      "mean_token_accuracy": 0.1844901666045189,
      "num_tokens": 46189264.0,
      "step": 24770
    },
    {
      "entropy": 6.109082126617432,
      "epoch": 2.1878311550688805,
      "grad_norm": 2.125,
      "learning_rate": 0.0004519878767348792,
      "loss": 5.8639,
      "mean_token_accuracy": 0.16633577942848204,
      "num_tokens": 46199417.0,
      "step": 24775
    },
    {
      "entropy": 6.030775260925293,
      "epoch": 2.1882726951607205,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004519684306774378,
      "loss": 5.6734,
      "mean_token_accuracy": 0.17879429012537001,
      "num_tokens": 46209360.0,
      "step": 24780
    },
    {
      "entropy": 6.0273580074310305,
      "epoch": 2.188714235252561,
      "grad_norm": 4.03125,
      "learning_rate": 0.00045194898115328555,
      "loss": 5.6878,
      "mean_token_accuracy": 0.1733464404940605,
      "num_tokens": 46218414.0,
      "step": 24785
    },
    {
      "entropy": 6.049801397323608,
      "epoch": 2.189155775344401,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004519295281628033,
      "loss": 5.6891,
      "mean_token_accuracy": 0.1817675545811653,
      "num_tokens": 46227526.0,
      "step": 24790
    },
    {
      "entropy": 6.115029239654541,
      "epoch": 2.1895973154362416,
      "grad_norm": 4.59375,
      "learning_rate": 0.0004519100717063723,
      "loss": 5.6428,
      "mean_token_accuracy": 0.1805925875902176,
      "num_tokens": 46236041.0,
      "step": 24795
    },
    {
      "entropy": 6.018642377853394,
      "epoch": 2.190038855528082,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00045189061178437363,
      "loss": 5.6426,
      "mean_token_accuracy": 0.17868036776781082,
      "num_tokens": 46244557.0,
      "step": 24800
    },
    {
      "entropy": 6.031788015365601,
      "epoch": 2.190480395619922,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004518711483971885,
      "loss": 5.6638,
      "mean_token_accuracy": 0.18209255635738372,
      "num_tokens": 46253957.0,
      "step": 24805
    },
    {
      "entropy": 6.072386884689331,
      "epoch": 2.1909219357117626,
      "grad_norm": 2.28125,
      "learning_rate": 0.000451851681545198,
      "loss": 5.7743,
      "mean_token_accuracy": 0.17269730269908906,
      "num_tokens": 46263110.0,
      "step": 24810
    },
    {
      "entropy": 6.001974582672119,
      "epoch": 2.191363475803603,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045183221122878375,
      "loss": 5.7639,
      "mean_token_accuracy": 0.17331134676933288,
      "num_tokens": 46272163.0,
      "step": 24815
    },
    {
      "entropy": 6.076730060577392,
      "epoch": 2.191805015895443,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045181273744832707,
      "loss": 5.7429,
      "mean_token_accuracy": 0.17555874735116958,
      "num_tokens": 46281097.0,
      "step": 24820
    },
    {
      "entropy": 6.148045921325684,
      "epoch": 2.1922465559872837,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004517932602042094,
      "loss": 5.7119,
      "mean_token_accuracy": 0.1703383207321167,
      "num_tokens": 46291552.0,
      "step": 24825
    },
    {
      "entropy": 6.094576787948609,
      "epoch": 2.192688096079124,
      "grad_norm": 2.703125,
      "learning_rate": 0.00045177377949681235,
      "loss": 5.7211,
      "mean_token_accuracy": 0.17696198970079421,
      "num_tokens": 46300661.0,
      "step": 24830
    },
    {
      "entropy": 6.035983085632324,
      "epoch": 2.1931296361709642,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045175429532651743,
      "loss": 5.6686,
      "mean_token_accuracy": 0.17993694990873338,
      "num_tokens": 46309705.0,
      "step": 24835
    },
    {
      "entropy": 5.990501165390015,
      "epoch": 2.1935711762628047,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004517348076937064,
      "loss": 5.6837,
      "mean_token_accuracy": 0.17819372713565826,
      "num_tokens": 46319701.0,
      "step": 24840
    },
    {
      "entropy": 5.9667503356933596,
      "epoch": 2.194012716354645,
      "grad_norm": 1.890625,
      "learning_rate": 0.00045171531659876104,
      "loss": 5.6899,
      "mean_token_accuracy": 0.1794554829597473,
      "num_tokens": 46328195.0,
      "step": 24845
    },
    {
      "entropy": 6.162162256240845,
      "epoch": 2.1944542564464853,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004516958220420631,
      "loss": 5.7507,
      "mean_token_accuracy": 0.17743850946426393,
      "num_tokens": 46337518.0,
      "step": 24850
    },
    {
      "entropy": 6.009145450592041,
      "epoch": 2.194895796538326,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004516763240239945,
      "loss": 5.6383,
      "mean_token_accuracy": 0.18466827869415284,
      "num_tokens": 46346406.0,
      "step": 24855
    },
    {
      "entropy": 5.9547020435333256,
      "epoch": 2.195337336630166,
      "grad_norm": 2.125,
      "learning_rate": 0.0004516568225449372,
      "loss": 5.6734,
      "mean_token_accuracy": 0.17965957671403884,
      "num_tokens": 46355284.0,
      "step": 24860
    },
    {
      "entropy": 6.073467445373535,
      "epoch": 2.1957788767220063,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00045163731760527326,
      "loss": 5.7805,
      "mean_token_accuracy": 0.17209069281816483,
      "num_tokens": 46364982.0,
      "step": 24865
    },
    {
      "entropy": 5.996128988265991,
      "epoch": 2.196220416813847,
      "grad_norm": 2.625,
      "learning_rate": 0.0004516178092053847,
      "loss": 5.612,
      "mean_token_accuracy": 0.18363979756832122,
      "num_tokens": 46373923.0,
      "step": 24870
    },
    {
      "entropy": 6.012650442123413,
      "epoch": 2.196661956905687,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004515982973456537,
      "loss": 5.6242,
      "mean_token_accuracy": 0.18654758632183074,
      "num_tokens": 46382194.0,
      "step": 24875
    },
    {
      "entropy": 5.995025014877319,
      "epoch": 2.1971034969975274,
      "grad_norm": 3.375,
      "learning_rate": 0.0004515787820264624,
      "loss": 5.7056,
      "mean_token_accuracy": 0.17500094026327134,
      "num_tokens": 46391463.0,
      "step": 24880
    },
    {
      "entropy": 6.042351961135864,
      "epoch": 2.197545037089368,
      "grad_norm": 1.625,
      "learning_rate": 0.00045155926324819335,
      "loss": 5.7414,
      "mean_token_accuracy": 0.17896852344274522,
      "num_tokens": 46401235.0,
      "step": 24885
    },
    {
      "entropy": 6.049805927276611,
      "epoch": 2.197986577181208,
      "grad_norm": 2.046875,
      "learning_rate": 0.00045153974101122866,
      "loss": 5.7447,
      "mean_token_accuracy": 0.17498043328523635,
      "num_tokens": 46410631.0,
      "step": 24890
    },
    {
      "entropy": 6.1298009872436525,
      "epoch": 2.1984281172730484,
      "grad_norm": 1.5,
      "learning_rate": 0.00045152021531595085,
      "loss": 5.7665,
      "mean_token_accuracy": 0.16962603032588958,
      "num_tokens": 46420467.0,
      "step": 24895
    },
    {
      "entropy": 6.028362560272217,
      "epoch": 2.198869657364889,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004515006861627424,
      "loss": 5.6687,
      "mean_token_accuracy": 0.17999161183834075,
      "num_tokens": 46430276.0,
      "step": 24900
    },
    {
      "entropy": 6.056849718093872,
      "epoch": 2.199311197456729,
      "grad_norm": 1.828125,
      "learning_rate": 0.00045148115355198596,
      "loss": 5.7987,
      "mean_token_accuracy": 0.1747591659426689,
      "num_tokens": 46439227.0,
      "step": 24905
    },
    {
      "entropy": 6.012310600280761,
      "epoch": 2.1997527375485695,
      "grad_norm": 2.453125,
      "learning_rate": 0.000451461617484064,
      "loss": 5.7463,
      "mean_token_accuracy": 0.17650977969169618,
      "num_tokens": 46449043.0,
      "step": 24910
    },
    {
      "entropy": 6.0541338443756105,
      "epoch": 2.20019427764041,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004514420779593594,
      "loss": 5.7169,
      "mean_token_accuracy": 0.17532592564821242,
      "num_tokens": 46458332.0,
      "step": 24915
    },
    {
      "entropy": 6.1258392333984375,
      "epoch": 2.20063581773225,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045142253497825477,
      "loss": 5.6551,
      "mean_token_accuracy": 0.17967853397130967,
      "num_tokens": 46467208.0,
      "step": 24920
    },
    {
      "entropy": 6.036689758300781,
      "epoch": 2.2010773578240905,
      "grad_norm": 2.0,
      "learning_rate": 0.00045140298854113306,
      "loss": 5.6803,
      "mean_token_accuracy": 0.18510361909866332,
      "num_tokens": 46476715.0,
      "step": 24925
    },
    {
      "entropy": 5.878598737716675,
      "epoch": 2.2015188979159306,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00045138343864837703,
      "loss": 5.6293,
      "mean_token_accuracy": 0.18931738287210464,
      "num_tokens": 46484783.0,
      "step": 24930
    },
    {
      "entropy": 5.968104600906372,
      "epoch": 2.201960438007771,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004513638853003699,
      "loss": 5.6387,
      "mean_token_accuracy": 0.18209874927997588,
      "num_tokens": 46493616.0,
      "step": 24935
    },
    {
      "entropy": 6.092381381988526,
      "epoch": 2.2024019780996116,
      "grad_norm": 2.375,
      "learning_rate": 0.00045134432849749446,
      "loss": 5.7363,
      "mean_token_accuracy": 0.1727239429950714,
      "num_tokens": 46502779.0,
      "step": 24940
    },
    {
      "entropy": 6.013078308105468,
      "epoch": 2.2028435181914516,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00045132476824013396,
      "loss": 5.6985,
      "mean_token_accuracy": 0.1767430916428566,
      "num_tokens": 46511506.0,
      "step": 24945
    },
    {
      "entropy": 6.074885702133178,
      "epoch": 2.203285058283292,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004513052045286715,
      "loss": 5.7367,
      "mean_token_accuracy": 0.1675547331571579,
      "num_tokens": 46520219.0,
      "step": 24950
    },
    {
      "entropy": 6.114455938339233,
      "epoch": 2.2037265983751326,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004512856373634904,
      "loss": 5.7209,
      "mean_token_accuracy": 0.17937218695878981,
      "num_tokens": 46528737.0,
      "step": 24955
    },
    {
      "entropy": 6.0671905994415285,
      "epoch": 2.2041681384669727,
      "grad_norm": 2.03125,
      "learning_rate": 0.00045126606674497394,
      "loss": 5.805,
      "mean_token_accuracy": 0.17324816882610322,
      "num_tokens": 46538802.0,
      "step": 24960
    },
    {
      "entropy": 6.042453336715698,
      "epoch": 2.204609678558813,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004512464926735054,
      "loss": 5.7313,
      "mean_token_accuracy": 0.17435165196657182,
      "num_tokens": 46548369.0,
      "step": 24965
    },
    {
      "entropy": 5.994946527481079,
      "epoch": 2.2050512186506537,
      "grad_norm": 2.53125,
      "learning_rate": 0.00045122691514946835,
      "loss": 5.7699,
      "mean_token_accuracy": 0.1749277338385582,
      "num_tokens": 46557876.0,
      "step": 24970
    },
    {
      "entropy": 6.006895637512207,
      "epoch": 2.2054927587424937,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004512073341732463,
      "loss": 5.7191,
      "mean_token_accuracy": 0.1837351441383362,
      "num_tokens": 46566747.0,
      "step": 24975
    },
    {
      "entropy": 6.075999164581299,
      "epoch": 2.2059342988343342,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004511877497452227,
      "loss": 5.7103,
      "mean_token_accuracy": 0.18222341537475586,
      "num_tokens": 46576196.0,
      "step": 24980
    },
    {
      "entropy": 6.13926739692688,
      "epoch": 2.2063758389261743,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004511681618657813,
      "loss": 5.7832,
      "mean_token_accuracy": 0.17218048870563507,
      "num_tokens": 46585734.0,
      "step": 24985
    },
    {
      "entropy": 5.988838481903076,
      "epoch": 2.206817379018015,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00045114857053530586,
      "loss": 5.6565,
      "mean_token_accuracy": 0.17738207429647446,
      "num_tokens": 46595931.0,
      "step": 24990
    },
    {
      "entropy": 6.080410718917847,
      "epoch": 2.2072589191098553,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00045112897575418006,
      "loss": 5.7162,
      "mean_token_accuracy": 0.17486781477928162,
      "num_tokens": 46604688.0,
      "step": 24995
    },
    {
      "entropy": 5.968244791030884,
      "epoch": 2.2077004592016953,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004511093775227878,
      "loss": 5.704,
      "mean_token_accuracy": 0.18291618525981904,
      "num_tokens": 46613860.0,
      "step": 25000
    },
    {
      "entropy": 6.101986265182495,
      "epoch": 2.208141999293536,
      "grad_norm": 2.359375,
      "learning_rate": 0.000451089775841513,
      "loss": 5.7765,
      "mean_token_accuracy": 0.1699947863817215,
      "num_tokens": 46623041.0,
      "step": 25005
    },
    {
      "entropy": 6.0303644180297855,
      "epoch": 2.2085835393853763,
      "grad_norm": 3.21875,
      "learning_rate": 0.00045107017071073963,
      "loss": 5.7295,
      "mean_token_accuracy": 0.17352001070976258,
      "num_tokens": 46633025.0,
      "step": 25010
    },
    {
      "entropy": 6.067092370986939,
      "epoch": 2.2090250794772164,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004510505621308517,
      "loss": 5.7277,
      "mean_token_accuracy": 0.18025416731834412,
      "num_tokens": 46641826.0,
      "step": 25015
    },
    {
      "entropy": 6.127798366546631,
      "epoch": 2.209466619569057,
      "grad_norm": 1.609375,
      "learning_rate": 0.00045103095010223354,
      "loss": 5.8232,
      "mean_token_accuracy": 0.17143535912036895,
      "num_tokens": 46652028.0,
      "step": 25020
    },
    {
      "entropy": 6.040845346450806,
      "epoch": 2.2099081596608974,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00045101133462526903,
      "loss": 5.8005,
      "mean_token_accuracy": 0.17549717724323272,
      "num_tokens": 46660763.0,
      "step": 25025
    },
    {
      "entropy": 6.0685248374938965,
      "epoch": 2.2103496997527374,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004509917157003427,
      "loss": 5.7705,
      "mean_token_accuracy": 0.17176493108272553,
      "num_tokens": 46670213.0,
      "step": 25030
    },
    {
      "entropy": 6.123567724227906,
      "epoch": 2.210791239844578,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004509720933278386,
      "loss": 5.7329,
      "mean_token_accuracy": 0.17670647501945497,
      "num_tokens": 46679658.0,
      "step": 25035
    },
    {
      "entropy": 5.996911144256591,
      "epoch": 2.2112327799364184,
      "grad_norm": 2.625,
      "learning_rate": 0.0004509524675081414,
      "loss": 5.678,
      "mean_token_accuracy": 0.1824314698576927,
      "num_tokens": 46687920.0,
      "step": 25040
    },
    {
      "entropy": 6.034418201446533,
      "epoch": 2.2116743200282585,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004509328382416354,
      "loss": 5.7732,
      "mean_token_accuracy": 0.16758892089128494,
      "num_tokens": 46696779.0,
      "step": 25045
    },
    {
      "entropy": 6.088092279434204,
      "epoch": 2.212115860120099,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004509132055287051,
      "loss": 5.7215,
      "mean_token_accuracy": 0.17092274874448776,
      "num_tokens": 46706568.0,
      "step": 25050
    },
    {
      "entropy": 6.11254358291626,
      "epoch": 2.2125574002119395,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004508935693697352,
      "loss": 5.7614,
      "mean_token_accuracy": 0.17542219907045364,
      "num_tokens": 46715369.0,
      "step": 25055
    },
    {
      "entropy": 6.002852201461792,
      "epoch": 2.2129989403037795,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004508739297651103,
      "loss": 5.7001,
      "mean_token_accuracy": 0.1736611805856228,
      "num_tokens": 46724867.0,
      "step": 25060
    },
    {
      "entropy": 6.034820985794068,
      "epoch": 2.21344048039562,
      "grad_norm": 2.125,
      "learning_rate": 0.00045085428671521514,
      "loss": 5.757,
      "mean_token_accuracy": 0.17864349037408828,
      "num_tokens": 46734611.0,
      "step": 25065
    },
    {
      "entropy": 6.059646844863892,
      "epoch": 2.21388202048746,
      "grad_norm": 2.0,
      "learning_rate": 0.00045083464022043456,
      "loss": 5.7467,
      "mean_token_accuracy": 0.17570917308330536,
      "num_tokens": 46744031.0,
      "step": 25070
    },
    {
      "entropy": 6.004491329193115,
      "epoch": 2.2143235605793006,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00045081499028115337,
      "loss": 5.6826,
      "mean_token_accuracy": 0.18481325656175612,
      "num_tokens": 46753379.0,
      "step": 25075
    },
    {
      "entropy": 6.105690050125122,
      "epoch": 2.214765100671141,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004507953368977565,
      "loss": 5.822,
      "mean_token_accuracy": 0.16593809723854064,
      "num_tokens": 46762625.0,
      "step": 25080
    },
    {
      "entropy": 6.076445436477661,
      "epoch": 2.215206640762981,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00045077568007062893,
      "loss": 5.777,
      "mean_token_accuracy": 0.17136968076229095,
      "num_tokens": 46772577.0,
      "step": 25085
    },
    {
      "entropy": 6.041543960571289,
      "epoch": 2.2156481808548216,
      "grad_norm": 1.859375,
      "learning_rate": 0.00045075601980015573,
      "loss": 5.742,
      "mean_token_accuracy": 0.17774572521448134,
      "num_tokens": 46782226.0,
      "step": 25090
    },
    {
      "entropy": 6.038024663925171,
      "epoch": 2.216089720946662,
      "grad_norm": 1.578125,
      "learning_rate": 0.00045073635608672215,
      "loss": 5.6336,
      "mean_token_accuracy": 0.19029305279254913,
      "num_tokens": 46789988.0,
      "step": 25095
    },
    {
      "entropy": 6.097830152511596,
      "epoch": 2.216531261038502,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045071668893071324,
      "loss": 5.769,
      "mean_token_accuracy": 0.1623351752758026,
      "num_tokens": 46799448.0,
      "step": 25100
    },
    {
      "entropy": 6.079371118545533,
      "epoch": 2.2169728011303427,
      "grad_norm": 2.953125,
      "learning_rate": 0.00045069701833251434,
      "loss": 5.7745,
      "mean_token_accuracy": 0.17346482574939728,
      "num_tokens": 46809829.0,
      "step": 25105
    },
    {
      "entropy": 5.985646677017212,
      "epoch": 2.217414341222183,
      "grad_norm": 1.890625,
      "learning_rate": 0.00045067734429251086,
      "loss": 5.6943,
      "mean_token_accuracy": 0.1781681016087532,
      "num_tokens": 46817975.0,
      "step": 25110
    },
    {
      "entropy": 5.913733720779419,
      "epoch": 2.2178558813140232,
      "grad_norm": 2.015625,
      "learning_rate": 0.00045065766681108793,
      "loss": 5.6843,
      "mean_token_accuracy": 0.1804809257388115,
      "num_tokens": 46827350.0,
      "step": 25115
    },
    {
      "entropy": 6.067097568511963,
      "epoch": 2.2182974214058637,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004506379858886314,
      "loss": 5.7195,
      "mean_token_accuracy": 0.1797318384051323,
      "num_tokens": 46835944.0,
      "step": 25120
    },
    {
      "entropy": 6.046725130081176,
      "epoch": 2.2187389614977038,
      "grad_norm": 2.375,
      "learning_rate": 0.00045061830152552646,
      "loss": 5.6696,
      "mean_token_accuracy": 0.1813129886984825,
      "num_tokens": 46844843.0,
      "step": 25125
    },
    {
      "entropy": 6.032180643081665,
      "epoch": 2.2191805015895443,
      "grad_norm": 2.375,
      "learning_rate": 0.000450598613722159,
      "loss": 5.7997,
      "mean_token_accuracy": 0.1676554873585701,
      "num_tokens": 46854188.0,
      "step": 25130
    },
    {
      "entropy": 6.042804813385009,
      "epoch": 2.2196220416813848,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004505789224789145,
      "loss": 5.7687,
      "mean_token_accuracy": 0.17505763620138168,
      "num_tokens": 46863279.0,
      "step": 25135
    },
    {
      "entropy": 6.072401094436645,
      "epoch": 2.220063581773225,
      "grad_norm": 1.96875,
      "learning_rate": 0.00045055922779617874,
      "loss": 5.7205,
      "mean_token_accuracy": 0.1736467197537422,
      "num_tokens": 46873336.0,
      "step": 25140
    },
    {
      "entropy": 6.081298923492431,
      "epoch": 2.2205051218650653,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004505395296743376,
      "loss": 5.7269,
      "mean_token_accuracy": 0.17711414992809296,
      "num_tokens": 46882497.0,
      "step": 25145
    },
    {
      "entropy": 6.043819570541382,
      "epoch": 2.220946661956906,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00045051982811377696,
      "loss": 5.6584,
      "mean_token_accuracy": 0.18041330128908156,
      "num_tokens": 46891260.0,
      "step": 25150
    },
    {
      "entropy": 6.000118446350098,
      "epoch": 2.221388202048746,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00045050012311488265,
      "loss": 5.634,
      "mean_token_accuracy": 0.18501115292310716,
      "num_tokens": 46899969.0,
      "step": 25155
    },
    {
      "entropy": 5.931546640396118,
      "epoch": 2.2218297421405864,
      "grad_norm": 2.796875,
      "learning_rate": 0.00045048041467804067,
      "loss": 5.6443,
      "mean_token_accuracy": 0.1877874344587326,
      "num_tokens": 46909206.0,
      "step": 25160
    },
    {
      "entropy": 6.0348677158355715,
      "epoch": 2.222271282232427,
      "grad_norm": 2.3125,
      "learning_rate": 0.00045046070280363725,
      "loss": 5.668,
      "mean_token_accuracy": 0.18541376441717147,
      "num_tokens": 46918423.0,
      "step": 25165
    },
    {
      "entropy": 6.0928514957427975,
      "epoch": 2.222712822324267,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004504409874920585,
      "loss": 5.7089,
      "mean_token_accuracy": 0.17581377774477006,
      "num_tokens": 46927817.0,
      "step": 25170
    },
    {
      "entropy": 6.115260314941406,
      "epoch": 2.2231543624161074,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004504212687436905,
      "loss": 5.8006,
      "mean_token_accuracy": 0.17105025947093963,
      "num_tokens": 46936880.0,
      "step": 25175
    },
    {
      "entropy": 6.0691810131073,
      "epoch": 2.223595902507948,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004504015465589196,
      "loss": 5.7375,
      "mean_token_accuracy": 0.1711537465453148,
      "num_tokens": 46946722.0,
      "step": 25180
    },
    {
      "entropy": 6.005175590515137,
      "epoch": 2.224037442599788,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004503818209381322,
      "loss": 5.6852,
      "mean_token_accuracy": 0.17629962414503098,
      "num_tokens": 46956660.0,
      "step": 25185
    },
    {
      "entropy": 6.051830244064331,
      "epoch": 2.2244789826916285,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004503620918817147,
      "loss": 5.6988,
      "mean_token_accuracy": 0.17148443609476088,
      "num_tokens": 46965823.0,
      "step": 25190
    },
    {
      "entropy": 6.019561433792115,
      "epoch": 2.224920522783469,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00045034235939005353,
      "loss": 5.7441,
      "mean_token_accuracy": 0.1735680878162384,
      "num_tokens": 46974933.0,
      "step": 25195
    },
    {
      "entropy": 6.062387084960937,
      "epoch": 2.225362062875309,
      "grad_norm": 2.078125,
      "learning_rate": 0.00045032262346353526,
      "loss": 5.8071,
      "mean_token_accuracy": 0.16900740712881088,
      "num_tokens": 46983506.0,
      "step": 25200
    },
    {
      "entropy": 6.018953895568847,
      "epoch": 2.2258036029671495,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004503028841025465,
      "loss": 5.7096,
      "mean_token_accuracy": 0.1760956257581711,
      "num_tokens": 46992234.0,
      "step": 25205
    },
    {
      "entropy": 5.956487703323364,
      "epoch": 2.2262451430589896,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00045028314130747397,
      "loss": 5.6521,
      "mean_token_accuracy": 0.18242706954479218,
      "num_tokens": 47002249.0,
      "step": 25210
    },
    {
      "entropy": 6.0558051586151125,
      "epoch": 2.22668668315083,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045026339507870435,
      "loss": 5.7872,
      "mean_token_accuracy": 0.17403094917535783,
      "num_tokens": 47010721.0,
      "step": 25215
    },
    {
      "entropy": 6.170264768600464,
      "epoch": 2.2271282232426706,
      "grad_norm": 1.578125,
      "learning_rate": 0.00045024364541662457,
      "loss": 5.7622,
      "mean_token_accuracy": 0.17530993223190308,
      "num_tokens": 47019675.0,
      "step": 25220
    },
    {
      "entropy": 6.062171506881714,
      "epoch": 2.2275697633345106,
      "grad_norm": 1.625,
      "learning_rate": 0.0004502238923216213,
      "loss": 5.7243,
      "mean_token_accuracy": 0.1810107946395874,
      "num_tokens": 47028293.0,
      "step": 25225
    },
    {
      "entropy": 6.0736016750335695,
      "epoch": 2.228011303426351,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004502041357940817,
      "loss": 5.767,
      "mean_token_accuracy": 0.17240528017282486,
      "num_tokens": 47037605.0,
      "step": 25230
    },
    {
      "entropy": 5.998285484313965,
      "epoch": 2.2284528435181916,
      "grad_norm": 2.734375,
      "learning_rate": 0.00045018437583439275,
      "loss": 5.6897,
      "mean_token_accuracy": 0.18019328713417054,
      "num_tokens": 47045889.0,
      "step": 25235
    },
    {
      "entropy": 6.029602432250977,
      "epoch": 2.2288943836100317,
      "grad_norm": 2.71875,
      "learning_rate": 0.00045016461244294154,
      "loss": 5.7168,
      "mean_token_accuracy": 0.17958432734012603,
      "num_tokens": 47055896.0,
      "step": 25240
    },
    {
      "entropy": 6.069905567169189,
      "epoch": 2.229335923701872,
      "grad_norm": 1.5,
      "learning_rate": 0.00045014484562011514,
      "loss": 5.7679,
      "mean_token_accuracy": 0.17534936666488649,
      "num_tokens": 47064980.0,
      "step": 25245
    },
    {
      "entropy": 6.031907606124878,
      "epoch": 2.2297774637937127,
      "grad_norm": 1.75,
      "learning_rate": 0.0004501250753663008,
      "loss": 5.6521,
      "mean_token_accuracy": 0.17913320213556289,
      "num_tokens": 47074045.0,
      "step": 25250
    },
    {
      "entropy": 6.029669666290284,
      "epoch": 2.2302190038855527,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004501053016818859,
      "loss": 5.6718,
      "mean_token_accuracy": 0.1768660753965378,
      "num_tokens": 47083439.0,
      "step": 25255
    },
    {
      "entropy": 6.005933666229248,
      "epoch": 2.230660543977393,
      "grad_norm": 1.65625,
      "learning_rate": 0.00045008552456725766,
      "loss": 5.6767,
      "mean_token_accuracy": 0.17887476533651353,
      "num_tokens": 47091977.0,
      "step": 25260
    },
    {
      "entropy": 5.967308235168457,
      "epoch": 2.2311020840692333,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004500657440228037,
      "loss": 5.7554,
      "mean_token_accuracy": 0.17420131266117095,
      "num_tokens": 47102197.0,
      "step": 25265
    },
    {
      "entropy": 6.093684434890747,
      "epoch": 2.2315436241610738,
      "grad_norm": 2.125,
      "learning_rate": 0.00045004596004891124,
      "loss": 5.7514,
      "mean_token_accuracy": 0.1834702968597412,
      "num_tokens": 47111563.0,
      "step": 25270
    },
    {
      "entropy": 6.021131134033203,
      "epoch": 2.2319851642529143,
      "grad_norm": 1.859375,
      "learning_rate": 0.00045002617264596807,
      "loss": 5.7254,
      "mean_token_accuracy": 0.182077619433403,
      "num_tokens": 47120272.0,
      "step": 25275
    },
    {
      "entropy": 6.023928689956665,
      "epoch": 2.2324267043447543,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004500063818143617,
      "loss": 5.7151,
      "mean_token_accuracy": 0.17729026526212693,
      "num_tokens": 47131023.0,
      "step": 25280
    },
    {
      "entropy": 5.978570508956909,
      "epoch": 2.232868244436595,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004499865875544797,
      "loss": 5.6619,
      "mean_token_accuracy": 0.1788106694817543,
      "num_tokens": 47139968.0,
      "step": 25285
    },
    {
      "entropy": 6.006287622451782,
      "epoch": 2.2333097845284353,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004499667898667101,
      "loss": 5.6794,
      "mean_token_accuracy": 0.18414108604192733,
      "num_tokens": 47149843.0,
      "step": 25290
    },
    {
      "entropy": 6.038427686691284,
      "epoch": 2.2337513246202754,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044994698875144064,
      "loss": 5.7049,
      "mean_token_accuracy": 0.17712724953889847,
      "num_tokens": 47159025.0,
      "step": 25295
    },
    {
      "entropy": 6.0461290836334225,
      "epoch": 2.234192864712116,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004499271842090591,
      "loss": 5.759,
      "mean_token_accuracy": 0.1738530784845352,
      "num_tokens": 47168644.0,
      "step": 25300
    },
    {
      "entropy": 6.058022594451904,
      "epoch": 2.2346344048039564,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00044990737623995357,
      "loss": 5.6705,
      "mean_token_accuracy": 0.18085331320762635,
      "num_tokens": 47178051.0,
      "step": 25305
    },
    {
      "entropy": 6.140856456756592,
      "epoch": 2.2350759448957964,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004498875648445119,
      "loss": 5.7955,
      "mean_token_accuracy": 0.17289526462554933,
      "num_tokens": 47187855.0,
      "step": 25310
    },
    {
      "entropy": 5.987267303466797,
      "epoch": 2.235517484987637,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004498677500231224,
      "loss": 5.7219,
      "mean_token_accuracy": 0.179219987988472,
      "num_tokens": 47196700.0,
      "step": 25315
    },
    {
      "entropy": 6.085606718063355,
      "epoch": 2.2359590250794774,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004498479317761731,
      "loss": 5.7728,
      "mean_token_accuracy": 0.1699157863855362,
      "num_tokens": 47205202.0,
      "step": 25320
    },
    {
      "entropy": 6.02024474143982,
      "epoch": 2.2364005651713175,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004498281101040522,
      "loss": 5.6938,
      "mean_token_accuracy": 0.18322307020425796,
      "num_tokens": 47214803.0,
      "step": 25325
    },
    {
      "entropy": 6.104619550704956,
      "epoch": 2.236842105263158,
      "grad_norm": 2.078125,
      "learning_rate": 0.00044980828500714814,
      "loss": 5.805,
      "mean_token_accuracy": 0.1675583764910698,
      "num_tokens": 47224641.0,
      "step": 25330
    },
    {
      "entropy": 6.041909837722779,
      "epoch": 2.2372836453549985,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004497884564858492,
      "loss": 5.768,
      "mean_token_accuracy": 0.16843321472406386,
      "num_tokens": 47233905.0,
      "step": 25335
    },
    {
      "entropy": 6.040278434753418,
      "epoch": 2.2377251854468385,
      "grad_norm": 1.640625,
      "learning_rate": 0.00044976862454054376,
      "loss": 5.6825,
      "mean_token_accuracy": 0.17843635231256486,
      "num_tokens": 47242209.0,
      "step": 25340
    },
    {
      "entropy": 6.16157374382019,
      "epoch": 2.238166725538679,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004497487891716204,
      "loss": 5.7729,
      "mean_token_accuracy": 0.18140608668327332,
      "num_tokens": 47252609.0,
      "step": 25345
    },
    {
      "entropy": 6.0986353874206545,
      "epoch": 2.238608265630519,
      "grad_norm": 2.046875,
      "learning_rate": 0.00044972895037946755,
      "loss": 5.7559,
      "mean_token_accuracy": 0.1717152863740921,
      "num_tokens": 47261868.0,
      "step": 25350
    },
    {
      "entropy": 6.0343550682067875,
      "epoch": 2.2390498057223596,
      "grad_norm": 2.1875,
      "learning_rate": 0.00044970910816447403,
      "loss": 5.6964,
      "mean_token_accuracy": 0.18116360604763032,
      "num_tokens": 47271404.0,
      "step": 25355
    },
    {
      "entropy": 6.0645795345306395,
      "epoch": 2.2394913458142,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044968926252702835,
      "loss": 5.7759,
      "mean_token_accuracy": 0.17029038816690445,
      "num_tokens": 47280669.0,
      "step": 25360
    },
    {
      "entropy": 6.030475902557373,
      "epoch": 2.23993288590604,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004496694134675195,
      "loss": 5.7695,
      "mean_token_accuracy": 0.16790039986371993,
      "num_tokens": 47290096.0,
      "step": 25365
    },
    {
      "entropy": 6.05766487121582,
      "epoch": 2.2403744259978806,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004496495609863361,
      "loss": 5.7508,
      "mean_token_accuracy": 0.17855436950922013,
      "num_tokens": 47298767.0,
      "step": 25370
    },
    {
      "entropy": 5.95639591217041,
      "epoch": 2.240815966089721,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00044962970508386716,
      "loss": 5.6359,
      "mean_token_accuracy": 0.18819597065448762,
      "num_tokens": 47307950.0,
      "step": 25375
    },
    {
      "entropy": 6.060315847396851,
      "epoch": 2.241257506181561,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004496098457605016,
      "loss": 5.7618,
      "mean_token_accuracy": 0.17924597561359407,
      "num_tokens": 47317963.0,
      "step": 25380
    },
    {
      "entropy": 6.082465648651123,
      "epoch": 2.2416990462734017,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004495899830166284,
      "loss": 5.7242,
      "mean_token_accuracy": 0.1778642162680626,
      "num_tokens": 47326728.0,
      "step": 25385
    },
    {
      "entropy": 6.158141994476319,
      "epoch": 2.242140586365242,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004495701168526368,
      "loss": 5.7565,
      "mean_token_accuracy": 0.17387374490499496,
      "num_tokens": 47336386.0,
      "step": 25390
    },
    {
      "entropy": 6.039257764816284,
      "epoch": 2.242582126457082,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00044955024726891585,
      "loss": 5.7484,
      "mean_token_accuracy": 0.16686405539512633,
      "num_tokens": 47345958.0,
      "step": 25395
    },
    {
      "entropy": 6.040530872344971,
      "epoch": 2.2430236665489227,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004495303742658549,
      "loss": 5.7631,
      "mean_token_accuracy": 0.17331896722316742,
      "num_tokens": 47355758.0,
      "step": 25400
    },
    {
      "entropy": 6.115247011184692,
      "epoch": 2.2434652066407628,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004495104978438431,
      "loss": 5.7557,
      "mean_token_accuracy": 0.17392015755176543,
      "num_tokens": 47365186.0,
      "step": 25405
    },
    {
      "entropy": 6.033462381362915,
      "epoch": 2.2439067467326033,
      "grad_norm": 1.78125,
      "learning_rate": 0.00044949061800327,
      "loss": 5.7273,
      "mean_token_accuracy": 0.17783308476209642,
      "num_tokens": 47374443.0,
      "step": 25410
    },
    {
      "entropy": 5.98583574295044,
      "epoch": 2.2443482868244438,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004494707347445248,
      "loss": 5.6454,
      "mean_token_accuracy": 0.1788348898291588,
      "num_tokens": 47383369.0,
      "step": 25415
    },
    {
      "entropy": 6.003458404541016,
      "epoch": 2.244789826916284,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004494508480679972,
      "loss": 5.7131,
      "mean_token_accuracy": 0.1813594028353691,
      "num_tokens": 47393104.0,
      "step": 25420
    },
    {
      "entropy": 5.977770233154297,
      "epoch": 2.2452313670081243,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00044943095797407676,
      "loss": 5.6967,
      "mean_token_accuracy": 0.18148788064718246,
      "num_tokens": 47401954.0,
      "step": 25425
    },
    {
      "entropy": 6.056436681747437,
      "epoch": 2.245672907099965,
      "grad_norm": 1.53125,
      "learning_rate": 0.000449411064463153,
      "loss": 5.6813,
      "mean_token_accuracy": 0.1783305823802948,
      "num_tokens": 47411463.0,
      "step": 25430
    },
    {
      "entropy": 6.022914743423462,
      "epoch": 2.246114447191805,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00044939116753561564,
      "loss": 5.6151,
      "mean_token_accuracy": 0.1769454747438431,
      "num_tokens": 47421405.0,
      "step": 25435
    },
    {
      "entropy": 5.994136714935303,
      "epoch": 2.2465559872836454,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004493712671918546,
      "loss": 5.7311,
      "mean_token_accuracy": 0.18067601174116135,
      "num_tokens": 47430268.0,
      "step": 25440
    },
    {
      "entropy": 6.053057861328125,
      "epoch": 2.246997527375486,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004493513634322595,
      "loss": 5.7571,
      "mean_token_accuracy": 0.165557399392128,
      "num_tokens": 47439571.0,
      "step": 25445
    },
    {
      "entropy": 6.0487090110778805,
      "epoch": 2.247439067467326,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004493314562572204,
      "loss": 5.7219,
      "mean_token_accuracy": 0.1736926406621933,
      "num_tokens": 47449085.0,
      "step": 25450
    },
    {
      "entropy": 5.989612674713134,
      "epoch": 2.2478806075591664,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004493115456671272,
      "loss": 5.6251,
      "mean_token_accuracy": 0.18210221827030182,
      "num_tokens": 47458740.0,
      "step": 25455
    },
    {
      "entropy": 6.044020318984986,
      "epoch": 2.248322147651007,
      "grad_norm": 1.625,
      "learning_rate": 0.00044929163166236997,
      "loss": 5.7655,
      "mean_token_accuracy": 0.17686106413602828,
      "num_tokens": 47467458.0,
      "step": 25460
    },
    {
      "entropy": 6.080802631378174,
      "epoch": 2.248763687742847,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004492717142433388,
      "loss": 5.7346,
      "mean_token_accuracy": 0.1746361032128334,
      "num_tokens": 47476845.0,
      "step": 25465
    },
    {
      "entropy": 5.99481520652771,
      "epoch": 2.2492052278346875,
      "grad_norm": 1.5,
      "learning_rate": 0.00044925179341042385,
      "loss": 5.711,
      "mean_token_accuracy": 0.18098821938037873,
      "num_tokens": 47485632.0,
      "step": 25470
    },
    {
      "entropy": 6.0276538848876955,
      "epoch": 2.249646767926528,
      "grad_norm": 2.125,
      "learning_rate": 0.00044923186916401536,
      "loss": 5.7073,
      "mean_token_accuracy": 0.17928725481033325,
      "num_tokens": 47494681.0,
      "step": 25475
    },
    {
      "entropy": 5.99576621055603,
      "epoch": 2.250088308018368,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004492119415045036,
      "loss": 5.7491,
      "mean_token_accuracy": 0.1786305859684944,
      "num_tokens": 47504482.0,
      "step": 25480
    },
    {
      "entropy": 6.042805004119873,
      "epoch": 2.2505298481102085,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004491920104322791,
      "loss": 5.7291,
      "mean_token_accuracy": 0.18202611804008484,
      "num_tokens": 47513601.0,
      "step": 25485
    },
    {
      "entropy": 6.070801067352295,
      "epoch": 2.2509713882020486,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004491720759477321,
      "loss": 5.6693,
      "mean_token_accuracy": 0.18063899725675583,
      "num_tokens": 47522811.0,
      "step": 25490
    },
    {
      "entropy": 6.004248571395874,
      "epoch": 2.251412928293889,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004491521380512533,
      "loss": 5.6924,
      "mean_token_accuracy": 0.17462234646081926,
      "num_tokens": 47532471.0,
      "step": 25495
    },
    {
      "entropy": 5.964924955368042,
      "epoch": 2.2518544683857296,
      "grad_norm": 1.5625,
      "learning_rate": 0.000449132196743233,
      "loss": 5.7049,
      "mean_token_accuracy": 0.1864021822810173,
      "num_tokens": 47541336.0,
      "step": 25500
    },
    {
      "entropy": 6.030114555358887,
      "epoch": 2.2522960084775696,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004491122520240621,
      "loss": 5.661,
      "mean_token_accuracy": 0.1809897854924202,
      "num_tokens": 47551446.0,
      "step": 25505
    },
    {
      "entropy": 6.033249473571777,
      "epoch": 2.25273754856941,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004490923038941312,
      "loss": 5.717,
      "mean_token_accuracy": 0.17551203370094298,
      "num_tokens": 47560596.0,
      "step": 25510
    },
    {
      "entropy": 6.088245630264282,
      "epoch": 2.2531790886612506,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000449072352353831,
      "loss": 5.8067,
      "mean_token_accuracy": 0.17544334679841994,
      "num_tokens": 47570612.0,
      "step": 25515
    },
    {
      "entropy": 6.045509195327758,
      "epoch": 2.2536206287530907,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004490523974035525,
      "loss": 5.7429,
      "mean_token_accuracy": 0.17220479398965835,
      "num_tokens": 47580234.0,
      "step": 25520
    },
    {
      "entropy": 6.093022346496582,
      "epoch": 2.254062168844931,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004490324390436866,
      "loss": 5.7431,
      "mean_token_accuracy": 0.17862291783094406,
      "num_tokens": 47589777.0,
      "step": 25525
    },
    {
      "entropy": 6.083073472976684,
      "epoch": 2.254503708936771,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000449012477274624,
      "loss": 5.6701,
      "mean_token_accuracy": 0.1803273841738701,
      "num_tokens": 47597951.0,
      "step": 25530
    },
    {
      "entropy": 5.974893522262573,
      "epoch": 2.2549452490286117,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00044899251209675606,
      "loss": 5.7625,
      "mean_token_accuracy": 0.17730558216571807,
      "num_tokens": 47606054.0,
      "step": 25535
    },
    {
      "entropy": 6.072769546508789,
      "epoch": 2.255386789120452,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004489725435104737,
      "loss": 5.7607,
      "mean_token_accuracy": 0.17374195009469987,
      "num_tokens": 47614704.0,
      "step": 25540
    },
    {
      "entropy": 6.091206693649292,
      "epoch": 2.2558283292122923,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004489525715161682,
      "loss": 5.7227,
      "mean_token_accuracy": 0.1768797978758812,
      "num_tokens": 47623189.0,
      "step": 25545
    },
    {
      "entropy": 6.033403778076172,
      "epoch": 2.2562698693041328,
      "grad_norm": 2.0,
      "learning_rate": 0.00044893259611423076,
      "loss": 5.7223,
      "mean_token_accuracy": 0.17251829206943511,
      "num_tokens": 47633831.0,
      "step": 25550
    },
    {
      "entropy": 6.090437507629394,
      "epoch": 2.2567114093959733,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004489126173050527,
      "loss": 5.6463,
      "mean_token_accuracy": 0.1779806688427925,
      "num_tokens": 47642155.0,
      "step": 25555
    },
    {
      "entropy": 6.14357180595398,
      "epoch": 2.2571529494878133,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004488926350890253,
      "loss": 5.8476,
      "mean_token_accuracy": 0.16462046504020691,
      "num_tokens": 47651042.0,
      "step": 25560
    },
    {
      "entropy": 6.099903726577759,
      "epoch": 2.257594489579654,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004488726494665402,
      "loss": 5.7952,
      "mean_token_accuracy": 0.1733235478401184,
      "num_tokens": 47661424.0,
      "step": 25565
    },
    {
      "entropy": 6.087075662612915,
      "epoch": 2.2580360296714943,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044885266043798866,
      "loss": 5.7603,
      "mean_token_accuracy": 0.17191237658262254,
      "num_tokens": 47671603.0,
      "step": 25570
    },
    {
      "entropy": 6.059715318679809,
      "epoch": 2.2584775697633344,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004488326680037624,
      "loss": 5.7235,
      "mean_token_accuracy": 0.17308178693056106,
      "num_tokens": 47679861.0,
      "step": 25575
    },
    {
      "entropy": 5.911651611328125,
      "epoch": 2.258919109855175,
      "grad_norm": 1.625,
      "learning_rate": 0.0004488126721642531,
      "loss": 5.6359,
      "mean_token_accuracy": 0.18296304643154143,
      "num_tokens": 47689702.0,
      "step": 25580
    },
    {
      "entropy": 5.9859757900238035,
      "epoch": 2.2593606499470154,
      "grad_norm": 2.28125,
      "learning_rate": 0.00044879267291985243,
      "loss": 5.704,
      "mean_token_accuracy": 0.1712062820792198,
      "num_tokens": 47699261.0,
      "step": 25585
    },
    {
      "entropy": 6.1207578659057615,
      "epoch": 2.2598021900388554,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044877267027095203,
      "loss": 5.7927,
      "mean_token_accuracy": 0.17183718532323838,
      "num_tokens": 47709079.0,
      "step": 25590
    },
    {
      "entropy": 6.012593126296997,
      "epoch": 2.260243730130696,
      "grad_norm": 1.625,
      "learning_rate": 0.0004487526642179439,
      "loss": 5.6589,
      "mean_token_accuracy": 0.17956391274929046,
      "num_tokens": 47718641.0,
      "step": 25595
    },
    {
      "entropy": 6.092954492568969,
      "epoch": 2.2606852702225364,
      "grad_norm": 2.15625,
      "learning_rate": 0.00044873265476121993,
      "loss": 5.8255,
      "mean_token_accuracy": 0.16886094957590103,
      "num_tokens": 47729156.0,
      "step": 25600
    },
    {
      "entropy": 6.030070781707764,
      "epoch": 2.2611268103143765,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044871264190117196,
      "loss": 5.7521,
      "mean_token_accuracy": 0.17237618565559387,
      "num_tokens": 47738424.0,
      "step": 25605
    },
    {
      "entropy": 6.078490447998047,
      "epoch": 2.261568350406217,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004486926256381922,
      "loss": 5.8536,
      "mean_token_accuracy": 0.16277689784765242,
      "num_tokens": 47749032.0,
      "step": 25610
    },
    {
      "entropy": 6.06685094833374,
      "epoch": 2.2620098904980575,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004486726059726727,
      "loss": 5.7509,
      "mean_token_accuracy": 0.1752706915140152,
      "num_tokens": 47758979.0,
      "step": 25615
    },
    {
      "entropy": 6.036699533462524,
      "epoch": 2.2624514305898975,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004486525829050056,
      "loss": 5.7124,
      "mean_token_accuracy": 0.17764170467853546,
      "num_tokens": 47768752.0,
      "step": 25620
    },
    {
      "entropy": 6.163960933685303,
      "epoch": 2.262892970681738,
      "grad_norm": 2.25,
      "learning_rate": 0.0004486325564355831,
      "loss": 5.7477,
      "mean_token_accuracy": 0.17114284932613372,
      "num_tokens": 47778175.0,
      "step": 25625
    },
    {
      "entropy": 6.004239416122436,
      "epoch": 2.263334510773578,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004486125265647977,
      "loss": 5.7218,
      "mean_token_accuracy": 0.18178464323282242,
      "num_tokens": 47787864.0,
      "step": 25630
    },
    {
      "entropy": 5.982497882843018,
      "epoch": 2.2637760508654186,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004485924932930416,
      "loss": 5.7681,
      "mean_token_accuracy": 0.17088536620140077,
      "num_tokens": 47796912.0,
      "step": 25635
    },
    {
      "entropy": 6.123118782043457,
      "epoch": 2.264217590957259,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004485724566207073,
      "loss": 5.8046,
      "mean_token_accuracy": 0.17066890150308608,
      "num_tokens": 47807452.0,
      "step": 25640
    },
    {
      "entropy": 6.132375812530517,
      "epoch": 2.264659131049099,
      "grad_norm": 1.4375,
      "learning_rate": 0.00044855241654818723,
      "loss": 5.8297,
      "mean_token_accuracy": 0.17049993127584456,
      "num_tokens": 47817575.0,
      "step": 25645
    },
    {
      "entropy": 6.108438777923584,
      "epoch": 2.2651006711409396,
      "grad_norm": 3.375,
      "learning_rate": 0.0004485323730758741,
      "loss": 5.7217,
      "mean_token_accuracy": 0.18726795464754104,
      "num_tokens": 47826177.0,
      "step": 25650
    },
    {
      "entropy": 5.972574853897095,
      "epoch": 2.26554221123278,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004485123262041604,
      "loss": 5.6678,
      "mean_token_accuracy": 0.175558240711689,
      "num_tokens": 47835622.0,
      "step": 25655
    },
    {
      "entropy": 6.040759515762329,
      "epoch": 2.26598375132462,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004484922759334389,
      "loss": 5.7138,
      "mean_token_accuracy": 0.17904701083898544,
      "num_tokens": 47845552.0,
      "step": 25660
    },
    {
      "entropy": 6.0338891506195065,
      "epoch": 2.2664252914164607,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004484722222641024,
      "loss": 5.6932,
      "mean_token_accuracy": 0.17729313522577286,
      "num_tokens": 47853820.0,
      "step": 25665
    },
    {
      "entropy": 5.955590391159058,
      "epoch": 2.2668668315083007,
      "grad_norm": 4.125,
      "learning_rate": 0.0004484521651965438,
      "loss": 5.6644,
      "mean_token_accuracy": 0.18444336652755738,
      "num_tokens": 47863416.0,
      "step": 25670
    },
    {
      "entropy": 6.099171543121338,
      "epoch": 2.267308371600141,
      "grad_norm": 1.734375,
      "learning_rate": 0.00044843210473115584,
      "loss": 5.74,
      "mean_token_accuracy": 0.17818276584148407,
      "num_tokens": 47871407.0,
      "step": 25675
    },
    {
      "entropy": 6.123391914367676,
      "epoch": 2.2677499116919817,
      "grad_norm": 2.0,
      "learning_rate": 0.0004484120408683316,
      "loss": 5.7105,
      "mean_token_accuracy": 0.17047855854034424,
      "num_tokens": 47880742.0,
      "step": 25680
    },
    {
      "entropy": 6.013138389587402,
      "epoch": 2.2681914517838218,
      "grad_norm": 1.703125,
      "learning_rate": 0.00044839197360846416,
      "loss": 5.7635,
      "mean_token_accuracy": 0.16930958926677703,
      "num_tokens": 47889573.0,
      "step": 25685
    },
    {
      "entropy": 6.0612798690795895,
      "epoch": 2.2686329918756623,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00044837190295194654,
      "loss": 5.7138,
      "mean_token_accuracy": 0.17488819658756255,
      "num_tokens": 47898855.0,
      "step": 25690
    },
    {
      "entropy": 6.09538779258728,
      "epoch": 2.2690745319675028,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044835182889917195,
      "loss": 5.7402,
      "mean_token_accuracy": 0.1772387742996216,
      "num_tokens": 47908981.0,
      "step": 25695
    },
    {
      "entropy": 5.9487221240997314,
      "epoch": 2.269516072059343,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004483317514505336,
      "loss": 5.7188,
      "mean_token_accuracy": 0.18079887479543685,
      "num_tokens": 47918313.0,
      "step": 25700
    },
    {
      "entropy": 5.980592489242554,
      "epoch": 2.2699576121511833,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004483116706064248,
      "loss": 5.6817,
      "mean_token_accuracy": 0.1793721690773964,
      "num_tokens": 47927838.0,
      "step": 25705
    },
    {
      "entropy": 6.071624994277954,
      "epoch": 2.270399152243024,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004482915863672389,
      "loss": 5.7074,
      "mean_token_accuracy": 0.17920632362365724,
      "num_tokens": 47937774.0,
      "step": 25710
    },
    {
      "entropy": 6.109913492202759,
      "epoch": 2.270840692334864,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004482714987333695,
      "loss": 5.6833,
      "mean_token_accuracy": 0.182609623670578,
      "num_tokens": 47947091.0,
      "step": 25715
    },
    {
      "entropy": 5.99449462890625,
      "epoch": 2.2712822324267044,
      "grad_norm": 1.578125,
      "learning_rate": 0.00044825140770520997,
      "loss": 5.6299,
      "mean_token_accuracy": 0.18508990108966827,
      "num_tokens": 47955715.0,
      "step": 25720
    },
    {
      "entropy": 5.916472959518432,
      "epoch": 2.271723772518545,
      "grad_norm": 1.78125,
      "learning_rate": 0.00044823131328315387,
      "loss": 5.6057,
      "mean_token_accuracy": 0.19356101453304292,
      "num_tokens": 47965991.0,
      "step": 25725
    },
    {
      "entropy": 6.138766813278198,
      "epoch": 2.272165312610385,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004482112154675948,
      "loss": 5.7591,
      "mean_token_accuracy": 0.1790761500597,
      "num_tokens": 47975610.0,
      "step": 25730
    },
    {
      "entropy": 6.017828130722046,
      "epoch": 2.2726068527022254,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00044819111425892667,
      "loss": 5.6964,
      "mean_token_accuracy": 0.184140445291996,
      "num_tokens": 47984977.0,
      "step": 25735
    },
    {
      "entropy": 6.008146047592163,
      "epoch": 2.273048392794066,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00044817100965754297,
      "loss": 5.6528,
      "mean_token_accuracy": 0.17787908315658568,
      "num_tokens": 47994205.0,
      "step": 25740
    },
    {
      "entropy": 6.0235217094421385,
      "epoch": 2.273489932885906,
      "grad_norm": 1.640625,
      "learning_rate": 0.00044815090166383775,
      "loss": 5.7344,
      "mean_token_accuracy": 0.17910091876983641,
      "num_tokens": 48003715.0,
      "step": 25745
    },
    {
      "entropy": 6.072096443176269,
      "epoch": 2.2739314729777464,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004481307902782048,
      "loss": 5.8187,
      "mean_token_accuracy": 0.1711183562874794,
      "num_tokens": 48013200.0,
      "step": 25750
    },
    {
      "entropy": 6.081849002838135,
      "epoch": 2.274373013069587,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004481106755010382,
      "loss": 5.73,
      "mean_token_accuracy": 0.17965518087148666,
      "num_tokens": 48022366.0,
      "step": 25755
    },
    {
      "entropy": 6.096280479431153,
      "epoch": 2.274814553161427,
      "grad_norm": 2.390625,
      "learning_rate": 0.00044809055733273183,
      "loss": 5.7571,
      "mean_token_accuracy": 0.17860091477632523,
      "num_tokens": 48031968.0,
      "step": 25760
    },
    {
      "entropy": 6.068123769760132,
      "epoch": 2.2752560932532675,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00044807043577367996,
      "loss": 5.6898,
      "mean_token_accuracy": 0.17814256399869918,
      "num_tokens": 48041069.0,
      "step": 25765
    },
    {
      "entropy": 5.902412557601929,
      "epoch": 2.2756976333451076,
      "grad_norm": 2.171875,
      "learning_rate": 0.00044805031082427666,
      "loss": 5.6178,
      "mean_token_accuracy": 0.17989975214004517,
      "num_tokens": 48050550.0,
      "step": 25770
    },
    {
      "entropy": 6.050845289230347,
      "epoch": 2.276139173436948,
      "grad_norm": 1.5,
      "learning_rate": 0.00044803018248491615,
      "loss": 5.6933,
      "mean_token_accuracy": 0.18261670023202897,
      "num_tokens": 48059971.0,
      "step": 25775
    },
    {
      "entropy": 6.074761533737183,
      "epoch": 2.2765807135287885,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004480100507559928,
      "loss": 5.5447,
      "mean_token_accuracy": 0.1863307312130928,
      "num_tokens": 48068566.0,
      "step": 25780
    },
    {
      "entropy": 5.909382057189942,
      "epoch": 2.2770222536206286,
      "grad_norm": 3.296875,
      "learning_rate": 0.000447989915637901,
      "loss": 5.631,
      "mean_token_accuracy": 0.1862158939242363,
      "num_tokens": 48078352.0,
      "step": 25785
    },
    {
      "entropy": 5.944258499145508,
      "epoch": 2.277463793712469,
      "grad_norm": 2.0,
      "learning_rate": 0.0004479697771310351,
      "loss": 5.7152,
      "mean_token_accuracy": 0.178833244740963,
      "num_tokens": 48087089.0,
      "step": 25790
    },
    {
      "entropy": 6.058592653274536,
      "epoch": 2.2779053338043096,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044794963523578965,
      "loss": 5.7683,
      "mean_token_accuracy": 0.17567681670188903,
      "num_tokens": 48096762.0,
      "step": 25795
    },
    {
      "entropy": 6.064421463012695,
      "epoch": 2.2783468738961496,
      "grad_norm": 1.796875,
      "learning_rate": 0.00044792948995255917,
      "loss": 5.7067,
      "mean_token_accuracy": 0.18671585619449615,
      "num_tokens": 48105994.0,
      "step": 25800
    },
    {
      "entropy": 5.9654576778411865,
      "epoch": 2.27878841398799,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004479093412817383,
      "loss": 5.6707,
      "mean_token_accuracy": 0.1860273152589798,
      "num_tokens": 48115148.0,
      "step": 25805
    },
    {
      "entropy": 5.964911603927613,
      "epoch": 2.27922995407983,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00044788918922372196,
      "loss": 5.76,
      "mean_token_accuracy": 0.17485454678535461,
      "num_tokens": 48124696.0,
      "step": 25810
    },
    {
      "entropy": 6.0326180934906,
      "epoch": 2.2796714941716707,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004478690337789046,
      "loss": 5.6676,
      "mean_token_accuracy": 0.18102737367153168,
      "num_tokens": 48134165.0,
      "step": 25815
    },
    {
      "entropy": 6.133547353744507,
      "epoch": 2.280113034263511,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044784887494768113,
      "loss": 5.7094,
      "mean_token_accuracy": 0.1731340691447258,
      "num_tokens": 48144080.0,
      "step": 25820
    },
    {
      "entropy": 5.99552001953125,
      "epoch": 2.2805545743553512,
      "grad_norm": 2.390625,
      "learning_rate": 0.00044782871273044654,
      "loss": 5.7505,
      "mean_token_accuracy": 0.18244427293539048,
      "num_tokens": 48156161.0,
      "step": 25825
    },
    {
      "entropy": 6.0382905960083,
      "epoch": 2.2809961144471917,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004478085471275958,
      "loss": 5.7342,
      "mean_token_accuracy": 0.1706133618950844,
      "num_tokens": 48165317.0,
      "step": 25830
    },
    {
      "entropy": 6.063380336761474,
      "epoch": 2.2814376545390322,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004477883781395239,
      "loss": 5.6841,
      "mean_token_accuracy": 0.17859037220478058,
      "num_tokens": 48173639.0,
      "step": 25835
    },
    {
      "entropy": 6.020235681533814,
      "epoch": 2.2818791946308723,
      "grad_norm": 1.84375,
      "learning_rate": 0.00044776820576662593,
      "loss": 5.7264,
      "mean_token_accuracy": 0.17918055802583693,
      "num_tokens": 48183596.0,
      "step": 25840
    },
    {
      "entropy": 6.118922090530395,
      "epoch": 2.282320734722713,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004477480300092971,
      "loss": 5.7931,
      "mean_token_accuracy": 0.17734775096178054,
      "num_tokens": 48194269.0,
      "step": 25845
    },
    {
      "entropy": 6.0384268283844,
      "epoch": 2.2827622748145533,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00044772785086793257,
      "loss": 5.7016,
      "mean_token_accuracy": 0.18077051788568496,
      "num_tokens": 48203180.0,
      "step": 25850
    },
    {
      "entropy": 5.98937201499939,
      "epoch": 2.2832038149063933,
      "grad_norm": 2.1875,
      "learning_rate": 0.00044770766834292765,
      "loss": 5.7283,
      "mean_token_accuracy": 0.1785341680049896,
      "num_tokens": 48212768.0,
      "step": 25855
    },
    {
      "entropy": 6.041850852966308,
      "epoch": 2.283645354998234,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004476874824346777,
      "loss": 5.7496,
      "mean_token_accuracy": 0.17557932436466217,
      "num_tokens": 48221620.0,
      "step": 25860
    },
    {
      "entropy": 6.11845850944519,
      "epoch": 2.2840868950900743,
      "grad_norm": 1.84375,
      "learning_rate": 0.00044766729314357827,
      "loss": 5.7715,
      "mean_token_accuracy": 0.1685803398489952,
      "num_tokens": 48231379.0,
      "step": 25865
    },
    {
      "entropy": 6.09968934059143,
      "epoch": 2.2845284351819144,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004476471004700248,
      "loss": 5.7885,
      "mean_token_accuracy": 0.17317433506250382,
      "num_tokens": 48241827.0,
      "step": 25870
    },
    {
      "entropy": 6.022066450119018,
      "epoch": 2.284969975273755,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004476269044144127,
      "loss": 5.7131,
      "mean_token_accuracy": 0.1716233089566231,
      "num_tokens": 48250747.0,
      "step": 25875
    },
    {
      "entropy": 6.091340255737305,
      "epoch": 2.2854115153655954,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004476067049771378,
      "loss": 5.7138,
      "mean_token_accuracy": 0.17380787134170533,
      "num_tokens": 48260909.0,
      "step": 25880
    },
    {
      "entropy": 6.1534827709197994,
      "epoch": 2.2858530554574354,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004475865021585957,
      "loss": 5.7725,
      "mean_token_accuracy": 0.17312343269586564,
      "num_tokens": 48270974.0,
      "step": 25885
    },
    {
      "entropy": 5.966058778762817,
      "epoch": 2.286294595549276,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004475662959591822,
      "loss": 5.6297,
      "mean_token_accuracy": 0.17795832455158234,
      "num_tokens": 48279466.0,
      "step": 25890
    },
    {
      "entropy": 6.003415536880493,
      "epoch": 2.2867361356411164,
      "grad_norm": 2.390625,
      "learning_rate": 0.000447546086379293,
      "loss": 5.7701,
      "mean_token_accuracy": 0.1756591796875,
      "num_tokens": 48289830.0,
      "step": 25895
    },
    {
      "entropy": 6.040771436691284,
      "epoch": 2.2871776757329565,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004475258734193242,
      "loss": 5.6993,
      "mean_token_accuracy": 0.17698077708482743,
      "num_tokens": 48299493.0,
      "step": 25900
    },
    {
      "entropy": 6.033964061737061,
      "epoch": 2.287619215824797,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004475056570796717,
      "loss": 5.7192,
      "mean_token_accuracy": 0.18498658239841462,
      "num_tokens": 48308366.0,
      "step": 25905
    },
    {
      "entropy": 5.970059871673584,
      "epoch": 2.288060755916637,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004474854373607314,
      "loss": 5.7086,
      "mean_token_accuracy": 0.18232780694961548,
      "num_tokens": 48317760.0,
      "step": 25910
    },
    {
      "entropy": 6.019486427307129,
      "epoch": 2.2885022960084775,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004474652142628995,
      "loss": 5.6958,
      "mean_token_accuracy": 0.17325934618711472,
      "num_tokens": 48325847.0,
      "step": 25915
    },
    {
      "entropy": 6.1122478485107425,
      "epoch": 2.288943836100318,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004474449877865722,
      "loss": 5.7474,
      "mean_token_accuracy": 0.1746838629245758,
      "num_tokens": 48334938.0,
      "step": 25920
    },
    {
      "entropy": 6.0091766834259035,
      "epoch": 2.289385376192158,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004474247579321456,
      "loss": 5.7106,
      "mean_token_accuracy": 0.17145783752202987,
      "num_tokens": 48344414.0,
      "step": 25925
    },
    {
      "entropy": 6.04113998413086,
      "epoch": 2.2898269162839986,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000447404524700016,
      "loss": 5.6785,
      "mean_token_accuracy": 0.17755413204431533,
      "num_tokens": 48353711.0,
      "step": 25930
    },
    {
      "entropy": 5.981535005569458,
      "epoch": 2.290268456375839,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004473842880905799,
      "loss": 5.6291,
      "mean_token_accuracy": 0.18702974319458007,
      "num_tokens": 48362907.0,
      "step": 25935
    },
    {
      "entropy": 6.090908336639404,
      "epoch": 2.290709996467679,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044736404810423366,
      "loss": 5.7021,
      "mean_token_accuracy": 0.18581179529428482,
      "num_tokens": 48372450.0,
      "step": 25940
    },
    {
      "entropy": 6.004093933105469,
      "epoch": 2.2911515365595196,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004473438047413737,
      "loss": 5.6336,
      "mean_token_accuracy": 0.18622260838747023,
      "num_tokens": 48381521.0,
      "step": 25945
    },
    {
      "entropy": 6.003949403762817,
      "epoch": 2.2915930766513597,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004473235580023966,
      "loss": 5.6879,
      "mean_token_accuracy": 0.18274352252483367,
      "num_tokens": 48390635.0,
      "step": 25950
    },
    {
      "entropy": 6.06712965965271,
      "epoch": 2.2920346167432,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044730330788769903,
      "loss": 5.7599,
      "mean_token_accuracy": 0.16867355704307557,
      "num_tokens": 48400199.0,
      "step": 25955
    },
    {
      "entropy": 6.010861587524414,
      "epoch": 2.2924761568350407,
      "grad_norm": 1.703125,
      "learning_rate": 0.00044728305439767764,
      "loss": 5.6932,
      "mean_token_accuracy": 0.1807205617427826,
      "num_tokens": 48409266.0,
      "step": 25960
    },
    {
      "entropy": 6.019666910171509,
      "epoch": 2.2929176969268807,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004472627975327293,
      "loss": 5.7334,
      "mean_token_accuracy": 0.176639886200428,
      "num_tokens": 48419195.0,
      "step": 25965
    },
    {
      "entropy": 6.100021886825561,
      "epoch": 2.2933592370187212,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004472425372932506,
      "loss": 5.8185,
      "mean_token_accuracy": 0.17281925976276397,
      "num_tokens": 48430637.0,
      "step": 25970
    },
    {
      "entropy": 5.971015357971192,
      "epoch": 2.2938007771105617,
      "grad_norm": 1.875,
      "learning_rate": 0.0004472222736796386,
      "loss": 5.576,
      "mean_token_accuracy": 0.189948470890522,
      "num_tokens": 48439929.0,
      "step": 25975
    },
    {
      "entropy": 6.073219871520996,
      "epoch": 2.294242317202402,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004472020066922902,
      "loss": 5.8297,
      "mean_token_accuracy": 0.16815374195575714,
      "num_tokens": 48449398.0,
      "step": 25980
    },
    {
      "entropy": 6.0374846935272215,
      "epoch": 2.2946838572942423,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004471817363316024,
      "loss": 5.7166,
      "mean_token_accuracy": 0.17762890905141832,
      "num_tokens": 48458843.0,
      "step": 25985
    },
    {
      "entropy": 6.0636509418487545,
      "epoch": 2.295125397386083,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004471614625979723,
      "loss": 5.6583,
      "mean_token_accuracy": 0.1794382408261299,
      "num_tokens": 48468111.0,
      "step": 25990
    },
    {
      "entropy": 6.039353609085083,
      "epoch": 2.295566937477923,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004471411854917972,
      "loss": 5.7045,
      "mean_token_accuracy": 0.18084288984537125,
      "num_tokens": 48477501.0,
      "step": 25995
    },
    {
      "entropy": 6.134314775466919,
      "epoch": 2.2960084775697633,
      "grad_norm": 1.578125,
      "learning_rate": 0.000447120905013474,
      "loss": 5.8418,
      "mean_token_accuracy": 0.1654657706618309,
      "num_tokens": 48487576.0,
      "step": 26000
    },
    {
      "entropy": 5.967932224273682,
      "epoch": 2.296450017661604,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004471006211634002,
      "loss": 5.6276,
      "mean_token_accuracy": 0.19373249113559723,
      "num_tokens": 48496230.0,
      "step": 26005
    },
    {
      "entropy": 6.072743034362793,
      "epoch": 2.296891557753444,
      "grad_norm": 2.171875,
      "learning_rate": 0.00044708033394197323,
      "loss": 5.7246,
      "mean_token_accuracy": 0.17704427391290664,
      "num_tokens": 48505871.0,
      "step": 26010
    },
    {
      "entropy": 6.0695713520050045,
      "epoch": 2.2973330978452844,
      "grad_norm": 1.8125,
      "learning_rate": 0.00044706004334959024,
      "loss": 5.746,
      "mean_token_accuracy": 0.17189155220985414,
      "num_tokens": 48515298.0,
      "step": 26015
    },
    {
      "entropy": 5.990046405792237,
      "epoch": 2.297774637937125,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004470397493866489,
      "loss": 5.7375,
      "mean_token_accuracy": 0.17514027953147887,
      "num_tokens": 48524573.0,
      "step": 26020
    },
    {
      "entropy": 6.072584962844848,
      "epoch": 2.298216178028965,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004470194520535467,
      "loss": 5.7914,
      "mean_token_accuracy": 0.17177725881338118,
      "num_tokens": 48533244.0,
      "step": 26025
    },
    {
      "entropy": 6.127796792984009,
      "epoch": 2.2986577181208054,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004469991513506814,
      "loss": 5.6652,
      "mean_token_accuracy": 0.1795307070016861,
      "num_tokens": 48542464.0,
      "step": 26030
    },
    {
      "entropy": 6.002179670333862,
      "epoch": 2.299099258212646,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004469788472784504,
      "loss": 5.742,
      "mean_token_accuracy": 0.18320469856262206,
      "num_tokens": 48551947.0,
      "step": 26035
    },
    {
      "entropy": 6.014916896820068,
      "epoch": 2.299540798304486,
      "grad_norm": 1.8125,
      "learning_rate": 0.00044695853983725163,
      "loss": 5.6722,
      "mean_token_accuracy": 0.18374090194702147,
      "num_tokens": 48561287.0,
      "step": 26040
    },
    {
      "entropy": 6.014718866348266,
      "epoch": 2.2999823383963265,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004469382290274829,
      "loss": 5.7918,
      "mean_token_accuracy": 0.17048033326864243,
      "num_tokens": 48570800.0,
      "step": 26045
    },
    {
      "entropy": 6.022571039199829,
      "epoch": 2.3004238784881665,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044691791484954195,
      "loss": 5.6368,
      "mean_token_accuracy": 0.18488933593034745,
      "num_tokens": 48579705.0,
      "step": 26050
    },
    {
      "entropy": 6.060318326950073,
      "epoch": 2.300865418580007,
      "grad_norm": 1.46875,
      "learning_rate": 0.000446897597303827,
      "loss": 5.7517,
      "mean_token_accuracy": 0.18043670505285264,
      "num_tokens": 48588793.0,
      "step": 26055
    },
    {
      "entropy": 6.04321174621582,
      "epoch": 2.3013069586718475,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004468772763907357,
      "loss": 5.6279,
      "mean_token_accuracy": 0.18335917294025422,
      "num_tokens": 48598361.0,
      "step": 26060
    },
    {
      "entropy": 6.0546125888824465,
      "epoch": 2.3017484987636876,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004468569521106664,
      "loss": 5.7179,
      "mean_token_accuracy": 0.17276445031166077,
      "num_tokens": 48607703.0,
      "step": 26065
    },
    {
      "entropy": 5.948693370819091,
      "epoch": 2.302190038855528,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004468366244640171,
      "loss": 5.6682,
      "mean_token_accuracy": 0.18436662554740907,
      "num_tokens": 48616345.0,
      "step": 26070
    },
    {
      "entropy": 6.057308101654053,
      "epoch": 2.3026315789473686,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044681629345118615,
      "loss": 5.7378,
      "mean_token_accuracy": 0.17485498487949372,
      "num_tokens": 48624907.0,
      "step": 26075
    },
    {
      "entropy": 6.032192087173462,
      "epoch": 2.3030731190392086,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004467959590725716,
      "loss": 5.6945,
      "mean_token_accuracy": 0.183023601770401,
      "num_tokens": 48634766.0,
      "step": 26080
    },
    {
      "entropy": 6.05084114074707,
      "epoch": 2.303514659131049,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044677562132857204,
      "loss": 5.729,
      "mean_token_accuracy": 0.17914541512727739,
      "num_tokens": 48644321.0,
      "step": 26085
    },
    {
      "entropy": 6.0157674789428714,
      "epoch": 2.303956199222889,
      "grad_norm": 1.546875,
      "learning_rate": 0.00044675528021958567,
      "loss": 5.7314,
      "mean_token_accuracy": 0.17830923348665237,
      "num_tokens": 48653507.0,
      "step": 26090
    },
    {
      "entropy": 5.9760575771331785,
      "epoch": 2.3043977393147297,
      "grad_norm": 1.453125,
      "learning_rate": 0.000446734935746011,
      "loss": 5.7009,
      "mean_token_accuracy": 0.18286084830760957,
      "num_tokens": 48662186.0,
      "step": 26095
    },
    {
      "entropy": 6.001360082626343,
      "epoch": 2.30483927940657,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004467145879082467,
      "loss": 5.7011,
      "mean_token_accuracy": 0.17985236942768096,
      "num_tokens": 48671605.0,
      "step": 26100
    },
    {
      "entropy": 6.014617300033569,
      "epoch": 2.3052808194984102,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004466942367066912,
      "loss": 5.6355,
      "mean_token_accuracy": 0.18278767317533492,
      "num_tokens": 48679873.0,
      "step": 26105
    },
    {
      "entropy": 6.034718322753906,
      "epoch": 2.3057223595902507,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00044667388214174336,
      "loss": 5.6063,
      "mean_token_accuracy": 0.18254622370004653,
      "num_tokens": 48688678.0,
      "step": 26110
    },
    {
      "entropy": 5.9953807353973385,
      "epoch": 2.3061638996820912,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004466535242138018,
      "loss": 5.7075,
      "mean_token_accuracy": 0.1776547312736511,
      "num_tokens": 48697794.0,
      "step": 26115
    },
    {
      "entropy": 6.039161014556885,
      "epoch": 2.3066054397739313,
      "grad_norm": 1.921875,
      "learning_rate": 0.00044663316292326526,
      "loss": 5.8377,
      "mean_token_accuracy": 0.17383372634649277,
      "num_tokens": 48706859.0,
      "step": 26120
    },
    {
      "entropy": 6.061024713516235,
      "epoch": 2.307046979865772,
      "grad_norm": 2.15625,
      "learning_rate": 0.00044661279827053267,
      "loss": 5.6693,
      "mean_token_accuracy": 0.1792992800474167,
      "num_tokens": 48716627.0,
      "step": 26125
    },
    {
      "entropy": 6.044272375106812,
      "epoch": 2.3074885199576123,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044659243025600297,
      "loss": 5.7205,
      "mean_token_accuracy": 0.17327332645654678,
      "num_tokens": 48726514.0,
      "step": 26130
    },
    {
      "entropy": 6.0350518226623535,
      "epoch": 2.3079300600494523,
      "grad_norm": 1.53125,
      "learning_rate": 0.00044657205888007525,
      "loss": 5.6878,
      "mean_token_accuracy": 0.1819120779633522,
      "num_tokens": 48735546.0,
      "step": 26135
    },
    {
      "entropy": 5.950460386276245,
      "epoch": 2.308371600141293,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004465516841431483,
      "loss": 5.6085,
      "mean_token_accuracy": 0.17919116020202636,
      "num_tokens": 48744719.0,
      "step": 26140
    },
    {
      "entropy": 6.023001623153687,
      "epoch": 2.3088131402331333,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004465313060456216,
      "loss": 5.7126,
      "mean_token_accuracy": 0.17930537313222886,
      "num_tokens": 48754201.0,
      "step": 26145
    },
    {
      "entropy": 6.062580013275147,
      "epoch": 2.3092546803249734,
      "grad_norm": 1.5,
      "learning_rate": 0.00044651092458789405,
      "loss": 5.7323,
      "mean_token_accuracy": 0.1711074635386467,
      "num_tokens": 48763552.0,
      "step": 26150
    },
    {
      "entropy": 6.041425561904907,
      "epoch": 2.309696220416814,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004464905397703651,
      "loss": 5.703,
      "mean_token_accuracy": 0.1778918743133545,
      "num_tokens": 48772494.0,
      "step": 26155
    },
    {
      "entropy": 6.044746589660645,
      "epoch": 2.3101377605086544,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044647015159343396,
      "loss": 5.7272,
      "mean_token_accuracy": 0.1709308758378029,
      "num_tokens": 48782121.0,
      "step": 26160
    },
    {
      "entropy": 5.985767984390259,
      "epoch": 2.3105793006004944,
      "grad_norm": 2.234375,
      "learning_rate": 0.00044644976005750003,
      "loss": 5.6207,
      "mean_token_accuracy": 0.18565240055322646,
      "num_tokens": 48792040.0,
      "step": 26165
    },
    {
      "entropy": 5.99914026260376,
      "epoch": 2.311020840692335,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004464293651629629,
      "loss": 5.6972,
      "mean_token_accuracy": 0.17987836599349977,
      "num_tokens": 48801429.0,
      "step": 26170
    },
    {
      "entropy": 5.997629308700562,
      "epoch": 2.3114623807841754,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00044640896691022195,
      "loss": 5.7104,
      "mean_token_accuracy": 0.18102218061685563,
      "num_tokens": 48811185.0,
      "step": 26175
    },
    {
      "entropy": 6.075421571731567,
      "epoch": 2.3119039208760155,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004463885652996768,
      "loss": 5.7694,
      "mean_token_accuracy": 0.17211493700742722,
      "num_tokens": 48820691.0,
      "step": 26180
    },
    {
      "entropy": 6.135617017745972,
      "epoch": 2.312345460967856,
      "grad_norm": 1.8125,
      "learning_rate": 0.00044636816033172716,
      "loss": 5.7807,
      "mean_token_accuracy": 0.1711033046245575,
      "num_tokens": 48829960.0,
      "step": 26185
    },
    {
      "entropy": 6.058924961090088,
      "epoch": 2.312787001059696,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004463477520067728,
      "loss": 5.6724,
      "mean_token_accuracy": 0.1721493661403656,
      "num_tokens": 48839107.0,
      "step": 26190
    },
    {
      "entropy": 6.0629807472229,
      "epoch": 2.3132285411515365,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004463273403252133,
      "loss": 5.778,
      "mean_token_accuracy": 0.1760726734995842,
      "num_tokens": 48848430.0,
      "step": 26195
    },
    {
      "entropy": 6.001303768157959,
      "epoch": 2.313670081243377,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004463069252874487,
      "loss": 5.6534,
      "mean_token_accuracy": 0.1880018025636673,
      "num_tokens": 48857869.0,
      "step": 26200
    },
    {
      "entropy": 6.097454071044922,
      "epoch": 2.314111621335217,
      "grad_norm": 1.5,
      "learning_rate": 0.00044628650689387887,
      "loss": 5.7537,
      "mean_token_accuracy": 0.1769058182835579,
      "num_tokens": 48867730.0,
      "step": 26205
    },
    {
      "entropy": 6.111022996902466,
      "epoch": 2.3145531614270576,
      "grad_norm": 2.53125,
      "learning_rate": 0.00044626608514490384,
      "loss": 5.7697,
      "mean_token_accuracy": 0.17922558337450029,
      "num_tokens": 48876615.0,
      "step": 26210
    },
    {
      "entropy": 6.02492995262146,
      "epoch": 2.314994701518898,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004462456600409235,
      "loss": 5.7377,
      "mean_token_accuracy": 0.1795410171151161,
      "num_tokens": 48885732.0,
      "step": 26215
    },
    {
      "entropy": 6.070626878738404,
      "epoch": 2.315436241610738,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004462252315823381,
      "loss": 5.6656,
      "mean_token_accuracy": 0.17950979620218277,
      "num_tokens": 48894372.0,
      "step": 26220
    },
    {
      "entropy": 6.047794008255005,
      "epoch": 2.3158777817025786,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004462047997695478,
      "loss": 5.8091,
      "mean_token_accuracy": 0.1736207753419876,
      "num_tokens": 48904321.0,
      "step": 26225
    },
    {
      "entropy": 6.029712104797364,
      "epoch": 2.3163193217944187,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044618436460295294,
      "loss": 5.7494,
      "mean_token_accuracy": 0.17450084835290908,
      "num_tokens": 48914215.0,
      "step": 26230
    },
    {
      "entropy": 6.0652186393737795,
      "epoch": 2.316760861886259,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004461639260829535,
      "loss": 5.6986,
      "mean_token_accuracy": 0.16969432830810546,
      "num_tokens": 48923383.0,
      "step": 26235
    },
    {
      "entropy": 6.089188289642334,
      "epoch": 2.3172024019780997,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004461434842099503,
      "loss": 5.731,
      "mean_token_accuracy": 0.17582143396139144,
      "num_tokens": 48933560.0,
      "step": 26240
    },
    {
      "entropy": 5.942011547088623,
      "epoch": 2.3176439420699397,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004461230389843435,
      "loss": 5.641,
      "mean_token_accuracy": 0.1857260897755623,
      "num_tokens": 48943111.0,
      "step": 26245
    },
    {
      "entropy": 6.052125406265259,
      "epoch": 2.3180854821617802,
      "grad_norm": 1.875,
      "learning_rate": 0.0004461025904065338,
      "loss": 5.7021,
      "mean_token_accuracy": 0.18272095769643784,
      "num_tokens": 48952470.0,
      "step": 26250
    },
    {
      "entropy": 6.160805177688599,
      "epoch": 2.3185270222536207,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044608213847692154,
      "loss": 5.766,
      "mean_token_accuracy": 0.17007261663675308,
      "num_tokens": 48961273.0,
      "step": 26255
    },
    {
      "entropy": 6.171191120147705,
      "epoch": 2.318968562345461,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004460616831959075,
      "loss": 5.7909,
      "mean_token_accuracy": 0.17495279312133788,
      "num_tokens": 48971078.0,
      "step": 26260
    },
    {
      "entropy": 6.020609331130982,
      "epoch": 2.3194101024373013,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004460412245638924,
      "loss": 5.7593,
      "mean_token_accuracy": 0.1779707983136177,
      "num_tokens": 48980711.0,
      "step": 26265
    },
    {
      "entropy": 5.983940839767456,
      "epoch": 2.3198516425291418,
      "grad_norm": 1.8125,
      "learning_rate": 0.000446020762581277,
      "loss": 5.6742,
      "mean_token_accuracy": 0.17693228870630265,
      "num_tokens": 48990195.0,
      "step": 26270
    },
    {
      "entropy": 5.890049934387207,
      "epoch": 2.320293182620982,
      "grad_norm": 2.796875,
      "learning_rate": 0.00044600029724846206,
      "loss": 5.5563,
      "mean_token_accuracy": 0.19065133184194566,
      "num_tokens": 49000271.0,
      "step": 26275
    },
    {
      "entropy": 6.001823425292969,
      "epoch": 2.3207347227128223,
      "grad_norm": 1.875,
      "learning_rate": 0.0004459798285658487,
      "loss": 5.6735,
      "mean_token_accuracy": 0.17491325289011,
      "num_tokens": 49009624.0,
      "step": 26280
    },
    {
      "entropy": 6.005040550231934,
      "epoch": 2.321176262804663,
      "grad_norm": 1.640625,
      "learning_rate": 0.00044595935653383756,
      "loss": 5.6673,
      "mean_token_accuracy": 0.1774981886148453,
      "num_tokens": 49018845.0,
      "step": 26285
    },
    {
      "entropy": 6.008721542358399,
      "epoch": 2.321617802896503,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044593888115282995,
      "loss": 5.6258,
      "mean_token_accuracy": 0.19009407013654708,
      "num_tokens": 49028352.0,
      "step": 26290
    },
    {
      "entropy": 6.003528833389282,
      "epoch": 2.3220593429883434,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044591840242322686,
      "loss": 5.7425,
      "mean_token_accuracy": 0.17763376235961914,
      "num_tokens": 49038180.0,
      "step": 26295
    },
    {
      "entropy": 6.003032207489014,
      "epoch": 2.322500883080184,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004458979203454295,
      "loss": 5.6116,
      "mean_token_accuracy": 0.1915464147925377,
      "num_tokens": 49048038.0,
      "step": 26300
    },
    {
      "entropy": 5.979941368103027,
      "epoch": 2.322942423172024,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00044587743491983903,
      "loss": 5.7186,
      "mean_token_accuracy": 0.1757347047328949,
      "num_tokens": 49057304.0,
      "step": 26305
    },
    {
      "entropy": 5.997381925582886,
      "epoch": 2.3233839632638644,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004458569461468568,
      "loss": 5.8039,
      "mean_token_accuracy": 0.16884294599294664,
      "num_tokens": 49067071.0,
      "step": 26310
    },
    {
      "entropy": 5.981875467300415,
      "epoch": 2.323825503355705,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044583645402688414,
      "loss": 5.7414,
      "mean_token_accuracy": 0.18366419821977614,
      "num_tokens": 49075814.0,
      "step": 26315
    },
    {
      "entropy": 6.145153427124024,
      "epoch": 2.324267043447545,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004458159585603226,
      "loss": 5.7982,
      "mean_token_accuracy": 0.1717921733856201,
      "num_tokens": 49085831.0,
      "step": 26320
    },
    {
      "entropy": 6.155538940429688,
      "epoch": 2.3247085835393855,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00044579545974757353,
      "loss": 5.7454,
      "mean_token_accuracy": 0.1772438108921051,
      "num_tokens": 49095893.0,
      "step": 26325
    },
    {
      "entropy": 5.937009382247925,
      "epoch": 2.3251501236312255,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004457749575890385,
      "loss": 5.6688,
      "mean_token_accuracy": 0.18532393425703048,
      "num_tokens": 49104810.0,
      "step": 26330
    },
    {
      "entropy": 5.942334222793579,
      "epoch": 2.325591663723066,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004457544520851192,
      "loss": 5.7413,
      "mean_token_accuracy": 0.17340340167284013,
      "num_tokens": 49114108.0,
      "step": 26335
    },
    {
      "entropy": 6.0957090854644775,
      "epoch": 2.3260332038149065,
      "grad_norm": 1.5,
      "learning_rate": 0.00044573394323621735,
      "loss": 5.7002,
      "mean_token_accuracy": 0.17529976218938828,
      "num_tokens": 49123371.0,
      "step": 26340
    },
    {
      "entropy": 6.065323114395142,
      "epoch": 2.3264747439067466,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004457134310427346,
      "loss": 5.7136,
      "mean_token_accuracy": 0.17843590378761293,
      "num_tokens": 49132056.0,
      "step": 26345
    },
    {
      "entropy": 6.013977909088135,
      "epoch": 2.326916283998587,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004456929155050729,
      "loss": 5.7202,
      "mean_token_accuracy": 0.17297108322381974,
      "num_tokens": 49141316.0,
      "step": 26350
    },
    {
      "entropy": 5.9694360256195065,
      "epoch": 2.3273578240904276,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044567239662363393,
      "loss": 5.6584,
      "mean_token_accuracy": 0.18449953943490982,
      "num_tokens": 49151337.0,
      "step": 26355
    },
    {
      "entropy": 6.087922859191894,
      "epoch": 2.3277993641822676,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004456518743988199,
      "loss": 5.6965,
      "mean_token_accuracy": 0.1741637781262398,
      "num_tokens": 49160970.0,
      "step": 26360
    },
    {
      "entropy": 6.04336895942688,
      "epoch": 2.328240904274108,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004456313488310326,
      "loss": 5.6979,
      "mean_token_accuracy": 0.18464103490114211,
      "num_tokens": 49170727.0,
      "step": 26365
    },
    {
      "entropy": 5.993464660644531,
      "epoch": 2.328682444365948,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00044561081992067425,
      "loss": 5.736,
      "mean_token_accuracy": 0.17921882271766662,
      "num_tokens": 49179485.0,
      "step": 26370
    },
    {
      "entropy": 6.087097263336181,
      "epoch": 2.3291239844577887,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000445590287668147,
      "loss": 5.7746,
      "mean_token_accuracy": 0.16599512696266175,
      "num_tokens": 49189339.0,
      "step": 26375
    },
    {
      "entropy": 6.050902080535889,
      "epoch": 2.329565524549629,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00044556975207385305,
      "loss": 5.6652,
      "mean_token_accuracy": 0.18101859241724014,
      "num_tokens": 49198387.0,
      "step": 26380
    },
    {
      "entropy": 6.011427402496338,
      "epoch": 2.3300070646414692,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00044554921313819465,
      "loss": 5.7389,
      "mean_token_accuracy": 0.17790881544351578,
      "num_tokens": 49207518.0,
      "step": 26385
    },
    {
      "entropy": 5.961977434158325,
      "epoch": 2.3304486047333097,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004455286708615742,
      "loss": 5.6487,
      "mean_token_accuracy": 0.1875054121017456,
      "num_tokens": 49217031.0,
      "step": 26390
    },
    {
      "entropy": 6.0387269973754885,
      "epoch": 2.3308901448251502,
      "grad_norm": 1.40625,
      "learning_rate": 0.000445508125244394,
      "loss": 5.711,
      "mean_token_accuracy": 0.17383566349744797,
      "num_tokens": 49225500.0,
      "step": 26395
    },
    {
      "entropy": 6.045146656036377,
      "epoch": 2.3313316849169903,
      "grad_norm": 1.484375,
      "learning_rate": 0.00044548757628705654,
      "loss": 5.633,
      "mean_token_accuracy": 0.18326546996831894,
      "num_tokens": 49235088.0,
      "step": 26400
    },
    {
      "entropy": 5.983602285385132,
      "epoch": 2.3317732250088308,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004454670239899646,
      "loss": 5.7528,
      "mean_token_accuracy": 0.16998885124921798,
      "num_tokens": 49245597.0,
      "step": 26405
    },
    {
      "entropy": 5.9689473628997805,
      "epoch": 2.3322147651006713,
      "grad_norm": 1.53125,
      "learning_rate": 0.00044544646835352047,
      "loss": 5.575,
      "mean_token_accuracy": 0.19097760021686555,
      "num_tokens": 49254284.0,
      "step": 26410
    },
    {
      "entropy": 6.0040665626525875,
      "epoch": 2.3326563051925113,
      "grad_norm": 1.6875,
      "learning_rate": 0.000445425909378127,
      "loss": 5.7701,
      "mean_token_accuracy": 0.17682541459798812,
      "num_tokens": 49264226.0,
      "step": 26415
    },
    {
      "entropy": 5.963019227981567,
      "epoch": 2.333097845284352,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00044540534706418697,
      "loss": 5.7141,
      "mean_token_accuracy": 0.1803745895624161,
      "num_tokens": 49274022.0,
      "step": 26420
    },
    {
      "entropy": 6.125445747375489,
      "epoch": 2.3335393853761923,
      "grad_norm": 1.609375,
      "learning_rate": 0.000445384781412103,
      "loss": 5.7145,
      "mean_token_accuracy": 0.18100140392780303,
      "num_tokens": 49283376.0,
      "step": 26425
    },
    {
      "entropy": 6.0841151714324955,
      "epoch": 2.3339809254680324,
      "grad_norm": 2.328125,
      "learning_rate": 0.000445364212422278,
      "loss": 5.7687,
      "mean_token_accuracy": 0.17032209485769273,
      "num_tokens": 49292994.0,
      "step": 26430
    },
    {
      "entropy": 6.056407833099366,
      "epoch": 2.334422465559873,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004453436400951152,
      "loss": 5.8174,
      "mean_token_accuracy": 0.17708782702684403,
      "num_tokens": 49303599.0,
      "step": 26435
    },
    {
      "entropy": 6.060115718841553,
      "epoch": 2.3348640056517134,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004453230644310172,
      "loss": 5.7762,
      "mean_token_accuracy": 0.16907054632902146,
      "num_tokens": 49313397.0,
      "step": 26440
    },
    {
      "entropy": 6.101541137695312,
      "epoch": 2.3353055457435534,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044530248543038725,
      "loss": 5.7151,
      "mean_token_accuracy": 0.1789648115634918,
      "num_tokens": 49323288.0,
      "step": 26445
    },
    {
      "entropy": 5.942186594009399,
      "epoch": 2.335747085835394,
      "grad_norm": 2.390625,
      "learning_rate": 0.00044528190309362853,
      "loss": 5.5914,
      "mean_token_accuracy": 0.1923166409134865,
      "num_tokens": 49332153.0,
      "step": 26450
    },
    {
      "entropy": 6.000648307800293,
      "epoch": 2.3361886259272344,
      "grad_norm": 2.375,
      "learning_rate": 0.0004452613174211441,
      "loss": 5.6932,
      "mean_token_accuracy": 0.17484316378831863,
      "num_tokens": 49340773.0,
      "step": 26455
    },
    {
      "entropy": 6.01386137008667,
      "epoch": 2.3366301660190745,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00044524072841333736,
      "loss": 5.7292,
      "mean_token_accuracy": 0.17926356047391892,
      "num_tokens": 49349863.0,
      "step": 26460
    },
    {
      "entropy": 5.983629465103149,
      "epoch": 2.337071706110915,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004452201360706115,
      "loss": 5.5592,
      "mean_token_accuracy": 0.18919723629951476,
      "num_tokens": 49358518.0,
      "step": 26465
    },
    {
      "entropy": 5.968180561065674,
      "epoch": 2.337513246202755,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004451995403933701,
      "loss": 5.6612,
      "mean_token_accuracy": 0.18555983304977416,
      "num_tokens": 49367509.0,
      "step": 26470
    },
    {
      "entropy": 6.011783218383789,
      "epoch": 2.3379547862945955,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004451789413820164,
      "loss": 5.7235,
      "mean_token_accuracy": 0.1789129078388214,
      "num_tokens": 49377157.0,
      "step": 26475
    },
    {
      "entropy": 6.138460779190064,
      "epoch": 2.338396326386436,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004451583390369541,
      "loss": 5.7744,
      "mean_token_accuracy": 0.1719501033425331,
      "num_tokens": 49386580.0,
      "step": 26480
    },
    {
      "entropy": 6.01350679397583,
      "epoch": 2.338837866478276,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004451377333585866,
      "loss": 5.6344,
      "mean_token_accuracy": 0.18286568969488143,
      "num_tokens": 49395894.0,
      "step": 26485
    },
    {
      "entropy": 5.954947662353516,
      "epoch": 2.3392794065701166,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044511712434731774,
      "loss": 5.668,
      "mean_token_accuracy": 0.19009291231632233,
      "num_tokens": 49405475.0,
      "step": 26490
    },
    {
      "entropy": 6.045060729980468,
      "epoch": 2.339720946661957,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004450965120035511,
      "loss": 5.7074,
      "mean_token_accuracy": 0.17707425802946092,
      "num_tokens": 49414982.0,
      "step": 26495
    },
    {
      "entropy": 6.012358713150024,
      "epoch": 2.340162486753797,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004450758963276907,
      "loss": 5.6522,
      "mean_token_accuracy": 0.18616560250520706,
      "num_tokens": 49425502.0,
      "step": 26500
    },
    {
      "entropy": 6.040039968490601,
      "epoch": 2.3406040268456376,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00044505527732014,
      "loss": 5.6309,
      "mean_token_accuracy": 0.1747712790966034,
      "num_tokens": 49435278.0,
      "step": 26505
    },
    {
      "entropy": 5.938655042648316,
      "epoch": 2.3410455669374777,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004450346549813033,
      "loss": 5.6867,
      "mean_token_accuracy": 0.18396445661783217,
      "num_tokens": 49444505.0,
      "step": 26510
    },
    {
      "entropy": 5.936652183532715,
      "epoch": 2.341487107029318,
      "grad_norm": 1.9375,
      "learning_rate": 0.00044501402931158433,
      "loss": 5.6574,
      "mean_token_accuracy": 0.18270938098430634,
      "num_tokens": 49453485.0,
      "step": 26515
    },
    {
      "entropy": 6.014896583557129,
      "epoch": 2.3419286471211587,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004449934003113872,
      "loss": 5.6134,
      "mean_token_accuracy": 0.18438885658979415,
      "num_tokens": 49462319.0,
      "step": 26520
    },
    {
      "entropy": 6.027523422241211,
      "epoch": 2.3423701872129987,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004449727679811161,
      "loss": 5.6799,
      "mean_token_accuracy": 0.18641518354415892,
      "num_tokens": 49470949.0,
      "step": 26525
    },
    {
      "entropy": 6.036988735198975,
      "epoch": 2.342811727304839,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004449521323211751,
      "loss": 5.6716,
      "mean_token_accuracy": 0.17946166247129441,
      "num_tokens": 49480269.0,
      "step": 26530
    },
    {
      "entropy": 6.039339590072632,
      "epoch": 2.3432532673966797,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004449314933319685,
      "loss": 5.6894,
      "mean_token_accuracy": 0.17937133610248565,
      "num_tokens": 49489364.0,
      "step": 26535
    },
    {
      "entropy": 6.039367151260376,
      "epoch": 2.3436948074885198,
      "grad_norm": 2.125,
      "learning_rate": 0.00044491085101390065,
      "loss": 5.6903,
      "mean_token_accuracy": 0.1775499016046524,
      "num_tokens": 49499360.0,
      "step": 26540
    },
    {
      "entropy": 5.9291881084442135,
      "epoch": 2.3441363475803603,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004448902053673758,
      "loss": 5.5886,
      "mean_token_accuracy": 0.18055369555950165,
      "num_tokens": 49507770.0,
      "step": 26545
    },
    {
      "entropy": 5.918485975265503,
      "epoch": 2.3445778876722008,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044486955639279845,
      "loss": 5.6682,
      "mean_token_accuracy": 0.18261973857879638,
      "num_tokens": 49516824.0,
      "step": 26550
    },
    {
      "entropy": 6.008398151397705,
      "epoch": 2.345019427764041,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004448489040905731,
      "loss": 5.6725,
      "mean_token_accuracy": 0.1794450893998146,
      "num_tokens": 49525176.0,
      "step": 26555
    },
    {
      "entropy": 6.0761150360107425,
      "epoch": 2.3454609678558813,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004448282484611043,
      "loss": 5.7929,
      "mean_token_accuracy": 0.17517825812101365,
      "num_tokens": 49535701.0,
      "step": 26560
    },
    {
      "entropy": 5.987164402008057,
      "epoch": 2.345902507947722,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044480758950479673,
      "loss": 5.7,
      "mean_token_accuracy": 0.1909013569355011,
      "num_tokens": 49545473.0,
      "step": 26565
    },
    {
      "entropy": 5.946458530426026,
      "epoch": 2.346344048039562,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00044478692722205507,
      "loss": 5.6136,
      "mean_token_accuracy": 0.1805594563484192,
      "num_tokens": 49553955.0,
      "step": 26570
    },
    {
      "entropy": 6.03220009803772,
      "epoch": 2.3467855881314024,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000444766261613284,
      "loss": 5.7399,
      "mean_token_accuracy": 0.17075813114643096,
      "num_tokens": 49563305.0,
      "step": 26575
    },
    {
      "entropy": 6.0108795166015625,
      "epoch": 2.347227128223243,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044474559267888847,
      "loss": 5.7166,
      "mean_token_accuracy": 0.18201910555362702,
      "num_tokens": 49571841.0,
      "step": 26580
    },
    {
      "entropy": 6.04520845413208,
      "epoch": 2.347668668315083,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004447249204192733,
      "loss": 5.7104,
      "mean_token_accuracy": 0.17642883211374283,
      "num_tokens": 49582862.0,
      "step": 26585
    },
    {
      "entropy": 6.071958303451538,
      "epoch": 2.3481102084069234,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004447042448348435,
      "loss": 5.738,
      "mean_token_accuracy": 0.1785326525568962,
      "num_tokens": 49592410.0,
      "step": 26590
    },
    {
      "entropy": 6.006961011886597,
      "epoch": 2.348551748498764,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044468356592600405,
      "loss": 5.6887,
      "mean_token_accuracy": 0.18324469178915023,
      "num_tokens": 49600582.0,
      "step": 26595
    },
    {
      "entropy": 6.010526418685913,
      "epoch": 2.348993288590604,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044466288369316006,
      "loss": 5.6906,
      "mean_token_accuracy": 0.18069593906402587,
      "num_tokens": 49610165.0,
      "step": 26600
    },
    {
      "entropy": 5.934121990203858,
      "epoch": 2.3494348286824445,
      "grad_norm": 1.515625,
      "learning_rate": 0.00044464219813671664,
      "loss": 5.6752,
      "mean_token_accuracy": 0.19198964387178422,
      "num_tokens": 49619231.0,
      "step": 26605
    },
    {
      "entropy": 5.979853773117066,
      "epoch": 2.3498763687742845,
      "grad_norm": 1.5625,
      "learning_rate": 0.000444621509257079,
      "loss": 5.5836,
      "mean_token_accuracy": 0.1912298619747162,
      "num_tokens": 49627899.0,
      "step": 26610
    },
    {
      "entropy": 6.029573678970337,
      "epoch": 2.350317908866125,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044460081705465247,
      "loss": 5.7854,
      "mean_token_accuracy": 0.16748727560043336,
      "num_tokens": 49637866.0,
      "step": 26615
    },
    {
      "entropy": 6.097955846786499,
      "epoch": 2.3507594489579655,
      "grad_norm": 2.5,
      "learning_rate": 0.00044458012152984253,
      "loss": 5.7529,
      "mean_token_accuracy": 0.18221017122268676,
      "num_tokens": 49648273.0,
      "step": 26620
    },
    {
      "entropy": 6.100955629348755,
      "epoch": 2.3512009890498056,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004445594226830544,
      "loss": 5.718,
      "mean_token_accuracy": 0.17429618686437606,
      "num_tokens": 49657944.0,
      "step": 26625
    },
    {
      "entropy": 6.021671199798584,
      "epoch": 2.351642529141646,
      "grad_norm": 1.359375,
      "learning_rate": 0.00044453872051469345,
      "loss": 5.7102,
      "mean_token_accuracy": 0.1835184872150421,
      "num_tokens": 49667158.0,
      "step": 26630
    },
    {
      "entropy": 5.974455976486206,
      "epoch": 2.3520840692334866,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044451801502516555,
      "loss": 5.6222,
      "mean_token_accuracy": 0.18277681469917298,
      "num_tokens": 49675535.0,
      "step": 26635
    },
    {
      "entropy": 5.985066509246826,
      "epoch": 2.3525256093253266,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004444973062148761,
      "loss": 5.7238,
      "mean_token_accuracy": 0.17867843061685562,
      "num_tokens": 49684922.0,
      "step": 26640
    },
    {
      "entropy": 6.055995941162109,
      "epoch": 2.352967149417167,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004444765940842308,
      "loss": 5.6826,
      "mean_token_accuracy": 0.1748145580291748,
      "num_tokens": 49695302.0,
      "step": 26645
    },
    {
      "entropy": 6.039875030517578,
      "epoch": 2.3534086895090076,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004444558786336355,
      "loss": 5.7044,
      "mean_token_accuracy": 0.18561569154262542,
      "num_tokens": 49704668.0,
      "step": 26650
    },
    {
      "entropy": 5.941536140441895,
      "epoch": 2.3538502296008477,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004444351598634958,
      "loss": 5.6461,
      "mean_token_accuracy": 0.1846553862094879,
      "num_tokens": 49713704.0,
      "step": 26655
    },
    {
      "entropy": 6.0303122997283936,
      "epoch": 2.354291769692688,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004444144377742178,
      "loss": 5.7231,
      "mean_token_accuracy": 0.1781400039792061,
      "num_tokens": 49723743.0,
      "step": 26660
    },
    {
      "entropy": 6.0707885265350345,
      "epoch": 2.354733309784528,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004443937123662071,
      "loss": 5.8022,
      "mean_token_accuracy": 0.17360570579767226,
      "num_tokens": 49734760.0,
      "step": 26665
    },
    {
      "entropy": 6.013079023361206,
      "epoch": 2.3551748498763687,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004443729836398702,
      "loss": 5.6825,
      "mean_token_accuracy": 0.17483025938272476,
      "num_tokens": 49744836.0,
      "step": 26670
    },
    {
      "entropy": 6.01318588256836,
      "epoch": 2.355616389968209,
      "grad_norm": 1.515625,
      "learning_rate": 0.00044435225159561263,
      "loss": 5.6453,
      "mean_token_accuracy": 0.18602761775255203,
      "num_tokens": 49753646.0,
      "step": 26675
    },
    {
      "entropy": 5.988765001296997,
      "epoch": 2.3560579300600493,
      "grad_norm": 1.25,
      "learning_rate": 0.0004443315162338409,
      "loss": 5.6949,
      "mean_token_accuracy": 0.18096723407506943,
      "num_tokens": 49762576.0,
      "step": 26680
    },
    {
      "entropy": 5.999406242370606,
      "epoch": 2.3564994701518898,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000444310777554961,
      "loss": 5.6789,
      "mean_token_accuracy": 0.17883919030427933,
      "num_tokens": 49772614.0,
      "step": 26685
    },
    {
      "entropy": 5.998238229751587,
      "epoch": 2.3569410102437303,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004442900355593793,
      "loss": 5.6463,
      "mean_token_accuracy": 0.18457995653152465,
      "num_tokens": 49781005.0,
      "step": 26690
    },
    {
      "entropy": 5.96651258468628,
      "epoch": 2.3573825503355703,
      "grad_norm": 1.4375,
      "learning_rate": 0.00044426929024750204,
      "loss": 5.6429,
      "mean_token_accuracy": 0.18095150887966155,
      "num_tokens": 49790639.0,
      "step": 26695
    },
    {
      "entropy": 6.016487979888916,
      "epoch": 2.357824090427411,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004442485416197357,
      "loss": 5.7657,
      "mean_token_accuracy": 0.173963762819767,
      "num_tokens": 49800010.0,
      "step": 26700
    },
    {
      "entropy": 6.012135648727417,
      "epoch": 2.3582656305192513,
      "grad_norm": 1.375,
      "learning_rate": 0.00044422778967648675,
      "loss": 5.6557,
      "mean_token_accuracy": 0.17883078753948212,
      "num_tokens": 49808510.0,
      "step": 26705
    },
    {
      "entropy": 6.045469188690186,
      "epoch": 2.3587071706110914,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044420703441816153,
      "loss": 5.7074,
      "mean_token_accuracy": 0.1754029929637909,
      "num_tokens": 49818017.0,
      "step": 26710
    },
    {
      "entropy": 5.989930105209351,
      "epoch": 2.359148710702932,
      "grad_norm": 1.75,
      "learning_rate": 0.0004441862758451668,
      "loss": 5.6572,
      "mean_token_accuracy": 0.18117285072803496,
      "num_tokens": 49826797.0,
      "step": 26715
    },
    {
      "entropy": 5.978230094909668,
      "epoch": 2.3595902507947724,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044416551395790916,
      "loss": 5.6507,
      "mean_token_accuracy": 0.18111518770456314,
      "num_tokens": 49835632.0,
      "step": 26720
    },
    {
      "entropy": 5.998672389984131,
      "epoch": 2.3600317908866124,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004441447487567953,
      "loss": 5.6997,
      "mean_token_accuracy": 0.18032974004745483,
      "num_tokens": 49844770.0,
      "step": 26725
    },
    {
      "entropy": 6.018344402313232,
      "epoch": 2.360473330978453,
      "grad_norm": 1.59375,
      "learning_rate": 0.00044412398024223194,
      "loss": 5.7362,
      "mean_token_accuracy": 0.17834548205137252,
      "num_tokens": 49854170.0,
      "step": 26730
    },
    {
      "entropy": 6.086949062347412,
      "epoch": 2.3609148710702934,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004441032084146261,
      "loss": 5.7467,
      "mean_token_accuracy": 0.17796375453472138,
      "num_tokens": 49863541.0,
      "step": 26735
    },
    {
      "entropy": 6.031700229644775,
      "epoch": 2.3613564111621335,
      "grad_norm": 2.109375,
      "learning_rate": 0.00044408243327438446,
      "loss": 5.6591,
      "mean_token_accuracy": 0.18156271576881408,
      "num_tokens": 49872556.0,
      "step": 26740
    },
    {
      "entropy": 6.006669998168945,
      "epoch": 2.361797951253974,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00044406165482191424,
      "loss": 5.7533,
      "mean_token_accuracy": 0.17658618539571763,
      "num_tokens": 49882188.0,
      "step": 26745
    },
    {
      "entropy": 6.03294734954834,
      "epoch": 2.362239491345814,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004440408730576223,
      "loss": 5.7775,
      "mean_token_accuracy": 0.1678992658853531,
      "num_tokens": 49891980.0,
      "step": 26750
    },
    {
      "entropy": 6.110069513320923,
      "epoch": 2.3626810314376545,
      "grad_norm": 1.421875,
      "learning_rate": 0.00044402008798191575,
      "loss": 5.6913,
      "mean_token_accuracy": 0.18307857364416122,
      "num_tokens": 49900783.0,
      "step": 26755
    },
    {
      "entropy": 6.013420486450196,
      "epoch": 2.363122571529495,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004439992995952019,
      "loss": 5.7173,
      "mean_token_accuracy": 0.17956802099943162,
      "num_tokens": 49910525.0,
      "step": 26760
    },
    {
      "entropy": 5.95426115989685,
      "epoch": 2.363564111621335,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044397850789788774,
      "loss": 5.6905,
      "mean_token_accuracy": 0.18085169047117233,
      "num_tokens": 49919286.0,
      "step": 26765
    },
    {
      "entropy": 5.978308200836182,
      "epoch": 2.3640056517131756,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004439577128903808,
      "loss": 5.6396,
      "mean_token_accuracy": 0.187736576795578,
      "num_tokens": 49927882.0,
      "step": 26770
    },
    {
      "entropy": 6.0808434009552,
      "epoch": 2.364447191805016,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004439369145730884,
      "loss": 5.7305,
      "mean_token_accuracy": 0.17391123473644257,
      "num_tokens": 49936289.0,
      "step": 26775
    },
    {
      "entropy": 6.080900001525879,
      "epoch": 2.364888731896856,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004439161129464179,
      "loss": 5.7192,
      "mean_token_accuracy": 0.17568335235118865,
      "num_tokens": 49945356.0,
      "step": 26780
    },
    {
      "entropy": 5.99698839187622,
      "epoch": 2.3653302719886966,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044389530801077675,
      "loss": 5.6442,
      "mean_token_accuracy": 0.18462553024291992,
      "num_tokens": 49954216.0,
      "step": 26785
    },
    {
      "entropy": 6.00535135269165,
      "epoch": 2.365771812080537,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004438744997665726,
      "loss": 5.7737,
      "mean_token_accuracy": 0.1794869512319565,
      "num_tokens": 49963179.0,
      "step": 26790
    },
    {
      "entropy": 6.017975950241089,
      "epoch": 2.366213352172377,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004438536882142131,
      "loss": 5.7663,
      "mean_token_accuracy": 0.17260024845600128,
      "num_tokens": 49973108.0,
      "step": 26795
    },
    {
      "entropy": 6.099006223678589,
      "epoch": 2.3666548922642177,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004438328733541059,
      "loss": 5.678,
      "mean_token_accuracy": 0.17383593022823335,
      "num_tokens": 49982003.0,
      "step": 26800
    },
    {
      "entropy": 6.052188205718994,
      "epoch": 2.3670964323560577,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004438120551866586,
      "loss": 5.6992,
      "mean_token_accuracy": 0.18377901762723922,
      "num_tokens": 49992034.0,
      "step": 26805
    },
    {
      "entropy": 5.947895097732544,
      "epoch": 2.367537972447898,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004437912337122792,
      "loss": 5.6783,
      "mean_token_accuracy": 0.1837504342198372,
      "num_tokens": 50001379.0,
      "step": 26810
    },
    {
      "entropy": 6.058159208297729,
      "epoch": 2.3679795125397387,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004437704089313756,
      "loss": 5.7693,
      "mean_token_accuracy": 0.1737336203455925,
      "num_tokens": 50010721.0,
      "step": 26815
    },
    {
      "entropy": 6.058268356323242,
      "epoch": 2.3684210526315788,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00044374958084435564,
      "loss": 5.7505,
      "mean_token_accuracy": 0.1772601455450058,
      "num_tokens": 50019571.0,
      "step": 26820
    },
    {
      "entropy": 5.992744970321655,
      "epoch": 2.3688625927234193,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044372874945162734,
      "loss": 5.6657,
      "mean_token_accuracy": 0.17877335995435714,
      "num_tokens": 50028045.0,
      "step": 26825
    },
    {
      "entropy": 6.083613014221191,
      "epoch": 2.3693041328152598,
      "grad_norm": 1.546875,
      "learning_rate": 0.00044370791475359877,
      "loss": 5.6635,
      "mean_token_accuracy": 0.18022848516702653,
      "num_tokens": 50036769.0,
      "step": 26830
    },
    {
      "entropy": 5.983336639404297,
      "epoch": 2.3697456729071,
      "grad_norm": 1.515625,
      "learning_rate": 0.00044368707675067814,
      "loss": 5.7007,
      "mean_token_accuracy": 0.18008812218904496,
      "num_tokens": 50045973.0,
      "step": 26835
    },
    {
      "entropy": 5.94638204574585,
      "epoch": 2.3701872129989403,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004436662354432736,
      "loss": 5.6226,
      "mean_token_accuracy": 0.18298470228910446,
      "num_tokens": 50055243.0,
      "step": 26840
    },
    {
      "entropy": 6.0585705757141115,
      "epoch": 2.370628753090781,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044364539083179347,
      "loss": 5.6989,
      "mean_token_accuracy": 0.18291028887033461,
      "num_tokens": 50064573.0,
      "step": 26845
    },
    {
      "entropy": 5.9375170230865475,
      "epoch": 2.371070293182621,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000443624542916646,
      "loss": 5.5455,
      "mean_token_accuracy": 0.18975622355937957,
      "num_tokens": 50073578.0,
      "step": 26850
    },
    {
      "entropy": 5.947583532333374,
      "epoch": 2.3715118332744614,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004436036916982397,
      "loss": 5.7041,
      "mean_token_accuracy": 0.1793969064950943,
      "num_tokens": 50083620.0,
      "step": 26855
    },
    {
      "entropy": 5.9521886825561525,
      "epoch": 2.371953373366302,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004435828371769829,
      "loss": 5.6268,
      "mean_token_accuracy": 0.18917134404182434,
      "num_tokens": 50092050.0,
      "step": 26860
    },
    {
      "entropy": 6.075591945648194,
      "epoch": 2.372394913458142,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044356197935328426,
      "loss": 5.7378,
      "mean_token_accuracy": 0.17460485100746154,
      "num_tokens": 50102425.0,
      "step": 26865
    },
    {
      "entropy": 6.020492172241211,
      "epoch": 2.3728364535499824,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044354111822755225,
      "loss": 5.6062,
      "mean_token_accuracy": 0.1781404510140419,
      "num_tokens": 50112338.0,
      "step": 26870
    },
    {
      "entropy": 5.967376661300659,
      "epoch": 2.373277993641823,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00044352025380019563,
      "loss": 5.7194,
      "mean_token_accuracy": 0.17801481038331984,
      "num_tokens": 50121302.0,
      "step": 26875
    },
    {
      "entropy": 5.981369495391846,
      "epoch": 2.373719533733663,
      "grad_norm": 1.984375,
      "learning_rate": 0.00044349938607162305,
      "loss": 5.6266,
      "mean_token_accuracy": 0.182475683093071,
      "num_tokens": 50129820.0,
      "step": 26880
    },
    {
      "entropy": 6.034530878067017,
      "epoch": 2.3741610738255035,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00044347851504224336,
      "loss": 5.7442,
      "mean_token_accuracy": 0.17407308965921403,
      "num_tokens": 50138860.0,
      "step": 26885
    },
    {
      "entropy": 6.00489616394043,
      "epoch": 2.3746026139173435,
      "grad_norm": 1.84375,
      "learning_rate": 0.00044345764071246536,
      "loss": 5.659,
      "mean_token_accuracy": 0.18287645727396012,
      "num_tokens": 50148257.0,
      "step": 26890
    },
    {
      "entropy": 5.980583000183105,
      "epoch": 2.375044154009184,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00044343676308269796,
      "loss": 5.6501,
      "mean_token_accuracy": 0.18523019850254058,
      "num_tokens": 50157380.0,
      "step": 26895
    },
    {
      "entropy": 6.106829214096069,
      "epoch": 2.3754856941010245,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044341588215335016,
      "loss": 5.7014,
      "mean_token_accuracy": 0.181707900762558,
      "num_tokens": 50167806.0,
      "step": 26900
    },
    {
      "entropy": 6.070315313339234,
      "epoch": 2.3759272341928646,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044339499792483104,
      "loss": 5.7372,
      "mean_token_accuracy": 0.17850941568613052,
      "num_tokens": 50176763.0,
      "step": 26905
    },
    {
      "entropy": 5.890826272964477,
      "epoch": 2.376368774284705,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004433741103975498,
      "loss": 5.6947,
      "mean_token_accuracy": 0.1843548059463501,
      "num_tokens": 50186735.0,
      "step": 26910
    },
    {
      "entropy": 5.947778558731079,
      "epoch": 2.3768103143765456,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044335321957191535,
      "loss": 5.6576,
      "mean_token_accuracy": 0.1808363601565361,
      "num_tokens": 50195206.0,
      "step": 26915
    },
    {
      "entropy": 6.089373636245727,
      "epoch": 2.3772518544683856,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044333232544833707,
      "loss": 5.7413,
      "mean_token_accuracy": 0.1750466838479042,
      "num_tokens": 50203706.0,
      "step": 26920
    },
    {
      "entropy": 6.025934171676636,
      "epoch": 2.377693394560226,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004433114280272243,
      "loss": 5.7183,
      "mean_token_accuracy": 0.1740570545196533,
      "num_tokens": 50213007.0,
      "step": 26925
    },
    {
      "entropy": 6.0382524013519285,
      "epoch": 2.3781349346520666,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004432905273089864,
      "loss": 5.6494,
      "mean_token_accuracy": 0.18398965299129486,
      "num_tokens": 50223155.0,
      "step": 26930
    },
    {
      "entropy": 5.901139163970948,
      "epoch": 2.3785764747439067,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00044326962329403275,
      "loss": 5.5649,
      "mean_token_accuracy": 0.19197474420070648,
      "num_tokens": 50231788.0,
      "step": 26935
    },
    {
      "entropy": 5.968618869781494,
      "epoch": 2.379018014835747,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044324871598277294,
      "loss": 5.6434,
      "mean_token_accuracy": 0.18300013691186906,
      "num_tokens": 50241557.0,
      "step": 26940
    },
    {
      "entropy": 6.030010318756103,
      "epoch": 2.379459554927587,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044322780537561643,
      "loss": 5.7598,
      "mean_token_accuracy": 0.18020653873682022,
      "num_tokens": 50252141.0,
      "step": 26945
    },
    {
      "entropy": 6.121658611297607,
      "epoch": 2.3799010950194277,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004432068914729729,
      "loss": 5.7615,
      "mean_token_accuracy": 0.1744946599006653,
      "num_tokens": 50261361.0,
      "step": 26950
    },
    {
      "entropy": 6.0255224227905275,
      "epoch": 2.380342635111268,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00044318597427525197,
      "loss": 5.7533,
      "mean_token_accuracy": 0.17378995269536973,
      "num_tokens": 50270193.0,
      "step": 26955
    },
    {
      "entropy": 6.071790552139282,
      "epoch": 2.3807841752031083,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00044316505378286356,
      "loss": 5.7195,
      "mean_token_accuracy": 0.17362090796232224,
      "num_tokens": 50280217.0,
      "step": 26960
    },
    {
      "entropy": 5.994744777679443,
      "epoch": 2.3812257152949488,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004431441299962173,
      "loss": 5.6296,
      "mean_token_accuracy": 0.1793859988451004,
      "num_tokens": 50289610.0,
      "step": 26965
    },
    {
      "entropy": 5.966510963439942,
      "epoch": 2.3816672553867893,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004431232029157232,
      "loss": 5.5849,
      "mean_token_accuracy": 0.18556904792785645,
      "num_tokens": 50298734.0,
      "step": 26970
    },
    {
      "entropy": 5.97077054977417,
      "epoch": 2.3821087954786293,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004431022725417912,
      "loss": 5.6096,
      "mean_token_accuracy": 0.18704057335853577,
      "num_tokens": 50307679.0,
      "step": 26975
    },
    {
      "entropy": 6.066034889221191,
      "epoch": 2.38255033557047,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004430813388748312,
      "loss": 5.6112,
      "mean_token_accuracy": 0.18847079277038575,
      "num_tokens": 50315983.0,
      "step": 26980
    },
    {
      "entropy": 5.963740730285645,
      "epoch": 2.3829918756623103,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00044306040191525347,
      "loss": 5.7235,
      "mean_token_accuracy": 0.17836444526910783,
      "num_tokens": 50325222.0,
      "step": 26985
    },
    {
      "entropy": 5.916135358810425,
      "epoch": 2.3834334157541504,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000443039461663468,
      "loss": 5.6533,
      "mean_token_accuracy": 0.1851285830140114,
      "num_tokens": 50333417.0,
      "step": 26990
    },
    {
      "entropy": 5.981042957305908,
      "epoch": 2.383874955845991,
      "grad_norm": 1.765625,
      "learning_rate": 0.00044301851811988504,
      "loss": 5.6351,
      "mean_token_accuracy": 0.19270563423633574,
      "num_tokens": 50342836.0,
      "step": 26995
    },
    {
      "entropy": 6.027767038345337,
      "epoch": 2.3843164959378313,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00044299757128491487,
      "loss": 5.7744,
      "mean_token_accuracy": 0.1751588225364685,
      "num_tokens": 50351988.0,
      "step": 27000
    },
    {
      "epoch": 2.3843164959378313,
      "eval_entropy": 5.759912882898834,
      "eval_loss": 5.775855541229248,
      "eval_mean_token_accuracy": 0.1834229815923762,
      "eval_num_tokens": 50351988.0,
      "eval_runtime": 26.2883,
      "eval_samples_per_second": 1343.373,
      "eval_steps_per_second": 167.945,
      "step": 27000
    },
    {
      "entropy": 6.0086253643035885,
      "epoch": 2.3847580360296714,
      "grad_norm": 1.578125,
      "learning_rate": 0.00044297662115896783,
      "loss": 5.7497,
      "mean_token_accuracy": 0.1768230065703392,
      "num_tokens": 50361087.0,
      "step": 27005
    },
    {
      "entropy": 6.047536945343017,
      "epoch": 2.385199576121512,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00044295566774245435,
      "loss": 5.7091,
      "mean_token_accuracy": 0.17741555720567703,
      "num_tokens": 50369842.0,
      "step": 27010
    },
    {
      "entropy": 6.094355916976928,
      "epoch": 2.3856411162133524,
      "grad_norm": 1.375,
      "learning_rate": 0.00044293471103578475,
      "loss": 5.7815,
      "mean_token_accuracy": 0.17727210819721223,
      "num_tokens": 50379125.0,
      "step": 27015
    },
    {
      "entropy": 5.908969259262085,
      "epoch": 2.3860826563051925,
      "grad_norm": 1.484375,
      "learning_rate": 0.00044291375103936977,
      "loss": 5.6052,
      "mean_token_accuracy": 0.18916090726852416,
      "num_tokens": 50388031.0,
      "step": 27020
    },
    {
      "entropy": 5.995253610610962,
      "epoch": 2.386524196397033,
      "grad_norm": 1.75,
      "learning_rate": 0.00044289278775361984,
      "loss": 5.7217,
      "mean_token_accuracy": 0.17574053555727004,
      "num_tokens": 50397348.0,
      "step": 27025
    },
    {
      "entropy": 6.080354452133179,
      "epoch": 2.386965736488873,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044287182117894576,
      "loss": 5.7222,
      "mean_token_accuracy": 0.18347825855016708,
      "num_tokens": 50406782.0,
      "step": 27030
    },
    {
      "entropy": 5.99479193687439,
      "epoch": 2.3874072765807135,
      "grad_norm": 1.515625,
      "learning_rate": 0.00044285085131575807,
      "loss": 5.6376,
      "mean_token_accuracy": 0.18587710559368134,
      "num_tokens": 50415527.0,
      "step": 27035
    },
    {
      "entropy": 6.013674211502075,
      "epoch": 2.387848816672554,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044282987816446774,
      "loss": 5.7458,
      "mean_token_accuracy": 0.17529134452342987,
      "num_tokens": 50424852.0,
      "step": 27040
    },
    {
      "entropy": 5.992533540725708,
      "epoch": 2.388290356764394,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004428089017254856,
      "loss": 5.7238,
      "mean_token_accuracy": 0.17698214799165726,
      "num_tokens": 50434371.0,
      "step": 27045
    },
    {
      "entropy": 6.093685865402222,
      "epoch": 2.3887318968562345,
      "grad_norm": 1.859375,
      "learning_rate": 0.00044278792199922245,
      "loss": 5.7624,
      "mean_token_accuracy": 0.17501270323991774,
      "num_tokens": 50443358.0,
      "step": 27050
    },
    {
      "entropy": 6.036582040786743,
      "epoch": 2.389173436948075,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00044276693898608933,
      "loss": 5.6826,
      "mean_token_accuracy": 0.18023095279932022,
      "num_tokens": 50453132.0,
      "step": 27055
    },
    {
      "entropy": 5.993710041046143,
      "epoch": 2.389614977039915,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044274595268649727,
      "loss": 5.6363,
      "mean_token_accuracy": 0.18962339013814927,
      "num_tokens": 50461837.0,
      "step": 27060
    },
    {
      "entropy": 5.981816720962525,
      "epoch": 2.3900565171317556,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044272496310085745,
      "loss": 5.6592,
      "mean_token_accuracy": 0.18713430166244507,
      "num_tokens": 50470482.0,
      "step": 27065
    },
    {
      "entropy": 5.98534836769104,
      "epoch": 2.390498057223596,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00044270397022958094,
      "loss": 5.7009,
      "mean_token_accuracy": 0.17315630316734315,
      "num_tokens": 50480168.0,
      "step": 27070
    },
    {
      "entropy": 5.979889631271362,
      "epoch": 2.390939597315436,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00044268297407307905,
      "loss": 5.6872,
      "mean_token_accuracy": 0.18153275698423385,
      "num_tokens": 50489937.0,
      "step": 27075
    },
    {
      "entropy": 5.977669620513916,
      "epoch": 2.3913811374072766,
      "grad_norm": 1.609375,
      "learning_rate": 0.00044266197463176305,
      "loss": 5.7084,
      "mean_token_accuracy": 0.17394983172416686,
      "num_tokens": 50499151.0,
      "step": 27080
    },
    {
      "entropy": 6.110803127288818,
      "epoch": 2.3918226774991167,
      "grad_norm": 1.65625,
      "learning_rate": 0.00044264097190604436,
      "loss": 5.7777,
      "mean_token_accuracy": 0.17507931292057038,
      "num_tokens": 50508413.0,
      "step": 27085
    },
    {
      "entropy": 6.095498275756836,
      "epoch": 2.392264217590957,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004426199658963344,
      "loss": 5.771,
      "mean_token_accuracy": 0.17893251925706863,
      "num_tokens": 50517513.0,
      "step": 27090
    },
    {
      "entropy": 6.000348520278931,
      "epoch": 2.3927057576827977,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044259895660304455,
      "loss": 5.6237,
      "mean_token_accuracy": 0.18997481912374498,
      "num_tokens": 50526057.0,
      "step": 27095
    },
    {
      "entropy": 5.8898537158966064,
      "epoch": 2.3931472977746377,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044257794402658655,
      "loss": 5.6043,
      "mean_token_accuracy": 0.18234999626874923,
      "num_tokens": 50535388.0,
      "step": 27100
    },
    {
      "entropy": 6.057642936706543,
      "epoch": 2.3935888378664782,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004425569281673719,
      "loss": 5.7028,
      "mean_token_accuracy": 0.17300032079219818,
      "num_tokens": 50544606.0,
      "step": 27105
    },
    {
      "entropy": 5.987120342254639,
      "epoch": 2.3940303779583187,
      "grad_norm": 1.53125,
      "learning_rate": 0.00044253590902581223,
      "loss": 5.6891,
      "mean_token_accuracy": 0.17663753777742386,
      "num_tokens": 50553822.0,
      "step": 27110
    },
    {
      "entropy": 5.975704145431519,
      "epoch": 2.394471918050159,
      "grad_norm": 1.5,
      "learning_rate": 0.0004425148866023195,
      "loss": 5.6814,
      "mean_token_accuracy": 0.18400569409132003,
      "num_tokens": 50562879.0,
      "step": 27115
    },
    {
      "entropy": 5.939655590057373,
      "epoch": 2.3949134581419993,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044249386089730534,
      "loss": 5.6097,
      "mean_token_accuracy": 0.189388208091259,
      "num_tokens": 50571536.0,
      "step": 27120
    },
    {
      "entropy": 5.930270719528198,
      "epoch": 2.39535499823384,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004424728319111818,
      "loss": 5.6109,
      "mean_token_accuracy": 0.18388313204050064,
      "num_tokens": 50580692.0,
      "step": 27125
    },
    {
      "entropy": 6.075661706924438,
      "epoch": 2.39579653832568,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004424517996443606,
      "loss": 5.8331,
      "mean_token_accuracy": 0.1792971208691597,
      "num_tokens": 50589456.0,
      "step": 27130
    },
    {
      "entropy": 6.033675909042358,
      "epoch": 2.3962380784175203,
      "grad_norm": 1.453125,
      "learning_rate": 0.00044243076409725395,
      "loss": 5.6608,
      "mean_token_accuracy": 0.1824326753616333,
      "num_tokens": 50598947.0,
      "step": 27135
    },
    {
      "entropy": 6.0575621128082275,
      "epoch": 2.396679618509361,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044240972527027385,
      "loss": 5.7091,
      "mean_token_accuracy": 0.18370827585458754,
      "num_tokens": 50607373.0,
      "step": 27140
    },
    {
      "entropy": 6.0993749618530275,
      "epoch": 2.397121158601201,
      "grad_norm": 1.84375,
      "learning_rate": 0.00044238868316383246,
      "loss": 5.8003,
      "mean_token_accuracy": 0.17599960565567016,
      "num_tokens": 50616930.0,
      "step": 27145
    },
    {
      "entropy": 5.9961060047149655,
      "epoch": 2.3975626986930414,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000442367637778342,
      "loss": 5.6991,
      "mean_token_accuracy": 0.1875371217727661,
      "num_tokens": 50626658.0,
      "step": 27150
    },
    {
      "entropy": 5.963674736022949,
      "epoch": 2.398004238784882,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004423465891142146,
      "loss": 5.7415,
      "mean_token_accuracy": 0.17402851283550264,
      "num_tokens": 50635692.0,
      "step": 27155
    },
    {
      "entropy": 6.030885505676269,
      "epoch": 2.398445778876722,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00044232553717186274,
      "loss": 5.6081,
      "mean_token_accuracy": 0.19091918766498567,
      "num_tokens": 50644036.0,
      "step": 27160
    },
    {
      "entropy": 5.992510366439819,
      "epoch": 2.3988873189685624,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004423044819516988,
      "loss": 5.7876,
      "mean_token_accuracy": 0.17234549671411514,
      "num_tokens": 50652363.0,
      "step": 27165
    },
    {
      "entropy": 5.980995464324951,
      "epoch": 2.3993288590604025,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004422834234541352,
      "loss": 5.6939,
      "mean_token_accuracy": 0.17581460624933243,
      "num_tokens": 50661985.0,
      "step": 27170
    },
    {
      "entropy": 5.991255807876587,
      "epoch": 2.399770399152243,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004422623616795845,
      "loss": 5.6609,
      "mean_token_accuracy": 0.18304557204246522,
      "num_tokens": 50671171.0,
      "step": 27175
    },
    {
      "entropy": 5.983360147476196,
      "epoch": 2.4002119392440835,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044224129662845916,
      "loss": 5.6434,
      "mean_token_accuracy": 0.17933208346366883,
      "num_tokens": 50679792.0,
      "step": 27180
    },
    {
      "entropy": 5.978426504135132,
      "epoch": 2.4006534793359235,
      "grad_norm": 1.59375,
      "learning_rate": 0.000442220228301172,
      "loss": 5.7681,
      "mean_token_accuracy": 0.17417300939559938,
      "num_tokens": 50690082.0,
      "step": 27185
    },
    {
      "entropy": 5.9925824165344235,
      "epoch": 2.401095019427764,
      "grad_norm": 2.0,
      "learning_rate": 0.0004421991566981357,
      "loss": 5.6322,
      "mean_token_accuracy": 0.19228113889694215,
      "num_tokens": 50698351.0,
      "step": 27190
    },
    {
      "entropy": 5.991819858551025,
      "epoch": 2.4015365595196045,
      "grad_norm": 1.21875,
      "learning_rate": 0.000442178081819763,
      "loss": 5.612,
      "mean_token_accuracy": 0.1869213879108429,
      "num_tokens": 50706977.0,
      "step": 27195
    },
    {
      "entropy": 5.970295667648315,
      "epoch": 2.4019780996114446,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004421570036664667,
      "loss": 5.6859,
      "mean_token_accuracy": 0.18570122122764587,
      "num_tokens": 50716168.0,
      "step": 27200
    },
    {
      "entropy": 6.027118444442749,
      "epoch": 2.402419639703285,
      "grad_norm": 1.9375,
      "learning_rate": 0.00044213592223865976,
      "loss": 5.7074,
      "mean_token_accuracy": 0.17995236814022064,
      "num_tokens": 50725301.0,
      "step": 27205
    },
    {
      "entropy": 6.043588352203369,
      "epoch": 2.4028611797951256,
      "grad_norm": 1.84375,
      "learning_rate": 0.00044211483753675524,
      "loss": 5.6832,
      "mean_token_accuracy": 0.18072980791330337,
      "num_tokens": 50734494.0,
      "step": 27210
    },
    {
      "entropy": 6.012661981582641,
      "epoch": 2.4033027198869656,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004420937495611661,
      "loss": 5.732,
      "mean_token_accuracy": 0.1784653455018997,
      "num_tokens": 50743650.0,
      "step": 27215
    },
    {
      "entropy": 5.939428520202637,
      "epoch": 2.403744259978806,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004420726583123054,
      "loss": 5.633,
      "mean_token_accuracy": 0.17786276638507842,
      "num_tokens": 50753538.0,
      "step": 27220
    },
    {
      "entropy": 6.069830465316772,
      "epoch": 2.404185800070646,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004420515637905863,
      "loss": 5.7625,
      "mean_token_accuracy": 0.1760679855942726,
      "num_tokens": 50763349.0,
      "step": 27225
    },
    {
      "entropy": 6.096151399612427,
      "epoch": 2.4046273401624867,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004420304659964221,
      "loss": 5.7183,
      "mean_token_accuracy": 0.17282784730196,
      "num_tokens": 50772348.0,
      "step": 27230
    },
    {
      "entropy": 6.088141870498657,
      "epoch": 2.405068880254327,
      "grad_norm": 2.078125,
      "learning_rate": 0.00044200936493022616,
      "loss": 5.7838,
      "mean_token_accuracy": 0.1734149530529976,
      "num_tokens": 50782054.0,
      "step": 27235
    },
    {
      "entropy": 6.0886373043060305,
      "epoch": 2.4055104203461672,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004419882605924116,
      "loss": 5.7213,
      "mean_token_accuracy": 0.1818305030465126,
      "num_tokens": 50791955.0,
      "step": 27240
    },
    {
      "entropy": 5.981839561462403,
      "epoch": 2.4059519604380077,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004419671529833921,
      "loss": 5.7266,
      "mean_token_accuracy": 0.17957244962453842,
      "num_tokens": 50801907.0,
      "step": 27245
    },
    {
      "entropy": 6.035249853134156,
      "epoch": 2.4063935005298482,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004419460421035809,
      "loss": 5.8241,
      "mean_token_accuracy": 0.16976408958435057,
      "num_tokens": 50811413.0,
      "step": 27250
    },
    {
      "entropy": 6.040361976623535,
      "epoch": 2.4068350406216883,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004419249279533918,
      "loss": 5.6572,
      "mean_token_accuracy": 0.1823677107691765,
      "num_tokens": 50820080.0,
      "step": 27255
    },
    {
      "entropy": 6.05166654586792,
      "epoch": 2.407276580713529,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044190381053323824,
      "loss": 5.7597,
      "mean_token_accuracy": 0.1742892563343048,
      "num_tokens": 50829622.0,
      "step": 27260
    },
    {
      "entropy": 6.001866388320923,
      "epoch": 2.4077181208053693,
      "grad_norm": 2.703125,
      "learning_rate": 0.00044188268984353396,
      "loss": 5.5996,
      "mean_token_accuracy": 0.19217129349708556,
      "num_tokens": 50838112.0,
      "step": 27265
    },
    {
      "entropy": 5.980332946777343,
      "epoch": 2.4081596608972093,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004418615658846927,
      "loss": 5.7232,
      "mean_token_accuracy": 0.17578680962324142,
      "num_tokens": 50848014.0,
      "step": 27270
    },
    {
      "entropy": 6.13927845954895,
      "epoch": 2.40860120098905,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004418404386571283,
      "loss": 5.7939,
      "mean_token_accuracy": 0.16240599155426025,
      "num_tokens": 50857103.0,
      "step": 27275
    },
    {
      "entropy": 6.005201005935669,
      "epoch": 2.4090427410808903,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004418193081612545,
      "loss": 5.6641,
      "mean_token_accuracy": 0.18026067614555358,
      "num_tokens": 50865866.0,
      "step": 27280
    },
    {
      "entropy": 6.07469367980957,
      "epoch": 2.4094842811727304,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00044179817439748547,
      "loss": 5.7257,
      "mean_token_accuracy": 0.1731061667203903,
      "num_tokens": 50874694.0,
      "step": 27285
    },
    {
      "entropy": 5.9928089618682865,
      "epoch": 2.409925821264571,
      "grad_norm": 2.78125,
      "learning_rate": 0.00044177703736623497,
      "loss": 5.6487,
      "mean_token_accuracy": 0.18003278523683547,
      "num_tokens": 50884965.0,
      "step": 27290
    },
    {
      "entropy": 6.084443664550781,
      "epoch": 2.4103673613564114,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044175589706791724,
      "loss": 5.7626,
      "mean_token_accuracy": 0.17449579536914825,
      "num_tokens": 50894192.0,
      "step": 27295
    },
    {
      "entropy": 6.0650848865509035,
      "epoch": 2.4108089014482514,
      "grad_norm": 2.171875,
      "learning_rate": 0.00044173475350294623,
      "loss": 5.6488,
      "mean_token_accuracy": 0.179318605363369,
      "num_tokens": 50902458.0,
      "step": 27300
    },
    {
      "entropy": 6.019539833068848,
      "epoch": 2.411250441540092,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044171360667173633,
      "loss": 5.662,
      "mean_token_accuracy": 0.1816403478384018,
      "num_tokens": 50912038.0,
      "step": 27305
    },
    {
      "entropy": 6.013415431976318,
      "epoch": 2.411691981631932,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004416924565747016,
      "loss": 5.7739,
      "mean_token_accuracy": 0.1675596281886101,
      "num_tokens": 50921164.0,
      "step": 27310
    },
    {
      "entropy": 6.022168922424316,
      "epoch": 2.4121335217237725,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044167130321225644,
      "loss": 5.6826,
      "mean_token_accuracy": 0.17629795670509338,
      "num_tokens": 50930359.0,
      "step": 27315
    },
    {
      "entropy": 5.991133832931519,
      "epoch": 2.412575061815613,
      "grad_norm": 2.984375,
      "learning_rate": 0.00044165014658481533,
      "loss": 5.7065,
      "mean_token_accuracy": 0.17644907683134078,
      "num_tokens": 50940675.0,
      "step": 27320
    },
    {
      "entropy": 6.044725370407105,
      "epoch": 2.413016601907453,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004416289866927926,
      "loss": 5.7962,
      "mean_token_accuracy": 0.17021893709897995,
      "num_tokens": 50949402.0,
      "step": 27325
    },
    {
      "entropy": 6.012938404083252,
      "epoch": 2.4134581419992935,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004416078235366027,
      "loss": 5.6614,
      "mean_token_accuracy": 0.17948836982250213,
      "num_tokens": 50958348.0,
      "step": 27330
    },
    {
      "entropy": 6.0405426025390625,
      "epoch": 2.413899682091134,
      "grad_norm": 1.734375,
      "learning_rate": 0.00044158665711666034,
      "loss": 5.6603,
      "mean_token_accuracy": 0.18076151907444,
      "num_tokens": 50967348.0,
      "step": 27335
    },
    {
      "entropy": 6.046257877349854,
      "epoch": 2.414341222182974,
      "grad_norm": 1.75,
      "learning_rate": 0.00044156548743338016,
      "loss": 5.7021,
      "mean_token_accuracy": 0.18524545282125474,
      "num_tokens": 50975882.0,
      "step": 27340
    },
    {
      "entropy": 6.056058597564697,
      "epoch": 2.4147827622748146,
      "grad_norm": 2.0,
      "learning_rate": 0.00044154431448717674,
      "loss": 5.6632,
      "mean_token_accuracy": 0.1808573544025421,
      "num_tokens": 50985241.0,
      "step": 27345
    },
    {
      "entropy": 6.014345645904541,
      "epoch": 2.415224302366655,
      "grad_norm": 3.359375,
      "learning_rate": 0.00044152313827846494,
      "loss": 5.6844,
      "mean_token_accuracy": 0.18167378306388854,
      "num_tokens": 50994834.0,
      "step": 27350
    },
    {
      "entropy": 5.994855976104736,
      "epoch": 2.415665842458495,
      "grad_norm": 1.8125,
      "learning_rate": 0.00044150195880765964,
      "loss": 5.6684,
      "mean_token_accuracy": 0.18250108063220977,
      "num_tokens": 51003919.0,
      "step": 27355
    },
    {
      "entropy": 6.048073244094849,
      "epoch": 2.4161073825503356,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00044148077607517554,
      "loss": 5.6465,
      "mean_token_accuracy": 0.17823680937290193,
      "num_tokens": 51012868.0,
      "step": 27360
    },
    {
      "entropy": 6.037415885925293,
      "epoch": 2.4165489226421757,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00044145959008142784,
      "loss": 5.6958,
      "mean_token_accuracy": 0.18039046376943588,
      "num_tokens": 51021693.0,
      "step": 27365
    },
    {
      "entropy": 6.051009941101074,
      "epoch": 2.416990462734016,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004414384008268314,
      "loss": 5.6928,
      "mean_token_accuracy": 0.17742778211832047,
      "num_tokens": 51031100.0,
      "step": 27370
    },
    {
      "entropy": 5.991608142852783,
      "epoch": 2.4174320028258567,
      "grad_norm": 1.828125,
      "learning_rate": 0.00044141720831180134,
      "loss": 5.7389,
      "mean_token_accuracy": 0.17887755781412124,
      "num_tokens": 51041077.0,
      "step": 27375
    },
    {
      "entropy": 5.977350473403931,
      "epoch": 2.4178735429176967,
      "grad_norm": 2.875,
      "learning_rate": 0.0004413960125367527,
      "loss": 5.6502,
      "mean_token_accuracy": 0.18177531510591508,
      "num_tokens": 51050403.0,
      "step": 27380
    },
    {
      "entropy": 5.976207590103149,
      "epoch": 2.4183150830095372,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044137481350210093,
      "loss": 5.6786,
      "mean_token_accuracy": 0.1779630795121193,
      "num_tokens": 51060119.0,
      "step": 27385
    },
    {
      "entropy": 6.0355541706085205,
      "epoch": 2.4187566231013777,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00044135361120826116,
      "loss": 5.7663,
      "mean_token_accuracy": 0.17036932557821274,
      "num_tokens": 51070355.0,
      "step": 27390
    },
    {
      "entropy": 6.013363265991211,
      "epoch": 2.419198163193218,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004413324056556488,
      "loss": 5.6182,
      "mean_token_accuracy": 0.18682178407907485,
      "num_tokens": 51079386.0,
      "step": 27395
    },
    {
      "entropy": 6.041614675521851,
      "epoch": 2.4196397032850583,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004413111968446792,
      "loss": 5.7698,
      "mean_token_accuracy": 0.17045081555843353,
      "num_tokens": 51089234.0,
      "step": 27400
    },
    {
      "entropy": 6.0915132522583,
      "epoch": 2.420081243376899,
      "grad_norm": 8.0625,
      "learning_rate": 0.0004412899847757678,
      "loss": 5.6943,
      "mean_token_accuracy": 0.17894820123910904,
      "num_tokens": 51099062.0,
      "step": 27405
    },
    {
      "entropy": 6.0592084407806395,
      "epoch": 2.420522783468739,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004412687694493302,
      "loss": 5.6432,
      "mean_token_accuracy": 0.17716228365898132,
      "num_tokens": 51108278.0,
      "step": 27410
    },
    {
      "entropy": 6.045741367340088,
      "epoch": 2.4209643235605793,
      "grad_norm": 1.46875,
      "learning_rate": 0.000441247550865782,
      "loss": 5.7407,
      "mean_token_accuracy": 0.17982436418533326,
      "num_tokens": 51117378.0,
      "step": 27415
    },
    {
      "entropy": 5.985430574417114,
      "epoch": 2.42140586365242,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044122632902553877,
      "loss": 5.6656,
      "mean_token_accuracy": 0.1841372400522232,
      "num_tokens": 51126323.0,
      "step": 27420
    },
    {
      "entropy": 6.010504531860351,
      "epoch": 2.42184740374426,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004412051039290163,
      "loss": 5.6819,
      "mean_token_accuracy": 0.1864933654665947,
      "num_tokens": 51135734.0,
      "step": 27425
    },
    {
      "entropy": 5.8645295143127445,
      "epoch": 2.4222889438361004,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00044118387557663034,
      "loss": 5.5781,
      "mean_token_accuracy": 0.19034167528152465,
      "num_tokens": 51145020.0,
      "step": 27430
    },
    {
      "entropy": 6.0371440887451175,
      "epoch": 2.422730483927941,
      "grad_norm": 3.34375,
      "learning_rate": 0.00044116264396879684,
      "loss": 5.7044,
      "mean_token_accuracy": 0.17526839673519135,
      "num_tokens": 51153908.0,
      "step": 27435
    },
    {
      "entropy": 6.0168369770050045,
      "epoch": 2.423172024019781,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004411414091059316,
      "loss": 5.6288,
      "mean_token_accuracy": 0.18286663740873338,
      "num_tokens": 51163547.0,
      "step": 27440
    },
    {
      "entropy": 6.061150026321411,
      "epoch": 2.4236135641116214,
      "grad_norm": 1.921875,
      "learning_rate": 0.00044112017098845064,
      "loss": 5.7689,
      "mean_token_accuracy": 0.17154446989297867,
      "num_tokens": 51172559.0,
      "step": 27445
    },
    {
      "entropy": 5.996813774108887,
      "epoch": 2.4240551042034615,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004410989296167701,
      "loss": 5.6254,
      "mean_token_accuracy": 0.1879034534096718,
      "num_tokens": 51181455.0,
      "step": 27450
    },
    {
      "entropy": 6.01319580078125,
      "epoch": 2.424496644295302,
      "grad_norm": 2.09375,
      "learning_rate": 0.00044107768499130587,
      "loss": 5.7288,
      "mean_token_accuracy": 0.17621323168277742,
      "num_tokens": 51191149.0,
      "step": 27455
    },
    {
      "entropy": 5.979516983032227,
      "epoch": 2.4249381843871425,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004410564371124743,
      "loss": 5.6436,
      "mean_token_accuracy": 0.18926046192646026,
      "num_tokens": 51200702.0,
      "step": 27460
    },
    {
      "entropy": 6.032996129989624,
      "epoch": 2.4253797244789825,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044103518598069156,
      "loss": 5.7066,
      "mean_token_accuracy": 0.17337646782398225,
      "num_tokens": 51209058.0,
      "step": 27465
    },
    {
      "entropy": 6.0600262641906735,
      "epoch": 2.425821264570823,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00044101393159637396,
      "loss": 5.7153,
      "mean_token_accuracy": 0.18146573901176452,
      "num_tokens": 51219295.0,
      "step": 27470
    },
    {
      "entropy": 5.945971870422364,
      "epoch": 2.4262628046626635,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044099267395993785,
      "loss": 5.6198,
      "mean_token_accuracy": 0.1809125065803528,
      "num_tokens": 51228430.0,
      "step": 27475
    },
    {
      "entropy": 5.9368500232696535,
      "epoch": 2.4267043447545036,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004409714130717997,
      "loss": 5.6146,
      "mean_token_accuracy": 0.18475380390882493,
      "num_tokens": 51237804.0,
      "step": 27480
    },
    {
      "entropy": 6.006909132003784,
      "epoch": 2.427145884846344,
      "grad_norm": 1.5625,
      "learning_rate": 0.000440950148932376,
      "loss": 5.6624,
      "mean_token_accuracy": 0.18511331528425218,
      "num_tokens": 51247313.0,
      "step": 27485
    },
    {
      "entropy": 5.982644557952881,
      "epoch": 2.4275874249381846,
      "grad_norm": 1.5625,
      "learning_rate": 0.00044092888154208314,
      "loss": 5.6976,
      "mean_token_accuracy": 0.1787732347846031,
      "num_tokens": 51257310.0,
      "step": 27490
    },
    {
      "entropy": 6.001168918609619,
      "epoch": 2.4280289650300246,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004409076109013379,
      "loss": 5.6688,
      "mean_token_accuracy": 0.18232283145189285,
      "num_tokens": 51267028.0,
      "step": 27495
    },
    {
      "entropy": 6.048819303512573,
      "epoch": 2.428470505121865,
      "grad_norm": 2.046875,
      "learning_rate": 0.000440886337010557,
      "loss": 5.6418,
      "mean_token_accuracy": 0.18200089931488037,
      "num_tokens": 51276756.0,
      "step": 27500
    },
    {
      "entropy": 6.0154965877532955,
      "epoch": 2.428912045213705,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000440865059870157,
      "loss": 5.7085,
      "mean_token_accuracy": 0.1747611716389656,
      "num_tokens": 51286491.0,
      "step": 27505
    },
    {
      "entropy": 6.073829555511475,
      "epoch": 2.4293535853055457,
      "grad_norm": 3.609375,
      "learning_rate": 0.00044084377948055487,
      "loss": 5.7534,
      "mean_token_accuracy": 0.17314067780971526,
      "num_tokens": 51295221.0,
      "step": 27510
    },
    {
      "entropy": 6.032900953292847,
      "epoch": 2.429795125397386,
      "grad_norm": 1.4375,
      "learning_rate": 0.00044082249584216735,
      "loss": 5.7144,
      "mean_token_accuracy": 0.18261484801769257,
      "num_tokens": 51304637.0,
      "step": 27515
    },
    {
      "entropy": 5.894020366668701,
      "epoch": 2.4302366654892262,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004408012089554115,
      "loss": 5.6029,
      "mean_token_accuracy": 0.1871160238981247,
      "num_tokens": 51314107.0,
      "step": 27520
    },
    {
      "entropy": 6.048101615905762,
      "epoch": 2.4306782055810667,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00044077991882070423,
      "loss": 5.6294,
      "mean_token_accuracy": 0.1871064692735672,
      "num_tokens": 51323353.0,
      "step": 27525
    },
    {
      "entropy": 6.014856147766113,
      "epoch": 2.4311197456729072,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004407586254384627,
      "loss": 5.6412,
      "mean_token_accuracy": 0.1815165564417839,
      "num_tokens": 51333771.0,
      "step": 27530
    },
    {
      "entropy": 6.0330699443817135,
      "epoch": 2.4315612857647473,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004407373288091039,
      "loss": 5.6869,
      "mean_token_accuracy": 0.1740817353129387,
      "num_tokens": 51343476.0,
      "step": 27535
    },
    {
      "entropy": 6.024708843231201,
      "epoch": 2.432002825856588,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004407160289330452,
      "loss": 5.6371,
      "mean_token_accuracy": 0.18633682876825333,
      "num_tokens": 51352903.0,
      "step": 27540
    },
    {
      "entropy": 6.00507082939148,
      "epoch": 2.4324443659484283,
      "grad_norm": 4.25,
      "learning_rate": 0.00044069472581070363,
      "loss": 5.689,
      "mean_token_accuracy": 0.18009292632341384,
      "num_tokens": 51362922.0,
      "step": 27545
    },
    {
      "entropy": 6.013424348831177,
      "epoch": 2.4328859060402683,
      "grad_norm": 2.015625,
      "learning_rate": 0.00044067341944249667,
      "loss": 5.7114,
      "mean_token_accuracy": 0.1760515257716179,
      "num_tokens": 51371789.0,
      "step": 27550
    },
    {
      "entropy": 6.104208755493164,
      "epoch": 2.433327446132109,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044065210982884164,
      "loss": 5.7415,
      "mean_token_accuracy": 0.17633882761001587,
      "num_tokens": 51380989.0,
      "step": 27555
    },
    {
      "entropy": 6.011701011657715,
      "epoch": 2.4337689862239493,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004406307969701559,
      "loss": 5.7219,
      "mean_token_accuracy": 0.17904194295406342,
      "num_tokens": 51390191.0,
      "step": 27560
    },
    {
      "entropy": 6.017073392868042,
      "epoch": 2.4342105263157894,
      "grad_norm": 1.71875,
      "learning_rate": 0.00044060948086685714,
      "loss": 5.7215,
      "mean_token_accuracy": 0.1722428247332573,
      "num_tokens": 51399133.0,
      "step": 27565
    },
    {
      "entropy": 5.988965845108032,
      "epoch": 2.43465206640763,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044058816151936283,
      "loss": 5.5956,
      "mean_token_accuracy": 0.18592584431171416,
      "num_tokens": 51408473.0,
      "step": 27570
    },
    {
      "entropy": 5.958673048019409,
      "epoch": 2.4350936064994704,
      "grad_norm": 1.5,
      "learning_rate": 0.00044056683892809055,
      "loss": 5.5841,
      "mean_token_accuracy": 0.19225268959999084,
      "num_tokens": 51416944.0,
      "step": 27575
    },
    {
      "entropy": 5.954107999801636,
      "epoch": 2.4355351465913104,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044054551309345816,
      "loss": 5.8003,
      "mean_token_accuracy": 0.1759103536605835,
      "num_tokens": 51426193.0,
      "step": 27580
    },
    {
      "entropy": 6.058848857879639,
      "epoch": 2.435976686683151,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004405241840158833,
      "loss": 5.7367,
      "mean_token_accuracy": 0.18485459238290786,
      "num_tokens": 51435718.0,
      "step": 27585
    },
    {
      "entropy": 6.020826959609986,
      "epoch": 2.436418226774991,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004405028516957837,
      "loss": 5.6316,
      "mean_token_accuracy": 0.18230602741241456,
      "num_tokens": 51445127.0,
      "step": 27590
    },
    {
      "entropy": 6.084327554702758,
      "epoch": 2.4368597668668315,
      "grad_norm": 2.5625,
      "learning_rate": 0.00044048151613357745,
      "loss": 5.7485,
      "mean_token_accuracy": 0.17602858543395997,
      "num_tokens": 51454813.0,
      "step": 27595
    },
    {
      "entropy": 6.053457927703858,
      "epoch": 2.437301306958672,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004404601773296824,
      "loss": 5.6779,
      "mean_token_accuracy": 0.18343334943056105,
      "num_tokens": 51464454.0,
      "step": 27600
    },
    {
      "entropy": 6.019262981414795,
      "epoch": 2.437742847050512,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004404388352845166,
      "loss": 5.7171,
      "mean_token_accuracy": 0.16698646396398545,
      "num_tokens": 51473535.0,
      "step": 27605
    },
    {
      "entropy": 6.077666234970093,
      "epoch": 2.4381843871423525,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00044041748999849807,
      "loss": 5.6675,
      "mean_token_accuracy": 0.17623393535614013,
      "num_tokens": 51482490.0,
      "step": 27610
    },
    {
      "entropy": 6.009787273406983,
      "epoch": 2.438625927234193,
      "grad_norm": 1.84375,
      "learning_rate": 0.000440396141472045,
      "loss": 5.7071,
      "mean_token_accuracy": 0.18481629341840744,
      "num_tokens": 51493604.0,
      "step": 27615
    },
    {
      "entropy": 5.981246614456177,
      "epoch": 2.439067467326033,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004403747897055756,
      "loss": 5.6896,
      "mean_token_accuracy": 0.1831749141216278,
      "num_tokens": 51502980.0,
      "step": 27620
    },
    {
      "entropy": 5.99896936416626,
      "epoch": 2.4395090074178736,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004403534346995081,
      "loss": 5.6793,
      "mean_token_accuracy": 0.18574691712856292,
      "num_tokens": 51512131.0,
      "step": 27625
    },
    {
      "entropy": 6.0507410049438475,
      "epoch": 2.439950547509714,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044033207645426093,
      "loss": 5.7417,
      "mean_token_accuracy": 0.17231928557157516,
      "num_tokens": 51521600.0,
      "step": 27630
    },
    {
      "entropy": 5.994101238250733,
      "epoch": 2.440392087601554,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004403107149702524,
      "loss": 5.5893,
      "mean_token_accuracy": 0.19420057833194732,
      "num_tokens": 51531123.0,
      "step": 27635
    },
    {
      "entropy": 6.029895496368408,
      "epoch": 2.4408336276933946,
      "grad_norm": 1.5,
      "learning_rate": 0.0004402893502479009,
      "loss": 5.6448,
      "mean_token_accuracy": 0.1864178866147995,
      "num_tokens": 51540395.0,
      "step": 27640
    },
    {
      "entropy": 5.940986585617066,
      "epoch": 2.4412751677852347,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000440267982287625,
      "loss": 5.642,
      "mean_token_accuracy": 0.18068654239177703,
      "num_tokens": 51549683.0,
      "step": 27645
    },
    {
      "entropy": 6.09458966255188,
      "epoch": 2.441716707877075,
      "grad_norm": 2.0,
      "learning_rate": 0.0004402466110898435,
      "loss": 5.8335,
      "mean_token_accuracy": 0.16762491017580033,
      "num_tokens": 51559800.0,
      "step": 27650
    },
    {
      "entropy": 6.058006525039673,
      "epoch": 2.4421582479689157,
      "grad_norm": 2.453125,
      "learning_rate": 0.00044022523665497473,
      "loss": 5.6689,
      "mean_token_accuracy": 0.1812430053949356,
      "num_tokens": 51569788.0,
      "step": 27655
    },
    {
      "entropy": 6.024569988250732,
      "epoch": 2.4425997880607557,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004402038589834375,
      "loss": 5.6993,
      "mean_token_accuracy": 0.17760098427534105,
      "num_tokens": 51578598.0,
      "step": 27660
    },
    {
      "entropy": 6.074063396453857,
      "epoch": 2.4430413281525962,
      "grad_norm": 1.75,
      "learning_rate": 0.0004401824780756507,
      "loss": 5.7883,
      "mean_token_accuracy": 0.1690832145512104,
      "num_tokens": 51589203.0,
      "step": 27665
    },
    {
      "entropy": 6.035926008224488,
      "epoch": 2.4434828682444367,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004401610939320331,
      "loss": 5.6681,
      "mean_token_accuracy": 0.18196702152490615,
      "num_tokens": 51597813.0,
      "step": 27670
    },
    {
      "entropy": 6.039413642883301,
      "epoch": 2.4439244083362768,
      "grad_norm": 2.0,
      "learning_rate": 0.00044013970655300364,
      "loss": 5.7417,
      "mean_token_accuracy": 0.17807915657758713,
      "num_tokens": 51607386.0,
      "step": 27675
    },
    {
      "entropy": 6.079324674606323,
      "epoch": 2.4443659484281173,
      "grad_norm": 1.875,
      "learning_rate": 0.00044011831593898117,
      "loss": 5.7643,
      "mean_token_accuracy": 0.17291515022516252,
      "num_tokens": 51617091.0,
      "step": 27680
    },
    {
      "entropy": 6.03662075996399,
      "epoch": 2.4448074885199578,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044009692209038485,
      "loss": 5.759,
      "mean_token_accuracy": 0.1782648429274559,
      "num_tokens": 51625616.0,
      "step": 27685
    },
    {
      "entropy": 6.000855827331543,
      "epoch": 2.445249028611798,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004400755250076337,
      "loss": 5.5801,
      "mean_token_accuracy": 0.19268137067556382,
      "num_tokens": 51634693.0,
      "step": 27690
    },
    {
      "entropy": 6.005915307998658,
      "epoch": 2.4456905687036383,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00044005412469114684,
      "loss": 5.5852,
      "mean_token_accuracy": 0.1964389279484749,
      "num_tokens": 51642549.0,
      "step": 27695
    },
    {
      "entropy": 5.954552841186524,
      "epoch": 2.446132108795479,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044003272114134367,
      "loss": 5.6404,
      "mean_token_accuracy": 0.18122755885124206,
      "num_tokens": 51651742.0,
      "step": 27700
    },
    {
      "entropy": 6.022280168533325,
      "epoch": 2.446573648887319,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004400113143586433,
      "loss": 5.7105,
      "mean_token_accuracy": 0.18515589982271194,
      "num_tokens": 51660929.0,
      "step": 27705
    },
    {
      "entropy": 6.101159143447876,
      "epoch": 2.4470151889791594,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043998990434346514,
      "loss": 5.7228,
      "mean_token_accuracy": 0.1754469782114029,
      "num_tokens": 51670580.0,
      "step": 27710
    },
    {
      "entropy": 6.033937597274781,
      "epoch": 2.447456729071,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004399684910962286,
      "loss": 5.6951,
      "mean_token_accuracy": 0.18234473913908006,
      "num_tokens": 51680184.0,
      "step": 27715
    },
    {
      "entropy": 5.971181631088257,
      "epoch": 2.44789826916284,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004399470746173531,
      "loss": 5.6901,
      "mean_token_accuracy": 0.17557148933410643,
      "num_tokens": 51690277.0,
      "step": 27720
    },
    {
      "entropy": 6.009972286224365,
      "epoch": 2.4483398092546804,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004399256549072583,
      "loss": 5.6613,
      "mean_token_accuracy": 0.1875368356704712,
      "num_tokens": 51699316.0,
      "step": 27725
    },
    {
      "entropy": 5.971741247177124,
      "epoch": 2.4487813493465205,
      "grad_norm": 1.484375,
      "learning_rate": 0.00043990423196636357,
      "loss": 5.6726,
      "mean_token_accuracy": 0.18746046125888824,
      "num_tokens": 51708552.0,
      "step": 27730
    },
    {
      "entropy": 6.0631028175354,
      "epoch": 2.449222889438361,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004398828057950888,
      "loss": 5.6597,
      "mean_token_accuracy": 0.17921168208122254,
      "num_tokens": 51717082.0,
      "step": 27735
    },
    {
      "entropy": 5.9979578971862795,
      "epoch": 2.4496644295302015,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004398613763938537,
      "loss": 5.7181,
      "mean_token_accuracy": 0.18242692649364473,
      "num_tokens": 51727059.0,
      "step": 27740
    },
    {
      "entropy": 5.968033409118652,
      "epoch": 2.4501059696220415,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043983994376307797,
      "loss": 5.6057,
      "mean_token_accuracy": 0.19051026701927185,
      "num_tokens": 51735889.0,
      "step": 27745
    },
    {
      "entropy": 6.061326408386231,
      "epoch": 2.450547509713882,
      "grad_norm": 1.375,
      "learning_rate": 0.00043981850790318145,
      "loss": 5.7611,
      "mean_token_accuracy": 0.17450565546751023,
      "num_tokens": 51744856.0,
      "step": 27750
    },
    {
      "entropy": 5.960956144332886,
      "epoch": 2.4509890498057225,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043979706881458415,
      "loss": 5.6437,
      "mean_token_accuracy": 0.18993865251541137,
      "num_tokens": 51753552.0,
      "step": 27755
    },
    {
      "entropy": 6.050931787490844,
      "epoch": 2.4514305898975626,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00043977562649770587,
      "loss": 5.7608,
      "mean_token_accuracy": 0.1734069287776947,
      "num_tokens": 51762929.0,
      "step": 27760
    },
    {
      "entropy": 6.132037496566772,
      "epoch": 2.451872129989403,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043975418095296696,
      "loss": 5.7989,
      "mean_token_accuracy": 0.1748652935028076,
      "num_tokens": 51772732.0,
      "step": 27765
    },
    {
      "entropy": 6.032432985305786,
      "epoch": 2.4523136700812436,
      "grad_norm": 1.671875,
      "learning_rate": 0.00043973273218078716,
      "loss": 5.7841,
      "mean_token_accuracy": 0.17532563656568528,
      "num_tokens": 51782304.0,
      "step": 27770
    },
    {
      "entropy": 5.99051308631897,
      "epoch": 2.4527552101730836,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00043971128018158693,
      "loss": 5.607,
      "mean_token_accuracy": 0.18618471920490265,
      "num_tokens": 51791492.0,
      "step": 27775
    },
    {
      "entropy": 6.0140709400177,
      "epoch": 2.453196750264924,
      "grad_norm": 1.5,
      "learning_rate": 0.0004396898249557864,
      "loss": 5.724,
      "mean_token_accuracy": 0.18036556988954544,
      "num_tokens": 51800906.0,
      "step": 27780
    },
    {
      "entropy": 5.970389318466187,
      "epoch": 2.453638290356764,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043966836650380587,
      "loss": 5.5692,
      "mean_token_accuracy": 0.1917752206325531,
      "num_tokens": 51809071.0,
      "step": 27785
    },
    {
      "entropy": 6.03619909286499,
      "epoch": 2.4540798304486047,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043964690482606566,
      "loss": 5.6621,
      "mean_token_accuracy": 0.18502136468887329,
      "num_tokens": 51817450.0,
      "step": 27790
    },
    {
      "entropy": 6.0069481372833256,
      "epoch": 2.454521370540445,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043962543992298624,
      "loss": 5.6844,
      "mean_token_accuracy": 0.18013445436954498,
      "num_tokens": 51826400.0,
      "step": 27795
    },
    {
      "entropy": 6.0172319412231445,
      "epoch": 2.4549629106322852,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004396039717949881,
      "loss": 5.6799,
      "mean_token_accuracy": 0.18197934776544572,
      "num_tokens": 51835415.0,
      "step": 27800
    },
    {
      "entropy": 6.035876798629761,
      "epoch": 2.4554044507241257,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004395825004424917,
      "loss": 5.7226,
      "mean_token_accuracy": 0.1870769292116165,
      "num_tokens": 51844914.0,
      "step": 27805
    },
    {
      "entropy": 5.978488254547119,
      "epoch": 2.455845990815966,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004395610258659178,
      "loss": 5.6832,
      "mean_token_accuracy": 0.1799481377005577,
      "num_tokens": 51854178.0,
      "step": 27810
    },
    {
      "entropy": 6.0703939437866214,
      "epoch": 2.4562875309078063,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043953954806568693,
      "loss": 5.7926,
      "mean_token_accuracy": 0.17147888690233232,
      "num_tokens": 51863184.0,
      "step": 27815
    },
    {
      "entropy": 6.0169415950775145,
      "epoch": 2.4567290709996468,
      "grad_norm": 2.03125,
      "learning_rate": 0.00043951806704221994,
      "loss": 5.7265,
      "mean_token_accuracy": 0.18409039825201035,
      "num_tokens": 51872651.0,
      "step": 27820
    },
    {
      "entropy": 5.90775237083435,
      "epoch": 2.4571706110914873,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004394965827959376,
      "loss": 5.6549,
      "mean_token_accuracy": 0.1819853812456131,
      "num_tokens": 51881527.0,
      "step": 27825
    },
    {
      "entropy": 6.029032230377197,
      "epoch": 2.4576121511833273,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043947509532726067,
      "loss": 5.7656,
      "mean_token_accuracy": 0.17554882913827896,
      "num_tokens": 51892083.0,
      "step": 27830
    },
    {
      "entropy": 6.024727869033813,
      "epoch": 2.458053691275168,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004394536046366102,
      "loss": 5.662,
      "mean_token_accuracy": 0.1802817016839981,
      "num_tokens": 51900758.0,
      "step": 27835
    },
    {
      "entropy": 6.06628646850586,
      "epoch": 2.4584952313670083,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004394321107244072,
      "loss": 5.7407,
      "mean_token_accuracy": 0.18045576959848403,
      "num_tokens": 51910624.0,
      "step": 27840
    },
    {
      "entropy": 5.826616096496582,
      "epoch": 2.4589367714588484,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004394106135910726,
      "loss": 5.4731,
      "mean_token_accuracy": 0.18951869904994964,
      "num_tokens": 51919306.0,
      "step": 27845
    },
    {
      "entropy": 6.03258056640625,
      "epoch": 2.459378311550689,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00043938911323702766,
      "loss": 5.7134,
      "mean_token_accuracy": 0.17827011346817018,
      "num_tokens": 51928650.0,
      "step": 27850
    },
    {
      "entropy": 6.032146310806274,
      "epoch": 2.4598198516425294,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004393676096626934,
      "loss": 5.7591,
      "mean_token_accuracy": 0.1722305580973625,
      "num_tokens": 51939198.0,
      "step": 27855
    },
    {
      "entropy": 5.85827898979187,
      "epoch": 2.4602613917343694,
      "grad_norm": 1.53125,
      "learning_rate": 0.00043934610286849114,
      "loss": 5.4399,
      "mean_token_accuracy": 0.20920738130807875,
      "num_tokens": 51948641.0,
      "step": 27860
    },
    {
      "entropy": 5.991836833953857,
      "epoch": 2.46070293182621,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004393245928548423,
      "loss": 5.6589,
      "mean_token_accuracy": 0.18816886693239213,
      "num_tokens": 51958258.0,
      "step": 27865
    },
    {
      "entropy": 5.9744655132293705,
      "epoch": 2.46114447191805,
      "grad_norm": 1.53125,
      "learning_rate": 0.00043930307962216806,
      "loss": 5.6295,
      "mean_token_accuracy": 0.1810983747243881,
      "num_tokens": 51967295.0,
      "step": 27870
    },
    {
      "entropy": 5.825023221969604,
      "epoch": 2.4615860120098905,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043928156317088994,
      "loss": 5.5788,
      "mean_token_accuracy": 0.18742819130420685,
      "num_tokens": 51976450.0,
      "step": 27875
    },
    {
      "entropy": 6.034429693222046,
      "epoch": 2.462027552101731,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004392600435014295,
      "loss": 5.6622,
      "mean_token_accuracy": 0.18059876710176467,
      "num_tokens": 51985782.0,
      "step": 27880
    },
    {
      "entropy": 6.1168581485748295,
      "epoch": 2.462469092193571,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043923852061420815,
      "loss": 5.7406,
      "mean_token_accuracy": 0.17970819920301437,
      "num_tokens": 51994745.0,
      "step": 27885
    },
    {
      "entropy": 6.056033420562744,
      "epoch": 2.4629106322854115,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004392169945096476,
      "loss": 5.7437,
      "mean_token_accuracy": 0.17689338326454163,
      "num_tokens": 52004369.0,
      "step": 27890
    },
    {
      "entropy": 5.964374113082886,
      "epoch": 2.463352172377252,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043919546518816953,
      "loss": 5.6972,
      "mean_token_accuracy": 0.17950916439294815,
      "num_tokens": 52012556.0,
      "step": 27895
    },
    {
      "entropy": 6.047852277755737,
      "epoch": 2.463793712469092,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004391739326501957,
      "loss": 5.7654,
      "mean_token_accuracy": 0.17130647897720336,
      "num_tokens": 52022854.0,
      "step": 27900
    },
    {
      "entropy": 6.092540597915649,
      "epoch": 2.4642352525609326,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004391523968961479,
      "loss": 5.6969,
      "mean_token_accuracy": 0.17704804241657257,
      "num_tokens": 52032924.0,
      "step": 27905
    },
    {
      "entropy": 6.009565210342407,
      "epoch": 2.464676792652773,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000439130857926448,
      "loss": 5.6668,
      "mean_token_accuracy": 0.17972676306962967,
      "num_tokens": 52042625.0,
      "step": 27910
    },
    {
      "entropy": 5.9539541721344,
      "epoch": 2.465118332744613,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004391093157415179,
      "loss": 5.6754,
      "mean_token_accuracy": 0.18312485218048097,
      "num_tokens": 52051964.0,
      "step": 27915
    },
    {
      "entropy": 5.972874164581299,
      "epoch": 2.4655598728364536,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004390877703417797,
      "loss": 5.6829,
      "mean_token_accuracy": 0.1860860124230385,
      "num_tokens": 52060877.0,
      "step": 27920
    },
    {
      "entropy": 6.057538175582886,
      "epoch": 2.4660014129282937,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00043906622172765533,
      "loss": 5.7863,
      "mean_token_accuracy": 0.17769789844751357,
      "num_tokens": 52069585.0,
      "step": 27925
    },
    {
      "entropy": 6.0815596103668215,
      "epoch": 2.466442953020134,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00043904466989956706,
      "loss": 5.7392,
      "mean_token_accuracy": 0.17975265085697173,
      "num_tokens": 52078435.0,
      "step": 27930
    },
    {
      "entropy": 6.077190828323364,
      "epoch": 2.4668844931119747,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00043902311485793694,
      "loss": 5.7214,
      "mean_token_accuracy": 0.1769708886742592,
      "num_tokens": 52088227.0,
      "step": 27935
    },
    {
      "entropy": 6.003769302368164,
      "epoch": 2.4673260332038147,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004390015566031873,
      "loss": 5.6945,
      "mean_token_accuracy": 0.17424077838659285,
      "num_tokens": 52098101.0,
      "step": 27940
    },
    {
      "entropy": 5.986034917831421,
      "epoch": 2.467767573295655,
      "grad_norm": 1.71875,
      "learning_rate": 0.00043897999513574045,
      "loss": 5.6325,
      "mean_token_accuracy": 0.1884421780705452,
      "num_tokens": 52107370.0,
      "step": 27945
    },
    {
      "entropy": 5.918498611450195,
      "epoch": 2.4682091133874957,
      "grad_norm": 1.53125,
      "learning_rate": 0.00043895843045601874,
      "loss": 5.6476,
      "mean_token_accuracy": 0.18534681499004363,
      "num_tokens": 52116808.0,
      "step": 27950
    },
    {
      "entropy": 6.02683801651001,
      "epoch": 2.4686506534793358,
      "grad_norm": 1.5625,
      "learning_rate": 0.00043893686256444463,
      "loss": 5.5941,
      "mean_token_accuracy": 0.18783729523420334,
      "num_tokens": 52125724.0,
      "step": 27955
    },
    {
      "entropy": 6.070071649551392,
      "epoch": 2.4690921935711763,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004389152914614407,
      "loss": 5.7618,
      "mean_token_accuracy": 0.1733543023467064,
      "num_tokens": 52135530.0,
      "step": 27960
    },
    {
      "entropy": 6.006943225860596,
      "epoch": 2.4695337336630168,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004388937171474293,
      "loss": 5.7022,
      "mean_token_accuracy": 0.17935902029275894,
      "num_tokens": 52144195.0,
      "step": 27965
    },
    {
      "entropy": 5.986231851577759,
      "epoch": 2.469975273754857,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00043887213962283335,
      "loss": 5.7506,
      "mean_token_accuracy": 0.1755823314189911,
      "num_tokens": 52153675.0,
      "step": 27970
    },
    {
      "entropy": 5.982085227966309,
      "epoch": 2.4704168138466973,
      "grad_norm": 2.5,
      "learning_rate": 0.00043885055888807516,
      "loss": 5.6915,
      "mean_token_accuracy": 0.18293244987726212,
      "num_tokens": 52163455.0,
      "step": 27975
    },
    {
      "entropy": 6.035530090332031,
      "epoch": 2.470858353938538,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004388289749435779,
      "loss": 5.6902,
      "mean_token_accuracy": 0.18066394180059434,
      "num_tokens": 52172054.0,
      "step": 27980
    },
    {
      "entropy": 6.04333724975586,
      "epoch": 2.471299894030378,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004388073877897641,
      "loss": 5.7765,
      "mean_token_accuracy": 0.1702347844839096,
      "num_tokens": 52181667.0,
      "step": 27985
    },
    {
      "entropy": 6.049409818649292,
      "epoch": 2.4717414341222184,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043878579742705675,
      "loss": 5.6612,
      "mean_token_accuracy": 0.17754931449890138,
      "num_tokens": 52190472.0,
      "step": 27990
    },
    {
      "entropy": 6.033915662765503,
      "epoch": 2.472182974214059,
      "grad_norm": 1.734375,
      "learning_rate": 0.00043876420385587886,
      "loss": 5.6029,
      "mean_token_accuracy": 0.19050088673830032,
      "num_tokens": 52199905.0,
      "step": 27995
    },
    {
      "entropy": 5.997323226928711,
      "epoch": 2.472624514305899,
      "grad_norm": 2.8125,
      "learning_rate": 0.00043874260707665327,
      "loss": 5.6994,
      "mean_token_accuracy": 0.17748301178216935,
      "num_tokens": 52209270.0,
      "step": 28000
    },
    {
      "entropy": 5.932402420043945,
      "epoch": 2.4730660543977394,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004387210070898032,
      "loss": 5.6934,
      "mean_token_accuracy": 0.17913610637187957,
      "num_tokens": 52218144.0,
      "step": 28005
    },
    {
      "entropy": 6.091369199752807,
      "epoch": 2.4735075944895795,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004386994038957516,
      "loss": 5.6581,
      "mean_token_accuracy": 0.18313466608524323,
      "num_tokens": 52227128.0,
      "step": 28010
    },
    {
      "entropy": 6.034490871429443,
      "epoch": 2.47394913458142,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004386777974949219,
      "loss": 5.6694,
      "mean_token_accuracy": 0.1803836390376091,
      "num_tokens": 52235823.0,
      "step": 28015
    },
    {
      "entropy": 5.995304107666016,
      "epoch": 2.4743906746732605,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004386561878877372,
      "loss": 5.8066,
      "mean_token_accuracy": 0.17541635930538177,
      "num_tokens": 52245510.0,
      "step": 28020
    },
    {
      "entropy": 6.05365309715271,
      "epoch": 2.4748322147651005,
      "grad_norm": 3.0625,
      "learning_rate": 0.00043863457507462076,
      "loss": 5.6592,
      "mean_token_accuracy": 0.18340984880924224,
      "num_tokens": 52254583.0,
      "step": 28025
    },
    {
      "entropy": 6.080927133560181,
      "epoch": 2.475273754856941,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004386129590559962,
      "loss": 5.6993,
      "mean_token_accuracy": 0.1799377754330635,
      "num_tokens": 52264778.0,
      "step": 28030
    },
    {
      "entropy": 6.011626291275024,
      "epoch": 2.4757152949487815,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004385913398322867,
      "loss": 5.7134,
      "mean_token_accuracy": 0.18047615140676498,
      "num_tokens": 52274959.0,
      "step": 28035
    },
    {
      "entropy": 6.032221937179566,
      "epoch": 2.4761568350406216,
      "grad_norm": 1.421875,
      "learning_rate": 0.000438569717403916,
      "loss": 5.663,
      "mean_token_accuracy": 0.17703675478696823,
      "num_tokens": 52284606.0,
      "step": 28040
    },
    {
      "entropy": 5.997014808654785,
      "epoch": 2.476598375132462,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004385480917713075,
      "loss": 5.7146,
      "mean_token_accuracy": 0.17445535063743592,
      "num_tokens": 52294571.0,
      "step": 28045
    },
    {
      "entropy": 6.023624753952026,
      "epoch": 2.4770399152243026,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004385264629348849,
      "loss": 5.6691,
      "mean_token_accuracy": 0.18106569051742555,
      "num_tokens": 52303575.0,
      "step": 28050
    },
    {
      "entropy": 5.936225366592407,
      "epoch": 2.4774814553161426,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004385048308950719,
      "loss": 5.6466,
      "mean_token_accuracy": 0.18381939083337784,
      "num_tokens": 52312863.0,
      "step": 28055
    },
    {
      "entropy": 6.036515474319458,
      "epoch": 2.477922995407983,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004384831956522922,
      "loss": 5.7043,
      "mean_token_accuracy": 0.1802392363548279,
      "num_tokens": 52322493.0,
      "step": 28060
    },
    {
      "entropy": 6.009575366973877,
      "epoch": 2.478364535499823,
      "grad_norm": 1.390625,
      "learning_rate": 0.00043846155720696976,
      "loss": 5.6172,
      "mean_token_accuracy": 0.19134356528520585,
      "num_tokens": 52331159.0,
      "step": 28065
    },
    {
      "entropy": 5.993453884124756,
      "epoch": 2.4788060755916637,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043843991555952847,
      "loss": 5.6582,
      "mean_token_accuracy": 0.17914120703935624,
      "num_tokens": 52340276.0,
      "step": 28070
    },
    {
      "entropy": 5.9973915100097654,
      "epoch": 2.479247615683504,
      "grad_norm": 1.5,
      "learning_rate": 0.00043841827071039197,
      "loss": 5.6946,
      "mean_token_accuracy": 0.184036186337471,
      "num_tokens": 52349666.0,
      "step": 28075
    },
    {
      "entropy": 5.939842176437378,
      "epoch": 2.479689155775344,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043839662265998463,
      "loss": 5.6354,
      "mean_token_accuracy": 0.19466277360916137,
      "num_tokens": 52358238.0,
      "step": 28080
    },
    {
      "entropy": 5.976932001113892,
      "epoch": 2.4801306958671847,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043837497140873037,
      "loss": 5.6484,
      "mean_token_accuracy": 0.1879303425550461,
      "num_tokens": 52367663.0,
      "step": 28085
    },
    {
      "entropy": 5.998005533218384,
      "epoch": 2.480572235959025,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043835331695705326,
      "loss": 5.6715,
      "mean_token_accuracy": 0.1851142778992653,
      "num_tokens": 52377320.0,
      "step": 28090
    },
    {
      "entropy": 6.010440683364868,
      "epoch": 2.4810137760508653,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004383316593053777,
      "loss": 5.7054,
      "mean_token_accuracy": 0.17916995733976365,
      "num_tokens": 52386231.0,
      "step": 28095
    },
    {
      "entropy": 6.020002174377441,
      "epoch": 2.4814553161427058,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004383099984541277,
      "loss": 5.7541,
      "mean_token_accuracy": 0.17172861695289612,
      "num_tokens": 52395487.0,
      "step": 28100
    },
    {
      "entropy": 6.01707091331482,
      "epoch": 2.4818968562345463,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043828833440372775,
      "loss": 5.6785,
      "mean_token_accuracy": 0.17727337926626205,
      "num_tokens": 52404372.0,
      "step": 28105
    },
    {
      "entropy": 5.875301027297974,
      "epoch": 2.4823383963263863,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004382666671546022,
      "loss": 5.5293,
      "mean_token_accuracy": 0.1927002862095833,
      "num_tokens": 52413156.0,
      "step": 28110
    },
    {
      "entropy": 5.999047708511353,
      "epoch": 2.482779936418227,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043824499670717547,
      "loss": 5.6897,
      "mean_token_accuracy": 0.17450736314058304,
      "num_tokens": 52422782.0,
      "step": 28115
    },
    {
      "entropy": 6.0854192733764645,
      "epoch": 2.4832214765100673,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00043822332306187213,
      "loss": 5.7055,
      "mean_token_accuracy": 0.1743354544043541,
      "num_tokens": 52432363.0,
      "step": 28120
    },
    {
      "entropy": 5.990893363952637,
      "epoch": 2.4836630166019074,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004382016462191167,
      "loss": 5.6397,
      "mean_token_accuracy": 0.18156102895736695,
      "num_tokens": 52440995.0,
      "step": 28125
    },
    {
      "entropy": 5.972017526626587,
      "epoch": 2.484104556693748,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00043817996617933393,
      "loss": 5.6568,
      "mean_token_accuracy": 0.18004069179296495,
      "num_tokens": 52450135.0,
      "step": 28130
    },
    {
      "entropy": 6.022070693969726,
      "epoch": 2.4845460967855884,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004381582829429483,
      "loss": 5.6482,
      "mean_token_accuracy": 0.18566809296607972,
      "num_tokens": 52459065.0,
      "step": 28135
    },
    {
      "entropy": 5.954772138595581,
      "epoch": 2.4849876368774284,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004381365965103848,
      "loss": 5.6591,
      "mean_token_accuracy": 0.18655849993228912,
      "num_tokens": 52468382.0,
      "step": 28140
    },
    {
      "entropy": 5.944567489624023,
      "epoch": 2.485429176969269,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004381149068820681,
      "loss": 5.6605,
      "mean_token_accuracy": 0.1830165907740593,
      "num_tokens": 52476940.0,
      "step": 28145
    },
    {
      "entropy": 6.012415790557862,
      "epoch": 2.485870717061109,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00043809321405842314,
      "loss": 5.7266,
      "mean_token_accuracy": 0.17527499943971633,
      "num_tokens": 52486496.0,
      "step": 28150
    },
    {
      "entropy": 6.089297819137573,
      "epoch": 2.4863122571529495,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004380715180398749,
      "loss": 5.7293,
      "mean_token_accuracy": 0.1765411987900734,
      "num_tokens": 52495778.0,
      "step": 28155
    },
    {
      "entropy": 6.023750877380371,
      "epoch": 2.48675379724479,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004380498188268483,
      "loss": 5.5645,
      "mean_token_accuracy": 0.19164239317178727,
      "num_tokens": 52504855.0,
      "step": 28160
    },
    {
      "entropy": 5.902591133117676,
      "epoch": 2.48719533733663,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043802811641976855,
      "loss": 5.6124,
      "mean_token_accuracy": 0.17852247059345244,
      "num_tokens": 52513159.0,
      "step": 28165
    },
    {
      "entropy": 6.020082187652588,
      "epoch": 2.4876368774284705,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004380064108190607,
      "loss": 5.6901,
      "mean_token_accuracy": 0.17939042150974274,
      "num_tokens": 52521869.0,
      "step": 28170
    },
    {
      "entropy": 6.043376874923706,
      "epoch": 2.488078417520311,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004379847020251499,
      "loss": 5.766,
      "mean_token_accuracy": 0.18183800876140593,
      "num_tokens": 52531886.0,
      "step": 28175
    },
    {
      "entropy": 5.965773200988769,
      "epoch": 2.488519957612151,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00043796299003846155,
      "loss": 5.6285,
      "mean_token_accuracy": 0.1867879346013069,
      "num_tokens": 52540331.0,
      "step": 28180
    },
    {
      "entropy": 6.021798467636108,
      "epoch": 2.4889614977039916,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043794127485942094,
      "loss": 5.7684,
      "mean_token_accuracy": 0.17352186143398285,
      "num_tokens": 52549686.0,
      "step": 28185
    },
    {
      "entropy": 5.956199359893799,
      "epoch": 2.489403037795832,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004379195564884534,
      "loss": 5.6303,
      "mean_token_accuracy": 0.18834291398525238,
      "num_tokens": 52558722.0,
      "step": 28190
    },
    {
      "entropy": 6.03195219039917,
      "epoch": 2.489844577887672,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004378978349259844,
      "loss": 5.7331,
      "mean_token_accuracy": 0.17743021696805955,
      "num_tokens": 52568884.0,
      "step": 28195
    },
    {
      "entropy": 6.034687137603759,
      "epoch": 2.4902861179795126,
      "grad_norm": 1.390625,
      "learning_rate": 0.00043787611017243944,
      "loss": 5.6773,
      "mean_token_accuracy": 0.17937147468328477,
      "num_tokens": 52578519.0,
      "step": 28200
    },
    {
      "entropy": 5.973254537582397,
      "epoch": 2.4907276580713527,
      "grad_norm": 1.53125,
      "learning_rate": 0.00043785438222824406,
      "loss": 5.6,
      "mean_token_accuracy": 0.18943917155265808,
      "num_tokens": 52588496.0,
      "step": 28205
    },
    {
      "entropy": 5.984907341003418,
      "epoch": 2.491169198163193,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00043783265109382407,
      "loss": 5.7006,
      "mean_token_accuracy": 0.1770885095000267,
      "num_tokens": 52598463.0,
      "step": 28210
    },
    {
      "entropy": 5.895063257217407,
      "epoch": 2.4916107382550337,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043781091676960507,
      "loss": 5.5954,
      "mean_token_accuracy": 0.18750732988119126,
      "num_tokens": 52607257.0,
      "step": 28215
    },
    {
      "entropy": 6.06453161239624,
      "epoch": 2.4920522783468737,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004377891792560127,
      "loss": 5.6813,
      "mean_token_accuracy": 0.18437850922346116,
      "num_tokens": 52616434.0,
      "step": 28220
    },
    {
      "entropy": 6.025432586669922,
      "epoch": 2.492493818438714,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043776743855347287,
      "loss": 5.7818,
      "mean_token_accuracy": 0.1790674462914467,
      "num_tokens": 52625564.0,
      "step": 28225
    },
    {
      "entropy": 6.070011901855469,
      "epoch": 2.4929353585305547,
      "grad_norm": 2.046875,
      "learning_rate": 0.00043774569466241165,
      "loss": 5.8006,
      "mean_token_accuracy": 0.17404087036848068,
      "num_tokens": 52634081.0,
      "step": 28230
    },
    {
      "entropy": 5.9712881565094,
      "epoch": 2.4933768986223948,
      "grad_norm": 1.625,
      "learning_rate": 0.0004377239475832547,
      "loss": 5.6726,
      "mean_token_accuracy": 0.1881624460220337,
      "num_tokens": 52643722.0,
      "step": 28235
    },
    {
      "entropy": 6.029793977737427,
      "epoch": 2.4938184387142353,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004377021973164281,
      "loss": 5.6564,
      "mean_token_accuracy": 0.18697115927934646,
      "num_tokens": 52652831.0,
      "step": 28240
    },
    {
      "entropy": 6.119189262390137,
      "epoch": 2.4942599788060758,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004376804438623582,
      "loss": 5.8239,
      "mean_token_accuracy": 0.17056341022253035,
      "num_tokens": 52663139.0,
      "step": 28245
    },
    {
      "entropy": 6.047054672241211,
      "epoch": 2.494701518897916,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043765868722147076,
      "loss": 5.6646,
      "mean_token_accuracy": 0.18619194328784944,
      "num_tokens": 52672281.0,
      "step": 28250
    },
    {
      "entropy": 5.976760387420654,
      "epoch": 2.4951430589897563,
      "grad_norm": 1.796875,
      "learning_rate": 0.00043763692739419224,
      "loss": 5.6351,
      "mean_token_accuracy": 0.18199677318334578,
      "num_tokens": 52681644.0,
      "step": 28255
    },
    {
      "entropy": 6.06539363861084,
      "epoch": 2.495584599081597,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004376151643809487,
      "loss": 5.738,
      "mean_token_accuracy": 0.17487376779317856,
      "num_tokens": 52690706.0,
      "step": 28260
    },
    {
      "entropy": 6.072233819961548,
      "epoch": 2.496026139173437,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004375933981821667,
      "loss": 5.6955,
      "mean_token_accuracy": 0.17919690757989884,
      "num_tokens": 52700391.0,
      "step": 28265
    },
    {
      "entropy": 6.015890312194824,
      "epoch": 2.4964676792652774,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043757162879827245,
      "loss": 5.6652,
      "mean_token_accuracy": 0.18011298924684524,
      "num_tokens": 52709586.0,
      "step": 28270
    },
    {
      "entropy": 5.941529750823975,
      "epoch": 2.496909219357118,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004375498562296924,
      "loss": 5.6524,
      "mean_token_accuracy": 0.18512852638959884,
      "num_tokens": 52718926.0,
      "step": 28275
    },
    {
      "entropy": 6.045276308059693,
      "epoch": 2.497350759448958,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004375280804768532,
      "loss": 5.6121,
      "mean_token_accuracy": 0.1885722115635872,
      "num_tokens": 52728311.0,
      "step": 28280
    },
    {
      "entropy": 6.12320237159729,
      "epoch": 2.4977922995407984,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004375063015401813,
      "loss": 5.7117,
      "mean_token_accuracy": 0.17864732444286346,
      "num_tokens": 52737758.0,
      "step": 28285
    },
    {
      "entropy": 6.027470922470092,
      "epoch": 2.4982338396326385,
      "grad_norm": 1.703125,
      "learning_rate": 0.00043748451942010337,
      "loss": 5.743,
      "mean_token_accuracy": 0.18265011310577392,
      "num_tokens": 52746959.0,
      "step": 28290
    },
    {
      "entropy": 5.994355154037476,
      "epoch": 2.498675379724479,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004374627341170461,
      "loss": 5.7038,
      "mean_token_accuracy": 0.17584479749202728,
      "num_tokens": 52757315.0,
      "step": 28295
    },
    {
      "entropy": 6.029384279251099,
      "epoch": 2.4991169198163194,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004374409456314362,
      "loss": 5.7049,
      "mean_token_accuracy": 0.18482485711574553,
      "num_tokens": 52766927.0,
      "step": 28300
    },
    {
      "entropy": 5.9974202632904055,
      "epoch": 2.4995584599081595,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004374191539637007,
      "loss": 5.7086,
      "mean_token_accuracy": 0.1818643942475319,
      "num_tokens": 52775103.0,
      "step": 28305
    },
    {
      "entropy": 6.008070373535157,
      "epoch": 2.5,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004373973591142663,
      "loss": 5.731,
      "mean_token_accuracy": 0.18226670026779174,
      "num_tokens": 52785281.0,
      "step": 28310
    },
    {
      "entropy": 5.978436660766602,
      "epoch": 2.5004415400918405,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004373755610835599,
      "loss": 5.5937,
      "mean_token_accuracy": 0.18277614414691926,
      "num_tokens": 52793593.0,
      "step": 28315
    },
    {
      "entropy": 6.057453298568726,
      "epoch": 2.5008830801836806,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043735375987200877,
      "loss": 5.6789,
      "mean_token_accuracy": 0.17840089350938798,
      "num_tokens": 52803103.0,
      "step": 28320
    },
    {
      "entropy": 6.0859002590179445,
      "epoch": 2.501324620275521,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004373319554800397,
      "loss": 5.7159,
      "mean_token_accuracy": 0.1787596672773361,
      "num_tokens": 52811928.0,
      "step": 28325
    },
    {
      "entropy": 5.921882486343383,
      "epoch": 2.501766160367361,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00043731014790808,
      "loss": 5.5527,
      "mean_token_accuracy": 0.1956298142671585,
      "num_tokens": 52821352.0,
      "step": 28330
    },
    {
      "entropy": 5.919241523742675,
      "epoch": 2.5022077004592016,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004372883371565568,
      "loss": 5.5783,
      "mean_token_accuracy": 0.1942017197608948,
      "num_tokens": 52830688.0,
      "step": 28335
    },
    {
      "entropy": 5.970051527023315,
      "epoch": 2.502649240551042,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00043726652322589735,
      "loss": 5.6568,
      "mean_token_accuracy": 0.18236907124519347,
      "num_tokens": 52840804.0,
      "step": 28340
    },
    {
      "entropy": 6.012986850738526,
      "epoch": 2.503090780642882,
      "grad_norm": 5.0,
      "learning_rate": 0.000437244706116529,
      "loss": 5.6897,
      "mean_token_accuracy": 0.17058936655521392,
      "num_tokens": 52850476.0,
      "step": 28345
    },
    {
      "entropy": 6.020309925079346,
      "epoch": 2.5035323207347226,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00043722288582887913,
      "loss": 5.7531,
      "mean_token_accuracy": 0.1743412733078003,
      "num_tokens": 52859254.0,
      "step": 28350
    },
    {
      "entropy": 5.985603713989258,
      "epoch": 2.503973860826563,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004372010623633753,
      "loss": 5.678,
      "mean_token_accuracy": 0.17937082946300506,
      "num_tokens": 52867814.0,
      "step": 28355
    },
    {
      "entropy": 5.9818957328796385,
      "epoch": 2.504415400918403,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00043717923572044473,
      "loss": 5.5698,
      "mean_token_accuracy": 0.1820481687784195,
      "num_tokens": 52876958.0,
      "step": 28360
    },
    {
      "entropy": 5.936850357055664,
      "epoch": 2.5048569410102437,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043715740590051527,
      "loss": 5.5876,
      "mean_token_accuracy": 0.18870676755905152,
      "num_tokens": 52885978.0,
      "step": 28365
    },
    {
      "entropy": 5.939035844802857,
      "epoch": 2.505298481102084,
      "grad_norm": 1.5,
      "learning_rate": 0.0004371355729040144,
      "loss": 5.6118,
      "mean_token_accuracy": 0.18848535567522048,
      "num_tokens": 52894081.0,
      "step": 28370
    },
    {
      "entropy": 6.0074159622192385,
      "epoch": 2.5057400211939242,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004371137367313699,
      "loss": 5.7115,
      "mean_token_accuracy": 0.1805637300014496,
      "num_tokens": 52904012.0,
      "step": 28375
    },
    {
      "entropy": 6.038396644592285,
      "epoch": 2.5061815612857647,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043709189738300956,
      "loss": 5.6595,
      "mean_token_accuracy": 0.17811064571142196,
      "num_tokens": 52913319.0,
      "step": 28380
    },
    {
      "entropy": 5.984468030929565,
      "epoch": 2.5066231013776052,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004370700548593611,
      "loss": 5.6458,
      "mean_token_accuracy": 0.1806642398238182,
      "num_tokens": 52923539.0,
      "step": 28385
    },
    {
      "entropy": 5.972508382797241,
      "epoch": 2.5070646414694453,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004370482091608523,
      "loss": 5.5772,
      "mean_token_accuracy": 0.1892811194062233,
      "num_tokens": 52931856.0,
      "step": 28390
    },
    {
      "entropy": 5.980058336257935,
      "epoch": 2.507506181561286,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00043702636028791135,
      "loss": 5.7462,
      "mean_token_accuracy": 0.1737753629684448,
      "num_tokens": 52942270.0,
      "step": 28395
    },
    {
      "entropy": 5.9803338050842285,
      "epoch": 2.5079477216531263,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004370045082409662,
      "loss": 5.6253,
      "mean_token_accuracy": 0.1919827491044998,
      "num_tokens": 52951503.0,
      "step": 28400
    },
    {
      "entropy": 5.999129915237427,
      "epoch": 2.5083892617449663,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004369826530204449,
      "loss": 5.6681,
      "mean_token_accuracy": 0.1839463457465172,
      "num_tokens": 52960288.0,
      "step": 28405
    },
    {
      "entropy": 5.985926103591919,
      "epoch": 2.508830801836807,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00043696079462677537,
      "loss": 5.7094,
      "mean_token_accuracy": 0.18013712465763093,
      "num_tokens": 52969313.0,
      "step": 28410
    },
    {
      "entropy": 6.007653617858887,
      "epoch": 2.5092723419286473,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004369389330603861,
      "loss": 5.7018,
      "mean_token_accuracy": 0.1777064397931099,
      "num_tokens": 52978924.0,
      "step": 28415
    },
    {
      "entropy": 6.070556402206421,
      "epoch": 2.5097138820204874,
      "grad_norm": 1.46875,
      "learning_rate": 0.00043691706832170527,
      "loss": 5.6952,
      "mean_token_accuracy": 0.18293579071760177,
      "num_tokens": 52989027.0,
      "step": 28420
    },
    {
      "entropy": 6.077523231506348,
      "epoch": 2.510155422112328,
      "grad_norm": 1.53125,
      "learning_rate": 0.00043689520041116116,
      "loss": 5.7017,
      "mean_token_accuracy": 0.1791674792766571,
      "num_tokens": 52999436.0,
      "step": 28425
    },
    {
      "entropy": 5.983181810379028,
      "epoch": 2.5105969622041684,
      "grad_norm": 1.6875,
      "learning_rate": 0.00043687332932918204,
      "loss": 5.6933,
      "mean_token_accuracy": 0.1868923783302307,
      "num_tokens": 53008153.0,
      "step": 28430
    },
    {
      "entropy": 5.965483283996582,
      "epoch": 2.5110385022960084,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004368514550761965,
      "loss": 5.672,
      "mean_token_accuracy": 0.1876717507839203,
      "num_tokens": 53017344.0,
      "step": 28435
    },
    {
      "entropy": 5.948480749130249,
      "epoch": 2.511480042387849,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043682957765263304,
      "loss": 5.6021,
      "mean_token_accuracy": 0.1904551148414612,
      "num_tokens": 53026966.0,
      "step": 28440
    },
    {
      "entropy": 6.01513090133667,
      "epoch": 2.511921582479689,
      "grad_norm": 1.6875,
      "learning_rate": 0.00043680769705892013,
      "loss": 5.6546,
      "mean_token_accuracy": 0.1791793867945671,
      "num_tokens": 53035879.0,
      "step": 28445
    },
    {
      "entropy": 5.97525110244751,
      "epoch": 2.5123631225715295,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004367858132954866,
      "loss": 5.6812,
      "mean_token_accuracy": 0.1811175301671028,
      "num_tokens": 53044727.0,
      "step": 28450
    },
    {
      "entropy": 5.8855931758880615,
      "epoch": 2.51280466266337,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004367639263627609,
      "loss": 5.4752,
      "mean_token_accuracy": 0.19637687504291534,
      "num_tokens": 53054152.0,
      "step": 28455
    },
    {
      "entropy": 5.994456672668457,
      "epoch": 2.51324620275521,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004367420362611718,
      "loss": 5.7131,
      "mean_token_accuracy": 0.17413085848093032,
      "num_tokens": 53063259.0,
      "step": 28460
    },
    {
      "entropy": 6.064922285079956,
      "epoch": 2.5136877428470505,
      "grad_norm": 1.625,
      "learning_rate": 0.0004367201429911483,
      "loss": 5.6554,
      "mean_token_accuracy": 0.1807147443294525,
      "num_tokens": 53072139.0,
      "step": 28465
    },
    {
      "entropy": 6.019908094406128,
      "epoch": 2.5141292829388906,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004366982465531192,
      "loss": 5.6763,
      "mean_token_accuracy": 0.17581669241189957,
      "num_tokens": 53081833.0,
      "step": 28470
    },
    {
      "entropy": 5.886303997039795,
      "epoch": 2.514570823030731,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004366763469475134,
      "loss": 5.636,
      "mean_token_accuracy": 0.19482362121343613,
      "num_tokens": 53091952.0,
      "step": 28475
    },
    {
      "entropy": 5.95730881690979,
      "epoch": 2.5150123631225716,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004366544441747599,
      "loss": 5.6744,
      "mean_token_accuracy": 0.1875257283449173,
      "num_tokens": 53101619.0,
      "step": 28480
    },
    {
      "entropy": 6.083674335479737,
      "epoch": 2.5154539032144116,
      "grad_norm": 3.25,
      "learning_rate": 0.0004366325382352878,
      "loss": 5.6425,
      "mean_token_accuracy": 0.18222219049930571,
      "num_tokens": 53110931.0,
      "step": 28485
    },
    {
      "entropy": 5.996267557144165,
      "epoch": 2.515895443306252,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00043661062912952607,
      "loss": 5.6568,
      "mean_token_accuracy": 0.1810712307691574,
      "num_tokens": 53119949.0,
      "step": 28490
    },
    {
      "entropy": 5.950575351715088,
      "epoch": 2.5163369833980926,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004365887168579042,
      "loss": 5.6496,
      "mean_token_accuracy": 0.18754176050424576,
      "num_tokens": 53129048.0,
      "step": 28495
    },
    {
      "entropy": 6.0792640209197994,
      "epoch": 2.5167785234899327,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004365668014208512,
      "loss": 5.7255,
      "mean_token_accuracy": 0.17398383915424348,
      "num_tokens": 53139121.0,
      "step": 28500
    },
    {
      "entropy": 6.056385564804077,
      "epoch": 2.517220063581773,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043654488281879645,
      "loss": 5.6187,
      "mean_token_accuracy": 0.1864268124103546,
      "num_tokens": 53148098.0,
      "step": 28505
    },
    {
      "entropy": 5.986588335037231,
      "epoch": 2.5176616036736137,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004365229610521693,
      "loss": 5.6608,
      "mean_token_accuracy": 0.181496062874794,
      "num_tokens": 53156992.0,
      "step": 28510
    },
    {
      "entropy": 5.965496587753296,
      "epoch": 2.5181031437654537,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00043650103612139917,
      "loss": 5.6798,
      "mean_token_accuracy": 0.17819141447544098,
      "num_tokens": 53166985.0,
      "step": 28515
    },
    {
      "entropy": 6.039875125885009,
      "epoch": 2.5185446838572942,
      "grad_norm": 1.5,
      "learning_rate": 0.0004364791080269156,
      "loss": 5.703,
      "mean_token_accuracy": 0.18063240498304367,
      "num_tokens": 53176924.0,
      "step": 28520
    },
    {
      "entropy": 5.972803831100464,
      "epoch": 2.5189862239491347,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004364571767691482,
      "loss": 5.6985,
      "mean_token_accuracy": 0.17949337661266326,
      "num_tokens": 53186966.0,
      "step": 28525
    },
    {
      "entropy": 6.034223937988282,
      "epoch": 2.519427764040975,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043643524234852655,
      "loss": 5.7266,
      "mean_token_accuracy": 0.18016164600849152,
      "num_tokens": 53196147.0,
      "step": 28530
    },
    {
      "entropy": 5.981148958206177,
      "epoch": 2.5198693041328153,
      "grad_norm": 1.515625,
      "learning_rate": 0.00043641330476548024,
      "loss": 5.566,
      "mean_token_accuracy": 0.19192690253257752,
      "num_tokens": 53204446.0,
      "step": 28535
    },
    {
      "entropy": 5.977843618392944,
      "epoch": 2.520310844224656,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004363913640204392,
      "loss": 5.6954,
      "mean_token_accuracy": 0.17357856780290604,
      "num_tokens": 53213528.0,
      "step": 28540
    },
    {
      "entropy": 6.059853649139404,
      "epoch": 2.520752384316496,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000436369420113833,
      "loss": 5.7187,
      "mean_token_accuracy": 0.18058066070079803,
      "num_tokens": 53223574.0,
      "step": 28545
    },
    {
      "entropy": 5.945732879638672,
      "epoch": 2.5211939244083363,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00043634747304609173,
      "loss": 5.5776,
      "mean_token_accuracy": 0.182159523665905,
      "num_tokens": 53233011.0,
      "step": 28550
    },
    {
      "entropy": 6.008843851089478,
      "epoch": 2.521635464500177,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004363255228176452,
      "loss": 5.7502,
      "mean_token_accuracy": 0.1772286504507065,
      "num_tokens": 53241835.0,
      "step": 28555
    },
    {
      "entropy": 6.105554151535034,
      "epoch": 2.522077004592017,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004363035694289234,
      "loss": 5.692,
      "mean_token_accuracy": 0.1817841500043869,
      "num_tokens": 53250666.0,
      "step": 28560
    },
    {
      "entropy": 5.938131999969483,
      "epoch": 2.5225185446838574,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043628161288035644,
      "loss": 5.6331,
      "mean_token_accuracy": 0.18257057815790176,
      "num_tokens": 53259386.0,
      "step": 28565
    },
    {
      "entropy": 5.9675696849822994,
      "epoch": 2.522960084775698,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043625965317237435,
      "loss": 5.6908,
      "mean_token_accuracy": 0.1792156770825386,
      "num_tokens": 53268760.0,
      "step": 28570
    },
    {
      "entropy": 6.05617733001709,
      "epoch": 2.523401624867538,
      "grad_norm": 1.515625,
      "learning_rate": 0.00043623769030540747,
      "loss": 5.6702,
      "mean_token_accuracy": 0.18034393340349197,
      "num_tokens": 53277817.0,
      "step": 28575
    },
    {
      "entropy": 6.0017341613769535,
      "epoch": 2.5238431649593784,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043621572427988585,
      "loss": 5.5818,
      "mean_token_accuracy": 0.19740607887506484,
      "num_tokens": 53286699.0,
      "step": 28580
    },
    {
      "entropy": 6.014805746078491,
      "epoch": 2.5242847050512185,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004361937550962399,
      "loss": 5.6136,
      "mean_token_accuracy": 0.18569284826517105,
      "num_tokens": 53295431.0,
      "step": 28585
    },
    {
      "entropy": 5.921599912643432,
      "epoch": 2.524726245143059,
      "grad_norm": 1.578125,
      "learning_rate": 0.00043617178275489997,
      "loss": 5.6289,
      "mean_token_accuracy": 0.17996378093957902,
      "num_tokens": 53304986.0,
      "step": 28590
    },
    {
      "entropy": 5.99004373550415,
      "epoch": 2.5251677852348995,
      "grad_norm": 1.375,
      "learning_rate": 0.00043614980725629663,
      "loss": 5.7569,
      "mean_token_accuracy": 0.16748713552951813,
      "num_tokens": 53313908.0,
      "step": 28595
    },
    {
      "entropy": 6.024456977844238,
      "epoch": 2.5256093253267395,
      "grad_norm": 1.65625,
      "learning_rate": 0.00043612782860086,
      "loss": 5.6424,
      "mean_token_accuracy": 0.1813490256667137,
      "num_tokens": 53323553.0,
      "step": 28600
    },
    {
      "entropy": 6.054202508926392,
      "epoch": 2.52605086541858,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043610584678902096,
      "loss": 5.7237,
      "mean_token_accuracy": 0.17681425511837007,
      "num_tokens": 53333236.0,
      "step": 28605
    },
    {
      "entropy": 6.024262189865112,
      "epoch": 2.52649240551042,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043608386182120997,
      "loss": 5.6823,
      "mean_token_accuracy": 0.1807365521788597,
      "num_tokens": 53342344.0,
      "step": 28610
    },
    {
      "entropy": 5.9488075256347654,
      "epoch": 2.5269339456022606,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004360618736978577,
      "loss": 5.6797,
      "mean_token_accuracy": 0.1889035299420357,
      "num_tokens": 53351764.0,
      "step": 28615
    },
    {
      "entropy": 5.891925287246704,
      "epoch": 2.527375485694101,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000436039882419395,
      "loss": 5.4977,
      "mean_token_accuracy": 0.19460216909646988,
      "num_tokens": 53360667.0,
      "step": 28620
    },
    {
      "entropy": 6.0324952602386475,
      "epoch": 2.527817025785941,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043601788798625253,
      "loss": 5.6799,
      "mean_token_accuracy": 0.18051612973213196,
      "num_tokens": 53370011.0,
      "step": 28625
    },
    {
      "entropy": 6.048724746704101,
      "epoch": 2.5282585658777816,
      "grad_norm": 1.40625,
      "learning_rate": 0.00043599589039886125,
      "loss": 5.7771,
      "mean_token_accuracy": 0.1791238456964493,
      "num_tokens": 53379035.0,
      "step": 28630
    },
    {
      "entropy": 5.97433180809021,
      "epoch": 2.528700105969622,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000435973889657652,
      "loss": 5.6306,
      "mean_token_accuracy": 0.1895334914326668,
      "num_tokens": 53387739.0,
      "step": 28635
    },
    {
      "entropy": 6.038815879821778,
      "epoch": 2.529141646061462,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043595188576305585,
      "loss": 5.7547,
      "mean_token_accuracy": 0.17679326832294465,
      "num_tokens": 53398219.0,
      "step": 28640
    },
    {
      "entropy": 6.03806209564209,
      "epoch": 2.5295831861533027,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043592987871550373,
      "loss": 5.6998,
      "mean_token_accuracy": 0.1790256455540657,
      "num_tokens": 53407337.0,
      "step": 28645
    },
    {
      "entropy": 6.104467916488647,
      "epoch": 2.530024726245143,
      "grad_norm": 3.25,
      "learning_rate": 0.00043590786851542687,
      "loss": 5.7552,
      "mean_token_accuracy": 0.1811181053519249,
      "num_tokens": 53416223.0,
      "step": 28650
    },
    {
      "entropy": 5.9697356700897215,
      "epoch": 2.5304662663369832,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043588585516325635,
      "loss": 5.6422,
      "mean_token_accuracy": 0.1821350112557411,
      "num_tokens": 53424719.0,
      "step": 28655
    },
    {
      "entropy": 6.048151922225952,
      "epoch": 2.5309078064288237,
      "grad_norm": 4.6875,
      "learning_rate": 0.00043586383865942347,
      "loss": 5.6495,
      "mean_token_accuracy": 0.18389957994222642,
      "num_tokens": 53433763.0,
      "step": 28660
    },
    {
      "entropy": 6.040418910980224,
      "epoch": 2.5313493465206642,
      "grad_norm": 1.484375,
      "learning_rate": 0.00043584181900435947,
      "loss": 5.6688,
      "mean_token_accuracy": 0.18089215457439423,
      "num_tokens": 53443270.0,
      "step": 28665
    },
    {
      "entropy": 6.029223394393921,
      "epoch": 2.5317908866125043,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004358197961984957,
      "loss": 5.6295,
      "mean_token_accuracy": 0.18291924446821212,
      "num_tokens": 53451555.0,
      "step": 28670
    },
    {
      "entropy": 5.9353282928466795,
      "epoch": 2.532232426704345,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004357977702422635,
      "loss": 5.7118,
      "mean_token_accuracy": 0.1738572135567665,
      "num_tokens": 53460845.0,
      "step": 28675
    },
    {
      "entropy": 5.950704431533813,
      "epoch": 2.5326739667961853,
      "grad_norm": 2.828125,
      "learning_rate": 0.00043577574113609456,
      "loss": 5.612,
      "mean_token_accuracy": 0.19280239045619965,
      "num_tokens": 53470169.0,
      "step": 28680
    },
    {
      "entropy": 5.990049171447754,
      "epoch": 2.5331155068880253,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004357537088804203,
      "loss": 5.6489,
      "mean_token_accuracy": 0.18302438110113145,
      "num_tokens": 53478484.0,
      "step": 28685
    },
    {
      "entropy": 5.994502353668213,
      "epoch": 2.533557046979866,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004357316734756722,
      "loss": 5.6596,
      "mean_token_accuracy": 0.18297844976186753,
      "num_tokens": 53488791.0,
      "step": 28690
    },
    {
      "entropy": 5.983919763565064,
      "epoch": 2.5339985870717063,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004357096349222821,
      "loss": 5.6256,
      "mean_token_accuracy": 0.18925434052944184,
      "num_tokens": 53498310.0,
      "step": 28695
    },
    {
      "entropy": 6.0070556640625,
      "epoch": 2.5344401271635464,
      "grad_norm": 1.9375,
      "learning_rate": 0.00043568759322068165,
      "loss": 5.6676,
      "mean_token_accuracy": 0.1874783605337143,
      "num_tokens": 53508145.0,
      "step": 28700
    },
    {
      "entropy": 6.053684997558594,
      "epoch": 2.534881667255387,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004356655483713026,
      "loss": 5.7132,
      "mean_token_accuracy": 0.18509092479944228,
      "num_tokens": 53517473.0,
      "step": 28705
    },
    {
      "entropy": 6.015948057174683,
      "epoch": 2.5353232073472274,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043564350037457683,
      "loss": 5.6805,
      "mean_token_accuracy": 0.17594987899065018,
      "num_tokens": 53526625.0,
      "step": 28710
    },
    {
      "entropy": 6.0102745532989506,
      "epoch": 2.5357647474390674,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00043562144923093627,
      "loss": 5.5694,
      "mean_token_accuracy": 0.19317377507686614,
      "num_tokens": 53535480.0,
      "step": 28715
    },
    {
      "entropy": 5.94033637046814,
      "epoch": 2.536206287530908,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043559939494081286,
      "loss": 5.6899,
      "mean_token_accuracy": 0.18017374575138093,
      "num_tokens": 53544368.0,
      "step": 28720
    },
    {
      "entropy": 5.957814502716064,
      "epoch": 2.536647827622748,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004355773375046387,
      "loss": 5.6613,
      "mean_token_accuracy": 0.18719397783279418,
      "num_tokens": 53553235.0,
      "step": 28725
    },
    {
      "entropy": 6.030424642562866,
      "epoch": 2.5370893677145885,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004355552769228457,
      "loss": 5.708,
      "mean_token_accuracy": 0.17459890097379685,
      "num_tokens": 53562635.0,
      "step": 28730
    },
    {
      "entropy": 6.106467437744141,
      "epoch": 2.537530907806429,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004355332131958663,
      "loss": 5.6151,
      "mean_token_accuracy": 0.17895655930042267,
      "num_tokens": 53571394.0,
      "step": 28735
    },
    {
      "entropy": 6.0697715282440186,
      "epoch": 2.537972447898269,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004355111463241324,
      "loss": 5.7288,
      "mean_token_accuracy": 0.17590746879577637,
      "num_tokens": 53580718.0,
      "step": 28740
    },
    {
      "entropy": 5.989678859710693,
      "epoch": 2.5384139879901095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004354890763080765,
      "loss": 5.7538,
      "mean_token_accuracy": 0.181534942984581,
      "num_tokens": 53590120.0,
      "step": 28745
    },
    {
      "entropy": 5.985000467300415,
      "epoch": 2.5388555280819496,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004354670031481309,
      "loss": 5.6791,
      "mean_token_accuracy": 0.18006689101457596,
      "num_tokens": 53599783.0,
      "step": 28750
    },
    {
      "entropy": 6.0283331871032715,
      "epoch": 2.53929706817379,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004354449268447279,
      "loss": 5.6603,
      "mean_token_accuracy": 0.1818864658474922,
      "num_tokens": 53608663.0,
      "step": 28755
    },
    {
      "entropy": 6.031127309799194,
      "epoch": 2.5397386082656306,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00043542284739830006,
      "loss": 5.7554,
      "mean_token_accuracy": 0.17744542211294173,
      "num_tokens": 53618239.0,
      "step": 28760
    },
    {
      "entropy": 6.100083255767823,
      "epoch": 2.5401801483574706,
      "grad_norm": 1.65625,
      "learning_rate": 0.00043540076480927994,
      "loss": 5.7,
      "mean_token_accuracy": 0.18314384669065475,
      "num_tokens": 53627683.0,
      "step": 28765
    },
    {
      "entropy": 5.980447435379029,
      "epoch": 2.540621688449311,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00043537867907809995,
      "loss": 5.5806,
      "mean_token_accuracy": 0.18541384041309356,
      "num_tokens": 53636980.0,
      "step": 28770
    },
    {
      "entropy": 5.958779573440552,
      "epoch": 2.5410632285411516,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004353565902051929,
      "loss": 5.6765,
      "mean_token_accuracy": 0.17880996316671371,
      "num_tokens": 53647310.0,
      "step": 28775
    },
    {
      "entropy": 5.987650060653687,
      "epoch": 2.5415047686329917,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043533449819099145,
      "loss": 5.7458,
      "mean_token_accuracy": 0.18107657581567765,
      "num_tokens": 53657433.0,
      "step": 28780
    },
    {
      "entropy": 6.000500297546386,
      "epoch": 2.541946308724832,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004353124030359284,
      "loss": 5.6384,
      "mean_token_accuracy": 0.18552677929401398,
      "num_tokens": 53666969.0,
      "step": 28785
    },
    {
      "entropy": 6.025702428817749,
      "epoch": 2.5423878488166727,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043529030474043657,
      "loss": 5.6771,
      "mean_token_accuracy": 0.18051717877388002,
      "num_tokens": 53676332.0,
      "step": 28790
    },
    {
      "entropy": 6.00628628730774,
      "epoch": 2.5428293889085127,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043526820330494877,
      "loss": 5.7289,
      "mean_token_accuracy": 0.18168286830186844,
      "num_tokens": 53685766.0,
      "step": 28795
    },
    {
      "entropy": 6.093632602691651,
      "epoch": 2.5432709290003532,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004352460987298981,
      "loss": 5.6572,
      "mean_token_accuracy": 0.17932604849338532,
      "num_tokens": 53694744.0,
      "step": 28800
    },
    {
      "entropy": 5.966246032714844,
      "epoch": 2.5437124690921937,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004352239910157174,
      "loss": 5.5505,
      "mean_token_accuracy": 0.1967330738902092,
      "num_tokens": 53704266.0,
      "step": 28805
    },
    {
      "entropy": 5.863087177276611,
      "epoch": 2.544154009184034,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00043520188016283977,
      "loss": 5.5946,
      "mean_token_accuracy": 0.18683600425720215,
      "num_tokens": 53712070.0,
      "step": 28810
    },
    {
      "entropy": 5.902587509155273,
      "epoch": 2.5445955492758743,
      "grad_norm": 1.40625,
      "learning_rate": 0.00043517976617169857,
      "loss": 5.661,
      "mean_token_accuracy": 0.17774668484926223,
      "num_tokens": 53720152.0,
      "step": 28815
    },
    {
      "entropy": 5.910110378265381,
      "epoch": 2.5450370893677148,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004351576490427268,
      "loss": 5.4831,
      "mean_token_accuracy": 0.1979513868689537,
      "num_tokens": 53729451.0,
      "step": 28820
    },
    {
      "entropy": 5.955924606323242,
      "epoch": 2.545478629459555,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004351355287763578,
      "loss": 5.6742,
      "mean_token_accuracy": 0.18105957210063933,
      "num_tokens": 53738866.0,
      "step": 28825
    },
    {
      "entropy": 5.944352436065674,
      "epoch": 2.5459201695513953,
      "grad_norm": 1.25,
      "learning_rate": 0.00043511340537302485,
      "loss": 5.5399,
      "mean_token_accuracy": 0.18893066346645354,
      "num_tokens": 53748006.0,
      "step": 28830
    },
    {
      "entropy": 6.00185260772705,
      "epoch": 2.546361709643236,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004350912788331613,
      "loss": 5.5706,
      "mean_token_accuracy": 0.19442579746246338,
      "num_tokens": 53756412.0,
      "step": 28835
    },
    {
      "entropy": 5.936732578277588,
      "epoch": 2.546803249735076,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004350691491572007,
      "loss": 5.6903,
      "mean_token_accuracy": 0.18233014345169068,
      "num_tokens": 53765509.0,
      "step": 28840
    },
    {
      "entropy": 5.960292673110962,
      "epoch": 2.5472447898269164,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004350470163455764,
      "loss": 5.6938,
      "mean_token_accuracy": 0.17737482935190202,
      "num_tokens": 53774883.0,
      "step": 28845
    },
    {
      "entropy": 6.0534404754638675,
      "epoch": 2.547686329918757,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00043502488039872224,
      "loss": 5.6717,
      "mean_token_accuracy": 0.17328175082802771,
      "num_tokens": 53783650.0,
      "step": 28850
    },
    {
      "entropy": 6.041615915298462,
      "epoch": 2.548127870010597,
      "grad_norm": 1.578125,
      "learning_rate": 0.00043500274131707156,
      "loss": 5.7405,
      "mean_token_accuracy": 0.17559683173894883,
      "num_tokens": 53793031.0,
      "step": 28855
    },
    {
      "entropy": 6.0642026424407955,
      "epoch": 2.5485694101024374,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004349805991010582,
      "loss": 5.7719,
      "mean_token_accuracy": 0.169334976375103,
      "num_tokens": 53803967.0,
      "step": 28860
    },
    {
      "entropy": 6.025951194763183,
      "epoch": 2.5490109501942775,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00043495845375111594,
      "loss": 5.6904,
      "mean_token_accuracy": 0.1768988624215126,
      "num_tokens": 53813575.0,
      "step": 28865
    },
    {
      "entropy": 5.972794151306152,
      "epoch": 2.549452490286118,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043493630526767846,
      "loss": 5.6396,
      "mean_token_accuracy": 0.1857285901904106,
      "num_tokens": 53823250.0,
      "step": 28870
    },
    {
      "entropy": 5.988069486618042,
      "epoch": 2.5498940303779585,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043491415365117984,
      "loss": 5.6335,
      "mean_token_accuracy": 0.18591178357601165,
      "num_tokens": 53832300.0,
      "step": 28875
    },
    {
      "entropy": 5.968751239776611,
      "epoch": 2.5503355704697985,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043489199890205376,
      "loss": 5.5902,
      "mean_token_accuracy": 0.18379128128290176,
      "num_tokens": 53841385.0,
      "step": 28880
    },
    {
      "entropy": 6.007231521606445,
      "epoch": 2.550777110561639,
      "grad_norm": 1.65625,
      "learning_rate": 0.00043486984102073445,
      "loss": 5.7003,
      "mean_token_accuracy": 0.1785671055316925,
      "num_tokens": 53850349.0,
      "step": 28885
    },
    {
      "entropy": 5.92902979850769,
      "epoch": 2.551218650653479,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043484768000765585,
      "loss": 5.5522,
      "mean_token_accuracy": 0.19365454763174056,
      "num_tokens": 53859547.0,
      "step": 28890
    },
    {
      "entropy": 5.9923125267028805,
      "epoch": 2.5516601907453196,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004348255158632521,
      "loss": 5.6696,
      "mean_token_accuracy": 0.18060100674629212,
      "num_tokens": 53868497.0,
      "step": 28895
    },
    {
      "entropy": 5.90680160522461,
      "epoch": 2.55210173083716,
      "grad_norm": 1.765625,
      "learning_rate": 0.00043480334858795737,
      "loss": 5.5993,
      "mean_token_accuracy": 0.18016465604305268,
      "num_tokens": 53877200.0,
      "step": 28900
    },
    {
      "entropy": 5.997800683975219,
      "epoch": 2.552543270929,
      "grad_norm": 1.265625,
      "learning_rate": 0.000434781178182206,
      "loss": 5.5694,
      "mean_token_accuracy": 0.19634449928998948,
      "num_tokens": 53887081.0,
      "step": 28905
    },
    {
      "entropy": 5.958229398727417,
      "epoch": 2.5529848110208406,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004347590046464322,
      "loss": 5.6201,
      "mean_token_accuracy": 0.18186433464288712,
      "num_tokens": 53896334.0,
      "step": 28910
    },
    {
      "entropy": 5.968129682540893,
      "epoch": 2.553426351112681,
      "grad_norm": 1.40625,
      "learning_rate": 0.00043473682798107034,
      "loss": 5.652,
      "mean_token_accuracy": 0.1840098813176155,
      "num_tokens": 53905658.0,
      "step": 28915
    },
    {
      "entropy": 6.063820838928223,
      "epoch": 2.553867891204521,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004347146481865549,
      "loss": 5.7527,
      "mean_token_accuracy": 0.17833008319139482,
      "num_tokens": 53915747.0,
      "step": 28920
    },
    {
      "entropy": 6.026525926589966,
      "epoch": 2.5543094312963617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004346924652633203,
      "loss": 5.7047,
      "mean_token_accuracy": 0.18329222053289412,
      "num_tokens": 53924636.0,
      "step": 28925
    },
    {
      "entropy": 5.935042476654052,
      "epoch": 2.554750971388202,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004346702792118011,
      "loss": 5.6082,
      "mean_token_accuracy": 0.1872940257191658,
      "num_tokens": 53934104.0,
      "step": 28930
    },
    {
      "entropy": 6.031295251846314,
      "epoch": 2.5551925114800422,
      "grad_norm": 1.46875,
      "learning_rate": 0.00043464809003243197,
      "loss": 5.6929,
      "mean_token_accuracy": 0.1868533030152321,
      "num_tokens": 53943055.0,
      "step": 28935
    },
    {
      "entropy": 5.9709357738494875,
      "epoch": 2.5556340515718827,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004346258977256475,
      "loss": 5.647,
      "mean_token_accuracy": 0.18951558023691178,
      "num_tokens": 53952489.0,
      "step": 28940
    },
    {
      "entropy": 5.992350912094116,
      "epoch": 2.5560755916637232,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043460370229188257,
      "loss": 5.5971,
      "mean_token_accuracy": 0.18298600018024444,
      "num_tokens": 53962274.0,
      "step": 28945
    },
    {
      "entropy": 5.987404870986938,
      "epoch": 2.5565171317555633,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043458150373157183,
      "loss": 5.6873,
      "mean_token_accuracy": 0.18119214624166488,
      "num_tokens": 53971152.0,
      "step": 28950
    },
    {
      "entropy": 6.00222487449646,
      "epoch": 2.5569586718474038,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004345593020451503,
      "loss": 5.632,
      "mean_token_accuracy": 0.18355462402105333,
      "num_tokens": 53979510.0,
      "step": 28955
    },
    {
      "entropy": 5.882973861694336,
      "epoch": 2.5574002119392443,
      "grad_norm": 1.46875,
      "learning_rate": 0.00043453709723305264,
      "loss": 5.6021,
      "mean_token_accuracy": 0.1868735060095787,
      "num_tokens": 53987917.0,
      "step": 28960
    },
    {
      "entropy": 5.989379215240478,
      "epoch": 2.5578417520310843,
      "grad_norm": 1.484375,
      "learning_rate": 0.00043451488929571405,
      "loss": 5.7017,
      "mean_token_accuracy": 0.1765973150730133,
      "num_tokens": 53997924.0,
      "step": 28965
    },
    {
      "entropy": 5.9721132755279545,
      "epoch": 2.558283292122925,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004344926782335695,
      "loss": 5.6591,
      "mean_token_accuracy": 0.17810166329145433,
      "num_tokens": 54006505.0,
      "step": 28970
    },
    {
      "entropy": 6.054075717926025,
      "epoch": 2.5587248322147653,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043447046404705404,
      "loss": 5.7531,
      "mean_token_accuracy": 0.1781377375125885,
      "num_tokens": 54016794.0,
      "step": 28975
    },
    {
      "entropy": 5.921995973587036,
      "epoch": 2.5591663723066054,
      "grad_norm": 1.375,
      "learning_rate": 0.0004344482467366029,
      "loss": 5.5498,
      "mean_token_accuracy": 0.19297775328159333,
      "num_tokens": 54025995.0,
      "step": 28980
    },
    {
      "entropy": 6.0108420848846436,
      "epoch": 2.559607912398446,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004344260263026513,
      "loss": 5.633,
      "mean_token_accuracy": 0.1874069944024086,
      "num_tokens": 54035501.0,
      "step": 28985
    },
    {
      "entropy": 6.0065735340118405,
      "epoch": 2.5600494524902864,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004344038027456346,
      "loss": 5.7172,
      "mean_token_accuracy": 0.17747959941625596,
      "num_tokens": 54045283.0,
      "step": 28990
    },
    {
      "entropy": 6.000532770156861,
      "epoch": 2.5604909925821264,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004343815760659879,
      "loss": 5.7522,
      "mean_token_accuracy": 0.1822257846593857,
      "num_tokens": 54056250.0,
      "step": 28995
    },
    {
      "entropy": 6.005073308944702,
      "epoch": 2.560932532673967,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004343593462641469,
      "loss": 5.6202,
      "mean_token_accuracy": 0.18864733278751372,
      "num_tokens": 54065225.0,
      "step": 29000
    },
    {
      "entropy": 6.002578258514404,
      "epoch": 2.561374072765807,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004343371133405469,
      "loss": 5.7454,
      "mean_token_accuracy": 0.18204960376024246,
      "num_tokens": 54074056.0,
      "step": 29005
    },
    {
      "entropy": 6.01514310836792,
      "epoch": 2.5618156128576475,
      "grad_norm": 1.375,
      "learning_rate": 0.00043431487729562345,
      "loss": 5.6164,
      "mean_token_accuracy": 0.18590531945228578,
      "num_tokens": 54083202.0,
      "step": 29010
    },
    {
      "entropy": 5.949268770217896,
      "epoch": 2.562257152949488,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004342926381298122,
      "loss": 5.5219,
      "mean_token_accuracy": 0.1993756577372551,
      "num_tokens": 54092088.0,
      "step": 29015
    },
    {
      "entropy": 6.0165760040283205,
      "epoch": 2.562698693041328,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004342703958435487,
      "loss": 5.747,
      "mean_token_accuracy": 0.17796616554260253,
      "num_tokens": 54101113.0,
      "step": 29020
    },
    {
      "entropy": 5.961110162734985,
      "epoch": 2.5631402331331685,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004342481504372688,
      "loss": 5.6459,
      "mean_token_accuracy": 0.19603005945682525,
      "num_tokens": 54109832.0,
      "step": 29025
    },
    {
      "entropy": 5.975558805465698,
      "epoch": 2.5635817732250086,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043422590191140813,
      "loss": 5.5972,
      "mean_token_accuracy": 0.18803804963827134,
      "num_tokens": 54117783.0,
      "step": 29030
    },
    {
      "entropy": 6.049834442138672,
      "epoch": 2.564023313316849,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004342036502664026,
      "loss": 5.7255,
      "mean_token_accuracy": 0.17522048503160476,
      "num_tokens": 54126539.0,
      "step": 29035
    },
    {
      "entropy": 6.0046533107757565,
      "epoch": 2.5644648534086896,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004341813955026882,
      "loss": 5.5738,
      "mean_token_accuracy": 0.19059089571237564,
      "num_tokens": 54135316.0,
      "step": 29040
    },
    {
      "entropy": 5.931634330749512,
      "epoch": 2.5649063935005296,
      "grad_norm": 1.390625,
      "learning_rate": 0.00043415913762070074,
      "loss": 5.633,
      "mean_token_accuracy": 0.19018080085515976,
      "num_tokens": 54144164.0,
      "step": 29045
    },
    {
      "entropy": 5.976301431655884,
      "epoch": 2.56534793359237,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004341368766208763,
      "loss": 5.6914,
      "mean_token_accuracy": 0.18106572180986405,
      "num_tokens": 54154030.0,
      "step": 29050
    },
    {
      "entropy": 5.898052024841308,
      "epoch": 2.5657894736842106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004341146125036509,
      "loss": 5.6014,
      "mean_token_accuracy": 0.19873946607112886,
      "num_tokens": 54163507.0,
      "step": 29055
    },
    {
      "entropy": 5.886259746551514,
      "epoch": 2.5662310137760507,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043409234526946077,
      "loss": 5.5692,
      "mean_token_accuracy": 0.18955798745155333,
      "num_tokens": 54173016.0,
      "step": 29060
    },
    {
      "entropy": 6.016428422927857,
      "epoch": 2.566672553867891,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00043407007491874216,
      "loss": 5.6865,
      "mean_token_accuracy": 0.18022528290748596,
      "num_tokens": 54181965.0,
      "step": 29065
    },
    {
      "entropy": 5.998293066024781,
      "epoch": 2.5671140939597317,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043404780145193115,
      "loss": 5.6303,
      "mean_token_accuracy": 0.18432801961898804,
      "num_tokens": 54190574.0,
      "step": 29070
    },
    {
      "entropy": 6.07862229347229,
      "epoch": 2.5675556340515717,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043402552486946424,
      "loss": 5.7208,
      "mean_token_accuracy": 0.18121076822280885,
      "num_tokens": 54200105.0,
      "step": 29075
    },
    {
      "entropy": 6.080522727966309,
      "epoch": 2.567997174143412,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00043400324517177765,
      "loss": 5.6846,
      "mean_token_accuracy": 0.17416635602712632,
      "num_tokens": 54209247.0,
      "step": 29080
    },
    {
      "entropy": 6.010126161575317,
      "epoch": 2.5684387142352527,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043398096235930795,
      "loss": 5.6147,
      "mean_token_accuracy": 0.18561898320913314,
      "num_tokens": 54218122.0,
      "step": 29085
    },
    {
      "entropy": 5.866766834259034,
      "epoch": 2.5688802543270928,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00043395867643249164,
      "loss": 5.5435,
      "mean_token_accuracy": 0.19400385171175002,
      "num_tokens": 54226706.0,
      "step": 29090
    },
    {
      "entropy": 5.923044109344483,
      "epoch": 2.5693217944189333,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004339363873917652,
      "loss": 5.6381,
      "mean_token_accuracy": 0.18082132339477539,
      "num_tokens": 54235927.0,
      "step": 29095
    },
    {
      "entropy": 6.009950780868531,
      "epoch": 2.5697633345107738,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043391409523756537,
      "loss": 5.7417,
      "mean_token_accuracy": 0.17794918417930602,
      "num_tokens": 54245939.0,
      "step": 29100
    },
    {
      "entropy": 6.0023821830749515,
      "epoch": 2.570204874602614,
      "grad_norm": 1.765625,
      "learning_rate": 0.00043389179997032885,
      "loss": 5.7252,
      "mean_token_accuracy": 0.17691311240196228,
      "num_tokens": 54255726.0,
      "step": 29105
    },
    {
      "entropy": 5.93682222366333,
      "epoch": 2.5706464146944543,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004338695015904922,
      "loss": 5.5764,
      "mean_token_accuracy": 0.18897178173065185,
      "num_tokens": 54264485.0,
      "step": 29110
    },
    {
      "entropy": 5.923153495788574,
      "epoch": 2.571087954786295,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004338472000984924,
      "loss": 5.6287,
      "mean_token_accuracy": 0.1824408605694771,
      "num_tokens": 54272717.0,
      "step": 29115
    },
    {
      "entropy": 5.940501070022583,
      "epoch": 2.571529494878135,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043382489549476636,
      "loss": 5.5622,
      "mean_token_accuracy": 0.1885616585612297,
      "num_tokens": 54281806.0,
      "step": 29120
    },
    {
      "entropy": 5.959068727493286,
      "epoch": 2.5719710349699754,
      "grad_norm": 1.9375,
      "learning_rate": 0.000433802587779751,
      "loss": 5.7113,
      "mean_token_accuracy": 0.18067944347858428,
      "num_tokens": 54291147.0,
      "step": 29125
    },
    {
      "entropy": 6.034480333328247,
      "epoch": 2.572412575061816,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00043378027695388315,
      "loss": 5.6984,
      "mean_token_accuracy": 0.18355059176683425,
      "num_tokens": 54299696.0,
      "step": 29130
    },
    {
      "entropy": 6.060085964202881,
      "epoch": 2.572854115153656,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00043375796301760007,
      "loss": 5.646,
      "mean_token_accuracy": 0.18928551226854323,
      "num_tokens": 54309320.0,
      "step": 29135
    },
    {
      "entropy": 6.007713651657104,
      "epoch": 2.5732956552454964,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004337356459713387,
      "loss": 5.7486,
      "mean_token_accuracy": 0.17688731402158736,
      "num_tokens": 54318794.0,
      "step": 29140
    },
    {
      "entropy": 5.990982294082642,
      "epoch": 2.5737371953373365,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004337133258155363,
      "loss": 5.6753,
      "mean_token_accuracy": 0.1786198854446411,
      "num_tokens": 54327792.0,
      "step": 29145
    },
    {
      "entropy": 6.024251317977905,
      "epoch": 2.574178735429177,
      "grad_norm": 1.59375,
      "learning_rate": 0.00043369100255063016,
      "loss": 5.6875,
      "mean_token_accuracy": 0.174262335896492,
      "num_tokens": 54337524.0,
      "step": 29150
    },
    {
      "entropy": 5.933884143829346,
      "epoch": 2.5746202755210175,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004336686761770575,
      "loss": 5.5814,
      "mean_token_accuracy": 0.19294020384550095,
      "num_tokens": 54347409.0,
      "step": 29155
    },
    {
      "entropy": 5.844069528579712,
      "epoch": 2.5750618156128575,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00043364634669525577,
      "loss": 5.584,
      "mean_token_accuracy": 0.18969229906797408,
      "num_tokens": 54356744.0,
      "step": 29160
    },
    {
      "entropy": 5.97955436706543,
      "epoch": 2.575503355704698,
      "grad_norm": 1.375,
      "learning_rate": 0.00043362401410566224,
      "loss": 5.6587,
      "mean_token_accuracy": 0.1802041783928871,
      "num_tokens": 54365274.0,
      "step": 29165
    },
    {
      "entropy": 5.955103778839112,
      "epoch": 2.575944895796538,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043360167840871455,
      "loss": 5.6084,
      "mean_token_accuracy": 0.18904134780168533,
      "num_tokens": 54374448.0,
      "step": 29170
    },
    {
      "entropy": 5.932705307006836,
      "epoch": 2.5763864358883786,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043357933960485017,
      "loss": 5.6129,
      "mean_token_accuracy": 0.17995439618825912,
      "num_tokens": 54383599.0,
      "step": 29175
    },
    {
      "entropy": 5.900909852981568,
      "epoch": 2.576827975980219,
      "grad_norm": 2.90625,
      "learning_rate": 0.00043355699769450674,
      "loss": 5.5357,
      "mean_token_accuracy": 0.1961402490735054,
      "num_tokens": 54392294.0,
      "step": 29180
    },
    {
      "entropy": 6.039721250534058,
      "epoch": 2.577269516072059,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043353465267812185,
      "loss": 5.655,
      "mean_token_accuracy": 0.18502010256052018,
      "num_tokens": 54401854.0,
      "step": 29185
    },
    {
      "entropy": 6.001086139678955,
      "epoch": 2.5777110561638996,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004335123045561333,
      "loss": 5.7494,
      "mean_token_accuracy": 0.1779002696275711,
      "num_tokens": 54411255.0,
      "step": 29190
    },
    {
      "entropy": 6.002755355834961,
      "epoch": 2.57815259625574,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043348995332897887,
      "loss": 5.6858,
      "mean_token_accuracy": 0.1748967796564102,
      "num_tokens": 54420437.0,
      "step": 29195
    },
    {
      "entropy": 5.908544063568115,
      "epoch": 2.57859413634758,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043346759899709643,
      "loss": 5.5107,
      "mean_token_accuracy": 0.19591933041810988,
      "num_tokens": 54429166.0,
      "step": 29200
    },
    {
      "entropy": 5.983400821685791,
      "epoch": 2.5790356764394207,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004334452415609238,
      "loss": 5.6079,
      "mean_token_accuracy": 0.18183168768882751,
      "num_tokens": 54438060.0,
      "step": 29205
    },
    {
      "entropy": 5.854114627838134,
      "epoch": 2.579477216531261,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000433422881020899,
      "loss": 5.5978,
      "mean_token_accuracy": 0.19172147214412688,
      "num_tokens": 54447503.0,
      "step": 29210
    },
    {
      "entropy": 5.895774984359742,
      "epoch": 2.579918756623101,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043340051737746,
      "loss": 5.5991,
      "mean_token_accuracy": 0.18887109011411668,
      "num_tokens": 54456741.0,
      "step": 29215
    },
    {
      "entropy": 6.009028816223145,
      "epoch": 2.5803602967149417,
      "grad_norm": 1.265625,
      "learning_rate": 0.000433378150631045,
      "loss": 5.6911,
      "mean_token_accuracy": 0.17968419045209885,
      "num_tokens": 54466614.0,
      "step": 29220
    },
    {
      "entropy": 6.051503658294678,
      "epoch": 2.580801836806782,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004333557807820922,
      "loss": 5.652,
      "mean_token_accuracy": 0.18574352711439132,
      "num_tokens": 54475191.0,
      "step": 29225
    },
    {
      "entropy": 5.90380449295044,
      "epoch": 2.5812433768986223,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004333334078310396,
      "loss": 5.5897,
      "mean_token_accuracy": 0.1856099396944046,
      "num_tokens": 54483924.0,
      "step": 29230
    },
    {
      "entropy": 5.852296257019043,
      "epoch": 2.5816849169904628,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043331103177832566,
      "loss": 5.5895,
      "mean_token_accuracy": 0.18682182729244232,
      "num_tokens": 54493739.0,
      "step": 29235
    },
    {
      "entropy": 5.955192708969117,
      "epoch": 2.5821264570823033,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004332886526243886,
      "loss": 5.6281,
      "mean_token_accuracy": 0.1799652487039566,
      "num_tokens": 54503760.0,
      "step": 29240
    },
    {
      "entropy": 6.063754558563232,
      "epoch": 2.5825679971741433,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004332662703696669,
      "loss": 5.6192,
      "mean_token_accuracy": 0.19206175506114959,
      "num_tokens": 54512884.0,
      "step": 29245
    },
    {
      "entropy": 5.9932887077331545,
      "epoch": 2.583009537265984,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004332438850145989,
      "loss": 5.6927,
      "mean_token_accuracy": 0.18177761286497116,
      "num_tokens": 54522747.0,
      "step": 29250
    },
    {
      "entropy": 5.899727153778076,
      "epoch": 2.5834510773578243,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004332214965596232,
      "loss": 5.5824,
      "mean_token_accuracy": 0.19063006341457367,
      "num_tokens": 54532109.0,
      "step": 29255
    },
    {
      "entropy": 5.949883031845093,
      "epoch": 2.5838926174496644,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004331991050051785,
      "loss": 5.6334,
      "mean_token_accuracy": 0.19144455790519715,
      "num_tokens": 54541160.0,
      "step": 29260
    },
    {
      "entropy": 5.988670873641968,
      "epoch": 2.584334157541505,
      "grad_norm": 1.53125,
      "learning_rate": 0.00043317671035170315,
      "loss": 5.6396,
      "mean_token_accuracy": 0.19214773774147034,
      "num_tokens": 54550162.0,
      "step": 29265
    },
    {
      "entropy": 5.914722347259522,
      "epoch": 2.5847756976333454,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043315431259963606,
      "loss": 5.494,
      "mean_token_accuracy": 0.19684784561395646,
      "num_tokens": 54559355.0,
      "step": 29270
    },
    {
      "entropy": 5.899127578735351,
      "epoch": 2.5852172377251854,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043313191174941594,
      "loss": 5.5653,
      "mean_token_accuracy": 0.19111478179693223,
      "num_tokens": 54568628.0,
      "step": 29275
    },
    {
      "entropy": 5.905876684188843,
      "epoch": 2.585658777817026,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004331095078014816,
      "loss": 5.6427,
      "mean_token_accuracy": 0.18528647124767303,
      "num_tokens": 54578744.0,
      "step": 29280
    },
    {
      "entropy": 6.063348579406738,
      "epoch": 2.586100317908866,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043308710075627195,
      "loss": 5.6731,
      "mean_token_accuracy": 0.18462103903293609,
      "num_tokens": 54588654.0,
      "step": 29285
    },
    {
      "entropy": 5.96837100982666,
      "epoch": 2.5865418580007065,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004330646906142259,
      "loss": 5.6164,
      "mean_token_accuracy": 0.18716322034597396,
      "num_tokens": 54597085.0,
      "step": 29290
    },
    {
      "entropy": 5.923666048049927,
      "epoch": 2.586983398092547,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004330422773757824,
      "loss": 5.7109,
      "mean_token_accuracy": 0.18280919045209884,
      "num_tokens": 54606151.0,
      "step": 29295
    },
    {
      "entropy": 5.93885760307312,
      "epoch": 2.587424938184387,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004330198610413806,
      "loss": 5.6027,
      "mean_token_accuracy": 0.18462304770946503,
      "num_tokens": 54615085.0,
      "step": 29300
    },
    {
      "entropy": 5.969778728485108,
      "epoch": 2.5878664782762275,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004329974416114596,
      "loss": 5.5868,
      "mean_token_accuracy": 0.18825413137674332,
      "num_tokens": 54623895.0,
      "step": 29305
    },
    {
      "entropy": 5.896623039245606,
      "epoch": 2.5883080183680676,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004329750190864586,
      "loss": 5.5551,
      "mean_token_accuracy": 0.19428563863039017,
      "num_tokens": 54633096.0,
      "step": 29310
    },
    {
      "entropy": 5.966534996032715,
      "epoch": 2.588749558459908,
      "grad_norm": 1.46875,
      "learning_rate": 0.00043295259346681677,
      "loss": 5.7259,
      "mean_token_accuracy": 0.17377757728099824,
      "num_tokens": 54642210.0,
      "step": 29315
    },
    {
      "entropy": 6.054153966903686,
      "epoch": 2.5891910985517486,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004329301647529734,
      "loss": 5.7845,
      "mean_token_accuracy": 0.16754352897405625,
      "num_tokens": 54651960.0,
      "step": 29320
    },
    {
      "entropy": 6.051436901092529,
      "epoch": 2.5896326386435886,
      "grad_norm": 2.359375,
      "learning_rate": 0.00043290773294536814,
      "loss": 5.6545,
      "mean_token_accuracy": 0.18441399931907654,
      "num_tokens": 54661560.0,
      "step": 29325
    },
    {
      "entropy": 6.00860276222229,
      "epoch": 2.590074178735429,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043288529804443997,
      "loss": 5.6715,
      "mean_token_accuracy": 0.17880164831876755,
      "num_tokens": 54671302.0,
      "step": 29330
    },
    {
      "entropy": 5.95245852470398,
      "epoch": 2.5905157188272696,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043286286005062875,
      "loss": 5.6881,
      "mean_token_accuracy": 0.1796528920531273,
      "num_tokens": 54681641.0,
      "step": 29335
    },
    {
      "entropy": 5.9470391273498535,
      "epoch": 2.5909572589191097,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043284041896437386,
      "loss": 5.6407,
      "mean_token_accuracy": 0.17904799729585646,
      "num_tokens": 54690190.0,
      "step": 29340
    },
    {
      "entropy": 6.062044858932495,
      "epoch": 2.59139879901095,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004328179747861149,
      "loss": 5.7293,
      "mean_token_accuracy": 0.1761363849043846,
      "num_tokens": 54699164.0,
      "step": 29345
    },
    {
      "entropy": 5.958253526687622,
      "epoch": 2.5918403391027907,
      "grad_norm": 1.390625,
      "learning_rate": 0.00043279552751629153,
      "loss": 5.5359,
      "mean_token_accuracy": 0.19833561331033706,
      "num_tokens": 54708509.0,
      "step": 29350
    },
    {
      "entropy": 5.975539207458496,
      "epoch": 2.5922818791946307,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043277307715534365,
      "loss": 5.6783,
      "mean_token_accuracy": 0.178049498796463,
      "num_tokens": 54717382.0,
      "step": 29355
    },
    {
      "entropy": 5.918326187133789,
      "epoch": 2.592723419286471,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004327506237037109,
      "loss": 5.6491,
      "mean_token_accuracy": 0.19091008454561234,
      "num_tokens": 54726191.0,
      "step": 29360
    },
    {
      "entropy": 6.0595705032348635,
      "epoch": 2.5931649593783117,
      "grad_norm": 1.1875,
      "learning_rate": 0.000432728167161833,
      "loss": 5.7118,
      "mean_token_accuracy": 0.1813634678721428,
      "num_tokens": 54736748.0,
      "step": 29365
    },
    {
      "entropy": 5.950178384780884,
      "epoch": 2.5936064994701518,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004327057075301501,
      "loss": 5.6226,
      "mean_token_accuracy": 0.18651025444269181,
      "num_tokens": 54746623.0,
      "step": 29370
    },
    {
      "entropy": 5.974226760864258,
      "epoch": 2.5940480395619923,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043268324480910203,
      "loss": 5.5704,
      "mean_token_accuracy": 0.19207168221473694,
      "num_tokens": 54755778.0,
      "step": 29375
    },
    {
      "entropy": 6.0194634914398195,
      "epoch": 2.5944895796538328,
      "grad_norm": 1.5,
      "learning_rate": 0.000432660778999129,
      "loss": 5.7606,
      "mean_token_accuracy": 0.17940463572740556,
      "num_tokens": 54765410.0,
      "step": 29380
    },
    {
      "entropy": 6.035443115234375,
      "epoch": 2.594931119745673,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004326383101006709,
      "loss": 5.7875,
      "mean_token_accuracy": 0.17225972265005113,
      "num_tokens": 54776115.0,
      "step": 29385
    },
    {
      "entropy": 5.956208753585815,
      "epoch": 2.5953726598375133,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000432615838114168,
      "loss": 5.6103,
      "mean_token_accuracy": 0.18337967842817307,
      "num_tokens": 54784719.0,
      "step": 29390
    },
    {
      "entropy": 6.00891900062561,
      "epoch": 2.595814199929354,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043259336304006036,
      "loss": 5.5895,
      "mean_token_accuracy": 0.18668196052312852,
      "num_tokens": 54794003.0,
      "step": 29395
    },
    {
      "entropy": 6.007985210418701,
      "epoch": 2.596255740021194,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004325708848787884,
      "loss": 5.6593,
      "mean_token_accuracy": 0.1876630261540413,
      "num_tokens": 54803634.0,
      "step": 29400
    },
    {
      "entropy": 5.981464624404907,
      "epoch": 2.5966972801130344,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004325484036307925,
      "loss": 5.708,
      "mean_token_accuracy": 0.19029986262321472,
      "num_tokens": 54813466.0,
      "step": 29405
    },
    {
      "entropy": 6.032210779190064,
      "epoch": 2.597138820204875,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004325259192965129,
      "loss": 5.6559,
      "mean_token_accuracy": 0.17944552898406982,
      "num_tokens": 54823855.0,
      "step": 29410
    },
    {
      "entropy": 6.029872608184815,
      "epoch": 2.597580360296715,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004325034318763902,
      "loss": 5.6417,
      "mean_token_accuracy": 0.18733559399843216,
      "num_tokens": 54832696.0,
      "step": 29415
    },
    {
      "entropy": 5.911052703857422,
      "epoch": 2.5980219003885554,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004324809413708648,
      "loss": 5.7112,
      "mean_token_accuracy": 0.1766494631767273,
      "num_tokens": 54841552.0,
      "step": 29420
    },
    {
      "entropy": 5.873369312286377,
      "epoch": 2.5984634404803955,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004324584477803773,
      "loss": 5.6404,
      "mean_token_accuracy": 0.18759051710367203,
      "num_tokens": 54851196.0,
      "step": 29425
    },
    {
      "entropy": 6.063150262832641,
      "epoch": 2.598904980572236,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043243595110536846,
      "loss": 5.5933,
      "mean_token_accuracy": 0.18205299526453017,
      "num_tokens": 54860334.0,
      "step": 29430
    },
    {
      "entropy": 5.957345390319825,
      "epoch": 2.5993465206640765,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043241345134627885,
      "loss": 5.6302,
      "mean_token_accuracy": 0.1853907361626625,
      "num_tokens": 54870056.0,
      "step": 29435
    },
    {
      "entropy": 5.995867443084717,
      "epoch": 2.5997880607559165,
      "grad_norm": 1.578125,
      "learning_rate": 0.00043239094850354914,
      "loss": 5.7222,
      "mean_token_accuracy": 0.17646201997995375,
      "num_tokens": 54880788.0,
      "step": 29440
    },
    {
      "entropy": 5.990607881546021,
      "epoch": 2.600229600847757,
      "grad_norm": 2.0,
      "learning_rate": 0.0004323684425776203,
      "loss": 5.6702,
      "mean_token_accuracy": 0.1812853291630745,
      "num_tokens": 54890021.0,
      "step": 29445
    },
    {
      "entropy": 5.925124788284302,
      "epoch": 2.600671140939597,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043234593356893325,
      "loss": 5.5291,
      "mean_token_accuracy": 0.1988020047545433,
      "num_tokens": 54898884.0,
      "step": 29450
    },
    {
      "entropy": 5.998127365112305,
      "epoch": 2.6011126810314376,
      "grad_norm": 2.0,
      "learning_rate": 0.00043232342147792887,
      "loss": 5.705,
      "mean_token_accuracy": 0.1769766628742218,
      "num_tokens": 54907742.0,
      "step": 29455
    },
    {
      "entropy": 6.019182443618774,
      "epoch": 2.601554221123278,
      "grad_norm": 1.484375,
      "learning_rate": 0.000432300906305048,
      "loss": 5.6714,
      "mean_token_accuracy": 0.18467034846544267,
      "num_tokens": 54916838.0,
      "step": 29460
    },
    {
      "entropy": 6.030784177780151,
      "epoch": 2.601995761215118,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004322783880507319,
      "loss": 5.731,
      "mean_token_accuracy": 0.17412341833114625,
      "num_tokens": 54925942.0,
      "step": 29465
    },
    {
      "entropy": 5.9011870384216305,
      "epoch": 2.6024373013069586,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043225586671542164,
      "loss": 5.5649,
      "mean_token_accuracy": 0.18323589414358138,
      "num_tokens": 54934611.0,
      "step": 29470
    },
    {
      "entropy": 6.056069993972779,
      "epoch": 2.602878841398799,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004322333422995584,
      "loss": 5.6637,
      "mean_token_accuracy": 0.17987721860408784,
      "num_tokens": 54944354.0,
      "step": 29475
    },
    {
      "entropy": 5.96573805809021,
      "epoch": 2.603320381490639,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043221081480358337,
      "loss": 5.7289,
      "mean_token_accuracy": 0.1765023350715637,
      "num_tokens": 54954070.0,
      "step": 29480
    },
    {
      "entropy": 6.069894456863404,
      "epoch": 2.6037619215824797,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004321882842279379,
      "loss": 5.7349,
      "mean_token_accuracy": 0.17792805135250092,
      "num_tokens": 54963922.0,
      "step": 29485
    },
    {
      "entropy": 6.002144479751587,
      "epoch": 2.60420346167432,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004321657505730634,
      "loss": 5.5953,
      "mean_token_accuracy": 0.1927308037877083,
      "num_tokens": 54973202.0,
      "step": 29490
    },
    {
      "entropy": 5.973252487182617,
      "epoch": 2.60464500176616,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043214321383940117,
      "loss": 5.6574,
      "mean_token_accuracy": 0.18608666360378265,
      "num_tokens": 54982318.0,
      "step": 29495
    },
    {
      "entropy": 5.907000637054443,
      "epoch": 2.6050865418580007,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043212067402739274,
      "loss": 5.5111,
      "mean_token_accuracy": 0.1952067047357559,
      "num_tokens": 54991397.0,
      "step": 29500
    },
    {
      "entropy": 6.058016586303711,
      "epoch": 2.605528081949841,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004320981311374797,
      "loss": 5.7113,
      "mean_token_accuracy": 0.18254433423280716,
      "num_tokens": 55001160.0,
      "step": 29505
    },
    {
      "entropy": 6.027294254302978,
      "epoch": 2.6059696220416813,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043207558517010364,
      "loss": 5.6909,
      "mean_token_accuracy": 0.18099435567855834,
      "num_tokens": 55010311.0,
      "step": 29510
    },
    {
      "entropy": 5.908308982849121,
      "epoch": 2.6064111621335218,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004320530361257061,
      "loss": 5.6364,
      "mean_token_accuracy": 0.18992536514997482,
      "num_tokens": 55020017.0,
      "step": 29515
    },
    {
      "entropy": 6.024827146530152,
      "epoch": 2.6068527022253623,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000432030484004729,
      "loss": 5.733,
      "mean_token_accuracy": 0.17896700650453568,
      "num_tokens": 55029930.0,
      "step": 29520
    },
    {
      "entropy": 5.976856422424317,
      "epoch": 2.6072942423172023,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043200792880761403,
      "loss": 5.6288,
      "mean_token_accuracy": 0.18019764721393586,
      "num_tokens": 55039065.0,
      "step": 29525
    },
    {
      "entropy": 5.914311313629151,
      "epoch": 2.607735782409043,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000431985370534803,
      "loss": 5.5352,
      "mean_token_accuracy": 0.188973468542099,
      "num_tokens": 55047592.0,
      "step": 29530
    },
    {
      "entropy": 5.977641487121582,
      "epoch": 2.6081773225008833,
      "grad_norm": 2.25,
      "learning_rate": 0.00043196280918673783,
      "loss": 5.6954,
      "mean_token_accuracy": 0.17552240043878556,
      "num_tokens": 55058031.0,
      "step": 29535
    },
    {
      "entropy": 6.017150735855102,
      "epoch": 2.6086188625927234,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043194024476386045,
      "loss": 5.7198,
      "mean_token_accuracy": 0.18693213909864426,
      "num_tokens": 55068283.0,
      "step": 29540
    },
    {
      "entropy": 5.994935703277588,
      "epoch": 2.609060402684564,
      "grad_norm": 1.34375,
      "learning_rate": 0.000431917677266613,
      "loss": 5.5113,
      "mean_token_accuracy": 0.1894264355301857,
      "num_tokens": 55077480.0,
      "step": 29545
    },
    {
      "entropy": 5.923508787155152,
      "epoch": 2.6095019427764043,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043189510669543754,
      "loss": 5.691,
      "mean_token_accuracy": 0.1820530042052269,
      "num_tokens": 55086783.0,
      "step": 29550
    },
    {
      "entropy": 5.97606635093689,
      "epoch": 2.6099434828682444,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004318725330507762,
      "loss": 5.7187,
      "mean_token_accuracy": 0.17214056998491287,
      "num_tokens": 55096861.0,
      "step": 29555
    },
    {
      "entropy": 6.016105222702026,
      "epoch": 2.610385022960085,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00043184995633307094,
      "loss": 5.6759,
      "mean_token_accuracy": 0.18988265693187714,
      "num_tokens": 55107504.0,
      "step": 29560
    },
    {
      "entropy": 6.07467131614685,
      "epoch": 2.610826563051925,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004318273765427644,
      "loss": 5.7992,
      "mean_token_accuracy": 0.1806540906429291,
      "num_tokens": 55116620.0,
      "step": 29565
    },
    {
      "entropy": 6.032183456420898,
      "epoch": 2.6112681031437655,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004318047936802987,
      "loss": 5.6372,
      "mean_token_accuracy": 0.17960902750492097,
      "num_tokens": 55126428.0,
      "step": 29570
    },
    {
      "entropy": 5.983462524414063,
      "epoch": 2.611709643235606,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004317822077461164,
      "loss": 5.6383,
      "mean_token_accuracy": 0.17865613251924514,
      "num_tokens": 55135369.0,
      "step": 29575
    },
    {
      "entropy": 5.989570760726929,
      "epoch": 2.612151183327446,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004317596187406596,
      "loss": 5.6589,
      "mean_token_accuracy": 0.17457776367664338,
      "num_tokens": 55144742.0,
      "step": 29580
    },
    {
      "entropy": 5.972332382202149,
      "epoch": 2.6125927234192865,
      "grad_norm": 1.703125,
      "learning_rate": 0.00043173702666437114,
      "loss": 5.72,
      "mean_token_accuracy": 0.18247637897729874,
      "num_tokens": 55154665.0,
      "step": 29585
    },
    {
      "entropy": 6.018748140335083,
      "epoch": 2.6130342635111266,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043171443151769347,
      "loss": 5.6448,
      "mean_token_accuracy": 0.18438758701086044,
      "num_tokens": 55163288.0,
      "step": 29590
    },
    {
      "entropy": 6.023061084747314,
      "epoch": 2.613475803602967,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004316918333010692,
      "loss": 5.7425,
      "mean_token_accuracy": 0.1729889839887619,
      "num_tokens": 55172775.0,
      "step": 29595
    },
    {
      "entropy": 6.042168807983399,
      "epoch": 2.6139173436948075,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004316692320149411,
      "loss": 5.6428,
      "mean_token_accuracy": 0.18525144308805466,
      "num_tokens": 55182227.0,
      "step": 29600
    },
    {
      "entropy": 6.044992303848266,
      "epoch": 2.6143588837866476,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004316466276597517,
      "loss": 5.6824,
      "mean_token_accuracy": 0.18241609036922454,
      "num_tokens": 55191562.0,
      "step": 29605
    },
    {
      "entropy": 5.930574131011963,
      "epoch": 2.614800423878488,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000431624020235944,
      "loss": 5.5451,
      "mean_token_accuracy": 0.20359638929367066,
      "num_tokens": 55200439.0,
      "step": 29610
    },
    {
      "entropy": 5.8656974792480465,
      "epoch": 2.6152419639703286,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004316014097439609,
      "loss": 5.5837,
      "mean_token_accuracy": 0.19438981860876084,
      "num_tokens": 55208850.0,
      "step": 29615
    },
    {
      "entropy": 6.022376251220703,
      "epoch": 2.6156835040621687,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004315787961842452,
      "loss": 5.7143,
      "mean_token_accuracy": 0.17937727719545365,
      "num_tokens": 55218364.0,
      "step": 29620
    },
    {
      "entropy": 5.967840671539307,
      "epoch": 2.616125044154009,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043155617955723994,
      "loss": 5.6233,
      "mean_token_accuracy": 0.18989232629537584,
      "num_tokens": 55227201.0,
      "step": 29625
    },
    {
      "entropy": 5.99395227432251,
      "epoch": 2.6165665842458496,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004315335598633882,
      "loss": 5.7458,
      "mean_token_accuracy": 0.17454001307487488,
      "num_tokens": 55236907.0,
      "step": 29630
    },
    {
      "entropy": 6.1154131412506105,
      "epoch": 2.6170081243376897,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043151093710313296,
      "loss": 5.6846,
      "mean_token_accuracy": 0.17881130874156953,
      "num_tokens": 55246795.0,
      "step": 29635
    },
    {
      "entropy": 5.976366376876831,
      "epoch": 2.61744966442953,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043148831127691754,
      "loss": 5.6762,
      "mean_token_accuracy": 0.1812344893813133,
      "num_tokens": 55255313.0,
      "step": 29640
    },
    {
      "entropy": 5.916363859176636,
      "epoch": 2.6178912045213707,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043146568238518505,
      "loss": 5.6149,
      "mean_token_accuracy": 0.18174317628145217,
      "num_tokens": 55264698.0,
      "step": 29645
    },
    {
      "entropy": 6.018360280990601,
      "epoch": 2.6183327446132107,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004314430504283789,
      "loss": 5.7315,
      "mean_token_accuracy": 0.1802513673901558,
      "num_tokens": 55274180.0,
      "step": 29650
    },
    {
      "entropy": 6.025973606109619,
      "epoch": 2.6187742847050512,
      "grad_norm": 1.875,
      "learning_rate": 0.0004314204154069423,
      "loss": 5.7063,
      "mean_token_accuracy": 0.1757657751441002,
      "num_tokens": 55283552.0,
      "step": 29655
    },
    {
      "entropy": 5.9820887565612795,
      "epoch": 2.6192158247968917,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004313977773213188,
      "loss": 5.7251,
      "mean_token_accuracy": 0.1791217714548111,
      "num_tokens": 55293929.0,
      "step": 29660
    },
    {
      "entropy": 6.000983047485351,
      "epoch": 2.619657364888732,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004313751361719517,
      "loss": 5.5921,
      "mean_token_accuracy": 0.18733823746442796,
      "num_tokens": 55303586.0,
      "step": 29665
    },
    {
      "entropy": 6.073283386230469,
      "epoch": 2.6200989049805723,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043135249195928464,
      "loss": 5.6017,
      "mean_token_accuracy": 0.18689574301242828,
      "num_tokens": 55313032.0,
      "step": 29670
    },
    {
      "entropy": 5.9196757793426515,
      "epoch": 2.620540445072413,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043132984468376124,
      "loss": 5.7081,
      "mean_token_accuracy": 0.1827259823679924,
      "num_tokens": 55323046.0,
      "step": 29675
    },
    {
      "entropy": 6.0232947826385494,
      "epoch": 2.620981985164253,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043130719434582505,
      "loss": 5.6344,
      "mean_token_accuracy": 0.18820732086896896,
      "num_tokens": 55331956.0,
      "step": 29680
    },
    {
      "entropy": 5.89790153503418,
      "epoch": 2.6214235252560933,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004312845409459198,
      "loss": 5.5734,
      "mean_token_accuracy": 0.18734740763902663,
      "num_tokens": 55341573.0,
      "step": 29685
    },
    {
      "entropy": 5.932088899612427,
      "epoch": 2.621865065347934,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043126188448448934,
      "loss": 5.5969,
      "mean_token_accuracy": 0.1879306599497795,
      "num_tokens": 55350325.0,
      "step": 29690
    },
    {
      "entropy": 6.026337623596191,
      "epoch": 2.622306605439774,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004312392249619774,
      "loss": 5.7505,
      "mean_token_accuracy": 0.1769897922873497,
      "num_tokens": 55359716.0,
      "step": 29695
    },
    {
      "entropy": 6.100197410583496,
      "epoch": 2.6227481455316144,
      "grad_norm": 2.109375,
      "learning_rate": 0.00043121656237882786,
      "loss": 5.7492,
      "mean_token_accuracy": 0.17495867013931274,
      "num_tokens": 55369443.0,
      "step": 29700
    },
    {
      "entropy": 6.039412546157837,
      "epoch": 2.6231896856234544,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043119389673548477,
      "loss": 5.6964,
      "mean_token_accuracy": 0.17697400450706482,
      "num_tokens": 55379018.0,
      "step": 29705
    },
    {
      "entropy": 6.016566801071167,
      "epoch": 2.623631225715295,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000431171228032392,
      "loss": 5.6956,
      "mean_token_accuracy": 0.18063931167125702,
      "num_tokens": 55388681.0,
      "step": 29710
    },
    {
      "entropy": 6.026255035400391,
      "epoch": 2.6240727658071354,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004311485562699938,
      "loss": 5.6404,
      "mean_token_accuracy": 0.1800203263759613,
      "num_tokens": 55398495.0,
      "step": 29715
    },
    {
      "entropy": 5.930399084091187,
      "epoch": 2.6245143058989755,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004311258814487342,
      "loss": 5.6637,
      "mean_token_accuracy": 0.18674989342689513,
      "num_tokens": 55408498.0,
      "step": 29720
    },
    {
      "entropy": 5.9209105491638185,
      "epoch": 2.624955845990816,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004311032035690572,
      "loss": 5.6728,
      "mean_token_accuracy": 0.1814674809575081,
      "num_tokens": 55417336.0,
      "step": 29725
    },
    {
      "entropy": 6.041076040267944,
      "epoch": 2.625397386082656,
      "grad_norm": 1.875,
      "learning_rate": 0.0004310805226314074,
      "loss": 5.6508,
      "mean_token_accuracy": 0.18447604030370712,
      "num_tokens": 55426828.0,
      "step": 29730
    },
    {
      "entropy": 6.016793203353882,
      "epoch": 2.6258389261744965,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004310578386362288,
      "loss": 5.5726,
      "mean_token_accuracy": 0.18441122770309448,
      "num_tokens": 55435325.0,
      "step": 29735
    },
    {
      "entropy": 5.9789080142974855,
      "epoch": 2.626280466266337,
      "grad_norm": 1.4375,
      "learning_rate": 0.000431035151583966,
      "loss": 5.7268,
      "mean_token_accuracy": 0.17142765522003173,
      "num_tokens": 55444892.0,
      "step": 29740
    },
    {
      "entropy": 5.879587268829345,
      "epoch": 2.626722006358177,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004310124614750632,
      "loss": 5.5283,
      "mean_token_accuracy": 0.19676176309585572,
      "num_tokens": 55453816.0,
      "step": 29745
    },
    {
      "entropy": 6.022342395782471,
      "epoch": 2.6271635464500176,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004309897683099652,
      "loss": 5.7232,
      "mean_token_accuracy": 0.17657838612794877,
      "num_tokens": 55463411.0,
      "step": 29750
    },
    {
      "entropy": 6.074942016601563,
      "epoch": 2.627605086541858,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004309670720891162,
      "loss": 5.7826,
      "mean_token_accuracy": 0.169261172413826,
      "num_tokens": 55472939.0,
      "step": 29755
    },
    {
      "entropy": 5.942177438735962,
      "epoch": 2.628046626633698,
      "grad_norm": 1.59375,
      "learning_rate": 0.000430944372812961,
      "loss": 5.4866,
      "mean_token_accuracy": 0.19269143790006638,
      "num_tokens": 55481149.0,
      "step": 29760
    },
    {
      "entropy": 5.98435606956482,
      "epoch": 2.6284881667255386,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004309216704819442,
      "loss": 5.5614,
      "mean_token_accuracy": 0.18755978494882583,
      "num_tokens": 55489858.0,
      "step": 29765
    },
    {
      "entropy": 5.916850805282593,
      "epoch": 2.628929706817379,
      "grad_norm": 1.875,
      "learning_rate": 0.0004308989650965106,
      "loss": 5.6393,
      "mean_token_accuracy": 0.1864780828356743,
      "num_tokens": 55498835.0,
      "step": 29770
    },
    {
      "entropy": 5.910834074020386,
      "epoch": 2.629371246909219,
      "grad_norm": 1.625,
      "learning_rate": 0.0004308762566571049,
      "loss": 5.5364,
      "mean_token_accuracy": 0.1890222042798996,
      "num_tokens": 55507437.0,
      "step": 29775
    },
    {
      "entropy": 5.967218065261841,
      "epoch": 2.6298127870010597,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004308535451641721,
      "loss": 5.6649,
      "mean_token_accuracy": 0.18149922639131547,
      "num_tokens": 55516304.0,
      "step": 29780
    },
    {
      "entropy": 6.012179517745972,
      "epoch": 2.6302543270929,
      "grad_norm": 1.625,
      "learning_rate": 0.0004308308306181569,
      "loss": 5.5327,
      "mean_token_accuracy": 0.19417689442634584,
      "num_tokens": 55525720.0,
      "step": 29785
    },
    {
      "entropy": 5.9895923137664795,
      "epoch": 2.6306958671847402,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004308081130195042,
      "loss": 5.7241,
      "mean_token_accuracy": 0.17865446954965591,
      "num_tokens": 55534882.0,
      "step": 29790
    },
    {
      "entropy": 5.904460620880127,
      "epoch": 2.6311374072765807,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004307853923686594,
      "loss": 5.6125,
      "mean_token_accuracy": 0.18575142472982406,
      "num_tokens": 55544337.0,
      "step": 29795
    },
    {
      "entropy": 5.976598310470581,
      "epoch": 2.6315789473684212,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004307626686660673,
      "loss": 5.5972,
      "mean_token_accuracy": 0.18665324300527572,
      "num_tokens": 55553616.0,
      "step": 29800
    },
    {
      "entropy": 6.0816679954528805,
      "epoch": 2.6320204874602613,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00043073994191217307,
      "loss": 5.6858,
      "mean_token_accuracy": 0.181797157227993,
      "num_tokens": 55564066.0,
      "step": 29805
    },
    {
      "entropy": 5.962124681472778,
      "epoch": 2.632462027552102,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043071721210742196,
      "loss": 5.5677,
      "mean_token_accuracy": 0.19343754202127456,
      "num_tokens": 55573189.0,
      "step": 29810
    },
    {
      "entropy": 5.899409389495849,
      "epoch": 2.6329035676439423,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004306944792522593,
      "loss": 5.5966,
      "mean_token_accuracy": 0.1872014507651329,
      "num_tokens": 55581421.0,
      "step": 29815
    },
    {
      "entropy": 5.949683380126953,
      "epoch": 2.6333451077357823,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00043067174334713025,
      "loss": 5.6747,
      "mean_token_accuracy": 0.17413582652807236,
      "num_tokens": 55591323.0,
      "step": 29820
    },
    {
      "entropy": 5.998705959320068,
      "epoch": 2.633786647827623,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004306490043924803,
      "loss": 5.5856,
      "mean_token_accuracy": 0.19483235627412795,
      "num_tokens": 55600352.0,
      "step": 29825
    },
    {
      "entropy": 5.979430913925171,
      "epoch": 2.6342281879194633,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004306262623887549,
      "loss": 5.6603,
      "mean_token_accuracy": 0.18546532541513444,
      "num_tokens": 55609678.0,
      "step": 29830
    },
    {
      "entropy": 5.965701770782471,
      "epoch": 2.6346697280113034,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004306035173363995,
      "loss": 5.6981,
      "mean_token_accuracy": 0.18078883439302446,
      "num_tokens": 55619401.0,
      "step": 29835
    },
    {
      "entropy": 5.963696002960205,
      "epoch": 2.635111268103144,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043058076923585976,
      "loss": 5.6955,
      "mean_token_accuracy": 0.17777763456106185,
      "num_tokens": 55629208.0,
      "step": 29840
    },
    {
      "entropy": 5.964335918426514,
      "epoch": 2.635552808194984,
      "grad_norm": 1.5,
      "learning_rate": 0.00043055801808758123,
      "loss": 5.594,
      "mean_token_accuracy": 0.18333418816328048,
      "num_tokens": 55638304.0,
      "step": 29845
    },
    {
      "entropy": 5.9732481956481935,
      "epoch": 2.6359943482868244,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00043053526389200955,
      "loss": 5.6775,
      "mean_token_accuracy": 0.18471531867980956,
      "num_tokens": 55647982.0,
      "step": 29850
    },
    {
      "entropy": 5.942673397064209,
      "epoch": 2.636435888378665,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043051250664959053,
      "loss": 5.5904,
      "mean_token_accuracy": 0.18898673951625825,
      "num_tokens": 55657088.0,
      "step": 29855
    },
    {
      "entropy": 6.0154783725738525,
      "epoch": 2.636877428470505,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004304897463607699,
      "loss": 5.6731,
      "mean_token_accuracy": 0.1864723414182663,
      "num_tokens": 55666451.0,
      "step": 29860
    },
    {
      "entropy": 5.901706504821777,
      "epoch": 2.6373189685623455,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004304669830259936,
      "loss": 5.5859,
      "mean_token_accuracy": 0.18823574632406234,
      "num_tokens": 55675636.0,
      "step": 29865
    },
    {
      "entropy": 5.987693881988525,
      "epoch": 2.6377605086541855,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043044421664570755,
      "loss": 5.6498,
      "mean_token_accuracy": 0.18345523923635482,
      "num_tokens": 55686092.0,
      "step": 29870
    },
    {
      "entropy": 5.910808038711548,
      "epoch": 2.638202048746026,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00043042144722035765,
      "loss": 5.5933,
      "mean_token_accuracy": 0.1887090176343918,
      "num_tokens": 55695212.0,
      "step": 29875
    },
    {
      "entropy": 5.901487207412719,
      "epoch": 2.6386435888378665,
      "grad_norm": 1.578125,
      "learning_rate": 0.00043039867475039,
      "loss": 5.6231,
      "mean_token_accuracy": 0.19054369926452636,
      "num_tokens": 55704471.0,
      "step": 29880
    },
    {
      "entropy": 6.016774034500122,
      "epoch": 2.6390851289297066,
      "grad_norm": 1.5,
      "learning_rate": 0.0004303758992362507,
      "loss": 5.6776,
      "mean_token_accuracy": 0.17843406796455383,
      "num_tokens": 55712915.0,
      "step": 29885
    },
    {
      "entropy": 6.015658903121948,
      "epoch": 2.639526669021547,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004303531206783859,
      "loss": 5.6712,
      "mean_token_accuracy": 0.17948143929243088,
      "num_tokens": 55722045.0,
      "step": 29890
    },
    {
      "entropy": 5.957853746414185,
      "epoch": 2.6399682091133876,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004303303390772419,
      "loss": 5.627,
      "mean_token_accuracy": 0.1872076705098152,
      "num_tokens": 55731548.0,
      "step": 29895
    },
    {
      "entropy": 6.066502094268799,
      "epoch": 2.6404097492052276,
      "grad_norm": 1.5,
      "learning_rate": 0.00043030755443326476,
      "loss": 5.7237,
      "mean_token_accuracy": 0.17837947905063628,
      "num_tokens": 55741494.0,
      "step": 29900
    },
    {
      "entropy": 6.062026023864746,
      "epoch": 2.640851289297068,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043028476674690107,
      "loss": 5.6675,
      "mean_token_accuracy": 0.18555330038070678,
      "num_tokens": 55750000.0,
      "step": 29905
    },
    {
      "entropy": 5.976464128494262,
      "epoch": 2.6412928293889086,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004302619760185971,
      "loss": 5.6979,
      "mean_token_accuracy": 0.1822872504591942,
      "num_tokens": 55760375.0,
      "step": 29910
    },
    {
      "entropy": 5.979368257522583,
      "epoch": 2.6417343694807487,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004302391822487993,
      "loss": 5.6298,
      "mean_token_accuracy": 0.18376773446798325,
      "num_tokens": 55768726.0,
      "step": 29915
    },
    {
      "entropy": 5.96920371055603,
      "epoch": 2.642175909572589,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004302163854379541,
      "loss": 5.6928,
      "mean_token_accuracy": 0.186072501540184,
      "num_tokens": 55778203.0,
      "step": 29920
    },
    {
      "entropy": 5.962170839309692,
      "epoch": 2.6426174496644297,
      "grad_norm": 1.5,
      "learning_rate": 0.00043019358558650837,
      "loss": 5.5837,
      "mean_token_accuracy": 0.19004447609186173,
      "num_tokens": 55787719.0,
      "step": 29925
    },
    {
      "entropy": 5.894108438491822,
      "epoch": 2.6430589897562697,
      "grad_norm": 1.546875,
      "learning_rate": 0.00043017078269490846,
      "loss": 5.6155,
      "mean_token_accuracy": 0.18323493152856826,
      "num_tokens": 55796824.0,
      "step": 29930
    },
    {
      "entropy": 6.0078949451446535,
      "epoch": 2.6435005298481102,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043014797676360124,
      "loss": 5.6248,
      "mean_token_accuracy": 0.1904333934187889,
      "num_tokens": 55805460.0,
      "step": 29935
    },
    {
      "entropy": 5.862185621261597,
      "epoch": 2.6439420699399507,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004301251677930332,
      "loss": 5.6211,
      "mean_token_accuracy": 0.18867418467998504,
      "num_tokens": 55815112.0,
      "step": 29940
    },
    {
      "entropy": 5.969110107421875,
      "epoch": 2.644383610031791,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004301023557836516,
      "loss": 5.6175,
      "mean_token_accuracy": 0.1896526798605919,
      "num_tokens": 55823815.0,
      "step": 29945
    },
    {
      "entropy": 5.995712184906006,
      "epoch": 2.6448251501236313,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004300795407359029,
      "loss": 5.5981,
      "mean_token_accuracy": 0.1891697198152542,
      "num_tokens": 55833513.0,
      "step": 29950
    },
    {
      "entropy": 5.9644904136657715,
      "epoch": 2.645266690215472,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043005672265023425,
      "loss": 5.5923,
      "mean_token_accuracy": 0.19112141877412797,
      "num_tokens": 55841765.0,
      "step": 29955
    },
    {
      "entropy": 5.960331439971924,
      "epoch": 2.645708230307312,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043003390152709264,
      "loss": 5.6786,
      "mean_token_accuracy": 0.18605118244886398,
      "num_tokens": 55851125.0,
      "step": 29960
    },
    {
      "entropy": 5.971808242797851,
      "epoch": 2.6461497703991523,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043001107736692504,
      "loss": 5.6425,
      "mean_token_accuracy": 0.18813201785087585,
      "num_tokens": 55861826.0,
      "step": 29965
    },
    {
      "entropy": 6.057372283935547,
      "epoch": 2.646591310490993,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004299882501701785,
      "loss": 5.6615,
      "mean_token_accuracy": 0.18393197655677795,
      "num_tokens": 55870658.0,
      "step": 29970
    },
    {
      "entropy": 5.953923845291138,
      "epoch": 2.647032850582833,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042996541993730035,
      "loss": 5.6572,
      "mean_token_accuracy": 0.18310242295265197,
      "num_tokens": 55880228.0,
      "step": 29975
    },
    {
      "entropy": 5.928289985656738,
      "epoch": 2.6474743906746734,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042994258666873773,
      "loss": 5.7066,
      "mean_token_accuracy": 0.17898865640163422,
      "num_tokens": 55890101.0,
      "step": 29980
    },
    {
      "entropy": 6.010761547088623,
      "epoch": 2.6479159307665134,
      "grad_norm": 1.234375,
      "learning_rate": 0.000429919750364938,
      "loss": 5.588,
      "mean_token_accuracy": 0.19042495340108873,
      "num_tokens": 55899164.0,
      "step": 29985
    },
    {
      "entropy": 6.006376314163208,
      "epoch": 2.648357470858354,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042989691102634847,
      "loss": 5.6113,
      "mean_token_accuracy": 0.18787094652652742,
      "num_tokens": 55907735.0,
      "step": 29990
    },
    {
      "entropy": 6.006076049804688,
      "epoch": 2.6487990109501944,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042987406865341657,
      "loss": 5.7478,
      "mean_token_accuracy": 0.17543884813785554,
      "num_tokens": 55917374.0,
      "step": 29995
    },
    {
      "entropy": 5.986318922042846,
      "epoch": 2.6492405510420345,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004298512232465897,
      "loss": 5.676,
      "mean_token_accuracy": 0.17448736131191253,
      "num_tokens": 55926430.0,
      "step": 30000
    },
    {
      "epoch": 2.6492405510420345,
      "eval_entropy": 5.766285487830571,
      "eval_loss": 5.712087154388428,
      "eval_mean_token_accuracy": 0.1895079952062425,
      "eval_num_tokens": 55926430.0,
      "eval_runtime": 26.2902,
      "eval_samples_per_second": 1343.275,
      "eval_steps_per_second": 167.933,
      "step": 30000
    },
    {
      "entropy": 5.980271816253662,
      "epoch": 2.649682091133875,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004298283748063154,
      "loss": 5.659,
      "mean_token_accuracy": 0.18224591314792632,
      "num_tokens": 55935805.0,
      "step": 30005
    },
    {
      "entropy": 6.004607915878296,
      "epoch": 2.650123631225715,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004298055233330414,
      "loss": 5.6608,
      "mean_token_accuracy": 0.18580012619495392,
      "num_tokens": 55945278.0,
      "step": 30010
    },
    {
      "entropy": 5.958535861968994,
      "epoch": 2.6505651713175555,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004297826688272151,
      "loss": 5.6186,
      "mean_token_accuracy": 0.19107327461242676,
      "num_tokens": 55954316.0,
      "step": 30015
    },
    {
      "entropy": 5.918033218383789,
      "epoch": 2.651006711409396,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004297598112892845,
      "loss": 5.5357,
      "mean_token_accuracy": 0.19617455303668976,
      "num_tokens": 55963249.0,
      "step": 30020
    },
    {
      "entropy": 5.9412986755371096,
      "epoch": 2.651448251501236,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042973695071969715,
      "loss": 5.608,
      "mean_token_accuracy": 0.19030253887176513,
      "num_tokens": 55972010.0,
      "step": 30025
    },
    {
      "entropy": 5.91542911529541,
      "epoch": 2.6518897915930766,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004297140871189009,
      "loss": 5.5381,
      "mean_token_accuracy": 0.19405810832977294,
      "num_tokens": 55981197.0,
      "step": 30030
    },
    {
      "entropy": 6.008786582946778,
      "epoch": 2.652331331684917,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004296912204873437,
      "loss": 5.7187,
      "mean_token_accuracy": 0.178082737326622,
      "num_tokens": 55991367.0,
      "step": 30035
    },
    {
      "entropy": 6.024305200576782,
      "epoch": 2.652772871776757,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042966835082547346,
      "loss": 5.5839,
      "mean_token_accuracy": 0.19056221097707748,
      "num_tokens": 56000614.0,
      "step": 30040
    },
    {
      "entropy": 5.843510293960572,
      "epoch": 2.6532144118685976,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004296454781337382,
      "loss": 5.5679,
      "mean_token_accuracy": 0.1937951475381851,
      "num_tokens": 56009786.0,
      "step": 30045
    },
    {
      "entropy": 5.973918724060058,
      "epoch": 2.653655951960438,
      "grad_norm": 1.203125,
      "learning_rate": 0.000429622602412586,
      "loss": 5.6546,
      "mean_token_accuracy": 0.18457321524620057,
      "num_tokens": 56019696.0,
      "step": 30050
    },
    {
      "entropy": 6.041146659851075,
      "epoch": 2.654097492052278,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042959972366246493,
      "loss": 5.5962,
      "mean_token_accuracy": 0.19066880345344545,
      "num_tokens": 56027960.0,
      "step": 30055
    },
    {
      "entropy": 5.98619441986084,
      "epoch": 2.6545390321441187,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004295768418838232,
      "loss": 5.6929,
      "mean_token_accuracy": 0.173959681391716,
      "num_tokens": 56036875.0,
      "step": 30060
    },
    {
      "entropy": 5.999112462997436,
      "epoch": 2.654980572235959,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004295539570771091,
      "loss": 5.686,
      "mean_token_accuracy": 0.18115017116069793,
      "num_tokens": 56045916.0,
      "step": 30065
    },
    {
      "entropy": 6.014813804626465,
      "epoch": 2.6554221123277992,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004295310692427708,
      "loss": 5.6874,
      "mean_token_accuracy": 0.18150765746831893,
      "num_tokens": 56056048.0,
      "step": 30070
    },
    {
      "entropy": 5.974394416809082,
      "epoch": 2.6558636524196397,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042950817838125675,
      "loss": 5.6162,
      "mean_token_accuracy": 0.18476945608854295,
      "num_tokens": 56065455.0,
      "step": 30075
    },
    {
      "entropy": 5.971304559707642,
      "epoch": 2.6563051925114802,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004294852844930154,
      "loss": 5.646,
      "mean_token_accuracy": 0.18608921021223068,
      "num_tokens": 56074070.0,
      "step": 30080
    },
    {
      "entropy": 6.004162359237671,
      "epoch": 2.6567467326033203,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042946238757849515,
      "loss": 5.6734,
      "mean_token_accuracy": 0.18697351813316346,
      "num_tokens": 56083518.0,
      "step": 30085
    },
    {
      "entropy": 5.986919975280761,
      "epoch": 2.657188272695161,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042943948763814465,
      "loss": 5.6321,
      "mean_token_accuracy": 0.1873683825135231,
      "num_tokens": 56092591.0,
      "step": 30090
    },
    {
      "entropy": 6.049514722824097,
      "epoch": 2.6576298127870013,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004294165846724123,
      "loss": 5.8351,
      "mean_token_accuracy": 0.1728338047862053,
      "num_tokens": 56102120.0,
      "step": 30095
    },
    {
      "entropy": 6.07698073387146,
      "epoch": 2.6580713528788413,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000429393678681747,
      "loss": 5.641,
      "mean_token_accuracy": 0.18589580953121185,
      "num_tokens": 56111885.0,
      "step": 30100
    },
    {
      "entropy": 5.90182933807373,
      "epoch": 2.658512892970682,
      "grad_norm": 1.578125,
      "learning_rate": 0.00042937076966659726,
      "loss": 5.6009,
      "mean_token_accuracy": 0.19249369353055953,
      "num_tokens": 56122070.0,
      "step": 30105
    },
    {
      "entropy": 5.949805974960327,
      "epoch": 2.6589544330625223,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042934785762741197,
      "loss": 5.6293,
      "mean_token_accuracy": 0.19270943999290466,
      "num_tokens": 56132142.0,
      "step": 30110
    },
    {
      "entropy": 6.011702632904052,
      "epoch": 2.6593959731543624,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004293249425646399,
      "loss": 5.6564,
      "mean_token_accuracy": 0.18297690749168397,
      "num_tokens": 56141860.0,
      "step": 30115
    },
    {
      "entropy": 5.959278106689453,
      "epoch": 2.659837513246203,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042930202447873003,
      "loss": 5.5512,
      "mean_token_accuracy": 0.1891790360212326,
      "num_tokens": 56152035.0,
      "step": 30120
    },
    {
      "entropy": 6.001845598220825,
      "epoch": 2.660279053338043,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004292791033701312,
      "loss": 5.6821,
      "mean_token_accuracy": 0.1833362877368927,
      "num_tokens": 56161266.0,
      "step": 30125
    },
    {
      "entropy": 5.986358547210694,
      "epoch": 2.6607205934298834,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004292561792392926,
      "loss": 5.6945,
      "mean_token_accuracy": 0.17946010828018188,
      "num_tokens": 56171860.0,
      "step": 30130
    },
    {
      "entropy": 6.000863265991211,
      "epoch": 2.661162133521724,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042923325208666307,
      "loss": 5.6775,
      "mean_token_accuracy": 0.1923846945166588,
      "num_tokens": 56180492.0,
      "step": 30135
    },
    {
      "entropy": 5.939663600921631,
      "epoch": 2.661603673613564,
      "grad_norm": 4.71875,
      "learning_rate": 0.00042921032191269183,
      "loss": 5.598,
      "mean_token_accuracy": 0.18502826988697052,
      "num_tokens": 56190292.0,
      "step": 30140
    },
    {
      "entropy": 6.058265590667725,
      "epoch": 2.6620452137054045,
      "grad_norm": 2.140625,
      "learning_rate": 0.00042918738871782814,
      "loss": 5.7454,
      "mean_token_accuracy": 0.17608426064252852,
      "num_tokens": 56200270.0,
      "step": 30145
    },
    {
      "entropy": 6.010371160507202,
      "epoch": 2.6624867537972445,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004291644525025212,
      "loss": 5.6997,
      "mean_token_accuracy": 0.18205874562263488,
      "num_tokens": 56210876.0,
      "step": 30150
    },
    {
      "entropy": 5.9958912372589115,
      "epoch": 2.662928293889085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042914151326722025,
      "loss": 5.6425,
      "mean_token_accuracy": 0.18458113223314285,
      "num_tokens": 56221100.0,
      "step": 30155
    },
    {
      "entropy": 6.006061124801636,
      "epoch": 2.6633698339809255,
      "grad_norm": 1.375,
      "learning_rate": 0.0004291185710123748,
      "loss": 5.6913,
      "mean_token_accuracy": 0.1766015335917473,
      "num_tokens": 56230404.0,
      "step": 30160
    },
    {
      "entropy": 5.98223991394043,
      "epoch": 2.6638113740727656,
      "grad_norm": 1.5,
      "learning_rate": 0.00042909562573843416,
      "loss": 5.5344,
      "mean_token_accuracy": 0.20183046758174897,
      "num_tokens": 56239120.0,
      "step": 30165
    },
    {
      "entropy": 5.996742582321167,
      "epoch": 2.664252914164606,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00042907267744584775,
      "loss": 5.6483,
      "mean_token_accuracy": 0.19058796614408494,
      "num_tokens": 56248979.0,
      "step": 30170
    },
    {
      "entropy": 5.92452039718628,
      "epoch": 2.6646944542564466,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00042904972613506536,
      "loss": 5.6866,
      "mean_token_accuracy": 0.1813066080212593,
      "num_tokens": 56258133.0,
      "step": 30175
    },
    {
      "entropy": 6.022087526321411,
      "epoch": 2.6651359943482866,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042902677180653643,
      "loss": 5.7412,
      "mean_token_accuracy": 0.1800127163529396,
      "num_tokens": 56267806.0,
      "step": 30180
    },
    {
      "entropy": 6.050100231170655,
      "epoch": 2.665577534440127,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004290038144607106,
      "loss": 5.6832,
      "mean_token_accuracy": 0.18016895949840545,
      "num_tokens": 56276993.0,
      "step": 30185
    },
    {
      "entropy": 5.849536466598511,
      "epoch": 2.6660190745319676,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00042898085409803766,
      "loss": 5.4375,
      "mean_token_accuracy": 0.2016384243965149,
      "num_tokens": 56286716.0,
      "step": 30190
    },
    {
      "entropy": 5.894965934753418,
      "epoch": 2.6664606146238077,
      "grad_norm": 1.75,
      "learning_rate": 0.0004289578907189673,
      "loss": 5.5893,
      "mean_token_accuracy": 0.1888311207294464,
      "num_tokens": 56296194.0,
      "step": 30195
    },
    {
      "entropy": 5.929652833938599,
      "epoch": 2.666902154715648,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042893492432394953,
      "loss": 5.595,
      "mean_token_accuracy": 0.1883564129471779,
      "num_tokens": 56304855.0,
      "step": 30200
    },
    {
      "entropy": 6.028350973129273,
      "epoch": 2.6673436948074887,
      "grad_norm": 1.515625,
      "learning_rate": 0.00042891195491343405,
      "loss": 5.704,
      "mean_token_accuracy": 0.17397406548261643,
      "num_tokens": 56313869.0,
      "step": 30205
    },
    {
      "entropy": 6.0456747055053714,
      "epoch": 2.6677852348993287,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00042888898248787094,
      "loss": 5.6452,
      "mean_token_accuracy": 0.18239085674285888,
      "num_tokens": 56323551.0,
      "step": 30210
    },
    {
      "entropy": 6.048734712600708,
      "epoch": 2.6682267749911692,
      "grad_norm": 1.5,
      "learning_rate": 0.0004288660070477101,
      "loss": 5.654,
      "mean_token_accuracy": 0.18312640786170958,
      "num_tokens": 56333068.0,
      "step": 30215
    },
    {
      "entropy": 5.9555257797241214,
      "epoch": 2.6686683150830097,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042884302859340183,
      "loss": 5.5627,
      "mean_token_accuracy": 0.19285908788442613,
      "num_tokens": 56341952.0,
      "step": 30220
    },
    {
      "entropy": 5.83759241104126,
      "epoch": 2.6691098551748498,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00042882004712539615,
      "loss": 5.5016,
      "mean_token_accuracy": 0.18705224692821504,
      "num_tokens": 56351829.0,
      "step": 30225
    },
    {
      "entropy": 6.002959251403809,
      "epoch": 2.6695513952666903,
      "grad_norm": 1.25,
      "learning_rate": 0.00042879706264414304,
      "loss": 5.6841,
      "mean_token_accuracy": 0.18468658924102782,
      "num_tokens": 56361290.0,
      "step": 30230
    },
    {
      "entropy": 5.904333448410034,
      "epoch": 2.6699929353585308,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004287740751500931,
      "loss": 5.6412,
      "mean_token_accuracy": 0.19715138971805574,
      "num_tokens": 56371582.0,
      "step": 30235
    },
    {
      "entropy": 5.898732614517212,
      "epoch": 2.670434475450371,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042875108464369646,
      "loss": 5.506,
      "mean_token_accuracy": 0.19524157792329788,
      "num_tokens": 56380637.0,
      "step": 30240
    },
    {
      "entropy": 5.949707269668579,
      "epoch": 2.6708760155422113,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042872809112540347,
      "loss": 5.6866,
      "mean_token_accuracy": 0.18220334947109224,
      "num_tokens": 56390332.0,
      "step": 30245
    },
    {
      "entropy": 5.967242050170898,
      "epoch": 2.671317555634052,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004287050945956646,
      "loss": 5.6041,
      "mean_token_accuracy": 0.18118558675050736,
      "num_tokens": 56399379.0,
      "step": 30250
    },
    {
      "entropy": 5.886063957214356,
      "epoch": 2.671759095725892,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042868209505493034,
      "loss": 5.5653,
      "mean_token_accuracy": 0.1936924695968628,
      "num_tokens": 56408645.0,
      "step": 30255
    },
    {
      "entropy": 5.893675851821899,
      "epoch": 2.6722006358177324,
      "grad_norm": 1.875,
      "learning_rate": 0.0004286590925036513,
      "loss": 5.5749,
      "mean_token_accuracy": 0.19342378079891204,
      "num_tokens": 56417119.0,
      "step": 30260
    },
    {
      "entropy": 5.946434783935547,
      "epoch": 2.6726421759095724,
      "grad_norm": 1.796875,
      "learning_rate": 0.000428636086942278,
      "loss": 5.6542,
      "mean_token_accuracy": 0.1853835478425026,
      "num_tokens": 56426443.0,
      "step": 30265
    },
    {
      "entropy": 6.010185194015503,
      "epoch": 2.673083716001413,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004286130783712611,
      "loss": 5.7579,
      "mean_token_accuracy": 0.17260287702083588,
      "num_tokens": 56435675.0,
      "step": 30270
    },
    {
      "entropy": 6.006824207305908,
      "epoch": 2.6735252560932534,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004285900667910514,
      "loss": 5.6445,
      "mean_token_accuracy": 0.17835439890623092,
      "num_tokens": 56445787.0,
      "step": 30275
    },
    {
      "entropy": 6.055143213272094,
      "epoch": 2.6739667961850935,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004285670522020996,
      "loss": 5.6313,
      "mean_token_accuracy": 0.19182975739240646,
      "num_tokens": 56455029.0,
      "step": 30280
    },
    {
      "entropy": 6.05564866065979,
      "epoch": 2.674408336276934,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004285440346048566,
      "loss": 5.6725,
      "mean_token_accuracy": 0.17741392999887468,
      "num_tokens": 56464566.0,
      "step": 30285
    },
    {
      "entropy": 5.940537691116333,
      "epoch": 2.674849876368774,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004285210139997733,
      "loss": 5.6714,
      "mean_token_accuracy": 0.1845533162355423,
      "num_tokens": 56473350.0,
      "step": 30290
    },
    {
      "entropy": 5.954421901702881,
      "epoch": 2.6752914164606145,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004284979903873007,
      "loss": 5.6094,
      "mean_token_accuracy": 0.1917117342352867,
      "num_tokens": 56482841.0,
      "step": 30295
    },
    {
      "entropy": 6.038290882110596,
      "epoch": 2.675732956552455,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042847496376788964,
      "loss": 5.7357,
      "mean_token_accuracy": 0.17877039760351182,
      "num_tokens": 56492851.0,
      "step": 30300
    },
    {
      "entropy": 6.080140590667725,
      "epoch": 2.676174496644295,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042845193414199146,
      "loss": 5.7165,
      "mean_token_accuracy": 0.1785028949379921,
      "num_tokens": 56501514.0,
      "step": 30305
    },
    {
      "entropy": 6.0035364627838135,
      "epoch": 2.6766160367361356,
      "grad_norm": 1.578125,
      "learning_rate": 0.00042842890151005705,
      "loss": 5.5864,
      "mean_token_accuracy": 0.18511435836553575,
      "num_tokens": 56509963.0,
      "step": 30310
    },
    {
      "entropy": 5.881657505035401,
      "epoch": 2.677057576827976,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042840586587253776,
      "loss": 5.6495,
      "mean_token_accuracy": 0.1839665040373802,
      "num_tokens": 56519598.0,
      "step": 30315
    },
    {
      "entropy": 5.935123109817505,
      "epoch": 2.677499116919816,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042838282722988484,
      "loss": 5.5905,
      "mean_token_accuracy": 0.1845237284898758,
      "num_tokens": 56528275.0,
      "step": 30320
    },
    {
      "entropy": 6.016073274612427,
      "epoch": 2.6779406570116566,
      "grad_norm": 1.4375,
      "learning_rate": 0.00042835978558254954,
      "loss": 5.658,
      "mean_token_accuracy": 0.18424589633941652,
      "num_tokens": 56537974.0,
      "step": 30325
    },
    {
      "entropy": 6.009022855758667,
      "epoch": 2.678382197103497,
      "grad_norm": 1.25,
      "learning_rate": 0.0004283367409309833,
      "loss": 5.6906,
      "mean_token_accuracy": 0.18134088665246964,
      "num_tokens": 56546262.0,
      "step": 30330
    },
    {
      "entropy": 6.009977149963379,
      "epoch": 2.678823737195337,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004283136932756375,
      "loss": 5.6426,
      "mean_token_accuracy": 0.18215729296207428,
      "num_tokens": 56556564.0,
      "step": 30335
    },
    {
      "entropy": 5.901270627975464,
      "epoch": 2.6792652772871777,
      "grad_norm": 1.25,
      "learning_rate": 0.0004282906426169636,
      "loss": 5.4982,
      "mean_token_accuracy": 0.20102401822805405,
      "num_tokens": 56565693.0,
      "step": 30340
    },
    {
      "entropy": 5.947522401809692,
      "epoch": 2.679706817379018,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004282675889554132,
      "loss": 5.633,
      "mean_token_accuracy": 0.19330906569957734,
      "num_tokens": 56575693.0,
      "step": 30345
    },
    {
      "entropy": 6.020604085922241,
      "epoch": 2.6801483574708582,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004282445322914379,
      "loss": 5.6556,
      "mean_token_accuracy": 0.18457947075366973,
      "num_tokens": 56585147.0,
      "step": 30350
    },
    {
      "entropy": 5.98465313911438,
      "epoch": 2.6805898975626987,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042822147262548937,
      "loss": 5.7196,
      "mean_token_accuracy": 0.17533523887395858,
      "num_tokens": 56595841.0,
      "step": 30355
    },
    {
      "entropy": 5.938784790039063,
      "epoch": 2.681031437654539,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004281984099580194,
      "loss": 5.6234,
      "mean_token_accuracy": 0.17628976702690125,
      "num_tokens": 56606226.0,
      "step": 30360
    },
    {
      "entropy": 6.054877138137817,
      "epoch": 2.6814729777463793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004281753442894796,
      "loss": 5.6342,
      "mean_token_accuracy": 0.1862315282225609,
      "num_tokens": 56614821.0,
      "step": 30365
    },
    {
      "entropy": 5.897002935409546,
      "epoch": 2.6819145178382198,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000428152275620322,
      "loss": 5.5662,
      "mean_token_accuracy": 0.1962735489010811,
      "num_tokens": 56623489.0,
      "step": 30370
    },
    {
      "entropy": 5.918709135055542,
      "epoch": 2.6823560579300603,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004281292039509984,
      "loss": 5.6444,
      "mean_token_accuracy": 0.1909630924463272,
      "num_tokens": 56633071.0,
      "step": 30375
    },
    {
      "entropy": 6.097145128250122,
      "epoch": 2.6827975980219003,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004281061292819608,
      "loss": 5.748,
      "mean_token_accuracy": 0.17383075952529908,
      "num_tokens": 56643316.0,
      "step": 30380
    },
    {
      "entropy": 6.0514171600341795,
      "epoch": 2.683239138113741,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004280830516136611,
      "loss": 5.5989,
      "mean_token_accuracy": 0.1918766126036644,
      "num_tokens": 56652690.0,
      "step": 30385
    },
    {
      "entropy": 5.853613424301147,
      "epoch": 2.6836806782055813,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042805997094655157,
      "loss": 5.5358,
      "mean_token_accuracy": 0.19114202708005906,
      "num_tokens": 56662442.0,
      "step": 30390
    },
    {
      "entropy": 5.930671691894531,
      "epoch": 2.6841222182974214,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004280368872810843,
      "loss": 5.6629,
      "mean_token_accuracy": 0.1811872750520706,
      "num_tokens": 56672308.0,
      "step": 30395
    },
    {
      "entropy": 5.996845769882202,
      "epoch": 2.684563758389262,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004280138006177114,
      "loss": 5.6413,
      "mean_token_accuracy": 0.1843100905418396,
      "num_tokens": 56682193.0,
      "step": 30400
    },
    {
      "entropy": 6.048449611663818,
      "epoch": 2.685005298481102,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042799071095688516,
      "loss": 5.6453,
      "mean_token_accuracy": 0.1797408491373062,
      "num_tokens": 56690856.0,
      "step": 30405
    },
    {
      "entropy": 5.936465883255005,
      "epoch": 2.6854468385729424,
      "grad_norm": 1.546875,
      "learning_rate": 0.00042796761829905797,
      "loss": 5.6543,
      "mean_token_accuracy": 0.18128087520599365,
      "num_tokens": 56699998.0,
      "step": 30410
    },
    {
      "entropy": 5.894102668762207,
      "epoch": 2.685888378664783,
      "grad_norm": 1.46875,
      "learning_rate": 0.000427944522644682,
      "loss": 5.6221,
      "mean_token_accuracy": 0.18873292356729507,
      "num_tokens": 56708977.0,
      "step": 30415
    },
    {
      "entropy": 6.024966478347778,
      "epoch": 2.686329918756623,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004279214239942099,
      "loss": 5.6639,
      "mean_token_accuracy": 0.18986508399248123,
      "num_tokens": 56718141.0,
      "step": 30420
    },
    {
      "entropy": 6.055437803268433,
      "epoch": 2.6867714588484635,
      "grad_norm": 1.84375,
      "learning_rate": 0.00042789832234809414,
      "loss": 5.6627,
      "mean_token_accuracy": 0.1791806325316429,
      "num_tokens": 56727211.0,
      "step": 30425
    },
    {
      "entropy": 5.970885992050171,
      "epoch": 2.6872129989403035,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00042787521770678704,
      "loss": 5.7346,
      "mean_token_accuracy": 0.17968770414590834,
      "num_tokens": 56737027.0,
      "step": 30430
    },
    {
      "entropy": 5.9266167163848875,
      "epoch": 2.687654539032144,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042785211007074147,
      "loss": 5.5563,
      "mean_token_accuracy": 0.1767723262310028,
      "num_tokens": 56745957.0,
      "step": 30435
    },
    {
      "entropy": 5.988842105865478,
      "epoch": 2.6880960791239845,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042782899944041004,
      "loss": 5.5575,
      "mean_token_accuracy": 0.1837415099143982,
      "num_tokens": 56755487.0,
      "step": 30440
    },
    {
      "entropy": 5.957640695571899,
      "epoch": 2.6885376192158246,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00042780588581624535,
      "loss": 5.702,
      "mean_token_accuracy": 0.18367446064949036,
      "num_tokens": 56765606.0,
      "step": 30445
    },
    {
      "entropy": 6.003440046310425,
      "epoch": 2.688979159307665,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004277827691987003,
      "loss": 5.6096,
      "mean_token_accuracy": 0.18620555400848388,
      "num_tokens": 56775015.0,
      "step": 30450
    },
    {
      "entropy": 6.000748109817505,
      "epoch": 2.6894206993995056,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042775964958822766,
      "loss": 5.7714,
      "mean_token_accuracy": 0.17311763465404512,
      "num_tokens": 56784166.0,
      "step": 30455
    },
    {
      "entropy": 6.03564567565918,
      "epoch": 2.6898622394913456,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004277365269852804,
      "loss": 5.6907,
      "mean_token_accuracy": 0.18298422396183014,
      "num_tokens": 56793146.0,
      "step": 30460
    },
    {
      "entropy": 5.935494136810303,
      "epoch": 2.690303779583186,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004277134013903114,
      "loss": 5.5363,
      "mean_token_accuracy": 0.20109089761972426,
      "num_tokens": 56802041.0,
      "step": 30465
    },
    {
      "entropy": 5.967216110229492,
      "epoch": 2.6907453196750266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042769027280377376,
      "loss": 5.702,
      "mean_token_accuracy": 0.1822405606508255,
      "num_tokens": 56810981.0,
      "step": 30470
    },
    {
      "entropy": 5.948998928070068,
      "epoch": 2.6911868597668667,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004276671412261205,
      "loss": 5.5591,
      "mean_token_accuracy": 0.18956681936979294,
      "num_tokens": 56820591.0,
      "step": 30475
    },
    {
      "entropy": 5.983471918106079,
      "epoch": 2.691628399858707,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004276440066578048,
      "loss": 5.6667,
      "mean_token_accuracy": 0.19120879769325255,
      "num_tokens": 56830376.0,
      "step": 30480
    },
    {
      "entropy": 6.060791301727295,
      "epoch": 2.6920699399505477,
      "grad_norm": 1.25,
      "learning_rate": 0.00042762086909927977,
      "loss": 5.7315,
      "mean_token_accuracy": 0.17655886858701705,
      "num_tokens": 56839487.0,
      "step": 30485
    },
    {
      "entropy": 5.983880662918091,
      "epoch": 2.6925114800423877,
      "grad_norm": 1.25,
      "learning_rate": 0.0004275977285509987,
      "loss": 5.6774,
      "mean_token_accuracy": 0.18822723627090454,
      "num_tokens": 56849570.0,
      "step": 30490
    },
    {
      "entropy": 5.943186044692993,
      "epoch": 2.692953020134228,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000427574585013415,
      "loss": 5.6345,
      "mean_token_accuracy": 0.1905589669942856,
      "num_tokens": 56858506.0,
      "step": 30495
    },
    {
      "entropy": 6.007855749130249,
      "epoch": 2.6933945602260687,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042755143848698194,
      "loss": 5.6296,
      "mean_token_accuracy": 0.1872812733054161,
      "num_tokens": 56868276.0,
      "step": 30500
    },
    {
      "entropy": 5.958961343765258,
      "epoch": 2.6938361003179088,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042752828897215295,
      "loss": 5.6381,
      "mean_token_accuracy": 0.1882803499698639,
      "num_tokens": 56877434.0,
      "step": 30505
    },
    {
      "entropy": 6.052465534210205,
      "epoch": 2.6942776404097493,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004275051364693815,
      "loss": 5.6733,
      "mean_token_accuracy": 0.17874758392572404,
      "num_tokens": 56888186.0,
      "step": 30510
    },
    {
      "entropy": 5.967326831817627,
      "epoch": 2.6947191805015898,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00042748198097912113,
      "loss": 5.5313,
      "mean_token_accuracy": 0.1905076324939728,
      "num_tokens": 56897125.0,
      "step": 30515
    },
    {
      "entropy": 5.921041774749756,
      "epoch": 2.69516072059343,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004274588225018255,
      "loss": 5.6574,
      "mean_token_accuracy": 0.1769066035747528,
      "num_tokens": 56906305.0,
      "step": 30520
    },
    {
      "entropy": 6.047793292999268,
      "epoch": 2.6956022606852703,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004274356610379483,
      "loss": 5.6436,
      "mean_token_accuracy": 0.17551755011081696,
      "num_tokens": 56915645.0,
      "step": 30525
    },
    {
      "entropy": 5.966585874557495,
      "epoch": 2.696043800777111,
      "grad_norm": 1.875,
      "learning_rate": 0.0004274124965879431,
      "loss": 5.6009,
      "mean_token_accuracy": 0.19003041237592697,
      "num_tokens": 56924567.0,
      "step": 30530
    },
    {
      "entropy": 5.991842365264892,
      "epoch": 2.696485340868951,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004273893291522638,
      "loss": 5.6524,
      "mean_token_accuracy": 0.17876754254102706,
      "num_tokens": 56933864.0,
      "step": 30535
    },
    {
      "entropy": 5.949660062789917,
      "epoch": 2.6969268809607914,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042736615873136425,
      "loss": 5.606,
      "mean_token_accuracy": 0.1926262080669403,
      "num_tokens": 56943150.0,
      "step": 30540
    },
    {
      "entropy": 5.892768478393554,
      "epoch": 2.6973684210526314,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00042734298532569825,
      "loss": 5.5818,
      "mean_token_accuracy": 0.19171915352344512,
      "num_tokens": 56951775.0,
      "step": 30545
    },
    {
      "entropy": 6.014314222335815,
      "epoch": 2.697809961144472,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042731980893571985,
      "loss": 5.6994,
      "mean_token_accuracy": 0.1829576835036278,
      "num_tokens": 56960827.0,
      "step": 30550
    },
    {
      "entropy": 6.074637413024902,
      "epoch": 2.6982515012363124,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000427296629561883,
      "loss": 5.6067,
      "mean_token_accuracy": 0.1907670497894287,
      "num_tokens": 56968966.0,
      "step": 30555
    },
    {
      "entropy": 6.021270036697388,
      "epoch": 2.6986930413281525,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004272734472046419,
      "loss": 5.6349,
      "mean_token_accuracy": 0.1908511236310005,
      "num_tokens": 56977818.0,
      "step": 30560
    },
    {
      "entropy": 5.9298337459564205,
      "epoch": 2.699134581419993,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042725026186445043,
      "loss": 5.596,
      "mean_token_accuracy": 0.19260309636592865,
      "num_tokens": 56986641.0,
      "step": 30565
    },
    {
      "entropy": 6.001609373092651,
      "epoch": 2.699576121511833,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042722707354176303,
      "loss": 5.6595,
      "mean_token_accuracy": 0.18372992873191835,
      "num_tokens": 56995806.0,
      "step": 30570
    },
    {
      "entropy": 5.9681604385375975,
      "epoch": 2.7000176616036735,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004272038822370338,
      "loss": 5.6699,
      "mean_token_accuracy": 0.1817266047000885,
      "num_tokens": 57004701.0,
      "step": 30575
    },
    {
      "entropy": 5.974004697799683,
      "epoch": 2.700459201695514,
      "grad_norm": 1.234375,
      "learning_rate": 0.000427180687950717,
      "loss": 5.5486,
      "mean_token_accuracy": 0.19064116477966309,
      "num_tokens": 57014260.0,
      "step": 30580
    },
    {
      "entropy": 5.966245031356811,
      "epoch": 2.700900741787354,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042715749068326713,
      "loss": 5.6726,
      "mean_token_accuracy": 0.183436881005764,
      "num_tokens": 57023057.0,
      "step": 30585
    },
    {
      "entropy": 6.022824239730835,
      "epoch": 2.7013422818791946,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00042713429043513863,
      "loss": 5.6637,
      "mean_token_accuracy": 0.18408044278621674,
      "num_tokens": 57032046.0,
      "step": 30590
    },
    {
      "entropy": 5.940027332305908,
      "epoch": 2.701783821971035,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004271110872067858,
      "loss": 5.5847,
      "mean_token_accuracy": 0.18672652691602706,
      "num_tokens": 57041473.0,
      "step": 30595
    },
    {
      "entropy": 5.946277809143067,
      "epoch": 2.702225362062875,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042708788099866333,
      "loss": 5.6444,
      "mean_token_accuracy": 0.19151431918144227,
      "num_tokens": 57050098.0,
      "step": 30600
    },
    {
      "entropy": 5.983521556854248,
      "epoch": 2.7026669021547156,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00042706467181122574,
      "loss": 5.5329,
      "mean_token_accuracy": 0.19342079758644104,
      "num_tokens": 57059922.0,
      "step": 30605
    },
    {
      "entropy": 5.973619651794434,
      "epoch": 2.703108442246556,
      "grad_norm": 1.671875,
      "learning_rate": 0.00042704145964492766,
      "loss": 5.6374,
      "mean_token_accuracy": 0.18598273098468782,
      "num_tokens": 57069841.0,
      "step": 30610
    },
    {
      "entropy": 5.9694733142852785,
      "epoch": 2.703549982338396,
      "grad_norm": 1.453125,
      "learning_rate": 0.000427018244500224,
      "loss": 5.6543,
      "mean_token_accuracy": 0.18320136666297912,
      "num_tokens": 57078891.0,
      "step": 30615
    },
    {
      "entropy": 5.935983371734619,
      "epoch": 2.7039915224302367,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042699502637756924,
      "loss": 5.5396,
      "mean_token_accuracy": 0.20322292149066926,
      "num_tokens": 57087899.0,
      "step": 30620
    },
    {
      "entropy": 5.928938293457032,
      "epoch": 2.704433062522077,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004269718052774185,
      "loss": 5.6891,
      "mean_token_accuracy": 0.18264323472976685,
      "num_tokens": 57096169.0,
      "step": 30625
    },
    {
      "entropy": 5.986781740188599,
      "epoch": 2.704874602613917,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004269485812002264,
      "loss": 5.597,
      "mean_token_accuracy": 0.1918404906988144,
      "num_tokens": 57106423.0,
      "step": 30630
    },
    {
      "entropy": 5.943233394622803,
      "epoch": 2.7053161427057577,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00042692535414644804,
      "loss": 5.6187,
      "mean_token_accuracy": 0.18764855116605758,
      "num_tokens": 57115308.0,
      "step": 30635
    },
    {
      "entropy": 5.908019590377807,
      "epoch": 2.705757682797598,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042690212411653844,
      "loss": 5.6371,
      "mean_token_accuracy": 0.189467391371727,
      "num_tokens": 57125108.0,
      "step": 30640
    },
    {
      "entropy": 5.997734403610229,
      "epoch": 2.7061992228894383,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004268788911109526,
      "loss": 5.5717,
      "mean_token_accuracy": 0.19357537031173705,
      "num_tokens": 57134392.0,
      "step": 30645
    },
    {
      "entropy": 5.968635702133179,
      "epoch": 2.7066407629812788,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004268556551301457,
      "loss": 5.6207,
      "mean_token_accuracy": 0.18628529608249664,
      "num_tokens": 57143245.0,
      "step": 30650
    },
    {
      "entropy": 5.927993583679199,
      "epoch": 2.7070823030731193,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004268324161745728,
      "loss": 5.604,
      "mean_token_accuracy": 0.1888149857521057,
      "num_tokens": 57151691.0,
      "step": 30655
    },
    {
      "entropy": 5.927881145477295,
      "epoch": 2.7075238431649593,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042680917424468926,
      "loss": 5.6133,
      "mean_token_accuracy": 0.19303995072841645,
      "num_tokens": 57161157.0,
      "step": 30660
    },
    {
      "entropy": 6.016458415985108,
      "epoch": 2.7079653832568,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004267859293409504,
      "loss": 5.5995,
      "mean_token_accuracy": 0.18782916963100432,
      "num_tokens": 57169599.0,
      "step": 30665
    },
    {
      "entropy": 5.977391862869263,
      "epoch": 2.7084069233486403,
      "grad_norm": 1.546875,
      "learning_rate": 0.00042676268146381134,
      "loss": 5.5983,
      "mean_token_accuracy": 0.187615829706192,
      "num_tokens": 57178788.0,
      "step": 30670
    },
    {
      "entropy": 5.960860538482666,
      "epoch": 2.7088484634404804,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042673943061372776,
      "loss": 5.6367,
      "mean_token_accuracy": 0.18715501427650452,
      "num_tokens": 57187634.0,
      "step": 30675
    },
    {
      "entropy": 5.915687990188599,
      "epoch": 2.709290003532321,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004267161767911551,
      "loss": 5.6291,
      "mean_token_accuracy": 0.18380187302827836,
      "num_tokens": 57196777.0,
      "step": 30680
    },
    {
      "entropy": 6.049829387664795,
      "epoch": 2.709731543624161,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042669291999654867,
      "loss": 5.7318,
      "mean_token_accuracy": 0.1729646772146225,
      "num_tokens": 57206092.0,
      "step": 30685
    },
    {
      "entropy": 6.016754817962647,
      "epoch": 2.7101730837160014,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00042666966023036443,
      "loss": 5.601,
      "mean_token_accuracy": 0.1921016752719879,
      "num_tokens": 57215717.0,
      "step": 30690
    },
    {
      "entropy": 5.948001050949097,
      "epoch": 2.710614623807842,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004266463974930576,
      "loss": 5.6364,
      "mean_token_accuracy": 0.1834026724100113,
      "num_tokens": 57224172.0,
      "step": 30695
    },
    {
      "entropy": 5.944686555862427,
      "epoch": 2.711056163899682,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004266231317850842,
      "loss": 5.5768,
      "mean_token_accuracy": 0.18903705328702927,
      "num_tokens": 57233532.0,
      "step": 30700
    },
    {
      "entropy": 6.006374979019165,
      "epoch": 2.7114977039915225,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004265998631068998,
      "loss": 5.5734,
      "mean_token_accuracy": 0.19131691455841066,
      "num_tokens": 57243348.0,
      "step": 30705
    },
    {
      "entropy": 5.999321603775025,
      "epoch": 2.7119392440833625,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042657659145896044,
      "loss": 5.7381,
      "mean_token_accuracy": 0.17957108169794084,
      "num_tokens": 57253028.0,
      "step": 30710
    },
    {
      "entropy": 5.896546220779419,
      "epoch": 2.712380784175203,
      "grad_norm": 2.3125,
      "learning_rate": 0.00042655331684172173,
      "loss": 5.5567,
      "mean_token_accuracy": 0.1905011534690857,
      "num_tokens": 57261843.0,
      "step": 30715
    },
    {
      "entropy": 6.034680032730103,
      "epoch": 2.7128223242670435,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042653003925563975,
      "loss": 5.6714,
      "mean_token_accuracy": 0.18043527156114578,
      "num_tokens": 57271370.0,
      "step": 30720
    },
    {
      "entropy": 5.944211053848266,
      "epoch": 2.7132638643588836,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00042650675870117057,
      "loss": 5.5404,
      "mean_token_accuracy": 0.19144391119480134,
      "num_tokens": 57280107.0,
      "step": 30725
    },
    {
      "entropy": 5.870504903793335,
      "epoch": 2.713705404450724,
      "grad_norm": 1.546875,
      "learning_rate": 0.00042648347517877,
      "loss": 5.5438,
      "mean_token_accuracy": 0.19743940830230713,
      "num_tokens": 57289878.0,
      "step": 30730
    },
    {
      "entropy": 5.9481956481933596,
      "epoch": 2.7141469445425646,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042646018868889446,
      "loss": 5.6286,
      "mean_token_accuracy": 0.18180225044488907,
      "num_tokens": 57298591.0,
      "step": 30735
    },
    {
      "entropy": 5.9482780456542965,
      "epoch": 2.7145884846344046,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004264368992319999,
      "loss": 5.5269,
      "mean_token_accuracy": 0.19202844500541688,
      "num_tokens": 57307476.0,
      "step": 30740
    },
    {
      "entropy": 5.942166376113891,
      "epoch": 2.715030024726245,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004264136068085426,
      "loss": 5.654,
      "mean_token_accuracy": 0.18510958999395372,
      "num_tokens": 57316953.0,
      "step": 30745
    },
    {
      "entropy": 5.951453590393067,
      "epoch": 2.7154715648180856,
      "grad_norm": 1.515625,
      "learning_rate": 0.00042639031141897884,
      "loss": 5.5778,
      "mean_token_accuracy": 0.1934559762477875,
      "num_tokens": 57326317.0,
      "step": 30750
    },
    {
      "entropy": 5.912520980834961,
      "epoch": 2.7159131049099257,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000426367013063765,
      "loss": 5.5341,
      "mean_token_accuracy": 0.18569480776786804,
      "num_tokens": 57335706.0,
      "step": 30755
    },
    {
      "entropy": 6.0305835723876955,
      "epoch": 2.716354645001766,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00042634371174335743,
      "loss": 5.6813,
      "mean_token_accuracy": 0.18481729477643966,
      "num_tokens": 57345491.0,
      "step": 30760
    },
    {
      "entropy": 5.917489624023437,
      "epoch": 2.7167961850936067,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00042632040745821273,
      "loss": 5.6602,
      "mean_token_accuracy": 0.17707966417074203,
      "num_tokens": 57355158.0,
      "step": 30765
    },
    {
      "entropy": 5.991963052749634,
      "epoch": 2.7172377251854467,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004262971002087872,
      "loss": 5.6815,
      "mean_token_accuracy": 0.1882403701543808,
      "num_tokens": 57364327.0,
      "step": 30770
    },
    {
      "entropy": 6.051650094985962,
      "epoch": 2.717679265277287,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004262737899955376,
      "loss": 5.6461,
      "mean_token_accuracy": 0.17792246192693711,
      "num_tokens": 57373366.0,
      "step": 30775
    },
    {
      "entropy": 5.956535053253174,
      "epoch": 2.7181208053691277,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00042625047681892043,
      "loss": 5.6329,
      "mean_token_accuracy": 0.18962212949991225,
      "num_tokens": 57382221.0,
      "step": 30780
    },
    {
      "entropy": 5.972223901748658,
      "epoch": 2.7185623454609678,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004262271606793924,
      "loss": 5.7339,
      "mean_token_accuracy": 0.1778600662946701,
      "num_tokens": 57391937.0,
      "step": 30785
    },
    {
      "entropy": 6.044479274749756,
      "epoch": 2.7190038855528083,
      "grad_norm": 1.375,
      "learning_rate": 0.00042620384157741035,
      "loss": 5.634,
      "mean_token_accuracy": 0.18854504972696304,
      "num_tokens": 57401189.0,
      "step": 30790
    },
    {
      "entropy": 6.0384971618652346,
      "epoch": 2.7194454256446488,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042618051951343104,
      "loss": 5.6078,
      "mean_token_accuracy": 0.1914215862751007,
      "num_tokens": 57410248.0,
      "step": 30795
    },
    {
      "entropy": 5.972090482711792,
      "epoch": 2.719886965736489,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004261571944879114,
      "loss": 5.7039,
      "mean_token_accuracy": 0.1793148010969162,
      "num_tokens": 57420450.0,
      "step": 30800
    },
    {
      "entropy": 5.962462854385376,
      "epoch": 2.7203285058283293,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004261338665013082,
      "loss": 5.6863,
      "mean_token_accuracy": 0.17641791701316833,
      "num_tokens": 57430381.0,
      "step": 30805
    },
    {
      "entropy": 6.092658233642578,
      "epoch": 2.72077004592017,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004261105355540785,
      "loss": 5.7296,
      "mean_token_accuracy": 0.18172670602798463,
      "num_tokens": 57439252.0,
      "step": 30810
    },
    {
      "entropy": 6.090515565872193,
      "epoch": 2.72121158601201,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004260872016466794,
      "loss": 5.8351,
      "mean_token_accuracy": 0.169037763774395,
      "num_tokens": 57449143.0,
      "step": 30815
    },
    {
      "entropy": 5.9724047660827635,
      "epoch": 2.7216531261038504,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00042606386477956796,
      "loss": 5.6605,
      "mean_token_accuracy": 0.1818341240286827,
      "num_tokens": 57458608.0,
      "step": 30820
    },
    {
      "entropy": 6.0326948165893555,
      "epoch": 2.7220946661956904,
      "grad_norm": 1.375,
      "learning_rate": 0.0004260405249532013,
      "loss": 5.7022,
      "mean_token_accuracy": 0.17688195854425431,
      "num_tokens": 57467616.0,
      "step": 30825
    },
    {
      "entropy": 6.071791410446167,
      "epoch": 2.722536206287531,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042601718216803657,
      "loss": 5.6362,
      "mean_token_accuracy": 0.18519724905490875,
      "num_tokens": 57477460.0,
      "step": 30830
    },
    {
      "entropy": 6.0072253227233885,
      "epoch": 2.7229777463793714,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042599383642453125,
      "loss": 5.7042,
      "mean_token_accuracy": 0.17946138828992844,
      "num_tokens": 57486287.0,
      "step": 30835
    },
    {
      "entropy": 5.977075719833374,
      "epoch": 2.7234192864712115,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004259704877231425,
      "loss": 5.6439,
      "mean_token_accuracy": 0.18561745285987855,
      "num_tokens": 57495891.0,
      "step": 30840
    },
    {
      "entropy": 5.97886848449707,
      "epoch": 2.723860826563052,
      "grad_norm": 1.421875,
      "learning_rate": 0.00042594713606432775,
      "loss": 5.6532,
      "mean_token_accuracy": 0.1853910818696022,
      "num_tokens": 57505384.0,
      "step": 30845
    },
    {
      "entropy": 6.036773443222046,
      "epoch": 2.724302366654892,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042592378144854447,
      "loss": 5.6699,
      "mean_token_accuracy": 0.17888214141130448,
      "num_tokens": 57514106.0,
      "step": 30850
    },
    {
      "entropy": 5.937573480606079,
      "epoch": 2.7247439067467325,
      "grad_norm": 1.453125,
      "learning_rate": 0.00042590042387625023,
      "loss": 5.5821,
      "mean_token_accuracy": 0.1839797779917717,
      "num_tokens": 57524018.0,
      "step": 30855
    },
    {
      "entropy": 5.961933755874634,
      "epoch": 2.725185446838573,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004258770633479024,
      "loss": 5.57,
      "mean_token_accuracy": 0.19216573983430862,
      "num_tokens": 57533393.0,
      "step": 30860
    },
    {
      "entropy": 5.852818727493286,
      "epoch": 2.725626986930413,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004258536998639588,
      "loss": 5.4949,
      "mean_token_accuracy": 0.2011329188942909,
      "num_tokens": 57542442.0,
      "step": 30865
    },
    {
      "entropy": 6.017074298858643,
      "epoch": 2.7260685270222536,
      "grad_norm": 1.375,
      "learning_rate": 0.0004258303334248769,
      "loss": 5.6326,
      "mean_token_accuracy": 0.1916551873087883,
      "num_tokens": 57552367.0,
      "step": 30870
    },
    {
      "entropy": 5.948402261734008,
      "epoch": 2.726510067114094,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004258069640311146,
      "loss": 5.6824,
      "mean_token_accuracy": 0.17628440111875535,
      "num_tokens": 57562291.0,
      "step": 30875
    },
    {
      "entropy": 6.04799108505249,
      "epoch": 2.726951607205934,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004257835916831296,
      "loss": 5.7083,
      "mean_token_accuracy": 0.18058827221393586,
      "num_tokens": 57571490.0,
      "step": 30880
    },
    {
      "entropy": 5.993978595733642,
      "epoch": 2.7273931472977746,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00042576021638137984,
      "loss": 5.6773,
      "mean_token_accuracy": 0.18339404314756394,
      "num_tokens": 57580010.0,
      "step": 30885
    },
    {
      "entropy": 6.025324773788452,
      "epoch": 2.727834687389615,
      "grad_norm": 1.25,
      "learning_rate": 0.0004257368381263232,
      "loss": 5.6821,
      "mean_token_accuracy": 0.1871342658996582,
      "num_tokens": 57589773.0,
      "step": 30890
    },
    {
      "entropy": 6.00528621673584,
      "epoch": 2.728276227481455,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042571345691841763,
      "loss": 5.6297,
      "mean_token_accuracy": 0.19033428728580476,
      "num_tokens": 57598358.0,
      "step": 30895
    },
    {
      "entropy": 5.966138458251953,
      "epoch": 2.7287177675732956,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042569007275812103,
      "loss": 5.6019,
      "mean_token_accuracy": 0.1858905777335167,
      "num_tokens": 57607447.0,
      "step": 30900
    },
    {
      "entropy": 5.900591039657593,
      "epoch": 2.729159307665136,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004256666856458918,
      "loss": 5.598,
      "mean_token_accuracy": 0.1885889396071434,
      "num_tokens": 57617253.0,
      "step": 30905
    },
    {
      "entropy": 5.976289558410644,
      "epoch": 2.729600847756976,
      "grad_norm": 1.8125,
      "learning_rate": 0.00042564329558218776,
      "loss": 5.6934,
      "mean_token_accuracy": 0.177902452647686,
      "num_tokens": 57627304.0,
      "step": 30910
    },
    {
      "entropy": 6.016786384582519,
      "epoch": 2.7300423878488167,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004256199025674673,
      "loss": 5.6331,
      "mean_token_accuracy": 0.1846248373389244,
      "num_tokens": 57635639.0,
      "step": 30915
    },
    {
      "entropy": 6.005493640899658,
      "epoch": 2.730483927940657,
      "grad_norm": 1.453125,
      "learning_rate": 0.00042559650660218857,
      "loss": 5.6623,
      "mean_token_accuracy": 0.18239661902189255,
      "num_tokens": 57644986.0,
      "step": 30920
    },
    {
      "entropy": 5.920060873031616,
      "epoch": 2.7309254680324972,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042557310768680994,
      "loss": 5.5893,
      "mean_token_accuracy": 0.18288650810718537,
      "num_tokens": 57654506.0,
      "step": 30925
    },
    {
      "entropy": 6.035576725006104,
      "epoch": 2.7313670081243377,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004255497058217898,
      "loss": 5.6171,
      "mean_token_accuracy": 0.18566060513257981,
      "num_tokens": 57663075.0,
      "step": 30930
    },
    {
      "entropy": 5.989269113540649,
      "epoch": 2.7318085482161782,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004255263010075865,
      "loss": 5.6677,
      "mean_token_accuracy": 0.19247149378061296,
      "num_tokens": 57672483.0,
      "step": 30935
    },
    {
      "entropy": 5.998298168182373,
      "epoch": 2.7322500883080183,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004255028932446586,
      "loss": 5.6879,
      "mean_token_accuracy": 0.18263897001743318,
      "num_tokens": 57681624.0,
      "step": 30940
    },
    {
      "entropy": 6.014601612091065,
      "epoch": 2.732691628399859,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004254794825334645,
      "loss": 5.6527,
      "mean_token_accuracy": 0.17964722365140914,
      "num_tokens": 57692166.0,
      "step": 30945
    },
    {
      "entropy": 6.0724564552307125,
      "epoch": 2.7331331684916993,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004254560688744631,
      "loss": 5.7268,
      "mean_token_accuracy": 0.1739753767848015,
      "num_tokens": 57701984.0,
      "step": 30950
    },
    {
      "entropy": 5.8979163646698,
      "epoch": 2.7335747085835393,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042543265226811273,
      "loss": 5.4866,
      "mean_token_accuracy": 0.2021907940506935,
      "num_tokens": 57710870.0,
      "step": 30955
    },
    {
      "entropy": 5.972800588607788,
      "epoch": 2.73401624867538,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004254092327148723,
      "loss": 5.6709,
      "mean_token_accuracy": 0.18408658653497695,
      "num_tokens": 57719381.0,
      "step": 30960
    },
    {
      "entropy": 6.027007246017456,
      "epoch": 2.73445778876722,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042538581021520056,
      "loss": 5.6986,
      "mean_token_accuracy": 0.17925811409950257,
      "num_tokens": 57728551.0,
      "step": 30965
    },
    {
      "entropy": 5.981102371215821,
      "epoch": 2.7348993288590604,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042536238476955637,
      "loss": 5.6057,
      "mean_token_accuracy": 0.1899806097149849,
      "num_tokens": 57737532.0,
      "step": 30970
    },
    {
      "entropy": 5.9212370872497555,
      "epoch": 2.735340868950901,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004253389563783986,
      "loss": 5.5724,
      "mean_token_accuracy": 0.18416908830404283,
      "num_tokens": 57746207.0,
      "step": 30975
    },
    {
      "entropy": 5.989662599563599,
      "epoch": 2.735782409042741,
      "grad_norm": 1.71875,
      "learning_rate": 0.00042531552504218605,
      "loss": 5.7215,
      "mean_token_accuracy": 0.17589859664440155,
      "num_tokens": 57755876.0,
      "step": 30980
    },
    {
      "entropy": 5.973957538604736,
      "epoch": 2.7362239491345814,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004252920907613779,
      "loss": 5.6107,
      "mean_token_accuracy": 0.18410394787788392,
      "num_tokens": 57764862.0,
      "step": 30985
    },
    {
      "entropy": 5.937624073028564,
      "epoch": 2.7366654892264215,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042526865353643316,
      "loss": 5.6018,
      "mean_token_accuracy": 0.18664489388465882,
      "num_tokens": 57773910.0,
      "step": 30990
    },
    {
      "entropy": 6.0411612033844,
      "epoch": 2.737107029318262,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004252452133678111,
      "loss": 5.7075,
      "mean_token_accuracy": 0.1742665871977806,
      "num_tokens": 57783373.0,
      "step": 30995
    },
    {
      "entropy": 5.982658767700196,
      "epoch": 2.7375485694101025,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042522177025597065,
      "loss": 5.5949,
      "mean_token_accuracy": 0.19074948728084565,
      "num_tokens": 57792214.0,
      "step": 31000
    },
    {
      "entropy": 5.882665729522705,
      "epoch": 2.7379901095019425,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004251983242013711,
      "loss": 5.5825,
      "mean_token_accuracy": 0.18636007905006408,
      "num_tokens": 57801250.0,
      "step": 31005
    },
    {
      "entropy": 5.917698240280151,
      "epoch": 2.738431649593783,
      "grad_norm": 1.375,
      "learning_rate": 0.00042517487520447183,
      "loss": 5.572,
      "mean_token_accuracy": 0.1875057563185692,
      "num_tokens": 57810814.0,
      "step": 31010
    },
    {
      "entropy": 5.943295860290528,
      "epoch": 2.7388731896856235,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004251514232657323,
      "loss": 5.5809,
      "mean_token_accuracy": 0.18885659277439118,
      "num_tokens": 57819804.0,
      "step": 31015
    },
    {
      "entropy": 6.006222915649414,
      "epoch": 2.7393147297774636,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042512796838561163,
      "loss": 5.6434,
      "mean_token_accuracy": 0.1824524834752083,
      "num_tokens": 57829321.0,
      "step": 31020
    },
    {
      "entropy": 5.94835352897644,
      "epoch": 2.739756269869304,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004251045105645695,
      "loss": 5.58,
      "mean_token_accuracy": 0.19206538647413254,
      "num_tokens": 57838885.0,
      "step": 31025
    },
    {
      "entropy": 5.955958366394043,
      "epoch": 2.7401978099611446,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00042508104980306527,
      "loss": 5.628,
      "mean_token_accuracy": 0.1816286951303482,
      "num_tokens": 57848856.0,
      "step": 31030
    },
    {
      "entropy": 5.9027605056762695,
      "epoch": 2.7406393500529846,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042505758610155873,
      "loss": 5.6052,
      "mean_token_accuracy": 0.1851982906460762,
      "num_tokens": 57857527.0,
      "step": 31035
    },
    {
      "entropy": 5.931025981903076,
      "epoch": 2.741080890144825,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042503411946050943,
      "loss": 5.6491,
      "mean_token_accuracy": 0.18131913095712662,
      "num_tokens": 57866946.0,
      "step": 31040
    },
    {
      "entropy": 6.080127716064453,
      "epoch": 2.7415224302366656,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004250106498803771,
      "loss": 5.6746,
      "mean_token_accuracy": 0.1828841134905815,
      "num_tokens": 57876400.0,
      "step": 31045
    },
    {
      "entropy": 5.8837004661560055,
      "epoch": 2.7419639703285057,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004249871773616214,
      "loss": 5.5142,
      "mean_token_accuracy": 0.19583750516176224,
      "num_tokens": 57885224.0,
      "step": 31050
    },
    {
      "entropy": 5.923188352584839,
      "epoch": 2.742405510420346,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042496370190470216,
      "loss": 5.6662,
      "mean_token_accuracy": 0.18348006159067154,
      "num_tokens": 57894379.0,
      "step": 31055
    },
    {
      "entropy": 5.941050815582275,
      "epoch": 2.7428470505121867,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004249402235100793,
      "loss": 5.6663,
      "mean_token_accuracy": 0.18205907940864563,
      "num_tokens": 57903851.0,
      "step": 31060
    },
    {
      "entropy": 6.034585285186767,
      "epoch": 2.7432885906040267,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004249167421782127,
      "loss": 5.745,
      "mean_token_accuracy": 0.1750384822487831,
      "num_tokens": 57912705.0,
      "step": 31065
    },
    {
      "entropy": 5.961391735076904,
      "epoch": 2.7437301306958672,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004248932579095625,
      "loss": 5.5572,
      "mean_token_accuracy": 0.19971191734075547,
      "num_tokens": 57921623.0,
      "step": 31070
    },
    {
      "entropy": 5.88310227394104,
      "epoch": 2.7441716707877077,
      "grad_norm": 2.015625,
      "learning_rate": 0.00042486977070458856,
      "loss": 5.58,
      "mean_token_accuracy": 0.18503732830286027,
      "num_tokens": 57930673.0,
      "step": 31075
    },
    {
      "entropy": 5.881405973434449,
      "epoch": 2.744613210879548,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004248462805637511,
      "loss": 5.5973,
      "mean_token_accuracy": 0.1920399472117424,
      "num_tokens": 57939727.0,
      "step": 31080
    },
    {
      "entropy": 6.038373708724976,
      "epoch": 2.7450547509713883,
      "grad_norm": 1.125,
      "learning_rate": 0.00042482278748751014,
      "loss": 5.6137,
      "mean_token_accuracy": 0.1811959370970726,
      "num_tokens": 57949830.0,
      "step": 31085
    },
    {
      "entropy": 5.989900302886963,
      "epoch": 2.745496291063229,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000424799291476326,
      "loss": 5.6626,
      "mean_token_accuracy": 0.18816177248954774,
      "num_tokens": 57959577.0,
      "step": 31090
    },
    {
      "entropy": 5.830651044845581,
      "epoch": 2.745937831155069,
      "grad_norm": 1.140625,
      "learning_rate": 0.000424775792530659,
      "loss": 5.6104,
      "mean_token_accuracy": 0.1913789004087448,
      "num_tokens": 57968569.0,
      "step": 31095
    },
    {
      "entropy": 5.9668717861175535,
      "epoch": 2.7463793712469093,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004247522906509693,
      "loss": 5.6656,
      "mean_token_accuracy": 0.18386841267347337,
      "num_tokens": 57978285.0,
      "step": 31100
    },
    {
      "entropy": 5.963809585571289,
      "epoch": 2.7468209113387494,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004247287858377176,
      "loss": 5.6718,
      "mean_token_accuracy": 0.1916997417807579,
      "num_tokens": 57987561.0,
      "step": 31105
    },
    {
      "entropy": 5.954493713378906,
      "epoch": 2.74726245143059,
      "grad_norm": 1.34375,
      "learning_rate": 0.000424705278091364,
      "loss": 5.6078,
      "mean_token_accuracy": 0.1886981964111328,
      "num_tokens": 57997999.0,
      "step": 31110
    },
    {
      "entropy": 5.985544872283936,
      "epoch": 2.7477039915224304,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004246817674123691,
      "loss": 5.6499,
      "mean_token_accuracy": 0.18516538441181182,
      "num_tokens": 58007655.0,
      "step": 31115
    },
    {
      "entropy": 6.011958265304566,
      "epoch": 2.7481455316142704,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004246582538011935,
      "loss": 5.6205,
      "mean_token_accuracy": 0.18839472830295562,
      "num_tokens": 58016542.0,
      "step": 31120
    },
    {
      "entropy": 5.9205464839935305,
      "epoch": 2.748587071706111,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004246347372582979,
      "loss": 5.6126,
      "mean_token_accuracy": 0.18844369053840637,
      "num_tokens": 58026169.0,
      "step": 31125
    },
    {
      "entropy": 5.942705965042114,
      "epoch": 2.749028611797951,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042461121778414286,
      "loss": 5.6174,
      "mean_token_accuracy": 0.18884463608264923,
      "num_tokens": 58036006.0,
      "step": 31130
    },
    {
      "entropy": 5.988434791564941,
      "epoch": 2.7494701518897915,
      "grad_norm": 1.453125,
      "learning_rate": 0.00042458769537918916,
      "loss": 5.5827,
      "mean_token_accuracy": 0.18963656425476075,
      "num_tokens": 58044601.0,
      "step": 31135
    },
    {
      "entropy": 5.980305480957031,
      "epoch": 2.749911691981632,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004245641700438976,
      "loss": 5.6066,
      "mean_token_accuracy": 0.18723456114530562,
      "num_tokens": 58053492.0,
      "step": 31140
    },
    {
      "entropy": 5.9523557186126705,
      "epoch": 2.750353232073472,
      "grad_norm": 1.359375,
      "learning_rate": 0.000424540641778729,
      "loss": 5.5945,
      "mean_token_accuracy": 0.19141431301832199,
      "num_tokens": 58062457.0,
      "step": 31145
    },
    {
      "entropy": 5.890523529052734,
      "epoch": 2.7507947721653125,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004245171105841443,
      "loss": 5.5298,
      "mean_token_accuracy": 0.19438455402851104,
      "num_tokens": 58071432.0,
      "step": 31150
    },
    {
      "entropy": 6.011520433425903,
      "epoch": 2.751236312257153,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042449357646060437,
      "loss": 5.7377,
      "mean_token_accuracy": 0.18238672614097595,
      "num_tokens": 58080588.0,
      "step": 31155
    },
    {
      "entropy": 5.942641353607177,
      "epoch": 2.751677852348993,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004244700394085704,
      "loss": 5.5513,
      "mean_token_accuracy": 0.19018156528472902,
      "num_tokens": 58090467.0,
      "step": 31160
    },
    {
      "entropy": 5.994507789611816,
      "epoch": 2.7521193924408336,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042444649942850325,
      "loss": 5.6822,
      "mean_token_accuracy": 0.18303484469652176,
      "num_tokens": 58099746.0,
      "step": 31165
    },
    {
      "entropy": 5.951570320129394,
      "epoch": 2.752560932532674,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042442295652086434,
      "loss": 5.6288,
      "mean_token_accuracy": 0.1868046760559082,
      "num_tokens": 58109039.0,
      "step": 31170
    },
    {
      "entropy": 5.92836480140686,
      "epoch": 2.753002472624514,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004243994106861145,
      "loss": 5.5407,
      "mean_token_accuracy": 0.19299046844244003,
      "num_tokens": 58117567.0,
      "step": 31175
    },
    {
      "entropy": 5.870970916748047,
      "epoch": 2.7534440127163546,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004243758619247153,
      "loss": 5.5234,
      "mean_token_accuracy": 0.18935272246599197,
      "num_tokens": 58125918.0,
      "step": 31180
    },
    {
      "entropy": 5.965092515945434,
      "epoch": 2.753885552808195,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042435231023712784,
      "loss": 5.6688,
      "mean_token_accuracy": 0.18243522942066193,
      "num_tokens": 58135862.0,
      "step": 31185
    },
    {
      "entropy": 6.037359619140625,
      "epoch": 2.754327092900035,
      "grad_norm": 2.5,
      "learning_rate": 0.00042432875562381354,
      "loss": 5.7446,
      "mean_token_accuracy": 0.17408138364553452,
      "num_tokens": 58145182.0,
      "step": 31190
    },
    {
      "entropy": 6.021795701980591,
      "epoch": 2.7547686329918757,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004243051980852339,
      "loss": 5.6295,
      "mean_token_accuracy": 0.18413154780864716,
      "num_tokens": 58153764.0,
      "step": 31195
    },
    {
      "entropy": 5.967591190338135,
      "epoch": 2.755210173083716,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004242816376218504,
      "loss": 5.585,
      "mean_token_accuracy": 0.19205904603004456,
      "num_tokens": 58162982.0,
      "step": 31200
    },
    {
      "entropy": 5.924546623229981,
      "epoch": 2.7556517131755562,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042425807423412444,
      "loss": 5.6058,
      "mean_token_accuracy": 0.1847239315509796,
      "num_tokens": 58172300.0,
      "step": 31205
    },
    {
      "entropy": 5.992878675460815,
      "epoch": 2.7560932532673967,
      "grad_norm": 1.25,
      "learning_rate": 0.00042423450792251774,
      "loss": 5.6373,
      "mean_token_accuracy": 0.18666552603244782,
      "num_tokens": 58181494.0,
      "step": 31210
    },
    {
      "entropy": 5.977120590209961,
      "epoch": 2.7565347933592372,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042421093868749195,
      "loss": 5.6464,
      "mean_token_accuracy": 0.19044464230537414,
      "num_tokens": 58191500.0,
      "step": 31215
    },
    {
      "entropy": 5.955701923370361,
      "epoch": 2.7569763334510773,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004241873665295086,
      "loss": 5.6382,
      "mean_token_accuracy": 0.19295424073934556,
      "num_tokens": 58201012.0,
      "step": 31220
    },
    {
      "entropy": 5.903749132156372,
      "epoch": 2.757417873542918,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004241637914490297,
      "loss": 5.5848,
      "mean_token_accuracy": 0.190072900056839,
      "num_tokens": 58209901.0,
      "step": 31225
    },
    {
      "entropy": 5.997347021102906,
      "epoch": 2.7578594136347583,
      "grad_norm": 1.375,
      "learning_rate": 0.00042414021344651685,
      "loss": 5.6606,
      "mean_token_accuracy": 0.18099277466535568,
      "num_tokens": 58219809.0,
      "step": 31230
    },
    {
      "entropy": 5.926541805267334,
      "epoch": 2.7583009537265983,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004241166325224321,
      "loss": 5.6028,
      "mean_token_accuracy": 0.18304303586483,
      "num_tokens": 58229626.0,
      "step": 31235
    },
    {
      "entropy": 5.978384304046631,
      "epoch": 2.758742493818439,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004240930486772374,
      "loss": 5.6122,
      "mean_token_accuracy": 0.1865341693162918,
      "num_tokens": 58239533.0,
      "step": 31240
    },
    {
      "entropy": 5.957190752029419,
      "epoch": 2.759184033910279,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00042406946191139456,
      "loss": 5.5672,
      "mean_token_accuracy": 0.1876811772584915,
      "num_tokens": 58248998.0,
      "step": 31245
    },
    {
      "entropy": 5.95540189743042,
      "epoch": 2.7596255740021194,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00042404587222536573,
      "loss": 5.581,
      "mean_token_accuracy": 0.184913732111454,
      "num_tokens": 58258024.0,
      "step": 31250
    },
    {
      "entropy": 5.937592697143555,
      "epoch": 2.76006711409396,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042402227961961314,
      "loss": 5.6562,
      "mean_token_accuracy": 0.18264322429895402,
      "num_tokens": 58267937.0,
      "step": 31255
    },
    {
      "entropy": 5.9668927669525145,
      "epoch": 2.7605086541858,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004239986840945988,
      "loss": 5.6615,
      "mean_token_accuracy": 0.18322503715753555,
      "num_tokens": 58276834.0,
      "step": 31260
    },
    {
      "entropy": 5.997774839401245,
      "epoch": 2.7609501942776404,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004239750856507849,
      "loss": 5.6881,
      "mean_token_accuracy": 0.17923223078250886,
      "num_tokens": 58286816.0,
      "step": 31265
    },
    {
      "entropy": 5.948994159698486,
      "epoch": 2.7613917343694805,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004239514842886339,
      "loss": 5.5828,
      "mean_token_accuracy": 0.19694940894842147,
      "num_tokens": 58296086.0,
      "step": 31270
    },
    {
      "entropy": 5.955049324035644,
      "epoch": 2.761833274461321,
      "grad_norm": 1.25,
      "learning_rate": 0.00042392788000860797,
      "loss": 5.6462,
      "mean_token_accuracy": 0.18194997012615205,
      "num_tokens": 58305851.0,
      "step": 31275
    },
    {
      "entropy": 5.840792894363403,
      "epoch": 2.7622748145531615,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042390427281116966,
      "loss": 5.525,
      "mean_token_accuracy": 0.18868571370840073,
      "num_tokens": 58314693.0,
      "step": 31280
    },
    {
      "entropy": 5.963656949996948,
      "epoch": 2.7627163546450015,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042388066269678115,
      "loss": 5.6002,
      "mean_token_accuracy": 0.18901156336069108,
      "num_tokens": 58323425.0,
      "step": 31285
    },
    {
      "entropy": 5.968524551391601,
      "epoch": 2.763157894736842,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00042385704966590537,
      "loss": 5.5806,
      "mean_token_accuracy": 0.19151457399129868,
      "num_tokens": 58332651.0,
      "step": 31290
    },
    {
      "entropy": 5.977307081222534,
      "epoch": 2.7635994348286825,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004238334337190045,
      "loss": 5.6524,
      "mean_token_accuracy": 0.18239837288856506,
      "num_tokens": 58342366.0,
      "step": 31295
    },
    {
      "entropy": 5.948267698287964,
      "epoch": 2.7640409749205226,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00042380981485654127,
      "loss": 5.6307,
      "mean_token_accuracy": 0.18985143154859543,
      "num_tokens": 58352370.0,
      "step": 31300
    },
    {
      "entropy": 5.993791627883911,
      "epoch": 2.764482515012363,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042378619307897853,
      "loss": 5.6835,
      "mean_token_accuracy": 0.18657415509223937,
      "num_tokens": 58360819.0,
      "step": 31305
    },
    {
      "entropy": 6.024039268493652,
      "epoch": 2.7649240551042036,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004237625683867788,
      "loss": 5.7146,
      "mean_token_accuracy": 0.17843483984470368,
      "num_tokens": 58370744.0,
      "step": 31310
    },
    {
      "entropy": 6.076037073135376,
      "epoch": 2.7653655951960436,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042373894078040497,
      "loss": 5.6429,
      "mean_token_accuracy": 0.1845259577035904,
      "num_tokens": 58380965.0,
      "step": 31315
    },
    {
      "entropy": 5.967077684402466,
      "epoch": 2.765807135287884,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042371531026031995,
      "loss": 5.6083,
      "mean_token_accuracy": 0.18792535364627838,
      "num_tokens": 58390386.0,
      "step": 31320
    },
    {
      "entropy": 5.823403310775757,
      "epoch": 2.7662486753797246,
      "grad_norm": 1.125,
      "learning_rate": 0.0004236916768269865,
      "loss": 5.5237,
      "mean_token_accuracy": 0.1931168630719185,
      "num_tokens": 58399603.0,
      "step": 31325
    },
    {
      "entropy": 5.9449831485748295,
      "epoch": 2.7666902154715647,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004236680404808677,
      "loss": 5.5724,
      "mean_token_accuracy": 0.19446683079004287,
      "num_tokens": 58408627.0,
      "step": 31330
    },
    {
      "entropy": 5.994288921356201,
      "epoch": 2.767131755563405,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004236444012224266,
      "loss": 5.6455,
      "mean_token_accuracy": 0.18777942061424255,
      "num_tokens": 58418132.0,
      "step": 31335
    },
    {
      "entropy": 5.8854516506195065,
      "epoch": 2.7675732956552457,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004236207590521261,
      "loss": 5.6504,
      "mean_token_accuracy": 0.18475256264209747,
      "num_tokens": 58427298.0,
      "step": 31340
    },
    {
      "entropy": 5.912888097763061,
      "epoch": 2.7680148357470857,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004235971139704295,
      "loss": 5.6314,
      "mean_token_accuracy": 0.19295356720685958,
      "num_tokens": 58436977.0,
      "step": 31345
    },
    {
      "entropy": 6.0328062057495115,
      "epoch": 2.7684563758389262,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004235734659777999,
      "loss": 5.7141,
      "mean_token_accuracy": 0.18218026757240297,
      "num_tokens": 58446801.0,
      "step": 31350
    },
    {
      "entropy": 5.988728046417236,
      "epoch": 2.7688979159307667,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004235498150747007,
      "loss": 5.6148,
      "mean_token_accuracy": 0.18763121366500854,
      "num_tokens": 58456625.0,
      "step": 31355
    },
    {
      "entropy": 5.99283652305603,
      "epoch": 2.769339456022607,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004235261612615951,
      "loss": 5.5908,
      "mean_token_accuracy": 0.18264820277690888,
      "num_tokens": 58465989.0,
      "step": 31360
    },
    {
      "entropy": 5.854195165634155,
      "epoch": 2.7697809961144473,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004235025045389464,
      "loss": 5.6111,
      "mean_token_accuracy": 0.19598504453897475,
      "num_tokens": 58475703.0,
      "step": 31365
    },
    {
      "entropy": 5.925386714935303,
      "epoch": 2.7702225362062878,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004234788449072181,
      "loss": 5.5506,
      "mean_token_accuracy": 0.19017844945192336,
      "num_tokens": 58485192.0,
      "step": 31370
    },
    {
      "entropy": 6.021717357635498,
      "epoch": 2.770664076298128,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042345518236687376,
      "loss": 5.6815,
      "mean_token_accuracy": 0.17737847566604614,
      "num_tokens": 58494275.0,
      "step": 31375
    },
    {
      "entropy": 5.932120227813721,
      "epoch": 2.7711056163899683,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00042343151691837675,
      "loss": 5.5596,
      "mean_token_accuracy": 0.19443225115537643,
      "num_tokens": 58504513.0,
      "step": 31380
    },
    {
      "entropy": 5.938470315933228,
      "epoch": 2.7715471564818084,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004234078485621907,
      "loss": 5.6687,
      "mean_token_accuracy": 0.18048936128616333,
      "num_tokens": 58514190.0,
      "step": 31385
    },
    {
      "entropy": 5.844409418106079,
      "epoch": 2.771988696573649,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004233841772987794,
      "loss": 5.4885,
      "mean_token_accuracy": 0.20470507591962814,
      "num_tokens": 58524189.0,
      "step": 31390
    },
    {
      "entropy": 5.96816463470459,
      "epoch": 2.7724302366654894,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004233605031286065,
      "loss": 5.609,
      "mean_token_accuracy": 0.18174556493759156,
      "num_tokens": 58533899.0,
      "step": 31395
    },
    {
      "entropy": 5.989956378936768,
      "epoch": 2.7728717767573294,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042333682605213556,
      "loss": 5.5879,
      "mean_token_accuracy": 0.1921074703335762,
      "num_tokens": 58543523.0,
      "step": 31400
    },
    {
      "entropy": 5.893774032592773,
      "epoch": 2.77331331684917,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042331314606983063,
      "loss": 5.5568,
      "mean_token_accuracy": 0.18648988753557205,
      "num_tokens": 58552623.0,
      "step": 31405
    },
    {
      "entropy": 6.0094812393188475,
      "epoch": 2.77375485694101,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004232894631821556,
      "loss": 5.6839,
      "mean_token_accuracy": 0.1847570240497589,
      "num_tokens": 58562194.0,
      "step": 31410
    },
    {
      "entropy": 5.997704792022705,
      "epoch": 2.7741963970328505,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042326577738957424,
      "loss": 5.5774,
      "mean_token_accuracy": 0.18725328594446183,
      "num_tokens": 58571715.0,
      "step": 31415
    },
    {
      "entropy": 5.955130338668823,
      "epoch": 2.774637937124691,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042324208869255064,
      "loss": 5.5893,
      "mean_token_accuracy": 0.19009992629289627,
      "num_tokens": 58582305.0,
      "step": 31420
    },
    {
      "entropy": 5.901904582977295,
      "epoch": 2.775079477216531,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004232183970915489,
      "loss": 5.6893,
      "mean_token_accuracy": 0.18112677186727524,
      "num_tokens": 58592825.0,
      "step": 31425
    },
    {
      "entropy": 5.936890077590943,
      "epoch": 2.7755210173083715,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042319470258703293,
      "loss": 5.5653,
      "mean_token_accuracy": 0.19370200335979462,
      "num_tokens": 58601985.0,
      "step": 31430
    },
    {
      "entropy": 5.869934988021851,
      "epoch": 2.775962557400212,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042317100517946706,
      "loss": 5.5594,
      "mean_token_accuracy": 0.19836499094963073,
      "num_tokens": 58611390.0,
      "step": 31435
    },
    {
      "entropy": 6.011073446273803,
      "epoch": 2.776404097492052,
      "grad_norm": 1.25,
      "learning_rate": 0.0004231473048693155,
      "loss": 5.6249,
      "mean_token_accuracy": 0.1881101459264755,
      "num_tokens": 58621154.0,
      "step": 31440
    },
    {
      "entropy": 5.879802083969116,
      "epoch": 2.7768456375838926,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042312360165704246,
      "loss": 5.453,
      "mean_token_accuracy": 0.20201350897550582,
      "num_tokens": 58629346.0,
      "step": 31445
    },
    {
      "entropy": 5.886003923416138,
      "epoch": 2.777287177675733,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004230998955431124,
      "loss": 5.6589,
      "mean_token_accuracy": 0.1850365847349167,
      "num_tokens": 58638472.0,
      "step": 31450
    },
    {
      "entropy": 5.962806606292725,
      "epoch": 2.777728717767573,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004230761865279895,
      "loss": 5.627,
      "mean_token_accuracy": 0.19151421040296554,
      "num_tokens": 58648016.0,
      "step": 31455
    },
    {
      "entropy": 5.982826566696167,
      "epoch": 2.7781702578594136,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042305247461213837,
      "loss": 5.6642,
      "mean_token_accuracy": 0.17935896813869476,
      "num_tokens": 58657667.0,
      "step": 31460
    },
    {
      "entropy": 5.982476902008057,
      "epoch": 2.778611797951254,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042302875979602345,
      "loss": 5.692,
      "mean_token_accuracy": 0.17984830290079118,
      "num_tokens": 58667465.0,
      "step": 31465
    },
    {
      "entropy": 5.933959722518921,
      "epoch": 2.779053338043094,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004230050420801093,
      "loss": 5.5235,
      "mean_token_accuracy": 0.1968450054526329,
      "num_tokens": 58676037.0,
      "step": 31470
    },
    {
      "entropy": 5.845414876937866,
      "epoch": 2.7794948781349347,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042298132146486054,
      "loss": 5.5505,
      "mean_token_accuracy": 0.20441123843193054,
      "num_tokens": 58684807.0,
      "step": 31475
    },
    {
      "entropy": 5.8375749588012695,
      "epoch": 2.779936418226775,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004229575979507419,
      "loss": 5.5899,
      "mean_token_accuracy": 0.18530622124671936,
      "num_tokens": 58694013.0,
      "step": 31480
    },
    {
      "entropy": 5.984395456314087,
      "epoch": 2.7803779583186152,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042293387153821807,
      "loss": 5.6175,
      "mean_token_accuracy": 0.18821204602718353,
      "num_tokens": 58703328.0,
      "step": 31485
    },
    {
      "entropy": 5.981878089904785,
      "epoch": 2.7808194984104557,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004229101422277538,
      "loss": 5.5807,
      "mean_token_accuracy": 0.18789077848196029,
      "num_tokens": 58712130.0,
      "step": 31490
    },
    {
      "entropy": 5.975812816619873,
      "epoch": 2.7812610385022962,
      "grad_norm": 1.515625,
      "learning_rate": 0.000422886410019814,
      "loss": 5.6433,
      "mean_token_accuracy": 0.18366542756557463,
      "num_tokens": 58721386.0,
      "step": 31495
    },
    {
      "entropy": 5.949778318405151,
      "epoch": 2.7817025785941363,
      "grad_norm": 1.25,
      "learning_rate": 0.00042286267491486345,
      "loss": 5.5789,
      "mean_token_accuracy": 0.18930310755968094,
      "num_tokens": 58730572.0,
      "step": 31500
    },
    {
      "entropy": 5.892454481124878,
      "epoch": 2.7821441186859768,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004228389369133673,
      "loss": 5.4979,
      "mean_token_accuracy": 0.19123108088970184,
      "num_tokens": 58738605.0,
      "step": 31505
    },
    {
      "entropy": 5.884854412078857,
      "epoch": 2.7825856587778173,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042281519601579035,
      "loss": 5.6568,
      "mean_token_accuracy": 0.18215292543172837,
      "num_tokens": 58747429.0,
      "step": 31510
    },
    {
      "entropy": 5.930685806274414,
      "epoch": 2.7830271988696573,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042279145222259785,
      "loss": 5.569,
      "mean_token_accuracy": 0.18782996237277985,
      "num_tokens": 58756213.0,
      "step": 31515
    },
    {
      "entropy": 6.046748447418213,
      "epoch": 2.783468738961498,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00042276770553425493,
      "loss": 5.6013,
      "mean_token_accuracy": 0.1925421729683876,
      "num_tokens": 58765025.0,
      "step": 31520
    },
    {
      "entropy": 6.00041766166687,
      "epoch": 2.783910279053338,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042274395595122654,
      "loss": 5.6816,
      "mean_token_accuracy": 0.18154319375753403,
      "num_tokens": 58775237.0,
      "step": 31525
    },
    {
      "entropy": 5.934314203262329,
      "epoch": 2.7843518191451784,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004227202034739782,
      "loss": 5.575,
      "mean_token_accuracy": 0.19348684698343277,
      "num_tokens": 58784555.0,
      "step": 31530
    },
    {
      "entropy": 5.935778331756592,
      "epoch": 2.784793359237019,
      "grad_norm": 1.234375,
      "learning_rate": 0.000422696448102975,
      "loss": 5.6172,
      "mean_token_accuracy": 0.18711372017860411,
      "num_tokens": 58793277.0,
      "step": 31535
    },
    {
      "entropy": 5.93610725402832,
      "epoch": 2.785234899328859,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042267268983868246,
      "loss": 5.5824,
      "mean_token_accuracy": 0.19202755987644196,
      "num_tokens": 58802872.0,
      "step": 31540
    },
    {
      "entropy": 5.889186191558838,
      "epoch": 2.7856764394206994,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004226489286815658,
      "loss": 5.5766,
      "mean_token_accuracy": 0.19579940438270568,
      "num_tokens": 58811728.0,
      "step": 31545
    },
    {
      "entropy": 5.973244094848633,
      "epoch": 2.7861179795125395,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004226251646320907,
      "loss": 5.7059,
      "mean_token_accuracy": 0.18227566182613372,
      "num_tokens": 58820689.0,
      "step": 31550
    },
    {
      "entropy": 6.025663423538208,
      "epoch": 2.78655951960438,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004226013976907225,
      "loss": 5.6187,
      "mean_token_accuracy": 0.19434286057949066,
      "num_tokens": 58829062.0,
      "step": 31555
    },
    {
      "entropy": 5.969925785064698,
      "epoch": 2.7870010596962205,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042257762785792693,
      "loss": 5.6025,
      "mean_token_accuracy": 0.18736793249845504,
      "num_tokens": 58838183.0,
      "step": 31560
    },
    {
      "entropy": 5.865938329696656,
      "epoch": 2.7874425997880605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042255385513416963,
      "loss": 5.5658,
      "mean_token_accuracy": 0.19105178266763687,
      "num_tokens": 58847305.0,
      "step": 31565
    },
    {
      "entropy": 5.923144388198852,
      "epoch": 2.787884139879901,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004225300795199161,
      "loss": 5.5554,
      "mean_token_accuracy": 0.18911392241716385,
      "num_tokens": 58856161.0,
      "step": 31570
    },
    {
      "entropy": 5.975593757629395,
      "epoch": 2.7883256799717415,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00042250630101563224,
      "loss": 5.599,
      "mean_token_accuracy": 0.18796436339616776,
      "num_tokens": 58865620.0,
      "step": 31575
    },
    {
      "entropy": 5.987051773071289,
      "epoch": 2.7887672200635816,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042248251962178387,
      "loss": 5.6313,
      "mean_token_accuracy": 0.19245840162038802,
      "num_tokens": 58873997.0,
      "step": 31580
    },
    {
      "entropy": 5.938768815994263,
      "epoch": 2.789208760155422,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004224587353388368,
      "loss": 5.6385,
      "mean_token_accuracy": 0.18858134299516677,
      "num_tokens": 58882905.0,
      "step": 31585
    },
    {
      "entropy": 5.9310302734375,
      "epoch": 2.7896503002472626,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042243494816725696,
      "loss": 5.5663,
      "mean_token_accuracy": 0.18271249830722808,
      "num_tokens": 58892551.0,
      "step": 31590
    },
    {
      "entropy": 5.857443380355835,
      "epoch": 2.7900918403391026,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004224111581075104,
      "loss": 5.4786,
      "mean_token_accuracy": 0.19551650434732437,
      "num_tokens": 58902195.0,
      "step": 31595
    },
    {
      "entropy": 5.941458606719971,
      "epoch": 2.790533380430943,
      "grad_norm": 1.25,
      "learning_rate": 0.00042238736516006303,
      "loss": 5.5881,
      "mean_token_accuracy": 0.1936073362827301,
      "num_tokens": 58911055.0,
      "step": 31600
    },
    {
      "entropy": 5.934753036499023,
      "epoch": 2.7909749205227836,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000422363569325381,
      "loss": 5.5868,
      "mean_token_accuracy": 0.1941434398293495,
      "num_tokens": 58920931.0,
      "step": 31605
    },
    {
      "entropy": 6.023649120330811,
      "epoch": 2.7914164606146237,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004223397706039304,
      "loss": 5.728,
      "mean_token_accuracy": 0.17771221548318863,
      "num_tokens": 58931571.0,
      "step": 31610
    },
    {
      "entropy": 6.0711362838745115,
      "epoch": 2.791858000706464,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004223159689961777,
      "loss": 5.6881,
      "mean_token_accuracy": 0.17795897275209427,
      "num_tokens": 58940323.0,
      "step": 31615
    },
    {
      "entropy": 5.942058753967285,
      "epoch": 2.7922995407983047,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042229216450258874,
      "loss": 5.6314,
      "mean_token_accuracy": 0.1847422942519188,
      "num_tokens": 58949865.0,
      "step": 31620
    },
    {
      "entropy": 5.869816303253174,
      "epoch": 2.7927410808901447,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042226835712363015,
      "loss": 5.5886,
      "mean_token_accuracy": 0.19377538114786147,
      "num_tokens": 58958075.0,
      "step": 31625
    },
    {
      "entropy": 6.014513969421387,
      "epoch": 2.793182620981985,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004222445468597681,
      "loss": 5.66,
      "mean_token_accuracy": 0.18310250639915465,
      "num_tokens": 58966985.0,
      "step": 31630
    },
    {
      "entropy": 5.969236707687378,
      "epoch": 2.7936241610738257,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042222073371146925,
      "loss": 5.5358,
      "mean_token_accuracy": 0.18960415720939636,
      "num_tokens": 58976668.0,
      "step": 31635
    },
    {
      "entropy": 5.965900945663452,
      "epoch": 2.7940657011656658,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004221969176791999,
      "loss": 5.6671,
      "mean_token_accuracy": 0.183441361784935,
      "num_tokens": 58986867.0,
      "step": 31640
    },
    {
      "entropy": 5.947656106948853,
      "epoch": 2.7945072412575063,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00042217309876342655,
      "loss": 5.6255,
      "mean_token_accuracy": 0.18421817868947982,
      "num_tokens": 58996626.0,
      "step": 31645
    },
    {
      "entropy": 5.935919284820557,
      "epoch": 2.7949487813493468,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042214927696461604,
      "loss": 5.5845,
      "mean_token_accuracy": 0.19364119619131087,
      "num_tokens": 59005291.0,
      "step": 31650
    },
    {
      "entropy": 6.041907835006714,
      "epoch": 2.795390321441187,
      "grad_norm": 1.515625,
      "learning_rate": 0.00042212545228323476,
      "loss": 5.6715,
      "mean_token_accuracy": 0.18472568243741988,
      "num_tokens": 59016308.0,
      "step": 31655
    },
    {
      "entropy": 5.966346073150635,
      "epoch": 2.7958318615330273,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004221016247197497,
      "loss": 5.634,
      "mean_token_accuracy": 0.17886050194501876,
      "num_tokens": 59026069.0,
      "step": 31660
    },
    {
      "entropy": 6.011286020278931,
      "epoch": 2.7962734016248674,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00042207779427462733,
      "loss": 5.667,
      "mean_token_accuracy": 0.18129089027643203,
      "num_tokens": 59036163.0,
      "step": 31665
    },
    {
      "entropy": 5.96740312576294,
      "epoch": 2.796714941716708,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004220539609483347,
      "loss": 5.6107,
      "mean_token_accuracy": 0.19428400248289107,
      "num_tokens": 59045752.0,
      "step": 31670
    },
    {
      "entropy": 5.9350017547607425,
      "epoch": 2.7971564818085484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042203012474133857,
      "loss": 5.5384,
      "mean_token_accuracy": 0.19526515007019044,
      "num_tokens": 59054515.0,
      "step": 31675
    },
    {
      "entropy": 5.931487607955932,
      "epoch": 2.7975980219003884,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042200628565410594,
      "loss": 5.6233,
      "mean_token_accuracy": 0.18954141885042192,
      "num_tokens": 59063166.0,
      "step": 31680
    },
    {
      "entropy": 6.007737684249878,
      "epoch": 2.798039561992229,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042198244368710377,
      "loss": 5.7286,
      "mean_token_accuracy": 0.16826617270708083,
      "num_tokens": 59072133.0,
      "step": 31685
    },
    {
      "entropy": 5.983328008651734,
      "epoch": 2.798481102084069,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042195859884079905,
      "loss": 5.6745,
      "mean_token_accuracy": 0.18700529783964157,
      "num_tokens": 59081968.0,
      "step": 31690
    },
    {
      "entropy": 6.048095464706421,
      "epoch": 2.7989226421759095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042193475111565915,
      "loss": 5.7113,
      "mean_token_accuracy": 0.1725090801715851,
      "num_tokens": 59091606.0,
      "step": 31695
    },
    {
      "entropy": 5.9926722049713135,
      "epoch": 2.79936418226775,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042191090051215085,
      "loss": 5.5406,
      "mean_token_accuracy": 0.19217709302902222,
      "num_tokens": 59099768.0,
      "step": 31700
    },
    {
      "entropy": 5.946174287796021,
      "epoch": 2.79980572235959,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004218870470307417,
      "loss": 5.6713,
      "mean_token_accuracy": 0.1863571360707283,
      "num_tokens": 59109071.0,
      "step": 31705
    },
    {
      "entropy": 5.892638540267944,
      "epoch": 2.8002472624514305,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004218631906718987,
      "loss": 5.6474,
      "mean_token_accuracy": 0.18154931515455247,
      "num_tokens": 59117634.0,
      "step": 31710
    },
    {
      "entropy": 6.016881704330444,
      "epoch": 2.800688802543271,
      "grad_norm": 1.53125,
      "learning_rate": 0.00042183933143608943,
      "loss": 5.6512,
      "mean_token_accuracy": 0.18941683322191238,
      "num_tokens": 59127785.0,
      "step": 31715
    },
    {
      "entropy": 5.999285697937012,
      "epoch": 2.801130342635111,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042181546932378114,
      "loss": 5.659,
      "mean_token_accuracy": 0.1920042261481285,
      "num_tokens": 59137599.0,
      "step": 31720
    },
    {
      "entropy": 5.974074840545654,
      "epoch": 2.8015718827269516,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042179160433544126,
      "loss": 5.6484,
      "mean_token_accuracy": 0.18666889667510986,
      "num_tokens": 59146089.0,
      "step": 31725
    },
    {
      "entropy": 5.9618659019470215,
      "epoch": 2.802013422818792,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004217677364715373,
      "loss": 5.669,
      "mean_token_accuracy": 0.17953677475452423,
      "num_tokens": 59155749.0,
      "step": 31730
    },
    {
      "entropy": 6.038573360443115,
      "epoch": 2.802454962910632,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042174386573253696,
      "loss": 5.6696,
      "mean_token_accuracy": 0.18178068101406097,
      "num_tokens": 59165714.0,
      "step": 31735
    },
    {
      "entropy": 6.046713399887085,
      "epoch": 2.8028965030024726,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042171999211890773,
      "loss": 5.7207,
      "mean_token_accuracy": 0.1782287687063217,
      "num_tokens": 59174738.0,
      "step": 31740
    },
    {
      "entropy": 5.902861213684082,
      "epoch": 2.803338043094313,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004216961156311173,
      "loss": 5.5353,
      "mean_token_accuracy": 0.20054279416799545,
      "num_tokens": 59183380.0,
      "step": 31745
    },
    {
      "entropy": 5.912584638595581,
      "epoch": 2.803779583186153,
      "grad_norm": 1.375,
      "learning_rate": 0.0004216722362696334,
      "loss": 5.5172,
      "mean_token_accuracy": 0.19979477822780609,
      "num_tokens": 59193265.0,
      "step": 31750
    },
    {
      "entropy": 5.95393648147583,
      "epoch": 2.8042211232779937,
      "grad_norm": 1.546875,
      "learning_rate": 0.00042164835403492377,
      "loss": 5.6286,
      "mean_token_accuracy": 0.18237049877643585,
      "num_tokens": 59202543.0,
      "step": 31755
    },
    {
      "entropy": 5.969543313980102,
      "epoch": 2.804662663369834,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042162446892745633,
      "loss": 5.6178,
      "mean_token_accuracy": 0.1832516998052597,
      "num_tokens": 59211564.0,
      "step": 31760
    },
    {
      "entropy": 5.922652006149292,
      "epoch": 2.805104203461674,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00042160058094769886,
      "loss": 5.6199,
      "mean_token_accuracy": 0.19059715420007706,
      "num_tokens": 59220908.0,
      "step": 31765
    },
    {
      "entropy": 6.00909194946289,
      "epoch": 2.8055457435535147,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042157669009611947,
      "loss": 5.6536,
      "mean_token_accuracy": 0.18101921677589417,
      "num_tokens": 59231230.0,
      "step": 31770
    },
    {
      "entropy": 5.874865007400513,
      "epoch": 2.805987283645355,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042155279637318604,
      "loss": 5.447,
      "mean_token_accuracy": 0.20754718780517578,
      "num_tokens": 59240356.0,
      "step": 31775
    },
    {
      "entropy": 5.854852533340454,
      "epoch": 2.8064288237371953,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042152889977936664,
      "loss": 5.5637,
      "mean_token_accuracy": 0.19218663424253463,
      "num_tokens": 59248834.0,
      "step": 31780
    },
    {
      "entropy": 6.00075740814209,
      "epoch": 2.8068703638290358,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004215050003151295,
      "loss": 5.7012,
      "mean_token_accuracy": 0.17917584478855134,
      "num_tokens": 59258683.0,
      "step": 31785
    },
    {
      "entropy": 6.054208183288575,
      "epoch": 2.8073119039208763,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004214810979809427,
      "loss": 5.7296,
      "mean_token_accuracy": 0.17749354392290115,
      "num_tokens": 59268291.0,
      "step": 31790
    },
    {
      "entropy": 6.024962663650513,
      "epoch": 2.8077534440127163,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042145719277727453,
      "loss": 5.7753,
      "mean_token_accuracy": 0.17067302614450455,
      "num_tokens": 59278608.0,
      "step": 31795
    },
    {
      "entropy": 5.91436014175415,
      "epoch": 2.808194984104557,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004214332847045932,
      "loss": 5.5227,
      "mean_token_accuracy": 0.1955579102039337,
      "num_tokens": 59287859.0,
      "step": 31800
    },
    {
      "entropy": 5.8302771091461185,
      "epoch": 2.808636524196397,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004214093737633671,
      "loss": 5.5006,
      "mean_token_accuracy": 0.1974669948220253,
      "num_tokens": 59296996.0,
      "step": 31805
    },
    {
      "entropy": 5.937426805496216,
      "epoch": 2.8090780642882374,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042138545995406467,
      "loss": 5.5918,
      "mean_token_accuracy": 0.18784001469612122,
      "num_tokens": 59306776.0,
      "step": 31810
    },
    {
      "entropy": 5.917514657974243,
      "epoch": 2.809519604380078,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004213615432771543,
      "loss": 5.5581,
      "mean_token_accuracy": 0.18747795671224593,
      "num_tokens": 59316271.0,
      "step": 31815
    },
    {
      "entropy": 5.845629167556763,
      "epoch": 2.809961144471918,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004213376237331046,
      "loss": 5.499,
      "mean_token_accuracy": 0.19036347419023514,
      "num_tokens": 59324654.0,
      "step": 31820
    },
    {
      "entropy": 5.9219683647155765,
      "epoch": 2.8104026845637584,
      "grad_norm": 1.625,
      "learning_rate": 0.000421313701322384,
      "loss": 5.65,
      "mean_token_accuracy": 0.18348241299390794,
      "num_tokens": 59333681.0,
      "step": 31825
    },
    {
      "entropy": 5.984883260726929,
      "epoch": 2.8108442246555985,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004212897760454613,
      "loss": 5.6748,
      "mean_token_accuracy": 0.18306062817573548,
      "num_tokens": 59342455.0,
      "step": 31830
    },
    {
      "entropy": 5.97214412689209,
      "epoch": 2.811285764747439,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000421265847902805,
      "loss": 5.6164,
      "mean_token_accuracy": 0.1836322322487831,
      "num_tokens": 59351435.0,
      "step": 31835
    },
    {
      "entropy": 5.9629714012146,
      "epoch": 2.8117273048392795,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042124191689488395,
      "loss": 5.6571,
      "mean_token_accuracy": 0.18701203018426896,
      "num_tokens": 59360832.0,
      "step": 31840
    },
    {
      "entropy": 5.964438390731812,
      "epoch": 2.8121688449311195,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004212179830221669,
      "loss": 5.5436,
      "mean_token_accuracy": 0.18664126396179198,
      "num_tokens": 59368898.0,
      "step": 31845
    },
    {
      "entropy": 5.902813339233399,
      "epoch": 2.81261038502296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004211940462851227,
      "loss": 5.5146,
      "mean_token_accuracy": 0.19965990632772446,
      "num_tokens": 59377951.0,
      "step": 31850
    },
    {
      "entropy": 5.9256752014160154,
      "epoch": 2.8130519251148005,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042117010668422033,
      "loss": 5.5866,
      "mean_token_accuracy": 0.1928395375609398,
      "num_tokens": 59386988.0,
      "step": 31855
    },
    {
      "entropy": 5.883549165725708,
      "epoch": 2.8134934652066406,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004211461642199287,
      "loss": 5.5259,
      "mean_token_accuracy": 0.19796404540538787,
      "num_tokens": 59396658.0,
      "step": 31860
    },
    {
      "entropy": 5.857300472259522,
      "epoch": 2.813935005298481,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004211222188927168,
      "loss": 5.5526,
      "mean_token_accuracy": 0.19173527210950853,
      "num_tokens": 59407086.0,
      "step": 31865
    },
    {
      "entropy": 5.967449808120728,
      "epoch": 2.8143765453903216,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004210982707030538,
      "loss": 5.6491,
      "mean_token_accuracy": 0.182010480761528,
      "num_tokens": 59416141.0,
      "step": 31870
    },
    {
      "entropy": 5.973543739318847,
      "epoch": 2.8148180854821616,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042107431965140876,
      "loss": 5.6141,
      "mean_token_accuracy": 0.18771034181118013,
      "num_tokens": 59425856.0,
      "step": 31875
    },
    {
      "entropy": 6.02069616317749,
      "epoch": 2.815259625574002,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004210503657382508,
      "loss": 5.6988,
      "mean_token_accuracy": 0.1830527126789093,
      "num_tokens": 59435338.0,
      "step": 31880
    },
    {
      "entropy": 5.926805543899536,
      "epoch": 2.8157011656658426,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004210264089640493,
      "loss": 5.5767,
      "mean_token_accuracy": 0.19052637368440628,
      "num_tokens": 59444453.0,
      "step": 31885
    },
    {
      "entropy": 5.881967687606812,
      "epoch": 2.8161427057576827,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042100244932927346,
      "loss": 5.5246,
      "mean_token_accuracy": 0.19479372054338456,
      "num_tokens": 59453342.0,
      "step": 31890
    },
    {
      "entropy": 5.915037441253662,
      "epoch": 2.816584245849523,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004209784868343927,
      "loss": 5.641,
      "mean_token_accuracy": 0.1854898601770401,
      "num_tokens": 59462669.0,
      "step": 31895
    },
    {
      "entropy": 5.970892858505249,
      "epoch": 2.8170257859413637,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004209545214798764,
      "loss": 5.6315,
      "mean_token_accuracy": 0.18059488385915756,
      "num_tokens": 59472400.0,
      "step": 31900
    },
    {
      "entropy": 6.007495021820068,
      "epoch": 2.8174673260332037,
      "grad_norm": 1.3125,
      "learning_rate": 0.000420930553266194,
      "loss": 5.5627,
      "mean_token_accuracy": 0.18978487253189086,
      "num_tokens": 59482306.0,
      "step": 31905
    },
    {
      "entropy": 5.9364461421966555,
      "epoch": 2.817908866125044,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004209065821938151,
      "loss": 5.6235,
      "mean_token_accuracy": 0.18682666569948198,
      "num_tokens": 59491876.0,
      "step": 31910
    },
    {
      "entropy": 5.904759788513184,
      "epoch": 2.8183504062168847,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004208826082632092,
      "loss": 5.6681,
      "mean_token_accuracy": 0.18660243302583696,
      "num_tokens": 59502007.0,
      "step": 31915
    },
    {
      "entropy": 6.02505316734314,
      "epoch": 2.8187919463087248,
      "grad_norm": 1.265625,
      "learning_rate": 0.000420858631474846,
      "loss": 5.6739,
      "mean_token_accuracy": 0.1881355807185173,
      "num_tokens": 59511770.0,
      "step": 31920
    },
    {
      "entropy": 6.059689092636108,
      "epoch": 2.8192334864005653,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004208346518291952,
      "loss": 5.7604,
      "mean_token_accuracy": 0.1777110993862152,
      "num_tokens": 59521303.0,
      "step": 31925
    },
    {
      "entropy": 6.077617835998535,
      "epoch": 2.8196750264924058,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042081066932672645,
      "loss": 5.7616,
      "mean_token_accuracy": 0.17628901749849318,
      "num_tokens": 59532104.0,
      "step": 31930
    },
    {
      "entropy": 5.968693971633911,
      "epoch": 2.820116566584246,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042078668396790963,
      "loss": 5.5656,
      "mean_token_accuracy": 0.19093525856733323,
      "num_tokens": 59542163.0,
      "step": 31935
    },
    {
      "entropy": 5.942325067520142,
      "epoch": 2.8205581066760863,
      "grad_norm": 1.65625,
      "learning_rate": 0.00042076269575321457,
      "loss": 5.6391,
      "mean_token_accuracy": 0.18601624816656112,
      "num_tokens": 59552588.0,
      "step": 31940
    },
    {
      "entropy": 5.874893569946289,
      "epoch": 2.820999646767927,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004207387046831112,
      "loss": 5.6526,
      "mean_token_accuracy": 0.18343194723129272,
      "num_tokens": 59561531.0,
      "step": 31945
    },
    {
      "entropy": 5.99479341506958,
      "epoch": 2.821441186859767,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004207147107580695,
      "loss": 5.5322,
      "mean_token_accuracy": 0.1909334972500801,
      "num_tokens": 59570960.0,
      "step": 31950
    },
    {
      "entropy": 6.021224403381348,
      "epoch": 2.8218827269516074,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042069071397855953,
      "loss": 5.6938,
      "mean_token_accuracy": 0.1800004869699478,
      "num_tokens": 59579734.0,
      "step": 31955
    },
    {
      "entropy": 5.8883545875549315,
      "epoch": 2.8223242670434474,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042066671434505125,
      "loss": 5.6705,
      "mean_token_accuracy": 0.18708495944738388,
      "num_tokens": 59589260.0,
      "step": 31960
    },
    {
      "entropy": 5.9624950885772705,
      "epoch": 2.822765807135288,
      "grad_norm": 1.265625,
      "learning_rate": 0.000420642711858015,
      "loss": 5.5767,
      "mean_token_accuracy": 0.19451371431350709,
      "num_tokens": 59597972.0,
      "step": 31965
    },
    {
      "entropy": 5.893039751052856,
      "epoch": 2.823207347227128,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042061870651792076,
      "loss": 5.5423,
      "mean_token_accuracy": 0.19012268930673598,
      "num_tokens": 59607645.0,
      "step": 31970
    },
    {
      "entropy": 6.0507204055786135,
      "epoch": 2.8236488873189685,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042059469832523894,
      "loss": 5.6562,
      "mean_token_accuracy": 0.17935001105070114,
      "num_tokens": 59616405.0,
      "step": 31975
    },
    {
      "entropy": 5.86152400970459,
      "epoch": 2.824090427410809,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004205706872804397,
      "loss": 5.5428,
      "mean_token_accuracy": 0.19827741235494614,
      "num_tokens": 59625635.0,
      "step": 31980
    },
    {
      "entropy": 5.763715410232544,
      "epoch": 2.824531967502649,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004205466733839935,
      "loss": 5.3941,
      "mean_token_accuracy": 0.2046603664755821,
      "num_tokens": 59635115.0,
      "step": 31985
    },
    {
      "entropy": 5.951761770248413,
      "epoch": 2.8249735075944895,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004205226566363708,
      "loss": 5.6662,
      "mean_token_accuracy": 0.1872902825474739,
      "num_tokens": 59644249.0,
      "step": 31990
    },
    {
      "entropy": 5.981301546096802,
      "epoch": 2.82541504768633,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042049863703804205,
      "loss": 5.5768,
      "mean_token_accuracy": 0.19137880355119705,
      "num_tokens": 59653190.0,
      "step": 31995
    },
    {
      "entropy": 5.924276685714721,
      "epoch": 2.82585658777817,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004204746145894777,
      "loss": 5.5252,
      "mean_token_accuracy": 0.18904548436403273,
      "num_tokens": 59662181.0,
      "step": 32000
    },
    {
      "entropy": 5.875684595108032,
      "epoch": 2.8262981278700106,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004204505892911484,
      "loss": 5.6081,
      "mean_token_accuracy": 0.19375622868537903,
      "num_tokens": 59670949.0,
      "step": 32005
    },
    {
      "entropy": 5.934569883346557,
      "epoch": 2.826739667961851,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004204265611435247,
      "loss": 5.6586,
      "mean_token_accuracy": 0.18764194399118422,
      "num_tokens": 59680112.0,
      "step": 32010
    },
    {
      "entropy": 6.006656360626221,
      "epoch": 2.827181208053691,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042040253014707745,
      "loss": 5.5885,
      "mean_token_accuracy": 0.1856154352426529,
      "num_tokens": 59688910.0,
      "step": 32015
    },
    {
      "entropy": 5.990404176712036,
      "epoch": 2.8276227481455316,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004203784963022774,
      "loss": 5.6589,
      "mean_token_accuracy": 0.18785984814167023,
      "num_tokens": 59699047.0,
      "step": 32020
    },
    {
      "entropy": 5.953207015991211,
      "epoch": 2.828064288237372,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00042035445960959513,
      "loss": 5.6771,
      "mean_token_accuracy": 0.1859604611992836,
      "num_tokens": 59708545.0,
      "step": 32025
    },
    {
      "entropy": 6.061542129516601,
      "epoch": 2.828505828329212,
      "grad_norm": 1.25,
      "learning_rate": 0.0004203304200695017,
      "loss": 5.7302,
      "mean_token_accuracy": 0.18113005608320237,
      "num_tokens": 59718118.0,
      "step": 32030
    },
    {
      "entropy": 5.987319898605347,
      "epoch": 2.8289473684210527,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004203063776824681,
      "loss": 5.5351,
      "mean_token_accuracy": 0.19687182009220122,
      "num_tokens": 59726465.0,
      "step": 32035
    },
    {
      "entropy": 5.971376895904541,
      "epoch": 2.829388908512893,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042028233244896507,
      "loss": 5.6078,
      "mean_token_accuracy": 0.19194213151931763,
      "num_tokens": 59735399.0,
      "step": 32040
    },
    {
      "entropy": 5.819417715072632,
      "epoch": 2.829830448604733,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042025828436946386,
      "loss": 5.5318,
      "mean_token_accuracy": 0.18997405469417572,
      "num_tokens": 59744725.0,
      "step": 32045
    },
    {
      "entropy": 5.867546129226684,
      "epoch": 2.8302719886965737,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004202342334444355,
      "loss": 5.565,
      "mean_token_accuracy": 0.18989166915416716,
      "num_tokens": 59754317.0,
      "step": 32050
    },
    {
      "entropy": 6.0196089267730715,
      "epoch": 2.830713528788414,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042021017967435104,
      "loss": 5.7044,
      "mean_token_accuracy": 0.18116437792778015,
      "num_tokens": 59764951.0,
      "step": 32055
    },
    {
      "entropy": 6.003440237045288,
      "epoch": 2.8311550688802543,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004201861230596817,
      "loss": 5.6651,
      "mean_token_accuracy": 0.18125063478946685,
      "num_tokens": 59775116.0,
      "step": 32060
    },
    {
      "entropy": 5.923911285400391,
      "epoch": 2.8315966089720948,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042016206360089893,
      "loss": 5.4764,
      "mean_token_accuracy": 0.21052817404270172,
      "num_tokens": 59783253.0,
      "step": 32065
    },
    {
      "entropy": 5.9049779891967775,
      "epoch": 2.8320381490639353,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004201380012984738,
      "loss": 5.5893,
      "mean_token_accuracy": 0.1909286394715309,
      "num_tokens": 59793020.0,
      "step": 32070
    },
    {
      "entropy": 5.989395380020142,
      "epoch": 2.8324796891557753,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004201139361528777,
      "loss": 5.6883,
      "mean_token_accuracy": 0.180458228290081,
      "num_tokens": 59803860.0,
      "step": 32075
    },
    {
      "entropy": 5.925155162811279,
      "epoch": 2.832921229247616,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004200898681645822,
      "loss": 5.4936,
      "mean_token_accuracy": 0.19779870063066482,
      "num_tokens": 59813695.0,
      "step": 32080
    },
    {
      "entropy": 5.994637298583984,
      "epoch": 2.8333627693394563,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004200657973340588,
      "loss": 5.7193,
      "mean_token_accuracy": 0.1763078436255455,
      "num_tokens": 59823445.0,
      "step": 32085
    },
    {
      "entropy": 5.921116781234741,
      "epoch": 2.8338043094312964,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004200417236617788,
      "loss": 5.5694,
      "mean_token_accuracy": 0.19712032079696656,
      "num_tokens": 59833242.0,
      "step": 32090
    },
    {
      "entropy": 5.951837158203125,
      "epoch": 2.834245849523137,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042001764714821394,
      "loss": 5.5367,
      "mean_token_accuracy": 0.19230177402496337,
      "num_tokens": 59843364.0,
      "step": 32095
    },
    {
      "entropy": 5.9374815940856935,
      "epoch": 2.834687389614977,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004199935677938359,
      "loss": 5.5485,
      "mean_token_accuracy": 0.19829898774623872,
      "num_tokens": 59852940.0,
      "step": 32100
    },
    {
      "entropy": 5.950201463699341,
      "epoch": 2.8351289297068174,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004199694855991162,
      "loss": 5.6606,
      "mean_token_accuracy": 0.1846092164516449,
      "num_tokens": 59863313.0,
      "step": 32105
    },
    {
      "entropy": 6.039752721786499,
      "epoch": 2.8355704697986575,
      "grad_norm": 1.375,
      "learning_rate": 0.0004199454005645269,
      "loss": 5.7208,
      "mean_token_accuracy": 0.1775524213910103,
      "num_tokens": 59873307.0,
      "step": 32110
    },
    {
      "entropy": 5.951941204071045,
      "epoch": 2.836012009890498,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004199213126905395,
      "loss": 5.6481,
      "mean_token_accuracy": 0.19147749692201615,
      "num_tokens": 59882958.0,
      "step": 32115
    },
    {
      "entropy": 5.966812610626221,
      "epoch": 2.8364535499823385,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004198972219776261,
      "loss": 5.5811,
      "mean_token_accuracy": 0.19090806990861892,
      "num_tokens": 59892259.0,
      "step": 32120
    },
    {
      "entropy": 5.930962181091308,
      "epoch": 2.8368950900741785,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004198731284262585,
      "loss": 5.6352,
      "mean_token_accuracy": 0.18625596910715103,
      "num_tokens": 59901836.0,
      "step": 32125
    },
    {
      "entropy": 5.922631025314331,
      "epoch": 2.837336630166019,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041984903203690874,
      "loss": 5.4949,
      "mean_token_accuracy": 0.1888242706656456,
      "num_tokens": 59910751.0,
      "step": 32130
    },
    {
      "entropy": 5.968241357803345,
      "epoch": 2.8377781702578595,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004198249328100487,
      "loss": 5.5423,
      "mean_token_accuracy": 0.2004029706120491,
      "num_tokens": 59920101.0,
      "step": 32135
    },
    {
      "entropy": 5.9650777816772464,
      "epoch": 2.8382197103496996,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004198008307461507,
      "loss": 5.6087,
      "mean_token_accuracy": 0.18645275086164476,
      "num_tokens": 59928710.0,
      "step": 32140
    },
    {
      "entropy": 5.992299365997314,
      "epoch": 2.83866125044154,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004197767258456868,
      "loss": 5.6801,
      "mean_token_accuracy": 0.1770576611161232,
      "num_tokens": 59939277.0,
      "step": 32145
    },
    {
      "entropy": 6.066553974151612,
      "epoch": 2.8391027905333805,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041975261810912905,
      "loss": 5.7252,
      "mean_token_accuracy": 0.17570998519659042,
      "num_tokens": 59949093.0,
      "step": 32150
    },
    {
      "entropy": 5.955040454864502,
      "epoch": 2.8395443306252206,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041972850753695,
      "loss": 5.6104,
      "mean_token_accuracy": 0.18521528393030168,
      "num_tokens": 59958528.0,
      "step": 32155
    },
    {
      "entropy": 5.953333950042724,
      "epoch": 2.839985870717061,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004197043941296217,
      "loss": 5.6077,
      "mean_token_accuracy": 0.18795430660247803,
      "num_tokens": 59967444.0,
      "step": 32160
    },
    {
      "entropy": 5.956669425964355,
      "epoch": 2.8404274108089016,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004196802778876166,
      "loss": 5.6278,
      "mean_token_accuracy": 0.18976614326238633,
      "num_tokens": 59976997.0,
      "step": 32165
    },
    {
      "entropy": 5.894727182388306,
      "epoch": 2.8408689509007417,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004196561588114072,
      "loss": 5.5192,
      "mean_token_accuracy": 0.19536751359701157,
      "num_tokens": 59986352.0,
      "step": 32170
    },
    {
      "entropy": 6.009942245483399,
      "epoch": 2.841310490992582,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041963203690146585,
      "loss": 5.7291,
      "mean_token_accuracy": 0.17943375408649445,
      "num_tokens": 59995345.0,
      "step": 32175
    },
    {
      "entropy": 6.006143569946289,
      "epoch": 2.8417520310844226,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004196079121582652,
      "loss": 5.5995,
      "mean_token_accuracy": 0.189512237906456,
      "num_tokens": 60004792.0,
      "step": 32180
    },
    {
      "entropy": 5.933780908584595,
      "epoch": 2.8421935711762627,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004195837845822778,
      "loss": 5.598,
      "mean_token_accuracy": 0.192976276576519,
      "num_tokens": 60013472.0,
      "step": 32185
    },
    {
      "entropy": 5.948057222366333,
      "epoch": 2.842635111268103,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004195596541739763,
      "loss": 5.6217,
      "mean_token_accuracy": 0.1897917553782463,
      "num_tokens": 60023753.0,
      "step": 32190
    },
    {
      "entropy": 5.977219724655152,
      "epoch": 2.8430766513599437,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041953552093383336,
      "loss": 5.5736,
      "mean_token_accuracy": 0.18535277992486954,
      "num_tokens": 60032850.0,
      "step": 32195
    },
    {
      "entropy": 5.904814624786377,
      "epoch": 2.8435181914517837,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041951138486232177,
      "loss": 5.5765,
      "mean_token_accuracy": 0.1819789692759514,
      "num_tokens": 60041644.0,
      "step": 32200
    },
    {
      "entropy": 5.958833074569702,
      "epoch": 2.8439597315436242,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004194872459599144,
      "loss": 5.6466,
      "mean_token_accuracy": 0.19330279976129533,
      "num_tokens": 60051453.0,
      "step": 32205
    },
    {
      "entropy": 5.958893346786499,
      "epoch": 2.8444012716354647,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041946310422708396,
      "loss": 5.627,
      "mean_token_accuracy": 0.18326770514249802,
      "num_tokens": 60060408.0,
      "step": 32210
    },
    {
      "entropy": 5.951035594940185,
      "epoch": 2.844842811727305,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004194389596643035,
      "loss": 5.6722,
      "mean_token_accuracy": 0.18893194198608398,
      "num_tokens": 60069981.0,
      "step": 32215
    },
    {
      "entropy": 6.050418376922607,
      "epoch": 2.8452843518191453,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041941481227204596,
      "loss": 5.7333,
      "mean_token_accuracy": 0.180136975646019,
      "num_tokens": 60078785.0,
      "step": 32220
    },
    {
      "entropy": 6.012587356567383,
      "epoch": 2.845725891910986,
      "grad_norm": 1.25,
      "learning_rate": 0.00041939066205078443,
      "loss": 5.5407,
      "mean_token_accuracy": 0.19239441752433778,
      "num_tokens": 60087639.0,
      "step": 32225
    },
    {
      "entropy": 5.873753643035888,
      "epoch": 2.846167432002826,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041936650900099186,
      "loss": 5.5486,
      "mean_token_accuracy": 0.1880297392606735,
      "num_tokens": 60096385.0,
      "step": 32230
    },
    {
      "entropy": 5.945104742050171,
      "epoch": 2.8466089720946663,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041934235312314163,
      "loss": 5.6106,
      "mean_token_accuracy": 0.18740931004285813,
      "num_tokens": 60105449.0,
      "step": 32235
    },
    {
      "entropy": 5.963177967071533,
      "epoch": 2.8470505121865064,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004193181944177066,
      "loss": 5.6234,
      "mean_token_accuracy": 0.19402198046445845,
      "num_tokens": 60115060.0,
      "step": 32240
    },
    {
      "entropy": 5.969102811813355,
      "epoch": 2.847492052278347,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004192940328851604,
      "loss": 5.6324,
      "mean_token_accuracy": 0.1819182574748993,
      "num_tokens": 60124737.0,
      "step": 32245
    },
    {
      "entropy": 5.93684606552124,
      "epoch": 2.847933592370187,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004192698685259761,
      "loss": 5.645,
      "mean_token_accuracy": 0.1856759801506996,
      "num_tokens": 60133964.0,
      "step": 32250
    },
    {
      "entropy": 5.937462568283081,
      "epoch": 2.8483751324620274,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041924570134062707,
      "loss": 5.5463,
      "mean_token_accuracy": 0.19583469331264497,
      "num_tokens": 60143777.0,
      "step": 32255
    },
    {
      "entropy": 5.980064582824707,
      "epoch": 2.848816672553868,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004192215313295868,
      "loss": 5.5929,
      "mean_token_accuracy": 0.19009040147066117,
      "num_tokens": 60152813.0,
      "step": 32260
    },
    {
      "entropy": 5.933366918563843,
      "epoch": 2.849258212645708,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004191973584933288,
      "loss": 5.6375,
      "mean_token_accuracy": 0.18910027295351028,
      "num_tokens": 60161789.0,
      "step": 32265
    },
    {
      "entropy": 5.9062339782714846,
      "epoch": 2.8496997527375485,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041917318283232654,
      "loss": 5.6129,
      "mean_token_accuracy": 0.17977968007326126,
      "num_tokens": 60171171.0,
      "step": 32270
    },
    {
      "entropy": 6.022718143463135,
      "epoch": 2.850141292829389,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004191490043470536,
      "loss": 5.6199,
      "mean_token_accuracy": 0.191828091442585,
      "num_tokens": 60180512.0,
      "step": 32275
    },
    {
      "entropy": 6.047695350646973,
      "epoch": 2.850582832921229,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041912482303798373,
      "loss": 5.6702,
      "mean_token_accuracy": 0.18544241935014724,
      "num_tokens": 60189675.0,
      "step": 32280
    },
    {
      "entropy": 5.948505306243897,
      "epoch": 2.8510243730130695,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041910063890559043,
      "loss": 5.5674,
      "mean_token_accuracy": 0.1890127941966057,
      "num_tokens": 60199037.0,
      "step": 32285
    },
    {
      "entropy": 5.9508500576019285,
      "epoch": 2.85146591310491,
      "grad_norm": 1.125,
      "learning_rate": 0.00041907645195034764,
      "loss": 5.534,
      "mean_token_accuracy": 0.19887761175632476,
      "num_tokens": 60208631.0,
      "step": 32290
    },
    {
      "entropy": 5.889002943038941,
      "epoch": 2.85190745319675,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004190522621727291,
      "loss": 5.5278,
      "mean_token_accuracy": 0.19616444259881974,
      "num_tokens": 60217977.0,
      "step": 32295
    },
    {
      "entropy": 5.92779769897461,
      "epoch": 2.8523489932885906,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004190280695732086,
      "loss": 5.5935,
      "mean_token_accuracy": 0.19207098186016083,
      "num_tokens": 60227264.0,
      "step": 32300
    },
    {
      "entropy": 5.9761559009552006,
      "epoch": 2.852790533380431,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004190038741522602,
      "loss": 5.6367,
      "mean_token_accuracy": 0.18538013249635696,
      "num_tokens": 60236355.0,
      "step": 32305
    },
    {
      "entropy": 5.909020233154297,
      "epoch": 2.853232073472271,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041897967591035784,
      "loss": 5.5805,
      "mean_token_accuracy": 0.1836825653910637,
      "num_tokens": 60245993.0,
      "step": 32310
    },
    {
      "entropy": 5.96816086769104,
      "epoch": 2.8536736135641116,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004189554748479755,
      "loss": 5.5648,
      "mean_token_accuracy": 0.19841485917568208,
      "num_tokens": 60255449.0,
      "step": 32315
    },
    {
      "entropy": 5.945553112030029,
      "epoch": 2.854115153655952,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041893127096558714,
      "loss": 5.594,
      "mean_token_accuracy": 0.1909114897251129,
      "num_tokens": 60264708.0,
      "step": 32320
    },
    {
      "entropy": 6.017808055877685,
      "epoch": 2.854556693747792,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004189070642636672,
      "loss": 5.6865,
      "mean_token_accuracy": 0.18324354737997056,
      "num_tokens": 60273654.0,
      "step": 32325
    },
    {
      "entropy": 5.960073375701905,
      "epoch": 2.8549982338396327,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041888285474268964,
      "loss": 5.4966,
      "mean_token_accuracy": 0.20006991028785706,
      "num_tokens": 60282479.0,
      "step": 32330
    },
    {
      "entropy": 5.844095182418823,
      "epoch": 2.855439773931473,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041885864240312883,
      "loss": 5.5563,
      "mean_token_accuracy": 0.19630260318517684,
      "num_tokens": 60292622.0,
      "step": 32335
    },
    {
      "entropy": 5.9262189865112305,
      "epoch": 2.8558813140233132,
      "grad_norm": 1.125,
      "learning_rate": 0.00041883442724545894,
      "loss": 5.5755,
      "mean_token_accuracy": 0.18195835053920745,
      "num_tokens": 60302137.0,
      "step": 32340
    },
    {
      "entropy": 5.995759582519531,
      "epoch": 2.8563228541151537,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004188102092701546,
      "loss": 5.67,
      "mean_token_accuracy": 0.17401079535484315,
      "num_tokens": 60312253.0,
      "step": 32345
    },
    {
      "entropy": 6.004327726364136,
      "epoch": 2.8567643942069942,
      "grad_norm": 1.484375,
      "learning_rate": 0.00041878598847768986,
      "loss": 5.6476,
      "mean_token_accuracy": 0.18325887769460678,
      "num_tokens": 60321486.0,
      "step": 32350
    },
    {
      "entropy": 5.954946899414063,
      "epoch": 2.8572059342988343,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041876176486853946,
      "loss": 5.5787,
      "mean_token_accuracy": 0.18608253002166747,
      "num_tokens": 60330279.0,
      "step": 32355
    },
    {
      "entropy": 5.930724954605102,
      "epoch": 2.857647474390675,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004187375384431779,
      "loss": 5.6162,
      "mean_token_accuracy": 0.18078798800706863,
      "num_tokens": 60340915.0,
      "step": 32360
    },
    {
      "entropy": 5.898042345046997,
      "epoch": 2.8580890144825153,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004187133092020796,
      "loss": 5.6048,
      "mean_token_accuracy": 0.18242700397968292,
      "num_tokens": 60349849.0,
      "step": 32365
    },
    {
      "entropy": 6.010790491104126,
      "epoch": 2.8585305545743553,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004186890771457194,
      "loss": 5.6569,
      "mean_token_accuracy": 0.17905711531639099,
      "num_tokens": 60360072.0,
      "step": 32370
    },
    {
      "entropy": 6.060682773590088,
      "epoch": 2.858972094666196,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004186648422745718,
      "loss": 5.7012,
      "mean_token_accuracy": 0.17996898144483567,
      "num_tokens": 60369898.0,
      "step": 32375
    },
    {
      "entropy": 5.862459707260132,
      "epoch": 2.859413634758036,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041864060458911176,
      "loss": 5.5622,
      "mean_token_accuracy": 0.1977498710155487,
      "num_tokens": 60378573.0,
      "step": 32380
    },
    {
      "entropy": 5.966074085235595,
      "epoch": 2.8598551748498764,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004186163640898139,
      "loss": 5.6907,
      "mean_token_accuracy": 0.18508860915899278,
      "num_tokens": 60388479.0,
      "step": 32385
    },
    {
      "entropy": 5.929260158538819,
      "epoch": 2.8602967149417164,
      "grad_norm": 1.125,
      "learning_rate": 0.00041859212077715316,
      "loss": 5.5103,
      "mean_token_accuracy": 0.19432219713926316,
      "num_tokens": 60397854.0,
      "step": 32390
    },
    {
      "entropy": 5.947468519210815,
      "epoch": 2.860738255033557,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00041856787465160446,
      "loss": 5.654,
      "mean_token_accuracy": 0.18785181492567063,
      "num_tokens": 60407464.0,
      "step": 32395
    },
    {
      "entropy": 5.85146050453186,
      "epoch": 2.8611797951253974,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041854362571364274,
      "loss": 5.4586,
      "mean_token_accuracy": 0.20501975119113922,
      "num_tokens": 60415463.0,
      "step": 32400
    },
    {
      "entropy": 5.895313787460327,
      "epoch": 2.8616213352172375,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041851937396374296,
      "loss": 5.5531,
      "mean_token_accuracy": 0.18866271078586577,
      "num_tokens": 60424927.0,
      "step": 32405
    },
    {
      "entropy": 6.034522819519043,
      "epoch": 2.862062875309078,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004184951194023804,
      "loss": 5.6598,
      "mean_token_accuracy": 0.18780353516340256,
      "num_tokens": 60435353.0,
      "step": 32410
    },
    {
      "entropy": 5.960614728927612,
      "epoch": 2.8625044154009185,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041847086203003,
      "loss": 5.5734,
      "mean_token_accuracy": 0.1888716697692871,
      "num_tokens": 60445271.0,
      "step": 32415
    },
    {
      "entropy": 5.852257299423218,
      "epoch": 2.8629459554927585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041844660184716694,
      "loss": 5.4885,
      "mean_token_accuracy": 0.19409895837306976,
      "num_tokens": 60454817.0,
      "step": 32420
    },
    {
      "entropy": 5.964077186584473,
      "epoch": 2.863387495584599,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041842233885426664,
      "loss": 5.635,
      "mean_token_accuracy": 0.18542962670326232,
      "num_tokens": 60464796.0,
      "step": 32425
    },
    {
      "entropy": 5.944722318649292,
      "epoch": 2.8638290356764395,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041839807305180424,
      "loss": 5.5345,
      "mean_token_accuracy": 0.19689316153526307,
      "num_tokens": 60473766.0,
      "step": 32430
    },
    {
      "entropy": 6.067888307571411,
      "epoch": 2.8642705757682796,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004183738044402551,
      "loss": 5.7226,
      "mean_token_accuracy": 0.17441534996032715,
      "num_tokens": 60483564.0,
      "step": 32435
    },
    {
      "entropy": 6.000784540176392,
      "epoch": 2.86471211586012,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004183495330200947,
      "loss": 5.6575,
      "mean_token_accuracy": 0.18854375928640366,
      "num_tokens": 60492495.0,
      "step": 32440
    },
    {
      "entropy": 5.969932126998901,
      "epoch": 2.8651536559519606,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041832525879179853,
      "loss": 5.5654,
      "mean_token_accuracy": 0.19112169593572617,
      "num_tokens": 60502406.0,
      "step": 32445
    },
    {
      "entropy": 5.857871627807617,
      "epoch": 2.8655951960438006,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041830098175584196,
      "loss": 5.5323,
      "mean_token_accuracy": 0.19266361594200135,
      "num_tokens": 60511477.0,
      "step": 32450
    },
    {
      "entropy": 5.911859750747681,
      "epoch": 2.866036736135641,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041827670191270074,
      "loss": 5.6033,
      "mean_token_accuracy": 0.19218610376119613,
      "num_tokens": 60520709.0,
      "step": 32455
    },
    {
      "entropy": 5.941323947906494,
      "epoch": 2.8664782762274816,
      "grad_norm": 1.578125,
      "learning_rate": 0.00041825241926285033,
      "loss": 5.5991,
      "mean_token_accuracy": 0.19019549787044526,
      "num_tokens": 60529831.0,
      "step": 32460
    },
    {
      "entropy": 6.021410369873047,
      "epoch": 2.8669198163193217,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041822813380676665,
      "loss": 5.6587,
      "mean_token_accuracy": 0.18665360063314437,
      "num_tokens": 60539501.0,
      "step": 32465
    },
    {
      "entropy": 5.960210180282592,
      "epoch": 2.867361356411162,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041820384554492517,
      "loss": 5.6061,
      "mean_token_accuracy": 0.19107988029718398,
      "num_tokens": 60547533.0,
      "step": 32470
    },
    {
      "entropy": 5.901982164382934,
      "epoch": 2.8678028965030027,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004181795544778018,
      "loss": 5.6169,
      "mean_token_accuracy": 0.18312472254037857,
      "num_tokens": 60557047.0,
      "step": 32475
    },
    {
      "entropy": 6.038881826400757,
      "epoch": 2.8682444365948427,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004181552606058725,
      "loss": 5.5987,
      "mean_token_accuracy": 0.18889891505241393,
      "num_tokens": 60566333.0,
      "step": 32480
    },
    {
      "entropy": 5.956828832626343,
      "epoch": 2.8686859766866832,
      "grad_norm": 1.125,
      "learning_rate": 0.0004181309639296129,
      "loss": 5.5709,
      "mean_token_accuracy": 0.1929479420185089,
      "num_tokens": 60575090.0,
      "step": 32485
    },
    {
      "entropy": 5.825338506698609,
      "epoch": 2.8691275167785237,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004181066644494992,
      "loss": 5.5247,
      "mean_token_accuracy": 0.19184452444314956,
      "num_tokens": 60584000.0,
      "step": 32490
    },
    {
      "entropy": 5.80770092010498,
      "epoch": 2.869569056870364,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004180823621660073,
      "loss": 5.4787,
      "mean_token_accuracy": 0.19473305493593215,
      "num_tokens": 60593239.0,
      "step": 32495
    },
    {
      "entropy": 6.027522563934326,
      "epoch": 2.8700105969622043,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004180580570796134,
      "loss": 5.6692,
      "mean_token_accuracy": 0.1814475253224373,
      "num_tokens": 60601850.0,
      "step": 32500
    },
    {
      "entropy": 6.048062801361084,
      "epoch": 2.870452137054045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004180337491907934,
      "loss": 5.7122,
      "mean_token_accuracy": 0.1776737704873085,
      "num_tokens": 60611280.0,
      "step": 32505
    },
    {
      "entropy": 5.905357933044433,
      "epoch": 2.870893677145885,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004180094385000237,
      "loss": 5.6111,
      "mean_token_accuracy": 0.18927844911813735,
      "num_tokens": 60620738.0,
      "step": 32510
    },
    {
      "entropy": 5.930346870422364,
      "epoch": 2.8713352172377253,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041798512500778034,
      "loss": 5.6852,
      "mean_token_accuracy": 0.18300462067127227,
      "num_tokens": 60630472.0,
      "step": 32515
    },
    {
      "entropy": 5.966764688491821,
      "epoch": 2.8717767573295654,
      "grad_norm": 1.5,
      "learning_rate": 0.0004179608087145397,
      "loss": 5.6039,
      "mean_token_accuracy": 0.18637973219156265,
      "num_tokens": 60640555.0,
      "step": 32520
    },
    {
      "entropy": 6.029175567626953,
      "epoch": 2.872218297421406,
      "grad_norm": 1.25,
      "learning_rate": 0.00041793648962077816,
      "loss": 5.7089,
      "mean_token_accuracy": 0.18234895467758178,
      "num_tokens": 60649941.0,
      "step": 32525
    },
    {
      "entropy": 6.023813724517822,
      "epoch": 2.872659837513246,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041791216772697217,
      "loss": 5.649,
      "mean_token_accuracy": 0.18754871040582657,
      "num_tokens": 60659883.0,
      "step": 32530
    },
    {
      "entropy": 6.046991920471191,
      "epoch": 2.8731013776050864,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000417887843033598,
      "loss": 5.7014,
      "mean_token_accuracy": 0.18396992087364197,
      "num_tokens": 60669323.0,
      "step": 32535
    },
    {
      "entropy": 5.947601556777954,
      "epoch": 2.873542917696927,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004178635155411323,
      "loss": 5.6027,
      "mean_token_accuracy": 0.19489884227514268,
      "num_tokens": 60678497.0,
      "step": 32540
    },
    {
      "entropy": 5.903822708129883,
      "epoch": 2.873984457788767,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041783918525005155,
      "loss": 5.573,
      "mean_token_accuracy": 0.18946286886930466,
      "num_tokens": 60687312.0,
      "step": 32545
    },
    {
      "entropy": 5.937958002090454,
      "epoch": 2.8744259978806075,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041781485216083237,
      "loss": 5.5702,
      "mean_token_accuracy": 0.18705761283636094,
      "num_tokens": 60696765.0,
      "step": 32550
    },
    {
      "entropy": 5.993604421615601,
      "epoch": 2.874867537972448,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041779051627395146,
      "loss": 5.6452,
      "mean_token_accuracy": 0.20136420726776122,
      "num_tokens": 60705909.0,
      "step": 32555
    },
    {
      "entropy": 5.971966361999511,
      "epoch": 2.875309078064288,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041776617758988553,
      "loss": 5.499,
      "mean_token_accuracy": 0.19680954068899154,
      "num_tokens": 60715367.0,
      "step": 32560
    },
    {
      "entropy": 5.87441668510437,
      "epoch": 2.8757506181561285,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041774183610911143,
      "loss": 5.5762,
      "mean_token_accuracy": 0.18897462338209153,
      "num_tokens": 60724588.0,
      "step": 32565
    },
    {
      "entropy": 5.968279600143433,
      "epoch": 2.876192158247969,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004177174918321059,
      "loss": 5.619,
      "mean_token_accuracy": 0.19710659086704255,
      "num_tokens": 60732943.0,
      "step": 32570
    },
    {
      "entropy": 5.890835905075074,
      "epoch": 2.876633698339809,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041769314475934596,
      "loss": 5.5408,
      "mean_token_accuracy": 0.19482026100158692,
      "num_tokens": 60742378.0,
      "step": 32575
    },
    {
      "entropy": 5.904564237594604,
      "epoch": 2.8770752384316496,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004176687948913084,
      "loss": 5.5826,
      "mean_token_accuracy": 0.19332123994827272,
      "num_tokens": 60751175.0,
      "step": 32580
    },
    {
      "entropy": 6.017960405349731,
      "epoch": 2.87751677852349,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004176444422284703,
      "loss": 5.6501,
      "mean_token_accuracy": 0.18526534587144852,
      "num_tokens": 60760484.0,
      "step": 32585
    },
    {
      "entropy": 5.970536422729492,
      "epoch": 2.87795831861533,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004176200867713086,
      "loss": 5.5951,
      "mean_token_accuracy": 0.1883923292160034,
      "num_tokens": 60770055.0,
      "step": 32590
    },
    {
      "entropy": 5.931994199752808,
      "epoch": 2.8783998587071706,
      "grad_norm": 2.125,
      "learning_rate": 0.0004175957285203007,
      "loss": 5.6979,
      "mean_token_accuracy": 0.18375275135040284,
      "num_tokens": 60780347.0,
      "step": 32595
    },
    {
      "entropy": 5.96754412651062,
      "epoch": 2.878841398799011,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004175713674759235,
      "loss": 5.5352,
      "mean_token_accuracy": 0.1959535464644432,
      "num_tokens": 60789489.0,
      "step": 32600
    },
    {
      "entropy": 5.939057207107544,
      "epoch": 2.879282938890851,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00041754700363865426,
      "loss": 5.6345,
      "mean_token_accuracy": 0.1845103070139885,
      "num_tokens": 60797838.0,
      "step": 32605
    },
    {
      "entropy": 5.919227218627929,
      "epoch": 2.8797244789826917,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004175226370089703,
      "loss": 5.5582,
      "mean_token_accuracy": 0.1909296542406082,
      "num_tokens": 60808119.0,
      "step": 32610
    },
    {
      "entropy": 5.944025182723999,
      "epoch": 2.880166019074532,
      "grad_norm": 1.25,
      "learning_rate": 0.000417498267587349,
      "loss": 5.525,
      "mean_token_accuracy": 0.1911598727107048,
      "num_tokens": 60815905.0,
      "step": 32615
    },
    {
      "entropy": 5.92687840461731,
      "epoch": 2.8806075591663722,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004174738953742676,
      "loss": 5.6338,
      "mean_token_accuracy": 0.18932212889194489,
      "num_tokens": 60826258.0,
      "step": 32620
    },
    {
      "entropy": 5.971802759170532,
      "epoch": 2.8810490992582127,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004174495203702036,
      "loss": 5.5242,
      "mean_token_accuracy": 0.197598397731781,
      "num_tokens": 60834627.0,
      "step": 32625
    },
    {
      "entropy": 5.9781090259552006,
      "epoch": 2.8814906393500532,
      "grad_norm": 1.671875,
      "learning_rate": 0.00041742514257563464,
      "loss": 5.6445,
      "mean_token_accuracy": 0.18235762119293214,
      "num_tokens": 60844399.0,
      "step": 32630
    },
    {
      "entropy": 5.9155717372894285,
      "epoch": 2.8819321794418933,
      "grad_norm": 1.5,
      "learning_rate": 0.000417400761991038,
      "loss": 5.5731,
      "mean_token_accuracy": 0.1897434636950493,
      "num_tokens": 60852964.0,
      "step": 32635
    },
    {
      "entropy": 5.885687160491943,
      "epoch": 2.882373719533734,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041737637861689154,
      "loss": 5.5714,
      "mean_token_accuracy": 0.18865712434053422,
      "num_tokens": 60862100.0,
      "step": 32640
    },
    {
      "entropy": 6.068607807159424,
      "epoch": 2.8828152596255743,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004173519924536727,
      "loss": 5.6822,
      "mean_token_accuracy": 0.17989876866340637,
      "num_tokens": 60871761.0,
      "step": 32645
    },
    {
      "entropy": 5.897601652145386,
      "epoch": 2.8832567997174143,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004173276035018593,
      "loss": 5.568,
      "mean_token_accuracy": 0.1893341898918152,
      "num_tokens": 60880644.0,
      "step": 32650
    },
    {
      "entropy": 6.019065856933594,
      "epoch": 2.883698339809255,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041730321176192917,
      "loss": 5.608,
      "mean_token_accuracy": 0.18672902435064315,
      "num_tokens": 60889566.0,
      "step": 32655
    },
    {
      "entropy": 6.009263563156128,
      "epoch": 2.884139879901095,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00041727881723436,
      "loss": 5.7345,
      "mean_token_accuracy": 0.17933153510093688,
      "num_tokens": 60899591.0,
      "step": 32660
    },
    {
      "entropy": 5.990412473678589,
      "epoch": 2.8845814199929354,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041725441991962965,
      "loss": 5.6579,
      "mean_token_accuracy": 0.1788920894265175,
      "num_tokens": 60908928.0,
      "step": 32665
    },
    {
      "entropy": 5.966690921783448,
      "epoch": 2.8850229600847754,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004172300198182162,
      "loss": 5.6049,
      "mean_token_accuracy": 0.1872903361916542,
      "num_tokens": 60918105.0,
      "step": 32670
    },
    {
      "entropy": 5.969683599472046,
      "epoch": 2.885464500176616,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041720561693059756,
      "loss": 5.6186,
      "mean_token_accuracy": 0.18318091183900834,
      "num_tokens": 60927760.0,
      "step": 32675
    },
    {
      "entropy": 6.0098161697387695,
      "epoch": 2.8859060402684564,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041718121125725174,
      "loss": 5.6965,
      "mean_token_accuracy": 0.17538803964853286,
      "num_tokens": 60938551.0,
      "step": 32680
    },
    {
      "entropy": 5.929552125930786,
      "epoch": 2.8863475803602965,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041715680279865694,
      "loss": 5.5562,
      "mean_token_accuracy": 0.19478692710399628,
      "num_tokens": 60947457.0,
      "step": 32685
    },
    {
      "entropy": 5.813981056213379,
      "epoch": 2.886789120452137,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004171323915552911,
      "loss": 5.5624,
      "mean_token_accuracy": 0.19190119802951813,
      "num_tokens": 60957459.0,
      "step": 32690
    },
    {
      "entropy": 5.893312740325928,
      "epoch": 2.8872306605439775,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004171079775276326,
      "loss": 5.5644,
      "mean_token_accuracy": 0.18895624279975892,
      "num_tokens": 60966348.0,
      "step": 32695
    },
    {
      "entropy": 5.9883044242858885,
      "epoch": 2.8876722006358175,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041708356071615954,
      "loss": 5.5915,
      "mean_token_accuracy": 0.18891865462064744,
      "num_tokens": 60974778.0,
      "step": 32700
    },
    {
      "entropy": 5.966481161117554,
      "epoch": 2.888113740727658,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004170591411213504,
      "loss": 5.6752,
      "mean_token_accuracy": 0.1766636550426483,
      "num_tokens": 60984816.0,
      "step": 32705
    },
    {
      "entropy": 5.913921165466308,
      "epoch": 2.8885552808194985,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00041703471874368354,
      "loss": 5.5759,
      "mean_token_accuracy": 0.19210178256034852,
      "num_tokens": 60993723.0,
      "step": 32710
    },
    {
      "entropy": 5.9423504829406735,
      "epoch": 2.8889968209113386,
      "grad_norm": 1.390625,
      "learning_rate": 0.00041701029358363727,
      "loss": 5.6106,
      "mean_token_accuracy": 0.1958174392580986,
      "num_tokens": 61002363.0,
      "step": 32715
    },
    {
      "entropy": 6.001568603515625,
      "epoch": 2.889438361003179,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041698586564169014,
      "loss": 5.6219,
      "mean_token_accuracy": 0.18295260816812514,
      "num_tokens": 61011178.0,
      "step": 32720
    },
    {
      "entropy": 5.903592252731324,
      "epoch": 2.8898799010950196,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041696143491832063,
      "loss": 5.5872,
      "mean_token_accuracy": 0.19335272014141083,
      "num_tokens": 61020415.0,
      "step": 32725
    },
    {
      "entropy": 5.9593137264251705,
      "epoch": 2.8903214411868596,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004169370014140073,
      "loss": 5.6599,
      "mean_token_accuracy": 0.1885521486401558,
      "num_tokens": 61030537.0,
      "step": 32730
    },
    {
      "entropy": 5.989939165115357,
      "epoch": 2.8907629812787,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041691256512922897,
      "loss": 5.585,
      "mean_token_accuracy": 0.18989039361476898,
      "num_tokens": 61039666.0,
      "step": 32735
    },
    {
      "entropy": 5.989115619659424,
      "epoch": 2.8912045213705406,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041688812606446407,
      "loss": 5.6486,
      "mean_token_accuracy": 0.18491111993789672,
      "num_tokens": 61049066.0,
      "step": 32740
    },
    {
      "entropy": 5.95719199180603,
      "epoch": 2.8916460614623807,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00041686368422019153,
      "loss": 5.6814,
      "mean_token_accuracy": 0.17545851469039916,
      "num_tokens": 61059647.0,
      "step": 32745
    },
    {
      "entropy": 5.92459397315979,
      "epoch": 2.892087601554221,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004168392395968901,
      "loss": 5.5792,
      "mean_token_accuracy": 0.18575776219367982,
      "num_tokens": 61068814.0,
      "step": 32750
    },
    {
      "entropy": 5.986788034439087,
      "epoch": 2.8925291416460617,
      "grad_norm": 1.625,
      "learning_rate": 0.00041681479219503857,
      "loss": 5.6083,
      "mean_token_accuracy": 0.18563789874315262,
      "num_tokens": 61078480.0,
      "step": 32755
    },
    {
      "entropy": 5.963818693161011,
      "epoch": 2.8929706817379017,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000416790342015116,
      "loss": 5.6976,
      "mean_token_accuracy": 0.18775085359811783,
      "num_tokens": 61087830.0,
      "step": 32760
    },
    {
      "entropy": 5.924184942245484,
      "epoch": 2.8934122218297422,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041676588905760123,
      "loss": 5.5226,
      "mean_token_accuracy": 0.20092334002256393,
      "num_tokens": 61096675.0,
      "step": 32765
    },
    {
      "entropy": 5.928785181045532,
      "epoch": 2.8938537619215827,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004167414333229733,
      "loss": 5.5972,
      "mean_token_accuracy": 0.1966719925403595,
      "num_tokens": 61106526.0,
      "step": 32770
    },
    {
      "entropy": 5.96379451751709,
      "epoch": 2.8942953020134228,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004167169748117114,
      "loss": 5.6256,
      "mean_token_accuracy": 0.1905268833041191,
      "num_tokens": 61115851.0,
      "step": 32775
    },
    {
      "entropy": 5.894467353820801,
      "epoch": 2.8947368421052633,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004166925135242945,
      "loss": 5.5518,
      "mean_token_accuracy": 0.18872873932123185,
      "num_tokens": 61125222.0,
      "step": 32780
    },
    {
      "entropy": 5.959571599960327,
      "epoch": 2.8951783821971038,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041666804946120193,
      "loss": 5.6003,
      "mean_token_accuracy": 0.18774969577789308,
      "num_tokens": 61134672.0,
      "step": 32785
    },
    {
      "entropy": 5.91686978340149,
      "epoch": 2.895619922288944,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00041664358262291266,
      "loss": 5.5532,
      "mean_token_accuracy": 0.19384550154209138,
      "num_tokens": 61143809.0,
      "step": 32790
    },
    {
      "entropy": 5.933140611648559,
      "epoch": 2.8960614623807843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041661911300990625,
      "loss": 5.5597,
      "mean_token_accuracy": 0.18644407242536545,
      "num_tokens": 61153362.0,
      "step": 32795
    },
    {
      "entropy": 5.879608917236328,
      "epoch": 2.8965030024726244,
      "grad_norm": 1.171875,
      "learning_rate": 0.000416594640622662,
      "loss": 5.4565,
      "mean_token_accuracy": 0.2062701776623726,
      "num_tokens": 61162291.0,
      "step": 32800
    },
    {
      "entropy": 5.956330156326294,
      "epoch": 2.896944542564465,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041657016546165914,
      "loss": 5.6302,
      "mean_token_accuracy": 0.18704604506492614,
      "num_tokens": 61171335.0,
      "step": 32805
    },
    {
      "entropy": 5.940926122665405,
      "epoch": 2.897386082656305,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041654568752737736,
      "loss": 5.7119,
      "mean_token_accuracy": 0.180619977414608,
      "num_tokens": 61181137.0,
      "step": 32810
    },
    {
      "entropy": 5.990587949752808,
      "epoch": 2.8978276227481454,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041652120682029607,
      "loss": 5.6951,
      "mean_token_accuracy": 0.18461577445268632,
      "num_tokens": 61190743.0,
      "step": 32815
    },
    {
      "entropy": 5.903327894210816,
      "epoch": 2.898269162839986,
      "grad_norm": 1.25,
      "learning_rate": 0.0004164967233408948,
      "loss": 5.5345,
      "mean_token_accuracy": 0.1908460319042206,
      "num_tokens": 61200168.0,
      "step": 32820
    },
    {
      "entropy": 6.014680242538452,
      "epoch": 2.898710702931826,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004164722370896533,
      "loss": 5.6367,
      "mean_token_accuracy": 0.17846574783325195,
      "num_tokens": 61209930.0,
      "step": 32825
    },
    {
      "entropy": 5.875067806243896,
      "epoch": 2.8991522430236665,
      "grad_norm": 1.421875,
      "learning_rate": 0.000416447748067051,
      "loss": 5.535,
      "mean_token_accuracy": 0.19571694433689119,
      "num_tokens": 61219028.0,
      "step": 32830
    },
    {
      "entropy": 5.887606239318847,
      "epoch": 2.899593783115507,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004164232562735678,
      "loss": 5.54,
      "mean_token_accuracy": 0.1966194987297058,
      "num_tokens": 61228824.0,
      "step": 32835
    },
    {
      "entropy": 5.996668004989624,
      "epoch": 2.900035323207347,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004163987617096834,
      "loss": 5.6432,
      "mean_token_accuracy": 0.18085736483335496,
      "num_tokens": 61237862.0,
      "step": 32840
    },
    {
      "entropy": 5.925166654586792,
      "epoch": 2.9004768632991875,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004163742643758777,
      "loss": 5.5944,
      "mean_token_accuracy": 0.18978342711925505,
      "num_tokens": 61248779.0,
      "step": 32845
    },
    {
      "entropy": 5.96722445487976,
      "epoch": 2.900918403391028,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041634976427263064,
      "loss": 5.6288,
      "mean_token_accuracy": 0.18562151491641998,
      "num_tokens": 61259385.0,
      "step": 32850
    },
    {
      "entropy": 6.015726470947266,
      "epoch": 2.901359943482868,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000416325261400422,
      "loss": 5.6343,
      "mean_token_accuracy": 0.18677571266889573,
      "num_tokens": 61269789.0,
      "step": 32855
    },
    {
      "entropy": 6.0082540035247805,
      "epoch": 2.9018014835747086,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004163007557597319,
      "loss": 5.7064,
      "mean_token_accuracy": 0.18257752507925035,
      "num_tokens": 61279136.0,
      "step": 32860
    },
    {
      "entropy": 5.946903419494629,
      "epoch": 2.902243023666549,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004162762473510403,
      "loss": 5.5406,
      "mean_token_accuracy": 0.19605654776096343,
      "num_tokens": 61287448.0,
      "step": 32865
    },
    {
      "entropy": 5.904035520553589,
      "epoch": 2.902684563758389,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004162517361748275,
      "loss": 5.5576,
      "mean_token_accuracy": 0.18930386006832123,
      "num_tokens": 61296006.0,
      "step": 32870
    },
    {
      "entropy": 5.932345247268676,
      "epoch": 2.9031261038502296,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004162272222315734,
      "loss": 5.5734,
      "mean_token_accuracy": 0.187389075756073,
      "num_tokens": 61304420.0,
      "step": 32875
    },
    {
      "entropy": 5.934720897674561,
      "epoch": 2.90356764394207,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041620270552175843,
      "loss": 5.6938,
      "mean_token_accuracy": 0.18275884985923768,
      "num_tokens": 61314299.0,
      "step": 32880
    },
    {
      "entropy": 5.930859851837158,
      "epoch": 2.90400918403391,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041617818604586265,
      "loss": 5.53,
      "mean_token_accuracy": 0.1862326070666313,
      "num_tokens": 61323015.0,
      "step": 32885
    },
    {
      "entropy": 5.974299621582031,
      "epoch": 2.9044507241257507,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004161536638043665,
      "loss": 5.6168,
      "mean_token_accuracy": 0.18624694943428038,
      "num_tokens": 61332950.0,
      "step": 32890
    },
    {
      "entropy": 5.89360990524292,
      "epoch": 2.904892264217591,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004161291387977504,
      "loss": 5.5828,
      "mean_token_accuracy": 0.19856273382902145,
      "num_tokens": 61342212.0,
      "step": 32895
    },
    {
      "entropy": 5.90394811630249,
      "epoch": 2.9053338043094312,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004161046110264948,
      "loss": 5.532,
      "mean_token_accuracy": 0.19948505610227585,
      "num_tokens": 61350786.0,
      "step": 32900
    },
    {
      "entropy": 5.902687835693359,
      "epoch": 2.9057753444012717,
      "grad_norm": 1.125,
      "learning_rate": 0.0004160800804910801,
      "loss": 5.5815,
      "mean_token_accuracy": 0.19772440791130066,
      "num_tokens": 61360603.0,
      "step": 32905
    },
    {
      "entropy": 5.969352722167969,
      "epoch": 2.906216884493112,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041605554719198674,
      "loss": 5.5852,
      "mean_token_accuracy": 0.1923447296023369,
      "num_tokens": 61368552.0,
      "step": 32910
    },
    {
      "entropy": 5.826092147827149,
      "epoch": 2.9066584245849523,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004160310111296956,
      "loss": 5.4752,
      "mean_token_accuracy": 0.19468683302402495,
      "num_tokens": 61377833.0,
      "step": 32915
    },
    {
      "entropy": 5.966533041000366,
      "epoch": 2.9070999646767928,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004160064723046871,
      "loss": 5.6414,
      "mean_token_accuracy": 0.18559063971042633,
      "num_tokens": 61387002.0,
      "step": 32920
    },
    {
      "entropy": 5.933299255371094,
      "epoch": 2.9075415047686333,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041598193071744197,
      "loss": 5.605,
      "mean_token_accuracy": 0.18698446452617645,
      "num_tokens": 61396502.0,
      "step": 32925
    },
    {
      "entropy": 5.821740818023682,
      "epoch": 2.9079830448604733,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041595738636844093,
      "loss": 5.8278,
      "mean_token_accuracy": 0.18303808569908142,
      "num_tokens": 61405937.0,
      "step": 32930
    },
    {
      "entropy": 6.007824945449829,
      "epoch": 2.908424584952314,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000415932839258165,
      "loss": 5.5629,
      "mean_token_accuracy": 0.19231363087892533,
      "num_tokens": 61415504.0,
      "step": 32935
    },
    {
      "entropy": 5.896602153778076,
      "epoch": 2.908866125044154,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004159082893870948,
      "loss": 5.5934,
      "mean_token_accuracy": 0.18660977631807327,
      "num_tokens": 61424370.0,
      "step": 32940
    },
    {
      "entropy": 5.964948606491089,
      "epoch": 2.9093076651359944,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004158837367557114,
      "loss": 5.6014,
      "mean_token_accuracy": 0.18502584844827652,
      "num_tokens": 61433934.0,
      "step": 32945
    },
    {
      "entropy": 5.884669017791748,
      "epoch": 2.9097492052278344,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041585918136449566,
      "loss": 5.5445,
      "mean_token_accuracy": 0.1931694746017456,
      "num_tokens": 61443749.0,
      "step": 32950
    },
    {
      "entropy": 5.909134435653686,
      "epoch": 2.910190745319675,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004158346232139287,
      "loss": 5.6247,
      "mean_token_accuracy": 0.18276167809963226,
      "num_tokens": 61452939.0,
      "step": 32955
    },
    {
      "entropy": 5.969219064712524,
      "epoch": 2.9106322854115154,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004158100623044917,
      "loss": 5.5404,
      "mean_token_accuracy": 0.18202176541090012,
      "num_tokens": 61462037.0,
      "step": 32960
    },
    {
      "entropy": 5.93938946723938,
      "epoch": 2.9110738255033555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041578549863666555,
      "loss": 5.5581,
      "mean_token_accuracy": 0.19143824130296708,
      "num_tokens": 61470153.0,
      "step": 32965
    },
    {
      "entropy": 5.911550617218017,
      "epoch": 2.911515365595196,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004157609322109315,
      "loss": 5.7125,
      "mean_token_accuracy": 0.17789940237998964,
      "num_tokens": 61479834.0,
      "step": 32970
    },
    {
      "entropy": 6.000119972229004,
      "epoch": 2.9119569056870365,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004157363630277709,
      "loss": 5.641,
      "mean_token_accuracy": 0.18983963876962662,
      "num_tokens": 61488929.0,
      "step": 32975
    },
    {
      "entropy": 6.0358452796936035,
      "epoch": 2.9123984457788765,
      "grad_norm": 1.25,
      "learning_rate": 0.0004157117910876649,
      "loss": 5.5915,
      "mean_token_accuracy": 0.190521240234375,
      "num_tokens": 61498318.0,
      "step": 32980
    },
    {
      "entropy": 5.908922576904297,
      "epoch": 2.912839985870717,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041568721639109507,
      "loss": 5.5368,
      "mean_token_accuracy": 0.19629389345645903,
      "num_tokens": 61507898.0,
      "step": 32985
    },
    {
      "entropy": 5.861687755584716,
      "epoch": 2.9132815259625575,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004156626389385426,
      "loss": 5.5449,
      "mean_token_accuracy": 0.1948957249522209,
      "num_tokens": 61516275.0,
      "step": 32990
    },
    {
      "entropy": 5.930390977859497,
      "epoch": 2.9137230660543976,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000415638058730489,
      "loss": 5.526,
      "mean_token_accuracy": 0.1963529869914055,
      "num_tokens": 61525077.0,
      "step": 32995
    },
    {
      "entropy": 5.830012702941895,
      "epoch": 2.914164606146238,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000415613475767416,
      "loss": 5.4334,
      "mean_token_accuracy": 0.204247385263443,
      "num_tokens": 61533801.0,
      "step": 33000
    },
    {
      "epoch": 2.914164606146238,
      "eval_entropy": 5.624879054817986,
      "eval_loss": 5.659095764160156,
      "eval_mean_token_accuracy": 0.1940069488134754,
      "eval_num_tokens": 61533801.0,
      "eval_runtime": 26.759,
      "eval_samples_per_second": 1319.743,
      "eval_steps_per_second": 164.991,
      "step": 33000
    }
  ],
  "logging_steps": 5,
  "max_steps": 113230,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.1097478687744e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}