eus-latn-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407/checkpoint-30000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.6492405510420345,
  "eval_steps": 3000,
  "global_step": 30000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.790674829483033,
      "epoch": 0.0004415400918403391,
      "grad_norm": 13.0625,
      "learning_rate": 2e-06,
      "loss": 14.4349,
      "mean_token_accuracy": 0.0,
      "num_tokens": 9390.0,
      "step": 5
    },
    {
      "entropy": 4.818728256225586,
      "epoch": 0.0008830801836806782,
      "grad_norm": 14.1875,
      "learning_rate": 4.5e-06,
      "loss": 14.4117,
      "mean_token_accuracy": 0.0,
      "num_tokens": 18671.0,
      "step": 10
    },
    {
      "entropy": 4.857943296432495,
      "epoch": 0.0013246202755210173,
      "grad_norm": 16.25,
      "learning_rate": 7e-06,
      "loss": 14.1693,
      "mean_token_accuracy": 0.00014005602570250631,
      "num_tokens": 27614.0,
      "step": 15
    },
    {
      "entropy": 5.027469444274902,
      "epoch": 0.0017661603673613563,
      "grad_norm": 25.875,
      "learning_rate": 9.5e-06,
      "loss": 13.7713,
      "mean_token_accuracy": 8.547008619643747e-05,
      "num_tokens": 37850.0,
      "step": 20
    },
    {
      "entropy": 6.5380439281463625,
      "epoch": 0.0022077004592016957,
      "grad_norm": 26.375,
      "learning_rate": 1.2e-05,
      "loss": 12.2509,
      "mean_token_accuracy": 0.0,
      "num_tokens": 47166.0,
      "step": 25
    },
    {
      "entropy": 9.970003509521485,
      "epoch": 0.0026492405510420347,
      "grad_norm": 3.4375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.9371,
      "mean_token_accuracy": 0.00022374301915988325,
      "num_tokens": 55500.0,
      "step": 30
    },
    {
      "entropy": 10.680026626586914,
      "epoch": 0.0030907806428823736,
      "grad_norm": 3.15625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.6238,
      "mean_token_accuracy": 0.009453117521479726,
      "num_tokens": 63851.0,
      "step": 35
    },
    {
      "entropy": 10.703511428833007,
      "epoch": 0.0035323207347227126,
      "grad_norm": 3.21875,
      "learning_rate": 1.95e-05,
      "loss": 10.3602,
      "mean_token_accuracy": 0.029037438705563544,
      "num_tokens": 73697.0,
      "step": 40
    },
    {
      "entropy": 10.651962184906006,
      "epoch": 0.003973860826563052,
      "grad_norm": 2.625,
      "learning_rate": 2.2e-05,
      "loss": 10.0115,
      "mean_token_accuracy": 0.05894971713423729,
      "num_tokens": 83000.0,
      "step": 45
    },
    {
      "entropy": 10.439279747009277,
      "epoch": 0.004415400918403391,
      "grad_norm": 2.03125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.8132,
      "mean_token_accuracy": 0.05815875120460987,
      "num_tokens": 92982.0,
      "step": 50
    },
    {
      "entropy": 10.318083763122559,
      "epoch": 0.00485694101024373,
      "grad_norm": 1.8203125,
      "learning_rate": 2.7e-05,
      "loss": 9.6231,
      "mean_token_accuracy": 0.05530005097389221,
      "num_tokens": 101455.0,
      "step": 55
    },
    {
      "entropy": 10.381121063232422,
      "epoch": 0.005298481102084069,
      "grad_norm": 1.921875,
      "learning_rate": 2.95e-05,
      "loss": 9.5438,
      "mean_token_accuracy": 0.05805549845099449,
      "num_tokens": 110782.0,
      "step": 60
    },
    {
      "entropy": 10.360444736480712,
      "epoch": 0.005740021193924408,
      "grad_norm": 1.71875,
      "learning_rate": 3.2e-05,
      "loss": 9.4168,
      "mean_token_accuracy": 0.060499183088541034,
      "num_tokens": 119241.0,
      "step": 65
    },
    {
      "entropy": 10.300647163391114,
      "epoch": 0.006181561285764747,
      "grad_norm": 1.75,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.4178,
      "mean_token_accuracy": 0.055320289358496665,
      "num_tokens": 127903.0,
      "step": 70
    },
    {
      "entropy": 10.332123184204102,
      "epoch": 0.006623101377605086,
      "grad_norm": 1.6328125,
      "learning_rate": 3.7e-05,
      "loss": 9.3721,
      "mean_token_accuracy": 0.05736841931939125,
      "num_tokens": 137370.0,
      "step": 75
    },
    {
      "entropy": 10.290982055664063,
      "epoch": 0.007064641469445425,
      "grad_norm": 1.625,
      "learning_rate": 3.95e-05,
      "loss": 9.2214,
      "mean_token_accuracy": 0.06618293710052967,
      "num_tokens": 146582.0,
      "step": 80
    },
    {
      "entropy": 10.196907424926758,
      "epoch": 0.007506181561285765,
      "grad_norm": 1.5703125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.1585,
      "mean_token_accuracy": 0.05961471572518349,
      "num_tokens": 154933.0,
      "step": 85
    },
    {
      "entropy": 10.205323791503906,
      "epoch": 0.007947721653126103,
      "grad_norm": 1.5078125,
      "learning_rate": 4.45e-05,
      "loss": 9.1026,
      "mean_token_accuracy": 0.072137650847435,
      "num_tokens": 165157.0,
      "step": 90
    },
    {
      "entropy": 10.10411615371704,
      "epoch": 0.008389261744966443,
      "grad_norm": 1.328125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.9848,
      "mean_token_accuracy": 0.0728946004062891,
      "num_tokens": 174958.0,
      "step": 95
    },
    {
      "entropy": 10.01873140335083,
      "epoch": 0.008830801836806783,
      "grad_norm": 1.34375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.8889,
      "mean_token_accuracy": 0.07516518756747245,
      "num_tokens": 184256.0,
      "step": 100
    },
    {
      "entropy": 9.956882572174072,
      "epoch": 0.009272341928647121,
      "grad_norm": 1.546875,
      "learning_rate": 5.2e-05,
      "loss": 8.7839,
      "mean_token_accuracy": 0.067950439453125,
      "num_tokens": 192894.0,
      "step": 105
    },
    {
      "entropy": 9.884513092041015,
      "epoch": 0.00971388202048746,
      "grad_norm": 1.375,
      "learning_rate": 5.45e-05,
      "loss": 8.6868,
      "mean_token_accuracy": 0.07383731976151467,
      "num_tokens": 202675.0,
      "step": 110
    },
    {
      "entropy": 9.810705184936523,
      "epoch": 0.010155422112327799,
      "grad_norm": 1.1953125,
      "learning_rate": 5.7e-05,
      "loss": 8.624,
      "mean_token_accuracy": 0.07006355635821819,
      "num_tokens": 212261.0,
      "step": 115
    },
    {
      "entropy": 9.743825721740723,
      "epoch": 0.010596962204168139,
      "grad_norm": 1.171875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.599,
      "mean_token_accuracy": 0.06874018795788288,
      "num_tokens": 222329.0,
      "step": 120
    },
    {
      "entropy": 9.528209781646728,
      "epoch": 0.011038502296008477,
      "grad_norm": 1.109375,
      "learning_rate": 6.2e-05,
      "loss": 8.4477,
      "mean_token_accuracy": 0.06682575456798076,
      "num_tokens": 231247.0,
      "step": 125
    },
    {
      "entropy": 9.442446994781495,
      "epoch": 0.011480042387848817,
      "grad_norm": 1.15625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.3412,
      "mean_token_accuracy": 0.06921537183225154,
      "num_tokens": 239978.0,
      "step": 130
    },
    {
      "entropy": 9.272939491271973,
      "epoch": 0.011921582479689156,
      "grad_norm": 0.90625,
      "learning_rate": 6.7e-05,
      "loss": 8.3,
      "mean_token_accuracy": 0.06850462295114994,
      "num_tokens": 249735.0,
      "step": 135
    },
    {
      "entropy": 9.222266483306885,
      "epoch": 0.012363122571529495,
      "grad_norm": 1.15625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.2344,
      "mean_token_accuracy": 0.06959122642874718,
      "num_tokens": 259369.0,
      "step": 140
    },
    {
      "entropy": 8.956540203094482,
      "epoch": 0.012804662663369834,
      "grad_norm": 0.81640625,
      "learning_rate": 7.2e-05,
      "loss": 8.2305,
      "mean_token_accuracy": 0.06539506763219834,
      "num_tokens": 268645.0,
      "step": 145
    },
    {
      "entropy": 8.88605546951294,
      "epoch": 0.013246202755210172,
      "grad_norm": 1.1640625,
      "learning_rate": 7.45e-05,
      "loss": 8.0685,
      "mean_token_accuracy": 0.07155903875827789,
      "num_tokens": 276667.0,
      "step": 150
    },
    {
      "entropy": 8.623716259002686,
      "epoch": 0.013687742847050512,
      "grad_norm": 0.703125,
      "learning_rate": 7.7e-05,
      "loss": 8.0682,
      "mean_token_accuracy": 0.07539896108210087,
      "num_tokens": 286017.0,
      "step": 155
    },
    {
      "entropy": 8.590844440460206,
      "epoch": 0.01412928293889085,
      "grad_norm": 0.83984375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.0571,
      "mean_token_accuracy": 0.07278457470238209,
      "num_tokens": 295631.0,
      "step": 160
    },
    {
      "entropy": 8.562520027160645,
      "epoch": 0.01457082303073119,
      "grad_norm": 0.86328125,
      "learning_rate": 8.2e-05,
      "loss": 8.0486,
      "mean_token_accuracy": 0.06986252851784229,
      "num_tokens": 304704.0,
      "step": 165
    },
    {
      "entropy": 8.490843200683594,
      "epoch": 0.01501236312257153,
      "grad_norm": 0.7734375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 8.0665,
      "mean_token_accuracy": 0.07160350978374481,
      "num_tokens": 314195.0,
      "step": 170
    },
    {
      "entropy": 8.416227722167969,
      "epoch": 0.015453903214411868,
      "grad_norm": 1.0078125,
      "learning_rate": 8.7e-05,
      "loss": 8.0632,
      "mean_token_accuracy": 0.07028085552155972,
      "num_tokens": 323379.0,
      "step": 175
    },
    {
      "entropy": 8.398184299468994,
      "epoch": 0.015895443306252206,
      "grad_norm": 0.734375,
      "learning_rate": 8.95e-05,
      "loss": 7.9637,
      "mean_token_accuracy": 0.08185541778802871,
      "num_tokens": 332322.0,
      "step": 180
    },
    {
      "entropy": 8.336036014556885,
      "epoch": 0.016336983398092548,
      "grad_norm": 0.79296875,
      "learning_rate": 9.2e-05,
      "loss": 7.9427,
      "mean_token_accuracy": 0.08073886930942535,
      "num_tokens": 341735.0,
      "step": 185
    },
    {
      "entropy": 8.337114715576172,
      "epoch": 0.016778523489932886,
      "grad_norm": 0.92578125,
      "learning_rate": 9.45e-05,
      "loss": 8.0349,
      "mean_token_accuracy": 0.06938613168895244,
      "num_tokens": 351209.0,
      "step": 190
    },
    {
      "entropy": 8.39198350906372,
      "epoch": 0.017220063581773224,
      "grad_norm": 0.98046875,
      "learning_rate": 9.7e-05,
      "loss": 7.9792,
      "mean_token_accuracy": 0.07559169828891754,
      "num_tokens": 360467.0,
      "step": 195
    },
    {
      "entropy": 8.235328102111817,
      "epoch": 0.017661603673613566,
      "grad_norm": 0.875,
      "learning_rate": 9.95e-05,
      "loss": 7.9423,
      "mean_token_accuracy": 0.07669526152312756,
      "num_tokens": 370361.0,
      "step": 200
    },
    {
      "entropy": 8.374059581756592,
      "epoch": 0.018103143765453904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000102,
      "loss": 8.0075,
      "mean_token_accuracy": 0.07567069008946418,
      "num_tokens": 380366.0,
      "step": 205
    },
    {
      "entropy": 8.206629276275635,
      "epoch": 0.018544683857294242,
      "grad_norm": 1.0,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.9185,
      "mean_token_accuracy": 0.08159190192818641,
      "num_tokens": 390690.0,
      "step": 210
    },
    {
      "entropy": 8.24603796005249,
      "epoch": 0.01898622394913458,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000107,
      "loss": 7.9601,
      "mean_token_accuracy": 0.0793293446302414,
      "num_tokens": 400722.0,
      "step": 215
    },
    {
      "entropy": 8.157498931884765,
      "epoch": 0.01942776404097492,
      "grad_norm": 0.84375,
      "learning_rate": 0.0001095,
      "loss": 7.8501,
      "mean_token_accuracy": 0.08334142193198205,
      "num_tokens": 410223.0,
      "step": 220
    },
    {
      "entropy": 8.205572509765625,
      "epoch": 0.01986930413281526,
      "grad_norm": 1.0,
      "learning_rate": 0.000112,
      "loss": 7.9021,
      "mean_token_accuracy": 0.07716193869709968,
      "num_tokens": 420214.0,
      "step": 225
    },
    {
      "entropy": 8.172825717926026,
      "epoch": 0.020310844224655598,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0001145,
      "loss": 7.8564,
      "mean_token_accuracy": 0.08035471551120281,
      "num_tokens": 429407.0,
      "step": 230
    },
    {
      "entropy": 8.162760925292968,
      "epoch": 0.02075238431649594,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.8246,
      "mean_token_accuracy": 0.07542734369635581,
      "num_tokens": 438403.0,
      "step": 235
    },
    {
      "entropy": 8.177341651916503,
      "epoch": 0.021193924408336277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.8545,
      "mean_token_accuracy": 0.08535856604576111,
      "num_tokens": 447466.0,
      "step": 240
    },
    {
      "entropy": 8.069422197341918,
      "epoch": 0.021635464500176615,
      "grad_norm": 0.921875,
      "learning_rate": 0.000122,
      "loss": 7.9366,
      "mean_token_accuracy": 0.07459555268287658,
      "num_tokens": 457141.0,
      "step": 245
    },
    {
      "entropy": 8.160084056854249,
      "epoch": 0.022077004592016954,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0001245,
      "loss": 7.812,
      "mean_token_accuracy": 0.08223466873168946,
      "num_tokens": 465708.0,
      "step": 250
    },
    {
      "entropy": 8.142998504638673,
      "epoch": 0.022518544683857295,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000127,
      "loss": 7.8339,
      "mean_token_accuracy": 0.07565066292881965,
      "num_tokens": 475369.0,
      "step": 255
    },
    {
      "entropy": 8.075135421752929,
      "epoch": 0.022960084775697633,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001295,
      "loss": 7.7972,
      "mean_token_accuracy": 0.08645984381437302,
      "num_tokens": 484249.0,
      "step": 260
    },
    {
      "entropy": 8.122587871551513,
      "epoch": 0.02340162486753797,
      "grad_norm": 1.109375,
      "learning_rate": 0.000132,
      "loss": 7.8872,
      "mean_token_accuracy": 0.07687325775623322,
      "num_tokens": 493303.0,
      "step": 265
    },
    {
      "entropy": 8.101485538482667,
      "epoch": 0.023843164959378313,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.8664,
      "mean_token_accuracy": 0.0807331919670105,
      "num_tokens": 501503.0,
      "step": 270
    },
    {
      "entropy": 8.036290693283082,
      "epoch": 0.02428470505121865,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.8074,
      "mean_token_accuracy": 0.08591768592596054,
      "num_tokens": 509661.0,
      "step": 275
    },
    {
      "entropy": 8.045488977432251,
      "epoch": 0.02472624514305899,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001395,
      "loss": 7.7904,
      "mean_token_accuracy": 0.08441019728779793,
      "num_tokens": 519464.0,
      "step": 280
    },
    {
      "entropy": 8.107398653030396,
      "epoch": 0.025167785234899327,
      "grad_norm": 1.0625,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.7489,
      "mean_token_accuracy": 0.08773190379142762,
      "num_tokens": 527968.0,
      "step": 285
    },
    {
      "entropy": 8.081705808639526,
      "epoch": 0.02560932532673967,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001445,
      "loss": 7.7768,
      "mean_token_accuracy": 0.0868467777967453,
      "num_tokens": 537234.0,
      "step": 290
    },
    {
      "entropy": 7.99565052986145,
      "epoch": 0.026050865418580007,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000147,
      "loss": 7.7747,
      "mean_token_accuracy": 0.08527034223079681,
      "num_tokens": 546398.0,
      "step": 295
    },
    {
      "entropy": 8.011523675918578,
      "epoch": 0.026492405510420345,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001495,
      "loss": 7.7616,
      "mean_token_accuracy": 0.08982880860567093,
      "num_tokens": 555362.0,
      "step": 300
    },
    {
      "entropy": 8.107937812805176,
      "epoch": 0.026933945602260687,
      "grad_norm": 1.0,
      "learning_rate": 0.000152,
      "loss": 7.8221,
      "mean_token_accuracy": 0.07775180079042912,
      "num_tokens": 564575.0,
      "step": 305
    },
    {
      "entropy": 8.133016395568848,
      "epoch": 0.027375485694101025,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.8384,
      "mean_token_accuracy": 0.08304800540208816,
      "num_tokens": 573915.0,
      "step": 310
    },
    {
      "entropy": 8.016209363937378,
      "epoch": 0.027817025785941363,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000157,
      "loss": 7.7322,
      "mean_token_accuracy": 0.08581754639744758,
      "num_tokens": 583216.0,
      "step": 315
    },
    {
      "entropy": 7.982406425476074,
      "epoch": 0.0282585658777817,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001595,
      "loss": 7.7553,
      "mean_token_accuracy": 0.08679840788245201,
      "num_tokens": 591955.0,
      "step": 320
    },
    {
      "entropy": 7.9430736064910885,
      "epoch": 0.028700105969622042,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000162,
      "loss": 7.7588,
      "mean_token_accuracy": 0.08934888392686843,
      "num_tokens": 600999.0,
      "step": 325
    },
    {
      "entropy": 8.070584297180176,
      "epoch": 0.02914164606146238,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.6563,
      "mean_token_accuracy": 0.09217674285173416,
      "num_tokens": 609478.0,
      "step": 330
    },
    {
      "entropy": 7.987708568572998,
      "epoch": 0.02958318615330272,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.7225,
      "mean_token_accuracy": 0.08663035854697228,
      "num_tokens": 618348.0,
      "step": 335
    },
    {
      "entropy": 7.911137056350708,
      "epoch": 0.03002472624514306,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.7691,
      "mean_token_accuracy": 0.083287762850523,
      "num_tokens": 628548.0,
      "step": 340
    },
    {
      "entropy": 8.057271575927734,
      "epoch": 0.0304662663369834,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.7102,
      "mean_token_accuracy": 0.08196588605642319,
      "num_tokens": 637489.0,
      "step": 345
    },
    {
      "entropy": 7.939978122711182,
      "epoch": 0.030907806428823736,
      "grad_norm": 0.953125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.689,
      "mean_token_accuracy": 0.0814521424472332,
      "num_tokens": 646715.0,
      "step": 350
    },
    {
      "entropy": 7.897878551483155,
      "epoch": 0.031349346520664075,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000177,
      "loss": 7.6894,
      "mean_token_accuracy": 0.08998864293098449,
      "num_tokens": 656858.0,
      "step": 355
    },
    {
      "entropy": 8.019395637512208,
      "epoch": 0.03179088661250441,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0001795,
      "loss": 7.6755,
      "mean_token_accuracy": 0.08710955381393433,
      "num_tokens": 665968.0,
      "step": 360
    },
    {
      "entropy": 8.001319217681885,
      "epoch": 0.03223242670434476,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000182,
      "loss": 7.6655,
      "mean_token_accuracy": 0.08621685430407525,
      "num_tokens": 674295.0,
      "step": 365
    },
    {
      "entropy": 7.810992002487183,
      "epoch": 0.032673966796185096,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001845,
      "loss": 7.591,
      "mean_token_accuracy": 0.08370614722371102,
      "num_tokens": 683559.0,
      "step": 370
    },
    {
      "entropy": 7.816927337646485,
      "epoch": 0.033115506888025434,
      "grad_norm": 1.21875,
      "learning_rate": 0.000187,
      "loss": 7.6253,
      "mean_token_accuracy": 0.08996079638600349,
      "num_tokens": 692402.0,
      "step": 375
    },
    {
      "entropy": 7.967683601379394,
      "epoch": 0.03355704697986577,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001895,
      "loss": 7.7304,
      "mean_token_accuracy": 0.08065761215984821,
      "num_tokens": 702052.0,
      "step": 380
    },
    {
      "entropy": 8.058749055862426,
      "epoch": 0.03399858707170611,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000192,
      "loss": 7.6528,
      "mean_token_accuracy": 0.08705045655369759,
      "num_tokens": 711926.0,
      "step": 385
    },
    {
      "entropy": 7.8771873950958256,
      "epoch": 0.03444012716354645,
      "grad_norm": 0.96875,
      "learning_rate": 0.0001945,
      "loss": 7.612,
      "mean_token_accuracy": 0.08773906156420708,
      "num_tokens": 720948.0,
      "step": 390
    },
    {
      "entropy": 7.893786334991455,
      "epoch": 0.034881667255386786,
      "grad_norm": 1.5,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.6301,
      "mean_token_accuracy": 0.09444142654538154,
      "num_tokens": 729611.0,
      "step": 395
    },
    {
      "entropy": 7.892533588409424,
      "epoch": 0.03532320734722713,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.6187,
      "mean_token_accuracy": 0.08193654045462609,
      "num_tokens": 738433.0,
      "step": 400
    },
    {
      "entropy": 7.945340347290039,
      "epoch": 0.03576474743906747,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000202,
      "loss": 7.6397,
      "mean_token_accuracy": 0.08668759167194366,
      "num_tokens": 747310.0,
      "step": 405
    },
    {
      "entropy": 7.854477500915527,
      "epoch": 0.03620628753090781,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.5994,
      "mean_token_accuracy": 0.09020926207304,
      "num_tokens": 756362.0,
      "step": 410
    },
    {
      "entropy": 7.90778489112854,
      "epoch": 0.036647827622748146,
      "grad_norm": 0.984375,
      "learning_rate": 0.000207,
      "loss": 7.6034,
      "mean_token_accuracy": 0.08586042672395706,
      "num_tokens": 764978.0,
      "step": 415
    },
    {
      "entropy": 7.87300386428833,
      "epoch": 0.037089367714588484,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002095,
      "loss": 7.5707,
      "mean_token_accuracy": 0.09018185958266259,
      "num_tokens": 774058.0,
      "step": 420
    },
    {
      "entropy": 7.795767593383789,
      "epoch": 0.03753090780642882,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000212,
      "loss": 7.5492,
      "mean_token_accuracy": 0.0897379383444786,
      "num_tokens": 783332.0,
      "step": 425
    },
    {
      "entropy": 7.853004789352417,
      "epoch": 0.03797244789826916,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002145,
      "loss": 7.6298,
      "mean_token_accuracy": 0.08684360906481743,
      "num_tokens": 792481.0,
      "step": 430
    },
    {
      "entropy": 7.766995525360107,
      "epoch": 0.038413987990109505,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.5212,
      "mean_token_accuracy": 0.09301207512617111,
      "num_tokens": 801396.0,
      "step": 435
    },
    {
      "entropy": 7.8428326606750485,
      "epoch": 0.03885552808194984,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002195,
      "loss": 7.5865,
      "mean_token_accuracy": 0.08940735682845116,
      "num_tokens": 809903.0,
      "step": 440
    },
    {
      "entropy": 7.828377294540405,
      "epoch": 0.03929706817379018,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000222,
      "loss": 7.5389,
      "mean_token_accuracy": 0.0962544821202755,
      "num_tokens": 819144.0,
      "step": 445
    },
    {
      "entropy": 7.7183678150177,
      "epoch": 0.03973860826563052,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002245,
      "loss": 7.5608,
      "mean_token_accuracy": 0.08849129751324654,
      "num_tokens": 828881.0,
      "step": 450
    },
    {
      "entropy": 7.764478397369385,
      "epoch": 0.04018014835747086,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.4877,
      "mean_token_accuracy": 0.08765893578529357,
      "num_tokens": 837588.0,
      "step": 455
    },
    {
      "entropy": 7.767373847961426,
      "epoch": 0.040621688449311195,
      "grad_norm": 1.0,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.4849,
      "mean_token_accuracy": 0.09265839084982871,
      "num_tokens": 847002.0,
      "step": 460
    },
    {
      "entropy": 7.726333475112915,
      "epoch": 0.041063228541151534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.4945,
      "mean_token_accuracy": 0.09333177357912063,
      "num_tokens": 855791.0,
      "step": 465
    },
    {
      "entropy": 7.7462080955505375,
      "epoch": 0.04150476863299188,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.5242,
      "mean_token_accuracy": 0.0911882683634758,
      "num_tokens": 865392.0,
      "step": 470
    },
    {
      "entropy": 7.736569499969482,
      "epoch": 0.04194630872483222,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000237,
      "loss": 7.5873,
      "mean_token_accuracy": 0.08873779252171517,
      "num_tokens": 874807.0,
      "step": 475
    },
    {
      "entropy": 7.758917284011841,
      "epoch": 0.042387848816672555,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002395,
      "loss": 7.5409,
      "mean_token_accuracy": 0.09495326653122901,
      "num_tokens": 883928.0,
      "step": 480
    },
    {
      "entropy": 7.777913904190063,
      "epoch": 0.04282938890851289,
      "grad_norm": 1.140625,
      "learning_rate": 0.000242,
      "loss": 7.4436,
      "mean_token_accuracy": 0.09124857932329178,
      "num_tokens": 893047.0,
      "step": 485
    },
    {
      "entropy": 7.662859010696411,
      "epoch": 0.04327092900035323,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002445,
      "loss": 7.4593,
      "mean_token_accuracy": 0.09315531030297279,
      "num_tokens": 901645.0,
      "step": 490
    },
    {
      "entropy": 7.743328475952149,
      "epoch": 0.04371246909219357,
      "grad_norm": 1.109375,
      "learning_rate": 0.000247,
      "loss": 7.4727,
      "mean_token_accuracy": 0.09244368895888329,
      "num_tokens": 911169.0,
      "step": 495
    },
    {
      "entropy": 7.7239625453948975,
      "epoch": 0.04415400918403391,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002495,
      "loss": 7.4748,
      "mean_token_accuracy": 0.08498905003070831,
      "num_tokens": 921382.0,
      "step": 500
    },
    {
      "entropy": 7.544922304153443,
      "epoch": 0.04459554927587425,
      "grad_norm": 1.234375,
      "learning_rate": 0.000252,
      "loss": 7.4121,
      "mean_token_accuracy": 0.09429771155118942,
      "num_tokens": 930409.0,
      "step": 505
    },
    {
      "entropy": 7.67856912612915,
      "epoch": 0.04503708936771459,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002545,
      "loss": 7.3879,
      "mean_token_accuracy": 0.09879431128501892,
      "num_tokens": 939049.0,
      "step": 510
    },
    {
      "entropy": 7.718625736236572,
      "epoch": 0.04547862945955493,
      "grad_norm": 1.25,
      "learning_rate": 0.000257,
      "loss": 7.395,
      "mean_token_accuracy": 0.0960740551352501,
      "num_tokens": 947575.0,
      "step": 515
    },
    {
      "entropy": 7.709804058074951,
      "epoch": 0.045920169551395267,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002595,
      "loss": 7.5643,
      "mean_token_accuracy": 0.09047991409897804,
      "num_tokens": 957848.0,
      "step": 520
    },
    {
      "entropy": 7.655015087127685,
      "epoch": 0.046361709643235605,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000262,
      "loss": 7.3857,
      "mean_token_accuracy": 0.09998803585767746,
      "num_tokens": 966521.0,
      "step": 525
    },
    {
      "entropy": 7.688518905639649,
      "epoch": 0.04680324973507594,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.4461,
      "mean_token_accuracy": 0.09324755370616913,
      "num_tokens": 975827.0,
      "step": 530
    },
    {
      "entropy": 7.606715154647827,
      "epoch": 0.04724478982691628,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.4075,
      "mean_token_accuracy": 0.09566703587770461,
      "num_tokens": 985292.0,
      "step": 535
    },
    {
      "entropy": 7.616068124771118,
      "epoch": 0.047686329918756626,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.3841,
      "mean_token_accuracy": 0.09411159604787826,
      "num_tokens": 994791.0,
      "step": 540
    },
    {
      "entropy": 7.51567816734314,
      "epoch": 0.048127870010596964,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.3594,
      "mean_token_accuracy": 0.1026044063270092,
      "num_tokens": 1003700.0,
      "step": 545
    },
    {
      "entropy": 7.510391616821289,
      "epoch": 0.0485694101024373,
      "grad_norm": 1.0,
      "learning_rate": 0.0002745,
      "loss": 7.381,
      "mean_token_accuracy": 0.09829011410474778,
      "num_tokens": 1012682.0,
      "step": 550
    },
    {
      "entropy": 7.683912038803101,
      "epoch": 0.04901095019427764,
      "grad_norm": 1.28125,
      "learning_rate": 0.000277,
      "loss": 7.4075,
      "mean_token_accuracy": 0.09179475829005242,
      "num_tokens": 1021018.0,
      "step": 555
    },
    {
      "entropy": 7.570155191421509,
      "epoch": 0.04945249028611798,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002795,
      "loss": 7.2759,
      "mean_token_accuracy": 0.09699172824621201,
      "num_tokens": 1029744.0,
      "step": 560
    },
    {
      "entropy": 7.495694351196289,
      "epoch": 0.049894030377958316,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.3605,
      "mean_token_accuracy": 0.09879247918725013,
      "num_tokens": 1038805.0,
      "step": 565
    },
    {
      "entropy": 7.5144976615905765,
      "epoch": 0.050335570469798654,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002845,
      "loss": 7.4131,
      "mean_token_accuracy": 0.0988279327750206,
      "num_tokens": 1047656.0,
      "step": 570
    },
    {
      "entropy": 7.647522783279419,
      "epoch": 0.050777110561639,
      "grad_norm": 1.140625,
      "learning_rate": 0.000287,
      "loss": 7.4048,
      "mean_token_accuracy": 0.09629088416695594,
      "num_tokens": 1056598.0,
      "step": 575
    },
    {
      "entropy": 7.6095935821533205,
      "epoch": 0.05121865065347934,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002895,
      "loss": 7.3994,
      "mean_token_accuracy": 0.09847217947244644,
      "num_tokens": 1065226.0,
      "step": 580
    },
    {
      "entropy": 7.529495334625244,
      "epoch": 0.051660190745319676,
      "grad_norm": 1.4375,
      "learning_rate": 0.000292,
      "loss": 7.3208,
      "mean_token_accuracy": 0.10098938867449761,
      "num_tokens": 1074661.0,
      "step": 585
    },
    {
      "entropy": 7.503559398651123,
      "epoch": 0.052101730837160014,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002945,
      "loss": 7.4251,
      "mean_token_accuracy": 0.09441772177815437,
      "num_tokens": 1083921.0,
      "step": 590
    },
    {
      "entropy": 7.540312194824219,
      "epoch": 0.05254327092900035,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000297,
      "loss": 7.3162,
      "mean_token_accuracy": 0.104298634827137,
      "num_tokens": 1093399.0,
      "step": 595
    },
    {
      "entropy": 7.5133528232574465,
      "epoch": 0.05298481102084069,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002995,
      "loss": 7.391,
      "mean_token_accuracy": 0.09825902208685874,
      "num_tokens": 1104065.0,
      "step": 600
    },
    {
      "entropy": 7.433008003234863,
      "epoch": 0.05342635111268103,
      "grad_norm": 1.125,
      "learning_rate": 0.000302,
      "loss": 7.2778,
      "mean_token_accuracy": 0.1005440428853035,
      "num_tokens": 1112995.0,
      "step": 605
    },
    {
      "entropy": 7.47243971824646,
      "epoch": 0.05386789120452137,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003045,
      "loss": 7.3103,
      "mean_token_accuracy": 0.10175202563405036,
      "num_tokens": 1121637.0,
      "step": 610
    },
    {
      "entropy": 7.455365610122681,
      "epoch": 0.05430943129636171,
      "grad_norm": 1.140625,
      "learning_rate": 0.000307,
      "loss": 7.2549,
      "mean_token_accuracy": 0.09826337993144989,
      "num_tokens": 1131166.0,
      "step": 615
    },
    {
      "entropy": 7.4712036609649655,
      "epoch": 0.05475097138820205,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003095,
      "loss": 7.2562,
      "mean_token_accuracy": 0.10475531965494156,
      "num_tokens": 1140888.0,
      "step": 620
    },
    {
      "entropy": 7.551609897613526,
      "epoch": 0.05519251148004239,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000312,
      "loss": 7.4148,
      "mean_token_accuracy": 0.0961816966533661,
      "num_tokens": 1150278.0,
      "step": 625
    },
    {
      "entropy": 7.433546924591065,
      "epoch": 0.055634051571882726,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003145,
      "loss": 7.3742,
      "mean_token_accuracy": 0.0970606379210949,
      "num_tokens": 1159348.0,
      "step": 630
    },
    {
      "entropy": 7.624134588241577,
      "epoch": 0.056075591663723064,
      "grad_norm": 1.046875,
      "learning_rate": 0.000317,
      "loss": 7.3756,
      "mean_token_accuracy": 0.0949991799890995,
      "num_tokens": 1168883.0,
      "step": 635
    },
    {
      "entropy": 7.48681526184082,
      "epoch": 0.0565171317555634,
      "grad_norm": 1.0,
      "learning_rate": 0.0003195,
      "loss": 7.269,
      "mean_token_accuracy": 0.1064944364130497,
      "num_tokens": 1178572.0,
      "step": 640
    },
    {
      "entropy": 7.44178466796875,
      "epoch": 0.05695867184740375,
      "grad_norm": 1.046875,
      "learning_rate": 0.000322,
      "loss": 7.3576,
      "mean_token_accuracy": 0.0987204596400261,
      "num_tokens": 1188909.0,
      "step": 645
    },
    {
      "entropy": 7.466546869277954,
      "epoch": 0.057400211939244085,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.2649,
      "mean_token_accuracy": 0.09890259429812431,
      "num_tokens": 1197705.0,
      "step": 650
    },
    {
      "entropy": 7.450878572463989,
      "epoch": 0.05784175203108442,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.1061,
      "mean_token_accuracy": 0.10522415414452553,
      "num_tokens": 1206351.0,
      "step": 655
    },
    {
      "entropy": 7.340301847457885,
      "epoch": 0.05828329212292476,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.2237,
      "mean_token_accuracy": 0.09693196043372154,
      "num_tokens": 1214984.0,
      "step": 660
    },
    {
      "entropy": 7.4402018070220945,
      "epoch": 0.0587248322147651,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.2999,
      "mean_token_accuracy": 0.09738482013344765,
      "num_tokens": 1224485.0,
      "step": 665
    },
    {
      "entropy": 7.435847473144531,
      "epoch": 0.05916637230660544,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.3266,
      "mean_token_accuracy": 0.09173622950911522,
      "num_tokens": 1233560.0,
      "step": 670
    },
    {
      "entropy": 7.428315305709839,
      "epoch": 0.05960791239844578,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000337,
      "loss": 7.2436,
      "mean_token_accuracy": 0.09967414885759354,
      "num_tokens": 1242628.0,
      "step": 675
    },
    {
      "entropy": 7.388672494888306,
      "epoch": 0.06004945249028612,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003395,
      "loss": 7.1697,
      "mean_token_accuracy": 0.10538085550069809,
      "num_tokens": 1251004.0,
      "step": 680
    },
    {
      "entropy": 7.459445238113403,
      "epoch": 0.06049099258212646,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000342,
      "loss": 7.3463,
      "mean_token_accuracy": 0.09609238728880883,
      "num_tokens": 1260344.0,
      "step": 685
    },
    {
      "entropy": 7.343485164642334,
      "epoch": 0.0609325326739668,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.2517,
      "mean_token_accuracy": 0.09760257676243782,
      "num_tokens": 1269988.0,
      "step": 690
    },
    {
      "entropy": 7.340139007568359,
      "epoch": 0.061374072765807135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000347,
      "loss": 7.2126,
      "mean_token_accuracy": 0.10715288370847702,
      "num_tokens": 1280912.0,
      "step": 695
    },
    {
      "entropy": 7.350299119949341,
      "epoch": 0.06181561285764747,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003495,
      "loss": 7.2246,
      "mean_token_accuracy": 0.10604915320873261,
      "num_tokens": 1289684.0,
      "step": 700
    },
    {
      "entropy": 7.427703905105591,
      "epoch": 0.06225715294948781,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000352,
      "loss": 7.2534,
      "mean_token_accuracy": 0.09802542477846146,
      "num_tokens": 1298853.0,
      "step": 705
    },
    {
      "entropy": 7.3198949813842775,
      "epoch": 0.06269869304132815,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003545,
      "loss": 7.214,
      "mean_token_accuracy": 0.10874532908201218,
      "num_tokens": 1309112.0,
      "step": 710
    },
    {
      "entropy": 7.372763156890869,
      "epoch": 0.0631402331331685,
      "grad_norm": 1.296875,
      "learning_rate": 0.000357,
      "loss": 7.1981,
      "mean_token_accuracy": 0.10583075731992722,
      "num_tokens": 1319064.0,
      "step": 715
    },
    {
      "entropy": 7.269387340545654,
      "epoch": 0.06358177322500883,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003595,
      "loss": 7.1748,
      "mean_token_accuracy": 0.1100200168788433,
      "num_tokens": 1327889.0,
      "step": 720
    },
    {
      "entropy": 7.324726533889771,
      "epoch": 0.06402331331684917,
      "grad_norm": 1.28125,
      "learning_rate": 0.000362,
      "loss": 7.1938,
      "mean_token_accuracy": 0.10258080512285232,
      "num_tokens": 1337241.0,
      "step": 725
    },
    {
      "entropy": 7.319574499130249,
      "epoch": 0.06446485340868952,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003645,
      "loss": 7.2533,
      "mean_token_accuracy": 0.10085726305842399,
      "num_tokens": 1346527.0,
      "step": 730
    },
    {
      "entropy": 7.314885807037354,
      "epoch": 0.06490639350052985,
      "grad_norm": 1.125,
      "learning_rate": 0.000367,
      "loss": 7.2315,
      "mean_token_accuracy": 0.10445504561066628,
      "num_tokens": 1355677.0,
      "step": 735
    },
    {
      "entropy": 7.396700429916382,
      "epoch": 0.06534793359237019,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003695,
      "loss": 7.2163,
      "mean_token_accuracy": 0.10588330775499344,
      "num_tokens": 1364874.0,
      "step": 740
    },
    {
      "entropy": 7.285468482971192,
      "epoch": 0.06578947368421052,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000372,
      "loss": 7.1378,
      "mean_token_accuracy": 0.1090671844780445,
      "num_tokens": 1373717.0,
      "step": 745
    },
    {
      "entropy": 7.375531625747681,
      "epoch": 0.06623101377605087,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003745,
      "loss": 7.0955,
      "mean_token_accuracy": 0.10741576477885247,
      "num_tokens": 1382767.0,
      "step": 750
    },
    {
      "entropy": 7.1357104778289795,
      "epoch": 0.0666725538678912,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000377,
      "loss": 7.1389,
      "mean_token_accuracy": 0.10613262876868249,
      "num_tokens": 1391190.0,
      "step": 755
    },
    {
      "entropy": 7.234480524063111,
      "epoch": 0.06711409395973154,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003795,
      "loss": 7.1509,
      "mean_token_accuracy": 0.10508784130215645,
      "num_tokens": 1400722.0,
      "step": 760
    },
    {
      "entropy": 7.402392435073852,
      "epoch": 0.06755563405157189,
      "grad_norm": 1.03125,
      "learning_rate": 0.000382,
      "loss": 7.1779,
      "mean_token_accuracy": 0.10437385067343712,
      "num_tokens": 1409328.0,
      "step": 765
    },
    {
      "entropy": 7.06873927116394,
      "epoch": 0.06799717414341222,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003845,
      "loss": 7.0531,
      "mean_token_accuracy": 0.11192933171987533,
      "num_tokens": 1418504.0,
      "step": 770
    },
    {
      "entropy": 7.440014839172363,
      "epoch": 0.06843871423525257,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.1989,
      "mean_token_accuracy": 0.10317453742027283,
      "num_tokens": 1427690.0,
      "step": 775
    },
    {
      "entropy": 7.181108903884888,
      "epoch": 0.0688802543270929,
      "grad_norm": 1.125,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.1806,
      "mean_token_accuracy": 0.10798285007476807,
      "num_tokens": 1436798.0,
      "step": 780
    },
    {
      "entropy": 7.2046185493469235,
      "epoch": 0.06932179441893324,
      "grad_norm": 0.96875,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.1646,
      "mean_token_accuracy": 0.10358999595046044,
      "num_tokens": 1446357.0,
      "step": 785
    },
    {
      "entropy": 7.2555629253387455,
      "epoch": 0.06976333451077357,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.0882,
      "mean_token_accuracy": 0.11000654250383377,
      "num_tokens": 1455998.0,
      "step": 790
    },
    {
      "entropy": 7.207996559143067,
      "epoch": 0.07020487460261392,
      "grad_norm": 0.984375,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.145,
      "mean_token_accuracy": 0.09857687279582024,
      "num_tokens": 1465237.0,
      "step": 795
    },
    {
      "entropy": 7.24621729850769,
      "epoch": 0.07064641469445426,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003995,
      "loss": 7.0958,
      "mean_token_accuracy": 0.11087250858545303,
      "num_tokens": 1474363.0,
      "step": 800
    },
    {
      "entropy": 7.272359037399292,
      "epoch": 0.0710879547862946,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000402,
      "loss": 7.1713,
      "mean_token_accuracy": 0.10843008160591125,
      "num_tokens": 1483379.0,
      "step": 805
    },
    {
      "entropy": 7.28739447593689,
      "epoch": 0.07152949487813494,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004045,
      "loss": 7.1265,
      "mean_token_accuracy": 0.10922098532319069,
      "num_tokens": 1492507.0,
      "step": 810
    },
    {
      "entropy": 7.144436979293824,
      "epoch": 0.07197103496997527,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.0154,
      "mean_token_accuracy": 0.11775125116109848,
      "num_tokens": 1500888.0,
      "step": 815
    },
    {
      "entropy": 7.11500997543335,
      "epoch": 0.07241257506181561,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004095,
      "loss": 7.0709,
      "mean_token_accuracy": 0.10802061259746551,
      "num_tokens": 1510310.0,
      "step": 820
    },
    {
      "entropy": 7.1448290824890135,
      "epoch": 0.07285411515365595,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000412,
      "loss": 7.0494,
      "mean_token_accuracy": 0.11422519460320472,
      "num_tokens": 1519427.0,
      "step": 825
    },
    {
      "entropy": 7.2035074710845945,
      "epoch": 0.07329565524549629,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004145,
      "loss": 7.0679,
      "mean_token_accuracy": 0.1063395880162716,
      "num_tokens": 1529456.0,
      "step": 830
    },
    {
      "entropy": 7.131991720199585,
      "epoch": 0.07373719533733664,
      "grad_norm": 1.078125,
      "learning_rate": 0.000417,
      "loss": 7.0241,
      "mean_token_accuracy": 0.11403456106781959,
      "num_tokens": 1537695.0,
      "step": 835
    },
    {
      "entropy": 7.203299617767334,
      "epoch": 0.07417873542917697,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004195,
      "loss": 7.1456,
      "mean_token_accuracy": 0.10954299196600914,
      "num_tokens": 1547511.0,
      "step": 840
    },
    {
      "entropy": 7.255322885513306,
      "epoch": 0.07462027552101731,
      "grad_norm": 1.078125,
      "learning_rate": 0.000422,
      "loss": 7.1315,
      "mean_token_accuracy": 0.1110302060842514,
      "num_tokens": 1557035.0,
      "step": 845
    },
    {
      "entropy": 7.1888104438781735,
      "epoch": 0.07506181561285764,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004245,
      "loss": 7.0906,
      "mean_token_accuracy": 0.11411306262016296,
      "num_tokens": 1566773.0,
      "step": 850
    },
    {
      "entropy": 7.094766998291016,
      "epoch": 0.07550335570469799,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000427,
      "loss": 7.0787,
      "mean_token_accuracy": 0.10884842053055763,
      "num_tokens": 1576873.0,
      "step": 855
    },
    {
      "entropy": 7.178222894668579,
      "epoch": 0.07594489579653832,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004295,
      "loss": 7.111,
      "mean_token_accuracy": 0.10762306824326515,
      "num_tokens": 1586170.0,
      "step": 860
    },
    {
      "entropy": 7.286298131942749,
      "epoch": 0.07638643588837866,
      "grad_norm": 1.171875,
      "learning_rate": 0.000432,
      "loss": 7.154,
      "mean_token_accuracy": 0.10613771453499794,
      "num_tokens": 1596054.0,
      "step": 865
    },
    {
      "entropy": 7.1001307487487795,
      "epoch": 0.07682797598021901,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004345,
      "loss": 7.0262,
      "mean_token_accuracy": 0.11607334911823272,
      "num_tokens": 1604544.0,
      "step": 870
    },
    {
      "entropy": 7.172781848907471,
      "epoch": 0.07726951607205934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000437,
      "loss": 7.0446,
      "mean_token_accuracy": 0.11472792029380799,
      "num_tokens": 1614580.0,
      "step": 875
    },
    {
      "entropy": 7.132223224639892,
      "epoch": 0.07771105616389969,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004395,
      "loss": 7.1232,
      "mean_token_accuracy": 0.1109985999763012,
      "num_tokens": 1624701.0,
      "step": 880
    },
    {
      "entropy": 7.128903436660766,
      "epoch": 0.07815259625574002,
      "grad_norm": 1.015625,
      "learning_rate": 0.000442,
      "loss": 7.0573,
      "mean_token_accuracy": 0.10825628340244293,
      "num_tokens": 1634085.0,
      "step": 885
    },
    {
      "entropy": 7.123282432556152,
      "epoch": 0.07859413634758036,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004445,
      "loss": 7.0322,
      "mean_token_accuracy": 0.11617021560668946,
      "num_tokens": 1643190.0,
      "step": 890
    },
    {
      "entropy": 7.060208940505982,
      "epoch": 0.0790356764394207,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000447,
      "loss": 7.07,
      "mean_token_accuracy": 0.11254222765564918,
      "num_tokens": 1652705.0,
      "step": 895
    },
    {
      "entropy": 7.132848882675171,
      "epoch": 0.07947721653126104,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.0536,
      "mean_token_accuracy": 0.10692465007305145,
      "num_tokens": 1662210.0,
      "step": 900
    },
    {
      "entropy": 7.192712593078613,
      "epoch": 0.07991875662310138,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.1014,
      "mean_token_accuracy": 0.10652303621172905,
      "num_tokens": 1671893.0,
      "step": 905
    },
    {
      "entropy": 7.059550428390503,
      "epoch": 0.08036029671494171,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.0402,
      "mean_token_accuracy": 0.11181816533207893,
      "num_tokens": 1681217.0,
      "step": 910
    },
    {
      "entropy": 7.149940156936646,
      "epoch": 0.08080183680678206,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.053,
      "mean_token_accuracy": 0.11131602600216865,
      "num_tokens": 1690447.0,
      "step": 915
    },
    {
      "entropy": 7.081046295166016,
      "epoch": 0.08124337689862239,
      "grad_norm": 1.0,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.1332,
      "mean_token_accuracy": 0.10568991601467133,
      "num_tokens": 1700355.0,
      "step": 920
    },
    {
      "entropy": 7.16390905380249,
      "epoch": 0.08168491699046274,
      "grad_norm": 1.015625,
      "learning_rate": 0.000462,
      "loss": 7.028,
      "mean_token_accuracy": 0.10254786685109138,
      "num_tokens": 1709449.0,
      "step": 925
    },
    {
      "entropy": 7.091014242172241,
      "epoch": 0.08212645708230307,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004645,
      "loss": 7.058,
      "mean_token_accuracy": 0.10658924430608749,
      "num_tokens": 1718838.0,
      "step": 930
    },
    {
      "entropy": 7.023260927200317,
      "epoch": 0.08256799717414341,
      "grad_norm": 1.046875,
      "learning_rate": 0.000467,
      "loss": 7.0217,
      "mean_token_accuracy": 0.11323517858982086,
      "num_tokens": 1728594.0,
      "step": 935
    },
    {
      "entropy": 7.1843287467956545,
      "epoch": 0.08300953726598376,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004695,
      "loss": 7.0731,
      "mean_token_accuracy": 0.11138227805495263,
      "num_tokens": 1738814.0,
      "step": 940
    },
    {
      "entropy": 7.071042823791504,
      "epoch": 0.08345107735782409,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000472,
      "loss": 7.0089,
      "mean_token_accuracy": 0.11532488241791725,
      "num_tokens": 1747644.0,
      "step": 945
    },
    {
      "entropy": 7.104792213439941,
      "epoch": 0.08389261744966443,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004745,
      "loss": 7.0338,
      "mean_token_accuracy": 0.11352440416812896,
      "num_tokens": 1757489.0,
      "step": 950
    },
    {
      "entropy": 6.995518827438355,
      "epoch": 0.08433415754150476,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000477,
      "loss": 6.9488,
      "mean_token_accuracy": 0.11878458335995674,
      "num_tokens": 1767546.0,
      "step": 955
    },
    {
      "entropy": 7.094525289535523,
      "epoch": 0.08477569763334511,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004795,
      "loss": 7.0088,
      "mean_token_accuracy": 0.10509251430630684,
      "num_tokens": 1776035.0,
      "step": 960
    },
    {
      "entropy": 7.100050449371338,
      "epoch": 0.08521723772518544,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000482,
      "loss": 7.0869,
      "mean_token_accuracy": 0.10708501487970352,
      "num_tokens": 1786161.0,
      "step": 965
    },
    {
      "entropy": 7.161181020736694,
      "epoch": 0.08565877781702579,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004845,
      "loss": 7.1355,
      "mean_token_accuracy": 0.10680384710431098,
      "num_tokens": 1796093.0,
      "step": 970
    },
    {
      "entropy": 7.064108896255493,
      "epoch": 0.08610031790886613,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000487,
      "loss": 6.9829,
      "mean_token_accuracy": 0.1097193941473961,
      "num_tokens": 1805574.0,
      "step": 975
    },
    {
      "entropy": 7.07778491973877,
      "epoch": 0.08654185800070646,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004895,
      "loss": 7.0171,
      "mean_token_accuracy": 0.11008013710379601,
      "num_tokens": 1815175.0,
      "step": 980
    },
    {
      "entropy": 7.017868852615356,
      "epoch": 0.08698339809254681,
      "grad_norm": 1.03125,
      "learning_rate": 0.000492,
      "loss": 6.9322,
      "mean_token_accuracy": 0.11862852200865745,
      "num_tokens": 1824683.0,
      "step": 985
    },
    {
      "entropy": 7.004701948165893,
      "epoch": 0.08742493818438714,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004945,
      "loss": 6.9091,
      "mean_token_accuracy": 0.1145630083978176,
      "num_tokens": 1833174.0,
      "step": 990
    },
    {
      "entropy": 7.008507776260376,
      "epoch": 0.08786647827622748,
      "grad_norm": 1.203125,
      "learning_rate": 0.000497,
      "loss": 7.0124,
      "mean_token_accuracy": 0.1165225401520729,
      "num_tokens": 1842409.0,
      "step": 995
    },
    {
      "entropy": 6.900066137313843,
      "epoch": 0.08830801836806781,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995,
      "loss": 6.9172,
      "mean_token_accuracy": 0.1189465768635273,
      "num_tokens": 1851441.0,
      "step": 1000
    },
    {
      "entropy": 7.234589004516602,
      "epoch": 0.08874955845990816,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499999998589561,
      "loss": 7.0545,
      "mean_token_accuracy": 0.1098681665956974,
      "num_tokens": 1861188.0,
      "step": 1005
    },
    {
      "entropy": 6.916832828521729,
      "epoch": 0.0891910985517485,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999999928596523,
      "loss": 6.9934,
      "mean_token_accuracy": 0.1134356640279293,
      "num_tokens": 1870284.0,
      "step": 1010
    },
    {
      "entropy": 6.8979510307312015,
      "epoch": 0.08963263864358884,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999999827221219,
      "loss": 6.9508,
      "mean_token_accuracy": 0.11564922854304313,
      "num_tokens": 1879744.0,
      "step": 1015
    },
    {
      "entropy": 7.141992807388306,
      "epoch": 0.09007417873542918,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999999681769696,
      "loss": 6.9612,
      "mean_token_accuracy": 0.1163177601993084,
      "num_tokens": 1889241.0,
      "step": 1020
    },
    {
      "entropy": 6.971645736694336,
      "epoch": 0.09051571882726951,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499999949224196,
      "loss": 6.969,
      "mean_token_accuracy": 0.11504201143980027,
      "num_tokens": 1898247.0,
      "step": 1025
    },
    {
      "entropy": 7.006195020675659,
      "epoch": 0.09095725891910986,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999999258638013,
      "loss": 6.9244,
      "mean_token_accuracy": 0.11498644798994065,
      "num_tokens": 1907559.0,
      "step": 1030
    },
    {
      "entropy": 6.903467321395874,
      "epoch": 0.09139879901095019,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999998980957861,
      "loss": 6.924,
      "mean_token_accuracy": 0.11912157312035561,
      "num_tokens": 1917072.0,
      "step": 1035
    },
    {
      "entropy": 6.983310127258301,
      "epoch": 0.09184033910279053,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999998659201508,
      "loss": 6.8753,
      "mean_token_accuracy": 0.11874028518795968,
      "num_tokens": 1926597.0,
      "step": 1040
    },
    {
      "entropy": 7.006656980514526,
      "epoch": 0.09228187919463088,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999998293368961,
      "loss": 6.918,
      "mean_token_accuracy": 0.11443031057715417,
      "num_tokens": 1935978.0,
      "step": 1045
    },
    {
      "entropy": 6.924140882492066,
      "epoch": 0.09272341928647121,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999997883460227,
      "loss": 6.864,
      "mean_token_accuracy": 0.1171707384288311,
      "num_tokens": 1944424.0,
      "step": 1050
    },
    {
      "entropy": 6.984135913848877,
      "epoch": 0.09316495937831155,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999997429475314,
      "loss": 6.8384,
      "mean_token_accuracy": 0.12138021439313888,
      "num_tokens": 1953844.0,
      "step": 1055
    },
    {
      "entropy": 6.947112941741944,
      "epoch": 0.09360649947015189,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999996931414232,
      "loss": 6.9207,
      "mean_token_accuracy": 0.12120825350284577,
      "num_tokens": 1963974.0,
      "step": 1060
    },
    {
      "entropy": 6.977735805511474,
      "epoch": 0.09404803956199223,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999996389276988,
      "loss": 6.8969,
      "mean_token_accuracy": 0.12291403263807296,
      "num_tokens": 1973466.0,
      "step": 1065
    },
    {
      "entropy": 6.803595685958863,
      "epoch": 0.09448957965383256,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999995803063596,
      "loss": 6.918,
      "mean_token_accuracy": 0.12123456448316575,
      "num_tokens": 1983478.0,
      "step": 1070
    },
    {
      "entropy": 7.003172111511231,
      "epoch": 0.0949311197456729,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999995172774065,
      "loss": 6.9879,
      "mean_token_accuracy": 0.11715293675661087,
      "num_tokens": 1992775.0,
      "step": 1075
    },
    {
      "entropy": 7.030760860443115,
      "epoch": 0.09537265983751325,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999994498408408,
      "loss": 6.9419,
      "mean_token_accuracy": 0.11398354098200798,
      "num_tokens": 2002526.0,
      "step": 1080
    },
    {
      "entropy": 6.936420059204101,
      "epoch": 0.09581419992935358,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999993779966639,
      "loss": 6.9592,
      "mean_token_accuracy": 0.11412434950470925,
      "num_tokens": 2012476.0,
      "step": 1085
    },
    {
      "entropy": 6.838730955123902,
      "epoch": 0.09625574002119393,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999993017448771,
      "loss": 6.7924,
      "mean_token_accuracy": 0.13063410446047782,
      "num_tokens": 2021252.0,
      "step": 1090
    },
    {
      "entropy": 7.061969757080078,
      "epoch": 0.09669728011303426,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999992210854821,
      "loss": 6.9411,
      "mean_token_accuracy": 0.11837697625160218,
      "num_tokens": 2031438.0,
      "step": 1095
    },
    {
      "entropy": 6.9324125289917,
      "epoch": 0.0971388202048746,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999991360184801,
      "loss": 6.9789,
      "mean_token_accuracy": 0.11443927884101868,
      "num_tokens": 2041319.0,
      "step": 1100
    },
    {
      "entropy": 6.95938081741333,
      "epoch": 0.09758036029671494,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999990465438731,
      "loss": 6.9746,
      "mean_token_accuracy": 0.11487890034914017,
      "num_tokens": 2052060.0,
      "step": 1105
    },
    {
      "entropy": 6.891486358642578,
      "epoch": 0.09802190038855528,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999989526616628,
      "loss": 6.8643,
      "mean_token_accuracy": 0.12554761841893197,
      "num_tokens": 2061331.0,
      "step": 1110
    },
    {
      "entropy": 6.952145195007324,
      "epoch": 0.09846344048039563,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999988543718509,
      "loss": 6.8733,
      "mean_token_accuracy": 0.11660940647125244,
      "num_tokens": 2070006.0,
      "step": 1115
    },
    {
      "entropy": 7.005694484710693,
      "epoch": 0.09890498057223596,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999987516744394,
      "loss": 6.8403,
      "mean_token_accuracy": 0.12942354679107665,
      "num_tokens": 2079089.0,
      "step": 1120
    },
    {
      "entropy": 6.814391231536865,
      "epoch": 0.0993465206640763,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999986445694303,
      "loss": 6.8164,
      "mean_token_accuracy": 0.12300374433398246,
      "num_tokens": 2087237.0,
      "step": 1125
    },
    {
      "entropy": 6.877581930160522,
      "epoch": 0.09978806075591663,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999985330568258,
      "loss": 6.7838,
      "mean_token_accuracy": 0.12274843603372573,
      "num_tokens": 2096598.0,
      "step": 1130
    },
    {
      "entropy": 6.867826128005982,
      "epoch": 0.10022960084775698,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999984171366278,
      "loss": 6.8802,
      "mean_token_accuracy": 0.11098882853984833,
      "num_tokens": 2106106.0,
      "step": 1135
    },
    {
      "entropy": 6.928639554977417,
      "epoch": 0.10067114093959731,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499998296808839,
      "loss": 6.856,
      "mean_token_accuracy": 0.11591004803776742,
      "num_tokens": 2115478.0,
      "step": 1140
    },
    {
      "entropy": 6.942514848709107,
      "epoch": 0.10111268103143765,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999981720734615,
      "loss": 6.9007,
      "mean_token_accuracy": 0.11097749546170235,
      "num_tokens": 2124439.0,
      "step": 1145
    },
    {
      "entropy": 6.928429985046387,
      "epoch": 0.101554221123278,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999980429304977,
      "loss": 6.8983,
      "mean_token_accuracy": 0.11434343308210373,
      "num_tokens": 2133718.0,
      "step": 1150
    },
    {
      "entropy": 6.835229969024658,
      "epoch": 0.10199576121511833,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999979093799502,
      "loss": 6.801,
      "mean_token_accuracy": 0.12133207321166992,
      "num_tokens": 2142978.0,
      "step": 1155
    },
    {
      "entropy": 6.8165655612945555,
      "epoch": 0.10243730130695868,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999977714218217,
      "loss": 6.863,
      "mean_token_accuracy": 0.117961073666811,
      "num_tokens": 2152250.0,
      "step": 1160
    },
    {
      "entropy": 6.972959232330322,
      "epoch": 0.102878841398799,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999976290561147,
      "loss": 6.8608,
      "mean_token_accuracy": 0.11624824330210685,
      "num_tokens": 2161620.0,
      "step": 1165
    },
    {
      "entropy": 6.921932792663574,
      "epoch": 0.10332038149063935,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999974822828322,
      "loss": 6.8726,
      "mean_token_accuracy": 0.1173239678144455,
      "num_tokens": 2170856.0,
      "step": 1170
    },
    {
      "entropy": 6.885772609710694,
      "epoch": 0.10376192158247968,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499997331101977,
      "loss": 6.8626,
      "mean_token_accuracy": 0.11568826138973236,
      "num_tokens": 2180926.0,
      "step": 1175
    },
    {
      "entropy": 6.911135244369507,
      "epoch": 0.10420346167432003,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499997175513552,
      "loss": 6.8419,
      "mean_token_accuracy": 0.11412648186087608,
      "num_tokens": 2190248.0,
      "step": 1180
    },
    {
      "entropy": 6.86896710395813,
      "epoch": 0.10464500176616037,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999970155175603,
      "loss": 6.855,
      "mean_token_accuracy": 0.12222710996866226,
      "num_tokens": 2199833.0,
      "step": 1185
    },
    {
      "entropy": 6.856761360168457,
      "epoch": 0.1050865418580007,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499996851114005,
      "loss": 6.8115,
      "mean_token_accuracy": 0.12742498219013215,
      "num_tokens": 2208240.0,
      "step": 1190
    },
    {
      "entropy": 6.837878751754761,
      "epoch": 0.10552808194984105,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999966823028894,
      "loss": 6.8776,
      "mean_token_accuracy": 0.11117666661739349,
      "num_tokens": 2218758.0,
      "step": 1195
    },
    {
      "entropy": 6.94004077911377,
      "epoch": 0.10596962204168138,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999965090842168,
      "loss": 6.8665,
      "mean_token_accuracy": 0.12320348769426345,
      "num_tokens": 2228443.0,
      "step": 1200
    },
    {
      "entropy": 6.74250750541687,
      "epoch": 0.10641116213352173,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999963314579905,
      "loss": 6.7084,
      "mean_token_accuracy": 0.1319122113287449,
      "num_tokens": 2236787.0,
      "step": 1205
    },
    {
      "entropy": 6.956686353683471,
      "epoch": 0.10685270222536206,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999961494242139,
      "loss": 6.8901,
      "mean_token_accuracy": 0.11468368023633957,
      "num_tokens": 2247089.0,
      "step": 1210
    },
    {
      "entropy": 6.789958381652832,
      "epoch": 0.1072942423172024,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999959629828908,
      "loss": 6.816,
      "mean_token_accuracy": 0.11569953635334969,
      "num_tokens": 2256045.0,
      "step": 1215
    },
    {
      "entropy": 6.929527616500854,
      "epoch": 0.10773578240904275,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999957721340248,
      "loss": 6.8769,
      "mean_token_accuracy": 0.12202595993876457,
      "num_tokens": 2265835.0,
      "step": 1220
    },
    {
      "entropy": 6.807823610305786,
      "epoch": 0.10817732250088308,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999955768776195,
      "loss": 6.8076,
      "mean_token_accuracy": 0.11702087000012398,
      "num_tokens": 2275318.0,
      "step": 1225
    },
    {
      "entropy": 6.869472360610962,
      "epoch": 0.10861886259272342,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999953772136788,
      "loss": 6.7978,
      "mean_token_accuracy": 0.12102394551038742,
      "num_tokens": 2284821.0,
      "step": 1230
    },
    {
      "entropy": 6.885239315032959,
      "epoch": 0.10906040268456375,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999951731422068,
      "loss": 6.7645,
      "mean_token_accuracy": 0.12054353281855583,
      "num_tokens": 2294013.0,
      "step": 1235
    },
    {
      "entropy": 6.847450494766235,
      "epoch": 0.1095019427764041,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999949646632072,
      "loss": 6.7626,
      "mean_token_accuracy": 0.12177001982927323,
      "num_tokens": 2302727.0,
      "step": 1240
    },
    {
      "entropy": 6.8077342987060545,
      "epoch": 0.10994348286824443,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999947517766842,
      "loss": 6.8031,
      "mean_token_accuracy": 0.12163913846015931,
      "num_tokens": 2312032.0,
      "step": 1245
    },
    {
      "entropy": 6.986685180664063,
      "epoch": 0.11038502296008477,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499994534482642,
      "loss": 6.8748,
      "mean_token_accuracy": 0.1190544456243515,
      "num_tokens": 2321839.0,
      "step": 1250
    },
    {
      "entropy": 6.811014032363891,
      "epoch": 0.11082656305192512,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999943127810847,
      "loss": 6.8536,
      "mean_token_accuracy": 0.1122577242553234,
      "num_tokens": 2331255.0,
      "step": 1255
    },
    {
      "entropy": 6.794656276702881,
      "epoch": 0.11126810314376545,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999940866720169,
      "loss": 6.6705,
      "mean_token_accuracy": 0.12881582453846932,
      "num_tokens": 2340038.0,
      "step": 1260
    },
    {
      "entropy": 6.71243953704834,
      "epoch": 0.1117096432356058,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999938561554429,
      "loss": 6.7797,
      "mean_token_accuracy": 0.12242485880851746,
      "num_tokens": 2348901.0,
      "step": 1265
    },
    {
      "entropy": 6.858448314666748,
      "epoch": 0.11215118332744613,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999936212313672,
      "loss": 6.8659,
      "mean_token_accuracy": 0.11461173072457313,
      "num_tokens": 2358842.0,
      "step": 1270
    },
    {
      "entropy": 6.8239977836608885,
      "epoch": 0.11259272341928647,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999933818997943,
      "loss": 6.7596,
      "mean_token_accuracy": 0.12424605414271354,
      "num_tokens": 2368650.0,
      "step": 1275
    },
    {
      "entropy": 6.831825399398804,
      "epoch": 0.1130342635111268,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999931381607292,
      "loss": 6.8252,
      "mean_token_accuracy": 0.12058763056993485,
      "num_tokens": 2377916.0,
      "step": 1280
    },
    {
      "entropy": 6.818245553970337,
      "epoch": 0.11347580360296715,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999928900141764,
      "loss": 6.7698,
      "mean_token_accuracy": 0.12198482304811478,
      "num_tokens": 2387507.0,
      "step": 1285
    },
    {
      "entropy": 6.819052505493164,
      "epoch": 0.1139173436948075,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499992637460141,
      "loss": 6.8052,
      "mean_token_accuracy": 0.12523134648799897,
      "num_tokens": 2396148.0,
      "step": 1290
    },
    {
      "entropy": 6.816552209854126,
      "epoch": 0.11435888378664782,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999923804986275,
      "loss": 6.693,
      "mean_token_accuracy": 0.11803872361779214,
      "num_tokens": 2404891.0,
      "step": 1295
    },
    {
      "entropy": 6.840794086456299,
      "epoch": 0.11480042387848817,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999921191296415,
      "loss": 6.7153,
      "mean_token_accuracy": 0.12199744880199433,
      "num_tokens": 2414406.0,
      "step": 1300
    },
    {
      "entropy": 6.7452630519866945,
      "epoch": 0.1152419639703285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999918533531877,
      "loss": 6.8046,
      "mean_token_accuracy": 0.1228412576019764,
      "num_tokens": 2424363.0,
      "step": 1305
    },
    {
      "entropy": 6.852883148193359,
      "epoch": 0.11568350406216885,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999915831692714,
      "loss": 6.7419,
      "mean_token_accuracy": 0.1251549780368805,
      "num_tokens": 2433753.0,
      "step": 1310
    },
    {
      "entropy": 6.7218766689300535,
      "epoch": 0.11612504415400918,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999913085778981,
      "loss": 6.7685,
      "mean_token_accuracy": 0.1185051940381527,
      "num_tokens": 2443275.0,
      "step": 1315
    },
    {
      "entropy": 6.873374080657959,
      "epoch": 0.11656658424584952,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999910295790729,
      "loss": 6.7937,
      "mean_token_accuracy": 0.11835979968309403,
      "num_tokens": 2452510.0,
      "step": 1320
    },
    {
      "entropy": 6.8684648990631105,
      "epoch": 0.11700812433768987,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999907461728014,
      "loss": 6.8746,
      "mean_token_accuracy": 0.1169828750193119,
      "num_tokens": 2462742.0,
      "step": 1325
    },
    {
      "entropy": 6.740426445007325,
      "epoch": 0.1174496644295302,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999904583590893,
      "loss": 6.7434,
      "mean_token_accuracy": 0.12029099017381668,
      "num_tokens": 2471409.0,
      "step": 1330
    },
    {
      "entropy": 6.839800691604614,
      "epoch": 0.11789120452137054,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999901661379418,
      "loss": 6.6931,
      "mean_token_accuracy": 0.12773663252592088,
      "num_tokens": 2481011.0,
      "step": 1335
    },
    {
      "entropy": 6.690527105331421,
      "epoch": 0.11833274461321087,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999898695093652,
      "loss": 6.7866,
      "mean_token_accuracy": 0.12104339599609375,
      "num_tokens": 2490664.0,
      "step": 1340
    },
    {
      "entropy": 6.818962049484253,
      "epoch": 0.11877428470505122,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999895684733648,
      "loss": 6.7279,
      "mean_token_accuracy": 0.12824407517910003,
      "num_tokens": 2499799.0,
      "step": 1345
    },
    {
      "entropy": 6.8539299964904785,
      "epoch": 0.11921582479689156,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004999892630299467,
      "loss": 6.7045,
      "mean_token_accuracy": 0.1257259279489517,
      "num_tokens": 2508780.0,
      "step": 1350
    },
    {
      "entropy": 6.706318616867065,
      "epoch": 0.1196573648887319,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999889531791171,
      "loss": 6.7138,
      "mean_token_accuracy": 0.12127138078212737,
      "num_tokens": 2517741.0,
      "step": 1355
    },
    {
      "entropy": 6.766215896606445,
      "epoch": 0.12009890498057224,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999886389208817,
      "loss": 6.7972,
      "mean_token_accuracy": 0.11826895922422409,
      "num_tokens": 2528742.0,
      "step": 1360
    },
    {
      "entropy": 6.717579746246338,
      "epoch": 0.12054044507241257,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999883202552468,
      "loss": 6.7345,
      "mean_token_accuracy": 0.12455343306064606,
      "num_tokens": 2538609.0,
      "step": 1365
    },
    {
      "entropy": 6.882754898071289,
      "epoch": 0.12098198516425292,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999879971822189,
      "loss": 6.7157,
      "mean_token_accuracy": 0.11966117843985558,
      "num_tokens": 2547772.0,
      "step": 1370
    },
    {
      "entropy": 6.69037971496582,
      "epoch": 0.12142352525609325,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999876697018038,
      "loss": 6.6897,
      "mean_token_accuracy": 0.12769502475857736,
      "num_tokens": 2556114.0,
      "step": 1375
    },
    {
      "entropy": 6.8372406482696535,
      "epoch": 0.1218650653479336,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999873378140085,
      "loss": 6.7182,
      "mean_token_accuracy": 0.12253274098038673,
      "num_tokens": 2566814.0,
      "step": 1380
    },
    {
      "entropy": 6.714360284805298,
      "epoch": 0.12230660543977394,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999870015188389,
      "loss": 6.6914,
      "mean_token_accuracy": 0.12260655164718628,
      "num_tokens": 2576030.0,
      "step": 1385
    },
    {
      "entropy": 6.743939208984375,
      "epoch": 0.12274814553161427,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999866608163021,
      "loss": 6.7176,
      "mean_token_accuracy": 0.1260794699192047,
      "num_tokens": 2585756.0,
      "step": 1390
    },
    {
      "entropy": 6.800289154052734,
      "epoch": 0.12318968562345461,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999863157064045,
      "loss": 6.7797,
      "mean_token_accuracy": 0.1238692507147789,
      "num_tokens": 2595676.0,
      "step": 1395
    },
    {
      "entropy": 6.8109955310821535,
      "epoch": 0.12363122571529495,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999859661891529,
      "loss": 6.7624,
      "mean_token_accuracy": 0.1246532566845417,
      "num_tokens": 2606197.0,
      "step": 1400
    },
    {
      "entropy": 6.76338677406311,
      "epoch": 0.12407276580713529,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999856122645543,
      "loss": 6.6854,
      "mean_token_accuracy": 0.12518818601965903,
      "num_tokens": 2615311.0,
      "step": 1405
    },
    {
      "entropy": 6.725164937973022,
      "epoch": 0.12451430589897562,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999852539326154,
      "loss": 6.5931,
      "mean_token_accuracy": 0.12619537115097046,
      "num_tokens": 2624074.0,
      "step": 1410
    },
    {
      "entropy": 6.681712102890015,
      "epoch": 0.12495584599081597,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999848911933434,
      "loss": 6.7411,
      "mean_token_accuracy": 0.13295672461390495,
      "num_tokens": 2633877.0,
      "step": 1415
    },
    {
      "entropy": 6.7498420715332035,
      "epoch": 0.1253973860826563,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999845240467453,
      "loss": 6.6702,
      "mean_token_accuracy": 0.12471742331981658,
      "num_tokens": 2643330.0,
      "step": 1420
    },
    {
      "entropy": 6.74535722732544,
      "epoch": 0.12583892617449666,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999841524928282,
      "loss": 6.6543,
      "mean_token_accuracy": 0.12766205966472627,
      "num_tokens": 2652070.0,
      "step": 1425
    },
    {
      "entropy": 6.722480249404907,
      "epoch": 0.126280466266337,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999837765315997,
      "loss": 6.697,
      "mean_token_accuracy": 0.1277957484126091,
      "num_tokens": 2660546.0,
      "step": 1430
    },
    {
      "entropy": 6.598982477188111,
      "epoch": 0.12672200635817732,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999833961630669,
      "loss": 6.5999,
      "mean_token_accuracy": 0.1297621488571167,
      "num_tokens": 2669938.0,
      "step": 1435
    },
    {
      "entropy": 6.852873420715332,
      "epoch": 0.12716354645001765,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999830113872374,
      "loss": 6.7248,
      "mean_token_accuracy": 0.12213384285569191,
      "num_tokens": 2679814.0,
      "step": 1440
    },
    {
      "entropy": 6.585267829895019,
      "epoch": 0.127605086541858,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999826222041186,
      "loss": 6.6355,
      "mean_token_accuracy": 0.13383011817932128,
      "num_tokens": 2688733.0,
      "step": 1445
    },
    {
      "entropy": 6.706393527984619,
      "epoch": 0.12804662663369834,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999822286137182,
      "loss": 6.6188,
      "mean_token_accuracy": 0.12774292454123498,
      "num_tokens": 2697744.0,
      "step": 1450
    },
    {
      "entropy": 6.633390617370606,
      "epoch": 0.12848816672553867,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999818306160439,
      "loss": 6.5827,
      "mean_token_accuracy": 0.13278514444828032,
      "num_tokens": 2707037.0,
      "step": 1455
    },
    {
      "entropy": 6.808015394210815,
      "epoch": 0.12892970681737903,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999814282111034,
      "loss": 6.7453,
      "mean_token_accuracy": 0.12486135885119438,
      "num_tokens": 2717345.0,
      "step": 1460
    },
    {
      "entropy": 6.7304778575897215,
      "epoch": 0.12937124690921936,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999810213989047,
      "loss": 6.7017,
      "mean_token_accuracy": 0.12036777138710023,
      "num_tokens": 2726892.0,
      "step": 1465
    },
    {
      "entropy": 6.667453098297119,
      "epoch": 0.1298127870010597,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999806101794558,
      "loss": 6.6615,
      "mean_token_accuracy": 0.12705308422446251,
      "num_tokens": 2736479.0,
      "step": 1470
    },
    {
      "entropy": 6.809973049163818,
      "epoch": 0.13025432709290002,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999801945527648,
      "loss": 6.7078,
      "mean_token_accuracy": 0.12117967531085014,
      "num_tokens": 2745998.0,
      "step": 1475
    },
    {
      "entropy": 6.725075721740723,
      "epoch": 0.13069586718474038,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999797745188395,
      "loss": 6.6906,
      "mean_token_accuracy": 0.12821464985609055,
      "num_tokens": 2754346.0,
      "step": 1480
    },
    {
      "entropy": 6.686868619918823,
      "epoch": 0.13113740727658071,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999793500776886,
      "loss": 6.6285,
      "mean_token_accuracy": 0.1294437274336815,
      "num_tokens": 2763391.0,
      "step": 1485
    },
    {
      "entropy": 6.674964761734008,
      "epoch": 0.13157894736842105,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999789212293201,
      "loss": 6.6827,
      "mean_token_accuracy": 0.12898893728852273,
      "num_tokens": 2772764.0,
      "step": 1490
    },
    {
      "entropy": 6.775622892379761,
      "epoch": 0.1320204874602614,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999784879737423,
      "loss": 6.78,
      "mean_token_accuracy": 0.12160285860300064,
      "num_tokens": 2782312.0,
      "step": 1495
    },
    {
      "entropy": 6.745266675949097,
      "epoch": 0.13246202755210174,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999780503109642,
      "loss": 6.6798,
      "mean_token_accuracy": 0.1227384127676487,
      "num_tokens": 2791159.0,
      "step": 1500
    },
    {
      "entropy": 6.650141859054566,
      "epoch": 0.13290356764394207,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999776082409939,
      "loss": 6.5068,
      "mean_token_accuracy": 0.13456878885626794,
      "num_tokens": 2799319.0,
      "step": 1505
    },
    {
      "entropy": 6.6320771217346195,
      "epoch": 0.1333451077357824,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999771617638401,
      "loss": 6.6316,
      "mean_token_accuracy": 0.12712259590625763,
      "num_tokens": 2807401.0,
      "step": 1510
    },
    {
      "entropy": 6.758873414993286,
      "epoch": 0.13378664782762276,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999767108795118,
      "loss": 6.6961,
      "mean_token_accuracy": 0.12330949455499648,
      "num_tokens": 2817734.0,
      "step": 1515
    },
    {
      "entropy": 6.644004774093628,
      "epoch": 0.1342281879194631,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999762555880176,
      "loss": 6.6783,
      "mean_token_accuracy": 0.12152940481901169,
      "num_tokens": 2828235.0,
      "step": 1520
    },
    {
      "entropy": 6.743390846252441,
      "epoch": 0.13466972801130342,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999757958893666,
      "loss": 6.7124,
      "mean_token_accuracy": 0.1237283930182457,
      "num_tokens": 2837453.0,
      "step": 1525
    },
    {
      "entropy": 6.60786280632019,
      "epoch": 0.13511126810314378,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999753317835677,
      "loss": 6.6795,
      "mean_token_accuracy": 0.12087962031364441,
      "num_tokens": 2847055.0,
      "step": 1530
    },
    {
      "entropy": 6.7429241180419925,
      "epoch": 0.1355528081949841,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999748632706299,
      "loss": 6.6568,
      "mean_token_accuracy": 0.13167096227407454,
      "num_tokens": 2857101.0,
      "step": 1535
    },
    {
      "entropy": 6.673015403747558,
      "epoch": 0.13599434828682444,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999743903505626,
      "loss": 6.553,
      "mean_token_accuracy": 0.1336723633110523,
      "num_tokens": 2866685.0,
      "step": 1540
    },
    {
      "entropy": 6.676457214355469,
      "epoch": 0.13643588837866477,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999739130233749,
      "loss": 6.6731,
      "mean_token_accuracy": 0.12713466510176658,
      "num_tokens": 2876022.0,
      "step": 1545
    },
    {
      "entropy": 6.783061361312866,
      "epoch": 0.13687742847050513,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999734312890761,
      "loss": 6.6062,
      "mean_token_accuracy": 0.12626957073807715,
      "num_tokens": 2885560.0,
      "step": 1550
    },
    {
      "entropy": 6.59228024482727,
      "epoch": 0.13731896856234546,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999729451476757,
      "loss": 6.6439,
      "mean_token_accuracy": 0.12623701319098474,
      "num_tokens": 2894686.0,
      "step": 1555
    },
    {
      "entropy": 6.730476140975952,
      "epoch": 0.1377605086541858,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999724545991835,
      "loss": 6.6588,
      "mean_token_accuracy": 0.13341889455914496,
      "num_tokens": 2904390.0,
      "step": 1560
    },
    {
      "entropy": 6.662082195281982,
      "epoch": 0.13820204874602615,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999719596436086,
      "loss": 6.6982,
      "mean_token_accuracy": 0.12678939029574393,
      "num_tokens": 2913311.0,
      "step": 1565
    },
    {
      "entropy": 6.736054372787476,
      "epoch": 0.13864358883786648,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999714602809611,
      "loss": 6.578,
      "mean_token_accuracy": 0.13448369055986403,
      "num_tokens": 2923196.0,
      "step": 1570
    },
    {
      "entropy": 6.637621927261352,
      "epoch": 0.1390851289297068,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999709565112506,
      "loss": 6.6428,
      "mean_token_accuracy": 0.12785085365176202,
      "num_tokens": 2932813.0,
      "step": 1575
    },
    {
      "entropy": 6.658017587661743,
      "epoch": 0.13952666902154715,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499970448334487,
      "loss": 6.5634,
      "mean_token_accuracy": 0.119098000228405,
      "num_tokens": 2942694.0,
      "step": 1580
    },
    {
      "entropy": 6.5801304340362545,
      "epoch": 0.1399682091133875,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999699357506803,
      "loss": 6.5185,
      "mean_token_accuracy": 0.1330326519906521,
      "num_tokens": 2950932.0,
      "step": 1585
    },
    {
      "entropy": 6.701116371154785,
      "epoch": 0.14040974920522784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999694187598406,
      "loss": 6.6085,
      "mean_token_accuracy": 0.1314128704369068,
      "num_tokens": 2960102.0,
      "step": 1590
    },
    {
      "entropy": 6.627189350128174,
      "epoch": 0.14085128929706817,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004999688973619777,
      "loss": 6.6088,
      "mean_token_accuracy": 0.12604895159602164,
      "num_tokens": 2969968.0,
      "step": 1595
    },
    {
      "entropy": 6.588514184951782,
      "epoch": 0.14129282938890853,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999683715571022,
      "loss": 6.5542,
      "mean_token_accuracy": 0.1347218669950962,
      "num_tokens": 2978880.0,
      "step": 1600
    },
    {
      "entropy": 6.644829893112183,
      "epoch": 0.14173436948074886,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999678413452242,
      "loss": 6.5863,
      "mean_token_accuracy": 0.12890450209379195,
      "num_tokens": 2988369.0,
      "step": 1605
    },
    {
      "entropy": 6.667511320114135,
      "epoch": 0.1421759095725892,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999673067263542,
      "loss": 6.6373,
      "mean_token_accuracy": 0.12620072290301323,
      "num_tokens": 2997070.0,
      "step": 1610
    },
    {
      "entropy": 6.682367372512817,
      "epoch": 0.14261744966442952,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999667677005026,
      "loss": 6.5749,
      "mean_token_accuracy": 0.13310741856694222,
      "num_tokens": 3006547.0,
      "step": 1615
    },
    {
      "entropy": 6.584122562408448,
      "epoch": 0.14305898975626988,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999662242676799,
      "loss": 6.5986,
      "mean_token_accuracy": 0.1310425490140915,
      "num_tokens": 3015821.0,
      "step": 1620
    },
    {
      "entropy": 6.665965223312378,
      "epoch": 0.1435005298481102,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999656764278968,
      "loss": 6.5655,
      "mean_token_accuracy": 0.1309148021042347,
      "num_tokens": 3024750.0,
      "step": 1625
    },
    {
      "entropy": 6.623423147201538,
      "epoch": 0.14394206993995054,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999651241811642,
      "loss": 6.5389,
      "mean_token_accuracy": 0.1308048278093338,
      "num_tokens": 3033345.0,
      "step": 1630
    },
    {
      "entropy": 6.59770941734314,
      "epoch": 0.1443836100317909,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999645675274925,
      "loss": 6.5209,
      "mean_token_accuracy": 0.13443350791931152,
      "num_tokens": 3042060.0,
      "step": 1635
    },
    {
      "entropy": 6.660798358917236,
      "epoch": 0.14482515012363123,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999640064668931,
      "loss": 6.6684,
      "mean_token_accuracy": 0.12563745751976968,
      "num_tokens": 3052490.0,
      "step": 1640
    },
    {
      "entropy": 6.726674222946167,
      "epoch": 0.14526669021547156,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999634409993766,
      "loss": 6.6441,
      "mean_token_accuracy": 0.12772516757249833,
      "num_tokens": 3061934.0,
      "step": 1645
    },
    {
      "entropy": 6.556687307357788,
      "epoch": 0.1457082303073119,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999628711249544,
      "loss": 6.5591,
      "mean_token_accuracy": 0.13611432090401648,
      "num_tokens": 3070890.0,
      "step": 1650
    },
    {
      "entropy": 6.703302097320557,
      "epoch": 0.14614977039915225,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999622968436373,
      "loss": 6.5614,
      "mean_token_accuracy": 0.12631918862462044,
      "num_tokens": 3079933.0,
      "step": 1655
    },
    {
      "entropy": 6.5565461158752445,
      "epoch": 0.14659131049099258,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999617181554369,
      "loss": 6.6078,
      "mean_token_accuracy": 0.13254086449742317,
      "num_tokens": 3089910.0,
      "step": 1660
    },
    {
      "entropy": 6.718549633026123,
      "epoch": 0.1470328505828329,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999611350603643,
      "loss": 6.5916,
      "mean_token_accuracy": 0.13437702059745787,
      "num_tokens": 3098676.0,
      "step": 1665
    },
    {
      "entropy": 6.595283174514771,
      "epoch": 0.14747439067467327,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499960547558431,
      "loss": 6.7159,
      "mean_token_accuracy": 0.1256335400044918,
      "num_tokens": 3109055.0,
      "step": 1670
    },
    {
      "entropy": 6.624332904815674,
      "epoch": 0.1479159307665136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999599556496486,
      "loss": 6.5472,
      "mean_token_accuracy": 0.13295547068119049,
      "num_tokens": 3117517.0,
      "step": 1675
    },
    {
      "entropy": 6.573185825347901,
      "epoch": 0.14835747085835393,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999593593340286,
      "loss": 6.5544,
      "mean_token_accuracy": 0.13249536529183387,
      "num_tokens": 3126606.0,
      "step": 1680
    },
    {
      "entropy": 6.704092359542846,
      "epoch": 0.14879901095019427,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999587586115826,
      "loss": 6.5733,
      "mean_token_accuracy": 0.12993996366858482,
      "num_tokens": 3135444.0,
      "step": 1685
    },
    {
      "entropy": 6.57456955909729,
      "epoch": 0.14924055104203462,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999581534823226,
      "loss": 6.4927,
      "mean_token_accuracy": 0.14061653688549997,
      "num_tokens": 3144967.0,
      "step": 1690
    },
    {
      "entropy": 6.607442092895508,
      "epoch": 0.14968209113387496,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999575439462601,
      "loss": 6.522,
      "mean_token_accuracy": 0.12690635845065118,
      "num_tokens": 3153898.0,
      "step": 1695
    },
    {
      "entropy": 6.536995601654053,
      "epoch": 0.1501236312257153,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999569300034075,
      "loss": 6.5921,
      "mean_token_accuracy": 0.12649724259972572,
      "num_tokens": 3162341.0,
      "step": 1700
    },
    {
      "entropy": 6.60729718208313,
      "epoch": 0.15056517131755565,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999563116537764,
      "loss": 6.3826,
      "mean_token_accuracy": 0.14100785404443741,
      "num_tokens": 3171420.0,
      "step": 1705
    },
    {
      "entropy": 6.54091591835022,
      "epoch": 0.15100671140939598,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999556888973792,
      "loss": 6.4366,
      "mean_token_accuracy": 0.13033056780695915,
      "num_tokens": 3180221.0,
      "step": 1710
    },
    {
      "entropy": 6.6625391960144045,
      "epoch": 0.1514482515012363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999550617342279,
      "loss": 6.587,
      "mean_token_accuracy": 0.12798949852585792,
      "num_tokens": 3189156.0,
      "step": 1715
    },
    {
      "entropy": 6.539492607116699,
      "epoch": 0.15188979159307664,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499954430164335,
      "loss": 6.5149,
      "mean_token_accuracy": 0.13213447630405425,
      "num_tokens": 3199870.0,
      "step": 1720
    },
    {
      "entropy": 6.559759998321534,
      "epoch": 0.152331331684917,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999537941877127,
      "loss": 6.4883,
      "mean_token_accuracy": 0.13440720960497857,
      "num_tokens": 3208815.0,
      "step": 1725
    },
    {
      "entropy": 6.604632616043091,
      "epoch": 0.15277287177675733,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999531538043735,
      "loss": 6.5717,
      "mean_token_accuracy": 0.13279605731368066,
      "num_tokens": 3218692.0,
      "step": 1730
    },
    {
      "entropy": 6.584544372558594,
      "epoch": 0.15321441186859766,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999525090143298,
      "loss": 6.5367,
      "mean_token_accuracy": 0.14099612906575204,
      "num_tokens": 3227604.0,
      "step": 1735
    },
    {
      "entropy": 6.629879665374756,
      "epoch": 0.15365595196043802,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999518598175946,
      "loss": 6.5446,
      "mean_token_accuracy": 0.14111834466457368,
      "num_tokens": 3237912.0,
      "step": 1740
    },
    {
      "entropy": 6.453984022140503,
      "epoch": 0.15409749205227835,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999512062141805,
      "loss": 6.4239,
      "mean_token_accuracy": 0.13701695203781128,
      "num_tokens": 3246003.0,
      "step": 1745
    },
    {
      "entropy": 6.6436468124389645,
      "epoch": 0.15453903214411868,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999505482040999,
      "loss": 6.5558,
      "mean_token_accuracy": 0.13121686428785323,
      "num_tokens": 3256363.0,
      "step": 1750
    },
    {
      "entropy": 6.583334445953369,
      "epoch": 0.154980572235959,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999498857873662,
      "loss": 6.5518,
      "mean_token_accuracy": 0.13320463374257088,
      "num_tokens": 3265822.0,
      "step": 1755
    },
    {
      "entropy": 6.478305387496948,
      "epoch": 0.15542211232779937,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999492189639921,
      "loss": 6.4224,
      "mean_token_accuracy": 0.13528963476419448,
      "num_tokens": 3274614.0,
      "step": 1760
    },
    {
      "entropy": 6.61256685256958,
      "epoch": 0.1558636524196397,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999485477339907,
      "loss": 6.5201,
      "mean_token_accuracy": 0.13317029252648355,
      "num_tokens": 3283800.0,
      "step": 1765
    },
    {
      "entropy": 6.526872968673706,
      "epoch": 0.15630519251148003,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999478720973753,
      "loss": 6.4319,
      "mean_token_accuracy": 0.14009243845939637,
      "num_tokens": 3293221.0,
      "step": 1770
    },
    {
      "entropy": 6.5570995807647705,
      "epoch": 0.1567467326033204,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499947192054159,
      "loss": 6.6029,
      "mean_token_accuracy": 0.1324251540005207,
      "num_tokens": 3302852.0,
      "step": 1775
    },
    {
      "entropy": 6.650819587707519,
      "epoch": 0.15718827269516072,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499946507604355,
      "loss": 6.4641,
      "mean_token_accuracy": 0.14009604677557946,
      "num_tokens": 3311752.0,
      "step": 1780
    },
    {
      "entropy": 6.473453664779663,
      "epoch": 0.15762981278700106,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499945818747977,
      "loss": 6.5492,
      "mean_token_accuracy": 0.13722263872623444,
      "num_tokens": 3321339.0,
      "step": 1785
    },
    {
      "entropy": 6.671073293685913,
      "epoch": 0.1580713528788414,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999451254850383,
      "loss": 6.5514,
      "mean_token_accuracy": 0.1324629843235016,
      "num_tokens": 3330269.0,
      "step": 1790
    },
    {
      "entropy": 6.550148677825928,
      "epoch": 0.15851289297068175,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999444278155525,
      "loss": 6.4576,
      "mean_token_accuracy": 0.1376182422041893,
      "num_tokens": 3340770.0,
      "step": 1795
    },
    {
      "entropy": 6.544644594192505,
      "epoch": 0.15895443306252208,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999437257395333,
      "loss": 6.5753,
      "mean_token_accuracy": 0.13476464301347732,
      "num_tokens": 3349976.0,
      "step": 1800
    },
    {
      "entropy": 6.597910165786743,
      "epoch": 0.1593959731543624,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999430192569944,
      "loss": 6.6158,
      "mean_token_accuracy": 0.12608520165085793,
      "num_tokens": 3359764.0,
      "step": 1805
    },
    {
      "entropy": 6.539244031906128,
      "epoch": 0.15983751324620277,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999423083679498,
      "loss": 6.4868,
      "mean_token_accuracy": 0.13115186169743537,
      "num_tokens": 3369939.0,
      "step": 1810
    },
    {
      "entropy": 6.639003801345825,
      "epoch": 0.1602790533380431,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999415930724133,
      "loss": 6.6346,
      "mean_token_accuracy": 0.12994891554117202,
      "num_tokens": 3381326.0,
      "step": 1815
    },
    {
      "entropy": 6.659589242935181,
      "epoch": 0.16072059342988343,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999408733703988,
      "loss": 6.5949,
      "mean_token_accuracy": 0.12143847793340683,
      "num_tokens": 3391016.0,
      "step": 1820
    },
    {
      "entropy": 6.533053731918335,
      "epoch": 0.16116213352172376,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999401492619207,
      "loss": 6.3366,
      "mean_token_accuracy": 0.13945143148303032,
      "num_tokens": 3400294.0,
      "step": 1825
    },
    {
      "entropy": 6.587634897232055,
      "epoch": 0.16160367361356412,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999394207469928,
      "loss": 6.5447,
      "mean_token_accuracy": 0.12779992073774338,
      "num_tokens": 3409685.0,
      "step": 1830
    },
    {
      "entropy": 6.636557579040527,
      "epoch": 0.16204521370540445,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999386878256297,
      "loss": 6.4611,
      "mean_token_accuracy": 0.1320968374609947,
      "num_tokens": 3418946.0,
      "step": 1835
    },
    {
      "entropy": 6.423011112213135,
      "epoch": 0.16248675379724478,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999379504978457,
      "loss": 6.468,
      "mean_token_accuracy": 0.13338307663798332,
      "num_tokens": 3428245.0,
      "step": 1840
    },
    {
      "entropy": 6.578200912475586,
      "epoch": 0.16292829388908514,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499937208763655,
      "loss": 6.4741,
      "mean_token_accuracy": 0.1402788795530796,
      "num_tokens": 3437580.0,
      "step": 1845
    },
    {
      "entropy": 6.617180633544922,
      "epoch": 0.16336983398092547,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999364626230724,
      "loss": 6.4472,
      "mean_token_accuracy": 0.14007550328969956,
      "num_tokens": 3446393.0,
      "step": 1850
    },
    {
      "entropy": 6.451139450073242,
      "epoch": 0.1638113740727658,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999357120761124,
      "loss": 6.551,
      "mean_token_accuracy": 0.13571736514568328,
      "num_tokens": 3455956.0,
      "step": 1855
    },
    {
      "entropy": 6.6568724632263185,
      "epoch": 0.16425291416460613,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999349571227898,
      "loss": 6.5495,
      "mean_token_accuracy": 0.13430711701512338,
      "num_tokens": 3465722.0,
      "step": 1860
    },
    {
      "entropy": 6.55973629951477,
      "epoch": 0.1646944542564465,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999341977631193,
      "loss": 6.52,
      "mean_token_accuracy": 0.12885117009282113,
      "num_tokens": 3475467.0,
      "step": 1865
    },
    {
      "entropy": 6.513189268112183,
      "epoch": 0.16513599434828682,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999334339971157,
      "loss": 6.4162,
      "mean_token_accuracy": 0.13661258816719055,
      "num_tokens": 3484931.0,
      "step": 1870
    },
    {
      "entropy": 6.572188520431519,
      "epoch": 0.16557753444012716,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999326658247942,
      "loss": 6.5161,
      "mean_token_accuracy": 0.13407543525099755,
      "num_tokens": 3494001.0,
      "step": 1875
    },
    {
      "entropy": 6.522925567626953,
      "epoch": 0.16601907453196751,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999318932461696,
      "loss": 6.4558,
      "mean_token_accuracy": 0.1386028841137886,
      "num_tokens": 3503021.0,
      "step": 1880
    },
    {
      "entropy": 6.46082649230957,
      "epoch": 0.16646061462380785,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999311162612571,
      "loss": 6.4843,
      "mean_token_accuracy": 0.13625017702579498,
      "num_tokens": 3513005.0,
      "step": 1885
    },
    {
      "entropy": 6.661849021911621,
      "epoch": 0.16690215471564818,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499930334870072,
      "loss": 6.5629,
      "mean_token_accuracy": 0.13853515014052392,
      "num_tokens": 3523219.0,
      "step": 1890
    },
    {
      "entropy": 6.584684419631958,
      "epoch": 0.1673436948074885,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999295490726296,
      "loss": 6.51,
      "mean_token_accuracy": 0.13359814062714576,
      "num_tokens": 3532917.0,
      "step": 1895
    },
    {
      "entropy": 6.513212633132935,
      "epoch": 0.16778523489932887,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999287588689453,
      "loss": 6.5245,
      "mean_token_accuracy": 0.1389150969684124,
      "num_tokens": 3542845.0,
      "step": 1900
    },
    {
      "entropy": 6.599180459976196,
      "epoch": 0.1682267749911692,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999279642590344,
      "loss": 6.5438,
      "mean_token_accuracy": 0.13201173320412635,
      "num_tokens": 3552126.0,
      "step": 1905
    },
    {
      "entropy": 6.5960267066955565,
      "epoch": 0.16866831508300953,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999271652429127,
      "loss": 6.5561,
      "mean_token_accuracy": 0.12684691920876504,
      "num_tokens": 3561254.0,
      "step": 1910
    },
    {
      "entropy": 6.528163957595825,
      "epoch": 0.1691098551748499,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999263618205958,
      "loss": 6.3933,
      "mean_token_accuracy": 0.1403527893126011,
      "num_tokens": 3569781.0,
      "step": 1915
    },
    {
      "entropy": 6.555511140823365,
      "epoch": 0.16955139526669022,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999255539920993,
      "loss": 6.4866,
      "mean_token_accuracy": 0.13409090787172318,
      "num_tokens": 3579664.0,
      "step": 1920
    },
    {
      "entropy": 6.567546367645264,
      "epoch": 0.16999293535853055,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999247417574391,
      "loss": 6.5376,
      "mean_token_accuracy": 0.13570395410060881,
      "num_tokens": 3588671.0,
      "step": 1925
    },
    {
      "entropy": 6.54329285621643,
      "epoch": 0.17043447545037088,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999239251166312,
      "loss": 6.4281,
      "mean_token_accuracy": 0.13799721151590347,
      "num_tokens": 3597656.0,
      "step": 1930
    },
    {
      "entropy": 6.480014228820801,
      "epoch": 0.17087601554221124,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999231040696914,
      "loss": 6.4491,
      "mean_token_accuracy": 0.13938046917319297,
      "num_tokens": 3608017.0,
      "step": 1935
    },
    {
      "entropy": 6.529258060455322,
      "epoch": 0.17131755563405157,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999222786166361,
      "loss": 6.5236,
      "mean_token_accuracy": 0.1355483777821064,
      "num_tokens": 3618189.0,
      "step": 1940
    },
    {
      "entropy": 6.60916223526001,
      "epoch": 0.1717590957258919,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999214487574812,
      "loss": 6.4772,
      "mean_token_accuracy": 0.13526797890663148,
      "num_tokens": 3627211.0,
      "step": 1945
    },
    {
      "entropy": 6.498088455200195,
      "epoch": 0.17220063581773226,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999206144922431,
      "loss": 6.4181,
      "mean_token_accuracy": 0.13236208409070968,
      "num_tokens": 3636781.0,
      "step": 1950
    },
    {
      "entropy": 6.4924522876739506,
      "epoch": 0.1726421759095726,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499919775820938,
      "loss": 6.4985,
      "mean_token_accuracy": 0.13709900975227357,
      "num_tokens": 3644891.0,
      "step": 1955
    },
    {
      "entropy": 6.591422748565674,
      "epoch": 0.17308371600141292,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999189327435825,
      "loss": 6.5295,
      "mean_token_accuracy": 0.13574066162109374,
      "num_tokens": 3655477.0,
      "step": 1960
    },
    {
      "entropy": 6.5104138374328615,
      "epoch": 0.17352525609325326,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999180852601929,
      "loss": 6.5386,
      "mean_token_accuracy": 0.13653166219592094,
      "num_tokens": 3664542.0,
      "step": 1965
    },
    {
      "entropy": 6.577597141265869,
      "epoch": 0.17396679618509361,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000499917233370786,
      "loss": 6.4609,
      "mean_token_accuracy": 0.1293606199324131,
      "num_tokens": 3673806.0,
      "step": 1970
    },
    {
      "entropy": 6.484398126602173,
      "epoch": 0.17440833627693395,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999163770753784,
      "loss": 6.4524,
      "mean_token_accuracy": 0.13555625528097154,
      "num_tokens": 3683238.0,
      "step": 1975
    },
    {
      "entropy": 6.558987331390381,
      "epoch": 0.17484987636877428,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999155163739869,
      "loss": 6.4372,
      "mean_token_accuracy": 0.13345005139708518,
      "num_tokens": 3692161.0,
      "step": 1980
    },
    {
      "entropy": 6.510820627212524,
      "epoch": 0.17529141646061464,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999146512666284,
      "loss": 6.4535,
      "mean_token_accuracy": 0.13496344164013863,
      "num_tokens": 3701431.0,
      "step": 1985
    },
    {
      "entropy": 6.530817365646362,
      "epoch": 0.17573295655245497,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999137817533197,
      "loss": 6.4293,
      "mean_token_accuracy": 0.1400221474468708,
      "num_tokens": 3710963.0,
      "step": 1990
    },
    {
      "entropy": 6.48544750213623,
      "epoch": 0.1761744966442953,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999129078340779,
      "loss": 6.4221,
      "mean_token_accuracy": 0.14005866199731826,
      "num_tokens": 3720177.0,
      "step": 1995
    },
    {
      "entropy": 6.465721511840821,
      "epoch": 0.17661603673613563,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999120295089202,
      "loss": 6.3795,
      "mean_token_accuracy": 0.14514245688915253,
      "num_tokens": 3728123.0,
      "step": 2000
    },
    {
      "entropy": 6.571569442749023,
      "epoch": 0.177057576827976,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999111467778639,
      "loss": 6.4978,
      "mean_token_accuracy": 0.1349009484052658,
      "num_tokens": 3736869.0,
      "step": 2005
    },
    {
      "entropy": 6.4819518566131595,
      "epoch": 0.17749911691981632,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499910259640926,
      "loss": 6.3989,
      "mean_token_accuracy": 0.1346891440451145,
      "num_tokens": 3745830.0,
      "step": 2010
    },
    {
      "entropy": 6.486936283111572,
      "epoch": 0.17794065701165665,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499909368098124,
      "loss": 6.4328,
      "mean_token_accuracy": 0.13926308006048202,
      "num_tokens": 3755019.0,
      "step": 2015
    },
    {
      "entropy": 6.517996597290039,
      "epoch": 0.178382197103497,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999084721494754,
      "loss": 6.4076,
      "mean_token_accuracy": 0.13306454047560692,
      "num_tokens": 3764814.0,
      "step": 2020
    },
    {
      "entropy": 6.465504217147827,
      "epoch": 0.17882373719533734,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999075717949978,
      "loss": 6.384,
      "mean_token_accuracy": 0.14004691764712335,
      "num_tokens": 3774258.0,
      "step": 2025
    },
    {
      "entropy": 6.422070741653442,
      "epoch": 0.17926527728717767,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999066670347089,
      "loss": 6.4155,
      "mean_token_accuracy": 0.14247353076934816,
      "num_tokens": 3783961.0,
      "step": 2030
    },
    {
      "entropy": 6.416653776168824,
      "epoch": 0.179706817379018,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999057578686261,
      "loss": 6.3804,
      "mean_token_accuracy": 0.1390853337943554,
      "num_tokens": 3792417.0,
      "step": 2035
    },
    {
      "entropy": 6.496200704574585,
      "epoch": 0.18014835747085836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999048442967675,
      "loss": 6.3592,
      "mean_token_accuracy": 0.13822411596775055,
      "num_tokens": 3801479.0,
      "step": 2040
    },
    {
      "entropy": 6.456766891479492,
      "epoch": 0.1805898975626987,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999039263191508,
      "loss": 6.4305,
      "mean_token_accuracy": 0.133939179033041,
      "num_tokens": 3810799.0,
      "step": 2045
    },
    {
      "entropy": 6.5264753818511965,
      "epoch": 0.18103143765453902,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999030039357943,
      "loss": 6.48,
      "mean_token_accuracy": 0.1298608623445034,
      "num_tokens": 3820966.0,
      "step": 2050
    },
    {
      "entropy": 6.493787717819214,
      "epoch": 0.18147297774637938,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999020771467158,
      "loss": 6.4572,
      "mean_token_accuracy": 0.1327923409640789,
      "num_tokens": 3829247.0,
      "step": 2055
    },
    {
      "entropy": 6.596197843551636,
      "epoch": 0.1819145178382197,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999011459519335,
      "loss": 6.3948,
      "mean_token_accuracy": 0.13827238082885743,
      "num_tokens": 3838114.0,
      "step": 2060
    },
    {
      "entropy": 6.390679597854614,
      "epoch": 0.18235605793006004,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999002103514655,
      "loss": 6.4735,
      "mean_token_accuracy": 0.13869670927524566,
      "num_tokens": 3848075.0,
      "step": 2065
    },
    {
      "entropy": 6.585737419128418,
      "epoch": 0.18279759802190038,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998992703453304,
      "loss": 6.5106,
      "mean_token_accuracy": 0.13122646436095237,
      "num_tokens": 3857934.0,
      "step": 2070
    },
    {
      "entropy": 6.450649833679199,
      "epoch": 0.18323913811374073,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998983259335466,
      "loss": 6.3519,
      "mean_token_accuracy": 0.14236594662070273,
      "num_tokens": 3866707.0,
      "step": 2075
    },
    {
      "entropy": 6.58008508682251,
      "epoch": 0.18368067820558107,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998973771161324,
      "loss": 6.4244,
      "mean_token_accuracy": 0.1374949462711811,
      "num_tokens": 3875233.0,
      "step": 2080
    },
    {
      "entropy": 6.389654541015625,
      "epoch": 0.1841222182974214,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998964238931065,
      "loss": 6.4131,
      "mean_token_accuracy": 0.1403422772884369,
      "num_tokens": 3885173.0,
      "step": 2085
    },
    {
      "entropy": 6.564711236953736,
      "epoch": 0.18456375838926176,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998954662644876,
      "loss": 6.3803,
      "mean_token_accuracy": 0.13195990920066833,
      "num_tokens": 3894198.0,
      "step": 2090
    },
    {
      "entropy": 6.44653811454773,
      "epoch": 0.1850052984811021,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998945042302943,
      "loss": 6.382,
      "mean_token_accuracy": 0.1373509407043457,
      "num_tokens": 3904076.0,
      "step": 2095
    },
    {
      "entropy": 6.4586262702941895,
      "epoch": 0.18544683857294242,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998935377905457,
      "loss": 6.4943,
      "mean_token_accuracy": 0.13231708630919456,
      "num_tokens": 3913204.0,
      "step": 2100
    },
    {
      "entropy": 6.587808132171631,
      "epoch": 0.18588837866478275,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998925669452605,
      "loss": 6.4565,
      "mean_token_accuracy": 0.1339179016649723,
      "num_tokens": 3922148.0,
      "step": 2105
    },
    {
      "entropy": 6.401728963851928,
      "epoch": 0.1863299187566231,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998915916944579,
      "loss": 6.4234,
      "mean_token_accuracy": 0.1407448723912239,
      "num_tokens": 3931333.0,
      "step": 2110
    },
    {
      "entropy": 6.545709466934204,
      "epoch": 0.18677145884846344,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998906120381568,
      "loss": 6.3789,
      "mean_token_accuracy": 0.1448886923491955,
      "num_tokens": 3941061.0,
      "step": 2115
    },
    {
      "entropy": 6.505582046508789,
      "epoch": 0.18721299894030377,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998896279763766,
      "loss": 6.4761,
      "mean_token_accuracy": 0.13257319629192352,
      "num_tokens": 3950075.0,
      "step": 2120
    },
    {
      "entropy": 6.455861520767212,
      "epoch": 0.18765453903214413,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998886395091365,
      "loss": 6.3345,
      "mean_token_accuracy": 0.1409289576113224,
      "num_tokens": 3958885.0,
      "step": 2125
    },
    {
      "entropy": 6.417616128921509,
      "epoch": 0.18809607912398446,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998876466364559,
      "loss": 6.437,
      "mean_token_accuracy": 0.13843559697270394,
      "num_tokens": 3968218.0,
      "step": 2130
    },
    {
      "entropy": 6.400978994369507,
      "epoch": 0.1885376192158248,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998866493583541,
      "loss": 6.364,
      "mean_token_accuracy": 0.14499804973602295,
      "num_tokens": 3977435.0,
      "step": 2135
    },
    {
      "entropy": 6.388528203964233,
      "epoch": 0.18897915930766512,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998856476748509,
      "loss": 6.349,
      "mean_token_accuracy": 0.14326094537973405,
      "num_tokens": 3986608.0,
      "step": 2140
    },
    {
      "entropy": 6.440869951248169,
      "epoch": 0.18942069939950548,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998846415859656,
      "loss": 6.3602,
      "mean_token_accuracy": 0.14130929261445999,
      "num_tokens": 3996087.0,
      "step": 2145
    },
    {
      "entropy": 6.476541662216187,
      "epoch": 0.1898622394913458,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998836310917182,
      "loss": 6.4058,
      "mean_token_accuracy": 0.13654726892709732,
      "num_tokens": 4006257.0,
      "step": 2150
    },
    {
      "entropy": 6.478487300872803,
      "epoch": 0.19030377958318614,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998826161921282,
      "loss": 6.4072,
      "mean_token_accuracy": 0.14593140706419944,
      "num_tokens": 4015904.0,
      "step": 2155
    },
    {
      "entropy": 6.427832221984863,
      "epoch": 0.1907453196750265,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998815968872157,
      "loss": 6.4181,
      "mean_token_accuracy": 0.13653010204434396,
      "num_tokens": 4025417.0,
      "step": 2160
    },
    {
      "entropy": 6.392728614807129,
      "epoch": 0.19118685976686683,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998805731770007,
      "loss": 6.2628,
      "mean_token_accuracy": 0.15659967064857483,
      "num_tokens": 4035181.0,
      "step": 2165
    },
    {
      "entropy": 6.4461814880371096,
      "epoch": 0.19162839985870717,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499879545061503,
      "loss": 6.4504,
      "mean_token_accuracy": 0.13371687456965448,
      "num_tokens": 4045112.0,
      "step": 2170
    },
    {
      "entropy": 6.5883321285247805,
      "epoch": 0.1920699399505475,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998785125407432,
      "loss": 6.5425,
      "mean_token_accuracy": 0.12125966772437095,
      "num_tokens": 4054566.0,
      "step": 2175
    },
    {
      "entropy": 6.552111434936523,
      "epoch": 0.19251148004238786,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499877475614741,
      "loss": 6.3758,
      "mean_token_accuracy": 0.1340583384037018,
      "num_tokens": 4063960.0,
      "step": 2180
    },
    {
      "entropy": 6.3515486240386965,
      "epoch": 0.1929530201342282,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998764342835169,
      "loss": 6.3157,
      "mean_token_accuracy": 0.14617449343204497,
      "num_tokens": 4072620.0,
      "step": 2185
    },
    {
      "entropy": 6.406283712387085,
      "epoch": 0.19339456022606852,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998753885470915,
      "loss": 6.2789,
      "mean_token_accuracy": 0.14062159806489943,
      "num_tokens": 4081590.0,
      "step": 2190
    },
    {
      "entropy": 6.425132322311401,
      "epoch": 0.19383610031790888,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998743384054851,
      "loss": 6.4202,
      "mean_token_accuracy": 0.14044143706560136,
      "num_tokens": 4090758.0,
      "step": 2195
    },
    {
      "entropy": 6.3962568759918215,
      "epoch": 0.1942776404097492,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998732838587183,
      "loss": 6.2458,
      "mean_token_accuracy": 0.149021477997303,
      "num_tokens": 4099281.0,
      "step": 2200
    },
    {
      "entropy": 6.40768404006958,
      "epoch": 0.19471918050158954,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998722249068118,
      "loss": 6.3451,
      "mean_token_accuracy": 0.1404854990541935,
      "num_tokens": 4108953.0,
      "step": 2205
    },
    {
      "entropy": 6.396564531326294,
      "epoch": 0.19516072059342987,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998711615497863,
      "loss": 6.4535,
      "mean_token_accuracy": 0.13740591406822206,
      "num_tokens": 4118799.0,
      "step": 2210
    },
    {
      "entropy": 6.579476261138916,
      "epoch": 0.19560226068527023,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998700937876626,
      "loss": 6.4122,
      "mean_token_accuracy": 0.13571444600820542,
      "num_tokens": 4127862.0,
      "step": 2215
    },
    {
      "entropy": 6.4028857231140135,
      "epoch": 0.19604380077711056,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998690216204615,
      "loss": 6.5068,
      "mean_token_accuracy": 0.12704429849982263,
      "num_tokens": 4139029.0,
      "step": 2220
    },
    {
      "entropy": 6.5047478675842285,
      "epoch": 0.1964853408689509,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998679450482043,
      "loss": 6.3028,
      "mean_token_accuracy": 0.14782762974500657,
      "num_tokens": 4148257.0,
      "step": 2225
    },
    {
      "entropy": 6.341253662109375,
      "epoch": 0.19692688096079125,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499866864070912,
      "loss": 6.2627,
      "mean_token_accuracy": 0.14017492160201073,
      "num_tokens": 4157626.0,
      "step": 2230
    },
    {
      "entropy": 6.429310417175293,
      "epoch": 0.19736842105263158,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998657786886056,
      "loss": 6.4713,
      "mean_token_accuracy": 0.13993202298879623,
      "num_tokens": 4166804.0,
      "step": 2235
    },
    {
      "entropy": 6.554042720794678,
      "epoch": 0.1978099611444719,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998646889013066,
      "loss": 6.3607,
      "mean_token_accuracy": 0.14240839183330536,
      "num_tokens": 4175701.0,
      "step": 2240
    },
    {
      "entropy": 6.4098762512207035,
      "epoch": 0.19825150123631224,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998635947090362,
      "loss": 6.4425,
      "mean_token_accuracy": 0.14030273035168647,
      "num_tokens": 4184711.0,
      "step": 2245
    },
    {
      "entropy": 6.475190782546997,
      "epoch": 0.1986930413281526,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998624961118158,
      "loss": 6.4017,
      "mean_token_accuracy": 0.14099944159388542,
      "num_tokens": 4193931.0,
      "step": 2250
    },
    {
      "entropy": 6.44319109916687,
      "epoch": 0.19913458141999293,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499861393109667,
      "loss": 6.2899,
      "mean_token_accuracy": 0.14828752726316452,
      "num_tokens": 4203000.0,
      "step": 2255
    },
    {
      "entropy": 6.274944400787353,
      "epoch": 0.19957612151183327,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998602857026114,
      "loss": 6.2991,
      "mean_token_accuracy": 0.1458234503865242,
      "num_tokens": 4211977.0,
      "step": 2260
    },
    {
      "entropy": 6.472422647476196,
      "epoch": 0.20001766160367362,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998591738906708,
      "loss": 6.3814,
      "mean_token_accuracy": 0.15182094275951385,
      "num_tokens": 4220375.0,
      "step": 2265
    },
    {
      "entropy": 6.494894886016846,
      "epoch": 0.20045920169551396,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998580576738668,
      "loss": 6.3793,
      "mean_token_accuracy": 0.1367909237742424,
      "num_tokens": 4230506.0,
      "step": 2270
    },
    {
      "entropy": 6.404332447052002,
      "epoch": 0.2009007417873543,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998569370522213,
      "loss": 6.3524,
      "mean_token_accuracy": 0.13977290093898773,
      "num_tokens": 4240235.0,
      "step": 2275
    },
    {
      "entropy": 6.448247480392456,
      "epoch": 0.20134228187919462,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998558120257563,
      "loss": 6.4189,
      "mean_token_accuracy": 0.14339498728513717,
      "num_tokens": 4249775.0,
      "step": 2280
    },
    {
      "entropy": 6.419510173797607,
      "epoch": 0.20178382197103498,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998546825944938,
      "loss": 6.3504,
      "mean_token_accuracy": 0.141066338121891,
      "num_tokens": 4258523.0,
      "step": 2285
    },
    {
      "entropy": 6.419132709503174,
      "epoch": 0.2022253620628753,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499853548758456,
      "loss": 6.358,
      "mean_token_accuracy": 0.13891511633992196,
      "num_tokens": 4267683.0,
      "step": 2290
    },
    {
      "entropy": 6.495290946960449,
      "epoch": 0.20266690215471564,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000499852410517665,
      "loss": 6.417,
      "mean_token_accuracy": 0.13713881745934486,
      "num_tokens": 4276903.0,
      "step": 2295
    },
    {
      "entropy": 6.426017999649048,
      "epoch": 0.203108442246556,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004998512678721431,
      "loss": 6.3661,
      "mean_token_accuracy": 0.135909353941679,
      "num_tokens": 4287257.0,
      "step": 2300
    },
    {
      "entropy": 6.447468948364258,
      "epoch": 0.20354998233839633,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499850120821913,
      "loss": 6.3875,
      "mean_token_accuracy": 0.13645304143428802,
      "num_tokens": 4297345.0,
      "step": 2305
    },
    {
      "entropy": 6.528428220748902,
      "epoch": 0.20399152243023666,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998489693669967,
      "loss": 6.3613,
      "mean_token_accuracy": 0.13837311565876007,
      "num_tokens": 4306533.0,
      "step": 2310
    },
    {
      "entropy": 6.392482471466065,
      "epoch": 0.204433062522077,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499847813507417,
      "loss": 6.4023,
      "mean_token_accuracy": 0.14094773977994918,
      "num_tokens": 4316338.0,
      "step": 2315
    },
    {
      "entropy": 6.399602174758911,
      "epoch": 0.20487460261391735,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998466532431966,
      "loss": 6.3549,
      "mean_token_accuracy": 0.13948202207684518,
      "num_tokens": 4326585.0,
      "step": 2320
    },
    {
      "entropy": 6.544098567962647,
      "epoch": 0.20531614270575768,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998454885743581,
      "loss": 6.4795,
      "mean_token_accuracy": 0.1351695440709591,
      "num_tokens": 4336490.0,
      "step": 2325
    },
    {
      "entropy": 6.363640975952149,
      "epoch": 0.205757682797598,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998443195009242,
      "loss": 6.3589,
      "mean_token_accuracy": 0.1348782531917095,
      "num_tokens": 4346264.0,
      "step": 2330
    },
    {
      "entropy": 6.461185693740845,
      "epoch": 0.20619922288943837,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998431460229182,
      "loss": 6.3318,
      "mean_token_accuracy": 0.1448797807097435,
      "num_tokens": 4355102.0,
      "step": 2335
    },
    {
      "entropy": 6.482104396820068,
      "epoch": 0.2066407629812787,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998419681403627,
      "loss": 6.5133,
      "mean_token_accuracy": 0.13086750581860543,
      "num_tokens": 4365569.0,
      "step": 2340
    },
    {
      "entropy": 6.507464361190796,
      "epoch": 0.20708230307311903,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998407858532809,
      "loss": 6.36,
      "mean_token_accuracy": 0.14296017587184906,
      "num_tokens": 4375437.0,
      "step": 2345
    },
    {
      "entropy": 6.37006025314331,
      "epoch": 0.20752384316495937,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499839599161696,
      "loss": 6.3389,
      "mean_token_accuracy": 0.1450774312019348,
      "num_tokens": 4384420.0,
      "step": 2350
    },
    {
      "entropy": 6.460700035095215,
      "epoch": 0.20796538325679972,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998384080656314,
      "loss": 6.3106,
      "mean_token_accuracy": 0.14050144031643869,
      "num_tokens": 4393730.0,
      "step": 2355
    },
    {
      "entropy": 6.332716035842895,
      "epoch": 0.20840692334864006,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049983721256511,
      "loss": 6.2798,
      "mean_token_accuracy": 0.14351205080747603,
      "num_tokens": 4402731.0,
      "step": 2360
    },
    {
      "entropy": 6.3846518993377686,
      "epoch": 0.2088484634404804,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998360126601556,
      "loss": 6.3369,
      "mean_token_accuracy": 0.14289727210998535,
      "num_tokens": 4411606.0,
      "step": 2365
    },
    {
      "entropy": 6.3337644100189205,
      "epoch": 0.20929000353232075,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998348083507916,
      "loss": 6.4062,
      "mean_token_accuracy": 0.14081210866570473,
      "num_tokens": 4421685.0,
      "step": 2370
    },
    {
      "entropy": 6.558507633209229,
      "epoch": 0.20973154362416108,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998335996370416,
      "loss": 6.3782,
      "mean_token_accuracy": 0.14399294778704644,
      "num_tokens": 4431765.0,
      "step": 2375
    },
    {
      "entropy": 6.321137380599976,
      "epoch": 0.2101730837160014,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998323865189291,
      "loss": 6.3523,
      "mean_token_accuracy": 0.13661579713225364,
      "num_tokens": 4441191.0,
      "step": 2380
    },
    {
      "entropy": 6.481552457809448,
      "epoch": 0.21061462380784174,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998311689964781,
      "loss": 6.4322,
      "mean_token_accuracy": 0.13680859059095382,
      "num_tokens": 4450156.0,
      "step": 2385
    },
    {
      "entropy": 6.4807047843933105,
      "epoch": 0.2110561638996821,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998299470697125,
      "loss": 6.4163,
      "mean_token_accuracy": 0.14260546639561653,
      "num_tokens": 4459466.0,
      "step": 2390
    },
    {
      "entropy": 6.461113500595093,
      "epoch": 0.21149770399152243,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998287207386559,
      "loss": 6.4156,
      "mean_token_accuracy": 0.14247968047857285,
      "num_tokens": 4468539.0,
      "step": 2395
    },
    {
      "entropy": 6.525749444961548,
      "epoch": 0.21193924408336276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998274900033326,
      "loss": 6.315,
      "mean_token_accuracy": 0.14835015684366226,
      "num_tokens": 4477579.0,
      "step": 2400
    },
    {
      "entropy": 6.236664247512818,
      "epoch": 0.21238078417520312,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998262548637667,
      "loss": 6.2842,
      "mean_token_accuracy": 0.14991160482168198,
      "num_tokens": 4486800.0,
      "step": 2405
    },
    {
      "entropy": 6.444223546981812,
      "epoch": 0.21282232426704345,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998250153199822,
      "loss": 6.2465,
      "mean_token_accuracy": 0.14562757611274718,
      "num_tokens": 4495985.0,
      "step": 2410
    },
    {
      "entropy": 6.417825555801391,
      "epoch": 0.21326386435888378,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998237713720036,
      "loss": 6.4031,
      "mean_token_accuracy": 0.14113787487149237,
      "num_tokens": 4504944.0,
      "step": 2415
    },
    {
      "entropy": 6.344033908843994,
      "epoch": 0.2137054044507241,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998225230198552,
      "loss": 6.2875,
      "mean_token_accuracy": 0.14928205609321593,
      "num_tokens": 4515402.0,
      "step": 2420
    },
    {
      "entropy": 6.412223052978516,
      "epoch": 0.21414694454256447,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998212702635614,
      "loss": 6.355,
      "mean_token_accuracy": 0.14224686175584794,
      "num_tokens": 4525009.0,
      "step": 2425
    },
    {
      "entropy": 6.537710332870484,
      "epoch": 0.2145884846344048,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998200131031469,
      "loss": 6.4066,
      "mean_token_accuracy": 0.142412006855011,
      "num_tokens": 4534460.0,
      "step": 2430
    },
    {
      "entropy": 6.355252885818482,
      "epoch": 0.21503002472624513,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998187515386361,
      "loss": 6.2239,
      "mean_token_accuracy": 0.15292632952332497,
      "num_tokens": 4543748.0,
      "step": 2435
    },
    {
      "entropy": 6.32159743309021,
      "epoch": 0.2154715648180855,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998174855700538,
      "loss": 6.3515,
      "mean_token_accuracy": 0.14536840543150903,
      "num_tokens": 4552722.0,
      "step": 2440
    },
    {
      "entropy": 6.455835342407227,
      "epoch": 0.21591310490992582,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998162151974248,
      "loss": 6.2244,
      "mean_token_accuracy": 0.1433302193880081,
      "num_tokens": 4561607.0,
      "step": 2445
    },
    {
      "entropy": 6.376346635818481,
      "epoch": 0.21635464500176615,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000499814940420774,
      "loss": 6.4926,
      "mean_token_accuracy": 0.13043315410614015,
      "num_tokens": 4572524.0,
      "step": 2450
    },
    {
      "entropy": 6.473890399932861,
      "epoch": 0.21679618509360649,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998136612401266,
      "loss": 6.306,
      "mean_token_accuracy": 0.13808697760105132,
      "num_tokens": 4581601.0,
      "step": 2455
    },
    {
      "entropy": 6.362298917770386,
      "epoch": 0.21723772518544684,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998123776555071,
      "loss": 6.355,
      "mean_token_accuracy": 0.13434374257922171,
      "num_tokens": 4591795.0,
      "step": 2460
    },
    {
      "entropy": 6.4316198348999025,
      "epoch": 0.21767926527728718,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998110896669412,
      "loss": 6.3127,
      "mean_token_accuracy": 0.14232389852404595,
      "num_tokens": 4600745.0,
      "step": 2465
    },
    {
      "entropy": 6.403714847564697,
      "epoch": 0.2181208053691275,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998097972744539,
      "loss": 6.3668,
      "mean_token_accuracy": 0.13958390951156616,
      "num_tokens": 4610490.0,
      "step": 2470
    },
    {
      "entropy": 6.421542739868164,
      "epoch": 0.21856234546096787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998085004780705,
      "loss": 6.3027,
      "mean_token_accuracy": 0.14644000679254532,
      "num_tokens": 4619511.0,
      "step": 2475
    },
    {
      "entropy": 6.411759996414185,
      "epoch": 0.2190038855528082,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998071992778164,
      "loss": 6.3536,
      "mean_token_accuracy": 0.13926490917801856,
      "num_tokens": 4628186.0,
      "step": 2480
    },
    {
      "entropy": 6.416973400115967,
      "epoch": 0.21944542564464853,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499805893673717,
      "loss": 6.3574,
      "mean_token_accuracy": 0.14341954439878463,
      "num_tokens": 4637431.0,
      "step": 2485
    },
    {
      "entropy": 6.347478723526001,
      "epoch": 0.21988696573648886,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998045836657982,
      "loss": 6.2093,
      "mean_token_accuracy": 0.14213306605815887,
      "num_tokens": 4646627.0,
      "step": 2490
    },
    {
      "entropy": 6.368200349807739,
      "epoch": 0.22032850582832922,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998032692540853,
      "loss": 6.3012,
      "mean_token_accuracy": 0.14045739471912383,
      "num_tokens": 4656095.0,
      "step": 2495
    },
    {
      "entropy": 6.476345205307007,
      "epoch": 0.22077004592016955,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998019504386044,
      "loss": 6.3876,
      "mean_token_accuracy": 0.13640450164675713,
      "num_tokens": 4665807.0,
      "step": 2500
    },
    {
      "entropy": 6.444914436340332,
      "epoch": 0.22121158601200988,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998006272193809,
      "loss": 6.3493,
      "mean_token_accuracy": 0.13855071663856505,
      "num_tokens": 4674459.0,
      "step": 2505
    },
    {
      "entropy": 6.361084604263306,
      "epoch": 0.22165312610385024,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997992995964412,
      "loss": 6.4325,
      "mean_token_accuracy": 0.13779560700058938,
      "num_tokens": 4684063.0,
      "step": 2510
    },
    {
      "entropy": 6.45024299621582,
      "epoch": 0.22209466619569057,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997979675698109,
      "loss": 6.3029,
      "mean_token_accuracy": 0.14212062656879426,
      "num_tokens": 4692807.0,
      "step": 2515
    },
    {
      "entropy": 6.438331031799317,
      "epoch": 0.2225362062875309,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997966311395164,
      "loss": 6.2422,
      "mean_token_accuracy": 0.14746622294187545,
      "num_tokens": 4701100.0,
      "step": 2520
    },
    {
      "entropy": 6.284497547149658,
      "epoch": 0.22297774637937123,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997952903055836,
      "loss": 6.3071,
      "mean_token_accuracy": 0.13971827551722527,
      "num_tokens": 4710697.0,
      "step": 2525
    },
    {
      "entropy": 6.4242840766906735,
      "epoch": 0.2234192864712116,
      "grad_norm": 1.125,
      "learning_rate": 0.000499793945068039,
      "loss": 6.2875,
      "mean_token_accuracy": 0.14202770590782166,
      "num_tokens": 4718745.0,
      "step": 2530
    },
    {
      "entropy": 6.358328342437744,
      "epoch": 0.22386082656305192,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997925954269088,
      "loss": 6.2493,
      "mean_token_accuracy": 0.15010830983519555,
      "num_tokens": 4728056.0,
      "step": 2535
    },
    {
      "entropy": 6.390234851837159,
      "epoch": 0.22430236665489225,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997912413822196,
      "loss": 6.3892,
      "mean_token_accuracy": 0.14226726815104485,
      "num_tokens": 4737605.0,
      "step": 2540
    },
    {
      "entropy": 6.333422613143921,
      "epoch": 0.2247439067467326,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997898829339979,
      "loss": 6.216,
      "mean_token_accuracy": 0.15019772350788116,
      "num_tokens": 4746168.0,
      "step": 2545
    },
    {
      "entropy": 6.414393615722656,
      "epoch": 0.22518544683857294,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049978852008227,
      "loss": 6.2827,
      "mean_token_accuracy": 0.14390757903456688,
      "num_tokens": 4755072.0,
      "step": 2550
    },
    {
      "entropy": 6.292112064361572,
      "epoch": 0.22562698693041328,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499787152827063,
      "loss": 6.2887,
      "mean_token_accuracy": 0.14927180036902427,
      "num_tokens": 4764580.0,
      "step": 2555
    },
    {
      "entropy": 6.362084197998047,
      "epoch": 0.2260685270222536,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997857811684035,
      "loss": 6.3109,
      "mean_token_accuracy": 0.14619807451963424,
      "num_tokens": 4774135.0,
      "step": 2560
    },
    {
      "entropy": 6.443705081939697,
      "epoch": 0.22651006711409397,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997844051063183,
      "loss": 6.3931,
      "mean_token_accuracy": 0.1407366193830967,
      "num_tokens": 4784733.0,
      "step": 2565
    },
    {
      "entropy": 6.393408250808716,
      "epoch": 0.2269516072059343,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997830246408346,
      "loss": 6.3304,
      "mean_token_accuracy": 0.14791636019945145,
      "num_tokens": 4795327.0,
      "step": 2570
    },
    {
      "entropy": 6.342170095443725,
      "epoch": 0.22739314729777463,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997816397719791,
      "loss": 6.3028,
      "mean_token_accuracy": 0.14114121049642564,
      "num_tokens": 4804314.0,
      "step": 2575
    },
    {
      "entropy": 6.458865118026734,
      "epoch": 0.227834687389615,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997802504997792,
      "loss": 6.3913,
      "mean_token_accuracy": 0.13652418628335,
      "num_tokens": 4813637.0,
      "step": 2580
    },
    {
      "entropy": 6.382553291320801,
      "epoch": 0.22827622748145532,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997788568242621,
      "loss": 6.2591,
      "mean_token_accuracy": 0.14338775500655174,
      "num_tokens": 4823094.0,
      "step": 2585
    },
    {
      "entropy": 6.327674865722656,
      "epoch": 0.22871776757329565,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499777458745455,
      "loss": 6.1969,
      "mean_token_accuracy": 0.1477431207895279,
      "num_tokens": 4833199.0,
      "step": 2590
    },
    {
      "entropy": 6.392719554901123,
      "epoch": 0.22915930766513598,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997760562633853,
      "loss": 6.2909,
      "mean_token_accuracy": 0.14219107255339622,
      "num_tokens": 4842970.0,
      "step": 2595
    },
    {
      "entropy": 6.394578695297241,
      "epoch": 0.22960084775697634,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997746493780804,
      "loss": 6.3788,
      "mean_token_accuracy": 0.13738251477479935,
      "num_tokens": 4852043.0,
      "step": 2600
    },
    {
      "entropy": 6.376160097122193,
      "epoch": 0.23004238784881667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499773238089568,
      "loss": 6.2693,
      "mean_token_accuracy": 0.13766007199883462,
      "num_tokens": 4862232.0,
      "step": 2605
    },
    {
      "entropy": 6.345012950897217,
      "epoch": 0.230483927940657,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997718223978758,
      "loss": 6.2081,
      "mean_token_accuracy": 0.1466532751917839,
      "num_tokens": 4871186.0,
      "step": 2610
    },
    {
      "entropy": 6.343753385543823,
      "epoch": 0.23092546803249736,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997704023030315,
      "loss": 6.3059,
      "mean_token_accuracy": 0.150559451431036,
      "num_tokens": 4879974.0,
      "step": 2615
    },
    {
      "entropy": 6.393126726150513,
      "epoch": 0.2313670081243377,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997689778050627,
      "loss": 6.3617,
      "mean_token_accuracy": 0.14292784333229064,
      "num_tokens": 4890300.0,
      "step": 2620
    },
    {
      "entropy": 6.369229030609131,
      "epoch": 0.23180854821617802,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997675489039975,
      "loss": 6.3301,
      "mean_token_accuracy": 0.1426799289882183,
      "num_tokens": 4900428.0,
      "step": 2625
    },
    {
      "entropy": 6.3974145412445065,
      "epoch": 0.23225008830801835,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997661155998638,
      "loss": 6.3245,
      "mean_token_accuracy": 0.1442883849143982,
      "num_tokens": 4910092.0,
      "step": 2630
    },
    {
      "entropy": 6.384046411514282,
      "epoch": 0.2326916283998587,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997646778926898,
      "loss": 6.3247,
      "mean_token_accuracy": 0.13738622814416884,
      "num_tokens": 4919593.0,
      "step": 2635
    },
    {
      "entropy": 6.373585891723633,
      "epoch": 0.23313316849169904,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997632357825035,
      "loss": 6.3001,
      "mean_token_accuracy": 0.14202155098319053,
      "num_tokens": 4929098.0,
      "step": 2640
    },
    {
      "entropy": 6.43012547492981,
      "epoch": 0.23357470858353938,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997617892693333,
      "loss": 6.3657,
      "mean_token_accuracy": 0.1421157017350197,
      "num_tokens": 4938265.0,
      "step": 2645
    },
    {
      "entropy": 6.394748878479004,
      "epoch": 0.23401624867537973,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997603383532075,
      "loss": 6.3016,
      "mean_token_accuracy": 0.14679210409522056,
      "num_tokens": 4946694.0,
      "step": 2650
    },
    {
      "entropy": 6.368328046798706,
      "epoch": 0.23445778876722007,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997588830341545,
      "loss": 6.3132,
      "mean_token_accuracy": 0.1434150867164135,
      "num_tokens": 4955296.0,
      "step": 2655
    },
    {
      "entropy": 6.32787938117981,
      "epoch": 0.2348993288590604,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997574233122028,
      "loss": 6.2759,
      "mean_token_accuracy": 0.14597226828336715,
      "num_tokens": 4964409.0,
      "step": 2660
    },
    {
      "entropy": 6.3884584426879885,
      "epoch": 0.23534086895090076,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997559591873809,
      "loss": 6.284,
      "mean_token_accuracy": 0.1481903851032257,
      "num_tokens": 4973449.0,
      "step": 2665
    },
    {
      "entropy": 6.345009279251099,
      "epoch": 0.2357824090427411,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997544906597178,
      "loss": 6.2779,
      "mean_token_accuracy": 0.1470661997795105,
      "num_tokens": 4983057.0,
      "step": 2670
    },
    {
      "entropy": 6.2613893985748295,
      "epoch": 0.23622394913458142,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004997530177292418,
      "loss": 6.3532,
      "mean_token_accuracy": 0.13861697241663934,
      "num_tokens": 4991950.0,
      "step": 2675
    },
    {
      "entropy": 6.461032247543335,
      "epoch": 0.23666548922642175,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997515403959823,
      "loss": 6.2857,
      "mean_token_accuracy": 0.14603266417980193,
      "num_tokens": 5001042.0,
      "step": 2680
    },
    {
      "entropy": 6.356680679321289,
      "epoch": 0.2371070293182621,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997500586599677,
      "loss": 6.2198,
      "mean_token_accuracy": 0.15022996366024016,
      "num_tokens": 5009827.0,
      "step": 2685
    },
    {
      "entropy": 6.292784547805786,
      "epoch": 0.23754856941010244,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997485725212274,
      "loss": 6.2662,
      "mean_token_accuracy": 0.1465997129678726,
      "num_tokens": 5018708.0,
      "step": 2690
    },
    {
      "entropy": 6.334398937225342,
      "epoch": 0.23799010950194277,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997470819797903,
      "loss": 6.1678,
      "mean_token_accuracy": 0.149826068431139,
      "num_tokens": 5027522.0,
      "step": 2695
    },
    {
      "entropy": 6.312096786499024,
      "epoch": 0.23843164959378313,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997455870356857,
      "loss": 6.2858,
      "mean_token_accuracy": 0.14754335582256317,
      "num_tokens": 5035755.0,
      "step": 2700
    },
    {
      "entropy": 6.38718318939209,
      "epoch": 0.23887318968562346,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997440876889429,
      "loss": 6.2373,
      "mean_token_accuracy": 0.14902258217334746,
      "num_tokens": 5045289.0,
      "step": 2705
    },
    {
      "entropy": 6.2602025985717775,
      "epoch": 0.2393147297774638,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997425839395913,
      "loss": 6.2623,
      "mean_token_accuracy": 0.14851141721010208,
      "num_tokens": 5053774.0,
      "step": 2710
    },
    {
      "entropy": 6.4446056365966795,
      "epoch": 0.23975626986930412,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997410757876602,
      "loss": 6.3368,
      "mean_token_accuracy": 0.13768139705061913,
      "num_tokens": 5062911.0,
      "step": 2715
    },
    {
      "entropy": 6.34313178062439,
      "epoch": 0.24019780996114448,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997395632331793,
      "loss": 6.1974,
      "mean_token_accuracy": 0.15056394785642624,
      "num_tokens": 5072107.0,
      "step": 2720
    },
    {
      "entropy": 6.232982730865478,
      "epoch": 0.2406393500529848,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997380462761781,
      "loss": 6.1744,
      "mean_token_accuracy": 0.15013156086206436,
      "num_tokens": 5080588.0,
      "step": 2725
    },
    {
      "entropy": 6.360136985778809,
      "epoch": 0.24108089014482514,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997365249166864,
      "loss": 6.3571,
      "mean_token_accuracy": 0.1455472856760025,
      "num_tokens": 5090262.0,
      "step": 2730
    },
    {
      "entropy": 6.398047304153442,
      "epoch": 0.2415224302366655,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997349991547342,
      "loss": 6.2776,
      "mean_token_accuracy": 0.15021264627575875,
      "num_tokens": 5099285.0,
      "step": 2735
    },
    {
      "entropy": 6.356108903884888,
      "epoch": 0.24196397032850583,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997334689903509,
      "loss": 6.3226,
      "mean_token_accuracy": 0.14855852872133254,
      "num_tokens": 5109115.0,
      "step": 2740
    },
    {
      "entropy": 6.388672256469727,
      "epoch": 0.24240551042034617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004997319344235668,
      "loss": 6.3429,
      "mean_token_accuracy": 0.14180680066347123,
      "num_tokens": 5117977.0,
      "step": 2745
    },
    {
      "entropy": 6.393019914627075,
      "epoch": 0.2428470505121865,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499730395454412,
      "loss": 6.3025,
      "mean_token_accuracy": 0.1451731264591217,
      "num_tokens": 5127457.0,
      "step": 2750
    },
    {
      "entropy": 6.300089502334595,
      "epoch": 0.24328859060402686,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997288520829166,
      "loss": 6.3466,
      "mean_token_accuracy": 0.14065672382712363,
      "num_tokens": 5137310.0,
      "step": 2755
    },
    {
      "entropy": 6.394161605834961,
      "epoch": 0.2437301306958672,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997273043091107,
      "loss": 6.2725,
      "mean_token_accuracy": 0.14155926927924156,
      "num_tokens": 5146963.0,
      "step": 2760
    },
    {
      "entropy": 6.31079797744751,
      "epoch": 0.24417167078770752,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997257521330248,
      "loss": 6.2601,
      "mean_token_accuracy": 0.14022860154509545,
      "num_tokens": 5155521.0,
      "step": 2765
    },
    {
      "entropy": 6.40408935546875,
      "epoch": 0.24461321087954788,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997241955546892,
      "loss": 6.237,
      "mean_token_accuracy": 0.14580907300114632,
      "num_tokens": 5165182.0,
      "step": 2770
    },
    {
      "entropy": 6.289978647232056,
      "epoch": 0.2450547509713882,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997226345741343,
      "loss": 6.2649,
      "mean_token_accuracy": 0.13975519686937332,
      "num_tokens": 5175511.0,
      "step": 2775
    },
    {
      "entropy": 6.354159593582153,
      "epoch": 0.24549629106322854,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499721069191391,
      "loss": 6.2126,
      "mean_token_accuracy": 0.14699607565999032,
      "num_tokens": 5184772.0,
      "step": 2780
    },
    {
      "entropy": 6.288126516342163,
      "epoch": 0.24593783115506887,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997194994064896,
      "loss": 6.3014,
      "mean_token_accuracy": 0.14003223031759263,
      "num_tokens": 5195136.0,
      "step": 2785
    },
    {
      "entropy": 6.391946315765381,
      "epoch": 0.24637937124690923,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499717925219461,
      "loss": 6.2279,
      "mean_token_accuracy": 0.15093553364276885,
      "num_tokens": 5203163.0,
      "step": 2790
    },
    {
      "entropy": 6.346267318725586,
      "epoch": 0.24682091133874956,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997163466303362,
      "loss": 6.3224,
      "mean_token_accuracy": 0.14383373707532882,
      "num_tokens": 5213233.0,
      "step": 2795
    },
    {
      "entropy": 6.329722881317139,
      "epoch": 0.2472624514305899,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499714763639146,
      "loss": 6.2346,
      "mean_token_accuracy": 0.14386766105890275,
      "num_tokens": 5222940.0,
      "step": 2800
    },
    {
      "entropy": 6.36411566734314,
      "epoch": 0.24770399152243025,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997131762459211,
      "loss": 6.2596,
      "mean_token_accuracy": 0.14432956129312516,
      "num_tokens": 5232263.0,
      "step": 2805
    },
    {
      "entropy": 6.369926023483276,
      "epoch": 0.24814553161427058,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997115844506932,
      "loss": 6.2334,
      "mean_token_accuracy": 0.14295720756053926,
      "num_tokens": 5241536.0,
      "step": 2810
    },
    {
      "entropy": 6.348195028305054,
      "epoch": 0.2485870717061109,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997099882534929,
      "loss": 6.2732,
      "mean_token_accuracy": 0.14211497604846954,
      "num_tokens": 5250702.0,
      "step": 2815
    },
    {
      "entropy": 6.359058141708374,
      "epoch": 0.24902861179795124,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997083876543519,
      "loss": 6.2763,
      "mean_token_accuracy": 0.14498503208160402,
      "num_tokens": 5259811.0,
      "step": 2820
    },
    {
      "entropy": 6.397128582000732,
      "epoch": 0.2494701518897916,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997067826533014,
      "loss": 6.3615,
      "mean_token_accuracy": 0.13723283037543296,
      "num_tokens": 5270518.0,
      "step": 2825
    },
    {
      "entropy": 6.375804328918457,
      "epoch": 0.24991169198163193,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004997051732503726,
      "loss": 6.2458,
      "mean_token_accuracy": 0.14747673273086548,
      "num_tokens": 5279538.0,
      "step": 2830
    },
    {
      "entropy": 6.313772678375244,
      "epoch": 0.25035323207347226,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997035594455975,
      "loss": 6.2702,
      "mean_token_accuracy": 0.13872402533888817,
      "num_tokens": 5289633.0,
      "step": 2835
    },
    {
      "entropy": 6.370833015441894,
      "epoch": 0.2507947721653126,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997019412390074,
      "loss": 6.3603,
      "mean_token_accuracy": 0.1444901555776596,
      "num_tokens": 5299148.0,
      "step": 2840
    },
    {
      "entropy": 6.36065092086792,
      "epoch": 0.2512363122571529,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499700318630634,
      "loss": 6.2527,
      "mean_token_accuracy": 0.1457889422774315,
      "num_tokens": 5309090.0,
      "step": 2845
    },
    {
      "entropy": 6.384716939926148,
      "epoch": 0.2516778523489933,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996986916205092,
      "loss": 6.3297,
      "mean_token_accuracy": 0.14129810705780982,
      "num_tokens": 5318798.0,
      "step": 2850
    },
    {
      "entropy": 6.313976621627807,
      "epoch": 0.25211939244083365,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996970602086648,
      "loss": 6.1848,
      "mean_token_accuracy": 0.15023760497570038,
      "num_tokens": 5327915.0,
      "step": 2855
    },
    {
      "entropy": 6.251888847351074,
      "epoch": 0.252560932532674,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996954243951327,
      "loss": 6.2192,
      "mean_token_accuracy": 0.15385923832654952,
      "num_tokens": 5336970.0,
      "step": 2860
    },
    {
      "entropy": 6.299870014190674,
      "epoch": 0.2530024726245143,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996937841799451,
      "loss": 6.1821,
      "mean_token_accuracy": 0.15226729065179825,
      "num_tokens": 5345167.0,
      "step": 2865
    },
    {
      "entropy": 6.1627833366394045,
      "epoch": 0.25344401271635464,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004996921395631342,
      "loss": 6.1711,
      "mean_token_accuracy": 0.14804726019501685,
      "num_tokens": 5353399.0,
      "step": 2870
    },
    {
      "entropy": 6.380198335647583,
      "epoch": 0.25388555280819497,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499690490544732,
      "loss": 6.285,
      "mean_token_accuracy": 0.146748573333025,
      "num_tokens": 5363224.0,
      "step": 2875
    },
    {
      "entropy": 6.372404766082764,
      "epoch": 0.2543270929000353,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996888371247707,
      "loss": 6.2862,
      "mean_token_accuracy": 0.14108646661043167,
      "num_tokens": 5372274.0,
      "step": 2880
    },
    {
      "entropy": 6.302175998687744,
      "epoch": 0.2547686329918757,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499687179303283,
      "loss": 6.2746,
      "mean_token_accuracy": 0.15104661732912064,
      "num_tokens": 5380240.0,
      "step": 2885
    },
    {
      "entropy": 6.269204902648926,
      "epoch": 0.255210173083716,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996855170803012,
      "loss": 6.138,
      "mean_token_accuracy": 0.15041064321994782,
      "num_tokens": 5389390.0,
      "step": 2890
    },
    {
      "entropy": 6.343308639526367,
      "epoch": 0.25565171317555635,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996838504558581,
      "loss": 6.2986,
      "mean_token_accuracy": 0.14492825120687486,
      "num_tokens": 5399425.0,
      "step": 2895
    },
    {
      "entropy": 6.379653215408325,
      "epoch": 0.2560932532673967,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499682179429986,
      "loss": 6.3089,
      "mean_token_accuracy": 0.14139388352632523,
      "num_tokens": 5408717.0,
      "step": 2900
    },
    {
      "entropy": 6.292103576660156,
      "epoch": 0.256534793359237,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996805040027178,
      "loss": 6.2399,
      "mean_token_accuracy": 0.1403766691684723,
      "num_tokens": 5418475.0,
      "step": 2905
    },
    {
      "entropy": 6.395513296127319,
      "epoch": 0.25697633345107734,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996788241740863,
      "loss": 6.2884,
      "mean_token_accuracy": 0.143946073949337,
      "num_tokens": 5428403.0,
      "step": 2910
    },
    {
      "entropy": 6.366812467575073,
      "epoch": 0.2574178735429177,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996771399441243,
      "loss": 6.3188,
      "mean_token_accuracy": 0.14125285297632217,
      "num_tokens": 5437347.0,
      "step": 2915
    },
    {
      "entropy": 6.395107555389404,
      "epoch": 0.25785941363475806,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996754513128652,
      "loss": 6.2216,
      "mean_token_accuracy": 0.1553879424929619,
      "num_tokens": 5446804.0,
      "step": 2920
    },
    {
      "entropy": 6.245992279052734,
      "epoch": 0.2583009537265984,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996737582803416,
      "loss": 6.1701,
      "mean_token_accuracy": 0.14774591475725174,
      "num_tokens": 5455888.0,
      "step": 2925
    },
    {
      "entropy": 6.349690961837768,
      "epoch": 0.2587424938184387,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996720608465868,
      "loss": 6.1785,
      "mean_token_accuracy": 0.14454589933156967,
      "num_tokens": 5463977.0,
      "step": 2930
    },
    {
      "entropy": 6.251695680618286,
      "epoch": 0.25918403391027905,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996703590116342,
      "loss": 6.2901,
      "mean_token_accuracy": 0.1413638859987259,
      "num_tokens": 5473780.0,
      "step": 2935
    },
    {
      "entropy": 6.343139219284057,
      "epoch": 0.2596255740021194,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996686527755171,
      "loss": 6.1747,
      "mean_token_accuracy": 0.15054369121789932,
      "num_tokens": 5482151.0,
      "step": 2940
    },
    {
      "entropy": 6.287330961227417,
      "epoch": 0.2600671140939597,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996669421382687,
      "loss": 6.181,
      "mean_token_accuracy": 0.15408090725541115,
      "num_tokens": 5491103.0,
      "step": 2945
    },
    {
      "entropy": 6.23843822479248,
      "epoch": 0.26050865418580005,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996652270999228,
      "loss": 6.2051,
      "mean_token_accuracy": 0.1455566719174385,
      "num_tokens": 5500367.0,
      "step": 2950
    },
    {
      "entropy": 6.401996898651123,
      "epoch": 0.26095019427764043,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996635076605128,
      "loss": 6.2392,
      "mean_token_accuracy": 0.1509515941143036,
      "num_tokens": 5509631.0,
      "step": 2955
    },
    {
      "entropy": 6.3384003162384035,
      "epoch": 0.26139173436948077,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996617838200725,
      "loss": 6.2572,
      "mean_token_accuracy": 0.14331620335578918,
      "num_tokens": 5518635.0,
      "step": 2960
    },
    {
      "entropy": 6.241027069091797,
      "epoch": 0.2618332744613211,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996600555786357,
      "loss": 6.2142,
      "mean_token_accuracy": 0.1464727446436882,
      "num_tokens": 5527696.0,
      "step": 2965
    },
    {
      "entropy": 6.348132085800171,
      "epoch": 0.26227481455316143,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996583229362362,
      "loss": 6.1834,
      "mean_token_accuracy": 0.14780823439359664,
      "num_tokens": 5536632.0,
      "step": 2970
    },
    {
      "entropy": 6.378821849822998,
      "epoch": 0.26271635464500176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996565858929078,
      "loss": 6.2627,
      "mean_token_accuracy": 0.14528179541230202,
      "num_tokens": 5545825.0,
      "step": 2975
    },
    {
      "entropy": 6.259585618972778,
      "epoch": 0.2631578947368421,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996548444486847,
      "loss": 6.1389,
      "mean_token_accuracy": 0.15060140788555146,
      "num_tokens": 5555158.0,
      "step": 2980
    },
    {
      "entropy": 6.116889953613281,
      "epoch": 0.2635994348286824,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996530986036008,
      "loss": 6.0795,
      "mean_token_accuracy": 0.15272270664572715,
      "num_tokens": 5564218.0,
      "step": 2985
    },
    {
      "entropy": 6.255494451522827,
      "epoch": 0.2640409749205228,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996513483576907,
      "loss": 6.2219,
      "mean_token_accuracy": 0.14951637461781503,
      "num_tokens": 5572760.0,
      "step": 2990
    },
    {
      "entropy": 6.423755645751953,
      "epoch": 0.26448251501236314,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996495937109884,
      "loss": 6.2825,
      "mean_token_accuracy": 0.14191085398197173,
      "num_tokens": 5581660.0,
      "step": 2995
    },
    {
      "entropy": 6.250067615509034,
      "epoch": 0.26492405510420347,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996478346635283,
      "loss": 6.1968,
      "mean_token_accuracy": 0.1436440147459507,
      "num_tokens": 5590664.0,
      "step": 3000
    },
    {
      "epoch": 0.26492405510420347,
      "eval_entropy": 6.077911184598204,
      "eval_loss": 6.2711029052734375,
      "eval_mean_token_accuracy": 0.15016848111384373,
      "eval_num_tokens": 5590664.0,
      "eval_runtime": 26.2453,
      "eval_samples_per_second": 1345.574,
      "eval_steps_per_second": 168.221,
      "step": 3000
    },
    {
      "entropy": 6.211904573440552,
      "epoch": 0.2653655951960438,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996460712153448,
      "loss": 6.0603,
      "mean_token_accuracy": 0.1621384307742119,
      "num_tokens": 5598727.0,
      "step": 3005
    },
    {
      "entropy": 6.220605993270874,
      "epoch": 0.26580713528788413,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004996443033664726,
      "loss": 6.2746,
      "mean_token_accuracy": 0.1402523137629032,
      "num_tokens": 5608549.0,
      "step": 3010
    },
    {
      "entropy": 6.372675228118896,
      "epoch": 0.26624867537972446,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996425311169463,
      "loss": 6.1935,
      "mean_token_accuracy": 0.15231838524341584,
      "num_tokens": 5617628.0,
      "step": 3015
    },
    {
      "entropy": 6.362924861907959,
      "epoch": 0.2666902154715648,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996407544668005,
      "loss": 6.2649,
      "mean_token_accuracy": 0.14236303716897963,
      "num_tokens": 5627182.0,
      "step": 3020
    },
    {
      "entropy": 6.325074291229248,
      "epoch": 0.2671317555634052,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996389734160701,
      "loss": 6.293,
      "mean_token_accuracy": 0.14348414838314055,
      "num_tokens": 5636413.0,
      "step": 3025
    },
    {
      "entropy": 6.389656829833984,
      "epoch": 0.2675732956552455,
      "grad_norm": 1.078125,
      "learning_rate": 0.00049963718796479,
      "loss": 6.2355,
      "mean_token_accuracy": 0.14448917284607887,
      "num_tokens": 5646025.0,
      "step": 3030
    },
    {
      "entropy": 6.272923135757447,
      "epoch": 0.26801483574708584,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996353981129952,
      "loss": 6.2316,
      "mean_token_accuracy": 0.15121424347162246,
      "num_tokens": 5656577.0,
      "step": 3035
    },
    {
      "entropy": 6.341066169738769,
      "epoch": 0.2684563758389262,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996336038607206,
      "loss": 6.2194,
      "mean_token_accuracy": 0.1483485922217369,
      "num_tokens": 5665163.0,
      "step": 3040
    },
    {
      "entropy": 6.198571634292603,
      "epoch": 0.2688979159307665,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996318052080015,
      "loss": 6.2208,
      "mean_token_accuracy": 0.14988780170679092,
      "num_tokens": 5675196.0,
      "step": 3045
    },
    {
      "entropy": 6.395411729812622,
      "epoch": 0.26933945602260684,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996300021548731,
      "loss": 6.163,
      "mean_token_accuracy": 0.14734217673540115,
      "num_tokens": 5684361.0,
      "step": 3050
    },
    {
      "entropy": 6.142444658279419,
      "epoch": 0.26978099611444717,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996281947013707,
      "loss": 6.1584,
      "mean_token_accuracy": 0.1533804029226303,
      "num_tokens": 5694187.0,
      "step": 3055
    },
    {
      "entropy": 6.307489728927612,
      "epoch": 0.27022253620628756,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004996263828475298,
      "loss": 6.2235,
      "mean_token_accuracy": 0.1459271177649498,
      "num_tokens": 5702858.0,
      "step": 3060
    },
    {
      "entropy": 6.287201023101806,
      "epoch": 0.2706640762981279,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996245665933857,
      "loss": 6.124,
      "mean_token_accuracy": 0.1559446483850479,
      "num_tokens": 5712727.0,
      "step": 3065
    },
    {
      "entropy": 6.197343635559082,
      "epoch": 0.2711056163899682,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996227459389741,
      "loss": 6.2573,
      "mean_token_accuracy": 0.14863042607903482,
      "num_tokens": 5722904.0,
      "step": 3070
    },
    {
      "entropy": 6.400126695632935,
      "epoch": 0.27154715648180855,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996209208843307,
      "loss": 6.3578,
      "mean_token_accuracy": 0.14644555673003196,
      "num_tokens": 5732228.0,
      "step": 3075
    },
    {
      "entropy": 6.322457504272461,
      "epoch": 0.2719886965736489,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996190914294912,
      "loss": 6.2574,
      "mean_token_accuracy": 0.1481010966002941,
      "num_tokens": 5743548.0,
      "step": 3080
    },
    {
      "entropy": 6.271878862380982,
      "epoch": 0.2724302366654892,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996172575744914,
      "loss": 6.2867,
      "mean_token_accuracy": 0.13876855000853539,
      "num_tokens": 5754112.0,
      "step": 3085
    },
    {
      "entropy": 6.337986612319947,
      "epoch": 0.27287177675732954,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996154193193673,
      "loss": 6.0993,
      "mean_token_accuracy": 0.15423450618982315,
      "num_tokens": 5763377.0,
      "step": 3090
    },
    {
      "entropy": 6.137074375152588,
      "epoch": 0.27331331684916993,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996135766641549,
      "loss": 6.0806,
      "mean_token_accuracy": 0.15594624429941178,
      "num_tokens": 5772000.0,
      "step": 3095
    },
    {
      "entropy": 6.32679500579834,
      "epoch": 0.27375485694101026,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996117296088903,
      "loss": 6.2131,
      "mean_token_accuracy": 0.14269956201314926,
      "num_tokens": 5782192.0,
      "step": 3100
    },
    {
      "entropy": 6.341950845718384,
      "epoch": 0.2741963970328506,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996098781536095,
      "loss": 6.2614,
      "mean_token_accuracy": 0.1522984981536865,
      "num_tokens": 5791163.0,
      "step": 3105
    },
    {
      "entropy": 6.3308337211608885,
      "epoch": 0.2746379371246909,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996080222983492,
      "loss": 6.259,
      "mean_token_accuracy": 0.14087159857153891,
      "num_tokens": 5801815.0,
      "step": 3110
    },
    {
      "entropy": 6.410363006591797,
      "epoch": 0.27507947721653125,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996061620431453,
      "loss": 6.2805,
      "mean_token_accuracy": 0.1418459102511406,
      "num_tokens": 5811545.0,
      "step": 3115
    },
    {
      "entropy": 6.3260101795196535,
      "epoch": 0.2755210173083716,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996042973880344,
      "loss": 6.2846,
      "mean_token_accuracy": 0.14676327556371688,
      "num_tokens": 5820285.0,
      "step": 3120
    },
    {
      "entropy": 6.260831451416015,
      "epoch": 0.2759625574002119,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996024283330532,
      "loss": 6.1242,
      "mean_token_accuracy": 0.14824864715337754,
      "num_tokens": 5829284.0,
      "step": 3125
    },
    {
      "entropy": 6.268897342681885,
      "epoch": 0.2764040974920523,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499600554878238,
      "loss": 6.2951,
      "mean_token_accuracy": 0.1431872047483921,
      "num_tokens": 5838973.0,
      "step": 3130
    },
    {
      "entropy": 6.278724241256714,
      "epoch": 0.27684563758389263,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004995986770236258,
      "loss": 6.1837,
      "mean_token_accuracy": 0.15050409361720085,
      "num_tokens": 5847434.0,
      "step": 3135
    },
    {
      "entropy": 6.304096031188965,
      "epoch": 0.27728717767573297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995967947692533,
      "loss": 6.1808,
      "mean_token_accuracy": 0.14390211701393127,
      "num_tokens": 5856464.0,
      "step": 3140
    },
    {
      "entropy": 6.311007452011109,
      "epoch": 0.2777287177675733,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995949081151571,
      "loss": 6.2402,
      "mean_token_accuracy": 0.14435049369931222,
      "num_tokens": 5865892.0,
      "step": 3145
    },
    {
      "entropy": 6.353293609619141,
      "epoch": 0.2781702578594136,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995930170613746,
      "loss": 6.2075,
      "mean_token_accuracy": 0.1576365649700165,
      "num_tokens": 5874606.0,
      "step": 3150
    },
    {
      "entropy": 6.275713014602661,
      "epoch": 0.27861179795125396,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995911216079425,
      "loss": 6.2116,
      "mean_token_accuracy": 0.15098029375076294,
      "num_tokens": 5883853.0,
      "step": 3155
    },
    {
      "entropy": 6.34206337928772,
      "epoch": 0.2790533380430943,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995892217548981,
      "loss": 6.2474,
      "mean_token_accuracy": 0.1448906570672989,
      "num_tokens": 5892841.0,
      "step": 3160
    },
    {
      "entropy": 6.324929285049438,
      "epoch": 0.2794948781349347,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995873175022786,
      "loss": 6.2295,
      "mean_token_accuracy": 0.15006719902157784,
      "num_tokens": 5901966.0,
      "step": 3165
    },
    {
      "entropy": 6.260264158248901,
      "epoch": 0.279936418226775,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995854088501213,
      "loss": 6.1281,
      "mean_token_accuracy": 0.14603292495012282,
      "num_tokens": 5910003.0,
      "step": 3170
    },
    {
      "entropy": 6.189469957351685,
      "epoch": 0.28037795831861534,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995834957984634,
      "loss": 6.2606,
      "mean_token_accuracy": 0.14485765993595123,
      "num_tokens": 5920307.0,
      "step": 3175
    },
    {
      "entropy": 6.290289306640625,
      "epoch": 0.28081949841045567,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995815783473428,
      "loss": 6.1153,
      "mean_token_accuracy": 0.15536403357982637,
      "num_tokens": 5929875.0,
      "step": 3180
    },
    {
      "entropy": 6.247414398193359,
      "epoch": 0.281261038502296,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995796564967967,
      "loss": 6.1021,
      "mean_token_accuracy": 0.14264860302209853,
      "num_tokens": 5938570.0,
      "step": 3185
    },
    {
      "entropy": 6.209772968292237,
      "epoch": 0.28170257859413633,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995777302468628,
      "loss": 6.2353,
      "mean_token_accuracy": 0.14704401940107345,
      "num_tokens": 5947693.0,
      "step": 3190
    },
    {
      "entropy": 6.332046413421631,
      "epoch": 0.28214411868597666,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995757995975789,
      "loss": 6.3286,
      "mean_token_accuracy": 0.14671236276626587,
      "num_tokens": 5957377.0,
      "step": 3195
    },
    {
      "entropy": 6.426817464828491,
      "epoch": 0.28258565877781705,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995738645489828,
      "loss": 6.2285,
      "mean_token_accuracy": 0.14824536591768264,
      "num_tokens": 5966443.0,
      "step": 3200
    },
    {
      "entropy": 6.258016681671142,
      "epoch": 0.2830271988696574,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995719251011124,
      "loss": 6.1471,
      "mean_token_accuracy": 0.14848615527153014,
      "num_tokens": 5975027.0,
      "step": 3205
    },
    {
      "entropy": 6.288780975341797,
      "epoch": 0.2834687389614977,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995699812540058,
      "loss": 6.2244,
      "mean_token_accuracy": 0.15162927508354188,
      "num_tokens": 5983722.0,
      "step": 3210
    },
    {
      "entropy": 6.282262182235717,
      "epoch": 0.28391027905333804,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499568033007701,
      "loss": 6.2324,
      "mean_token_accuracy": 0.1441471680998802,
      "num_tokens": 5993358.0,
      "step": 3215
    },
    {
      "entropy": 6.377847146987915,
      "epoch": 0.2843518191451784,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995660803622361,
      "loss": 6.161,
      "mean_token_accuracy": 0.1480330415070057,
      "num_tokens": 6002743.0,
      "step": 3220
    },
    {
      "entropy": 6.323432493209839,
      "epoch": 0.2847933592370187,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995641233176494,
      "loss": 6.3205,
      "mean_token_accuracy": 0.14193187803030013,
      "num_tokens": 6013727.0,
      "step": 3225
    },
    {
      "entropy": 6.404472589492798,
      "epoch": 0.28523489932885904,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995621618739792,
      "loss": 6.2652,
      "mean_token_accuracy": 0.14536072462797164,
      "num_tokens": 6023397.0,
      "step": 3230
    },
    {
      "entropy": 6.271181678771972,
      "epoch": 0.2856764394206994,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499560196031264,
      "loss": 6.1638,
      "mean_token_accuracy": 0.15034203678369523,
      "num_tokens": 6032453.0,
      "step": 3235
    },
    {
      "entropy": 6.19854884147644,
      "epoch": 0.28611797951253976,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995582257895423,
      "loss": 6.1258,
      "mean_token_accuracy": 0.14395371079444885,
      "num_tokens": 6042441.0,
      "step": 3240
    },
    {
      "entropy": 6.274320507049561,
      "epoch": 0.2865595196043801,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995562511488528,
      "loss": 6.1197,
      "mean_token_accuracy": 0.15483347177505494,
      "num_tokens": 6050881.0,
      "step": 3245
    },
    {
      "entropy": 6.220722341537476,
      "epoch": 0.2870010596962204,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995542721092337,
      "loss": 6.1872,
      "mean_token_accuracy": 0.15040701180696486,
      "num_tokens": 6060804.0,
      "step": 3250
    },
    {
      "entropy": 6.368199014663697,
      "epoch": 0.28744259978806075,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995522886707244,
      "loss": 6.2847,
      "mean_token_accuracy": 0.14250023737549783,
      "num_tokens": 6070774.0,
      "step": 3255
    },
    {
      "entropy": 6.330855846405029,
      "epoch": 0.2878841398799011,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995503008333634,
      "loss": 6.2491,
      "mean_token_accuracy": 0.14379776269197464,
      "num_tokens": 6080158.0,
      "step": 3260
    },
    {
      "entropy": 6.32779221534729,
      "epoch": 0.2883256799717414,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995483085971897,
      "loss": 6.16,
      "mean_token_accuracy": 0.15112278908491134,
      "num_tokens": 6089183.0,
      "step": 3265
    },
    {
      "entropy": 6.200159549713135,
      "epoch": 0.2887672200635818,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995463119622424,
      "loss": 6.2524,
      "mean_token_accuracy": 0.14461245387792587,
      "num_tokens": 6098536.0,
      "step": 3270
    },
    {
      "entropy": 6.196738433837891,
      "epoch": 0.28920876015542213,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995443109285604,
      "loss": 6.0112,
      "mean_token_accuracy": 0.1629092276096344,
      "num_tokens": 6107745.0,
      "step": 3275
    },
    {
      "entropy": 6.289895725250244,
      "epoch": 0.28965030024726246,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995423054961832,
      "loss": 6.2158,
      "mean_token_accuracy": 0.15729496926069259,
      "num_tokens": 6117512.0,
      "step": 3280
    },
    {
      "entropy": 6.227946424484253,
      "epoch": 0.2900918403391028,
      "grad_norm": 1.078125,
      "learning_rate": 0.00049954029566515,
      "loss": 6.2178,
      "mean_token_accuracy": 0.15285916179418563,
      "num_tokens": 6126030.0,
      "step": 3285
    },
    {
      "entropy": 6.346371364593506,
      "epoch": 0.2905333804309431,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995382814355,
      "loss": 6.2625,
      "mean_token_accuracy": 0.14053603783249854,
      "num_tokens": 6134888.0,
      "step": 3290
    },
    {
      "entropy": 6.351688432693481,
      "epoch": 0.29097492052278345,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995362628072728,
      "loss": 6.2117,
      "mean_token_accuracy": 0.1513692669570446,
      "num_tokens": 6144274.0,
      "step": 3295
    },
    {
      "entropy": 6.218261671066284,
      "epoch": 0.2914164606146238,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995342397805078,
      "loss": 6.2217,
      "mean_token_accuracy": 0.1503726065158844,
      "num_tokens": 6153406.0,
      "step": 3300
    },
    {
      "entropy": 6.262264680862427,
      "epoch": 0.29185800070646417,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995322123552448,
      "loss": 6.1233,
      "mean_token_accuracy": 0.1533094823360443,
      "num_tokens": 6162743.0,
      "step": 3305
    },
    {
      "entropy": 6.2753763675689695,
      "epoch": 0.2922995407983045,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995301805315235,
      "loss": 6.1201,
      "mean_token_accuracy": 0.15415377020835877,
      "num_tokens": 6171997.0,
      "step": 3310
    },
    {
      "entropy": 6.181772375106812,
      "epoch": 0.29274108089014483,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995281443093837,
      "loss": 6.1422,
      "mean_token_accuracy": 0.15062253326177596,
      "num_tokens": 6181275.0,
      "step": 3315
    },
    {
      "entropy": 6.289641571044922,
      "epoch": 0.29318262098198516,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995261036888652,
      "loss": 6.2135,
      "mean_token_accuracy": 0.14647497087717057,
      "num_tokens": 6191640.0,
      "step": 3320
    },
    {
      "entropy": 6.357338809967041,
      "epoch": 0.2936241610738255,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995240586700081,
      "loss": 6.2137,
      "mean_token_accuracy": 0.14479927867650985,
      "num_tokens": 6201508.0,
      "step": 3325
    },
    {
      "entropy": 6.220925617218017,
      "epoch": 0.2940657011656658,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995220092528522,
      "loss": 6.1079,
      "mean_token_accuracy": 0.15776659697294235,
      "num_tokens": 6209902.0,
      "step": 3330
    },
    {
      "entropy": 6.257762432098389,
      "epoch": 0.29450724125750616,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499519955437438,
      "loss": 6.2514,
      "mean_token_accuracy": 0.14023935049772263,
      "num_tokens": 6219760.0,
      "step": 3335
    },
    {
      "entropy": 6.248654699325561,
      "epoch": 0.29494878134934654,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995178972238054,
      "loss": 6.2308,
      "mean_token_accuracy": 0.14278148710727692,
      "num_tokens": 6228721.0,
      "step": 3340
    },
    {
      "entropy": 6.228511571884155,
      "epoch": 0.2953903214411869,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499515834611995,
      "loss": 6.0645,
      "mean_token_accuracy": 0.16055997535586358,
      "num_tokens": 6237070.0,
      "step": 3345
    },
    {
      "entropy": 6.2931300640106205,
      "epoch": 0.2958318615330272,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995137676020472,
      "loss": 6.2051,
      "mean_token_accuracy": 0.1466339647769928,
      "num_tokens": 6245659.0,
      "step": 3350
    },
    {
      "entropy": 6.231089639663696,
      "epoch": 0.29627340162486754,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995116961940023,
      "loss": 6.1736,
      "mean_token_accuracy": 0.15321153849363328,
      "num_tokens": 6255175.0,
      "step": 3355
    },
    {
      "entropy": 6.241381883621216,
      "epoch": 0.29671494171670787,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995096203879009,
      "loss": 6.1761,
      "mean_token_accuracy": 0.1443149983882904,
      "num_tokens": 6264962.0,
      "step": 3360
    },
    {
      "entropy": 6.29106593132019,
      "epoch": 0.2971564818085482,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995075401837837,
      "loss": 6.0989,
      "mean_token_accuracy": 0.15684471875429154,
      "num_tokens": 6273411.0,
      "step": 3365
    },
    {
      "entropy": 6.135526657104492,
      "epoch": 0.29759802190038853,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995054555816915,
      "loss": 6.0795,
      "mean_token_accuracy": 0.15481019616127015,
      "num_tokens": 6282618.0,
      "step": 3370
    },
    {
      "entropy": 6.155969190597534,
      "epoch": 0.2980395619922289,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995033665816651,
      "loss": 6.0726,
      "mean_token_accuracy": 0.15434197783470155,
      "num_tokens": 6292008.0,
      "step": 3375
    },
    {
      "entropy": 6.147941923141479,
      "epoch": 0.29848110208406925,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995012731837454,
      "loss": 6.0441,
      "mean_token_accuracy": 0.15588683634996414,
      "num_tokens": 6301079.0,
      "step": 3380
    },
    {
      "entropy": 6.323877429962158,
      "epoch": 0.2989226421759096,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994991753879736,
      "loss": 6.2273,
      "mean_token_accuracy": 0.14481185078620912,
      "num_tokens": 6310543.0,
      "step": 3385
    },
    {
      "entropy": 6.276103544235229,
      "epoch": 0.2993641822677499,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004994970731943904,
      "loss": 6.1692,
      "mean_token_accuracy": 0.1476268857717514,
      "num_tokens": 6320037.0,
      "step": 3390
    },
    {
      "entropy": 6.177714109420776,
      "epoch": 0.29980572235959024,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994949666030374,
      "loss": 6.1795,
      "mean_token_accuracy": 0.15479264855384828,
      "num_tokens": 6328948.0,
      "step": 3395
    },
    {
      "entropy": 6.212888717651367,
      "epoch": 0.3002472624514306,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994928556139557,
      "loss": 6.0425,
      "mean_token_accuracy": 0.1581359773874283,
      "num_tokens": 6337449.0,
      "step": 3400
    },
    {
      "entropy": 6.162185001373291,
      "epoch": 0.3006888025432709,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994907402271865,
      "loss": 6.1028,
      "mean_token_accuracy": 0.15499115511775016,
      "num_tokens": 6346130.0,
      "step": 3405
    },
    {
      "entropy": 6.15036244392395,
      "epoch": 0.3011303426351113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994886204427715,
      "loss": 6.1344,
      "mean_token_accuracy": 0.15851569175720215,
      "num_tokens": 6354973.0,
      "step": 3410
    },
    {
      "entropy": 6.328536748886108,
      "epoch": 0.3015718827269516,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994864962607519,
      "loss": 6.2148,
      "mean_token_accuracy": 0.14824963808059693,
      "num_tokens": 6364107.0,
      "step": 3415
    },
    {
      "entropy": 6.26984658241272,
      "epoch": 0.30201342281879195,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994843676811697,
      "loss": 6.1438,
      "mean_token_accuracy": 0.14578014612197876,
      "num_tokens": 6372859.0,
      "step": 3420
    },
    {
      "entropy": 6.184666395187378,
      "epoch": 0.3024549629106323,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994822347040664,
      "loss": 6.0282,
      "mean_token_accuracy": 0.1636571153998375,
      "num_tokens": 6381818.0,
      "step": 3425
    },
    {
      "entropy": 6.257387351989746,
      "epoch": 0.3028965030024726,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994800973294837,
      "loss": 6.1797,
      "mean_token_accuracy": 0.1482721135020256,
      "num_tokens": 6391460.0,
      "step": 3430
    },
    {
      "entropy": 6.25637993812561,
      "epoch": 0.30333804309431295,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994779555574636,
      "loss": 6.2086,
      "mean_token_accuracy": 0.15168215036392213,
      "num_tokens": 6401461.0,
      "step": 3435
    },
    {
      "entropy": 6.262447023391724,
      "epoch": 0.3037795831861533,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000499475809388048,
      "loss": 6.0842,
      "mean_token_accuracy": 0.14643362984061242,
      "num_tokens": 6410435.0,
      "step": 3440
    },
    {
      "entropy": 6.279139566421509,
      "epoch": 0.30422112327799367,
      "grad_norm": 1.125,
      "learning_rate": 0.000499473658821279,
      "loss": 6.2844,
      "mean_token_accuracy": 0.14141586795449257,
      "num_tokens": 6420710.0,
      "step": 3445
    },
    {
      "entropy": 6.305064296722412,
      "epoch": 0.304662663369834,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994715038571986,
      "loss": 6.1276,
      "mean_token_accuracy": 0.149975299090147,
      "num_tokens": 6429882.0,
      "step": 3450
    },
    {
      "entropy": 6.167029666900635,
      "epoch": 0.30510420346167433,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994693444958493,
      "loss": 6.1142,
      "mean_token_accuracy": 0.15782218649983407,
      "num_tokens": 6439183.0,
      "step": 3455
    },
    {
      "entropy": 6.307245588302612,
      "epoch": 0.30554574355351466,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499467180737273,
      "loss": 6.288,
      "mean_token_accuracy": 0.14309904649853705,
      "num_tokens": 6448460.0,
      "step": 3460
    },
    {
      "entropy": 6.270366525650024,
      "epoch": 0.305987283645355,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994650125815124,
      "loss": 6.1656,
      "mean_token_accuracy": 0.14977657794952393,
      "num_tokens": 6457687.0,
      "step": 3465
    },
    {
      "entropy": 6.241862058639526,
      "epoch": 0.3064288237371953,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994628400286097,
      "loss": 6.1063,
      "mean_token_accuracy": 0.15189075246453285,
      "num_tokens": 6467057.0,
      "step": 3470
    },
    {
      "entropy": 6.209890508651734,
      "epoch": 0.30687036382903565,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994606630786078,
      "loss": 6.1745,
      "mean_token_accuracy": 0.15533267706632614,
      "num_tokens": 6477484.0,
      "step": 3475
    },
    {
      "entropy": 6.248384809494018,
      "epoch": 0.30731190392087604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994584817315492,
      "loss": 6.233,
      "mean_token_accuracy": 0.14967372938990592,
      "num_tokens": 6488381.0,
      "step": 3480
    },
    {
      "entropy": 6.30739917755127,
      "epoch": 0.30775344401271637,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994562959874765,
      "loss": 6.1277,
      "mean_token_accuracy": 0.1523176297545433,
      "num_tokens": 6498279.0,
      "step": 3485
    },
    {
      "entropy": 6.166545104980469,
      "epoch": 0.3081949841045567,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994541058464326,
      "loss": 6.1476,
      "mean_token_accuracy": 0.14837286472320557,
      "num_tokens": 6508008.0,
      "step": 3490
    },
    {
      "entropy": 6.321815013885498,
      "epoch": 0.30863652419639703,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994519113084605,
      "loss": 6.1947,
      "mean_token_accuracy": 0.1496584579348564,
      "num_tokens": 6517263.0,
      "step": 3495
    },
    {
      "entropy": 6.262754631042481,
      "epoch": 0.30907806428823736,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994497123736029,
      "loss": 6.2631,
      "mean_token_accuracy": 0.1426304429769516,
      "num_tokens": 6527682.0,
      "step": 3500
    },
    {
      "entropy": 6.267001819610596,
      "epoch": 0.3095196043800777,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994475090419034,
      "loss": 6.1298,
      "mean_token_accuracy": 0.15242742300033568,
      "num_tokens": 6537143.0,
      "step": 3505
    },
    {
      "entropy": 6.250511837005615,
      "epoch": 0.309961144471918,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994453013134047,
      "loss": 6.1352,
      "mean_token_accuracy": 0.14925057888031007,
      "num_tokens": 6546561.0,
      "step": 3510
    },
    {
      "entropy": 6.185484886169434,
      "epoch": 0.3104026845637584,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994430891881502,
      "loss": 6.124,
      "mean_token_accuracy": 0.15090959072113036,
      "num_tokens": 6555806.0,
      "step": 3515
    },
    {
      "entropy": 6.202518653869629,
      "epoch": 0.31084422465559874,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994408726661832,
      "loss": 6.166,
      "mean_token_accuracy": 0.1501057654619217,
      "num_tokens": 6564836.0,
      "step": 3520
    },
    {
      "entropy": 6.245743179321289,
      "epoch": 0.3112857647474391,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994386517475472,
      "loss": 6.1713,
      "mean_token_accuracy": 0.1496300369501114,
      "num_tokens": 6574997.0,
      "step": 3525
    },
    {
      "entropy": 6.246868562698364,
      "epoch": 0.3117273048392794,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994364264322856,
      "loss": 6.1715,
      "mean_token_accuracy": 0.14895060658454895,
      "num_tokens": 6584589.0,
      "step": 3530
    },
    {
      "entropy": 6.162112808227539,
      "epoch": 0.31216884493111974,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994341967204421,
      "loss": 6.1199,
      "mean_token_accuracy": 0.15284974724054337,
      "num_tokens": 6592898.0,
      "step": 3535
    },
    {
      "entropy": 6.326304626464844,
      "epoch": 0.31261038502296007,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994319626120603,
      "loss": 6.1268,
      "mean_token_accuracy": 0.16072259843349457,
      "num_tokens": 6602454.0,
      "step": 3540
    },
    {
      "entropy": 6.1308966159820555,
      "epoch": 0.3130519251148004,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994297241071841,
      "loss": 6.1467,
      "mean_token_accuracy": 0.1551404133439064,
      "num_tokens": 6612454.0,
      "step": 3545
    },
    {
      "entropy": 6.29251446723938,
      "epoch": 0.3134934652066408,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499427481205857,
      "loss": 6.1423,
      "mean_token_accuracy": 0.15539032369852065,
      "num_tokens": 6621286.0,
      "step": 3550
    },
    {
      "entropy": 6.253758335113526,
      "epoch": 0.3139350052984811,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004994252339081234,
      "loss": 6.0651,
      "mean_token_accuracy": 0.15716515630483627,
      "num_tokens": 6629684.0,
      "step": 3555
    },
    {
      "entropy": 6.167656517028808,
      "epoch": 0.31437654539032145,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499422982214027,
      "loss": 6.1775,
      "mean_token_accuracy": 0.15429823398590087,
      "num_tokens": 6638836.0,
      "step": 3560
    },
    {
      "entropy": 6.254745101928711,
      "epoch": 0.3148180854821618,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994207261236121,
      "loss": 6.1189,
      "mean_token_accuracy": 0.14906142503023148,
      "num_tokens": 6647337.0,
      "step": 3565
    },
    {
      "entropy": 6.145832633972168,
      "epoch": 0.3152596255740021,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994184656369227,
      "loss": 6.1026,
      "mean_token_accuracy": 0.145336801558733,
      "num_tokens": 6657246.0,
      "step": 3570
    },
    {
      "entropy": 6.310924911499024,
      "epoch": 0.31570116566584244,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994162007540033,
      "loss": 6.2505,
      "mean_token_accuracy": 0.14196238815784454,
      "num_tokens": 6667981.0,
      "step": 3575
    },
    {
      "entropy": 6.2675220489501955,
      "epoch": 0.3161427057576828,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994139314748981,
      "loss": 6.0974,
      "mean_token_accuracy": 0.15776362121105195,
      "num_tokens": 6677462.0,
      "step": 3580
    },
    {
      "entropy": 6.107239675521851,
      "epoch": 0.31658424584952316,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994116577996517,
      "loss": 6.1152,
      "mean_token_accuracy": 0.15776521414518357,
      "num_tokens": 6687225.0,
      "step": 3585
    },
    {
      "entropy": 6.2434509754180905,
      "epoch": 0.3170257859413635,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994093797283084,
      "loss": 6.2122,
      "mean_token_accuracy": 0.1487804666161537,
      "num_tokens": 6696345.0,
      "step": 3590
    },
    {
      "entropy": 6.313143348693847,
      "epoch": 0.3174673260332038,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994070972609132,
      "loss": 6.1867,
      "mean_token_accuracy": 0.14744829311966895,
      "num_tokens": 6706207.0,
      "step": 3595
    },
    {
      "entropy": 6.24591326713562,
      "epoch": 0.31790886612504415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004994048103975103,
      "loss": 6.0904,
      "mean_token_accuracy": 0.15303746610879898,
      "num_tokens": 6714430.0,
      "step": 3600
    },
    {
      "entropy": 6.204165840148926,
      "epoch": 0.3183504062168845,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499402519138145,
      "loss": 6.1197,
      "mean_token_accuracy": 0.1557806834578514,
      "num_tokens": 6724139.0,
      "step": 3605
    },
    {
      "entropy": 6.151504611968994,
      "epoch": 0.3187919463087248,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994002234828619,
      "loss": 6.1566,
      "mean_token_accuracy": 0.1508852459490299,
      "num_tokens": 6733525.0,
      "step": 3610
    },
    {
      "entropy": 6.321462392807007,
      "epoch": 0.31923348640056515,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499397923431706,
      "loss": 6.2666,
      "mean_token_accuracy": 0.1465996690094471,
      "num_tokens": 6743738.0,
      "step": 3615
    },
    {
      "entropy": 6.337309741973877,
      "epoch": 0.31967502649240553,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004993956189847226,
      "loss": 6.1012,
      "mean_token_accuracy": 0.15459158420562744,
      "num_tokens": 6752900.0,
      "step": 3620
    },
    {
      "entropy": 6.150611686706543,
      "epoch": 0.32011656658424587,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993933101419565,
      "loss": 6.106,
      "mean_token_accuracy": 0.1578878253698349,
      "num_tokens": 6761776.0,
      "step": 3625
    },
    {
      "entropy": 6.23021969795227,
      "epoch": 0.3205581066760862,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993909969034531,
      "loss": 6.2486,
      "mean_token_accuracy": 0.1432959534227848,
      "num_tokens": 6771543.0,
      "step": 3630
    },
    {
      "entropy": 6.232846355438232,
      "epoch": 0.3209996467679265,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004993886792692576,
      "loss": 6.1261,
      "mean_token_accuracy": 0.15022226572036743,
      "num_tokens": 6780010.0,
      "step": 3635
    },
    {
      "entropy": 6.273227500915527,
      "epoch": 0.32144118685976686,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993863572394156,
      "loss": 6.186,
      "mean_token_accuracy": 0.14826097190380097,
      "num_tokens": 6788830.0,
      "step": 3640
    },
    {
      "entropy": 6.266057395935059,
      "epoch": 0.3218827269516072,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993840308139724,
      "loss": 6.163,
      "mean_token_accuracy": 0.15327871441841126,
      "num_tokens": 6799121.0,
      "step": 3645
    },
    {
      "entropy": 6.232186365127563,
      "epoch": 0.3223242670434475,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993816999929738,
      "loss": 6.1253,
      "mean_token_accuracy": 0.15162717401981354,
      "num_tokens": 6808519.0,
      "step": 3650
    },
    {
      "entropy": 6.129374217987061,
      "epoch": 0.3227658071352879,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993793647764651,
      "loss": 6.1462,
      "mean_token_accuracy": 0.16090914756059646,
      "num_tokens": 6817583.0,
      "step": 3655
    },
    {
      "entropy": 6.254270458221436,
      "epoch": 0.32320734722712824,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993770251644923,
      "loss": 6.1315,
      "mean_token_accuracy": 0.15279360860586166,
      "num_tokens": 6827732.0,
      "step": 3660
    },
    {
      "entropy": 6.188901424407959,
      "epoch": 0.32364888731896857,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993746811571013,
      "loss": 6.1101,
      "mean_token_accuracy": 0.15757904648780824,
      "num_tokens": 6836903.0,
      "step": 3665
    },
    {
      "entropy": 6.073100471496582,
      "epoch": 0.3240904274108089,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993723327543379,
      "loss": 6.119,
      "mean_token_accuracy": 0.15876784324645996,
      "num_tokens": 6845684.0,
      "step": 3670
    },
    {
      "entropy": 6.200423860549927,
      "epoch": 0.32453196750264923,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499369979956248,
      "loss": 6.1035,
      "mean_token_accuracy": 0.14955462887883186,
      "num_tokens": 6854650.0,
      "step": 3675
    },
    {
      "entropy": 6.267183446884156,
      "epoch": 0.32497350759448956,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993676227628779,
      "loss": 6.0977,
      "mean_token_accuracy": 0.1460177183151245,
      "num_tokens": 6864340.0,
      "step": 3680
    },
    {
      "entropy": 6.208888530731201,
      "epoch": 0.3254150476863299,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004993652611742736,
      "loss": 6.1343,
      "mean_token_accuracy": 0.15321332961320877,
      "num_tokens": 6873244.0,
      "step": 3685
    },
    {
      "entropy": 6.15152006149292,
      "epoch": 0.3258565877781703,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993628951904815,
      "loss": 6.0176,
      "mean_token_accuracy": 0.15752123296260834,
      "num_tokens": 6882392.0,
      "step": 3690
    },
    {
      "entropy": 6.2300208568572994,
      "epoch": 0.3262981278700106,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004993605248115479,
      "loss": 6.2446,
      "mean_token_accuracy": 0.1435020685195923,
      "num_tokens": 6891515.0,
      "step": 3695
    },
    {
      "entropy": 6.330176734924317,
      "epoch": 0.32673966796185094,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993581500375191,
      "loss": 6.1177,
      "mean_token_accuracy": 0.14757276177406312,
      "num_tokens": 6900104.0,
      "step": 3700
    },
    {
      "entropy": 6.2001354694366455,
      "epoch": 0.3271812080536913,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004993557708684417,
      "loss": 6.2227,
      "mean_token_accuracy": 0.14519331306219102,
      "num_tokens": 6910443.0,
      "step": 3705
    },
    {
      "entropy": 6.208211469650268,
      "epoch": 0.3276227481455316,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004993533873043625,
      "loss": 6.1062,
      "mean_token_accuracy": 0.15348025262355805,
      "num_tokens": 6919882.0,
      "step": 3710
    },
    {
      "entropy": 6.281739521026611,
      "epoch": 0.32806428823737194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499350999345328,
      "loss": 6.2142,
      "mean_token_accuracy": 0.14149869233369827,
      "num_tokens": 6929757.0,
      "step": 3715
    },
    {
      "entropy": 6.222444868087768,
      "epoch": 0.32850582832921227,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499348606991385,
      "loss": 6.201,
      "mean_token_accuracy": 0.1466899633407593,
      "num_tokens": 6938578.0,
      "step": 3720
    },
    {
      "entropy": 6.181999588012696,
      "epoch": 0.32894736842105265,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993462102425805,
      "loss": 6.0957,
      "mean_token_accuracy": 0.14744215086102486,
      "num_tokens": 6947920.0,
      "step": 3725
    },
    {
      "entropy": 6.170785808563233,
      "epoch": 0.329388908512893,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993438090989612,
      "loss": 6.0325,
      "mean_token_accuracy": 0.15373560339212416,
      "num_tokens": 6957291.0,
      "step": 3730
    },
    {
      "entropy": 6.204143905639649,
      "epoch": 0.3298304486047333,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004993414035605743,
      "loss": 6.0944,
      "mean_token_accuracy": 0.16135938167572023,
      "num_tokens": 6966504.0,
      "step": 3735
    },
    {
      "entropy": 6.26701283454895,
      "epoch": 0.33027198869657365,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993389936274669,
      "loss": 6.1381,
      "mean_token_accuracy": 0.15607366263866423,
      "num_tokens": 6976234.0,
      "step": 3740
    },
    {
      "entropy": 6.179308748245239,
      "epoch": 0.330713528788414,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993365792996862,
      "loss": 6.1882,
      "mean_token_accuracy": 0.1475161299109459,
      "num_tokens": 6986372.0,
      "step": 3745
    },
    {
      "entropy": 6.258656454086304,
      "epoch": 0.3311550688802543,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993341605772795,
      "loss": 6.0768,
      "mean_token_accuracy": 0.155860435962677,
      "num_tokens": 6995553.0,
      "step": 3750
    },
    {
      "entropy": 6.148722696304321,
      "epoch": 0.33159660897209464,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993317374602941,
      "loss": 5.9902,
      "mean_token_accuracy": 0.16303742080926895,
      "num_tokens": 7005675.0,
      "step": 3755
    },
    {
      "entropy": 6.081928777694702,
      "epoch": 0.33203814906393503,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004993293099487777,
      "loss": 6.1198,
      "mean_token_accuracy": 0.14885973036289216,
      "num_tokens": 7014953.0,
      "step": 3760
    },
    {
      "entropy": 6.263726997375488,
      "epoch": 0.33247968915577536,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993268780427776,
      "loss": 6.1445,
      "mean_token_accuracy": 0.1564795732498169,
      "num_tokens": 7025075.0,
      "step": 3765
    },
    {
      "entropy": 6.201370096206665,
      "epoch": 0.3329212292476157,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004993244417423416,
      "loss": 6.1224,
      "mean_token_accuracy": 0.15349680185317993,
      "num_tokens": 7034286.0,
      "step": 3770
    },
    {
      "entropy": 6.201808977127075,
      "epoch": 0.333362769339456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004993220010475174,
      "loss": 6.1767,
      "mean_token_accuracy": 0.14727241545915604,
      "num_tokens": 7043957.0,
      "step": 3775
    },
    {
      "entropy": 6.2423442840576175,
      "epoch": 0.33380430943129635,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993195559583526,
      "loss": 6.091,
      "mean_token_accuracy": 0.15680659636855127,
      "num_tokens": 7053170.0,
      "step": 3780
    },
    {
      "entropy": 6.163934850692749,
      "epoch": 0.3342458495231367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993171064748954,
      "loss": 6.0909,
      "mean_token_accuracy": 0.15337430387735368,
      "num_tokens": 7062469.0,
      "step": 3785
    },
    {
      "entropy": 6.245288276672364,
      "epoch": 0.334687389614977,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993146525971937,
      "loss": 6.0751,
      "mean_token_accuracy": 0.15703266561031343,
      "num_tokens": 7071463.0,
      "step": 3790
    },
    {
      "entropy": 6.30460000038147,
      "epoch": 0.3351289297068174,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993121943252955,
      "loss": 6.1685,
      "mean_token_accuracy": 0.1484996944665909,
      "num_tokens": 7081574.0,
      "step": 3795
    },
    {
      "entropy": 6.146984243392945,
      "epoch": 0.33557046979865773,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993097316592489,
      "loss": 6.0826,
      "mean_token_accuracy": 0.15749146193265914,
      "num_tokens": 7090835.0,
      "step": 3800
    },
    {
      "entropy": 6.072944211959839,
      "epoch": 0.33601200989049806,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004993072645991023,
      "loss": 6.0092,
      "mean_token_accuracy": 0.14467609971761702,
      "num_tokens": 7100521.0,
      "step": 3805
    },
    {
      "entropy": 6.172339153289795,
      "epoch": 0.3364535499823384,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000499304793144904,
      "loss": 6.108,
      "mean_token_accuracy": 0.15104963332414628,
      "num_tokens": 7109011.0,
      "step": 3810
    },
    {
      "entropy": 6.258471536636352,
      "epoch": 0.3368950900741787,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993023172967022,
      "loss": 6.1451,
      "mean_token_accuracy": 0.14512295573949813,
      "num_tokens": 7119128.0,
      "step": 3815
    },
    {
      "entropy": 6.266326761245727,
      "epoch": 0.33733663016601906,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992998370545458,
      "loss": 6.2073,
      "mean_token_accuracy": 0.14363647550344466,
      "num_tokens": 7128313.0,
      "step": 3820
    },
    {
      "entropy": 6.234633827209473,
      "epoch": 0.3377781702578594,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004992973524184831,
      "loss": 6.1686,
      "mean_token_accuracy": 0.14797609224915503,
      "num_tokens": 7137567.0,
      "step": 3825
    },
    {
      "entropy": 6.209169197082519,
      "epoch": 0.3382197103496998,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004992948633885627,
      "loss": 6.1355,
      "mean_token_accuracy": 0.15170362889766692,
      "num_tokens": 7147254.0,
      "step": 3830
    },
    {
      "entropy": 6.123815822601318,
      "epoch": 0.3386612504415401,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992923699648335,
      "loss": 6.055,
      "mean_token_accuracy": 0.15891509354114533,
      "num_tokens": 7156215.0,
      "step": 3835
    },
    {
      "entropy": 6.286255836486816,
      "epoch": 0.33910279053338044,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992898721473445,
      "loss": 6.1341,
      "mean_token_accuracy": 0.14742862731218337,
      "num_tokens": 7165653.0,
      "step": 3840
    },
    {
      "entropy": 6.107093811035156,
      "epoch": 0.33954433062522077,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992873699361444,
      "loss": 6.0116,
      "mean_token_accuracy": 0.15364348739385605,
      "num_tokens": 7174797.0,
      "step": 3845
    },
    {
      "entropy": 6.231772947311401,
      "epoch": 0.3399858707170611,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992848633312822,
      "loss": 6.076,
      "mean_token_accuracy": 0.15550567209720612,
      "num_tokens": 7184040.0,
      "step": 3850
    },
    {
      "entropy": 6.206028461456299,
      "epoch": 0.34042741080890143,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004992823523328071,
      "loss": 6.0162,
      "mean_token_accuracy": 0.156949782371521,
      "num_tokens": 7193533.0,
      "step": 3855
    },
    {
      "entropy": 6.040725946426392,
      "epoch": 0.34086895090074176,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004992798369407684,
      "loss": 6.1271,
      "mean_token_accuracy": 0.1540011927485466,
      "num_tokens": 7203355.0,
      "step": 3860
    },
    {
      "entropy": 6.231026697158813,
      "epoch": 0.34131049099258215,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992773171552152,
      "loss": 6.0662,
      "mean_token_accuracy": 0.15278246700763704,
      "num_tokens": 7212569.0,
      "step": 3865
    },
    {
      "entropy": 6.1147346019744875,
      "epoch": 0.3417520310844225,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992747929761968,
      "loss": 6.0862,
      "mean_token_accuracy": 0.1568584769964218,
      "num_tokens": 7221886.0,
      "step": 3870
    },
    {
      "entropy": 6.20992078781128,
      "epoch": 0.3421935711762628,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004992722644037628,
      "loss": 6.1399,
      "mean_token_accuracy": 0.14878576919436454,
      "num_tokens": 7232065.0,
      "step": 3875
    },
    {
      "entropy": 6.2117961883544925,
      "epoch": 0.34263511126810314,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992697314379628,
      "loss": 6.0802,
      "mean_token_accuracy": 0.1546872690320015,
      "num_tokens": 7242287.0,
      "step": 3880
    },
    {
      "entropy": 6.163637351989746,
      "epoch": 0.3430766513599435,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004992671940788462,
      "loss": 6.0744,
      "mean_token_accuracy": 0.15107770562171935,
      "num_tokens": 7250638.0,
      "step": 3885
    },
    {
      "entropy": 6.146958065032959,
      "epoch": 0.3435181914517838,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004992646523264628,
      "loss": 6.0492,
      "mean_token_accuracy": 0.1611901268362999,
      "num_tokens": 7259526.0,
      "step": 3890
    },
    {
      "entropy": 6.119075059890747,
      "epoch": 0.34395973154362414,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992621061808625,
      "loss": 6.0604,
      "mean_token_accuracy": 0.15892861932516097,
      "num_tokens": 7269260.0,
      "step": 3895
    },
    {
      "entropy": 6.161959505081176,
      "epoch": 0.3444012716354645,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499259555642095,
      "loss": 6.0786,
      "mean_token_accuracy": 0.15186367481946944,
      "num_tokens": 7279061.0,
      "step": 3900
    },
    {
      "entropy": 6.21967043876648,
      "epoch": 0.34484281172730485,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992570007102104,
      "loss": 6.0831,
      "mean_token_accuracy": 0.15683933049440385,
      "num_tokens": 7289721.0,
      "step": 3905
    },
    {
      "entropy": 6.214530658721924,
      "epoch": 0.3452843518191452,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004992544413852587,
      "loss": 6.2136,
      "mean_token_accuracy": 0.14554179906845094,
      "num_tokens": 7299836.0,
      "step": 3910
    },
    {
      "entropy": 6.225806951522827,
      "epoch": 0.3457258919109855,
      "grad_norm": 1.125,
      "learning_rate": 0.00049925187766729,
      "loss": 6.2249,
      "mean_token_accuracy": 0.14765020608901977,
      "num_tokens": 7309306.0,
      "step": 3915
    },
    {
      "entropy": 6.249141550064087,
      "epoch": 0.34616743200282585,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004992493095563545,
      "loss": 6.1238,
      "mean_token_accuracy": 0.16257761269807816,
      "num_tokens": 7318486.0,
      "step": 3920
    },
    {
      "entropy": 6.111485767364502,
      "epoch": 0.3466089720946662,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992467370525026,
      "loss": 6.083,
      "mean_token_accuracy": 0.1557182028889656,
      "num_tokens": 7327764.0,
      "step": 3925
    },
    {
      "entropy": 6.248839044570923,
      "epoch": 0.3470505121865065,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992441601557848,
      "loss": 6.168,
      "mean_token_accuracy": 0.15208059847354888,
      "num_tokens": 7337118.0,
      "step": 3930
    },
    {
      "entropy": 6.197002506256103,
      "epoch": 0.3474920522783469,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004992415788662514,
      "loss": 6.0958,
      "mean_token_accuracy": 0.1525207430124283,
      "num_tokens": 7345973.0,
      "step": 3935
    },
    {
      "entropy": 6.194469833374024,
      "epoch": 0.34793359237018723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992389931839529,
      "loss": 6.0928,
      "mean_token_accuracy": 0.1534672871232033,
      "num_tokens": 7355776.0,
      "step": 3940
    },
    {
      "entropy": 6.168269205093384,
      "epoch": 0.34837513246202756,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992364031089401,
      "loss": 6.1161,
      "mean_token_accuracy": 0.1555505856871605,
      "num_tokens": 7364820.0,
      "step": 3945
    },
    {
      "entropy": 6.2373566150665285,
      "epoch": 0.3488166725538679,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004992338086412636,
      "loss": 6.1218,
      "mean_token_accuracy": 0.14930969327688218,
      "num_tokens": 7373881.0,
      "step": 3950
    },
    {
      "entropy": 6.135814476013183,
      "epoch": 0.3492582126457082,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004992312097809744,
      "loss": 6.0892,
      "mean_token_accuracy": 0.15562115162611007,
      "num_tokens": 7383095.0,
      "step": 3955
    },
    {
      "entropy": 6.195784330368042,
      "epoch": 0.34969975273754855,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992286065281234,
      "loss": 6.1216,
      "mean_token_accuracy": 0.15000923871994018,
      "num_tokens": 7392702.0,
      "step": 3960
    },
    {
      "entropy": 6.247405576705932,
      "epoch": 0.3501412928293889,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992259988827614,
      "loss": 6.2769,
      "mean_token_accuracy": 0.14470015615224838,
      "num_tokens": 7402560.0,
      "step": 3965
    },
    {
      "entropy": 6.284795522689819,
      "epoch": 0.35058283292122927,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992233868449397,
      "loss": 6.051,
      "mean_token_accuracy": 0.15953975468873977,
      "num_tokens": 7412375.0,
      "step": 3970
    },
    {
      "entropy": 6.195076513290405,
      "epoch": 0.3510243730130696,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992207704147093,
      "loss": 6.0928,
      "mean_token_accuracy": 0.15187776535749437,
      "num_tokens": 7420734.0,
      "step": 3975
    },
    {
      "entropy": 6.101401853561401,
      "epoch": 0.35146591310490993,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992181495921216,
      "loss": 6.0919,
      "mean_token_accuracy": 0.15785269439220428,
      "num_tokens": 7430156.0,
      "step": 3980
    },
    {
      "entropy": 6.222190999984742,
      "epoch": 0.35190745319675026,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004992155243772277,
      "loss": 6.1007,
      "mean_token_accuracy": 0.1574430137872696,
      "num_tokens": 7439189.0,
      "step": 3985
    },
    {
      "entropy": 6.2206672668457035,
      "epoch": 0.3523489932885906,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992128947700795,
      "loss": 6.1249,
      "mean_token_accuracy": 0.1455397441983223,
      "num_tokens": 7448768.0,
      "step": 3990
    },
    {
      "entropy": 6.180650806427002,
      "epoch": 0.3527905333804309,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499210260770728,
      "loss": 6.064,
      "mean_token_accuracy": 0.15497558265924455,
      "num_tokens": 7458468.0,
      "step": 3995
    },
    {
      "entropy": 6.150880622863769,
      "epoch": 0.35323207347227126,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499207622379225,
      "loss": 6.1475,
      "mean_token_accuracy": 0.1480465464293957,
      "num_tokens": 7467351.0,
      "step": 4000
    },
    {
      "entropy": 6.2564455509185795,
      "epoch": 0.35367361356411164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992049795956222,
      "loss": 6.1012,
      "mean_token_accuracy": 0.14711347222328186,
      "num_tokens": 7477116.0,
      "step": 4005
    },
    {
      "entropy": 6.234343814849853,
      "epoch": 0.354115153655952,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004992023324199715,
      "loss": 6.0249,
      "mean_token_accuracy": 0.15656134784221648,
      "num_tokens": 7485216.0,
      "step": 4010
    },
    {
      "entropy": 6.075209522247315,
      "epoch": 0.3545566937477923,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991996808523245,
      "loss": 6.1798,
      "mean_token_accuracy": 0.1469147637486458,
      "num_tokens": 7495084.0,
      "step": 4015
    },
    {
      "entropy": 6.306225442886353,
      "epoch": 0.35499823383963264,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991970248927332,
      "loss": 6.1973,
      "mean_token_accuracy": 0.14604026302695275,
      "num_tokens": 7503219.0,
      "step": 4020
    },
    {
      "entropy": 6.196821165084839,
      "epoch": 0.35543977393147297,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004991943645412498,
      "loss": 5.9766,
      "mean_token_accuracy": 0.1642112761735916,
      "num_tokens": 7511961.0,
      "step": 4025
    },
    {
      "entropy": 5.985270690917969,
      "epoch": 0.3558813140233133,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991916997979263,
      "loss": 6.1096,
      "mean_token_accuracy": 0.1546097069978714,
      "num_tokens": 7521927.0,
      "step": 4030
    },
    {
      "entropy": 6.103000640869141,
      "epoch": 0.35632285411515363,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991890306628149,
      "loss": 6.0451,
      "mean_token_accuracy": 0.15554122179746627,
      "num_tokens": 7531093.0,
      "step": 4035
    },
    {
      "entropy": 6.198269557952881,
      "epoch": 0.356764394206994,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991863571359678,
      "loss": 6.0078,
      "mean_token_accuracy": 0.15710966363549234,
      "num_tokens": 7539813.0,
      "step": 4040
    },
    {
      "entropy": 6.244335699081421,
      "epoch": 0.35720593429883435,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004991836792174376,
      "loss": 6.1875,
      "mean_token_accuracy": 0.1476308137178421,
      "num_tokens": 7548954.0,
      "step": 4045
    },
    {
      "entropy": 6.100079536437988,
      "epoch": 0.3576474743906747,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991809969072765,
      "loss": 6.0322,
      "mean_token_accuracy": 0.1548540085554123,
      "num_tokens": 7558275.0,
      "step": 4050
    },
    {
      "entropy": 6.13811764717102,
      "epoch": 0.358089014482515,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004991783102055371,
      "loss": 6.0688,
      "mean_token_accuracy": 0.15873494520783424,
      "num_tokens": 7567870.0,
      "step": 4055
    },
    {
      "entropy": 6.249936914443969,
      "epoch": 0.35853055457435534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991756191122723,
      "loss": 6.0385,
      "mean_token_accuracy": 0.15568251237273217,
      "num_tokens": 7577485.0,
      "step": 4060
    },
    {
      "entropy": 6.142808294296264,
      "epoch": 0.3589720946661957,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991729236275346,
      "loss": 6.0888,
      "mean_token_accuracy": 0.150615693628788,
      "num_tokens": 7587398.0,
      "step": 4065
    },
    {
      "entropy": 6.10182294845581,
      "epoch": 0.359413634758036,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991702237513768,
      "loss": 6.0896,
      "mean_token_accuracy": 0.1515856146812439,
      "num_tokens": 7596310.0,
      "step": 4070
    },
    {
      "entropy": 6.236705541610718,
      "epoch": 0.3598551748498764,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991675194838517,
      "loss": 6.1214,
      "mean_token_accuracy": 0.14487750828266144,
      "num_tokens": 7606971.0,
      "step": 4075
    },
    {
      "entropy": 6.157118654251098,
      "epoch": 0.3602967149417167,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004991648108250125,
      "loss": 6.1073,
      "mean_token_accuracy": 0.15116591304540633,
      "num_tokens": 7616636.0,
      "step": 4080
    },
    {
      "entropy": 6.23088059425354,
      "epoch": 0.36073825503355705,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499162097774912,
      "loss": 6.1325,
      "mean_token_accuracy": 0.15135489255189896,
      "num_tokens": 7625128.0,
      "step": 4085
    },
    {
      "entropy": 6.168088626861572,
      "epoch": 0.3611797951253974,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991593803336037,
      "loss": 6.0464,
      "mean_token_accuracy": 0.15914286002516748,
      "num_tokens": 7634224.0,
      "step": 4090
    },
    {
      "entropy": 6.109174299240112,
      "epoch": 0.3616213352172377,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991566585011405,
      "loss": 6.0204,
      "mean_token_accuracy": 0.1594786301255226,
      "num_tokens": 7643392.0,
      "step": 4095
    },
    {
      "entropy": 6.046089458465576,
      "epoch": 0.36206287530907805,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004991539322775758,
      "loss": 6.0402,
      "mean_token_accuracy": 0.15390734672546386,
      "num_tokens": 7652491.0,
      "step": 4100
    },
    {
      "entropy": 6.333108568191529,
      "epoch": 0.3625044154009184,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991512016629632,
      "loss": 6.2631,
      "mean_token_accuracy": 0.1374726377427578,
      "num_tokens": 7663192.0,
      "step": 4105
    },
    {
      "entropy": 6.233335781097412,
      "epoch": 0.36294595549275877,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991484666573558,
      "loss": 6.0951,
      "mean_token_accuracy": 0.1491971492767334,
      "num_tokens": 7672801.0,
      "step": 4110
    },
    {
      "entropy": 6.150677967071533,
      "epoch": 0.3633874955845991,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991457272608077,
      "loss": 6.1107,
      "mean_token_accuracy": 0.1492188058793545,
      "num_tokens": 7683013.0,
      "step": 4115
    },
    {
      "entropy": 6.215946054458618,
      "epoch": 0.3638290356764394,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991429834733721,
      "loss": 6.1377,
      "mean_token_accuracy": 0.14726671427488328,
      "num_tokens": 7692760.0,
      "step": 4120
    },
    {
      "entropy": 6.212368583679199,
      "epoch": 0.36427057576827976,
      "grad_norm": 1.390625,
      "learning_rate": 0.000499140235295103,
      "loss": 6.0806,
      "mean_token_accuracy": 0.15191175639629365,
      "num_tokens": 7702005.0,
      "step": 4125
    },
    {
      "entropy": 6.164025735855103,
      "epoch": 0.3647121158601201,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004991374827260542,
      "loss": 6.129,
      "mean_token_accuracy": 0.1487259477376938,
      "num_tokens": 7711604.0,
      "step": 4130
    },
    {
      "entropy": 6.113088750839234,
      "epoch": 0.3651536559519604,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004991347257662795,
      "loss": 6.0471,
      "mean_token_accuracy": 0.153825144469738,
      "num_tokens": 7720434.0,
      "step": 4135
    },
    {
      "entropy": 6.218751907348633,
      "epoch": 0.36559519604380075,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499131964415833,
      "loss": 6.0877,
      "mean_token_accuracy": 0.1532296895980835,
      "num_tokens": 7730733.0,
      "step": 4140
    },
    {
      "entropy": 6.158688402175903,
      "epoch": 0.36603673613564114,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004991291986747689,
      "loss": 6.1581,
      "mean_token_accuracy": 0.153475009649992,
      "num_tokens": 7740394.0,
      "step": 4145
    },
    {
      "entropy": 6.191565799713135,
      "epoch": 0.36647827622748147,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991264285431412,
      "loss": 6.1045,
      "mean_token_accuracy": 0.15052587389945984,
      "num_tokens": 7748919.0,
      "step": 4150
    },
    {
      "entropy": 6.293946027755737,
      "epoch": 0.3669198163193218,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991236540210041,
      "loss": 6.1009,
      "mean_token_accuracy": 0.1537343256175518,
      "num_tokens": 7758591.0,
      "step": 4155
    },
    {
      "entropy": 6.244499015808105,
      "epoch": 0.36736135641116213,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991208751084122,
      "loss": 6.1485,
      "mean_token_accuracy": 0.1491132453083992,
      "num_tokens": 7768269.0,
      "step": 4160
    },
    {
      "entropy": 6.249349975585938,
      "epoch": 0.36780289650300246,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004991180918054199,
      "loss": 6.1767,
      "mean_token_accuracy": 0.1463681861758232,
      "num_tokens": 7776980.0,
      "step": 4165
    },
    {
      "entropy": 6.171654605865479,
      "epoch": 0.3682444365948428,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991153041120815,
      "loss": 6.0811,
      "mean_token_accuracy": 0.15733396261930466,
      "num_tokens": 7786664.0,
      "step": 4170
    },
    {
      "entropy": 6.12093939781189,
      "epoch": 0.3686859766866831,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991125120284519,
      "loss": 5.9525,
      "mean_token_accuracy": 0.1650825932621956,
      "num_tokens": 7794981.0,
      "step": 4175
    },
    {
      "entropy": 6.110823106765747,
      "epoch": 0.3691275167785235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991097155545856,
      "loss": 6.0387,
      "mean_token_accuracy": 0.1560913234949112,
      "num_tokens": 7804850.0,
      "step": 4180
    },
    {
      "entropy": 6.106446075439453,
      "epoch": 0.36956905687036384,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004991069146905374,
      "loss": 6.052,
      "mean_token_accuracy": 0.15863640755414962,
      "num_tokens": 7814117.0,
      "step": 4185
    },
    {
      "entropy": 6.183217334747314,
      "epoch": 0.3700105969622042,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991041094363621,
      "loss": 6.0928,
      "mean_token_accuracy": 0.15649753212928771,
      "num_tokens": 7823384.0,
      "step": 4190
    },
    {
      "entropy": 6.2000589847564695,
      "epoch": 0.3704521370540445,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991012997921149,
      "loss": 6.0863,
      "mean_token_accuracy": 0.154317145049572,
      "num_tokens": 7834072.0,
      "step": 4195
    },
    {
      "entropy": 6.225698471069336,
      "epoch": 0.37089367714588484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990984857578506,
      "loss": 6.0718,
      "mean_token_accuracy": 0.1523341119289398,
      "num_tokens": 7843334.0,
      "step": 4200
    },
    {
      "entropy": 6.040940427780152,
      "epoch": 0.37133521723772517,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004990956673336245,
      "loss": 6.0132,
      "mean_token_accuracy": 0.15985633432865143,
      "num_tokens": 7853187.0,
      "step": 4205
    },
    {
      "entropy": 6.121350336074829,
      "epoch": 0.3717767573295655,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990928445194917,
      "loss": 6.0836,
      "mean_token_accuracy": 0.1519768550992012,
      "num_tokens": 7862262.0,
      "step": 4210
    },
    {
      "entropy": 6.15352463722229,
      "epoch": 0.3722182974214059,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990900173155074,
      "loss": 6.0184,
      "mean_token_accuracy": 0.15704632550477982,
      "num_tokens": 7870725.0,
      "step": 4215
    },
    {
      "entropy": 6.124208784103393,
      "epoch": 0.3726598375132462,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990871857217273,
      "loss": 5.9447,
      "mean_token_accuracy": 0.15963577330112458,
      "num_tokens": 7879492.0,
      "step": 4220
    },
    {
      "entropy": 6.087597894668579,
      "epoch": 0.37310137760508655,
      "grad_norm": 1.375,
      "learning_rate": 0.0004990843497382066,
      "loss": 6.0649,
      "mean_token_accuracy": 0.15454517751932145,
      "num_tokens": 7888710.0,
      "step": 4225
    },
    {
      "entropy": 6.171204519271851,
      "epoch": 0.3735429176969269,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990815093650009,
      "loss": 6.0846,
      "mean_token_accuracy": 0.15838453769683838,
      "num_tokens": 7898313.0,
      "step": 4230
    },
    {
      "entropy": 6.163542985916138,
      "epoch": 0.3739844577887672,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004990786646021659,
      "loss": 6.0972,
      "mean_token_accuracy": 0.15132492929697036,
      "num_tokens": 7906726.0,
      "step": 4235
    },
    {
      "entropy": 6.143102693557739,
      "epoch": 0.37442599788060754,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990758154497573,
      "loss": 6.0083,
      "mean_token_accuracy": 0.16228035539388658,
      "num_tokens": 7915554.0,
      "step": 4240
    },
    {
      "entropy": 6.224862670898437,
      "epoch": 0.3748675379724479,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004990729619078309,
      "loss": 6.1357,
      "mean_token_accuracy": 0.15337878912687303,
      "num_tokens": 7924732.0,
      "step": 4245
    },
    {
      "entropy": 6.230613327026367,
      "epoch": 0.37530907806428826,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990701039764427,
      "loss": 6.061,
      "mean_token_accuracy": 0.1594015821814537,
      "num_tokens": 7933808.0,
      "step": 4250
    },
    {
      "entropy": 6.101580095291138,
      "epoch": 0.3757506181561286,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990672416556485,
      "loss": 6.0319,
      "mean_token_accuracy": 0.15845490843057633,
      "num_tokens": 7942558.0,
      "step": 4255
    },
    {
      "entropy": 6.129454803466797,
      "epoch": 0.3761921582479689,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004990643749455045,
      "loss": 6.1321,
      "mean_token_accuracy": 0.14685340449213982,
      "num_tokens": 7951951.0,
      "step": 4260
    },
    {
      "entropy": 6.2848834037780765,
      "epoch": 0.37663369833980925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990615038460667,
      "loss": 6.0247,
      "mean_token_accuracy": 0.1576920345425606,
      "num_tokens": 7960543.0,
      "step": 4265
    },
    {
      "entropy": 6.126399230957031,
      "epoch": 0.3770752384316496,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990586283573916,
      "loss": 6.007,
      "mean_token_accuracy": 0.155757275223732,
      "num_tokens": 7970273.0,
      "step": 4270
    },
    {
      "entropy": 6.158202123641968,
      "epoch": 0.3775167785234899,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990557484795355,
      "loss": 6.1038,
      "mean_token_accuracy": 0.1517634019255638,
      "num_tokens": 7978888.0,
      "step": 4275
    },
    {
      "entropy": 6.186517333984375,
      "epoch": 0.37795831861533025,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990528642125545,
      "loss": 6.0938,
      "mean_token_accuracy": 0.15008659660816193,
      "num_tokens": 7988843.0,
      "step": 4280
    },
    {
      "entropy": 6.1609944820404055,
      "epoch": 0.37839985870717063,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990499755565055,
      "loss": 6.0944,
      "mean_token_accuracy": 0.15281789302825927,
      "num_tokens": 7998403.0,
      "step": 4285
    },
    {
      "entropy": 6.190595436096191,
      "epoch": 0.37884139879901096,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990470825114448,
      "loss": 6.1108,
      "mean_token_accuracy": 0.15413309782743453,
      "num_tokens": 8007493.0,
      "step": 4290
    },
    {
      "entropy": 6.112000417709351,
      "epoch": 0.3792829388908513,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990441850774292,
      "loss": 6.0581,
      "mean_token_accuracy": 0.1584509640932083,
      "num_tokens": 8016399.0,
      "step": 4295
    },
    {
      "entropy": 6.215829849243164,
      "epoch": 0.3797244789826916,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990412832545155,
      "loss": 6.0591,
      "mean_token_accuracy": 0.1544952630996704,
      "num_tokens": 8025891.0,
      "step": 4300
    },
    {
      "entropy": 6.077403736114502,
      "epoch": 0.38016601907453196,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004990383770427603,
      "loss": 6.0097,
      "mean_token_accuracy": 0.15796091556549072,
      "num_tokens": 8034733.0,
      "step": 4305
    },
    {
      "entropy": 6.135886240005493,
      "epoch": 0.3806075591663723,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990354664422209,
      "loss": 6.0527,
      "mean_token_accuracy": 0.15982279628515245,
      "num_tokens": 8044443.0,
      "step": 4310
    },
    {
      "entropy": 6.160210800170899,
      "epoch": 0.3810490992582126,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004990325514529541,
      "loss": 6.0403,
      "mean_token_accuracy": 0.16555903106927872,
      "num_tokens": 8053920.0,
      "step": 4315
    },
    {
      "entropy": 6.1733404159545895,
      "epoch": 0.381490639350053,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990296320750169,
      "loss": 6.0242,
      "mean_token_accuracy": 0.15823598951101303,
      "num_tokens": 8063455.0,
      "step": 4320
    },
    {
      "entropy": 6.164728498458862,
      "epoch": 0.38193217944189334,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990267083084667,
      "loss": 6.1096,
      "mean_token_accuracy": 0.15378031879663467,
      "num_tokens": 8073353.0,
      "step": 4325
    },
    {
      "entropy": 6.216868829727173,
      "epoch": 0.38237371953373367,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990237801533607,
      "loss": 6.1198,
      "mean_token_accuracy": 0.1568579077720642,
      "num_tokens": 8082164.0,
      "step": 4330
    },
    {
      "entropy": 6.222427320480347,
      "epoch": 0.382815259625574,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004990208476097562,
      "loss": 6.126,
      "mean_token_accuracy": 0.15226055085659027,
      "num_tokens": 8090790.0,
      "step": 4335
    },
    {
      "entropy": 6.293479156494141,
      "epoch": 0.38325679971741433,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004990179106777109,
      "loss": 6.1346,
      "mean_token_accuracy": 0.14660235047340392,
      "num_tokens": 8100412.0,
      "step": 4340
    },
    {
      "entropy": 6.121231889724731,
      "epoch": 0.38369833980925466,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004990149693572819,
      "loss": 6.1328,
      "mean_token_accuracy": 0.14986878782510757,
      "num_tokens": 8110002.0,
      "step": 4345
    },
    {
      "entropy": 6.1944701194763185,
      "epoch": 0.384139879901095,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004990120236485271,
      "loss": 6.0262,
      "mean_token_accuracy": 0.15945157259702683,
      "num_tokens": 8118898.0,
      "step": 4350
    },
    {
      "entropy": 6.093376398086548,
      "epoch": 0.3845814199929354,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990090735515043,
      "loss": 6.0696,
      "mean_token_accuracy": 0.1521653488278389,
      "num_tokens": 8128207.0,
      "step": 4355
    },
    {
      "entropy": 6.162809753417969,
      "epoch": 0.3850229600847757,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499006119066271,
      "loss": 6.0659,
      "mean_token_accuracy": 0.15488530248403548,
      "num_tokens": 8137317.0,
      "step": 4360
    },
    {
      "entropy": 6.234450483322144,
      "epoch": 0.38546450017661604,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990031601928854,
      "loss": 6.0901,
      "mean_token_accuracy": 0.15392402857542037,
      "num_tokens": 8146519.0,
      "step": 4365
    },
    {
      "entropy": 6.081244421005249,
      "epoch": 0.3859060402684564,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990001969314051,
      "loss": 6.1156,
      "mean_token_accuracy": 0.150378455221653,
      "num_tokens": 8155916.0,
      "step": 4370
    },
    {
      "entropy": 6.188456583023071,
      "epoch": 0.3863475803602967,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989972292818884,
      "loss": 6.1306,
      "mean_token_accuracy": 0.15578001141548156,
      "num_tokens": 8165251.0,
      "step": 4375
    },
    {
      "entropy": 6.154488277435303,
      "epoch": 0.38678912045213704,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004989942572443934,
      "loss": 6.1046,
      "mean_token_accuracy": 0.154154072701931,
      "num_tokens": 8174761.0,
      "step": 4380
    },
    {
      "entropy": 6.2399732112884525,
      "epoch": 0.38723066054397737,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004989912808189784,
      "loss": 6.0624,
      "mean_token_accuracy": 0.1586390733718872,
      "num_tokens": 8182980.0,
      "step": 4385
    },
    {
      "entropy": 6.286262083053589,
      "epoch": 0.38767220063581775,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989883000057013,
      "loss": 6.1013,
      "mean_token_accuracy": 0.15117012858390808,
      "num_tokens": 8193306.0,
      "step": 4390
    },
    {
      "entropy": 6.023512887954712,
      "epoch": 0.3881137407276581,
      "grad_norm": 1.203125,
      "learning_rate": 0.000498985314804621,
      "loss": 5.9827,
      "mean_token_accuracy": 0.16275162994861603,
      "num_tokens": 8203446.0,
      "step": 4395
    },
    {
      "entropy": 6.129458618164063,
      "epoch": 0.3885552808194984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989823252157958,
      "loss": 6.0411,
      "mean_token_accuracy": 0.14927417337894439,
      "num_tokens": 8212414.0,
      "step": 4400
    },
    {
      "entropy": 6.107169580459595,
      "epoch": 0.38899682091133875,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989793312392841,
      "loss": 5.9633,
      "mean_token_accuracy": 0.16139311194419861,
      "num_tokens": 8221509.0,
      "step": 4405
    },
    {
      "entropy": 6.121396112442016,
      "epoch": 0.3894383610031791,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004989763328751448,
      "loss": 6.0825,
      "mean_token_accuracy": 0.15815991312265396,
      "num_tokens": 8230017.0,
      "step": 4410
    },
    {
      "entropy": 6.1887977600097654,
      "epoch": 0.3898799010950194,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004989733301234365,
      "loss": 6.0249,
      "mean_token_accuracy": 0.15967779457569123,
      "num_tokens": 8238853.0,
      "step": 4415
    },
    {
      "entropy": 6.224899578094482,
      "epoch": 0.39032144118685974,
      "grad_norm": 1.078125,
      "learning_rate": 0.000498970322984218,
      "loss": 6.1023,
      "mean_token_accuracy": 0.1519481733441353,
      "num_tokens": 8248112.0,
      "step": 4420
    },
    {
      "entropy": 6.203225469589233,
      "epoch": 0.39076298127870013,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004989673114575483,
      "loss": 6.0589,
      "mean_token_accuracy": 0.1536906696856022,
      "num_tokens": 8257166.0,
      "step": 4425
    },
    {
      "entropy": 6.128660726547241,
      "epoch": 0.39120452137054046,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004989642955434863,
      "loss": 6.0714,
      "mean_token_accuracy": 0.1515656217932701,
      "num_tokens": 8266229.0,
      "step": 4430
    },
    {
      "entropy": 6.163424253463745,
      "epoch": 0.3916460614623808,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004989612752420912,
      "loss": 6.039,
      "mean_token_accuracy": 0.15465213656425475,
      "num_tokens": 8275102.0,
      "step": 4435
    },
    {
      "entropy": 6.022084140777588,
      "epoch": 0.3920876015542211,
      "grad_norm": 1.203125,
      "learning_rate": 0.000498958250553422,
      "loss": 5.9631,
      "mean_token_accuracy": 0.16303362101316451,
      "num_tokens": 8283847.0,
      "step": 4440
    },
    {
      "entropy": 6.193567132949829,
      "epoch": 0.39252914164606145,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004989552214775381,
      "loss": 6.0871,
      "mean_token_accuracy": 0.15131851583719252,
      "num_tokens": 8292622.0,
      "step": 4445
    },
    {
      "entropy": 6.194511032104492,
      "epoch": 0.3929706817379018,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004989521880144988,
      "loss": 5.9982,
      "mean_token_accuracy": 0.17004551142454147,
      "num_tokens": 8301026.0,
      "step": 4450
    },
    {
      "entropy": 6.184074640274048,
      "epoch": 0.3934122218297421,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004989491501643635,
      "loss": 6.2578,
      "mean_token_accuracy": 0.14575279951095582,
      "num_tokens": 8310977.0,
      "step": 4455
    },
    {
      "entropy": 6.20936131477356,
      "epoch": 0.3938537619215825,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004989461079271916,
      "loss": 6.0296,
      "mean_token_accuracy": 0.16169211864471436,
      "num_tokens": 8319391.0,
      "step": 4460
    },
    {
      "entropy": 6.169527339935303,
      "epoch": 0.39429530201342283,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989430613030429,
      "loss": 5.9922,
      "mean_token_accuracy": 0.15554805994033813,
      "num_tokens": 8328639.0,
      "step": 4465
    },
    {
      "entropy": 6.117289400100708,
      "epoch": 0.39473684210526316,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000498940010291977,
      "loss": 6.096,
      "mean_token_accuracy": 0.15155849754810333,
      "num_tokens": 8338190.0,
      "step": 4470
    },
    {
      "entropy": 6.100512361526489,
      "epoch": 0.3951783821971035,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004989369548940536,
      "loss": 6.0395,
      "mean_token_accuracy": 0.15572706907987593,
      "num_tokens": 8346874.0,
      "step": 4475
    },
    {
      "entropy": 6.190318775177002,
      "epoch": 0.3956199222889438,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004989338951093327,
      "loss": 6.0178,
      "mean_token_accuracy": 0.16222874522209169,
      "num_tokens": 8356446.0,
      "step": 4480
    },
    {
      "entropy": 6.1148622035980225,
      "epoch": 0.39606146238078416,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989308309378741,
      "loss": 6.0199,
      "mean_token_accuracy": 0.15970418155193328,
      "num_tokens": 8365694.0,
      "step": 4485
    },
    {
      "entropy": 6.157066297531128,
      "epoch": 0.3965030024726245,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004989277623797379,
      "loss": 6.0754,
      "mean_token_accuracy": 0.1534503474831581,
      "num_tokens": 8374282.0,
      "step": 4490
    },
    {
      "entropy": 6.160027027130127,
      "epoch": 0.3969445425644649,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004989246894349841,
      "loss": 6.0372,
      "mean_token_accuracy": 0.1563117504119873,
      "num_tokens": 8383315.0,
      "step": 4495
    },
    {
      "entropy": 6.214316082000733,
      "epoch": 0.3973860826563052,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989216121036732,
      "loss": 6.0236,
      "mean_token_accuracy": 0.15833714008331298,
      "num_tokens": 8392263.0,
      "step": 4500
    },
    {
      "entropy": 6.106395578384399,
      "epoch": 0.39782762274814554,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004989185303858651,
      "loss": 6.0696,
      "mean_token_accuracy": 0.14981550127267837,
      "num_tokens": 8400734.0,
      "step": 4505
    },
    {
      "entropy": 6.156090450286865,
      "epoch": 0.39826916283998587,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004989154442816203,
      "loss": 6.0793,
      "mean_token_accuracy": 0.15283239632844925,
      "num_tokens": 8410635.0,
      "step": 4510
    },
    {
      "entropy": 6.205539274215698,
      "epoch": 0.3987107029318262,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004989123537909994,
      "loss": 6.0664,
      "mean_token_accuracy": 0.15485348254442216,
      "num_tokens": 8420111.0,
      "step": 4515
    },
    {
      "entropy": 6.103611946105957,
      "epoch": 0.39915224302366653,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004989092589140629,
      "loss": 6.0177,
      "mean_token_accuracy": 0.1484901040792465,
      "num_tokens": 8429459.0,
      "step": 4520
    },
    {
      "entropy": 6.117227792739868,
      "epoch": 0.39959378311550686,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004989061596508712,
      "loss": 6.0403,
      "mean_token_accuracy": 0.16086821481585503,
      "num_tokens": 8438083.0,
      "step": 4525
    },
    {
      "entropy": 6.07325234413147,
      "epoch": 0.40003532320734725,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004989030560014853,
      "loss": 6.0506,
      "mean_token_accuracy": 0.15744656324386597,
      "num_tokens": 8447713.0,
      "step": 4530
    },
    {
      "entropy": 6.1523223400115965,
      "epoch": 0.4004768632991876,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004988999479659657,
      "loss": 6.0226,
      "mean_token_accuracy": 0.158653824031353,
      "num_tokens": 8457394.0,
      "step": 4535
    },
    {
      "entropy": 6.117883396148682,
      "epoch": 0.4009184033910279,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004988968355443737,
      "loss": 5.9913,
      "mean_token_accuracy": 0.16181344538927078,
      "num_tokens": 8467333.0,
      "step": 4540
    },
    {
      "entropy": 6.124732971191406,
      "epoch": 0.40135994348286824,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004988937187367699,
      "loss": 6.1032,
      "mean_token_accuracy": 0.15110900700092317,
      "num_tokens": 8477530.0,
      "step": 4545
    },
    {
      "entropy": 6.217570829391479,
      "epoch": 0.4018014835747086,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988905975432154,
      "loss": 6.1802,
      "mean_token_accuracy": 0.15269524306058885,
      "num_tokens": 8486861.0,
      "step": 4550
    },
    {
      "entropy": 6.170765733718872,
      "epoch": 0.4022430236665489,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988874719637715,
      "loss": 5.9867,
      "mean_token_accuracy": 0.158355513215065,
      "num_tokens": 8496541.0,
      "step": 4555
    },
    {
      "entropy": 6.052766561508179,
      "epoch": 0.40268456375838924,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004988843419984994,
      "loss": 6.064,
      "mean_token_accuracy": 0.15494307354092599,
      "num_tokens": 8505667.0,
      "step": 4560
    },
    {
      "entropy": 6.24229063987732,
      "epoch": 0.4031261038502296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004988812076474604,
      "loss": 6.052,
      "mean_token_accuracy": 0.15319542214274406,
      "num_tokens": 8515133.0,
      "step": 4565
    },
    {
      "entropy": 6.170625066757202,
      "epoch": 0.40356764394206995,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004988780689107158,
      "loss": 6.007,
      "mean_token_accuracy": 0.1620650038123131,
      "num_tokens": 8524797.0,
      "step": 4570
    },
    {
      "entropy": 6.0161412239074705,
      "epoch": 0.4040091840339103,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004988749257883271,
      "loss": 6.0288,
      "mean_token_accuracy": 0.15016194060444832,
      "num_tokens": 8534667.0,
      "step": 4575
    },
    {
      "entropy": 6.143600559234619,
      "epoch": 0.4044507241257506,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498871778280356,
      "loss": 6.0166,
      "mean_token_accuracy": 0.15943924337625504,
      "num_tokens": 8543874.0,
      "step": 4580
    },
    {
      "entropy": 6.1106805324554445,
      "epoch": 0.40489226421759095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988686263868641,
      "loss": 6.0353,
      "mean_token_accuracy": 0.1529652863740921,
      "num_tokens": 8553620.0,
      "step": 4585
    },
    {
      "entropy": 6.17731556892395,
      "epoch": 0.4053338043094313,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004988654701079131,
      "loss": 6.1113,
      "mean_token_accuracy": 0.15474483817815782,
      "num_tokens": 8563857.0,
      "step": 4590
    },
    {
      "entropy": 6.2202249526977536,
      "epoch": 0.4057753444012716,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004988623094435649,
      "loss": 6.0898,
      "mean_token_accuracy": 0.1531184583902359,
      "num_tokens": 8572677.0,
      "step": 4595
    },
    {
      "entropy": 6.140727758407593,
      "epoch": 0.406216884493112,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988591443938813,
      "loss": 6.016,
      "mean_token_accuracy": 0.15535678565502167,
      "num_tokens": 8581907.0,
      "step": 4600
    },
    {
      "entropy": 6.138199281692505,
      "epoch": 0.4066584245849523,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004988559749589244,
      "loss": 6.0609,
      "mean_token_accuracy": 0.15539143681526185,
      "num_tokens": 8591701.0,
      "step": 4605
    },
    {
      "entropy": 6.19168872833252,
      "epoch": 0.40709996467679266,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988528011387563,
      "loss": 6.1128,
      "mean_token_accuracy": 0.15955741629004477,
      "num_tokens": 8601054.0,
      "step": 4610
    },
    {
      "entropy": 6.115007019042968,
      "epoch": 0.407541504768633,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004988496229334392,
      "loss": 6.0089,
      "mean_token_accuracy": 0.16271338164806365,
      "num_tokens": 8610346.0,
      "step": 4615
    },
    {
      "entropy": 6.082312822341919,
      "epoch": 0.4079830448604733,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004988464403430352,
      "loss": 6.0114,
      "mean_token_accuracy": 0.1530932977795601,
      "num_tokens": 8620823.0,
      "step": 4620
    },
    {
      "entropy": 6.228973913192749,
      "epoch": 0.40842458495231365,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988432533676067,
      "loss": 6.1457,
      "mean_token_accuracy": 0.14871701523661612,
      "num_tokens": 8630184.0,
      "step": 4625
    },
    {
      "entropy": 6.265304517745972,
      "epoch": 0.408866125044154,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004988400620072163,
      "loss": 6.1541,
      "mean_token_accuracy": 0.14276653826236724,
      "num_tokens": 8640064.0,
      "step": 4630
    },
    {
      "entropy": 6.173982954025268,
      "epoch": 0.40930766513599437,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004988368662619263,
      "loss": 6.0413,
      "mean_token_accuracy": 0.15715423077344895,
      "num_tokens": 8650503.0,
      "step": 4635
    },
    {
      "entropy": 6.191125011444091,
      "epoch": 0.4097492052278347,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004988336661317994,
      "loss": 6.0652,
      "mean_token_accuracy": 0.15463789254426957,
      "num_tokens": 8659125.0,
      "step": 4640
    },
    {
      "entropy": 6.157907199859619,
      "epoch": 0.41019074531967503,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988304616168984,
      "loss": 6.0841,
      "mean_token_accuracy": 0.15441161543130874,
      "num_tokens": 8668193.0,
      "step": 4645
    },
    {
      "entropy": 6.116786527633667,
      "epoch": 0.41063228541151536,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004988272527172858,
      "loss": 6.0384,
      "mean_token_accuracy": 0.15524010509252548,
      "num_tokens": 8677515.0,
      "step": 4650
    },
    {
      "entropy": 6.21679277420044,
      "epoch": 0.4110738255033557,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004988240394330246,
      "loss": 6.0764,
      "mean_token_accuracy": 0.14973534047603607,
      "num_tokens": 8687549.0,
      "step": 4655
    },
    {
      "entropy": 6.157021522521973,
      "epoch": 0.411515365595196,
      "grad_norm": 1.375,
      "learning_rate": 0.0004988208217641778,
      "loss": 6.0639,
      "mean_token_accuracy": 0.15594548732042313,
      "num_tokens": 8697371.0,
      "step": 4660
    },
    {
      "entropy": 6.13969612121582,
      "epoch": 0.41195690568703636,
      "grad_norm": 1.5,
      "learning_rate": 0.0004988175997108086,
      "loss": 6.0735,
      "mean_token_accuracy": 0.15609194859862327,
      "num_tokens": 8707193.0,
      "step": 4665
    },
    {
      "entropy": 6.1602945804595945,
      "epoch": 0.41239844577887674,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004988143732729797,
      "loss": 6.0644,
      "mean_token_accuracy": 0.15450926274061202,
      "num_tokens": 8716052.0,
      "step": 4670
    },
    {
      "entropy": 6.081143188476562,
      "epoch": 0.4128399858707171,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004988111424507546,
      "loss": 6.0209,
      "mean_token_accuracy": 0.1590244859457016,
      "num_tokens": 8726140.0,
      "step": 4675
    },
    {
      "entropy": 6.134521245956421,
      "epoch": 0.4132815259625574,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004988079072441964,
      "loss": 6.0028,
      "mean_token_accuracy": 0.16453344523906707,
      "num_tokens": 8735299.0,
      "step": 4680
    },
    {
      "entropy": 6.157465314865112,
      "epoch": 0.41372306605439774,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004988046676533687,
      "loss": 6.0908,
      "mean_token_accuracy": 0.15370513945817948,
      "num_tokens": 8744686.0,
      "step": 4685
    },
    {
      "entropy": 6.140842294692993,
      "epoch": 0.41416460614623807,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004988014236783347,
      "loss": 6.063,
      "mean_token_accuracy": 0.1632717102766037,
      "num_tokens": 8754942.0,
      "step": 4690
    },
    {
      "entropy": 6.232805824279785,
      "epoch": 0.4146061462380784,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004987981753191582,
      "loss": 6.0523,
      "mean_token_accuracy": 0.15246548503637314,
      "num_tokens": 8764054.0,
      "step": 4695
    },
    {
      "entropy": 6.0288361549377445,
      "epoch": 0.41504768632991873,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004987949225759027,
      "loss": 5.9405,
      "mean_token_accuracy": 0.15906094312667846,
      "num_tokens": 8773050.0,
      "step": 4700
    },
    {
      "entropy": 6.1974766731262205,
      "epoch": 0.4154892264217591,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004987916654486321,
      "loss": 6.0544,
      "mean_token_accuracy": 0.1537775442004204,
      "num_tokens": 8782476.0,
      "step": 4705
    },
    {
      "entropy": 6.254334783554077,
      "epoch": 0.41593076651359945,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004987884039374099,
      "loss": 6.1212,
      "mean_token_accuracy": 0.1502728283405304,
      "num_tokens": 8791147.0,
      "step": 4710
    },
    {
      "entropy": 6.237736749649048,
      "epoch": 0.4163723066054398,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004987851380423001,
      "loss": 6.1069,
      "mean_token_accuracy": 0.1571350358426571,
      "num_tokens": 8801151.0,
      "step": 4715
    },
    {
      "entropy": 6.067666149139404,
      "epoch": 0.4168138466972801,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004987818677633668,
      "loss": 6.0577,
      "mean_token_accuracy": 0.14970119222998618,
      "num_tokens": 8809587.0,
      "step": 4720
    },
    {
      "entropy": 6.103376960754394,
      "epoch": 0.41725538678912044,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000498778593100674,
      "loss": 5.94,
      "mean_token_accuracy": 0.15202507078647615,
      "num_tokens": 8818168.0,
      "step": 4725
    },
    {
      "entropy": 6.131261110305786,
      "epoch": 0.4176969268809608,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987753140542857,
      "loss": 6.0105,
      "mean_token_accuracy": 0.16188293397426606,
      "num_tokens": 8827477.0,
      "step": 4730
    },
    {
      "entropy": 6.104963493347168,
      "epoch": 0.4181384669728011,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004987720306242664,
      "loss": 5.9847,
      "mean_token_accuracy": 0.16286925673484803,
      "num_tokens": 8837067.0,
      "step": 4735
    },
    {
      "entropy": 6.133073472976685,
      "epoch": 0.4185800070646415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004987687428106803,
      "loss": 6.0872,
      "mean_token_accuracy": 0.16172488033771515,
      "num_tokens": 8845790.0,
      "step": 4740
    },
    {
      "entropy": 6.185379123687744,
      "epoch": 0.4190215471564818,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004987654506135917,
      "loss": 6.0226,
      "mean_token_accuracy": 0.15944662541151047,
      "num_tokens": 8855242.0,
      "step": 4745
    },
    {
      "entropy": 6.192673587799073,
      "epoch": 0.41946308724832215,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004987621540330652,
      "loss": 6.0723,
      "mean_token_accuracy": 0.15567026063799858,
      "num_tokens": 8864459.0,
      "step": 4750
    },
    {
      "entropy": 6.19939341545105,
      "epoch": 0.4199046273401625,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004987588530691653,
      "loss": 6.1385,
      "mean_token_accuracy": 0.14711003005504608,
      "num_tokens": 8875028.0,
      "step": 4755
    },
    {
      "entropy": 6.135413789749146,
      "epoch": 0.4203461674320028,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004987555477219569,
      "loss": 5.9786,
      "mean_token_accuracy": 0.15931818783283233,
      "num_tokens": 8883857.0,
      "step": 4760
    },
    {
      "entropy": 6.110503196716309,
      "epoch": 0.42078770752384315,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004987522379915045,
      "loss": 5.9705,
      "mean_token_accuracy": 0.15814343243837356,
      "num_tokens": 8893499.0,
      "step": 4765
    },
    {
      "entropy": 6.095895624160766,
      "epoch": 0.4212292476156835,
      "grad_norm": 1.25,
      "learning_rate": 0.000498748923877873,
      "loss": 5.9891,
      "mean_token_accuracy": 0.1591852620244026,
      "num_tokens": 8903368.0,
      "step": 4770
    },
    {
      "entropy": 6.126567029953003,
      "epoch": 0.42167078770752386,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004987456053811273,
      "loss": 6.0966,
      "mean_token_accuracy": 0.1539273589849472,
      "num_tokens": 8912701.0,
      "step": 4775
    },
    {
      "entropy": 6.1158490657806395,
      "epoch": 0.4221123277993642,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004987422825013325,
      "loss": 6.0444,
      "mean_token_accuracy": 0.15109995752573013,
      "num_tokens": 8921962.0,
      "step": 4780
    },
    {
      "entropy": 6.1833864688873295,
      "epoch": 0.4225538678912045,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004987389552385536,
      "loss": 6.0307,
      "mean_token_accuracy": 0.15406385958194732,
      "num_tokens": 8931923.0,
      "step": 4785
    },
    {
      "entropy": 6.131430625915527,
      "epoch": 0.42299540798304486,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004987356235928558,
      "loss": 6.0635,
      "mean_token_accuracy": 0.1470622941851616,
      "num_tokens": 8940403.0,
      "step": 4790
    },
    {
      "entropy": 6.131550025939942,
      "epoch": 0.4234369480748852,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004987322875643044,
      "loss": 5.9887,
      "mean_token_accuracy": 0.16142310500144957,
      "num_tokens": 8949377.0,
      "step": 4795
    },
    {
      "entropy": 6.2069591045379635,
      "epoch": 0.4238784881667255,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004987289471529647,
      "loss": 6.1304,
      "mean_token_accuracy": 0.1479768604040146,
      "num_tokens": 8958719.0,
      "step": 4800
    },
    {
      "entropy": 6.165348720550537,
      "epoch": 0.42432002825856585,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004987256023589022,
      "loss": 6.1048,
      "mean_token_accuracy": 0.14835046380758285,
      "num_tokens": 8968226.0,
      "step": 4805
    },
    {
      "entropy": 6.13749942779541,
      "epoch": 0.42476156835040624,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004987222531821824,
      "loss": 6.0528,
      "mean_token_accuracy": 0.14842675924301146,
      "num_tokens": 8976670.0,
      "step": 4810
    },
    {
      "entropy": 6.09095401763916,
      "epoch": 0.42520310844224657,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987188996228709,
      "loss": 5.9901,
      "mean_token_accuracy": 0.16065036058425902,
      "num_tokens": 8986185.0,
      "step": 4815
    },
    {
      "entropy": 6.155765628814697,
      "epoch": 0.4256446485340869,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004987155416810334,
      "loss": 5.9644,
      "mean_token_accuracy": 0.16488435715436936,
      "num_tokens": 8995124.0,
      "step": 4820
    },
    {
      "entropy": 6.094462633132935,
      "epoch": 0.42608618862592723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004987121793567356,
      "loss": 6.0704,
      "mean_token_accuracy": 0.15586088821291924,
      "num_tokens": 9004380.0,
      "step": 4825
    },
    {
      "entropy": 6.142639970779419,
      "epoch": 0.42652772871776756,
      "grad_norm": 1.25,
      "learning_rate": 0.0004987088126500436,
      "loss": 5.963,
      "mean_token_accuracy": 0.16008084118366242,
      "num_tokens": 9013791.0,
      "step": 4830
    },
    {
      "entropy": 6.069281530380249,
      "epoch": 0.4269692688096079,
      "grad_norm": 1.234375,
      "learning_rate": 0.000498705441561023,
      "loss": 5.9998,
      "mean_token_accuracy": 0.16714757829904556,
      "num_tokens": 9023076.0,
      "step": 4835
    },
    {
      "entropy": 6.315524005889893,
      "epoch": 0.4274108089014482,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004987020660897401,
      "loss": 6.1089,
      "mean_token_accuracy": 0.14946657419204712,
      "num_tokens": 9032720.0,
      "step": 4840
    },
    {
      "entropy": 6.199988222122192,
      "epoch": 0.4278523489932886,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000498698686236261,
      "loss": 6.0577,
      "mean_token_accuracy": 0.1480626255273819,
      "num_tokens": 9042652.0,
      "step": 4845
    },
    {
      "entropy": 6.085298490524292,
      "epoch": 0.42829388908512894,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004986953020006519,
      "loss": 6.1376,
      "mean_token_accuracy": 0.14969860166311263,
      "num_tokens": 9052172.0,
      "step": 4850
    },
    {
      "entropy": 6.134743070602417,
      "epoch": 0.4287354291769693,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986919133829788,
      "loss": 5.9956,
      "mean_token_accuracy": 0.15985623747110367,
      "num_tokens": 9061798.0,
      "step": 4855
    },
    {
      "entropy": 6.040613985061645,
      "epoch": 0.4291769692688096,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004986885203833086,
      "loss": 5.9499,
      "mean_token_accuracy": 0.15626863837242128,
      "num_tokens": 9070429.0,
      "step": 4860
    },
    {
      "entropy": 6.098531293869018,
      "epoch": 0.42961850936064994,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004986851230017075,
      "loss": 6.0492,
      "mean_token_accuracy": 0.16342882812023163,
      "num_tokens": 9080200.0,
      "step": 4865
    },
    {
      "entropy": 6.163259077072143,
      "epoch": 0.43006004945249027,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004986817212382419,
      "loss": 6.0382,
      "mean_token_accuracy": 0.15508455336093901,
      "num_tokens": 9089555.0,
      "step": 4870
    },
    {
      "entropy": 6.126989889144897,
      "epoch": 0.4305015895443306,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986783150929786,
      "loss": 6.0263,
      "mean_token_accuracy": 0.15248029232025145,
      "num_tokens": 9099091.0,
      "step": 4875
    },
    {
      "entropy": 6.172487115859985,
      "epoch": 0.430943129636171,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986749045659845,
      "loss": 6.1075,
      "mean_token_accuracy": 0.14817112535238267,
      "num_tokens": 9109008.0,
      "step": 4880
    },
    {
      "entropy": 6.152895927429199,
      "epoch": 0.4313846697280113,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004986714896573261,
      "loss": 5.9513,
      "mean_token_accuracy": 0.16084639877080917,
      "num_tokens": 9117882.0,
      "step": 4885
    },
    {
      "entropy": 6.139704322814941,
      "epoch": 0.43182620981985165,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004986680703670704,
      "loss": 6.1215,
      "mean_token_accuracy": 0.15487379878759383,
      "num_tokens": 9126860.0,
      "step": 4890
    },
    {
      "entropy": 6.141199541091919,
      "epoch": 0.432267749911692,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004986646466952845,
      "loss": 6.0119,
      "mean_token_accuracy": 0.15635189563035964,
      "num_tokens": 9135819.0,
      "step": 4895
    },
    {
      "entropy": 6.10592885017395,
      "epoch": 0.4327092900035323,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004986612186420353,
      "loss": 5.9934,
      "mean_token_accuracy": 0.15510470867156984,
      "num_tokens": 9145302.0,
      "step": 4900
    },
    {
      "entropy": 6.044864749908447,
      "epoch": 0.43315083009537264,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986577862073901,
      "loss": 6.0575,
      "mean_token_accuracy": 0.15394357293844224,
      "num_tokens": 9154667.0,
      "step": 4905
    },
    {
      "entropy": 6.2363903522491455,
      "epoch": 0.43359237018721297,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004986543493914159,
      "loss": 6.098,
      "mean_token_accuracy": 0.14959411323070526,
      "num_tokens": 9164562.0,
      "step": 4910
    },
    {
      "entropy": 6.208088731765747,
      "epoch": 0.43403391027905336,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004986509081941805,
      "loss": 6.1071,
      "mean_token_accuracy": 0.15276289731264114,
      "num_tokens": 9174872.0,
      "step": 4915
    },
    {
      "entropy": 6.079789733886718,
      "epoch": 0.4344754503708937,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004986474626157507,
      "loss": 5.862,
      "mean_token_accuracy": 0.17531196177005767,
      "num_tokens": 9184322.0,
      "step": 4920
    },
    {
      "entropy": 6.058652734756469,
      "epoch": 0.434916990462734,
      "grad_norm": 1.25,
      "learning_rate": 0.0004986440126561945,
      "loss": 5.9894,
      "mean_token_accuracy": 0.15689075142145156,
      "num_tokens": 9194450.0,
      "step": 4925
    },
    {
      "entropy": 6.151858711242676,
      "epoch": 0.43535853055457435,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004986405583155792,
      "loss": 6.0305,
      "mean_token_accuracy": 0.15203123837709426,
      "num_tokens": 9203658.0,
      "step": 4930
    },
    {
      "entropy": 6.100604057312012,
      "epoch": 0.4358000706464147,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986370995939725,
      "loss": 6.0239,
      "mean_token_accuracy": 0.15403898507356645,
      "num_tokens": 9213609.0,
      "step": 4935
    },
    {
      "entropy": 6.112157011032105,
      "epoch": 0.436241610738255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004986336364914423,
      "loss": 6.0414,
      "mean_token_accuracy": 0.15200137123465537,
      "num_tokens": 9222704.0,
      "step": 4940
    },
    {
      "entropy": 6.147404766082763,
      "epoch": 0.43668315083009535,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004986301690080564,
      "loss": 6.0472,
      "mean_token_accuracy": 0.1494756668806076,
      "num_tokens": 9231599.0,
      "step": 4945
    },
    {
      "entropy": 6.176404428482056,
      "epoch": 0.43712469092193573,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986266971438826,
      "loss": 6.0761,
      "mean_token_accuracy": 0.15215079635381698,
      "num_tokens": 9241886.0,
      "step": 4950
    },
    {
      "entropy": 6.142639207839966,
      "epoch": 0.43756623101377606,
      "grad_norm": 1.34375,
      "learning_rate": 0.000498623220898989,
      "loss": 6.0651,
      "mean_token_accuracy": 0.1492701292037964,
      "num_tokens": 9251085.0,
      "step": 4955
    },
    {
      "entropy": 6.155983543395996,
      "epoch": 0.4380077711056164,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004986197402734436,
      "loss": 6.0272,
      "mean_token_accuracy": 0.15382544845342636,
      "num_tokens": 9259601.0,
      "step": 4960
    },
    {
      "entropy": 6.18567385673523,
      "epoch": 0.4384493111974567,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004986162552673148,
      "loss": 6.0699,
      "mean_token_accuracy": 0.15705521255731583,
      "num_tokens": 9268935.0,
      "step": 4965
    },
    {
      "entropy": 6.133043241500855,
      "epoch": 0.43889085128929706,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004986127658806706,
      "loss": 6.0813,
      "mean_token_accuracy": 0.15293170362710953,
      "num_tokens": 9277647.0,
      "step": 4970
    },
    {
      "entropy": 6.120587873458862,
      "epoch": 0.4393323913811374,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004986092721135796,
      "loss": 6.0199,
      "mean_token_accuracy": 0.15239207521080972,
      "num_tokens": 9286610.0,
      "step": 4975
    },
    {
      "entropy": 6.135066556930542,
      "epoch": 0.4397739314729777,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004986057739661101,
      "loss": 6.1032,
      "mean_token_accuracy": 0.1508208692073822,
      "num_tokens": 9295946.0,
      "step": 4980
    },
    {
      "entropy": 6.1115028858184814,
      "epoch": 0.4402154715648181,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004986022714383307,
      "loss": 6.0042,
      "mean_token_accuracy": 0.1543491631746292,
      "num_tokens": 9304903.0,
      "step": 4985
    },
    {
      "entropy": 6.15944766998291,
      "epoch": 0.44065701165665844,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004985987645303099,
      "loss": 5.9185,
      "mean_token_accuracy": 0.16130532771348954,
      "num_tokens": 9313606.0,
      "step": 4990
    },
    {
      "entropy": 6.016908359527588,
      "epoch": 0.44109855174849877,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004985952532421164,
      "loss": 6.0275,
      "mean_token_accuracy": 0.15621849447488784,
      "num_tokens": 9322815.0,
      "step": 4995
    },
    {
      "entropy": 6.17736701965332,
      "epoch": 0.4415400918403391,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004985917375738193,
      "loss": 5.9593,
      "mean_token_accuracy": 0.15536014586687089,
      "num_tokens": 9332630.0,
      "step": 5000
    },
    {
      "entropy": 6.067361211776733,
      "epoch": 0.44198163193217943,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004985882175254871,
      "loss": 5.9846,
      "mean_token_accuracy": 0.15410226881504058,
      "num_tokens": 9342216.0,
      "step": 5005
    },
    {
      "entropy": 6.146738815307617,
      "epoch": 0.44242317202401976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004985846930971887,
      "loss": 6.0431,
      "mean_token_accuracy": 0.1559235379099846,
      "num_tokens": 9352295.0,
      "step": 5010
    },
    {
      "entropy": 6.107598447799683,
      "epoch": 0.4428647121158601,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004985811642889937,
      "loss": 6.0348,
      "mean_token_accuracy": 0.15718846172094345,
      "num_tokens": 9361462.0,
      "step": 5015
    },
    {
      "entropy": 6.144611740112305,
      "epoch": 0.4433062522077005,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004985776311009705,
      "loss": 5.9684,
      "mean_token_accuracy": 0.15345567613840103,
      "num_tokens": 9370379.0,
      "step": 5020
    },
    {
      "entropy": 6.103248453140258,
      "epoch": 0.4437477922995408,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004985740935331888,
      "loss": 6.0098,
      "mean_token_accuracy": 0.15823583900928498,
      "num_tokens": 9379255.0,
      "step": 5025
    },
    {
      "entropy": 6.085464525222778,
      "epoch": 0.44418933239138114,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004985705515857177,
      "loss": 5.99,
      "mean_token_accuracy": 0.14899933189153672,
      "num_tokens": 9389313.0,
      "step": 5030
    },
    {
      "entropy": 6.020923233032226,
      "epoch": 0.4446308724832215,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004985670052586268,
      "loss": 5.7916,
      "mean_token_accuracy": 0.17029385417699813,
      "num_tokens": 9397778.0,
      "step": 5035
    },
    {
      "entropy": 6.082318449020386,
      "epoch": 0.4450724125750618,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004985634545519853,
      "loss": 6.0589,
      "mean_token_accuracy": 0.15298160612583162,
      "num_tokens": 9407831.0,
      "step": 5040
    },
    {
      "entropy": 6.0772803783416744,
      "epoch": 0.44551395266690214,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004985598994658629,
      "loss": 5.9669,
      "mean_token_accuracy": 0.15956881046295165,
      "num_tokens": 9418458.0,
      "step": 5045
    },
    {
      "entropy": 6.016619396209717,
      "epoch": 0.44595549275874247,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004985563400003291,
      "loss": 5.8707,
      "mean_token_accuracy": 0.17422997653484346,
      "num_tokens": 9426911.0,
      "step": 5050
    },
    {
      "entropy": 6.14707236289978,
      "epoch": 0.44639703285058285,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004985527761554539,
      "loss": 6.0541,
      "mean_token_accuracy": 0.1595864400267601,
      "num_tokens": 9435896.0,
      "step": 5055
    },
    {
      "entropy": 6.207449245452881,
      "epoch": 0.4468385729424232,
      "grad_norm": 1.296875,
      "learning_rate": 0.000498549207931307,
      "loss": 6.0612,
      "mean_token_accuracy": 0.15011950582265854,
      "num_tokens": 9445272.0,
      "step": 5060
    },
    {
      "entropy": 6.06727409362793,
      "epoch": 0.4472801130342635,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004985456353279581,
      "loss": 6.0228,
      "mean_token_accuracy": 0.15027147233486177,
      "num_tokens": 9455377.0,
      "step": 5065
    },
    {
      "entropy": 6.108924150466919,
      "epoch": 0.44772165312610385,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004985420583454774,
      "loss": 6.0477,
      "mean_token_accuracy": 0.1528654247522354,
      "num_tokens": 9464918.0,
      "step": 5070
    },
    {
      "entropy": 6.118135070800781,
      "epoch": 0.4481631932179442,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004985384769839349,
      "loss": 6.0416,
      "mean_token_accuracy": 0.15041410326957702,
      "num_tokens": 9473322.0,
      "step": 5075
    },
    {
      "entropy": 6.191469526290893,
      "epoch": 0.4486047333097845,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004985348912434008,
      "loss": 5.969,
      "mean_token_accuracy": 0.16189600080251693,
      "num_tokens": 9482255.0,
      "step": 5080
    },
    {
      "entropy": 6.092324352264404,
      "epoch": 0.44904627340162484,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004985313011239452,
      "loss": 5.9959,
      "mean_token_accuracy": 0.1581245869398117,
      "num_tokens": 9491709.0,
      "step": 5085
    },
    {
      "entropy": 6.024058151245117,
      "epoch": 0.4494878134934652,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004985277066256388,
      "loss": 5.9807,
      "mean_token_accuracy": 0.16034325063228608,
      "num_tokens": 9500594.0,
      "step": 5090
    },
    {
      "entropy": 6.112436008453369,
      "epoch": 0.44992935358530556,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004985241077485515,
      "loss": 6.0018,
      "mean_token_accuracy": 0.15738717019557952,
      "num_tokens": 9509088.0,
      "step": 5095
    },
    {
      "entropy": 6.149496126174927,
      "epoch": 0.4503708936771459,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004985205044927541,
      "loss": 5.9725,
      "mean_token_accuracy": 0.15938366055488587,
      "num_tokens": 9517776.0,
      "step": 5100
    },
    {
      "entropy": 6.069735097885132,
      "epoch": 0.4508124337689862,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004985168968583173,
      "loss": 6.0348,
      "mean_token_accuracy": 0.15837667435407637,
      "num_tokens": 9527080.0,
      "step": 5105
    },
    {
      "entropy": 6.106137084960937,
      "epoch": 0.45125397386082655,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004985132848453114,
      "loss": 5.8952,
      "mean_token_accuracy": 0.16738586127758026,
      "num_tokens": 9536358.0,
      "step": 5110
    },
    {
      "entropy": 6.04980616569519,
      "epoch": 0.4516955139526669,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004985096684538075,
      "loss": 5.9813,
      "mean_token_accuracy": 0.15662433505058287,
      "num_tokens": 9545528.0,
      "step": 5115
    },
    {
      "entropy": 6.060785722732544,
      "epoch": 0.4521370540445072,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004985060476838763,
      "loss": 6.0113,
      "mean_token_accuracy": 0.15891691744327546,
      "num_tokens": 9554433.0,
      "step": 5120
    },
    {
      "entropy": 6.073312520980835,
      "epoch": 0.4525785941363476,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004985024225355887,
      "loss": 5.9656,
      "mean_token_accuracy": 0.1543935567140579,
      "num_tokens": 9563932.0,
      "step": 5125
    },
    {
      "entropy": 6.100546646118164,
      "epoch": 0.45302013422818793,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004984987930090158,
      "loss": 5.9552,
      "mean_token_accuracy": 0.15897123962640763,
      "num_tokens": 9572829.0,
      "step": 5130
    },
    {
      "entropy": 6.025969839096069,
      "epoch": 0.45346167432002826,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004984951591042285,
      "loss": 6.0045,
      "mean_token_accuracy": 0.16426790058612822,
      "num_tokens": 9583597.0,
      "step": 5135
    },
    {
      "entropy": 6.24551477432251,
      "epoch": 0.4539032144118686,
      "grad_norm": 1.25,
      "learning_rate": 0.0004984915208212983,
      "loss": 6.0912,
      "mean_token_accuracy": 0.15156230181455613,
      "num_tokens": 9593114.0,
      "step": 5140
    },
    {
      "entropy": 6.167973184585572,
      "epoch": 0.4543447545037089,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004984878781602964,
      "loss": 5.9722,
      "mean_token_accuracy": 0.1515617176890373,
      "num_tokens": 9601454.0,
      "step": 5145
    },
    {
      "entropy": 6.117550992965699,
      "epoch": 0.45478629459554926,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004984842311212939,
      "loss": 6.0266,
      "mean_token_accuracy": 0.1547075927257538,
      "num_tokens": 9611994.0,
      "step": 5150
    },
    {
      "entropy": 6.099160146713257,
      "epoch": 0.4552278346873896,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004984805797043625,
      "loss": 5.9819,
      "mean_token_accuracy": 0.1554704263806343,
      "num_tokens": 9621318.0,
      "step": 5155
    },
    {
      "entropy": 6.086650943756103,
      "epoch": 0.45566937477923,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004984769239095736,
      "loss": 5.9871,
      "mean_token_accuracy": 0.1622908428311348,
      "num_tokens": 9630270.0,
      "step": 5160
    },
    {
      "entropy": 6.161284017562866,
      "epoch": 0.4561109148710703,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004984732637369989,
      "loss": 6.0164,
      "mean_token_accuracy": 0.15295830443501474,
      "num_tokens": 9640391.0,
      "step": 5165
    },
    {
      "entropy": 6.111026573181152,
      "epoch": 0.45655245496291064,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004984695991867099,
      "loss": 6.0302,
      "mean_token_accuracy": 0.15423648655414582,
      "num_tokens": 9648827.0,
      "step": 5170
    },
    {
      "entropy": 6.066336679458618,
      "epoch": 0.45699399505475097,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004984659302587788,
      "loss": 5.9651,
      "mean_token_accuracy": 0.1540757015347481,
      "num_tokens": 9657940.0,
      "step": 5175
    },
    {
      "entropy": 6.132708024978638,
      "epoch": 0.4574355351465913,
      "grad_norm": 1.25,
      "learning_rate": 0.000498462256953277,
      "loss": 6.0864,
      "mean_token_accuracy": 0.15138714611530305,
      "num_tokens": 9668296.0,
      "step": 5180
    },
    {
      "entropy": 6.187513446807861,
      "epoch": 0.45787707523843163,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004984585792702767,
      "loss": 6.0401,
      "mean_token_accuracy": 0.14838093519210815,
      "num_tokens": 9677914.0,
      "step": 5185
    },
    {
      "entropy": 6.183677721023559,
      "epoch": 0.45831861533027196,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004984548972098501,
      "loss": 6.174,
      "mean_token_accuracy": 0.13970668166875838,
      "num_tokens": 9687339.0,
      "step": 5190
    },
    {
      "entropy": 6.133589363098144,
      "epoch": 0.45876015542211235,
      "grad_norm": 1.546875,
      "learning_rate": 0.000498451210772069,
      "loss": 5.9454,
      "mean_token_accuracy": 0.16496401354670526,
      "num_tokens": 9696889.0,
      "step": 5195
    },
    {
      "entropy": 6.0567436695098875,
      "epoch": 0.4592016955139527,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004984475199570058,
      "loss": 5.9136,
      "mean_token_accuracy": 0.1610390767455101,
      "num_tokens": 9705498.0,
      "step": 5200
    },
    {
      "entropy": 6.099144554138183,
      "epoch": 0.459643235605793,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004984438247647329,
      "loss": 5.9902,
      "mean_token_accuracy": 0.15816803127527237,
      "num_tokens": 9713936.0,
      "step": 5205
    },
    {
      "entropy": 6.152586174011231,
      "epoch": 0.46008477569763334,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004984401251953223,
      "loss": 6.06,
      "mean_token_accuracy": 0.15645882338285447,
      "num_tokens": 9723924.0,
      "step": 5210
    },
    {
      "entropy": 6.179743766784668,
      "epoch": 0.4605263157894737,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004984364212488469,
      "loss": 6.0873,
      "mean_token_accuracy": 0.15282048285007477,
      "num_tokens": 9733427.0,
      "step": 5215
    },
    {
      "entropy": 6.035888767242431,
      "epoch": 0.460967855881314,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004984327129253789,
      "loss": 5.8837,
      "mean_token_accuracy": 0.16696648448705673,
      "num_tokens": 9742054.0,
      "step": 5220
    },
    {
      "entropy": 6.049908781051636,
      "epoch": 0.46140939597315433,
      "grad_norm": 1.25,
      "learning_rate": 0.0004984290002249914,
      "loss": 6.063,
      "mean_token_accuracy": 0.1488596171140671,
      "num_tokens": 9752447.0,
      "step": 5225
    },
    {
      "entropy": 6.131543159484863,
      "epoch": 0.4618509360649947,
      "grad_norm": 1.5,
      "learning_rate": 0.0004984252831477567,
      "loss": 5.8886,
      "mean_token_accuracy": 0.16800648123025894,
      "num_tokens": 9760878.0,
      "step": 5230
    },
    {
      "entropy": 5.97862868309021,
      "epoch": 0.46229247615683505,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004984215616937477,
      "loss": 5.9523,
      "mean_token_accuracy": 0.1603931352496147,
      "num_tokens": 9770200.0,
      "step": 5235
    },
    {
      "entropy": 6.214956617355346,
      "epoch": 0.4627340162486754,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004984178358630374,
      "loss": 6.0469,
      "mean_token_accuracy": 0.15920519679784775,
      "num_tokens": 9780303.0,
      "step": 5240
    },
    {
      "entropy": 6.130382776260376,
      "epoch": 0.4631755563405157,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004984141056556989,
      "loss": 5.9529,
      "mean_token_accuracy": 0.16305534839630126,
      "num_tokens": 9790248.0,
      "step": 5245
    },
    {
      "entropy": 6.11019229888916,
      "epoch": 0.46361709643235605,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004984103710718051,
      "loss": 6.0503,
      "mean_token_accuracy": 0.1460764303803444,
      "num_tokens": 9799345.0,
      "step": 5250
    },
    {
      "entropy": 6.091036796569824,
      "epoch": 0.4640586365241964,
      "grad_norm": 1.390625,
      "learning_rate": 0.000498406632111429,
      "loss": 5.9308,
      "mean_token_accuracy": 0.15862552225589752,
      "num_tokens": 9808561.0,
      "step": 5255
    },
    {
      "entropy": 6.106261253356934,
      "epoch": 0.4645001766160367,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004984028887746443,
      "loss": 5.9739,
      "mean_token_accuracy": 0.16280067563056946,
      "num_tokens": 9818324.0,
      "step": 5260
    },
    {
      "entropy": 5.984898710250855,
      "epoch": 0.4649417167078771,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004983991410615239,
      "loss": 5.9645,
      "mean_token_accuracy": 0.16320008635520936,
      "num_tokens": 9827935.0,
      "step": 5265
    },
    {
      "entropy": 6.089527177810669,
      "epoch": 0.4653832567997174,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004983953889721414,
      "loss": 5.9782,
      "mean_token_accuracy": 0.15890799909830094,
      "num_tokens": 9837118.0,
      "step": 5270
    },
    {
      "entropy": 6.149358510971069,
      "epoch": 0.46582479689155776,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004983916325065703,
      "loss": 6.0455,
      "mean_token_accuracy": 0.1533641681075096,
      "num_tokens": 9846197.0,
      "step": 5275
    },
    {
      "entropy": 6.196761894226074,
      "epoch": 0.4662663369833981,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004983878716648842,
      "loss": 6.107,
      "mean_token_accuracy": 0.14858163744211197,
      "num_tokens": 9856120.0,
      "step": 5280
    },
    {
      "entropy": 6.119464254379272,
      "epoch": 0.4667078770752384,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004983841064471567,
      "loss": 6.056,
      "mean_token_accuracy": 0.1537718027830124,
      "num_tokens": 9865599.0,
      "step": 5285
    },
    {
      "entropy": 6.096899557113647,
      "epoch": 0.46714941716707875,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004983803368534617,
      "loss": 5.9688,
      "mean_token_accuracy": 0.15628019720315933,
      "num_tokens": 9876471.0,
      "step": 5290
    },
    {
      "entropy": 6.196185064315796,
      "epoch": 0.4675909572589191,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004983765628838728,
      "loss": 6.0124,
      "mean_token_accuracy": 0.15632506608963012,
      "num_tokens": 9887680.0,
      "step": 5295
    },
    {
      "entropy": 6.205327701568604,
      "epoch": 0.46803249735075947,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004983727845384641,
      "loss": 6.0798,
      "mean_token_accuracy": 0.15046066045761108,
      "num_tokens": 9897366.0,
      "step": 5300
    },
    {
      "entropy": 6.112229824066162,
      "epoch": 0.4684740374425998,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004983690018173096,
      "loss": 6.0473,
      "mean_token_accuracy": 0.15282203108072281,
      "num_tokens": 9907437.0,
      "step": 5305
    },
    {
      "entropy": 6.03530478477478,
      "epoch": 0.46891557753444013,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004983652147204834,
      "loss": 5.8493,
      "mean_token_accuracy": 0.16888897120952606,
      "num_tokens": 9915766.0,
      "step": 5310
    },
    {
      "entropy": 6.039136505126953,
      "epoch": 0.46935711762628046,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004983614232480598,
      "loss": 6.0557,
      "mean_token_accuracy": 0.1539039731025696,
      "num_tokens": 9925389.0,
      "step": 5315
    },
    {
      "entropy": 6.1289163589477536,
      "epoch": 0.4697986577181208,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983576274001127,
      "loss": 6.0973,
      "mean_token_accuracy": 0.15069840773940085,
      "num_tokens": 9935798.0,
      "step": 5320
    },
    {
      "entropy": 6.164807653427124,
      "epoch": 0.4702401978099611,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000498353827176717,
      "loss": 5.9409,
      "mean_token_accuracy": 0.16221913695335388,
      "num_tokens": 9945579.0,
      "step": 5325
    },
    {
      "entropy": 6.1347403049469,
      "epoch": 0.4706817379018015,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004983500225779466,
      "loss": 5.9683,
      "mean_token_accuracy": 0.15579652935266494,
      "num_tokens": 9955248.0,
      "step": 5330
    },
    {
      "entropy": 6.085616779327393,
      "epoch": 0.47112327799364184,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004983462136038764,
      "loss": 5.9757,
      "mean_token_accuracy": 0.16684675961732864,
      "num_tokens": 9965078.0,
      "step": 5335
    },
    {
      "entropy": 6.127761745452881,
      "epoch": 0.4715648180854822,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004983424002545809,
      "loss": 5.9417,
      "mean_token_accuracy": 0.16493815779685975,
      "num_tokens": 9973932.0,
      "step": 5340
    },
    {
      "entropy": 6.168215322494507,
      "epoch": 0.4720063581773225,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004983385825301348,
      "loss": 6.0002,
      "mean_token_accuracy": 0.1548996612429619,
      "num_tokens": 9983315.0,
      "step": 5345
    },
    {
      "entropy": 6.058854579925537,
      "epoch": 0.47244789826916284,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004983347604306129,
      "loss": 5.9641,
      "mean_token_accuracy": 0.15613725483417512,
      "num_tokens": 9992987.0,
      "step": 5350
    },
    {
      "entropy": 6.019079732894897,
      "epoch": 0.47288943836100317,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004983309339560899,
      "loss": 6.0237,
      "mean_token_accuracy": 0.15275818705558777,
      "num_tokens": 10002268.0,
      "step": 5355
    },
    {
      "entropy": 6.169502782821655,
      "epoch": 0.4733309784528435,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004983271031066412,
      "loss": 6.0223,
      "mean_token_accuracy": 0.14990446120500564,
      "num_tokens": 10011772.0,
      "step": 5360
    },
    {
      "entropy": 6.224719524383545,
      "epoch": 0.4737725185446839,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004983232678823414,
      "loss": 6.0393,
      "mean_token_accuracy": 0.15731042325496675,
      "num_tokens": 10021069.0,
      "step": 5365
    },
    {
      "entropy": 6.063888311386108,
      "epoch": 0.4742140586365242,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004983194282832657,
      "loss": 5.9557,
      "mean_token_accuracy": 0.161125111579895,
      "num_tokens": 10029706.0,
      "step": 5370
    },
    {
      "entropy": 5.960548067092896,
      "epoch": 0.47465559872836455,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004983155843094895,
      "loss": 5.8997,
      "mean_token_accuracy": 0.16574549674987793,
      "num_tokens": 10039633.0,
      "step": 5375
    },
    {
      "entropy": 6.088515663146973,
      "epoch": 0.4750971388202049,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004983117359610881,
      "loss": 5.9683,
      "mean_token_accuracy": 0.1593565970659256,
      "num_tokens": 10048675.0,
      "step": 5380
    },
    {
      "entropy": 6.142480039596558,
      "epoch": 0.4755386789120452,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983078832381367,
      "loss": 5.9571,
      "mean_token_accuracy": 0.16286074072122575,
      "num_tokens": 10057447.0,
      "step": 5385
    },
    {
      "entropy": 6.040254259109497,
      "epoch": 0.47598021900388554,
      "grad_norm": 1.75,
      "learning_rate": 0.0004983040261407109,
      "loss": 6.0401,
      "mean_token_accuracy": 0.15268651247024537,
      "num_tokens": 10067020.0,
      "step": 5390
    },
    {
      "entropy": 6.183468341827393,
      "epoch": 0.47642175909572587,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004983001646688863,
      "loss": 6.0737,
      "mean_token_accuracy": 0.15229557305574418,
      "num_tokens": 10076466.0,
      "step": 5395
    },
    {
      "entropy": 6.156057071685791,
      "epoch": 0.47686329918756626,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004982962988227383,
      "loss": 5.9383,
      "mean_token_accuracy": 0.1562927931547165,
      "num_tokens": 10085384.0,
      "step": 5400
    },
    {
      "entropy": 6.085453033447266,
      "epoch": 0.4773048392794066,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000498292428602343,
      "loss": 6.0543,
      "mean_token_accuracy": 0.15518272593617438,
      "num_tokens": 10094015.0,
      "step": 5405
    },
    {
      "entropy": 6.131287050247193,
      "epoch": 0.4777463793712469,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004982885540077758,
      "loss": 5.9489,
      "mean_token_accuracy": 0.16021449863910675,
      "num_tokens": 10103107.0,
      "step": 5410
    },
    {
      "entropy": 6.061293983459473,
      "epoch": 0.47818791946308725,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004982846750391129,
      "loss": 5.883,
      "mean_token_accuracy": 0.165081886947155,
      "num_tokens": 10112408.0,
      "step": 5415
    },
    {
      "entropy": 6.103114128112793,
      "epoch": 0.4786294595549276,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004982807916964303,
      "loss": 5.9587,
      "mean_token_accuracy": 0.15787963271141053,
      "num_tokens": 10121605.0,
      "step": 5420
    },
    {
      "entropy": 6.012549638748169,
      "epoch": 0.4790709996467679,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000498276903979804,
      "loss": 5.9534,
      "mean_token_accuracy": 0.1598804622888565,
      "num_tokens": 10130484.0,
      "step": 5425
    },
    {
      "entropy": 5.984193134307861,
      "epoch": 0.47951253973860825,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00049827301188931,
      "loss": 5.906,
      "mean_token_accuracy": 0.16575224399566652,
      "num_tokens": 10140002.0,
      "step": 5430
    },
    {
      "entropy": 6.203652048110962,
      "epoch": 0.47995407983044863,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004982691154250247,
      "loss": 6.0681,
      "mean_token_accuracy": 0.1568043977022171,
      "num_tokens": 10150287.0,
      "step": 5435
    },
    {
      "entropy": 6.098856639862061,
      "epoch": 0.48039561992228896,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004982652145870245,
      "loss": 5.9027,
      "mean_token_accuracy": 0.16656555682420732,
      "num_tokens": 10160615.0,
      "step": 5440
    },
    {
      "entropy": 5.997969436645508,
      "epoch": 0.4808371600141293,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004982613093753856,
      "loss": 5.9701,
      "mean_token_accuracy": 0.16098668649792672,
      "num_tokens": 10170271.0,
      "step": 5445
    },
    {
      "entropy": 6.051625394821167,
      "epoch": 0.4812787001059696,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004982573997901847,
      "loss": 5.9498,
      "mean_token_accuracy": 0.16217143833637238,
      "num_tokens": 10179663.0,
      "step": 5450
    },
    {
      "entropy": 6.167967081069946,
      "epoch": 0.48172024019780996,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004982534858314982,
      "loss": 6.0302,
      "mean_token_accuracy": 0.1533224031329155,
      "num_tokens": 10188608.0,
      "step": 5455
    },
    {
      "entropy": 6.123821449279785,
      "epoch": 0.4821617802896503,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004982495674994031,
      "loss": 5.9888,
      "mean_token_accuracy": 0.1592209592461586,
      "num_tokens": 10198462.0,
      "step": 5460
    },
    {
      "entropy": 5.982554721832275,
      "epoch": 0.4826033203814906,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004982456447939758,
      "loss": 5.9323,
      "mean_token_accuracy": 0.16437650620937347,
      "num_tokens": 10208362.0,
      "step": 5465
    },
    {
      "entropy": 6.075630807876587,
      "epoch": 0.483044860473331,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004982417177152933,
      "loss": 5.955,
      "mean_token_accuracy": 0.1612927421927452,
      "num_tokens": 10217915.0,
      "step": 5470
    },
    {
      "entropy": 6.136414337158203,
      "epoch": 0.48348640056517134,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004982377862634325,
      "loss": 6.0141,
      "mean_token_accuracy": 0.16026580333709717,
      "num_tokens": 10226913.0,
      "step": 5475
    },
    {
      "entropy": 6.0771276473999025,
      "epoch": 0.48392794065701167,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004982338504384705,
      "loss": 5.9819,
      "mean_token_accuracy": 0.15748471468687059,
      "num_tokens": 10236516.0,
      "step": 5480
    },
    {
      "entropy": 6.1290271282196045,
      "epoch": 0.484369480748852,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004982299102404843,
      "loss": 5.9977,
      "mean_token_accuracy": 0.15856396406888962,
      "num_tokens": 10245492.0,
      "step": 5485
    },
    {
      "entropy": 5.977679443359375,
      "epoch": 0.48481102084069233,
      "grad_norm": 1.734375,
      "learning_rate": 0.000498225965669551,
      "loss": 5.8679,
      "mean_token_accuracy": 0.17316461503505706,
      "num_tokens": 10254094.0,
      "step": 5490
    },
    {
      "entropy": 6.03317437171936,
      "epoch": 0.48525256093253266,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004982220167257482,
      "loss": 5.9069,
      "mean_token_accuracy": 0.1637963816523552,
      "num_tokens": 10263645.0,
      "step": 5495
    },
    {
      "entropy": 6.013968801498413,
      "epoch": 0.485694101024373,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004982180634091529,
      "loss": 5.8919,
      "mean_token_accuracy": 0.16816270351409912,
      "num_tokens": 10273416.0,
      "step": 5500
    },
    {
      "entropy": 6.0709892272949215,
      "epoch": 0.4861356411162134,
      "grad_norm": 1.25,
      "learning_rate": 0.0004982141057198427,
      "loss": 5.9987,
      "mean_token_accuracy": 0.15827764123678206,
      "num_tokens": 10283194.0,
      "step": 5505
    },
    {
      "entropy": 6.133906984329224,
      "epoch": 0.4865771812080537,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004982101436578952,
      "loss": 5.9737,
      "mean_token_accuracy": 0.15922945886850357,
      "num_tokens": 10292188.0,
      "step": 5510
    },
    {
      "entropy": 6.0081017971038815,
      "epoch": 0.48701872129989404,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004982061772233878,
      "loss": 5.9009,
      "mean_token_accuracy": 0.17360175549983978,
      "num_tokens": 10301539.0,
      "step": 5515
    },
    {
      "entropy": 6.000823640823365,
      "epoch": 0.4874602613917344,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004982022064163984,
      "loss": 5.9093,
      "mean_token_accuracy": 0.16490670889616013,
      "num_tokens": 10310379.0,
      "step": 5520
    },
    {
      "entropy": 6.19787015914917,
      "epoch": 0.4879018014835747,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004981982312370047,
      "loss": 6.1108,
      "mean_token_accuracy": 0.1488179437816143,
      "num_tokens": 10320622.0,
      "step": 5525
    },
    {
      "entropy": 6.114509153366089,
      "epoch": 0.48834334157541504,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004981942516852847,
      "loss": 5.9356,
      "mean_token_accuracy": 0.15839738100767137,
      "num_tokens": 10330525.0,
      "step": 5530
    },
    {
      "entropy": 6.041282749176025,
      "epoch": 0.48878488166725537,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004981902677613161,
      "loss": 5.9671,
      "mean_token_accuracy": 0.1586508110165596,
      "num_tokens": 10340288.0,
      "step": 5535
    },
    {
      "entropy": 6.1574572086334225,
      "epoch": 0.48922642175909575,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004981862794651771,
      "loss": 6.0482,
      "mean_token_accuracy": 0.15438254177570343,
      "num_tokens": 10349836.0,
      "step": 5540
    },
    {
      "entropy": 6.124531698226929,
      "epoch": 0.4896679618509361,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004981822867969459,
      "loss": 5.9835,
      "mean_token_accuracy": 0.1561306193470955,
      "num_tokens": 10359481.0,
      "step": 5545
    },
    {
      "entropy": 6.077211523056031,
      "epoch": 0.4901095019427764,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004981782897567006,
      "loss": 5.9606,
      "mean_token_accuracy": 0.16075108498334884,
      "num_tokens": 10369837.0,
      "step": 5550
    },
    {
      "entropy": 6.108371019363403,
      "epoch": 0.49055104203461675,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004981742883445195,
      "loss": 5.927,
      "mean_token_accuracy": 0.16440120637416838,
      "num_tokens": 10379276.0,
      "step": 5555
    },
    {
      "entropy": 6.127094459533692,
      "epoch": 0.4909925821264571,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000498170282560481,
      "loss": 6.0562,
      "mean_token_accuracy": 0.15287164598703384,
      "num_tokens": 10388878.0,
      "step": 5560
    },
    {
      "entropy": 6.158031272888183,
      "epoch": 0.4914341222182974,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004981662724046637,
      "loss": 6.0416,
      "mean_token_accuracy": 0.15418365895748137,
      "num_tokens": 10399148.0,
      "step": 5565
    },
    {
      "entropy": 6.072212409973145,
      "epoch": 0.49187566231013774,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000498162257877146,
      "loss": 5.9705,
      "mean_token_accuracy": 0.16276238560676576,
      "num_tokens": 10409002.0,
      "step": 5570
    },
    {
      "entropy": 6.052406740188599,
      "epoch": 0.4923172024019781,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004981582389780065,
      "loss": 6.0182,
      "mean_token_accuracy": 0.15543360412120819,
      "num_tokens": 10418083.0,
      "step": 5575
    },
    {
      "entropy": 6.2516388416290285,
      "epoch": 0.49275874249381846,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004981542157073241,
      "loss": 6.0216,
      "mean_token_accuracy": 0.15745319724082946,
      "num_tokens": 10428134.0,
      "step": 5580
    },
    {
      "entropy": 6.102121114730835,
      "epoch": 0.4932002825856588,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004981501880651775,
      "loss": 6.0144,
      "mean_token_accuracy": 0.15700092390179635,
      "num_tokens": 10437436.0,
      "step": 5585
    },
    {
      "entropy": 6.011652183532715,
      "epoch": 0.4936418226774991,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004981461560516457,
      "loss": 5.9437,
      "mean_token_accuracy": 0.15731440335512162,
      "num_tokens": 10446176.0,
      "step": 5590
    },
    {
      "entropy": 6.185711860656738,
      "epoch": 0.49408336276933945,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004981421196668075,
      "loss": 6.1086,
      "mean_token_accuracy": 0.15340851247310638,
      "num_tokens": 10455953.0,
      "step": 5595
    },
    {
      "entropy": 6.20690336227417,
      "epoch": 0.4945249028611798,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004981380789107422,
      "loss": 5.9589,
      "mean_token_accuracy": 0.15672969669103623,
      "num_tokens": 10465129.0,
      "step": 5600
    },
    {
      "entropy": 6.003274393081665,
      "epoch": 0.4949664429530201,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004981340337835287,
      "loss": 5.9608,
      "mean_token_accuracy": 0.1587096706032753,
      "num_tokens": 10474376.0,
      "step": 5605
    },
    {
      "entropy": 6.04667649269104,
      "epoch": 0.4954079830448605,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004981299842852464,
      "loss": 6.02,
      "mean_token_accuracy": 0.1601344585418701,
      "num_tokens": 10484849.0,
      "step": 5610
    },
    {
      "entropy": 6.119048452377319,
      "epoch": 0.49584952313670083,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004981259304159747,
      "loss": 5.9104,
      "mean_token_accuracy": 0.17004732936620712,
      "num_tokens": 10493950.0,
      "step": 5615
    },
    {
      "entropy": 6.0551595211029055,
      "epoch": 0.49629106322854116,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004981218721757929,
      "loss": 5.877,
      "mean_token_accuracy": 0.16873401552438735,
      "num_tokens": 10502890.0,
      "step": 5620
    },
    {
      "entropy": 5.9873803615570065,
      "epoch": 0.4967326033203815,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004981178095647805,
      "loss": 5.8707,
      "mean_token_accuracy": 0.160433566570282,
      "num_tokens": 10511702.0,
      "step": 5625
    },
    {
      "entropy": 6.003235149383545,
      "epoch": 0.4971741434122218,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004981137425830171,
      "loss": 5.9481,
      "mean_token_accuracy": 0.16074998080730438,
      "num_tokens": 10520212.0,
      "step": 5630
    },
    {
      "entropy": 6.066579341888428,
      "epoch": 0.49761568350406216,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004981096712305825,
      "loss": 5.971,
      "mean_token_accuracy": 0.1613484501838684,
      "num_tokens": 10529652.0,
      "step": 5635
    },
    {
      "entropy": 6.099881601333618,
      "epoch": 0.4980572235959025,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004981055955075561,
      "loss": 6.0225,
      "mean_token_accuracy": 0.158574703335762,
      "num_tokens": 10537980.0,
      "step": 5640
    },
    {
      "entropy": 6.089739847183227,
      "epoch": 0.4984987636877429,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004981015154140181,
      "loss": 5.8995,
      "mean_token_accuracy": 0.16764541566371918,
      "num_tokens": 10546854.0,
      "step": 5645
    },
    {
      "entropy": 5.998034429550171,
      "epoch": 0.4989403037795832,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004980974309500483,
      "loss": 5.923,
      "mean_token_accuracy": 0.16142944097518921,
      "num_tokens": 10555691.0,
      "step": 5650
    },
    {
      "entropy": 6.080042123794556,
      "epoch": 0.49938184387142354,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004980933421157267,
      "loss": 6.0191,
      "mean_token_accuracy": 0.1503999724984169,
      "num_tokens": 10564750.0,
      "step": 5655
    },
    {
      "entropy": 6.1378124237060545,
      "epoch": 0.49982338396326387,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004980892489111334,
      "loss": 6.0389,
      "mean_token_accuracy": 0.15747978240251542,
      "num_tokens": 10573515.0,
      "step": 5660
    },
    {
      "entropy": 6.055855131149292,
      "epoch": 0.5002649240551043,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004980851513363486,
      "loss": 5.9161,
      "mean_token_accuracy": 0.1634524017572403,
      "num_tokens": 10582730.0,
      "step": 5665
    },
    {
      "entropy": 5.963370323181152,
      "epoch": 0.5007064641469445,
      "grad_norm": 5.15625,
      "learning_rate": 0.0004980810493914526,
      "loss": 5.8141,
      "mean_token_accuracy": 0.18174145370721817,
      "num_tokens": 10591267.0,
      "step": 5670
    },
    {
      "entropy": 6.040016841888428,
      "epoch": 0.5011480042387849,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004980769430765256,
      "loss": 5.918,
      "mean_token_accuracy": 0.15779468268156052,
      "num_tokens": 10599776.0,
      "step": 5675
    },
    {
      "entropy": 6.018471956253052,
      "epoch": 0.5015895443306252,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004980728323916484,
      "loss": 5.9294,
      "mean_token_accuracy": 0.15198549777269363,
      "num_tokens": 10609670.0,
      "step": 5680
    },
    {
      "entropy": 6.0253712177276615,
      "epoch": 0.5020310844224656,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004980687173369009,
      "loss": 5.8857,
      "mean_token_accuracy": 0.16054306030273438,
      "num_tokens": 10619032.0,
      "step": 5685
    },
    {
      "entropy": 6.13008975982666,
      "epoch": 0.5024726245143059,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004980645979123644,
      "loss": 5.9696,
      "mean_token_accuracy": 0.15761574804782869,
      "num_tokens": 10628851.0,
      "step": 5690
    },
    {
      "entropy": 6.1254706382751465,
      "epoch": 0.5029141646061462,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004980604741181192,
      "loss": 5.9384,
      "mean_token_accuracy": 0.1661163553595543,
      "num_tokens": 10638335.0,
      "step": 5695
    },
    {
      "entropy": 6.104246044158936,
      "epoch": 0.5033557046979866,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004980563459542461,
      "loss": 6.0288,
      "mean_token_accuracy": 0.15877759456634521,
      "num_tokens": 10648450.0,
      "step": 5700
    },
    {
      "entropy": 6.129976463317871,
      "epoch": 0.5037972447898269,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004980522134208261,
      "loss": 6.0499,
      "mean_token_accuracy": 0.15289226770401002,
      "num_tokens": 10658700.0,
      "step": 5705
    },
    {
      "entropy": 6.125590181350708,
      "epoch": 0.5042387848816673,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004980480765179401,
      "loss": 6.0498,
      "mean_token_accuracy": 0.15848255753517151,
      "num_tokens": 10667952.0,
      "step": 5710
    },
    {
      "entropy": 6.015667390823364,
      "epoch": 0.5046803249735076,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004980439352456692,
      "loss": 5.8856,
      "mean_token_accuracy": 0.16176477670669556,
      "num_tokens": 10677228.0,
      "step": 5715
    },
    {
      "entropy": 6.1053516387939455,
      "epoch": 0.505121865065348,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004980397896040944,
      "loss": 5.974,
      "mean_token_accuracy": 0.16118671298027037,
      "num_tokens": 10686183.0,
      "step": 5720
    },
    {
      "entropy": 6.190397262573242,
      "epoch": 0.5055634051571882,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004980356395932969,
      "loss": 5.9997,
      "mean_token_accuracy": 0.15813823491334916,
      "num_tokens": 10695293.0,
      "step": 5725
    },
    {
      "entropy": 6.0877281665802006,
      "epoch": 0.5060049452490286,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004980314852133581,
      "loss": 5.9647,
      "mean_token_accuracy": 0.1628525137901306,
      "num_tokens": 10704853.0,
      "step": 5730
    },
    {
      "entropy": 6.098628950119019,
      "epoch": 0.506446485340869,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004980273264643594,
      "loss": 5.9596,
      "mean_token_accuracy": 0.15651024580001832,
      "num_tokens": 10714307.0,
      "step": 5735
    },
    {
      "entropy": 5.91804838180542,
      "epoch": 0.5068880254327093,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004980231633463822,
      "loss": 5.8195,
      "mean_token_accuracy": 0.16056904792785645,
      "num_tokens": 10723513.0,
      "step": 5740
    },
    {
      "entropy": 6.041822719573974,
      "epoch": 0.5073295655245497,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004980189958595081,
      "loss": 6.009,
      "mean_token_accuracy": 0.15504895150661469,
      "num_tokens": 10732809.0,
      "step": 5745
    },
    {
      "entropy": 6.180126619338989,
      "epoch": 0.5077711056163899,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004980148240038186,
      "loss": 6.0285,
      "mean_token_accuracy": 0.16252532303333284,
      "num_tokens": 10742960.0,
      "step": 5750
    },
    {
      "entropy": 6.167002391815186,
      "epoch": 0.5082126457082303,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004980106477793957,
      "loss": 6.0962,
      "mean_token_accuracy": 0.15922853201627732,
      "num_tokens": 10752835.0,
      "step": 5755
    },
    {
      "entropy": 6.090708923339844,
      "epoch": 0.5086541858000706,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004980064671863209,
      "loss": 5.9568,
      "mean_token_accuracy": 0.16255878955125808,
      "num_tokens": 10762139.0,
      "step": 5760
    },
    {
      "entropy": 6.121759986877441,
      "epoch": 0.509095725891911,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004980022822246763,
      "loss": 6.0284,
      "mean_token_accuracy": 0.1527095004916191,
      "num_tokens": 10771961.0,
      "step": 5765
    },
    {
      "entropy": 6.189744853973389,
      "epoch": 0.5095372659837514,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004979980928945439,
      "loss": 6.0799,
      "mean_token_accuracy": 0.14973168522119523,
      "num_tokens": 10780163.0,
      "step": 5770
    },
    {
      "entropy": 6.200748443603516,
      "epoch": 0.5099788060755917,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004979938991960056,
      "loss": 6.0327,
      "mean_token_accuracy": 0.156923408806324,
      "num_tokens": 10788339.0,
      "step": 5775
    },
    {
      "entropy": 6.119477415084839,
      "epoch": 0.510420346167432,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004979897011291436,
      "loss": 6.0257,
      "mean_token_accuracy": 0.15160492956638336,
      "num_tokens": 10797315.0,
      "step": 5780
    },
    {
      "entropy": 6.097667026519775,
      "epoch": 0.5108618862592723,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004979854986940402,
      "loss": 5.9344,
      "mean_token_accuracy": 0.15801928341388702,
      "num_tokens": 10806864.0,
      "step": 5785
    },
    {
      "entropy": 6.078370189666748,
      "epoch": 0.5113034263511127,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004979812918907777,
      "loss": 5.971,
      "mean_token_accuracy": 0.15703734010457993,
      "num_tokens": 10816417.0,
      "step": 5790
    },
    {
      "entropy": 6.111645221710205,
      "epoch": 0.511744966442953,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004979770807194385,
      "loss": 6.0491,
      "mean_token_accuracy": 0.15200948417186738,
      "num_tokens": 10825692.0,
      "step": 5795
    },
    {
      "entropy": 6.125649499893188,
      "epoch": 0.5121865065347934,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004979728651801051,
      "loss": 6.0184,
      "mean_token_accuracy": 0.1593027725815773,
      "num_tokens": 10834652.0,
      "step": 5800
    },
    {
      "entropy": 6.142418766021729,
      "epoch": 0.5126280466266337,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004979686452728602,
      "loss": 6.0045,
      "mean_token_accuracy": 0.15610153079032899,
      "num_tokens": 10844491.0,
      "step": 5805
    },
    {
      "entropy": 6.076530647277832,
      "epoch": 0.513069586718474,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004979644209977863,
      "loss": 5.9576,
      "mean_token_accuracy": 0.16532048285007478,
      "num_tokens": 10853948.0,
      "step": 5810
    },
    {
      "entropy": 6.112010383605957,
      "epoch": 0.5135111268103144,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004979601923549661,
      "loss": 5.9547,
      "mean_token_accuracy": 0.16036146879196167,
      "num_tokens": 10861745.0,
      "step": 5815
    },
    {
      "entropy": 6.1195940494537355,
      "epoch": 0.5139526669021547,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004979559593444826,
      "loss": 6.0383,
      "mean_token_accuracy": 0.1518269196152687,
      "num_tokens": 10871239.0,
      "step": 5820
    },
    {
      "entropy": 6.060505533218384,
      "epoch": 0.5143942069939951,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004979517219664187,
      "loss": 5.8893,
      "mean_token_accuracy": 0.16240942627191543,
      "num_tokens": 10880883.0,
      "step": 5825
    },
    {
      "entropy": 6.087244081497192,
      "epoch": 0.5148357470858353,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004979474802208572,
      "loss": 5.9569,
      "mean_token_accuracy": 0.16193697676062585,
      "num_tokens": 10890002.0,
      "step": 5830
    },
    {
      "entropy": 6.015149307250977,
      "epoch": 0.5152772871776757,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004979432341078816,
      "loss": 5.8864,
      "mean_token_accuracy": 0.15554073452949524,
      "num_tokens": 10898299.0,
      "step": 5835
    },
    {
      "entropy": 6.089050912857056,
      "epoch": 0.5157188272695161,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004979389836275746,
      "loss": 6.0774,
      "mean_token_accuracy": 0.14996080696582795,
      "num_tokens": 10907918.0,
      "step": 5840
    },
    {
      "entropy": 6.133493900299072,
      "epoch": 0.5161603673613564,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004979347287800198,
      "loss": 5.9596,
      "mean_token_accuracy": 0.15501177459955215,
      "num_tokens": 10917318.0,
      "step": 5845
    },
    {
      "entropy": 6.060728597640991,
      "epoch": 0.5166019074531968,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004979304695653005,
      "loss": 5.9144,
      "mean_token_accuracy": 0.16151465103030205,
      "num_tokens": 10926338.0,
      "step": 5850
    },
    {
      "entropy": 5.939118814468384,
      "epoch": 0.5170434475450371,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004979262059835001,
      "loss": 5.9432,
      "mean_token_accuracy": 0.16033163890242577,
      "num_tokens": 10935244.0,
      "step": 5855
    },
    {
      "entropy": 6.143280649185181,
      "epoch": 0.5174849876368774,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004979219380347021,
      "loss": 5.9905,
      "mean_token_accuracy": 0.15961788594722748,
      "num_tokens": 10943764.0,
      "step": 5860
    },
    {
      "entropy": 6.160701942443848,
      "epoch": 0.5179265277287177,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004979176657189901,
      "loss": 5.9547,
      "mean_token_accuracy": 0.16057219803333284,
      "num_tokens": 10952580.0,
      "step": 5865
    },
    {
      "entropy": 5.9959807872772215,
      "epoch": 0.5183680678205581,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004979133890364477,
      "loss": 5.9825,
      "mean_token_accuracy": 0.1650010645389557,
      "num_tokens": 10961745.0,
      "step": 5870
    },
    {
      "entropy": 5.959691524505615,
      "epoch": 0.5188096079123985,
      "grad_norm": 1.296875,
      "learning_rate": 0.000497909107987159,
      "loss": 5.9087,
      "mean_token_accuracy": 0.15640757530927657,
      "num_tokens": 10971018.0,
      "step": 5875
    },
    {
      "entropy": 6.110277318954468,
      "epoch": 0.5192511480042388,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004979048225712074,
      "loss": 5.9408,
      "mean_token_accuracy": 0.1647150531411171,
      "num_tokens": 10980070.0,
      "step": 5880
    },
    {
      "entropy": 6.112304973602295,
      "epoch": 0.5196926880960792,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004979005327886772,
      "loss": 5.9115,
      "mean_token_accuracy": 0.15777890086174012,
      "num_tokens": 10989090.0,
      "step": 5885
    },
    {
      "entropy": 6.028247594833374,
      "epoch": 0.5201342281879194,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004978962386396525,
      "loss": 5.9676,
      "mean_token_accuracy": 0.16010279804468155,
      "num_tokens": 10998652.0,
      "step": 5890
    },
    {
      "entropy": 5.936372900009156,
      "epoch": 0.5205757682797598,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004978919401242171,
      "loss": 5.8702,
      "mean_token_accuracy": 0.17004906684160231,
      "num_tokens": 11006323.0,
      "step": 5895
    },
    {
      "entropy": 5.953699111938477,
      "epoch": 0.5210173083716001,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004978876372424553,
      "loss": 5.8013,
      "mean_token_accuracy": 0.17337664365768432,
      "num_tokens": 11015054.0,
      "step": 5900
    },
    {
      "entropy": 6.08972134590149,
      "epoch": 0.5214588484634405,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004978833299944515,
      "loss": 5.9875,
      "mean_token_accuracy": 0.15530108660459518,
      "num_tokens": 11025787.0,
      "step": 5905
    },
    {
      "entropy": 6.201382303237915,
      "epoch": 0.5219003885552809,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004978790183802901,
      "loss": 6.0957,
      "mean_token_accuracy": 0.15121813490986824,
      "num_tokens": 11036091.0,
      "step": 5910
    },
    {
      "entropy": 6.078470230102539,
      "epoch": 0.5223419286471211,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004978747024000554,
      "loss": 6.0798,
      "mean_token_accuracy": 0.15142182558774947,
      "num_tokens": 11046442.0,
      "step": 5915
    },
    {
      "entropy": 6.137741947174073,
      "epoch": 0.5227834687389615,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004978703820538321,
      "loss": 6.0767,
      "mean_token_accuracy": 0.15347778424620628,
      "num_tokens": 11056907.0,
      "step": 5920
    },
    {
      "entropy": 6.088362789154052,
      "epoch": 0.5232250088308018,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004978660573417048,
      "loss": 5.9226,
      "mean_token_accuracy": 0.165810264647007,
      "num_tokens": 11065298.0,
      "step": 5925
    },
    {
      "entropy": 6.02788200378418,
      "epoch": 0.5236665489226422,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000497861728263758,
      "loss": 5.9277,
      "mean_token_accuracy": 0.16316543370485306,
      "num_tokens": 11074509.0,
      "step": 5930
    },
    {
      "entropy": 6.140137195587158,
      "epoch": 0.5241080890144825,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004978573948200769,
      "loss": 5.9728,
      "mean_token_accuracy": 0.16266270875930786,
      "num_tokens": 11082433.0,
      "step": 5935
    },
    {
      "entropy": 6.069718503952027,
      "epoch": 0.5245496291063229,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004978530570107461,
      "loss": 5.9336,
      "mean_token_accuracy": 0.16215700507164002,
      "num_tokens": 11090838.0,
      "step": 5940
    },
    {
      "entropy": 6.054668140411377,
      "epoch": 0.5249911691981632,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004978487148358506,
      "loss": 5.9288,
      "mean_token_accuracy": 0.1637731358408928,
      "num_tokens": 11100048.0,
      "step": 5945
    },
    {
      "entropy": 5.982475614547729,
      "epoch": 0.5254327092900035,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004978443682954756,
      "loss": 5.9389,
      "mean_token_accuracy": 0.16071433573961258,
      "num_tokens": 11109871.0,
      "step": 5950
    },
    {
      "entropy": 6.093014669418335,
      "epoch": 0.5258742493818439,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004978400173897061,
      "loss": 5.924,
      "mean_token_accuracy": 0.1567403718829155,
      "num_tokens": 11119450.0,
      "step": 5955
    },
    {
      "entropy": 6.094651889801026,
      "epoch": 0.5263157894736842,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004978356621186275,
      "loss": 5.8755,
      "mean_token_accuracy": 0.17035853564739228,
      "num_tokens": 11128321.0,
      "step": 5960
    },
    {
      "entropy": 6.019528484344482,
      "epoch": 0.5267573295655246,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004978313024823249,
      "loss": 5.9397,
      "mean_token_accuracy": 0.15585954636335372,
      "num_tokens": 11138657.0,
      "step": 5965
    },
    {
      "entropy": 6.009906387329101,
      "epoch": 0.5271988696573648,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004978269384808839,
      "loss": 5.9209,
      "mean_token_accuracy": 0.16706853806972505,
      "num_tokens": 11147704.0,
      "step": 5970
    },
    {
      "entropy": 6.089446401596069,
      "epoch": 0.5276404097492052,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004978225701143898,
      "loss": 6.0138,
      "mean_token_accuracy": 0.1562186986207962,
      "num_tokens": 11156979.0,
      "step": 5975
    },
    {
      "entropy": 6.104236221313476,
      "epoch": 0.5280819498410456,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004978181973829284,
      "loss": 5.8985,
      "mean_token_accuracy": 0.15810604989528657,
      "num_tokens": 11165539.0,
      "step": 5980
    },
    {
      "entropy": 6.11846137046814,
      "epoch": 0.5285234899328859,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004978138202865851,
      "loss": 5.9758,
      "mean_token_accuracy": 0.1503559224307537,
      "num_tokens": 11175911.0,
      "step": 5985
    },
    {
      "entropy": 6.050204277038574,
      "epoch": 0.5289650300247263,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004978094388254459,
      "loss": 5.9883,
      "mean_token_accuracy": 0.15720559507608414,
      "num_tokens": 11185365.0,
      "step": 5990
    },
    {
      "entropy": 6.095414018630981,
      "epoch": 0.5294065701165666,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004978050529995965,
      "loss": 5.8979,
      "mean_token_accuracy": 0.16393667906522752,
      "num_tokens": 11195318.0,
      "step": 5995
    },
    {
      "entropy": 6.074192810058594,
      "epoch": 0.5298481102084069,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004978006628091228,
      "loss": 6.0079,
      "mean_token_accuracy": 0.15748494416475295,
      "num_tokens": 11205246.0,
      "step": 6000
    },
    {
      "epoch": 0.5298481102084069,
      "eval_entropy": 5.929435700441406,
      "eval_loss": 5.984138011932373,
      "eval_mean_token_accuracy": 0.16603976909265278,
      "eval_num_tokens": 11205246.0,
      "eval_runtime": 26.1397,
      "eval_samples_per_second": 1351.009,
      "eval_steps_per_second": 168.9,
      "step": 6000
    },
    {
      "entropy": 6.097824764251709,
      "epoch": 0.5302896503002472,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004977962682541108,
      "loss": 5.9636,
      "mean_token_accuracy": 0.15681301951408386,
      "num_tokens": 11215183.0,
      "step": 6005
    },
    {
      "entropy": 6.0872814655303955,
      "epoch": 0.5307311903920876,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004977918693346467,
      "loss": 5.9167,
      "mean_token_accuracy": 0.15939552038908006,
      "num_tokens": 11224127.0,
      "step": 6010
    },
    {
      "entropy": 6.113217401504516,
      "epoch": 0.531172730483928,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004977874660508165,
      "loss": 5.9723,
      "mean_token_accuracy": 0.16238286793231965,
      "num_tokens": 11233014.0,
      "step": 6015
    },
    {
      "entropy": 6.072064352035523,
      "epoch": 0.5316142705757683,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004977830584027065,
      "loss": 5.9707,
      "mean_token_accuracy": 0.16605914533138275,
      "num_tokens": 11242251.0,
      "step": 6020
    },
    {
      "entropy": 6.053228425979614,
      "epoch": 0.5320558106676087,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004977786463904033,
      "loss": 5.9956,
      "mean_token_accuracy": 0.1588194712996483,
      "num_tokens": 11251947.0,
      "step": 6025
    },
    {
      "entropy": 6.073520278930664,
      "epoch": 0.5324973507594489,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004977742300139929,
      "loss": 6.0406,
      "mean_token_accuracy": 0.1539533868432045,
      "num_tokens": 11261125.0,
      "step": 6030
    },
    {
      "entropy": 6.160229969024658,
      "epoch": 0.5329388908512893,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004977698092735621,
      "loss": 5.9402,
      "mean_token_accuracy": 0.16421253234148026,
      "num_tokens": 11270334.0,
      "step": 6035
    },
    {
      "entropy": 5.976454734802246,
      "epoch": 0.5333804309431296,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977653841691975,
      "loss": 5.9399,
      "mean_token_accuracy": 0.15946147739887237,
      "num_tokens": 11280497.0,
      "step": 6040
    },
    {
      "entropy": 6.087053728103638,
      "epoch": 0.53382197103497,
      "grad_norm": 1.375,
      "learning_rate": 0.0004977609547009857,
      "loss": 5.8659,
      "mean_token_accuracy": 0.16440399140119552,
      "num_tokens": 11288636.0,
      "step": 6045
    },
    {
      "entropy": 6.129591178894043,
      "epoch": 0.5342635111268104,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004977565208690134,
      "loss": 5.9827,
      "mean_token_accuracy": 0.15665870159864426,
      "num_tokens": 11297519.0,
      "step": 6050
    },
    {
      "entropy": 6.083321666717529,
      "epoch": 0.5347050512186506,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004977520826733677,
      "loss": 5.9588,
      "mean_token_accuracy": 0.1661783814430237,
      "num_tokens": 11306280.0,
      "step": 6055
    },
    {
      "entropy": 6.075170660018921,
      "epoch": 0.535146591310491,
      "grad_norm": 1.25,
      "learning_rate": 0.0004977476401141353,
      "loss": 5.9527,
      "mean_token_accuracy": 0.15530341863632202,
      "num_tokens": 11315366.0,
      "step": 6060
    },
    {
      "entropy": 6.076067924499512,
      "epoch": 0.5355881314023313,
      "grad_norm": 1.25,
      "learning_rate": 0.0004977431931914032,
      "loss": 5.9595,
      "mean_token_accuracy": 0.15634357184171677,
      "num_tokens": 11324664.0,
      "step": 6065
    },
    {
      "entropy": 6.069175720214844,
      "epoch": 0.5360296714941717,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004977387419052586,
      "loss": 5.9339,
      "mean_token_accuracy": 0.1536666676402092,
      "num_tokens": 11334118.0,
      "step": 6070
    },
    {
      "entropy": 6.2017174243927,
      "epoch": 0.536471211586012,
      "grad_norm": 1.234375,
      "learning_rate": 0.000497734286255789,
      "loss": 6.0281,
      "mean_token_accuracy": 0.15356491655111312,
      "num_tokens": 11342684.0,
      "step": 6075
    },
    {
      "entropy": 6.07715802192688,
      "epoch": 0.5369127516778524,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004977298262430813,
      "loss": 5.9634,
      "mean_token_accuracy": 0.16605139821767806,
      "num_tokens": 11351298.0,
      "step": 6080
    },
    {
      "entropy": 6.011256170272827,
      "epoch": 0.5373542917696927,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004977253618672229,
      "loss": 5.9153,
      "mean_token_accuracy": 0.16742231994867324,
      "num_tokens": 11360264.0,
      "step": 6085
    },
    {
      "entropy": 5.972763013839722,
      "epoch": 0.537795831861533,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977208931283013,
      "loss": 5.894,
      "mean_token_accuracy": 0.16407648473978043,
      "num_tokens": 11369815.0,
      "step": 6090
    },
    {
      "entropy": 6.075697612762451,
      "epoch": 0.5382373719533734,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004977164200264042,
      "loss": 5.9606,
      "mean_token_accuracy": 0.16177425235509874,
      "num_tokens": 11379131.0,
      "step": 6095
    },
    {
      "entropy": 6.092071437835694,
      "epoch": 0.5386789120452137,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004977119425616191,
      "loss": 5.8475,
      "mean_token_accuracy": 0.16356392502784728,
      "num_tokens": 11387692.0,
      "step": 6100
    },
    {
      "entropy": 6.020279884338379,
      "epoch": 0.5391204521370541,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004977074607340336,
      "loss": 5.9208,
      "mean_token_accuracy": 0.16179520785808563,
      "num_tokens": 11396278.0,
      "step": 6105
    },
    {
      "entropy": 5.9607339859008786,
      "epoch": 0.5395619922288943,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004977029745437357,
      "loss": 5.8561,
      "mean_token_accuracy": 0.16862278580665588,
      "num_tokens": 11404790.0,
      "step": 6110
    },
    {
      "entropy": 6.074794054031372,
      "epoch": 0.5400035323207347,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000497698483990813,
      "loss": 5.9448,
      "mean_token_accuracy": 0.15622814297676085,
      "num_tokens": 11414614.0,
      "step": 6115
    },
    {
      "entropy": 6.05021300315857,
      "epoch": 0.5404450724125751,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004976939890753539,
      "loss": 5.9789,
      "mean_token_accuracy": 0.15662158280611038,
      "num_tokens": 11423696.0,
      "step": 6120
    },
    {
      "entropy": 5.987365674972534,
      "epoch": 0.5408866125044154,
      "grad_norm": 1.375,
      "learning_rate": 0.000497689489797446,
      "loss": 5.9184,
      "mean_token_accuracy": 0.1647222951054573,
      "num_tokens": 11432669.0,
      "step": 6125
    },
    {
      "entropy": 6.090789747238159,
      "epoch": 0.5413281525962558,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004976849861571778,
      "loss": 5.9464,
      "mean_token_accuracy": 0.15996651202440262,
      "num_tokens": 11442132.0,
      "step": 6130
    },
    {
      "entropy": 6.047543382644653,
      "epoch": 0.541769692688096,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004976804781546373,
      "loss": 5.9234,
      "mean_token_accuracy": 0.1532900132238865,
      "num_tokens": 11451821.0,
      "step": 6135
    },
    {
      "entropy": 6.068201351165771,
      "epoch": 0.5422112327799364,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004976759657899129,
      "loss": 5.9426,
      "mean_token_accuracy": 0.16034639477729798,
      "num_tokens": 11460242.0,
      "step": 6140
    },
    {
      "entropy": 5.969129514694214,
      "epoch": 0.5426527728717767,
      "grad_norm": 1.234375,
      "learning_rate": 0.000497671449063093,
      "loss": 5.9507,
      "mean_token_accuracy": 0.16280617713928222,
      "num_tokens": 11469588.0,
      "step": 6145
    },
    {
      "entropy": 6.102783298492431,
      "epoch": 0.5430943129636171,
      "grad_norm": 1.359375,
      "learning_rate": 0.000497666927974266,
      "loss": 6.0042,
      "mean_token_accuracy": 0.15202597826719283,
      "num_tokens": 11480401.0,
      "step": 6150
    },
    {
      "entropy": 6.1373790264129635,
      "epoch": 0.5435358530554575,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004976624025235205,
      "loss": 5.938,
      "mean_token_accuracy": 0.16415768265724182,
      "num_tokens": 11489117.0,
      "step": 6155
    },
    {
      "entropy": 5.957103109359741,
      "epoch": 0.5439773931472978,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004976578727109453,
      "loss": 5.871,
      "mean_token_accuracy": 0.16513819396495819,
      "num_tokens": 11497612.0,
      "step": 6160
    },
    {
      "entropy": 5.988688659667969,
      "epoch": 0.5444189332391381,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004976533385366289,
      "loss": 5.9763,
      "mean_token_accuracy": 0.15876150280237197,
      "num_tokens": 11507024.0,
      "step": 6165
    },
    {
      "entropy": 6.076825761795044,
      "epoch": 0.5448604733309784,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004976488000006602,
      "loss": 5.9042,
      "mean_token_accuracy": 0.1657929614186287,
      "num_tokens": 11516873.0,
      "step": 6170
    },
    {
      "entropy": 6.166196537017822,
      "epoch": 0.5453020134228188,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004976442571031282,
      "loss": 5.9943,
      "mean_token_accuracy": 0.16310707181692125,
      "num_tokens": 11526128.0,
      "step": 6175
    },
    {
      "entropy": 5.990091466903687,
      "epoch": 0.5457435535146591,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004976397098441218,
      "loss": 6.0027,
      "mean_token_accuracy": 0.16429250240325927,
      "num_tokens": 11535030.0,
      "step": 6180
    },
    {
      "entropy": 5.9520035743713375,
      "epoch": 0.5461850936064995,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004976351582237301,
      "loss": 5.885,
      "mean_token_accuracy": 0.16240569949150085,
      "num_tokens": 11543113.0,
      "step": 6185
    },
    {
      "entropy": 6.077959585189819,
      "epoch": 0.5466266336983399,
      "grad_norm": 1.375,
      "learning_rate": 0.0004976306022420423,
      "loss": 5.948,
      "mean_token_accuracy": 0.15816203504800797,
      "num_tokens": 11552744.0,
      "step": 6190
    },
    {
      "entropy": 6.0480293273925785,
      "epoch": 0.5470681737901801,
      "grad_norm": 1.75,
      "learning_rate": 0.0004976260418991476,
      "loss": 5.9521,
      "mean_token_accuracy": 0.15910596996545792,
      "num_tokens": 11561274.0,
      "step": 6195
    },
    {
      "entropy": 6.120669841766357,
      "epoch": 0.5475097138820205,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004976214771951354,
      "loss": 5.9849,
      "mean_token_accuracy": 0.15998436361551285,
      "num_tokens": 11570485.0,
      "step": 6200
    },
    {
      "entropy": 6.119882488250733,
      "epoch": 0.5479512539738608,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004976169081300949,
      "loss": 5.9707,
      "mean_token_accuracy": 0.15495690554380417,
      "num_tokens": 11579380.0,
      "step": 6205
    },
    {
      "entropy": 6.016612339019775,
      "epoch": 0.5483927940657012,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004976123347041159,
      "loss": 5.9419,
      "mean_token_accuracy": 0.15940989702939987,
      "num_tokens": 11588686.0,
      "step": 6210
    },
    {
      "entropy": 5.955654191970825,
      "epoch": 0.5488343341575415,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004976077569172879,
      "loss": 5.8544,
      "mean_token_accuracy": 0.16704583019018174,
      "num_tokens": 11598547.0,
      "step": 6215
    },
    {
      "entropy": 6.076641321182251,
      "epoch": 0.5492758742493818,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004976031747697006,
      "loss": 5.9632,
      "mean_token_accuracy": 0.16434568166732788,
      "num_tokens": 11608407.0,
      "step": 6220
    },
    {
      "entropy": 6.077345609664917,
      "epoch": 0.5497174143412222,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004975985882614435,
      "loss": 5.9317,
      "mean_token_accuracy": 0.15894413441419603,
      "num_tokens": 11618641.0,
      "step": 6225
    },
    {
      "entropy": 6.108517217636108,
      "epoch": 0.5501589544330625,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004975939973926068,
      "loss": 5.9658,
      "mean_token_accuracy": 0.15470803529024124,
      "num_tokens": 11627082.0,
      "step": 6230
    },
    {
      "entropy": 6.035842847824097,
      "epoch": 0.5506004945249029,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004975894021632803,
      "loss": 5.8591,
      "mean_token_accuracy": 0.16572816520929337,
      "num_tokens": 11636045.0,
      "step": 6235
    },
    {
      "entropy": 6.017991161346435,
      "epoch": 0.5510420346167432,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000497584802573554,
      "loss": 5.9476,
      "mean_token_accuracy": 0.1594747096300125,
      "num_tokens": 11646516.0,
      "step": 6240
    },
    {
      "entropy": 5.969410848617554,
      "epoch": 0.5514835747085836,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000497580198623518,
      "loss": 5.8793,
      "mean_token_accuracy": 0.1672035813331604,
      "num_tokens": 11655512.0,
      "step": 6245
    },
    {
      "entropy": 6.07194766998291,
      "epoch": 0.5519251148004238,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004975755903132623,
      "loss": 5.9988,
      "mean_token_accuracy": 0.16115525662899016,
      "num_tokens": 11665641.0,
      "step": 6250
    },
    {
      "entropy": 6.070268392562866,
      "epoch": 0.5523666548922642,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004975709776428776,
      "loss": 5.9626,
      "mean_token_accuracy": 0.14910453855991362,
      "num_tokens": 11674585.0,
      "step": 6255
    },
    {
      "entropy": 6.0138458728790285,
      "epoch": 0.5528081949841046,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000497566360612454,
      "loss": 5.8206,
      "mean_token_accuracy": 0.1709229715168476,
      "num_tokens": 11683498.0,
      "step": 6260
    },
    {
      "entropy": 6.020624494552612,
      "epoch": 0.5532497350759449,
      "grad_norm": 1.421875,
      "learning_rate": 0.000497561739222082,
      "loss": 5.8708,
      "mean_token_accuracy": 0.16468418687582015,
      "num_tokens": 11692663.0,
      "step": 6265
    },
    {
      "entropy": 6.071487998962402,
      "epoch": 0.5536912751677853,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004975571134718521,
      "loss": 6.0272,
      "mean_token_accuracy": 0.1560715228319168,
      "num_tokens": 11701906.0,
      "step": 6270
    },
    {
      "entropy": 6.1155585765838625,
      "epoch": 0.5541328152596255,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004975524833618548,
      "loss": 5.939,
      "mean_token_accuracy": 0.16798952966928482,
      "num_tokens": 11710932.0,
      "step": 6275
    },
    {
      "entropy": 6.000746726989746,
      "epoch": 0.5545743553514659,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000497547848892181,
      "loss": 5.9412,
      "mean_token_accuracy": 0.16227794140577317,
      "num_tokens": 11719953.0,
      "step": 6280
    },
    {
      "entropy": 6.077520656585693,
      "epoch": 0.5550158954433062,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004975432100629215,
      "loss": 5.8256,
      "mean_token_accuracy": 0.16723808497190476,
      "num_tokens": 11728706.0,
      "step": 6285
    },
    {
      "entropy": 6.107032203674317,
      "epoch": 0.5554574355351466,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000497538566874167,
      "loss": 6.0226,
      "mean_token_accuracy": 0.15363354086875916,
      "num_tokens": 11738537.0,
      "step": 6290
    },
    {
      "entropy": 6.031191444396972,
      "epoch": 0.555898975626987,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004975339193260084,
      "loss": 5.839,
      "mean_token_accuracy": 0.16909542083740234,
      "num_tokens": 11747503.0,
      "step": 6295
    },
    {
      "entropy": 6.056135463714599,
      "epoch": 0.5563405157188273,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004975292674185371,
      "loss": 5.9422,
      "mean_token_accuracy": 0.1652400091290474,
      "num_tokens": 11756501.0,
      "step": 6300
    },
    {
      "entropy": 6.039711570739746,
      "epoch": 0.5567820558106676,
      "grad_norm": 1.375,
      "learning_rate": 0.0004975246111518439,
      "loss": 5.9083,
      "mean_token_accuracy": 0.15950924158096313,
      "num_tokens": 11765483.0,
      "step": 6305
    },
    {
      "entropy": 6.124545240402222,
      "epoch": 0.5572235959025079,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004975199505260202,
      "loss": 6.0394,
      "mean_token_accuracy": 0.14992669820785523,
      "num_tokens": 11774941.0,
      "step": 6310
    },
    {
      "entropy": 6.157734012603759,
      "epoch": 0.5576651359943483,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004975152855411571,
      "loss": 6.0228,
      "mean_token_accuracy": 0.15619456619024277,
      "num_tokens": 11785080.0,
      "step": 6315
    },
    {
      "entropy": 6.149894714355469,
      "epoch": 0.5581066760861886,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004975106161973462,
      "loss": 5.898,
      "mean_token_accuracy": 0.16184557229280472,
      "num_tokens": 11793933.0,
      "step": 6320
    },
    {
      "entropy": 6.058248376846313,
      "epoch": 0.558548216178029,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000497505942494679,
      "loss": 5.9241,
      "mean_token_accuracy": 0.16193547546863557,
      "num_tokens": 11803592.0,
      "step": 6325
    },
    {
      "entropy": 5.964681005477905,
      "epoch": 0.5589897562698694,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004975012644332467,
      "loss": 5.8295,
      "mean_token_accuracy": 0.1708283841609955,
      "num_tokens": 11812397.0,
      "step": 6330
    },
    {
      "entropy": 5.99154577255249,
      "epoch": 0.5594312963617096,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004974965820131413,
      "loss": 5.9663,
      "mean_token_accuracy": 0.15441266000270842,
      "num_tokens": 11821734.0,
      "step": 6335
    },
    {
      "entropy": 6.028355646133423,
      "epoch": 0.55987283645355,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004974918952344544,
      "loss": 5.8438,
      "mean_token_accuracy": 0.1648928314447403,
      "num_tokens": 11832107.0,
      "step": 6340
    },
    {
      "entropy": 6.0491269588470455,
      "epoch": 0.5603143765453903,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004974872040972777,
      "loss": 5.894,
      "mean_token_accuracy": 0.15779831409454345,
      "num_tokens": 11841983.0,
      "step": 6345
    },
    {
      "entropy": 5.933441162109375,
      "epoch": 0.5607559166372307,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004974825086017033,
      "loss": 5.8688,
      "mean_token_accuracy": 0.1653630644083023,
      "num_tokens": 11851836.0,
      "step": 6350
    },
    {
      "entropy": 6.113220262527466,
      "epoch": 0.561197456729071,
      "grad_norm": 1.28125,
      "learning_rate": 0.000497477808747823,
      "loss": 5.9227,
      "mean_token_accuracy": 0.16370657682418824,
      "num_tokens": 11861109.0,
      "step": 6355
    },
    {
      "entropy": 6.086468172073364,
      "epoch": 0.5616389968209113,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004974731045357292,
      "loss": 6.0077,
      "mean_token_accuracy": 0.15728640407323838,
      "num_tokens": 11871015.0,
      "step": 6360
    },
    {
      "entropy": 5.971633386611939,
      "epoch": 0.5620805369127517,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004974683959655136,
      "loss": 5.9026,
      "mean_token_accuracy": 0.16690124869346618,
      "num_tokens": 11881306.0,
      "step": 6365
    },
    {
      "entropy": 6.044920015335083,
      "epoch": 0.562522077004592,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004974636830372686,
      "loss": 5.8321,
      "mean_token_accuracy": 0.1697683110833168,
      "num_tokens": 11890251.0,
      "step": 6370
    },
    {
      "entropy": 6.0341572761535645,
      "epoch": 0.5629636170964324,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004974589657510866,
      "loss": 5.9548,
      "mean_token_accuracy": 0.1595368355512619,
      "num_tokens": 11899662.0,
      "step": 6375
    },
    {
      "entropy": 6.0119726181030275,
      "epoch": 0.5634051571882727,
      "grad_norm": 1.25,
      "learning_rate": 0.0004974542441070601,
      "loss": 5.9088,
      "mean_token_accuracy": 0.164400115609169,
      "num_tokens": 11908699.0,
      "step": 6380
    },
    {
      "entropy": 6.074244689941406,
      "epoch": 0.563846697280113,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004974495181052814,
      "loss": 5.9153,
      "mean_token_accuracy": 0.16292343586683272,
      "num_tokens": 11916637.0,
      "step": 6385
    },
    {
      "entropy": 6.057909202575684,
      "epoch": 0.5642882373719533,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004974447877458433,
      "loss": 5.978,
      "mean_token_accuracy": 0.15960414558649064,
      "num_tokens": 11926063.0,
      "step": 6390
    },
    {
      "entropy": 6.042461633682251,
      "epoch": 0.5647297774637937,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004974400530288381,
      "loss": 5.8396,
      "mean_token_accuracy": 0.16759493798017502,
      "num_tokens": 11934861.0,
      "step": 6395
    },
    {
      "entropy": 6.064510822296143,
      "epoch": 0.5651713175556341,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004974353139543589,
      "loss": 5.9472,
      "mean_token_accuracy": 0.1592862457036972,
      "num_tokens": 11944403.0,
      "step": 6400
    },
    {
      "entropy": 6.09717435836792,
      "epoch": 0.5656128576474744,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004974305705224983,
      "loss": 6.0273,
      "mean_token_accuracy": 0.16373891979455948,
      "num_tokens": 11953306.0,
      "step": 6405
    },
    {
      "entropy": 6.048258686065674,
      "epoch": 0.5660543977393148,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004974258227333493,
      "loss": 5.9229,
      "mean_token_accuracy": 0.16490714997053146,
      "num_tokens": 11962727.0,
      "step": 6410
    },
    {
      "entropy": 6.083757448196411,
      "epoch": 0.566495937831155,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004974210705870052,
      "loss": 6.0653,
      "mean_token_accuracy": 0.1529408350586891,
      "num_tokens": 11972375.0,
      "step": 6415
    },
    {
      "entropy": 5.9368589401245115,
      "epoch": 0.5669374779229954,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004974163140835586,
      "loss": 5.8357,
      "mean_token_accuracy": 0.16640966832637788,
      "num_tokens": 11981989.0,
      "step": 6420
    },
    {
      "entropy": 6.001610279083252,
      "epoch": 0.5673790180148357,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004974115532231029,
      "loss": 5.8865,
      "mean_token_accuracy": 0.16079047322273254,
      "num_tokens": 11990865.0,
      "step": 6425
    },
    {
      "entropy": 6.038493490219116,
      "epoch": 0.5678205581066761,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004974067880057315,
      "loss": 5.8903,
      "mean_token_accuracy": 0.16845848858356477,
      "num_tokens": 12000697.0,
      "step": 6430
    },
    {
      "entropy": 6.008808898925781,
      "epoch": 0.5682620981985165,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004974020184315375,
      "loss": 5.9645,
      "mean_token_accuracy": 0.16037520170211791,
      "num_tokens": 12010560.0,
      "step": 6435
    },
    {
      "entropy": 6.135006332397461,
      "epoch": 0.5687036382903567,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004973972445006144,
      "loss": 5.9804,
      "mean_token_accuracy": 0.16373104155063628,
      "num_tokens": 12020760.0,
      "step": 6440
    },
    {
      "entropy": 5.968979454040527,
      "epoch": 0.5691451783821971,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004973924662130559,
      "loss": 5.8425,
      "mean_token_accuracy": 0.1618216186761856,
      "num_tokens": 12030467.0,
      "step": 6445
    },
    {
      "entropy": 6.0167624950408936,
      "epoch": 0.5695867184740374,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004973876835689554,
      "loss": 5.9458,
      "mean_token_accuracy": 0.15894586592912674,
      "num_tokens": 12039897.0,
      "step": 6450
    },
    {
      "entropy": 6.1160314083099365,
      "epoch": 0.5700282585658778,
      "grad_norm": 1.375,
      "learning_rate": 0.0004973828965684066,
      "loss": 5.9589,
      "mean_token_accuracy": 0.1576131209731102,
      "num_tokens": 12048997.0,
      "step": 6455
    },
    {
      "entropy": 6.03399133682251,
      "epoch": 0.5704697986577181,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004973781052115035,
      "loss": 5.8459,
      "mean_token_accuracy": 0.17132855206727982,
      "num_tokens": 12057493.0,
      "step": 6460
    },
    {
      "entropy": 5.896988725662231,
      "epoch": 0.5709113387495585,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004973733094983396,
      "loss": 5.917,
      "mean_token_accuracy": 0.16548199504613875,
      "num_tokens": 12066329.0,
      "step": 6465
    },
    {
      "entropy": 6.044803476333618,
      "epoch": 0.5713528788413988,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004973685094290092,
      "loss": 5.966,
      "mean_token_accuracy": 0.15556841790676118,
      "num_tokens": 12076022.0,
      "step": 6470
    },
    {
      "entropy": 6.063698959350586,
      "epoch": 0.5717944189332391,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000497363705003606,
      "loss": 5.9132,
      "mean_token_accuracy": 0.15902747064828873,
      "num_tokens": 12085750.0,
      "step": 6475
    },
    {
      "entropy": 6.0139764785766605,
      "epoch": 0.5722359590250795,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004973588962222244,
      "loss": 5.9232,
      "mean_token_accuracy": 0.16362606585025788,
      "num_tokens": 12094935.0,
      "step": 6480
    },
    {
      "entropy": 5.97671914100647,
      "epoch": 0.5726774991169198,
      "grad_norm": 1.375,
      "learning_rate": 0.0004973540830849585,
      "loss": 5.8763,
      "mean_token_accuracy": 0.16199066191911698,
      "num_tokens": 12104970.0,
      "step": 6485
    },
    {
      "entropy": 6.06100058555603,
      "epoch": 0.5731190392087602,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004973492655919025,
      "loss": 5.9691,
      "mean_token_accuracy": 0.16144821047782898,
      "num_tokens": 12116195.0,
      "step": 6490
    },
    {
      "entropy": 6.093713283538818,
      "epoch": 0.5735605793006004,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004973444437431508,
      "loss": 5.9069,
      "mean_token_accuracy": 0.16268889904022216,
      "num_tokens": 12125485.0,
      "step": 6495
    },
    {
      "entropy": 5.996689748764038,
      "epoch": 0.5740021193924408,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004973396175387981,
      "loss": 5.9323,
      "mean_token_accuracy": 0.16517322808504104,
      "num_tokens": 12135474.0,
      "step": 6500
    },
    {
      "entropy": 6.063704156875611,
      "epoch": 0.5744436594842812,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004973347869789385,
      "loss": 5.9733,
      "mean_token_accuracy": 0.15778698474168779,
      "num_tokens": 12144945.0,
      "step": 6505
    },
    {
      "entropy": 6.049719285964966,
      "epoch": 0.5748851995761215,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004973299520636671,
      "loss": 5.9195,
      "mean_token_accuracy": 0.1623150423169136,
      "num_tokens": 12154893.0,
      "step": 6510
    },
    {
      "entropy": 6.064568901062012,
      "epoch": 0.5753267396679619,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004973251127930782,
      "loss": 5.991,
      "mean_token_accuracy": 0.15995137244462967,
      "num_tokens": 12164047.0,
      "step": 6515
    },
    {
      "entropy": 6.106995487213135,
      "epoch": 0.5757682797598022,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004973202691672669,
      "loss": 5.971,
      "mean_token_accuracy": 0.15734548568725587,
      "num_tokens": 12172965.0,
      "step": 6520
    },
    {
      "entropy": 6.065262413024902,
      "epoch": 0.5762098198516425,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004973154211863279,
      "loss": 5.9858,
      "mean_token_accuracy": 0.15848653614521027,
      "num_tokens": 12183534.0,
      "step": 6525
    },
    {
      "entropy": 6.019828367233276,
      "epoch": 0.5766513599434828,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004973105688503563,
      "loss": 5.9443,
      "mean_token_accuracy": 0.16066363602876663,
      "num_tokens": 12194370.0,
      "step": 6530
    },
    {
      "entropy": 6.086677503585816,
      "epoch": 0.5770929000353232,
      "grad_norm": 1.375,
      "learning_rate": 0.000497305712159447,
      "loss": 5.953,
      "mean_token_accuracy": 0.15858365148305892,
      "num_tokens": 12203765.0,
      "step": 6535
    },
    {
      "entropy": 6.064221906661987,
      "epoch": 0.5775344401271636,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004973008511136954,
      "loss": 5.8947,
      "mean_token_accuracy": 0.16712932288646698,
      "num_tokens": 12212889.0,
      "step": 6540
    },
    {
      "entropy": 6.027991771697998,
      "epoch": 0.5779759802190039,
      "grad_norm": 1.625,
      "learning_rate": 0.0004972959857131963,
      "loss": 5.961,
      "mean_token_accuracy": 0.15869880318641663,
      "num_tokens": 12222055.0,
      "step": 6545
    },
    {
      "entropy": 6.09398717880249,
      "epoch": 0.5784175203108443,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004972911159580454,
      "loss": 5.9716,
      "mean_token_accuracy": 0.15896429717540742,
      "num_tokens": 12232092.0,
      "step": 6550
    },
    {
      "entropy": 6.056658220291138,
      "epoch": 0.5788590604026845,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000497286241848338,
      "loss": 5.9823,
      "mean_token_accuracy": 0.15855540782213212,
      "num_tokens": 12241411.0,
      "step": 6555
    },
    {
      "entropy": 6.101971817016602,
      "epoch": 0.5793006004945249,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004972813633841696,
      "loss": 5.8669,
      "mean_token_accuracy": 0.16081757098436356,
      "num_tokens": 12249735.0,
      "step": 6560
    },
    {
      "entropy": 6.011325693130493,
      "epoch": 0.5797421405863652,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004972764805656355,
      "loss": 5.8807,
      "mean_token_accuracy": 0.16948403120040895,
      "num_tokens": 12259132.0,
      "step": 6565
    },
    {
      "entropy": 5.9479166030883786,
      "epoch": 0.5801836806782056,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004972715933928318,
      "loss": 5.9587,
      "mean_token_accuracy": 0.1681016907095909,
      "num_tokens": 12269401.0,
      "step": 6570
    },
    {
      "entropy": 6.1060340881347654,
      "epoch": 0.580625220770046,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004972667018658539,
      "loss": 5.9837,
      "mean_token_accuracy": 0.15532615929841995,
      "num_tokens": 12279163.0,
      "step": 6575
    },
    {
      "entropy": 6.030529546737671,
      "epoch": 0.5810667608618862,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004972618059847976,
      "loss": 5.872,
      "mean_token_accuracy": 0.16901613771915436,
      "num_tokens": 12287942.0,
      "step": 6580
    },
    {
      "entropy": 5.988468074798584,
      "epoch": 0.5815083009537266,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004972569057497591,
      "loss": 5.8788,
      "mean_token_accuracy": 0.16304004341363906,
      "num_tokens": 12296717.0,
      "step": 6585
    },
    {
      "entropy": 6.02855339050293,
      "epoch": 0.5819498410455669,
      "grad_norm": 1.375,
      "learning_rate": 0.0004972520011608341,
      "loss": 5.8562,
      "mean_token_accuracy": 0.17002730965614318,
      "num_tokens": 12305983.0,
      "step": 6590
    },
    {
      "entropy": 5.991772890090942,
      "epoch": 0.5823913811374073,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004972470922181189,
      "loss": 5.932,
      "mean_token_accuracy": 0.15614658892154692,
      "num_tokens": 12315338.0,
      "step": 6595
    },
    {
      "entropy": 6.037394952774048,
      "epoch": 0.5828329212292476,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004972421789217094,
      "loss": 6.0198,
      "mean_token_accuracy": 0.15138231888413428,
      "num_tokens": 12324828.0,
      "step": 6600
    },
    {
      "entropy": 6.185899257659912,
      "epoch": 0.583274461321088,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004972372612717022,
      "loss": 5.893,
      "mean_token_accuracy": 0.1675851821899414,
      "num_tokens": 12332844.0,
      "step": 6605
    },
    {
      "entropy": 6.062373638153076,
      "epoch": 0.5837160014129283,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004972323392681934,
      "loss": 5.9112,
      "mean_token_accuracy": 0.1606480211019516,
      "num_tokens": 12342062.0,
      "step": 6610
    },
    {
      "entropy": 5.956027936935425,
      "epoch": 0.5841575415047686,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004972274129112794,
      "loss": 5.9835,
      "mean_token_accuracy": 0.1534557655453682,
      "num_tokens": 12352071.0,
      "step": 6615
    },
    {
      "entropy": 6.217044639587402,
      "epoch": 0.584599081596609,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004972224822010567,
      "loss": 6.0609,
      "mean_token_accuracy": 0.15077690929174423,
      "num_tokens": 12362025.0,
      "step": 6620
    },
    {
      "entropy": 6.157152462005615,
      "epoch": 0.5850406216884493,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004972175471376222,
      "loss": 6.043,
      "mean_token_accuracy": 0.15202385187149048,
      "num_tokens": 12371547.0,
      "step": 6625
    },
    {
      "entropy": 6.0660265445709225,
      "epoch": 0.5854821617802897,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004972126077210722,
      "loss": 5.9528,
      "mean_token_accuracy": 0.156717449426651,
      "num_tokens": 12381385.0,
      "step": 6630
    },
    {
      "entropy": 6.107764101028442,
      "epoch": 0.5859237018721299,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004972076639515035,
      "loss": 5.9961,
      "mean_token_accuracy": 0.1575634926557541,
      "num_tokens": 12391302.0,
      "step": 6635
    },
    {
      "entropy": 5.931721448898315,
      "epoch": 0.5863652419639703,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004972027158290131,
      "loss": 5.8827,
      "mean_token_accuracy": 0.16473877429962158,
      "num_tokens": 12400388.0,
      "step": 6640
    },
    {
      "entropy": 6.0112042903900145,
      "epoch": 0.5868067820558107,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000497197763353698,
      "loss": 5.9466,
      "mean_token_accuracy": 0.15398259609937667,
      "num_tokens": 12409695.0,
      "step": 6645
    },
    {
      "entropy": 6.130546712875367,
      "epoch": 0.587248322147651,
      "grad_norm": 1.3125,
      "learning_rate": 0.000497192806525655,
      "loss": 5.8142,
      "mean_token_accuracy": 0.1708827257156372,
      "num_tokens": 12418728.0,
      "step": 6650
    },
    {
      "entropy": 6.000955820083618,
      "epoch": 0.5876898622394914,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004971878453449811,
      "loss": 5.9827,
      "mean_token_accuracy": 0.16348764598369597,
      "num_tokens": 12428072.0,
      "step": 6655
    },
    {
      "entropy": 6.079076623916626,
      "epoch": 0.5881314023313317,
      "grad_norm": 1.296875,
      "learning_rate": 0.000497182879811774,
      "loss": 5.9081,
      "mean_token_accuracy": 0.1675763249397278,
      "num_tokens": 12436624.0,
      "step": 6660
    },
    {
      "entropy": 6.015807867050171,
      "epoch": 0.588572942423172,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004971779099261304,
      "loss": 5.9174,
      "mean_token_accuracy": 0.1616780251264572,
      "num_tokens": 12446370.0,
      "step": 6665
    },
    {
      "entropy": 6.032747650146485,
      "epoch": 0.5890144825150123,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000497172935688148,
      "loss": 5.9717,
      "mean_token_accuracy": 0.15881636887788772,
      "num_tokens": 12456705.0,
      "step": 6670
    },
    {
      "entropy": 6.0455207347869875,
      "epoch": 0.5894560226068527,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004971679570979242,
      "loss": 5.9891,
      "mean_token_accuracy": 0.16214286386966706,
      "num_tokens": 12466150.0,
      "step": 6675
    },
    {
      "entropy": 6.019035196304321,
      "epoch": 0.5898975626986931,
      "grad_norm": 1.25,
      "learning_rate": 0.0004971629741555565,
      "loss": 5.8321,
      "mean_token_accuracy": 0.16692282557487487,
      "num_tokens": 12475051.0,
      "step": 6680
    },
    {
      "entropy": 6.051440095901489,
      "epoch": 0.5903391027905334,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004971579868611423,
      "loss": 5.908,
      "mean_token_accuracy": 0.16642434149980545,
      "num_tokens": 12485427.0,
      "step": 6685
    },
    {
      "entropy": 5.968716716766357,
      "epoch": 0.5907806428823738,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004971529952147797,
      "loss": 5.866,
      "mean_token_accuracy": 0.16762274205684663,
      "num_tokens": 12494489.0,
      "step": 6690
    },
    {
      "entropy": 6.060288572311402,
      "epoch": 0.591222182974214,
      "grad_norm": 1.640625,
      "learning_rate": 0.000497147999216566,
      "loss": 5.8809,
      "mean_token_accuracy": 0.16746745109558106,
      "num_tokens": 12503656.0,
      "step": 6695
    },
    {
      "entropy": 6.088530397415161,
      "epoch": 0.5916637230660544,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004971429988665996,
      "loss": 5.9356,
      "mean_token_accuracy": 0.15473153740167617,
      "num_tokens": 12512940.0,
      "step": 6700
    },
    {
      "entropy": 6.049561214447022,
      "epoch": 0.5921052631578947,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000497137994164978,
      "loss": 5.8734,
      "mean_token_accuracy": 0.16000210046768187,
      "num_tokens": 12523186.0,
      "step": 6705
    },
    {
      "entropy": 6.0405772686004635,
      "epoch": 0.5925468032497351,
      "grad_norm": 1.5,
      "learning_rate": 0.0004971329851117996,
      "loss": 5.9207,
      "mean_token_accuracy": 0.1611115738749504,
      "num_tokens": 12533018.0,
      "step": 6710
    },
    {
      "entropy": 6.150906324386597,
      "epoch": 0.5929883433415755,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004971279717071622,
      "loss": 5.9854,
      "mean_token_accuracy": 0.15813910067081452,
      "num_tokens": 12542889.0,
      "step": 6715
    },
    {
      "entropy": 6.118391656875611,
      "epoch": 0.5934298834334157,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004971229539511643,
      "loss": 5.9951,
      "mean_token_accuracy": 0.15295567288994788,
      "num_tokens": 12553625.0,
      "step": 6720
    },
    {
      "entropy": 5.998600721359253,
      "epoch": 0.5938714235252561,
      "grad_norm": 1.21875,
      "learning_rate": 0.000497117931843904,
      "loss": 5.9109,
      "mean_token_accuracy": 0.16377592980861663,
      "num_tokens": 12563012.0,
      "step": 6725
    },
    {
      "entropy": 6.154142999649048,
      "epoch": 0.5943129636170964,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004971129053854797,
      "loss": 6.0368,
      "mean_token_accuracy": 0.14983948394656182,
      "num_tokens": 12572892.0,
      "step": 6730
    },
    {
      "entropy": 6.081307983398437,
      "epoch": 0.5947545037089368,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00049710787457599,
      "loss": 5.8738,
      "mean_token_accuracy": 0.16873954683542253,
      "num_tokens": 12581550.0,
      "step": 6735
    },
    {
      "entropy": 6.030810117721558,
      "epoch": 0.5951960438007771,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004971028394155332,
      "loss": 5.9572,
      "mean_token_accuracy": 0.15993183553218843,
      "num_tokens": 12590599.0,
      "step": 6740
    },
    {
      "entropy": 6.106176710128784,
      "epoch": 0.5956375838926175,
      "grad_norm": 1.375,
      "learning_rate": 0.0004970977999042082,
      "loss": 5.9152,
      "mean_token_accuracy": 0.16037029176950454,
      "num_tokens": 12599497.0,
      "step": 6745
    },
    {
      "entropy": 6.025887393951416,
      "epoch": 0.5960791239844578,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004970927560421137,
      "loss": 5.9749,
      "mean_token_accuracy": 0.16369700878858567,
      "num_tokens": 12608273.0,
      "step": 6750
    },
    {
      "entropy": 5.978644990921021,
      "epoch": 0.5965206640762981,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004970877078293483,
      "loss": 5.8404,
      "mean_token_accuracy": 0.16632604449987412,
      "num_tokens": 12617287.0,
      "step": 6755
    },
    {
      "entropy": 6.0520916938781735,
      "epoch": 0.5969622041681385,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004970826552660111,
      "loss": 5.9189,
      "mean_token_accuracy": 0.1645960807800293,
      "num_tokens": 12626874.0,
      "step": 6760
    },
    {
      "entropy": 6.046804714202881,
      "epoch": 0.5974037442599788,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004970775983522009,
      "loss": 5.9547,
      "mean_token_accuracy": 0.15649548768997193,
      "num_tokens": 12635218.0,
      "step": 6765
    },
    {
      "entropy": 6.0162660598754885,
      "epoch": 0.5978452843518192,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004970725370880168,
      "loss": 5.927,
      "mean_token_accuracy": 0.1587716147303581,
      "num_tokens": 12644395.0,
      "step": 6770
    },
    {
      "entropy": 6.0681376457214355,
      "epoch": 0.5982868244436594,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004970674714735581,
      "loss": 5.9416,
      "mean_token_accuracy": 0.16704253554344178,
      "num_tokens": 12654540.0,
      "step": 6775
    },
    {
      "entropy": 6.038317584991455,
      "epoch": 0.5987283645354998,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004970624015089239,
      "loss": 5.8046,
      "mean_token_accuracy": 0.17055938690900802,
      "num_tokens": 12663546.0,
      "step": 6780
    },
    {
      "entropy": 5.958972406387329,
      "epoch": 0.5991699046273402,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004970573271942135,
      "loss": 5.9125,
      "mean_token_accuracy": 0.1605516865849495,
      "num_tokens": 12672353.0,
      "step": 6785
    },
    {
      "entropy": 5.948368644714355,
      "epoch": 0.5996114447191805,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004970522485295265,
      "loss": 5.8077,
      "mean_token_accuracy": 0.1750718891620636,
      "num_tokens": 12680833.0,
      "step": 6790
    },
    {
      "entropy": 6.008566856384277,
      "epoch": 0.6000529848110209,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004970471655149621,
      "loss": 5.7693,
      "mean_token_accuracy": 0.168387308716774,
      "num_tokens": 12690033.0,
      "step": 6795
    },
    {
      "entropy": 5.891954612731934,
      "epoch": 0.6004945249028611,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00049704207815062,
      "loss": 5.8498,
      "mean_token_accuracy": 0.17069175988435745,
      "num_tokens": 12700127.0,
      "step": 6800
    },
    {
      "entropy": 6.0572412490844725,
      "epoch": 0.6009360649947015,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004970369864366,
      "loss": 5.8864,
      "mean_token_accuracy": 0.1621399462223053,
      "num_tokens": 12709803.0,
      "step": 6805
    },
    {
      "entropy": 6.027445650100708,
      "epoch": 0.6013776050865418,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004970318903730017,
      "loss": 5.8683,
      "mean_token_accuracy": 0.16281817108392715,
      "num_tokens": 12718876.0,
      "step": 6810
    },
    {
      "entropy": 6.07098970413208,
      "epoch": 0.6018191451783822,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004970267899599249,
      "loss": 5.9846,
      "mean_token_accuracy": 0.1596504732966423,
      "num_tokens": 12728745.0,
      "step": 6815
    },
    {
      "entropy": 6.074090528488159,
      "epoch": 0.6022606852702226,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004970216851974695,
      "loss": 5.9778,
      "mean_token_accuracy": 0.15287121534347534,
      "num_tokens": 12738764.0,
      "step": 6820
    },
    {
      "entropy": 5.971881341934204,
      "epoch": 0.6027022253620629,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004970165760857356,
      "loss": 5.9192,
      "mean_token_accuracy": 0.15215612798929215,
      "num_tokens": 12747888.0,
      "step": 6825
    },
    {
      "entropy": 6.147901725769043,
      "epoch": 0.6031437654539032,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004970114626248232,
      "loss": 5.9269,
      "mean_token_accuracy": 0.16107769459486007,
      "num_tokens": 12757763.0,
      "step": 6830
    },
    {
      "entropy": 6.051682806015014,
      "epoch": 0.6035853055457435,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004970063448148326,
      "loss": 5.965,
      "mean_token_accuracy": 0.16029387414455415,
      "num_tokens": 12767924.0,
      "step": 6835
    },
    {
      "entropy": 6.023601818084717,
      "epoch": 0.6040268456375839,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004970012226558639,
      "loss": 5.977,
      "mean_token_accuracy": 0.1597777470946312,
      "num_tokens": 12777140.0,
      "step": 6840
    },
    {
      "entropy": 6.055026721954346,
      "epoch": 0.6044683857294242,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004969960961480176,
      "loss": 5.9521,
      "mean_token_accuracy": 0.16517617404460908,
      "num_tokens": 12786731.0,
      "step": 6845
    },
    {
      "entropy": 6.070271015167236,
      "epoch": 0.6049099258212646,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004969909652913939,
      "loss": 5.8877,
      "mean_token_accuracy": 0.16475060284137727,
      "num_tokens": 12796839.0,
      "step": 6850
    },
    {
      "entropy": 6.030538511276245,
      "epoch": 0.605351465913105,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004969858300860935,
      "loss": 5.9737,
      "mean_token_accuracy": 0.1638708546757698,
      "num_tokens": 12805944.0,
      "step": 6855
    },
    {
      "entropy": 6.017725276947021,
      "epoch": 0.6057930060049452,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004969806905322169,
      "loss": 5.9286,
      "mean_token_accuracy": 0.16286252290010453,
      "num_tokens": 12814692.0,
      "step": 6860
    },
    {
      "entropy": 6.016694974899292,
      "epoch": 0.6062345460967856,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004969755466298649,
      "loss": 5.8609,
      "mean_token_accuracy": 0.1611609011888504,
      "num_tokens": 12824506.0,
      "step": 6865
    },
    {
      "entropy": 6.07229380607605,
      "epoch": 0.6066760861886259,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004969703983791382,
      "loss": 5.9186,
      "mean_token_accuracy": 0.16030610352754593,
      "num_tokens": 12835157.0,
      "step": 6870
    },
    {
      "entropy": 6.005796813964844,
      "epoch": 0.6071176262804663,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004969652457801376,
      "loss": 5.846,
      "mean_token_accuracy": 0.16162244081497193,
      "num_tokens": 12844281.0,
      "step": 6875
    },
    {
      "entropy": 6.094640493392944,
      "epoch": 0.6075591663723066,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004969600888329641,
      "loss": 5.9086,
      "mean_token_accuracy": 0.16073376536369324,
      "num_tokens": 12854486.0,
      "step": 6880
    },
    {
      "entropy": 6.048177909851074,
      "epoch": 0.608000706464147,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004969549275377186,
      "loss": 5.8907,
      "mean_token_accuracy": 0.16041633486747742,
      "num_tokens": 12863574.0,
      "step": 6885
    },
    {
      "entropy": 5.933614730834961,
      "epoch": 0.6084422465559873,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004969497618945024,
      "loss": 5.8647,
      "mean_token_accuracy": 0.16501740068197251,
      "num_tokens": 12872450.0,
      "step": 6890
    },
    {
      "entropy": 5.981774759292603,
      "epoch": 0.6088837866478276,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004969445919034165,
      "loss": 5.8105,
      "mean_token_accuracy": 0.17405697256326674,
      "num_tokens": 12881299.0,
      "step": 6895
    },
    {
      "entropy": 5.927428197860718,
      "epoch": 0.609325326739668,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004969394175645623,
      "loss": 5.7886,
      "mean_token_accuracy": 0.1706787884235382,
      "num_tokens": 12889943.0,
      "step": 6900
    },
    {
      "entropy": 5.946518993377685,
      "epoch": 0.6097668668315083,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004969342388780411,
      "loss": 5.8944,
      "mean_token_accuracy": 0.16340020298957825,
      "num_tokens": 12899497.0,
      "step": 6905
    },
    {
      "entropy": 6.05028657913208,
      "epoch": 0.6102084069233487,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004969290558439546,
      "loss": 5.8661,
      "mean_token_accuracy": 0.16563009470701218,
      "num_tokens": 12908677.0,
      "step": 6910
    },
    {
      "entropy": 6.065933132171631,
      "epoch": 0.6106499470151889,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004969238684624038,
      "loss": 5.9824,
      "mean_token_accuracy": 0.15356329083442688,
      "num_tokens": 12918468.0,
      "step": 6915
    },
    {
      "entropy": 6.017871952056884,
      "epoch": 0.6110914871070293,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004969186767334908,
      "loss": 5.8613,
      "mean_token_accuracy": 0.17257402390241622,
      "num_tokens": 12928527.0,
      "step": 6920
    },
    {
      "entropy": 6.052063798904419,
      "epoch": 0.6115330271988697,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004969134806573172,
      "loss": 5.8645,
      "mean_token_accuracy": 0.1648426666855812,
      "num_tokens": 12937021.0,
      "step": 6925
    },
    {
      "entropy": 5.897130489349365,
      "epoch": 0.61197456729071,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004969082802339846,
      "loss": 5.8116,
      "mean_token_accuracy": 0.17037201374769212,
      "num_tokens": 12945570.0,
      "step": 6930
    },
    {
      "entropy": 5.988294363021851,
      "epoch": 0.6124161073825504,
      "grad_norm": 1.3125,
      "learning_rate": 0.000496903075463595,
      "loss": 5.9244,
      "mean_token_accuracy": 0.16518771648406982,
      "num_tokens": 12954594.0,
      "step": 6935
    },
    {
      "entropy": 6.077741241455078,
      "epoch": 0.6128576474743906,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004968978663462504,
      "loss": 5.9046,
      "mean_token_accuracy": 0.16602875143289567,
      "num_tokens": 12963328.0,
      "step": 6940
    },
    {
      "entropy": 5.99314284324646,
      "epoch": 0.613299187566231,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004968926528820529,
      "loss": 5.8949,
      "mean_token_accuracy": 0.16651069819927217,
      "num_tokens": 12972806.0,
      "step": 6945
    },
    {
      "entropy": 6.007574129104614,
      "epoch": 0.6137407276580713,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004968874350711044,
      "loss": 5.8937,
      "mean_token_accuracy": 0.1642751067876816,
      "num_tokens": 12981940.0,
      "step": 6950
    },
    {
      "entropy": 6.007917213439941,
      "epoch": 0.6141822677499117,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004968822129135072,
      "loss": 5.8541,
      "mean_token_accuracy": 0.16383134424686432,
      "num_tokens": 12990501.0,
      "step": 6955
    },
    {
      "entropy": 6.093807888031006,
      "epoch": 0.6146238078417521,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004968769864093637,
      "loss": 5.9088,
      "mean_token_accuracy": 0.16012341976165773,
      "num_tokens": 12999734.0,
      "step": 6960
    },
    {
      "entropy": 6.101202297210693,
      "epoch": 0.6150653479335924,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004968717555587762,
      "loss": 5.9861,
      "mean_token_accuracy": 0.15358458906412126,
      "num_tokens": 13009323.0,
      "step": 6965
    },
    {
      "entropy": 6.119562292098999,
      "epoch": 0.6155068880254327,
      "grad_norm": 1.5,
      "learning_rate": 0.0004968665203618472,
      "loss": 5.9908,
      "mean_token_accuracy": 0.1654104083776474,
      "num_tokens": 13018136.0,
      "step": 6970
    },
    {
      "entropy": 6.0345367908477785,
      "epoch": 0.615948428117273,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004968612808186793,
      "loss": 5.9365,
      "mean_token_accuracy": 0.16047066450119019,
      "num_tokens": 13026494.0,
      "step": 6975
    },
    {
      "entropy": 6.053148508071899,
      "epoch": 0.6163899682091134,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004968560369293751,
      "loss": 5.9433,
      "mean_token_accuracy": 0.16229794472455977,
      "num_tokens": 13036548.0,
      "step": 6980
    },
    {
      "entropy": 6.089382314682007,
      "epoch": 0.6168315083009537,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004968507886940372,
      "loss": 5.954,
      "mean_token_accuracy": 0.16225423216819762,
      "num_tokens": 13045653.0,
      "step": 6985
    },
    {
      "entropy": 6.10148777961731,
      "epoch": 0.6172730483927941,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004968455361127686,
      "loss": 6.0024,
      "mean_token_accuracy": 0.1586548089981079,
      "num_tokens": 13056204.0,
      "step": 6990
    },
    {
      "entropy": 6.0787682056427,
      "epoch": 0.6177145884846345,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004968402791856721,
      "loss": 5.9765,
      "mean_token_accuracy": 0.16385324150323868,
      "num_tokens": 13067215.0,
      "step": 6995
    },
    {
      "entropy": 6.088096475601196,
      "epoch": 0.6181561285764747,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004968350179128507,
      "loss": 5.9221,
      "mean_token_accuracy": 0.1585768058896065,
      "num_tokens": 13076824.0,
      "step": 7000
    },
    {
      "entropy": 5.991998910903931,
      "epoch": 0.6185976686683151,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004968297522944075,
      "loss": 5.8726,
      "mean_token_accuracy": 0.16714178770780563,
      "num_tokens": 13086205.0,
      "step": 7005
    },
    {
      "entropy": 5.973235273361206,
      "epoch": 0.6190392087601554,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004968244823304455,
      "loss": 5.8218,
      "mean_token_accuracy": 0.16783057302236556,
      "num_tokens": 13096381.0,
      "step": 7010
    },
    {
      "entropy": 6.062705564498901,
      "epoch": 0.6194807488519958,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004968192080210683,
      "loss": 5.9077,
      "mean_token_accuracy": 0.1670522004365921,
      "num_tokens": 13106533.0,
      "step": 7015
    },
    {
      "entropy": 5.984185743331909,
      "epoch": 0.619922288943836,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004968139293663787,
      "loss": 5.8625,
      "mean_token_accuracy": 0.1657854288816452,
      "num_tokens": 13116605.0,
      "step": 7020
    },
    {
      "entropy": 5.997422504425049,
      "epoch": 0.6203638290356764,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004968086463664806,
      "loss": 5.8531,
      "mean_token_accuracy": 0.17010919004678726,
      "num_tokens": 13125316.0,
      "step": 7025
    },
    {
      "entropy": 6.074321842193603,
      "epoch": 0.6208053691275168,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000496803359021477,
      "loss": 6.0405,
      "mean_token_accuracy": 0.16099716424942018,
      "num_tokens": 13135595.0,
      "step": 7030
    },
    {
      "entropy": 6.097686433792115,
      "epoch": 0.6212469092193571,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004967980673314719,
      "loss": 5.9492,
      "mean_token_accuracy": 0.15932941138744355,
      "num_tokens": 13144395.0,
      "step": 7035
    },
    {
      "entropy": 6.023861026763916,
      "epoch": 0.6216884493111975,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004967927712965688,
      "loss": 5.8975,
      "mean_token_accuracy": 0.16490717828273774,
      "num_tokens": 13153906.0,
      "step": 7040
    },
    {
      "entropy": 6.061697959899902,
      "epoch": 0.6221299894030378,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004967874709168714,
      "loss": 5.853,
      "mean_token_accuracy": 0.16454191356897355,
      "num_tokens": 13163840.0,
      "step": 7045
    },
    {
      "entropy": 6.029821729660034,
      "epoch": 0.6225715294948782,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004967821661924837,
      "loss": 5.8585,
      "mean_token_accuracy": 0.16540730595588685,
      "num_tokens": 13173224.0,
      "step": 7050
    },
    {
      "entropy": 6.043258905410767,
      "epoch": 0.6230130695867184,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004967768571235094,
      "loss": 5.9062,
      "mean_token_accuracy": 0.15961679369211196,
      "num_tokens": 13182664.0,
      "step": 7055
    },
    {
      "entropy": 5.974499702453613,
      "epoch": 0.6234546096785588,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004967715437100526,
      "loss": 5.8154,
      "mean_token_accuracy": 0.16828686892986297,
      "num_tokens": 13191076.0,
      "step": 7060
    },
    {
      "entropy": 6.139584732055664,
      "epoch": 0.6238961497703992,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004967662259522174,
      "loss": 6.0221,
      "mean_token_accuracy": 0.15660104751586915,
      "num_tokens": 13201075.0,
      "step": 7065
    },
    {
      "entropy": 6.096574211120606,
      "epoch": 0.6243376898622395,
      "grad_norm": 1.265625,
      "learning_rate": 0.000496760903850108,
      "loss": 5.93,
      "mean_token_accuracy": 0.1620198294520378,
      "num_tokens": 13210602.0,
      "step": 7070
    },
    {
      "entropy": 6.080165433883667,
      "epoch": 0.6247792299540799,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004967555774038286,
      "loss": 5.9076,
      "mean_token_accuracy": 0.15945767164230346,
      "num_tokens": 13219601.0,
      "step": 7075
    },
    {
      "entropy": 6.0232421398162845,
      "epoch": 0.6252207700459201,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004967502466134836,
      "loss": 5.9811,
      "mean_token_accuracy": 0.15633575469255448,
      "num_tokens": 13230675.0,
      "step": 7080
    },
    {
      "entropy": 6.057520818710327,
      "epoch": 0.6256623101377605,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004967449114791773,
      "loss": 5.9156,
      "mean_token_accuracy": 0.16074651181697847,
      "num_tokens": 13240349.0,
      "step": 7085
    },
    {
      "entropy": 6.054627895355225,
      "epoch": 0.6261038502296008,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004967395720010143,
      "loss": 5.9756,
      "mean_token_accuracy": 0.15755234509706498,
      "num_tokens": 13249867.0,
      "step": 7090
    },
    {
      "entropy": 6.089642667770386,
      "epoch": 0.6265453903214412,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004967342281790992,
      "loss": 5.9095,
      "mean_token_accuracy": 0.16455434262752533,
      "num_tokens": 13259143.0,
      "step": 7095
    },
    {
      "entropy": 6.015751934051513,
      "epoch": 0.6269869304132816,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004967288800135368,
      "loss": 5.84,
      "mean_token_accuracy": 0.16161815971136093,
      "num_tokens": 13268673.0,
      "step": 7100
    },
    {
      "entropy": 6.036752605438233,
      "epoch": 0.6274284705051218,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004967235275044315,
      "loss": 5.8735,
      "mean_token_accuracy": 0.16451816260814667,
      "num_tokens": 13278035.0,
      "step": 7105
    },
    {
      "entropy": 6.101490259170532,
      "epoch": 0.6278700105969622,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004967181706518886,
      "loss": 5.9788,
      "mean_token_accuracy": 0.16411617249250413,
      "num_tokens": 13286950.0,
      "step": 7110
    },
    {
      "entropy": 5.9993843078613285,
      "epoch": 0.6283115506888025,
      "grad_norm": 2.0,
      "learning_rate": 0.0004967128094560128,
      "loss": 5.8609,
      "mean_token_accuracy": 0.17262431532144545,
      "num_tokens": 13296425.0,
      "step": 7115
    },
    {
      "entropy": 6.1313232421875,
      "epoch": 0.6287530907806429,
      "grad_norm": 1.5,
      "learning_rate": 0.0004967074439169092,
      "loss": 5.9902,
      "mean_token_accuracy": 0.1502525493502617,
      "num_tokens": 13306094.0,
      "step": 7120
    },
    {
      "entropy": 6.051422643661499,
      "epoch": 0.6291946308724832,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004967020740346827,
      "loss": 5.9086,
      "mean_token_accuracy": 0.16666829138994216,
      "num_tokens": 13315181.0,
      "step": 7125
    },
    {
      "entropy": 5.989902448654175,
      "epoch": 0.6296361709643236,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004966966998094387,
      "loss": 5.8307,
      "mean_token_accuracy": 0.17334404289722444,
      "num_tokens": 13324108.0,
      "step": 7130
    },
    {
      "entropy": 6.002046394348144,
      "epoch": 0.630077711056164,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004966913212412825,
      "loss": 5.8927,
      "mean_token_accuracy": 0.17191413789987564,
      "num_tokens": 13333101.0,
      "step": 7135
    },
    {
      "entropy": 5.960778665542603,
      "epoch": 0.6305192511480042,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004966859383303194,
      "loss": 5.8222,
      "mean_token_accuracy": 0.16842588186264038,
      "num_tokens": 13341804.0,
      "step": 7140
    },
    {
      "entropy": 6.031595945358276,
      "epoch": 0.6309607912398446,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004966805510766549,
      "loss": 5.8624,
      "mean_token_accuracy": 0.16456822901964188,
      "num_tokens": 13350911.0,
      "step": 7145
    },
    {
      "entropy": 6.114427614212036,
      "epoch": 0.6314023313316849,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004966751594803944,
      "loss": 5.8576,
      "mean_token_accuracy": 0.16239054948091508,
      "num_tokens": 13360657.0,
      "step": 7150
    },
    {
      "entropy": 5.901859426498413,
      "epoch": 0.6318438714235253,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004966697635416437,
      "loss": 5.807,
      "mean_token_accuracy": 0.1658569395542145,
      "num_tokens": 13370462.0,
      "step": 7155
    },
    {
      "entropy": 6.063317108154297,
      "epoch": 0.6322854115153655,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004966643632605082,
      "loss": 5.922,
      "mean_token_accuracy": 0.15854725539684295,
      "num_tokens": 13379446.0,
      "step": 7160
    },
    {
      "entropy": 6.1069611549377445,
      "epoch": 0.6327269516072059,
      "grad_norm": 1.375,
      "learning_rate": 0.000496658958637094,
      "loss": 5.9265,
      "mean_token_accuracy": 0.16306125223636628,
      "num_tokens": 13389238.0,
      "step": 7165
    },
    {
      "entropy": 6.04985294342041,
      "epoch": 0.6331684916990463,
      "grad_norm": 1.390625,
      "learning_rate": 0.000496653549671507,
      "loss": 5.926,
      "mean_token_accuracy": 0.16360241323709487,
      "num_tokens": 13398613.0,
      "step": 7170
    },
    {
      "entropy": 6.058578968048096,
      "epoch": 0.6336100317908866,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004966481363638529,
      "loss": 5.9238,
      "mean_token_accuracy": 0.16620850712060928,
      "num_tokens": 13407600.0,
      "step": 7175
    },
    {
      "entropy": 5.953970432281494,
      "epoch": 0.634051571882727,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004966427187142378,
      "loss": 5.8066,
      "mean_token_accuracy": 0.17403708547353744,
      "num_tokens": 13415974.0,
      "step": 7180
    },
    {
      "entropy": 6.028492593765259,
      "epoch": 0.6344931119745673,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004966372967227681,
      "loss": 5.8296,
      "mean_token_accuracy": 0.1693430557847023,
      "num_tokens": 13425628.0,
      "step": 7185
    },
    {
      "entropy": 5.933087348937988,
      "epoch": 0.6349346520664076,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004966318703895496,
      "loss": 5.8403,
      "mean_token_accuracy": 0.17187251299619674,
      "num_tokens": 13435573.0,
      "step": 7190
    },
    {
      "entropy": 6.019541263580322,
      "epoch": 0.6353761921582479,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004966264397146888,
      "loss": 5.8092,
      "mean_token_accuracy": 0.16915347278118134,
      "num_tokens": 13443969.0,
      "step": 7195
    },
    {
      "entropy": 6.018271732330322,
      "epoch": 0.6358177322500883,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004966210046982921,
      "loss": 5.9069,
      "mean_token_accuracy": 0.1629061296582222,
      "num_tokens": 13452823.0,
      "step": 7200
    },
    {
      "entropy": 6.068755483627319,
      "epoch": 0.6362592723419287,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004966155653404661,
      "loss": 5.9196,
      "mean_token_accuracy": 0.16197239011526107,
      "num_tokens": 13461090.0,
      "step": 7205
    },
    {
      "entropy": 6.133015775680542,
      "epoch": 0.636700812433769,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004966101216413172,
      "loss": 5.9121,
      "mean_token_accuracy": 0.1573682501912117,
      "num_tokens": 13471425.0,
      "step": 7210
    },
    {
      "entropy": 6.068574714660644,
      "epoch": 0.6371423525256094,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004966046736009519,
      "loss": 5.9723,
      "mean_token_accuracy": 0.1632949158549309,
      "num_tokens": 13479755.0,
      "step": 7215
    },
    {
      "entropy": 5.951725959777832,
      "epoch": 0.6375838926174496,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004965992212194772,
      "loss": 5.8577,
      "mean_token_accuracy": 0.17126143872737884,
      "num_tokens": 13488854.0,
      "step": 7220
    },
    {
      "entropy": 6.062816572189331,
      "epoch": 0.63802543270929,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004965937644969998,
      "loss": 5.9191,
      "mean_token_accuracy": 0.1607258602976799,
      "num_tokens": 13497987.0,
      "step": 7225
    },
    {
      "entropy": 6.078987693786621,
      "epoch": 0.6384669728011303,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004965883034336266,
      "loss": 5.9421,
      "mean_token_accuracy": 0.16266340464353563,
      "num_tokens": 13506960.0,
      "step": 7230
    },
    {
      "entropy": 6.036382436752319,
      "epoch": 0.6389085128929707,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004965828380294645,
      "loss": 5.8821,
      "mean_token_accuracy": 0.1660477951169014,
      "num_tokens": 13515907.0,
      "step": 7235
    },
    {
      "entropy": 6.054521989822388,
      "epoch": 0.6393500529848111,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004965773682846207,
      "loss": 5.9641,
      "mean_token_accuracy": 0.16726862788200378,
      "num_tokens": 13524876.0,
      "step": 7240
    },
    {
      "entropy": 6.0579798221588135,
      "epoch": 0.6397915930766513,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004965718941992021,
      "loss": 5.8624,
      "mean_token_accuracy": 0.1678209349513054,
      "num_tokens": 13532891.0,
      "step": 7245
    },
    {
      "entropy": 6.108337450027466,
      "epoch": 0.6402331331684917,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004965664157733162,
      "loss": 5.9374,
      "mean_token_accuracy": 0.16674156337976456,
      "num_tokens": 13542409.0,
      "step": 7250
    },
    {
      "entropy": 5.97765622138977,
      "epoch": 0.640674673260332,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004965609330070703,
      "loss": 5.9069,
      "mean_token_accuracy": 0.16326847672462463,
      "num_tokens": 13551947.0,
      "step": 7255
    },
    {
      "entropy": 6.088966798782349,
      "epoch": 0.6411162133521724,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004965554459005717,
      "loss": 5.9212,
      "mean_token_accuracy": 0.16018670350313186,
      "num_tokens": 13560812.0,
      "step": 7260
    },
    {
      "entropy": 6.0528840065002445,
      "epoch": 0.6415577534440127,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004965499544539278,
      "loss": 5.9344,
      "mean_token_accuracy": 0.162466648966074,
      "num_tokens": 13570197.0,
      "step": 7265
    },
    {
      "entropy": 5.996675062179565,
      "epoch": 0.641999293535853,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004965444586672463,
      "loss": 5.8443,
      "mean_token_accuracy": 0.16122349947690964,
      "num_tokens": 13579614.0,
      "step": 7270
    },
    {
      "entropy": 6.066436815261841,
      "epoch": 0.6424408336276934,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000496538958540635,
      "loss": 5.8921,
      "mean_token_accuracy": 0.16424771398305893,
      "num_tokens": 13589511.0,
      "step": 7275
    },
    {
      "entropy": 6.036352348327637,
      "epoch": 0.6428823737195337,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004965334540742015,
      "loss": 5.8698,
      "mean_token_accuracy": 0.1678945705294609,
      "num_tokens": 13597718.0,
      "step": 7280
    },
    {
      "entropy": 6.0071446895599365,
      "epoch": 0.6433239138113741,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004965279452680535,
      "loss": 5.962,
      "mean_token_accuracy": 0.15827914625406264,
      "num_tokens": 13607599.0,
      "step": 7285
    },
    {
      "entropy": 6.15809268951416,
      "epoch": 0.6437654539032144,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004965224321222991,
      "loss": 6.0109,
      "mean_token_accuracy": 0.1586283728480339,
      "num_tokens": 13616513.0,
      "step": 7290
    },
    {
      "entropy": 6.08264536857605,
      "epoch": 0.6442069939950548,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004965169146370462,
      "loss": 5.8812,
      "mean_token_accuracy": 0.16384359896183015,
      "num_tokens": 13625505.0,
      "step": 7295
    },
    {
      "entropy": 6.07037844657898,
      "epoch": 0.644648534086895,
      "grad_norm": 1.625,
      "learning_rate": 0.0004965113928124031,
      "loss": 5.9287,
      "mean_token_accuracy": 0.16326516717672349,
      "num_tokens": 13635340.0,
      "step": 7300
    },
    {
      "entropy": 5.908147239685059,
      "epoch": 0.6450900741787354,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004965058666484775,
      "loss": 5.8545,
      "mean_token_accuracy": 0.16143952161073685,
      "num_tokens": 13644886.0,
      "step": 7305
    },
    {
      "entropy": 5.967869806289673,
      "epoch": 0.6455316142705758,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004965003361453782,
      "loss": 5.8811,
      "mean_token_accuracy": 0.16908676475286483,
      "num_tokens": 13654093.0,
      "step": 7310
    },
    {
      "entropy": 6.116286134719848,
      "epoch": 0.6459731543624161,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004964948013032132,
      "loss": 5.9554,
      "mean_token_accuracy": 0.1531238943338394,
      "num_tokens": 13663095.0,
      "step": 7315
    },
    {
      "entropy": 6.0913684368133545,
      "epoch": 0.6464146944542565,
      "grad_norm": 1.234375,
      "learning_rate": 0.000496489262122091,
      "loss": 5.9779,
      "mean_token_accuracy": 0.16273966878652574,
      "num_tokens": 13673624.0,
      "step": 7320
    },
    {
      "entropy": 6.033328199386597,
      "epoch": 0.6468562345460968,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004964837186021201,
      "loss": 5.8581,
      "mean_token_accuracy": 0.16317969858646392,
      "num_tokens": 13683145.0,
      "step": 7325
    },
    {
      "entropy": 5.987786388397216,
      "epoch": 0.6472977746379371,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004964781707434093,
      "loss": 5.8507,
      "mean_token_accuracy": 0.15873554050922395,
      "num_tokens": 13691963.0,
      "step": 7330
    },
    {
      "entropy": 5.932110071182251,
      "epoch": 0.6477393147297774,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004964726185460669,
      "loss": 5.7666,
      "mean_token_accuracy": 0.1703936919569969,
      "num_tokens": 13700926.0,
      "step": 7335
    },
    {
      "entropy": 6.086005926132202,
      "epoch": 0.6481808548216178,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004964670620102019,
      "loss": 5.9927,
      "mean_token_accuracy": 0.15364788472652435,
      "num_tokens": 13709824.0,
      "step": 7340
    },
    {
      "entropy": 6.05243673324585,
      "epoch": 0.6486223949134582,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004964615011359231,
      "loss": 5.8899,
      "mean_token_accuracy": 0.16376349031925203,
      "num_tokens": 13718789.0,
      "step": 7345
    },
    {
      "entropy": 6.070804214477539,
      "epoch": 0.6490639350052985,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004964559359233395,
      "loss": 5.9055,
      "mean_token_accuracy": 0.16012013852596282,
      "num_tokens": 13728180.0,
      "step": 7350
    },
    {
      "entropy": 6.069525623321534,
      "epoch": 0.6495054750971389,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004964503663725601,
      "loss": 5.9942,
      "mean_token_accuracy": 0.1554926410317421,
      "num_tokens": 13737819.0,
      "step": 7355
    },
    {
      "entropy": 6.066304063796997,
      "epoch": 0.6499470151889791,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004964447924836938,
      "loss": 5.9771,
      "mean_token_accuracy": 0.1574745424091816,
      "num_tokens": 13747519.0,
      "step": 7360
    },
    {
      "entropy": 6.091936492919922,
      "epoch": 0.6503885552808195,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004964392142568501,
      "loss": 5.9718,
      "mean_token_accuracy": 0.15019024461507796,
      "num_tokens": 13757821.0,
      "step": 7365
    },
    {
      "entropy": 5.972079277038574,
      "epoch": 0.6508300953726598,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004964336316921381,
      "loss": 5.7573,
      "mean_token_accuracy": 0.16695429980754853,
      "num_tokens": 13766560.0,
      "step": 7370
    },
    {
      "entropy": 6.025388288497925,
      "epoch": 0.6512716354645002,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004964280447896672,
      "loss": 5.8072,
      "mean_token_accuracy": 0.16919275969266892,
      "num_tokens": 13775560.0,
      "step": 7375
    },
    {
      "entropy": 5.9751911640167235,
      "epoch": 0.6517131755563406,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004964224535495468,
      "loss": 5.8937,
      "mean_token_accuracy": 0.16387373208999634,
      "num_tokens": 13784391.0,
      "step": 7380
    },
    {
      "entropy": 5.957211112976074,
      "epoch": 0.6521547156481808,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004964168579718864,
      "loss": 5.869,
      "mean_token_accuracy": 0.1669353127479553,
      "num_tokens": 13794008.0,
      "step": 7385
    },
    {
      "entropy": 6.05711817741394,
      "epoch": 0.6525962557400212,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004964112580567959,
      "loss": 5.9144,
      "mean_token_accuracy": 0.159195776283741,
      "num_tokens": 13804321.0,
      "step": 7390
    },
    {
      "entropy": 6.070773077011109,
      "epoch": 0.6530377958318615,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004964056538043845,
      "loss": 5.9157,
      "mean_token_accuracy": 0.15635670274496077,
      "num_tokens": 13813492.0,
      "step": 7395
    },
    {
      "entropy": 6.008130598068237,
      "epoch": 0.6534793359237019,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004964000452147624,
      "loss": 5.9007,
      "mean_token_accuracy": 0.169648377597332,
      "num_tokens": 13822471.0,
      "step": 7400
    },
    {
      "entropy": 5.992099714279175,
      "epoch": 0.6539208760155422,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004963944322880394,
      "loss": 5.96,
      "mean_token_accuracy": 0.15306852012872696,
      "num_tokens": 13832789.0,
      "step": 7405
    },
    {
      "entropy": 6.087279605865478,
      "epoch": 0.6543624161073825,
      "grad_norm": 1.5,
      "learning_rate": 0.0004963888150243253,
      "loss": 5.9298,
      "mean_token_accuracy": 0.1591775357723236,
      "num_tokens": 13842230.0,
      "step": 7410
    },
    {
      "entropy": 6.0844536304473875,
      "epoch": 0.6548039561992229,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004963831934237302,
      "loss": 5.8461,
      "mean_token_accuracy": 0.16865865141153336,
      "num_tokens": 13851542.0,
      "step": 7415
    },
    {
      "entropy": 6.002407360076904,
      "epoch": 0.6552454962910632,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004963775674863643,
      "loss": 5.8748,
      "mean_token_accuracy": 0.16350056529045104,
      "num_tokens": 13860753.0,
      "step": 7420
    },
    {
      "entropy": 6.004277229309082,
      "epoch": 0.6556870363829036,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004963719372123378,
      "loss": 5.862,
      "mean_token_accuracy": 0.1674501046538353,
      "num_tokens": 13870133.0,
      "step": 7425
    },
    {
      "entropy": 6.038403463363648,
      "epoch": 0.6561285764747439,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004963663026017609,
      "loss": 5.941,
      "mean_token_accuracy": 0.16310749650001527,
      "num_tokens": 13879086.0,
      "step": 7430
    },
    {
      "entropy": 6.047044372558593,
      "epoch": 0.6565701165665843,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004963606636547441,
      "loss": 5.8789,
      "mean_token_accuracy": 0.16871441453695296,
      "num_tokens": 13888706.0,
      "step": 7435
    },
    {
      "entropy": 6.038377714157105,
      "epoch": 0.6570116566584245,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004963550203713976,
      "loss": 5.9301,
      "mean_token_accuracy": 0.15731319785118103,
      "num_tokens": 13897455.0,
      "step": 7440
    },
    {
      "entropy": 6.029115009307861,
      "epoch": 0.6574531967502649,
      "grad_norm": 1.5,
      "learning_rate": 0.0004963493727518323,
      "loss": 5.9129,
      "mean_token_accuracy": 0.16222378611564636,
      "num_tokens": 13907697.0,
      "step": 7445
    },
    {
      "entropy": 5.996784257888794,
      "epoch": 0.6578947368421053,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004963437207961585,
      "loss": 5.9135,
      "mean_token_accuracy": 0.16317444592714309,
      "num_tokens": 13917124.0,
      "step": 7450
    },
    {
      "entropy": 6.048440885543823,
      "epoch": 0.6583362769339456,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004963380645044874,
      "loss": 5.8641,
      "mean_token_accuracy": 0.16341730505228041,
      "num_tokens": 13926434.0,
      "step": 7455
    },
    {
      "entropy": 5.950565671920776,
      "epoch": 0.658777817025786,
      "grad_norm": 1.75,
      "learning_rate": 0.0004963324038769292,
      "loss": 5.8284,
      "mean_token_accuracy": 0.16859227418899536,
      "num_tokens": 13934883.0,
      "step": 7460
    },
    {
      "entropy": 6.026736736297607,
      "epoch": 0.6592193571176262,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004963267389135953,
      "loss": 5.9366,
      "mean_token_accuracy": 0.16433704197406768,
      "num_tokens": 13944514.0,
      "step": 7465
    },
    {
      "entropy": 6.020615196228027,
      "epoch": 0.6596608972094666,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004963210696145964,
      "loss": 5.8521,
      "mean_token_accuracy": 0.16204071938991546,
      "num_tokens": 13953434.0,
      "step": 7470
    },
    {
      "entropy": 6.0246869087219235,
      "epoch": 0.6601024373013069,
      "grad_norm": 1.375,
      "learning_rate": 0.0004963153959800436,
      "loss": 5.8636,
      "mean_token_accuracy": 0.16973265260457993,
      "num_tokens": 13962493.0,
      "step": 7475
    },
    {
      "entropy": 6.054371166229248,
      "epoch": 0.6605439773931473,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000496309718010048,
      "loss": 5.9044,
      "mean_token_accuracy": 0.1610116109251976,
      "num_tokens": 13971193.0,
      "step": 7480
    },
    {
      "entropy": 5.995823526382447,
      "epoch": 0.6609855174849877,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004963040357047209,
      "loss": 5.9127,
      "mean_token_accuracy": 0.16409173905849456,
      "num_tokens": 13981829.0,
      "step": 7485
    },
    {
      "entropy": 5.979941177368164,
      "epoch": 0.661427057576828,
      "grad_norm": 1.5,
      "learning_rate": 0.0004962983490641737,
      "loss": 5.8615,
      "mean_token_accuracy": 0.1629578024148941,
      "num_tokens": 13990262.0,
      "step": 7490
    },
    {
      "entropy": 5.9936957359313965,
      "epoch": 0.6618685976686683,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004962926580885177,
      "loss": 5.9007,
      "mean_token_accuracy": 0.16571170836687088,
      "num_tokens": 14000100.0,
      "step": 7495
    },
    {
      "entropy": 6.05236439704895,
      "epoch": 0.6623101377605086,
      "grad_norm": 1.5,
      "learning_rate": 0.0004962869627778642,
      "loss": 5.9336,
      "mean_token_accuracy": 0.16470786482095717,
      "num_tokens": 14009330.0,
      "step": 7500
    },
    {
      "entropy": 5.993536043167114,
      "epoch": 0.662751677852349,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004962812631323252,
      "loss": 5.8976,
      "mean_token_accuracy": 0.15885231047868728,
      "num_tokens": 14018109.0,
      "step": 7505
    },
    {
      "entropy": 6.01054573059082,
      "epoch": 0.6631932179441893,
      "grad_norm": 1.359375,
      "learning_rate": 0.000496275559152012,
      "loss": 5.7961,
      "mean_token_accuracy": 0.1680995926260948,
      "num_tokens": 14027318.0,
      "step": 7510
    },
    {
      "entropy": 6.015023469924927,
      "epoch": 0.6636347580360297,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004962698508370365,
      "loss": 5.9339,
      "mean_token_accuracy": 0.15832750350236893,
      "num_tokens": 14037061.0,
      "step": 7515
    },
    {
      "entropy": 5.948602676391602,
      "epoch": 0.6640762981278701,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004962641381875103,
      "loss": 5.8786,
      "mean_token_accuracy": 0.15682309120893478,
      "num_tokens": 14046008.0,
      "step": 7520
    },
    {
      "entropy": 6.026252365112304,
      "epoch": 0.6645178382197103,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004962584212035456,
      "loss": 5.8877,
      "mean_token_accuracy": 0.16163944005966185,
      "num_tokens": 14054535.0,
      "step": 7525
    },
    {
      "entropy": 6.092637443542481,
      "epoch": 0.6649593783115507,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004962526998852543,
      "loss": 5.8989,
      "mean_token_accuracy": 0.16449156999588013,
      "num_tokens": 14063680.0,
      "step": 7530
    },
    {
      "entropy": 6.065542888641358,
      "epoch": 0.665400918403391,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004962469742327483,
      "loss": 5.9386,
      "mean_token_accuracy": 0.16228194236755372,
      "num_tokens": 14072206.0,
      "step": 7535
    },
    {
      "entropy": 6.043942975997925,
      "epoch": 0.6658424584952314,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00049624124424614,
      "loss": 5.9271,
      "mean_token_accuracy": 0.16253331303596497,
      "num_tokens": 14081943.0,
      "step": 7540
    },
    {
      "entropy": 6.0551399230957035,
      "epoch": 0.6662839985870717,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004962355099255415,
      "loss": 5.8902,
      "mean_token_accuracy": 0.16782213896512985,
      "num_tokens": 14091695.0,
      "step": 7545
    },
    {
      "entropy": 5.9793314933776855,
      "epoch": 0.666725538678912,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004962297712710653,
      "loss": 5.815,
      "mean_token_accuracy": 0.17347454726696016,
      "num_tokens": 14100811.0,
      "step": 7550
    },
    {
      "entropy": 6.09973931312561,
      "epoch": 0.6671670787707524,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004962240282828236,
      "loss": 5.9826,
      "mean_token_accuracy": 0.14768042415380478,
      "num_tokens": 14109818.0,
      "step": 7555
    },
    {
      "entropy": 6.081058645248413,
      "epoch": 0.6676086188625927,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000496218280960929,
      "loss": 5.9184,
      "mean_token_accuracy": 0.15994459986686707,
      "num_tokens": 14119811.0,
      "step": 7560
    },
    {
      "entropy": 6.110054540634155,
      "epoch": 0.6680501589544331,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004962125293054941,
      "loss": 5.8684,
      "mean_token_accuracy": 0.16850226670503615,
      "num_tokens": 14128963.0,
      "step": 7565
    },
    {
      "entropy": 5.9815596580505375,
      "epoch": 0.6684916990462734,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004962067733166317,
      "loss": 5.9451,
      "mean_token_accuracy": 0.15114885419607163,
      "num_tokens": 14138532.0,
      "step": 7570
    },
    {
      "entropy": 5.973617649078369,
      "epoch": 0.6689332391381138,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004962010129944542,
      "loss": 5.8884,
      "mean_token_accuracy": 0.1637090712785721,
      "num_tokens": 14148107.0,
      "step": 7575
    },
    {
      "entropy": 6.079353713989258,
      "epoch": 0.669374779229954,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004961952483390747,
      "loss": 5.9456,
      "mean_token_accuracy": 0.15459064096212388,
      "num_tokens": 14158823.0,
      "step": 7580
    },
    {
      "entropy": 6.022123670578003,
      "epoch": 0.6698163193217944,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004961894793506062,
      "loss": 5.8854,
      "mean_token_accuracy": 0.1703451931476593,
      "num_tokens": 14168147.0,
      "step": 7585
    },
    {
      "entropy": 5.947274494171142,
      "epoch": 0.6702578594136348,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004961837060291615,
      "loss": 5.9241,
      "mean_token_accuracy": 0.1599956676363945,
      "num_tokens": 14177123.0,
      "step": 7590
    },
    {
      "entropy": 6.023081350326538,
      "epoch": 0.6706993995054751,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004961779283748538,
      "loss": 5.908,
      "mean_token_accuracy": 0.1616277053952217,
      "num_tokens": 14186614.0,
      "step": 7595
    },
    {
      "entropy": 6.018943977355957,
      "epoch": 0.6711409395973155,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004961721463877963,
      "loss": 5.9096,
      "mean_token_accuracy": 0.16158800423145295,
      "num_tokens": 14195310.0,
      "step": 7600
    },
    {
      "entropy": 6.002673530578614,
      "epoch": 0.6715824796891557,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004961663600681023,
      "loss": 5.8765,
      "mean_token_accuracy": 0.16197859048843383,
      "num_tokens": 14204321.0,
      "step": 7605
    },
    {
      "entropy": 5.991839742660522,
      "epoch": 0.6720240197809961,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000496160569415885,
      "loss": 5.9103,
      "mean_token_accuracy": 0.16719352900981904,
      "num_tokens": 14213615.0,
      "step": 7610
    },
    {
      "entropy": 6.088936424255371,
      "epoch": 0.6724655598728364,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004961547744312579,
      "loss": 5.785,
      "mean_token_accuracy": 0.17092676907777787,
      "num_tokens": 14222715.0,
      "step": 7615
    },
    {
      "entropy": 6.0617317199707035,
      "epoch": 0.6729070999646768,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004961489751143346,
      "loss": 5.9181,
      "mean_token_accuracy": 0.16617656946182252,
      "num_tokens": 14232117.0,
      "step": 7620
    },
    {
      "entropy": 5.968132638931275,
      "epoch": 0.6733486400565172,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004961431714652287,
      "loss": 5.7999,
      "mean_token_accuracy": 0.1727928102016449,
      "num_tokens": 14240606.0,
      "step": 7625
    },
    {
      "entropy": 6.0396177768707275,
      "epoch": 0.6737901801483575,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004961373634840538,
      "loss": 5.9223,
      "mean_token_accuracy": 0.15404981672763823,
      "num_tokens": 14249844.0,
      "step": 7630
    },
    {
      "entropy": 6.007670164108276,
      "epoch": 0.6742317202401978,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004961315511709237,
      "loss": 5.9178,
      "mean_token_accuracy": 0.1644563376903534,
      "num_tokens": 14259231.0,
      "step": 7635
    },
    {
      "entropy": 6.081117105484009,
      "epoch": 0.6746732603320381,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004961257345259524,
      "loss": 5.9229,
      "mean_token_accuracy": 0.16527654528617858,
      "num_tokens": 14268508.0,
      "step": 7640
    },
    {
      "entropy": 6.0169459819793705,
      "epoch": 0.6751148004238785,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004961199135492535,
      "loss": 5.839,
      "mean_token_accuracy": 0.17000525146722795,
      "num_tokens": 14278057.0,
      "step": 7645
    },
    {
      "entropy": 5.990185642242432,
      "epoch": 0.6755563405157188,
      "grad_norm": 1.625,
      "learning_rate": 0.0004961140882409414,
      "loss": 5.9137,
      "mean_token_accuracy": 0.16065546423196791,
      "num_tokens": 14287225.0,
      "step": 7650
    },
    {
      "entropy": 6.092997264862061,
      "epoch": 0.6759978806075592,
      "grad_norm": 1.5,
      "learning_rate": 0.00049610825860113,
      "loss": 5.8698,
      "mean_token_accuracy": 0.16446682810783386,
      "num_tokens": 14296529.0,
      "step": 7655
    },
    {
      "entropy": 5.976733684539795,
      "epoch": 0.6764394206993996,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004961024246299336,
      "loss": 5.8333,
      "mean_token_accuracy": 0.16461792439222336,
      "num_tokens": 14305675.0,
      "step": 7660
    },
    {
      "entropy": 5.9833447456359865,
      "epoch": 0.6768809607912398,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004960965863274665,
      "loss": 5.7966,
      "mean_token_accuracy": 0.16425360292196273,
      "num_tokens": 14313632.0,
      "step": 7665
    },
    {
      "entropy": 5.937602996826172,
      "epoch": 0.6773225008830802,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000496090743693843,
      "loss": 5.9136,
      "mean_token_accuracy": 0.16260362714529036,
      "num_tokens": 14323353.0,
      "step": 7670
    },
    {
      "entropy": 6.1345940113067625,
      "epoch": 0.6777640409749205,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004960848967291775,
      "loss": 5.9051,
      "mean_token_accuracy": 0.1604089170694351,
      "num_tokens": 14331953.0,
      "step": 7675
    },
    {
      "entropy": 6.091330718994141,
      "epoch": 0.6782055810667609,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004960790454335846,
      "loss": 5.9563,
      "mean_token_accuracy": 0.15599895119667054,
      "num_tokens": 14342456.0,
      "step": 7680
    },
    {
      "entropy": 5.974487972259522,
      "epoch": 0.6786471211586012,
      "grad_norm": 1.421875,
      "learning_rate": 0.000496073189807179,
      "loss": 5.7375,
      "mean_token_accuracy": 0.17251136004924775,
      "num_tokens": 14351013.0,
      "step": 7685
    },
    {
      "entropy": 6.053620958328247,
      "epoch": 0.6790886612504415,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004960673298500753,
      "loss": 6.0012,
      "mean_token_accuracy": 0.1551806792616844,
      "num_tokens": 14360887.0,
      "step": 7690
    },
    {
      "entropy": 6.065280342102051,
      "epoch": 0.6795302013422819,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004960614655623883,
      "loss": 5.9223,
      "mean_token_accuracy": 0.1608433708548546,
      "num_tokens": 14370510.0,
      "step": 7695
    },
    {
      "entropy": 5.98971471786499,
      "epoch": 0.6799717414341222,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000496055596944233,
      "loss": 5.9103,
      "mean_token_accuracy": 0.16423211097717286,
      "num_tokens": 14379315.0,
      "step": 7700
    },
    {
      "entropy": 6.040924167633056,
      "epoch": 0.6804132815259626,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004960497239957243,
      "loss": 5.8872,
      "mean_token_accuracy": 0.159690023958683,
      "num_tokens": 14388537.0,
      "step": 7705
    },
    {
      "entropy": 6.125636291503906,
      "epoch": 0.6808548216178029,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004960438467169772,
      "loss": 5.929,
      "mean_token_accuracy": 0.16286856234073638,
      "num_tokens": 14398347.0,
      "step": 7710
    },
    {
      "entropy": 6.0843593120574955,
      "epoch": 0.6812963617096433,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004960379651081068,
      "loss": 5.855,
      "mean_token_accuracy": 0.16956682354211808,
      "num_tokens": 14407304.0,
      "step": 7715
    },
    {
      "entropy": 5.984658145904541,
      "epoch": 0.6817379018014835,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004960320791692283,
      "loss": 5.875,
      "mean_token_accuracy": 0.162070694565773,
      "num_tokens": 14416641.0,
      "step": 7720
    },
    {
      "entropy": 6.045535898208618,
      "epoch": 0.6821794418933239,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004960261889004572,
      "loss": 5.9795,
      "mean_token_accuracy": 0.15513098388910293,
      "num_tokens": 14426095.0,
      "step": 7725
    },
    {
      "entropy": 6.097327518463135,
      "epoch": 0.6826209819851643,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004960202943019088,
      "loss": 5.8766,
      "mean_token_accuracy": 0.16527304351329802,
      "num_tokens": 14434563.0,
      "step": 7730
    },
    {
      "entropy": 5.9518779754638675,
      "epoch": 0.6830625220770046,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004960143953736984,
      "loss": 5.7974,
      "mean_token_accuracy": 0.17335865646600723,
      "num_tokens": 14444319.0,
      "step": 7735
    },
    {
      "entropy": 5.9988618850708,
      "epoch": 0.683504062168845,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004960084921159419,
      "loss": 5.8591,
      "mean_token_accuracy": 0.167277492582798,
      "num_tokens": 14453381.0,
      "step": 7740
    },
    {
      "entropy": 5.874141550064087,
      "epoch": 0.6839456022606852,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004960025845287546,
      "loss": 5.8492,
      "mean_token_accuracy": 0.1701892837882042,
      "num_tokens": 14462660.0,
      "step": 7745
    },
    {
      "entropy": 5.956099653244019,
      "epoch": 0.6843871423525256,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004959966726122523,
      "loss": 5.7854,
      "mean_token_accuracy": 0.17432601302862166,
      "num_tokens": 14471820.0,
      "step": 7750
    },
    {
      "entropy": 6.101538801193238,
      "epoch": 0.6848286824443659,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000495990756366551,
      "loss": 5.9344,
      "mean_token_accuracy": 0.1623198062181473,
      "num_tokens": 14481663.0,
      "step": 7755
    },
    {
      "entropy": 6.123762798309326,
      "epoch": 0.6852702225362063,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004959848357917664,
      "loss": 5.9611,
      "mean_token_accuracy": 0.1574322536587715,
      "num_tokens": 14491515.0,
      "step": 7760
    },
    {
      "entropy": 6.004442834854126,
      "epoch": 0.6857117626280467,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004959789108880145,
      "loss": 5.7873,
      "mean_token_accuracy": 0.164418064057827,
      "num_tokens": 14500852.0,
      "step": 7765
    },
    {
      "entropy": 6.017218780517578,
      "epoch": 0.686153302719887,
      "grad_norm": 1.5,
      "learning_rate": 0.0004959729816554115,
      "loss": 5.8648,
      "mean_token_accuracy": 0.16977741122245787,
      "num_tokens": 14509771.0,
      "step": 7770
    },
    {
      "entropy": 5.983751487731934,
      "epoch": 0.6865948428117273,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004959670480940734,
      "loss": 5.8261,
      "mean_token_accuracy": 0.16761362850666045,
      "num_tokens": 14519377.0,
      "step": 7775
    },
    {
      "entropy": 5.9449629306793215,
      "epoch": 0.6870363829035676,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004959611102041165,
      "loss": 5.8723,
      "mean_token_accuracy": 0.16630145758390427,
      "num_tokens": 14528998.0,
      "step": 7780
    },
    {
      "entropy": 6.026974439620972,
      "epoch": 0.687477922995408,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004959551679856571,
      "loss": 5.9144,
      "mean_token_accuracy": 0.1634315237402916,
      "num_tokens": 14538153.0,
      "step": 7785
    },
    {
      "entropy": 6.042395067214966,
      "epoch": 0.6879194630872483,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004959492214388117,
      "loss": 5.8503,
      "mean_token_accuracy": 0.16629589945077897,
      "num_tokens": 14547631.0,
      "step": 7790
    },
    {
      "entropy": 5.997350263595581,
      "epoch": 0.6883610031790887,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004959432705636968,
      "loss": 5.7851,
      "mean_token_accuracy": 0.171341934800148,
      "num_tokens": 14557685.0,
      "step": 7795
    },
    {
      "entropy": 6.041592073440552,
      "epoch": 0.688802543270929,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004959373153604287,
      "loss": 5.9302,
      "mean_token_accuracy": 0.16370871067047119,
      "num_tokens": 14567360.0,
      "step": 7800
    },
    {
      "entropy": 6.000632572174072,
      "epoch": 0.6892440833627693,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004959313558291243,
      "loss": 5.8529,
      "mean_token_accuracy": 0.1705251455307007,
      "num_tokens": 14576916.0,
      "step": 7805
    },
    {
      "entropy": 6.014446067810058,
      "epoch": 0.6896856234546097,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004959253919699004,
      "loss": 5.7764,
      "mean_token_accuracy": 0.17379749268293382,
      "num_tokens": 14586716.0,
      "step": 7810
    },
    {
      "entropy": 5.978111124038696,
      "epoch": 0.69012716354645,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004959194237828735,
      "loss": 5.8253,
      "mean_token_accuracy": 0.1650472640991211,
      "num_tokens": 14595690.0,
      "step": 7815
    },
    {
      "entropy": 5.980210590362549,
      "epoch": 0.6905687036382904,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004959134512681609,
      "loss": 5.8995,
      "mean_token_accuracy": 0.1592855393886566,
      "num_tokens": 14605481.0,
      "step": 7820
    },
    {
      "entropy": 6.018238973617554,
      "epoch": 0.6910102437301306,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004959074744258794,
      "loss": 5.9594,
      "mean_token_accuracy": 0.15670410841703414,
      "num_tokens": 14615242.0,
      "step": 7825
    },
    {
      "entropy": 6.039451837539673,
      "epoch": 0.691451783821971,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004959014932561461,
      "loss": 5.7923,
      "mean_token_accuracy": 0.16917330920696258,
      "num_tokens": 14624343.0,
      "step": 7830
    },
    {
      "entropy": 5.9386570930480955,
      "epoch": 0.6918933239138114,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004958955077590782,
      "loss": 5.7665,
      "mean_token_accuracy": 0.17301661819219588,
      "num_tokens": 14633014.0,
      "step": 7835
    },
    {
      "entropy": 5.955418539047241,
      "epoch": 0.6923348640056517,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004958895179347929,
      "loss": 5.8864,
      "mean_token_accuracy": 0.1625771328806877,
      "num_tokens": 14642796.0,
      "step": 7840
    },
    {
      "entropy": 6.032704639434814,
      "epoch": 0.6927764040974921,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004958835237834075,
      "loss": 5.9372,
      "mean_token_accuracy": 0.15462349355220795,
      "num_tokens": 14652388.0,
      "step": 7845
    },
    {
      "entropy": 6.132510662078857,
      "epoch": 0.6932179441893324,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004958775253050396,
      "loss": 5.8631,
      "mean_token_accuracy": 0.16805050820112227,
      "num_tokens": 14661219.0,
      "step": 7850
    },
    {
      "entropy": 6.07625560760498,
      "epoch": 0.6936594842811727,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004958715224998066,
      "loss": 5.885,
      "mean_token_accuracy": 0.16949716359376907,
      "num_tokens": 14670878.0,
      "step": 7855
    },
    {
      "entropy": 5.988736534118653,
      "epoch": 0.694101024373013,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004958655153678261,
      "loss": 5.9343,
      "mean_token_accuracy": 0.1602768912911415,
      "num_tokens": 14680893.0,
      "step": 7860
    },
    {
      "entropy": 6.038381433486938,
      "epoch": 0.6945425644648534,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004958595039092156,
      "loss": 5.8472,
      "mean_token_accuracy": 0.16352402567863464,
      "num_tokens": 14689249.0,
      "step": 7865
    },
    {
      "entropy": 6.069445562362671,
      "epoch": 0.6949841045566938,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004958534881240932,
      "loss": 5.8468,
      "mean_token_accuracy": 0.16557328253984452,
      "num_tokens": 14698724.0,
      "step": 7870
    },
    {
      "entropy": 6.03548789024353,
      "epoch": 0.6954256446485341,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004958474680125766,
      "loss": 5.7999,
      "mean_token_accuracy": 0.1614638313651085,
      "num_tokens": 14707756.0,
      "step": 7875
    },
    {
      "entropy": 5.9449488639831545,
      "epoch": 0.6958671847403745,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004958414435747837,
      "loss": 5.8556,
      "mean_token_accuracy": 0.16810656636953353,
      "num_tokens": 14717416.0,
      "step": 7880
    },
    {
      "entropy": 5.9693653106689455,
      "epoch": 0.6963087248322147,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004958354148108324,
      "loss": 5.8907,
      "mean_token_accuracy": 0.16461798548698425,
      "num_tokens": 14727321.0,
      "step": 7885
    },
    {
      "entropy": 5.969771814346314,
      "epoch": 0.6967502649240551,
      "grad_norm": 2.359375,
      "learning_rate": 0.000495829381720841,
      "loss": 5.7638,
      "mean_token_accuracy": 0.17590563893318176,
      "num_tokens": 14736298.0,
      "step": 7890
    },
    {
      "entropy": 6.0774918556213375,
      "epoch": 0.6971918050158954,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004958233443049275,
      "loss": 5.8168,
      "mean_token_accuracy": 0.1612874150276184,
      "num_tokens": 14745893.0,
      "step": 7895
    },
    {
      "entropy": 5.994456005096436,
      "epoch": 0.6976333451077358,
      "grad_norm": 1.5,
      "learning_rate": 0.0004958173025632103,
      "loss": 5.8764,
      "mean_token_accuracy": 0.1590081751346588,
      "num_tokens": 14755277.0,
      "step": 7900
    },
    {
      "entropy": 6.05733437538147,
      "epoch": 0.6980748851995762,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004958112564958079,
      "loss": 5.99,
      "mean_token_accuracy": 0.15567466169595717,
      "num_tokens": 14764369.0,
      "step": 7905
    },
    {
      "entropy": 6.000176811218262,
      "epoch": 0.6985164252914164,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004958052061028385,
      "loss": 5.8249,
      "mean_token_accuracy": 0.17366339713335038,
      "num_tokens": 14772970.0,
      "step": 7910
    },
    {
      "entropy": 6.0693926334381105,
      "epoch": 0.6989579653832568,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004957991513844205,
      "loss": 5.9008,
      "mean_token_accuracy": 0.1680343359708786,
      "num_tokens": 14782862.0,
      "step": 7915
    },
    {
      "entropy": 6.051274967193604,
      "epoch": 0.6993995054750971,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004957930923406729,
      "loss": 5.8969,
      "mean_token_accuracy": 0.16636960506439208,
      "num_tokens": 14792042.0,
      "step": 7920
    },
    {
      "entropy": 5.996286153793335,
      "epoch": 0.6998410455669375,
      "grad_norm": 1.5,
      "learning_rate": 0.0004957870289717142,
      "loss": 5.8408,
      "mean_token_accuracy": 0.16046467274427414,
      "num_tokens": 14801032.0,
      "step": 7925
    },
    {
      "entropy": 6.02137451171875,
      "epoch": 0.7002825856587778,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004957809612776631,
      "loss": 5.8596,
      "mean_token_accuracy": 0.16641456931829451,
      "num_tokens": 14810594.0,
      "step": 7930
    },
    {
      "entropy": 6.00620470046997,
      "epoch": 0.7007241257506182,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004957748892586384,
      "loss": 5.9501,
      "mean_token_accuracy": 0.165498748421669,
      "num_tokens": 14820570.0,
      "step": 7935
    },
    {
      "entropy": 5.895208406448364,
      "epoch": 0.7011656658424585,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004957688129147594,
      "loss": 5.7706,
      "mean_token_accuracy": 0.1790166676044464,
      "num_tokens": 14830078.0,
      "step": 7940
    },
    {
      "entropy": 6.026910591125488,
      "epoch": 0.7016072059342988,
      "grad_norm": 1.375,
      "learning_rate": 0.0004957627322461448,
      "loss": 5.8544,
      "mean_token_accuracy": 0.16862896382808684,
      "num_tokens": 14839974.0,
      "step": 7945
    },
    {
      "entropy": 6.061175012588501,
      "epoch": 0.7020487460261392,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004957566472529139,
      "loss": 5.954,
      "mean_token_accuracy": 0.15304158478975297,
      "num_tokens": 14849551.0,
      "step": 7950
    },
    {
      "entropy": 6.011351013183594,
      "epoch": 0.7024902861179795,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004957505579351858,
      "loss": 5.862,
      "mean_token_accuracy": 0.16445884555578233,
      "num_tokens": 14859011.0,
      "step": 7955
    },
    {
      "entropy": 5.97165355682373,
      "epoch": 0.7029318262098199,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004957444642930798,
      "loss": 5.8183,
      "mean_token_accuracy": 0.16486039459705354,
      "num_tokens": 14867948.0,
      "step": 7960
    },
    {
      "entropy": 6.02838978767395,
      "epoch": 0.7033733663016601,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004957383663267152,
      "loss": 5.8825,
      "mean_token_accuracy": 0.16174710541963577,
      "num_tokens": 14877233.0,
      "step": 7965
    },
    {
      "entropy": 6.0542638301849365,
      "epoch": 0.7038149063935005,
      "grad_norm": 1.5,
      "learning_rate": 0.0004957322640362118,
      "loss": 5.8992,
      "mean_token_accuracy": 0.16317424327135086,
      "num_tokens": 14887439.0,
      "step": 7970
    },
    {
      "entropy": 6.137719011306762,
      "epoch": 0.7042564464853409,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004957261574216887,
      "loss": 5.9772,
      "mean_token_accuracy": 0.16205019503831863,
      "num_tokens": 14897064.0,
      "step": 7975
    },
    {
      "entropy": 6.055337810516358,
      "epoch": 0.7046979865771812,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004957200464832656,
      "loss": 5.9176,
      "mean_token_accuracy": 0.15968191921710967,
      "num_tokens": 14907054.0,
      "step": 7980
    },
    {
      "entropy": 5.990302705764771,
      "epoch": 0.7051395266690216,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004957139312210626,
      "loss": 5.8819,
      "mean_token_accuracy": 0.1649075925350189,
      "num_tokens": 14916969.0,
      "step": 7985
    },
    {
      "entropy": 6.02733588218689,
      "epoch": 0.7055810667608619,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000495707811635199,
      "loss": 5.907,
      "mean_token_accuracy": 0.16429728716611863,
      "num_tokens": 14926404.0,
      "step": 7990
    },
    {
      "entropy": 6.033973550796508,
      "epoch": 0.7060226068527022,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004957016877257949,
      "loss": 5.8792,
      "mean_token_accuracy": 0.1650012642145157,
      "num_tokens": 14935693.0,
      "step": 7995
    },
    {
      "entropy": 5.978080940246582,
      "epoch": 0.7064641469445425,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004956955594929704,
      "loss": 5.8245,
      "mean_token_accuracy": 0.16376616731286048,
      "num_tokens": 14945147.0,
      "step": 8000
    },
    {
      "entropy": 5.948422574996949,
      "epoch": 0.7069056870363829,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004956894269368454,
      "loss": 5.8657,
      "mean_token_accuracy": 0.1638239175081253,
      "num_tokens": 14954022.0,
      "step": 8005
    },
    {
      "entropy": 5.952633428573608,
      "epoch": 0.7073472271282233,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004956832900575399,
      "loss": 5.8716,
      "mean_token_accuracy": 0.16031511574983598,
      "num_tokens": 14962977.0,
      "step": 8010
    },
    {
      "entropy": 6.035693979263305,
      "epoch": 0.7077887672200636,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004956771488551742,
      "loss": 5.8928,
      "mean_token_accuracy": 0.15938186645507812,
      "num_tokens": 14973420.0,
      "step": 8015
    },
    {
      "entropy": 6.132956552505493,
      "epoch": 0.708230307311904,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004956710033298689,
      "loss": 5.8627,
      "mean_token_accuracy": 0.16708310693502426,
      "num_tokens": 14981762.0,
      "step": 8020
    },
    {
      "entropy": 6.101843976974488,
      "epoch": 0.7086718474037442,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000495664853481744,
      "loss": 5.9503,
      "mean_token_accuracy": 0.16041337698698044,
      "num_tokens": 14991449.0,
      "step": 8025
    },
    {
      "entropy": 6.028578805923462,
      "epoch": 0.7091133874955846,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004956586993109201,
      "loss": 5.8799,
      "mean_token_accuracy": 0.1627985119819641,
      "num_tokens": 15000802.0,
      "step": 8030
    },
    {
      "entropy": 6.014289045333863,
      "epoch": 0.7095549275874249,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004956525408175179,
      "loss": 5.8875,
      "mean_token_accuracy": 0.16235641092061998,
      "num_tokens": 15010580.0,
      "step": 8035
    },
    {
      "entropy": 5.969857788085937,
      "epoch": 0.7099964676792653,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004956463780016578,
      "loss": 5.9324,
      "mean_token_accuracy": 0.15980477035045623,
      "num_tokens": 15020059.0,
      "step": 8040
    },
    {
      "entropy": 6.070868968963623,
      "epoch": 0.7104380077711057,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004956402108634607,
      "loss": 5.8696,
      "mean_token_accuracy": 0.16538775116205215,
      "num_tokens": 15030268.0,
      "step": 8045
    },
    {
      "entropy": 5.997285652160644,
      "epoch": 0.7108795478629459,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004956340394030473,
      "loss": 5.8496,
      "mean_token_accuracy": 0.16604703515768052,
      "num_tokens": 15038865.0,
      "step": 8050
    },
    {
      "entropy": 6.069686794281006,
      "epoch": 0.7113210879547863,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004956278636205386,
      "loss": 5.991,
      "mean_token_accuracy": 0.1484125152230263,
      "num_tokens": 15048427.0,
      "step": 8055
    },
    {
      "entropy": 5.9873466968536375,
      "epoch": 0.7117626280466266,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004956216835160556,
      "loss": 5.8458,
      "mean_token_accuracy": 0.17582572996616364,
      "num_tokens": 15057366.0,
      "step": 8060
    },
    {
      "entropy": 6.044218683242798,
      "epoch": 0.712204168138467,
      "grad_norm": 1.625,
      "learning_rate": 0.0004956154990897191,
      "loss": 5.8795,
      "mean_token_accuracy": 0.15993637442588807,
      "num_tokens": 15067098.0,
      "step": 8065
    },
    {
      "entropy": 6.051040840148926,
      "epoch": 0.7126457082303073,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004956093103416505,
      "loss": 5.8797,
      "mean_token_accuracy": 0.16278290301561354,
      "num_tokens": 15075734.0,
      "step": 8070
    },
    {
      "entropy": 6.057865238189697,
      "epoch": 0.7130872483221476,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004956031172719711,
      "loss": 5.9255,
      "mean_token_accuracy": 0.16511817574501036,
      "num_tokens": 15084625.0,
      "step": 8075
    },
    {
      "entropy": 5.939253950119019,
      "epoch": 0.713528788413988,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004955969198808021,
      "loss": 5.7378,
      "mean_token_accuracy": 0.18222922831773758,
      "num_tokens": 15093256.0,
      "step": 8080
    },
    {
      "entropy": 5.952800703048706,
      "epoch": 0.7139703285058283,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004955907181682648,
      "loss": 5.8272,
      "mean_token_accuracy": 0.17085774391889572,
      "num_tokens": 15103175.0,
      "step": 8085
    },
    {
      "entropy": 6.015474128723144,
      "epoch": 0.7144118685976687,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004955845121344809,
      "loss": 5.9234,
      "mean_token_accuracy": 0.16088522225618362,
      "num_tokens": 15112789.0,
      "step": 8090
    },
    {
      "entropy": 6.063271713256836,
      "epoch": 0.714853408689509,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004955783017795717,
      "loss": 5.936,
      "mean_token_accuracy": 0.1662163808941841,
      "num_tokens": 15121869.0,
      "step": 8095
    },
    {
      "entropy": 6.000880479812622,
      "epoch": 0.7152949487813494,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000495572087103659,
      "loss": 5.8404,
      "mean_token_accuracy": 0.16656978577375411,
      "num_tokens": 15131441.0,
      "step": 8100
    },
    {
      "entropy": 5.968098402023315,
      "epoch": 0.7157364888731896,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004955658681068647,
      "loss": 5.8773,
      "mean_token_accuracy": 0.16107980757951737,
      "num_tokens": 15140641.0,
      "step": 8105
    },
    {
      "entropy": 6.045904302597046,
      "epoch": 0.71617802896503,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004955596447893105,
      "loss": 5.8728,
      "mean_token_accuracy": 0.16488367021083833,
      "num_tokens": 15150073.0,
      "step": 8110
    },
    {
      "entropy": 5.916897344589233,
      "epoch": 0.7166195690568704,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004955534171511181,
      "loss": 5.6291,
      "mean_token_accuracy": 0.18246244937181472,
      "num_tokens": 15159879.0,
      "step": 8115
    },
    {
      "entropy": 5.921168994903565,
      "epoch": 0.7170611091487107,
      "grad_norm": 1.875,
      "learning_rate": 0.00049554718519241,
      "loss": 5.8187,
      "mean_token_accuracy": 0.17096499651670455,
      "num_tokens": 15168316.0,
      "step": 8120
    },
    {
      "entropy": 5.946758127212524,
      "epoch": 0.7175026492405511,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004955409489133078,
      "loss": 5.8865,
      "mean_token_accuracy": 0.1636739045381546,
      "num_tokens": 15176700.0,
      "step": 8125
    },
    {
      "entropy": 6.089166688919067,
      "epoch": 0.7179441893323913,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004955347083139338,
      "loss": 5.8745,
      "mean_token_accuracy": 0.16882388591766356,
      "num_tokens": 15185997.0,
      "step": 8130
    },
    {
      "entropy": 5.963169384002685,
      "epoch": 0.7183857294242317,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004955284633944104,
      "loss": 5.8886,
      "mean_token_accuracy": 0.16364263594150544,
      "num_tokens": 15195341.0,
      "step": 8135
    },
    {
      "entropy": 5.988012361526489,
      "epoch": 0.718827269516072,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004955222141548597,
      "loss": 5.9061,
      "mean_token_accuracy": 0.1646143302321434,
      "num_tokens": 15205501.0,
      "step": 8140
    },
    {
      "entropy": 6.043216848373413,
      "epoch": 0.7192688096079124,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004955159605954043,
      "loss": 5.8431,
      "mean_token_accuracy": 0.1658138707280159,
      "num_tokens": 15214778.0,
      "step": 8145
    },
    {
      "entropy": 6.04915189743042,
      "epoch": 0.7197103496997528,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004955097027161666,
      "loss": 5.8873,
      "mean_token_accuracy": 0.16919119507074357,
      "num_tokens": 15224201.0,
      "step": 8150
    },
    {
      "entropy": 5.954038476943969,
      "epoch": 0.7201518897915931,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004955034405172693,
      "loss": 5.8114,
      "mean_token_accuracy": 0.17064497172832488,
      "num_tokens": 15233472.0,
      "step": 8155
    },
    {
      "entropy": 6.0233845710754395,
      "epoch": 0.7205934298834334,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000495497173998835,
      "loss": 5.8847,
      "mean_token_accuracy": 0.16386122703552247,
      "num_tokens": 15242323.0,
      "step": 8160
    },
    {
      "entropy": 6.042276906967163,
      "epoch": 0.7210349699752737,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004954909031609864,
      "loss": 5.9032,
      "mean_token_accuracy": 0.1618262857198715,
      "num_tokens": 15252952.0,
      "step": 8165
    },
    {
      "entropy": 6.041112995147705,
      "epoch": 0.7214765100671141,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004954846280038465,
      "loss": 5.8937,
      "mean_token_accuracy": 0.16644167602062226,
      "num_tokens": 15261815.0,
      "step": 8170
    },
    {
      "entropy": 6.043339776992798,
      "epoch": 0.7219180501589544,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004954783485275381,
      "loss": 5.929,
      "mean_token_accuracy": 0.15943233221769332,
      "num_tokens": 15271026.0,
      "step": 8175
    },
    {
      "entropy": 5.986793136596679,
      "epoch": 0.7223595902507948,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004954720647321843,
      "loss": 5.7721,
      "mean_token_accuracy": 0.1739621266722679,
      "num_tokens": 15279795.0,
      "step": 8180
    },
    {
      "entropy": 6.092932939529419,
      "epoch": 0.7228011303426352,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004954657766179081,
      "loss": 5.9041,
      "mean_token_accuracy": 0.1600492849946022,
      "num_tokens": 15289113.0,
      "step": 8185
    },
    {
      "entropy": 5.976881504058838,
      "epoch": 0.7232426704344754,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004954594841848327,
      "loss": 5.767,
      "mean_token_accuracy": 0.17719005048274994,
      "num_tokens": 15298271.0,
      "step": 8190
    },
    {
      "entropy": 6.073885917663574,
      "epoch": 0.7236842105263158,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004954531874330814,
      "loss": 5.8562,
      "mean_token_accuracy": 0.16308169662952424,
      "num_tokens": 15307677.0,
      "step": 8195
    },
    {
      "entropy": 6.005601215362549,
      "epoch": 0.7241257506181561,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004954468863627774,
      "loss": 5.8728,
      "mean_token_accuracy": 0.16475293338298796,
      "num_tokens": 15316637.0,
      "step": 8200
    },
    {
      "entropy": 5.967337942123413,
      "epoch": 0.7245672907099965,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004954405809740445,
      "loss": 5.8599,
      "mean_token_accuracy": 0.16259262561798096,
      "num_tokens": 15325221.0,
      "step": 8205
    },
    {
      "entropy": 6.023098516464233,
      "epoch": 0.7250088308018368,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004954342712670058,
      "loss": 5.8724,
      "mean_token_accuracy": 0.16482951641082763,
      "num_tokens": 15335877.0,
      "step": 8210
    },
    {
      "entropy": 6.053239488601685,
      "epoch": 0.7254503708936771,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004954279572417852,
      "loss": 5.8843,
      "mean_token_accuracy": 0.16335032880306244,
      "num_tokens": 15345096.0,
      "step": 8215
    },
    {
      "entropy": 5.979609155654908,
      "epoch": 0.7258919109855175,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004954216388985063,
      "loss": 5.7252,
      "mean_token_accuracy": 0.17507985085248948,
      "num_tokens": 15353785.0,
      "step": 8220
    },
    {
      "entropy": 5.9983946800231935,
      "epoch": 0.7263334510773578,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004954153162372928,
      "loss": 5.8308,
      "mean_token_accuracy": 0.1683450922369957,
      "num_tokens": 15362761.0,
      "step": 8225
    },
    {
      "entropy": 6.0386522769927975,
      "epoch": 0.7267749911691982,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004954089892582686,
      "loss": 5.8215,
      "mean_token_accuracy": 0.16770410239696504,
      "num_tokens": 15371951.0,
      "step": 8230
    },
    {
      "entropy": 6.033671808242798,
      "epoch": 0.7272165312610385,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004954026579615577,
      "loss": 5.8935,
      "mean_token_accuracy": 0.1658299371600151,
      "num_tokens": 15380277.0,
      "step": 8235
    },
    {
      "entropy": 5.973099327087402,
      "epoch": 0.7276580713528789,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004953963223472841,
      "loss": 5.8997,
      "mean_token_accuracy": 0.16341323554515838,
      "num_tokens": 15390951.0,
      "step": 8240
    },
    {
      "entropy": 6.110719299316406,
      "epoch": 0.7280996114447191,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004953899824155718,
      "loss": 5.8614,
      "mean_token_accuracy": 0.16662085205316543,
      "num_tokens": 15399394.0,
      "step": 8245
    },
    {
      "entropy": 5.979340410232544,
      "epoch": 0.7285411515365595,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004953836381665452,
      "loss": 5.8164,
      "mean_token_accuracy": 0.16606194376945496,
      "num_tokens": 15408644.0,
      "step": 8250
    },
    {
      "entropy": 6.033501100540161,
      "epoch": 0.7289826916283999,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004953772896003284,
      "loss": 5.8773,
      "mean_token_accuracy": 0.16229377537965775,
      "num_tokens": 15417398.0,
      "step": 8255
    },
    {
      "entropy": 6.058249139785767,
      "epoch": 0.7294242317202402,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004953709367170458,
      "loss": 5.9495,
      "mean_token_accuracy": 0.1523979589343071,
      "num_tokens": 15426532.0,
      "step": 8260
    },
    {
      "entropy": 6.039269495010376,
      "epoch": 0.7298657718120806,
      "grad_norm": 1.5,
      "learning_rate": 0.000495364579516822,
      "loss": 5.9197,
      "mean_token_accuracy": 0.16389914900064467,
      "num_tokens": 15436371.0,
      "step": 8265
    },
    {
      "entropy": 6.098191547393799,
      "epoch": 0.7303073119039208,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004953582179997813,
      "loss": 5.9687,
      "mean_token_accuracy": 0.15681514143943787,
      "num_tokens": 15446600.0,
      "step": 8270
    },
    {
      "entropy": 6.082651376724243,
      "epoch": 0.7307488519957612,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004953518521660484,
      "loss": 5.8208,
      "mean_token_accuracy": 0.16526079773902894,
      "num_tokens": 15455547.0,
      "step": 8275
    },
    {
      "entropy": 6.003137111663818,
      "epoch": 0.7311903920876015,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004953454820157481,
      "loss": 5.8614,
      "mean_token_accuracy": 0.16507752984762192,
      "num_tokens": 15464543.0,
      "step": 8280
    },
    {
      "entropy": 6.047074937820435,
      "epoch": 0.7316319321794419,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004953391075490051,
      "loss": 5.9886,
      "mean_token_accuracy": 0.15476133823394775,
      "num_tokens": 15473626.0,
      "step": 8285
    },
    {
      "entropy": 6.047168159484864,
      "epoch": 0.7320734722712823,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004953327287659444,
      "loss": 5.9098,
      "mean_token_accuracy": 0.15806199312210084,
      "num_tokens": 15484164.0,
      "step": 8290
    },
    {
      "entropy": 6.163697624206543,
      "epoch": 0.7325150123631226,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004953263456666907,
      "loss": 5.9146,
      "mean_token_accuracy": 0.15813378542661666,
      "num_tokens": 15493718.0,
      "step": 8295
    },
    {
      "entropy": 6.036547660827637,
      "epoch": 0.7329565524549629,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004953199582513693,
      "loss": 5.8568,
      "mean_token_accuracy": 0.1601982071995735,
      "num_tokens": 15503339.0,
      "step": 8300
    },
    {
      "entropy": 6.029874324798584,
      "epoch": 0.7333980925468032,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004953135665201052,
      "loss": 5.8401,
      "mean_token_accuracy": 0.17256525456905364,
      "num_tokens": 15513314.0,
      "step": 8305
    },
    {
      "entropy": 6.069580602645874,
      "epoch": 0.7338396326386436,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004953071704730237,
      "loss": 5.9203,
      "mean_token_accuracy": 0.15911492258310317,
      "num_tokens": 15522943.0,
      "step": 8310
    },
    {
      "entropy": 6.0191961288452145,
      "epoch": 0.7342811727304839,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004953007701102499,
      "loss": 5.8915,
      "mean_token_accuracy": 0.16012327522039413,
      "num_tokens": 15531975.0,
      "step": 8315
    },
    {
      "entropy": 5.931882619857788,
      "epoch": 0.7347227128223243,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004952943654319094,
      "loss": 5.8235,
      "mean_token_accuracy": 0.17004168182611465,
      "num_tokens": 15541139.0,
      "step": 8320
    },
    {
      "entropy": 5.996765184402466,
      "epoch": 0.7351642529141647,
      "grad_norm": 2.125,
      "learning_rate": 0.0004952879564381276,
      "loss": 5.9014,
      "mean_token_accuracy": 0.16608232408761978,
      "num_tokens": 15550604.0,
      "step": 8325
    },
    {
      "entropy": 6.023161125183106,
      "epoch": 0.7356057930060049,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004952815431290298,
      "loss": 5.8069,
      "mean_token_accuracy": 0.17140596508979797,
      "num_tokens": 15558995.0,
      "step": 8330
    },
    {
      "entropy": 6.1181238174438475,
      "epoch": 0.7360473330978453,
      "grad_norm": 1.609375,
      "learning_rate": 0.000495275125504742,
      "loss": 5.9002,
      "mean_token_accuracy": 0.16549504548311234,
      "num_tokens": 15568229.0,
      "step": 8335
    },
    {
      "entropy": 5.97317328453064,
      "epoch": 0.7364888731896856,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004952687035653899,
      "loss": 5.8214,
      "mean_token_accuracy": 0.167140843719244,
      "num_tokens": 15577042.0,
      "step": 8340
    },
    {
      "entropy": 5.926223802566528,
      "epoch": 0.736930413281526,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004952622773110989,
      "loss": 5.7381,
      "mean_token_accuracy": 0.17849351614713668,
      "num_tokens": 15585825.0,
      "step": 8345
    },
    {
      "entropy": 5.940134286880493,
      "epoch": 0.7373719533733663,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004952558467419953,
      "loss": 5.7658,
      "mean_token_accuracy": 0.17126924693584442,
      "num_tokens": 15595486.0,
      "step": 8350
    },
    {
      "entropy": 5.950185823440552,
      "epoch": 0.7378134934652066,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000495249411858205,
      "loss": 5.813,
      "mean_token_accuracy": 0.15817922800779344,
      "num_tokens": 15605312.0,
      "step": 8355
    },
    {
      "entropy": 5.981673097610473,
      "epoch": 0.738255033557047,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004952429726598538,
      "loss": 5.8119,
      "mean_token_accuracy": 0.17498987764120102,
      "num_tokens": 15615185.0,
      "step": 8360
    },
    {
      "entropy": 6.078198766708374,
      "epoch": 0.7386965736488873,
      "grad_norm": 1.5,
      "learning_rate": 0.0004952365291470682,
      "loss": 5.9209,
      "mean_token_accuracy": 0.16283219456672668,
      "num_tokens": 15625259.0,
      "step": 8365
    },
    {
      "entropy": 6.13894739151001,
      "epoch": 0.7391381137407277,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004952300813199742,
      "loss": 5.8934,
      "mean_token_accuracy": 0.16778266727924346,
      "num_tokens": 15635040.0,
      "step": 8370
    },
    {
      "entropy": 6.093637466430664,
      "epoch": 0.739579653832568,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004952236291786981,
      "loss": 5.9225,
      "mean_token_accuracy": 0.15750904381275177,
      "num_tokens": 15644996.0,
      "step": 8375
    },
    {
      "entropy": 5.882752561569214,
      "epoch": 0.7400211939244083,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004952171727233665,
      "loss": 5.8031,
      "mean_token_accuracy": 0.16644213497638702,
      "num_tokens": 15654598.0,
      "step": 8380
    },
    {
      "entropy": 5.981708002090454,
      "epoch": 0.7404627340162486,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004952107119541055,
      "loss": 5.8277,
      "mean_token_accuracy": 0.1631438508629799,
      "num_tokens": 15664213.0,
      "step": 8385
    },
    {
      "entropy": 6.015055465698242,
      "epoch": 0.740904274108089,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004952042468710421,
      "loss": 5.898,
      "mean_token_accuracy": 0.16343553215265275,
      "num_tokens": 15673595.0,
      "step": 8390
    },
    {
      "entropy": 6.0489397048950195,
      "epoch": 0.7413458141999294,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004951977774743027,
      "loss": 5.9483,
      "mean_token_accuracy": 0.15604591816663743,
      "num_tokens": 15682753.0,
      "step": 8395
    },
    {
      "entropy": 6.007998657226563,
      "epoch": 0.7417873542917697,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004951913037640139,
      "loss": 5.7714,
      "mean_token_accuracy": 0.17713254690170288,
      "num_tokens": 15691508.0,
      "step": 8400
    },
    {
      "entropy": 5.9607525825500485,
      "epoch": 0.7422288943836101,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004951848257403029,
      "loss": 5.7621,
      "mean_token_accuracy": 0.16910529732704163,
      "num_tokens": 15701110.0,
      "step": 8405
    },
    {
      "entropy": 5.835663938522339,
      "epoch": 0.7426704344754503,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004951783434032963,
      "loss": 5.7001,
      "mean_token_accuracy": 0.17653533816337585,
      "num_tokens": 15710565.0,
      "step": 8410
    },
    {
      "entropy": 5.945553207397461,
      "epoch": 0.7431119745672907,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004951718567531211,
      "loss": 5.8933,
      "mean_token_accuracy": 0.16108815222978592,
      "num_tokens": 15718364.0,
      "step": 8415
    },
    {
      "entropy": 6.040232849121094,
      "epoch": 0.743553514659131,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004951653657899045,
      "loss": 5.9572,
      "mean_token_accuracy": 0.15713704228401185,
      "num_tokens": 15729127.0,
      "step": 8420
    },
    {
      "entropy": 6.014114904403686,
      "epoch": 0.7439950547509714,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004951588705137737,
      "loss": 5.9172,
      "mean_token_accuracy": 0.160747928917408,
      "num_tokens": 15738491.0,
      "step": 8425
    },
    {
      "entropy": 6.051564168930054,
      "epoch": 0.7444365948428118,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004951523709248557,
      "loss": 5.812,
      "mean_token_accuracy": 0.17417716681957246,
      "num_tokens": 15748346.0,
      "step": 8430
    },
    {
      "entropy": 5.936352062225342,
      "epoch": 0.744878134934652,
      "grad_norm": 1.5625,
      "learning_rate": 0.000495145867023278,
      "loss": 5.8178,
      "mean_token_accuracy": 0.17726168632507325,
      "num_tokens": 15757242.0,
      "step": 8435
    },
    {
      "entropy": 6.062502765655518,
      "epoch": 0.7453196750264924,
      "grad_norm": 1.921875,
      "learning_rate": 0.000495139358809168,
      "loss": 5.8817,
      "mean_token_accuracy": 0.16312602609395982,
      "num_tokens": 15767236.0,
      "step": 8440
    },
    {
      "entropy": 6.0330291271209715,
      "epoch": 0.7457612151183327,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004951328462826532,
      "loss": 5.917,
      "mean_token_accuracy": 0.15931661278009415,
      "num_tokens": 15776977.0,
      "step": 8445
    },
    {
      "entropy": 5.967469167709351,
      "epoch": 0.7462027552101731,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004951263294438611,
      "loss": 5.8146,
      "mean_token_accuracy": 0.16552847474813462,
      "num_tokens": 15787333.0,
      "step": 8450
    },
    {
      "entropy": 5.8979497909545895,
      "epoch": 0.7466442953020134,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004951198082929194,
      "loss": 5.6991,
      "mean_token_accuracy": 0.182376991212368,
      "num_tokens": 15794773.0,
      "step": 8455
    },
    {
      "entropy": 5.887519073486328,
      "epoch": 0.7470858353938538,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004951132828299558,
      "loss": 5.8119,
      "mean_token_accuracy": 0.17694963216781617,
      "num_tokens": 15803803.0,
      "step": 8460
    },
    {
      "entropy": 6.003292751312256,
      "epoch": 0.7475273754856941,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004951067530550982,
      "loss": 5.8686,
      "mean_token_accuracy": 0.1630527436733246,
      "num_tokens": 15814021.0,
      "step": 8465
    },
    {
      "entropy": 6.018431043624878,
      "epoch": 0.7479689155775344,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004951002189684745,
      "loss": 5.847,
      "mean_token_accuracy": 0.1591825008392334,
      "num_tokens": 15822974.0,
      "step": 8470
    },
    {
      "entropy": 5.89467420578003,
      "epoch": 0.7484104556693748,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004950936805702127,
      "loss": 5.7536,
      "mean_token_accuracy": 0.17445838451385498,
      "num_tokens": 15832202.0,
      "step": 8475
    },
    {
      "entropy": 6.020446825027466,
      "epoch": 0.7488519957612151,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004950871378604409,
      "loss": 5.8714,
      "mean_token_accuracy": 0.16651229858398436,
      "num_tokens": 15841223.0,
      "step": 8480
    },
    {
      "entropy": 6.053658437728882,
      "epoch": 0.7492935358530555,
      "grad_norm": 2.125,
      "learning_rate": 0.0004950805908392872,
      "loss": 5.8103,
      "mean_token_accuracy": 0.1726827636361122,
      "num_tokens": 15850260.0,
      "step": 8485
    },
    {
      "entropy": 5.962708044052124,
      "epoch": 0.7497350759448957,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004950740395068799,
      "loss": 5.9007,
      "mean_token_accuracy": 0.16698147505521774,
      "num_tokens": 15859780.0,
      "step": 8490
    },
    {
      "entropy": 5.943686819076538,
      "epoch": 0.7501766160367361,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004950674838633473,
      "loss": 5.8261,
      "mean_token_accuracy": 0.16707435846328736,
      "num_tokens": 15868331.0,
      "step": 8495
    },
    {
      "entropy": 6.012507152557373,
      "epoch": 0.7506181561285765,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004950609239088178,
      "loss": 5.8834,
      "mean_token_accuracy": 0.16294038891792298,
      "num_tokens": 15877359.0,
      "step": 8500
    },
    {
      "entropy": 6.04980845451355,
      "epoch": 0.7510596962204168,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004950543596434201,
      "loss": 5.7905,
      "mean_token_accuracy": 0.1808181643486023,
      "num_tokens": 15886976.0,
      "step": 8505
    },
    {
      "entropy": 5.95252537727356,
      "epoch": 0.7515012363122572,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004950477910672825,
      "loss": 5.8655,
      "mean_token_accuracy": 0.1658195436000824,
      "num_tokens": 15896139.0,
      "step": 8510
    },
    {
      "entropy": 5.946612977981568,
      "epoch": 0.7519427764040975,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000495041218180534,
      "loss": 5.8504,
      "mean_token_accuracy": 0.17413663417100905,
      "num_tokens": 15904672.0,
      "step": 8515
    },
    {
      "entropy": 5.9041801452636715,
      "epoch": 0.7523843164959378,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000495034640983303,
      "loss": 5.7024,
      "mean_token_accuracy": 0.176799039542675,
      "num_tokens": 15912716.0,
      "step": 8520
    },
    {
      "entropy": 6.003086757659912,
      "epoch": 0.7528258565877781,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004950280594757186,
      "loss": 5.8215,
      "mean_token_accuracy": 0.16853666454553604,
      "num_tokens": 15921974.0,
      "step": 8525
    },
    {
      "entropy": 6.049868392944336,
      "epoch": 0.7532673966796185,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004950214736579097,
      "loss": 5.8433,
      "mean_token_accuracy": 0.16325998157262803,
      "num_tokens": 15930859.0,
      "step": 8530
    },
    {
      "entropy": 5.95236120223999,
      "epoch": 0.7537089367714589,
      "grad_norm": 1.75,
      "learning_rate": 0.0004950148835300051,
      "loss": 5.7595,
      "mean_token_accuracy": 0.1746041879057884,
      "num_tokens": 15940052.0,
      "step": 8535
    },
    {
      "entropy": 5.976876878738404,
      "epoch": 0.7541504768632992,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004950082890921341,
      "loss": 5.833,
      "mean_token_accuracy": 0.16658651530742646,
      "num_tokens": 15948422.0,
      "step": 8540
    },
    {
      "entropy": 5.991146755218506,
      "epoch": 0.7545920169551396,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004950016903444261,
      "loss": 5.9212,
      "mean_token_accuracy": 0.16847308576107026,
      "num_tokens": 15957433.0,
      "step": 8545
    },
    {
      "entropy": 6.095296812057495,
      "epoch": 0.7550335570469798,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004949950872870099,
      "loss": 5.8565,
      "mean_token_accuracy": 0.16426337659358978,
      "num_tokens": 15966542.0,
      "step": 8550
    },
    {
      "entropy": 6.052794790267944,
      "epoch": 0.7554750971388202,
      "grad_norm": 1.96875,
      "learning_rate": 0.000494988479920015,
      "loss": 5.9687,
      "mean_token_accuracy": 0.15761574506759643,
      "num_tokens": 15977067.0,
      "step": 8555
    },
    {
      "entropy": 6.062600803375244,
      "epoch": 0.7559166372306605,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000494981868243571,
      "loss": 5.8892,
      "mean_token_accuracy": 0.16307247430086136,
      "num_tokens": 15986502.0,
      "step": 8560
    },
    {
      "entropy": 6.134724473953247,
      "epoch": 0.7563581773225009,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004949752522578073,
      "loss": 5.9426,
      "mean_token_accuracy": 0.15969116240739822,
      "num_tokens": 15996788.0,
      "step": 8565
    },
    {
      "entropy": 6.0735067367553714,
      "epoch": 0.7567997174143413,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004949686319628535,
      "loss": 5.846,
      "mean_token_accuracy": 0.16205350160598755,
      "num_tokens": 16007212.0,
      "step": 8570
    },
    {
      "entropy": 5.9686089038848875,
      "epoch": 0.7572412575061815,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004949620073588394,
      "loss": 5.8694,
      "mean_token_accuracy": 0.16372007131576538,
      "num_tokens": 16017000.0,
      "step": 8575
    },
    {
      "entropy": 6.056089782714844,
      "epoch": 0.7576827975980219,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004949553784458945,
      "loss": 5.9647,
      "mean_token_accuracy": 0.15608513206243516,
      "num_tokens": 16026647.0,
      "step": 8580
    },
    {
      "entropy": 5.986236476898194,
      "epoch": 0.7581243376898622,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004949487452241489,
      "loss": 5.6913,
      "mean_token_accuracy": 0.1788015127182007,
      "num_tokens": 16034773.0,
      "step": 8585
    },
    {
      "entropy": 6.041558456420899,
      "epoch": 0.7585658777817026,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004949421076937325,
      "loss": 5.8507,
      "mean_token_accuracy": 0.1660602033138275,
      "num_tokens": 16043787.0,
      "step": 8590
    },
    {
      "entropy": 5.984687471389771,
      "epoch": 0.7590074178735429,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004949354658547753,
      "loss": 5.8063,
      "mean_token_accuracy": 0.17274508327245713,
      "num_tokens": 16053130.0,
      "step": 8595
    },
    {
      "entropy": 5.994377183914184,
      "epoch": 0.7594489579653833,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004949288197074074,
      "loss": 5.9551,
      "mean_token_accuracy": 0.15927992463111879,
      "num_tokens": 16062529.0,
      "step": 8600
    },
    {
      "entropy": 5.955754947662354,
      "epoch": 0.7598904980572236,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004949221692517589,
      "loss": 5.8028,
      "mean_token_accuracy": 0.16142662912607192,
      "num_tokens": 16071932.0,
      "step": 8605
    },
    {
      "entropy": 6.0386292934417725,
      "epoch": 0.7603320381490639,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004949155144879602,
      "loss": 5.7827,
      "mean_token_accuracy": 0.17397554367780685,
      "num_tokens": 16081077.0,
      "step": 8610
    },
    {
      "entropy": 5.993456697463989,
      "epoch": 0.7607735782409043,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004949088554161417,
      "loss": 5.7304,
      "mean_token_accuracy": 0.17781465798616408,
      "num_tokens": 16089725.0,
      "step": 8615
    },
    {
      "entropy": 6.007733106613159,
      "epoch": 0.7612151183327446,
      "grad_norm": 1.5,
      "learning_rate": 0.0004949021920364339,
      "loss": 5.8982,
      "mean_token_accuracy": 0.1661226361989975,
      "num_tokens": 16099362.0,
      "step": 8620
    },
    {
      "entropy": 6.0902595043182375,
      "epoch": 0.761656658424585,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004948955243489671,
      "loss": 5.917,
      "mean_token_accuracy": 0.15821752846240997,
      "num_tokens": 16108746.0,
      "step": 8625
    },
    {
      "entropy": 6.114125061035156,
      "epoch": 0.7620981985164252,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004948888523538721,
      "loss": 6.0579,
      "mean_token_accuracy": 0.15333025604486467,
      "num_tokens": 16118902.0,
      "step": 8630
    },
    {
      "entropy": 5.9824565887451175,
      "epoch": 0.7625397386082656,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004948821760512795,
      "loss": 5.7935,
      "mean_token_accuracy": 0.1688033312559128,
      "num_tokens": 16127781.0,
      "step": 8635
    },
    {
      "entropy": 5.991695070266724,
      "epoch": 0.762981278700106,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004948754954413201,
      "loss": 5.7665,
      "mean_token_accuracy": 0.1775361105799675,
      "num_tokens": 16136004.0,
      "step": 8640
    },
    {
      "entropy": 5.9446573734283445,
      "epoch": 0.7634228187919463,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004948688105241248,
      "loss": 5.8371,
      "mean_token_accuracy": 0.1675607979297638,
      "num_tokens": 16145023.0,
      "step": 8645
    },
    {
      "entropy": 5.97957935333252,
      "epoch": 0.7638643588837867,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004948621212998246,
      "loss": 5.8866,
      "mean_token_accuracy": 0.1627459466457367,
      "num_tokens": 16154242.0,
      "step": 8650
    },
    {
      "entropy": 6.039345216751099,
      "epoch": 0.764305898975627,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004948554277685505,
      "loss": 5.8596,
      "mean_token_accuracy": 0.1614910364151001,
      "num_tokens": 16163252.0,
      "step": 8655
    },
    {
      "entropy": 6.080992984771728,
      "epoch": 0.7647474390674673,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004948487299304335,
      "loss": 5.7838,
      "mean_token_accuracy": 0.1723628520965576,
      "num_tokens": 16171363.0,
      "step": 8660
    },
    {
      "entropy": 6.038664770126343,
      "epoch": 0.7651889791593076,
      "grad_norm": 2.375,
      "learning_rate": 0.000494842027785605,
      "loss": 5.9707,
      "mean_token_accuracy": 0.16340688914060592,
      "num_tokens": 16181234.0,
      "step": 8665
    },
    {
      "entropy": 6.008198308944702,
      "epoch": 0.765630519251148,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004948353213341961,
      "loss": 5.8662,
      "mean_token_accuracy": 0.16622024774551392,
      "num_tokens": 16189420.0,
      "step": 8670
    },
    {
      "entropy": 6.095108985900879,
      "epoch": 0.7660720593429884,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004948286105763385,
      "loss": 6.0349,
      "mean_token_accuracy": 0.14946414679288864,
      "num_tokens": 16199202.0,
      "step": 8675
    },
    {
      "entropy": 6.088163900375366,
      "epoch": 0.7665135994348287,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004948218955121632,
      "loss": 5.8595,
      "mean_token_accuracy": 0.16536297798156738,
      "num_tokens": 16208481.0,
      "step": 8680
    },
    {
      "entropy": 6.020900392532349,
      "epoch": 0.766955139526669,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004948151761418021,
      "loss": 5.928,
      "mean_token_accuracy": 0.16450031846761703,
      "num_tokens": 16217282.0,
      "step": 8685
    },
    {
      "entropy": 6.010745429992676,
      "epoch": 0.7673966796185093,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004948084524653868,
      "loss": 5.786,
      "mean_token_accuracy": 0.17689766138792037,
      "num_tokens": 16226567.0,
      "step": 8690
    },
    {
      "entropy": 6.038272476196289,
      "epoch": 0.7678382197103497,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004948017244830489,
      "loss": 5.9912,
      "mean_token_accuracy": 0.1531811460852623,
      "num_tokens": 16236255.0,
      "step": 8695
    },
    {
      "entropy": 6.012173700332641,
      "epoch": 0.76827975980219,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004947949921949201,
      "loss": 5.7611,
      "mean_token_accuracy": 0.18096890598535537,
      "num_tokens": 16244838.0,
      "step": 8700
    },
    {
      "entropy": 5.877737808227539,
      "epoch": 0.7687212998940304,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004947882556011325,
      "loss": 5.6674,
      "mean_token_accuracy": 0.17592710703611375,
      "num_tokens": 16254101.0,
      "step": 8705
    },
    {
      "entropy": 5.790978288650512,
      "epoch": 0.7691628399858708,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000494781514701818,
      "loss": 5.7549,
      "mean_token_accuracy": 0.17472999691963195,
      "num_tokens": 16263496.0,
      "step": 8710
    },
    {
      "entropy": 6.015494537353516,
      "epoch": 0.769604380077711,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004947747694971085,
      "loss": 5.8577,
      "mean_token_accuracy": 0.17157624363899232,
      "num_tokens": 16271884.0,
      "step": 8715
    },
    {
      "entropy": 6.020488214492798,
      "epoch": 0.7700459201695514,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004947680199871363,
      "loss": 5.7849,
      "mean_token_accuracy": 0.17175298631191255,
      "num_tokens": 16281197.0,
      "step": 8720
    },
    {
      "entropy": 5.941747856140137,
      "epoch": 0.7704874602613917,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004947612661720336,
      "loss": 5.8585,
      "mean_token_accuracy": 0.16881806254386902,
      "num_tokens": 16289908.0,
      "step": 8725
    },
    {
      "entropy": 5.970186948776245,
      "epoch": 0.7709290003532321,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004947545080519327,
      "loss": 5.8431,
      "mean_token_accuracy": 0.16303220093250276,
      "num_tokens": 16299237.0,
      "step": 8730
    },
    {
      "entropy": 6.07957649230957,
      "epoch": 0.7713705404450724,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000494747745626966,
      "loss": 5.8625,
      "mean_token_accuracy": 0.17015912532806396,
      "num_tokens": 16308175.0,
      "step": 8735
    },
    {
      "entropy": 5.939646482467651,
      "epoch": 0.7718120805369127,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004947409788972659,
      "loss": 5.8134,
      "mean_token_accuracy": 0.17128169685602188,
      "num_tokens": 16317121.0,
      "step": 8740
    },
    {
      "entropy": 5.888267755508423,
      "epoch": 0.7722536206287531,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004947342078629649,
      "loss": 5.7836,
      "mean_token_accuracy": 0.1751389503479004,
      "num_tokens": 16325764.0,
      "step": 8745
    },
    {
      "entropy": 6.0162074089050295,
      "epoch": 0.7726951607205934,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004947274325241959,
      "loss": 5.7971,
      "mean_token_accuracy": 0.1663631409406662,
      "num_tokens": 16333823.0,
      "step": 8750
    },
    {
      "entropy": 6.007051038742065,
      "epoch": 0.7731367008124338,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004947206528810913,
      "loss": 5.7953,
      "mean_token_accuracy": 0.1734223335981369,
      "num_tokens": 16341918.0,
      "step": 8755
    },
    {
      "entropy": 5.964048862457275,
      "epoch": 0.7735782409042741,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004947138689337843,
      "loss": 5.8454,
      "mean_token_accuracy": 0.1691407859325409,
      "num_tokens": 16351904.0,
      "step": 8760
    },
    {
      "entropy": 6.049382305145263,
      "epoch": 0.7740197809961145,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004947070806824074,
      "loss": 5.863,
      "mean_token_accuracy": 0.16452886164188385,
      "num_tokens": 16360921.0,
      "step": 8765
    },
    {
      "entropy": 6.076323413848877,
      "epoch": 0.7744613210879547,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004947002881270937,
      "loss": 5.882,
      "mean_token_accuracy": 0.16276332437992097,
      "num_tokens": 16370929.0,
      "step": 8770
    },
    {
      "entropy": 6.061858701705932,
      "epoch": 0.7749028611797951,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004946934912679764,
      "loss": 5.8834,
      "mean_token_accuracy": 0.1623873770236969,
      "num_tokens": 16380099.0,
      "step": 8775
    },
    {
      "entropy": 6.058119440078736,
      "epoch": 0.7753444012716355,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004946866901051885,
      "loss": 5.8984,
      "mean_token_accuracy": 0.166536608338356,
      "num_tokens": 16389109.0,
      "step": 8780
    },
    {
      "entropy": 5.995142030715942,
      "epoch": 0.7757859413634758,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004946798846388634,
      "loss": 5.9407,
      "mean_token_accuracy": 0.1601017102599144,
      "num_tokens": 16398601.0,
      "step": 8785
    },
    {
      "entropy": 5.908467817306518,
      "epoch": 0.7762274814553162,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004946730748691342,
      "loss": 5.804,
      "mean_token_accuracy": 0.17755288928747176,
      "num_tokens": 16406950.0,
      "step": 8790
    },
    {
      "entropy": 6.086664867401123,
      "epoch": 0.7766690215471564,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004946662607961344,
      "loss": 5.812,
      "mean_token_accuracy": 0.16931003481149673,
      "num_tokens": 16414680.0,
      "step": 8795
    },
    {
      "entropy": 6.029450368881226,
      "epoch": 0.7771105616389968,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004946594424199976,
      "loss": 5.8757,
      "mean_token_accuracy": 0.16709612607955932,
      "num_tokens": 16424060.0,
      "step": 8800
    },
    {
      "entropy": 6.009132289886475,
      "epoch": 0.7775521017308371,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004946526197408571,
      "loss": 5.8871,
      "mean_token_accuracy": 0.16631227284669875,
      "num_tokens": 16433691.0,
      "step": 8805
    },
    {
      "entropy": 6.034335660934448,
      "epoch": 0.7779936418226775,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004946457927588467,
      "loss": 5.9038,
      "mean_token_accuracy": 0.16435332745313644,
      "num_tokens": 16443234.0,
      "step": 8810
    },
    {
      "entropy": 5.976377153396607,
      "epoch": 0.7784351819145179,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004946389614741002,
      "loss": 5.828,
      "mean_token_accuracy": 0.17055855989456176,
      "num_tokens": 16452084.0,
      "step": 8815
    },
    {
      "entropy": 6.041907787322998,
      "epoch": 0.7788767220063582,
      "grad_norm": 1.375,
      "learning_rate": 0.0004946321258867513,
      "loss": 5.8151,
      "mean_token_accuracy": 0.16775723546743393,
      "num_tokens": 16460135.0,
      "step": 8820
    },
    {
      "entropy": 6.004779911041259,
      "epoch": 0.7793182620981985,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000494625285996934,
      "loss": 5.9481,
      "mean_token_accuracy": 0.16491457521915437,
      "num_tokens": 16469919.0,
      "step": 8825
    },
    {
      "entropy": 5.999904203414917,
      "epoch": 0.7797598021900388,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004946184418047823,
      "loss": 5.8509,
      "mean_token_accuracy": 0.16316285580396653,
      "num_tokens": 16478473.0,
      "step": 8830
    },
    {
      "entropy": 6.058293390274048,
      "epoch": 0.7802013422818792,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004946115933104301,
      "loss": 5.8679,
      "mean_token_accuracy": 0.15920920595526694,
      "num_tokens": 16486878.0,
      "step": 8835
    },
    {
      "entropy": 6.025899887084961,
      "epoch": 0.7806428823737195,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004946047405140119,
      "loss": 5.901,
      "mean_token_accuracy": 0.16094491481781006,
      "num_tokens": 16496798.0,
      "step": 8840
    },
    {
      "entropy": 6.0621325969696045,
      "epoch": 0.7810844224655599,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004945978834156616,
      "loss": 5.8335,
      "mean_token_accuracy": 0.16948919147253036,
      "num_tokens": 16506508.0,
      "step": 8845
    },
    {
      "entropy": 5.898591375350952,
      "epoch": 0.7815259625574003,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004945910220155136,
      "loss": 5.8036,
      "mean_token_accuracy": 0.1735559344291687,
      "num_tokens": 16514979.0,
      "step": 8850
    },
    {
      "entropy": 6.0083986759185795,
      "epoch": 0.7819675026492405,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004945841563137025,
      "loss": 5.8737,
      "mean_token_accuracy": 0.1645631179213524,
      "num_tokens": 16524654.0,
      "step": 8855
    },
    {
      "entropy": 6.1006495475769045,
      "epoch": 0.7824090427410809,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004945772863103626,
      "loss": 5.9451,
      "mean_token_accuracy": 0.15826726704835892,
      "num_tokens": 16534486.0,
      "step": 8860
    },
    {
      "entropy": 6.079514646530152,
      "epoch": 0.7828505828329212,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004945704120056286,
      "loss": 5.8814,
      "mean_token_accuracy": 0.17055715173482894,
      "num_tokens": 16543718.0,
      "step": 8865
    },
    {
      "entropy": 5.943536615371704,
      "epoch": 0.7832921229247616,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000494563533399635,
      "loss": 5.8278,
      "mean_token_accuracy": 0.16279819309711457,
      "num_tokens": 16553159.0,
      "step": 8870
    },
    {
      "entropy": 6.071165704727173,
      "epoch": 0.7837336630166019,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004945566504925167,
      "loss": 5.8979,
      "mean_token_accuracy": 0.16369519382715225,
      "num_tokens": 16562398.0,
      "step": 8875
    },
    {
      "entropy": 6.046672582626343,
      "epoch": 0.7841752031084422,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004945497632844084,
      "loss": 5.8434,
      "mean_token_accuracy": 0.1709023430943489,
      "num_tokens": 16571530.0,
      "step": 8880
    },
    {
      "entropy": 5.891947078704834,
      "epoch": 0.7846167432002826,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004945428717754453,
      "loss": 5.8451,
      "mean_token_accuracy": 0.16528365015983582,
      "num_tokens": 16581074.0,
      "step": 8885
    },
    {
      "entropy": 5.903192663192749,
      "epoch": 0.7850582832921229,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000494535975965762,
      "loss": 5.7092,
      "mean_token_accuracy": 0.17383420020341872,
      "num_tokens": 16589574.0,
      "step": 8890
    },
    {
      "entropy": 5.908491230010986,
      "epoch": 0.7854998233839633,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004945290758554939,
      "loss": 5.7966,
      "mean_token_accuracy": 0.17097795158624648,
      "num_tokens": 16599623.0,
      "step": 8895
    },
    {
      "entropy": 5.951535558700561,
      "epoch": 0.7859413634758036,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000494522171444776,
      "loss": 5.8771,
      "mean_token_accuracy": 0.1554115429520607,
      "num_tokens": 16608559.0,
      "step": 8900
    },
    {
      "entropy": 6.007721900939941,
      "epoch": 0.786382903567644,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004945152627337437,
      "loss": 5.7946,
      "mean_token_accuracy": 0.17543937712907792,
      "num_tokens": 16617148.0,
      "step": 8905
    },
    {
      "entropy": 5.975952243804931,
      "epoch": 0.7868244436594842,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004945083497225321,
      "loss": 5.8109,
      "mean_token_accuracy": 0.16906408369541168,
      "num_tokens": 16626405.0,
      "step": 8910
    },
    {
      "entropy": 6.001080751419067,
      "epoch": 0.7872659837513246,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004945014324112769,
      "loss": 5.833,
      "mean_token_accuracy": 0.17065125554800034,
      "num_tokens": 16635116.0,
      "step": 8915
    },
    {
      "entropy": 5.936459684371949,
      "epoch": 0.787707523843165,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004944945108001134,
      "loss": 5.8052,
      "mean_token_accuracy": 0.1653392180800438,
      "num_tokens": 16644381.0,
      "step": 8920
    },
    {
      "entropy": 5.9898097038269045,
      "epoch": 0.7881490639350053,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004944875848891773,
      "loss": 5.8612,
      "mean_token_accuracy": 0.1636720508337021,
      "num_tokens": 16654275.0,
      "step": 8925
    },
    {
      "entropy": 6.067295503616333,
      "epoch": 0.7885906040268457,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004944806546786042,
      "loss": 5.9303,
      "mean_token_accuracy": 0.1632734939455986,
      "num_tokens": 16663484.0,
      "step": 8930
    },
    {
      "entropy": 6.037387943267822,
      "epoch": 0.7890321441186859,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00049447372016853,
      "loss": 5.8665,
      "mean_token_accuracy": 0.17046364545822143,
      "num_tokens": 16672490.0,
      "step": 8935
    },
    {
      "entropy": 6.083343410491944,
      "epoch": 0.7894736842105263,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004944667813590904,
      "loss": 5.8409,
      "mean_token_accuracy": 0.165067557990551,
      "num_tokens": 16681242.0,
      "step": 8940
    },
    {
      "entropy": 5.971105098724365,
      "epoch": 0.7899152243023666,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004944598382504214,
      "loss": 5.7775,
      "mean_token_accuracy": 0.16900418400764466,
      "num_tokens": 16690359.0,
      "step": 8945
    },
    {
      "entropy": 5.975606966018677,
      "epoch": 0.790356764394207,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004944528908426589,
      "loss": 5.8643,
      "mean_token_accuracy": 0.1650958314538002,
      "num_tokens": 16700774.0,
      "step": 8950
    },
    {
      "entropy": 6.042025375366211,
      "epoch": 0.7907983044860474,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004944459391359391,
      "loss": 5.9374,
      "mean_token_accuracy": 0.16598828583955766,
      "num_tokens": 16710513.0,
      "step": 8955
    },
    {
      "entropy": 5.985045576095581,
      "epoch": 0.7912398445778877,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004944389831303982,
      "loss": 5.8374,
      "mean_token_accuracy": 0.16597743779420854,
      "num_tokens": 16720785.0,
      "step": 8960
    },
    {
      "entropy": 6.096461009979248,
      "epoch": 0.791681384669728,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004944320228261723,
      "loss": 5.8489,
      "mean_token_accuracy": 0.16366840451955794,
      "num_tokens": 16729083.0,
      "step": 8965
    },
    {
      "entropy": 5.978036975860595,
      "epoch": 0.7921229247615683,
      "grad_norm": 1.9140625,
      "learning_rate": 0.000494425058223398,
      "loss": 5.8412,
      "mean_token_accuracy": 0.16302479505538942,
      "num_tokens": 16738313.0,
      "step": 8970
    },
    {
      "entropy": 5.952572774887085,
      "epoch": 0.7925644648534087,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004944180893222117,
      "loss": 5.7871,
      "mean_token_accuracy": 0.17404089868068695,
      "num_tokens": 16747191.0,
      "step": 8975
    },
    {
      "entropy": 5.940504741668701,
      "epoch": 0.793006004945249,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004944111161227496,
      "loss": 5.8295,
      "mean_token_accuracy": 0.16756651401519776,
      "num_tokens": 16755921.0,
      "step": 8980
    },
    {
      "entropy": 6.005930614471436,
      "epoch": 0.7934475450370894,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004944041386251486,
      "loss": 5.9722,
      "mean_token_accuracy": 0.162301617115736,
      "num_tokens": 16765394.0,
      "step": 8985
    },
    {
      "entropy": 6.021315050125122,
      "epoch": 0.7938890851289298,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004943971568295454,
      "loss": 5.851,
      "mean_token_accuracy": 0.17589876651763917,
      "num_tokens": 16774824.0,
      "step": 8990
    },
    {
      "entropy": 6.0411216735839846,
      "epoch": 0.79433062522077,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004943901707360765,
      "loss": 5.9653,
      "mean_token_accuracy": 0.15621765553951264,
      "num_tokens": 16783945.0,
      "step": 8995
    },
    {
      "entropy": 5.95935206413269,
      "epoch": 0.7947721653126104,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000494383180344879,
      "loss": 5.7439,
      "mean_token_accuracy": 0.17004499435424805,
      "num_tokens": 16792618.0,
      "step": 9000
    },
    {
      "epoch": 0.7947721653126104,
      "eval_entropy": 5.774197348921917,
      "eval_loss": 5.872763156890869,
      "eval_mean_token_accuracy": 0.17264041615408,
      "eval_num_tokens": 16792618.0,
      "eval_runtime": 26.2257,
      "eval_samples_per_second": 1346.581,
      "eval_steps_per_second": 168.346,
      "step": 9000
    },
    {
      "entropy": 5.941266775131226,
      "epoch": 0.7952137054044507,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004943761856560898,
      "loss": 5.8412,
      "mean_token_accuracy": 0.1739298015832901,
      "num_tokens": 16801210.0,
      "step": 9005
    },
    {
      "entropy": 6.00039587020874,
      "epoch": 0.7956552454962911,
      "grad_norm": 1.59375,
      "learning_rate": 0.000494369186669846,
      "loss": 5.8748,
      "mean_token_accuracy": 0.16477510184049607,
      "num_tokens": 16810371.0,
      "step": 9010
    },
    {
      "entropy": 6.072319173812867,
      "epoch": 0.7960967855881314,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004943621833862844,
      "loss": 5.7852,
      "mean_token_accuracy": 0.16916722655296326,
      "num_tokens": 16818886.0,
      "step": 9015
    },
    {
      "entropy": 6.060644626617432,
      "epoch": 0.7965383256799717,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004943551758055425,
      "loss": 5.8459,
      "mean_token_accuracy": 0.1706278830766678,
      "num_tokens": 16827705.0,
      "step": 9020
    },
    {
      "entropy": 6.0029925346374515,
      "epoch": 0.7969798657718121,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004943481639277574,
      "loss": 5.8133,
      "mean_token_accuracy": 0.16445554345846175,
      "num_tokens": 16836313.0,
      "step": 9025
    },
    {
      "entropy": 5.99051866531372,
      "epoch": 0.7974214058636524,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004943411477530665,
      "loss": 5.9,
      "mean_token_accuracy": 0.15801046937704086,
      "num_tokens": 16845016.0,
      "step": 9030
    },
    {
      "entropy": 6.039083003997803,
      "epoch": 0.7978629459554928,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004943341272816073,
      "loss": 5.8104,
      "mean_token_accuracy": 0.16541790813207627,
      "num_tokens": 16854025.0,
      "step": 9035
    },
    {
      "entropy": 6.015139532089234,
      "epoch": 0.7983044860473331,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004943271025135173,
      "loss": 5.7965,
      "mean_token_accuracy": 0.16925208270549774,
      "num_tokens": 16862859.0,
      "step": 9040
    },
    {
      "entropy": 5.987388277053833,
      "epoch": 0.7987460261391734,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004943200734489341,
      "loss": 5.9064,
      "mean_token_accuracy": 0.16620191037654877,
      "num_tokens": 16872226.0,
      "step": 9045
    },
    {
      "entropy": 5.935065841674804,
      "epoch": 0.7991875662310137,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004943130400879953,
      "loss": 5.7871,
      "mean_token_accuracy": 0.17395144551992417,
      "num_tokens": 16881977.0,
      "step": 9050
    },
    {
      "entropy": 6.003419685363769,
      "epoch": 0.7996291063228541,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004943060024308389,
      "loss": 5.879,
      "mean_token_accuracy": 0.16055256724357606,
      "num_tokens": 16891656.0,
      "step": 9055
    },
    {
      "entropy": 6.008896398544311,
      "epoch": 0.8000706464146945,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004942989604776026,
      "loss": 5.7544,
      "mean_token_accuracy": 0.1753992348909378,
      "num_tokens": 16900944.0,
      "step": 9060
    },
    {
      "entropy": 5.880461978912353,
      "epoch": 0.8005121865065348,
      "grad_norm": 2.0,
      "learning_rate": 0.0004942919142284244,
      "loss": 5.7738,
      "mean_token_accuracy": 0.17210271209478378,
      "num_tokens": 16910154.0,
      "step": 9065
    },
    {
      "entropy": 6.027874279022217,
      "epoch": 0.8009537265983752,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004942848636834423,
      "loss": 5.9076,
      "mean_token_accuracy": 0.16426340490579605,
      "num_tokens": 16919694.0,
      "step": 9070
    },
    {
      "entropy": 6.09167332649231,
      "epoch": 0.8013952666902154,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004942778088427944,
      "loss": 5.8551,
      "mean_token_accuracy": 0.16622861176729203,
      "num_tokens": 16928530.0,
      "step": 9075
    },
    {
      "entropy": 6.067292928695679,
      "epoch": 0.8018368067820558,
      "grad_norm": 1.375,
      "learning_rate": 0.000494270749706619,
      "loss": 5.8796,
      "mean_token_accuracy": 0.16309232711791993,
      "num_tokens": 16938614.0,
      "step": 9080
    },
    {
      "entropy": 6.004611015319824,
      "epoch": 0.8022783468738961,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004942636862750544,
      "loss": 5.8232,
      "mean_token_accuracy": 0.1672021970152855,
      "num_tokens": 16947655.0,
      "step": 9085
    },
    {
      "entropy": 6.013993453979492,
      "epoch": 0.8027198869657365,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004942566185482387,
      "loss": 5.9213,
      "mean_token_accuracy": 0.1601979151368141,
      "num_tokens": 16957101.0,
      "step": 9090
    },
    {
      "entropy": 6.012250661849976,
      "epoch": 0.8031614270575769,
      "grad_norm": 10.5,
      "learning_rate": 0.0004942495465263106,
      "loss": 5.9408,
      "mean_token_accuracy": 0.15858723670244218,
      "num_tokens": 16966421.0,
      "step": 9095
    },
    {
      "entropy": 6.113315486907959,
      "epoch": 0.8036029671494171,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004942424702094085,
      "loss": 5.841,
      "mean_token_accuracy": 0.1748947262763977,
      "num_tokens": 16975656.0,
      "step": 9100
    },
    {
      "entropy": 6.038866281509399,
      "epoch": 0.8040445072412575,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004942353895976713,
      "loss": 5.8954,
      "mean_token_accuracy": 0.16503171473741532,
      "num_tokens": 16984409.0,
      "step": 9105
    },
    {
      "entropy": 5.946926879882812,
      "epoch": 0.8044860473330978,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004942283046912374,
      "loss": 5.7914,
      "mean_token_accuracy": 0.1615387961268425,
      "num_tokens": 16992728.0,
      "step": 9110
    },
    {
      "entropy": 5.988664960861206,
      "epoch": 0.8049275874249382,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004942212154902456,
      "loss": 5.7094,
      "mean_token_accuracy": 0.17891569435596466,
      "num_tokens": 17001616.0,
      "step": 9115
    },
    {
      "entropy": 6.016485023498535,
      "epoch": 0.8053691275167785,
      "grad_norm": 1.609375,
      "learning_rate": 0.000494214121994835,
      "loss": 5.8815,
      "mean_token_accuracy": 0.1644422948360443,
      "num_tokens": 17010655.0,
      "step": 9120
    },
    {
      "entropy": 5.981185865402222,
      "epoch": 0.8058106676086189,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004942070242051444,
      "loss": 5.8206,
      "mean_token_accuracy": 0.16579540222883224,
      "num_tokens": 17019819.0,
      "step": 9125
    },
    {
      "entropy": 6.017875623703003,
      "epoch": 0.8062522077004592,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004941999221213129,
      "loss": 5.869,
      "mean_token_accuracy": 0.16917242109775543,
      "num_tokens": 17029218.0,
      "step": 9130
    },
    {
      "entropy": 6.040101957321167,
      "epoch": 0.8066937477922995,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004941928157434796,
      "loss": 5.7759,
      "mean_token_accuracy": 0.1664850026369095,
      "num_tokens": 17037880.0,
      "step": 9135
    },
    {
      "entropy": 5.899328088760376,
      "epoch": 0.8071352878841399,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004941857050717836,
      "loss": 5.8518,
      "mean_token_accuracy": 0.16521087139844895,
      "num_tokens": 17047934.0,
      "step": 9140
    },
    {
      "entropy": 6.061554431915283,
      "epoch": 0.8075768279759802,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004941785901063642,
      "loss": 5.9161,
      "mean_token_accuracy": 0.16436673253774642,
      "num_tokens": 17057258.0,
      "step": 9145
    },
    {
      "entropy": 6.060186815261841,
      "epoch": 0.8080183680678206,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004941714708473611,
      "loss": 5.8742,
      "mean_token_accuracy": 0.16407700031995773,
      "num_tokens": 17066387.0,
      "step": 9150
    },
    {
      "entropy": 5.970388793945313,
      "epoch": 0.8084599081596608,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004941643472949134,
      "loss": 5.8069,
      "mean_token_accuracy": 0.16943138539791108,
      "num_tokens": 17075952.0,
      "step": 9155
    },
    {
      "entropy": 5.959314489364624,
      "epoch": 0.8089014482515012,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004941572194491608,
      "loss": 5.6841,
      "mean_token_accuracy": 0.17572013437747955,
      "num_tokens": 17084209.0,
      "step": 9160
    },
    {
      "entropy": 5.960428619384766,
      "epoch": 0.8093429883433416,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000494150087310243,
      "loss": 5.903,
      "mean_token_accuracy": 0.16461869776248933,
      "num_tokens": 17093395.0,
      "step": 9165
    },
    {
      "entropy": 6.011094474792481,
      "epoch": 0.8097845284351819,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004941429508782995,
      "loss": 5.8087,
      "mean_token_accuracy": 0.16933547407388688,
      "num_tokens": 17102646.0,
      "step": 9170
    },
    {
      "entropy": 6.0214849472045895,
      "epoch": 0.8102260685270223,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004941358101534703,
      "loss": 5.8163,
      "mean_token_accuracy": 0.1657823845744133,
      "num_tokens": 17111530.0,
      "step": 9175
    },
    {
      "entropy": 5.9327880859375,
      "epoch": 0.8106676086188626,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004941286651358952,
      "loss": 5.7754,
      "mean_token_accuracy": 0.17723568230867387,
      "num_tokens": 17119659.0,
      "step": 9180
    },
    {
      "entropy": 5.851162385940552,
      "epoch": 0.8111091487107029,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004941215158257142,
      "loss": 5.7937,
      "mean_token_accuracy": 0.16790302991867065,
      "num_tokens": 17127931.0,
      "step": 9185
    },
    {
      "entropy": 5.973969125747681,
      "epoch": 0.8115506888025432,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004941143622230672,
      "loss": 5.8164,
      "mean_token_accuracy": 0.17351385802030564,
      "num_tokens": 17137374.0,
      "step": 9190
    },
    {
      "entropy": 6.0743568420410154,
      "epoch": 0.8119922288943836,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004941072043280946,
      "loss": 5.885,
      "mean_token_accuracy": 0.16001378148794174,
      "num_tokens": 17147257.0,
      "step": 9195
    },
    {
      "entropy": 6.005939292907715,
      "epoch": 0.812433768986224,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004941000421409365,
      "loss": 5.7516,
      "mean_token_accuracy": 0.17450112253427505,
      "num_tokens": 17156233.0,
      "step": 9200
    },
    {
      "entropy": 6.042790842056275,
      "epoch": 0.8128753090780643,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004940928756617331,
      "loss": 5.9396,
      "mean_token_accuracy": 0.15833428949117662,
      "num_tokens": 17166364.0,
      "step": 9205
    },
    {
      "entropy": 5.958063983917237,
      "epoch": 0.8133168491699047,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004940857048906248,
      "loss": 5.8384,
      "mean_token_accuracy": 0.1704106017947197,
      "num_tokens": 17175853.0,
      "step": 9210
    },
    {
      "entropy": 6.029957246780396,
      "epoch": 0.8137583892617449,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004940785298277523,
      "loss": 5.9398,
      "mean_token_accuracy": 0.1621903046965599,
      "num_tokens": 17187849.0,
      "step": 9215
    },
    {
      "entropy": 6.07569317817688,
      "epoch": 0.8141999293535853,
      "grad_norm": 1.65625,
      "learning_rate": 0.000494071350473256,
      "loss": 5.8259,
      "mean_token_accuracy": 0.1650031879544258,
      "num_tokens": 17196646.0,
      "step": 9220
    },
    {
      "entropy": 6.061435317993164,
      "epoch": 0.8146414694454256,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004940641668272765,
      "loss": 5.9578,
      "mean_token_accuracy": 0.16093774735927582,
      "num_tokens": 17205919.0,
      "step": 9225
    },
    {
      "entropy": 6.044738626480102,
      "epoch": 0.815083009537266,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004940569788899544,
      "loss": 5.9068,
      "mean_token_accuracy": 0.16127809137105942,
      "num_tokens": 17215085.0,
      "step": 9230
    },
    {
      "entropy": 5.978901338577271,
      "epoch": 0.8155245496291064,
      "grad_norm": 2.0,
      "learning_rate": 0.0004940497866614309,
      "loss": 5.886,
      "mean_token_accuracy": 0.15986076593399048,
      "num_tokens": 17223633.0,
      "step": 9235
    },
    {
      "entropy": 6.01971402168274,
      "epoch": 0.8159660897209466,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004940425901418465,
      "loss": 5.7953,
      "mean_token_accuracy": 0.1701249048113823,
      "num_tokens": 17232915.0,
      "step": 9240
    },
    {
      "entropy": 6.049149703979492,
      "epoch": 0.816407629812787,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004940353893313423,
      "loss": 5.9239,
      "mean_token_accuracy": 0.1602263942360878,
      "num_tokens": 17242392.0,
      "step": 9245
    },
    {
      "entropy": 6.114197826385498,
      "epoch": 0.8168491699046273,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004940281842300596,
      "loss": 5.9946,
      "mean_token_accuracy": 0.15723732411861419,
      "num_tokens": 17252259.0,
      "step": 9250
    },
    {
      "entropy": 6.034798717498779,
      "epoch": 0.8172907099964677,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004940209748381392,
      "loss": 5.8621,
      "mean_token_accuracy": 0.16262195855379105,
      "num_tokens": 17261584.0,
      "step": 9255
    },
    {
      "entropy": 6.006899690628051,
      "epoch": 0.817732250088308,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004940137611557225,
      "loss": 5.8059,
      "mean_token_accuracy": 0.16907447427511216,
      "num_tokens": 17270901.0,
      "step": 9260
    },
    {
      "entropy": 6.030629062652588,
      "epoch": 0.8181737901801484,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004940065431829508,
      "loss": 5.8586,
      "mean_token_accuracy": 0.16188233494758605,
      "num_tokens": 17280024.0,
      "step": 9265
    },
    {
      "entropy": 6.03720874786377,
      "epoch": 0.8186153302719887,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004939993209199654,
      "loss": 5.9058,
      "mean_token_accuracy": 0.16143932938575745,
      "num_tokens": 17290131.0,
      "step": 9270
    },
    {
      "entropy": 6.105310440063477,
      "epoch": 0.819056870363829,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004939920943669079,
      "loss": 5.9044,
      "mean_token_accuracy": 0.1629559114575386,
      "num_tokens": 17299453.0,
      "step": 9275
    },
    {
      "entropy": 6.005107450485229,
      "epoch": 0.8194984104556694,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004939848635239198,
      "loss": 5.8754,
      "mean_token_accuracy": 0.163715136051178,
      "num_tokens": 17309344.0,
      "step": 9280
    },
    {
      "entropy": 6.007379484176636,
      "epoch": 0.8199399505475097,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004939776283911429,
      "loss": 5.7864,
      "mean_token_accuracy": 0.17647817730903625,
      "num_tokens": 17318636.0,
      "step": 9285
    },
    {
      "entropy": 5.983161783218383,
      "epoch": 0.8203814906393501,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004939703889687186,
      "loss": 5.8451,
      "mean_token_accuracy": 0.16662515699863434,
      "num_tokens": 17328146.0,
      "step": 9290
    },
    {
      "entropy": 5.976247024536133,
      "epoch": 0.8208230307311903,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004939631452567891,
      "loss": 5.9131,
      "mean_token_accuracy": 0.16368812918663025,
      "num_tokens": 17338542.0,
      "step": 9295
    },
    {
      "entropy": 6.122719049453735,
      "epoch": 0.8212645708230307,
      "grad_norm": 2.03125,
      "learning_rate": 0.000493955897255496,
      "loss": 5.8842,
      "mean_token_accuracy": 0.16635640561580659,
      "num_tokens": 17347577.0,
      "step": 9300
    },
    {
      "entropy": 6.022024011611938,
      "epoch": 0.8217061109148711,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004939486449649814,
      "loss": 5.8624,
      "mean_token_accuracy": 0.16558304727077483,
      "num_tokens": 17356554.0,
      "step": 9305
    },
    {
      "entropy": 5.960217761993408,
      "epoch": 0.8221476510067114,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004939413883853873,
      "loss": 5.8814,
      "mean_token_accuracy": 0.1689998045563698,
      "num_tokens": 17366432.0,
      "step": 9310
    },
    {
      "entropy": 6.087877368927002,
      "epoch": 0.8225891910985518,
      "grad_norm": 1.78125,
      "learning_rate": 0.000493934127516856,
      "loss": 5.8458,
      "mean_token_accuracy": 0.1692266508936882,
      "num_tokens": 17376556.0,
      "step": 9315
    },
    {
      "entropy": 6.018303537368775,
      "epoch": 0.823030731190392,
      "grad_norm": 1.875,
      "learning_rate": 0.0004939268623595297,
      "loss": 5.8919,
      "mean_token_accuracy": 0.15746195390820503,
      "num_tokens": 17385382.0,
      "step": 9320
    },
    {
      "entropy": 5.992043781280517,
      "epoch": 0.8234722712822324,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004939195929135506,
      "loss": 5.7691,
      "mean_token_accuracy": 0.1613371714949608,
      "num_tokens": 17393229.0,
      "step": 9325
    },
    {
      "entropy": 5.988037776947022,
      "epoch": 0.8239138113740727,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004939123191790612,
      "loss": 5.7709,
      "mean_token_accuracy": 0.16014604717493058,
      "num_tokens": 17402439.0,
      "step": 9330
    },
    {
      "entropy": 5.987378358840942,
      "epoch": 0.8243553514659131,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004939050411562041,
      "loss": 5.9294,
      "mean_token_accuracy": 0.15827521085739135,
      "num_tokens": 17412349.0,
      "step": 9335
    },
    {
      "entropy": 5.954100561141968,
      "epoch": 0.8247968915577535,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004938977588451216,
      "loss": 5.6967,
      "mean_token_accuracy": 0.17938645631074907,
      "num_tokens": 17420842.0,
      "step": 9340
    },
    {
      "entropy": 6.057862091064453,
      "epoch": 0.8252384316495938,
      "grad_norm": 2.75,
      "learning_rate": 0.0004938904722459565,
      "loss": 5.7685,
      "mean_token_accuracy": 0.1707105204463005,
      "num_tokens": 17429877.0,
      "step": 9345
    },
    {
      "entropy": 6.008994388580322,
      "epoch": 0.8256799717414341,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004938831813588515,
      "loss": 5.8971,
      "mean_token_accuracy": 0.1633854404091835,
      "num_tokens": 17439508.0,
      "step": 9350
    },
    {
      "entropy": 5.926471376419068,
      "epoch": 0.8261215118332744,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004938758861839495,
      "loss": 5.8169,
      "mean_token_accuracy": 0.16402943134307862,
      "num_tokens": 17449004.0,
      "step": 9355
    },
    {
      "entropy": 5.991986894607544,
      "epoch": 0.8265630519251148,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004938685867213934,
      "loss": 5.8272,
      "mean_token_accuracy": 0.1690111994743347,
      "num_tokens": 17458462.0,
      "step": 9360
    },
    {
      "entropy": 6.053527164459228,
      "epoch": 0.8270045920169551,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004938612829713262,
      "loss": 5.8479,
      "mean_token_accuracy": 0.16346956491470338,
      "num_tokens": 17466808.0,
      "step": 9365
    },
    {
      "entropy": 6.0850182056427,
      "epoch": 0.8274461321087955,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004938539749338909,
      "loss": 5.8984,
      "mean_token_accuracy": 0.1645541086792946,
      "num_tokens": 17476441.0,
      "step": 9370
    },
    {
      "entropy": 6.065642690658569,
      "epoch": 0.8278876722006359,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004938466626092308,
      "loss": 5.8183,
      "mean_token_accuracy": 0.1734413832426071,
      "num_tokens": 17485393.0,
      "step": 9375
    },
    {
      "entropy": 5.940054988861084,
      "epoch": 0.8283292122924761,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000493839345997489,
      "loss": 5.8528,
      "mean_token_accuracy": 0.16922394186258316,
      "num_tokens": 17494904.0,
      "step": 9380
    },
    {
      "entropy": 6.036320209503174,
      "epoch": 0.8287707523843165,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004938320250988087,
      "loss": 5.8517,
      "mean_token_accuracy": 0.1592379555106163,
      "num_tokens": 17505137.0,
      "step": 9385
    },
    {
      "entropy": 6.0248651027679445,
      "epoch": 0.8292122924761568,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004938246999133337,
      "loss": 5.9001,
      "mean_token_accuracy": 0.1642383113503456,
      "num_tokens": 17514039.0,
      "step": 9390
    },
    {
      "entropy": 5.958837652206421,
      "epoch": 0.8296538325679972,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004938173704412071,
      "loss": 5.7857,
      "mean_token_accuracy": 0.1705437883734703,
      "num_tokens": 17522423.0,
      "step": 9395
    },
    {
      "entropy": 5.987477111816406,
      "epoch": 0.8300953726598375,
      "grad_norm": 2.125,
      "learning_rate": 0.0004938100366825728,
      "loss": 5.7645,
      "mean_token_accuracy": 0.17269083708524705,
      "num_tokens": 17531958.0,
      "step": 9400
    },
    {
      "entropy": 6.023804759979248,
      "epoch": 0.8305369127516778,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004938026986375742,
      "loss": 5.8286,
      "mean_token_accuracy": 0.1699989214539528,
      "num_tokens": 17540905.0,
      "step": 9405
    },
    {
      "entropy": 5.9978249073028564,
      "epoch": 0.8309784528435182,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004937953563063553,
      "loss": 5.8802,
      "mean_token_accuracy": 0.1642656847834587,
      "num_tokens": 17551007.0,
      "step": 9410
    },
    {
      "entropy": 5.944204664230346,
      "epoch": 0.8314199929353585,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004937880096890598,
      "loss": 5.7058,
      "mean_token_accuracy": 0.17777763903141022,
      "num_tokens": 17559403.0,
      "step": 9415
    },
    {
      "entropy": 5.989461374282837,
      "epoch": 0.8318615330271989,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004937806587858315,
      "loss": 5.8403,
      "mean_token_accuracy": 0.17147947996854782,
      "num_tokens": 17569191.0,
      "step": 9420
    },
    {
      "entropy": 6.078926372528076,
      "epoch": 0.8323030731190392,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004937733035968147,
      "loss": 5.9022,
      "mean_token_accuracy": 0.15950139611959457,
      "num_tokens": 17578538.0,
      "step": 9425
    },
    {
      "entropy": 5.990520238876343,
      "epoch": 0.8327446132108796,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004937659441221531,
      "loss": 5.794,
      "mean_token_accuracy": 0.17096205651760102,
      "num_tokens": 17587344.0,
      "step": 9430
    },
    {
      "entropy": 5.9082780361175535,
      "epoch": 0.8331861533027198,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004937585803619912,
      "loss": 5.7814,
      "mean_token_accuracy": 0.17017468810081482,
      "num_tokens": 17596033.0,
      "step": 9435
    },
    {
      "entropy": 6.001602983474731,
      "epoch": 0.8336276933945602,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004937512123164731,
      "loss": 5.7366,
      "mean_token_accuracy": 0.1792237713932991,
      "num_tokens": 17605860.0,
      "step": 9440
    },
    {
      "entropy": 6.037835311889649,
      "epoch": 0.8340692334864006,
      "grad_norm": 1.875,
      "learning_rate": 0.0004937438399857431,
      "loss": 5.8755,
      "mean_token_accuracy": 0.16186862140893937,
      "num_tokens": 17615819.0,
      "step": 9445
    },
    {
      "entropy": 6.026955842971802,
      "epoch": 0.8345107735782409,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004937364633699459,
      "loss": 5.8441,
      "mean_token_accuracy": 0.16839220821857454,
      "num_tokens": 17625478.0,
      "step": 9450
    },
    {
      "entropy": 6.022288179397583,
      "epoch": 0.8349523136700813,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004937290824692255,
      "loss": 5.8191,
      "mean_token_accuracy": 0.1682550862431526,
      "num_tokens": 17635329.0,
      "step": 9455
    },
    {
      "entropy": 5.94835467338562,
      "epoch": 0.8353938537619215,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004937216972837269,
      "loss": 5.8152,
      "mean_token_accuracy": 0.16099725067615508,
      "num_tokens": 17644129.0,
      "step": 9460
    },
    {
      "entropy": 5.91757493019104,
      "epoch": 0.8358353938537619,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004937143078135946,
      "loss": 5.7547,
      "mean_token_accuracy": 0.17431456297636033,
      "num_tokens": 17652836.0,
      "step": 9465
    },
    {
      "entropy": 6.034531450271606,
      "epoch": 0.8362769339456022,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004937069140589735,
      "loss": 5.7935,
      "mean_token_accuracy": 0.16793447434902192,
      "num_tokens": 17661222.0,
      "step": 9470
    },
    {
      "entropy": 5.951354742050171,
      "epoch": 0.8367184740374426,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004936995160200083,
      "loss": 5.7551,
      "mean_token_accuracy": 0.17580796182155609,
      "num_tokens": 17669637.0,
      "step": 9475
    },
    {
      "entropy": 6.0513585090637205,
      "epoch": 0.837160014129283,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004936921136968439,
      "loss": 5.9381,
      "mean_token_accuracy": 0.15244681164622306,
      "num_tokens": 17678589.0,
      "step": 9480
    },
    {
      "entropy": 6.065783834457397,
      "epoch": 0.8376015542211233,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004936847070896254,
      "loss": 5.9067,
      "mean_token_accuracy": 0.16115776300430298,
      "num_tokens": 17689129.0,
      "step": 9485
    },
    {
      "entropy": 6.0849708080291744,
      "epoch": 0.8380430943129636,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004936772961984979,
      "loss": 5.9027,
      "mean_token_accuracy": 0.1689728707075119,
      "num_tokens": 17698864.0,
      "step": 9490
    },
    {
      "entropy": 6.058166265487671,
      "epoch": 0.8384846344048039,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004936698810236065,
      "loss": 5.9882,
      "mean_token_accuracy": 0.15745993703603745,
      "num_tokens": 17707900.0,
      "step": 9495
    },
    {
      "entropy": 6.042195796966553,
      "epoch": 0.8389261744966443,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004936624615650964,
      "loss": 5.808,
      "mean_token_accuracy": 0.17014045864343644,
      "num_tokens": 17716983.0,
      "step": 9500
    },
    {
      "entropy": 5.999377870559693,
      "epoch": 0.8393677145884846,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004936550378231133,
      "loss": 5.8048,
      "mean_token_accuracy": 0.16574602127075194,
      "num_tokens": 17726503.0,
      "step": 9505
    },
    {
      "entropy": 6.001359987258911,
      "epoch": 0.839809254680325,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004936476097978023,
      "loss": 5.8939,
      "mean_token_accuracy": 0.1637062907218933,
      "num_tokens": 17735708.0,
      "step": 9510
    },
    {
      "entropy": 5.994287729263306,
      "epoch": 0.8402507947721654,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004936401774893088,
      "loss": 5.8038,
      "mean_token_accuracy": 0.16697103083133696,
      "num_tokens": 17745169.0,
      "step": 9515
    },
    {
      "entropy": 6.0431797981262205,
      "epoch": 0.8406923348640056,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004936327408977786,
      "loss": 5.9432,
      "mean_token_accuracy": 0.1595461219549179,
      "num_tokens": 17755630.0,
      "step": 9520
    },
    {
      "entropy": 6.0142663478851315,
      "epoch": 0.841133874955846,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004936253000233574,
      "loss": 5.8283,
      "mean_token_accuracy": 0.17293956726789475,
      "num_tokens": 17764469.0,
      "step": 9525
    },
    {
      "entropy": 5.962830114364624,
      "epoch": 0.8415754150476863,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004936178548661909,
      "loss": 5.9207,
      "mean_token_accuracy": 0.1591649293899536,
      "num_tokens": 17774977.0,
      "step": 9530
    },
    {
      "entropy": 6.046345520019531,
      "epoch": 0.8420169551395267,
      "grad_norm": 1.40625,
      "learning_rate": 0.000493610405426425,
      "loss": 5.8196,
      "mean_token_accuracy": 0.16513993442058564,
      "num_tokens": 17784250.0,
      "step": 9535
    },
    {
      "entropy": 6.038363361358643,
      "epoch": 0.842458495231367,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004936029517042056,
      "loss": 5.8518,
      "mean_token_accuracy": 0.16764956265687941,
      "num_tokens": 17793858.0,
      "step": 9540
    },
    {
      "entropy": 6.0372391700744625,
      "epoch": 0.8429000353232073,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004935954936996787,
      "loss": 5.9554,
      "mean_token_accuracy": 0.16116243451833726,
      "num_tokens": 17804330.0,
      "step": 9545
    },
    {
      "entropy": 6.033014535903931,
      "epoch": 0.8433415754150477,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004935880314129903,
      "loss": 5.8408,
      "mean_token_accuracy": 0.17326375991106033,
      "num_tokens": 17813750.0,
      "step": 9550
    },
    {
      "entropy": 5.969830417633057,
      "epoch": 0.843783115506888,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004935805648442868,
      "loss": 5.8038,
      "mean_token_accuracy": 0.167977911233902,
      "num_tokens": 17823537.0,
      "step": 9555
    },
    {
      "entropy": 5.982773494720459,
      "epoch": 0.8442246555987284,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004935730939937143,
      "loss": 5.8136,
      "mean_token_accuracy": 0.17517689615488052,
      "num_tokens": 17833298.0,
      "step": 9560
    },
    {
      "entropy": 5.976465034484863,
      "epoch": 0.8446661956905687,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004935656188614192,
      "loss": 5.737,
      "mean_token_accuracy": 0.1733606703579426,
      "num_tokens": 17842339.0,
      "step": 9565
    },
    {
      "entropy": 6.008795309066772,
      "epoch": 0.845107735782409,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004935581394475479,
      "loss": 5.9289,
      "mean_token_accuracy": 0.16886893808841705,
      "num_tokens": 17851827.0,
      "step": 9570
    },
    {
      "entropy": 5.95584626197815,
      "epoch": 0.8455492758742493,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004935506557522469,
      "loss": 5.7982,
      "mean_token_accuracy": 0.17455206960439681,
      "num_tokens": 17860788.0,
      "step": 9575
    },
    {
      "entropy": 6.097922229766846,
      "epoch": 0.8459908159660897,
      "grad_norm": 2.0,
      "learning_rate": 0.000493543167775663,
      "loss": 5.7847,
      "mean_token_accuracy": 0.1724754050374031,
      "num_tokens": 17869849.0,
      "step": 9580
    },
    {
      "entropy": 6.013970279693604,
      "epoch": 0.8464323560579301,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004935356755179426,
      "loss": 5.8191,
      "mean_token_accuracy": 0.16670462042093276,
      "num_tokens": 17879431.0,
      "step": 9585
    },
    {
      "entropy": 6.042088985443115,
      "epoch": 0.8468738961497704,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004935281789792326,
      "loss": 5.9476,
      "mean_token_accuracy": 0.16087115854024886,
      "num_tokens": 17888599.0,
      "step": 9590
    },
    {
      "entropy": 5.985098457336425,
      "epoch": 0.8473154362416108,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004935206781596799,
      "loss": 5.8266,
      "mean_token_accuracy": 0.17410103529691695,
      "num_tokens": 17898662.0,
      "step": 9595
    },
    {
      "entropy": 6.038266277313232,
      "epoch": 0.847756976333451,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004935131730594315,
      "loss": 5.8564,
      "mean_token_accuracy": 0.16674958169460297,
      "num_tokens": 17907810.0,
      "step": 9600
    },
    {
      "entropy": 5.981097459793091,
      "epoch": 0.8481985164252914,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004935056636786342,
      "loss": 5.8282,
      "mean_token_accuracy": 0.16734828054904938,
      "num_tokens": 17917611.0,
      "step": 9605
    },
    {
      "entropy": 5.966639995574951,
      "epoch": 0.8486400565171317,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004934981500174353,
      "loss": 5.8433,
      "mean_token_accuracy": 0.16487517058849335,
      "num_tokens": 17926076.0,
      "step": 9610
    },
    {
      "entropy": 6.038370132446289,
      "epoch": 0.8490815966089721,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004934906320759818,
      "loss": 5.8761,
      "mean_token_accuracy": 0.16527412384748458,
      "num_tokens": 17935521.0,
      "step": 9615
    },
    {
      "entropy": 6.072547054290771,
      "epoch": 0.8495231367008125,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004934831098544211,
      "loss": 5.8459,
      "mean_token_accuracy": 0.17391809970140457,
      "num_tokens": 17944849.0,
      "step": 9620
    },
    {
      "entropy": 6.06336669921875,
      "epoch": 0.8499646767926528,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004934755833529007,
      "loss": 5.8252,
      "mean_token_accuracy": 0.16894840747117995,
      "num_tokens": 17953988.0,
      "step": 9625
    },
    {
      "entropy": 5.969831466674805,
      "epoch": 0.8504062168844931,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004934680525715677,
      "loss": 5.8247,
      "mean_token_accuracy": 0.1663273498415947,
      "num_tokens": 17963319.0,
      "step": 9630
    },
    {
      "entropy": 6.004313802719116,
      "epoch": 0.8508477569763334,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004934605175105698,
      "loss": 5.7655,
      "mean_token_accuracy": 0.1794113963842392,
      "num_tokens": 17972054.0,
      "step": 9635
    },
    {
      "entropy": 6.0897088050842285,
      "epoch": 0.8512892970681738,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004934529781700546,
      "loss": 5.9241,
      "mean_token_accuracy": 0.16022397577762604,
      "num_tokens": 17981630.0,
      "step": 9640
    },
    {
      "entropy": 6.045695877075195,
      "epoch": 0.8517308371600141,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004934454345501698,
      "loss": 5.8614,
      "mean_token_accuracy": 0.16092253774404525,
      "num_tokens": 17990987.0,
      "step": 9645
    },
    {
      "entropy": 5.906431531906128,
      "epoch": 0.8521723772518545,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004934378866510633,
      "loss": 5.764,
      "mean_token_accuracy": 0.17493847757577896,
      "num_tokens": 17999284.0,
      "step": 9650
    },
    {
      "entropy": 5.998637628555298,
      "epoch": 0.8526139173436948,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004934303344728828,
      "loss": 5.7818,
      "mean_token_accuracy": 0.16770309656858445,
      "num_tokens": 18008670.0,
      "step": 9655
    },
    {
      "entropy": 5.967482423782348,
      "epoch": 0.8530554574355351,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004934227780157761,
      "loss": 5.7853,
      "mean_token_accuracy": 0.17448310256004335,
      "num_tokens": 18018374.0,
      "step": 9660
    },
    {
      "entropy": 6.040644407272339,
      "epoch": 0.8534969975273755,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004934152172798916,
      "loss": 5.8803,
      "mean_token_accuracy": 0.1725458726286888,
      "num_tokens": 18026943.0,
      "step": 9665
    },
    {
      "entropy": 5.996393537521362,
      "epoch": 0.8539385376192158,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004934076522653772,
      "loss": 5.8137,
      "mean_token_accuracy": 0.17032251507043839,
      "num_tokens": 18036270.0,
      "step": 9670
    },
    {
      "entropy": 5.9290753364562985,
      "epoch": 0.8543800777110562,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000493400082972381,
      "loss": 5.8042,
      "mean_token_accuracy": 0.16725092232227326,
      "num_tokens": 18045611.0,
      "step": 9675
    },
    {
      "entropy": 5.9702893733978275,
      "epoch": 0.8548216178028964,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004933925094010516,
      "loss": 5.8499,
      "mean_token_accuracy": 0.16466162502765655,
      "num_tokens": 18055152.0,
      "step": 9680
    },
    {
      "entropy": 6.040339088439941,
      "epoch": 0.8552631578947368,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004933849315515369,
      "loss": 5.8251,
      "mean_token_accuracy": 0.16667694002389907,
      "num_tokens": 18064376.0,
      "step": 9685
    },
    {
      "entropy": 5.933894348144531,
      "epoch": 0.8557046979865772,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004933773494239857,
      "loss": 5.742,
      "mean_token_accuracy": 0.17865219563245774,
      "num_tokens": 18073569.0,
      "step": 9690
    },
    {
      "entropy": 5.9812424659729,
      "epoch": 0.8561462380784175,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004933697630185464,
      "loss": 5.8365,
      "mean_token_accuracy": 0.1723334863781929,
      "num_tokens": 18083264.0,
      "step": 9695
    },
    {
      "entropy": 6.035292339324951,
      "epoch": 0.8565877781702579,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004933621723353677,
      "loss": 5.8311,
      "mean_token_accuracy": 0.16717321500182153,
      "num_tokens": 18093151.0,
      "step": 9700
    },
    {
      "entropy": 5.95260443687439,
      "epoch": 0.8570293182620982,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000493354577374598,
      "loss": 5.7354,
      "mean_token_accuracy": 0.1816457837820053,
      "num_tokens": 18102087.0,
      "step": 9705
    },
    {
      "entropy": 5.994460964202881,
      "epoch": 0.8574708583539385,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004933469781363865,
      "loss": 5.8703,
      "mean_token_accuracy": 0.16676534563302994,
      "num_tokens": 18111854.0,
      "step": 9710
    },
    {
      "entropy": 6.052928924560547,
      "epoch": 0.8579123984457788,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004933393746208819,
      "loss": 5.8892,
      "mean_token_accuracy": 0.16378687620162963,
      "num_tokens": 18120520.0,
      "step": 9715
    },
    {
      "entropy": 6.051138830184937,
      "epoch": 0.8583539385376192,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004933317668282331,
      "loss": 5.9426,
      "mean_token_accuracy": 0.16159225404262542,
      "num_tokens": 18130557.0,
      "step": 9720
    },
    {
      "entropy": 6.075818586349487,
      "epoch": 0.8587954786294596,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004933241547585891,
      "loss": 5.7244,
      "mean_token_accuracy": 0.17409752905368805,
      "num_tokens": 18139199.0,
      "step": 9725
    },
    {
      "entropy": 5.959064102172851,
      "epoch": 0.8592370187212999,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004933165384120991,
      "loss": 5.7982,
      "mean_token_accuracy": 0.16895681619644165,
      "num_tokens": 18150039.0,
      "step": 9730
    },
    {
      "entropy": 5.912999486923217,
      "epoch": 0.8596785588131403,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004933089177889124,
      "loss": 5.7993,
      "mean_token_accuracy": 0.16626023799180983,
      "num_tokens": 18159451.0,
      "step": 9735
    },
    {
      "entropy": 6.015667581558228,
      "epoch": 0.8601200989049805,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004933012928891781,
      "loss": 5.8742,
      "mean_token_accuracy": 0.16694561541080474,
      "num_tokens": 18169896.0,
      "step": 9740
    },
    {
      "entropy": 6.040274524688721,
      "epoch": 0.8605616389968209,
      "grad_norm": 1.625,
      "learning_rate": 0.0004932936637130456,
      "loss": 5.7726,
      "mean_token_accuracy": 0.17514863312244416,
      "num_tokens": 18177764.0,
      "step": 9745
    },
    {
      "entropy": 6.000326204299927,
      "epoch": 0.8610031790886612,
      "grad_norm": 1.75,
      "learning_rate": 0.0004932860302606644,
      "loss": 5.8506,
      "mean_token_accuracy": 0.1685821920633316,
      "num_tokens": 18186940.0,
      "step": 9750
    },
    {
      "entropy": 5.984031820297242,
      "epoch": 0.8614447191805016,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004932783925321838,
      "loss": 5.7285,
      "mean_token_accuracy": 0.17475786805152893,
      "num_tokens": 18195374.0,
      "step": 9755
    },
    {
      "entropy": 5.933980989456177,
      "epoch": 0.861886259272342,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000493270750527754,
      "loss": 5.8251,
      "mean_token_accuracy": 0.1692585453391075,
      "num_tokens": 18204779.0,
      "step": 9760
    },
    {
      "entropy": 5.94135684967041,
      "epoch": 0.8623277993641822,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004932631042475241,
      "loss": 5.7685,
      "mean_token_accuracy": 0.1741400569677353,
      "num_tokens": 18214099.0,
      "step": 9765
    },
    {
      "entropy": 5.960148763656616,
      "epoch": 0.8627693394560226,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004932554536916441,
      "loss": 5.8428,
      "mean_token_accuracy": 0.17174284309148788,
      "num_tokens": 18223190.0,
      "step": 9770
    },
    {
      "entropy": 6.038312864303589,
      "epoch": 0.8632108795478629,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000493247798860264,
      "loss": 5.8336,
      "mean_token_accuracy": 0.17261188477277756,
      "num_tokens": 18232380.0,
      "step": 9775
    },
    {
      "entropy": 6.0753703117370605,
      "epoch": 0.8636524196397033,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004932401397535337,
      "loss": 5.9178,
      "mean_token_accuracy": 0.1631627470254898,
      "num_tokens": 18242621.0,
      "step": 9780
    },
    {
      "entropy": 6.08184962272644,
      "epoch": 0.8640939597315436,
      "grad_norm": 2.1875,
      "learning_rate": 0.000493232476371603,
      "loss": 5.9481,
      "mean_token_accuracy": 0.1636492282152176,
      "num_tokens": 18253183.0,
      "step": 9785
    },
    {
      "entropy": 6.007740831375122,
      "epoch": 0.864535499823384,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004932248087146223,
      "loss": 5.7486,
      "mean_token_accuracy": 0.17018086612224578,
      "num_tokens": 18261967.0,
      "step": 9790
    },
    {
      "entropy": 5.943189096450806,
      "epoch": 0.8649770399152243,
      "grad_norm": 1.625,
      "learning_rate": 0.0004932171367827417,
      "loss": 5.8268,
      "mean_token_accuracy": 0.16698989272117615,
      "num_tokens": 18270902.0,
      "step": 9795
    },
    {
      "entropy": 5.992133235931396,
      "epoch": 0.8654185800070646,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004932094605761115,
      "loss": 5.8579,
      "mean_token_accuracy": 0.17358374893665313,
      "num_tokens": 18279387.0,
      "step": 9800
    },
    {
      "entropy": 5.937605714797973,
      "epoch": 0.865860120098905,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004932017800948821,
      "loss": 5.6603,
      "mean_token_accuracy": 0.1796998143196106,
      "num_tokens": 18288363.0,
      "step": 9805
    },
    {
      "entropy": 6.07945613861084,
      "epoch": 0.8663016601907453,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004931940953392038,
      "loss": 5.9728,
      "mean_token_accuracy": 0.15800556987524034,
      "num_tokens": 18297398.0,
      "step": 9810
    },
    {
      "entropy": 6.066021203994751,
      "epoch": 0.8667432002825857,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004931864063092274,
      "loss": 5.8072,
      "mean_token_accuracy": 0.1684715837240219,
      "num_tokens": 18306230.0,
      "step": 9815
    },
    {
      "entropy": 6.009411478042603,
      "epoch": 0.8671847403744259,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004931787130051034,
      "loss": 5.9412,
      "mean_token_accuracy": 0.1657043479382992,
      "num_tokens": 18316395.0,
      "step": 9820
    },
    {
      "entropy": 5.963434362411499,
      "epoch": 0.8676262804662663,
      "grad_norm": 1.5,
      "learning_rate": 0.0004931710154269824,
      "loss": 5.8386,
      "mean_token_accuracy": 0.1774977833032608,
      "num_tokens": 18326020.0,
      "step": 9825
    },
    {
      "entropy": 5.950678396224975,
      "epoch": 0.8680678205581067,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004931633135750154,
      "loss": 5.8482,
      "mean_token_accuracy": 0.1690017983317375,
      "num_tokens": 18335320.0,
      "step": 9830
    },
    {
      "entropy": 6.013969230651855,
      "epoch": 0.868509360649947,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004931556074493531,
      "loss": 5.8672,
      "mean_token_accuracy": 0.1664042592048645,
      "num_tokens": 18345137.0,
      "step": 9835
    },
    {
      "entropy": 6.006626081466675,
      "epoch": 0.8689509007417874,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004931478970501465,
      "loss": 5.8628,
      "mean_token_accuracy": 0.1689472183585167,
      "num_tokens": 18354289.0,
      "step": 9840
    },
    {
      "entropy": 6.098352861404419,
      "epoch": 0.8693924408336277,
      "grad_norm": 2.0,
      "learning_rate": 0.0004931401823775466,
      "loss": 5.9391,
      "mean_token_accuracy": 0.15878856778144837,
      "num_tokens": 18363701.0,
      "step": 9845
    },
    {
      "entropy": 6.120831251144409,
      "epoch": 0.869833980925468,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004931324634317047,
      "loss": 5.8467,
      "mean_token_accuracy": 0.16003623902797698,
      "num_tokens": 18373336.0,
      "step": 9850
    },
    {
      "entropy": 6.053362512588501,
      "epoch": 0.8702755210173083,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004931247402127719,
      "loss": 5.8407,
      "mean_token_accuracy": 0.16822702288627625,
      "num_tokens": 18383233.0,
      "step": 9855
    },
    {
      "entropy": 6.035002946853638,
      "epoch": 0.8707170611091487,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004931170127208995,
      "loss": 5.8865,
      "mean_token_accuracy": 0.16325117498636246,
      "num_tokens": 18393214.0,
      "step": 9860
    },
    {
      "entropy": 5.987590885162353,
      "epoch": 0.8711586012009891,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004931092809562388,
      "loss": 5.7788,
      "mean_token_accuracy": 0.1711319714784622,
      "num_tokens": 18401773.0,
      "step": 9865
    },
    {
      "entropy": 5.970164728164673,
      "epoch": 0.8716001412928294,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004931015449189414,
      "loss": 5.828,
      "mean_token_accuracy": 0.16339541971683502,
      "num_tokens": 18411748.0,
      "step": 9870
    },
    {
      "entropy": 6.00556640625,
      "epoch": 0.8720416813846698,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004930938046091587,
      "loss": 5.8756,
      "mean_token_accuracy": 0.16804593652486802,
      "num_tokens": 18421399.0,
      "step": 9875
    },
    {
      "entropy": 6.005237674713134,
      "epoch": 0.87248322147651,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004930860600270425,
      "loss": 5.8802,
      "mean_token_accuracy": 0.16329103857278823,
      "num_tokens": 18430052.0,
      "step": 9880
    },
    {
      "entropy": 6.0280194759368895,
      "epoch": 0.8729247615683504,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004930783111727443,
      "loss": 5.7106,
      "mean_token_accuracy": 0.1797136604785919,
      "num_tokens": 18438268.0,
      "step": 9885
    },
    {
      "entropy": 6.017422676086426,
      "epoch": 0.8733663016601907,
      "grad_norm": 1.90625,
      "learning_rate": 0.000493070558046416,
      "loss": 5.8469,
      "mean_token_accuracy": 0.16759492307901383,
      "num_tokens": 18446964.0,
      "step": 9890
    },
    {
      "entropy": 6.02519154548645,
      "epoch": 0.8738078417520311,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004930628006482097,
      "loss": 5.842,
      "mean_token_accuracy": 0.1586953066289425,
      "num_tokens": 18455701.0,
      "step": 9895
    },
    {
      "entropy": 5.978132152557373,
      "epoch": 0.8742493818438715,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004930550389782769,
      "loss": 5.8038,
      "mean_token_accuracy": 0.16593412458896636,
      "num_tokens": 18465635.0,
      "step": 9900
    },
    {
      "entropy": 5.958258533477784,
      "epoch": 0.8746909219357117,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004930472730367701,
      "loss": 5.7267,
      "mean_token_accuracy": 0.16820934116840364,
      "num_tokens": 18474833.0,
      "step": 9905
    },
    {
      "entropy": 6.048409032821655,
      "epoch": 0.8751324620275521,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004930395028238412,
      "loss": 5.8346,
      "mean_token_accuracy": 0.16550887376070023,
      "num_tokens": 18482987.0,
      "step": 9910
    },
    {
      "entropy": 5.955231666564941,
      "epoch": 0.8755740021193924,
      "grad_norm": 1.75,
      "learning_rate": 0.0004930317283396423,
      "loss": 5.7227,
      "mean_token_accuracy": 0.1746671810746193,
      "num_tokens": 18491274.0,
      "step": 9915
    },
    {
      "entropy": 5.987809944152832,
      "epoch": 0.8760155422112328,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000493023949584326,
      "loss": 5.8937,
      "mean_token_accuracy": 0.16021096110343933,
      "num_tokens": 18500811.0,
      "step": 9920
    },
    {
      "entropy": 5.892692232131958,
      "epoch": 0.8764570823030731,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004930161665580445,
      "loss": 5.6721,
      "mean_token_accuracy": 0.18408720344305038,
      "num_tokens": 18509475.0,
      "step": 9925
    },
    {
      "entropy": 6.078322219848633,
      "epoch": 0.8768986223949135,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004930083792609502,
      "loss": 5.8238,
      "mean_token_accuracy": 0.16401641070842743,
      "num_tokens": 18519410.0,
      "step": 9930
    },
    {
      "entropy": 5.8636486530303955,
      "epoch": 0.8773401624867538,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004930005876931958,
      "loss": 5.6611,
      "mean_token_accuracy": 0.18206960707902908,
      "num_tokens": 18529194.0,
      "step": 9935
    },
    {
      "entropy": 6.004857063293457,
      "epoch": 0.8777817025785941,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004929927918549339,
      "loss": 5.8735,
      "mean_token_accuracy": 0.15991197973489762,
      "num_tokens": 18539166.0,
      "step": 9940
    },
    {
      "entropy": 5.99868106842041,
      "epoch": 0.8782232426704345,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004929849917463171,
      "loss": 5.7955,
      "mean_token_accuracy": 0.17093802392482757,
      "num_tokens": 18548166.0,
      "step": 9945
    },
    {
      "entropy": 5.960470199584961,
      "epoch": 0.8786647827622748,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004929771873674984,
      "loss": 5.8056,
      "mean_token_accuracy": 0.16528980284929276,
      "num_tokens": 18557041.0,
      "step": 9950
    },
    {
      "entropy": 5.992664957046509,
      "epoch": 0.8791063228541152,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004929693787186305,
      "loss": 5.8483,
      "mean_token_accuracy": 0.1624258577823639,
      "num_tokens": 18566461.0,
      "step": 9955
    },
    {
      "entropy": 5.950482988357544,
      "epoch": 0.8795478629459554,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004929615657998664,
      "loss": 5.7632,
      "mean_token_accuracy": 0.1686630889773369,
      "num_tokens": 18576334.0,
      "step": 9960
    },
    {
      "entropy": 5.985661172866822,
      "epoch": 0.8799894030377958,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004929537486113593,
      "loss": 5.7676,
      "mean_token_accuracy": 0.16868792176246644,
      "num_tokens": 18584882.0,
      "step": 9965
    },
    {
      "entropy": 6.04607048034668,
      "epoch": 0.8804309431296362,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004929459271532621,
      "loss": 5.8249,
      "mean_token_accuracy": 0.17223169654607773,
      "num_tokens": 18593669.0,
      "step": 9970
    },
    {
      "entropy": 6.0269121646881105,
      "epoch": 0.8808724832214765,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004929381014257282,
      "loss": 5.8604,
      "mean_token_accuracy": 0.16619434505701064,
      "num_tokens": 18603357.0,
      "step": 9975
    },
    {
      "entropy": 6.008830833435058,
      "epoch": 0.8813140233133169,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004929302714289107,
      "loss": 5.7762,
      "mean_token_accuracy": 0.17586284428834914,
      "num_tokens": 18611856.0,
      "step": 9980
    },
    {
      "entropy": 5.925434350967407,
      "epoch": 0.8817555634051572,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004929224371629634,
      "loss": 5.8302,
      "mean_token_accuracy": 0.1711333855986595,
      "num_tokens": 18620500.0,
      "step": 9985
    },
    {
      "entropy": 5.9265233993530275,
      "epoch": 0.8821971034969975,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004929145986280394,
      "loss": 5.7739,
      "mean_token_accuracy": 0.17708952575922013,
      "num_tokens": 18628836.0,
      "step": 9990
    },
    {
      "entropy": 5.947731924057007,
      "epoch": 0.8826386435888378,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004929067558242922,
      "loss": 5.7026,
      "mean_token_accuracy": 0.17609767168760299,
      "num_tokens": 18636831.0,
      "step": 9995
    },
    {
      "entropy": 5.9934173107147215,
      "epoch": 0.8830801836806782,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004928989087518758,
      "loss": 5.8545,
      "mean_token_accuracy": 0.17121524959802628,
      "num_tokens": 18646327.0,
      "step": 10000
    },
    {
      "entropy": 5.985447359085083,
      "epoch": 0.8835217237725186,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004928910574109435,
      "loss": 5.8655,
      "mean_token_accuracy": 0.16686370521783828,
      "num_tokens": 18656125.0,
      "step": 10005
    },
    {
      "entropy": 6.0014176845550535,
      "epoch": 0.8839632638643589,
      "grad_norm": 2.375,
      "learning_rate": 0.0004928832018016495,
      "loss": 5.8732,
      "mean_token_accuracy": 0.16861837357282639,
      "num_tokens": 18665853.0,
      "step": 10010
    },
    {
      "entropy": 6.076805973052979,
      "epoch": 0.8844048039561992,
      "grad_norm": 2.75,
      "learning_rate": 0.0004928753419241472,
      "loss": 5.9193,
      "mean_token_accuracy": 0.15759100615978242,
      "num_tokens": 18676674.0,
      "step": 10015
    },
    {
      "entropy": 6.0625269412994385,
      "epoch": 0.8848463440480395,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004928674777785912,
      "loss": 5.749,
      "mean_token_accuracy": 0.17859821617603303,
      "num_tokens": 18685535.0,
      "step": 10020
    },
    {
      "entropy": 5.986581325531006,
      "epoch": 0.8852878841398799,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004928596093651351,
      "loss": 5.7937,
      "mean_token_accuracy": 0.16889239251613616,
      "num_tokens": 18694186.0,
      "step": 10025
    },
    {
      "entropy": 5.949480724334717,
      "epoch": 0.8857294242317202,
      "grad_norm": 2.375,
      "learning_rate": 0.000492851736683933,
      "loss": 5.7872,
      "mean_token_accuracy": 0.17278547137975692,
      "num_tokens": 18703061.0,
      "step": 10030
    },
    {
      "entropy": 6.004718446731568,
      "epoch": 0.8861709643235606,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004928438597351395,
      "loss": 5.7324,
      "mean_token_accuracy": 0.17449625134468078,
      "num_tokens": 18712466.0,
      "step": 10035
    },
    {
      "entropy": 5.996918392181397,
      "epoch": 0.886612504415401,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004928359785189086,
      "loss": 5.848,
      "mean_token_accuracy": 0.16393670737743377,
      "num_tokens": 18721522.0,
      "step": 10040
    },
    {
      "entropy": 5.895922231674194,
      "epoch": 0.8870540445072412,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004928280930353948,
      "loss": 5.7844,
      "mean_token_accuracy": 0.168227019906044,
      "num_tokens": 18730488.0,
      "step": 10045
    },
    {
      "entropy": 5.964899063110352,
      "epoch": 0.8874955845990816,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004928202032847525,
      "loss": 5.7105,
      "mean_token_accuracy": 0.17312824875116348,
      "num_tokens": 18740318.0,
      "step": 10050
    },
    {
      "entropy": 5.997185611724854,
      "epoch": 0.8879371246909219,
      "grad_norm": 13.0625,
      "learning_rate": 0.0004928123092671362,
      "loss": 5.8393,
      "mean_token_accuracy": 0.16409815549850465,
      "num_tokens": 18748731.0,
      "step": 10055
    },
    {
      "entropy": 5.89947476387024,
      "epoch": 0.8883786647827623,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004928044109827008,
      "loss": 5.7635,
      "mean_token_accuracy": 0.1718330979347229,
      "num_tokens": 18757847.0,
      "step": 10060
    },
    {
      "entropy": 6.0889490127563475,
      "epoch": 0.8888202048746026,
      "grad_norm": 2.0,
      "learning_rate": 0.0004927965084316007,
      "loss": 5.8568,
      "mean_token_accuracy": 0.15744939744472503,
      "num_tokens": 18767429.0,
      "step": 10065
    },
    {
      "entropy": 6.042394828796387,
      "epoch": 0.889261744966443,
      "grad_norm": 2.359375,
      "learning_rate": 0.000492788601613991,
      "loss": 5.773,
      "mean_token_accuracy": 0.1722065582871437,
      "num_tokens": 18776652.0,
      "step": 10070
    },
    {
      "entropy": 6.038338375091553,
      "epoch": 0.8897032850582833,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004927806905300265,
      "loss": 5.8307,
      "mean_token_accuracy": 0.16547014862298964,
      "num_tokens": 18785687.0,
      "step": 10075
    },
    {
      "entropy": 5.9673412322998045,
      "epoch": 0.8901448251501236,
      "grad_norm": 1.703125,
      "learning_rate": 0.000492772775179862,
      "loss": 5.867,
      "mean_token_accuracy": 0.1659177601337433,
      "num_tokens": 18794924.0,
      "step": 10080
    },
    {
      "entropy": 5.940322780609131,
      "epoch": 0.890586365241964,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004927648555636528,
      "loss": 5.7223,
      "mean_token_accuracy": 0.18146456331014632,
      "num_tokens": 18803359.0,
      "step": 10085
    },
    {
      "entropy": 6.010080289840698,
      "epoch": 0.8910279053338043,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004927569316815539,
      "loss": 5.9499,
      "mean_token_accuracy": 0.1580890327692032,
      "num_tokens": 18813496.0,
      "step": 10090
    },
    {
      "entropy": 6.023469352722168,
      "epoch": 0.8914694454256447,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004927490035337205,
      "loss": 5.7941,
      "mean_token_accuracy": 0.16975688487291335,
      "num_tokens": 18823274.0,
      "step": 10095
    },
    {
      "entropy": 5.9998383045196535,
      "epoch": 0.8919109855174849,
      "grad_norm": 1.609375,
      "learning_rate": 0.000492741071120308,
      "loss": 5.8082,
      "mean_token_accuracy": 0.16879395693540572,
      "num_tokens": 18832436.0,
      "step": 10100
    },
    {
      "entropy": 6.113572883605957,
      "epoch": 0.8923525256093253,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004927331344414717,
      "loss": 5.8595,
      "mean_token_accuracy": 0.16608996838331222,
      "num_tokens": 18841923.0,
      "step": 10105
    },
    {
      "entropy": 6.045321798324585,
      "epoch": 0.8927940657011657,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004927251934973672,
      "loss": 5.8983,
      "mean_token_accuracy": 0.1701177567243576,
      "num_tokens": 18850957.0,
      "step": 10110
    },
    {
      "entropy": 6.070820665359497,
      "epoch": 0.893235605793006,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004927172482881499,
      "loss": 5.8935,
      "mean_token_accuracy": 0.1602569743990898,
      "num_tokens": 18859622.0,
      "step": 10115
    },
    {
      "entropy": 6.068447256088257,
      "epoch": 0.8936771458848464,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004927092988139755,
      "loss": 5.8756,
      "mean_token_accuracy": 0.163049054145813,
      "num_tokens": 18868555.0,
      "step": 10120
    },
    {
      "entropy": 6.048940849304199,
      "epoch": 0.8941186859766866,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004927013450749999,
      "loss": 5.8479,
      "mean_token_accuracy": 0.16745685636997223,
      "num_tokens": 18877445.0,
      "step": 10125
    },
    {
      "entropy": 5.962615346908569,
      "epoch": 0.894560226068527,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004926933870713787,
      "loss": 5.7496,
      "mean_token_accuracy": 0.17192499935626984,
      "num_tokens": 18886386.0,
      "step": 10130
    },
    {
      "entropy": 6.001466941833496,
      "epoch": 0.8950017661603673,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004926854248032678,
      "loss": 5.8533,
      "mean_token_accuracy": 0.16207423508167268,
      "num_tokens": 18895016.0,
      "step": 10135
    },
    {
      "entropy": 5.983141565322876,
      "epoch": 0.8954433062522077,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004926774582708234,
      "loss": 5.7258,
      "mean_token_accuracy": 0.17000181078910828,
      "num_tokens": 18904126.0,
      "step": 10140
    },
    {
      "entropy": 6.032631301879883,
      "epoch": 0.8958848463440481,
      "grad_norm": 4.75,
      "learning_rate": 0.0004926694874742012,
      "loss": 5.794,
      "mean_token_accuracy": 0.1646634042263031,
      "num_tokens": 18913064.0,
      "step": 10145
    },
    {
      "entropy": 6.034622526168823,
      "epoch": 0.8963263864358884,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004926615124135577,
      "loss": 5.8985,
      "mean_token_accuracy": 0.1604196771979332,
      "num_tokens": 18921189.0,
      "step": 10150
    },
    {
      "entropy": 5.966013050079345,
      "epoch": 0.8967679265277287,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004926535330890488,
      "loss": 5.8128,
      "mean_token_accuracy": 0.17230157554149628,
      "num_tokens": 18930818.0,
      "step": 10155
    },
    {
      "entropy": 6.057120084762573,
      "epoch": 0.897209466619569,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004926455495008311,
      "loss": 5.8235,
      "mean_token_accuracy": 0.1639510676264763,
      "num_tokens": 18940376.0,
      "step": 10160
    },
    {
      "entropy": 6.042151069641113,
      "epoch": 0.8976510067114094,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004926375616490608,
      "loss": 5.8589,
      "mean_token_accuracy": 0.16974166482686998,
      "num_tokens": 18949418.0,
      "step": 10165
    },
    {
      "entropy": 6.0127623081207275,
      "epoch": 0.8980925468032497,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004926295695338945,
      "loss": 5.8013,
      "mean_token_accuracy": 0.1742484077811241,
      "num_tokens": 18959315.0,
      "step": 10170
    },
    {
      "entropy": 5.948454856872559,
      "epoch": 0.8985340868950901,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004926215731554887,
      "loss": 5.7142,
      "mean_token_accuracy": 0.17966544330120088,
      "num_tokens": 18968336.0,
      "step": 10175
    },
    {
      "entropy": 6.029808568954468,
      "epoch": 0.8989756269869305,
      "grad_norm": 2.234375,
      "learning_rate": 0.000492613572514,
      "loss": 5.7844,
      "mean_token_accuracy": 0.17160264104604722,
      "num_tokens": 18977123.0,
      "step": 10180
    },
    {
      "entropy": 6.027696847915649,
      "epoch": 0.8994171670787707,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004926055676095851,
      "loss": 5.9058,
      "mean_token_accuracy": 0.16375845670700073,
      "num_tokens": 18986150.0,
      "step": 10185
    },
    {
      "entropy": 6.090052270889283,
      "epoch": 0.8998587071706111,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004925975584424012,
      "loss": 5.9086,
      "mean_token_accuracy": 0.16778819710016252,
      "num_tokens": 18996010.0,
      "step": 10190
    },
    {
      "entropy": 6.057377338409424,
      "epoch": 0.9003002472624514,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004925895450126046,
      "loss": 5.8869,
      "mean_token_accuracy": 0.1585620239377022,
      "num_tokens": 19005100.0,
      "step": 10195
    },
    {
      "entropy": 5.996817922592163,
      "epoch": 0.9007417873542918,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004925815273203526,
      "loss": 5.7848,
      "mean_token_accuracy": 0.17036335468292235,
      "num_tokens": 19013844.0,
      "step": 10200
    },
    {
      "entropy": 6.0277472019195555,
      "epoch": 0.901183327446132,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004925735053658022,
      "loss": 5.9292,
      "mean_token_accuracy": 0.16351503431797026,
      "num_tokens": 19024218.0,
      "step": 10205
    },
    {
      "entropy": 6.083570337295532,
      "epoch": 0.9016248675379724,
      "grad_norm": 1.75,
      "learning_rate": 0.0004925654791491106,
      "loss": 5.8506,
      "mean_token_accuracy": 0.1659012630581856,
      "num_tokens": 19034725.0,
      "step": 10210
    },
    {
      "entropy": 6.1107401847839355,
      "epoch": 0.9020664076298128,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004925574486704351,
      "loss": 5.8669,
      "mean_token_accuracy": 0.16051332503557206,
      "num_tokens": 19044597.0,
      "step": 10215
    },
    {
      "entropy": 6.169135284423828,
      "epoch": 0.9025079477216531,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004925494139299329,
      "loss": 6.0053,
      "mean_token_accuracy": 0.15389579087495803,
      "num_tokens": 19055124.0,
      "step": 10220
    },
    {
      "entropy": 6.02846851348877,
      "epoch": 0.9029494878134935,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004925413749277613,
      "loss": 5.7504,
      "mean_token_accuracy": 0.17126532644033432,
      "num_tokens": 19064516.0,
      "step": 10225
    },
    {
      "entropy": 5.943942546844482,
      "epoch": 0.9033910279053338,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004925333316640779,
      "loss": 5.8318,
      "mean_token_accuracy": 0.16919440776109695,
      "num_tokens": 19072926.0,
      "step": 10230
    },
    {
      "entropy": 5.965893793106079,
      "epoch": 0.9038325679971742,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004925252841390403,
      "loss": 5.8513,
      "mean_token_accuracy": 0.16273195445537567,
      "num_tokens": 19082313.0,
      "step": 10235
    },
    {
      "entropy": 5.933097791671753,
      "epoch": 0.9042741080890144,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004925172323528061,
      "loss": 5.807,
      "mean_token_accuracy": 0.1742863968014717,
      "num_tokens": 19091891.0,
      "step": 10240
    },
    {
      "entropy": 6.075602293014526,
      "epoch": 0.9047156481808548,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000492509176305533,
      "loss": 5.8499,
      "mean_token_accuracy": 0.16614402830600739,
      "num_tokens": 19100311.0,
      "step": 10245
    },
    {
      "entropy": 6.047270584106445,
      "epoch": 0.9051571882726952,
      "grad_norm": 2.5,
      "learning_rate": 0.0004925011159973788,
      "loss": 5.8418,
      "mean_token_accuracy": 0.16825296431779863,
      "num_tokens": 19110463.0,
      "step": 10250
    },
    {
      "entropy": 6.04560227394104,
      "epoch": 0.9055987283645355,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004924930514285015,
      "loss": 5.8287,
      "mean_token_accuracy": 0.16916860342025758,
      "num_tokens": 19120723.0,
      "step": 10255
    },
    {
      "entropy": 5.956229400634766,
      "epoch": 0.9060402684563759,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000492484982599059,
      "loss": 5.8134,
      "mean_token_accuracy": 0.16655978858470916,
      "num_tokens": 19130259.0,
      "step": 10260
    },
    {
      "entropy": 6.0474998474121096,
      "epoch": 0.9064818085482161,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004924769095092094,
      "loss": 5.8794,
      "mean_token_accuracy": 0.16543149054050446,
      "num_tokens": 19140256.0,
      "step": 10265
    },
    {
      "entropy": 5.9884788513183596,
      "epoch": 0.9069233486400565,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004924688321591109,
      "loss": 5.8728,
      "mean_token_accuracy": 0.1729671910405159,
      "num_tokens": 19149165.0,
      "step": 10270
    },
    {
      "entropy": 5.965931797027588,
      "epoch": 0.9073648887318968,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004924607505489215,
      "loss": 5.8013,
      "mean_token_accuracy": 0.17080118060111998,
      "num_tokens": 19159254.0,
      "step": 10275
    },
    {
      "entropy": 6.103870820999146,
      "epoch": 0.9078064288237372,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004924526646787998,
      "loss": 5.8222,
      "mean_token_accuracy": 0.16772038340568543,
      "num_tokens": 19168166.0,
      "step": 10280
    },
    {
      "entropy": 6.075332069396973,
      "epoch": 0.9082479689155776,
      "grad_norm": 1.8984375,
      "learning_rate": 0.000492444574548904,
      "loss": 5.8405,
      "mean_token_accuracy": 0.16294437795877456,
      "num_tokens": 19176887.0,
      "step": 10285
    },
    {
      "entropy": 5.963015031814575,
      "epoch": 0.9086895090074179,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004924364801593927,
      "loss": 5.7515,
      "mean_token_accuracy": 0.17720825374126434,
      "num_tokens": 19185510.0,
      "step": 10290
    },
    {
      "entropy": 5.904071235656739,
      "epoch": 0.9091310490992582,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004924283815104243,
      "loss": 5.7939,
      "mean_token_accuracy": 0.17455934584140778,
      "num_tokens": 19195459.0,
      "step": 10295
    },
    {
      "entropy": 6.018945026397705,
      "epoch": 0.9095725891910985,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004924202786021578,
      "loss": 5.821,
      "mean_token_accuracy": 0.16881232410669328,
      "num_tokens": 19204590.0,
      "step": 10300
    },
    {
      "entropy": 5.987010765075683,
      "epoch": 0.9100141292829389,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004924121714347515,
      "loss": 5.7679,
      "mean_token_accuracy": 0.1634682223200798,
      "num_tokens": 19213302.0,
      "step": 10305
    },
    {
      "entropy": 5.972708415985108,
      "epoch": 0.9104556693747792,
      "grad_norm": 2.375,
      "learning_rate": 0.0004924040600083644,
      "loss": 5.8845,
      "mean_token_accuracy": 0.16732324361801149,
      "num_tokens": 19223172.0,
      "step": 10310
    },
    {
      "entropy": 6.043222093582154,
      "epoch": 0.9108972094666196,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004923959443231554,
      "loss": 5.7881,
      "mean_token_accuracy": 0.16711824387311935,
      "num_tokens": 19232234.0,
      "step": 10315
    },
    {
      "entropy": 5.945252847671509,
      "epoch": 0.91133874955846,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004923878243792836,
      "loss": 5.7717,
      "mean_token_accuracy": 0.17137450724840164,
      "num_tokens": 19241070.0,
      "step": 10320
    },
    {
      "entropy": 5.989360427856445,
      "epoch": 0.9117802896503002,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004923797001769079,
      "loss": 5.8564,
      "mean_token_accuracy": 0.16664954423904418,
      "num_tokens": 19251094.0,
      "step": 10325
    },
    {
      "entropy": 6.135943174362183,
      "epoch": 0.9122218297421406,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004923715717161875,
      "loss": 5.9542,
      "mean_token_accuracy": 0.15768681317567826,
      "num_tokens": 19260588.0,
      "step": 10330
    },
    {
      "entropy": 6.101527309417724,
      "epoch": 0.9126633698339809,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004923634389972817,
      "loss": 5.9774,
      "mean_token_accuracy": 0.15897405296564102,
      "num_tokens": 19270687.0,
      "step": 10335
    },
    {
      "entropy": 5.965734195709229,
      "epoch": 0.9131049099258213,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004923553020203497,
      "loss": 5.7596,
      "mean_token_accuracy": 0.1712796226143837,
      "num_tokens": 19278739.0,
      "step": 10340
    },
    {
      "entropy": 6.115788078308105,
      "epoch": 0.9135464500176615,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004923471607855511,
      "loss": 5.9297,
      "mean_token_accuracy": 0.1538059964776039,
      "num_tokens": 19289184.0,
      "step": 10345
    },
    {
      "entropy": 5.970717334747315,
      "epoch": 0.9139879901095019,
      "grad_norm": 1.8203125,
      "learning_rate": 0.000492339015293045,
      "loss": 5.7746,
      "mean_token_accuracy": 0.1836891993880272,
      "num_tokens": 19298390.0,
      "step": 10350
    },
    {
      "entropy": 5.948292970657349,
      "epoch": 0.9144295302013423,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004923308655429913,
      "loss": 5.7355,
      "mean_token_accuracy": 0.17486001551151276,
      "num_tokens": 19307227.0,
      "step": 10355
    },
    {
      "entropy": 6.03751015663147,
      "epoch": 0.9148710702931826,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004923227115355495,
      "loss": 5.7478,
      "mean_token_accuracy": 0.16704521477222442,
      "num_tokens": 19317701.0,
      "step": 10360
    },
    {
      "entropy": 6.031579113006591,
      "epoch": 0.915312610385023,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004923145532708794,
      "loss": 5.8614,
      "mean_token_accuracy": 0.16852820217609404,
      "num_tokens": 19326983.0,
      "step": 10365
    },
    {
      "entropy": 6.007206058502197,
      "epoch": 0.9157541504768633,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004923063907491408,
      "loss": 5.8839,
      "mean_token_accuracy": 0.16169315129518508,
      "num_tokens": 19337157.0,
      "step": 10370
    },
    {
      "entropy": 6.0474934577941895,
      "epoch": 0.9161956905687036,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004922982239704936,
      "loss": 5.9639,
      "mean_token_accuracy": 0.16408284157514572,
      "num_tokens": 19346845.0,
      "step": 10375
    },
    {
      "entropy": 6.107244873046875,
      "epoch": 0.9166372306605439,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004922900529350978,
      "loss": 5.8326,
      "mean_token_accuracy": 0.17033075243234636,
      "num_tokens": 19356905.0,
      "step": 10380
    },
    {
      "entropy": 6.042671203613281,
      "epoch": 0.9170787707523843,
      "grad_norm": 1.375,
      "learning_rate": 0.0004922818776431134,
      "loss": 5.8214,
      "mean_token_accuracy": 0.15983829498291016,
      "num_tokens": 19366595.0,
      "step": 10385
    },
    {
      "entropy": 6.018132495880127,
      "epoch": 0.9175203108442247,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004922736980947005,
      "loss": 5.8848,
      "mean_token_accuracy": 0.16535525321960448,
      "num_tokens": 19377213.0,
      "step": 10390
    },
    {
      "entropy": 6.001478481292724,
      "epoch": 0.917961850936065,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004922655142900196,
      "loss": 5.742,
      "mean_token_accuracy": 0.16831236183643342,
      "num_tokens": 19385510.0,
      "step": 10395
    },
    {
      "entropy": 5.9540716171264645,
      "epoch": 0.9184033910279054,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004922573262292308,
      "loss": 5.7815,
      "mean_token_accuracy": 0.16358954459428787,
      "num_tokens": 19394505.0,
      "step": 10400
    },
    {
      "entropy": 5.962457704544067,
      "epoch": 0.9188449311197456,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004922491339124945,
      "loss": 5.6794,
      "mean_token_accuracy": 0.17455343455076217,
      "num_tokens": 19403140.0,
      "step": 10405
    },
    {
      "entropy": 5.9841881275177,
      "epoch": 0.919286471211586,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004922409373399713,
      "loss": 5.8875,
      "mean_token_accuracy": 0.16283275336027145,
      "num_tokens": 19412709.0,
      "step": 10410
    },
    {
      "entropy": 5.9437761306762695,
      "epoch": 0.9197280113034263,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004922327365118216,
      "loss": 5.7306,
      "mean_token_accuracy": 0.17520128786563874,
      "num_tokens": 19422442.0,
      "step": 10415
    },
    {
      "entropy": 6.013984823226929,
      "epoch": 0.9201695513952667,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004922245314282062,
      "loss": 5.8573,
      "mean_token_accuracy": 0.16567598134279252,
      "num_tokens": 19432153.0,
      "step": 10420
    },
    {
      "entropy": 5.988616323471069,
      "epoch": 0.9206110914871071,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004922163220892858,
      "loss": 5.6832,
      "mean_token_accuracy": 0.1823098748922348,
      "num_tokens": 19440811.0,
      "step": 10425
    },
    {
      "entropy": 5.9555412292480465,
      "epoch": 0.9210526315789473,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000492208108495221,
      "loss": 5.8588,
      "mean_token_accuracy": 0.16761756986379622,
      "num_tokens": 19451330.0,
      "step": 10430
    },
    {
      "entropy": 5.9230536937713625,
      "epoch": 0.9214941716707877,
      "grad_norm": 1.703125,
      "learning_rate": 0.000492199890646173,
      "loss": 5.6983,
      "mean_token_accuracy": 0.1846833571791649,
      "num_tokens": 19459689.0,
      "step": 10435
    },
    {
      "entropy": 6.030665397644043,
      "epoch": 0.921935711762628,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004921916685423028,
      "loss": 5.7551,
      "mean_token_accuracy": 0.17246931791305542,
      "num_tokens": 19468110.0,
      "step": 10440
    },
    {
      "entropy": 5.983535623550415,
      "epoch": 0.9223772518544684,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004921834421837711,
      "loss": 5.8192,
      "mean_token_accuracy": 0.16404144763946532,
      "num_tokens": 19477190.0,
      "step": 10445
    },
    {
      "entropy": 6.092430591583252,
      "epoch": 0.9228187919463087,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004921752115707394,
      "loss": 5.9195,
      "mean_token_accuracy": 0.16309409886598586,
      "num_tokens": 19485757.0,
      "step": 10450
    },
    {
      "entropy": 6.059690475463867,
      "epoch": 0.9232603320381491,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004921669767033688,
      "loss": 5.8539,
      "mean_token_accuracy": 0.16356213837862016,
      "num_tokens": 19494846.0,
      "step": 10455
    },
    {
      "entropy": 5.986880207061768,
      "epoch": 0.9237018721299894,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004921587375818207,
      "loss": 5.8418,
      "mean_token_accuracy": 0.16628807932138442,
      "num_tokens": 19504307.0,
      "step": 10460
    },
    {
      "entropy": 6.010497951507569,
      "epoch": 0.9241434122218297,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004921504942062564,
      "loss": 5.7781,
      "mean_token_accuracy": 0.16801584959030152,
      "num_tokens": 19512741.0,
      "step": 10465
    },
    {
      "entropy": 5.956787252426148,
      "epoch": 0.9245849523136701,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004921422465768374,
      "loss": 5.6549,
      "mean_token_accuracy": 0.17740724086761475,
      "num_tokens": 19521823.0,
      "step": 10470
    },
    {
      "entropy": 6.001222944259643,
      "epoch": 0.9250264924055104,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004921339946937253,
      "loss": 5.8591,
      "mean_token_accuracy": 0.16669148355722427,
      "num_tokens": 19531048.0,
      "step": 10475
    },
    {
      "entropy": 6.044188165664673,
      "epoch": 0.9254680324973508,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004921257385570818,
      "loss": 5.8162,
      "mean_token_accuracy": 0.17120853662490845,
      "num_tokens": 19539543.0,
      "step": 10480
    },
    {
      "entropy": 6.096062421798706,
      "epoch": 0.925909572589191,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004921174781670685,
      "loss": 5.9247,
      "mean_token_accuracy": 0.16498176455497743,
      "num_tokens": 19549242.0,
      "step": 10485
    },
    {
      "entropy": 6.070945692062378,
      "epoch": 0.9263511126810314,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004921092135238473,
      "loss": 5.8422,
      "mean_token_accuracy": 0.1619497060775757,
      "num_tokens": 19558259.0,
      "step": 10490
    },
    {
      "entropy": 6.021557855606079,
      "epoch": 0.9267926527728718,
      "grad_norm": 1.578125,
      "learning_rate": 0.00049210094462758,
      "loss": 5.8189,
      "mean_token_accuracy": 0.16899777501821517,
      "num_tokens": 19567327.0,
      "step": 10495
    },
    {
      "entropy": 6.036462640762329,
      "epoch": 0.9272341928647121,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004920926714784288,
      "loss": 5.855,
      "mean_token_accuracy": 0.16469730883836747,
      "num_tokens": 19576696.0,
      "step": 10500
    },
    {
      "entropy": 6.01403579711914,
      "epoch": 0.9276757329565525,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004920843940765555,
      "loss": 5.7437,
      "mean_token_accuracy": 0.17097172141075134,
      "num_tokens": 19585708.0,
      "step": 10505
    },
    {
      "entropy": 6.025144863128662,
      "epoch": 0.9281172730483928,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004920761124221224,
      "loss": 5.8908,
      "mean_token_accuracy": 0.1659647688269615,
      "num_tokens": 19595199.0,
      "step": 10510
    },
    {
      "entropy": 6.00489616394043,
      "epoch": 0.9285588131402331,
      "grad_norm": 1.5,
      "learning_rate": 0.0004920678265152919,
      "loss": 5.8256,
      "mean_token_accuracy": 0.1605701707303524,
      "num_tokens": 19604781.0,
      "step": 10515
    },
    {
      "entropy": 6.026356172561646,
      "epoch": 0.9290003532320734,
      "grad_norm": 1.640625,
      "learning_rate": 0.000492059536356226,
      "loss": 5.7732,
      "mean_token_accuracy": 0.1747516244649887,
      "num_tokens": 19614203.0,
      "step": 10520
    },
    {
      "entropy": 6.019496011734009,
      "epoch": 0.9294418933239138,
      "grad_norm": 1.625,
      "learning_rate": 0.0004920512419450872,
      "loss": 5.8025,
      "mean_token_accuracy": 0.17150902897119522,
      "num_tokens": 19623664.0,
      "step": 10525
    },
    {
      "entropy": 5.990518093109131,
      "epoch": 0.9298834334157542,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000492042943282038,
      "loss": 5.8108,
      "mean_token_accuracy": 0.17041826993227005,
      "num_tokens": 19632960.0,
      "step": 10530
    },
    {
      "entropy": 5.9955668449401855,
      "epoch": 0.9303249735075945,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000492034640367241,
      "loss": 5.9308,
      "mean_token_accuracy": 0.15933274179697038,
      "num_tokens": 19643087.0,
      "step": 10535
    },
    {
      "entropy": 6.070011329650879,
      "epoch": 0.9307665135994349,
      "grad_norm": 1.59375,
      "learning_rate": 0.000492026333200859,
      "loss": 5.8737,
      "mean_token_accuracy": 0.16645802557468414,
      "num_tokens": 19653015.0,
      "step": 10540
    },
    {
      "entropy": 6.028250122070313,
      "epoch": 0.9312080536912751,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004920180217830543,
      "loss": 5.8954,
      "mean_token_accuracy": 0.16525914818048476,
      "num_tokens": 19661569.0,
      "step": 10545
    },
    {
      "entropy": 5.893161964416504,
      "epoch": 0.9316495937831155,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00049200970611399,
      "loss": 5.7027,
      "mean_token_accuracy": 0.17925852835178374,
      "num_tokens": 19669559.0,
      "step": 10550
    },
    {
      "entropy": 5.921646547317505,
      "epoch": 0.9320911338749558,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004920013861938291,
      "loss": 5.7284,
      "mean_token_accuracy": 0.18288709819316865,
      "num_tokens": 19678766.0,
      "step": 10555
    },
    {
      "entropy": 5.988527011871338,
      "epoch": 0.9325326739667962,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004919930620227345,
      "loss": 5.8548,
      "mean_token_accuracy": 0.15858465284109116,
      "num_tokens": 19688422.0,
      "step": 10560
    },
    {
      "entropy": 6.099614858627319,
      "epoch": 0.9329742140586366,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004919847336008692,
      "loss": 5.8865,
      "mean_token_accuracy": 0.16525954753160477,
      "num_tokens": 19697206.0,
      "step": 10565
    },
    {
      "entropy": 6.0650975704193115,
      "epoch": 0.9334157541504768,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004919764009283963,
      "loss": 5.8669,
      "mean_token_accuracy": 0.16111723333597183,
      "num_tokens": 19706065.0,
      "step": 10570
    },
    {
      "entropy": 6.030399179458618,
      "epoch": 0.9338572942423172,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004919680640054792,
      "loss": 5.7794,
      "mean_token_accuracy": 0.17192087024450303,
      "num_tokens": 19715402.0,
      "step": 10575
    },
    {
      "entropy": 6.004525375366211,
      "epoch": 0.9342988343341575,
      "grad_norm": 1.90625,
      "learning_rate": 0.000491959722832281,
      "loss": 5.7321,
      "mean_token_accuracy": 0.1754280924797058,
      "num_tokens": 19724328.0,
      "step": 10580
    },
    {
      "entropy": 5.910653209686279,
      "epoch": 0.9347403744259979,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004919513774089653,
      "loss": 5.8325,
      "mean_token_accuracy": 0.1720520257949829,
      "num_tokens": 19734087.0,
      "step": 10585
    },
    {
      "entropy": 6.058596134185791,
      "epoch": 0.9351819145178382,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004919430277356955,
      "loss": 5.8104,
      "mean_token_accuracy": 0.1688505083322525,
      "num_tokens": 19743010.0,
      "step": 10590
    },
    {
      "entropy": 5.9669163703918455,
      "epoch": 0.9356234546096786,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004919346738126351,
      "loss": 5.7427,
      "mean_token_accuracy": 0.175693878531456,
      "num_tokens": 19752212.0,
      "step": 10595
    },
    {
      "entropy": 5.984468603134156,
      "epoch": 0.9360649947015189,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004919263156399478,
      "loss": 5.8503,
      "mean_token_accuracy": 0.164433790743351,
      "num_tokens": 19762496.0,
      "step": 10600
    },
    {
      "entropy": 6.021150827407837,
      "epoch": 0.9365065347933592,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004919179532177976,
      "loss": 5.846,
      "mean_token_accuracy": 0.16879568845033646,
      "num_tokens": 19772665.0,
      "step": 10605
    },
    {
      "entropy": 5.892103052139282,
      "epoch": 0.9369480748851996,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004919095865463479,
      "loss": 5.6709,
      "mean_token_accuracy": 0.17989853024482727,
      "num_tokens": 19781639.0,
      "step": 10610
    },
    {
      "entropy": 5.981200265884399,
      "epoch": 0.9373896149770399,
      "grad_norm": 2.25,
      "learning_rate": 0.0004919012156257627,
      "loss": 5.7667,
      "mean_token_accuracy": 0.16907234638929367,
      "num_tokens": 19791475.0,
      "step": 10615
    },
    {
      "entropy": 6.02192645072937,
      "epoch": 0.9378311550688803,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004918928404562061,
      "loss": 5.7774,
      "mean_token_accuracy": 0.1712593361735344,
      "num_tokens": 19800967.0,
      "step": 10620
    },
    {
      "entropy": 5.947515249252319,
      "epoch": 0.9382726951607206,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004918844610378421,
      "loss": 5.9131,
      "mean_token_accuracy": 0.16362484246492387,
      "num_tokens": 19810303.0,
      "step": 10625
    },
    {
      "entropy": 6.076837348937988,
      "epoch": 0.9387142352525609,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004918760773708348,
      "loss": 5.8193,
      "mean_token_accuracy": 0.1612846091389656,
      "num_tokens": 19819049.0,
      "step": 10630
    },
    {
      "entropy": 6.073859214782715,
      "epoch": 0.9391557753444013,
      "grad_norm": 1.5,
      "learning_rate": 0.0004918676894553485,
      "loss": 5.8531,
      "mean_token_accuracy": 0.1676570177078247,
      "num_tokens": 19828253.0,
      "step": 10635
    },
    {
      "entropy": 5.961663913726807,
      "epoch": 0.9395973154362416,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004918592972915476,
      "loss": 5.8106,
      "mean_token_accuracy": 0.1745462715625763,
      "num_tokens": 19837673.0,
      "step": 10640
    },
    {
      "entropy": 5.932157897949219,
      "epoch": 0.940038855528082,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004918509008795963,
      "loss": 5.8512,
      "mean_token_accuracy": 0.16296276599168777,
      "num_tokens": 19846289.0,
      "step": 10645
    },
    {
      "entropy": 5.946554470062256,
      "epoch": 0.9404803956199222,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004918425002196594,
      "loss": 5.8133,
      "mean_token_accuracy": 0.16478978991508483,
      "num_tokens": 19855384.0,
      "step": 10650
    },
    {
      "entropy": 5.981734752655029,
      "epoch": 0.9409219357117626,
      "grad_norm": 2.296875,
      "learning_rate": 0.000491834095311901,
      "loss": 5.7256,
      "mean_token_accuracy": 0.17702484130859375,
      "num_tokens": 19864688.0,
      "step": 10655
    },
    {
      "entropy": 5.9941754817962645,
      "epoch": 0.941363475803603,
      "grad_norm": 2.03125,
      "learning_rate": 0.000491825686156486,
      "loss": 5.7347,
      "mean_token_accuracy": 0.171346752345562,
      "num_tokens": 19873692.0,
      "step": 10660
    },
    {
      "entropy": 6.019399261474609,
      "epoch": 0.9418050158954433,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004918172727535792,
      "loss": 5.8573,
      "mean_token_accuracy": 0.16431866139173507,
      "num_tokens": 19882063.0,
      "step": 10665
    },
    {
      "entropy": 6.005358266830444,
      "epoch": 0.9422465559872837,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004918088551033454,
      "loss": 5.8143,
      "mean_token_accuracy": 0.16709273606538771,
      "num_tokens": 19891727.0,
      "step": 10670
    },
    {
      "entropy": 5.932406759262085,
      "epoch": 0.942688096079124,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004918004332059493,
      "loss": 5.8489,
      "mean_token_accuracy": 0.17194831222295762,
      "num_tokens": 19901851.0,
      "step": 10675
    },
    {
      "entropy": 6.038689661026001,
      "epoch": 0.9431296361709643,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004917920070615561,
      "loss": 5.7874,
      "mean_token_accuracy": 0.16205894947052002,
      "num_tokens": 19911092.0,
      "step": 10680
    },
    {
      "entropy": 5.9625006198883055,
      "epoch": 0.9435711762628046,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004917835766703307,
      "loss": 5.7588,
      "mean_token_accuracy": 0.17523420453071595,
      "num_tokens": 19920842.0,
      "step": 10685
    },
    {
      "entropy": 6.018384742736816,
      "epoch": 0.944012716354645,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004917751420324384,
      "loss": 5.8684,
      "mean_token_accuracy": 0.16540421098470687,
      "num_tokens": 19930833.0,
      "step": 10690
    },
    {
      "entropy": 6.0669207096099855,
      "epoch": 0.9444542564464854,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004917667031480444,
      "loss": 5.8884,
      "mean_token_accuracy": 0.16578007638454437,
      "num_tokens": 19940514.0,
      "step": 10695
    },
    {
      "entropy": 6.007361221313476,
      "epoch": 0.9448957965383257,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004917582600173138,
      "loss": 5.7396,
      "mean_token_accuracy": 0.17524563521146774,
      "num_tokens": 19948949.0,
      "step": 10700
    },
    {
      "entropy": 5.867607307434082,
      "epoch": 0.9453373366301661,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004917498126404123,
      "loss": 5.717,
      "mean_token_accuracy": 0.17226629108190536,
      "num_tokens": 19958410.0,
      "step": 10705
    },
    {
      "entropy": 5.93734655380249,
      "epoch": 0.9457788767220063,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004917413610175052,
      "loss": 5.7132,
      "mean_token_accuracy": 0.18073296248912812,
      "num_tokens": 19968160.0,
      "step": 10710
    },
    {
      "entropy": 6.001311540603638,
      "epoch": 0.9462204168138467,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004917329051487582,
      "loss": 5.8792,
      "mean_token_accuracy": 0.15905057042837142,
      "num_tokens": 19976811.0,
      "step": 10715
    },
    {
      "entropy": 6.1178806781768795,
      "epoch": 0.946661956905687,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004917244450343368,
      "loss": 5.9291,
      "mean_token_accuracy": 0.162720163166523,
      "num_tokens": 19986326.0,
      "step": 10720
    },
    {
      "entropy": 5.937975215911865,
      "epoch": 0.9471034969975274,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004917159806744067,
      "loss": 5.6805,
      "mean_token_accuracy": 0.171738201379776,
      "num_tokens": 19996255.0,
      "step": 10725
    },
    {
      "entropy": 5.956577587127685,
      "epoch": 0.9475450370893678,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004917075120691338,
      "loss": 5.8767,
      "mean_token_accuracy": 0.16906536221504212,
      "num_tokens": 20004805.0,
      "step": 10730
    },
    {
      "entropy": 5.96188440322876,
      "epoch": 0.947986577181208,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004916990392186841,
      "loss": 5.7767,
      "mean_token_accuracy": 0.17188985794782638,
      "num_tokens": 20014189.0,
      "step": 10735
    },
    {
      "entropy": 6.049740886688232,
      "epoch": 0.9484281172730484,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004916905621232233,
      "loss": 5.7779,
      "mean_token_accuracy": 0.1737987980246544,
      "num_tokens": 20023261.0,
      "step": 10740
    },
    {
      "entropy": 5.931927347183228,
      "epoch": 0.9488696573648887,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004916820807829177,
      "loss": 5.7308,
      "mean_token_accuracy": 0.1741984859108925,
      "num_tokens": 20032433.0,
      "step": 10745
    },
    {
      "entropy": 5.957098436355591,
      "epoch": 0.9493111974567291,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004916735951979335,
      "loss": 5.8473,
      "mean_token_accuracy": 0.16521918177604675,
      "num_tokens": 20041455.0,
      "step": 10750
    },
    {
      "entropy": 6.013712930679321,
      "epoch": 0.9497527375485694,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004916651053684366,
      "loss": 5.8583,
      "mean_token_accuracy": 0.16902563199400902,
      "num_tokens": 20050088.0,
      "step": 10755
    },
    {
      "entropy": 6.03648190498352,
      "epoch": 0.9501942776404098,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004916566112945936,
      "loss": 5.8403,
      "mean_token_accuracy": 0.16953138560056685,
      "num_tokens": 20059114.0,
      "step": 10760
    },
    {
      "entropy": 6.005605697631836,
      "epoch": 0.9506358177322501,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004916481129765708,
      "loss": 5.8434,
      "mean_token_accuracy": 0.16606313288211821,
      "num_tokens": 20068692.0,
      "step": 10765
    },
    {
      "entropy": 5.988512325286865,
      "epoch": 0.9510773578240904,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004916396104145347,
      "loss": 5.8705,
      "mean_token_accuracy": 0.16754111647605896,
      "num_tokens": 20078380.0,
      "step": 10770
    },
    {
      "entropy": 5.896767950057983,
      "epoch": 0.9515188979159308,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004916311036086518,
      "loss": 5.716,
      "mean_token_accuracy": 0.18125931322574615,
      "num_tokens": 20087849.0,
      "step": 10775
    },
    {
      "entropy": 5.938125896453857,
      "epoch": 0.9519604380077711,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004916225925590889,
      "loss": 5.6547,
      "mean_token_accuracy": 0.18121337294578552,
      "num_tokens": 20097253.0,
      "step": 10780
    },
    {
      "entropy": 5.971297359466552,
      "epoch": 0.9524019780996115,
      "grad_norm": 2.0,
      "learning_rate": 0.0004916140772660124,
      "loss": 5.8462,
      "mean_token_accuracy": 0.172994901239872,
      "num_tokens": 20105931.0,
      "step": 10785
    },
    {
      "entropy": 6.003271102905273,
      "epoch": 0.9528435181914517,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004916055577295895,
      "loss": 5.7553,
      "mean_token_accuracy": 0.17517725229263306,
      "num_tokens": 20115546.0,
      "step": 10790
    },
    {
      "entropy": 5.985776948928833,
      "epoch": 0.9532850582832921,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004915970339499868,
      "loss": 5.7738,
      "mean_token_accuracy": 0.1728436455130577,
      "num_tokens": 20125009.0,
      "step": 10795
    },
    {
      "entropy": 5.862307453155518,
      "epoch": 0.9537265983751325,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004915885059273714,
      "loss": 5.7374,
      "mean_token_accuracy": 0.17673692107200623,
      "num_tokens": 20134007.0,
      "step": 10800
    },
    {
      "entropy": 5.971608638763428,
      "epoch": 0.9541681384669728,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004915799736619104,
      "loss": 5.7595,
      "mean_token_accuracy": 0.173075895011425,
      "num_tokens": 20142617.0,
      "step": 10805
    },
    {
      "entropy": 5.951097583770752,
      "epoch": 0.9546096785588132,
      "grad_norm": 2.09375,
      "learning_rate": 0.000491571437153771,
      "loss": 5.7702,
      "mean_token_accuracy": 0.17575397044420243,
      "num_tokens": 20151732.0,
      "step": 10810
    },
    {
      "entropy": 6.080293321609497,
      "epoch": 0.9550512186506535,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004915628964031202,
      "loss": 5.8827,
      "mean_token_accuracy": 0.1621328040957451,
      "num_tokens": 20161733.0,
      "step": 10815
    },
    {
      "entropy": 6.004674625396729,
      "epoch": 0.9554927587424938,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004915543514101253,
      "loss": 5.8016,
      "mean_token_accuracy": 0.16970572173595427,
      "num_tokens": 20171022.0,
      "step": 10820
    },
    {
      "entropy": 5.930174112319946,
      "epoch": 0.9559342988343341,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004915458021749541,
      "loss": 5.8132,
      "mean_token_accuracy": 0.1605657383799553,
      "num_tokens": 20180809.0,
      "step": 10825
    },
    {
      "entropy": 5.951100921630859,
      "epoch": 0.9563758389261745,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004915372486977736,
      "loss": 5.7428,
      "mean_token_accuracy": 0.17625252157449722,
      "num_tokens": 20189614.0,
      "step": 10830
    },
    {
      "entropy": 6.029724359512329,
      "epoch": 0.9568173790180149,
      "grad_norm": 1.375,
      "learning_rate": 0.0004915286909787516,
      "loss": 5.8935,
      "mean_token_accuracy": 0.1590195283293724,
      "num_tokens": 20199136.0,
      "step": 10835
    },
    {
      "entropy": 5.989223146438599,
      "epoch": 0.9572589191098552,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004915201290180556,
      "loss": 5.7761,
      "mean_token_accuracy": 0.16439834833145142,
      "num_tokens": 20208768.0,
      "step": 10840
    },
    {
      "entropy": 5.914671230316162,
      "epoch": 0.9577004592016956,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004915115628158535,
      "loss": 5.7724,
      "mean_token_accuracy": 0.17083653658628464,
      "num_tokens": 20218109.0,
      "step": 10845
    },
    {
      "entropy": 6.004290294647217,
      "epoch": 0.9581419992935358,
      "grad_norm": 1.5,
      "learning_rate": 0.000491502992372313,
      "loss": 5.7731,
      "mean_token_accuracy": 0.16725710779428482,
      "num_tokens": 20226768.0,
      "step": 10850
    },
    {
      "entropy": 5.991324758529663,
      "epoch": 0.9585835393853762,
      "grad_norm": 2.015625,
      "learning_rate": 0.000491494417687602,
      "loss": 5.7184,
      "mean_token_accuracy": 0.17218151092529296,
      "num_tokens": 20236519.0,
      "step": 10855
    },
    {
      "entropy": 5.94624695777893,
      "epoch": 0.9590250794772165,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004914858387618886,
      "loss": 5.7573,
      "mean_token_accuracy": 0.1722206875681877,
      "num_tokens": 20245611.0,
      "step": 10860
    },
    {
      "entropy": 5.907992362976074,
      "epoch": 0.9594666195690569,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004914772555953406,
      "loss": 5.6558,
      "mean_token_accuracy": 0.1819604218006134,
      "num_tokens": 20255089.0,
      "step": 10865
    },
    {
      "entropy": 5.935388612747192,
      "epoch": 0.9599081596608973,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004914686681881264,
      "loss": 5.8238,
      "mean_token_accuracy": 0.16650898456573487,
      "num_tokens": 20264794.0,
      "step": 10870
    },
    {
      "entropy": 5.934653568267822,
      "epoch": 0.9603496997527375,
      "grad_norm": 1.46875,
      "learning_rate": 0.000491460076540414,
      "loss": 5.6954,
      "mean_token_accuracy": 0.18213992118835448,
      "num_tokens": 20274088.0,
      "step": 10875
    },
    {
      "entropy": 5.9760363578796385,
      "epoch": 0.9607912398445779,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004914514806523718,
      "loss": 5.859,
      "mean_token_accuracy": 0.16019129902124404,
      "num_tokens": 20283837.0,
      "step": 10880
    },
    {
      "entropy": 6.003293085098266,
      "epoch": 0.9612327799364182,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004914428805241683,
      "loss": 5.7622,
      "mean_token_accuracy": 0.16625455170869827,
      "num_tokens": 20293287.0,
      "step": 10885
    },
    {
      "entropy": 5.907747411727906,
      "epoch": 0.9616743200282586,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004914342761559718,
      "loss": 5.719,
      "mean_token_accuracy": 0.17710004597902299,
      "num_tokens": 20302262.0,
      "step": 10890
    },
    {
      "entropy": 6.0164556980133055,
      "epoch": 0.9621158601200989,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004914256675479509,
      "loss": 5.8292,
      "mean_token_accuracy": 0.1677974209189415,
      "num_tokens": 20312475.0,
      "step": 10895
    },
    {
      "entropy": 6.0269428253173825,
      "epoch": 0.9625574002119393,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004914170547002742,
      "loss": 5.833,
      "mean_token_accuracy": 0.16870647370815278,
      "num_tokens": 20322243.0,
      "step": 10900
    },
    {
      "entropy": 6.016332721710205,
      "epoch": 0.9629989403037796,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004914084376131106,
      "loss": 5.7569,
      "mean_token_accuracy": 0.16687582582235336,
      "num_tokens": 20331299.0,
      "step": 10905
    },
    {
      "entropy": 5.913394355773926,
      "epoch": 0.9634404803956199,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004913998162866288,
      "loss": 5.6931,
      "mean_token_accuracy": 0.17729486376047135,
      "num_tokens": 20340800.0,
      "step": 10910
    },
    {
      "entropy": 5.959266328811646,
      "epoch": 0.9638820204874603,
      "grad_norm": 1.625,
      "learning_rate": 0.0004913911907209976,
      "loss": 5.7692,
      "mean_token_accuracy": 0.17721022814512252,
      "num_tokens": 20350292.0,
      "step": 10915
    },
    {
      "entropy": 6.043052721023559,
      "epoch": 0.9643235605793006,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004913825609163861,
      "loss": 5.8679,
      "mean_token_accuracy": 0.16006149500608444,
      "num_tokens": 20360109.0,
      "step": 10920
    },
    {
      "entropy": 5.860322570800781,
      "epoch": 0.964765100671141,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004913739268729633,
      "loss": 5.6917,
      "mean_token_accuracy": 0.1784888446331024,
      "num_tokens": 20368917.0,
      "step": 10925
    },
    {
      "entropy": 6.006972455978394,
      "epoch": 0.9652066407629812,
      "grad_norm": 1.75,
      "learning_rate": 0.0004913652885908983,
      "loss": 5.8962,
      "mean_token_accuracy": 0.16221495121717452,
      "num_tokens": 20378580.0,
      "step": 10930
    },
    {
      "entropy": 5.97238073348999,
      "epoch": 0.9656481808548216,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004913566460703603,
      "loss": 5.7268,
      "mean_token_accuracy": 0.17603309750556945,
      "num_tokens": 20387497.0,
      "step": 10935
    },
    {
      "entropy": 5.99204421043396,
      "epoch": 0.966089720946662,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004913479993115187,
      "loss": 5.8727,
      "mean_token_accuracy": 0.1619122162461281,
      "num_tokens": 20396586.0,
      "step": 10940
    },
    {
      "entropy": 6.0532196998596195,
      "epoch": 0.9665312610385023,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004913393483145428,
      "loss": 5.9377,
      "mean_token_accuracy": 0.15912597626447678,
      "num_tokens": 20406183.0,
      "step": 10945
    },
    {
      "entropy": 6.163526821136474,
      "epoch": 0.9669728011303427,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004913306930796022,
      "loss": 5.8877,
      "mean_token_accuracy": 0.16394948363304138,
      "num_tokens": 20415716.0,
      "step": 10950
    },
    {
      "entropy": 6.033840751647949,
      "epoch": 0.967414341222183,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004913220336068662,
      "loss": 5.801,
      "mean_token_accuracy": 0.16864629238843917,
      "num_tokens": 20424715.0,
      "step": 10955
    },
    {
      "entropy": 5.9165130138397215,
      "epoch": 0.9678558813140233,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004913133698965048,
      "loss": 5.7995,
      "mean_token_accuracy": 0.1718275874853134,
      "num_tokens": 20434686.0,
      "step": 10960
    },
    {
      "entropy": 6.020512485504151,
      "epoch": 0.9682974214058636,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004913047019486874,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1689111992716789,
      "num_tokens": 20443619.0,
      "step": 10965
    },
    {
      "entropy": 5.928170108795166,
      "epoch": 0.968738961497704,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004912960297635839,
      "loss": 5.7247,
      "mean_token_accuracy": 0.18071752190589904,
      "num_tokens": 20453540.0,
      "step": 10970
    },
    {
      "entropy": 5.973074340820313,
      "epoch": 0.9691805015895444,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004912873533413641,
      "loss": 5.8386,
      "mean_token_accuracy": 0.16714389324188234,
      "num_tokens": 20463395.0,
      "step": 10975
    },
    {
      "entropy": 6.01979546546936,
      "epoch": 0.9696220416813847,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004912786726821982,
      "loss": 5.7723,
      "mean_token_accuracy": 0.1673387110233307,
      "num_tokens": 20472302.0,
      "step": 10980
    },
    {
      "entropy": 5.932343435287476,
      "epoch": 0.970063581773225,
      "grad_norm": 1.953125,
      "learning_rate": 0.000491269987786256,
      "loss": 5.8032,
      "mean_token_accuracy": 0.16516162306070328,
      "num_tokens": 20482077.0,
      "step": 10985
    },
    {
      "entropy": 5.987225818634033,
      "epoch": 0.9705051218650653,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004912612986537078,
      "loss": 5.7917,
      "mean_token_accuracy": 0.16847066432237626,
      "num_tokens": 20492177.0,
      "step": 10990
    },
    {
      "entropy": 6.021402645111084,
      "epoch": 0.9709466619569057,
      "grad_norm": 1.75,
      "learning_rate": 0.0004912526052847238,
      "loss": 5.8207,
      "mean_token_accuracy": 0.17326410710811616,
      "num_tokens": 20502063.0,
      "step": 10995
    },
    {
      "entropy": 5.953307962417602,
      "epoch": 0.971388202048746,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004912439076794742,
      "loss": 5.7591,
      "mean_token_accuracy": 0.1668146625161171,
      "num_tokens": 20510631.0,
      "step": 11000
    },
    {
      "entropy": 5.895603370666504,
      "epoch": 0.9718297421405864,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004912352058381293,
      "loss": 5.7888,
      "mean_token_accuracy": 0.1695307195186615,
      "num_tokens": 20519978.0,
      "step": 11005
    },
    {
      "entropy": 5.953231716156006,
      "epoch": 0.9722712822324268,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004912264997608599,
      "loss": 5.7773,
      "mean_token_accuracy": 0.1737328365445137,
      "num_tokens": 20529346.0,
      "step": 11010
    },
    {
      "entropy": 6.022437715530396,
      "epoch": 0.972712822324267,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004912177894478364,
      "loss": 5.7437,
      "mean_token_accuracy": 0.1649177610874176,
      "num_tokens": 20537619.0,
      "step": 11015
    },
    {
      "entropy": 5.978044605255127,
      "epoch": 0.9731543624161074,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004912090748992291,
      "loss": 5.7933,
      "mean_token_accuracy": 0.17317971140146254,
      "num_tokens": 20547011.0,
      "step": 11020
    },
    {
      "entropy": 5.990376043319702,
      "epoch": 0.9735959025079477,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004912003561152092,
      "loss": 5.7147,
      "mean_token_accuracy": 0.17324843108654023,
      "num_tokens": 20556425.0,
      "step": 11025
    },
    {
      "entropy": 5.928810358047485,
      "epoch": 0.9740374425997881,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004911916330959472,
      "loss": 5.7334,
      "mean_token_accuracy": 0.1760266274213791,
      "num_tokens": 20565330.0,
      "step": 11030
    },
    {
      "entropy": 5.9286705493927006,
      "epoch": 0.9744789826916284,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004911829058416141,
      "loss": 5.8313,
      "mean_token_accuracy": 0.1810968041419983,
      "num_tokens": 20575320.0,
      "step": 11035
    },
    {
      "entropy": 5.994808006286621,
      "epoch": 0.9749205227834687,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004911741743523808,
      "loss": 5.7431,
      "mean_token_accuracy": 0.17848343253135682,
      "num_tokens": 20585105.0,
      "step": 11040
    },
    {
      "entropy": 5.966033029556274,
      "epoch": 0.9753620628753091,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004911654386284184,
      "loss": 5.7916,
      "mean_token_accuracy": 0.17322781980037688,
      "num_tokens": 20593889.0,
      "step": 11045
    },
    {
      "entropy": 5.978029823303222,
      "epoch": 0.9758036029671494,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000491156698669898,
      "loss": 5.8134,
      "mean_token_accuracy": 0.17286840006709098,
      "num_tokens": 20602322.0,
      "step": 11050
    },
    {
      "entropy": 6.022975111007691,
      "epoch": 0.9762451430589898,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004911479544769908,
      "loss": 5.8974,
      "mean_token_accuracy": 0.15766229033470153,
      "num_tokens": 20612077.0,
      "step": 11055
    },
    {
      "entropy": 6.059322118759155,
      "epoch": 0.9766866831508301,
      "grad_norm": 1.8984375,
      "learning_rate": 0.000491139206049868,
      "loss": 5.8469,
      "mean_token_accuracy": 0.17113145887851716,
      "num_tokens": 20620683.0,
      "step": 11060
    },
    {
      "entropy": 5.986019706726074,
      "epoch": 0.9771282232426705,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004911304533887011,
      "loss": 5.9007,
      "mean_token_accuracy": 0.16220422834157944,
      "num_tokens": 20631124.0,
      "step": 11065
    },
    {
      "entropy": 5.98687162399292,
      "epoch": 0.9775697633345107,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004911216964936616,
      "loss": 5.7466,
      "mean_token_accuracy": 0.17153911888599396,
      "num_tokens": 20640515.0,
      "step": 11070
    },
    {
      "entropy": 5.96340537071228,
      "epoch": 0.9780113034263511,
      "grad_norm": 3.84375,
      "learning_rate": 0.0004911129353649211,
      "loss": 5.7365,
      "mean_token_accuracy": 0.174886654317379,
      "num_tokens": 20649819.0,
      "step": 11075
    },
    {
      "entropy": 5.944764804840088,
      "epoch": 0.9784528435181915,
      "grad_norm": 2.28125,
      "learning_rate": 0.000491104170002651,
      "loss": 5.7713,
      "mean_token_accuracy": 0.17416697293519973,
      "num_tokens": 20657831.0,
      "step": 11080
    },
    {
      "entropy": 6.038313531875611,
      "epoch": 0.9788943836100318,
      "grad_norm": 1.53125,
      "learning_rate": 0.000491095400407023,
      "loss": 5.8761,
      "mean_token_accuracy": 0.17019579261541368,
      "num_tokens": 20666807.0,
      "step": 11085
    },
    {
      "entropy": 6.0149578094482425,
      "epoch": 0.9793359237018722,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004910866265782091,
      "loss": 5.7491,
      "mean_token_accuracy": 0.17066479325294495,
      "num_tokens": 20674859.0,
      "step": 11090
    },
    {
      "entropy": 6.064696598052978,
      "epoch": 0.9797774637937124,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004910778485163811,
      "loss": 5.8275,
      "mean_token_accuracy": 0.16826068758964538,
      "num_tokens": 20684060.0,
      "step": 11095
    },
    {
      "entropy": 5.976723861694336,
      "epoch": 0.9802190038855528,
      "grad_norm": 2.609375,
      "learning_rate": 0.000491069066221711,
      "loss": 5.7471,
      "mean_token_accuracy": 0.1807445615530014,
      "num_tokens": 20693446.0,
      "step": 11100
    },
    {
      "entropy": 6.011556386947632,
      "epoch": 0.9806605439773931,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004910602796943707,
      "loss": 5.8807,
      "mean_token_accuracy": 0.1663028821349144,
      "num_tokens": 20702635.0,
      "step": 11105
    },
    {
      "entropy": 6.090901565551758,
      "epoch": 0.9811020840692335,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004910514889345323,
      "loss": 5.8358,
      "mean_token_accuracy": 0.16323625594377517,
      "num_tokens": 20712256.0,
      "step": 11110
    },
    {
      "entropy": 5.991445112228393,
      "epoch": 0.9815436241610739,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004910426939423682,
      "loss": 5.7859,
      "mean_token_accuracy": 0.16746611446142196,
      "num_tokens": 20720175.0,
      "step": 11115
    },
    {
      "entropy": 5.970749759674073,
      "epoch": 0.9819851642529142,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004910338947180506,
      "loss": 5.8054,
      "mean_token_accuracy": 0.16871384233236314,
      "num_tokens": 20729849.0,
      "step": 11120
    },
    {
      "entropy": 5.9835960388183596,
      "epoch": 0.9824267043447545,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004910250912617519,
      "loss": 5.7546,
      "mean_token_accuracy": 0.17132087796926498,
      "num_tokens": 20739129.0,
      "step": 11125
    },
    {
      "entropy": 6.016697931289673,
      "epoch": 0.9828682444365948,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004910162835736445,
      "loss": 5.709,
      "mean_token_accuracy": 0.17840465158224106,
      "num_tokens": 20748632.0,
      "step": 11130
    },
    {
      "entropy": 5.988829517364502,
      "epoch": 0.9833097845284352,
      "grad_norm": 2.421875,
      "learning_rate": 0.000491007471653901,
      "loss": 5.7538,
      "mean_token_accuracy": 0.1671848103404045,
      "num_tokens": 20758535.0,
      "step": 11135
    },
    {
      "entropy": 5.963017082214355,
      "epoch": 0.9837513246202755,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004909986555026939,
      "loss": 5.8005,
      "mean_token_accuracy": 0.17092676162719728,
      "num_tokens": 20768982.0,
      "step": 11140
    },
    {
      "entropy": 5.994900512695312,
      "epoch": 0.9841928647121159,
      "grad_norm": 2.21875,
      "learning_rate": 0.000490989835120196,
      "loss": 5.862,
      "mean_token_accuracy": 0.16857540905475615,
      "num_tokens": 20778936.0,
      "step": 11145
    },
    {
      "entropy": 6.042771339416504,
      "epoch": 0.9846344048039563,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004909810105065801,
      "loss": 5.848,
      "mean_token_accuracy": 0.16031375378370286,
      "num_tokens": 20788618.0,
      "step": 11150
    },
    {
      "entropy": 6.0676703453063965,
      "epoch": 0.9850759448957965,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004909721816620191,
      "loss": 5.8365,
      "mean_token_accuracy": 0.17483314573764802,
      "num_tokens": 20797721.0,
      "step": 11155
    },
    {
      "entropy": 6.043312644958496,
      "epoch": 0.9855174849876369,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004909633485866857,
      "loss": 5.7721,
      "mean_token_accuracy": 0.16904150396585466,
      "num_tokens": 20807305.0,
      "step": 11160
    },
    {
      "entropy": 5.888973760604858,
      "epoch": 0.9859590250794772,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004909545112807533,
      "loss": 5.7272,
      "mean_token_accuracy": 0.1731989398598671,
      "num_tokens": 20816984.0,
      "step": 11165
    },
    {
      "entropy": 6.0584144115448,
      "epoch": 0.9864005651713176,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004909456697443948,
      "loss": 5.8929,
      "mean_token_accuracy": 0.15854351669549943,
      "num_tokens": 20825738.0,
      "step": 11170
    },
    {
      "entropy": 6.052678203582763,
      "epoch": 0.9868421052631579,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004909368239777835,
      "loss": 5.8344,
      "mean_token_accuracy": 0.17343858778476715,
      "num_tokens": 20834135.0,
      "step": 11175
    },
    {
      "entropy": 6.089842700958252,
      "epoch": 0.9872836453549982,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004909279739810926,
      "loss": 5.9038,
      "mean_token_accuracy": 0.16258951276540756,
      "num_tokens": 20843689.0,
      "step": 11180
    },
    {
      "entropy": 6.050971508026123,
      "epoch": 0.9877251854468386,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004909191197544954,
      "loss": 5.7843,
      "mean_token_accuracy": 0.16911011934280396,
      "num_tokens": 20851724.0,
      "step": 11185
    },
    {
      "entropy": 6.092201471328735,
      "epoch": 0.9881667255386789,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004909102612981655,
      "loss": 5.9519,
      "mean_token_accuracy": 0.1617167681455612,
      "num_tokens": 20862325.0,
      "step": 11190
    },
    {
      "entropy": 6.003147792816162,
      "epoch": 0.9886082656305193,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004909013986122763,
      "loss": 5.7471,
      "mean_token_accuracy": 0.17576643824577332,
      "num_tokens": 20871046.0,
      "step": 11195
    },
    {
      "entropy": 6.023550748825073,
      "epoch": 0.9890498057223596,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004908925316970016,
      "loss": 5.7915,
      "mean_token_accuracy": 0.16211405843496324,
      "num_tokens": 20881605.0,
      "step": 11200
    },
    {
      "entropy": 5.917197895050049,
      "epoch": 0.9894913458142,
      "grad_norm": 2.265625,
      "learning_rate": 0.000490883660552515,
      "loss": 5.7289,
      "mean_token_accuracy": 0.16584061533212663,
      "num_tokens": 20891170.0,
      "step": 11205
    },
    {
      "entropy": 5.991090631484985,
      "epoch": 0.9899328859060402,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004908747851789902,
      "loss": 5.7403,
      "mean_token_accuracy": 0.15918954312801362,
      "num_tokens": 20900307.0,
      "step": 11210
    },
    {
      "entropy": 5.9911338806152346,
      "epoch": 0.9903744259978806,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004908659055766011,
      "loss": 5.8943,
      "mean_token_accuracy": 0.16472139954566956,
      "num_tokens": 20910623.0,
      "step": 11215
    },
    {
      "entropy": 6.066866540908814,
      "epoch": 0.990815966089721,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004908570217455218,
      "loss": 5.8166,
      "mean_token_accuracy": 0.1658707246184349,
      "num_tokens": 20919744.0,
      "step": 11220
    },
    {
      "entropy": 6.003478002548218,
      "epoch": 0.9912575061815613,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004908481336859262,
      "loss": 5.8639,
      "mean_token_accuracy": 0.171516814827919,
      "num_tokens": 20928701.0,
      "step": 11225
    },
    {
      "entropy": 5.999844121932983,
      "epoch": 0.9916990462734017,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004908392413979884,
      "loss": 5.8468,
      "mean_token_accuracy": 0.15534245297312738,
      "num_tokens": 20937786.0,
      "step": 11230
    },
    {
      "entropy": 5.970778274536133,
      "epoch": 0.9921405863652419,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004908303448818825,
      "loss": 5.7455,
      "mean_token_accuracy": 0.17506744116544723,
      "num_tokens": 20947214.0,
      "step": 11235
    },
    {
      "entropy": 5.949552249908447,
      "epoch": 0.9925821264570823,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000490821444137783,
      "loss": 5.854,
      "mean_token_accuracy": 0.1731862783432007,
      "num_tokens": 20956041.0,
      "step": 11240
    },
    {
      "entropy": 6.05905122756958,
      "epoch": 0.9930236665489226,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004908125391658641,
      "loss": 5.8267,
      "mean_token_accuracy": 0.16466867923736572,
      "num_tokens": 20966162.0,
      "step": 11245
    },
    {
      "entropy": 6.021831130981445,
      "epoch": 0.993465206640763,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004908036299663003,
      "loss": 5.8266,
      "mean_token_accuracy": 0.16720837503671646,
      "num_tokens": 20976025.0,
      "step": 11250
    },
    {
      "entropy": 6.02084493637085,
      "epoch": 0.9939067467326034,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004907947165392662,
      "loss": 5.8584,
      "mean_token_accuracy": 0.16399545222520828,
      "num_tokens": 20985606.0,
      "step": 11255
    },
    {
      "entropy": 6.0483697891235355,
      "epoch": 0.9943482868244437,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004907857988849363,
      "loss": 5.8316,
      "mean_token_accuracy": 0.16402145475149155,
      "num_tokens": 20994308.0,
      "step": 11260
    },
    {
      "entropy": 5.948052406311035,
      "epoch": 0.994789826916284,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004907768770034853,
      "loss": 5.756,
      "mean_token_accuracy": 0.17144669741392135,
      "num_tokens": 21003783.0,
      "step": 11265
    },
    {
      "entropy": 6.023220729827881,
      "epoch": 0.9952313670081243,
      "grad_norm": 1.578125,
      "learning_rate": 0.000490767950895088,
      "loss": 5.7298,
      "mean_token_accuracy": 0.16765106618404388,
      "num_tokens": 21012870.0,
      "step": 11270
    },
    {
      "entropy": 6.064298534393311,
      "epoch": 0.9956729070999647,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004907590205599192,
      "loss": 5.8202,
      "mean_token_accuracy": 0.16320745795965194,
      "num_tokens": 21021987.0,
      "step": 11275
    },
    {
      "entropy": 5.932501220703125,
      "epoch": 0.996114447191805,
      "grad_norm": 2.3125,
      "learning_rate": 0.000490750085998154,
      "loss": 5.8329,
      "mean_token_accuracy": 0.16234722584486008,
      "num_tokens": 21030977.0,
      "step": 11280
    },
    {
      "entropy": 5.986550569534302,
      "epoch": 0.9965559872836454,
      "grad_norm": 1.5,
      "learning_rate": 0.0004907411472099672,
      "loss": 5.756,
      "mean_token_accuracy": 0.16745672076940538,
      "num_tokens": 21039259.0,
      "step": 11285
    },
    {
      "entropy": 6.030724573135376,
      "epoch": 0.9969975273754857,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004907322041955341,
      "loss": 5.8425,
      "mean_token_accuracy": 0.1665562704205513,
      "num_tokens": 21049321.0,
      "step": 11290
    },
    {
      "entropy": 6.031997394561768,
      "epoch": 0.997439067467326,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004907232569550298,
      "loss": 5.9026,
      "mean_token_accuracy": 0.15812076702713967,
      "num_tokens": 21059469.0,
      "step": 11295
    },
    {
      "entropy": 6.024282503128052,
      "epoch": 0.9978806075591664,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004907143054886297,
      "loss": 5.8388,
      "mean_token_accuracy": 0.16478039026260377,
      "num_tokens": 21068694.0,
      "step": 11300
    },
    {
      "entropy": 6.020016002655029,
      "epoch": 0.9983221476510067,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000490705349796509,
      "loss": 5.7754,
      "mean_token_accuracy": 0.1716444805264473,
      "num_tokens": 21077232.0,
      "step": 11305
    },
    {
      "entropy": 6.047950696945191,
      "epoch": 0.9987636877428471,
      "grad_norm": 1.5,
      "learning_rate": 0.0004906963898788431,
      "loss": 5.7888,
      "mean_token_accuracy": 0.16969122588634492,
      "num_tokens": 21086517.0,
      "step": 11310
    },
    {
      "entropy": 5.983556652069092,
      "epoch": 0.9992052278346873,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004906874257358075,
      "loss": 5.8444,
      "mean_token_accuracy": 0.17099157720804214,
      "num_tokens": 21096969.0,
      "step": 11315
    },
    {
      "entropy": 5.980427312850952,
      "epoch": 0.9996467679265277,
      "grad_norm": 1.9140625,
      "learning_rate": 0.000490678457367578,
      "loss": 5.7915,
      "mean_token_accuracy": 0.16400003731250762,
      "num_tokens": 21106554.0,
      "step": 11320
    },
    {
      "entropy": 6.017829036712646,
      "epoch": 1.000088308018368,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004906694847743302,
      "loss": 5.7764,
      "mean_token_accuracy": 0.16667501181364058,
      "num_tokens": 21114829.0,
      "step": 11325
    },
    {
      "entropy": 6.032284355163574,
      "epoch": 1.0005298481102085,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004906605079562399,
      "loss": 5.7214,
      "mean_token_accuracy": 0.1728486344218254,
      "num_tokens": 21123568.0,
      "step": 11330
    },
    {
      "entropy": 5.998350667953491,
      "epoch": 1.0009713882020488,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004906515269134827,
      "loss": 5.7827,
      "mean_token_accuracy": 0.16798163652420045,
      "num_tokens": 21132811.0,
      "step": 11335
    },
    {
      "entropy": 5.996047115325927,
      "epoch": 1.001412928293889,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004906425416462349,
      "loss": 5.8278,
      "mean_token_accuracy": 0.16611848026514053,
      "num_tokens": 21141549.0,
      "step": 11340
    },
    {
      "entropy": 6.016817903518676,
      "epoch": 1.0018544683857293,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004906335521546723,
      "loss": 5.7236,
      "mean_token_accuracy": 0.17209179252386092,
      "num_tokens": 21150603.0,
      "step": 11345
    },
    {
      "entropy": 5.8438413619995115,
      "epoch": 1.0022960084775698,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004906245584389712,
      "loss": 5.5472,
      "mean_token_accuracy": 0.18766358941793443,
      "num_tokens": 21159842.0,
      "step": 11350
    },
    {
      "entropy": 5.922256517410278,
      "epoch": 1.00273754856941,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004906155604993075,
      "loss": 5.7026,
      "mean_token_accuracy": 0.1780036374926567,
      "num_tokens": 21169139.0,
      "step": 11355
    },
    {
      "entropy": 5.837229537963867,
      "epoch": 1.0031790886612504,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004906065583358576,
      "loss": 5.6767,
      "mean_token_accuracy": 0.17509741634130477,
      "num_tokens": 21178059.0,
      "step": 11360
    },
    {
      "entropy": 5.94249906539917,
      "epoch": 1.0036206287530909,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004905975519487978,
      "loss": 5.7086,
      "mean_token_accuracy": 0.17003520131111144,
      "num_tokens": 21187966.0,
      "step": 11365
    },
    {
      "entropy": 6.005231332778931,
      "epoch": 1.0040621688449312,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004905885413383046,
      "loss": 5.7315,
      "mean_token_accuracy": 0.16974558532238007,
      "num_tokens": 21197460.0,
      "step": 11370
    },
    {
      "entropy": 6.004400444030762,
      "epoch": 1.0045037089367714,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004905795265045546,
      "loss": 5.7716,
      "mean_token_accuracy": 0.16958380937576295,
      "num_tokens": 21206977.0,
      "step": 11375
    },
    {
      "entropy": 5.978320980072022,
      "epoch": 1.0049452490286117,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004905705074477242,
      "loss": 5.7035,
      "mean_token_accuracy": 0.16527058482170104,
      "num_tokens": 21216046.0,
      "step": 11380
    },
    {
      "entropy": 5.938219118118286,
      "epoch": 1.0053867891204522,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004905614841679901,
      "loss": 5.7557,
      "mean_token_accuracy": 0.1728820785880089,
      "num_tokens": 21225612.0,
      "step": 11385
    },
    {
      "entropy": 5.942444133758545,
      "epoch": 1.0058283292122925,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004905524566655292,
      "loss": 5.6532,
      "mean_token_accuracy": 0.17415272444486618,
      "num_tokens": 21233641.0,
      "step": 11390
    },
    {
      "entropy": 5.9322888374328615,
      "epoch": 1.0062698693041328,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004905434249405183,
      "loss": 5.7228,
      "mean_token_accuracy": 0.1694304198026657,
      "num_tokens": 21243444.0,
      "step": 11395
    },
    {
      "entropy": 5.924720191955567,
      "epoch": 1.0067114093959733,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004905343889931342,
      "loss": 5.6807,
      "mean_token_accuracy": 0.17966057658195494,
      "num_tokens": 21253473.0,
      "step": 11400
    },
    {
      "entropy": 5.926651477813721,
      "epoch": 1.0071529494878135,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000490525348823554,
      "loss": 5.6558,
      "mean_token_accuracy": 0.18528196215629578,
      "num_tokens": 21262049.0,
      "step": 11405
    },
    {
      "entropy": 5.916750860214234,
      "epoch": 1.0075944895796538,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004905163044319549,
      "loss": 5.6808,
      "mean_token_accuracy": 0.17246959656476973,
      "num_tokens": 21270909.0,
      "step": 11410
    },
    {
      "entropy": 5.938456106185913,
      "epoch": 1.008036029671494,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004905072558185139,
      "loss": 5.7681,
      "mean_token_accuracy": 0.17287649512290953,
      "num_tokens": 21280503.0,
      "step": 11415
    },
    {
      "entropy": 6.024928188323974,
      "epoch": 1.0084775697633346,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004904982029834083,
      "loss": 5.7682,
      "mean_token_accuracy": 0.1688655585050583,
      "num_tokens": 21289957.0,
      "step": 11420
    },
    {
      "entropy": 5.956943130493164,
      "epoch": 1.0089191098551749,
      "grad_norm": 1.75,
      "learning_rate": 0.0004904891459268154,
      "loss": 5.7434,
      "mean_token_accuracy": 0.1691091001033783,
      "num_tokens": 21298799.0,
      "step": 11425
    },
    {
      "entropy": 5.936780691146851,
      "epoch": 1.0093606499470151,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004904800846489128,
      "loss": 5.7479,
      "mean_token_accuracy": 0.16977997422218322,
      "num_tokens": 21308251.0,
      "step": 11430
    },
    {
      "entropy": 6.027180004119873,
      "epoch": 1.0098021900388556,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004904710191498779,
      "loss": 5.7704,
      "mean_token_accuracy": 0.16857996135950087,
      "num_tokens": 21317705.0,
      "step": 11435
    },
    {
      "entropy": 5.952654600143433,
      "epoch": 1.010243730130696,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004904619494298882,
      "loss": 5.744,
      "mean_token_accuracy": 0.16214617639780043,
      "num_tokens": 21326867.0,
      "step": 11440
    },
    {
      "entropy": 5.956965589523316,
      "epoch": 1.0106852702225362,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004904528754891215,
      "loss": 5.7537,
      "mean_token_accuracy": 0.17429810911417007,
      "num_tokens": 21336054.0,
      "step": 11445
    },
    {
      "entropy": 5.939871788024902,
      "epoch": 1.0111268103143765,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004904437973277555,
      "loss": 5.693,
      "mean_token_accuracy": 0.17177852243185043,
      "num_tokens": 21344457.0,
      "step": 11450
    },
    {
      "entropy": 6.033869028091431,
      "epoch": 1.011568350406217,
      "grad_norm": 2.125,
      "learning_rate": 0.0004904347149459681,
      "loss": 5.7803,
      "mean_token_accuracy": 0.17063615769147872,
      "num_tokens": 21353639.0,
      "step": 11455
    },
    {
      "entropy": 5.8367383003234865,
      "epoch": 1.0120098904980572,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004904256283439371,
      "loss": 5.5881,
      "mean_token_accuracy": 0.1793370082974434,
      "num_tokens": 21362005.0,
      "step": 11460
    },
    {
      "entropy": 5.990533638000488,
      "epoch": 1.0124514305898975,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004904165375218405,
      "loss": 5.756,
      "mean_token_accuracy": 0.17171766459941865,
      "num_tokens": 21371917.0,
      "step": 11465
    },
    {
      "entropy": 6.030205631256104,
      "epoch": 1.012892970681738,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004904074424798565,
      "loss": 5.7774,
      "mean_token_accuracy": 0.16853656023740768,
      "num_tokens": 21381500.0,
      "step": 11470
    },
    {
      "entropy": 5.9031003475189205,
      "epoch": 1.0133345107735783,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004903983432181632,
      "loss": 5.6027,
      "mean_token_accuracy": 0.18293141275644303,
      "num_tokens": 21391280.0,
      "step": 11475
    },
    {
      "entropy": 5.960421705245972,
      "epoch": 1.0137760508654186,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004903892397369388,
      "loss": 5.7214,
      "mean_token_accuracy": 0.17304562628269196,
      "num_tokens": 21400371.0,
      "step": 11480
    },
    {
      "entropy": 5.857281732559204,
      "epoch": 1.0142175909572588,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004903801320363618,
      "loss": 5.7377,
      "mean_token_accuracy": 0.17329668998718262,
      "num_tokens": 21410493.0,
      "step": 11485
    },
    {
      "entropy": 6.04967622756958,
      "epoch": 1.0146591310490993,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004903710201166105,
      "loss": 5.8371,
      "mean_token_accuracy": 0.15925003439188004,
      "num_tokens": 21420516.0,
      "step": 11490
    },
    {
      "entropy": 6.017249536514282,
      "epoch": 1.0151006711409396,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004903619039778634,
      "loss": 5.7631,
      "mean_token_accuracy": 0.16286804229021073,
      "num_tokens": 21429614.0,
      "step": 11495
    },
    {
      "entropy": 5.978137826919555,
      "epoch": 1.0155422112327799,
      "grad_norm": 1.921875,
      "learning_rate": 0.000490352783620299,
      "loss": 5.7261,
      "mean_token_accuracy": 0.16925765424966813,
      "num_tokens": 21438595.0,
      "step": 11500
    },
    {
      "entropy": 6.019017267227173,
      "epoch": 1.0159837513246204,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004903436590440961,
      "loss": 5.758,
      "mean_token_accuracy": 0.1755782648921013,
      "num_tokens": 21447708.0,
      "step": 11505
    },
    {
      "entropy": 5.963498783111572,
      "epoch": 1.0164252914164607,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004903345302494334,
      "loss": 5.7184,
      "mean_token_accuracy": 0.17244312018156052,
      "num_tokens": 21457210.0,
      "step": 11510
    },
    {
      "entropy": 5.895744132995605,
      "epoch": 1.016866831508301,
      "grad_norm": 1.75,
      "learning_rate": 0.0004903253972364897,
      "loss": 5.7202,
      "mean_token_accuracy": 0.16742899119853974,
      "num_tokens": 21466247.0,
      "step": 11515
    },
    {
      "entropy": 5.960920906066894,
      "epoch": 1.0173083716001412,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004903162600054439,
      "loss": 5.6751,
      "mean_token_accuracy": 0.17459392100572585,
      "num_tokens": 21475634.0,
      "step": 11520
    },
    {
      "entropy": 6.021289539337158,
      "epoch": 1.0177499116919817,
      "grad_norm": 1.875,
      "learning_rate": 0.0004903071185564751,
      "loss": 5.8107,
      "mean_token_accuracy": 0.17013828456401825,
      "num_tokens": 21485523.0,
      "step": 11525
    },
    {
      "entropy": 5.998600625991822,
      "epoch": 1.018191451783822,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004902979728897623,
      "loss": 5.7563,
      "mean_token_accuracy": 0.17158716171979904,
      "num_tokens": 21494909.0,
      "step": 11530
    },
    {
      "entropy": 6.029300785064697,
      "epoch": 1.0186329918756623,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004902888230054846,
      "loss": 5.7698,
      "mean_token_accuracy": 0.17082602232694627,
      "num_tokens": 21504413.0,
      "step": 11535
    },
    {
      "entropy": 5.883685827255249,
      "epoch": 1.0190745319675028,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004902796689038213,
      "loss": 5.6508,
      "mean_token_accuracy": 0.17973359674215317,
      "num_tokens": 21514068.0,
      "step": 11540
    },
    {
      "entropy": 5.902773189544678,
      "epoch": 1.019516072059343,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004902705105849517,
      "loss": 5.6712,
      "mean_token_accuracy": 0.17633067667484284,
      "num_tokens": 21523536.0,
      "step": 11545
    },
    {
      "entropy": 5.916357421875,
      "epoch": 1.0199576121511833,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004902613480490553,
      "loss": 5.7553,
      "mean_token_accuracy": 0.16829731017351152,
      "num_tokens": 21532554.0,
      "step": 11550
    },
    {
      "entropy": 5.904748344421387,
      "epoch": 1.0203991522430236,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004902521812963116,
      "loss": 5.6412,
      "mean_token_accuracy": 0.17432893067598343,
      "num_tokens": 21542508.0,
      "step": 11555
    },
    {
      "entropy": 6.001231288909912,
      "epoch": 1.020840692334864,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004902430103268998,
      "loss": 5.7336,
      "mean_token_accuracy": 0.16080355644226074,
      "num_tokens": 21551954.0,
      "step": 11560
    },
    {
      "entropy": 5.9578272819519045,
      "epoch": 1.0212822324267044,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004902338351410002,
      "loss": 5.6766,
      "mean_token_accuracy": 0.17667103558778763,
      "num_tokens": 21561676.0,
      "step": 11565
    },
    {
      "entropy": 5.925234651565551,
      "epoch": 1.0217237725185446,
      "grad_norm": 2.34375,
      "learning_rate": 0.000490224655738792,
      "loss": 5.7927,
      "mean_token_accuracy": 0.1612503260374069,
      "num_tokens": 21571916.0,
      "step": 11570
    },
    {
      "entropy": 5.990692281723023,
      "epoch": 1.0221653126103851,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004902154721204552,
      "loss": 5.7359,
      "mean_token_accuracy": 0.1818292886018753,
      "num_tokens": 21581892.0,
      "step": 11575
    },
    {
      "entropy": 5.984904384613037,
      "epoch": 1.0226068527022254,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004902062842861696,
      "loss": 5.6871,
      "mean_token_accuracy": 0.17149700671434404,
      "num_tokens": 21591390.0,
      "step": 11580
    },
    {
      "entropy": 5.986855792999267,
      "epoch": 1.0230483927940657,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004901970922361154,
      "loss": 5.7468,
      "mean_token_accuracy": 0.18060247749090194,
      "num_tokens": 21600766.0,
      "step": 11585
    },
    {
      "entropy": 5.908822011947632,
      "epoch": 1.023489932885906,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004901878959704726,
      "loss": 5.6908,
      "mean_token_accuracy": 0.18095418363809584,
      "num_tokens": 21609264.0,
      "step": 11590
    },
    {
      "entropy": 5.983219861984253,
      "epoch": 1.0239314729777464,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004901786954894212,
      "loss": 5.7692,
      "mean_token_accuracy": 0.1688540369272232,
      "num_tokens": 21618409.0,
      "step": 11595
    },
    {
      "entropy": 5.90957407951355,
      "epoch": 1.0243730130695867,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004901694907931416,
      "loss": 5.6592,
      "mean_token_accuracy": 0.1680929258465767,
      "num_tokens": 21628158.0,
      "step": 11600
    },
    {
      "entropy": 5.991635131835937,
      "epoch": 1.024814553161427,
      "grad_norm": 1.71875,
      "learning_rate": 0.000490160281881814,
      "loss": 5.7875,
      "mean_token_accuracy": 0.17219406813383104,
      "num_tokens": 21637859.0,
      "step": 11605
    },
    {
      "entropy": 6.008704280853271,
      "epoch": 1.0252560932532675,
      "grad_norm": 1.875,
      "learning_rate": 0.0004901510687556188,
      "loss": 5.7749,
      "mean_token_accuracy": 0.16721676737070085,
      "num_tokens": 21647389.0,
      "step": 11610
    },
    {
      "entropy": 6.004774379730224,
      "epoch": 1.0256976333451078,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004901418514147367,
      "loss": 5.7352,
      "mean_token_accuracy": 0.16795544773340226,
      "num_tokens": 21655857.0,
      "step": 11615
    },
    {
      "entropy": 5.949306964874268,
      "epoch": 1.026139173436948,
      "grad_norm": 2.53125,
      "learning_rate": 0.000490132629859348,
      "loss": 5.6933,
      "mean_token_accuracy": 0.17819422036409377,
      "num_tokens": 21666150.0,
      "step": 11620
    },
    {
      "entropy": 5.954881620407105,
      "epoch": 1.0265807135287883,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004901234040896334,
      "loss": 5.7126,
      "mean_token_accuracy": 0.1698005437850952,
      "num_tokens": 21675662.0,
      "step": 11625
    },
    {
      "entropy": 6.002849912643432,
      "epoch": 1.0270222536206288,
      "grad_norm": 1.625,
      "learning_rate": 0.0004901141741057737,
      "loss": 5.7774,
      "mean_token_accuracy": 0.16578176319599153,
      "num_tokens": 21685721.0,
      "step": 11630
    },
    {
      "entropy": 6.012971496582031,
      "epoch": 1.027463793712469,
      "grad_norm": 1.5,
      "learning_rate": 0.0004901049399079495,
      "loss": 5.7504,
      "mean_token_accuracy": 0.1728194072842598,
      "num_tokens": 21694507.0,
      "step": 11635
    },
    {
      "entropy": 5.99225344657898,
      "epoch": 1.0279053338043094,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004900957014963421,
      "loss": 5.7583,
      "mean_token_accuracy": 0.18061111122369766,
      "num_tokens": 21704092.0,
      "step": 11640
    },
    {
      "entropy": 5.809380340576172,
      "epoch": 1.0283468738961499,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004900864588711321,
      "loss": 5.6694,
      "mean_token_accuracy": 0.18092458546161652,
      "num_tokens": 21713313.0,
      "step": 11645
    },
    {
      "entropy": 5.9611798286437985,
      "epoch": 1.0287884139879901,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004900772120325008,
      "loss": 5.7366,
      "mean_token_accuracy": 0.17457195073366166,
      "num_tokens": 21721622.0,
      "step": 11650
    },
    {
      "entropy": 5.97767071723938,
      "epoch": 1.0292299540798304,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004900679609806291,
      "loss": 5.6707,
      "mean_token_accuracy": 0.17381463348865508,
      "num_tokens": 21730914.0,
      "step": 11655
    },
    {
      "entropy": 5.980806922912597,
      "epoch": 1.0296714941716707,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004900587057156985,
      "loss": 5.6692,
      "mean_token_accuracy": 0.17887462824583053,
      "num_tokens": 21740124.0,
      "step": 11660
    },
    {
      "entropy": 5.933705759048462,
      "epoch": 1.0301130342635112,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004900494462378902,
      "loss": 5.7525,
      "mean_token_accuracy": 0.16688884198665618,
      "num_tokens": 21750013.0,
      "step": 11665
    },
    {
      "entropy": 5.890878772735595,
      "epoch": 1.0305545743553515,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004900401825473854,
      "loss": 5.7507,
      "mean_token_accuracy": 0.17413905560970305,
      "num_tokens": 21758842.0,
      "step": 11670
    },
    {
      "entropy": 5.927459049224853,
      "epoch": 1.0309961144471917,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004900309146443658,
      "loss": 5.6536,
      "mean_token_accuracy": 0.1740245521068573,
      "num_tokens": 21767002.0,
      "step": 11675
    },
    {
      "entropy": 5.974818420410156,
      "epoch": 1.0314376545390322,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004900216425290128,
      "loss": 5.7696,
      "mean_token_accuracy": 0.16744465827941896,
      "num_tokens": 21775372.0,
      "step": 11680
    },
    {
      "entropy": 5.955351209640503,
      "epoch": 1.0318791946308725,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004900123662015082,
      "loss": 5.7685,
      "mean_token_accuracy": 0.16843886077404022,
      "num_tokens": 21784541.0,
      "step": 11685
    },
    {
      "entropy": 5.948904180526734,
      "epoch": 1.0323207347227128,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004900030856620336,
      "loss": 5.6529,
      "mean_token_accuracy": 0.18304046094417573,
      "num_tokens": 21794525.0,
      "step": 11690
    },
    {
      "entropy": 5.963394546508789,
      "epoch": 1.032762274814553,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004899938009107708,
      "loss": 5.758,
      "mean_token_accuracy": 0.17228179275989533,
      "num_tokens": 21804469.0,
      "step": 11695
    },
    {
      "entropy": 5.917418718338013,
      "epoch": 1.0332038149063936,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004899845119479019,
      "loss": 5.6985,
      "mean_token_accuracy": 0.1747218430042267,
      "num_tokens": 21813986.0,
      "step": 11700
    },
    {
      "entropy": 5.9892199516296385,
      "epoch": 1.0336453549982338,
      "grad_norm": 1.625,
      "learning_rate": 0.0004899752187736085,
      "loss": 5.7045,
      "mean_token_accuracy": 0.17282750010490416,
      "num_tokens": 21824083.0,
      "step": 11705
    },
    {
      "entropy": 5.891262626647949,
      "epoch": 1.0340868950900741,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004899659213880728,
      "loss": 5.6508,
      "mean_token_accuracy": 0.1831414967775345,
      "num_tokens": 21833526.0,
      "step": 11710
    },
    {
      "entropy": 5.944592094421386,
      "epoch": 1.0345284351819146,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004899566197914771,
      "loss": 5.6768,
      "mean_token_accuracy": 0.17587029933929443,
      "num_tokens": 21842537.0,
      "step": 11715
    },
    {
      "entropy": 5.985992670059204,
      "epoch": 1.034969975273755,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004899473139840033,
      "loss": 5.7212,
      "mean_token_accuracy": 0.17308437824249268,
      "num_tokens": 21851604.0,
      "step": 11720
    },
    {
      "entropy": 5.953021097183227,
      "epoch": 1.0354115153655952,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004899380039658339,
      "loss": 5.7686,
      "mean_token_accuracy": 0.16968014240264892,
      "num_tokens": 21861480.0,
      "step": 11725
    },
    {
      "entropy": 5.939228630065918,
      "epoch": 1.0358530554574354,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004899286897371513,
      "loss": 5.7478,
      "mean_token_accuracy": 0.17233823537826537,
      "num_tokens": 21870901.0,
      "step": 11730
    },
    {
      "entropy": 5.8766419887542725,
      "epoch": 1.036294595549276,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000489919371298138,
      "loss": 5.6283,
      "mean_token_accuracy": 0.18169922679662703,
      "num_tokens": 21880491.0,
      "step": 11735
    },
    {
      "entropy": 5.890109491348267,
      "epoch": 1.0367361356411162,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004899100486489763,
      "loss": 5.6946,
      "mean_token_accuracy": 0.17681025713682175,
      "num_tokens": 21889833.0,
      "step": 11740
    },
    {
      "entropy": 5.895658111572265,
      "epoch": 1.0371776757329565,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004899007217898491,
      "loss": 5.7354,
      "mean_token_accuracy": 0.17619029581546783,
      "num_tokens": 21899222.0,
      "step": 11745
    },
    {
      "entropy": 6.013517904281616,
      "epoch": 1.037619215824797,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004898913907209389,
      "loss": 5.7485,
      "mean_token_accuracy": 0.17189059853553773,
      "num_tokens": 21908848.0,
      "step": 11750
    },
    {
      "entropy": 5.9578900814056395,
      "epoch": 1.0380607559166373,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004898820554424285,
      "loss": 5.6781,
      "mean_token_accuracy": 0.16759795993566512,
      "num_tokens": 21917867.0,
      "step": 11755
    },
    {
      "entropy": 5.976204299926758,
      "epoch": 1.0385022960084775,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000489872715954501,
      "loss": 5.7706,
      "mean_token_accuracy": 0.16450867652893067,
      "num_tokens": 21927476.0,
      "step": 11760
    },
    {
      "entropy": 6.0786598205566404,
      "epoch": 1.0389438361003178,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004898633722573391,
      "loss": 5.8377,
      "mean_token_accuracy": 0.1652936100959778,
      "num_tokens": 21937068.0,
      "step": 11765
    },
    {
      "entropy": 5.9219505310058596,
      "epoch": 1.0393853761921583,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000489854024351126,
      "loss": 5.7114,
      "mean_token_accuracy": 0.17316484302282334,
      "num_tokens": 21946144.0,
      "step": 11770
    },
    {
      "entropy": 5.972825527191162,
      "epoch": 1.0398269162839986,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004898446722360446,
      "loss": 5.6542,
      "mean_token_accuracy": 0.17927841544151307,
      "num_tokens": 21955346.0,
      "step": 11775
    },
    {
      "entropy": 5.917709684371948,
      "epoch": 1.0402684563758389,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004898353159122784,
      "loss": 5.7345,
      "mean_token_accuracy": 0.1762217789888382,
      "num_tokens": 21965034.0,
      "step": 11780
    },
    {
      "entropy": 5.915597438812256,
      "epoch": 1.0407099964676794,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004898259553800106,
      "loss": 5.7021,
      "mean_token_accuracy": 0.17157121747732162,
      "num_tokens": 21975167.0,
      "step": 11785
    },
    {
      "entropy": 6.001184701919556,
      "epoch": 1.0411515365595196,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004898165906394244,
      "loss": 5.7073,
      "mean_token_accuracy": 0.17095027565956117,
      "num_tokens": 21984372.0,
      "step": 11790
    },
    {
      "entropy": 5.937566328048706,
      "epoch": 1.04159307665136,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004898072216907033,
      "loss": 5.7015,
      "mean_token_accuracy": 0.17414507269859314,
      "num_tokens": 21993513.0,
      "step": 11795
    },
    {
      "entropy": 5.936249399185181,
      "epoch": 1.0420346167432002,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004897978485340311,
      "loss": 5.7734,
      "mean_token_accuracy": 0.16741596907377243,
      "num_tokens": 22002593.0,
      "step": 11800
    },
    {
      "entropy": 6.005516290664673,
      "epoch": 1.0424761568350407,
      "grad_norm": 1.671875,
      "learning_rate": 0.000489788471169591,
      "loss": 5.7907,
      "mean_token_accuracy": 0.16941866427659988,
      "num_tokens": 22012296.0,
      "step": 11805
    },
    {
      "entropy": 5.9861159324646,
      "epoch": 1.042917696926881,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004897790895975671,
      "loss": 5.8051,
      "mean_token_accuracy": 0.16327016949653625,
      "num_tokens": 22021267.0,
      "step": 11810
    },
    {
      "entropy": 5.980770444869995,
      "epoch": 1.0433592370187212,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004897697038181428,
      "loss": 5.7768,
      "mean_token_accuracy": 0.1721298351883888,
      "num_tokens": 22030878.0,
      "step": 11815
    },
    {
      "entropy": 5.923258638381958,
      "epoch": 1.0438007771105617,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004897603138315022,
      "loss": 5.7554,
      "mean_token_accuracy": 0.17359738200902938,
      "num_tokens": 22040245.0,
      "step": 11820
    },
    {
      "entropy": 5.975063180923462,
      "epoch": 1.044242317202402,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004897509196378293,
      "loss": 5.7395,
      "mean_token_accuracy": 0.1754832774400711,
      "num_tokens": 22050169.0,
      "step": 11825
    },
    {
      "entropy": 6.012202167510987,
      "epoch": 1.0446838572942423,
      "grad_norm": 2.5,
      "learning_rate": 0.000489741521237308,
      "loss": 5.7602,
      "mean_token_accuracy": 0.17230364978313445,
      "num_tokens": 22059168.0,
      "step": 11830
    },
    {
      "entropy": 5.910138893127441,
      "epoch": 1.0451253973860826,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004897321186301223,
      "loss": 5.6953,
      "mean_token_accuracy": 0.16782840788364412,
      "num_tokens": 22068326.0,
      "step": 11835
    },
    {
      "entropy": 5.933950471878052,
      "epoch": 1.045566937477923,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004897227118164566,
      "loss": 5.7332,
      "mean_token_accuracy": 0.17260353565216063,
      "num_tokens": 22078026.0,
      "step": 11840
    },
    {
      "entropy": 5.97835111618042,
      "epoch": 1.0460084775697633,
      "grad_norm": 2.5625,
      "learning_rate": 0.000489713300796495,
      "loss": 5.7714,
      "mean_token_accuracy": 0.16629578769207,
      "num_tokens": 22088230.0,
      "step": 11845
    },
    {
      "entropy": 5.906348705291748,
      "epoch": 1.0464500176616036,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004897038855704221,
      "loss": 5.5866,
      "mean_token_accuracy": 0.18455611318349838,
      "num_tokens": 22096667.0,
      "step": 11850
    },
    {
      "entropy": 5.899099445343017,
      "epoch": 1.0468915577534441,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004896944661384222,
      "loss": 5.7253,
      "mean_token_accuracy": 0.1725448414683342,
      "num_tokens": 22106871.0,
      "step": 11855
    },
    {
      "entropy": 5.929962873458862,
      "epoch": 1.0473330978452844,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004896850425006797,
      "loss": 5.6834,
      "mean_token_accuracy": 0.17062857151031494,
      "num_tokens": 22114954.0,
      "step": 11860
    },
    {
      "entropy": 5.962481164932251,
      "epoch": 1.0477746379371247,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004896756146573794,
      "loss": 5.7015,
      "mean_token_accuracy": 0.17050741910934447,
      "num_tokens": 22125293.0,
      "step": 11865
    },
    {
      "entropy": 6.005695104598999,
      "epoch": 1.048216178028965,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004896661826087059,
      "loss": 5.7319,
      "mean_token_accuracy": 0.17089567333459854,
      "num_tokens": 22133731.0,
      "step": 11870
    },
    {
      "entropy": 5.907718896865845,
      "epoch": 1.0486577181208054,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004896567463548439,
      "loss": 5.732,
      "mean_token_accuracy": 0.16580215096473694,
      "num_tokens": 22143007.0,
      "step": 11875
    },
    {
      "entropy": 5.963164615631103,
      "epoch": 1.0490992582126457,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004896473058959783,
      "loss": 5.736,
      "mean_token_accuracy": 0.17010460942983627,
      "num_tokens": 22152044.0,
      "step": 11880
    },
    {
      "entropy": 6.050265264511109,
      "epoch": 1.049540798304486,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004896378612322942,
      "loss": 5.8042,
      "mean_token_accuracy": 0.16925871819257737,
      "num_tokens": 22161473.0,
      "step": 11885
    },
    {
      "entropy": 5.987928819656372,
      "epoch": 1.0499823383963265,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004896284123639763,
      "loss": 5.7358,
      "mean_token_accuracy": 0.16610346138477325,
      "num_tokens": 22171704.0,
      "step": 11890
    },
    {
      "entropy": 5.939279222488404,
      "epoch": 1.0504238784881668,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00048961895929121,
      "loss": 5.7271,
      "mean_token_accuracy": 0.16976826190948485,
      "num_tokens": 22180593.0,
      "step": 11895
    },
    {
      "entropy": 5.901366949081421,
      "epoch": 1.050865418580007,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004896095020141802,
      "loss": 5.6908,
      "mean_token_accuracy": 0.17781361937522888,
      "num_tokens": 22189552.0,
      "step": 11900
    },
    {
      "entropy": 5.954809141159058,
      "epoch": 1.0513069586718473,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004896000405330724,
      "loss": 5.7257,
      "mean_token_accuracy": 0.16914680749177932,
      "num_tokens": 22199314.0,
      "step": 11905
    },
    {
      "entropy": 5.900572395324707,
      "epoch": 1.0517484987636878,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000489590574848072,
      "loss": 5.6724,
      "mean_token_accuracy": 0.17794393748044968,
      "num_tokens": 22208128.0,
      "step": 11910
    },
    {
      "entropy": 5.897657966613769,
      "epoch": 1.052190038855528,
      "grad_norm": 2.25,
      "learning_rate": 0.000489581104959364,
      "loss": 5.6537,
      "mean_token_accuracy": 0.1821790114045143,
      "num_tokens": 22217700.0,
      "step": 11915
    },
    {
      "entropy": 5.941938734054565,
      "epoch": 1.0526315789473684,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004895716308671343,
      "loss": 5.7075,
      "mean_token_accuracy": 0.1744215413928032,
      "num_tokens": 22225941.0,
      "step": 11920
    },
    {
      "entropy": 5.920284128189087,
      "epoch": 1.0530731190392089,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004895621525715685,
      "loss": 5.6923,
      "mean_token_accuracy": 0.17301340997219086,
      "num_tokens": 22234935.0,
      "step": 11925
    },
    {
      "entropy": 5.934741544723511,
      "epoch": 1.0535146591310491,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004895526700728521,
      "loss": 5.7668,
      "mean_token_accuracy": 0.17099023312330247,
      "num_tokens": 22244087.0,
      "step": 11930
    },
    {
      "entropy": 5.9216385841369625,
      "epoch": 1.0539561992228894,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004895431833711708,
      "loss": 5.7086,
      "mean_token_accuracy": 0.16913665533065797,
      "num_tokens": 22254277.0,
      "step": 11935
    },
    {
      "entropy": 5.9343325138092045,
      "epoch": 1.0543977393147297,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004895336924667107,
      "loss": 5.6855,
      "mean_token_accuracy": 0.17650166749954224,
      "num_tokens": 22263150.0,
      "step": 11940
    },
    {
      "entropy": 5.946196985244751,
      "epoch": 1.0548392794065702,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004895241973596576,
      "loss": 5.6883,
      "mean_token_accuracy": 0.1718298003077507,
      "num_tokens": 22271973.0,
      "step": 11945
    },
    {
      "entropy": 5.884443902969361,
      "epoch": 1.0552808194984105,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004895146980501973,
      "loss": 5.6761,
      "mean_token_accuracy": 0.17968894690275192,
      "num_tokens": 22280637.0,
      "step": 11950
    },
    {
      "entropy": 5.889499378204346,
      "epoch": 1.0557223595902507,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004895051945385163,
      "loss": 5.6969,
      "mean_token_accuracy": 0.169900843501091,
      "num_tokens": 22290601.0,
      "step": 11955
    },
    {
      "entropy": 5.975777435302734,
      "epoch": 1.0561638996820912,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004894956868248004,
      "loss": 5.7397,
      "mean_token_accuracy": 0.17498461306095123,
      "num_tokens": 22300212.0,
      "step": 11960
    },
    {
      "entropy": 6.008127927780151,
      "epoch": 1.0566054397739315,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004894861749092359,
      "loss": 5.7115,
      "mean_token_accuracy": 0.17412784695625305,
      "num_tokens": 22309171.0,
      "step": 11965
    },
    {
      "entropy": 5.898822069168091,
      "epoch": 1.0570469798657718,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004894766587920094,
      "loss": 5.7267,
      "mean_token_accuracy": 0.18330826759338378,
      "num_tokens": 22318457.0,
      "step": 11970
    },
    {
      "entropy": 5.986385107040405,
      "epoch": 1.057488519957612,
      "grad_norm": 1.78125,
      "learning_rate": 0.000489467138473307,
      "loss": 5.747,
      "mean_token_accuracy": 0.1688412234187126,
      "num_tokens": 22327077.0,
      "step": 11975
    },
    {
      "entropy": 5.930625486373901,
      "epoch": 1.0579300600494526,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004894576139533154,
      "loss": 5.7338,
      "mean_token_accuracy": 0.17051164656877518,
      "num_tokens": 22336259.0,
      "step": 11980
    },
    {
      "entropy": 5.942082929611206,
      "epoch": 1.0583716001412928,
      "grad_norm": 2.03125,
      "learning_rate": 0.000489448085232221,
      "loss": 5.7088,
      "mean_token_accuracy": 0.18380282819271088,
      "num_tokens": 22344978.0,
      "step": 11985
    },
    {
      "entropy": 5.929968166351318,
      "epoch": 1.058813140233133,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004894385523102107,
      "loss": 5.6986,
      "mean_token_accuracy": 0.1768606498837471,
      "num_tokens": 22353950.0,
      "step": 11990
    },
    {
      "entropy": 5.964371824264527,
      "epoch": 1.0592546803249736,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004894290151874711,
      "loss": 5.6826,
      "mean_token_accuracy": 0.1693834498524666,
      "num_tokens": 22363100.0,
      "step": 11995
    },
    {
      "entropy": 5.892598438262939,
      "epoch": 1.0596962204168139,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000489419473864189,
      "loss": 5.6899,
      "mean_token_accuracy": 0.17939914166927337,
      "num_tokens": 22371979.0,
      "step": 12000
    },
    {
      "epoch": 1.0596962204168139,
      "eval_entropy": 5.767613719840822,
      "eval_loss": 5.8206987380981445,
      "eval_mean_token_accuracy": 0.1758889344688568,
      "eval_num_tokens": 22371979.0,
      "eval_runtime": 26.119,
      "eval_samples_per_second": 1352.083,
      "eval_steps_per_second": 169.034,
      "step": 12000
    },
    {
      "entropy": 5.919700241088867,
      "epoch": 1.0601377605086542,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004894099283405515,
      "loss": 5.6757,
      "mean_token_accuracy": 0.17638276666402816,
      "num_tokens": 22380987.0,
      "step": 12005
    },
    {
      "entropy": 5.980326318740845,
      "epoch": 1.0605793006004944,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004894003786167452,
      "loss": 5.7468,
      "mean_token_accuracy": 0.16735122948884965,
      "num_tokens": 22389367.0,
      "step": 12010
    },
    {
      "entropy": 5.976982879638672,
      "epoch": 1.061020840692335,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004893908246929576,
      "loss": 5.7823,
      "mean_token_accuracy": 0.1713136374950409,
      "num_tokens": 22399158.0,
      "step": 12015
    },
    {
      "entropy": 5.973134565353393,
      "epoch": 1.0614623807841752,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004893812665693755,
      "loss": 5.7811,
      "mean_token_accuracy": 0.16887140274047852,
      "num_tokens": 22408245.0,
      "step": 12020
    },
    {
      "entropy": 5.943129444122315,
      "epoch": 1.0619039208760155,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004893717042461865,
      "loss": 5.7571,
      "mean_token_accuracy": 0.1705349326133728,
      "num_tokens": 22417014.0,
      "step": 12025
    },
    {
      "entropy": 6.045736169815063,
      "epoch": 1.062345460967856,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004893621377235777,
      "loss": 5.6995,
      "mean_token_accuracy": 0.1726825013756752,
      "num_tokens": 22426504.0,
      "step": 12030
    },
    {
      "entropy": 6.051658868789673,
      "epoch": 1.0627870010596963,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004893525670017364,
      "loss": 5.7933,
      "mean_token_accuracy": 0.16473076939582826,
      "num_tokens": 22435859.0,
      "step": 12035
    },
    {
      "entropy": 5.930938053131103,
      "epoch": 1.0632285411515365,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004893429920808503,
      "loss": 5.7018,
      "mean_token_accuracy": 0.17337859123945237,
      "num_tokens": 22445112.0,
      "step": 12040
    },
    {
      "entropy": 5.90281457901001,
      "epoch": 1.0636700812433768,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000489333412961107,
      "loss": 5.786,
      "mean_token_accuracy": 0.17001413106918334,
      "num_tokens": 22454602.0,
      "step": 12045
    },
    {
      "entropy": 6.043345880508423,
      "epoch": 1.0641116213352173,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004893238296426939,
      "loss": 5.7973,
      "mean_token_accuracy": 0.16596632152795793,
      "num_tokens": 22463918.0,
      "step": 12050
    },
    {
      "entropy": 5.952414703369141,
      "epoch": 1.0645531614270576,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004893142421257989,
      "loss": 5.6781,
      "mean_token_accuracy": 0.17010308653116227,
      "num_tokens": 22473059.0,
      "step": 12055
    },
    {
      "entropy": 5.897744941711426,
      "epoch": 1.0649947015188979,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004893046504106099,
      "loss": 5.7481,
      "mean_token_accuracy": 0.1664930209517479,
      "num_tokens": 22482615.0,
      "step": 12060
    },
    {
      "entropy": 5.875802803039551,
      "epoch": 1.0654362416107384,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004892950544973145,
      "loss": 5.6698,
      "mean_token_accuracy": 0.17800179570913316,
      "num_tokens": 22492141.0,
      "step": 12065
    },
    {
      "entropy": 5.946538019180298,
      "epoch": 1.0658777817025786,
      "grad_norm": 1.875,
      "learning_rate": 0.0004892854543861009,
      "loss": 5.6317,
      "mean_token_accuracy": 0.17720572054386138,
      "num_tokens": 22502084.0,
      "step": 12070
    },
    {
      "entropy": 5.932238674163818,
      "epoch": 1.066319321794419,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004892758500771572,
      "loss": 5.6619,
      "mean_token_accuracy": 0.1740851432085037,
      "num_tokens": 22510565.0,
      "step": 12075
    },
    {
      "entropy": 5.950931024551392,
      "epoch": 1.0667608618862592,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004892662415706714,
      "loss": 5.7587,
      "mean_token_accuracy": 0.1679132416844368,
      "num_tokens": 22520089.0,
      "step": 12080
    },
    {
      "entropy": 5.912334871292114,
      "epoch": 1.0672024019780997,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004892566288668319,
      "loss": 5.6903,
      "mean_token_accuracy": 0.16954240649938584,
      "num_tokens": 22529008.0,
      "step": 12085
    },
    {
      "entropy": 5.942786121368409,
      "epoch": 1.06764394206994,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004892470119658267,
      "loss": 5.6735,
      "mean_token_accuracy": 0.17632627934217454,
      "num_tokens": 22538040.0,
      "step": 12090
    },
    {
      "entropy": 5.979012155532837,
      "epoch": 1.0680854821617802,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004892373908678445,
      "loss": 5.753,
      "mean_token_accuracy": 0.1679326117038727,
      "num_tokens": 22548786.0,
      "step": 12095
    },
    {
      "entropy": 5.959107971191406,
      "epoch": 1.0685270222536207,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004892277655730736,
      "loss": 5.7105,
      "mean_token_accuracy": 0.17167818546295166,
      "num_tokens": 22559841.0,
      "step": 12100
    },
    {
      "entropy": 5.910323715209961,
      "epoch": 1.068968562345461,
      "grad_norm": 1.75,
      "learning_rate": 0.0004892181360817027,
      "loss": 5.6675,
      "mean_token_accuracy": 0.17487546056509018,
      "num_tokens": 22569338.0,
      "step": 12105
    },
    {
      "entropy": 5.875791454315186,
      "epoch": 1.0694101024373013,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004892085023939204,
      "loss": 5.7322,
      "mean_token_accuracy": 0.1684942737221718,
      "num_tokens": 22578184.0,
      "step": 12110
    },
    {
      "entropy": 6.012448835372925,
      "epoch": 1.0698516425291416,
      "grad_norm": 2.0,
      "learning_rate": 0.0004891988645099151,
      "loss": 5.8592,
      "mean_token_accuracy": 0.16393927484750748,
      "num_tokens": 22587007.0,
      "step": 12115
    },
    {
      "entropy": 6.025312280654907,
      "epoch": 1.070293182620982,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004891892224298761,
      "loss": 5.7532,
      "mean_token_accuracy": 0.1650542750954628,
      "num_tokens": 22596898.0,
      "step": 12120
    },
    {
      "entropy": 5.928552913665771,
      "epoch": 1.0707347227128223,
      "grad_norm": 5.28125,
      "learning_rate": 0.000489179576153992,
      "loss": 5.6883,
      "mean_token_accuracy": 0.1798458367586136,
      "num_tokens": 22605627.0,
      "step": 12125
    },
    {
      "entropy": 5.90749773979187,
      "epoch": 1.0711762628046626,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004891699256824518,
      "loss": 5.6228,
      "mean_token_accuracy": 0.17563726752996445,
      "num_tokens": 22615269.0,
      "step": 12130
    },
    {
      "entropy": 5.898746538162231,
      "epoch": 1.071617802896503,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004891602710154446,
      "loss": 5.6687,
      "mean_token_accuracy": 0.1752561166882515,
      "num_tokens": 22625048.0,
      "step": 12135
    },
    {
      "entropy": 5.888989448547363,
      "epoch": 1.0720593429883434,
      "grad_norm": 2.5,
      "learning_rate": 0.0004891506121531593,
      "loss": 5.6703,
      "mean_token_accuracy": 0.1775853231549263,
      "num_tokens": 22634146.0,
      "step": 12140
    },
    {
      "entropy": 5.983624744415283,
      "epoch": 1.0725008830801837,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004891409490957855,
      "loss": 5.7559,
      "mean_token_accuracy": 0.17233321517705918,
      "num_tokens": 22642915.0,
      "step": 12145
    },
    {
      "entropy": 5.990595531463623,
      "epoch": 1.072942423172024,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004891312818435122,
      "loss": 5.6981,
      "mean_token_accuracy": 0.17558740675449372,
      "num_tokens": 22651546.0,
      "step": 12150
    },
    {
      "entropy": 5.943798589706421,
      "epoch": 1.0733839632638644,
      "grad_norm": 1.875,
      "learning_rate": 0.000489121610396529,
      "loss": 5.6931,
      "mean_token_accuracy": 0.17555972039699555,
      "num_tokens": 22660712.0,
      "step": 12155
    },
    {
      "entropy": 5.945981073379516,
      "epoch": 1.0738255033557047,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004891119347550251,
      "loss": 5.6832,
      "mean_token_accuracy": 0.17613542973995208,
      "num_tokens": 22669428.0,
      "step": 12160
    },
    {
      "entropy": 5.953300619125367,
      "epoch": 1.074267043447545,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004891022549191902,
      "loss": 5.7762,
      "mean_token_accuracy": 0.16566626131534576,
      "num_tokens": 22679126.0,
      "step": 12165
    },
    {
      "entropy": 5.877469491958618,
      "epoch": 1.0747085835393855,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004890925708892138,
      "loss": 5.6714,
      "mean_token_accuracy": 0.17717462480068208,
      "num_tokens": 22688640.0,
      "step": 12170
    },
    {
      "entropy": 5.944130086898804,
      "epoch": 1.0751501236312258,
      "grad_norm": 1.5625,
      "learning_rate": 0.000489082882665286,
      "loss": 5.6254,
      "mean_token_accuracy": 0.17992755323648452,
      "num_tokens": 22698158.0,
      "step": 12175
    },
    {
      "entropy": 5.812044382095337,
      "epoch": 1.075591663723066,
      "grad_norm": 1.953125,
      "learning_rate": 0.000489073190247596,
      "loss": 5.5892,
      "mean_token_accuracy": 0.18361220955848695,
      "num_tokens": 22707649.0,
      "step": 12180
    },
    {
      "entropy": 5.933582448959351,
      "epoch": 1.0760332038149063,
      "grad_norm": 2.078125,
      "learning_rate": 0.000489063493636334,
      "loss": 5.7743,
      "mean_token_accuracy": 0.17365689873695372,
      "num_tokens": 22717123.0,
      "step": 12185
    },
    {
      "entropy": 5.931095981597901,
      "epoch": 1.0764747439067468,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004890537928316901,
      "loss": 5.6975,
      "mean_token_accuracy": 0.1747051805257797,
      "num_tokens": 22725921.0,
      "step": 12190
    },
    {
      "entropy": 5.9415693283081055,
      "epoch": 1.076916283998587,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004890440878338538,
      "loss": 5.8082,
      "mean_token_accuracy": 0.1634167417883873,
      "num_tokens": 22735798.0,
      "step": 12195
    },
    {
      "entropy": 6.019251966476441,
      "epoch": 1.0773578240904274,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004890343786430158,
      "loss": 5.7808,
      "mean_token_accuracy": 0.17191626131534576,
      "num_tokens": 22745583.0,
      "step": 12200
    },
    {
      "entropy": 5.960968542098999,
      "epoch": 1.0777993641822678,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000489024665259366,
      "loss": 5.6097,
      "mean_token_accuracy": 0.18140915036201477,
      "num_tokens": 22754693.0,
      "step": 12205
    },
    {
      "entropy": 5.799593734741211,
      "epoch": 1.0782409042741081,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004890149476830947,
      "loss": 5.6203,
      "mean_token_accuracy": 0.18247604072093965,
      "num_tokens": 22763416.0,
      "step": 12210
    },
    {
      "entropy": 5.934555768966675,
      "epoch": 1.0786824443659484,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004890052259143924,
      "loss": 5.6949,
      "mean_token_accuracy": 0.1773104950785637,
      "num_tokens": 22772411.0,
      "step": 12215
    },
    {
      "entropy": 5.870721960067749,
      "epoch": 1.0791239844577887,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004889954999534493,
      "loss": 5.5924,
      "mean_token_accuracy": 0.18232140690088272,
      "num_tokens": 22780148.0,
      "step": 12220
    },
    {
      "entropy": 5.856224203109742,
      "epoch": 1.0795655245496292,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004889857698004561,
      "loss": 5.6689,
      "mean_token_accuracy": 0.17569543719291686,
      "num_tokens": 22789453.0,
      "step": 12225
    },
    {
      "entropy": 6.016891574859619,
      "epoch": 1.0800070646414694,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004889760354556033,
      "loss": 5.7209,
      "mean_token_accuracy": 0.17619411647319794,
      "num_tokens": 22798800.0,
      "step": 12230
    },
    {
      "entropy": 5.937349891662597,
      "epoch": 1.0804486047333097,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004889662969190817,
      "loss": 5.7721,
      "mean_token_accuracy": 0.17256916165351868,
      "num_tokens": 22807766.0,
      "step": 12235
    },
    {
      "entropy": 5.869595193862915,
      "epoch": 1.0808901448251502,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004889565541910822,
      "loss": 5.667,
      "mean_token_accuracy": 0.17165149003267288,
      "num_tokens": 22817316.0,
      "step": 12240
    },
    {
      "entropy": 5.900118398666382,
      "epoch": 1.0813316849169905,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004889468072717953,
      "loss": 5.7308,
      "mean_token_accuracy": 0.1780767872929573,
      "num_tokens": 22827082.0,
      "step": 12245
    },
    {
      "entropy": 5.924513673782348,
      "epoch": 1.0817732250088308,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004889370561614122,
      "loss": 5.7025,
      "mean_token_accuracy": 0.17414399832487107,
      "num_tokens": 22836114.0,
      "step": 12250
    },
    {
      "entropy": 6.019763326644897,
      "epoch": 1.082214765100671,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004889273008601238,
      "loss": 5.7972,
      "mean_token_accuracy": 0.17175162583589554,
      "num_tokens": 22845653.0,
      "step": 12255
    },
    {
      "entropy": 6.0406163215637205,
      "epoch": 1.0826563051925115,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004889175413681213,
      "loss": 5.7428,
      "mean_token_accuracy": 0.17274274230003356,
      "num_tokens": 22855472.0,
      "step": 12260
    },
    {
      "entropy": 5.943841123580933,
      "epoch": 1.0830978452843518,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004889077776855957,
      "loss": 5.6703,
      "mean_token_accuracy": 0.1761347234249115,
      "num_tokens": 22864223.0,
      "step": 12265
    },
    {
      "entropy": 5.906883764266968,
      "epoch": 1.083539385376192,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004888980098127385,
      "loss": 5.7516,
      "mean_token_accuracy": 0.17228675931692122,
      "num_tokens": 22872868.0,
      "step": 12270
    },
    {
      "entropy": 5.9801044940948485,
      "epoch": 1.0839809254680326,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004888882377497409,
      "loss": 5.7784,
      "mean_token_accuracy": 0.16623635441064835,
      "num_tokens": 22881867.0,
      "step": 12275
    },
    {
      "entropy": 5.976600170135498,
      "epoch": 1.0844224655598729,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004888784614967943,
      "loss": 5.6907,
      "mean_token_accuracy": 0.1700669527053833,
      "num_tokens": 22890904.0,
      "step": 12280
    },
    {
      "entropy": 5.89644193649292,
      "epoch": 1.0848640056517131,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004888686810540905,
      "loss": 5.6445,
      "mean_token_accuracy": 0.17701754719018936,
      "num_tokens": 22900128.0,
      "step": 12285
    },
    {
      "entropy": 5.953851270675659,
      "epoch": 1.0853055457435534,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004888588964218208,
      "loss": 5.7387,
      "mean_token_accuracy": 0.17604176551103592,
      "num_tokens": 22910101.0,
      "step": 12290
    },
    {
      "entropy": 5.949468755722046,
      "epoch": 1.085747085835394,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004888491076001768,
      "loss": 5.6905,
      "mean_token_accuracy": 0.17611851245164872,
      "num_tokens": 22919372.0,
      "step": 12295
    },
    {
      "entropy": 5.942703628540039,
      "epoch": 1.0861886259272342,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004888393145893505,
      "loss": 5.705,
      "mean_token_accuracy": 0.17292705774307252,
      "num_tokens": 22929053.0,
      "step": 12300
    },
    {
      "entropy": 5.882792997360229,
      "epoch": 1.0866301660190745,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004888295173895335,
      "loss": 5.6524,
      "mean_token_accuracy": 0.17999105006456376,
      "num_tokens": 22938173.0,
      "step": 12305
    },
    {
      "entropy": 5.97471661567688,
      "epoch": 1.087071706110915,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004888197160009181,
      "loss": 5.7781,
      "mean_token_accuracy": 0.17068049311637878,
      "num_tokens": 22947076.0,
      "step": 12310
    },
    {
      "entropy": 5.918033456802368,
      "epoch": 1.0875132462027552,
      "grad_norm": 2.078125,
      "learning_rate": 0.000488809910423696,
      "loss": 5.6893,
      "mean_token_accuracy": 0.1804595783352852,
      "num_tokens": 22956267.0,
      "step": 12315
    },
    {
      "entropy": 6.048590850830078,
      "epoch": 1.0879547862945955,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004888001006580592,
      "loss": 5.7947,
      "mean_token_accuracy": 0.16591554135084152,
      "num_tokens": 22966751.0,
      "step": 12320
    },
    {
      "entropy": 5.871546602249145,
      "epoch": 1.0883963263864358,
      "grad_norm": 1.875,
      "learning_rate": 0.0004887902867042002,
      "loss": 5.7059,
      "mean_token_accuracy": 0.17762571573257446,
      "num_tokens": 22975862.0,
      "step": 12325
    },
    {
      "entropy": 5.946240997314453,
      "epoch": 1.0888378664782763,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004887804685623111,
      "loss": 5.6627,
      "mean_token_accuracy": 0.17397065907716752,
      "num_tokens": 22985085.0,
      "step": 12330
    },
    {
      "entropy": 5.970471715927124,
      "epoch": 1.0892794065701166,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004887706462325841,
      "loss": 5.797,
      "mean_token_accuracy": 0.1693852126598358,
      "num_tokens": 22993718.0,
      "step": 12335
    },
    {
      "entropy": 5.9191834926605225,
      "epoch": 1.0897209466619568,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004887608197152117,
      "loss": 5.639,
      "mean_token_accuracy": 0.1873803347349167,
      "num_tokens": 23002611.0,
      "step": 12340
    },
    {
      "entropy": 5.922781181335449,
      "epoch": 1.0901624867537973,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004887509890103864,
      "loss": 5.8302,
      "mean_token_accuracy": 0.16740358173847197,
      "num_tokens": 23012333.0,
      "step": 12345
    },
    {
      "entropy": 5.979339075088501,
      "epoch": 1.0906040268456376,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004887411541183009,
      "loss": 5.7477,
      "mean_token_accuracy": 0.1734440267086029,
      "num_tokens": 23020998.0,
      "step": 12350
    },
    {
      "entropy": 6.0557197570800785,
      "epoch": 1.091045566937478,
      "grad_norm": 1.75,
      "learning_rate": 0.0004887313150391477,
      "loss": 5.7275,
      "mean_token_accuracy": 0.17118511497974395,
      "num_tokens": 23029520.0,
      "step": 12355
    },
    {
      "entropy": 5.9565647602081295,
      "epoch": 1.0914871070293182,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004887214717731196,
      "loss": 5.806,
      "mean_token_accuracy": 0.16648072898387908,
      "num_tokens": 23039224.0,
      "step": 12360
    },
    {
      "entropy": 5.925967788696289,
      "epoch": 1.0919286471211587,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004887116243204094,
      "loss": 5.6849,
      "mean_token_accuracy": 0.16964706182479858,
      "num_tokens": 23047851.0,
      "step": 12365
    },
    {
      "entropy": 5.891302824020386,
      "epoch": 1.092370187212999,
      "grad_norm": 1.875,
      "learning_rate": 0.00048870177268121,
      "loss": 5.7123,
      "mean_token_accuracy": 0.16866123750805856,
      "num_tokens": 23057242.0,
      "step": 12370
    },
    {
      "entropy": 6.035675525665283,
      "epoch": 1.0928117273048392,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004886919168557145,
      "loss": 5.8327,
      "mean_token_accuracy": 0.16313546001911164,
      "num_tokens": 23066873.0,
      "step": 12375
    },
    {
      "entropy": 6.033766365051269,
      "epoch": 1.0932532673966797,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004886820568441158,
      "loss": 5.7918,
      "mean_token_accuracy": 0.16324014961719513,
      "num_tokens": 23076448.0,
      "step": 12380
    },
    {
      "entropy": 5.948325061798096,
      "epoch": 1.09369480748852,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004886721926466072,
      "loss": 5.6954,
      "mean_token_accuracy": 0.17317827939987182,
      "num_tokens": 23084759.0,
      "step": 12385
    },
    {
      "entropy": 5.922755718231201,
      "epoch": 1.0941363475803603,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004886623242633817,
      "loss": 5.6714,
      "mean_token_accuracy": 0.1725888043642044,
      "num_tokens": 23093236.0,
      "step": 12390
    },
    {
      "entropy": 5.954691410064697,
      "epoch": 1.0945778876722005,
      "grad_norm": 1.671875,
      "learning_rate": 0.000488652451694633,
      "loss": 5.7121,
      "mean_token_accuracy": 0.1719619080424309,
      "num_tokens": 23101621.0,
      "step": 12395
    },
    {
      "entropy": 6.012065982818603,
      "epoch": 1.095019427764041,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004886425749405542,
      "loss": 5.7617,
      "mean_token_accuracy": 0.1695653960108757,
      "num_tokens": 23110882.0,
      "step": 12400
    },
    {
      "entropy": 5.955191516876221,
      "epoch": 1.0954609678558813,
      "grad_norm": 2.0,
      "learning_rate": 0.0004886326940013388,
      "loss": 5.7215,
      "mean_token_accuracy": 0.17686995416879653,
      "num_tokens": 23120669.0,
      "step": 12405
    },
    {
      "entropy": 5.92799220085144,
      "epoch": 1.0959025079477216,
      "grad_norm": 1.375,
      "learning_rate": 0.0004886228088771805,
      "loss": 5.6495,
      "mean_token_accuracy": 0.17058452665805818,
      "num_tokens": 23129524.0,
      "step": 12410
    },
    {
      "entropy": 5.917777633666992,
      "epoch": 1.096344048039562,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004886129195682729,
      "loss": 5.694,
      "mean_token_accuracy": 0.17594977468252182,
      "num_tokens": 23139144.0,
      "step": 12415
    },
    {
      "entropy": 5.939364433288574,
      "epoch": 1.0967855881314024,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004886030260748096,
      "loss": 5.6377,
      "mean_token_accuracy": 0.18105735033750534,
      "num_tokens": 23148600.0,
      "step": 12420
    },
    {
      "entropy": 5.796615934371948,
      "epoch": 1.0972271282232426,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004885931283969845,
      "loss": 5.6221,
      "mean_token_accuracy": 0.18035324215888976,
      "num_tokens": 23156831.0,
      "step": 12425
    },
    {
      "entropy": 5.980561828613281,
      "epoch": 1.097668668315083,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004885832265349915,
      "loss": 5.7915,
      "mean_token_accuracy": 0.169062639772892,
      "num_tokens": 23166453.0,
      "step": 12430
    },
    {
      "entropy": 5.984948587417603,
      "epoch": 1.0981102084069234,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004885733204890246,
      "loss": 5.6794,
      "mean_token_accuracy": 0.17111412733793258,
      "num_tokens": 23175832.0,
      "step": 12435
    },
    {
      "entropy": 5.933304691314698,
      "epoch": 1.0985517484987637,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004885634102592778,
      "loss": 5.6904,
      "mean_token_accuracy": 0.17520835250616074,
      "num_tokens": 23185349.0,
      "step": 12440
    },
    {
      "entropy": 5.964191246032715,
      "epoch": 1.098993288590604,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004885534958459454,
      "loss": 5.8145,
      "mean_token_accuracy": 0.17049569487571717,
      "num_tokens": 23195391.0,
      "step": 12445
    },
    {
      "entropy": 5.868438673019409,
      "epoch": 1.0994348286824445,
      "grad_norm": 5.65625,
      "learning_rate": 0.0004885435772492213,
      "loss": 5.8236,
      "mean_token_accuracy": 0.1738370217382908,
      "num_tokens": 23204689.0,
      "step": 12450
    },
    {
      "entropy": 5.958547163009643,
      "epoch": 1.0998763687742847,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004885336544693,
      "loss": 5.7168,
      "mean_token_accuracy": 0.17147846519947052,
      "num_tokens": 23213182.0,
      "step": 12455
    },
    {
      "entropy": 5.83877854347229,
      "epoch": 1.100317908866125,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004885237275063758,
      "loss": 5.7271,
      "mean_token_accuracy": 0.16555773466825485,
      "num_tokens": 23222058.0,
      "step": 12460
    },
    {
      "entropy": 5.963030481338501,
      "epoch": 1.1007594489579653,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004885137963606433,
      "loss": 5.7571,
      "mean_token_accuracy": 0.17689058631658555,
      "num_tokens": 23231747.0,
      "step": 12465
    },
    {
      "entropy": 6.03634877204895,
      "epoch": 1.1012009890498058,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004885038610322969,
      "loss": 5.6972,
      "mean_token_accuracy": 0.17350972145795823,
      "num_tokens": 23240402.0,
      "step": 12470
    },
    {
      "entropy": 6.025196743011475,
      "epoch": 1.101642529141646,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004884939215215314,
      "loss": 5.7569,
      "mean_token_accuracy": 0.17686674147844314,
      "num_tokens": 23249344.0,
      "step": 12475
    },
    {
      "entropy": 5.974455213546753,
      "epoch": 1.1020840692334863,
      "grad_norm": 1.75,
      "learning_rate": 0.0004884839778285413,
      "loss": 5.7364,
      "mean_token_accuracy": 0.17291582226753235,
      "num_tokens": 23258157.0,
      "step": 12480
    },
    {
      "entropy": 5.927752065658569,
      "epoch": 1.1025256093253268,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004884740299535216,
      "loss": 5.7294,
      "mean_token_accuracy": 0.16331106573343276,
      "num_tokens": 23267798.0,
      "step": 12485
    },
    {
      "entropy": 5.968892335891724,
      "epoch": 1.1029671494171671,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000488464077896667,
      "loss": 5.7444,
      "mean_token_accuracy": 0.17355577200651168,
      "num_tokens": 23276429.0,
      "step": 12490
    },
    {
      "entropy": 5.9121135711669925,
      "epoch": 1.1034086895090074,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004884541216581724,
      "loss": 5.6531,
      "mean_token_accuracy": 0.17996444404125214,
      "num_tokens": 23285878.0,
      "step": 12495
    },
    {
      "entropy": 5.851118564605713,
      "epoch": 1.1038502296008477,
      "grad_norm": 2.515625,
      "learning_rate": 0.000488444161238233,
      "loss": 5.6749,
      "mean_token_accuracy": 0.18124425560235977,
      "num_tokens": 23295683.0,
      "step": 12500
    },
    {
      "entropy": 5.9666718482971195,
      "epoch": 1.1042917696926882,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004884341966370439,
      "loss": 5.7647,
      "mean_token_accuracy": 0.16938910335302354,
      "num_tokens": 23305539.0,
      "step": 12505
    },
    {
      "entropy": 5.9257011890411375,
      "epoch": 1.1047333097845284,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004884242278548003,
      "loss": 5.7121,
      "mean_token_accuracy": 0.17796731740236282,
      "num_tokens": 23314889.0,
      "step": 12510
    },
    {
      "entropy": 5.989439010620117,
      "epoch": 1.1051748498763687,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004884142548916975,
      "loss": 5.7761,
      "mean_token_accuracy": 0.17151159346103667,
      "num_tokens": 23324425.0,
      "step": 12515
    },
    {
      "entropy": 5.905883073806763,
      "epoch": 1.1056163899682092,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004884042777479308,
      "loss": 5.6401,
      "mean_token_accuracy": 0.1755518436431885,
      "num_tokens": 23332994.0,
      "step": 12520
    },
    {
      "entropy": 5.882895135879517,
      "epoch": 1.1060579300600495,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004883942964236957,
      "loss": 5.602,
      "mean_token_accuracy": 0.18103303760290146,
      "num_tokens": 23342108.0,
      "step": 12525
    },
    {
      "entropy": 5.879073524475098,
      "epoch": 1.1064994701518898,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004883843109191877,
      "loss": 5.6446,
      "mean_token_accuracy": 0.18046188354492188,
      "num_tokens": 23351439.0,
      "step": 12530
    },
    {
      "entropy": 5.947359895706176,
      "epoch": 1.10694101024373,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004883743212346021,
      "loss": 5.7708,
      "mean_token_accuracy": 0.17089190930128098,
      "num_tokens": 23360591.0,
      "step": 12535
    },
    {
      "entropy": 5.946292686462402,
      "epoch": 1.1073825503355705,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004883643273701353,
      "loss": 5.71,
      "mean_token_accuracy": 0.17205448150634767,
      "num_tokens": 23369854.0,
      "step": 12540
    },
    {
      "entropy": 5.940635251998901,
      "epoch": 1.1078240904274108,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004883543293259826,
      "loss": 5.7733,
      "mean_token_accuracy": 0.16475847363471985,
      "num_tokens": 23378071.0,
      "step": 12545
    },
    {
      "entropy": 6.010776376724243,
      "epoch": 1.108265630519251,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004883443271023397,
      "loss": 5.777,
      "mean_token_accuracy": 0.16283062994480133,
      "num_tokens": 23387083.0,
      "step": 12550
    },
    {
      "entropy": 5.976087999343872,
      "epoch": 1.1087071706110916,
      "grad_norm": 2.296875,
      "learning_rate": 0.000488334320699403,
      "loss": 5.598,
      "mean_token_accuracy": 0.1759439468383789,
      "num_tokens": 23395008.0,
      "step": 12555
    },
    {
      "entropy": 5.885503625869751,
      "epoch": 1.1091487107029319,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004883243101173681,
      "loss": 5.6946,
      "mean_token_accuracy": 0.17324723154306412,
      "num_tokens": 23404193.0,
      "step": 12560
    },
    {
      "entropy": 5.828455209732056,
      "epoch": 1.1095902507947721,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004883142953564315,
      "loss": 5.7079,
      "mean_token_accuracy": 0.17420599311590196,
      "num_tokens": 23413353.0,
      "step": 12565
    },
    {
      "entropy": 5.973459720611572,
      "epoch": 1.1100317908866124,
      "grad_norm": 2.015625,
      "learning_rate": 0.000488304276416789,
      "loss": 5.758,
      "mean_token_accuracy": 0.17255775779485702,
      "num_tokens": 23422797.0,
      "step": 12570
    },
    {
      "entropy": 6.015373706817627,
      "epoch": 1.110473330978453,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004882942532986372,
      "loss": 5.8106,
      "mean_token_accuracy": 0.17345898896455764,
      "num_tokens": 23432705.0,
      "step": 12575
    },
    {
      "entropy": 5.9600146293640135,
      "epoch": 1.1109148710702932,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004882842260021721,
      "loss": 5.6799,
      "mean_token_accuracy": 0.1780783489346504,
      "num_tokens": 23441613.0,
      "step": 12580
    },
    {
      "entropy": 5.897919368743897,
      "epoch": 1.1113564111621335,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048827419452759045,
      "loss": 5.7513,
      "mean_token_accuracy": 0.17219407111406326,
      "num_tokens": 23450975.0,
      "step": 12585
    },
    {
      "entropy": 5.969898509979248,
      "epoch": 1.111797951253974,
      "grad_norm": 1.859375,
      "learning_rate": 0.00048826415887508857,
      "loss": 5.7506,
      "mean_token_accuracy": 0.17060452699661255,
      "num_tokens": 23460646.0,
      "step": 12590
    },
    {
      "entropy": 6.081003952026367,
      "epoch": 1.1122394913458142,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048825411904486303,
      "loss": 5.8049,
      "mean_token_accuracy": 0.16477599442005159,
      "num_tokens": 23470454.0,
      "step": 12595
    },
    {
      "entropy": 5.942688035964966,
      "epoch": 1.1126810314376545,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004882440750371108,
      "loss": 5.7071,
      "mean_token_accuracy": 0.17348776757717133,
      "num_tokens": 23480013.0,
      "step": 12600
    },
    {
      "entropy": 5.961162185668945,
      "epoch": 1.1131225715294948,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048823402685202824,
      "loss": 5.7002,
      "mean_token_accuracy": 0.1738961637020111,
      "num_tokens": 23489485.0,
      "step": 12605
    },
    {
      "entropy": 5.890470743179321,
      "epoch": 1.1135641116213353,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004882239744898124,
      "loss": 5.6942,
      "mean_token_accuracy": 0.1804720491170883,
      "num_tokens": 23498147.0,
      "step": 12610
    },
    {
      "entropy": 5.825096511840821,
      "epoch": 1.1140056517131756,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004882139179506602,
      "loss": 5.5814,
      "mean_token_accuracy": 0.18521635234355927,
      "num_tokens": 23506614.0,
      "step": 12615
    },
    {
      "entropy": 5.976991748809814,
      "epoch": 1.1144471918050158,
      "grad_norm": 2.625,
      "learning_rate": 0.0004882038572347685,
      "loss": 5.8468,
      "mean_token_accuracy": 0.16263794600963594,
      "num_tokens": 23514927.0,
      "step": 12620
    },
    {
      "entropy": 6.01411657333374,
      "epoch": 1.1148887318968563,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048819379234233464,
      "loss": 5.877,
      "mean_token_accuracy": 0.16630604714155198,
      "num_tokens": 23525298.0,
      "step": 12625
    },
    {
      "entropy": 6.138945913314819,
      "epoch": 1.1153302719886966,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004881837232735556,
      "loss": 5.7449,
      "mean_token_accuracy": 0.17521167397499085,
      "num_tokens": 23533995.0,
      "step": 12630
    },
    {
      "entropy": 5.857966041564941,
      "epoch": 1.1157718120805369,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004881736500286286,
      "loss": 5.717,
      "mean_token_accuracy": 0.1721657171845436,
      "num_tokens": 23543900.0,
      "step": 12635
    },
    {
      "entropy": 5.938044118881225,
      "epoch": 1.1162133521723772,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048816357260775115,
      "loss": 5.8004,
      "mean_token_accuracy": 0.17061211913824081,
      "num_tokens": 23553790.0,
      "step": 12640
    },
    {
      "entropy": 5.947637844085693,
      "epoch": 1.1166548922642177,
      "grad_norm": 1.5,
      "learning_rate": 0.0004881534910111204,
      "loss": 5.7326,
      "mean_token_accuracy": 0.1732372149825096,
      "num_tokens": 23563473.0,
      "step": 12645
    },
    {
      "entropy": 5.99420018196106,
      "epoch": 1.117096432356058,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048814340523893406,
      "loss": 5.6207,
      "mean_token_accuracy": 0.18335351794958116,
      "num_tokens": 23572495.0,
      "step": 12650
    },
    {
      "entropy": 5.965641975402832,
      "epoch": 1.1175379724478982,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048813331529138965,
      "loss": 5.7951,
      "mean_token_accuracy": 0.160391403734684,
      "num_tokens": 23582945.0,
      "step": 12655
    },
    {
      "entropy": 6.047966384887696,
      "epoch": 1.1179795125397387,
      "grad_norm": 1.75,
      "learning_rate": 0.0004881232211686848,
      "loss": 5.8179,
      "mean_token_accuracy": 0.1590088739991188,
      "num_tokens": 23592370.0,
      "step": 12660
    },
    {
      "entropy": 6.0416038036346436,
      "epoch": 1.118421052631579,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004881131228710172,
      "loss": 5.7641,
      "mean_token_accuracy": 0.16688701063394545,
      "num_tokens": 23601589.0,
      "step": 12665
    },
    {
      "entropy": 5.959905099868775,
      "epoch": 1.1188625927234193,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004881030203985848,
      "loss": 5.6985,
      "mean_token_accuracy": 0.1751922994852066,
      "num_tokens": 23611456.0,
      "step": 12670
    },
    {
      "entropy": 5.951153945922852,
      "epoch": 1.1193041328152595,
      "grad_norm": 2.125,
      "learning_rate": 0.00048809291375158544,
      "loss": 5.7098,
      "mean_token_accuracy": 0.16594337671995163,
      "num_tokens": 23620362.0,
      "step": 12675
    },
    {
      "entropy": 5.971215915679932,
      "epoch": 1.1197456729071,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048808280293021703,
      "loss": 5.7406,
      "mean_token_accuracy": 0.17234519273042678,
      "num_tokens": 23629647.0,
      "step": 12680
    },
    {
      "entropy": 5.944452381134033,
      "epoch": 1.1201872129989403,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048807268793467774,
      "loss": 5.6552,
      "mean_token_accuracy": 0.17233465760946273,
      "num_tokens": 23639201.0,
      "step": 12685
    },
    {
      "entropy": 5.895379734039307,
      "epoch": 1.1206287530907806,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048806256876516565,
      "loss": 5.6469,
      "mean_token_accuracy": 0.1734580636024475,
      "num_tokens": 23648166.0,
      "step": 12690
    },
    {
      "entropy": 5.917275047302246,
      "epoch": 1.121070293182621,
      "grad_norm": 1.9375,
      "learning_rate": 0.00048805244542187906,
      "loss": 5.8244,
      "mean_token_accuracy": 0.17023787945508956,
      "num_tokens": 23658478.0,
      "step": 12695
    },
    {
      "entropy": 5.987471866607666,
      "epoch": 1.1215118332744614,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004880423179050161,
      "loss": 5.6408,
      "mean_token_accuracy": 0.17782986909151077,
      "num_tokens": 23667934.0,
      "step": 12700
    },
    {
      "entropy": 5.929670524597168,
      "epoch": 1.1219533733663016,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004880321862147755,
      "loss": 5.6864,
      "mean_token_accuracy": 0.17032040506601334,
      "num_tokens": 23676164.0,
      "step": 12705
    },
    {
      "entropy": 5.935270833969116,
      "epoch": 1.122394913458142,
      "grad_norm": 1.375,
      "learning_rate": 0.00048802205035135537,
      "loss": 5.7795,
      "mean_token_accuracy": 0.17152732759714126,
      "num_tokens": 23686223.0,
      "step": 12710
    },
    {
      "entropy": 5.934724283218384,
      "epoch": 1.1228364535499824,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048801191031495444,
      "loss": 5.7888,
      "mean_token_accuracy": 0.16620649844408036,
      "num_tokens": 23695043.0,
      "step": 12715
    },
    {
      "entropy": 5.9794168949127195,
      "epoch": 1.1232779936418227,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004880017661057714,
      "loss": 5.7838,
      "mean_token_accuracy": 0.16573376059532166,
      "num_tokens": 23704248.0,
      "step": 12720
    },
    {
      "entropy": 5.963555240631104,
      "epoch": 1.123719533733663,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004879916177240049,
      "loss": 5.6785,
      "mean_token_accuracy": 0.1726948618888855,
      "num_tokens": 23713724.0,
      "step": 12725
    },
    {
      "entropy": 6.058311653137207,
      "epoch": 1.1241610738255035,
      "grad_norm": 2.125,
      "learning_rate": 0.0004879814651698537,
      "loss": 5.8782,
      "mean_token_accuracy": 0.16043908894062042,
      "num_tokens": 23723039.0,
      "step": 12730
    },
    {
      "entropy": 5.927320528030395,
      "epoch": 1.1246026139173437,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048797130844351673,
      "loss": 5.6507,
      "mean_token_accuracy": 0.17061766237020493,
      "num_tokens": 23732024.0,
      "step": 12735
    },
    {
      "entropy": 5.931581783294678,
      "epoch": 1.125044154009184,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048796114754519294,
      "loss": 5.7696,
      "mean_token_accuracy": 0.17140851765871049,
      "num_tokens": 23740739.0,
      "step": 12740
    },
    {
      "entropy": 5.934288167953492,
      "epoch": 1.1254856941010243,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048795098247508145,
      "loss": 5.6912,
      "mean_token_accuracy": 0.1671063408255577,
      "num_tokens": 23750405.0,
      "step": 12745
    },
    {
      "entropy": 5.914581918716431,
      "epoch": 1.1259272341928648,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004879408132333813,
      "loss": 5.6898,
      "mean_token_accuracy": 0.17493831664323806,
      "num_tokens": 23759318.0,
      "step": 12750
    },
    {
      "entropy": 5.974626350402832,
      "epoch": 1.126368774284705,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00048793063982029166,
      "loss": 5.8483,
      "mean_token_accuracy": 0.16207719147205352,
      "num_tokens": 23769202.0,
      "step": 12755
    },
    {
      "entropy": 5.9084898948669435,
      "epoch": 1.1268103143765453,
      "grad_norm": 2.171875,
      "learning_rate": 0.000487920462236012,
      "loss": 5.6818,
      "mean_token_accuracy": 0.17751707136631012,
      "num_tokens": 23778478.0,
      "step": 12760
    },
    {
      "entropy": 5.984773063659668,
      "epoch": 1.1272518544683856,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004879102804807415,
      "loss": 5.7637,
      "mean_token_accuracy": 0.17513718754053115,
      "num_tokens": 23788044.0,
      "step": 12765
    },
    {
      "entropy": 5.987076330184936,
      "epoch": 1.127693394560226,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048790009455467967,
      "loss": 5.7592,
      "mean_token_accuracy": 0.16625713258981706,
      "num_tokens": 23796351.0,
      "step": 12770
    },
    {
      "entropy": 5.929470682144165,
      "epoch": 1.1281349346520664,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004878899044580261,
      "loss": 5.748,
      "mean_token_accuracy": 0.17512723356485366,
      "num_tokens": 23805553.0,
      "step": 12775
    },
    {
      "entropy": 5.876346778869629,
      "epoch": 1.1285764747439067,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048787971019098044,
      "loss": 5.6115,
      "mean_token_accuracy": 0.1818123921751976,
      "num_tokens": 23813742.0,
      "step": 12780
    },
    {
      "entropy": 5.894442701339722,
      "epoch": 1.1290180148357472,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004878695117537423,
      "loss": 5.7014,
      "mean_token_accuracy": 0.17298130095005035,
      "num_tokens": 23822135.0,
      "step": 12785
    },
    {
      "entropy": 5.974287843704223,
      "epoch": 1.1294595549275874,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004878593091465114,
      "loss": 5.7599,
      "mean_token_accuracy": 0.17514732033014296,
      "num_tokens": 23832021.0,
      "step": 12790
    },
    {
      "entropy": 5.967665338516236,
      "epoch": 1.1299010950194277,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004878491023694877,
      "loss": 5.7185,
      "mean_token_accuracy": 0.17528285682201386,
      "num_tokens": 23842432.0,
      "step": 12795
    },
    {
      "entropy": 5.969618225097657,
      "epoch": 1.1303426351112682,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004878388914228713,
      "loss": 5.7729,
      "mean_token_accuracy": 0.16340311318635942,
      "num_tokens": 23852172.0,
      "step": 12800
    },
    {
      "entropy": 5.88929967880249,
      "epoch": 1.1307841752031085,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004878286763068619,
      "loss": 5.6394,
      "mean_token_accuracy": 0.18231816440820695,
      "num_tokens": 23861178.0,
      "step": 12805
    },
    {
      "entropy": 5.9503199577331545,
      "epoch": 1.1312257152949488,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004878184570216598,
      "loss": 5.7581,
      "mean_token_accuracy": 0.16608208566904067,
      "num_tokens": 23869998.0,
      "step": 12810
    },
    {
      "entropy": 5.9207206726074215,
      "epoch": 1.131667255386789,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00048780823356746523,
      "loss": 5.6831,
      "mean_token_accuracy": 0.17149588614702224,
      "num_tokens": 23880404.0,
      "step": 12815
    },
    {
      "entropy": 5.856629419326782,
      "epoch": 1.1321087954786295,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004877980059444784,
      "loss": 5.6328,
      "mean_token_accuracy": 0.17987765967845917,
      "num_tokens": 23889643.0,
      "step": 12820
    },
    {
      "entropy": 5.968347454071045,
      "epoch": 1.1325503355704698,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004877877741528996,
      "loss": 5.7103,
      "mean_token_accuracy": 0.17564204037189485,
      "num_tokens": 23899264.0,
      "step": 12825
    },
    {
      "entropy": 5.965390396118164,
      "epoch": 1.13299187566231,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004877775381929294,
      "loss": 5.758,
      "mean_token_accuracy": 0.1675721138715744,
      "num_tokens": 23908685.0,
      "step": 12830
    },
    {
      "entropy": 5.911941957473755,
      "epoch": 1.1334334157541504,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004877672980647682,
      "loss": 5.7346,
      "mean_token_accuracy": 0.17184169143438338,
      "num_tokens": 23917838.0,
      "step": 12835
    },
    {
      "entropy": 5.936193466186523,
      "epoch": 1.1338749558459909,
      "grad_norm": 1.5,
      "learning_rate": 0.00048775705376861663,
      "loss": 5.7412,
      "mean_token_accuracy": 0.16978058815002442,
      "num_tokens": 23927498.0,
      "step": 12840
    },
    {
      "entropy": 5.999635648727417,
      "epoch": 1.1343164959378311,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004877468053046753,
      "loss": 5.7586,
      "mean_token_accuracy": 0.18179951459169388,
      "num_tokens": 23936205.0,
      "step": 12845
    },
    {
      "entropy": 5.885288572311401,
      "epoch": 1.1347580360296714,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048773655267314516,
      "loss": 5.6562,
      "mean_token_accuracy": 0.17996972054243088,
      "num_tokens": 23945897.0,
      "step": 12850
    },
    {
      "entropy": 5.8981927871704105,
      "epoch": 1.135199576121512,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004877262958742268,
      "loss": 5.6847,
      "mean_token_accuracy": 0.17190316468477249,
      "num_tokens": 23954873.0,
      "step": 12855
    },
    {
      "entropy": 5.921915054321289,
      "epoch": 1.1356411162133522,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048771603490812144,
      "loss": 5.7743,
      "mean_token_accuracy": 0.1694043219089508,
      "num_tokens": 23964631.0,
      "step": 12860
    },
    {
      "entropy": 5.929622507095337,
      "epoch": 1.1360826563051925,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004877057697750299,
      "loss": 5.6823,
      "mean_token_accuracy": 0.1771518960595131,
      "num_tokens": 23973935.0,
      "step": 12865
    },
    {
      "entropy": 6.004544258117676,
      "epoch": 1.136524196397033,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004876955004751533,
      "loss": 5.7899,
      "mean_token_accuracy": 0.16899486929178237,
      "num_tokens": 23984340.0,
      "step": 12870
    },
    {
      "entropy": 6.074613857269287,
      "epoch": 1.1369657364888732,
      "grad_norm": 1.71875,
      "learning_rate": 0.00048768522700869284,
      "loss": 5.8643,
      "mean_token_accuracy": 0.16745609939098358,
      "num_tokens": 23994387.0,
      "step": 12875
    },
    {
      "entropy": 5.947685718536377,
      "epoch": 1.1374072765807135,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048767494937584964,
      "loss": 5.6619,
      "mean_token_accuracy": 0.17775584012269974,
      "num_tokens": 24003391.0,
      "step": 12880
    },
    {
      "entropy": 5.88681583404541,
      "epoch": 1.1378488166725538,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048766466757682527,
      "loss": 5.643,
      "mean_token_accuracy": 0.1797345444560051,
      "num_tokens": 24013501.0,
      "step": 12885
    },
    {
      "entropy": 5.866731119155884,
      "epoch": 1.1382903567643943,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004876543816118209,
      "loss": 5.6166,
      "mean_token_accuracy": 0.17863497734069825,
      "num_tokens": 24022675.0,
      "step": 12890
    },
    {
      "entropy": 6.002243375778198,
      "epoch": 1.1387318968562345,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048764409148103815,
      "loss": 5.7953,
      "mean_token_accuracy": 0.16323062852025033,
      "num_tokens": 24032542.0,
      "step": 12895
    },
    {
      "entropy": 5.893810749053955,
      "epoch": 1.1391734369480748,
      "grad_norm": 1.625,
      "learning_rate": 0.0004876337971846787,
      "loss": 5.6832,
      "mean_token_accuracy": 0.1754581406712532,
      "num_tokens": 24041342.0,
      "step": 12900
    },
    {
      "entropy": 5.9228626728057865,
      "epoch": 1.139614977039915,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000487623498722944,
      "loss": 5.7042,
      "mean_token_accuracy": 0.1790216386318207,
      "num_tokens": 24051429.0,
      "step": 12905
    },
    {
      "entropy": 5.96890869140625,
      "epoch": 1.1400565171317556,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004876131960960358,
      "loss": 5.7067,
      "mean_token_accuracy": 0.17783437073230743,
      "num_tokens": 24061074.0,
      "step": 12910
    },
    {
      "entropy": 5.885454559326172,
      "epoch": 1.1404980572235959,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048760288930415615,
      "loss": 5.7174,
      "mean_token_accuracy": 0.17269231230020524,
      "num_tokens": 24069622.0,
      "step": 12915
    },
    {
      "entropy": 5.972828483581543,
      "epoch": 1.1409395973154361,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048759257834750673,
      "loss": 5.7358,
      "mean_token_accuracy": 0.1731975719332695,
      "num_tokens": 24078393.0,
      "step": 12920
    },
    {
      "entropy": 5.930322217941284,
      "epoch": 1.1413811374072766,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004875822632262897,
      "loss": 5.7173,
      "mean_token_accuracy": 0.1685234621167183,
      "num_tokens": 24089157.0,
      "step": 12925
    },
    {
      "entropy": 5.870923137664795,
      "epoch": 1.141822677499117,
      "grad_norm": 1.75,
      "learning_rate": 0.00048757194394070697,
      "loss": 5.6466,
      "mean_token_accuracy": 0.18034950643777847,
      "num_tokens": 24098022.0,
      "step": 12930
    },
    {
      "entropy": 5.860831880569458,
      "epoch": 1.1422642175909572,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048756162049096077,
      "loss": 5.6554,
      "mean_token_accuracy": 0.17355776131153106,
      "num_tokens": 24107629.0,
      "step": 12935
    },
    {
      "entropy": 5.93750147819519,
      "epoch": 1.1427057576827977,
      "grad_norm": 1.375,
      "learning_rate": 0.00048755129287725334,
      "loss": 5.7141,
      "mean_token_accuracy": 0.17552746087312698,
      "num_tokens": 24116903.0,
      "step": 12940
    },
    {
      "entropy": 5.901369571685791,
      "epoch": 1.143147297774638,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048754096109978684,
      "loss": 5.6632,
      "mean_token_accuracy": 0.1739124044775963,
      "num_tokens": 24126433.0,
      "step": 12945
    },
    {
      "entropy": 6.035658454895019,
      "epoch": 1.1435888378664782,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004875306251587639,
      "loss": 5.8778,
      "mean_token_accuracy": 0.16467935889959334,
      "num_tokens": 24136603.0,
      "step": 12950
    },
    {
      "entropy": 6.005431413650513,
      "epoch": 1.1440303779583185,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004875202850543869,
      "loss": 5.735,
      "mean_token_accuracy": 0.16973657459020614,
      "num_tokens": 24145656.0,
      "step": 12955
    },
    {
      "entropy": 5.918482542037964,
      "epoch": 1.144471918050159,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048750994078685837,
      "loss": 5.7746,
      "mean_token_accuracy": 0.16437962353229524,
      "num_tokens": 24155044.0,
      "step": 12960
    },
    {
      "entropy": 5.932920646667481,
      "epoch": 1.1449134581419993,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004874995923563809,
      "loss": 5.6978,
      "mean_token_accuracy": 0.17375231087207793,
      "num_tokens": 24163987.0,
      "step": 12965
    },
    {
      "entropy": 5.91859803199768,
      "epoch": 1.1453549982338396,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004874892397631573,
      "loss": 5.6768,
      "mean_token_accuracy": 0.16759736835956573,
      "num_tokens": 24173542.0,
      "step": 12970
    },
    {
      "entropy": 5.862208890914917,
      "epoch": 1.1457965383256798,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004874788830073904,
      "loss": 5.661,
      "mean_token_accuracy": 0.17140197157859802,
      "num_tokens": 24182785.0,
      "step": 12975
    },
    {
      "entropy": 5.935773849487305,
      "epoch": 1.1462380784175203,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000487468522089283,
      "loss": 5.7217,
      "mean_token_accuracy": 0.17053521126508714,
      "num_tokens": 24193477.0,
      "step": 12980
    },
    {
      "entropy": 5.982908773422241,
      "epoch": 1.1466796185093606,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048745815700903806,
      "loss": 5.7161,
      "mean_token_accuracy": 0.1672234997153282,
      "num_tokens": 24202289.0,
      "step": 12985
    },
    {
      "entropy": 5.838470888137818,
      "epoch": 1.147121158601201,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048744778776685865,
      "loss": 5.629,
      "mean_token_accuracy": 0.18295220136642457,
      "num_tokens": 24212199.0,
      "step": 12990
    },
    {
      "entropy": 5.949348592758179,
      "epoch": 1.1475626986930414,
      "grad_norm": 1.625,
      "learning_rate": 0.00048743741436294794,
      "loss": 5.7476,
      "mean_token_accuracy": 0.17771709710359573,
      "num_tokens": 24220984.0,
      "step": 12995
    },
    {
      "entropy": 5.947386741638184,
      "epoch": 1.1480042387848817,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004874270367975091,
      "loss": 5.6927,
      "mean_token_accuracy": 0.1741531863808632,
      "num_tokens": 24229599.0,
      "step": 13000
    },
    {
      "entropy": 5.934886121749878,
      "epoch": 1.148445778876722,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004874166550707454,
      "loss": 5.7528,
      "mean_token_accuracy": 0.17183105498552323,
      "num_tokens": 24238721.0,
      "step": 13005
    },
    {
      "entropy": 5.996446943283081,
      "epoch": 1.1488873189685624,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048740626918286026,
      "loss": 5.7596,
      "mean_token_accuracy": 0.16176596581935881,
      "num_tokens": 24248431.0,
      "step": 13010
    },
    {
      "entropy": 5.9642139911651615,
      "epoch": 1.1493288590604027,
      "grad_norm": 1.875,
      "learning_rate": 0.00048739587913405706,
      "loss": 5.7144,
      "mean_token_accuracy": 0.17013247013092042,
      "num_tokens": 24257770.0,
      "step": 13015
    },
    {
      "entropy": 5.961289358139038,
      "epoch": 1.149770399152243,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048738548492453944,
      "loss": 5.6962,
      "mean_token_accuracy": 0.17342830896377565,
      "num_tokens": 24266402.0,
      "step": 13020
    },
    {
      "entropy": 5.911700010299683,
      "epoch": 1.1502119392440833,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004873750865545109,
      "loss": 5.6723,
      "mean_token_accuracy": 0.172995662689209,
      "num_tokens": 24275164.0,
      "step": 13025
    },
    {
      "entropy": 5.918226957321167,
      "epoch": 1.1506534793359238,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048736468402417525,
      "loss": 5.7708,
      "mean_token_accuracy": 0.16361208260059357,
      "num_tokens": 24285650.0,
      "step": 13030
    },
    {
      "entropy": 5.984469699859619,
      "epoch": 1.151095019427764,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048735427733373617,
      "loss": 5.7396,
      "mean_token_accuracy": 0.16899943053722383,
      "num_tokens": 24294745.0,
      "step": 13035
    },
    {
      "entropy": 5.992206621170044,
      "epoch": 1.1515365595196043,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048734386648339767,
      "loss": 5.7007,
      "mean_token_accuracy": 0.17801126688718796,
      "num_tokens": 24303837.0,
      "step": 13040
    },
    {
      "entropy": 5.907953596115112,
      "epoch": 1.1519780996114446,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004873334514733634,
      "loss": 5.7742,
      "mean_token_accuracy": 0.170374695956707,
      "num_tokens": 24313417.0,
      "step": 13045
    },
    {
      "entropy": 5.950648450851441,
      "epoch": 1.152419639703285,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004873230323038378,
      "loss": 5.7035,
      "mean_token_accuracy": 0.17563520967960358,
      "num_tokens": 24322439.0,
      "step": 13050
    },
    {
      "entropy": 5.977941560745239,
      "epoch": 1.1528611797951254,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004873126089750246,
      "loss": 5.7621,
      "mean_token_accuracy": 0.1721951350569725,
      "num_tokens": 24332509.0,
      "step": 13055
    },
    {
      "entropy": 5.959326219558716,
      "epoch": 1.1533027198869656,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004873021814871282,
      "loss": 5.7658,
      "mean_token_accuracy": 0.16450584083795547,
      "num_tokens": 24342252.0,
      "step": 13060
    },
    {
      "entropy": 6.003157663345337,
      "epoch": 1.1537442599788061,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004872917498403528,
      "loss": 5.7612,
      "mean_token_accuracy": 0.17705838084220887,
      "num_tokens": 24352459.0,
      "step": 13065
    },
    {
      "entropy": 5.927250480651855,
      "epoch": 1.1541858000706464,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004872813140349028,
      "loss": 5.632,
      "mean_token_accuracy": 0.17713265866041183,
      "num_tokens": 24361788.0,
      "step": 13070
    },
    {
      "entropy": 5.915983772277832,
      "epoch": 1.1546273401624867,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004872708740709825,
      "loss": 5.7374,
      "mean_token_accuracy": 0.17037821859121322,
      "num_tokens": 24372117.0,
      "step": 13075
    },
    {
      "entropy": 5.889716339111328,
      "epoch": 1.1550688802543272,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048726042994879656,
      "loss": 5.672,
      "mean_token_accuracy": 0.17771655321121216,
      "num_tokens": 24380833.0,
      "step": 13080
    },
    {
      "entropy": 5.959154319763184,
      "epoch": 1.1555104203461675,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048724998166854944,
      "loss": 5.783,
      "mean_token_accuracy": 0.17464201897382736,
      "num_tokens": 24390258.0,
      "step": 13085
    },
    {
      "entropy": 5.958036470413208,
      "epoch": 1.1559519604380077,
      "grad_norm": 1.671875,
      "learning_rate": 0.000487239529230446,
      "loss": 5.7902,
      "mean_token_accuracy": 0.16594357937574386,
      "num_tokens": 24399359.0,
      "step": 13090
    },
    {
      "entropy": 5.936441278457641,
      "epoch": 1.156393500529848,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048722907263469084,
      "loss": 5.7314,
      "mean_token_accuracy": 0.17864323407411575,
      "num_tokens": 24408012.0,
      "step": 13095
    },
    {
      "entropy": 6.000053024291992,
      "epoch": 1.1568350406216885,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004872186118814889,
      "loss": 5.7294,
      "mean_token_accuracy": 0.171853169798851,
      "num_tokens": 24416804.0,
      "step": 13100
    },
    {
      "entropy": 5.825567245483398,
      "epoch": 1.1572765807135288,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048720814697104497,
      "loss": 5.5334,
      "mean_token_accuracy": 0.18620926141738892,
      "num_tokens": 24425401.0,
      "step": 13105
    },
    {
      "entropy": 5.8007519245147705,
      "epoch": 1.157718120805369,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048719767790356415,
      "loss": 5.609,
      "mean_token_accuracy": 0.1755686804652214,
      "num_tokens": 24434473.0,
      "step": 13110
    },
    {
      "entropy": 6.062527847290039,
      "epoch": 1.1581596608972093,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00048718720467925156,
      "loss": 5.8637,
      "mean_token_accuracy": 0.1599051460623741,
      "num_tokens": 24443154.0,
      "step": 13115
    },
    {
      "entropy": 5.982793712615967,
      "epoch": 1.1586012009890498,
      "grad_norm": 1.5,
      "learning_rate": 0.0004871767272983122,
      "loss": 5.6674,
      "mean_token_accuracy": 0.1824402317404747,
      "num_tokens": 24451308.0,
      "step": 13120
    },
    {
      "entropy": 5.9522705078125,
      "epoch": 1.1590427410808901,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004871662457609515,
      "loss": 5.6958,
      "mean_token_accuracy": 0.1748383268713951,
      "num_tokens": 24459997.0,
      "step": 13125
    },
    {
      "entropy": 5.948757743835449,
      "epoch": 1.1594842811727304,
      "grad_norm": 1.5,
      "learning_rate": 0.0004871557600673747,
      "loss": 5.7113,
      "mean_token_accuracy": 0.17272602915763854,
      "num_tokens": 24469448.0,
      "step": 13130
    },
    {
      "entropy": 5.889970588684082,
      "epoch": 1.1599258212645709,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048714527021778714,
      "loss": 5.6938,
      "mean_token_accuracy": 0.1770160377025604,
      "num_tokens": 24478304.0,
      "step": 13135
    },
    {
      "entropy": 5.915947437286377,
      "epoch": 1.1603673613564112,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004871347762123945,
      "loss": 5.5874,
      "mean_token_accuracy": 0.18575670719146728,
      "num_tokens": 24487300.0,
      "step": 13140
    },
    {
      "entropy": 5.9199957847595215,
      "epoch": 1.1608089014482514,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048712427805140224,
      "loss": 5.6582,
      "mean_token_accuracy": 0.17845288366079332,
      "num_tokens": 24495893.0,
      "step": 13145
    },
    {
      "entropy": 5.9128156185150145,
      "epoch": 1.161250441540092,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004871137757350159,
      "loss": 5.6472,
      "mean_token_accuracy": 0.17299401611089707,
      "num_tokens": 24504603.0,
      "step": 13150
    },
    {
      "entropy": 5.902621507644653,
      "epoch": 1.1616919816319322,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048710326926344153,
      "loss": 5.6393,
      "mean_token_accuracy": 0.1821286901831627,
      "num_tokens": 24513148.0,
      "step": 13155
    },
    {
      "entropy": 6.071345090866089,
      "epoch": 1.1621335217237725,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048709275863688455,
      "loss": 5.7908,
      "mean_token_accuracy": 0.16423829942941665,
      "num_tokens": 24524006.0,
      "step": 13160
    },
    {
      "entropy": 5.964014673233033,
      "epoch": 1.1625750618156128,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048708224385555116,
      "loss": 5.6496,
      "mean_token_accuracy": 0.1836867019534111,
      "num_tokens": 24532468.0,
      "step": 13165
    },
    {
      "entropy": 5.946556282043457,
      "epoch": 1.1630166019074533,
      "grad_norm": 1.5,
      "learning_rate": 0.0004870717249196472,
      "loss": 5.7339,
      "mean_token_accuracy": 0.16739621609449387,
      "num_tokens": 24540887.0,
      "step": 13170
    },
    {
      "entropy": 5.9652200698852536,
      "epoch": 1.1634581419992935,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004870612018293788,
      "loss": 5.6588,
      "mean_token_accuracy": 0.1730510860681534,
      "num_tokens": 24550565.0,
      "step": 13175
    },
    {
      "entropy": 5.914828681945801,
      "epoch": 1.1638996820911338,
      "grad_norm": 1.546875,
      "learning_rate": 0.000487050674584952,
      "loss": 5.6536,
      "mean_token_accuracy": 0.18037382811307906,
      "num_tokens": 24560283.0,
      "step": 13180
    },
    {
      "entropy": 5.816815567016602,
      "epoch": 1.164341222182974,
      "grad_norm": 1.578125,
      "learning_rate": 0.000487040143186573,
      "loss": 5.5958,
      "mean_token_accuracy": 0.17708114087581633,
      "num_tokens": 24569527.0,
      "step": 13185
    },
    {
      "entropy": 5.943865871429443,
      "epoch": 1.1647827622748146,
      "grad_norm": 1.5,
      "learning_rate": 0.00048702960763444833,
      "loss": 5.7106,
      "mean_token_accuracy": 0.17197999954223633,
      "num_tokens": 24578702.0,
      "step": 13190
    },
    {
      "entropy": 5.984963655471802,
      "epoch": 1.1652243023666549,
      "grad_norm": 1.75,
      "learning_rate": 0.0004870190679287842,
      "loss": 5.6662,
      "mean_token_accuracy": 0.17837640643119812,
      "num_tokens": 24588252.0,
      "step": 13195
    },
    {
      "entropy": 5.878225994110108,
      "epoch": 1.1656658424584951,
      "grad_norm": 2.5,
      "learning_rate": 0.0004870085240697871,
      "loss": 5.6646,
      "mean_token_accuracy": 0.17542481571435928,
      "num_tokens": 24596989.0,
      "step": 13200
    },
    {
      "entropy": 5.888234043121338,
      "epoch": 1.1661073825503356,
      "grad_norm": 1.75,
      "learning_rate": 0.00048699797605766365,
      "loss": 5.7824,
      "mean_token_accuracy": 0.17070534229278564,
      "num_tokens": 24607051.0,
      "step": 13205
    },
    {
      "entropy": 6.010226392745972,
      "epoch": 1.166548922642176,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048698742389262036,
      "loss": 5.7756,
      "mean_token_accuracy": 0.16238270848989486,
      "num_tokens": 24616561.0,
      "step": 13210
    },
    {
      "entropy": 5.918627309799194,
      "epoch": 1.1669904627340162,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048697686757486397,
      "loss": 5.7338,
      "mean_token_accuracy": 0.1744931161403656,
      "num_tokens": 24625942.0,
      "step": 13215
    },
    {
      "entropy": 5.974892950057983,
      "epoch": 1.1674320028258567,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004869663071046013,
      "loss": 5.7049,
      "mean_token_accuracy": 0.16854122430086135,
      "num_tokens": 24635646.0,
      "step": 13220
    },
    {
      "entropy": 5.919003295898437,
      "epoch": 1.167873542917697,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048695574248203935,
      "loss": 5.7047,
      "mean_token_accuracy": 0.17816849797964096,
      "num_tokens": 24645147.0,
      "step": 13225
    },
    {
      "entropy": 5.935742664337158,
      "epoch": 1.1683150830095372,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004869451737073848,
      "loss": 5.6916,
      "mean_token_accuracy": 0.17469227612018584,
      "num_tokens": 24654796.0,
      "step": 13230
    },
    {
      "entropy": 5.914208221435547,
      "epoch": 1.1687566231013775,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004869346007808449,
      "loss": 5.6556,
      "mean_token_accuracy": 0.17688250094652175,
      "num_tokens": 24663765.0,
      "step": 13235
    },
    {
      "entropy": 5.934512281417847,
      "epoch": 1.169198163193218,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00048692402370262674,
      "loss": 5.7573,
      "mean_token_accuracy": 0.1694234326481819,
      "num_tokens": 24673758.0,
      "step": 13240
    },
    {
      "entropy": 5.923940229415893,
      "epoch": 1.1696397032850583,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048691344247293747,
      "loss": 5.6232,
      "mean_token_accuracy": 0.17605926245450973,
      "num_tokens": 24682751.0,
      "step": 13245
    },
    {
      "entropy": 5.915913057327271,
      "epoch": 1.1700812433768986,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048690285709198435,
      "loss": 5.7339,
      "mean_token_accuracy": 0.17603832185268403,
      "num_tokens": 24691375.0,
      "step": 13250
    },
    {
      "entropy": 5.924881792068481,
      "epoch": 1.1705227834687388,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004868922675599747,
      "loss": 5.6955,
      "mean_token_accuracy": 0.17576789259910583,
      "num_tokens": 24699852.0,
      "step": 13255
    },
    {
      "entropy": 5.877166986465454,
      "epoch": 1.1709643235605793,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004868816738771162,
      "loss": 5.5888,
      "mean_token_accuracy": 0.1834797441959381,
      "num_tokens": 24708857.0,
      "step": 13260
    },
    {
      "entropy": 5.936138868331909,
      "epoch": 1.1714058636524196,
      "grad_norm": 1.375,
      "learning_rate": 0.0004868710760436161,
      "loss": 5.7894,
      "mean_token_accuracy": 0.16602290123701097,
      "num_tokens": 24718262.0,
      "step": 13265
    },
    {
      "entropy": 6.020315408706665,
      "epoch": 1.1718474037442599,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048686047405968217,
      "loss": 5.7314,
      "mean_token_accuracy": 0.1739261880517006,
      "num_tokens": 24727458.0,
      "step": 13270
    },
    {
      "entropy": 6.006884813308716,
      "epoch": 1.1722889438361004,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048684986792552206,
      "loss": 5.799,
      "mean_token_accuracy": 0.16788618564605712,
      "num_tokens": 24737297.0,
      "step": 13275
    },
    {
      "entropy": 5.937396049499512,
      "epoch": 1.1727304839279407,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048683925764134346,
      "loss": 5.6967,
      "mean_token_accuracy": 0.17942382395267487,
      "num_tokens": 24746939.0,
      "step": 13280
    },
    {
      "entropy": 5.9933762550354,
      "epoch": 1.173172024019781,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004868286432073543,
      "loss": 5.8328,
      "mean_token_accuracy": 0.17054160088300704,
      "num_tokens": 24756907.0,
      "step": 13285
    },
    {
      "entropy": 5.9321976661682125,
      "epoch": 1.1736135641116214,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048681802462376253,
      "loss": 5.6318,
      "mean_token_accuracy": 0.1822195202112198,
      "num_tokens": 24765997.0,
      "step": 13290
    },
    {
      "entropy": 5.961455202102661,
      "epoch": 1.1740551042034617,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048680740189077605,
      "loss": 5.7093,
      "mean_token_accuracy": 0.1633992999792099,
      "num_tokens": 24775090.0,
      "step": 13295
    },
    {
      "entropy": 5.866019010543823,
      "epoch": 1.174496644295302,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048679677500860314,
      "loss": 5.6614,
      "mean_token_accuracy": 0.17434234023094178,
      "num_tokens": 24784709.0,
      "step": 13300
    },
    {
      "entropy": 5.941660690307617,
      "epoch": 1.1749381843871423,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004867861439774519,
      "loss": 5.7135,
      "mean_token_accuracy": 0.17504084408283233,
      "num_tokens": 24794473.0,
      "step": 13305
    },
    {
      "entropy": 5.991569375991821,
      "epoch": 1.1753797244789828,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00048677550879753037,
      "loss": 5.8241,
      "mean_token_accuracy": 0.1695534408092499,
      "num_tokens": 24804718.0,
      "step": 13310
    },
    {
      "entropy": 6.037198066711426,
      "epoch": 1.175821264570823,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048676486946904715,
      "loss": 5.7102,
      "mean_token_accuracy": 0.17287857979536056,
      "num_tokens": 24813632.0,
      "step": 13315
    },
    {
      "entropy": 5.895893192291259,
      "epoch": 1.1762628046626633,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004867542259922105,
      "loss": 5.6325,
      "mean_token_accuracy": 0.184901861846447,
      "num_tokens": 24823395.0,
      "step": 13320
    },
    {
      "entropy": 5.96007776260376,
      "epoch": 1.1767043447545038,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004867435783672291,
      "loss": 5.7383,
      "mean_token_accuracy": 0.1684446766972542,
      "num_tokens": 24832869.0,
      "step": 13325
    },
    {
      "entropy": 5.995143175125122,
      "epoch": 1.177145884846344,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004867329265943114,
      "loss": 5.704,
      "mean_token_accuracy": 0.17244094163179396,
      "num_tokens": 24841706.0,
      "step": 13330
    },
    {
      "entropy": 6.005990505218506,
      "epoch": 1.1775874249381844,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004867222706736661,
      "loss": 5.7313,
      "mean_token_accuracy": 0.16982605755329133,
      "num_tokens": 24851149.0,
      "step": 13335
    },
    {
      "entropy": 5.9855142593383786,
      "epoch": 1.1780289650300246,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004867116106055019,
      "loss": 5.6887,
      "mean_token_accuracy": 0.17659004181623458,
      "num_tokens": 24860002.0,
      "step": 13340
    },
    {
      "entropy": 5.93112530708313,
      "epoch": 1.1784705051218651,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048670094639002766,
      "loss": 5.7528,
      "mean_token_accuracy": 0.16615783423185349,
      "num_tokens": 24869457.0,
      "step": 13345
    },
    {
      "entropy": 5.883618879318237,
      "epoch": 1.1789120452137054,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004866902780274523,
      "loss": 5.6839,
      "mean_token_accuracy": 0.176359660923481,
      "num_tokens": 24879527.0,
      "step": 13350
    },
    {
      "entropy": 5.898899126052856,
      "epoch": 1.1793535853055457,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004866796055179847,
      "loss": 5.715,
      "mean_token_accuracy": 0.17645428329706192,
      "num_tokens": 24889401.0,
      "step": 13355
    },
    {
      "entropy": 5.935476016998291,
      "epoch": 1.1797951253973862,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004866689288618342,
      "loss": 5.7016,
      "mean_token_accuracy": 0.18034995198249817,
      "num_tokens": 24897911.0,
      "step": 13360
    },
    {
      "entropy": 5.935819435119629,
      "epoch": 1.1802366654892265,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004866582480592096,
      "loss": 5.6749,
      "mean_token_accuracy": 0.1839087724685669,
      "num_tokens": 24906969.0,
      "step": 13365
    },
    {
      "entropy": 5.972019004821777,
      "epoch": 1.1806782055810667,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004866475631103203,
      "loss": 5.7084,
      "mean_token_accuracy": 0.17648442685604096,
      "num_tokens": 24915713.0,
      "step": 13370
    },
    {
      "entropy": 5.926186418533325,
      "epoch": 1.181119745672907,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048663687401537573,
      "loss": 5.7087,
      "mean_token_accuracy": 0.17153839021921158,
      "num_tokens": 24925484.0,
      "step": 13375
    },
    {
      "entropy": 5.935056543350219,
      "epoch": 1.1815612857647475,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048662618077458507,
      "loss": 5.7501,
      "mean_token_accuracy": 0.17804119288921355,
      "num_tokens": 24934751.0,
      "step": 13380
    },
    {
      "entropy": 5.969244194030762,
      "epoch": 1.1820028258565878,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004866154833881579,
      "loss": 5.7789,
      "mean_token_accuracy": 0.1714974656701088,
      "num_tokens": 24945059.0,
      "step": 13385
    },
    {
      "entropy": 5.98104305267334,
      "epoch": 1.182444365948428,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004866047818563038,
      "loss": 5.7239,
      "mean_token_accuracy": 0.17468527108430862,
      "num_tokens": 24954678.0,
      "step": 13390
    },
    {
      "entropy": 5.977413702011108,
      "epoch": 1.1828859060402686,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004865940761792323,
      "loss": 5.7239,
      "mean_token_accuracy": 0.16754123866558074,
      "num_tokens": 24964093.0,
      "step": 13395
    },
    {
      "entropy": 6.0326989650726315,
      "epoch": 1.1833274461321088,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004865833663571533,
      "loss": 5.6628,
      "mean_token_accuracy": 0.18191205561161042,
      "num_tokens": 24973903.0,
      "step": 13400
    },
    {
      "entropy": 5.868292999267578,
      "epoch": 1.183768986223949,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048657265239027635,
      "loss": 5.6553,
      "mean_token_accuracy": 0.18321623504161835,
      "num_tokens": 24982458.0,
      "step": 13405
    },
    {
      "entropy": 5.883187961578369,
      "epoch": 1.1842105263157894,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004865619342788116,
      "loss": 5.7349,
      "mean_token_accuracy": 0.16896335110068322,
      "num_tokens": 24992233.0,
      "step": 13410
    },
    {
      "entropy": 6.0136932849884035,
      "epoch": 1.1846520664076299,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048655121202296873,
      "loss": 5.6725,
      "mean_token_accuracy": 0.17194696813821791,
      "num_tokens": 25001855.0,
      "step": 13415
    },
    {
      "entropy": 5.900771141052246,
      "epoch": 1.1850936064994702,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048654048562295807,
      "loss": 5.5657,
      "mean_token_accuracy": 0.17896452248096467,
      "num_tokens": 25010950.0,
      "step": 13420
    },
    {
      "entropy": 5.89925365447998,
      "epoch": 1.1855351465913104,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004865297550789895,
      "loss": 5.7587,
      "mean_token_accuracy": 0.17006942629814148,
      "num_tokens": 25020816.0,
      "step": 13425
    },
    {
      "entropy": 6.043949270248413,
      "epoch": 1.185976686683151,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004865190203912734,
      "loss": 5.7848,
      "mean_token_accuracy": 0.17223084419965745,
      "num_tokens": 25030236.0,
      "step": 13430
    },
    {
      "entropy": 5.933374357223511,
      "epoch": 1.1864182267749912,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048650828156001995,
      "loss": 5.6607,
      "mean_token_accuracy": 0.17462195605039596,
      "num_tokens": 25040236.0,
      "step": 13435
    },
    {
      "entropy": 5.9386358737945555,
      "epoch": 1.1868597668668315,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004864975385854396,
      "loss": 5.8114,
      "mean_token_accuracy": 0.16821306347846984,
      "num_tokens": 25049835.0,
      "step": 13440
    },
    {
      "entropy": 5.990381908416748,
      "epoch": 1.1873013069586718,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048648679146774275,
      "loss": 5.7521,
      "mean_token_accuracy": 0.16655802428722383,
      "num_tokens": 25058947.0,
      "step": 13445
    },
    {
      "entropy": 6.000526189804077,
      "epoch": 1.1877428470505123,
      "grad_norm": 1.625,
      "learning_rate": 0.0004864760402071399,
      "loss": 5.7162,
      "mean_token_accuracy": 0.17131267488002777,
      "num_tokens": 25069128.0,
      "step": 13450
    },
    {
      "entropy": 5.908565711975098,
      "epoch": 1.1881843871423525,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048646528480384175,
      "loss": 5.7657,
      "mean_token_accuracy": 0.16495625972747802,
      "num_tokens": 25079717.0,
      "step": 13455
    },
    {
      "entropy": 5.930012369155884,
      "epoch": 1.1886259272341928,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004864545252580588,
      "loss": 5.6933,
      "mean_token_accuracy": 0.1795371189713478,
      "num_tokens": 25088656.0,
      "step": 13460
    },
    {
      "entropy": 5.990023231506347,
      "epoch": 1.1890674673260333,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048644376157000213,
      "loss": 5.6976,
      "mean_token_accuracy": 0.1663629502058029,
      "num_tokens": 25098066.0,
      "step": 13465
    },
    {
      "entropy": 5.902922916412353,
      "epoch": 1.1895090074178736,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004864329937398823,
      "loss": 5.7088,
      "mean_token_accuracy": 0.17520444989204406,
      "num_tokens": 25107530.0,
      "step": 13470
    },
    {
      "entropy": 5.955098676681518,
      "epoch": 1.1899505475097139,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048642222176791043,
      "loss": 5.6673,
      "mean_token_accuracy": 0.1741166204214096,
      "num_tokens": 25116226.0,
      "step": 13475
    },
    {
      "entropy": 5.996547985076904,
      "epoch": 1.1903920876015541,
      "grad_norm": 1.375,
      "learning_rate": 0.0004864114456542975,
      "loss": 5.8398,
      "mean_token_accuracy": 0.16004668325185775,
      "num_tokens": 25125619.0,
      "step": 13480
    },
    {
      "entropy": 5.955485486984253,
      "epoch": 1.1908336276933946,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004864006653992546,
      "loss": 5.6833,
      "mean_token_accuracy": 0.1734369292855263,
      "num_tokens": 25134782.0,
      "step": 13485
    },
    {
      "entropy": 5.996906709671021,
      "epoch": 1.191275167785235,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048638988100299283,
      "loss": 5.72,
      "mean_token_accuracy": 0.17814869582653045,
      "num_tokens": 25144709.0,
      "step": 13490
    },
    {
      "entropy": 5.9380110740661625,
      "epoch": 1.1917167078770752,
      "grad_norm": 1.5,
      "learning_rate": 0.0004863790924657236,
      "loss": 5.5987,
      "mean_token_accuracy": 0.1779467612504959,
      "num_tokens": 25153771.0,
      "step": 13495
    },
    {
      "entropy": 5.871428918838501,
      "epoch": 1.1921582479689157,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000486368299787658,
      "loss": 5.7586,
      "mean_token_accuracy": 0.171010385453701,
      "num_tokens": 25163217.0,
      "step": 13500
    },
    {
      "entropy": 5.973069477081299,
      "epoch": 1.192599788060756,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048635750296900784,
      "loss": 5.7866,
      "mean_token_accuracy": 0.1723656550049782,
      "num_tokens": 25172762.0,
      "step": 13505
    },
    {
      "entropy": 6.006270933151245,
      "epoch": 1.1930413281525962,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004863467020099842,
      "loss": 5.698,
      "mean_token_accuracy": 0.17801835536956787,
      "num_tokens": 25182349.0,
      "step": 13510
    },
    {
      "entropy": 5.954557275772094,
      "epoch": 1.1934828682444365,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048633589691079895,
      "loss": 5.7789,
      "mean_token_accuracy": 0.16486638635396958,
      "num_tokens": 25191971.0,
      "step": 13515
    },
    {
      "entropy": 5.817729377746582,
      "epoch": 1.193924408336277,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004863250876716637,
      "loss": 5.5438,
      "mean_token_accuracy": 0.18471172899007798,
      "num_tokens": 25200476.0,
      "step": 13520
    },
    {
      "entropy": 5.899543285369873,
      "epoch": 1.1943659484281173,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004863142742927902,
      "loss": 5.6619,
      "mean_token_accuracy": 0.1784363254904747,
      "num_tokens": 25209748.0,
      "step": 13525
    },
    {
      "entropy": 5.999986839294434,
      "epoch": 1.1948074885199576,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004863034567743901,
      "loss": 5.742,
      "mean_token_accuracy": 0.17008337676525115,
      "num_tokens": 25220140.0,
      "step": 13530
    },
    {
      "entropy": 5.874360752105713,
      "epoch": 1.195249028611798,
      "grad_norm": 2.0,
      "learning_rate": 0.0004862926351166756,
      "loss": 5.6143,
      "mean_token_accuracy": 0.18735847920179366,
      "num_tokens": 25228424.0,
      "step": 13535
    },
    {
      "entropy": 5.936167144775391,
      "epoch": 1.1956905687036383,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004862818093198585,
      "loss": 5.7309,
      "mean_token_accuracy": 0.17327931970357896,
      "num_tokens": 25237448.0,
      "step": 13540
    },
    {
      "entropy": 5.906550788879395,
      "epoch": 1.1961321087954786,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004862709793841509,
      "loss": 5.6424,
      "mean_token_accuracy": 0.179931178689003,
      "num_tokens": 25247124.0,
      "step": 13545
    },
    {
      "entropy": 5.861179399490356,
      "epoch": 1.1965736488873189,
      "grad_norm": 1.890625,
      "learning_rate": 0.00048626014530976495,
      "loss": 5.6754,
      "mean_token_accuracy": 0.17156106233596802,
      "num_tokens": 25257363.0,
      "step": 13550
    },
    {
      "entropy": 5.930722856521607,
      "epoch": 1.1970151889791594,
      "grad_norm": 1.84375,
      "learning_rate": 0.00048624930709691296,
      "loss": 5.7629,
      "mean_token_accuracy": 0.16417711973190308,
      "num_tokens": 25266837.0,
      "step": 13555
    },
    {
      "entropy": 5.962302589416504,
      "epoch": 1.1974567290709996,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048623846474580716,
      "loss": 5.7944,
      "mean_token_accuracy": 0.16165902316570283,
      "num_tokens": 25275767.0,
      "step": 13560
    },
    {
      "entropy": 5.926019859313965,
      "epoch": 1.19789826916284,
      "grad_norm": 1.8125,
      "learning_rate": 0.00048622761825665994,
      "loss": 5.7416,
      "mean_token_accuracy": 0.1642533838748932,
      "num_tokens": 25285212.0,
      "step": 13565
    },
    {
      "entropy": 5.914536857604981,
      "epoch": 1.1983398092546804,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004862167676296839,
      "loss": 5.6208,
      "mean_token_accuracy": 0.17389897257089615,
      "num_tokens": 25294698.0,
      "step": 13570
    },
    {
      "entropy": 5.937765741348267,
      "epoch": 1.1987813493465207,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004862059128650915,
      "loss": 5.6997,
      "mean_token_accuracy": 0.17571492195129396,
      "num_tokens": 25304343.0,
      "step": 13575
    },
    {
      "entropy": 5.973829126358032,
      "epoch": 1.199222889438361,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048619505396309525,
      "loss": 5.6328,
      "mean_token_accuracy": 0.17361133247613908,
      "num_tokens": 25312542.0,
      "step": 13580
    },
    {
      "entropy": 5.906447124481201,
      "epoch": 1.1996644295302012,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004861841909239081,
      "loss": 5.7389,
      "mean_token_accuracy": 0.17303294390439988,
      "num_tokens": 25322665.0,
      "step": 13585
    },
    {
      "entropy": 5.888718509674073,
      "epoch": 1.2001059696220417,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048617332374774276,
      "loss": 5.676,
      "mean_token_accuracy": 0.17804051339626312,
      "num_tokens": 25332234.0,
      "step": 13590
    },
    {
      "entropy": 5.939677715301514,
      "epoch": 1.200547509713882,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048616245243481216,
      "loss": 5.7597,
      "mean_token_accuracy": 0.1697024166584015,
      "num_tokens": 25341462.0,
      "step": 13595
    },
    {
      "entropy": 5.954547977447509,
      "epoch": 1.2009890498057223,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00048615157698532913,
      "loss": 5.7078,
      "mean_token_accuracy": 0.178890098631382,
      "num_tokens": 25351771.0,
      "step": 13600
    },
    {
      "entropy": 5.946116733551025,
      "epoch": 1.2014305898975628,
      "grad_norm": 2.640625,
      "learning_rate": 0.00048614069739950683,
      "loss": 5.6657,
      "mean_token_accuracy": 0.1788325145840645,
      "num_tokens": 25360108.0,
      "step": 13605
    },
    {
      "entropy": 5.873933267593384,
      "epoch": 1.201872129989403,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004861298136775583,
      "loss": 5.6711,
      "mean_token_accuracy": 0.17383701354265213,
      "num_tokens": 25369635.0,
      "step": 13610
    },
    {
      "entropy": 5.904291200637817,
      "epoch": 1.2023136700812433,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004861189258196969,
      "loss": 5.6571,
      "mean_token_accuracy": 0.1819812163710594,
      "num_tokens": 25379623.0,
      "step": 13615
    },
    {
      "entropy": 5.958040857315064,
      "epoch": 1.2027552101730836,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004861080338261357,
      "loss": 5.7584,
      "mean_token_accuracy": 0.17228599339723588,
      "num_tokens": 25388591.0,
      "step": 13620
    },
    {
      "entropy": 5.891545724868775,
      "epoch": 1.2031967502649241,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004860971376970882,
      "loss": 5.6209,
      "mean_token_accuracy": 0.1814316615462303,
      "num_tokens": 25397865.0,
      "step": 13625
    },
    {
      "entropy": 5.8059052467346195,
      "epoch": 1.2036382903567644,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048608623743276794,
      "loss": 5.5786,
      "mean_token_accuracy": 0.18578606247901916,
      "num_tokens": 25407143.0,
      "step": 13630
    },
    {
      "entropy": 5.928137874603271,
      "epoch": 1.2040798304486047,
      "grad_norm": 1.796875,
      "learning_rate": 0.00048607533303338823,
      "loss": 5.7275,
      "mean_token_accuracy": 0.1744789496064186,
      "num_tokens": 25415721.0,
      "step": 13635
    },
    {
      "entropy": 5.941731643676758,
      "epoch": 1.2045213705404452,
      "grad_norm": 1.375,
      "learning_rate": 0.00048606442449916284,
      "loss": 5.7556,
      "mean_token_accuracy": 0.1777839705348015,
      "num_tokens": 25425319.0,
      "step": 13640
    },
    {
      "entropy": 5.929516458511353,
      "epoch": 1.2049629106322854,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048605351183030537,
      "loss": 5.6983,
      "mean_token_accuracy": 0.17948974072933196,
      "num_tokens": 25435477.0,
      "step": 13645
    },
    {
      "entropy": 5.8874876499176025,
      "epoch": 1.2054044507241257,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004860425950270296,
      "loss": 5.697,
      "mean_token_accuracy": 0.1757381409406662,
      "num_tokens": 25445633.0,
      "step": 13650
    },
    {
      "entropy": 5.9028815746307375,
      "epoch": 1.205845990815966,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048603167408954953,
      "loss": 5.671,
      "mean_token_accuracy": 0.17585898488759993,
      "num_tokens": 25455228.0,
      "step": 13655
    },
    {
      "entropy": 5.931520223617554,
      "epoch": 1.2062875309078065,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048602074901807894,
      "loss": 5.6615,
      "mean_token_accuracy": 0.1745007261633873,
      "num_tokens": 25463992.0,
      "step": 13660
    },
    {
      "entropy": 5.950996828079224,
      "epoch": 1.2067290709996468,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048600981981283185,
      "loss": 5.6404,
      "mean_token_accuracy": 0.1851426988840103,
      "num_tokens": 25472985.0,
      "step": 13665
    },
    {
      "entropy": 5.909234952926636,
      "epoch": 1.207170611091487,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004859988864740225,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17261201441287993,
      "num_tokens": 25482165.0,
      "step": 13670
    },
    {
      "entropy": 5.866550636291504,
      "epoch": 1.2076121511833275,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004859879490018649,
      "loss": 5.7462,
      "mean_token_accuracy": 0.17762482017278672,
      "num_tokens": 25490643.0,
      "step": 13675
    },
    {
      "entropy": 5.938075685501099,
      "epoch": 1.2080536912751678,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004859770073965733,
      "loss": 5.6513,
      "mean_token_accuracy": 0.17934835106134414,
      "num_tokens": 25499954.0,
      "step": 13680
    },
    {
      "entropy": 5.9424785614013675,
      "epoch": 1.208495231367008,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048596606165836223,
      "loss": 5.6885,
      "mean_token_accuracy": 0.17619566172361373,
      "num_tokens": 25509512.0,
      "step": 13685
    },
    {
      "entropy": 5.908395051956177,
      "epoch": 1.2089367714588484,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048595511178744594,
      "loss": 5.6197,
      "mean_token_accuracy": 0.17326670438051223,
      "num_tokens": 25518909.0,
      "step": 13690
    },
    {
      "entropy": 5.904380035400391,
      "epoch": 1.2093783115506889,
      "grad_norm": 1.4375,
      "learning_rate": 0.000485944157784039,
      "loss": 5.6978,
      "mean_token_accuracy": 0.17091204673051835,
      "num_tokens": 25528221.0,
      "step": 13695
    },
    {
      "entropy": 5.899301290512085,
      "epoch": 1.2098198516425291,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000485933199648356,
      "loss": 5.6908,
      "mean_token_accuracy": 0.16989997029304504,
      "num_tokens": 25537428.0,
      "step": 13700
    },
    {
      "entropy": 5.908147192001342,
      "epoch": 1.2102613917343694,
      "grad_norm": 1.859375,
      "learning_rate": 0.00048592223738061155,
      "loss": 5.694,
      "mean_token_accuracy": 0.17192309647798537,
      "num_tokens": 25547608.0,
      "step": 13705
    },
    {
      "entropy": 5.9348273277282715,
      "epoch": 1.21070293182621,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00048591127098102045,
      "loss": 5.787,
      "mean_token_accuracy": 0.1691073402762413,
      "num_tokens": 25557482.0,
      "step": 13710
    },
    {
      "entropy": 5.88905143737793,
      "epoch": 1.2111444719180502,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004859003004497974,
      "loss": 5.6282,
      "mean_token_accuracy": 0.17827312499284745,
      "num_tokens": 25566967.0,
      "step": 13715
    },
    {
      "entropy": 5.946938610076904,
      "epoch": 1.2115860120098905,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048588932578715754,
      "loss": 5.6399,
      "mean_token_accuracy": 0.17730989009141923,
      "num_tokens": 25576114.0,
      "step": 13720
    },
    {
      "entropy": 5.970891094207763,
      "epoch": 1.2120275521017307,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048587834699331567,
      "loss": 5.8427,
      "mean_token_accuracy": 0.1615636706352234,
      "num_tokens": 25585953.0,
      "step": 13725
    },
    {
      "entropy": 5.948155069351197,
      "epoch": 1.2124690921935712,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000485867364068487,
      "loss": 5.8226,
      "mean_token_accuracy": 0.1646987482905388,
      "num_tokens": 25595445.0,
      "step": 13730
    },
    {
      "entropy": 5.975005578994751,
      "epoch": 1.2129106322854115,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048585637701288646,
      "loss": 5.676,
      "mean_token_accuracy": 0.1748975932598114,
      "num_tokens": 25605883.0,
      "step": 13735
    },
    {
      "entropy": 5.916291904449463,
      "epoch": 1.2133521723772518,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004858453858267295,
      "loss": 5.7304,
      "mean_token_accuracy": 0.16573525071144105,
      "num_tokens": 25615653.0,
      "step": 13740
    },
    {
      "entropy": 5.8757835865020756,
      "epoch": 1.2137937124690923,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004858343905102314,
      "loss": 5.6447,
      "mean_token_accuracy": 0.17673285752534867,
      "num_tokens": 25625043.0,
      "step": 13745
    },
    {
      "entropy": 5.973969173431397,
      "epoch": 1.2142352525609326,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048582339106360736,
      "loss": 5.6585,
      "mean_token_accuracy": 0.18139365017414094,
      "num_tokens": 25633642.0,
      "step": 13750
    },
    {
      "entropy": 5.966825723648071,
      "epoch": 1.2146767926527728,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048581238748707304,
      "loss": 5.7269,
      "mean_token_accuracy": 0.17123862206935883,
      "num_tokens": 25642154.0,
      "step": 13755
    },
    {
      "entropy": 5.893309116363525,
      "epoch": 1.2151183327446131,
      "grad_norm": 1.921875,
      "learning_rate": 0.000485801379780844,
      "loss": 5.6256,
      "mean_token_accuracy": 0.1780093789100647,
      "num_tokens": 25650250.0,
      "step": 13760
    },
    {
      "entropy": 5.88436541557312,
      "epoch": 1.2155598728364536,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048579036794513584,
      "loss": 5.6804,
      "mean_token_accuracy": 0.17928278148174287,
      "num_tokens": 25659789.0,
      "step": 13765
    },
    {
      "entropy": 5.890221643447876,
      "epoch": 1.216001412928294,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048577935198016413,
      "loss": 5.7178,
      "mean_token_accuracy": 0.17221244424581528,
      "num_tokens": 25668658.0,
      "step": 13770
    },
    {
      "entropy": 5.974110651016235,
      "epoch": 1.2164429530201342,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004857683318861449,
      "loss": 5.7321,
      "mean_token_accuracy": 0.17247796654701233,
      "num_tokens": 25679082.0,
      "step": 13775
    },
    {
      "entropy": 5.974181890487671,
      "epoch": 1.2168844931119747,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048575730766329384,
      "loss": 5.7108,
      "mean_token_accuracy": 0.16872587651014329,
      "num_tokens": 25688408.0,
      "step": 13780
    },
    {
      "entropy": 5.866273403167725,
      "epoch": 1.217326033203815,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048574627931182704,
      "loss": 5.6628,
      "mean_token_accuracy": 0.176899354159832,
      "num_tokens": 25697415.0,
      "step": 13785
    },
    {
      "entropy": 5.910143041610718,
      "epoch": 1.2177675732956552,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048573524683196047,
      "loss": 5.7271,
      "mean_token_accuracy": 0.17175474017858505,
      "num_tokens": 25706330.0,
      "step": 13790
    },
    {
      "entropy": 5.9862127780914305,
      "epoch": 1.2182091133874955,
      "grad_norm": 1.375,
      "learning_rate": 0.0004857242102239102,
      "loss": 5.7012,
      "mean_token_accuracy": 0.17438223958015442,
      "num_tokens": 25716198.0,
      "step": 13795
    },
    {
      "entropy": 5.9755308628082275,
      "epoch": 1.218650653479336,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004857131694878926,
      "loss": 5.6987,
      "mean_token_accuracy": 0.16604396253824233,
      "num_tokens": 25726486.0,
      "step": 13800
    },
    {
      "entropy": 5.88919734954834,
      "epoch": 1.2190921935711763,
      "grad_norm": 1.9375,
      "learning_rate": 0.00048570212462412376,
      "loss": 5.6028,
      "mean_token_accuracy": 0.17944130599498748,
      "num_tokens": 25735559.0,
      "step": 13805
    },
    {
      "entropy": 5.93406925201416,
      "epoch": 1.2195337336630165,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004856910756328202,
      "loss": 5.692,
      "mean_token_accuracy": 0.17372544407844542,
      "num_tokens": 25744781.0,
      "step": 13810
    },
    {
      "entropy": 5.920233058929443,
      "epoch": 1.219975273754857,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048568002251419835,
      "loss": 5.6623,
      "mean_token_accuracy": 0.18133081644773483,
      "num_tokens": 25753984.0,
      "step": 13815
    },
    {
      "entropy": 5.830778455734253,
      "epoch": 1.2204168138466973,
      "grad_norm": 1.375,
      "learning_rate": 0.0004856689652684746,
      "loss": 5.7435,
      "mean_token_accuracy": 0.17089764773845673,
      "num_tokens": 25763378.0,
      "step": 13820
    },
    {
      "entropy": 5.8689268112182615,
      "epoch": 1.2208583539385376,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004856579038958656,
      "loss": 5.6821,
      "mean_token_accuracy": 0.17523577362298964,
      "num_tokens": 25771907.0,
      "step": 13825
    },
    {
      "entropy": 6.024119186401367,
      "epoch": 1.2212998940303779,
      "grad_norm": 1.859375,
      "learning_rate": 0.00048564683839658805,
      "loss": 5.7287,
      "mean_token_accuracy": 0.17407168000936507,
      "num_tokens": 25780464.0,
      "step": 13830
    },
    {
      "entropy": 5.979361057281494,
      "epoch": 1.2217414341222184,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048563576877085887,
      "loss": 5.7156,
      "mean_token_accuracy": 0.17784766256809234,
      "num_tokens": 25790300.0,
      "step": 13835
    },
    {
      "entropy": 5.918692874908447,
      "epoch": 1.2221829742140586,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004856246950188947,
      "loss": 5.6405,
      "mean_token_accuracy": 0.17528326213359832,
      "num_tokens": 25799103.0,
      "step": 13840
    },
    {
      "entropy": 5.980585098266602,
      "epoch": 1.222624514305899,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048561361714091257,
      "loss": 5.7134,
      "mean_token_accuracy": 0.1715540662407875,
      "num_tokens": 25808506.0,
      "step": 13845
    },
    {
      "entropy": 5.997957515716553,
      "epoch": 1.2230660543977394,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048560253513712945,
      "loss": 5.697,
      "mean_token_accuracy": 0.1737539365887642,
      "num_tokens": 25817965.0,
      "step": 13850
    },
    {
      "entropy": 5.833532905578613,
      "epoch": 1.2235075944895797,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004855914490077624,
      "loss": 5.6854,
      "mean_token_accuracy": 0.17347604930400848,
      "num_tokens": 25827874.0,
      "step": 13855
    },
    {
      "entropy": 6.010499286651611,
      "epoch": 1.22394913458142,
      "grad_norm": 1.84375,
      "learning_rate": 0.00048558035875302864,
      "loss": 5.7931,
      "mean_token_accuracy": 0.1746525228023529,
      "num_tokens": 25836840.0,
      "step": 13860
    },
    {
      "entropy": 5.985385084152222,
      "epoch": 1.2243906746732602,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004855692643731455,
      "loss": 5.6761,
      "mean_token_accuracy": 0.18095205426216127,
      "num_tokens": 25846344.0,
      "step": 13865
    },
    {
      "entropy": 5.847048711776734,
      "epoch": 1.2248322147651007,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00048555816586833006,
      "loss": 5.6686,
      "mean_token_accuracy": 0.17741122394800185,
      "num_tokens": 25855336.0,
      "step": 13870
    },
    {
      "entropy": 5.95607271194458,
      "epoch": 1.225273754856941,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00048554706323880007,
      "loss": 5.7144,
      "mean_token_accuracy": 0.17336337119340897,
      "num_tokens": 25864402.0,
      "step": 13875
    },
    {
      "entropy": 5.978678560256958,
      "epoch": 1.2257152949487813,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004855359564847728,
      "loss": 5.6982,
      "mean_token_accuracy": 0.17589497715234756,
      "num_tokens": 25873326.0,
      "step": 13880
    },
    {
      "entropy": 5.9500470638275145,
      "epoch": 1.2261568350406218,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048552484560646585,
      "loss": 5.7873,
      "mean_token_accuracy": 0.1604952931404114,
      "num_tokens": 25882521.0,
      "step": 13885
    },
    {
      "entropy": 5.868669176101685,
      "epoch": 1.226598375132462,
      "grad_norm": 2.0,
      "learning_rate": 0.00048551373060409694,
      "loss": 5.5954,
      "mean_token_accuracy": 0.18266200125217438,
      "num_tokens": 25891087.0,
      "step": 13890
    },
    {
      "entropy": 5.979387617111206,
      "epoch": 1.2270399152243023,
      "grad_norm": 1.71875,
      "learning_rate": 0.00048550261147788375,
      "loss": 5.6595,
      "mean_token_accuracy": 0.17208310961723328,
      "num_tokens": 25900757.0,
      "step": 13895
    },
    {
      "entropy": 5.934347915649414,
      "epoch": 1.2274814553161426,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048549148822804415,
      "loss": 5.6559,
      "mean_token_accuracy": 0.18271127194166184,
      "num_tokens": 25909801.0,
      "step": 13900
    },
    {
      "entropy": 5.974944400787353,
      "epoch": 1.227922995407983,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000485480360854796,
      "loss": 5.7904,
      "mean_token_accuracy": 0.15782829523086547,
      "num_tokens": 25919142.0,
      "step": 13905
    },
    {
      "entropy": 5.957124423980713,
      "epoch": 1.2283645354998234,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004854692293583573,
      "loss": 5.6618,
      "mean_token_accuracy": 0.17944327592849732,
      "num_tokens": 25927560.0,
      "step": 13910
    },
    {
      "entropy": 5.9744116306304935,
      "epoch": 1.2288060755916637,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048545809373894605,
      "loss": 5.773,
      "mean_token_accuracy": 0.17189525067806244,
      "num_tokens": 25937184.0,
      "step": 13915
    },
    {
      "entropy": 6.042012166976929,
      "epoch": 1.2292476156835042,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048544695399678044,
      "loss": 5.7853,
      "mean_token_accuracy": 0.16290355771780013,
      "num_tokens": 25946641.0,
      "step": 13920
    },
    {
      "entropy": 6.060760974884033,
      "epoch": 1.2296891557753444,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004854358101320787,
      "loss": 5.8192,
      "mean_token_accuracy": 0.17275043725967407,
      "num_tokens": 25955910.0,
      "step": 13925
    },
    {
      "entropy": 5.917440700531006,
      "epoch": 1.2301306958671847,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048542466214505914,
      "loss": 5.7224,
      "mean_token_accuracy": 0.17595441043376922,
      "num_tokens": 25966580.0,
      "step": 13930
    },
    {
      "entropy": 5.892169713973999,
      "epoch": 1.230572235959025,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048541351003594015,
      "loss": 5.5946,
      "mean_token_accuracy": 0.18098146617412567,
      "num_tokens": 25975815.0,
      "step": 13935
    },
    {
      "entropy": 5.901758241653442,
      "epoch": 1.2310137760508655,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004854023538049401,
      "loss": 5.6714,
      "mean_token_accuracy": 0.1806225970387459,
      "num_tokens": 25984784.0,
      "step": 13940
    },
    {
      "entropy": 5.889666986465454,
      "epoch": 1.2314553161427058,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048539119345227774,
      "loss": 5.6251,
      "mean_token_accuracy": 0.18624144792556763,
      "num_tokens": 25994125.0,
      "step": 13945
    },
    {
      "entropy": 5.902031803131104,
      "epoch": 1.231896856234546,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004853800289781714,
      "loss": 5.6727,
      "mean_token_accuracy": 0.17630417495965958,
      "num_tokens": 26002876.0,
      "step": 13950
    },
    {
      "entropy": 5.878540658950806,
      "epoch": 1.2323383963263865,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048536886038284,
      "loss": 5.6878,
      "mean_token_accuracy": 0.17998367995023729,
      "num_tokens": 26012591.0,
      "step": 13955
    },
    {
      "entropy": 5.980203914642334,
      "epoch": 1.2327799364182268,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048535768766650225,
      "loss": 5.7973,
      "mean_token_accuracy": 0.16152193397283554,
      "num_tokens": 26022107.0,
      "step": 13960
    },
    {
      "entropy": 5.996450614929199,
      "epoch": 1.233221476510067,
      "grad_norm": 1.65625,
      "learning_rate": 0.000485346510829377,
      "loss": 5.7597,
      "mean_token_accuracy": 0.17914632111787795,
      "num_tokens": 26031035.0,
      "step": 13965
    },
    {
      "entropy": 5.906059455871582,
      "epoch": 1.2336630166019074,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048533532987168334,
      "loss": 5.5898,
      "mean_token_accuracy": 0.1870785176753998,
      "num_tokens": 26039980.0,
      "step": 13970
    },
    {
      "entropy": 5.996763038635254,
      "epoch": 1.2341045566937479,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048532414479364016,
      "loss": 5.811,
      "mean_token_accuracy": 0.16511304825544357,
      "num_tokens": 26049389.0,
      "step": 13975
    },
    {
      "entropy": 5.937296724319458,
      "epoch": 1.2345460967855881,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004853129555954666,
      "loss": 5.6577,
      "mean_token_accuracy": 0.17810503095388414,
      "num_tokens": 26058868.0,
      "step": 13980
    },
    {
      "entropy": 5.97762975692749,
      "epoch": 1.2349876368774284,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004853017622773818,
      "loss": 5.7264,
      "mean_token_accuracy": 0.16853921115398407,
      "num_tokens": 26068025.0,
      "step": 13985
    },
    {
      "entropy": 5.9750360488891605,
      "epoch": 1.235429176969269,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004852905648396051,
      "loss": 5.6949,
      "mean_token_accuracy": 0.16979873031377793,
      "num_tokens": 26076410.0,
      "step": 13990
    },
    {
      "entropy": 5.881725597381592,
      "epoch": 1.2358707170611092,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048527936328235583,
      "loss": 5.6477,
      "mean_token_accuracy": 0.18216056376695633,
      "num_tokens": 26085624.0,
      "step": 13995
    },
    {
      "entropy": 5.943519496917725,
      "epoch": 1.2363122571529495,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004852681576058534,
      "loss": 5.724,
      "mean_token_accuracy": 0.17696210443973542,
      "num_tokens": 26094898.0,
      "step": 14000
    },
    {
      "entropy": 5.939918136596679,
      "epoch": 1.2367537972447897,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004852569478103174,
      "loss": 5.6953,
      "mean_token_accuracy": 0.17627927660942078,
      "num_tokens": 26105041.0,
      "step": 14005
    },
    {
      "entropy": 5.952569150924683,
      "epoch": 1.2371953373366302,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048524573389596736,
      "loss": 5.6679,
      "mean_token_accuracy": 0.17571507692337035,
      "num_tokens": 26114140.0,
      "step": 14010
    },
    {
      "entropy": 5.873074102401733,
      "epoch": 1.2376368774284705,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048523451586302295,
      "loss": 5.5397,
      "mean_token_accuracy": 0.18858534395694732,
      "num_tokens": 26122611.0,
      "step": 14015
    },
    {
      "entropy": 5.8770781517028805,
      "epoch": 1.2380784175203108,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000485223293711704,
      "loss": 5.6103,
      "mean_token_accuracy": 0.17926859557628633,
      "num_tokens": 26131548.0,
      "step": 14020
    },
    {
      "entropy": 5.882212781906128,
      "epoch": 1.2385199576121513,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048521206744223024,
      "loss": 5.6092,
      "mean_token_accuracy": 0.17398136258125305,
      "num_tokens": 26139806.0,
      "step": 14025
    },
    {
      "entropy": 6.012924766540527,
      "epoch": 1.2389614977039916,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048520083705482165,
      "loss": 5.7327,
      "mean_token_accuracy": 0.17417315989732743,
      "num_tokens": 26148960.0,
      "step": 14030
    },
    {
      "entropy": 5.981174278259277,
      "epoch": 1.2394030377958318,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004851896025496982,
      "loss": 5.7331,
      "mean_token_accuracy": 0.1774987742304802,
      "num_tokens": 26158498.0,
      "step": 14035
    },
    {
      "entropy": 5.886492156982422,
      "epoch": 1.239844577887672,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048517836392707994,
      "loss": 5.6048,
      "mean_token_accuracy": 0.17816965878009797,
      "num_tokens": 26167319.0,
      "step": 14040
    },
    {
      "entropy": 5.882360935211182,
      "epoch": 1.2402861179795126,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048516712118718713,
      "loss": 5.6033,
      "mean_token_accuracy": 0.18233618885278702,
      "num_tokens": 26176059.0,
      "step": 14045
    },
    {
      "entropy": 5.93251781463623,
      "epoch": 1.2407276580713529,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004851558743302399,
      "loss": 5.6349,
      "mean_token_accuracy": 0.18068666011095047,
      "num_tokens": 26185102.0,
      "step": 14050
    },
    {
      "entropy": 5.894285774230957,
      "epoch": 1.2411691981631932,
      "grad_norm": 2.125,
      "learning_rate": 0.0004851446233564586,
      "loss": 5.6482,
      "mean_token_accuracy": 0.17463642805814744,
      "num_tokens": 26194873.0,
      "step": 14055
    },
    {
      "entropy": 5.922954654693603,
      "epoch": 1.2416107382550337,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004851333682660637,
      "loss": 5.7036,
      "mean_token_accuracy": 0.17505521774291993,
      "num_tokens": 26205362.0,
      "step": 14060
    },
    {
      "entropy": 5.836306190490722,
      "epoch": 1.242052278346874,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048512210905927555,
      "loss": 5.5997,
      "mean_token_accuracy": 0.17720614224672318,
      "num_tokens": 26214266.0,
      "step": 14065
    },
    {
      "entropy": 5.987836647033691,
      "epoch": 1.2424938184387142,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004851108457363149,
      "loss": 5.7601,
      "mean_token_accuracy": 0.17741140723228455,
      "num_tokens": 26223069.0,
      "step": 14070
    },
    {
      "entropy": 5.892253780364991,
      "epoch": 1.2429353585305545,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004850995782974022,
      "loss": 5.6946,
      "mean_token_accuracy": 0.17419238686561583,
      "num_tokens": 26232547.0,
      "step": 14075
    },
    {
      "entropy": 5.992074823379516,
      "epoch": 1.243376898622395,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004850883067427582,
      "loss": 5.6791,
      "mean_token_accuracy": 0.17146124243736266,
      "num_tokens": 26241014.0,
      "step": 14080
    },
    {
      "entropy": 5.909090995788574,
      "epoch": 1.2438184387142353,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048507703107260384,
      "loss": 5.7306,
      "mean_token_accuracy": 0.17725476026535034,
      "num_tokens": 26250784.0,
      "step": 14085
    },
    {
      "entropy": 5.94590802192688,
      "epoch": 1.2442599788060755,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048506575128715984,
      "loss": 5.6644,
      "mean_token_accuracy": 0.17474059909582138,
      "num_tokens": 26259769.0,
      "step": 14090
    },
    {
      "entropy": 5.942369413375855,
      "epoch": 1.244701518897916,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004850544673866474,
      "loss": 5.6807,
      "mean_token_accuracy": 0.17623415291309358,
      "num_tokens": 26269038.0,
      "step": 14095
    },
    {
      "entropy": 5.9494977474212645,
      "epoch": 1.2451430589897563,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004850431793712873,
      "loss": 5.6513,
      "mean_token_accuracy": 0.16738424003124236,
      "num_tokens": 26278657.0,
      "step": 14100
    },
    {
      "entropy": 5.883402347564697,
      "epoch": 1.2455845990815966,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048503188724130075,
      "loss": 5.6705,
      "mean_token_accuracy": 0.1724415823817253,
      "num_tokens": 26287458.0,
      "step": 14105
    },
    {
      "entropy": 5.9546013355255125,
      "epoch": 1.2460261391734369,
      "grad_norm": 1.453125,
      "learning_rate": 0.000485020590996909,
      "loss": 5.7142,
      "mean_token_accuracy": 0.1705986440181732,
      "num_tokens": 26296324.0,
      "step": 14110
    },
    {
      "entropy": 5.9133118152618405,
      "epoch": 1.2464676792652774,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048500929063833334,
      "loss": 5.6297,
      "mean_token_accuracy": 0.1805555820465088,
      "num_tokens": 26304799.0,
      "step": 14115
    },
    {
      "entropy": 5.9180983543396,
      "epoch": 1.2469092193571176,
      "grad_norm": 1.96875,
      "learning_rate": 0.00048499798616579505,
      "loss": 5.7605,
      "mean_token_accuracy": 0.17488011419773103,
      "num_tokens": 26314223.0,
      "step": 14120
    },
    {
      "entropy": 6.002673530578614,
      "epoch": 1.247350759448958,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00048498667757951574,
      "loss": 5.7647,
      "mean_token_accuracy": 0.16344328075647355,
      "num_tokens": 26323599.0,
      "step": 14125
    },
    {
      "entropy": 5.9948667049407955,
      "epoch": 1.2477922995407984,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004849753648797168,
      "loss": 5.7843,
      "mean_token_accuracy": 0.1757570594549179,
      "num_tokens": 26331929.0,
      "step": 14130
    },
    {
      "entropy": 5.934284830093384,
      "epoch": 1.2482338396326387,
      "grad_norm": 1.625,
      "learning_rate": 0.0004849640480666199,
      "loss": 5.6688,
      "mean_token_accuracy": 0.17607956975698472,
      "num_tokens": 26340929.0,
      "step": 14135
    },
    {
      "entropy": 5.977491283416748,
      "epoch": 1.248675379724479,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004849527271404467,
      "loss": 5.6509,
      "mean_token_accuracy": 0.17545788884162902,
      "num_tokens": 26350393.0,
      "step": 14140
    },
    {
      "entropy": 6.036669445037842,
      "epoch": 1.2491169198163192,
      "grad_norm": 1.625,
      "learning_rate": 0.00048494140210141897,
      "loss": 5.8434,
      "mean_token_accuracy": 0.170181305706501,
      "num_tokens": 26360709.0,
      "step": 14145
    },
    {
      "entropy": 5.926997232437134,
      "epoch": 1.2495584599081597,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048493007294975855,
      "loss": 5.6437,
      "mean_token_accuracy": 0.1822052776813507,
      "num_tokens": 26369447.0,
      "step": 14150
    },
    {
      "entropy": 5.930496454238892,
      "epoch": 1.25,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048491873968568746,
      "loss": 5.6294,
      "mean_token_accuracy": 0.18320029079914094,
      "num_tokens": 26379281.0,
      "step": 14155
    },
    {
      "entropy": 5.811459541320801,
      "epoch": 1.2504415400918403,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004849074023094276,
      "loss": 5.583,
      "mean_token_accuracy": 0.187502658367157,
      "num_tokens": 26387664.0,
      "step": 14160
    },
    {
      "entropy": 5.849246835708618,
      "epoch": 1.2508830801836806,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048489606082120106,
      "loss": 5.6134,
      "mean_token_accuracy": 0.17633943557739257,
      "num_tokens": 26397485.0,
      "step": 14165
    },
    {
      "entropy": 5.960401487350464,
      "epoch": 1.251324620275521,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004848847152212301,
      "loss": 5.567,
      "mean_token_accuracy": 0.18389885872602463,
      "num_tokens": 26406307.0,
      "step": 14170
    },
    {
      "entropy": 5.98303542137146,
      "epoch": 1.2517661603673613,
      "grad_norm": 1.28125,
      "learning_rate": 0.000484873365509737,
      "loss": 5.7824,
      "mean_token_accuracy": 0.172248375415802,
      "num_tokens": 26416155.0,
      "step": 14175
    },
    {
      "entropy": 5.940882587432862,
      "epoch": 1.2522077004592016,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048486201168694397,
      "loss": 5.6747,
      "mean_token_accuracy": 0.17158570438623427,
      "num_tokens": 26424147.0,
      "step": 14180
    },
    {
      "entropy": 5.931738710403442,
      "epoch": 1.252649240551042,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048485065375307353,
      "loss": 5.672,
      "mean_token_accuracy": 0.17452831715345382,
      "num_tokens": 26433249.0,
      "step": 14185
    },
    {
      "entropy": 5.961124038696289,
      "epoch": 1.2530907806428824,
      "grad_norm": 2.96875,
      "learning_rate": 0.00048483929170834805,
      "loss": 5.7093,
      "mean_token_accuracy": 0.17901448607444764,
      "num_tokens": 26442346.0,
      "step": 14190
    },
    {
      "entropy": 5.94314136505127,
      "epoch": 1.2535323207347226,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048482792555299027,
      "loss": 5.6671,
      "mean_token_accuracy": 0.17966310381889344,
      "num_tokens": 26450885.0,
      "step": 14195
    },
    {
      "entropy": 5.895475673675537,
      "epoch": 1.2539738608265631,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004848165552872227,
      "loss": 5.7232,
      "mean_token_accuracy": 0.17767656296491624,
      "num_tokens": 26459827.0,
      "step": 14200
    },
    {
      "entropy": 5.87793459892273,
      "epoch": 1.2544154009184034,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004848051809112682,
      "loss": 5.6932,
      "mean_token_accuracy": 0.1701851323246956,
      "num_tokens": 26468357.0,
      "step": 14205
    },
    {
      "entropy": 5.911648178100586,
      "epoch": 1.2548569410102437,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048479380242534955,
      "loss": 5.7945,
      "mean_token_accuracy": 0.17521611750125884,
      "num_tokens": 26477396.0,
      "step": 14210
    },
    {
      "entropy": 5.96491494178772,
      "epoch": 1.2552984811020842,
      "grad_norm": 2.125,
      "learning_rate": 0.0004847824198296896,
      "loss": 5.7238,
      "mean_token_accuracy": 0.1774645045399666,
      "num_tokens": 26486468.0,
      "step": 14215
    },
    {
      "entropy": 6.010610151290893,
      "epoch": 1.2557400211939245,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004847710331245113,
      "loss": 5.6789,
      "mean_token_accuracy": 0.17831059843301772,
      "num_tokens": 26496157.0,
      "step": 14220
    },
    {
      "entropy": 5.888781309127808,
      "epoch": 1.2561815612857647,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048475964231003776,
      "loss": 5.5541,
      "mean_token_accuracy": 0.18786979019641875,
      "num_tokens": 26504573.0,
      "step": 14225
    },
    {
      "entropy": 5.826602602005005,
      "epoch": 1.256623101377605,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004847482473864922,
      "loss": 5.5466,
      "mean_token_accuracy": 0.18551961630582808,
      "num_tokens": 26513441.0,
      "step": 14230
    },
    {
      "entropy": 5.929785585403442,
      "epoch": 1.2570646414694453,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004847368483540977,
      "loss": 5.7507,
      "mean_token_accuracy": 0.16561256498098373,
      "num_tokens": 26522911.0,
      "step": 14235
    },
    {
      "entropy": 5.903724288940429,
      "epoch": 1.2575061815612858,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048472544521307765,
      "loss": 5.6193,
      "mean_token_accuracy": 0.18439879417419433,
      "num_tokens": 26531388.0,
      "step": 14240
    },
    {
      "entropy": 5.931670904159546,
      "epoch": 1.257947721653126,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004847140379636554,
      "loss": 5.6192,
      "mean_token_accuracy": 0.17660903930664062,
      "num_tokens": 26540500.0,
      "step": 14245
    },
    {
      "entropy": 5.831917238235474,
      "epoch": 1.2583892617449663,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004847026266060545,
      "loss": 5.6325,
      "mean_token_accuracy": 0.1787765368819237,
      "num_tokens": 26550041.0,
      "step": 14250
    },
    {
      "entropy": 5.86501178741455,
      "epoch": 1.2588308018368068,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004846912111404983,
      "loss": 5.5956,
      "mean_token_accuracy": 0.1768846720457077,
      "num_tokens": 26559594.0,
      "step": 14255
    },
    {
      "entropy": 5.916268157958984,
      "epoch": 1.2592723419286471,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048467979156721055,
      "loss": 5.6679,
      "mean_token_accuracy": 0.18283335566520692,
      "num_tokens": 26569143.0,
      "step": 14260
    },
    {
      "entropy": 5.850358200073242,
      "epoch": 1.2597138820204874,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048466836788641496,
      "loss": 5.6495,
      "mean_token_accuracy": 0.17958984524011612,
      "num_tokens": 26578040.0,
      "step": 14265
    },
    {
      "entropy": 5.867253303527832,
      "epoch": 1.260155422112328,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048465694009833525,
      "loss": 5.6756,
      "mean_token_accuracy": 0.17919367253780366,
      "num_tokens": 26588283.0,
      "step": 14270
    },
    {
      "entropy": 5.93657546043396,
      "epoch": 1.2605969622041682,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048464550820319537,
      "loss": 5.7127,
      "mean_token_accuracy": 0.17615227848291398,
      "num_tokens": 26597815.0,
      "step": 14275
    },
    {
      "entropy": 5.868259572982788,
      "epoch": 1.2610385022960084,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004846340722012192,
      "loss": 5.6892,
      "mean_token_accuracy": 0.17518152594566344,
      "num_tokens": 26606752.0,
      "step": 14280
    },
    {
      "entropy": 5.924601125717163,
      "epoch": 1.261480042387849,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004846226320926308,
      "loss": 5.6885,
      "mean_token_accuracy": 0.17523885071277617,
      "num_tokens": 26615312.0,
      "step": 14285
    },
    {
      "entropy": 5.924990510940551,
      "epoch": 1.2619215824796892,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004846111878776542,
      "loss": 5.6689,
      "mean_token_accuracy": 0.17463979572057725,
      "num_tokens": 26624642.0,
      "step": 14290
    },
    {
      "entropy": 5.875468730926514,
      "epoch": 1.2623631225715295,
      "grad_norm": 1.875,
      "learning_rate": 0.0004845997395565137,
      "loss": 5.5452,
      "mean_token_accuracy": 0.18670294880867006,
      "num_tokens": 26633250.0,
      "step": 14295
    },
    {
      "entropy": 5.84554352760315,
      "epoch": 1.2628046626633698,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004845882871294335,
      "loss": 5.684,
      "mean_token_accuracy": 0.17404927909374238,
      "num_tokens": 26642422.0,
      "step": 14300
    },
    {
      "entropy": 5.946943664550782,
      "epoch": 1.26324620275521,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004845768305966379,
      "loss": 5.7445,
      "mean_token_accuracy": 0.17241754978895188,
      "num_tokens": 26651978.0,
      "step": 14305
    },
    {
      "entropy": 5.9360181331634525,
      "epoch": 1.2636877428470505,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004845653699583515,
      "loss": 5.7302,
      "mean_token_accuracy": 0.16796547025442124,
      "num_tokens": 26662075.0,
      "step": 14310
    },
    {
      "entropy": 5.907423591613769,
      "epoch": 1.2641292829388908,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048455390521479855,
      "loss": 5.6509,
      "mean_token_accuracy": 0.17645078152418137,
      "num_tokens": 26670813.0,
      "step": 14315
    },
    {
      "entropy": 5.978578901290893,
      "epoch": 1.264570823030731,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048454243636620383,
      "loss": 5.6829,
      "mean_token_accuracy": 0.18011376708745958,
      "num_tokens": 26679867.0,
      "step": 14320
    },
    {
      "entropy": 5.884921073913574,
      "epoch": 1.2650123631225716,
      "grad_norm": 1.546875,
      "learning_rate": 0.000484530963412792,
      "loss": 5.6995,
      "mean_token_accuracy": 0.17008066773414612,
      "num_tokens": 26688523.0,
      "step": 14325
    },
    {
      "entropy": 5.898282098770141,
      "epoch": 1.2654539032144119,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048451948635478765,
      "loss": 5.7486,
      "mean_token_accuracy": 0.17235429286956788,
      "num_tokens": 26697922.0,
      "step": 14330
    },
    {
      "entropy": 6.008814716339112,
      "epoch": 1.2658954433062521,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004845080051924158,
      "loss": 5.6982,
      "mean_token_accuracy": 0.1649021327495575,
      "num_tokens": 26707058.0,
      "step": 14335
    },
    {
      "entropy": 5.9452167510986325,
      "epoch": 1.2663369833980926,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004844965199259013,
      "loss": 5.7029,
      "mean_token_accuracy": 0.17543908059597016,
      "num_tokens": 26715873.0,
      "step": 14340
    },
    {
      "entropy": 5.944910335540771,
      "epoch": 1.266778523489933,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004844850305554692,
      "loss": 5.6896,
      "mean_token_accuracy": 0.1786316677927971,
      "num_tokens": 26726253.0,
      "step": 14345
    },
    {
      "entropy": 5.924415397644043,
      "epoch": 1.2672200635817732,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048447353708134426,
      "loss": 5.708,
      "mean_token_accuracy": 0.17837436944246293,
      "num_tokens": 26737077.0,
      "step": 14350
    },
    {
      "entropy": 5.9020439147949215,
      "epoch": 1.2676616036736137,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000484462039503752,
      "loss": 5.7032,
      "mean_token_accuracy": 0.1751269981265068,
      "num_tokens": 26746603.0,
      "step": 14355
    },
    {
      "entropy": 5.8973283767700195,
      "epoch": 1.268103143765454,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048445053782291753,
      "loss": 5.6945,
      "mean_token_accuracy": 0.17376383095979692,
      "num_tokens": 26756174.0,
      "step": 14360
    },
    {
      "entropy": 5.915427017211914,
      "epoch": 1.2685446838572942,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048443903203906614,
      "loss": 5.6882,
      "mean_token_accuracy": 0.17778382450342178,
      "num_tokens": 26765124.0,
      "step": 14365
    },
    {
      "entropy": 5.921507835388184,
      "epoch": 1.2689862239491345,
      "grad_norm": 1.5,
      "learning_rate": 0.0004844275221524232,
      "loss": 5.6485,
      "mean_token_accuracy": 0.18363571017980576,
      "num_tokens": 26774438.0,
      "step": 14370
    },
    {
      "entropy": 5.984895372390747,
      "epoch": 1.2694277640409748,
      "grad_norm": 1.734375,
      "learning_rate": 0.00048441600816321424,
      "loss": 5.7212,
      "mean_token_accuracy": 0.16797182559967042,
      "num_tokens": 26783613.0,
      "step": 14375
    },
    {
      "entropy": 5.869450664520263,
      "epoch": 1.2698693041328153,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004844044900716647,
      "loss": 5.5546,
      "mean_token_accuracy": 0.18659719228744506,
      "num_tokens": 26792136.0,
      "step": 14380
    },
    {
      "entropy": 5.778707838058471,
      "epoch": 1.2703108442246556,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004843929678780004,
      "loss": 5.561,
      "mean_token_accuracy": 0.19016159176826478,
      "num_tokens": 26801481.0,
      "step": 14385
    },
    {
      "entropy": 5.8489889144897464,
      "epoch": 1.2707523843164958,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004843814415824469,
      "loss": 5.6529,
      "mean_token_accuracy": 0.18183598071336746,
      "num_tokens": 26810503.0,
      "step": 14390
    },
    {
      "entropy": 6.001218175888061,
      "epoch": 1.2711939244083363,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00048436991118523003,
      "loss": 5.6766,
      "mean_token_accuracy": 0.17590827941894532,
      "num_tokens": 26820007.0,
      "step": 14395
    },
    {
      "entropy": 5.929369258880615,
      "epoch": 1.2716354645001766,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048435837668657565,
      "loss": 5.6113,
      "mean_token_accuracy": 0.1845945104956627,
      "num_tokens": 26829088.0,
      "step": 14400
    },
    {
      "entropy": 5.848558521270752,
      "epoch": 1.272077004592017,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004843468380867098,
      "loss": 5.7042,
      "mean_token_accuracy": 0.17948804944753646,
      "num_tokens": 26838619.0,
      "step": 14405
    },
    {
      "entropy": 5.92393684387207,
      "epoch": 1.2725185446838574,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004843352953858585,
      "loss": 5.7288,
      "mean_token_accuracy": 0.17729997187852858,
      "num_tokens": 26848728.0,
      "step": 14410
    },
    {
      "entropy": 5.976728296279907,
      "epoch": 1.2729600847756977,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004843237485842477,
      "loss": 5.6388,
      "mean_token_accuracy": 0.18478863537311555,
      "num_tokens": 26859293.0,
      "step": 14415
    },
    {
      "entropy": 5.9146209239959715,
      "epoch": 1.273401624867538,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048431219768210374,
      "loss": 5.6921,
      "mean_token_accuracy": 0.17229849994182586,
      "num_tokens": 26868818.0,
      "step": 14420
    },
    {
      "entropy": 5.932367610931396,
      "epoch": 1.2738431649593784,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004843006426796529,
      "loss": 5.7673,
      "mean_token_accuracy": 0.16835840344429015,
      "num_tokens": 26879245.0,
      "step": 14425
    },
    {
      "entropy": 5.936654806137085,
      "epoch": 1.2742847050512187,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004842890835771215,
      "loss": 5.6611,
      "mean_token_accuracy": 0.17658648937940596,
      "num_tokens": 26888074.0,
      "step": 14430
    },
    {
      "entropy": 5.898121786117554,
      "epoch": 1.274726245143059,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000484277520374736,
      "loss": 5.6941,
      "mean_token_accuracy": 0.17338912934064865,
      "num_tokens": 26897324.0,
      "step": 14435
    },
    {
      "entropy": 5.996611452102661,
      "epoch": 1.2751677852348993,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048426595307272285,
      "loss": 5.7904,
      "mean_token_accuracy": 0.16970804184675217,
      "num_tokens": 26906797.0,
      "step": 14440
    },
    {
      "entropy": 6.012749528884887,
      "epoch": 1.2756093253267395,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048425438167130875,
      "loss": 5.7117,
      "mean_token_accuracy": 0.17792679816484452,
      "num_tokens": 26915737.0,
      "step": 14445
    },
    {
      "entropy": 5.836005449295044,
      "epoch": 1.27605086541858,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048424280617072026,
      "loss": 5.6652,
      "mean_token_accuracy": 0.17732300013303756,
      "num_tokens": 26925364.0,
      "step": 14450
    },
    {
      "entropy": 5.853279638290405,
      "epoch": 1.2764924055104203,
      "grad_norm": 1.5,
      "learning_rate": 0.00048423122657118424,
      "loss": 5.6654,
      "mean_token_accuracy": 0.17952002882957457,
      "num_tokens": 26935188.0,
      "step": 14455
    },
    {
      "entropy": 5.942749261856079,
      "epoch": 1.2769339456022606,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048421964287292745,
      "loss": 5.6405,
      "mean_token_accuracy": 0.17386209815740586,
      "num_tokens": 26944606.0,
      "step": 14460
    },
    {
      "entropy": 6.005256605148316,
      "epoch": 1.277375485694101,
      "grad_norm": 1.5,
      "learning_rate": 0.0004842080550761769,
      "loss": 5.7723,
      "mean_token_accuracy": 0.16954320073127746,
      "num_tokens": 26955147.0,
      "step": 14465
    },
    {
      "entropy": 5.905995273590088,
      "epoch": 1.2778170257859414,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004841964631811595,
      "loss": 5.7109,
      "mean_token_accuracy": 0.17932043820619584,
      "num_tokens": 26964243.0,
      "step": 14470
    },
    {
      "entropy": 5.927519035339356,
      "epoch": 1.2782585658777816,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048418486718810244,
      "loss": 5.6768,
      "mean_token_accuracy": 0.1723976492881775,
      "num_tokens": 26973284.0,
      "step": 14475
    },
    {
      "entropy": 5.941860389709473,
      "epoch": 1.2787001059696221,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004841732670972327,
      "loss": 5.7811,
      "mean_token_accuracy": 0.17325678765773772,
      "num_tokens": 26982757.0,
      "step": 14480
    },
    {
      "entropy": 5.964514446258545,
      "epoch": 1.2791416460614624,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004841616629087776,
      "loss": 5.7247,
      "mean_token_accuracy": 0.17363827675580978,
      "num_tokens": 26991784.0,
      "step": 14485
    },
    {
      "entropy": 5.9475287914276125,
      "epoch": 1.2795831861533027,
      "grad_norm": 1.8125,
      "learning_rate": 0.00048415005462296455,
      "loss": 5.7202,
      "mean_token_accuracy": 0.1646709680557251,
      "num_tokens": 27000700.0,
      "step": 14490
    },
    {
      "entropy": 5.946364402770996,
      "epoch": 1.2800247262451432,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004841384422400209,
      "loss": 5.6739,
      "mean_token_accuracy": 0.1856359899044037,
      "num_tokens": 27010089.0,
      "step": 14495
    },
    {
      "entropy": 5.869690561294556,
      "epoch": 1.2804662663369835,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004841268257601741,
      "loss": 5.6807,
      "mean_token_accuracy": 0.17967394292354583,
      "num_tokens": 27018961.0,
      "step": 14500
    },
    {
      "entropy": 5.925314664840698,
      "epoch": 1.2809078064288237,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004841152051836517,
      "loss": 5.7423,
      "mean_token_accuracy": 0.17542578279972076,
      "num_tokens": 27027905.0,
      "step": 14505
    },
    {
      "entropy": 5.865329694747925,
      "epoch": 1.281349346520664,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048410358051068135,
      "loss": 5.6562,
      "mean_token_accuracy": 0.17330632954835892,
      "num_tokens": 27036969.0,
      "step": 14510
    },
    {
      "entropy": 5.810550737380981,
      "epoch": 1.2817908866125043,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004840919517414908,
      "loss": 5.5997,
      "mean_token_accuracy": 0.1836143761873245,
      "num_tokens": 27046654.0,
      "step": 14515
    },
    {
      "entropy": 6.010165119171143,
      "epoch": 1.2822324267043448,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048408031887630775,
      "loss": 5.7212,
      "mean_token_accuracy": 0.1710189923644066,
      "num_tokens": 27058492.0,
      "step": 14520
    },
    {
      "entropy": 5.967662811279297,
      "epoch": 1.282673966796185,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004840686819153604,
      "loss": 5.7301,
      "mean_token_accuracy": 0.1760622352361679,
      "num_tokens": 27068863.0,
      "step": 14525
    },
    {
      "entropy": 5.921542119979859,
      "epoch": 1.2831155068880253,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004840570408588762,
      "loss": 5.6744,
      "mean_token_accuracy": 0.1786969155073166,
      "num_tokens": 27078793.0,
      "step": 14530
    },
    {
      "entropy": 5.833789920806884,
      "epoch": 1.2835570469798658,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004840453957070836,
      "loss": 5.4971,
      "mean_token_accuracy": 0.18880011141300201,
      "num_tokens": 27087093.0,
      "step": 14535
    },
    {
      "entropy": 5.837006044387818,
      "epoch": 1.283998587071706,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048403374646021057,
      "loss": 5.602,
      "mean_token_accuracy": 0.17864794731140138,
      "num_tokens": 27096601.0,
      "step": 14540
    },
    {
      "entropy": 5.903562116622925,
      "epoch": 1.2844401271635464,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048402209311848533,
      "loss": 5.6637,
      "mean_token_accuracy": 0.17818803638219832,
      "num_tokens": 27105041.0,
      "step": 14545
    },
    {
      "entropy": 5.958106422424317,
      "epoch": 1.2848816672553869,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048401043568213614,
      "loss": 5.7263,
      "mean_token_accuracy": 0.1716931104660034,
      "num_tokens": 27114252.0,
      "step": 14550
    },
    {
      "entropy": 5.942597484588623,
      "epoch": 1.2853232073472272,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004839987741513914,
      "loss": 5.6434,
      "mean_token_accuracy": 0.1784108981490135,
      "num_tokens": 27122480.0,
      "step": 14555
    },
    {
      "entropy": 5.912863540649414,
      "epoch": 1.2857647474390674,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048398710852647954,
      "loss": 5.7138,
      "mean_token_accuracy": 0.16936530619859697,
      "num_tokens": 27132322.0,
      "step": 14560
    },
    {
      "entropy": 5.83943886756897,
      "epoch": 1.286206287530908,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048397543880762905,
      "loss": 5.5576,
      "mean_token_accuracy": 0.191128146648407,
      "num_tokens": 27140782.0,
      "step": 14565
    },
    {
      "entropy": 5.875221109390258,
      "epoch": 1.2866478276227482,
      "grad_norm": 1.6875,
      "learning_rate": 0.00048396376499506866,
      "loss": 5.6028,
      "mean_token_accuracy": 0.1811915785074234,
      "num_tokens": 27150058.0,
      "step": 14570
    },
    {
      "entropy": 5.847175359725952,
      "epoch": 1.2870893677145885,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048395208708902686,
      "loss": 5.6431,
      "mean_token_accuracy": 0.17920541018247604,
      "num_tokens": 27159574.0,
      "step": 14575
    },
    {
      "entropy": 5.8994134902954105,
      "epoch": 1.2875309078064288,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048394040508973257,
      "loss": 5.6548,
      "mean_token_accuracy": 0.17913998514413834,
      "num_tokens": 27168025.0,
      "step": 14580
    },
    {
      "entropy": 5.85661039352417,
      "epoch": 1.287972447898269,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004839287189974145,
      "loss": 5.6529,
      "mean_token_accuracy": 0.17940210700035095,
      "num_tokens": 27177108.0,
      "step": 14585
    },
    {
      "entropy": 5.95959358215332,
      "epoch": 1.2884139879901095,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048391702881230177,
      "loss": 5.7287,
      "mean_token_accuracy": 0.17021486312150955,
      "num_tokens": 27187486.0,
      "step": 14590
    },
    {
      "entropy": 5.988849115371704,
      "epoch": 1.2888555280819498,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004839053345346232,
      "loss": 5.6998,
      "mean_token_accuracy": 0.17103867530822753,
      "num_tokens": 27195962.0,
      "step": 14595
    },
    {
      "entropy": 5.891149377822876,
      "epoch": 1.28929706817379,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004838936361646079,
      "loss": 5.6933,
      "mean_token_accuracy": 0.18065960109233856,
      "num_tokens": 27206354.0,
      "step": 14600
    },
    {
      "entropy": 5.85495400428772,
      "epoch": 1.2897386082656306,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004838819337024852,
      "loss": 5.5771,
      "mean_token_accuracy": 0.18725096434354782,
      "num_tokens": 27215326.0,
      "step": 14605
    },
    {
      "entropy": 5.930504035949707,
      "epoch": 1.2901801483574709,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048387022714848406,
      "loss": 5.6664,
      "mean_token_accuracy": 0.1745573714375496,
      "num_tokens": 27224253.0,
      "step": 14610
    },
    {
      "entropy": 5.942274522781372,
      "epoch": 1.2906216884493111,
      "grad_norm": 1.484375,
      "learning_rate": 0.000483858516502834,
      "loss": 5.6702,
      "mean_token_accuracy": 0.17803490161895752,
      "num_tokens": 27233007.0,
      "step": 14615
    },
    {
      "entropy": 5.887301015853882,
      "epoch": 1.2910632285411516,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004838468017657644,
      "loss": 5.5926,
      "mean_token_accuracy": 0.17960865944623947,
      "num_tokens": 27242127.0,
      "step": 14620
    },
    {
      "entropy": 5.873813962936401,
      "epoch": 1.291504768632992,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00048383508293750474,
      "loss": 5.7395,
      "mean_token_accuracy": 0.17251099795103073,
      "num_tokens": 27252753.0,
      "step": 14625
    },
    {
      "entropy": 5.995630502700806,
      "epoch": 1.2919463087248322,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004838233600182846,
      "loss": 5.7351,
      "mean_token_accuracy": 0.17319215089082718,
      "num_tokens": 27262498.0,
      "step": 14630
    },
    {
      "entropy": 6.061165857315063,
      "epoch": 1.2923878488166727,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004838116330083336,
      "loss": 5.7474,
      "mean_token_accuracy": 0.16970619112253188,
      "num_tokens": 27271524.0,
      "step": 14635
    },
    {
      "entropy": 5.878019857406616,
      "epoch": 1.292829388908513,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004837999019078814,
      "loss": 5.5861,
      "mean_token_accuracy": 0.19177144765853882,
      "num_tokens": 27280649.0,
      "step": 14640
    },
    {
      "entropy": 5.900376081466675,
      "epoch": 1.2932709290003532,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004837881667171579,
      "loss": 5.6915,
      "mean_token_accuracy": 0.17113495916128157,
      "num_tokens": 27289593.0,
      "step": 14645
    },
    {
      "entropy": 5.896522474288941,
      "epoch": 1.2937124690921935,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00048377642743639297,
      "loss": 5.6771,
      "mean_token_accuracy": 0.17260036915540694,
      "num_tokens": 27298070.0,
      "step": 14650
    },
    {
      "entropy": 5.849561071395874,
      "epoch": 1.2941540091840338,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004837646840658166,
      "loss": 5.6185,
      "mean_token_accuracy": 0.1822208732366562,
      "num_tokens": 27306738.0,
      "step": 14655
    },
    {
      "entropy": 5.9113161087036135,
      "epoch": 1.2945955492758743,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048375293660565875,
      "loss": 5.6052,
      "mean_token_accuracy": 0.18277504444122314,
      "num_tokens": 27315243.0,
      "step": 14660
    },
    {
      "entropy": 5.879252004623413,
      "epoch": 1.2950370893677146,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004837411850561496,
      "loss": 5.6172,
      "mean_token_accuracy": 0.18300796896219254,
      "num_tokens": 27325327.0,
      "step": 14665
    },
    {
      "entropy": 5.897431945800781,
      "epoch": 1.2954786294595548,
      "grad_norm": 1.625,
      "learning_rate": 0.00048372942941751935,
      "loss": 5.6292,
      "mean_token_accuracy": 0.18231210112571716,
      "num_tokens": 27333436.0,
      "step": 14670
    },
    {
      "entropy": 5.89586501121521,
      "epoch": 1.2959201695513953,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004837176696899984,
      "loss": 5.7112,
      "mean_token_accuracy": 0.16630406826734542,
      "num_tokens": 27342689.0,
      "step": 14675
    },
    {
      "entropy": 5.939995145797729,
      "epoch": 1.2963617096432356,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048370590587381684,
      "loss": 5.724,
      "mean_token_accuracy": 0.17956029921770095,
      "num_tokens": 27352347.0,
      "step": 14680
    },
    {
      "entropy": 5.9825788021087645,
      "epoch": 1.2968032497350759,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048369413796920535,
      "loss": 5.7128,
      "mean_token_accuracy": 0.1747731849551201,
      "num_tokens": 27362153.0,
      "step": 14685
    },
    {
      "entropy": 5.884520435333252,
      "epoch": 1.2972447898269164,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004836823659763944,
      "loss": 5.6513,
      "mean_token_accuracy": 0.173822221159935,
      "num_tokens": 27371126.0,
      "step": 14690
    },
    {
      "entropy": 5.8241064071655275,
      "epoch": 1.2976863299187567,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048367058989561457,
      "loss": 5.5898,
      "mean_token_accuracy": 0.18091725260019303,
      "num_tokens": 27380727.0,
      "step": 14695
    },
    {
      "entropy": 5.898044490814209,
      "epoch": 1.298127870010597,
      "grad_norm": 1.5,
      "learning_rate": 0.0004836588097270966,
      "loss": 5.6498,
      "mean_token_accuracy": 0.180899915099144,
      "num_tokens": 27389970.0,
      "step": 14700
    },
    {
      "entropy": 5.922196912765503,
      "epoch": 1.2985694101024374,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048364702547107113,
      "loss": 5.6053,
      "mean_token_accuracy": 0.1764370620250702,
      "num_tokens": 27399177.0,
      "step": 14705
    },
    {
      "entropy": 5.915177297592163,
      "epoch": 1.2990109501942777,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004836352371277692,
      "loss": 5.6876,
      "mean_token_accuracy": 0.17970783561468123,
      "num_tokens": 27408776.0,
      "step": 14710
    },
    {
      "entropy": 5.9213825225830075,
      "epoch": 1.299452490286118,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004836234446974216,
      "loss": 5.7071,
      "mean_token_accuracy": 0.1746725931763649,
      "num_tokens": 27418708.0,
      "step": 14715
    },
    {
      "entropy": 5.918446779251099,
      "epoch": 1.2998940303779583,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004836116481802593,
      "loss": 5.6334,
      "mean_token_accuracy": 0.17416589409112931,
      "num_tokens": 27428027.0,
      "step": 14720
    },
    {
      "entropy": 5.8794636726379395,
      "epoch": 1.3003355704697985,
      "grad_norm": 1.734375,
      "learning_rate": 0.00048359984757651356,
      "loss": 5.6514,
      "mean_token_accuracy": 0.17965403497219085,
      "num_tokens": 27438214.0,
      "step": 14725
    },
    {
      "entropy": 5.911548280715943,
      "epoch": 1.300777110561639,
      "grad_norm": 1.875,
      "learning_rate": 0.00048358804288641536,
      "loss": 5.5798,
      "mean_token_accuracy": 0.18196048438549042,
      "num_tokens": 27447413.0,
      "step": 14730
    },
    {
      "entropy": 5.8685160160064695,
      "epoch": 1.3012186506534793,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048357623411019607,
      "loss": 5.6004,
      "mean_token_accuracy": 0.18419422656297685,
      "num_tokens": 27456316.0,
      "step": 14735
    },
    {
      "entropy": 5.863255834579467,
      "epoch": 1.3016601907453196,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000483564421248087,
      "loss": 5.6618,
      "mean_token_accuracy": 0.18219218850135804,
      "num_tokens": 27465346.0,
      "step": 14740
    },
    {
      "entropy": 5.8458709716796875,
      "epoch": 1.30210173083716,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048355260430031947,
      "loss": 5.5194,
      "mean_token_accuracy": 0.18867741972208024,
      "num_tokens": 27473367.0,
      "step": 14745
    },
    {
      "entropy": 5.970284748077392,
      "epoch": 1.3025432709290004,
      "grad_norm": 1.515625,
      "learning_rate": 0.000483540783267125,
      "loss": 5.6924,
      "mean_token_accuracy": 0.17441833317279815,
      "num_tokens": 27482187.0,
      "step": 14750
    },
    {
      "entropy": 5.8374794006347654,
      "epoch": 1.3029848110208406,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048352895814873526,
      "loss": 5.6342,
      "mean_token_accuracy": 0.17687694281339644,
      "num_tokens": 27492406.0,
      "step": 14755
    },
    {
      "entropy": 5.900814008712769,
      "epoch": 1.3034263511126811,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004835171289453818,
      "loss": 5.7054,
      "mean_token_accuracy": 0.18307556658983232,
      "num_tokens": 27501929.0,
      "step": 14760
    },
    {
      "entropy": 5.869705867767334,
      "epoch": 1.3038678912045214,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004835052956572964,
      "loss": 5.5564,
      "mean_token_accuracy": 0.1927289515733719,
      "num_tokens": 27510657.0,
      "step": 14765
    },
    {
      "entropy": 5.837793159484863,
      "epoch": 1.3043094312963617,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048349345828471084,
      "loss": 5.6985,
      "mean_token_accuracy": 0.17013710737228394,
      "num_tokens": 27519726.0,
      "step": 14770
    },
    {
      "entropy": 5.871703052520752,
      "epoch": 1.3047509713882022,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048348161682785704,
      "loss": 5.6335,
      "mean_token_accuracy": 0.18101432174444199,
      "num_tokens": 27528876.0,
      "step": 14775
    },
    {
      "entropy": 5.988534832000733,
      "epoch": 1.3051925114800424,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004834697712869669,
      "loss": 5.6632,
      "mean_token_accuracy": 0.1759288415312767,
      "num_tokens": 27537782.0,
      "step": 14780
    },
    {
      "entropy": 5.920468473434449,
      "epoch": 1.3056340515718827,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004834579216622725,
      "loss": 5.6984,
      "mean_token_accuracy": 0.17012072652578353,
      "num_tokens": 27547322.0,
      "step": 14785
    },
    {
      "entropy": 5.8022815704345705,
      "epoch": 1.306075591663723,
      "grad_norm": 1.9609375,
      "learning_rate": 0.000483446067954006,
      "loss": 5.6468,
      "mean_token_accuracy": 0.17586903423070907,
      "num_tokens": 27556070.0,
      "step": 14790
    },
    {
      "entropy": 5.8747233867645265,
      "epoch": 1.3065171317555633,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048343421016239945,
      "loss": 5.5601,
      "mean_token_accuracy": 0.18754286915063859,
      "num_tokens": 27565951.0,
      "step": 14795
    },
    {
      "entropy": 5.941370820999145,
      "epoch": 1.3069586718474038,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004834223482876854,
      "loss": 5.621,
      "mean_token_accuracy": 0.17698982059955598,
      "num_tokens": 27575102.0,
      "step": 14800
    },
    {
      "entropy": 5.933369445800781,
      "epoch": 1.307400211939244,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004834104823300961,
      "loss": 5.7115,
      "mean_token_accuracy": 0.17401691973209382,
      "num_tokens": 27584854.0,
      "step": 14805
    },
    {
      "entropy": 5.850229215621948,
      "epoch": 1.3078417520310843,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048339861228986395,
      "loss": 5.6843,
      "mean_token_accuracy": 0.18363939225673676,
      "num_tokens": 27593403.0,
      "step": 14810
    },
    {
      "entropy": 5.93776650428772,
      "epoch": 1.3082832921229248,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048338673816722145,
      "loss": 5.7291,
      "mean_token_accuracy": 0.17736464142799377,
      "num_tokens": 27602162.0,
      "step": 14815
    },
    {
      "entropy": 5.944648504257202,
      "epoch": 1.308724832214765,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048337485996240127,
      "loss": 5.7355,
      "mean_token_accuracy": 0.16883742511272432,
      "num_tokens": 27611433.0,
      "step": 14820
    },
    {
      "entropy": 5.898706722259521,
      "epoch": 1.3091663723066054,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004833629776756362,
      "loss": 5.564,
      "mean_token_accuracy": 0.18379827439785004,
      "num_tokens": 27620443.0,
      "step": 14825
    },
    {
      "entropy": 5.972296571731567,
      "epoch": 1.3096079123984459,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048335109130715875,
      "loss": 5.6512,
      "mean_token_accuracy": 0.17572671175003052,
      "num_tokens": 27629507.0,
      "step": 14830
    },
    {
      "entropy": 5.896455526351929,
      "epoch": 1.3100494524902861,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004833392008572021,
      "loss": 5.6523,
      "mean_token_accuracy": 0.1792852133512497,
      "num_tokens": 27638355.0,
      "step": 14835
    },
    {
      "entropy": 5.910127115249634,
      "epoch": 1.3104909925821264,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048332730632599883,
      "loss": 5.7077,
      "mean_token_accuracy": 0.16892997920513153,
      "num_tokens": 27649074.0,
      "step": 14840
    },
    {
      "entropy": 5.925766134262085,
      "epoch": 1.310932532673967,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004833154077137821,
      "loss": 5.6943,
      "mean_token_accuracy": 0.17281589061021804,
      "num_tokens": 27658564.0,
      "step": 14845
    },
    {
      "entropy": 5.905432176589966,
      "epoch": 1.3113740727658072,
      "grad_norm": 1.375,
      "learning_rate": 0.00048330350502078515,
      "loss": 5.6318,
      "mean_token_accuracy": 0.1766461580991745,
      "num_tokens": 27667963.0,
      "step": 14850
    },
    {
      "entropy": 5.9142780780792235,
      "epoch": 1.3118156128576475,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048329159824724087,
      "loss": 5.7191,
      "mean_token_accuracy": 0.1759385034441948,
      "num_tokens": 27677528.0,
      "step": 14855
    },
    {
      "entropy": 5.96515154838562,
      "epoch": 1.3122571529494877,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048327968739338275,
      "loss": 5.6383,
      "mean_token_accuracy": 0.18272966891527176,
      "num_tokens": 27687659.0,
      "step": 14860
    },
    {
      "entropy": 5.88814787864685,
      "epoch": 1.312698693041328,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048326777245944396,
      "loss": 5.5837,
      "mean_token_accuracy": 0.18439485132694244,
      "num_tokens": 27697222.0,
      "step": 14865
    },
    {
      "entropy": 5.909781646728516,
      "epoch": 1.3131402331331685,
      "grad_norm": 1.75,
      "learning_rate": 0.000483255853445658,
      "loss": 5.6681,
      "mean_token_accuracy": 0.17490864396095276,
      "num_tokens": 27705664.0,
      "step": 14870
    },
    {
      "entropy": 5.8378948211669925,
      "epoch": 1.3135817732250088,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048324393035225824,
      "loss": 5.5951,
      "mean_token_accuracy": 0.18595435172319413,
      "num_tokens": 27714879.0,
      "step": 14875
    },
    {
      "entropy": 5.906320381164551,
      "epoch": 1.314023313316849,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004832320031794783,
      "loss": 5.7364,
      "mean_token_accuracy": 0.16977429986000062,
      "num_tokens": 27724749.0,
      "step": 14880
    },
    {
      "entropy": 5.944372844696045,
      "epoch": 1.3144648534086896,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004832200719275519,
      "loss": 5.6809,
      "mean_token_accuracy": 0.1787618339061737,
      "num_tokens": 27733109.0,
      "step": 14885
    },
    {
      "entropy": 5.7291277885437015,
      "epoch": 1.3149063935005298,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004832081365967128,
      "loss": 5.4683,
      "mean_token_accuracy": 0.18973643481731414,
      "num_tokens": 27742079.0,
      "step": 14890
    },
    {
      "entropy": 5.880386781692505,
      "epoch": 1.3153479335923701,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048319619718719455,
      "loss": 5.7486,
      "mean_token_accuracy": 0.16478121429681777,
      "num_tokens": 27750900.0,
      "step": 14895
    },
    {
      "entropy": 5.940733194351196,
      "epoch": 1.3157894736842106,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048318425369923133,
      "loss": 5.6076,
      "mean_token_accuracy": 0.18223343938589096,
      "num_tokens": 27760392.0,
      "step": 14900
    },
    {
      "entropy": 5.875930738449097,
      "epoch": 1.316231013776051,
      "grad_norm": 1.25,
      "learning_rate": 0.0004831723061330569,
      "loss": 5.6395,
      "mean_token_accuracy": 0.1776714578270912,
      "num_tokens": 27768805.0,
      "step": 14905
    },
    {
      "entropy": 5.956908273696899,
      "epoch": 1.3166725538678912,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048316035448890546,
      "loss": 5.7461,
      "mean_token_accuracy": 0.1674629658460617,
      "num_tokens": 27777165.0,
      "step": 14910
    },
    {
      "entropy": 5.986285018920898,
      "epoch": 1.3171140939597317,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000483148398767011,
      "loss": 5.69,
      "mean_token_accuracy": 0.17757479399442672,
      "num_tokens": 27786709.0,
      "step": 14915
    },
    {
      "entropy": 5.867564058303833,
      "epoch": 1.317555634051572,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048313643896760776,
      "loss": 5.6147,
      "mean_token_accuracy": 0.18132674992084502,
      "num_tokens": 27795977.0,
      "step": 14920
    },
    {
      "entropy": 5.814897775650024,
      "epoch": 1.3179971741434122,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048312447509093017,
      "loss": 5.6441,
      "mean_token_accuracy": 0.1736961543560028,
      "num_tokens": 27804838.0,
      "step": 14925
    },
    {
      "entropy": 5.942633581161499,
      "epoch": 1.3184387142352525,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004831125071372124,
      "loss": 5.6195,
      "mean_token_accuracy": 0.18223338574171066,
      "num_tokens": 27813245.0,
      "step": 14930
    },
    {
      "entropy": 5.995274686813355,
      "epoch": 1.3188802543270928,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004831005351066889,
      "loss": 5.7593,
      "mean_token_accuracy": 0.1679219126701355,
      "num_tokens": 27822509.0,
      "step": 14935
    },
    {
      "entropy": 5.807715892791748,
      "epoch": 1.3193217944189333,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004830885589995944,
      "loss": 5.5407,
      "mean_token_accuracy": 0.1878236785531044,
      "num_tokens": 27831140.0,
      "step": 14940
    },
    {
      "entropy": 5.929039669036865,
      "epoch": 1.3197633345107735,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048307657881616333,
      "loss": 5.6699,
      "mean_token_accuracy": 0.17543411999940872,
      "num_tokens": 27839537.0,
      "step": 14945
    },
    {
      "entropy": 5.926523733139038,
      "epoch": 1.3202048746026138,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00048306459455663045,
      "loss": 5.6301,
      "mean_token_accuracy": 0.1794668212532997,
      "num_tokens": 27848888.0,
      "step": 14950
    },
    {
      "entropy": 5.941968393325806,
      "epoch": 1.3206464146944543,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004830526062212305,
      "loss": 5.6929,
      "mean_token_accuracy": 0.1747799724340439,
      "num_tokens": 27858910.0,
      "step": 14955
    },
    {
      "entropy": 5.844124174118042,
      "epoch": 1.3210879547862946,
      "grad_norm": 1.890625,
      "learning_rate": 0.00048304061381019826,
      "loss": 5.6062,
      "mean_token_accuracy": 0.18478866368532182,
      "num_tokens": 27867967.0,
      "step": 14960
    },
    {
      "entropy": 5.836395120620727,
      "epoch": 1.3215294948781349,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048302861732376875,
      "loss": 5.6302,
      "mean_token_accuracy": 0.18093497753143312,
      "num_tokens": 27877317.0,
      "step": 14965
    },
    {
      "entropy": 5.919211196899414,
      "epoch": 1.3219710349699754,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048301661676217693,
      "loss": 5.6656,
      "mean_token_accuracy": 0.17825794965028763,
      "num_tokens": 27886400.0,
      "step": 14970
    },
    {
      "entropy": 5.921872949600219,
      "epoch": 1.3224125750618156,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004830046121256579,
      "loss": 5.6608,
      "mean_token_accuracy": 0.17208372354507445,
      "num_tokens": 27895816.0,
      "step": 14975
    },
    {
      "entropy": 5.931504535675049,
      "epoch": 1.322854115153656,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004829926034144468,
      "loss": 5.7058,
      "mean_token_accuracy": 0.17726118713617325,
      "num_tokens": 27905614.0,
      "step": 14980
    },
    {
      "entropy": 5.922978639602661,
      "epoch": 1.3232956552454964,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004829805906287789,
      "loss": 5.6844,
      "mean_token_accuracy": 0.1741032585501671,
      "num_tokens": 27914714.0,
      "step": 14985
    },
    {
      "entropy": 5.900981521606445,
      "epoch": 1.3237371953373367,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004829685737688895,
      "loss": 5.6399,
      "mean_token_accuracy": 0.18235090523958206,
      "num_tokens": 27924821.0,
      "step": 14990
    },
    {
      "entropy": 5.931923484802246,
      "epoch": 1.324178735429177,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048295655283501403,
      "loss": 5.6916,
      "mean_token_accuracy": 0.1738789990544319,
      "num_tokens": 27934028.0,
      "step": 14995
    },
    {
      "entropy": 5.853687810897827,
      "epoch": 1.3246202755210172,
      "grad_norm": 2.046875,
      "learning_rate": 0.000482944527827388,
      "loss": 5.6351,
      "mean_token_accuracy": 0.17382654100656508,
      "num_tokens": 27943550.0,
      "step": 15000
    },
    {
      "epoch": 1.3246202755210172,
      "eval_entropy": 5.741849998871575,
      "eval_loss": 5.744239330291748,
      "eval_mean_token_accuracy": 0.1815548952197443,
      "eval_num_tokens": 27943550.0,
      "eval_runtime": 26.3036,
      "eval_samples_per_second": 1342.592,
      "eval_steps_per_second": 167.848,
      "step": 15000
    },
    {
      "entropy": 5.934597110748291,
      "epoch": 1.3250618156128575,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004829324987462469,
      "loss": 5.6347,
      "mean_token_accuracy": 0.17356710284948348,
      "num_tokens": 27952752.0,
      "step": 15005
    },
    {
      "entropy": 5.969898700714111,
      "epoch": 1.325503355704698,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004829204655918264,
      "loss": 5.7102,
      "mean_token_accuracy": 0.17254795879125595,
      "num_tokens": 27961535.0,
      "step": 15010
    },
    {
      "entropy": 5.92507495880127,
      "epoch": 1.3259448957965383,
      "grad_norm": 1.734375,
      "learning_rate": 0.00048290842836436226,
      "loss": 5.6259,
      "mean_token_accuracy": 0.18013035655021667,
      "num_tokens": 27969644.0,
      "step": 15015
    },
    {
      "entropy": 5.9352500438690186,
      "epoch": 1.3263864358883786,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004828963870640902,
      "loss": 5.7668,
      "mean_token_accuracy": 0.16760257482528687,
      "num_tokens": 27978859.0,
      "step": 15020
    },
    {
      "entropy": 5.959953498840332,
      "epoch": 1.326827975980219,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048288434169124625,
      "loss": 5.6873,
      "mean_token_accuracy": 0.1775441735982895,
      "num_tokens": 27988693.0,
      "step": 15025
    },
    {
      "entropy": 5.952336025238037,
      "epoch": 1.3272695160720593,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004828722922460662,
      "loss": 5.6227,
      "mean_token_accuracy": 0.18492192327976226,
      "num_tokens": 27998196.0,
      "step": 15030
    },
    {
      "entropy": 5.932687044143677,
      "epoch": 1.3277110561638996,
      "grad_norm": 1.84375,
      "learning_rate": 0.00048286023872878614,
      "loss": 5.769,
      "mean_token_accuracy": 0.16737345308065416,
      "num_tokens": 28008743.0,
      "step": 15035
    },
    {
      "entropy": 5.95484619140625,
      "epoch": 1.3281525962557401,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004828481811396423,
      "loss": 5.7425,
      "mean_token_accuracy": 0.16793157458305358,
      "num_tokens": 28017684.0,
      "step": 15040
    },
    {
      "entropy": 5.858897304534912,
      "epoch": 1.3285941363475804,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004828361194788708,
      "loss": 5.7151,
      "mean_token_accuracy": 0.17845116704702377,
      "num_tokens": 28027992.0,
      "step": 15045
    },
    {
      "entropy": 6.038978958129883,
      "epoch": 1.3290356764394207,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048282405374670785,
      "loss": 5.771,
      "mean_token_accuracy": 0.17045434862375258,
      "num_tokens": 28037457.0,
      "step": 15050
    },
    {
      "entropy": 5.995742511749268,
      "epoch": 1.3294772165312612,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048281198394338986,
      "loss": 5.7401,
      "mean_token_accuracy": 0.17402165979146958,
      "num_tokens": 28046881.0,
      "step": 15055
    },
    {
      "entropy": 5.918190670013428,
      "epoch": 1.3299187566231014,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048279991006915335,
      "loss": 5.5908,
      "mean_token_accuracy": 0.17629314213991165,
      "num_tokens": 28056932.0,
      "step": 15060
    },
    {
      "entropy": 5.9013776779174805,
      "epoch": 1.3303602967149417,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048278783212423484,
      "loss": 5.7646,
      "mean_token_accuracy": 0.17475581169128418,
      "num_tokens": 28066843.0,
      "step": 15065
    },
    {
      "entropy": 5.901216220855713,
      "epoch": 1.330801836806782,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004827757501088708,
      "loss": 5.6534,
      "mean_token_accuracy": 0.1732030063867569,
      "num_tokens": 28076329.0,
      "step": 15070
    },
    {
      "entropy": 5.924569702148437,
      "epoch": 1.3312433768986223,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048276366402329803,
      "loss": 5.73,
      "mean_token_accuracy": 0.17494574785232545,
      "num_tokens": 28086402.0,
      "step": 15075
    },
    {
      "entropy": 5.948095703125,
      "epoch": 1.3316849169904628,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004827515738677532,
      "loss": 5.659,
      "mean_token_accuracy": 0.1753866419196129,
      "num_tokens": 28095030.0,
      "step": 15080
    },
    {
      "entropy": 5.831193828582764,
      "epoch": 1.332126457082303,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004827394796424732,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17747818678617477,
      "num_tokens": 28105138.0,
      "step": 15085
    },
    {
      "entropy": 5.9256102561950685,
      "epoch": 1.3325679971741433,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048272738134769503,
      "loss": 5.682,
      "mean_token_accuracy": 0.1733996331691742,
      "num_tokens": 28114359.0,
      "step": 15090
    },
    {
      "entropy": 5.942698955535889,
      "epoch": 1.3330095372659838,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048271527898365545,
      "loss": 5.6379,
      "mean_token_accuracy": 0.18408717960119247,
      "num_tokens": 28124045.0,
      "step": 15095
    },
    {
      "entropy": 5.860558319091797,
      "epoch": 1.333451077357824,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004827031725505917,
      "loss": 5.6761,
      "mean_token_accuracy": 0.18470071256160736,
      "num_tokens": 28132990.0,
      "step": 15100
    },
    {
      "entropy": 5.888694763183594,
      "epoch": 1.3338926174496644,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004826910620487411,
      "loss": 5.6504,
      "mean_token_accuracy": 0.1784898519515991,
      "num_tokens": 28142504.0,
      "step": 15105
    },
    {
      "entropy": 5.906008577346801,
      "epoch": 1.3343341575415049,
      "grad_norm": 2.125,
      "learning_rate": 0.0004826789474783405,
      "loss": 5.7365,
      "mean_token_accuracy": 0.17005418688058854,
      "num_tokens": 28151227.0,
      "step": 15110
    },
    {
      "entropy": 5.962719058990478,
      "epoch": 1.3347756976333451,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004826668288396276,
      "loss": 5.6566,
      "mean_token_accuracy": 0.1813158392906189,
      "num_tokens": 28160244.0,
      "step": 15115
    },
    {
      "entropy": 5.849847412109375,
      "epoch": 1.3352172377251854,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048265470613283954,
      "loss": 5.6617,
      "mean_token_accuracy": 0.17557185441255568,
      "num_tokens": 28169295.0,
      "step": 15120
    },
    {
      "entropy": 5.981151390075683,
      "epoch": 1.335658777817026,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048264257935821394,
      "loss": 5.7027,
      "mean_token_accuracy": 0.1736290767788887,
      "num_tokens": 28179199.0,
      "step": 15125
    },
    {
      "entropy": 5.949420022964477,
      "epoch": 1.3361003179088662,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004826304485159882,
      "loss": 5.7221,
      "mean_token_accuracy": 0.18347188383340834,
      "num_tokens": 28188315.0,
      "step": 15130
    },
    {
      "entropy": 5.94394998550415,
      "epoch": 1.3365418580007065,
      "grad_norm": 1.625,
      "learning_rate": 0.00048261831360640016,
      "loss": 5.7183,
      "mean_token_accuracy": 0.1760927066206932,
      "num_tokens": 28197776.0,
      "step": 15135
    },
    {
      "entropy": 5.9113178730010985,
      "epoch": 1.3369833980925467,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048260617462968735,
      "loss": 5.704,
      "mean_token_accuracy": 0.1794688045978546,
      "num_tokens": 28207985.0,
      "step": 15140
    },
    {
      "entropy": 5.904776573181152,
      "epoch": 1.337424938184387,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048259403158608773,
      "loss": 5.5243,
      "mean_token_accuracy": 0.18667106330394745,
      "num_tokens": 28216321.0,
      "step": 15145
    },
    {
      "entropy": 5.92898998260498,
      "epoch": 1.3378664782762275,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004825818844758391,
      "loss": 5.7397,
      "mean_token_accuracy": 0.17152916938066481,
      "num_tokens": 28226763.0,
      "step": 15150
    },
    {
      "entropy": 5.864075565338135,
      "epoch": 1.3383080183680678,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048256973329917943,
      "loss": 5.6553,
      "mean_token_accuracy": 0.17790609747171401,
      "num_tokens": 28236123.0,
      "step": 15155
    },
    {
      "entropy": 5.98751745223999,
      "epoch": 1.338749558459908,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048255757805634665,
      "loss": 5.7488,
      "mean_token_accuracy": 0.17575381994247435,
      "num_tokens": 28245848.0,
      "step": 15160
    },
    {
      "entropy": 5.84109206199646,
      "epoch": 1.3391910985517486,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048254541874757895,
      "loss": 5.72,
      "mean_token_accuracy": 0.17593135088682174,
      "num_tokens": 28255414.0,
      "step": 15165
    },
    {
      "entropy": 5.866649198532104,
      "epoch": 1.3396326386435888,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048253325537311463,
      "loss": 5.575,
      "mean_token_accuracy": 0.18223771154880525,
      "num_tokens": 28264162.0,
      "step": 15170
    },
    {
      "entropy": 5.899988651275635,
      "epoch": 1.3400741787354291,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048252108793319184,
      "loss": 5.6395,
      "mean_token_accuracy": 0.1787200778722763,
      "num_tokens": 28273146.0,
      "step": 15175
    },
    {
      "entropy": 5.885097551345825,
      "epoch": 1.3405157188272696,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004825089164280489,
      "loss": 5.5805,
      "mean_token_accuracy": 0.17420858293771743,
      "num_tokens": 28281865.0,
      "step": 15180
    },
    {
      "entropy": 5.8747031688690186,
      "epoch": 1.3409572589191099,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048249674085792433,
      "loss": 5.5458,
      "mean_token_accuracy": 0.17954271137714387,
      "num_tokens": 28291197.0,
      "step": 15185
    },
    {
      "entropy": 5.89144377708435,
      "epoch": 1.3413987990109502,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048248456122305667,
      "loss": 5.7434,
      "mean_token_accuracy": 0.17421520501375198,
      "num_tokens": 28300283.0,
      "step": 15190
    },
    {
      "entropy": 5.8786975860595705,
      "epoch": 1.3418403391027907,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004824723775236844,
      "loss": 5.7992,
      "mean_token_accuracy": 0.1683397799730301,
      "num_tokens": 28309121.0,
      "step": 15195
    },
    {
      "entropy": 5.9606489658355715,
      "epoch": 1.342281879194631,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004824601897600463,
      "loss": 5.5971,
      "mean_token_accuracy": 0.1837773695588112,
      "num_tokens": 28318377.0,
      "step": 15200
    },
    {
      "entropy": 5.965206575393677,
      "epoch": 1.3427234192864712,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048244799793238104,
      "loss": 5.5882,
      "mean_token_accuracy": 0.180729478597641,
      "num_tokens": 28328341.0,
      "step": 15205
    },
    {
      "entropy": 5.9311988830566404,
      "epoch": 1.3431649593783115,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004824358020409275,
      "loss": 5.7912,
      "mean_token_accuracy": 0.17183338552713395,
      "num_tokens": 28337340.0,
      "step": 15210
    },
    {
      "entropy": 5.864830350875854,
      "epoch": 1.3436064994701518,
      "grad_norm": 1.875,
      "learning_rate": 0.0004824236020859246,
      "loss": 5.723,
      "mean_token_accuracy": 0.1779323160648346,
      "num_tokens": 28347653.0,
      "step": 15215
    },
    {
      "entropy": 6.054541015625,
      "epoch": 1.3440480395619923,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048241139806761124,
      "loss": 5.759,
      "mean_token_accuracy": 0.17057999819517136,
      "num_tokens": 28356596.0,
      "step": 15220
    },
    {
      "entropy": 5.959856367111206,
      "epoch": 1.3444895796538325,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048239918998622657,
      "loss": 5.7408,
      "mean_token_accuracy": 0.17471463233232498,
      "num_tokens": 28366253.0,
      "step": 15225
    },
    {
      "entropy": 5.8222959518432615,
      "epoch": 1.3449311197456728,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004823869778420097,
      "loss": 5.5834,
      "mean_token_accuracy": 0.18445563316345215,
      "num_tokens": 28374815.0,
      "step": 15230
    },
    {
      "entropy": 5.889749479293823,
      "epoch": 1.3453726598375133,
      "grad_norm": 1.75,
      "learning_rate": 0.00048237476163519994,
      "loss": 5.6875,
      "mean_token_accuracy": 0.1735275000333786,
      "num_tokens": 28384221.0,
      "step": 15235
    },
    {
      "entropy": 5.901077032089233,
      "epoch": 1.3458141999293536,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004823625413660365,
      "loss": 5.6543,
      "mean_token_accuracy": 0.17886954694986343,
      "num_tokens": 28393453.0,
      "step": 15240
    },
    {
      "entropy": 5.885251998901367,
      "epoch": 1.3462557400211939,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048235031703475884,
      "loss": 5.5907,
      "mean_token_accuracy": 0.188381527364254,
      "num_tokens": 28402343.0,
      "step": 15245
    },
    {
      "entropy": 5.877451992034912,
      "epoch": 1.3466972801130344,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004823380886416064,
      "loss": 5.6157,
      "mean_token_accuracy": 0.1855811893939972,
      "num_tokens": 28411527.0,
      "step": 15250
    },
    {
      "entropy": 5.915568161010742,
      "epoch": 1.3471388202048746,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004823258561868187,
      "loss": 5.7292,
      "mean_token_accuracy": 0.17174500226974487,
      "num_tokens": 28422518.0,
      "step": 15255
    },
    {
      "entropy": 5.952272987365722,
      "epoch": 1.347580360296715,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048231361967063534,
      "loss": 5.6766,
      "mean_token_accuracy": 0.17071190029382705,
      "num_tokens": 28431692.0,
      "step": 15260
    },
    {
      "entropy": 5.95924186706543,
      "epoch": 1.3480219003885554,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004823013790932961,
      "loss": 5.6876,
      "mean_token_accuracy": 0.17967732697725297,
      "num_tokens": 28440304.0,
      "step": 15265
    },
    {
      "entropy": 5.974655485153198,
      "epoch": 1.3484634404803957,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004822891344550408,
      "loss": 5.7536,
      "mean_token_accuracy": 0.16915396004915237,
      "num_tokens": 28450265.0,
      "step": 15270
    },
    {
      "entropy": 5.864753341674804,
      "epoch": 1.348904980572236,
      "grad_norm": 1.96875,
      "learning_rate": 0.00048227688575610915,
      "loss": 5.6476,
      "mean_token_accuracy": 0.17562443315982817,
      "num_tokens": 28459699.0,
      "step": 15275
    },
    {
      "entropy": 5.9265399932861325,
      "epoch": 1.3493465206640762,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004822646329967413,
      "loss": 5.6524,
      "mean_token_accuracy": 0.17180057764053344,
      "num_tokens": 28468523.0,
      "step": 15280
    },
    {
      "entropy": 6.004323720932007,
      "epoch": 1.3497880607559165,
      "grad_norm": 2.125,
      "learning_rate": 0.0004822523761771771,
      "loss": 5.7341,
      "mean_token_accuracy": 0.17033194303512572,
      "num_tokens": 28477629.0,
      "step": 15285
    },
    {
      "entropy": 5.882391548156738,
      "epoch": 1.350229600847757,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004822401152976568,
      "loss": 5.5999,
      "mean_token_accuracy": 0.1844799742102623,
      "num_tokens": 28486517.0,
      "step": 15290
    },
    {
      "entropy": 5.907935237884521,
      "epoch": 1.3506711409395973,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048222785035842045,
      "loss": 5.7145,
      "mean_token_accuracy": 0.17215269505977632,
      "num_tokens": 28496103.0,
      "step": 15295
    },
    {
      "entropy": 5.891082143783569,
      "epoch": 1.3511126810314376,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048221558135970834,
      "loss": 5.6879,
      "mean_token_accuracy": 0.17928629666566848,
      "num_tokens": 28505538.0,
      "step": 15300
    },
    {
      "entropy": 5.844837617874146,
      "epoch": 1.351554221123278,
      "grad_norm": 2.75,
      "learning_rate": 0.00048220330830176086,
      "loss": 5.5458,
      "mean_token_accuracy": 0.1818324625492096,
      "num_tokens": 28514322.0,
      "step": 15305
    },
    {
      "entropy": 6.035720014572144,
      "epoch": 1.3519957612151183,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004821910311848184,
      "loss": 5.6657,
      "mean_token_accuracy": 0.1800089344382286,
      "num_tokens": 28524298.0,
      "step": 15310
    },
    {
      "entropy": 5.921222448348999,
      "epoch": 1.3524373013069586,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004821787500091215,
      "loss": 5.6824,
      "mean_token_accuracy": 0.1783458724617958,
      "num_tokens": 28533567.0,
      "step": 15315
    },
    {
      "entropy": 5.802498912811279,
      "epoch": 1.352878841398799,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048216646477491074,
      "loss": 5.6684,
      "mean_token_accuracy": 0.17602547705173494,
      "num_tokens": 28541746.0,
      "step": 15320
    },
    {
      "entropy": 5.916517019271851,
      "epoch": 1.3533203814906394,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004821541754824267,
      "loss": 5.7181,
      "mean_token_accuracy": 0.1673661157488823,
      "num_tokens": 28551685.0,
      "step": 15325
    },
    {
      "entropy": 6.054804420471191,
      "epoch": 1.3537619215824797,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004821418821319102,
      "loss": 5.7354,
      "mean_token_accuracy": 0.17476488053798675,
      "num_tokens": 28560935.0,
      "step": 15330
    },
    {
      "entropy": 5.9118176937103275,
      "epoch": 1.3542034616743202,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004821295847236021,
      "loss": 5.5781,
      "mean_token_accuracy": 0.17985003143548967,
      "num_tokens": 28569506.0,
      "step": 15335
    },
    {
      "entropy": 5.844781017303466,
      "epoch": 1.3546450017661604,
      "grad_norm": 1.5,
      "learning_rate": 0.0004821172832577431,
      "loss": 5.6576,
      "mean_token_accuracy": 0.17644434571266174,
      "num_tokens": 28579792.0,
      "step": 15340
    },
    {
      "entropy": 5.918091106414795,
      "epoch": 1.3550865418580007,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004821049777345744,
      "loss": 5.7555,
      "mean_token_accuracy": 0.17478242069482802,
      "num_tokens": 28589756.0,
      "step": 15345
    },
    {
      "entropy": 5.933539915084839,
      "epoch": 1.355528081949841,
      "grad_norm": 1.4375,
      "learning_rate": 0.000482092668154337,
      "loss": 5.6665,
      "mean_token_accuracy": 0.17668498158454896,
      "num_tokens": 28599853.0,
      "step": 15350
    },
    {
      "entropy": 5.929706573486328,
      "epoch": 1.3559696220416813,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00048208035451727195,
      "loss": 5.686,
      "mean_token_accuracy": 0.17553935647010804,
      "num_tokens": 28608576.0,
      "step": 15355
    },
    {
      "entropy": 5.937180137634277,
      "epoch": 1.3564111621335218,
      "grad_norm": 1.375,
      "learning_rate": 0.0004820680368236206,
      "loss": 5.7246,
      "mean_token_accuracy": 0.16752175390720367,
      "num_tokens": 28618820.0,
      "step": 15360
    },
    {
      "entropy": 5.914182662963867,
      "epoch": 1.356852702225362,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004820557150736241,
      "loss": 5.6223,
      "mean_token_accuracy": 0.18305070847272872,
      "num_tokens": 28628092.0,
      "step": 15365
    },
    {
      "entropy": 5.9173197746276855,
      "epoch": 1.3572942423172023,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048204338926752396,
      "loss": 5.7237,
      "mean_token_accuracy": 0.1763898625969887,
      "num_tokens": 28637849.0,
      "step": 15370
    },
    {
      "entropy": 5.851296043395996,
      "epoch": 1.3577357824090428,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004820310594055616,
      "loss": 5.6152,
      "mean_token_accuracy": 0.1806778699159622,
      "num_tokens": 28646957.0,
      "step": 15375
    },
    {
      "entropy": 6.017248630523682,
      "epoch": 1.358177322500883,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004820187254879784,
      "loss": 5.7161,
      "mean_token_accuracy": 0.17495856285095215,
      "num_tokens": 28655768.0,
      "step": 15380
    },
    {
      "entropy": 5.944850158691406,
      "epoch": 1.3586188625927234,
      "grad_norm": 1.5,
      "learning_rate": 0.00048200638751501616,
      "loss": 5.6938,
      "mean_token_accuracy": 0.1770474225282669,
      "num_tokens": 28664719.0,
      "step": 15385
    },
    {
      "entropy": 5.88440203666687,
      "epoch": 1.3590604026845639,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004819940454869166,
      "loss": 5.6705,
      "mean_token_accuracy": 0.17429258972406386,
      "num_tokens": 28673568.0,
      "step": 15390
    },
    {
      "entropy": 5.843747615814209,
      "epoch": 1.3595019427764041,
      "grad_norm": 1.75,
      "learning_rate": 0.0004819816994039213,
      "loss": 5.6381,
      "mean_token_accuracy": 0.17764560133218765,
      "num_tokens": 28682558.0,
      "step": 15395
    },
    {
      "entropy": 5.8986372470855715,
      "epoch": 1.3599434828682444,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048196934926627227,
      "loss": 5.7348,
      "mean_token_accuracy": 0.17941066026687622,
      "num_tokens": 28691965.0,
      "step": 15400
    },
    {
      "entropy": 5.844736814498901,
      "epoch": 1.360385022960085,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004819569950742114,
      "loss": 5.6124,
      "mean_token_accuracy": 0.1838567301630974,
      "num_tokens": 28700805.0,
      "step": 15405
    },
    {
      "entropy": 5.880643129348755,
      "epoch": 1.3608265630519252,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004819446368279808,
      "loss": 5.6138,
      "mean_token_accuracy": 0.18501638174057006,
      "num_tokens": 28710169.0,
      "step": 15410
    },
    {
      "entropy": 6.016932535171509,
      "epoch": 1.3612681031437655,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00048193227452782237,
      "loss": 5.7768,
      "mean_token_accuracy": 0.1663575991988182,
      "num_tokens": 28718656.0,
      "step": 15415
    },
    {
      "entropy": 5.991364240646362,
      "epoch": 1.3617096432356057,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004819199081739783,
      "loss": 5.7239,
      "mean_token_accuracy": 0.17761796712875366,
      "num_tokens": 28728546.0,
      "step": 15420
    },
    {
      "entropy": 5.909090948104859,
      "epoch": 1.362151183327446,
      "grad_norm": 3.03125,
      "learning_rate": 0.000481907537766691,
      "loss": 5.6079,
      "mean_token_accuracy": 0.1793052986264229,
      "num_tokens": 28737267.0,
      "step": 15425
    },
    {
      "entropy": 5.856475734710694,
      "epoch": 1.3625927234192865,
      "grad_norm": 1.75,
      "learning_rate": 0.0004818951633062027,
      "loss": 5.6926,
      "mean_token_accuracy": 0.17626506090164185,
      "num_tokens": 28746721.0,
      "step": 15430
    },
    {
      "entropy": 5.941016721725464,
      "epoch": 1.3630342635111268,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048188278479275584,
      "loss": 5.7371,
      "mean_token_accuracy": 0.1762013852596283,
      "num_tokens": 28756099.0,
      "step": 15435
    },
    {
      "entropy": 5.972643232345581,
      "epoch": 1.363475803602967,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004818704022265928,
      "loss": 5.6697,
      "mean_token_accuracy": 0.17468469291925431,
      "num_tokens": 28765463.0,
      "step": 15440
    },
    {
      "entropy": 5.932384729385376,
      "epoch": 1.3639173436948075,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048185801560795627,
      "loss": 5.6293,
      "mean_token_accuracy": 0.18055523186922073,
      "num_tokens": 28775284.0,
      "step": 15445
    },
    {
      "entropy": 5.95625638961792,
      "epoch": 1.3643588837866478,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048184562493708895,
      "loss": 5.6753,
      "mean_token_accuracy": 0.18499337285757064,
      "num_tokens": 28784524.0,
      "step": 15450
    },
    {
      "entropy": 5.885971021652222,
      "epoch": 1.364800423878488,
      "grad_norm": 1.75,
      "learning_rate": 0.00048183323021423334,
      "loss": 5.6353,
      "mean_token_accuracy": 0.17553602457046508,
      "num_tokens": 28792942.0,
      "step": 15455
    },
    {
      "entropy": 5.859358501434326,
      "epoch": 1.3652419639703286,
      "grad_norm": 3.296875,
      "learning_rate": 0.00048182083143963247,
      "loss": 5.6324,
      "mean_token_accuracy": 0.17992480844259262,
      "num_tokens": 28801746.0,
      "step": 15460
    },
    {
      "entropy": 5.960585260391236,
      "epoch": 1.3656835040621689,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004818084286135291,
      "loss": 5.699,
      "mean_token_accuracy": 0.17309850454330444,
      "num_tokens": 28810276.0,
      "step": 15465
    },
    {
      "entropy": 5.9734467506408695,
      "epoch": 1.3661250441540091,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004817960217361663,
      "loss": 5.6985,
      "mean_token_accuracy": 0.1776546284556389,
      "num_tokens": 28819545.0,
      "step": 15470
    },
    {
      "entropy": 5.849445581436157,
      "epoch": 1.3665665842458496,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000481783610807787,
      "loss": 5.5761,
      "mean_token_accuracy": 0.18362480849027635,
      "num_tokens": 28828163.0,
      "step": 15475
    },
    {
      "entropy": 5.911349058151245,
      "epoch": 1.36700812433769,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004817711958286343,
      "loss": 5.6902,
      "mean_token_accuracy": 0.17612197995185852,
      "num_tokens": 28837952.0,
      "step": 15480
    },
    {
      "entropy": 5.961468744277954,
      "epoch": 1.3674496644295302,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004817587767989516,
      "loss": 5.7423,
      "mean_token_accuracy": 0.16886923909187318,
      "num_tokens": 28847702.0,
      "step": 15485
    },
    {
      "entropy": 5.928764009475708,
      "epoch": 1.3678912045213705,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048174635371898197,
      "loss": 5.6948,
      "mean_token_accuracy": 0.17367339134216309,
      "num_tokens": 28856936.0,
      "step": 15490
    },
    {
      "entropy": 5.9199000835418705,
      "epoch": 1.3683327446132107,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004817339265889689,
      "loss": 5.6209,
      "mean_token_accuracy": 0.17979218065738678,
      "num_tokens": 28867072.0,
      "step": 15495
    },
    {
      "entropy": 5.840210103988648,
      "epoch": 1.3687742847050512,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004817214954091557,
      "loss": 5.5544,
      "mean_token_accuracy": 0.1836580216884613,
      "num_tokens": 28877546.0,
      "step": 15500
    },
    {
      "entropy": 5.918880796432495,
      "epoch": 1.3692158247968915,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048170906017978605,
      "loss": 5.658,
      "mean_token_accuracy": 0.17833959311246872,
      "num_tokens": 28886804.0,
      "step": 15505
    },
    {
      "entropy": 5.956535720825196,
      "epoch": 1.3696573648887318,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004816966209011034,
      "loss": 5.726,
      "mean_token_accuracy": 0.16565614342689514,
      "num_tokens": 28896476.0,
      "step": 15510
    },
    {
      "entropy": 6.0167333602905275,
      "epoch": 1.3700989049805723,
      "grad_norm": 1.75,
      "learning_rate": 0.00048168417757335155,
      "loss": 5.7064,
      "mean_token_accuracy": 0.1758236452937126,
      "num_tokens": 28906269.0,
      "step": 15515
    },
    {
      "entropy": 5.876074028015137,
      "epoch": 1.3705404450724126,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004816717301967743,
      "loss": 5.6399,
      "mean_token_accuracy": 0.17681920528411865,
      "num_tokens": 28915792.0,
      "step": 15520
    },
    {
      "entropy": 5.876484346389771,
      "epoch": 1.3709819851642528,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004816592787716152,
      "loss": 5.6848,
      "mean_token_accuracy": 0.1716994509100914,
      "num_tokens": 28925065.0,
      "step": 15525
    },
    {
      "entropy": 5.9178307056427,
      "epoch": 1.3714235252560933,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048164682329811855,
      "loss": 5.6949,
      "mean_token_accuracy": 0.16972355395555497,
      "num_tokens": 28933879.0,
      "step": 15530
    },
    {
      "entropy": 5.859436178207398,
      "epoch": 1.3718650653479336,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048163436377652804,
      "loss": 5.5338,
      "mean_token_accuracy": 0.18893634378910065,
      "num_tokens": 28943176.0,
      "step": 15535
    },
    {
      "entropy": 5.953158950805664,
      "epoch": 1.372306605439774,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004816219002070879,
      "loss": 5.7161,
      "mean_token_accuracy": 0.17766279578208924,
      "num_tokens": 28952770.0,
      "step": 15540
    },
    {
      "entropy": 5.782121562957764,
      "epoch": 1.3727481455316144,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004816094325900422,
      "loss": 5.547,
      "mean_token_accuracy": 0.18763880133628846,
      "num_tokens": 28961186.0,
      "step": 15545
    },
    {
      "entropy": 5.943677759170532,
      "epoch": 1.3731896856234547,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048159696092563533,
      "loss": 5.6687,
      "mean_token_accuracy": 0.17665790766477585,
      "num_tokens": 28971787.0,
      "step": 15550
    },
    {
      "entropy": 5.944572639465332,
      "epoch": 1.373631225715295,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004815844852141114,
      "loss": 5.777,
      "mean_token_accuracy": 0.17063609808683394,
      "num_tokens": 28982095.0,
      "step": 15555
    },
    {
      "entropy": 5.915904569625854,
      "epoch": 1.3740727658071352,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048157200545571497,
      "loss": 5.6677,
      "mean_token_accuracy": 0.18237135857343673,
      "num_tokens": 28991084.0,
      "step": 15560
    },
    {
      "entropy": 5.900287294387818,
      "epoch": 1.3745143058989755,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004815595216506903,
      "loss": 5.6655,
      "mean_token_accuracy": 0.1777072086930275,
      "num_tokens": 29000544.0,
      "step": 15565
    },
    {
      "entropy": 5.920689916610717,
      "epoch": 1.374955845990816,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004815470337992822,
      "loss": 5.5598,
      "mean_token_accuracy": 0.17998860776424408,
      "num_tokens": 29009366.0,
      "step": 15570
    },
    {
      "entropy": 5.872371816635132,
      "epoch": 1.3753973860826563,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048153454190173517,
      "loss": 5.7037,
      "mean_token_accuracy": 0.18155049979686738,
      "num_tokens": 29018730.0,
      "step": 15575
    },
    {
      "entropy": 6.013231182098389,
      "epoch": 1.3758389261744965,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048152204595829394,
      "loss": 5.7577,
      "mean_token_accuracy": 0.16663735657930373,
      "num_tokens": 29030255.0,
      "step": 15580
    },
    {
      "entropy": 5.973646116256714,
      "epoch": 1.376280466266337,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004815095459692032,
      "loss": 5.6078,
      "mean_token_accuracy": 0.1784512460231781,
      "num_tokens": 29039030.0,
      "step": 15585
    },
    {
      "entropy": 5.883748388290405,
      "epoch": 1.3767220063581773,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000481497041934708,
      "loss": 5.6443,
      "mean_token_accuracy": 0.18333866745233535,
      "num_tokens": 29047504.0,
      "step": 15590
    },
    {
      "entropy": 5.98466124534607,
      "epoch": 1.3771635464500176,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048148453385505317,
      "loss": 5.7329,
      "mean_token_accuracy": 0.17840099036693574,
      "num_tokens": 29056488.0,
      "step": 15595
    },
    {
      "entropy": 5.943624687194824,
      "epoch": 1.377605086541858,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048147202173048366,
      "loss": 5.66,
      "mean_token_accuracy": 0.18977410048246385,
      "num_tokens": 29066067.0,
      "step": 15600
    },
    {
      "entropy": 5.980694437026978,
      "epoch": 1.3780466266336984,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004814595055612447,
      "loss": 5.674,
      "mean_token_accuracy": 0.17975226789712906,
      "num_tokens": 29075136.0,
      "step": 15605
    },
    {
      "entropy": 5.842377853393555,
      "epoch": 1.3784881667255386,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048144698534758153,
      "loss": 5.636,
      "mean_token_accuracy": 0.18363478928804397,
      "num_tokens": 29084970.0,
      "step": 15610
    },
    {
      "entropy": 5.796068143844605,
      "epoch": 1.3789297068173791,
      "grad_norm": 1.71875,
      "learning_rate": 0.00048143446108973924,
      "loss": 5.5747,
      "mean_token_accuracy": 0.1790820524096489,
      "num_tokens": 29093234.0,
      "step": 15615
    },
    {
      "entropy": 5.92116641998291,
      "epoch": 1.3793712469092194,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048142193278796335,
      "loss": 5.6308,
      "mean_token_accuracy": 0.1782672271132469,
      "num_tokens": 29102115.0,
      "step": 15620
    },
    {
      "entropy": 5.972972106933594,
      "epoch": 1.3798127870010597,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004814094004424992,
      "loss": 5.6612,
      "mean_token_accuracy": 0.18050715923309327,
      "num_tokens": 29111065.0,
      "step": 15625
    },
    {
      "entropy": 5.849059772491455,
      "epoch": 1.3802543270929,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004813968640535922,
      "loss": 5.69,
      "mean_token_accuracy": 0.17682965993881225,
      "num_tokens": 29120238.0,
      "step": 15630
    },
    {
      "entropy": 5.89948582649231,
      "epoch": 1.3806958671847402,
      "grad_norm": 1.671875,
      "learning_rate": 0.000481384323621488,
      "loss": 5.6067,
      "mean_token_accuracy": 0.17593304067850113,
      "num_tokens": 29129062.0,
      "step": 15635
    },
    {
      "entropy": 5.95424075126648,
      "epoch": 1.3811374072765807,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004813717791464324,
      "loss": 5.7293,
      "mean_token_accuracy": 0.16904746294021605,
      "num_tokens": 29138990.0,
      "step": 15640
    },
    {
      "entropy": 5.892905282974243,
      "epoch": 1.381578947368421,
      "grad_norm": 1.5,
      "learning_rate": 0.00048135923062867094,
      "loss": 5.599,
      "mean_token_accuracy": 0.18791977018117906,
      "num_tokens": 29147507.0,
      "step": 15645
    },
    {
      "entropy": 5.865453052520752,
      "epoch": 1.3820204874602613,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004813466780684495,
      "loss": 5.7347,
      "mean_token_accuracy": 0.17258007377386092,
      "num_tokens": 29156277.0,
      "step": 15650
    },
    {
      "entropy": 5.974000549316406,
      "epoch": 1.3824620275521018,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004813341214660141,
      "loss": 5.8275,
      "mean_token_accuracy": 0.16973593533039094,
      "num_tokens": 29165925.0,
      "step": 15655
    },
    {
      "entropy": 6.026811218261718,
      "epoch": 1.382903567643942,
      "grad_norm": 2.5,
      "learning_rate": 0.0004813215608216105,
      "loss": 5.7386,
      "mean_token_accuracy": 0.17053484469652175,
      "num_tokens": 29175811.0,
      "step": 15660
    },
    {
      "entropy": 5.941466426849365,
      "epoch": 1.3833451077357823,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048130899613548487,
      "loss": 5.7218,
      "mean_token_accuracy": 0.17622524052858352,
      "num_tokens": 29185100.0,
      "step": 15665
    },
    {
      "entropy": 5.943367958068848,
      "epoch": 1.3837866478276228,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048129642740788337,
      "loss": 5.6885,
      "mean_token_accuracy": 0.17363928854465485,
      "num_tokens": 29194430.0,
      "step": 15670
    },
    {
      "entropy": 5.891606283187866,
      "epoch": 1.3842281879194631,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004812838546390521,
      "loss": 5.6352,
      "mean_token_accuracy": 0.1810683786869049,
      "num_tokens": 29203897.0,
      "step": 15675
    },
    {
      "entropy": 5.94944167137146,
      "epoch": 1.3846697280113034,
      "grad_norm": 1.375,
      "learning_rate": 0.0004812712778292375,
      "loss": 5.7635,
      "mean_token_accuracy": 0.17569662034511566,
      "num_tokens": 29213474.0,
      "step": 15680
    },
    {
      "entropy": 6.017031621932984,
      "epoch": 1.3851112681031439,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000481258696978686,
      "loss": 5.7191,
      "mean_token_accuracy": 0.17579027265310287,
      "num_tokens": 29222176.0,
      "step": 15685
    },
    {
      "entropy": 5.918281507492066,
      "epoch": 1.3855528081949842,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048124611208764375,
      "loss": 5.6351,
      "mean_token_accuracy": 0.1741022691130638,
      "num_tokens": 29231885.0,
      "step": 15690
    },
    {
      "entropy": 5.907310056686401,
      "epoch": 1.3859943482868244,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004812335231563576,
      "loss": 5.6403,
      "mean_token_accuracy": 0.18150193691253663,
      "num_tokens": 29241921.0,
      "step": 15695
    },
    {
      "entropy": 5.874593210220337,
      "epoch": 1.3864358883786647,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048122093018507396,
      "loss": 5.716,
      "mean_token_accuracy": 0.17979821264743806,
      "num_tokens": 29251308.0,
      "step": 15700
    },
    {
      "entropy": 5.888420724868775,
      "epoch": 1.386877428470505,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004812083331740396,
      "loss": 5.6676,
      "mean_token_accuracy": 0.18491006195545195,
      "num_tokens": 29260628.0,
      "step": 15705
    },
    {
      "entropy": 5.994103240966797,
      "epoch": 1.3873189685623455,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048119573212350127,
      "loss": 5.6878,
      "mean_token_accuracy": 0.17344592958688737,
      "num_tokens": 29269530.0,
      "step": 15710
    },
    {
      "entropy": 5.97232666015625,
      "epoch": 1.3877605086541858,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048118312703370576,
      "loss": 5.7147,
      "mean_token_accuracy": 0.17104984521865846,
      "num_tokens": 29279846.0,
      "step": 15715
    },
    {
      "entropy": 5.978313446044922,
      "epoch": 1.388202048746026,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048117051790490007,
      "loss": 5.6948,
      "mean_token_accuracy": 0.17586121708154678,
      "num_tokens": 29289822.0,
      "step": 15720
    },
    {
      "entropy": 5.845377826690674,
      "epoch": 1.3886435888378665,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048115790473733124,
      "loss": 5.6753,
      "mean_token_accuracy": 0.17724470496177674,
      "num_tokens": 29299454.0,
      "step": 15725
    },
    {
      "entropy": 5.879784393310547,
      "epoch": 1.3890851289297068,
      "grad_norm": 1.625,
      "learning_rate": 0.00048114528753124625,
      "loss": 5.7162,
      "mean_token_accuracy": 0.1759541869163513,
      "num_tokens": 29308398.0,
      "step": 15730
    },
    {
      "entropy": 5.9425211429595945,
      "epoch": 1.389526669021547,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048113266628689236,
      "loss": 5.6481,
      "mean_token_accuracy": 0.18479530811309813,
      "num_tokens": 29318968.0,
      "step": 15735
    },
    {
      "entropy": 5.996930837631226,
      "epoch": 1.3899682091133876,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004811200410045167,
      "loss": 5.7484,
      "mean_token_accuracy": 0.1635768473148346,
      "num_tokens": 29329745.0,
      "step": 15740
    },
    {
      "entropy": 5.884109306335449,
      "epoch": 1.3904097492052279,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004811074116843667,
      "loss": 5.546,
      "mean_token_accuracy": 0.18526943922042846,
      "num_tokens": 29338438.0,
      "step": 15745
    },
    {
      "entropy": 5.8488929271698,
      "epoch": 1.3908512892970681,
      "grad_norm": 1.75,
      "learning_rate": 0.00048109477832668977,
      "loss": 5.6182,
      "mean_token_accuracy": 0.17867804914712906,
      "num_tokens": 29347783.0,
      "step": 15750
    },
    {
      "entropy": 5.873744821548462,
      "epoch": 1.3912928293889086,
      "grad_norm": 1.75,
      "learning_rate": 0.0004810821409317334,
      "loss": 5.6928,
      "mean_token_accuracy": 0.17970954030752181,
      "num_tokens": 29357062.0,
      "step": 15755
    },
    {
      "entropy": 5.924177169799805,
      "epoch": 1.391734369480749,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048106949949974494,
      "loss": 5.6663,
      "mean_token_accuracy": 0.1808513030409813,
      "num_tokens": 29366361.0,
      "step": 15760
    },
    {
      "entropy": 5.827966928482056,
      "epoch": 1.3921759095725892,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004810568540309723,
      "loss": 5.6978,
      "mean_token_accuracy": 0.17053220570087432,
      "num_tokens": 29375992.0,
      "step": 15765
    },
    {
      "entropy": 5.930188417434692,
      "epoch": 1.3926174496644295,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048104420452566303,
      "loss": 5.7582,
      "mean_token_accuracy": 0.17517874538898467,
      "num_tokens": 29386408.0,
      "step": 15770
    },
    {
      "entropy": 6.037285566329956,
      "epoch": 1.3930589897562697,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048103155098406505,
      "loss": 5.7592,
      "mean_token_accuracy": 0.17286611795425416,
      "num_tokens": 29395881.0,
      "step": 15775
    },
    {
      "entropy": 5.921860647201538,
      "epoch": 1.3935005298481102,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048101889340642604,
      "loss": 5.5679,
      "mean_token_accuracy": 0.18698693215847015,
      "num_tokens": 29404397.0,
      "step": 15780
    },
    {
      "entropy": 5.818957805633545,
      "epoch": 1.3939420699399505,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004810062317929941,
      "loss": 5.6492,
      "mean_token_accuracy": 0.17857605069875718,
      "num_tokens": 29413219.0,
      "step": 15785
    },
    {
      "entropy": 5.936586380004883,
      "epoch": 1.3943836100317908,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004809935661440172,
      "loss": 5.7315,
      "mean_token_accuracy": 0.1693406492471695,
      "num_tokens": 29422727.0,
      "step": 15790
    },
    {
      "entropy": 6.007265615463257,
      "epoch": 1.3948251501236313,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004809808964597436,
      "loss": 5.7292,
      "mean_token_accuracy": 0.17243069857358934,
      "num_tokens": 29432183.0,
      "step": 15795
    },
    {
      "entropy": 5.905940198898316,
      "epoch": 1.3952666902154716,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048096822274042145,
      "loss": 5.6324,
      "mean_token_accuracy": 0.17890364527702332,
      "num_tokens": 29440821.0,
      "step": 15800
    },
    {
      "entropy": 5.918467807769775,
      "epoch": 1.3957082303073118,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048095554498629885,
      "loss": 5.7528,
      "mean_token_accuracy": 0.16968773901462555,
      "num_tokens": 29450128.0,
      "step": 15805
    },
    {
      "entropy": 5.8856090068817135,
      "epoch": 1.3961497703991523,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048094286319762426,
      "loss": 5.6245,
      "mean_token_accuracy": 0.1809169813990593,
      "num_tokens": 29459271.0,
      "step": 15810
    },
    {
      "entropy": 5.914688301086426,
      "epoch": 1.3965913104909926,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048093017737464613,
      "loss": 5.5836,
      "mean_token_accuracy": 0.18467627912759782,
      "num_tokens": 29468517.0,
      "step": 15815
    },
    {
      "entropy": 5.907070827484131,
      "epoch": 1.3970328505828329,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048091748751761295,
      "loss": 5.751,
      "mean_token_accuracy": 0.16808104515075684,
      "num_tokens": 29478330.0,
      "step": 15820
    },
    {
      "entropy": 5.964093780517578,
      "epoch": 1.3974743906746734,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004809047936267732,
      "loss": 5.7073,
      "mean_token_accuracy": 0.17332876175642015,
      "num_tokens": 29487190.0,
      "step": 15825
    },
    {
      "entropy": 5.962989568710327,
      "epoch": 1.3979159307665137,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048089209570237573,
      "loss": 5.779,
      "mean_token_accuracy": 0.1710903450846672,
      "num_tokens": 29495885.0,
      "step": 15830
    },
    {
      "entropy": 5.904249048233032,
      "epoch": 1.398357470858354,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004808793937446692,
      "loss": 5.6426,
      "mean_token_accuracy": 0.171312353014946,
      "num_tokens": 29505524.0,
      "step": 15835
    },
    {
      "entropy": 5.866703271865845,
      "epoch": 1.3987990109501942,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004808666877539025,
      "loss": 5.6231,
      "mean_token_accuracy": 0.18337326645851135,
      "num_tokens": 29514003.0,
      "step": 15840
    },
    {
      "entropy": 5.908846807479859,
      "epoch": 1.3992405510420345,
      "grad_norm": 2.0,
      "learning_rate": 0.00048085397773032433,
      "loss": 5.6653,
      "mean_token_accuracy": 0.17538561522960663,
      "num_tokens": 29523236.0,
      "step": 15845
    },
    {
      "entropy": 5.958359622955323,
      "epoch": 1.399682091133875,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004808412636741839,
      "loss": 5.7105,
      "mean_token_accuracy": 0.17529277354478837,
      "num_tokens": 29532852.0,
      "step": 15850
    },
    {
      "entropy": 5.877608156204223,
      "epoch": 1.4001236312257153,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048082854558573017,
      "loss": 5.6741,
      "mean_token_accuracy": 0.18077631145715714,
      "num_tokens": 29542035.0,
      "step": 15855
    },
    {
      "entropy": 5.897961282730103,
      "epoch": 1.4005651713175555,
      "grad_norm": 3.34375,
      "learning_rate": 0.00048081582346521236,
      "loss": 5.6422,
      "mean_token_accuracy": 0.1795714780688286,
      "num_tokens": 29551978.0,
      "step": 15860
    },
    {
      "entropy": 6.027183151245117,
      "epoch": 1.401006711409396,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004808030973128795,
      "loss": 5.7959,
      "mean_token_accuracy": 0.16765390187501908,
      "num_tokens": 29562011.0,
      "step": 15865
    },
    {
      "entropy": 5.9084801197052,
      "epoch": 1.4014482515012363,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048079036712898106,
      "loss": 5.646,
      "mean_token_accuracy": 0.18198180198669434,
      "num_tokens": 29571498.0,
      "step": 15870
    },
    {
      "entropy": 5.867594432830811,
      "epoch": 1.4018897915930766,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048077763291376637,
      "loss": 5.6033,
      "mean_token_accuracy": 0.1884506642818451,
      "num_tokens": 29580704.0,
      "step": 15875
    },
    {
      "entropy": 5.911268949508667,
      "epoch": 1.402331331684917,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004807648946674849,
      "loss": 5.5928,
      "mean_token_accuracy": 0.18523462414741515,
      "num_tokens": 29590066.0,
      "step": 15880
    },
    {
      "entropy": 5.905565071105957,
      "epoch": 1.4027728717767574,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004807521523903862,
      "loss": 5.6941,
      "mean_token_accuracy": 0.1730918690562248,
      "num_tokens": 29599473.0,
      "step": 15885
    },
    {
      "entropy": 5.89090781211853,
      "epoch": 1.4032144118685976,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048073940608271974,
      "loss": 5.6932,
      "mean_token_accuracy": 0.17246467918157576,
      "num_tokens": 29609371.0,
      "step": 15890
    },
    {
      "entropy": 5.9210103988647464,
      "epoch": 1.4036559519604381,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004807266557447354,
      "loss": 5.569,
      "mean_token_accuracy": 0.1843820706009865,
      "num_tokens": 29618821.0,
      "step": 15895
    },
    {
      "entropy": 5.878189992904663,
      "epoch": 1.4040974920522784,
      "grad_norm": 2.125,
      "learning_rate": 0.0004807139013766829,
      "loss": 5.5832,
      "mean_token_accuracy": 0.18622346818447114,
      "num_tokens": 29627491.0,
      "step": 15900
    },
    {
      "entropy": 5.848198890686035,
      "epoch": 1.4045390321441187,
      "grad_norm": 1.625,
      "learning_rate": 0.00048070114297881205,
      "loss": 5.681,
      "mean_token_accuracy": 0.17946704030036925,
      "num_tokens": 29637699.0,
      "step": 15905
    },
    {
      "entropy": 5.9514282703399655,
      "epoch": 1.404980572235959,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004806883805513728,
      "loss": 5.6469,
      "mean_token_accuracy": 0.18110415637493132,
      "num_tokens": 29646439.0,
      "step": 15910
    },
    {
      "entropy": 5.95872130393982,
      "epoch": 1.4054221123277992,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004806756140946151,
      "loss": 5.7041,
      "mean_token_accuracy": 0.16715109646320342,
      "num_tokens": 29655704.0,
      "step": 15915
    },
    {
      "entropy": 5.998380565643311,
      "epoch": 1.4058636524196397,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048066284360878913,
      "loss": 5.7742,
      "mean_token_accuracy": 0.17472067028284072,
      "num_tokens": 29665163.0,
      "step": 15920
    },
    {
      "entropy": 5.8917937755584715,
      "epoch": 1.40630519251148,
      "grad_norm": 1.75,
      "learning_rate": 0.00048065006909414507,
      "loss": 5.6729,
      "mean_token_accuracy": 0.1777106076478958,
      "num_tokens": 29674803.0,
      "step": 15925
    },
    {
      "entropy": 5.9474962711334225,
      "epoch": 1.4067467326033203,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004806372905509331,
      "loss": 5.6924,
      "mean_token_accuracy": 0.17631972432136536,
      "num_tokens": 29685149.0,
      "step": 15930
    },
    {
      "entropy": 5.831018733978271,
      "epoch": 1.4071882726951608,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00048062450797940347,
      "loss": 5.5585,
      "mean_token_accuracy": 0.18526545614004136,
      "num_tokens": 29694224.0,
      "step": 15935
    },
    {
      "entropy": 5.897001028060913,
      "epoch": 1.407629812787001,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004806117213798068,
      "loss": 5.6631,
      "mean_token_accuracy": 0.18107045739889144,
      "num_tokens": 29704512.0,
      "step": 15940
    },
    {
      "entropy": 5.855301570892334,
      "epoch": 1.4080713528788413,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048059893075239334,
      "loss": 5.581,
      "mean_token_accuracy": 0.17911613881587982,
      "num_tokens": 29712990.0,
      "step": 15945
    },
    {
      "entropy": 5.948109531402588,
      "epoch": 1.4085128929706818,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048058613609741374,
      "loss": 5.7145,
      "mean_token_accuracy": 0.17642999738454818,
      "num_tokens": 29722844.0,
      "step": 15950
    },
    {
      "entropy": 5.953815221786499,
      "epoch": 1.408954433062522,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004805733374151188,
      "loss": 5.5999,
      "mean_token_accuracy": 0.17559873461723327,
      "num_tokens": 29731593.0,
      "step": 15955
    },
    {
      "entropy": 5.936201524734497,
      "epoch": 1.4093959731543624,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048056053470575894,
      "loss": 5.7044,
      "mean_token_accuracy": 0.1673717424273491,
      "num_tokens": 29741052.0,
      "step": 15960
    },
    {
      "entropy": 5.966942310333252,
      "epoch": 1.4098375132462029,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048054772796958517,
      "loss": 5.6901,
      "mean_token_accuracy": 0.17982660979032516,
      "num_tokens": 29750187.0,
      "step": 15965
    },
    {
      "entropy": 5.871349763870239,
      "epoch": 1.4102790533380432,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048053491720684836,
      "loss": 5.6371,
      "mean_token_accuracy": 0.1761351242661476,
      "num_tokens": 29759574.0,
      "step": 15970
    },
    {
      "entropy": 5.873283720016479,
      "epoch": 1.4107205934298834,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004805221024177994,
      "loss": 5.6742,
      "mean_token_accuracy": 0.17135443091392516,
      "num_tokens": 29769549.0,
      "step": 15975
    },
    {
      "entropy": 5.991179037094116,
      "epoch": 1.4111621335217237,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004805092836026893,
      "loss": 5.6967,
      "mean_token_accuracy": 0.17102856636047364,
      "num_tokens": 29778643.0,
      "step": 15980
    },
    {
      "entropy": 5.892449283599854,
      "epoch": 1.411603673613564,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004804964607617693,
      "loss": 5.6464,
      "mean_token_accuracy": 0.17699751853942872,
      "num_tokens": 29788685.0,
      "step": 15985
    },
    {
      "entropy": 5.827407026290894,
      "epoch": 1.4120452137054045,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00048048363389529045,
      "loss": 5.6019,
      "mean_token_accuracy": 0.18562958836555482,
      "num_tokens": 29798648.0,
      "step": 15990
    },
    {
      "entropy": 5.900378704071045,
      "epoch": 1.4124867537972448,
      "grad_norm": 1.640625,
      "learning_rate": 0.000480470803003504,
      "loss": 5.6807,
      "mean_token_accuracy": 0.17699760794639588,
      "num_tokens": 29807582.0,
      "step": 15995
    },
    {
      "entropy": 5.976664209365845,
      "epoch": 1.412928293889085,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004804579680866614,
      "loss": 5.6819,
      "mean_token_accuracy": 0.1775629550218582,
      "num_tokens": 29817443.0,
      "step": 16000
    },
    {
      "entropy": 5.954148244857788,
      "epoch": 1.4133698339809255,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004804451291450142,
      "loss": 5.6634,
      "mean_token_accuracy": 0.18039667904376983,
      "num_tokens": 29825796.0,
      "step": 16005
    },
    {
      "entropy": 5.893498659133911,
      "epoch": 1.4138113740727658,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048043228617881365,
      "loss": 5.628,
      "mean_token_accuracy": 0.17824345976114273,
      "num_tokens": 29834874.0,
      "step": 16010
    },
    {
      "entropy": 5.962541770935059,
      "epoch": 1.414252914164606,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004804194391883114,
      "loss": 5.6696,
      "mean_token_accuracy": 0.1731353923678398,
      "num_tokens": 29843884.0,
      "step": 16015
    },
    {
      "entropy": 6.001684713363647,
      "epoch": 1.4146944542564466,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048040658817375927,
      "loss": 5.7771,
      "mean_token_accuracy": 0.16440334171056747,
      "num_tokens": 29853180.0,
      "step": 16020
    },
    {
      "entropy": 5.829020977020264,
      "epoch": 1.4151359943482869,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048039373313540885,
      "loss": 5.6106,
      "mean_token_accuracy": 0.17595621198415756,
      "num_tokens": 29862926.0,
      "step": 16025
    },
    {
      "entropy": 5.815017127990723,
      "epoch": 1.4155775344401271,
      "grad_norm": 2.640625,
      "learning_rate": 0.000480380874073512,
      "loss": 5.64,
      "mean_token_accuracy": 0.18021250367164612,
      "num_tokens": 29871450.0,
      "step": 16030
    },
    {
      "entropy": 5.933086585998535,
      "epoch": 1.4160190745319676,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048036801098832067,
      "loss": 5.6139,
      "mean_token_accuracy": 0.18391968160867692,
      "num_tokens": 29880142.0,
      "step": 16035
    },
    {
      "entropy": 5.875817060470581,
      "epoch": 1.416460614623808,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004803551438800868,
      "loss": 5.6494,
      "mean_token_accuracy": 0.17990896254777908,
      "num_tokens": 29888668.0,
      "step": 16040
    },
    {
      "entropy": 5.9153553485870365,
      "epoch": 1.4169021547156482,
      "grad_norm": 1.5,
      "learning_rate": 0.00048034227274906235,
      "loss": 5.7822,
      "mean_token_accuracy": 0.1709391549229622,
      "num_tokens": 29898727.0,
      "step": 16045
    },
    {
      "entropy": 5.984368991851807,
      "epoch": 1.4173436948074885,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048032939759549964,
      "loss": 5.6847,
      "mean_token_accuracy": 0.16261391043663026,
      "num_tokens": 29909089.0,
      "step": 16050
    },
    {
      "entropy": 5.968800401687622,
      "epoch": 1.4177852348993287,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048031651841965085,
      "loss": 5.6443,
      "mean_token_accuracy": 0.17998259663581848,
      "num_tokens": 29918169.0,
      "step": 16055
    },
    {
      "entropy": 5.800526142120361,
      "epoch": 1.4182267749911692,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004803036352217682,
      "loss": 5.6713,
      "mean_token_accuracy": 0.18484727889299393,
      "num_tokens": 29928160.0,
      "step": 16060
    },
    {
      "entropy": 5.8660155773162845,
      "epoch": 1.4186683150830095,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048029074800210413,
      "loss": 5.6982,
      "mean_token_accuracy": 0.17152198404073715,
      "num_tokens": 29937695.0,
      "step": 16065
    },
    {
      "entropy": 6.063792657852173,
      "epoch": 1.4191098551748498,
      "grad_norm": 1.546875,
      "learning_rate": 0.000480277856760911,
      "loss": 5.7014,
      "mean_token_accuracy": 0.17309492230415344,
      "num_tokens": 29947380.0,
      "step": 16070
    },
    {
      "entropy": 5.98213300704956,
      "epoch": 1.4195513952666903,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048026496149844146,
      "loss": 5.7555,
      "mean_token_accuracy": 0.17132726013660432,
      "num_tokens": 29956313.0,
      "step": 16075
    },
    {
      "entropy": 5.847528028488159,
      "epoch": 1.4199929353585306,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048025206221494806,
      "loss": 5.602,
      "mean_token_accuracy": 0.1848934695124626,
      "num_tokens": 29965502.0,
      "step": 16080
    },
    {
      "entropy": 5.950541687011719,
      "epoch": 1.4204344754503708,
      "grad_norm": 1.5,
      "learning_rate": 0.0004802391589106835,
      "loss": 5.6161,
      "mean_token_accuracy": 0.17927108108997344,
      "num_tokens": 29974243.0,
      "step": 16085
    },
    {
      "entropy": 5.88813738822937,
      "epoch": 1.4208760155422113,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004802262515859005,
      "loss": 5.556,
      "mean_token_accuracy": 0.19431766122579575,
      "num_tokens": 29983966.0,
      "step": 16090
    },
    {
      "entropy": 5.817371034622193,
      "epoch": 1.4213175556340516,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004802133402408519,
      "loss": 5.6759,
      "mean_token_accuracy": 0.17723651379346847,
      "num_tokens": 29993365.0,
      "step": 16095
    },
    {
      "entropy": 5.9806629657745365,
      "epoch": 1.4217590957258919,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048020042487579076,
      "loss": 5.6949,
      "mean_token_accuracy": 0.17615126967430114,
      "num_tokens": 30001919.0,
      "step": 16100
    },
    {
      "entropy": 5.87896761894226,
      "epoch": 1.4222006358177324,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048018750549097,
      "loss": 5.6396,
      "mean_token_accuracy": 0.17911232858896256,
      "num_tokens": 30011353.0,
      "step": 16105
    },
    {
      "entropy": 5.951582050323486,
      "epoch": 1.4226421759095726,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048017458208664265,
      "loss": 5.7115,
      "mean_token_accuracy": 0.174793940782547,
      "num_tokens": 30019994.0,
      "step": 16110
    },
    {
      "entropy": 5.837914943695068,
      "epoch": 1.423083716001413,
      "grad_norm": 1.578125,
      "learning_rate": 0.000480161654663062,
      "loss": 5.5645,
      "mean_token_accuracy": 0.17787738740444184,
      "num_tokens": 30028934.0,
      "step": 16115
    },
    {
      "entropy": 5.885590124130249,
      "epoch": 1.4235252560932532,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004801487232204811,
      "loss": 5.5796,
      "mean_token_accuracy": 0.17927436381578446,
      "num_tokens": 30038069.0,
      "step": 16120
    },
    {
      "entropy": 5.930248022079468,
      "epoch": 1.4239667961850935,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004801357877591535,
      "loss": 5.7287,
      "mean_token_accuracy": 0.1704007238149643,
      "num_tokens": 30047307.0,
      "step": 16125
    },
    {
      "entropy": 5.911052417755127,
      "epoch": 1.424408336276934,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048012284827933245,
      "loss": 5.6629,
      "mean_token_accuracy": 0.17695229202508928,
      "num_tokens": 30057429.0,
      "step": 16130
    },
    {
      "entropy": 5.81870698928833,
      "epoch": 1.4248498763687742,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004801099047812715,
      "loss": 5.616,
      "mean_token_accuracy": 0.1796981394290924,
      "num_tokens": 30067327.0,
      "step": 16135
    },
    {
      "entropy": 5.955048322677612,
      "epoch": 1.4252914164606145,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004800969572652241,
      "loss": 5.7147,
      "mean_token_accuracy": 0.16946061998605727,
      "num_tokens": 30077103.0,
      "step": 16140
    },
    {
      "entropy": 5.970450019836425,
      "epoch": 1.425732956552455,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048008400573144394,
      "loss": 5.7951,
      "mean_token_accuracy": 0.16817922294139862,
      "num_tokens": 30087539.0,
      "step": 16145
    },
    {
      "entropy": 5.9440570831298825,
      "epoch": 1.4261744966442953,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048007105018018487,
      "loss": 5.6463,
      "mean_token_accuracy": 0.17323180586099624,
      "num_tokens": 30097153.0,
      "step": 16150
    },
    {
      "entropy": 5.89973669052124,
      "epoch": 1.4266160367361356,
      "grad_norm": 1.5,
      "learning_rate": 0.00048005809061170044,
      "loss": 5.7227,
      "mean_token_accuracy": 0.17570799291133882,
      "num_tokens": 30106291.0,
      "step": 16155
    },
    {
      "entropy": 5.898937749862671,
      "epoch": 1.427057576827976,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004800451270262447,
      "loss": 5.741,
      "mean_token_accuracy": 0.17294955402612686,
      "num_tokens": 30116441.0,
      "step": 16160
    },
    {
      "entropy": 5.918024158477783,
      "epoch": 1.4274991169198163,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004800321594240714,
      "loss": 5.6038,
      "mean_token_accuracy": 0.18953171670436858,
      "num_tokens": 30125040.0,
      "step": 16165
    },
    {
      "entropy": 5.931909704208374,
      "epoch": 1.4279406570116566,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004800191878054349,
      "loss": 5.6037,
      "mean_token_accuracy": 0.18564134687185288,
      "num_tokens": 30134284.0,
      "step": 16170
    },
    {
      "entropy": 5.944320249557495,
      "epoch": 1.4283821971034971,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048000621217058904,
      "loss": 5.5955,
      "mean_token_accuracy": 0.18816176950931549,
      "num_tokens": 30143679.0,
      "step": 16175
    },
    {
      "entropy": 5.791440343856811,
      "epoch": 1.4288237371953374,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047999323251978805,
      "loss": 5.5278,
      "mean_token_accuracy": 0.19174969494342803,
      "num_tokens": 30152011.0,
      "step": 16180
    },
    {
      "entropy": 5.795801115036011,
      "epoch": 1.4292652772871777,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00047998024885328625,
      "loss": 5.6368,
      "mean_token_accuracy": 0.18070774376392365,
      "num_tokens": 30161949.0,
      "step": 16185
    },
    {
      "entropy": 5.9326427459716795,
      "epoch": 1.429706817379018,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047996726117133795,
      "loss": 5.6571,
      "mean_token_accuracy": 0.18169366121292113,
      "num_tokens": 30172003.0,
      "step": 16190
    },
    {
      "entropy": 5.812878561019898,
      "epoch": 1.4301483574708582,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004799542694741975,
      "loss": 5.6171,
      "mean_token_accuracy": 0.18921156525611876,
      "num_tokens": 30181918.0,
      "step": 16195
    },
    {
      "entropy": 5.896188640594483,
      "epoch": 1.4305898975626987,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047994127376211953,
      "loss": 5.5991,
      "mean_token_accuracy": 0.18189290165901184,
      "num_tokens": 30191032.0,
      "step": 16200
    },
    {
      "entropy": 5.902913522720337,
      "epoch": 1.431031437654539,
      "grad_norm": 1.75,
      "learning_rate": 0.0004799282740353586,
      "loss": 5.6196,
      "mean_token_accuracy": 0.18093885034322738,
      "num_tokens": 30200276.0,
      "step": 16205
    },
    {
      "entropy": 5.958534383773804,
      "epoch": 1.4314729777463793,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047991527029416924,
      "loss": 5.6839,
      "mean_token_accuracy": 0.1810702860355377,
      "num_tokens": 30208574.0,
      "step": 16210
    },
    {
      "entropy": 5.849640226364135,
      "epoch": 1.4319145178382198,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004799022625388064,
      "loss": 5.6789,
      "mean_token_accuracy": 0.17807288616895675,
      "num_tokens": 30218912.0,
      "step": 16215
    },
    {
      "entropy": 5.930537128448487,
      "epoch": 1.43235605793006,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004798892507695247,
      "loss": 5.7564,
      "mean_token_accuracy": 0.17667900174856185,
      "num_tokens": 30228325.0,
      "step": 16220
    },
    {
      "entropy": 5.898623609542847,
      "epoch": 1.4327975980219003,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004798762349865791,
      "loss": 5.6356,
      "mean_token_accuracy": 0.18067460060119628,
      "num_tokens": 30237169.0,
      "step": 16225
    },
    {
      "entropy": 5.871915817260742,
      "epoch": 1.4332391381137408,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004798632151902246,
      "loss": 5.6181,
      "mean_token_accuracy": 0.18551321774721147,
      "num_tokens": 30246452.0,
      "step": 16230
    },
    {
      "entropy": 5.900274801254272,
      "epoch": 1.433680678205581,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047985019138071616,
      "loss": 5.6841,
      "mean_token_accuracy": 0.1756813108921051,
      "num_tokens": 30256055.0,
      "step": 16235
    },
    {
      "entropy": 5.837619638442993,
      "epoch": 1.4341222182974214,
      "grad_norm": 1.625,
      "learning_rate": 0.00047983716355830903,
      "loss": 5.5418,
      "mean_token_accuracy": 0.18319986164569854,
      "num_tokens": 30265764.0,
      "step": 16240
    },
    {
      "entropy": 5.906308364868164,
      "epoch": 1.4345637583892619,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047982413172325845,
      "loss": 5.7038,
      "mean_token_accuracy": 0.1829448848962784,
      "num_tokens": 30274490.0,
      "step": 16245
    },
    {
      "entropy": 5.883050489425659,
      "epoch": 1.4350052984811021,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047981109587581945,
      "loss": 5.6222,
      "mean_token_accuracy": 0.18116400837898256,
      "num_tokens": 30283330.0,
      "step": 16250
    },
    {
      "entropy": 5.955448722839355,
      "epoch": 1.4354468385729424,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004797980560162477,
      "loss": 5.7937,
      "mean_token_accuracy": 0.16233945786952972,
      "num_tokens": 30293542.0,
      "step": 16255
    },
    {
      "entropy": 5.942630577087402,
      "epoch": 1.4358883786647827,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004797850121447984,
      "loss": 5.7188,
      "mean_token_accuracy": 0.17953596264123917,
      "num_tokens": 30302122.0,
      "step": 16260
    },
    {
      "entropy": 5.887374353408814,
      "epoch": 1.436329918756623,
      "grad_norm": 1.765625,
      "learning_rate": 0.00047977196426172725,
      "loss": 5.7537,
      "mean_token_accuracy": 0.17787419855594636,
      "num_tokens": 30311490.0,
      "step": 16265
    },
    {
      "entropy": 5.841133308410645,
      "epoch": 1.4367714588484635,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047975891236728973,
      "loss": 5.616,
      "mean_token_accuracy": 0.18424348682165145,
      "num_tokens": 30321199.0,
      "step": 16270
    },
    {
      "entropy": 5.921429777145386,
      "epoch": 1.4372129989403037,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00047974585646174165,
      "loss": 5.6481,
      "mean_token_accuracy": 0.1792317137122154,
      "num_tokens": 30330114.0,
      "step": 16275
    },
    {
      "entropy": 5.934168815612793,
      "epoch": 1.437654539032144,
      "grad_norm": 1.5,
      "learning_rate": 0.00047973279654533866,
      "loss": 5.6558,
      "mean_token_accuracy": 0.18102662563323973,
      "num_tokens": 30339290.0,
      "step": 16280
    },
    {
      "entropy": 5.857350492477417,
      "epoch": 1.4380960791239845,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004797197326183365,
      "loss": 5.6203,
      "mean_token_accuracy": 0.18065737336874008,
      "num_tokens": 30349595.0,
      "step": 16285
    },
    {
      "entropy": 5.876824522018433,
      "epoch": 1.4385376192158248,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004797066646809914,
      "loss": 5.6748,
      "mean_token_accuracy": 0.17777315825223922,
      "num_tokens": 30358010.0,
      "step": 16290
    },
    {
      "entropy": 5.971915245056152,
      "epoch": 1.438979159307665,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047969359273355907,
      "loss": 5.7399,
      "mean_token_accuracy": 0.17310652881860733,
      "num_tokens": 30367408.0,
      "step": 16295
    },
    {
      "entropy": 5.930622196197509,
      "epoch": 1.4394206993995056,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047968051677629565,
      "loss": 5.7284,
      "mean_token_accuracy": 0.1734652489423752,
      "num_tokens": 30376475.0,
      "step": 16300
    },
    {
      "entropy": 5.913575315475464,
      "epoch": 1.4398622394913458,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047966743680945734,
      "loss": 5.6413,
      "mean_token_accuracy": 0.17763058543205262,
      "num_tokens": 30385361.0,
      "step": 16305
    },
    {
      "entropy": 5.93912091255188,
      "epoch": 1.4403037795831861,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004796543528333004,
      "loss": 5.6657,
      "mean_token_accuracy": 0.1784609168767929,
      "num_tokens": 30393594.0,
      "step": 16310
    },
    {
      "entropy": 5.888725996017456,
      "epoch": 1.4407453196750266,
      "grad_norm": 1.953125,
      "learning_rate": 0.000479641264848081,
      "loss": 5.6577,
      "mean_token_accuracy": 0.17895561158657075,
      "num_tokens": 30403444.0,
      "step": 16315
    },
    {
      "entropy": 5.840634536743164,
      "epoch": 1.441186859766867,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004796281728540556,
      "loss": 5.5605,
      "mean_token_accuracy": 0.1789017289876938,
      "num_tokens": 30412485.0,
      "step": 16320
    },
    {
      "entropy": 5.915186882019043,
      "epoch": 1.4416283998587072,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047961507685148077,
      "loss": 5.6153,
      "mean_token_accuracy": 0.18150109350681304,
      "num_tokens": 30421104.0,
      "step": 16325
    },
    {
      "entropy": 5.8485533714294435,
      "epoch": 1.4420699399505474,
      "grad_norm": 1.71875,
      "learning_rate": 0.00047960197684061287,
      "loss": 5.6214,
      "mean_token_accuracy": 0.18087481409311296,
      "num_tokens": 30430061.0,
      "step": 16330
    },
    {
      "entropy": 5.885530281066894,
      "epoch": 1.4425114800423877,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004795888728217086,
      "loss": 5.5461,
      "mean_token_accuracy": 0.1779729411005974,
      "num_tokens": 30437645.0,
      "step": 16335
    },
    {
      "entropy": 5.802228784561157,
      "epoch": 1.4429530201342282,
      "grad_norm": 1.875,
      "learning_rate": 0.00047957576479502466,
      "loss": 5.5428,
      "mean_token_accuracy": 0.19268881976604463,
      "num_tokens": 30447805.0,
      "step": 16340
    },
    {
      "entropy": 5.965684127807617,
      "epoch": 1.4433945602260685,
      "grad_norm": 1.75,
      "learning_rate": 0.0004795626527608179,
      "loss": 5.6931,
      "mean_token_accuracy": 0.18334885835647582,
      "num_tokens": 30457524.0,
      "step": 16345
    },
    {
      "entropy": 6.017973899841309,
      "epoch": 1.4438361003179088,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004795495367193451,
      "loss": 5.6682,
      "mean_token_accuracy": 0.18190035820007325,
      "num_tokens": 30467082.0,
      "step": 16350
    },
    {
      "entropy": 5.963843059539795,
      "epoch": 1.4442776404097493,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047953641667086315,
      "loss": 5.7233,
      "mean_token_accuracy": 0.17948476672172547,
      "num_tokens": 30476251.0,
      "step": 16355
    },
    {
      "entropy": 5.840140962600708,
      "epoch": 1.4447191805015895,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047952329261562914,
      "loss": 5.6143,
      "mean_token_accuracy": 0.18024616092443466,
      "num_tokens": 30485997.0,
      "step": 16360
    },
    {
      "entropy": 5.958113718032837,
      "epoch": 1.4451607205934298,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047951016455390023,
      "loss": 5.7195,
      "mean_token_accuracy": 0.17911118417978286,
      "num_tokens": 30494886.0,
      "step": 16365
    },
    {
      "entropy": 5.91194634437561,
      "epoch": 1.4456022606852703,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004794970324859334,
      "loss": 5.7314,
      "mean_token_accuracy": 0.16646508872509003,
      "num_tokens": 30504194.0,
      "step": 16370
    },
    {
      "entropy": 5.900558948516846,
      "epoch": 1.4460438007771106,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047948389641198615,
      "loss": 5.61,
      "mean_token_accuracy": 0.18098676949739456,
      "num_tokens": 30513077.0,
      "step": 16375
    },
    {
      "entropy": 5.912801074981689,
      "epoch": 1.4464853408689509,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047947075633231553,
      "loss": 5.6833,
      "mean_token_accuracy": 0.17289229184389115,
      "num_tokens": 30521999.0,
      "step": 16380
    },
    {
      "entropy": 5.946346950531006,
      "epoch": 1.4469268809607914,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004794576122471791,
      "loss": 5.7334,
      "mean_token_accuracy": 0.17191818803548814,
      "num_tokens": 30530138.0,
      "step": 16385
    },
    {
      "entropy": 5.839674854278565,
      "epoch": 1.4473684210526316,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004794444641568344,
      "loss": 5.5818,
      "mean_token_accuracy": 0.18059268742799758,
      "num_tokens": 30538850.0,
      "step": 16390
    },
    {
      "entropy": 5.874353456497192,
      "epoch": 1.447809961144472,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004794313120615388,
      "loss": 5.6777,
      "mean_token_accuracy": 0.17492283284664153,
      "num_tokens": 30547593.0,
      "step": 16395
    },
    {
      "entropy": 5.9027087688446045,
      "epoch": 1.4482515012363122,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004794181559615502,
      "loss": 5.6233,
      "mean_token_accuracy": 0.18063668459653853,
      "num_tokens": 30557154.0,
      "step": 16400
    },
    {
      "entropy": 5.963204669952392,
      "epoch": 1.4486930413281525,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047940499585712615,
      "loss": 5.6342,
      "mean_token_accuracy": 0.1795578420162201,
      "num_tokens": 30566379.0,
      "step": 16405
    },
    {
      "entropy": 5.823266839981079,
      "epoch": 1.449134581419993,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047939183174852446,
      "loss": 5.5731,
      "mean_token_accuracy": 0.19004496335983276,
      "num_tokens": 30575146.0,
      "step": 16410
    },
    {
      "entropy": 5.826879787445068,
      "epoch": 1.4495761215118332,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047937866363600295,
      "loss": 5.6577,
      "mean_token_accuracy": 0.17941279113292694,
      "num_tokens": 30584444.0,
      "step": 16415
    },
    {
      "entropy": 5.9460633277893065,
      "epoch": 1.4500176616036735,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004793654915198198,
      "loss": 5.6524,
      "mean_token_accuracy": 0.18458783030509948,
      "num_tokens": 30593779.0,
      "step": 16420
    },
    {
      "entropy": 5.98534665107727,
      "epoch": 1.450459201695514,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004793523154002328,
      "loss": 5.6665,
      "mean_token_accuracy": 0.18153499364852904,
      "num_tokens": 30603762.0,
      "step": 16425
    },
    {
      "entropy": 5.960584497451782,
      "epoch": 1.4509007417873543,
      "grad_norm": 2.375,
      "learning_rate": 0.00047933913527750015,
      "loss": 5.644,
      "mean_token_accuracy": 0.18152158856391906,
      "num_tokens": 30612907.0,
      "step": 16430
    },
    {
      "entropy": 5.845513200759887,
      "epoch": 1.4513422818791946,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004793259511518801,
      "loss": 5.5978,
      "mean_token_accuracy": 0.1895533472299576,
      "num_tokens": 30621537.0,
      "step": 16435
    },
    {
      "entropy": 5.839008378982544,
      "epoch": 1.451783821971035,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004793127630236309,
      "loss": 5.724,
      "mean_token_accuracy": 0.16866641938686372,
      "num_tokens": 30632143.0,
      "step": 16440
    },
    {
      "entropy": 5.996024227142334,
      "epoch": 1.4522253620628753,
      "grad_norm": 1.9375,
      "learning_rate": 0.00047929957089301085,
      "loss": 5.6838,
      "mean_token_accuracy": 0.17136377692222596,
      "num_tokens": 30641145.0,
      "step": 16445
    },
    {
      "entropy": 5.944555759429932,
      "epoch": 1.4526669021547156,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047928637476027836,
      "loss": 5.6634,
      "mean_token_accuracy": 0.17819404900074004,
      "num_tokens": 30650036.0,
      "step": 16450
    },
    {
      "entropy": 5.940585994720459,
      "epoch": 1.453108442246556,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047927317462569206,
      "loss": 5.6421,
      "mean_token_accuracy": 0.18208981305360794,
      "num_tokens": 30659552.0,
      "step": 16455
    },
    {
      "entropy": 5.937906980514526,
      "epoch": 1.4535499823383964,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004792599704895103,
      "loss": 5.6245,
      "mean_token_accuracy": 0.17942069619894027,
      "num_tokens": 30668652.0,
      "step": 16460
    },
    {
      "entropy": 5.931765985488892,
      "epoch": 1.4539915224302367,
      "grad_norm": 1.53125,
      "learning_rate": 0.000479246762351992,
      "loss": 5.709,
      "mean_token_accuracy": 0.17465552389621736,
      "num_tokens": 30678613.0,
      "step": 16465
    },
    {
      "entropy": 5.898845767974853,
      "epoch": 1.454433062522077,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047923355021339576,
      "loss": 5.6332,
      "mean_token_accuracy": 0.1851435735821724,
      "num_tokens": 30688259.0,
      "step": 16470
    },
    {
      "entropy": 5.910692262649536,
      "epoch": 1.4548746026139172,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004792203340739805,
      "loss": 5.6795,
      "mean_token_accuracy": 0.17299126386642455,
      "num_tokens": 30697309.0,
      "step": 16475
    },
    {
      "entropy": 6.005543899536133,
      "epoch": 1.4553161427057577,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004792071139340049,
      "loss": 5.7412,
      "mean_token_accuracy": 0.17424341291189194,
      "num_tokens": 30707278.0,
      "step": 16480
    },
    {
      "entropy": 5.890324592590332,
      "epoch": 1.455757682797598,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004791938897937281,
      "loss": 5.6301,
      "mean_token_accuracy": 0.17966896891593934,
      "num_tokens": 30716840.0,
      "step": 16485
    },
    {
      "entropy": 5.906936311721802,
      "epoch": 1.4561992228894383,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047918066165340926,
      "loss": 5.6993,
      "mean_token_accuracy": 0.17394667565822602,
      "num_tokens": 30727120.0,
      "step": 16490
    },
    {
      "entropy": 5.862757205963135,
      "epoch": 1.4566407629812788,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004791674295133072,
      "loss": 5.6201,
      "mean_token_accuracy": 0.18076477199792862,
      "num_tokens": 30735543.0,
      "step": 16495
    },
    {
      "entropy": 5.909570264816284,
      "epoch": 1.457082303073119,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047915419337368146,
      "loss": 5.6688,
      "mean_token_accuracy": 0.17290441542863846,
      "num_tokens": 30744971.0,
      "step": 16500
    },
    {
      "entropy": 5.899969005584717,
      "epoch": 1.4575238431649593,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047914095323479115,
      "loss": 5.5553,
      "mean_token_accuracy": 0.18456466794013976,
      "num_tokens": 30753444.0,
      "step": 16505
    },
    {
      "entropy": 5.90459566116333,
      "epoch": 1.4579653832567998,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004791277090968956,
      "loss": 5.6457,
      "mean_token_accuracy": 0.18387412428855895,
      "num_tokens": 30761469.0,
      "step": 16510
    },
    {
      "entropy": 5.803884744644165,
      "epoch": 1.45840692334864,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004791144609602544,
      "loss": 5.5821,
      "mean_token_accuracy": 0.18448227494955063,
      "num_tokens": 30770773.0,
      "step": 16515
    },
    {
      "entropy": 5.833355569839478,
      "epoch": 1.4588484634404804,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047910120882512696,
      "loss": 5.7074,
      "mean_token_accuracy": 0.17578627318143844,
      "num_tokens": 30779624.0,
      "step": 16520
    },
    {
      "entropy": 5.888250350952148,
      "epoch": 1.4592900035323209,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004790879526917729,
      "loss": 5.6114,
      "mean_token_accuracy": 0.17613171190023422,
      "num_tokens": 30788671.0,
      "step": 16525
    },
    {
      "entropy": 5.9667520999908445,
      "epoch": 1.4597315436241611,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004790746925604519,
      "loss": 5.6877,
      "mean_token_accuracy": 0.17987888306379318,
      "num_tokens": 30797470.0,
      "step": 16530
    },
    {
      "entropy": 5.945762729644775,
      "epoch": 1.4601730837160014,
      "grad_norm": 2.125,
      "learning_rate": 0.00047906142843142383,
      "loss": 5.6704,
      "mean_token_accuracy": 0.18114448338747025,
      "num_tokens": 30807209.0,
      "step": 16535
    },
    {
      "entropy": 5.963796329498291,
      "epoch": 1.4606146238078417,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047904816030494834,
      "loss": 5.7003,
      "mean_token_accuracy": 0.17103892862796782,
      "num_tokens": 30816156.0,
      "step": 16540
    },
    {
      "entropy": 5.884915161132812,
      "epoch": 1.461056163899682,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004790348881812855,
      "loss": 5.5794,
      "mean_token_accuracy": 0.1890389621257782,
      "num_tokens": 30825439.0,
      "step": 16545
    },
    {
      "entropy": 5.876064968109131,
      "epoch": 1.4614977039915225,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047902161206069514,
      "loss": 5.6174,
      "mean_token_accuracy": 0.1850159779191017,
      "num_tokens": 30834953.0,
      "step": 16550
    },
    {
      "entropy": 5.897009038925171,
      "epoch": 1.4619392440833627,
      "grad_norm": 2.25,
      "learning_rate": 0.00047900833194343764,
      "loss": 5.6723,
      "mean_token_accuracy": 0.17717111259698867,
      "num_tokens": 30843534.0,
      "step": 16555
    },
    {
      "entropy": 5.881877183914185,
      "epoch": 1.462380784175203,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004789950478297727,
      "loss": 5.5759,
      "mean_token_accuracy": 0.18852667659521102,
      "num_tokens": 30851869.0,
      "step": 16560
    },
    {
      "entropy": 5.892990970611573,
      "epoch": 1.4628223242670435,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047898175971996095,
      "loss": 5.7307,
      "mean_token_accuracy": 0.1760902598500252,
      "num_tokens": 30862262.0,
      "step": 16565
    },
    {
      "entropy": 5.911265754699707,
      "epoch": 1.4632638643588838,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00047896846761426255,
      "loss": 5.7056,
      "mean_token_accuracy": 0.17436186522245406,
      "num_tokens": 30872223.0,
      "step": 16570
    },
    {
      "entropy": 5.959250640869141,
      "epoch": 1.463705404450724,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004789551715129378,
      "loss": 5.6365,
      "mean_token_accuracy": 0.1781175673007965,
      "num_tokens": 30881834.0,
      "step": 16575
    },
    {
      "entropy": 5.926247024536133,
      "epoch": 1.4641469445425646,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004789418714162472,
      "loss": 5.71,
      "mean_token_accuracy": 0.17450368255376816,
      "num_tokens": 30893414.0,
      "step": 16580
    },
    {
      "entropy": 5.956064033508301,
      "epoch": 1.4645884846344048,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004789285673244514,
      "loss": 5.7534,
      "mean_token_accuracy": 0.16493886560201645,
      "num_tokens": 30902345.0,
      "step": 16585
    },
    {
      "entropy": 5.774648904800415,
      "epoch": 1.465030024726245,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004789152592378109,
      "loss": 5.4552,
      "mean_token_accuracy": 0.20286068320274353,
      "num_tokens": 30911619.0,
      "step": 16590
    },
    {
      "entropy": 5.917823076248169,
      "epoch": 1.4654715648180856,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047890194715658645,
      "loss": 5.6709,
      "mean_token_accuracy": 0.17277761399745942,
      "num_tokens": 30921515.0,
      "step": 16595
    },
    {
      "entropy": 5.927871894836426,
      "epoch": 1.4659131049099259,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004788886310810389,
      "loss": 5.6317,
      "mean_token_accuracy": 0.19329753369092942,
      "num_tokens": 30930807.0,
      "step": 16600
    },
    {
      "entropy": 5.975428104400635,
      "epoch": 1.4663546450017662,
      "grad_norm": 1.984375,
      "learning_rate": 0.00047887531101142893,
      "loss": 5.7094,
      "mean_token_accuracy": 0.1673653617501259,
      "num_tokens": 30939266.0,
      "step": 16605
    },
    {
      "entropy": 5.934521770477295,
      "epoch": 1.4667961850936064,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047886198694801764,
      "loss": 5.6047,
      "mean_token_accuracy": 0.1768937736749649,
      "num_tokens": 30948119.0,
      "step": 16610
    },
    {
      "entropy": 5.856646108627319,
      "epoch": 1.4672377251854467,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047884865889106593,
      "loss": 5.5547,
      "mean_token_accuracy": 0.18557057678699493,
      "num_tokens": 30957005.0,
      "step": 16615
    },
    {
      "entropy": 5.766826629638672,
      "epoch": 1.4676792652772872,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047883532684083493,
      "loss": 5.5883,
      "mean_token_accuracy": 0.19055165797472,
      "num_tokens": 30967144.0,
      "step": 16620
    },
    {
      "entropy": 5.851292419433594,
      "epoch": 1.4681208053691275,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004788219907975858,
      "loss": 5.5761,
      "mean_token_accuracy": 0.1818339318037033,
      "num_tokens": 30975315.0,
      "step": 16625
    },
    {
      "entropy": 5.947195053100586,
      "epoch": 1.4685623454609678,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004788086507615798,
      "loss": 5.7312,
      "mean_token_accuracy": 0.17409353852272033,
      "num_tokens": 30984705.0,
      "step": 16630
    },
    {
      "entropy": 5.924814224243164,
      "epoch": 1.4690038855528083,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004787953067330782,
      "loss": 5.7308,
      "mean_token_accuracy": 0.17230313271284103,
      "num_tokens": 30994936.0,
      "step": 16635
    },
    {
      "entropy": 5.828255605697632,
      "epoch": 1.4694454256446485,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004787819587123425,
      "loss": 5.6467,
      "mean_token_accuracy": 0.17073431015014648,
      "num_tokens": 31004011.0,
      "step": 16640
    },
    {
      "entropy": 6.0143214702606205,
      "epoch": 1.4698869657364888,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004787686066996341,
      "loss": 5.6846,
      "mean_token_accuracy": 0.17422959804534913,
      "num_tokens": 31013852.0,
      "step": 16645
    },
    {
      "entropy": 5.953846216201782,
      "epoch": 1.4703285058283293,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047875525069521457,
      "loss": 5.6297,
      "mean_token_accuracy": 0.1840740293264389,
      "num_tokens": 31023456.0,
      "step": 16650
    },
    {
      "entropy": 5.881595754623413,
      "epoch": 1.4707700459201696,
      "grad_norm": 1.875,
      "learning_rate": 0.0004787418906993456,
      "loss": 5.7084,
      "mean_token_accuracy": 0.1672755241394043,
      "num_tokens": 31032828.0,
      "step": 16655
    },
    {
      "entropy": 5.861470651626587,
      "epoch": 1.4712115860120099,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047872852671228884,
      "loss": 5.6437,
      "mean_token_accuracy": 0.182302625477314,
      "num_tokens": 31043017.0,
      "step": 16660
    },
    {
      "entropy": 5.840350246429443,
      "epoch": 1.4716531261038504,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047871515873430614,
      "loss": 5.502,
      "mean_token_accuracy": 0.19103837460279466,
      "num_tokens": 31051491.0,
      "step": 16665
    },
    {
      "entropy": 5.849490690231323,
      "epoch": 1.4720946661956906,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004787017867656593,
      "loss": 5.5447,
      "mean_token_accuracy": 0.19599919021129608,
      "num_tokens": 31060375.0,
      "step": 16670
    },
    {
      "entropy": 5.937414598464966,
      "epoch": 1.472536206287531,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004786884108066103,
      "loss": 5.7694,
      "mean_token_accuracy": 0.17301855832338334,
      "num_tokens": 31071410.0,
      "step": 16675
    },
    {
      "entropy": 5.946331405639649,
      "epoch": 1.4729777463793712,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004786750308574212,
      "loss": 5.4983,
      "mean_token_accuracy": 0.19388947635889053,
      "num_tokens": 31080123.0,
      "step": 16680
    },
    {
      "entropy": 5.907465648651123,
      "epoch": 1.4734192864712115,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004786616469183541,
      "loss": 5.6658,
      "mean_token_accuracy": 0.17594998925924302,
      "num_tokens": 31089870.0,
      "step": 16685
    },
    {
      "entropy": 5.8927075386047365,
      "epoch": 1.473860826563052,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004786482589896711,
      "loss": 5.6937,
      "mean_token_accuracy": 0.17556374669075012,
      "num_tokens": 31098682.0,
      "step": 16690
    },
    {
      "entropy": 5.932112407684326,
      "epoch": 1.4743023666548922,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004786348670716346,
      "loss": 5.679,
      "mean_token_accuracy": 0.1720130503177643,
      "num_tokens": 31107785.0,
      "step": 16695
    },
    {
      "entropy": 5.830134344100952,
      "epoch": 1.4747439067467325,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004786214711645068,
      "loss": 5.516,
      "mean_token_accuracy": 0.18768954873085023,
      "num_tokens": 31116853.0,
      "step": 16700
    },
    {
      "entropy": 5.876828622817993,
      "epoch": 1.475185446838573,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004786080712685503,
      "loss": 5.6195,
      "mean_token_accuracy": 0.1801268830895424,
      "num_tokens": 31126260.0,
      "step": 16705
    },
    {
      "entropy": 5.9241289615631105,
      "epoch": 1.4756269869304133,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00047859466738402743,
      "loss": 5.6896,
      "mean_token_accuracy": 0.17996252477169036,
      "num_tokens": 31135047.0,
      "step": 16710
    },
    {
      "entropy": 5.891602373123169,
      "epoch": 1.4760685270222536,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047858125951120073,
      "loss": 5.6254,
      "mean_token_accuracy": 0.18229165375232698,
      "num_tokens": 31144177.0,
      "step": 16715
    },
    {
      "entropy": 5.876333379745484,
      "epoch": 1.476510067114094,
      "grad_norm": 1.515625,
      "learning_rate": 0.000478567847650333,
      "loss": 5.6993,
      "mean_token_accuracy": 0.17407555729150773,
      "num_tokens": 31153586.0,
      "step": 16720
    },
    {
      "entropy": 5.953978395462036,
      "epoch": 1.4769516072059343,
      "grad_norm": 2.1875,
      "learning_rate": 0.000478554431801687,
      "loss": 5.8214,
      "mean_token_accuracy": 0.17402286678552628,
      "num_tokens": 31163252.0,
      "step": 16725
    },
    {
      "entropy": 5.872389936447144,
      "epoch": 1.4773931472977746,
      "grad_norm": 1.75,
      "learning_rate": 0.0004785410119655253,
      "loss": 5.5888,
      "mean_token_accuracy": 0.18309753984212876,
      "num_tokens": 31171947.0,
      "step": 16730
    },
    {
      "entropy": 5.893641710281372,
      "epoch": 1.477834687389615,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047852758814211104,
      "loss": 5.6217,
      "mean_token_accuracy": 0.18624928146600722,
      "num_tokens": 31180502.0,
      "step": 16735
    },
    {
      "entropy": 5.9891218662261965,
      "epoch": 1.4782762274814554,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047851416033170704,
      "loss": 5.7213,
      "mean_token_accuracy": 0.1674528643488884,
      "num_tokens": 31190744.0,
      "step": 16740
    },
    {
      "entropy": 5.954369688034058,
      "epoch": 1.4787177675732956,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004785007285345764,
      "loss": 5.7641,
      "mean_token_accuracy": 0.17248492389917375,
      "num_tokens": 31200869.0,
      "step": 16745
    },
    {
      "entropy": 5.9193824291229244,
      "epoch": 1.479159307665136,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004784872927509822,
      "loss": 5.6502,
      "mean_token_accuracy": 0.184397292137146,
      "num_tokens": 31210046.0,
      "step": 16750
    },
    {
      "entropy": 5.934375667572022,
      "epoch": 1.4796008477569762,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004784738529811877,
      "loss": 5.5808,
      "mean_token_accuracy": 0.1851020336151123,
      "num_tokens": 31219140.0,
      "step": 16755
    },
    {
      "entropy": 5.9033092021942135,
      "epoch": 1.4800423878488167,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047846040922545604,
      "loss": 5.5848,
      "mean_token_accuracy": 0.18748055398464203,
      "num_tokens": 31227993.0,
      "step": 16760
    },
    {
      "entropy": 5.870349025726318,
      "epoch": 1.480483927940657,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047844696148405074,
      "loss": 5.6414,
      "mean_token_accuracy": 0.181888385117054,
      "num_tokens": 31236959.0,
      "step": 16765
    },
    {
      "entropy": 5.874393796920776,
      "epoch": 1.4809254680324972,
      "grad_norm": 1.5,
      "learning_rate": 0.00047843350975723523,
      "loss": 5.541,
      "mean_token_accuracy": 0.18526301831007003,
      "num_tokens": 31246373.0,
      "step": 16770
    },
    {
      "entropy": 5.817453050613404,
      "epoch": 1.4813670081243377,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004784200540452729,
      "loss": 5.6718,
      "mean_token_accuracy": 0.17942648380994797,
      "num_tokens": 31255556.0,
      "step": 16775
    },
    {
      "entropy": 5.843888902664185,
      "epoch": 1.481808548216178,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004784065943484274,
      "loss": 5.5862,
      "mean_token_accuracy": 0.1893172010779381,
      "num_tokens": 31264824.0,
      "step": 16780
    },
    {
      "entropy": 5.798686695098877,
      "epoch": 1.4822500883080183,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00047839313066696246,
      "loss": 5.5572,
      "mean_token_accuracy": 0.19716596305370332,
      "num_tokens": 31273946.0,
      "step": 16785
    },
    {
      "entropy": 5.872301292419434,
      "epoch": 1.4826916283998588,
      "grad_norm": 1.671875,
      "learning_rate": 0.00047837966300114164,
      "loss": 5.6209,
      "mean_token_accuracy": 0.1806508183479309,
      "num_tokens": 31283492.0,
      "step": 16790
    },
    {
      "entropy": 6.007496643066406,
      "epoch": 1.483133168491699,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047836619135122906,
      "loss": 5.671,
      "mean_token_accuracy": 0.1754691481590271,
      "num_tokens": 31292250.0,
      "step": 16795
    },
    {
      "entropy": 5.973191118240356,
      "epoch": 1.4835747085835393,
      "grad_norm": 3.265625,
      "learning_rate": 0.00047835271571748834,
      "loss": 5.7806,
      "mean_token_accuracy": 0.1637235090136528,
      "num_tokens": 31302868.0,
      "step": 16800
    },
    {
      "entropy": 5.872155094146729,
      "epoch": 1.4840162486753798,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004783392361001836,
      "loss": 5.5686,
      "mean_token_accuracy": 0.18489944338798522,
      "num_tokens": 31311918.0,
      "step": 16805
    },
    {
      "entropy": 5.832064771652222,
      "epoch": 1.4844577887672201,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004783257524995788,
      "loss": 5.5747,
      "mean_token_accuracy": 0.19137714356184005,
      "num_tokens": 31321339.0,
      "step": 16810
    },
    {
      "entropy": 5.876286268234253,
      "epoch": 1.4848993288590604,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004783122649159383,
      "loss": 5.5618,
      "mean_token_accuracy": 0.17734655737876892,
      "num_tokens": 31330498.0,
      "step": 16815
    },
    {
      "entropy": 5.845732879638672,
      "epoch": 1.4853408689509007,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000478298773349526,
      "loss": 5.596,
      "mean_token_accuracy": 0.18656350374221803,
      "num_tokens": 31339676.0,
      "step": 16820
    },
    {
      "entropy": 5.981503677368164,
      "epoch": 1.485782409042741,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004782852778006064,
      "loss": 5.7717,
      "mean_token_accuracy": 0.1683503806591034,
      "num_tokens": 31350072.0,
      "step": 16825
    },
    {
      "entropy": 5.975645875930786,
      "epoch": 1.4862239491345814,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047827177826944396,
      "loss": 5.6176,
      "mean_token_accuracy": 0.18721563667058944,
      "num_tokens": 31359992.0,
      "step": 16830
    },
    {
      "entropy": 5.886351823806763,
      "epoch": 1.4866654892264217,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047825827475630285,
      "loss": 5.6382,
      "mean_token_accuracy": 0.18631134033203126,
      "num_tokens": 31369434.0,
      "step": 16835
    },
    {
      "entropy": 5.858706378936768,
      "epoch": 1.487107029318262,
      "grad_norm": 1.625,
      "learning_rate": 0.00047824476726144775,
      "loss": 5.6419,
      "mean_token_accuracy": 0.17988727390766143,
      "num_tokens": 31378773.0,
      "step": 16840
    },
    {
      "entropy": 5.935371828079224,
      "epoch": 1.4875485694101025,
      "grad_norm": 2.75,
      "learning_rate": 0.00047823125578514325,
      "loss": 5.666,
      "mean_token_accuracy": 0.1763545423746109,
      "num_tokens": 31388219.0,
      "step": 16845
    },
    {
      "entropy": 5.8763877868652346,
      "epoch": 1.4879901095019428,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047821774032765407,
      "loss": 5.5839,
      "mean_token_accuracy": 0.1845327541232109,
      "num_tokens": 31397216.0,
      "step": 16850
    },
    {
      "entropy": 5.813215541839599,
      "epoch": 1.488431649593783,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004782042208892449,
      "loss": 5.5618,
      "mean_token_accuracy": 0.19184888303279876,
      "num_tokens": 31406861.0,
      "step": 16855
    },
    {
      "entropy": 5.88200044631958,
      "epoch": 1.4888731896856235,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047819069747018065,
      "loss": 5.6552,
      "mean_token_accuracy": 0.17956829369068145,
      "num_tokens": 31416387.0,
      "step": 16860
    },
    {
      "entropy": 5.938643550872802,
      "epoch": 1.4893147297774638,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004781771700707262,
      "loss": 5.558,
      "mean_token_accuracy": 0.1852492496371269,
      "num_tokens": 31425056.0,
      "step": 16865
    },
    {
      "entropy": 5.936826086044311,
      "epoch": 1.489756269869304,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004781636386911465,
      "loss": 5.7506,
      "mean_token_accuracy": 0.17144275158643724,
      "num_tokens": 31435746.0,
      "step": 16870
    },
    {
      "entropy": 5.950125789642334,
      "epoch": 1.4901978099611446,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004781501033317067,
      "loss": 5.7466,
      "mean_token_accuracy": 0.17425078749656678,
      "num_tokens": 31444808.0,
      "step": 16875
    },
    {
      "entropy": 5.978330993652344,
      "epoch": 1.4906393500529849,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004781365639926718,
      "loss": 5.6572,
      "mean_token_accuracy": 0.18090200573205947,
      "num_tokens": 31454205.0,
      "step": 16880
    },
    {
      "entropy": 5.9266955852508545,
      "epoch": 1.4910808901448251,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004781230206743073,
      "loss": 5.6989,
      "mean_token_accuracy": 0.17569437474012375,
      "num_tokens": 31463446.0,
      "step": 16885
    },
    {
      "entropy": 5.916769790649414,
      "epoch": 1.4915224302366654,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004781094733768783,
      "loss": 5.6492,
      "mean_token_accuracy": 0.18272793740034105,
      "num_tokens": 31473168.0,
      "step": 16890
    },
    {
      "entropy": 5.871441793441773,
      "epoch": 1.4919639703285057,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047809592210065017,
      "loss": 5.5576,
      "mean_token_accuracy": 0.1865987330675125,
      "num_tokens": 31482899.0,
      "step": 16895
    },
    {
      "entropy": 5.860480546951294,
      "epoch": 1.4924055104203462,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047808236684588847,
      "loss": 5.5432,
      "mean_token_accuracy": 0.17813766300678252,
      "num_tokens": 31492503.0,
      "step": 16900
    },
    {
      "entropy": 5.777821969985962,
      "epoch": 1.4928470505121865,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004780688076128587,
      "loss": 5.5258,
      "mean_token_accuracy": 0.19518077075481416,
      "num_tokens": 31501520.0,
      "step": 16905
    },
    {
      "entropy": 5.861655569076538,
      "epoch": 1.4932885906040267,
      "grad_norm": 3.796875,
      "learning_rate": 0.00047805524440182654,
      "loss": 5.6768,
      "mean_token_accuracy": 0.18397715240716933,
      "num_tokens": 31510821.0,
      "step": 16910
    },
    {
      "entropy": 5.9625485897064205,
      "epoch": 1.4937301306958672,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047804167721305756,
      "loss": 5.6475,
      "mean_token_accuracy": 0.1826513573527336,
      "num_tokens": 31520472.0,
      "step": 16915
    },
    {
      "entropy": 5.887255096435547,
      "epoch": 1.4941716707877075,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047802810604681757,
      "loss": 5.6628,
      "mean_token_accuracy": 0.18334442973136902,
      "num_tokens": 31529129.0,
      "step": 16920
    },
    {
      "entropy": 5.792483568191528,
      "epoch": 1.4946132108795478,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047801453090337256,
      "loss": 5.5903,
      "mean_token_accuracy": 0.18630072921514512,
      "num_tokens": 31537897.0,
      "step": 16925
    },
    {
      "entropy": 5.876600933074951,
      "epoch": 1.4950547509713883,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004780009517829882,
      "loss": 5.6234,
      "mean_token_accuracy": 0.1787065789103508,
      "num_tokens": 31547656.0,
      "step": 16930
    },
    {
      "entropy": 5.893722438812256,
      "epoch": 1.4954962910632286,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004779873686859308,
      "loss": 5.5997,
      "mean_token_accuracy": 0.183150252699852,
      "num_tokens": 31557303.0,
      "step": 16935
    },
    {
      "entropy": 5.919459199905395,
      "epoch": 1.4959378311550688,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047797378161246613,
      "loss": 5.639,
      "mean_token_accuracy": 0.1785338595509529,
      "num_tokens": 31565749.0,
      "step": 16940
    },
    {
      "entropy": 5.898062419891358,
      "epoch": 1.4963793712469093,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047796019056286067,
      "loss": 5.679,
      "mean_token_accuracy": 0.1829585000872612,
      "num_tokens": 31575286.0,
      "step": 16945
    },
    {
      "entropy": 5.965204095840454,
      "epoch": 1.4968209113387496,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004779465955373804,
      "loss": 5.6152,
      "mean_token_accuracy": 0.1794959396123886,
      "num_tokens": 31583723.0,
      "step": 16950
    },
    {
      "entropy": 5.787666893005371,
      "epoch": 1.49726245143059,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004779329965362918,
      "loss": 5.468,
      "mean_token_accuracy": 0.20068585872650146,
      "num_tokens": 31593171.0,
      "step": 16955
    },
    {
      "entropy": 5.870197772979736,
      "epoch": 1.4977039915224302,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00047791939355986123,
      "loss": 5.6736,
      "mean_token_accuracy": 0.17519170492887498,
      "num_tokens": 31603958.0,
      "step": 16960
    },
    {
      "entropy": 5.862027740478515,
      "epoch": 1.4981455316142704,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004779057866083551,
      "loss": 5.6438,
      "mean_token_accuracy": 0.1784205198287964,
      "num_tokens": 31612460.0,
      "step": 16965
    },
    {
      "entropy": 5.973839044570923,
      "epoch": 1.498587071706111,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047789217568204004,
      "loss": 5.702,
      "mean_token_accuracy": 0.17805294543504716,
      "num_tokens": 31622114.0,
      "step": 16970
    },
    {
      "entropy": 5.976660919189453,
      "epoch": 1.4990286117979512,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047787856078118266,
      "loss": 5.6406,
      "mean_token_accuracy": 0.18262252658605577,
      "num_tokens": 31632272.0,
      "step": 16975
    },
    {
      "entropy": 5.892288732528686,
      "epoch": 1.4994701518897915,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004778649419060496,
      "loss": 5.5997,
      "mean_token_accuracy": 0.17830167710781097,
      "num_tokens": 31642235.0,
      "step": 16980
    },
    {
      "entropy": 5.888100481033325,
      "epoch": 1.499911691981632,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004778513190569077,
      "loss": 5.6968,
      "mean_token_accuracy": 0.17314873337745668,
      "num_tokens": 31652773.0,
      "step": 16985
    },
    {
      "entropy": 6.016780376434326,
      "epoch": 1.5003532320734723,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004778376922340239,
      "loss": 5.6856,
      "mean_token_accuracy": 0.17541701048612596,
      "num_tokens": 31662681.0,
      "step": 16990
    },
    {
      "entropy": 5.9019410610198975,
      "epoch": 1.5007947721653125,
      "grad_norm": 1.375,
      "learning_rate": 0.000477824061437665,
      "loss": 5.5734,
      "mean_token_accuracy": 0.1820012390613556,
      "num_tokens": 31671576.0,
      "step": 16995
    },
    {
      "entropy": 5.912090158462524,
      "epoch": 1.501236312257153,
      "grad_norm": 1.625,
      "learning_rate": 0.00047781042666809816,
      "loss": 5.7986,
      "mean_token_accuracy": 0.16452100425958632,
      "num_tokens": 31681149.0,
      "step": 17000
    },
    {
      "entropy": 5.851611614227295,
      "epoch": 1.5016778523489933,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004777967879255904,
      "loss": 5.612,
      "mean_token_accuracy": 0.177215838432312,
      "num_tokens": 31690007.0,
      "step": 17005
    },
    {
      "entropy": 6.022160577774048,
      "epoch": 1.5021193924408336,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004777831452104088,
      "loss": 5.7259,
      "mean_token_accuracy": 0.17231721431016922,
      "num_tokens": 31700055.0,
      "step": 17010
    },
    {
      "entropy": 5.917047691345215,
      "epoch": 1.502560932532674,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004777694985228208,
      "loss": 5.6394,
      "mean_token_accuracy": 0.18284562230110168,
      "num_tokens": 31708255.0,
      "step": 17015
    },
    {
      "entropy": 5.864719533920288,
      "epoch": 1.5030024726245141,
      "grad_norm": 2.296875,
      "learning_rate": 0.00047775584786309365,
      "loss": 5.6544,
      "mean_token_accuracy": 0.1791571319103241,
      "num_tokens": 31716855.0,
      "step": 17020
    },
    {
      "entropy": 5.857101821899414,
      "epoch": 1.5034440127163546,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047774219323149474,
      "loss": 5.6121,
      "mean_token_accuracy": 0.1821003556251526,
      "num_tokens": 31726153.0,
      "step": 17025
    },
    {
      "entropy": 5.864023160934448,
      "epoch": 1.5038855528081951,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004777285346282916,
      "loss": 5.6279,
      "mean_token_accuracy": 0.1840591996908188,
      "num_tokens": 31736935.0,
      "step": 17030
    },
    {
      "entropy": 5.90579833984375,
      "epoch": 1.5043270929000352,
      "grad_norm": 1.625,
      "learning_rate": 0.0004777148720537517,
      "loss": 5.701,
      "mean_token_accuracy": 0.17577412575483323,
      "num_tokens": 31746402.0,
      "step": 17035
    },
    {
      "entropy": 5.894345378875732,
      "epoch": 1.5047686329918757,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047770120550814276,
      "loss": 5.5418,
      "mean_token_accuracy": 0.1827808916568756,
      "num_tokens": 31755615.0,
      "step": 17040
    },
    {
      "entropy": 5.881766986846924,
      "epoch": 1.505210173083716,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004776875349917325,
      "loss": 5.7034,
      "mean_token_accuracy": 0.18207792341709136,
      "num_tokens": 31765701.0,
      "step": 17045
    },
    {
      "entropy": 5.9493495464324955,
      "epoch": 1.5056517131755562,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004776738605047887,
      "loss": 5.6548,
      "mean_token_accuracy": 0.1777200385928154,
      "num_tokens": 31774674.0,
      "step": 17050
    },
    {
      "entropy": 5.942049407958985,
      "epoch": 1.5060932532673967,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004776601820475793,
      "loss": 5.6092,
      "mean_token_accuracy": 0.17480422258377076,
      "num_tokens": 31784871.0,
      "step": 17055
    },
    {
      "entropy": 5.934363126754761,
      "epoch": 1.506534793359237,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004776464996203721,
      "loss": 5.575,
      "mean_token_accuracy": 0.188628950715065,
      "num_tokens": 31793274.0,
      "step": 17060
    },
    {
      "entropy": 5.88822340965271,
      "epoch": 1.5069763334510773,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004776328132234353,
      "loss": 5.7499,
      "mean_token_accuracy": 0.1672848120331764,
      "num_tokens": 31803543.0,
      "step": 17065
    },
    {
      "entropy": 5.968618583679199,
      "epoch": 1.5074178735429178,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004776191228570368,
      "loss": 5.7069,
      "mean_token_accuracy": 0.1694990873336792,
      "num_tokens": 31814529.0,
      "step": 17070
    },
    {
      "entropy": 5.980963039398193,
      "epoch": 1.507859413634758,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004776054285214451,
      "loss": 5.6958,
      "mean_token_accuracy": 0.18286251723766328,
      "num_tokens": 31824064.0,
      "step": 17075
    },
    {
      "entropy": 5.921520805358886,
      "epoch": 1.5083009537265983,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047759173021692813,
      "loss": 5.6769,
      "mean_token_accuracy": 0.17392075508832933,
      "num_tokens": 31833881.0,
      "step": 17080
    },
    {
      "entropy": 5.858774042129516,
      "epoch": 1.5087424938184388,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004775780279437545,
      "loss": 5.6058,
      "mean_token_accuracy": 0.18330678790807725,
      "num_tokens": 31842987.0,
      "step": 17085
    },
    {
      "entropy": 5.890638256072998,
      "epoch": 1.5091840339102789,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047756432170219243,
      "loss": 5.5606,
      "mean_token_accuracy": 0.1795755609869957,
      "num_tokens": 31851806.0,
      "step": 17090
    },
    {
      "entropy": 5.977456855773926,
      "epoch": 1.5096255740021194,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004775506114925106,
      "loss": 5.7434,
      "mean_token_accuracy": 0.1755651757121086,
      "num_tokens": 31861857.0,
      "step": 17095
    },
    {
      "entropy": 5.945065593719482,
      "epoch": 1.5100671140939599,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047753689731497734,
      "loss": 5.6168,
      "mean_token_accuracy": 0.17781486213207245,
      "num_tokens": 31871252.0,
      "step": 17100
    },
    {
      "entropy": 5.922310638427734,
      "epoch": 1.5105086541858,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004775231791698616,
      "loss": 5.7098,
      "mean_token_accuracy": 0.16988943964242936,
      "num_tokens": 31880734.0,
      "step": 17105
    },
    {
      "entropy": 5.938547945022583,
      "epoch": 1.5109501942776404,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047750945705743187,
      "loss": 5.6147,
      "mean_token_accuracy": 0.17599107921123505,
      "num_tokens": 31889304.0,
      "step": 17110
    },
    {
      "entropy": 5.9107990741729735,
      "epoch": 1.5113917343694807,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004774957309779571,
      "loss": 5.627,
      "mean_token_accuracy": 0.1774899646639824,
      "num_tokens": 31898586.0,
      "step": 17115
    },
    {
      "entropy": 5.898506832122803,
      "epoch": 1.511833274461321,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004774820009317061,
      "loss": 5.6905,
      "mean_token_accuracy": 0.17115868479013444,
      "num_tokens": 31908153.0,
      "step": 17120
    },
    {
      "entropy": 5.862042427062988,
      "epoch": 1.5122748145531615,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047746826691894787,
      "loss": 5.4736,
      "mean_token_accuracy": 0.19376207441091536,
      "num_tokens": 31916467.0,
      "step": 17125
    },
    {
      "entropy": 5.87470269203186,
      "epoch": 1.5127163546450018,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047745452893995145,
      "loss": 5.6128,
      "mean_token_accuracy": 0.17668602019548416,
      "num_tokens": 31925469.0,
      "step": 17130
    },
    {
      "entropy": 5.884097957611084,
      "epoch": 1.513157894736842,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047744078699498587,
      "loss": 5.6938,
      "mean_token_accuracy": 0.17215728759765625,
      "num_tokens": 31934747.0,
      "step": 17135
    },
    {
      "entropy": 5.917584657669067,
      "epoch": 1.5135994348286825,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047742704108432046,
      "loss": 5.5936,
      "mean_token_accuracy": 0.17751585841178893,
      "num_tokens": 31943773.0,
      "step": 17140
    },
    {
      "entropy": 5.877157974243164,
      "epoch": 1.5140409749205228,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047741329120822446,
      "loss": 5.68,
      "mean_token_accuracy": 0.18280746340751647,
      "num_tokens": 31952699.0,
      "step": 17145
    },
    {
      "entropy": 5.90375828742981,
      "epoch": 1.514482515012363,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004773995373669672,
      "loss": 5.663,
      "mean_token_accuracy": 0.1706971898674965,
      "num_tokens": 31962845.0,
      "step": 17150
    },
    {
      "entropy": 5.882090044021607,
      "epoch": 1.5149240551042036,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004773857795608181,
      "loss": 5.5314,
      "mean_token_accuracy": 0.190124574303627,
      "num_tokens": 31972437.0,
      "step": 17155
    },
    {
      "entropy": 5.918551206588745,
      "epoch": 1.5153655951960436,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047737201779004667,
      "loss": 5.7309,
      "mean_token_accuracy": 0.17588242441415786,
      "num_tokens": 31981652.0,
      "step": 17160
    },
    {
      "entropy": 5.912468910217285,
      "epoch": 1.5158071352878841,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004773582520549226,
      "loss": 5.6475,
      "mean_token_accuracy": 0.17843589633703233,
      "num_tokens": 31990088.0,
      "step": 17165
    },
    {
      "entropy": 5.922000789642334,
      "epoch": 1.5162486753797246,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004773444823557153,
      "loss": 5.662,
      "mean_token_accuracy": 0.18198143243789672,
      "num_tokens": 31999386.0,
      "step": 17170
    },
    {
      "entropy": 5.977784490585327,
      "epoch": 1.5166902154715647,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004773307086926948,
      "loss": 5.5775,
      "mean_token_accuracy": 0.18299761563539504,
      "num_tokens": 32007695.0,
      "step": 17175
    },
    {
      "entropy": 5.950079393386841,
      "epoch": 1.5171317555634052,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047731693106613075,
      "loss": 5.7253,
      "mean_token_accuracy": 0.17564723789691924,
      "num_tokens": 32018300.0,
      "step": 17180
    },
    {
      "entropy": 5.916191482543946,
      "epoch": 1.5175732956552455,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004773031494762931,
      "loss": 5.6932,
      "mean_token_accuracy": 0.1750804916024208,
      "num_tokens": 32028386.0,
      "step": 17185
    },
    {
      "entropy": 5.960503959655762,
      "epoch": 1.5180148357470857,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004772893639234518,
      "loss": 5.69,
      "mean_token_accuracy": 0.18017399311065674,
      "num_tokens": 32037260.0,
      "step": 17190
    },
    {
      "entropy": 5.945848751068115,
      "epoch": 1.5184563758389262,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004772755744078769,
      "loss": 5.7192,
      "mean_token_accuracy": 0.17664412707090377,
      "num_tokens": 32047654.0,
      "step": 17195
    },
    {
      "entropy": 5.950189733505249,
      "epoch": 1.5188979159307665,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004772617809298385,
      "loss": 5.6354,
      "mean_token_accuracy": 0.17891675233840942,
      "num_tokens": 32056791.0,
      "step": 17200
    },
    {
      "entropy": 5.921965408325195,
      "epoch": 1.5193394560226068,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004772479834896069,
      "loss": 5.6277,
      "mean_token_accuracy": 0.17675877660512923,
      "num_tokens": 32066561.0,
      "step": 17205
    },
    {
      "entropy": 6.006550931930542,
      "epoch": 1.5197809961144473,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047723418208745226,
      "loss": 5.6699,
      "mean_token_accuracy": 0.1777320235967636,
      "num_tokens": 32075801.0,
      "step": 17210
    },
    {
      "entropy": 5.84832534790039,
      "epoch": 1.5202225362062876,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004772203767236451,
      "loss": 5.6065,
      "mean_token_accuracy": 0.18875672221183776,
      "num_tokens": 32085274.0,
      "step": 17215
    },
    {
      "entropy": 5.84977970123291,
      "epoch": 1.5206640762981278,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047720656739845573,
      "loss": 5.637,
      "mean_token_accuracy": 0.17870444506406785,
      "num_tokens": 32094481.0,
      "step": 17220
    },
    {
      "entropy": 5.817449045181275,
      "epoch": 1.5211056163899683,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004771927541121547,
      "loss": 5.5975,
      "mean_token_accuracy": 0.1845959410071373,
      "num_tokens": 32103094.0,
      "step": 17225
    },
    {
      "entropy": 5.906315803527832,
      "epoch": 1.5215471564818084,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004771789368650126,
      "loss": 5.6105,
      "mean_token_accuracy": 0.18306406885385512,
      "num_tokens": 32112230.0,
      "step": 17230
    },
    {
      "entropy": 5.9626750469207765,
      "epoch": 1.5219886965736489,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047716511565730013,
      "loss": 5.6297,
      "mean_token_accuracy": 0.18487911820411682,
      "num_tokens": 32121472.0,
      "step": 17235
    },
    {
      "entropy": 5.963954114913941,
      "epoch": 1.5224302366654894,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000477151290489288,
      "loss": 5.6515,
      "mean_token_accuracy": 0.17959727048873902,
      "num_tokens": 32130849.0,
      "step": 17240
    },
    {
      "entropy": 5.875452613830566,
      "epoch": 1.5228717767573294,
      "grad_norm": 1.625,
      "learning_rate": 0.0004771374613612471,
      "loss": 5.5731,
      "mean_token_accuracy": 0.18088472187519072,
      "num_tokens": 32140874.0,
      "step": 17245
    },
    {
      "entropy": 5.868573331832886,
      "epoch": 1.52331331684917,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004771236282734482,
      "loss": 5.5809,
      "mean_token_accuracy": 0.17832313477993011,
      "num_tokens": 32150774.0,
      "step": 17250
    },
    {
      "entropy": 5.8150872707366945,
      "epoch": 1.5237548569410102,
      "grad_norm": 1.703125,
      "learning_rate": 0.00047710979122616247,
      "loss": 5.5904,
      "mean_token_accuracy": 0.18249335438013076,
      "num_tokens": 32159825.0,
      "step": 17255
    },
    {
      "entropy": 5.915825748443604,
      "epoch": 1.5241963970328505,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004770959502196609,
      "loss": 5.6582,
      "mean_token_accuracy": 0.1765471488237381,
      "num_tokens": 32168443.0,
      "step": 17260
    },
    {
      "entropy": 5.9387579441070555,
      "epoch": 1.524637937124691,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047708210525421445,
      "loss": 5.6053,
      "mean_token_accuracy": 0.18037229627370835,
      "num_tokens": 32178065.0,
      "step": 17265
    },
    {
      "entropy": 5.872773170471191,
      "epoch": 1.5250794772165313,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047706825633009466,
      "loss": 5.4916,
      "mean_token_accuracy": 0.1876641571521759,
      "num_tokens": 32187274.0,
      "step": 17270
    },
    {
      "entropy": 5.830988121032715,
      "epoch": 1.5255210173083715,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047705440344757264,
      "loss": 5.4589,
      "mean_token_accuracy": 0.19508485794067382,
      "num_tokens": 32195811.0,
      "step": 17275
    },
    {
      "entropy": 5.731103706359863,
      "epoch": 1.525962557400212,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00047704054660691964,
      "loss": 5.5658,
      "mean_token_accuracy": 0.1868984192609787,
      "num_tokens": 32204958.0,
      "step": 17280
    },
    {
      "entropy": 5.862072992324829,
      "epoch": 1.5264040974920523,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004770266858084074,
      "loss": 5.6018,
      "mean_token_accuracy": 0.18250273913145065,
      "num_tokens": 32214299.0,
      "step": 17285
    },
    {
      "entropy": 5.9188720226287845,
      "epoch": 1.5268456375838926,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047701282105230716,
      "loss": 5.6777,
      "mean_token_accuracy": 0.17532525062561036,
      "num_tokens": 32223353.0,
      "step": 17290
    },
    {
      "entropy": 5.94893479347229,
      "epoch": 1.527287177675733,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004769989523388906,
      "loss": 5.6616,
      "mean_token_accuracy": 0.17000703066587447,
      "num_tokens": 32232641.0,
      "step": 17295
    },
    {
      "entropy": 5.947882604598999,
      "epoch": 1.5277287177675731,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004769850796684296,
      "loss": 5.7219,
      "mean_token_accuracy": 0.17556276619434358,
      "num_tokens": 32243028.0,
      "step": 17300
    },
    {
      "entropy": 5.893780279159546,
      "epoch": 1.5281702578594136,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004769712030411957,
      "loss": 5.6549,
      "mean_token_accuracy": 0.179551300406456,
      "num_tokens": 32253660.0,
      "step": 17305
    },
    {
      "entropy": 5.907131147384644,
      "epoch": 1.5286117979512541,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047695732245746076,
      "loss": 5.6329,
      "mean_token_accuracy": 0.18480906933546065,
      "num_tokens": 32263267.0,
      "step": 17310
    },
    {
      "entropy": 5.877878141403198,
      "epoch": 1.5290533380430942,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004769434379174967,
      "loss": 5.5908,
      "mean_token_accuracy": 0.18816550970077514,
      "num_tokens": 32273929.0,
      "step": 17315
    },
    {
      "entropy": 5.855655717849731,
      "epoch": 1.5294948781349347,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004769295494215757,
      "loss": 5.6113,
      "mean_token_accuracy": 0.18302763551473616,
      "num_tokens": 32282717.0,
      "step": 17320
    },
    {
      "entropy": 5.939192247390747,
      "epoch": 1.529936418226775,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047691565696996954,
      "loss": 5.6573,
      "mean_token_accuracy": 0.1851394295692444,
      "num_tokens": 32293225.0,
      "step": 17325
    },
    {
      "entropy": 5.994198989868164,
      "epoch": 1.5303779583186152,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004769017605629505,
      "loss": 5.6088,
      "mean_token_accuracy": 0.18232553452253342,
      "num_tokens": 32301582.0,
      "step": 17330
    },
    {
      "entropy": 5.8350663661956785,
      "epoch": 1.5308194984104557,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004768878602007909,
      "loss": 5.5738,
      "mean_token_accuracy": 0.1818352907896042,
      "num_tokens": 32310131.0,
      "step": 17335
    },
    {
      "entropy": 5.81473970413208,
      "epoch": 1.531261038502296,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004768739558837628,
      "loss": 5.6319,
      "mean_token_accuracy": 0.18179805725812911,
      "num_tokens": 32320038.0,
      "step": 17340
    },
    {
      "entropy": 5.892063283920288,
      "epoch": 1.5317025785941363,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004768600476121388,
      "loss": 5.6942,
      "mean_token_accuracy": 0.17224355340003966,
      "num_tokens": 32329568.0,
      "step": 17345
    },
    {
      "entropy": 5.973785924911499,
      "epoch": 1.5321441186859768,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047684613538619125,
      "loss": 5.6364,
      "mean_token_accuracy": 0.1840396985411644,
      "num_tokens": 32338940.0,
      "step": 17350
    },
    {
      "entropy": 5.906280755996704,
      "epoch": 1.532585658777817,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004768322192061927,
      "loss": 5.5972,
      "mean_token_accuracy": 0.18160295039415358,
      "num_tokens": 32347757.0,
      "step": 17355
    },
    {
      "entropy": 5.811740684509277,
      "epoch": 1.5330271988696573,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004768182990724158,
      "loss": 5.5359,
      "mean_token_accuracy": 0.19202418625354767,
      "num_tokens": 32357099.0,
      "step": 17360
    },
    {
      "entropy": 5.87677321434021,
      "epoch": 1.5334687389614978,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047680437498513317,
      "loss": 5.6471,
      "mean_token_accuracy": 0.1761384904384613,
      "num_tokens": 32366548.0,
      "step": 17365
    },
    {
      "entropy": 5.949100637435913,
      "epoch": 1.5339102790533379,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047679044694461765,
      "loss": 5.766,
      "mean_token_accuracy": 0.17105017602443695,
      "num_tokens": 32377435.0,
      "step": 17370
    },
    {
      "entropy": 5.942914247512817,
      "epoch": 1.5343518191451784,
      "grad_norm": 1.40625,
      "learning_rate": 0.000476776514951142,
      "loss": 5.6463,
      "mean_token_accuracy": 0.17149488478899003,
      "num_tokens": 32386922.0,
      "step": 17375
    },
    {
      "entropy": 5.843531274795533,
      "epoch": 1.5347933592370189,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047676257900497915,
      "loss": 5.7106,
      "mean_token_accuracy": 0.17415758818387986,
      "num_tokens": 32396350.0,
      "step": 17380
    },
    {
      "entropy": 5.808764839172364,
      "epoch": 1.535234899328859,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047674863910640217,
      "loss": 5.4687,
      "mean_token_accuracy": 0.1968031957745552,
      "num_tokens": 32405678.0,
      "step": 17385
    },
    {
      "entropy": 5.920073986053467,
      "epoch": 1.5356764394206994,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047673469525568406,
      "loss": 5.6042,
      "mean_token_accuracy": 0.18699740320444108,
      "num_tokens": 32415793.0,
      "step": 17390
    },
    {
      "entropy": 5.860673141479492,
      "epoch": 1.5361179795125397,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047672074745309804,
      "loss": 5.489,
      "mean_token_accuracy": 0.19413916915655136,
      "num_tokens": 32424735.0,
      "step": 17395
    },
    {
      "entropy": 5.784531688690185,
      "epoch": 1.53655951960438,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004767067956989172,
      "loss": 5.6365,
      "mean_token_accuracy": 0.18361037373542785,
      "num_tokens": 32435213.0,
      "step": 17400
    },
    {
      "entropy": 5.828448486328125,
      "epoch": 1.5370010596962205,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004766928399934151,
      "loss": 5.5174,
      "mean_token_accuracy": 0.18660463094711305,
      "num_tokens": 32443932.0,
      "step": 17405
    },
    {
      "entropy": 5.879975080490112,
      "epoch": 1.5374425997880607,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004766788803368649,
      "loss": 5.5366,
      "mean_token_accuracy": 0.18072292059659958,
      "num_tokens": 32453851.0,
      "step": 17410
    },
    {
      "entropy": 5.868438673019409,
      "epoch": 1.537884139879901,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004766649167295401,
      "loss": 5.6185,
      "mean_token_accuracy": 0.18665646016597748,
      "num_tokens": 32462906.0,
      "step": 17415
    },
    {
      "entropy": 5.801081275939941,
      "epoch": 1.5383256799717415,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004766509491717144,
      "loss": 5.5918,
      "mean_token_accuracy": 0.18337966352701188,
      "num_tokens": 32471840.0,
      "step": 17420
    },
    {
      "entropy": 5.810268783569336,
      "epoch": 1.5387672200635818,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00047663697766366114,
      "loss": 5.5409,
      "mean_token_accuracy": 0.18577390015125275,
      "num_tokens": 32481111.0,
      "step": 17425
    },
    {
      "entropy": 5.906863832473755,
      "epoch": 1.539208760155422,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004766230022056542,
      "loss": 5.6411,
      "mean_token_accuracy": 0.17604240775108337,
      "num_tokens": 32489827.0,
      "step": 17430
    },
    {
      "entropy": 5.898635005950927,
      "epoch": 1.5396503002472626,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047660902279796736,
      "loss": 5.5658,
      "mean_token_accuracy": 0.19064524471759797,
      "num_tokens": 32499106.0,
      "step": 17435
    },
    {
      "entropy": 6.04878249168396,
      "epoch": 1.5400918403391026,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004765950394408744,
      "loss": 5.8234,
      "mean_token_accuracy": 0.15808669626712799,
      "num_tokens": 32507858.0,
      "step": 17440
    },
    {
      "entropy": 5.861137771606446,
      "epoch": 1.5405333804309431,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004765810521346493,
      "loss": 5.5884,
      "mean_token_accuracy": 0.1844190090894699,
      "num_tokens": 32516761.0,
      "step": 17445
    },
    {
      "entropy": 5.84430046081543,
      "epoch": 1.5409749205227836,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00047656706087956604,
      "loss": 5.6264,
      "mean_token_accuracy": 0.1886868566274643,
      "num_tokens": 32526585.0,
      "step": 17450
    },
    {
      "entropy": 5.903663969039917,
      "epoch": 1.5414164606146237,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047655306567589864,
      "loss": 5.6165,
      "mean_token_accuracy": 0.18131843954324722,
      "num_tokens": 32535607.0,
      "step": 17455
    },
    {
      "entropy": 5.855196475982666,
      "epoch": 1.5418580007064642,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047653906652392135,
      "loss": 5.5557,
      "mean_token_accuracy": 0.1892412468791008,
      "num_tokens": 32545401.0,
      "step": 17460
    },
    {
      "entropy": 5.907823324203491,
      "epoch": 1.5422995407983044,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004765250634239083,
      "loss": 5.6229,
      "mean_token_accuracy": 0.18683450073003768,
      "num_tokens": 32554421.0,
      "step": 17465
    },
    {
      "entropy": 5.935530757904052,
      "epoch": 1.5427410808901447,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00047651105637613395,
      "loss": 5.7078,
      "mean_token_accuracy": 0.1738913133740425,
      "num_tokens": 32563652.0,
      "step": 17470
    },
    {
      "entropy": 5.90299277305603,
      "epoch": 1.5431826209819852,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004764970453808726,
      "loss": 5.6761,
      "mean_token_accuracy": 0.1833822324872017,
      "num_tokens": 32572812.0,
      "step": 17475
    },
    {
      "entropy": 5.823865461349487,
      "epoch": 1.5436241610738255,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047648303043839875,
      "loss": 5.6018,
      "mean_token_accuracy": 0.1880460113286972,
      "num_tokens": 32581746.0,
      "step": 17480
    },
    {
      "entropy": 5.967733144760132,
      "epoch": 1.5440657011656658,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004764690115489869,
      "loss": 5.6173,
      "mean_token_accuracy": 0.18205758333206176,
      "num_tokens": 32591596.0,
      "step": 17485
    },
    {
      "entropy": 5.898574113845825,
      "epoch": 1.5445072412575063,
      "grad_norm": 1.828125,
      "learning_rate": 0.00047645498871291163,
      "loss": 5.6366,
      "mean_token_accuracy": 0.17731774598360062,
      "num_tokens": 32600610.0,
      "step": 17490
    },
    {
      "entropy": 5.9062690258026125,
      "epoch": 1.5449487813493465,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004764409619304478,
      "loss": 5.6573,
      "mean_token_accuracy": 0.17298922091722488,
      "num_tokens": 32609827.0,
      "step": 17495
    },
    {
      "entropy": 5.912159776687622,
      "epoch": 1.5453903214411868,
      "grad_norm": 1.5,
      "learning_rate": 0.00047642693120187007,
      "loss": 5.6591,
      "mean_token_accuracy": 0.17628840655088424,
      "num_tokens": 32618258.0,
      "step": 17500
    },
    {
      "entropy": 5.942746353149414,
      "epoch": 1.5458318615330273,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004764128965274533,
      "loss": 5.6351,
      "mean_token_accuracy": 0.18264712691307067,
      "num_tokens": 32628155.0,
      "step": 17505
    },
    {
      "entropy": 5.956526470184326,
      "epoch": 1.5462734016248674,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047639885790747244,
      "loss": 5.708,
      "mean_token_accuracy": 0.17549778670072555,
      "num_tokens": 32638347.0,
      "step": 17510
    },
    {
      "entropy": 5.898592901229859,
      "epoch": 1.5467149417167079,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047638481534220244,
      "loss": 5.5751,
      "mean_token_accuracy": 0.18309051990509034,
      "num_tokens": 32648636.0,
      "step": 17515
    },
    {
      "entropy": 5.882865762710571,
      "epoch": 1.5471564818085484,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004763707688319185,
      "loss": 5.6434,
      "mean_token_accuracy": 0.17818101048469542,
      "num_tokens": 32657762.0,
      "step": 17520
    },
    {
      "entropy": 5.864386940002442,
      "epoch": 1.5475980219003884,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004763567183768958,
      "loss": 5.6063,
      "mean_token_accuracy": 0.17635509818792344,
      "num_tokens": 32668265.0,
      "step": 17525
    },
    {
      "entropy": 5.852547025680542,
      "epoch": 1.548039561992229,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047634266397740934,
      "loss": 5.5421,
      "mean_token_accuracy": 0.18639024049043657,
      "num_tokens": 32678051.0,
      "step": 17530
    },
    {
      "entropy": 6.011360836029053,
      "epoch": 1.5484811020840692,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047632860563373466,
      "loss": 5.7168,
      "mean_token_accuracy": 0.18239671885967254,
      "num_tokens": 32687186.0,
      "step": 17535
    },
    {
      "entropy": 5.889007091522217,
      "epoch": 1.5489226421759095,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004763145433461472,
      "loss": 5.6235,
      "mean_token_accuracy": 0.18502212464809417,
      "num_tokens": 32697082.0,
      "step": 17540
    },
    {
      "entropy": 5.900610637664795,
      "epoch": 1.54936418226775,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004763004771149222,
      "loss": 5.6318,
      "mean_token_accuracy": 0.17766666561365127,
      "num_tokens": 32706502.0,
      "step": 17545
    },
    {
      "entropy": 5.938467025756836,
      "epoch": 1.5498057223595902,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004762864069403354,
      "loss": 5.6078,
      "mean_token_accuracy": 0.17404073625802993,
      "num_tokens": 32715035.0,
      "step": 17550
    },
    {
      "entropy": 5.8661096572875975,
      "epoch": 1.5502472624514305,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047627233282266236,
      "loss": 5.6224,
      "mean_token_accuracy": 0.17830966114997865,
      "num_tokens": 32724713.0,
      "step": 17555
    },
    {
      "entropy": 5.83719573020935,
      "epoch": 1.550688802543271,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004762582547621788,
      "loss": 5.6238,
      "mean_token_accuracy": 0.18238939642906188,
      "num_tokens": 32734584.0,
      "step": 17560
    },
    {
      "entropy": 5.89838752746582,
      "epoch": 1.5511303426351113,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047624417275916045,
      "loss": 5.5999,
      "mean_token_accuracy": 0.18029648661613465,
      "num_tokens": 32742744.0,
      "step": 17565
    },
    {
      "entropy": 5.867640686035156,
      "epoch": 1.5515718827269516,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047623008681388324,
      "loss": 5.5373,
      "mean_token_accuracy": 0.197463820874691,
      "num_tokens": 32752051.0,
      "step": 17570
    },
    {
      "entropy": 5.896329116821289,
      "epoch": 1.552013422818792,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047621599692662306,
      "loss": 5.6191,
      "mean_token_accuracy": 0.17850878834724426,
      "num_tokens": 32761740.0,
      "step": 17575
    },
    {
      "entropy": 5.7995617389678955,
      "epoch": 1.5524549629106321,
      "grad_norm": 1.765625,
      "learning_rate": 0.00047620190309765595,
      "loss": 5.557,
      "mean_token_accuracy": 0.18946147114038467,
      "num_tokens": 32770809.0,
      "step": 17580
    },
    {
      "entropy": 5.927500867843628,
      "epoch": 1.5528965030024726,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047618780532725805,
      "loss": 5.6602,
      "mean_token_accuracy": 0.17695012986660003,
      "num_tokens": 32780614.0,
      "step": 17585
    },
    {
      "entropy": 5.984373378753662,
      "epoch": 1.5533380430943131,
      "grad_norm": 2.75,
      "learning_rate": 0.00047617370361570534,
      "loss": 5.6409,
      "mean_token_accuracy": 0.17657425105571747,
      "num_tokens": 32790336.0,
      "step": 17590
    },
    {
      "entropy": 5.8811070919036865,
      "epoch": 1.5537795831861532,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004761595979632743,
      "loss": 5.6182,
      "mean_token_accuracy": 0.18347604423761368,
      "num_tokens": 32798991.0,
      "step": 17595
    },
    {
      "entropy": 5.937896108627319,
      "epoch": 1.5542211232779937,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004761454883702411,
      "loss": 5.6404,
      "mean_token_accuracy": 0.17994246631860733,
      "num_tokens": 32807616.0,
      "step": 17600
    },
    {
      "entropy": 5.890559387207031,
      "epoch": 1.554662663369834,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004761313748368821,
      "loss": 5.6783,
      "mean_token_accuracy": 0.17243186980485917,
      "num_tokens": 32817107.0,
      "step": 17605
    },
    {
      "entropy": 5.893438673019409,
      "epoch": 1.5551042034616742,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004761172573634739,
      "loss": 5.6404,
      "mean_token_accuracy": 0.18184627443552018,
      "num_tokens": 32827205.0,
      "step": 17610
    },
    {
      "entropy": 5.897537803649902,
      "epoch": 1.5555457435535147,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004761031359502931,
      "loss": 5.5437,
      "mean_token_accuracy": 0.1843265101313591,
      "num_tokens": 32836102.0,
      "step": 17615
    },
    {
      "entropy": 5.932310581207275,
      "epoch": 1.555987283645355,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00047608901059761617,
      "loss": 5.6784,
      "mean_token_accuracy": 0.1738116204738617,
      "num_tokens": 32845389.0,
      "step": 17620
    },
    {
      "entropy": 5.861991739273071,
      "epoch": 1.5564288237371953,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047607488130571997,
      "loss": 5.5497,
      "mean_token_accuracy": 0.18778333216905593,
      "num_tokens": 32854186.0,
      "step": 17625
    },
    {
      "entropy": 5.860267114639282,
      "epoch": 1.5568703638290358,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004760607480748811,
      "loss": 5.56,
      "mean_token_accuracy": 0.18250260651111602,
      "num_tokens": 32863498.0,
      "step": 17630
    },
    {
      "entropy": 5.893199014663696,
      "epoch": 1.557311903920876,
      "grad_norm": 1.859375,
      "learning_rate": 0.00047604661090537657,
      "loss": 5.6835,
      "mean_token_accuracy": 0.17352318465709687,
      "num_tokens": 32872483.0,
      "step": 17635
    },
    {
      "entropy": 5.961558055877686,
      "epoch": 1.5577534440127163,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004760324697974832,
      "loss": 5.6741,
      "mean_token_accuracy": 0.17394876331090928,
      "num_tokens": 32882739.0,
      "step": 17640
    },
    {
      "entropy": 5.981095170974731,
      "epoch": 1.5581949841045568,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004760183247514782,
      "loss": 5.7644,
      "mean_token_accuracy": 0.1764628916978836,
      "num_tokens": 32892707.0,
      "step": 17645
    },
    {
      "entropy": 5.928404808044434,
      "epoch": 1.5586365241963969,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047600417576763843,
      "loss": 5.6681,
      "mean_token_accuracy": 0.1751195788383484,
      "num_tokens": 32901916.0,
      "step": 17650
    },
    {
      "entropy": 5.856634140014648,
      "epoch": 1.5590780642882374,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004759900228462412,
      "loss": 5.6003,
      "mean_token_accuracy": 0.18466822654008866,
      "num_tokens": 32910656.0,
      "step": 17655
    },
    {
      "entropy": 6.010602331161499,
      "epoch": 1.5595196043800779,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004759758659875638,
      "loss": 5.7096,
      "mean_token_accuracy": 0.18000097423791886,
      "num_tokens": 32919579.0,
      "step": 17660
    },
    {
      "entropy": 5.903562259674072,
      "epoch": 1.559961144471918,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047596170519188335,
      "loss": 5.6505,
      "mean_token_accuracy": 0.17840010076761245,
      "num_tokens": 32928271.0,
      "step": 17665
    },
    {
      "entropy": 5.9111144065856935,
      "epoch": 1.5604026845637584,
      "grad_norm": 1.75,
      "learning_rate": 0.0004759475404594776,
      "loss": 5.5521,
      "mean_token_accuracy": 0.19078251719474792,
      "num_tokens": 32937671.0,
      "step": 17670
    },
    {
      "entropy": 5.923218631744385,
      "epoch": 1.5608442246555987,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004759333717906236,
      "loss": 5.7288,
      "mean_token_accuracy": 0.1720937356352806,
      "num_tokens": 32946896.0,
      "step": 17675
    },
    {
      "entropy": 5.864073371887207,
      "epoch": 1.561285764747439,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047591919918559927,
      "loss": 5.6738,
      "mean_token_accuracy": 0.17394649088382722,
      "num_tokens": 32956706.0,
      "step": 17680
    },
    {
      "entropy": 5.83187084197998,
      "epoch": 1.5617273048392795,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004759050226446821,
      "loss": 5.5209,
      "mean_token_accuracy": 0.19307021647691727,
      "num_tokens": 32965990.0,
      "step": 17685
    },
    {
      "entropy": 5.860781955718994,
      "epoch": 1.5621688449311197,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004758908421681497,
      "loss": 5.538,
      "mean_token_accuracy": 0.1857747793197632,
      "num_tokens": 32975311.0,
      "step": 17690
    },
    {
      "entropy": 5.9178320407867435,
      "epoch": 1.56261038502296,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047587665775627997,
      "loss": 5.6316,
      "mean_token_accuracy": 0.1809835687279701,
      "num_tokens": 32984641.0,
      "step": 17695
    },
    {
      "entropy": 5.894848155975342,
      "epoch": 1.5630519251148005,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047586246940935087,
      "loss": 5.6715,
      "mean_token_accuracy": 0.17661734968423842,
      "num_tokens": 32994696.0,
      "step": 17700
    },
    {
      "entropy": 5.858583736419678,
      "epoch": 1.5634934652066408,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00047584827712764013,
      "loss": 5.577,
      "mean_token_accuracy": 0.18670083731412887,
      "num_tokens": 33003392.0,
      "step": 17705
    },
    {
      "entropy": 5.898713731765747,
      "epoch": 1.563935005298481,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004758340809114259,
      "loss": 5.6397,
      "mean_token_accuracy": 0.17559218257665635,
      "num_tokens": 33012896.0,
      "step": 17710
    },
    {
      "entropy": 5.943633413314819,
      "epoch": 1.5643765453903216,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004758198807609862,
      "loss": 5.6838,
      "mean_token_accuracy": 0.17544018626213073,
      "num_tokens": 33022710.0,
      "step": 17715
    },
    {
      "entropy": 5.976519441604614,
      "epoch": 1.5648180854821618,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004758056766765993,
      "loss": 5.6142,
      "mean_token_accuracy": 0.17892904132604598,
      "num_tokens": 33031982.0,
      "step": 17720
    },
    {
      "entropy": 5.9029481410980225,
      "epoch": 1.5652596255740021,
      "grad_norm": 1.625,
      "learning_rate": 0.00047579146865854333,
      "loss": 5.7024,
      "mean_token_accuracy": 0.17993957102298735,
      "num_tokens": 33041496.0,
      "step": 17725
    },
    {
      "entropy": 5.86553316116333,
      "epoch": 1.5657011656658426,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047577725670709675,
      "loss": 5.5896,
      "mean_token_accuracy": 0.18756068795919417,
      "num_tokens": 33050764.0,
      "step": 17730
    },
    {
      "entropy": 5.869291210174561,
      "epoch": 1.5661427057576827,
      "grad_norm": 1.8984375,
      "learning_rate": 0.000475763040822538,
      "loss": 5.6499,
      "mean_token_accuracy": 0.18053606152534485,
      "num_tokens": 33060809.0,
      "step": 17735
    },
    {
      "entropy": 5.807592964172363,
      "epoch": 1.5665842458495232,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004757488210051453,
      "loss": 5.5149,
      "mean_token_accuracy": 0.18751915991306306,
      "num_tokens": 33070155.0,
      "step": 17740
    },
    {
      "entropy": 5.867930126190186,
      "epoch": 1.5670257859413634,
      "grad_norm": 2.625,
      "learning_rate": 0.0004757345972551974,
      "loss": 5.6223,
      "mean_token_accuracy": 0.18780549764633178,
      "num_tokens": 33079354.0,
      "step": 17745
    },
    {
      "entropy": 5.889115238189698,
      "epoch": 1.5674673260332037,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004757203695729729,
      "loss": 5.5787,
      "mean_token_accuracy": 0.1846635177731514,
      "num_tokens": 33087877.0,
      "step": 17750
    },
    {
      "entropy": 5.898090457916259,
      "epoch": 1.5679088661250442,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047570613795875055,
      "loss": 5.6725,
      "mean_token_accuracy": 0.1763919860124588,
      "num_tokens": 33097375.0,
      "step": 17755
    },
    {
      "entropy": 5.927813768386841,
      "epoch": 1.5683504062168845,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004756919024128091,
      "loss": 5.6403,
      "mean_token_accuracy": 0.18057384639978408,
      "num_tokens": 33106336.0,
      "step": 17760
    },
    {
      "entropy": 5.796157360076904,
      "epoch": 1.5687919463087248,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004756776629354274,
      "loss": 5.5024,
      "mean_token_accuracy": 0.19268492013216018,
      "num_tokens": 33115753.0,
      "step": 17765
    },
    {
      "entropy": 5.945907163619995,
      "epoch": 1.5692334864005653,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004756634195268844,
      "loss": 5.6532,
      "mean_token_accuracy": 0.17173600792884827,
      "num_tokens": 33125205.0,
      "step": 17770
    },
    {
      "entropy": 5.862942790985107,
      "epoch": 1.5696750264924055,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004756491721874592,
      "loss": 5.5752,
      "mean_token_accuracy": 0.18657755255699157,
      "num_tokens": 33134470.0,
      "step": 17775
    },
    {
      "entropy": 5.8235869884490965,
      "epoch": 1.5701165665842458,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047563492091743075,
      "loss": 5.5604,
      "mean_token_accuracy": 0.19287998378276824,
      "num_tokens": 33143346.0,
      "step": 17780
    },
    {
      "entropy": 5.818963098526001,
      "epoch": 1.5705581066760863,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047562066571707836,
      "loss": 5.5543,
      "mean_token_accuracy": 0.1843525782227516,
      "num_tokens": 33153291.0,
      "step": 17785
    },
    {
      "entropy": 5.960869455337525,
      "epoch": 1.5709996467679266,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047560640658668126,
      "loss": 5.7491,
      "mean_token_accuracy": 0.17322501689195632,
      "num_tokens": 33163642.0,
      "step": 17790
    },
    {
      "entropy": 5.9673668384552006,
      "epoch": 1.5714411868597669,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047559214352651863,
      "loss": 5.5709,
      "mean_token_accuracy": 0.1867929369211197,
      "num_tokens": 33173529.0,
      "step": 17795
    },
    {
      "entropy": 5.914347219467163,
      "epoch": 1.5718827269516074,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047557787653687014,
      "loss": 5.5571,
      "mean_token_accuracy": 0.1837488144636154,
      "num_tokens": 33180877.0,
      "step": 17800
    },
    {
      "entropy": 5.818093109130859,
      "epoch": 1.5723242670434474,
      "grad_norm": 1.625,
      "learning_rate": 0.000475563605618015,
      "loss": 5.6517,
      "mean_token_accuracy": 0.1773436889052391,
      "num_tokens": 33190831.0,
      "step": 17805
    },
    {
      "entropy": 5.887660169601441,
      "epoch": 1.572765807135288,
      "grad_norm": 1.78125,
      "learning_rate": 0.000475549330770233,
      "loss": 5.6687,
      "mean_token_accuracy": 0.17604399025440215,
      "num_tokens": 33200313.0,
      "step": 17810
    },
    {
      "entropy": 5.964189291000366,
      "epoch": 1.5732073472271282,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004755350519938036,
      "loss": 5.6238,
      "mean_token_accuracy": 0.18037100732326508,
      "num_tokens": 33209272.0,
      "step": 17815
    },
    {
      "entropy": 5.866541385650635,
      "epoch": 1.5736488873189685,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047552076928900663,
      "loss": 5.5741,
      "mean_token_accuracy": 0.1933285787701607,
      "num_tokens": 33218712.0,
      "step": 17820
    },
    {
      "entropy": 5.8714062690734865,
      "epoch": 1.574090427410809,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004755064826561217,
      "loss": 5.6759,
      "mean_token_accuracy": 0.1822512924671173,
      "num_tokens": 33227480.0,
      "step": 17825
    },
    {
      "entropy": 5.893762922286987,
      "epoch": 1.5745319675026492,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004754921920954289,
      "loss": 5.5772,
      "mean_token_accuracy": 0.18264857232570647,
      "num_tokens": 33236044.0,
      "step": 17830
    },
    {
      "entropy": 5.869308853149414,
      "epoch": 1.5749735075944895,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004754778976072081,
      "loss": 5.6368,
      "mean_token_accuracy": 0.1830672100186348,
      "num_tokens": 33244770.0,
      "step": 17835
    },
    {
      "entropy": 5.855055856704712,
      "epoch": 1.57541504768633,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047546359919173933,
      "loss": 5.569,
      "mean_token_accuracy": 0.18134605884552002,
      "num_tokens": 33254178.0,
      "step": 17840
    },
    {
      "entropy": 5.809454345703125,
      "epoch": 1.5758565877781703,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004754492968493026,
      "loss": 5.6165,
      "mean_token_accuracy": 0.18815515488386153,
      "num_tokens": 33264210.0,
      "step": 17845
    },
    {
      "entropy": 5.805385875701904,
      "epoch": 1.5762981278700106,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004754349905801782,
      "loss": 5.5357,
      "mean_token_accuracy": 0.18817970603704454,
      "num_tokens": 33273348.0,
      "step": 17850
    },
    {
      "entropy": 5.916689062118531,
      "epoch": 1.576739667961851,
      "grad_norm": 1.71875,
      "learning_rate": 0.00047542068038464625,
      "loss": 5.5718,
      "mean_token_accuracy": 0.19347279369831086,
      "num_tokens": 33281906.0,
      "step": 17855
    },
    {
      "entropy": 5.922132158279419,
      "epoch": 1.5771812080536913,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004754063662629872,
      "loss": 5.7052,
      "mean_token_accuracy": 0.1737179845571518,
      "num_tokens": 33291363.0,
      "step": 17860
    },
    {
      "entropy": 5.870745611190796,
      "epoch": 1.5776227481455316,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047539204821548146,
      "loss": 5.6201,
      "mean_token_accuracy": 0.18283369541168212,
      "num_tokens": 33299785.0,
      "step": 17865
    },
    {
      "entropy": 5.987374401092529,
      "epoch": 1.578064288237372,
      "grad_norm": 1.5,
      "learning_rate": 0.0004753777262424094,
      "loss": 5.7174,
      "mean_token_accuracy": 0.18006044626235962,
      "num_tokens": 33309655.0,
      "step": 17870
    },
    {
      "entropy": 5.921486854553223,
      "epoch": 1.5785058283292122,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047536340034405166,
      "loss": 5.6891,
      "mean_token_accuracy": 0.17361037731170653,
      "num_tokens": 33320158.0,
      "step": 17875
    },
    {
      "entropy": 5.8017274856567385,
      "epoch": 1.5789473684210527,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00047534907052068886,
      "loss": 5.5614,
      "mean_token_accuracy": 0.187873575091362,
      "num_tokens": 33328959.0,
      "step": 17880
    },
    {
      "entropy": 5.919699048995971,
      "epoch": 1.579388908512893,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004753347367726017,
      "loss": 5.6773,
      "mean_token_accuracy": 0.18052133917808533,
      "num_tokens": 33338624.0,
      "step": 17885
    },
    {
      "entropy": 5.96409592628479,
      "epoch": 1.5798304486047332,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047532039910007097,
      "loss": 5.6526,
      "mean_token_accuracy": 0.18409485071897508,
      "num_tokens": 33348083.0,
      "step": 17890
    },
    {
      "entropy": 5.909086894989014,
      "epoch": 1.5802719886965737,
      "grad_norm": 1.5,
      "learning_rate": 0.00047530605750337765,
      "loss": 5.584,
      "mean_token_accuracy": 0.18112163692712785,
      "num_tokens": 33358226.0,
      "step": 17895
    },
    {
      "entropy": 5.957688236236573,
      "epoch": 1.580713528788414,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047529171198280244,
      "loss": 5.7134,
      "mean_token_accuracy": 0.1772045373916626,
      "num_tokens": 33367997.0,
      "step": 17900
    },
    {
      "entropy": 5.934866333007813,
      "epoch": 1.5811550688802543,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00047527736253862664,
      "loss": 5.671,
      "mean_token_accuracy": 0.17616036683320999,
      "num_tokens": 33377574.0,
      "step": 17905
    },
    {
      "entropy": 5.912517023086548,
      "epoch": 1.5815966089720948,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047526300917113116,
      "loss": 5.686,
      "mean_token_accuracy": 0.18008253276348113,
      "num_tokens": 33387016.0,
      "step": 17910
    },
    {
      "entropy": 5.946104431152344,
      "epoch": 1.582038149063935,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004752486518805972,
      "loss": 5.6072,
      "mean_token_accuracy": 0.17755938470363616,
      "num_tokens": 33396239.0,
      "step": 17915
    },
    {
      "entropy": 5.842429542541504,
      "epoch": 1.5824796891557753,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047523429066730603,
      "loss": 5.5149,
      "mean_token_accuracy": 0.1916379600763321,
      "num_tokens": 33404738.0,
      "step": 17920
    },
    {
      "entropy": 5.75519437789917,
      "epoch": 1.5829212292476158,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047521992553153903,
      "loss": 5.5104,
      "mean_token_accuracy": 0.1923816442489624,
      "num_tokens": 33413670.0,
      "step": 17925
    },
    {
      "entropy": 5.8398042678833,
      "epoch": 1.583362769339456,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004752055564735775,
      "loss": 5.6378,
      "mean_token_accuracy": 0.17306254655122758,
      "num_tokens": 33422841.0,
      "step": 17930
    },
    {
      "entropy": 6.002101087570191,
      "epoch": 1.5838043094312964,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047519118349370306,
      "loss": 5.6658,
      "mean_token_accuracy": 0.17790588289499282,
      "num_tokens": 33432023.0,
      "step": 17935
    },
    {
      "entropy": 5.955539560317993,
      "epoch": 1.5842458495231369,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004751768065921971,
      "loss": 5.6606,
      "mean_token_accuracy": 0.16982008069753646,
      "num_tokens": 33442181.0,
      "step": 17940
    },
    {
      "entropy": 5.936857414245606,
      "epoch": 1.584687389614977,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047516242576934144,
      "loss": 5.7066,
      "mean_token_accuracy": 0.16783083379268646,
      "num_tokens": 33451847.0,
      "step": 17945
    },
    {
      "entropy": 5.951867008209229,
      "epoch": 1.5851289297068174,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047514804102541765,
      "loss": 5.6786,
      "mean_token_accuracy": 0.18039079159498214,
      "num_tokens": 33461587.0,
      "step": 17950
    },
    {
      "entropy": 5.970877456665039,
      "epoch": 1.5855704697986577,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047513365236070753,
      "loss": 5.7016,
      "mean_token_accuracy": 0.17196838408708573,
      "num_tokens": 33471553.0,
      "step": 17955
    },
    {
      "entropy": 5.831390428543091,
      "epoch": 1.586012009890498,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047511925977549303,
      "loss": 5.5456,
      "mean_token_accuracy": 0.188623283803463,
      "num_tokens": 33480984.0,
      "step": 17960
    },
    {
      "entropy": 5.906336927413941,
      "epoch": 1.5864535499823385,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047510486327005604,
      "loss": 5.6625,
      "mean_token_accuracy": 0.18237873464822768,
      "num_tokens": 33491196.0,
      "step": 17965
    },
    {
      "entropy": 5.794521760940552,
      "epoch": 1.5868950900741787,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047509046284467855,
      "loss": 5.5239,
      "mean_token_accuracy": 0.1890060856938362,
      "num_tokens": 33500404.0,
      "step": 17970
    },
    {
      "entropy": 5.935758495330811,
      "epoch": 1.587336630166019,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047507605849964274,
      "loss": 5.7157,
      "mean_token_accuracy": 0.17548144459724427,
      "num_tokens": 33509281.0,
      "step": 17975
    },
    {
      "entropy": 5.936355352401733,
      "epoch": 1.5877781702578595,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004750616502352307,
      "loss": 5.5891,
      "mean_token_accuracy": 0.17484840005636215,
      "num_tokens": 33518556.0,
      "step": 17980
    },
    {
      "entropy": 5.968524742126465,
      "epoch": 1.5882197103496998,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00047504723805172474,
      "loss": 5.7793,
      "mean_token_accuracy": 0.1691001236438751,
      "num_tokens": 33529266.0,
      "step": 17985
    },
    {
      "entropy": 5.864870214462281,
      "epoch": 1.58866125044154,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00047503282194940715,
      "loss": 5.6583,
      "mean_token_accuracy": 0.17974342852830888,
      "num_tokens": 33538972.0,
      "step": 17990
    },
    {
      "entropy": 5.87936110496521,
      "epoch": 1.5891027905333805,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004750184019285603,
      "loss": 5.5825,
      "mean_token_accuracy": 0.18118787556886673,
      "num_tokens": 33549528.0,
      "step": 17995
    },
    {
      "entropy": 5.828487920761108,
      "epoch": 1.5895443306252208,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004750039779894668,
      "loss": 5.5226,
      "mean_token_accuracy": 0.18468477725982665,
      "num_tokens": 33557828.0,
      "step": 18000
    },
    {
      "epoch": 1.5895443306252208,
      "eval_entropy": 5.653435604704601,
      "eval_loss": 5.6908159255981445,
      "eval_mean_token_accuracy": 0.18594363286547017,
      "eval_num_tokens": 33557828.0,
      "eval_runtime": 26.2431,
      "eval_samples_per_second": 1345.687,
      "eval_steps_per_second": 168.235,
      "step": 18000
    },
    {
      "entropy": 5.877416753768921,
      "epoch": 1.589985870717061,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00047498955013240907,
      "loss": 5.6762,
      "mean_token_accuracy": 0.17563886791467667,
      "num_tokens": 33567087.0,
      "step": 18005
    },
    {
      "entropy": 5.873627090454102,
      "epoch": 1.5904274108089016,
      "grad_norm": 1.890625,
      "learning_rate": 0.00047497511835766984,
      "loss": 5.6544,
      "mean_token_accuracy": 0.17707592099905015,
      "num_tokens": 33577442.0,
      "step": 18010
    },
    {
      "entropy": 5.965829467773437,
      "epoch": 1.5908689509007417,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047496068266553165,
      "loss": 5.6582,
      "mean_token_accuracy": 0.17760249078273774,
      "num_tokens": 33585655.0,
      "step": 18015
    },
    {
      "entropy": 5.922989845275879,
      "epoch": 1.5913104909925821,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047494624305627754,
      "loss": 5.6607,
      "mean_token_accuracy": 0.1773097559809685,
      "num_tokens": 33594808.0,
      "step": 18020
    },
    {
      "entropy": 5.94341025352478,
      "epoch": 1.5917520310844224,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004749317995301902,
      "loss": 5.7087,
      "mean_token_accuracy": 0.17429924309253692,
      "num_tokens": 33603924.0,
      "step": 18025
    },
    {
      "entropy": 5.892343521118164,
      "epoch": 1.5921935711762627,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004749173520875525,
      "loss": 5.5653,
      "mean_token_accuracy": 0.18425866216421127,
      "num_tokens": 33613870.0,
      "step": 18030
    },
    {
      "entropy": 5.865020608901977,
      "epoch": 1.5926351112681032,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047490290072864773,
      "loss": 5.623,
      "mean_token_accuracy": 0.18315306454896926,
      "num_tokens": 33622938.0,
      "step": 18035
    },
    {
      "entropy": 5.9211091041564945,
      "epoch": 1.5930766513599435,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00047488844545375886,
      "loss": 5.6808,
      "mean_token_accuracy": 0.1799934521317482,
      "num_tokens": 33631520.0,
      "step": 18040
    },
    {
      "entropy": 5.98555908203125,
      "epoch": 1.5935181914517837,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004748739862631689,
      "loss": 5.671,
      "mean_token_accuracy": 0.17576166689395906,
      "num_tokens": 33641365.0,
      "step": 18045
    },
    {
      "entropy": 5.856840562820435,
      "epoch": 1.5939597315436242,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047485952315716126,
      "loss": 5.4911,
      "mean_token_accuracy": 0.1923200771212578,
      "num_tokens": 33650425.0,
      "step": 18050
    },
    {
      "entropy": 5.904925918579101,
      "epoch": 1.5944012716354645,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047484505613601917,
      "loss": 5.6851,
      "mean_token_accuracy": 0.1761615455150604,
      "num_tokens": 33659928.0,
      "step": 18055
    },
    {
      "entropy": 5.868963432312012,
      "epoch": 1.5948428117273048,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004748305852000262,
      "loss": 5.6668,
      "mean_token_accuracy": 0.1844443276524544,
      "num_tokens": 33668860.0,
      "step": 18060
    },
    {
      "entropy": 5.9667566299438475,
      "epoch": 1.5952843518191453,
      "grad_norm": 2.578125,
      "learning_rate": 0.00047481611034946565,
      "loss": 5.6458,
      "mean_token_accuracy": 0.17786285579204558,
      "num_tokens": 33679343.0,
      "step": 18065
    },
    {
      "entropy": 5.882617044448852,
      "epoch": 1.5957258919109856,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004748016315846212,
      "loss": 5.5851,
      "mean_token_accuracy": 0.18951045721769333,
      "num_tokens": 33688749.0,
      "step": 18070
    },
    {
      "entropy": 5.921667861938476,
      "epoch": 1.5961674320028258,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047478714890577633,
      "loss": 5.654,
      "mean_token_accuracy": 0.17785474210977553,
      "num_tokens": 33698359.0,
      "step": 18075
    },
    {
      "entropy": 5.9411663055419925,
      "epoch": 1.5966089720946663,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047477266231321484,
      "loss": 5.6472,
      "mean_token_accuracy": 0.1725616052746773,
      "num_tokens": 33707402.0,
      "step": 18080
    },
    {
      "entropy": 5.908167695999145,
      "epoch": 1.5970505121865064,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047475817180722056,
      "loss": 5.5914,
      "mean_token_accuracy": 0.18342378586530686,
      "num_tokens": 33717043.0,
      "step": 18085
    },
    {
      "entropy": 5.847962856292725,
      "epoch": 1.597492052278347,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047474367738807734,
      "loss": 5.57,
      "mean_token_accuracy": 0.19307893216609956,
      "num_tokens": 33726527.0,
      "step": 18090
    },
    {
      "entropy": 5.84596962928772,
      "epoch": 1.5979335923701872,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047472917905606903,
      "loss": 5.5525,
      "mean_token_accuracy": 0.1862705186009407,
      "num_tokens": 33735357.0,
      "step": 18095
    },
    {
      "entropy": 5.776023530960083,
      "epoch": 1.5983751324620274,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047471467681147974,
      "loss": 5.4548,
      "mean_token_accuracy": 0.20128918588161468,
      "num_tokens": 33743660.0,
      "step": 18100
    },
    {
      "entropy": 5.8935459613800045,
      "epoch": 1.598816672553868,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004747001706545935,
      "loss": 5.6597,
      "mean_token_accuracy": 0.17620477378368377,
      "num_tokens": 33752899.0,
      "step": 18105
    },
    {
      "entropy": 5.890561580657959,
      "epoch": 1.5992582126457082,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004746856605856944,
      "loss": 5.6347,
      "mean_token_accuracy": 0.18205931782722473,
      "num_tokens": 33761333.0,
      "step": 18110
    },
    {
      "entropy": 5.900943183898926,
      "epoch": 1.5996997527375485,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047467114660506687,
      "loss": 5.6616,
      "mean_token_accuracy": 0.17888675034046173,
      "num_tokens": 33770507.0,
      "step": 18115
    },
    {
      "entropy": 5.833289241790771,
      "epoch": 1.600141292829389,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004746566287129951,
      "loss": 5.5834,
      "mean_token_accuracy": 0.1808208182454109,
      "num_tokens": 33781658.0,
      "step": 18120
    },
    {
      "entropy": 5.953752231597901,
      "epoch": 1.6005828329212293,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004746421069097636,
      "loss": 5.7097,
      "mean_token_accuracy": 0.17414340674877166,
      "num_tokens": 33792229.0,
      "step": 18125
    },
    {
      "entropy": 6.01095666885376,
      "epoch": 1.6010243730130695,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047462758119565663,
      "loss": 5.6601,
      "mean_token_accuracy": 0.18166436105966569,
      "num_tokens": 33801674.0,
      "step": 18130
    },
    {
      "entropy": 5.848023557662964,
      "epoch": 1.60146591310491,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000474613051570959,
      "loss": 5.5781,
      "mean_token_accuracy": 0.1909630626440048,
      "num_tokens": 33812066.0,
      "step": 18135
    },
    {
      "entropy": 5.927769565582276,
      "epoch": 1.6019074531967503,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047459851803595524,
      "loss": 5.7091,
      "mean_token_accuracy": 0.17546397745609282,
      "num_tokens": 33822206.0,
      "step": 18140
    },
    {
      "entropy": 5.957952356338501,
      "epoch": 1.6023489932885906,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047458398059092995,
      "loss": 5.6981,
      "mean_token_accuracy": 0.17649979442358016,
      "num_tokens": 33832075.0,
      "step": 18145
    },
    {
      "entropy": 5.916622734069824,
      "epoch": 1.602790533380431,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00047456943923616793,
      "loss": 5.5302,
      "mean_token_accuracy": 0.18238042593002318,
      "num_tokens": 33841523.0,
      "step": 18150
    },
    {
      "entropy": 5.893067026138306,
      "epoch": 1.6032320734722711,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047455489397195416,
      "loss": 5.5769,
      "mean_token_accuracy": 0.18152596950531005,
      "num_tokens": 33850756.0,
      "step": 18155
    },
    {
      "entropy": 5.8490166664123535,
      "epoch": 1.6036736135641116,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047454034479857356,
      "loss": 5.6431,
      "mean_token_accuracy": 0.1858445793390274,
      "num_tokens": 33859611.0,
      "step": 18160
    },
    {
      "entropy": 5.879465103149414,
      "epoch": 1.604115153655952,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000474525791716311,
      "loss": 5.6407,
      "mean_token_accuracy": 0.1807444676756859,
      "num_tokens": 33867993.0,
      "step": 18165
    },
    {
      "entropy": 5.938706254959106,
      "epoch": 1.6045566937477922,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047451123472545175,
      "loss": 5.7111,
      "mean_token_accuracy": 0.17588558793067932,
      "num_tokens": 33876785.0,
      "step": 18170
    },
    {
      "entropy": 5.921104192733765,
      "epoch": 1.6049982338396327,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00047449667382628083,
      "loss": 5.5554,
      "mean_token_accuracy": 0.18571677356958388,
      "num_tokens": 33885729.0,
      "step": 18175
    },
    {
      "entropy": 5.952887868881225,
      "epoch": 1.605439773931473,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004744821090190835,
      "loss": 5.5932,
      "mean_token_accuracy": 0.18684795647859573,
      "num_tokens": 33894622.0,
      "step": 18180
    },
    {
      "entropy": 5.850035572052002,
      "epoch": 1.6058813140233132,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047446754030414513,
      "loss": 5.6756,
      "mean_token_accuracy": 0.18364445269107818,
      "num_tokens": 33904187.0,
      "step": 18185
    },
    {
      "entropy": 5.854431867599487,
      "epoch": 1.6063228541151537,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004744529676817511,
      "loss": 5.6777,
      "mean_token_accuracy": 0.18205590695142745,
      "num_tokens": 33914327.0,
      "step": 18190
    },
    {
      "entropy": 5.959302520751953,
      "epoch": 1.606764394206994,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047443839115218685,
      "loss": 5.5158,
      "mean_token_accuracy": 0.18315639197826386,
      "num_tokens": 33923424.0,
      "step": 18195
    },
    {
      "entropy": 5.889602088928223,
      "epoch": 1.6072059342988343,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000474423810715738,
      "loss": 5.5788,
      "mean_token_accuracy": 0.1796111509203911,
      "num_tokens": 33932145.0,
      "step": 18200
    },
    {
      "entropy": 5.7743066310882565,
      "epoch": 1.6076474743906748,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004744092263726901,
      "loss": 5.5938,
      "mean_token_accuracy": 0.18472411334514618,
      "num_tokens": 33942505.0,
      "step": 18205
    },
    {
      "entropy": 5.879548358917236,
      "epoch": 1.608089014482515,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004743946381233288,
      "loss": 5.6543,
      "mean_token_accuracy": 0.1804282248020172,
      "num_tokens": 33952156.0,
      "step": 18210
    },
    {
      "entropy": 5.965029191970825,
      "epoch": 1.6085305545743553,
      "grad_norm": 1.703125,
      "learning_rate": 0.00047438004596794,
      "loss": 5.6686,
      "mean_token_accuracy": 0.18106010556221008,
      "num_tokens": 33962478.0,
      "step": 18215
    },
    {
      "entropy": 5.947190618515014,
      "epoch": 1.6089720946661958,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004743654499068095,
      "loss": 5.6018,
      "mean_token_accuracy": 0.18265584260225295,
      "num_tokens": 33971581.0,
      "step": 18220
    },
    {
      "entropy": 5.893946647644043,
      "epoch": 1.609413634758036,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004743508499402232,
      "loss": 5.6029,
      "mean_token_accuracy": 0.181997711956501,
      "num_tokens": 33981262.0,
      "step": 18225
    },
    {
      "entropy": 5.928482437133789,
      "epoch": 1.6098551748498764,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004743362460684671,
      "loss": 5.6485,
      "mean_token_accuracy": 0.17622361630201339,
      "num_tokens": 33989865.0,
      "step": 18230
    },
    {
      "entropy": 5.932847309112549,
      "epoch": 1.6102967149417167,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047432163829182737,
      "loss": 5.6464,
      "mean_token_accuracy": 0.174939426779747,
      "num_tokens": 33999261.0,
      "step": 18235
    },
    {
      "entropy": 5.973556852340698,
      "epoch": 1.610738255033557,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047430702661059007,
      "loss": 5.6374,
      "mean_token_accuracy": 0.17709564417600632,
      "num_tokens": 34008471.0,
      "step": 18240
    },
    {
      "entropy": 5.793225908279419,
      "epoch": 1.6111797951253974,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00047429241102504144,
      "loss": 5.5033,
      "mean_token_accuracy": 0.1940534994006157,
      "num_tokens": 34018305.0,
      "step": 18245
    },
    {
      "entropy": 5.880889225006103,
      "epoch": 1.6116213352172377,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047427779153546787,
      "loss": 5.6741,
      "mean_token_accuracy": 0.17414819151163102,
      "num_tokens": 34027393.0,
      "step": 18250
    },
    {
      "entropy": 5.918023777008057,
      "epoch": 1.612062875309078,
      "grad_norm": 3.421875,
      "learning_rate": 0.00047426316814215563,
      "loss": 5.6099,
      "mean_token_accuracy": 0.1820577621459961,
      "num_tokens": 34035835.0,
      "step": 18255
    },
    {
      "entropy": 5.875483417510987,
      "epoch": 1.6125044154009185,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047424854084539134,
      "loss": 5.673,
      "mean_token_accuracy": 0.1739334136247635,
      "num_tokens": 34045946.0,
      "step": 18260
    },
    {
      "entropy": 5.955771350860596,
      "epoch": 1.6129459554927588,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047423390964546145,
      "loss": 5.7247,
      "mean_token_accuracy": 0.17568904608488084,
      "num_tokens": 34055356.0,
      "step": 18265
    },
    {
      "entropy": 6.027866983413697,
      "epoch": 1.613387495584599,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004742192745426525,
      "loss": 5.783,
      "mean_token_accuracy": 0.16616266667842866,
      "num_tokens": 34065862.0,
      "step": 18270
    },
    {
      "entropy": 6.031091165542603,
      "epoch": 1.6138290356764395,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004742046355372513,
      "loss": 5.6584,
      "mean_token_accuracy": 0.18115150630474092,
      "num_tokens": 34074980.0,
      "step": 18275
    },
    {
      "entropy": 5.9205302715301515,
      "epoch": 1.6142705757682798,
      "grad_norm": 1.9375,
      "learning_rate": 0.00047418999262954466,
      "loss": 5.6437,
      "mean_token_accuracy": 0.17559588998556136,
      "num_tokens": 34085501.0,
      "step": 18280
    },
    {
      "entropy": 5.876504421234131,
      "epoch": 1.61471211586012,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004741753458198192,
      "loss": 5.6086,
      "mean_token_accuracy": 0.18046877831220626,
      "num_tokens": 34095564.0,
      "step": 18285
    },
    {
      "entropy": 5.887704658508301,
      "epoch": 1.6151536559519606,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047416069510836215,
      "loss": 5.6571,
      "mean_token_accuracy": 0.17717305570840836,
      "num_tokens": 34104571.0,
      "step": 18290
    },
    {
      "entropy": 5.938283777236938,
      "epoch": 1.6155951960438006,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004741460404954602,
      "loss": 5.6932,
      "mean_token_accuracy": 0.17122002094984054,
      "num_tokens": 34114272.0,
      "step": 18295
    },
    {
      "entropy": 5.868648862838745,
      "epoch": 1.6160367361356411,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004741313819814007,
      "loss": 5.6078,
      "mean_token_accuracy": 0.18391987681388855,
      "num_tokens": 34125521.0,
      "step": 18300
    },
    {
      "entropy": 5.894225215911865,
      "epoch": 1.6164782762274814,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004741167195664706,
      "loss": 5.6671,
      "mean_token_accuracy": 0.17441552728414536,
      "num_tokens": 34134447.0,
      "step": 18305
    },
    {
      "entropy": 5.904989385604859,
      "epoch": 1.6169198163193217,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047410205325095725,
      "loss": 5.5575,
      "mean_token_accuracy": 0.1844706118106842,
      "num_tokens": 34143207.0,
      "step": 18310
    },
    {
      "entropy": 5.857340526580811,
      "epoch": 1.6173613564111622,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004740873830351479,
      "loss": 5.5942,
      "mean_token_accuracy": 0.18009879142045976,
      "num_tokens": 34153025.0,
      "step": 18315
    },
    {
      "entropy": 5.87808051109314,
      "epoch": 1.6178028965030025,
      "grad_norm": 1.625,
      "learning_rate": 0.00047407270891932995,
      "loss": 5.6154,
      "mean_token_accuracy": 0.18018353581428528,
      "num_tokens": 34162793.0,
      "step": 18320
    },
    {
      "entropy": 5.8646728038787845,
      "epoch": 1.6182444365948427,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047405803090379083,
      "loss": 5.6534,
      "mean_token_accuracy": 0.17740414291620255,
      "num_tokens": 34171941.0,
      "step": 18325
    },
    {
      "entropy": 5.915258598327637,
      "epoch": 1.6186859766866832,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004740433489888182,
      "loss": 5.6319,
      "mean_token_accuracy": 0.18179358839988707,
      "num_tokens": 34180993.0,
      "step": 18330
    },
    {
      "entropy": 5.891985130310059,
      "epoch": 1.6191275167785235,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047402866317469944,
      "loss": 5.5683,
      "mean_token_accuracy": 0.1840633988380432,
      "num_tokens": 34190078.0,
      "step": 18335
    },
    {
      "entropy": 5.810507774353027,
      "epoch": 1.6195690568703638,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004740139734617224,
      "loss": 5.5548,
      "mean_token_accuracy": 0.1885581761598587,
      "num_tokens": 34199091.0,
      "step": 18340
    },
    {
      "entropy": 5.902998352050782,
      "epoch": 1.6200105969622043,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004739992798501748,
      "loss": 5.6851,
      "mean_token_accuracy": 0.17901372909545898,
      "num_tokens": 34209042.0,
      "step": 18345
    },
    {
      "entropy": 5.951916456222534,
      "epoch": 1.6204521370540446,
      "grad_norm": 3.078125,
      "learning_rate": 0.00047398458234034445,
      "loss": 5.6236,
      "mean_token_accuracy": 0.18292912244796752,
      "num_tokens": 34217311.0,
      "step": 18350
    },
    {
      "entropy": 5.816677904129028,
      "epoch": 1.6208936771458848,
      "grad_norm": 1.625,
      "learning_rate": 0.00047396988093251934,
      "loss": 5.5859,
      "mean_token_accuracy": 0.18733442574739456,
      "num_tokens": 34227001.0,
      "step": 18355
    },
    {
      "entropy": 5.887027025222778,
      "epoch": 1.6213352172377253,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004739551756269875,
      "loss": 5.6248,
      "mean_token_accuracy": 0.17762571275234224,
      "num_tokens": 34236214.0,
      "step": 18360
    },
    {
      "entropy": 5.90545539855957,
      "epoch": 1.6217767573295654,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004739404664240368,
      "loss": 5.5866,
      "mean_token_accuracy": 0.18065727055072783,
      "num_tokens": 34244808.0,
      "step": 18365
    },
    {
      "entropy": 5.839606142044067,
      "epoch": 1.6222182974214059,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047392575332395556,
      "loss": 5.5923,
      "mean_token_accuracy": 0.18739141523838043,
      "num_tokens": 34253355.0,
      "step": 18370
    },
    {
      "entropy": 5.969604730606079,
      "epoch": 1.6226598375132462,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004739110363270319,
      "loss": 5.7158,
      "mean_token_accuracy": 0.1712602213025093,
      "num_tokens": 34262235.0,
      "step": 18375
    },
    {
      "entropy": 5.9426065444946286,
      "epoch": 1.6231013776050864,
      "grad_norm": 1.90625,
      "learning_rate": 0.00047389631543355424,
      "loss": 5.5735,
      "mean_token_accuracy": 0.18347776979207991,
      "num_tokens": 34272174.0,
      "step": 18380
    },
    {
      "entropy": 5.8784623622894285,
      "epoch": 1.623542917696927,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004738815906438108,
      "loss": 5.589,
      "mean_token_accuracy": 0.18667105734348297,
      "num_tokens": 34281153.0,
      "step": 18385
    },
    {
      "entropy": 5.8625201225280765,
      "epoch": 1.6239844577887672,
      "grad_norm": 1.625,
      "learning_rate": 0.00047386686195809017,
      "loss": 5.6169,
      "mean_token_accuracy": 0.1805195316672325,
      "num_tokens": 34291045.0,
      "step": 18390
    },
    {
      "entropy": 5.900256872177124,
      "epoch": 1.6244259978806075,
      "grad_norm": 1.953125,
      "learning_rate": 0.00047385212937668084,
      "loss": 5.5773,
      "mean_token_accuracy": 0.18098017424345017,
      "num_tokens": 34299234.0,
      "step": 18395
    },
    {
      "entropy": 5.903788805007935,
      "epoch": 1.624867537972448,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00047383739289987134,
      "loss": 5.6447,
      "mean_token_accuracy": 0.1803504690527916,
      "num_tokens": 34309084.0,
      "step": 18400
    },
    {
      "entropy": 5.971255445480347,
      "epoch": 1.6253090780642883,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004738226525279504,
      "loss": 5.718,
      "mean_token_accuracy": 0.17027440816164016,
      "num_tokens": 34320039.0,
      "step": 18405
    },
    {
      "entropy": 5.872696208953857,
      "epoch": 1.6257506181561285,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004738079082612068,
      "loss": 5.5318,
      "mean_token_accuracy": 0.18494272083044053,
      "num_tokens": 34329020.0,
      "step": 18410
    },
    {
      "entropy": 5.923114347457886,
      "epoch": 1.626192158247969,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047379316009992927,
      "loss": 5.6661,
      "mean_token_accuracy": 0.17081080824136735,
      "num_tokens": 34337953.0,
      "step": 18415
    },
    {
      "entropy": 5.79700870513916,
      "epoch": 1.6266336983398093,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047377840804440686,
      "loss": 5.5162,
      "mean_token_accuracy": 0.18945104777812957,
      "num_tokens": 34346564.0,
      "step": 18420
    },
    {
      "entropy": 5.950593566894531,
      "epoch": 1.6270752384316496,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004737636520949285,
      "loss": 5.7628,
      "mean_token_accuracy": 0.17168935239315034,
      "num_tokens": 34357163.0,
      "step": 18425
    },
    {
      "entropy": 6.008922147750854,
      "epoch": 1.62751677852349,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004737488922517832,
      "loss": 5.7474,
      "mean_token_accuracy": 0.17189678251743318,
      "num_tokens": 34366731.0,
      "step": 18430
    },
    {
      "entropy": 5.895754957199097,
      "epoch": 1.6279583186153301,
      "grad_norm": 1.890625,
      "learning_rate": 0.00047373412851526013,
      "loss": 5.5912,
      "mean_token_accuracy": 0.18732481747865676,
      "num_tokens": 34375845.0,
      "step": 18435
    },
    {
      "entropy": 5.8833400249481205,
      "epoch": 1.6283998587071706,
      "grad_norm": 1.71875,
      "learning_rate": 0.00047371936088564856,
      "loss": 5.5951,
      "mean_token_accuracy": 0.1828589454293251,
      "num_tokens": 34385335.0,
      "step": 18440
    },
    {
      "entropy": 5.940899133682251,
      "epoch": 1.628841398799011,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004737045893632377,
      "loss": 5.6755,
      "mean_token_accuracy": 0.18172894567251205,
      "num_tokens": 34394356.0,
      "step": 18445
    },
    {
      "entropy": 5.904163932800293,
      "epoch": 1.6292829388908512,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047368981394831696,
      "loss": 5.6167,
      "mean_token_accuracy": 0.1807738721370697,
      "num_tokens": 34403727.0,
      "step": 18450
    },
    {
      "entropy": 5.98841667175293,
      "epoch": 1.6297244789826917,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047367503464117573,
      "loss": 5.67,
      "mean_token_accuracy": 0.18083688467741013,
      "num_tokens": 34413253.0,
      "step": 18455
    },
    {
      "entropy": 5.89859619140625,
      "epoch": 1.630166019074532,
      "grad_norm": 1.65625,
      "learning_rate": 0.00047366025144210356,
      "loss": 5.595,
      "mean_token_accuracy": 0.1850277826189995,
      "num_tokens": 34422430.0,
      "step": 18460
    },
    {
      "entropy": 5.897191333770752,
      "epoch": 1.6306075591663722,
      "grad_norm": 1.75,
      "learning_rate": 0.00047364546435139004,
      "loss": 5.6153,
      "mean_token_accuracy": 0.1783999353647232,
      "num_tokens": 34431111.0,
      "step": 18465
    },
    {
      "entropy": 5.896241760253906,
      "epoch": 1.6310490992582127,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047363067336932493,
      "loss": 5.5956,
      "mean_token_accuracy": 0.18560017496347428,
      "num_tokens": 34440044.0,
      "step": 18470
    },
    {
      "entropy": 5.879086828231811,
      "epoch": 1.631490639350053,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004736158784961978,
      "loss": 5.624,
      "mean_token_accuracy": 0.18534496873617173,
      "num_tokens": 34450315.0,
      "step": 18475
    },
    {
      "entropy": 5.87082200050354,
      "epoch": 1.6319321794418933,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047360107973229865,
      "loss": 5.5314,
      "mean_token_accuracy": 0.19719038009643555,
      "num_tokens": 34459380.0,
      "step": 18480
    },
    {
      "entropy": 5.9591797351837155,
      "epoch": 1.6323737195337338,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004735862770779172,
      "loss": 5.6556,
      "mean_token_accuracy": 0.17981008291244507,
      "num_tokens": 34468224.0,
      "step": 18485
    },
    {
      "entropy": 5.8714405536651615,
      "epoch": 1.632815259625574,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004735714705333436,
      "loss": 5.5381,
      "mean_token_accuracy": 0.18255700916051865,
      "num_tokens": 34476917.0,
      "step": 18490
    },
    {
      "entropy": 5.8377769947052,
      "epoch": 1.6332567997174143,
      "grad_norm": 1.5,
      "learning_rate": 0.00047355666009886785,
      "loss": 5.625,
      "mean_token_accuracy": 0.1847287267446518,
      "num_tokens": 34487192.0,
      "step": 18495
    },
    {
      "entropy": 5.871887063980102,
      "epoch": 1.6336983398092548,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047354184577478,
      "loss": 5.6029,
      "mean_token_accuracy": 0.18331348747015,
      "num_tokens": 34496658.0,
      "step": 18500
    },
    {
      "entropy": 5.911286449432373,
      "epoch": 1.6341398799010949,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047352702756137035,
      "loss": 5.5178,
      "mean_token_accuracy": 0.19093345254659652,
      "num_tokens": 34505126.0,
      "step": 18505
    },
    {
      "entropy": 5.914085817337036,
      "epoch": 1.6345814199929354,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004735122054589291,
      "loss": 5.5647,
      "mean_token_accuracy": 0.17961241006851197,
      "num_tokens": 34514236.0,
      "step": 18510
    },
    {
      "entropy": 5.757527780532837,
      "epoch": 1.6350229600847757,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004734973794677467,
      "loss": 5.5024,
      "mean_token_accuracy": 0.1865934506058693,
      "num_tokens": 34522609.0,
      "step": 18515
    },
    {
      "entropy": 5.822296476364135,
      "epoch": 1.635464500176616,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047348254958811347,
      "loss": 5.6415,
      "mean_token_accuracy": 0.18332186341285706,
      "num_tokens": 34531742.0,
      "step": 18520
    },
    {
      "entropy": 5.931294775009155,
      "epoch": 1.6359060402684564,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047346771582031994,
      "loss": 5.5591,
      "mean_token_accuracy": 0.18306408822536469,
      "num_tokens": 34540821.0,
      "step": 18525
    },
    {
      "entropy": 5.978801202774048,
      "epoch": 1.6363475803602967,
      "grad_norm": 1.5,
      "learning_rate": 0.0004734528781646568,
      "loss": 5.6517,
      "mean_token_accuracy": 0.18073801398277284,
      "num_tokens": 34549795.0,
      "step": 18530
    },
    {
      "entropy": 5.904628658294678,
      "epoch": 1.636789120452137,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004734380366214146,
      "loss": 5.6753,
      "mean_token_accuracy": 0.17446669787168503,
      "num_tokens": 34559455.0,
      "step": 18535
    },
    {
      "entropy": 5.923455476760864,
      "epoch": 1.6372306605439775,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004734231911908842,
      "loss": 5.6496,
      "mean_token_accuracy": 0.1754900574684143,
      "num_tokens": 34568832.0,
      "step": 18540
    },
    {
      "entropy": 5.927921199798584,
      "epoch": 1.6376722006358178,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047340834187335626,
      "loss": 5.6225,
      "mean_token_accuracy": 0.1871847167611122,
      "num_tokens": 34578630.0,
      "step": 18545
    },
    {
      "entropy": 5.968499851226807,
      "epoch": 1.638113740727658,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047339348866912173,
      "loss": 5.669,
      "mean_token_accuracy": 0.17591654360294343,
      "num_tokens": 34587782.0,
      "step": 18550
    },
    {
      "entropy": 5.830733394622802,
      "epoch": 1.6385552808194985,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004733786315784716,
      "loss": 5.5829,
      "mean_token_accuracy": 0.18597442358732225,
      "num_tokens": 34597133.0,
      "step": 18555
    },
    {
      "entropy": 5.848003673553467,
      "epoch": 1.6389968209113388,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047336377060169696,
      "loss": 5.6611,
      "mean_token_accuracy": 0.17179993987083436,
      "num_tokens": 34606697.0,
      "step": 18560
    },
    {
      "entropy": 5.96315746307373,
      "epoch": 1.639438361003179,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004733489057390888,
      "loss": 5.6142,
      "mean_token_accuracy": 0.18545187413692474,
      "num_tokens": 34615806.0,
      "step": 18565
    },
    {
      "entropy": 5.934550952911377,
      "epoch": 1.6398799010950196,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004733340369909384,
      "loss": 5.7662,
      "mean_token_accuracy": 0.1707201600074768,
      "num_tokens": 34626563.0,
      "step": 18570
    },
    {
      "entropy": 5.904636287689209,
      "epoch": 1.6403214411868596,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047331916435753706,
      "loss": 5.6555,
      "mean_token_accuracy": 0.18277918994426728,
      "num_tokens": 34636077.0,
      "step": 18575
    },
    {
      "entropy": 5.944668245315552,
      "epoch": 1.6407629812787001,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047330428783917607,
      "loss": 5.678,
      "mean_token_accuracy": 0.17595191895961762,
      "num_tokens": 34646164.0,
      "step": 18580
    },
    {
      "entropy": 5.901544427871704,
      "epoch": 1.6412045213705404,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047328940743614683,
      "loss": 5.6145,
      "mean_token_accuracy": 0.1762840673327446,
      "num_tokens": 34655577.0,
      "step": 18585
    },
    {
      "entropy": 5.940162944793701,
      "epoch": 1.6416460614623807,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00047327452314874094,
      "loss": 5.6646,
      "mean_token_accuracy": 0.17254654467105865,
      "num_tokens": 34664520.0,
      "step": 18590
    },
    {
      "entropy": 5.872901821136475,
      "epoch": 1.6420876015542212,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047325963497724983,
      "loss": 5.5501,
      "mean_token_accuracy": 0.1861518681049347,
      "num_tokens": 34673528.0,
      "step": 18595
    },
    {
      "entropy": 5.867902612686157,
      "epoch": 1.6425291416460615,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004732447429219652,
      "loss": 5.6675,
      "mean_token_accuracy": 0.18405531197786332,
      "num_tokens": 34682563.0,
      "step": 18600
    },
    {
      "entropy": 5.920318126678467,
      "epoch": 1.6429706817379017,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004732298469831789,
      "loss": 5.5558,
      "mean_token_accuracy": 0.18476105183362962,
      "num_tokens": 34691932.0,
      "step": 18605
    },
    {
      "entropy": 5.924129247665405,
      "epoch": 1.6434122218297422,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047321494716118256,
      "loss": 5.5696,
      "mean_token_accuracy": 0.1832903280854225,
      "num_tokens": 34701376.0,
      "step": 18610
    },
    {
      "entropy": 5.833938217163086,
      "epoch": 1.6438537619215825,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047320004345626816,
      "loss": 5.5881,
      "mean_token_accuracy": 0.18042815774679183,
      "num_tokens": 34711112.0,
      "step": 18615
    },
    {
      "entropy": 5.888959836959839,
      "epoch": 1.6442953020134228,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047318513586872766,
      "loss": 5.5869,
      "mean_token_accuracy": 0.18972935676574706,
      "num_tokens": 34720429.0,
      "step": 18620
    },
    {
      "entropy": 5.8986505508422855,
      "epoch": 1.6447368421052633,
      "grad_norm": 1.8671875,
      "learning_rate": 0.000473170224398853,
      "loss": 5.5994,
      "mean_token_accuracy": 0.18355555683374405,
      "num_tokens": 34729422.0,
      "step": 18625
    },
    {
      "entropy": 5.914456129074097,
      "epoch": 1.6451783821971036,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047315530904693643,
      "loss": 5.62,
      "mean_token_accuracy": 0.18128610104322435,
      "num_tokens": 34738231.0,
      "step": 18630
    },
    {
      "entropy": 5.899551057815552,
      "epoch": 1.6456199222889438,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00047314038981327,
      "loss": 5.6739,
      "mean_token_accuracy": 0.17467010766267776,
      "num_tokens": 34747002.0,
      "step": 18635
    },
    {
      "entropy": 5.793607378005982,
      "epoch": 1.6460614623807843,
      "grad_norm": 1.953125,
      "learning_rate": 0.000473125466698146,
      "loss": 5.5239,
      "mean_token_accuracy": 0.1925452917814255,
      "num_tokens": 34756093.0,
      "step": 18640
    },
    {
      "entropy": 5.901175260543823,
      "epoch": 1.6465030024726244,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004731105397018569,
      "loss": 5.6246,
      "mean_token_accuracy": 0.18271643966436385,
      "num_tokens": 34766031.0,
      "step": 18645
    },
    {
      "entropy": 5.943986797332764,
      "epoch": 1.6469445425644649,
      "grad_norm": 1.5,
      "learning_rate": 0.0004730956088246949,
      "loss": 5.7139,
      "mean_token_accuracy": 0.1752950832247734,
      "num_tokens": 34776487.0,
      "step": 18650
    },
    {
      "entropy": 5.929204273223877,
      "epoch": 1.6473860826563052,
      "grad_norm": 4.875,
      "learning_rate": 0.00047308067406695255,
      "loss": 5.5355,
      "mean_token_accuracy": 0.18798275738954545,
      "num_tokens": 34785372.0,
      "step": 18655
    },
    {
      "entropy": 5.8541929721832275,
      "epoch": 1.6478276227481454,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047306573542892257,
      "loss": 5.5493,
      "mean_token_accuracy": 0.18476638495922088,
      "num_tokens": 34795453.0,
      "step": 18660
    },
    {
      "entropy": 5.882027626037598,
      "epoch": 1.648269162839986,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047305079291089735,
      "loss": 5.6499,
      "mean_token_accuracy": 0.18040877133607863,
      "num_tokens": 34804818.0,
      "step": 18665
    },
    {
      "entropy": 5.91747260093689,
      "epoch": 1.6487107029318262,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00047303584651316985,
      "loss": 5.6829,
      "mean_token_accuracy": 0.17935716062784196,
      "num_tokens": 34814031.0,
      "step": 18670
    },
    {
      "entropy": 5.875995302200318,
      "epoch": 1.6491522430236665,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004730208962360327,
      "loss": 5.5571,
      "mean_token_accuracy": 0.1829606145620346,
      "num_tokens": 34823560.0,
      "step": 18675
    },
    {
      "entropy": 5.9430419445037845,
      "epoch": 1.649593783115507,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004730059420797788,
      "loss": 5.5865,
      "mean_token_accuracy": 0.181386736035347,
      "num_tokens": 34833662.0,
      "step": 18680
    },
    {
      "entropy": 5.951020908355713,
      "epoch": 1.6500353232073472,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004729909840447011,
      "loss": 5.7155,
      "mean_token_accuracy": 0.1703169196844101,
      "num_tokens": 34844264.0,
      "step": 18685
    },
    {
      "entropy": 5.9538764476776125,
      "epoch": 1.6504768632991875,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047297602213109265,
      "loss": 5.6006,
      "mean_token_accuracy": 0.1860039010643959,
      "num_tokens": 34853708.0,
      "step": 18690
    },
    {
      "entropy": 5.970012760162353,
      "epoch": 1.650918403391028,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004729610563392465,
      "loss": 5.6766,
      "mean_token_accuracy": 0.17713891267776488,
      "num_tokens": 34861926.0,
      "step": 18695
    },
    {
      "entropy": 5.819530725479126,
      "epoch": 1.6513599434828683,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004729460866694558,
      "loss": 5.5942,
      "mean_token_accuracy": 0.18716664612293243,
      "num_tokens": 34870919.0,
      "step": 18700
    },
    {
      "entropy": 5.878299713134766,
      "epoch": 1.6518014835747086,
      "grad_norm": 1.8203125,
      "learning_rate": 0.000472931113122014,
      "loss": 5.6269,
      "mean_token_accuracy": 0.1832582339644432,
      "num_tokens": 34879801.0,
      "step": 18705
    },
    {
      "entropy": 5.931321096420288,
      "epoch": 1.652243023666549,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004729161356972141,
      "loss": 5.6144,
      "mean_token_accuracy": 0.17733223289251326,
      "num_tokens": 34888605.0,
      "step": 18710
    },
    {
      "entropy": 5.922053575515747,
      "epoch": 1.6526845637583891,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047290115439534977,
      "loss": 5.6812,
      "mean_token_accuracy": 0.17114962339401246,
      "num_tokens": 34899294.0,
      "step": 18715
    },
    {
      "entropy": 5.95401816368103,
      "epoch": 1.6531261038502296,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004728861692167143,
      "loss": 5.6936,
      "mean_token_accuracy": 0.1761074557900429,
      "num_tokens": 34908624.0,
      "step": 18720
    },
    {
      "entropy": 5.9244489669799805,
      "epoch": 1.65356764394207,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004728711801616013,
      "loss": 5.5341,
      "mean_token_accuracy": 0.18459235578775407,
      "num_tokens": 34917390.0,
      "step": 18725
    },
    {
      "entropy": 5.9459198951721195,
      "epoch": 1.6540091840339102,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004728561872303045,
      "loss": 5.6406,
      "mean_token_accuracy": 0.17514788806438447,
      "num_tokens": 34926652.0,
      "step": 18730
    },
    {
      "entropy": 5.899163675308228,
      "epoch": 1.6544507241257507,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004728411904231175,
      "loss": 5.6752,
      "mean_token_accuracy": 0.17545907348394393,
      "num_tokens": 34936900.0,
      "step": 18735
    },
    {
      "entropy": 5.895047187805176,
      "epoch": 1.654892264217591,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004728261897403341,
      "loss": 5.6615,
      "mean_token_accuracy": 0.17931931614875793,
      "num_tokens": 34946827.0,
      "step": 18740
    },
    {
      "entropy": 5.983937072753906,
      "epoch": 1.6553338043094312,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047281118518224807,
      "loss": 5.6896,
      "mean_token_accuracy": 0.179614120721817,
      "num_tokens": 34955987.0,
      "step": 18745
    },
    {
      "entropy": 5.885171270370483,
      "epoch": 1.6557753444012717,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004727961767491535,
      "loss": 5.5726,
      "mean_token_accuracy": 0.1855631560087204,
      "num_tokens": 34965001.0,
      "step": 18750
    },
    {
      "entropy": 5.877617979049683,
      "epoch": 1.656216884493112,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004727811644413443,
      "loss": 5.6532,
      "mean_token_accuracy": 0.1787920832633972,
      "num_tokens": 34974113.0,
      "step": 18755
    },
    {
      "entropy": 5.843439435958862,
      "epoch": 1.6566584245849523,
      "grad_norm": 2.75,
      "learning_rate": 0.0004727661482591146,
      "loss": 5.5291,
      "mean_token_accuracy": 0.19087912291288375,
      "num_tokens": 34982535.0,
      "step": 18760
    },
    {
      "entropy": 5.876902437210083,
      "epoch": 1.6570999646767928,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004727511282027585,
      "loss": 5.634,
      "mean_token_accuracy": 0.1806099683046341,
      "num_tokens": 34991828.0,
      "step": 18765
    },
    {
      "entropy": 5.8969403266906735,
      "epoch": 1.657541504768633,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004727361042725703,
      "loss": 5.5828,
      "mean_token_accuracy": 0.18390600830316545,
      "num_tokens": 35001670.0,
      "step": 18770
    },
    {
      "entropy": 5.887630891799927,
      "epoch": 1.6579830448604733,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004727210764688442,
      "loss": 5.5514,
      "mean_token_accuracy": 0.18555099070072173,
      "num_tokens": 35011407.0,
      "step": 18775
    },
    {
      "entropy": 5.790446472167969,
      "epoch": 1.6584245849523138,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004727060447918748,
      "loss": 5.5,
      "mean_token_accuracy": 0.19463225603103637,
      "num_tokens": 35020753.0,
      "step": 18780
    },
    {
      "entropy": 5.913278675079345,
      "epoch": 1.6588661250441539,
      "grad_norm": 1.734375,
      "learning_rate": 0.00047269100924195623,
      "loss": 5.671,
      "mean_token_accuracy": 0.17666654288768768,
      "num_tokens": 35030599.0,
      "step": 18785
    },
    {
      "entropy": 5.919923162460327,
      "epoch": 1.6593076651359944,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047267596981938337,
      "loss": 5.6576,
      "mean_token_accuracy": 0.17745146453380584,
      "num_tokens": 35039098.0,
      "step": 18790
    },
    {
      "entropy": 5.976593255996704,
      "epoch": 1.6597492052278346,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047266092652445066,
      "loss": 5.6343,
      "mean_token_accuracy": 0.17812537550926208,
      "num_tokens": 35048286.0,
      "step": 18795
    },
    {
      "entropy": 6.001517629623413,
      "epoch": 1.660190745319675,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004726458793574528,
      "loss": 5.6865,
      "mean_token_accuracy": 0.17537646144628524,
      "num_tokens": 35057331.0,
      "step": 18800
    },
    {
      "entropy": 5.877080345153809,
      "epoch": 1.6606322854115154,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004726308283186846,
      "loss": 5.6377,
      "mean_token_accuracy": 0.17719049453735353,
      "num_tokens": 35066321.0,
      "step": 18805
    },
    {
      "entropy": 5.844009590148926,
      "epoch": 1.6610738255033557,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004726157734084409,
      "loss": 5.57,
      "mean_token_accuracy": 0.18537749499082565,
      "num_tokens": 35075218.0,
      "step": 18810
    },
    {
      "entropy": 5.916835355758667,
      "epoch": 1.661515365595196,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004726007146270164,
      "loss": 5.6323,
      "mean_token_accuracy": 0.1819580912590027,
      "num_tokens": 35084563.0,
      "step": 18815
    },
    {
      "entropy": 5.838687372207642,
      "epoch": 1.6619569056870365,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004725856519747065,
      "loss": 5.5414,
      "mean_token_accuracy": 0.18808421194553376,
      "num_tokens": 35093346.0,
      "step": 18820
    },
    {
      "entropy": 5.86430253982544,
      "epoch": 1.6623984457788767,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004725705854518059,
      "loss": 5.6134,
      "mean_token_accuracy": 0.1819564774632454,
      "num_tokens": 35102501.0,
      "step": 18825
    },
    {
      "entropy": 5.8794506072998045,
      "epoch": 1.662839985870717,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047255551505861007,
      "loss": 5.6184,
      "mean_token_accuracy": 0.1769965574145317,
      "num_tokens": 35112083.0,
      "step": 18830
    },
    {
      "entropy": 5.8454474925994875,
      "epoch": 1.6632815259625575,
      "grad_norm": 1.5,
      "learning_rate": 0.00047254044079541387,
      "loss": 5.5478,
      "mean_token_accuracy": 0.18820382356643678,
      "num_tokens": 35121380.0,
      "step": 18835
    },
    {
      "entropy": 5.9076399326324465,
      "epoch": 1.6637230660543978,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004725253626625129,
      "loss": 5.6763,
      "mean_token_accuracy": 0.1739080458879471,
      "num_tokens": 35130083.0,
      "step": 18840
    },
    {
      "entropy": 5.936508131027222,
      "epoch": 1.664164606146238,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047251028066020237,
      "loss": 5.5562,
      "mean_token_accuracy": 0.18594707399606705,
      "num_tokens": 35139001.0,
      "step": 18845
    },
    {
      "entropy": 5.85218300819397,
      "epoch": 1.6646061462380786,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047249519478877774,
      "loss": 5.5069,
      "mean_token_accuracy": 0.18826766014099122,
      "num_tokens": 35147701.0,
      "step": 18850
    },
    {
      "entropy": 5.795539093017578,
      "epoch": 1.6650476863299186,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004724801050485346,
      "loss": 5.5307,
      "mean_token_accuracy": 0.19010263681411743,
      "num_tokens": 35157334.0,
      "step": 18855
    },
    {
      "entropy": 5.874148035049439,
      "epoch": 1.6654892264217591,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004724650114397685,
      "loss": 5.5588,
      "mean_token_accuracy": 0.1873392328619957,
      "num_tokens": 35166647.0,
      "step": 18860
    },
    {
      "entropy": 5.9134190559387205,
      "epoch": 1.6659307665135994,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047244991396277515,
      "loss": 5.5582,
      "mean_token_accuracy": 0.17858651131391526,
      "num_tokens": 35176345.0,
      "step": 18865
    },
    {
      "entropy": 5.897709131240845,
      "epoch": 1.6663723066054397,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047243481261785027,
      "loss": 5.5687,
      "mean_token_accuracy": 0.18713643699884414,
      "num_tokens": 35185793.0,
      "step": 18870
    },
    {
      "entropy": 5.881582260131836,
      "epoch": 1.6668138466972802,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047241970740528975,
      "loss": 5.6537,
      "mean_token_accuracy": 0.18241412043571473,
      "num_tokens": 35195905.0,
      "step": 18875
    },
    {
      "entropy": 5.808372116088867,
      "epoch": 1.6672553867891204,
      "grad_norm": 1.625,
      "learning_rate": 0.00047240459832538934,
      "loss": 5.5053,
      "mean_token_accuracy": 0.19018662124872207,
      "num_tokens": 35204340.0,
      "step": 18880
    },
    {
      "entropy": 5.870752573013306,
      "epoch": 1.6676969268809607,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004723894853784452,
      "loss": 5.609,
      "mean_token_accuracy": 0.17730844765901566,
      "num_tokens": 35213270.0,
      "step": 18885
    },
    {
      "entropy": 5.869860982894897,
      "epoch": 1.6681384669728012,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004723743685647533,
      "loss": 5.5944,
      "mean_token_accuracy": 0.18141066133975983,
      "num_tokens": 35222652.0,
      "step": 18890
    },
    {
      "entropy": 5.915582132339478,
      "epoch": 1.6685800070646415,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004723592478846097,
      "loss": 5.5856,
      "mean_token_accuracy": 0.18400098830461503,
      "num_tokens": 35231684.0,
      "step": 18895
    },
    {
      "entropy": 5.885527896881103,
      "epoch": 1.6690215471564818,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047234412333831073,
      "loss": 5.6703,
      "mean_token_accuracy": 0.17749470174312593,
      "num_tokens": 35241589.0,
      "step": 18900
    },
    {
      "entropy": 6.031778335571289,
      "epoch": 1.6694630872483223,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004723289949261527,
      "loss": 5.7521,
      "mean_token_accuracy": 0.16931313574314116,
      "num_tokens": 35250903.0,
      "step": 18905
    },
    {
      "entropy": 5.940570974349976,
      "epoch": 1.6699046273401625,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004723138626484318,
      "loss": 5.6116,
      "mean_token_accuracy": 0.18052135407924652,
      "num_tokens": 35260556.0,
      "step": 18910
    },
    {
      "entropy": 5.7749754905700685,
      "epoch": 1.6703461674320028,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047229872650544446,
      "loss": 5.5074,
      "mean_token_accuracy": 0.19481605738401414,
      "num_tokens": 35270237.0,
      "step": 18915
    },
    {
      "entropy": 5.786803293228149,
      "epoch": 1.6707877075238433,
      "grad_norm": 1.84375,
      "learning_rate": 0.00047228358649748745,
      "loss": 5.5292,
      "mean_token_accuracy": 0.19226605594158172,
      "num_tokens": 35279017.0,
      "step": 18920
    },
    {
      "entropy": 5.842459630966187,
      "epoch": 1.6712292476156834,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004722684426248571,
      "loss": 5.5741,
      "mean_token_accuracy": 0.18441891819238662,
      "num_tokens": 35288150.0,
      "step": 18925
    },
    {
      "entropy": 5.868122625350952,
      "epoch": 1.6716707877075239,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047225329488785015,
      "loss": 5.6404,
      "mean_token_accuracy": 0.1853303760290146,
      "num_tokens": 35297463.0,
      "step": 18930
    },
    {
      "entropy": 5.890319156646728,
      "epoch": 1.6721123277993641,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047223814328676336,
      "loss": 5.6018,
      "mean_token_accuracy": 0.1801517501473427,
      "num_tokens": 35307409.0,
      "step": 18935
    },
    {
      "entropy": 5.818031978607178,
      "epoch": 1.6725538678912044,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004722229878218935,
      "loss": 5.5533,
      "mean_token_accuracy": 0.18438764065504074,
      "num_tokens": 35315613.0,
      "step": 18940
    },
    {
      "entropy": 5.851502513885498,
      "epoch": 1.672995407983045,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047220782849353743,
      "loss": 5.6126,
      "mean_token_accuracy": 0.18252478390932084,
      "num_tokens": 35325297.0,
      "step": 18945
    },
    {
      "entropy": 5.90099310874939,
      "epoch": 1.6734369480748852,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004721926653019922,
      "loss": 5.594,
      "mean_token_accuracy": 0.1777297481894493,
      "num_tokens": 35334014.0,
      "step": 18950
    },
    {
      "entropy": 5.88225154876709,
      "epoch": 1.6738784881667255,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004721774982475548,
      "loss": 5.5979,
      "mean_token_accuracy": 0.18687061816453934,
      "num_tokens": 35342784.0,
      "step": 18955
    },
    {
      "entropy": 5.937609767913818,
      "epoch": 1.674320028258566,
      "grad_norm": 1.75,
      "learning_rate": 0.00047216232733052237,
      "loss": 5.7191,
      "mean_token_accuracy": 0.17852629721164703,
      "num_tokens": 35351985.0,
      "step": 18960
    },
    {
      "entropy": 5.924067687988281,
      "epoch": 1.6747615683504062,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004721471525511921,
      "loss": 5.5679,
      "mean_token_accuracy": 0.1810382068157196,
      "num_tokens": 35360857.0,
      "step": 18965
    },
    {
      "entropy": 5.873107051849365,
      "epoch": 1.6752031084422465,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004721319739098611,
      "loss": 5.5615,
      "mean_token_accuracy": 0.1860722690820694,
      "num_tokens": 35370200.0,
      "step": 18970
    },
    {
      "entropy": 5.878828287124634,
      "epoch": 1.675644648534087,
      "grad_norm": 1.28125,
      "learning_rate": 0.000472116791406827,
      "loss": 5.6409,
      "mean_token_accuracy": 0.18122011423110962,
      "num_tokens": 35379481.0,
      "step": 18975
    },
    {
      "entropy": 5.822802543640137,
      "epoch": 1.6760861886259273,
      "grad_norm": 1.421875,
      "learning_rate": 0.000472101605042387,
      "loss": 5.572,
      "mean_token_accuracy": 0.1930101618170738,
      "num_tokens": 35387951.0,
      "step": 18980
    },
    {
      "entropy": 5.98285059928894,
      "epoch": 1.6765277287177676,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004720864148168387,
      "loss": 5.6061,
      "mean_token_accuracy": 0.17831956297159196,
      "num_tokens": 35396200.0,
      "step": 18985
    },
    {
      "entropy": 5.909352827072143,
      "epoch": 1.676969268809608,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004720712207304796,
      "loss": 5.6095,
      "mean_token_accuracy": 0.17886189222335816,
      "num_tokens": 35404474.0,
      "step": 18990
    },
    {
      "entropy": 5.915679454803467,
      "epoch": 1.6774108089014481,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004720560227836073,
      "loss": 5.7347,
      "mean_token_accuracy": 0.17303581833839415,
      "num_tokens": 35414419.0,
      "step": 18995
    },
    {
      "entropy": 5.879295539855957,
      "epoch": 1.6778523489932886,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004720408209765197,
      "loss": 5.5552,
      "mean_token_accuracy": 0.1843546837568283,
      "num_tokens": 35423727.0,
      "step": 19000
    },
    {
      "entropy": 5.8008311748504635,
      "epoch": 1.6782938890851289,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047202561530951443,
      "loss": 5.4929,
      "mean_token_accuracy": 0.19395811557769777,
      "num_tokens": 35433364.0,
      "step": 19005
    },
    {
      "entropy": 5.957615852355957,
      "epoch": 1.6787354291769692,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004720104057828894,
      "loss": 5.6569,
      "mean_token_accuracy": 0.17231166660785674,
      "num_tokens": 35442665.0,
      "step": 19010
    },
    {
      "entropy": 5.817198753356934,
      "epoch": 1.6791769692688097,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047199519239694257,
      "loss": 5.5557,
      "mean_token_accuracy": 0.18612952083349227,
      "num_tokens": 35451443.0,
      "step": 19015
    },
    {
      "entropy": 5.817591857910156,
      "epoch": 1.67961850936065,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047197997515197195,
      "loss": 5.5029,
      "mean_token_accuracy": 0.19969430267810823,
      "num_tokens": 35460622.0,
      "step": 19020
    },
    {
      "entropy": 5.812362194061279,
      "epoch": 1.6800600494524902,
      "grad_norm": 1.71875,
      "learning_rate": 0.00047196475404827565,
      "loss": 5.6546,
      "mean_token_accuracy": 0.1852271556854248,
      "num_tokens": 35470345.0,
      "step": 19025
    },
    {
      "entropy": 5.948154163360596,
      "epoch": 1.6805015895443307,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047194952908615183,
      "loss": 5.7333,
      "mean_token_accuracy": 0.17560504823923112,
      "num_tokens": 35480460.0,
      "step": 19030
    },
    {
      "entropy": 5.933764028549194,
      "epoch": 1.680943129636171,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004719343002658987,
      "loss": 5.6641,
      "mean_token_accuracy": 0.1781614303588867,
      "num_tokens": 35489983.0,
      "step": 19035
    },
    {
      "entropy": 5.879894781112671,
      "epoch": 1.6813846697280113,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047191906758781475,
      "loss": 5.5839,
      "mean_token_accuracy": 0.18786992877721786,
      "num_tokens": 35498963.0,
      "step": 19040
    },
    {
      "entropy": 5.870029067993164,
      "epoch": 1.6818262098198518,
      "grad_norm": 1.75,
      "learning_rate": 0.0004719038310521982,
      "loss": 5.6343,
      "mean_token_accuracy": 0.18965273052453996,
      "num_tokens": 35509054.0,
      "step": 19045
    },
    {
      "entropy": 5.93992018699646,
      "epoch": 1.682267749911692,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004718885906593476,
      "loss": 5.7467,
      "mean_token_accuracy": 0.168730728328228,
      "num_tokens": 35518093.0,
      "step": 19050
    },
    {
      "entropy": 5.968710803985596,
      "epoch": 1.6827092900035323,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004718733464095614,
      "loss": 5.6357,
      "mean_token_accuracy": 0.1828368693590164,
      "num_tokens": 35527330.0,
      "step": 19055
    },
    {
      "entropy": 5.9040357112884525,
      "epoch": 1.6831508300953728,
      "grad_norm": 1.625,
      "learning_rate": 0.00047185809830313844,
      "loss": 5.6315,
      "mean_token_accuracy": 0.1740979105234146,
      "num_tokens": 35538147.0,
      "step": 19060
    },
    {
      "entropy": 5.941530466079712,
      "epoch": 1.6835923701872129,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00047184284634037727,
      "loss": 5.7111,
      "mean_token_accuracy": 0.17096765339374542,
      "num_tokens": 35547930.0,
      "step": 19065
    },
    {
      "entropy": 5.864803838729858,
      "epoch": 1.6840339102790534,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047182759052157665,
      "loss": 5.5246,
      "mean_token_accuracy": 0.1837114378809929,
      "num_tokens": 35557111.0,
      "step": 19070
    },
    {
      "entropy": 5.8556122303009035,
      "epoch": 1.6844754503708936,
      "grad_norm": 1.375,
      "learning_rate": 0.0004718123308470355,
      "loss": 5.6991,
      "mean_token_accuracy": 0.1753170132637024,
      "num_tokens": 35565739.0,
      "step": 19075
    },
    {
      "entropy": 5.873278331756592,
      "epoch": 1.684916990462734,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00047179706731705263,
      "loss": 5.5488,
      "mean_token_accuracy": 0.18050973117351532,
      "num_tokens": 35574164.0,
      "step": 19080
    },
    {
      "entropy": 5.905177211761474,
      "epoch": 1.6853585305545744,
      "grad_norm": 1.875,
      "learning_rate": 0.00047178179993192723,
      "loss": 5.601,
      "mean_token_accuracy": 0.17910588383674622,
      "num_tokens": 35582813.0,
      "step": 19085
    },
    {
      "entropy": 5.939931201934814,
      "epoch": 1.6858000706464147,
      "grad_norm": 1.71875,
      "learning_rate": 0.00047176652869195825,
      "loss": 5.577,
      "mean_token_accuracy": 0.17906726747751237,
      "num_tokens": 35591884.0,
      "step": 19090
    },
    {
      "entropy": 5.9027632713317875,
      "epoch": 1.686241610738255,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047175125359744487,
      "loss": 5.6129,
      "mean_token_accuracy": 0.17895587384700776,
      "num_tokens": 35601824.0,
      "step": 19095
    },
    {
      "entropy": 5.85024094581604,
      "epoch": 1.6866831508300955,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047173597464868634,
      "loss": 5.5453,
      "mean_token_accuracy": 0.18780640214681626,
      "num_tokens": 35610756.0,
      "step": 19100
    },
    {
      "entropy": 5.795286512374878,
      "epoch": 1.6871246909219357,
      "grad_norm": 1.59375,
      "learning_rate": 0.000471720691845982,
      "loss": 5.5688,
      "mean_token_accuracy": 0.18757421672344207,
      "num_tokens": 35620160.0,
      "step": 19105
    },
    {
      "entropy": 5.92302622795105,
      "epoch": 1.687566231013776,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047170540518963114,
      "loss": 5.7123,
      "mean_token_accuracy": 0.17004642486572266,
      "num_tokens": 35630487.0,
      "step": 19110
    },
    {
      "entropy": 6.021547937393189,
      "epoch": 1.6880077711056165,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004716901146799333,
      "loss": 5.7717,
      "mean_token_accuracy": 0.16779865026474,
      "num_tokens": 35641924.0,
      "step": 19115
    },
    {
      "entropy": 5.9363110065460205,
      "epoch": 1.6884493111974568,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004716748203171879,
      "loss": 5.5787,
      "mean_token_accuracy": 0.17523247748613358,
      "num_tokens": 35651575.0,
      "step": 19120
    },
    {
      "entropy": 5.818540096282959,
      "epoch": 1.688890851289297,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047165952210169475,
      "loss": 5.5731,
      "mean_token_accuracy": 0.17990154772996902,
      "num_tokens": 35661481.0,
      "step": 19125
    },
    {
      "entropy": 5.7415002346038815,
      "epoch": 1.6893323913811376,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004716442200337533,
      "loss": 5.5553,
      "mean_token_accuracy": 0.18225292414426802,
      "num_tokens": 35672049.0,
      "step": 19130
    },
    {
      "entropy": 5.899270868301391,
      "epoch": 1.6897739314729776,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047162891411366343,
      "loss": 5.614,
      "mean_token_accuracy": 0.18708101511001587,
      "num_tokens": 35681653.0,
      "step": 19135
    },
    {
      "entropy": 5.946601295471192,
      "epoch": 1.690215471564818,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047161360434172506,
      "loss": 5.6588,
      "mean_token_accuracy": 0.18474891632795334,
      "num_tokens": 35690868.0,
      "step": 19140
    },
    {
      "entropy": 5.866607522964477,
      "epoch": 1.6906570116566584,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004715982907182379,
      "loss": 5.5714,
      "mean_token_accuracy": 0.1854156255722046,
      "num_tokens": 35699062.0,
      "step": 19145
    },
    {
      "entropy": 5.787081956863403,
      "epoch": 1.6910985517484987,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004715829732435021,
      "loss": 5.6461,
      "mean_token_accuracy": 0.17780405282974243,
      "num_tokens": 35708411.0,
      "step": 19150
    },
    {
      "entropy": 5.817020988464355,
      "epoch": 1.6915400918403392,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047156765191781763,
      "loss": 5.5513,
      "mean_token_accuracy": 0.18702746629714967,
      "num_tokens": 35717937.0,
      "step": 19155
    },
    {
      "entropy": 5.9095778465271,
      "epoch": 1.6919816319321794,
      "grad_norm": 1.375,
      "learning_rate": 0.00047155232674148465,
      "loss": 5.5609,
      "mean_token_accuracy": 0.18352209329605101,
      "num_tokens": 35727712.0,
      "step": 19160
    },
    {
      "entropy": 5.944918251037597,
      "epoch": 1.6924231720240197,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047153699771480335,
      "loss": 5.6529,
      "mean_token_accuracy": 0.18063961416482927,
      "num_tokens": 35737076.0,
      "step": 19165
    },
    {
      "entropy": 5.850956439971924,
      "epoch": 1.6928647121158602,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000471521664838074,
      "loss": 5.4659,
      "mean_token_accuracy": 0.18905059695243837,
      "num_tokens": 35746436.0,
      "step": 19170
    },
    {
      "entropy": 5.865803909301758,
      "epoch": 1.6933062522077005,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004715063281115971,
      "loss": 5.6537,
      "mean_token_accuracy": 0.18291245996952057,
      "num_tokens": 35755786.0,
      "step": 19175
    },
    {
      "entropy": 5.908975219726562,
      "epoch": 1.6937477922995408,
      "grad_norm": 1.546875,
      "learning_rate": 0.000471490987535673,
      "loss": 5.6364,
      "mean_token_accuracy": 0.1808708056807518,
      "num_tokens": 35764424.0,
      "step": 19180
    },
    {
      "entropy": 5.976862096786499,
      "epoch": 1.6941893323913813,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047147564311060216,
      "loss": 5.6109,
      "mean_token_accuracy": 0.18163918554782868,
      "num_tokens": 35773792.0,
      "step": 19185
    },
    {
      "entropy": 5.820072841644287,
      "epoch": 1.6946308724832215,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047146029483668517,
      "loss": 5.561,
      "mean_token_accuracy": 0.18100957721471786,
      "num_tokens": 35783285.0,
      "step": 19190
    },
    {
      "entropy": 5.86439471244812,
      "epoch": 1.6950724125750618,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004714449427142228,
      "loss": 5.6832,
      "mean_token_accuracy": 0.17643289417028427,
      "num_tokens": 35792476.0,
      "step": 19195
    },
    {
      "entropy": 5.936539936065674,
      "epoch": 1.6955139526669023,
      "grad_norm": 2.0,
      "learning_rate": 0.00047142958674351563,
      "loss": 5.5913,
      "mean_token_accuracy": 0.18860094398260116,
      "num_tokens": 35801640.0,
      "step": 19200
    },
    {
      "entropy": 5.910218143463135,
      "epoch": 1.6959554927587424,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047141422692486466,
      "loss": 5.6262,
      "mean_token_accuracy": 0.1802951768040657,
      "num_tokens": 35811073.0,
      "step": 19205
    },
    {
      "entropy": 5.864576435089111,
      "epoch": 1.6963970328505829,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047139886325857074,
      "loss": 5.5524,
      "mean_token_accuracy": 0.18887008279561995,
      "num_tokens": 35819814.0,
      "step": 19210
    },
    {
      "entropy": 5.872750520706177,
      "epoch": 1.6968385729424231,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004713834957449347,
      "loss": 5.5588,
      "mean_token_accuracy": 0.18682279139757157,
      "num_tokens": 35828363.0,
      "step": 19215
    },
    {
      "entropy": 5.857779216766358,
      "epoch": 1.6972801130342634,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004713681243842577,
      "loss": 5.5247,
      "mean_token_accuracy": 0.1891833171248436,
      "num_tokens": 35836868.0,
      "step": 19220
    },
    {
      "entropy": 5.900493001937866,
      "epoch": 1.697721653126104,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047135274917684087,
      "loss": 5.6187,
      "mean_token_accuracy": 0.18207525610923767,
      "num_tokens": 35846030.0,
      "step": 19225
    },
    {
      "entropy": 5.8308618545532225,
      "epoch": 1.6981631932179442,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004713373701229852,
      "loss": 5.4926,
      "mean_token_accuracy": 0.1902054503560066,
      "num_tokens": 35854862.0,
      "step": 19230
    },
    {
      "entropy": 5.8233579158782955,
      "epoch": 1.6986047333097845,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004713219872229923,
      "loss": 5.5382,
      "mean_token_accuracy": 0.18843249082565308,
      "num_tokens": 35864238.0,
      "step": 19235
    },
    {
      "entropy": 5.870490169525146,
      "epoch": 1.699046273401625,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004713066004771633,
      "loss": 5.541,
      "mean_token_accuracy": 0.18423995226621628,
      "num_tokens": 35872542.0,
      "step": 19240
    },
    {
      "entropy": 5.926501941680908,
      "epoch": 1.6994878134934652,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047129120988579953,
      "loss": 5.6451,
      "mean_token_accuracy": 0.17808191031217574,
      "num_tokens": 35881899.0,
      "step": 19245
    },
    {
      "entropy": 5.9293928146362305,
      "epoch": 1.6999293535853055,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00047127581544920263,
      "loss": 5.6257,
      "mean_token_accuracy": 0.1778813198208809,
      "num_tokens": 35891970.0,
      "step": 19250
    },
    {
      "entropy": 5.823053455352783,
      "epoch": 1.700370893677146,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004712604171676742,
      "loss": 5.5124,
      "mean_token_accuracy": 0.1820422187447548,
      "num_tokens": 35901286.0,
      "step": 19255
    },
    {
      "entropy": 5.812740898132324,
      "epoch": 1.7008124337689863,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00047124501504151574,
      "loss": 5.7293,
      "mean_token_accuracy": 0.17866381257772446,
      "num_tokens": 35912301.0,
      "step": 19260
    },
    {
      "entropy": 5.932995843887329,
      "epoch": 1.7012539738608266,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00047122960907102906,
      "loss": 5.6044,
      "mean_token_accuracy": 0.1805946260690689,
      "num_tokens": 35920767.0,
      "step": 19265
    },
    {
      "entropy": 5.947637128829956,
      "epoch": 1.701695513952667,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000471214199256516,
      "loss": 5.6981,
      "mean_token_accuracy": 0.1797356978058815,
      "num_tokens": 35930235.0,
      "step": 19270
    },
    {
      "entropy": 5.9566267967224125,
      "epoch": 1.702137054044507,
      "grad_norm": 1.703125,
      "learning_rate": 0.00047119878559827825,
      "loss": 5.6158,
      "mean_token_accuracy": 0.1753778800368309,
      "num_tokens": 35939213.0,
      "step": 19275
    },
    {
      "entropy": 5.952847671508789,
      "epoch": 1.7025785941363476,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047118336809661793,
      "loss": 5.6868,
      "mean_token_accuracy": 0.1693960189819336,
      "num_tokens": 35948221.0,
      "step": 19280
    },
    {
      "entropy": 5.916274547576904,
      "epoch": 1.7030201342281879,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004711679467518371,
      "loss": 5.6106,
      "mean_token_accuracy": 0.17386380732059478,
      "num_tokens": 35957961.0,
      "step": 19285
    },
    {
      "entropy": 5.903388166427613,
      "epoch": 1.7034616743200282,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047115252156423767,
      "loss": 5.6474,
      "mean_token_accuracy": 0.17280495762825013,
      "num_tokens": 35966742.0,
      "step": 19290
    },
    {
      "entropy": 5.910114431381226,
      "epoch": 1.7039032144118686,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004711370925341219,
      "loss": 5.5923,
      "mean_token_accuracy": 0.1781558722257614,
      "num_tokens": 35976312.0,
      "step": 19295
    },
    {
      "entropy": 5.888975477218628,
      "epoch": 1.704344754503709,
      "grad_norm": 1.5,
      "learning_rate": 0.00047112165966179206,
      "loss": 5.6671,
      "mean_token_accuracy": 0.1848137319087982,
      "num_tokens": 35985537.0,
      "step": 19300
    },
    {
      "entropy": 5.929999494552613,
      "epoch": 1.7047862945955492,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047110622294755045,
      "loss": 5.596,
      "mean_token_accuracy": 0.18230962455272676,
      "num_tokens": 35995004.0,
      "step": 19305
    },
    {
      "entropy": 5.882706832885742,
      "epoch": 1.7052278346873897,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047109078239169946,
      "loss": 5.6216,
      "mean_token_accuracy": 0.17474031895399095,
      "num_tokens": 36004786.0,
      "step": 19310
    },
    {
      "entropy": 5.820213413238525,
      "epoch": 1.70566937477923,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004710753379945415,
      "loss": 5.5542,
      "mean_token_accuracy": 0.19274654984474182,
      "num_tokens": 36014047.0,
      "step": 19315
    },
    {
      "entropy": 5.932874631881714,
      "epoch": 1.7061109148710702,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047105988975637927,
      "loss": 5.5814,
      "mean_token_accuracy": 0.18859634697437286,
      "num_tokens": 36022921.0,
      "step": 19320
    },
    {
      "entropy": 5.854664087295532,
      "epoch": 1.7065524549629107,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004710444376775153,
      "loss": 5.5876,
      "mean_token_accuracy": 0.17929477095603943,
      "num_tokens": 36032713.0,
      "step": 19325
    },
    {
      "entropy": 5.825881767272949,
      "epoch": 1.706993995054751,
      "grad_norm": 3.0625,
      "learning_rate": 0.00047102898175825224,
      "loss": 5.5851,
      "mean_token_accuracy": 0.17754208892583848,
      "num_tokens": 36041595.0,
      "step": 19330
    },
    {
      "entropy": 5.842580795288086,
      "epoch": 1.7074355351465913,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047101352199889295,
      "loss": 5.5841,
      "mean_token_accuracy": 0.17971374541521073,
      "num_tokens": 36051993.0,
      "step": 19335
    },
    {
      "entropy": 5.868608713150024,
      "epoch": 1.7078770752384318,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004709980583997402,
      "loss": 5.5849,
      "mean_token_accuracy": 0.18636864721775054,
      "num_tokens": 36061806.0,
      "step": 19340
    },
    {
      "entropy": 5.908734226226807,
      "epoch": 1.7083186153302719,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000470982590961097,
      "loss": 5.6311,
      "mean_token_accuracy": 0.17739569991827012,
      "num_tokens": 36071390.0,
      "step": 19345
    },
    {
      "entropy": 5.895809364318848,
      "epoch": 1.7087601554221123,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047096711968326626,
      "loss": 5.6184,
      "mean_token_accuracy": 0.18271129578351974,
      "num_tokens": 36080561.0,
      "step": 19350
    },
    {
      "entropy": 5.880069875717163,
      "epoch": 1.7092016955139526,
      "grad_norm": 1.75,
      "learning_rate": 0.0004709516445665512,
      "loss": 5.5524,
      "mean_token_accuracy": 0.17658421248197556,
      "num_tokens": 36089568.0,
      "step": 19355
    },
    {
      "entropy": 5.847279167175293,
      "epoch": 1.709643235605793,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004709361656112547,
      "loss": 5.5347,
      "mean_token_accuracy": 0.19057597517967223,
      "num_tokens": 36098881.0,
      "step": 19360
    },
    {
      "entropy": 5.864730882644653,
      "epoch": 1.7100847756976334,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004709206828176804,
      "loss": 5.5812,
      "mean_token_accuracy": 0.18632197231054307,
      "num_tokens": 36108209.0,
      "step": 19365
    },
    {
      "entropy": 5.856773233413696,
      "epoch": 1.7105263157894737,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004709051961861311,
      "loss": 5.6255,
      "mean_token_accuracy": 0.1823357731103897,
      "num_tokens": 36117573.0,
      "step": 19370
    },
    {
      "entropy": 5.884512186050415,
      "epoch": 1.710967855881314,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004708897057169105,
      "loss": 5.6289,
      "mean_token_accuracy": 0.1795087829232216,
      "num_tokens": 36127405.0,
      "step": 19375
    },
    {
      "entropy": 5.951234006881714,
      "epoch": 1.7114093959731544,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004708742114103221,
      "loss": 5.6083,
      "mean_token_accuracy": 0.1816836878657341,
      "num_tokens": 36136148.0,
      "step": 19380
    },
    {
      "entropy": 5.836689805984497,
      "epoch": 1.7118509360649947,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004708587132666692,
      "loss": 5.604,
      "mean_token_accuracy": 0.1854963555932045,
      "num_tokens": 36146164.0,
      "step": 19385
    },
    {
      "entropy": 5.934586524963379,
      "epoch": 1.712292476156835,
      "grad_norm": 3.578125,
      "learning_rate": 0.00047084321128625545,
      "loss": 5.592,
      "mean_token_accuracy": 0.1735032543540001,
      "num_tokens": 36155313.0,
      "step": 19390
    },
    {
      "entropy": 5.907535791397095,
      "epoch": 1.7127340162486755,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047082770546938457,
      "loss": 5.6138,
      "mean_token_accuracy": 0.18404117077589036,
      "num_tokens": 36164902.0,
      "step": 19395
    },
    {
      "entropy": 5.884841537475586,
      "epoch": 1.7131755563405158,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004708121958163604,
      "loss": 5.651,
      "mean_token_accuracy": 0.19012964963912965,
      "num_tokens": 36174128.0,
      "step": 19400
    },
    {
      "entropy": 5.95495023727417,
      "epoch": 1.713617096432356,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004707966823274866,
      "loss": 5.6829,
      "mean_token_accuracy": 0.17704345136880875,
      "num_tokens": 36183084.0,
      "step": 19405
    },
    {
      "entropy": 5.962893295288086,
      "epoch": 1.7140586365241965,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047078116500306723,
      "loss": 5.6281,
      "mean_token_accuracy": 0.18243843019008638,
      "num_tokens": 36191650.0,
      "step": 19410
    },
    {
      "entropy": 5.92274112701416,
      "epoch": 1.7145001766160366,
      "grad_norm": 1.25,
      "learning_rate": 0.00047076564384340616,
      "loss": 5.6169,
      "mean_token_accuracy": 0.17816449403762818,
      "num_tokens": 36200712.0,
      "step": 19415
    },
    {
      "entropy": 5.844231986999512,
      "epoch": 1.714941716707877,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004707501188488074,
      "loss": 5.5931,
      "mean_token_accuracy": 0.18204260766506195,
      "num_tokens": 36209347.0,
      "step": 19420
    },
    {
      "entropy": 5.8398932933807375,
      "epoch": 1.7153832567997174,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047073459001957524,
      "loss": 5.596,
      "mean_token_accuracy": 0.18137114197015763,
      "num_tokens": 36218233.0,
      "step": 19425
    },
    {
      "entropy": 5.927250719070434,
      "epoch": 1.7158247968915576,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004707190573560137,
      "loss": 5.6362,
      "mean_token_accuracy": 0.1832816481590271,
      "num_tokens": 36227627.0,
      "step": 19430
    },
    {
      "entropy": 5.876250743865967,
      "epoch": 1.7162663369833981,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004707035208584272,
      "loss": 5.6098,
      "mean_token_accuracy": 0.18303197920322417,
      "num_tokens": 36236298.0,
      "step": 19435
    },
    {
      "entropy": 5.8805841445922855,
      "epoch": 1.7167078770752384,
      "grad_norm": 1.703125,
      "learning_rate": 0.00047068798052711996,
      "loss": 5.6504,
      "mean_token_accuracy": 0.18288577497005462,
      "num_tokens": 36245623.0,
      "step": 19440
    },
    {
      "entropy": 5.86728048324585,
      "epoch": 1.7171494171670787,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004706724363623965,
      "loss": 5.5285,
      "mean_token_accuracy": 0.1902980700135231,
      "num_tokens": 36255117.0,
      "step": 19445
    },
    {
      "entropy": 5.856943893432617,
      "epoch": 1.7175909572589192,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047065688836456134,
      "loss": 5.584,
      "mean_token_accuracy": 0.18508364111185074,
      "num_tokens": 36264535.0,
      "step": 19450
    },
    {
      "entropy": 5.899629068374634,
      "epoch": 1.7180324973507595,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047064133653391895,
      "loss": 5.6018,
      "mean_token_accuracy": 0.17943939417600632,
      "num_tokens": 36274432.0,
      "step": 19455
    },
    {
      "entropy": 5.843019104003906,
      "epoch": 1.7184740374425997,
      "grad_norm": 2.125,
      "learning_rate": 0.0004706257808707741,
      "loss": 5.5334,
      "mean_token_accuracy": 0.18917024582624437,
      "num_tokens": 36282644.0,
      "step": 19460
    },
    {
      "entropy": 5.835893011093139,
      "epoch": 1.7189155775344402,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047061022137543144,
      "loss": 5.5825,
      "mean_token_accuracy": 0.18336371183395386,
      "num_tokens": 36290885.0,
      "step": 19465
    },
    {
      "entropy": 5.900510787963867,
      "epoch": 1.7193571176262805,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047059465804819576,
      "loss": 5.6792,
      "mean_token_accuracy": 0.17842910289764405,
      "num_tokens": 36299477.0,
      "step": 19470
    },
    {
      "entropy": 5.878560209274292,
      "epoch": 1.7197986577181208,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004705790908893721,
      "loss": 5.6562,
      "mean_token_accuracy": 0.1797110378742218,
      "num_tokens": 36309638.0,
      "step": 19475
    },
    {
      "entropy": 5.9505640983581545,
      "epoch": 1.7202401978099613,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004705635198992651,
      "loss": 5.7122,
      "mean_token_accuracy": 0.17385435104370117,
      "num_tokens": 36319027.0,
      "step": 19480
    },
    {
      "entropy": 5.900506067276001,
      "epoch": 1.7206817379018013,
      "grad_norm": 1.5,
      "learning_rate": 0.00047054794507818,
      "loss": 5.5596,
      "mean_token_accuracy": 0.18011519461870193,
      "num_tokens": 36327936.0,
      "step": 19485
    },
    {
      "entropy": 5.960008335113526,
      "epoch": 1.7211232779936418,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047053236642642193,
      "loss": 5.7519,
      "mean_token_accuracy": 0.17769253849983216,
      "num_tokens": 36337162.0,
      "step": 19490
    },
    {
      "entropy": 5.93083553314209,
      "epoch": 1.7215648180854821,
      "grad_norm": 1.515625,
      "learning_rate": 0.000470516783944296,
      "loss": 5.5982,
      "mean_token_accuracy": 0.18458424210548402,
      "num_tokens": 36345587.0,
      "step": 19495
    },
    {
      "entropy": 5.862644004821777,
      "epoch": 1.7220063581773224,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047050119763210747,
      "loss": 5.6393,
      "mean_token_accuracy": 0.1814238667488098,
      "num_tokens": 36354536.0,
      "step": 19500
    },
    {
      "entropy": 5.8919227600097654,
      "epoch": 1.722447898269163,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047048560749016167,
      "loss": 5.6867,
      "mean_token_accuracy": 0.17839696407318115,
      "num_tokens": 36363691.0,
      "step": 19505
    },
    {
      "entropy": 6.010924482345581,
      "epoch": 1.7228894383610032,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047047001351876403,
      "loss": 5.737,
      "mean_token_accuracy": 0.17599865943193435,
      "num_tokens": 36373898.0,
      "step": 19510
    },
    {
      "entropy": 5.900124025344849,
      "epoch": 1.7233309784528434,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004704544157182199,
      "loss": 5.7151,
      "mean_token_accuracy": 0.17715003043413163,
      "num_tokens": 36384345.0,
      "step": 19515
    },
    {
      "entropy": 5.858529186248779,
      "epoch": 1.723772518544684,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047043881408883497,
      "loss": 5.5917,
      "mean_token_accuracy": 0.18692393749952316,
      "num_tokens": 36393021.0,
      "step": 19520
    },
    {
      "entropy": 5.961220741271973,
      "epoch": 1.7242140586365242,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004704232086309148,
      "loss": 5.5932,
      "mean_token_accuracy": 0.17989325523376465,
      "num_tokens": 36402729.0,
      "step": 19525
    },
    {
      "entropy": 5.763304138183594,
      "epoch": 1.7246555987283645,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047040759934476517,
      "loss": 5.5118,
      "mean_token_accuracy": 0.18657368868589402,
      "num_tokens": 36412360.0,
      "step": 19530
    },
    {
      "entropy": 5.798563289642334,
      "epoch": 1.725097138820205,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004703919862306918,
      "loss": 5.5359,
      "mean_token_accuracy": 0.189731428027153,
      "num_tokens": 36420245.0,
      "step": 19535
    },
    {
      "entropy": 5.876319408416748,
      "epoch": 1.7255386789120453,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004703763692890006,
      "loss": 5.4979,
      "mean_token_accuracy": 0.1940944716334343,
      "num_tokens": 36428758.0,
      "step": 19540
    },
    {
      "entropy": 5.856221342086792,
      "epoch": 1.7259802190038855,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004703607485199973,
      "loss": 5.5838,
      "mean_token_accuracy": 0.18076451271772384,
      "num_tokens": 36437883.0,
      "step": 19545
    },
    {
      "entropy": 5.854744338989258,
      "epoch": 1.726421759095726,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047034512392398815,
      "loss": 5.6056,
      "mean_token_accuracy": 0.17785661220550536,
      "num_tokens": 36447950.0,
      "step": 19550
    },
    {
      "entropy": 5.925954055786133,
      "epoch": 1.726863299187566,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004703294955012791,
      "loss": 5.6332,
      "mean_token_accuracy": 0.17851487398147584,
      "num_tokens": 36456993.0,
      "step": 19555
    },
    {
      "entropy": 5.930733346939087,
      "epoch": 1.7273048392794066,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004703138632521762,
      "loss": 5.6789,
      "mean_token_accuracy": 0.1792685255408287,
      "num_tokens": 36465942.0,
      "step": 19560
    },
    {
      "entropy": 5.864876794815063,
      "epoch": 1.7277463793712469,
      "grad_norm": 1.4375,
      "learning_rate": 0.000470298227176986,
      "loss": 5.6602,
      "mean_token_accuracy": 0.18478365242481232,
      "num_tokens": 36474906.0,
      "step": 19565
    },
    {
      "entropy": 5.930994701385498,
      "epoch": 1.7281879194630871,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004702825872760145,
      "loss": 5.577,
      "mean_token_accuracy": 0.18360148817300798,
      "num_tokens": 36483579.0,
      "step": 19570
    },
    {
      "entropy": 5.878409337997437,
      "epoch": 1.7286294595549276,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004702669435495681,
      "loss": 5.5877,
      "mean_token_accuracy": 0.1865687683224678,
      "num_tokens": 36492541.0,
      "step": 19575
    },
    {
      "entropy": 5.93665132522583,
      "epoch": 1.729070999646768,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004702512959979535,
      "loss": 5.7271,
      "mean_token_accuracy": 0.17733775824308395,
      "num_tokens": 36502833.0,
      "step": 19580
    },
    {
      "entropy": 5.890898561477661,
      "epoch": 1.7295125397386082,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047023564462147697,
      "loss": 5.6063,
      "mean_token_accuracy": 0.18225155472755433,
      "num_tokens": 36513091.0,
      "step": 19585
    },
    {
      "entropy": 5.8899754047393795,
      "epoch": 1.7299540798304487,
      "grad_norm": 1.625,
      "learning_rate": 0.00047021998942044517,
      "loss": 5.6148,
      "mean_token_accuracy": 0.18287183344364166,
      "num_tokens": 36522911.0,
      "step": 19590
    },
    {
      "entropy": 5.9485105037689205,
      "epoch": 1.730395619922289,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047020433039516484,
      "loss": 5.6702,
      "mean_token_accuracy": 0.17794338017702102,
      "num_tokens": 36531934.0,
      "step": 19595
    },
    {
      "entropy": 5.914366579055786,
      "epoch": 1.7308371600141292,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004701886675459427,
      "loss": 5.6019,
      "mean_token_accuracy": 0.18016983121633529,
      "num_tokens": 36541823.0,
      "step": 19600
    },
    {
      "entropy": 5.948433208465576,
      "epoch": 1.7312787001059697,
      "grad_norm": 1.625,
      "learning_rate": 0.00047017300087308557,
      "loss": 5.6896,
      "mean_token_accuracy": 0.17846970409154891,
      "num_tokens": 36550954.0,
      "step": 19605
    },
    {
      "entropy": 5.854770660400391,
      "epoch": 1.73172024019781,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004701573303769004,
      "loss": 5.5943,
      "mean_token_accuracy": 0.18106776475906372,
      "num_tokens": 36560689.0,
      "step": 19610
    },
    {
      "entropy": 5.882445383071899,
      "epoch": 1.7321617802896503,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000470141656057694,
      "loss": 5.5902,
      "mean_token_accuracy": 0.18047057837247849,
      "num_tokens": 36569587.0,
      "step": 19615
    },
    {
      "entropy": 5.896106004714966,
      "epoch": 1.7326033203814908,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047012597791577364,
      "loss": 5.6211,
      "mean_token_accuracy": 0.1846356600522995,
      "num_tokens": 36579609.0,
      "step": 19620
    },
    {
      "entropy": 5.839815092086792,
      "epoch": 1.7330448604733308,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004701102959514463,
      "loss": 5.6117,
      "mean_token_accuracy": 0.1784946769475937,
      "num_tokens": 36589685.0,
      "step": 19625
    },
    {
      "entropy": 5.847561264038086,
      "epoch": 1.7334864005651713,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047009461016501926,
      "loss": 5.5818,
      "mean_token_accuracy": 0.18175967484712602,
      "num_tokens": 36598225.0,
      "step": 19630
    },
    {
      "entropy": 5.823268461227417,
      "epoch": 1.7339279406570116,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047007892055679976,
      "loss": 5.5094,
      "mean_token_accuracy": 0.18822446465492249,
      "num_tokens": 36607481.0,
      "step": 19635
    },
    {
      "entropy": 5.910299777984619,
      "epoch": 1.7343694807488519,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047006322712709513,
      "loss": 5.5959,
      "mean_token_accuracy": 0.1830284684896469,
      "num_tokens": 36616707.0,
      "step": 19640
    },
    {
      "entropy": 5.934399843215942,
      "epoch": 1.7348110208406924,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047004752987621286,
      "loss": 5.7341,
      "mean_token_accuracy": 0.17703829556703568,
      "num_tokens": 36626290.0,
      "step": 19645
    },
    {
      "entropy": 5.86400351524353,
      "epoch": 1.7352525609325327,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004700318288044603,
      "loss": 5.6195,
      "mean_token_accuracy": 0.1860916793346405,
      "num_tokens": 36635155.0,
      "step": 19650
    },
    {
      "entropy": 5.89954400062561,
      "epoch": 1.735694101024373,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047001612391214527,
      "loss": 5.5984,
      "mean_token_accuracy": 0.18723518997430802,
      "num_tokens": 36644632.0,
      "step": 19655
    },
    {
      "entropy": 5.9041290283203125,
      "epoch": 1.7361356411162134,
      "grad_norm": 1.375,
      "learning_rate": 0.0004700004151995753,
      "loss": 5.6489,
      "mean_token_accuracy": 0.17564476281404495,
      "num_tokens": 36654818.0,
      "step": 19660
    },
    {
      "entropy": 5.821690273284912,
      "epoch": 1.7365771812080537,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046998470266705795,
      "loss": 5.488,
      "mean_token_accuracy": 0.19151489436626434,
      "num_tokens": 36664149.0,
      "step": 19665
    },
    {
      "entropy": 5.841180038452149,
      "epoch": 1.737018721299894,
      "grad_norm": 1.6875,
      "learning_rate": 0.00046996898631490125,
      "loss": 5.5528,
      "mean_token_accuracy": 0.1915670946240425,
      "num_tokens": 36673447.0,
      "step": 19670
    },
    {
      "entropy": 5.98228645324707,
      "epoch": 1.7374602613917345,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000469953266143413,
      "loss": 5.7336,
      "mean_token_accuracy": 0.1730213060975075,
      "num_tokens": 36683426.0,
      "step": 19675
    },
    {
      "entropy": 5.952278995513916,
      "epoch": 1.7379018014835748,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046993754215290116,
      "loss": 5.6138,
      "mean_token_accuracy": 0.17700935751199723,
      "num_tokens": 36692708.0,
      "step": 19680
    },
    {
      "entropy": 5.7690471649169925,
      "epoch": 1.738343341575415,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004699218143436737,
      "loss": 5.53,
      "mean_token_accuracy": 0.18665760308504104,
      "num_tokens": 36701527.0,
      "step": 19685
    },
    {
      "entropy": 5.903882217407227,
      "epoch": 1.7387848816672555,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046990608271603874,
      "loss": 5.6876,
      "mean_token_accuracy": 0.17741188257932664,
      "num_tokens": 36711458.0,
      "step": 19690
    },
    {
      "entropy": 5.856635093688965,
      "epoch": 1.7392264217590956,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004698903472703045,
      "loss": 5.5958,
      "mean_token_accuracy": 0.1874726414680481,
      "num_tokens": 36721260.0,
      "step": 19695
    },
    {
      "entropy": 5.906638288497925,
      "epoch": 1.739667961850936,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046987460800677915,
      "loss": 5.6123,
      "mean_token_accuracy": 0.18867639750242232,
      "num_tokens": 36730496.0,
      "step": 19700
    },
    {
      "entropy": 5.879294776916504,
      "epoch": 1.7401095019427764,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046985886492577106,
      "loss": 5.6044,
      "mean_token_accuracy": 0.17800396233797072,
      "num_tokens": 36739016.0,
      "step": 19705
    },
    {
      "entropy": 5.955718231201172,
      "epoch": 1.7405510420346166,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004698431180275887,
      "loss": 5.6892,
      "mean_token_accuracy": 0.17189449220895767,
      "num_tokens": 36748259.0,
      "step": 19710
    },
    {
      "entropy": 6.001622104644776,
      "epoch": 1.7409925821264571,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046982736731254044,
      "loss": 5.7297,
      "mean_token_accuracy": 0.17238398492336274,
      "num_tokens": 36758174.0,
      "step": 19715
    },
    {
      "entropy": 5.934592533111572,
      "epoch": 1.7414341222182974,
      "grad_norm": 1.796875,
      "learning_rate": 0.00046981161278093487,
      "loss": 5.6834,
      "mean_token_accuracy": 0.1787327617406845,
      "num_tokens": 36767286.0,
      "step": 19720
    },
    {
      "entropy": 5.8986828327178955,
      "epoch": 1.7418756623101377,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046979585443308056,
      "loss": 5.5278,
      "mean_token_accuracy": 0.18513808995485306,
      "num_tokens": 36775818.0,
      "step": 19725
    },
    {
      "entropy": 5.859366130828858,
      "epoch": 1.7423172024019782,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00046978009226928633,
      "loss": 5.6094,
      "mean_token_accuracy": 0.18784332871437073,
      "num_tokens": 36785489.0,
      "step": 19730
    },
    {
      "entropy": 5.902548789978027,
      "epoch": 1.7427587424938185,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004697643262898608,
      "loss": 5.5859,
      "mean_token_accuracy": 0.18187982589006424,
      "num_tokens": 36794241.0,
      "step": 19735
    },
    {
      "entropy": 5.872925043106079,
      "epoch": 1.7432002825856587,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004697485564951128,
      "loss": 5.6106,
      "mean_token_accuracy": 0.17900541722774505,
      "num_tokens": 36803477.0,
      "step": 19740
    },
    {
      "entropy": 5.898757219314575,
      "epoch": 1.7436418226774992,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004697327828853515,
      "loss": 5.5982,
      "mean_token_accuracy": 0.17770358473062514,
      "num_tokens": 36811974.0,
      "step": 19745
    },
    {
      "entropy": 5.910835456848145,
      "epoch": 1.7440833627693395,
      "grad_norm": 9.0625,
      "learning_rate": 0.0004697170054608857,
      "loss": 5.6784,
      "mean_token_accuracy": 0.1786932334303856,
      "num_tokens": 36822049.0,
      "step": 19750
    },
    {
      "entropy": 5.907838916778564,
      "epoch": 1.7445249028611798,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004697012242220245,
      "loss": 5.5969,
      "mean_token_accuracy": 0.18512390106916427,
      "num_tokens": 36831989.0,
      "step": 19755
    },
    {
      "entropy": 5.8671647071838375,
      "epoch": 1.7449664429530203,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046968543916907704,
      "loss": 5.6412,
      "mean_token_accuracy": 0.17404000759124755,
      "num_tokens": 36841141.0,
      "step": 19760
    },
    {
      "entropy": 5.828925085067749,
      "epoch": 1.7454079830448603,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046966965030235253,
      "loss": 5.5032,
      "mean_token_accuracy": 0.19163623005151748,
      "num_tokens": 36850101.0,
      "step": 19765
    },
    {
      "entropy": 5.910364723205566,
      "epoch": 1.7458495231367008,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046965385762216024,
      "loss": 5.6252,
      "mean_token_accuracy": 0.18487497568130493,
      "num_tokens": 36859234.0,
      "step": 19770
    },
    {
      "entropy": 5.951223707199096,
      "epoch": 1.746291063228541,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004696380611288096,
      "loss": 5.6866,
      "mean_token_accuracy": 0.1793515980243683,
      "num_tokens": 36868937.0,
      "step": 19775
    },
    {
      "entropy": 5.8339017868042,
      "epoch": 1.7467326033203814,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046962226082261014,
      "loss": 5.5467,
      "mean_token_accuracy": 0.18861266672611238,
      "num_tokens": 36878380.0,
      "step": 19780
    },
    {
      "entropy": 5.84336485862732,
      "epoch": 1.7471741434122219,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004696064567038712,
      "loss": 5.6512,
      "mean_token_accuracy": 0.18254152536392212,
      "num_tokens": 36887702.0,
      "step": 19785
    },
    {
      "entropy": 5.869961786270141,
      "epoch": 1.7476156835040622,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046959064877290246,
      "loss": 5.5209,
      "mean_token_accuracy": 0.19536864310503005,
      "num_tokens": 36896777.0,
      "step": 19790
    },
    {
      "entropy": 5.910241317749024,
      "epoch": 1.7480572235959024,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046957483703001366,
      "loss": 5.548,
      "mean_token_accuracy": 0.18495236188173295,
      "num_tokens": 36905256.0,
      "step": 19795
    },
    {
      "entropy": 5.822274112701416,
      "epoch": 1.748498763687743,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004695590214755143,
      "loss": 5.579,
      "mean_token_accuracy": 0.18168120682239533,
      "num_tokens": 36913767.0,
      "step": 19800
    },
    {
      "entropy": 5.80119776725769,
      "epoch": 1.7489403037795832,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046954320210971444,
      "loss": 5.5801,
      "mean_token_accuracy": 0.18870102167129515,
      "num_tokens": 36922838.0,
      "step": 19805
    },
    {
      "entropy": 5.843250131607055,
      "epoch": 1.7493818438714235,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046952737893292396,
      "loss": 5.4664,
      "mean_token_accuracy": 0.1895427495241165,
      "num_tokens": 36931955.0,
      "step": 19810
    },
    {
      "entropy": 5.825087833404541,
      "epoch": 1.749823383963264,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046951155194545265,
      "loss": 5.5714,
      "mean_token_accuracy": 0.18421393632888794,
      "num_tokens": 36941505.0,
      "step": 19815
    },
    {
      "entropy": 5.858208131790161,
      "epoch": 1.7502649240551043,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00046949572114761076,
      "loss": 5.6591,
      "mean_token_accuracy": 0.17753167897462846,
      "num_tokens": 36951007.0,
      "step": 19820
    },
    {
      "entropy": 5.935961389541626,
      "epoch": 1.7507064641469445,
      "grad_norm": 1.5,
      "learning_rate": 0.00046947988653970826,
      "loss": 5.6162,
      "mean_token_accuracy": 0.1885736808180809,
      "num_tokens": 36961111.0,
      "step": 19825
    },
    {
      "entropy": 5.916716909408569,
      "epoch": 1.751148004238785,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004694640481220554,
      "loss": 5.6347,
      "mean_token_accuracy": 0.18360367864370347,
      "num_tokens": 36969605.0,
      "step": 19830
    },
    {
      "entropy": 5.943793487548828,
      "epoch": 1.751589544330625,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004694482058949624,
      "loss": 5.6206,
      "mean_token_accuracy": 0.17855388671159744,
      "num_tokens": 36979051.0,
      "step": 19835
    },
    {
      "entropy": 5.918580627441406,
      "epoch": 1.7520310844224656,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046943235985873975,
      "loss": 5.572,
      "mean_token_accuracy": 0.18514021188020707,
      "num_tokens": 36988807.0,
      "step": 19840
    },
    {
      "entropy": 5.83690299987793,
      "epoch": 1.7524726245143059,
      "grad_norm": 1.625,
      "learning_rate": 0.0004694165100136976,
      "loss": 5.502,
      "mean_token_accuracy": 0.19242252558469772,
      "num_tokens": 36997813.0,
      "step": 19845
    },
    {
      "entropy": 5.844189167022705,
      "epoch": 1.7529141646061461,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046940065636014665,
      "loss": 5.5764,
      "mean_token_accuracy": 0.1848168522119522,
      "num_tokens": 37007849.0,
      "step": 19850
    },
    {
      "entropy": 5.849606847763061,
      "epoch": 1.7533557046979866,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004693847988983974,
      "loss": 5.4834,
      "mean_token_accuracy": 0.1961677700281143,
      "num_tokens": 37016109.0,
      "step": 19855
    },
    {
      "entropy": 5.887281179428101,
      "epoch": 1.753797244789827,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004693689376287605,
      "loss": 5.5824,
      "mean_token_accuracy": 0.17898539900779725,
      "num_tokens": 37025156.0,
      "step": 19860
    },
    {
      "entropy": 5.939543104171753,
      "epoch": 1.7542387848816672,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046935307255154663,
      "loss": 5.718,
      "mean_token_accuracy": 0.17715738713741302,
      "num_tokens": 37035361.0,
      "step": 19865
    },
    {
      "entropy": 6.000812768936157,
      "epoch": 1.7546803249735077,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004693372036670667,
      "loss": 5.6331,
      "mean_token_accuracy": 0.17793141752481462,
      "num_tokens": 37045249.0,
      "step": 19870
    },
    {
      "entropy": 5.904667234420776,
      "epoch": 1.755121865065348,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046932133097563137,
      "loss": 5.6339,
      "mean_token_accuracy": 0.18670137971639633,
      "num_tokens": 37054637.0,
      "step": 19875
    },
    {
      "entropy": 5.858313274383545,
      "epoch": 1.7555634051571882,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046930545447755167,
      "loss": 5.6308,
      "mean_token_accuracy": 0.1864708587527275,
      "num_tokens": 37064557.0,
      "step": 19880
    },
    {
      "entropy": 5.87725338935852,
      "epoch": 1.7560049452490287,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004692895741731387,
      "loss": 5.6596,
      "mean_token_accuracy": 0.1730324372649193,
      "num_tokens": 37073915.0,
      "step": 19885
    },
    {
      "entropy": 5.971892499923706,
      "epoch": 1.756446485340869,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004692736900627034,
      "loss": 5.6216,
      "mean_token_accuracy": 0.17948929518461226,
      "num_tokens": 37083008.0,
      "step": 19890
    },
    {
      "entropy": 5.930801057815552,
      "epoch": 1.7568880254327093,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046925780214655713,
      "loss": 5.6387,
      "mean_token_accuracy": 0.17929941564798355,
      "num_tokens": 37092742.0,
      "step": 19895
    },
    {
      "entropy": 5.952897548675537,
      "epoch": 1.7573295655245498,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004692419104250108,
      "loss": 5.6782,
      "mean_token_accuracy": 0.17525677978992463,
      "num_tokens": 37102242.0,
      "step": 19900
    },
    {
      "entropy": 5.911886072158813,
      "epoch": 1.7577711056163898,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004692260148983761,
      "loss": 5.5837,
      "mean_token_accuracy": 0.19132448583841324,
      "num_tokens": 37111304.0,
      "step": 19905
    },
    {
      "entropy": 5.896725511550903,
      "epoch": 1.7582126457082303,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004692101155669641,
      "loss": 5.5927,
      "mean_token_accuracy": 0.1842863842844963,
      "num_tokens": 37121420.0,
      "step": 19910
    },
    {
      "entropy": 5.995995092391968,
      "epoch": 1.7586541858000706,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004691942124310865,
      "loss": 5.6445,
      "mean_token_accuracy": 0.1816993847489357,
      "num_tokens": 37131265.0,
      "step": 19915
    },
    {
      "entropy": 5.840927934646606,
      "epoch": 1.7590957258919109,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00046917830549105465,
      "loss": 5.5551,
      "mean_token_accuracy": 0.18360982537269593,
      "num_tokens": 37141097.0,
      "step": 19920
    },
    {
      "entropy": 5.758389139175415,
      "epoch": 1.7595372659837514,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004691623947471803,
      "loss": 5.4592,
      "mean_token_accuracy": 0.19595457166433333,
      "num_tokens": 37149687.0,
      "step": 19925
    },
    {
      "entropy": 5.884005975723267,
      "epoch": 1.7599788060755917,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046914648019977495,
      "loss": 5.5787,
      "mean_token_accuracy": 0.18759655505418776,
      "num_tokens": 37159219.0,
      "step": 19930
    },
    {
      "entropy": 5.896233797073364,
      "epoch": 1.760420346167432,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046913056184915054,
      "loss": 5.6109,
      "mean_token_accuracy": 0.18528549373149872,
      "num_tokens": 37168681.0,
      "step": 19935
    },
    {
      "entropy": 5.766439533233642,
      "epoch": 1.7608618862592724,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004691146396956188,
      "loss": 5.4928,
      "mean_token_accuracy": 0.19203969985246658,
      "num_tokens": 37177932.0,
      "step": 19940
    },
    {
      "entropy": 5.831232309341431,
      "epoch": 1.7613034263511127,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004690987137394917,
      "loss": 5.5005,
      "mean_token_accuracy": 0.18773321211338043,
      "num_tokens": 37186783.0,
      "step": 19945
    },
    {
      "entropy": 5.909509086608887,
      "epoch": 1.761744966442953,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004690827839810812,
      "loss": 5.6889,
      "mean_token_accuracy": 0.17385972589254378,
      "num_tokens": 37196290.0,
      "step": 19950
    },
    {
      "entropy": 5.898013734817505,
      "epoch": 1.7621865065347935,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046906685042069927,
      "loss": 5.5419,
      "mean_token_accuracy": 0.18542669117450714,
      "num_tokens": 37205057.0,
      "step": 19955
    },
    {
      "entropy": 5.919957256317138,
      "epoch": 1.7626280466266337,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004690509130586582,
      "loss": 5.6151,
      "mean_token_accuracy": 0.1796432241797447,
      "num_tokens": 37214078.0,
      "step": 19960
    },
    {
      "entropy": 5.848082733154297,
      "epoch": 1.763069586718474,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00046903497189527006,
      "loss": 5.5492,
      "mean_token_accuracy": 0.18913544118404388,
      "num_tokens": 37223232.0,
      "step": 19965
    },
    {
      "entropy": 5.840355968475341,
      "epoch": 1.7635111268103145,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004690190269308472,
      "loss": 5.572,
      "mean_token_accuracy": 0.19180630445480346,
      "num_tokens": 37232054.0,
      "step": 19970
    },
    {
      "entropy": 5.837779521942139,
      "epoch": 1.7639526669021546,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004690030781657019,
      "loss": 5.5244,
      "mean_token_accuracy": 0.18460294604301453,
      "num_tokens": 37241451.0,
      "step": 19975
    },
    {
      "entropy": 5.909575748443603,
      "epoch": 1.764394206993995,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004689871256001467,
      "loss": 5.5719,
      "mean_token_accuracy": 0.1807768389582634,
      "num_tokens": 37251029.0,
      "step": 19980
    },
    {
      "entropy": 5.839111948013306,
      "epoch": 1.7648357470858353,
      "grad_norm": 1.859375,
      "learning_rate": 0.00046897116923449403,
      "loss": 5.5891,
      "mean_token_accuracy": 0.1812613844871521,
      "num_tokens": 37259594.0,
      "step": 19985
    },
    {
      "entropy": 5.822681713104248,
      "epoch": 1.7652772871776756,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046895520906905635,
      "loss": 5.5322,
      "mean_token_accuracy": 0.18490878194570542,
      "num_tokens": 37268824.0,
      "step": 19990
    },
    {
      "entropy": 5.909144496917724,
      "epoch": 1.7657188272695161,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00046893924510414654,
      "loss": 5.5556,
      "mean_token_accuracy": 0.18339895457029343,
      "num_tokens": 37278494.0,
      "step": 19995
    },
    {
      "entropy": 5.698139190673828,
      "epoch": 1.7661603673613564,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004689232773400772,
      "loss": 5.3785,
      "mean_token_accuracy": 0.2125357061624527,
      "num_tokens": 37287648.0,
      "step": 20000
    },
    {
      "entropy": 5.8666986465454105,
      "epoch": 1.7666019074531967,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004689073057771611,
      "loss": 5.566,
      "mean_token_accuracy": 0.18253726959228517,
      "num_tokens": 37295938.0,
      "step": 20005
    },
    {
      "entropy": 5.849815750122071,
      "epoch": 1.7670434475450372,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00046889133041571126,
      "loss": 5.5623,
      "mean_token_accuracy": 0.18827050030231476,
      "num_tokens": 37305097.0,
      "step": 20010
    },
    {
      "entropy": 5.832500028610229,
      "epoch": 1.7674849876368774,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004688753512560405,
      "loss": 5.5862,
      "mean_token_accuracy": 0.18766206800937651,
      "num_tokens": 37313672.0,
      "step": 20015
    },
    {
      "entropy": 5.875202608108521,
      "epoch": 1.7679265277287177,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004688593682984619,
      "loss": 5.6488,
      "mean_token_accuracy": 0.18124972730875016,
      "num_tokens": 37323342.0,
      "step": 20020
    },
    {
      "entropy": 5.8576983451843265,
      "epoch": 1.7683680678205582,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004688433815432884,
      "loss": 5.6157,
      "mean_token_accuracy": 0.18202545642852783,
      "num_tokens": 37332578.0,
      "step": 20025
    },
    {
      "entropy": 5.89477276802063,
      "epoch": 1.7688096079123985,
      "grad_norm": 1.546875,
      "learning_rate": 0.00046882739099083335,
      "loss": 5.6186,
      "mean_token_accuracy": 0.17854406386613847,
      "num_tokens": 37341769.0,
      "step": 20030
    },
    {
      "entropy": 5.962050724029541,
      "epoch": 1.7692511480042388,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046881139664141006,
      "loss": 5.6643,
      "mean_token_accuracy": 0.1765436053276062,
      "num_tokens": 37351158.0,
      "step": 20035
    },
    {
      "entropy": 5.931488800048828,
      "epoch": 1.7696926880960793,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004687953984953316,
      "loss": 5.6081,
      "mean_token_accuracy": 0.18476668000221252,
      "num_tokens": 37360924.0,
      "step": 20040
    },
    {
      "entropy": 5.769788217544556,
      "epoch": 1.7701342281879193,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004687793965529115,
      "loss": 5.5443,
      "mean_token_accuracy": 0.18938349485397338,
      "num_tokens": 37369295.0,
      "step": 20045
    },
    {
      "entropy": 5.807572555541992,
      "epoch": 1.7705757682797598,
      "grad_norm": 1.5,
      "learning_rate": 0.0004687633908144633,
      "loss": 5.5824,
      "mean_token_accuracy": 0.18433805853128432,
      "num_tokens": 37379135.0,
      "step": 20050
    },
    {
      "entropy": 5.8438560485839846,
      "epoch": 1.7710173083716,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046874738128030045,
      "loss": 5.4798,
      "mean_token_accuracy": 0.19075767695903778,
      "num_tokens": 37387927.0,
      "step": 20055
    },
    {
      "entropy": 5.7925232410430905,
      "epoch": 1.7714588484634404,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004687313679507366,
      "loss": 5.5185,
      "mean_token_accuracy": 0.19077939093112944,
      "num_tokens": 37397629.0,
      "step": 20060
    },
    {
      "entropy": 5.80764422416687,
      "epoch": 1.7719003885552809,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046871535082608544,
      "loss": 5.5394,
      "mean_token_accuracy": 0.18452271223068237,
      "num_tokens": 37407688.0,
      "step": 20065
    },
    {
      "entropy": 5.850022029876709,
      "epoch": 1.7723419286471211,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004686993299066607,
      "loss": 5.5341,
      "mean_token_accuracy": 0.18827124834060668,
      "num_tokens": 37417019.0,
      "step": 20070
    },
    {
      "entropy": 5.883563041687012,
      "epoch": 1.7727834687389614,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004686833051927763,
      "loss": 5.5956,
      "mean_token_accuracy": 0.1811591476202011,
      "num_tokens": 37427685.0,
      "step": 20075
    },
    {
      "entropy": 5.796727418899536,
      "epoch": 1.773225008830802,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046866727668474605,
      "loss": 5.5068,
      "mean_token_accuracy": 0.193043515086174,
      "num_tokens": 37436844.0,
      "step": 20080
    },
    {
      "entropy": 5.763185358047485,
      "epoch": 1.7736665489226422,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000468651244382884,
      "loss": 5.4837,
      "mean_token_accuracy": 0.1927231326699257,
      "num_tokens": 37445755.0,
      "step": 20085
    },
    {
      "entropy": 5.827434444427491,
      "epoch": 1.7741080890144825,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004686352082875043,
      "loss": 5.6654,
      "mean_token_accuracy": 0.1747334510087967,
      "num_tokens": 37454176.0,
      "step": 20090
    },
    {
      "entropy": 5.900244092941284,
      "epoch": 1.774549629106323,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004686191683989209,
      "loss": 5.5561,
      "mean_token_accuracy": 0.19093385338783264,
      "num_tokens": 37463696.0,
      "step": 20095
    },
    {
      "entropy": 5.922184753417969,
      "epoch": 1.7749911691981632,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004686031247174482,
      "loss": 5.5885,
      "mean_token_accuracy": 0.18240967392921448,
      "num_tokens": 37473651.0,
      "step": 20100
    },
    {
      "entropy": 5.875768518447876,
      "epoch": 1.7754327092900035,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004685870772434002,
      "loss": 5.6663,
      "mean_token_accuracy": 0.17812173217535018,
      "num_tokens": 37483137.0,
      "step": 20105
    },
    {
      "entropy": 5.930988121032715,
      "epoch": 1.775874249381844,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004685710259770916,
      "loss": 5.6432,
      "mean_token_accuracy": 0.17713044732809066,
      "num_tokens": 37492029.0,
      "step": 20110
    },
    {
      "entropy": 5.874558067321777,
      "epoch": 1.776315789473684,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046855497091883664,
      "loss": 5.5598,
      "mean_token_accuracy": 0.18638107627630235,
      "num_tokens": 37500421.0,
      "step": 20115
    },
    {
      "entropy": 5.831426382064819,
      "epoch": 1.7767573295655246,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004685389120689499,
      "loss": 5.5549,
      "mean_token_accuracy": 0.18698069602251052,
      "num_tokens": 37510332.0,
      "step": 20120
    },
    {
      "entropy": 5.836037635803223,
      "epoch": 1.7771988696573648,
      "grad_norm": 2.25,
      "learning_rate": 0.000468522849427746,
      "loss": 5.5694,
      "mean_token_accuracy": 0.19214224219322204,
      "num_tokens": 37519802.0,
      "step": 20125
    },
    {
      "entropy": 6.006646919250488,
      "epoch": 1.7776404097492051,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046850678299553943,
      "loss": 5.7,
      "mean_token_accuracy": 0.1736489489674568,
      "num_tokens": 37529622.0,
      "step": 20130
    },
    {
      "entropy": 5.768665838241577,
      "epoch": 1.7780819498410456,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004684907127726451,
      "loss": 5.5316,
      "mean_token_accuracy": 0.19555853754282,
      "num_tokens": 37539446.0,
      "step": 20135
    },
    {
      "entropy": 5.784971714019775,
      "epoch": 1.778523489932886,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004684746387593778,
      "loss": 5.5041,
      "mean_token_accuracy": 0.1906916081905365,
      "num_tokens": 37547877.0,
      "step": 20140
    },
    {
      "entropy": 5.880180358886719,
      "epoch": 1.7789650300247262,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004684585609560523,
      "loss": 5.7293,
      "mean_token_accuracy": 0.1773300901055336,
      "num_tokens": 37557841.0,
      "step": 20145
    },
    {
      "entropy": 5.898041439056397,
      "epoch": 1.7794065701165667,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004684424793629836,
      "loss": 5.6423,
      "mean_token_accuracy": 0.17455686032772064,
      "num_tokens": 37566876.0,
      "step": 20150
    },
    {
      "entropy": 5.918929386138916,
      "epoch": 1.779848110208407,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046842639398048683,
      "loss": 5.6339,
      "mean_token_accuracy": 0.18124764859676362,
      "num_tokens": 37575571.0,
      "step": 20155
    },
    {
      "entropy": 5.98367714881897,
      "epoch": 1.7802896503002472,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004684103048088769,
      "loss": 5.7249,
      "mean_token_accuracy": 0.17889208048582078,
      "num_tokens": 37585972.0,
      "step": 20160
    },
    {
      "entropy": 5.9129632949829105,
      "epoch": 1.7807311903920877,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004683942118484692,
      "loss": 5.6383,
      "mean_token_accuracy": 0.17823220789432526,
      "num_tokens": 37594849.0,
      "step": 20165
    },
    {
      "entropy": 5.908299446105957,
      "epoch": 1.781172730483928,
      "grad_norm": 1.53125,
      "learning_rate": 0.00046837811509957883,
      "loss": 5.5606,
      "mean_token_accuracy": 0.1882735088467598,
      "num_tokens": 37603634.0,
      "step": 20170
    },
    {
      "entropy": 5.8329143047332765,
      "epoch": 1.7816142705757683,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046836201456252116,
      "loss": 5.5362,
      "mean_token_accuracy": 0.19093223065137863,
      "num_tokens": 37612518.0,
      "step": 20175
    },
    {
      "entropy": 5.823099088668823,
      "epoch": 1.7820558106676088,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004683459102376116,
      "loss": 5.5423,
      "mean_token_accuracy": 0.19289669245481492,
      "num_tokens": 37621553.0,
      "step": 20180
    },
    {
      "entropy": 5.8552470207214355,
      "epoch": 1.7824973507594488,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046832980212516557,
      "loss": 5.6343,
      "mean_token_accuracy": 0.1788346067070961,
      "num_tokens": 37631263.0,
      "step": 20185
    },
    {
      "entropy": 5.868450880050659,
      "epoch": 1.7829388908512893,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046831369022549876,
      "loss": 5.5212,
      "mean_token_accuracy": 0.19267574548721314,
      "num_tokens": 37640810.0,
      "step": 20190
    },
    {
      "entropy": 5.8666924953460695,
      "epoch": 1.7833804309431296,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004682975745389267,
      "loss": 5.6277,
      "mean_token_accuracy": 0.1824028417468071,
      "num_tokens": 37650848.0,
      "step": 20195
    },
    {
      "entropy": 5.905120515823365,
      "epoch": 1.7838219710349699,
      "grad_norm": 1.609375,
      "learning_rate": 0.00046828145506576504,
      "loss": 5.5853,
      "mean_token_accuracy": 0.18069953471422195,
      "num_tokens": 37660370.0,
      "step": 20200
    },
    {
      "entropy": 5.847794914245606,
      "epoch": 1.7842635111268104,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004682653318063296,
      "loss": 5.5427,
      "mean_token_accuracy": 0.18696775138378144,
      "num_tokens": 37669570.0,
      "step": 20205
    },
    {
      "entropy": 5.835547399520874,
      "epoch": 1.7847050512186506,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046824920476093624,
      "loss": 5.614,
      "mean_token_accuracy": 0.178546442091465,
      "num_tokens": 37679585.0,
      "step": 20210
    },
    {
      "entropy": 5.8897606372833256,
      "epoch": 1.785146591310491,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046823307392990086,
      "loss": 5.6315,
      "mean_token_accuracy": 0.18677870631217958,
      "num_tokens": 37688736.0,
      "step": 20215
    },
    {
      "entropy": 5.965898084640503,
      "epoch": 1.7855881314023314,
      "grad_norm": 1.671875,
      "learning_rate": 0.00046821693931353947,
      "loss": 5.6664,
      "mean_token_accuracy": 0.17419648468494414,
      "num_tokens": 37699104.0,
      "step": 20220
    },
    {
      "entropy": 5.963194561004639,
      "epoch": 1.7860296714941717,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004682008009121681,
      "loss": 5.6638,
      "mean_token_accuracy": 0.18209634572267533,
      "num_tokens": 37708260.0,
      "step": 20225
    },
    {
      "entropy": 5.832933950424194,
      "epoch": 1.786471211586012,
      "grad_norm": 1.359375,
      "learning_rate": 0.000468184658726103,
      "loss": 5.6004,
      "mean_token_accuracy": 0.18332876712083818,
      "num_tokens": 37717442.0,
      "step": 20230
    },
    {
      "entropy": 5.8584325313568115,
      "epoch": 1.7869127516778525,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004681685127556602,
      "loss": 5.6378,
      "mean_token_accuracy": 0.17870260626077653,
      "num_tokens": 37727063.0,
      "step": 20235
    },
    {
      "entropy": 5.950245475769043,
      "epoch": 1.7873542917696927,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046815236300115615,
      "loss": 5.5495,
      "mean_token_accuracy": 0.18133650571107865,
      "num_tokens": 37736441.0,
      "step": 20240
    },
    {
      "entropy": 5.749633026123047,
      "epoch": 1.787795831861533,
      "grad_norm": 1.59375,
      "learning_rate": 0.00046813620946290716,
      "loss": 5.499,
      "mean_token_accuracy": 0.18869638442993164,
      "num_tokens": 37745290.0,
      "step": 20245
    },
    {
      "entropy": 5.846426773071289,
      "epoch": 1.7882373719533735,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004681200521412296,
      "loss": 5.6006,
      "mean_token_accuracy": 0.18538093119859694,
      "num_tokens": 37754786.0,
      "step": 20250
    },
    {
      "entropy": 5.9335661888122555,
      "epoch": 1.7886789120452136,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046810389103644015,
      "loss": 5.6215,
      "mean_token_accuracy": 0.18044727444648742,
      "num_tokens": 37763513.0,
      "step": 20255
    },
    {
      "entropy": 5.9207408905029295,
      "epoch": 1.789120452137054,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004680877261488552,
      "loss": 5.595,
      "mean_token_accuracy": 0.18141841143369675,
      "num_tokens": 37771752.0,
      "step": 20260
    },
    {
      "entropy": 5.875630950927734,
      "epoch": 1.7895619922288943,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004680715574787916,
      "loss": 5.6155,
      "mean_token_accuracy": 0.17405181378126144,
      "num_tokens": 37781531.0,
      "step": 20265
    },
    {
      "entropy": 5.825566625595092,
      "epoch": 1.7900035323207346,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046805538502656596,
      "loss": 5.5437,
      "mean_token_accuracy": 0.18476777225732804,
      "num_tokens": 37791469.0,
      "step": 20270
    },
    {
      "entropy": 5.913211679458618,
      "epoch": 1.7904450724125751,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004680392087924952,
      "loss": 5.6144,
      "mean_token_accuracy": 0.18043319284915924,
      "num_tokens": 37800301.0,
      "step": 20275
    },
    {
      "entropy": 5.794536495208741,
      "epoch": 1.7908866125044154,
      "grad_norm": 1.578125,
      "learning_rate": 0.000468023028776896,
      "loss": 5.4553,
      "mean_token_accuracy": 0.1987692505121231,
      "num_tokens": 37808849.0,
      "step": 20280
    },
    {
      "entropy": 5.8400391101837155,
      "epoch": 1.7913281525962557,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004680068449800855,
      "loss": 5.6319,
      "mean_token_accuracy": 0.18375971615314485,
      "num_tokens": 37818632.0,
      "step": 20285
    },
    {
      "entropy": 5.9182170867919925,
      "epoch": 1.7917696926880962,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004679906574023807,
      "loss": 5.6126,
      "mean_token_accuracy": 0.18024830222129823,
      "num_tokens": 37828158.0,
      "step": 20290
    },
    {
      "entropy": 5.878743743896484,
      "epoch": 1.7922112327799364,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004679744660440987,
      "loss": 5.5659,
      "mean_token_accuracy": 0.18051768988370895,
      "num_tokens": 37836242.0,
      "step": 20295
    },
    {
      "entropy": 5.866732215881347,
      "epoch": 1.7926527728717767,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046795827090555654,
      "loss": 5.6008,
      "mean_token_accuracy": 0.18568928986787797,
      "num_tokens": 37844286.0,
      "step": 20300
    },
    {
      "entropy": 5.866362810134888,
      "epoch": 1.7930943129636172,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004679420719870717,
      "loss": 5.532,
      "mean_token_accuracy": 0.1852009892463684,
      "num_tokens": 37853440.0,
      "step": 20305
    },
    {
      "entropy": 5.860985040664673,
      "epoch": 1.7935358530554575,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004679258692889613,
      "loss": 5.4697,
      "mean_token_accuracy": 0.1885829508304596,
      "num_tokens": 37862511.0,
      "step": 20310
    },
    {
      "entropy": 5.907676267623901,
      "epoch": 1.7939773931472978,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000467909662811543,
      "loss": 5.674,
      "mean_token_accuracy": 0.174008047580719,
      "num_tokens": 37872797.0,
      "step": 20315
    },
    {
      "entropy": 5.835335826873779,
      "epoch": 1.7944189332391383,
      "grad_norm": 1.921875,
      "learning_rate": 0.000467893452555134,
      "loss": 5.5332,
      "mean_token_accuracy": 0.1956690415740013,
      "num_tokens": 37881859.0,
      "step": 20320
    },
    {
      "entropy": 5.847554445266724,
      "epoch": 1.7948604733309783,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046787723852005197,
      "loss": 5.5801,
      "mean_token_accuracy": 0.19073069840669632,
      "num_tokens": 37891381.0,
      "step": 20325
    },
    {
      "entropy": 5.757302474975586,
      "epoch": 1.7953020134228188,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004678610207066146,
      "loss": 5.4604,
      "mean_token_accuracy": 0.1941983327269554,
      "num_tokens": 37900554.0,
      "step": 20330
    },
    {
      "entropy": 5.892540645599365,
      "epoch": 1.795743553514659,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004678447991151395,
      "loss": 5.532,
      "mean_token_accuracy": 0.18936284333467485,
      "num_tokens": 37908965.0,
      "step": 20335
    },
    {
      "entropy": 5.786262941360474,
      "epoch": 1.7961850936064994,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046782857374594447,
      "loss": 5.5115,
      "mean_token_accuracy": 0.1876990169286728,
      "num_tokens": 37917555.0,
      "step": 20340
    },
    {
      "entropy": 5.8469147205352785,
      "epoch": 1.7966266336983399,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004678123445993473,
      "loss": 5.5866,
      "mean_token_accuracy": 0.1905738592147827,
      "num_tokens": 37926334.0,
      "step": 20345
    },
    {
      "entropy": 5.920217275619507,
      "epoch": 1.7970681737901801,
      "grad_norm": 1.28125,
      "learning_rate": 0.000467796111675666,
      "loss": 5.536,
      "mean_token_accuracy": 0.18720044642686845,
      "num_tokens": 37935912.0,
      "step": 20350
    },
    {
      "entropy": 5.816601800918579,
      "epoch": 1.7975097138820204,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046777987497521845,
      "loss": 5.5543,
      "mean_token_accuracy": 0.19533709287643433,
      "num_tokens": 37945238.0,
      "step": 20355
    },
    {
      "entropy": 5.812395524978638,
      "epoch": 1.797951253973861,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046776363449832286,
      "loss": 5.5536,
      "mean_token_accuracy": 0.18888749480247496,
      "num_tokens": 37954945.0,
      "step": 20360
    },
    {
      "entropy": 5.8124762058258055,
      "epoch": 1.7983927940657012,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046774739024529736,
      "loss": 5.4761,
      "mean_token_accuracy": 0.19225014746189117,
      "num_tokens": 37964114.0,
      "step": 20365
    },
    {
      "entropy": 5.835586929321289,
      "epoch": 1.7988343341575415,
      "grad_norm": 1.5,
      "learning_rate": 0.00046773114221645995,
      "loss": 5.6168,
      "mean_token_accuracy": 0.1724876269698143,
      "num_tokens": 37973774.0,
      "step": 20370
    },
    {
      "entropy": 5.90380597114563,
      "epoch": 1.799275874249382,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046771489041212916,
      "loss": 5.5969,
      "mean_token_accuracy": 0.18404897004365922,
      "num_tokens": 37982763.0,
      "step": 20375
    },
    {
      "entropy": 5.951628160476685,
      "epoch": 1.7997174143412222,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004676986348326233,
      "loss": 5.6805,
      "mean_token_accuracy": 0.17605172246694564,
      "num_tokens": 37992302.0,
      "step": 20380
    },
    {
      "entropy": 5.953120279312134,
      "epoch": 1.8001589544330625,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004676823754782607,
      "loss": 5.7217,
      "mean_token_accuracy": 0.17069627940654755,
      "num_tokens": 38001785.0,
      "step": 20385
    },
    {
      "entropy": 5.826594877243042,
      "epoch": 1.800600494524903,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046766611234936,
      "loss": 5.5073,
      "mean_token_accuracy": 0.18584761768579483,
      "num_tokens": 38010823.0,
      "step": 20390
    },
    {
      "entropy": 5.828123378753662,
      "epoch": 1.801042034616743,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004676498454462398,
      "loss": 5.5353,
      "mean_token_accuracy": 0.1918479546904564,
      "num_tokens": 38019718.0,
      "step": 20395
    },
    {
      "entropy": 5.887084150314331,
      "epoch": 1.8014835747085836,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046763357476921856,
      "loss": 5.5472,
      "mean_token_accuracy": 0.18508918732404708,
      "num_tokens": 38028962.0,
      "step": 20400
    },
    {
      "entropy": 5.82620530128479,
      "epoch": 1.8019251148004238,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004676173003186152,
      "loss": 5.5126,
      "mean_token_accuracy": 0.19645865261554718,
      "num_tokens": 38037824.0,
      "step": 20405
    },
    {
      "entropy": 5.8717855453491214,
      "epoch": 1.802366654892264,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046760102209474855,
      "loss": 5.6535,
      "mean_token_accuracy": 0.17923854887485505,
      "num_tokens": 38046764.0,
      "step": 20410
    },
    {
      "entropy": 5.919915580749512,
      "epoch": 1.8028081949841046,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004675847400979373,
      "loss": 5.5618,
      "mean_token_accuracy": 0.1878815233707428,
      "num_tokens": 38055363.0,
      "step": 20415
    },
    {
      "entropy": 5.899560070037841,
      "epoch": 1.8032497350759449,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046756845432850057,
      "loss": 5.6325,
      "mean_token_accuracy": 0.183535598218441,
      "num_tokens": 38065471.0,
      "step": 20420
    },
    {
      "entropy": 5.959843349456787,
      "epoch": 1.8036912751677852,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004675521647867573,
      "loss": 5.7146,
      "mean_token_accuracy": 0.1738400787115097,
      "num_tokens": 38075448.0,
      "step": 20425
    },
    {
      "entropy": 5.828608655929566,
      "epoch": 1.8041328152596257,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046753587147302665,
      "loss": 5.6428,
      "mean_token_accuracy": 0.19044111371040345,
      "num_tokens": 38085657.0,
      "step": 20430
    },
    {
      "entropy": 5.943186283111572,
      "epoch": 1.804574355351466,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004675195743876278,
      "loss": 5.5553,
      "mean_token_accuracy": 0.18860284686088563,
      "num_tokens": 38095580.0,
      "step": 20435
    },
    {
      "entropy": 5.963461637496948,
      "epoch": 1.8050158954433062,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046750327353088,
      "loss": 5.6879,
      "mean_token_accuracy": 0.17211780101060867,
      "num_tokens": 38104969.0,
      "step": 20440
    },
    {
      "entropy": 5.872720575332641,
      "epoch": 1.8054574355351467,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004674869689031026,
      "loss": 5.6403,
      "mean_token_accuracy": 0.1795252174139023,
      "num_tokens": 38114536.0,
      "step": 20445
    },
    {
      "entropy": 5.870252180099487,
      "epoch": 1.805898975626987,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00046747066050461486,
      "loss": 5.5448,
      "mean_token_accuracy": 0.18529100269079207,
      "num_tokens": 38122969.0,
      "step": 20450
    },
    {
      "entropy": 5.910367488861084,
      "epoch": 1.8063405157188273,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046745434833573635,
      "loss": 5.606,
      "mean_token_accuracy": 0.17730246037244796,
      "num_tokens": 38132892.0,
      "step": 20455
    },
    {
      "entropy": 5.761419200897217,
      "epoch": 1.8067820558106678,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004674380323967867,
      "loss": 5.5495,
      "mean_token_accuracy": 0.18705078214406967,
      "num_tokens": 38141950.0,
      "step": 20460
    },
    {
      "entropy": 5.905006647109985,
      "epoch": 1.8072235959025078,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004674217126880853,
      "loss": 5.6726,
      "mean_token_accuracy": 0.17529373615980148,
      "num_tokens": 38151314.0,
      "step": 20465
    },
    {
      "entropy": 6.055087947845459,
      "epoch": 1.8076651359943483,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004674053892099521,
      "loss": 5.6626,
      "mean_token_accuracy": 0.1847503513097763,
      "num_tokens": 38159992.0,
      "step": 20470
    },
    {
      "entropy": 5.920422983169556,
      "epoch": 1.8081066760861886,
      "grad_norm": 1.5625,
      "learning_rate": 0.00046738906196270663,
      "loss": 5.5539,
      "mean_token_accuracy": 0.183041051030159,
      "num_tokens": 38168436.0,
      "step": 20475
    },
    {
      "entropy": 5.825107574462891,
      "epoch": 1.8085482161780289,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000467372730946669,
      "loss": 5.5616,
      "mean_token_accuracy": 0.1833862617611885,
      "num_tokens": 38177932.0,
      "step": 20480
    },
    {
      "entropy": 5.884701824188232,
      "epoch": 1.8089897562698694,
      "grad_norm": 1.546875,
      "learning_rate": 0.00046735639616215884,
      "loss": 5.5755,
      "mean_token_accuracy": 0.18837995827198029,
      "num_tokens": 38187727.0,
      "step": 20485
    },
    {
      "entropy": 5.874946451187133,
      "epoch": 1.8094312963617096,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046734005760949634,
      "loss": 5.6363,
      "mean_token_accuracy": 0.17707834839820863,
      "num_tokens": 38198768.0,
      "step": 20490
    },
    {
      "entropy": 5.913182163238526,
      "epoch": 1.80987283645355,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046732371528900154,
      "loss": 5.6246,
      "mean_token_accuracy": 0.1820382758975029,
      "num_tokens": 38207538.0,
      "step": 20495
    },
    {
      "entropy": 5.948500299453736,
      "epoch": 1.8103143765453904,
      "grad_norm": 1.875,
      "learning_rate": 0.0004673073692009945,
      "loss": 5.6539,
      "mean_token_accuracy": 0.18114696592092513,
      "num_tokens": 38216254.0,
      "step": 20500
    },
    {
      "entropy": 5.948800468444825,
      "epoch": 1.8107559166372307,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004672910193457955,
      "loss": 5.6469,
      "mean_token_accuracy": 0.17362466603517532,
      "num_tokens": 38225356.0,
      "step": 20505
    },
    {
      "entropy": 5.859435367584228,
      "epoch": 1.811197456729071,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004672746657237248,
      "loss": 5.5033,
      "mean_token_accuracy": 0.19498553425073623,
      "num_tokens": 38234561.0,
      "step": 20510
    },
    {
      "entropy": 5.8265238285064695,
      "epoch": 1.8116389968209115,
      "grad_norm": 1.5,
      "learning_rate": 0.0004672583083351027,
      "loss": 5.5797,
      "mean_token_accuracy": 0.18321956396102906,
      "num_tokens": 38244418.0,
      "step": 20515
    },
    {
      "entropy": 5.865430021286011,
      "epoch": 1.8120805369127517,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004672419471802497,
      "loss": 5.5779,
      "mean_token_accuracy": 0.1864765927195549,
      "num_tokens": 38253639.0,
      "step": 20520
    },
    {
      "entropy": 5.890454053878784,
      "epoch": 1.812522077004592,
      "grad_norm": 1.734375,
      "learning_rate": 0.00046722558225948627,
      "loss": 5.4754,
      "mean_token_accuracy": 0.2007303535938263,
      "num_tokens": 38262401.0,
      "step": 20525
    },
    {
      "entropy": 5.794748783111572,
      "epoch": 1.8129636170964325,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046720921357313303,
      "loss": 5.5557,
      "mean_token_accuracy": 0.18777734637260438,
      "num_tokens": 38271070.0,
      "step": 20530
    },
    {
      "entropy": 5.882350921630859,
      "epoch": 1.8134051571882726,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004671928411215106,
      "loss": 5.5517,
      "mean_token_accuracy": 0.191727851331234,
      "num_tokens": 38280675.0,
      "step": 20535
    },
    {
      "entropy": 5.900142478942871,
      "epoch": 1.813846697280113,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046717646490493977,
      "loss": 5.637,
      "mean_token_accuracy": 0.18020148873329161,
      "num_tokens": 38290116.0,
      "step": 20540
    },
    {
      "entropy": 5.924641084671021,
      "epoch": 1.8142882373719533,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004671600849237412,
      "loss": 5.5868,
      "mean_token_accuracy": 0.18930605947971343,
      "num_tokens": 38299415.0,
      "step": 20545
    },
    {
      "entropy": 5.725792980194091,
      "epoch": 1.8147297774637936,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004671437011782359,
      "loss": 5.4882,
      "mean_token_accuracy": 0.1988699123263359,
      "num_tokens": 38308761.0,
      "step": 20550
    },
    {
      "entropy": 5.88196120262146,
      "epoch": 1.815171317555634,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004671273136687448,
      "loss": 5.6346,
      "mean_token_accuracy": 0.18313795179128647,
      "num_tokens": 38318212.0,
      "step": 20555
    },
    {
      "entropy": 5.9264246940612795,
      "epoch": 1.8156128576474744,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004671109223955889,
      "loss": 5.5898,
      "mean_token_accuracy": 0.18560122400522233,
      "num_tokens": 38327233.0,
      "step": 20560
    },
    {
      "entropy": 5.828034543991089,
      "epoch": 1.8160543977393147,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004670945273590892,
      "loss": 5.4781,
      "mean_token_accuracy": 0.19380053579807283,
      "num_tokens": 38336473.0,
      "step": 20565
    },
    {
      "entropy": 5.809112024307251,
      "epoch": 1.8164959378311551,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004670781285595671,
      "loss": 5.6269,
      "mean_token_accuracy": 0.18386404812335969,
      "num_tokens": 38345286.0,
      "step": 20570
    },
    {
      "entropy": 5.952305507659912,
      "epoch": 1.8169374779229954,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004670617259973435,
      "loss": 5.5351,
      "mean_token_accuracy": 0.18291568756103516,
      "num_tokens": 38355191.0,
      "step": 20575
    },
    {
      "entropy": 5.869803047180175,
      "epoch": 1.8173790180148357,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046704531967274007,
      "loss": 5.5146,
      "mean_token_accuracy": 0.18477482348680496,
      "num_tokens": 38363902.0,
      "step": 20580
    },
    {
      "entropy": 5.807870006561279,
      "epoch": 1.8178205581066762,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046702890958607804,
      "loss": 5.6886,
      "mean_token_accuracy": 0.17931094616651536,
      "num_tokens": 38374828.0,
      "step": 20585
    },
    {
      "entropy": 5.784125804901123,
      "epoch": 1.8182620981985165,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004670124957376788,
      "loss": 5.5685,
      "mean_token_accuracy": 0.1858455568552017,
      "num_tokens": 38384918.0,
      "step": 20590
    },
    {
      "entropy": 5.954828596115112,
      "epoch": 1.8187036382903567,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004669960781278641,
      "loss": 5.6092,
      "mean_token_accuracy": 0.17778640538454055,
      "num_tokens": 38394005.0,
      "step": 20595
    },
    {
      "entropy": 5.977911329269409,
      "epoch": 1.8191451783821972,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004669796567569553,
      "loss": 5.6576,
      "mean_token_accuracy": 0.18114194720983506,
      "num_tokens": 38402722.0,
      "step": 20600
    },
    {
      "entropy": 5.817265510559082,
      "epoch": 1.8195867184740373,
      "grad_norm": 1.625,
      "learning_rate": 0.0004669632316252743,
      "loss": 5.5635,
      "mean_token_accuracy": 0.18833845406770705,
      "num_tokens": 38411726.0,
      "step": 20605
    },
    {
      "entropy": 5.90852313041687,
      "epoch": 1.8200282585658778,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046694680273314275,
      "loss": 5.6308,
      "mean_token_accuracy": 0.1812708020210266,
      "num_tokens": 38421187.0,
      "step": 20610
    },
    {
      "entropy": 5.91398811340332,
      "epoch": 1.820469798657718,
      "grad_norm": 1.5,
      "learning_rate": 0.0004669303700808824,
      "loss": 5.6513,
      "mean_token_accuracy": 0.17700428813695906,
      "num_tokens": 38430722.0,
      "step": 20615
    },
    {
      "entropy": 5.847461748123169,
      "epoch": 1.8209113387495584,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004669139336688154,
      "loss": 5.5618,
      "mean_token_accuracy": 0.1859545275568962,
      "num_tokens": 38440763.0,
      "step": 20620
    },
    {
      "entropy": 5.895574426651001,
      "epoch": 1.8213528788413988,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046689749349726353,
      "loss": 5.6004,
      "mean_token_accuracy": 0.1838938370347023,
      "num_tokens": 38449298.0,
      "step": 20625
    },
    {
      "entropy": 5.887226343154907,
      "epoch": 1.8217944189332391,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004668810495665489,
      "loss": 5.5603,
      "mean_token_accuracy": 0.18632584512233735,
      "num_tokens": 38458941.0,
      "step": 20630
    },
    {
      "entropy": 5.8180702209472654,
      "epoch": 1.8222359590250794,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004668646018769936,
      "loss": 5.5596,
      "mean_token_accuracy": 0.19531823098659515,
      "num_tokens": 38468381.0,
      "step": 20635
    },
    {
      "entropy": 5.929505395889282,
      "epoch": 1.82267749911692,
      "grad_norm": 1.546875,
      "learning_rate": 0.00046684815042891993,
      "loss": 5.6042,
      "mean_token_accuracy": 0.1758841574192047,
      "num_tokens": 38477767.0,
      "step": 20640
    },
    {
      "entropy": 5.819535589218139,
      "epoch": 1.8231190392087602,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004668316952226501,
      "loss": 5.5192,
      "mean_token_accuracy": 0.18697240203619003,
      "num_tokens": 38486884.0,
      "step": 20645
    },
    {
      "entropy": 5.846132087707519,
      "epoch": 1.8235605793006004,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004668152362585064,
      "loss": 5.6177,
      "mean_token_accuracy": 0.1790119007229805,
      "num_tokens": 38496243.0,
      "step": 20650
    },
    {
      "entropy": 5.844096994400024,
      "epoch": 1.824002119392441,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004667987735368113,
      "loss": 5.5043,
      "mean_token_accuracy": 0.18455007970333098,
      "num_tokens": 38504447.0,
      "step": 20655
    },
    {
      "entropy": 5.9046063899993895,
      "epoch": 1.8244436594842812,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004667823070578874,
      "loss": 5.6268,
      "mean_token_accuracy": 0.183837890625,
      "num_tokens": 38514423.0,
      "step": 20660
    },
    {
      "entropy": 5.931299066543579,
      "epoch": 1.8248851995761215,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046676583682205713,
      "loss": 5.5466,
      "mean_token_accuracy": 0.18486977219581605,
      "num_tokens": 38522765.0,
      "step": 20665
    },
    {
      "entropy": 5.9328898906707765,
      "epoch": 1.825326739667962,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046674936282964326,
      "loss": 5.6916,
      "mean_token_accuracy": 0.17533372938632966,
      "num_tokens": 38532074.0,
      "step": 20670
    },
    {
      "entropy": 5.865649747848511,
      "epoch": 1.825768279759802,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004667328850809683,
      "loss": 5.5369,
      "mean_token_accuracy": 0.1850859358906746,
      "num_tokens": 38541423.0,
      "step": 20675
    },
    {
      "entropy": 5.814494943618774,
      "epoch": 1.8262098198516425,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004667164035763552,
      "loss": 5.6327,
      "mean_token_accuracy": 0.18360775113105773,
      "num_tokens": 38552369.0,
      "step": 20680
    },
    {
      "entropy": 5.925185203552246,
      "epoch": 1.8266513599434828,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004666999183161268,
      "loss": 5.5892,
      "mean_token_accuracy": 0.1861972764134407,
      "num_tokens": 38561310.0,
      "step": 20685
    },
    {
      "entropy": 5.926451730728149,
      "epoch": 1.827092900035323,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046668342930060604,
      "loss": 5.6447,
      "mean_token_accuracy": 0.180878147482872,
      "num_tokens": 38571157.0,
      "step": 20690
    },
    {
      "entropy": 5.93447093963623,
      "epoch": 1.8275344401271636,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046666693653011596,
      "loss": 5.5642,
      "mean_token_accuracy": 0.18415466845035552,
      "num_tokens": 38579862.0,
      "step": 20695
    },
    {
      "entropy": 5.926377582550049,
      "epoch": 1.8279759802190039,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046665044000497954,
      "loss": 5.63,
      "mean_token_accuracy": 0.18385003209114076,
      "num_tokens": 38589784.0,
      "step": 20700
    },
    {
      "entropy": 5.901109266281128,
      "epoch": 1.8284175203108441,
      "grad_norm": 1.6875,
      "learning_rate": 0.00046663393972551994,
      "loss": 5.5482,
      "mean_token_accuracy": 0.18796993792057037,
      "num_tokens": 38598925.0,
      "step": 20705
    },
    {
      "entropy": 5.902924108505249,
      "epoch": 1.8288590604026846,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046661743569206056,
      "loss": 5.6124,
      "mean_token_accuracy": 0.17928789258003236,
      "num_tokens": 38608245.0,
      "step": 20710
    },
    {
      "entropy": 5.919569826126098,
      "epoch": 1.829300600494525,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004666009279049245,
      "loss": 5.6753,
      "mean_token_accuracy": 0.18005069345235825,
      "num_tokens": 38617586.0,
      "step": 20715
    },
    {
      "entropy": 5.937134647369385,
      "epoch": 1.8297421405863652,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046658441636443526,
      "loss": 5.6197,
      "mean_token_accuracy": 0.18627241849899293,
      "num_tokens": 38626420.0,
      "step": 20720
    },
    {
      "entropy": 5.952634477615357,
      "epoch": 1.8301836806782057,
      "grad_norm": 6.9375,
      "learning_rate": 0.00046656790107091627,
      "loss": 5.5582,
      "mean_token_accuracy": 0.18574341237545014,
      "num_tokens": 38635585.0,
      "step": 20725
    },
    {
      "entropy": 5.90435004234314,
      "epoch": 1.830625220770046,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004665513820246911,
      "loss": 5.5679,
      "mean_token_accuracy": 0.18342121839523315,
      "num_tokens": 38643998.0,
      "step": 20730
    },
    {
      "entropy": 5.933841753005981,
      "epoch": 1.8310667608618862,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004665348592260833,
      "loss": 5.6114,
      "mean_token_accuracy": 0.18591436445713044,
      "num_tokens": 38653018.0,
      "step": 20735
    },
    {
      "entropy": 5.864882707595825,
      "epoch": 1.8315083009537267,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004665183326754165,
      "loss": 5.54,
      "mean_token_accuracy": 0.19055962711572647,
      "num_tokens": 38662079.0,
      "step": 20740
    },
    {
      "entropy": 5.866421365737915,
      "epoch": 1.8319498410455668,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004665018023730144,
      "loss": 5.5885,
      "mean_token_accuracy": 0.18704347163438798,
      "num_tokens": 38669928.0,
      "step": 20745
    },
    {
      "entropy": 5.833478546142578,
      "epoch": 1.8323913811374073,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000466485268319201,
      "loss": 5.5527,
      "mean_token_accuracy": 0.18304758369922638,
      "num_tokens": 38679475.0,
      "step": 20750
    },
    {
      "entropy": 5.922449207305908,
      "epoch": 1.8328329212292476,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004664687305143001,
      "loss": 5.6342,
      "mean_token_accuracy": 0.18515184968709947,
      "num_tokens": 38689057.0,
      "step": 20755
    },
    {
      "entropy": 5.902431535720825,
      "epoch": 1.8332744613210878,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046645218895863564,
      "loss": 5.4768,
      "mean_token_accuracy": 0.19860193878412247,
      "num_tokens": 38698362.0,
      "step": 20760
    },
    {
      "entropy": 5.793140172958374,
      "epoch": 1.8337160014129283,
      "grad_norm": 1.59375,
      "learning_rate": 0.00046643564365253173,
      "loss": 5.5057,
      "mean_token_accuracy": 0.19486292451620102,
      "num_tokens": 38708319.0,
      "step": 20765
    },
    {
      "entropy": 5.893560075759888,
      "epoch": 1.8341575415047686,
      "grad_norm": 1.375,
      "learning_rate": 0.0004664190945963124,
      "loss": 5.6266,
      "mean_token_accuracy": 0.18076992630958558,
      "num_tokens": 38717441.0,
      "step": 20770
    },
    {
      "entropy": 5.988618040084839,
      "epoch": 1.834599081596609,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046640254179030184,
      "loss": 5.6,
      "mean_token_accuracy": 0.17892884612083435,
      "num_tokens": 38726903.0,
      "step": 20775
    },
    {
      "entropy": 5.863165378570557,
      "epoch": 1.8350406216884494,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004663859852348244,
      "loss": 5.552,
      "mean_token_accuracy": 0.18922021239995956,
      "num_tokens": 38735983.0,
      "step": 20780
    },
    {
      "entropy": 5.855266761779785,
      "epoch": 1.8354821617802897,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046636942493020437,
      "loss": 5.6417,
      "mean_token_accuracy": 0.18647949397563934,
      "num_tokens": 38745017.0,
      "step": 20785
    },
    {
      "entropy": 5.816526317596436,
      "epoch": 1.83592370187213,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004663528608767661,
      "loss": 5.5104,
      "mean_token_accuracy": 0.18803058415651322,
      "num_tokens": 38754756.0,
      "step": 20790
    },
    {
      "entropy": 5.924617433547974,
      "epoch": 1.8363652419639704,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004663362930748341,
      "loss": 5.5739,
      "mean_token_accuracy": 0.18934491872787476,
      "num_tokens": 38763695.0,
      "step": 20795
    },
    {
      "entropy": 5.915267992019653,
      "epoch": 1.8368067820558107,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046631972152473295,
      "loss": 5.6668,
      "mean_token_accuracy": 0.1802893191576004,
      "num_tokens": 38773637.0,
      "step": 20800
    },
    {
      "entropy": 5.805772924423218,
      "epoch": 1.837248322147651,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046630314622678724,
      "loss": 5.4721,
      "mean_token_accuracy": 0.19938332736492156,
      "num_tokens": 38783004.0,
      "step": 20805
    },
    {
      "entropy": 5.850068187713623,
      "epoch": 1.8376898622394915,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004662865671813217,
      "loss": 5.5735,
      "mean_token_accuracy": 0.19182733744382857,
      "num_tokens": 38791342.0,
      "step": 20810
    },
    {
      "entropy": 5.8149511337280275,
      "epoch": 1.8381314023313315,
      "grad_norm": 1.671875,
      "learning_rate": 0.00046626998438866114,
      "loss": 5.5273,
      "mean_token_accuracy": 0.187217977643013,
      "num_tokens": 38801046.0,
      "step": 20815
    },
    {
      "entropy": 5.883310413360595,
      "epoch": 1.838572942423172,
      "grad_norm": 1.75,
      "learning_rate": 0.00046625339784913035,
      "loss": 5.5757,
      "mean_token_accuracy": 0.18511651456356049,
      "num_tokens": 38810806.0,
      "step": 20820
    },
    {
      "entropy": 5.920738649368286,
      "epoch": 1.8390144825150123,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046623680756305427,
      "loss": 5.6281,
      "mean_token_accuracy": 0.18006236851215363,
      "num_tokens": 38819362.0,
      "step": 20825
    },
    {
      "entropy": 5.923454856872558,
      "epoch": 1.8394560226068526,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004662202135307579,
      "loss": 5.6161,
      "mean_token_accuracy": 0.18292374014854432,
      "num_tokens": 38829337.0,
      "step": 20830
    },
    {
      "entropy": 5.883269548416138,
      "epoch": 1.839897562698693,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004662036157525663,
      "loss": 5.5808,
      "mean_token_accuracy": 0.1797638416290283,
      "num_tokens": 38838351.0,
      "step": 20835
    },
    {
      "entropy": 5.951020431518555,
      "epoch": 1.8403391027905334,
      "grad_norm": 1.78125,
      "learning_rate": 0.00046618701422880456,
      "loss": 5.6402,
      "mean_token_accuracy": 0.18052633255720138,
      "num_tokens": 38847830.0,
      "step": 20840
    },
    {
      "entropy": 5.943289804458618,
      "epoch": 1.8407806428823736,
      "grad_norm": 1.578125,
      "learning_rate": 0.000466170408959798,
      "loss": 5.6233,
      "mean_token_accuracy": 0.17599236518144606,
      "num_tokens": 38856945.0,
      "step": 20845
    },
    {
      "entropy": 5.881523942947387,
      "epoch": 1.8412221829742141,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004661537999458718,
      "loss": 5.5724,
      "mean_token_accuracy": 0.18212804943323135,
      "num_tokens": 38866493.0,
      "step": 20850
    },
    {
      "entropy": 5.821108198165893,
      "epoch": 1.8416637230660544,
      "grad_norm": 1.6875,
      "learning_rate": 0.00046613718718735143,
      "loss": 5.5217,
      "mean_token_accuracy": 0.18511007130146026,
      "num_tokens": 38875653.0,
      "step": 20855
    },
    {
      "entropy": 5.799018001556396,
      "epoch": 1.8421052631578947,
      "grad_norm": 1.59375,
      "learning_rate": 0.00046612057068456226,
      "loss": 5.4473,
      "mean_token_accuracy": 0.1961596354842186,
      "num_tokens": 38884740.0,
      "step": 20860
    },
    {
      "entropy": 5.823368835449219,
      "epoch": 1.8425468032497352,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00046610395043782983,
      "loss": 5.569,
      "mean_token_accuracy": 0.19108745008707045,
      "num_tokens": 38893801.0,
      "step": 20865
    },
    {
      "entropy": 5.859483146667481,
      "epoch": 1.8429883433415755,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046608732644747963,
      "loss": 5.5761,
      "mean_token_accuracy": 0.19179200530052185,
      "num_tokens": 38902298.0,
      "step": 20870
    },
    {
      "entropy": 5.87000002861023,
      "epoch": 1.8434298834334157,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046607069871383745,
      "loss": 5.4754,
      "mean_token_accuracy": 0.19641245901584625,
      "num_tokens": 38910904.0,
      "step": 20875
    },
    {
      "entropy": 5.811917734146118,
      "epoch": 1.8438714235252562,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004660540672372289,
      "loss": 5.4753,
      "mean_token_accuracy": 0.18979522883892058,
      "num_tokens": 38920405.0,
      "step": 20880
    },
    {
      "entropy": 5.822230243682862,
      "epoch": 1.8443129636170963,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004660374320179798,
      "loss": 5.4768,
      "mean_token_accuracy": 0.18495012819766998,
      "num_tokens": 38930118.0,
      "step": 20885
    },
    {
      "entropy": 5.860304832458496,
      "epoch": 1.8447545037089368,
      "grad_norm": 2.453125,
      "learning_rate": 0.00046602079305641615,
      "loss": 5.5871,
      "mean_token_accuracy": 0.18567434400320054,
      "num_tokens": 38939388.0,
      "step": 20890
    },
    {
      "entropy": 5.845680284500122,
      "epoch": 1.845196043800777,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046600415035286374,
      "loss": 5.5968,
      "mean_token_accuracy": 0.18590677082538604,
      "num_tokens": 38948632.0,
      "step": 20895
    },
    {
      "entropy": 5.91842679977417,
      "epoch": 1.8456375838926173,
      "grad_norm": 1.5,
      "learning_rate": 0.0004659875039076487,
      "loss": 5.6011,
      "mean_token_accuracy": 0.18062607944011688,
      "num_tokens": 38957504.0,
      "step": 20900
    },
    {
      "entropy": 5.844452667236328,
      "epoch": 1.8460791239844578,
      "grad_norm": 1.5625,
      "learning_rate": 0.000465970853721097,
      "loss": 5.5671,
      "mean_token_accuracy": 0.18301562517881392,
      "num_tokens": 38967713.0,
      "step": 20905
    },
    {
      "entropy": 5.877889585494995,
      "epoch": 1.8465206640762981,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00046595419979353494,
      "loss": 5.5781,
      "mean_token_accuracy": 0.18354347199201584,
      "num_tokens": 38976982.0,
      "step": 20910
    },
    {
      "entropy": 5.820690107345581,
      "epoch": 1.8469622041681384,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004659375421252887,
      "loss": 5.5477,
      "mean_token_accuracy": 0.18609475791454316,
      "num_tokens": 38985479.0,
      "step": 20915
    },
    {
      "entropy": 5.935887575149536,
      "epoch": 1.8474037442599789,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004659208807166846,
      "loss": 5.6361,
      "mean_token_accuracy": 0.1747157171368599,
      "num_tokens": 38995418.0,
      "step": 20920
    },
    {
      "entropy": 5.911508560180664,
      "epoch": 1.8478452843518192,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004659042155680491,
      "loss": 5.5546,
      "mean_token_accuracy": 0.1890340656042099,
      "num_tokens": 39004961.0,
      "step": 20925
    },
    {
      "entropy": 5.7955464839935305,
      "epoch": 1.8482868244436594,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004658875466797085,
      "loss": 5.541,
      "mean_token_accuracy": 0.19053864032030104,
      "num_tokens": 39014282.0,
      "step": 20930
    },
    {
      "entropy": 5.8516473293304445,
      "epoch": 1.8487283645355,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046587087405198943,
      "loss": 5.5848,
      "mean_token_accuracy": 0.18203202337026597,
      "num_tokens": 39024480.0,
      "step": 20935
    },
    {
      "entropy": 5.962622213363647,
      "epoch": 1.8491699046273402,
      "grad_norm": 1.375,
      "learning_rate": 0.00046585419768521854,
      "loss": 5.5903,
      "mean_token_accuracy": 0.1856498971581459,
      "num_tokens": 39034363.0,
      "step": 20940
    },
    {
      "entropy": 5.850033187866211,
      "epoch": 1.8496114447191805,
      "grad_norm": 1.921875,
      "learning_rate": 0.00046583751757972244,
      "loss": 5.5196,
      "mean_token_accuracy": 0.1892829954624176,
      "num_tokens": 39043823.0,
      "step": 20945
    },
    {
      "entropy": 5.800778007507324,
      "epoch": 1.850052984811021,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004658208337358279,
      "loss": 5.5257,
      "mean_token_accuracy": 0.1894927054643631,
      "num_tokens": 39052219.0,
      "step": 20950
    },
    {
      "entropy": 5.875915241241455,
      "epoch": 1.850494524902861,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046580414615386175,
      "loss": 5.6466,
      "mean_token_accuracy": 0.17950430363416672,
      "num_tokens": 39062974.0,
      "step": 20955
    },
    {
      "entropy": 5.724501705169677,
      "epoch": 1.8509360649947015,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00046578745483415087,
      "loss": 5.4032,
      "mean_token_accuracy": 0.20113253742456436,
      "num_tokens": 39072178.0,
      "step": 20960
    },
    {
      "entropy": 5.864728879928589,
      "epoch": 1.8513776050865418,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004657707597770223,
      "loss": 5.5835,
      "mean_token_accuracy": 0.18381558060646058,
      "num_tokens": 39081494.0,
      "step": 20965
    },
    {
      "entropy": 5.887950277328491,
      "epoch": 1.851819145178382,
      "grad_norm": 1.40625,
      "learning_rate": 0.000465754060982803,
      "loss": 5.5899,
      "mean_token_accuracy": 0.1775693863630295,
      "num_tokens": 39090783.0,
      "step": 20970
    },
    {
      "entropy": 5.905765008926392,
      "epoch": 1.8522606852702226,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004657373584518201,
      "loss": 5.5916,
      "mean_token_accuracy": 0.1888223335146904,
      "num_tokens": 39100150.0,
      "step": 20975
    },
    {
      "entropy": 5.905903625488281,
      "epoch": 1.8527022253620629,
      "grad_norm": 1.59375,
      "learning_rate": 0.000465720652184401,
      "loss": 5.6277,
      "mean_token_accuracy": 0.18231217861175536,
      "num_tokens": 39109419.0,
      "step": 20980
    },
    {
      "entropy": 5.790070056915283,
      "epoch": 1.8531437654539031,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004657039421808726,
      "loss": 5.5032,
      "mean_token_accuracy": 0.19081190824508668,
      "num_tokens": 39118312.0,
      "step": 20985
    },
    {
      "entropy": 5.9055784225463865,
      "epoch": 1.8535853055457436,
      "grad_norm": 2.5,
      "learning_rate": 0.00046568722844156254,
      "loss": 5.6415,
      "mean_token_accuracy": 0.17673494964838027,
      "num_tokens": 39128169.0,
      "step": 20990
    },
    {
      "entropy": 5.883026885986328,
      "epoch": 1.854026845637584,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046567051096679807,
      "loss": 5.5935,
      "mean_token_accuracy": 0.1900536298751831,
      "num_tokens": 39136790.0,
      "step": 20995
    },
    {
      "entropy": 5.874770212173462,
      "epoch": 1.8544683857294242,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004656537897569068,
      "loss": 5.5152,
      "mean_token_accuracy": 0.19180358350276946,
      "num_tokens": 39146974.0,
      "step": 21000
    },
    {
      "epoch": 1.8544683857294242,
      "eval_entropy": 5.700860997963599,
      "eval_loss": 5.643547058105469,
      "eval_mean_token_accuracy": 0.18990165647836496,
      "eval_num_tokens": 39146974.0,
      "eval_runtime": 26.1289,
      "eval_samples_per_second": 1351.57,
      "eval_steps_per_second": 168.97,
      "step": 21000
    },
    {
      "entropy": 5.943000078201294,
      "epoch": 1.8549099258212647,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004656370648122161,
      "loss": 5.7267,
      "mean_token_accuracy": 0.1755589485168457,
      "num_tokens": 39156910.0,
      "step": 21005
    },
    {
      "entropy": 5.934307003021241,
      "epoch": 1.855351465913105,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00046562033613305377,
      "loss": 5.662,
      "mean_token_accuracy": 0.17969456315040588,
      "num_tokens": 39165790.0,
      "step": 21010
    },
    {
      "entropy": 5.800854825973511,
      "epoch": 1.8557930060049452,
      "grad_norm": 1.375,
      "learning_rate": 0.0004656036037197475,
      "loss": 5.4553,
      "mean_token_accuracy": 0.20389885157346727,
      "num_tokens": 39174811.0,
      "step": 21015
    },
    {
      "entropy": 5.860662031173706,
      "epoch": 1.8562345460967857,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004655868675726251,
      "loss": 5.6031,
      "mean_token_accuracy": 0.17940495461225509,
      "num_tokens": 39184325.0,
      "step": 21020
    },
    {
      "entropy": 5.899876260757447,
      "epoch": 1.8566760861886258,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046557012769201425,
      "loss": 5.545,
      "mean_token_accuracy": 0.18737968504428865,
      "num_tokens": 39194879.0,
      "step": 21025
    },
    {
      "entropy": 5.853872966766358,
      "epoch": 1.8571176262804663,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000465553384078243,
      "loss": 5.5858,
      "mean_token_accuracy": 0.183933587372303,
      "num_tokens": 39204702.0,
      "step": 21030
    },
    {
      "entropy": 5.8990123748779295,
      "epoch": 1.8575591663723066,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004655366367316394,
      "loss": 5.6153,
      "mean_token_accuracy": 0.18828382045030595,
      "num_tokens": 39214561.0,
      "step": 21035
    },
    {
      "entropy": 5.959616756439209,
      "epoch": 1.8580007064641468,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00046551988565253136,
      "loss": 5.6792,
      "mean_token_accuracy": 0.17426469922065735,
      "num_tokens": 39224411.0,
      "step": 21040
    },
    {
      "entropy": 5.9067436218261715,
      "epoch": 1.8584422465559873,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004655031308412471,
      "loss": 5.5869,
      "mean_token_accuracy": 0.18858598470687865,
      "num_tokens": 39233422.0,
      "step": 21045
    },
    {
      "entropy": 5.956392097473144,
      "epoch": 1.8588837866478276,
      "grad_norm": 1.96875,
      "learning_rate": 0.00046548637229811495,
      "loss": 5.614,
      "mean_token_accuracy": 0.17941299378871917,
      "num_tokens": 39243919.0,
      "step": 21050
    },
    {
      "entropy": 5.948605680465699,
      "epoch": 1.8593253267396679,
      "grad_norm": 1.5625,
      "learning_rate": 0.00046546961002346304,
      "loss": 5.6651,
      "mean_token_accuracy": 0.1787140429019928,
      "num_tokens": 39253682.0,
      "step": 21055
    },
    {
      "entropy": 5.8835350513458256,
      "epoch": 1.8597668668315084,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004654528440176198,
      "loss": 5.6443,
      "mean_token_accuracy": 0.18680257499217987,
      "num_tokens": 39263855.0,
      "step": 21060
    },
    {
      "entropy": 5.914743041992187,
      "epoch": 1.8602084069233487,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004654360742809137,
      "loss": 5.5346,
      "mean_token_accuracy": 0.1903542011976242,
      "num_tokens": 39272292.0,
      "step": 21065
    },
    {
      "entropy": 5.8829957962036135,
      "epoch": 1.860649947015189,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004654193008136732,
      "loss": 5.5365,
      "mean_token_accuracy": 0.18821685016155243,
      "num_tokens": 39281321.0,
      "step": 21070
    },
    {
      "entropy": 5.93348798751831,
      "epoch": 1.8610914871070294,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046540252361622683,
      "loss": 5.6974,
      "mean_token_accuracy": 0.17625337988138198,
      "num_tokens": 39292255.0,
      "step": 21075
    },
    {
      "entropy": 5.940316152572632,
      "epoch": 1.8615330271988697,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046538574268890334,
      "loss": 5.6148,
      "mean_token_accuracy": 0.18176422119140626,
      "num_tokens": 39301841.0,
      "step": 21080
    },
    {
      "entropy": 5.900913858413697,
      "epoch": 1.86197456729071,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046536895803203144,
      "loss": 5.5843,
      "mean_token_accuracy": 0.17615684419870375,
      "num_tokens": 39311238.0,
      "step": 21085
    },
    {
      "entropy": 5.888750696182251,
      "epoch": 1.8624161073825505,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00046535216964593997,
      "loss": 5.5866,
      "mean_token_accuracy": 0.18416304737329484,
      "num_tokens": 39319945.0,
      "step": 21090
    },
    {
      "entropy": 5.843233823776245,
      "epoch": 1.8628576474743905,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004653353775309576,
      "loss": 5.565,
      "mean_token_accuracy": 0.18230943977832795,
      "num_tokens": 39329436.0,
      "step": 21095
    },
    {
      "entropy": 5.910763216018677,
      "epoch": 1.863299187566231,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004653185816874136,
      "loss": 5.598,
      "mean_token_accuracy": 0.18619284629821778,
      "num_tokens": 39338447.0,
      "step": 21100
    },
    {
      "entropy": 5.804048633575439,
      "epoch": 1.8637407276580713,
      "grad_norm": 1.375,
      "learning_rate": 0.0004653017821156368,
      "loss": 5.5027,
      "mean_token_accuracy": 0.19504960477352143,
      "num_tokens": 39346744.0,
      "step": 21105
    },
    {
      "entropy": 5.839439439773559,
      "epoch": 1.8641822677499116,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004652849788159563,
      "loss": 5.6266,
      "mean_token_accuracy": 0.17359012812376023,
      "num_tokens": 39357402.0,
      "step": 21110
    },
    {
      "entropy": 5.923741102218628,
      "epoch": 1.864623807841752,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046526817178870123,
      "loss": 5.616,
      "mean_token_accuracy": 0.18167375177145004,
      "num_tokens": 39366902.0,
      "step": 21115
    },
    {
      "entropy": 5.942162370681762,
      "epoch": 1.8650653479335924,
      "grad_norm": 1.53125,
      "learning_rate": 0.00046525136103420093,
      "loss": 5.5817,
      "mean_token_accuracy": 0.1864108219742775,
      "num_tokens": 39376330.0,
      "step": 21120
    },
    {
      "entropy": 5.835435914993286,
      "epoch": 1.8655068880254326,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046523454655278466,
      "loss": 5.5386,
      "mean_token_accuracy": 0.1944829210639,
      "num_tokens": 39385926.0,
      "step": 21125
    },
    {
      "entropy": 5.871251344680786,
      "epoch": 1.8659484281172731,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004652177283447818,
      "loss": 5.5136,
      "mean_token_accuracy": 0.18046933859586717,
      "num_tokens": 39395558.0,
      "step": 21130
    },
    {
      "entropy": 5.907923316955566,
      "epoch": 1.8663899682091134,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00046520090641052187,
      "loss": 5.5918,
      "mean_token_accuracy": 0.18496931195259095,
      "num_tokens": 39405072.0,
      "step": 21135
    },
    {
      "entropy": 5.898514938354492,
      "epoch": 1.8668315083009537,
      "grad_norm": 1.609375,
      "learning_rate": 0.00046518408075033437,
      "loss": 5.6906,
      "mean_token_accuracy": 0.1811235874891281,
      "num_tokens": 39414714.0,
      "step": 21140
    },
    {
      "entropy": 5.862201404571533,
      "epoch": 1.8672730483927942,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004651672513645488,
      "loss": 5.4859,
      "mean_token_accuracy": 0.187942835688591,
      "num_tokens": 39423189.0,
      "step": 21145
    },
    {
      "entropy": 5.842456436157226,
      "epoch": 1.8677145884846345,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046515041825349505,
      "loss": 5.6368,
      "mean_token_accuracy": 0.17876768559217454,
      "num_tokens": 39432791.0,
      "step": 21150
    },
    {
      "entropy": 5.898605060577393,
      "epoch": 1.8681561285764747,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004651335814175027,
      "loss": 5.5901,
      "mean_token_accuracy": 0.18209777176380157,
      "num_tokens": 39442849.0,
      "step": 21155
    },
    {
      "entropy": 5.929827404022217,
      "epoch": 1.8685976686683152,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046511674085690155,
      "loss": 5.5969,
      "mean_token_accuracy": 0.18350279331207275,
      "num_tokens": 39452171.0,
      "step": 21160
    },
    {
      "entropy": 5.893155336380005,
      "epoch": 1.8690392087601553,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004650998965720216,
      "loss": 5.5933,
      "mean_token_accuracy": 0.1811112254858017,
      "num_tokens": 39460351.0,
      "step": 21165
    },
    {
      "entropy": 5.927641487121582,
      "epoch": 1.8694807488519958,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046508304856319283,
      "loss": 5.652,
      "mean_token_accuracy": 0.18394415825605392,
      "num_tokens": 39470050.0,
      "step": 21170
    },
    {
      "entropy": 5.85578465461731,
      "epoch": 1.869922288943836,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046506619683074526,
      "loss": 5.5715,
      "mean_token_accuracy": 0.19082047343254088,
      "num_tokens": 39480192.0,
      "step": 21175
    },
    {
      "entropy": 5.843869066238403,
      "epoch": 1.8703638290356763,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046504934137500903,
      "loss": 5.6261,
      "mean_token_accuracy": 0.188265123963356,
      "num_tokens": 39490118.0,
      "step": 21180
    },
    {
      "entropy": 5.949170684814453,
      "epoch": 1.8708053691275168,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046503248219631417,
      "loss": 5.6502,
      "mean_token_accuracy": 0.1784625917673111,
      "num_tokens": 39499173.0,
      "step": 21185
    },
    {
      "entropy": 5.856217336654663,
      "epoch": 1.871246909219357,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046501561929499115,
      "loss": 5.4873,
      "mean_token_accuracy": 0.1948787659406662,
      "num_tokens": 39508595.0,
      "step": 21190
    },
    {
      "entropy": 5.8092328071594235,
      "epoch": 1.8716884493111974,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046499875267137016,
      "loss": 5.5341,
      "mean_token_accuracy": 0.19684279561042786,
      "num_tokens": 39519300.0,
      "step": 21195
    },
    {
      "entropy": 5.90353307723999,
      "epoch": 1.8721299894030379,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004649818823257817,
      "loss": 5.5912,
      "mean_token_accuracy": 0.18269855678081512,
      "num_tokens": 39527722.0,
      "step": 21200
    },
    {
      "entropy": 5.89766035079956,
      "epoch": 1.8725715294948782,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004649650082585562,
      "loss": 5.5669,
      "mean_token_accuracy": 0.1870953619480133,
      "num_tokens": 39537063.0,
      "step": 21205
    },
    {
      "entropy": 5.883340120315552,
      "epoch": 1.8730130695867184,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00046494813047002424,
      "loss": 5.5963,
      "mean_token_accuracy": 0.18584158420562744,
      "num_tokens": 39546402.0,
      "step": 21210
    },
    {
      "entropy": 5.955042171478271,
      "epoch": 1.873454609678559,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046493124896051634,
      "loss": 5.5991,
      "mean_token_accuracy": 0.18769053369760513,
      "num_tokens": 39555877.0,
      "step": 21215
    },
    {
      "entropy": 5.840330505371094,
      "epoch": 1.8738961497703992,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00046491436373036346,
      "loss": 5.5556,
      "mean_token_accuracy": 0.1861400380730629,
      "num_tokens": 39565153.0,
      "step": 21220
    },
    {
      "entropy": 5.872222089767456,
      "epoch": 1.8743376898622395,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004648974747798961,
      "loss": 5.5792,
      "mean_token_accuracy": 0.18427794724702834,
      "num_tokens": 39573775.0,
      "step": 21225
    },
    {
      "entropy": 5.895746469497681,
      "epoch": 1.87477922995408,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004648805821094452,
      "loss": 5.6545,
      "mean_token_accuracy": 0.176860012114048,
      "num_tokens": 39583159.0,
      "step": 21230
    },
    {
      "entropy": 5.9624122142791744,
      "epoch": 1.87522077004592,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046486368571934175,
      "loss": 5.6626,
      "mean_token_accuracy": 0.17608078718185424,
      "num_tokens": 39592545.0,
      "step": 21235
    },
    {
      "entropy": 5.705715465545654,
      "epoch": 1.8756623101377605,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004648467856099167,
      "loss": 5.4085,
      "mean_token_accuracy": 0.1982883706688881,
      "num_tokens": 39602059.0,
      "step": 21240
    },
    {
      "entropy": 5.855539798736572,
      "epoch": 1.8761038502296008,
      "grad_norm": 1.671875,
      "learning_rate": 0.000464829881781501,
      "loss": 5.5529,
      "mean_token_accuracy": 0.18427469879388808,
      "num_tokens": 39610777.0,
      "step": 21245
    },
    {
      "entropy": 5.870946836471558,
      "epoch": 1.876545390321441,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046481297423442593,
      "loss": 5.6314,
      "mean_token_accuracy": 0.18627751171588897,
      "num_tokens": 39620782.0,
      "step": 21250
    },
    {
      "entropy": 5.985285425186158,
      "epoch": 1.8769869304132816,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004647960629690226,
      "loss": 5.6666,
      "mean_token_accuracy": 0.17546181231737137,
      "num_tokens": 39631340.0,
      "step": 21255
    },
    {
      "entropy": 5.824886465072632,
      "epoch": 1.8774284705051218,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004647791479856223,
      "loss": 5.4917,
      "mean_token_accuracy": 0.19251861274242402,
      "num_tokens": 39640361.0,
      "step": 21260
    },
    {
      "entropy": 5.875917577743531,
      "epoch": 1.8778700105969621,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004647622292845566,
      "loss": 5.567,
      "mean_token_accuracy": 0.18602023124694825,
      "num_tokens": 39649726.0,
      "step": 21265
    },
    {
      "entropy": 5.864447021484375,
      "epoch": 1.8783115506888026,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004647453068661566,
      "loss": 5.6157,
      "mean_token_accuracy": 0.1876423180103302,
      "num_tokens": 39660226.0,
      "step": 21270
    },
    {
      "entropy": 5.821967601776123,
      "epoch": 1.878753090780643,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046472838073075397,
      "loss": 5.4195,
      "mean_token_accuracy": 0.1936606138944626,
      "num_tokens": 39668731.0,
      "step": 21275
    },
    {
      "entropy": 5.878619527816772,
      "epoch": 1.8791946308724832,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046471145087868034,
      "loss": 5.5845,
      "mean_token_accuracy": 0.19186805784702302,
      "num_tokens": 39678465.0,
      "step": 21280
    },
    {
      "entropy": 5.843048238754273,
      "epoch": 1.8796361709643237,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046469451731026725,
      "loss": 5.4334,
      "mean_token_accuracy": 0.19977840334177016,
      "num_tokens": 39687057.0,
      "step": 21285
    },
    {
      "entropy": 5.768490648269653,
      "epoch": 1.880077711056164,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004646775800258464,
      "loss": 5.5797,
      "mean_token_accuracy": 0.19160076081752778,
      "num_tokens": 39696387.0,
      "step": 21290
    },
    {
      "entropy": 5.8621796607971195,
      "epoch": 1.8805192511480042,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046466063902574973,
      "loss": 5.5366,
      "mean_token_accuracy": 0.18740716725587844,
      "num_tokens": 39705756.0,
      "step": 21295
    },
    {
      "entropy": 5.897675561904907,
      "epoch": 1.8809607912398447,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046464369431030894,
      "loss": 5.5578,
      "mean_token_accuracy": 0.18925928622484206,
      "num_tokens": 39713916.0,
      "step": 21300
    },
    {
      "entropy": 5.878945255279541,
      "epoch": 1.8814023313316848,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000464626745879856,
      "loss": 5.6524,
      "mean_token_accuracy": 0.1777836099267006,
      "num_tokens": 39722920.0,
      "step": 21305
    },
    {
      "entropy": 5.895596313476562,
      "epoch": 1.8818438714235253,
      "grad_norm": 1.5625,
      "learning_rate": 0.000464609793734723,
      "loss": 5.6391,
      "mean_token_accuracy": 0.17911608070135115,
      "num_tokens": 39732217.0,
      "step": 21310
    },
    {
      "entropy": 5.796049928665161,
      "epoch": 1.8822854115153655,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000464592837875242,
      "loss": 5.4718,
      "mean_token_accuracy": 0.1914995327591896,
      "num_tokens": 39741636.0,
      "step": 21315
    },
    {
      "entropy": 5.889091539382934,
      "epoch": 1.8827269516072058,
      "grad_norm": 1.5,
      "learning_rate": 0.0004645758783017451,
      "loss": 5.6082,
      "mean_token_accuracy": 0.1809345304965973,
      "num_tokens": 39751457.0,
      "step": 21320
    },
    {
      "entropy": 5.94678144454956,
      "epoch": 1.8831684916990463,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046455891501456456,
      "loss": 5.5615,
      "mean_token_accuracy": 0.18901580423116685,
      "num_tokens": 39760308.0,
      "step": 21325
    },
    {
      "entropy": 5.818446302413941,
      "epoch": 1.8836100317908866,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046454194801403274,
      "loss": 5.523,
      "mean_token_accuracy": 0.19438900649547577,
      "num_tokens": 39770002.0,
      "step": 21330
    },
    {
      "entropy": 5.725969314575195,
      "epoch": 1.8840515718827269,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004645249773004819,
      "loss": 5.5313,
      "mean_token_accuracy": 0.1864411324262619,
      "num_tokens": 39779043.0,
      "step": 21335
    },
    {
      "entropy": 5.937741470336914,
      "epoch": 1.8844931119745674,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046450800287424455,
      "loss": 5.5841,
      "mean_token_accuracy": 0.18159414678812028,
      "num_tokens": 39788545.0,
      "step": 21340
    },
    {
      "entropy": 5.917054986953735,
      "epoch": 1.8849346520664076,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004644910247356532,
      "loss": 5.4948,
      "mean_token_accuracy": 0.19415213018655778,
      "num_tokens": 39798063.0,
      "step": 21345
    },
    {
      "entropy": 5.81759819984436,
      "epoch": 1.885376192158248,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004644740428850404,
      "loss": 5.5189,
      "mean_token_accuracy": 0.1905117452144623,
      "num_tokens": 39808622.0,
      "step": 21350
    },
    {
      "entropy": 5.870694208145141,
      "epoch": 1.8858177322500884,
      "grad_norm": 1.375,
      "learning_rate": 0.0004644570573227389,
      "loss": 5.6497,
      "mean_token_accuracy": 0.17670899629592896,
      "num_tokens": 39817608.0,
      "step": 21355
    },
    {
      "entropy": 5.843390560150146,
      "epoch": 1.8862592723419287,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046444006804908144,
      "loss": 5.4452,
      "mean_token_accuracy": 0.19900553226470946,
      "num_tokens": 39826055.0,
      "step": 21360
    },
    {
      "entropy": 5.728752279281617,
      "epoch": 1.886700812433769,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004644230750644007,
      "loss": 5.4774,
      "mean_token_accuracy": 0.195243763923645,
      "num_tokens": 39835490.0,
      "step": 21365
    },
    {
      "entropy": 5.818202209472656,
      "epoch": 1.8871423525256095,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046440607836902974,
      "loss": 5.5777,
      "mean_token_accuracy": 0.18794382661581038,
      "num_tokens": 39845613.0,
      "step": 21370
    },
    {
      "entropy": 5.895964193344116,
      "epoch": 1.8875838926174495,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046438907796330145,
      "loss": 5.5663,
      "mean_token_accuracy": 0.18729329258203506,
      "num_tokens": 39854495.0,
      "step": 21375
    },
    {
      "entropy": 5.878222560882568,
      "epoch": 1.88802543270929,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00046437207384754874,
      "loss": 5.604,
      "mean_token_accuracy": 0.18756561428308488,
      "num_tokens": 39864528.0,
      "step": 21380
    },
    {
      "entropy": 5.9008660316467285,
      "epoch": 1.8884669728011303,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004643550660221048,
      "loss": 5.6497,
      "mean_token_accuracy": 0.17520724833011628,
      "num_tokens": 39873780.0,
      "step": 21385
    },
    {
      "entropy": 5.802047824859619,
      "epoch": 1.8889085128929706,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046433805448730287,
      "loss": 5.4534,
      "mean_token_accuracy": 0.1967767596244812,
      "num_tokens": 39882529.0,
      "step": 21390
    },
    {
      "entropy": 5.932577180862427,
      "epoch": 1.889350052984811,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004643210392434761,
      "loss": 5.5858,
      "mean_token_accuracy": 0.18206258863210678,
      "num_tokens": 39890688.0,
      "step": 21395
    },
    {
      "entropy": 5.840638065338135,
      "epoch": 1.8897915930766513,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046430402029095784,
      "loss": 5.6337,
      "mean_token_accuracy": 0.17786161601543427,
      "num_tokens": 39900849.0,
      "step": 21400
    },
    {
      "entropy": 5.8083068370819095,
      "epoch": 1.8902331331684916,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004642869976300815,
      "loss": 5.586,
      "mean_token_accuracy": 0.18372942358255387,
      "num_tokens": 39911297.0,
      "step": 21405
    },
    {
      "entropy": 5.8446259021759035,
      "epoch": 1.8906746732603321,
      "grad_norm": 1.75,
      "learning_rate": 0.0004642699712611806,
      "loss": 5.5263,
      "mean_token_accuracy": 0.1903154581785202,
      "num_tokens": 39920312.0,
      "step": 21410
    },
    {
      "entropy": 5.858526134490967,
      "epoch": 1.8911162133521724,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004642529411845885,
      "loss": 5.6016,
      "mean_token_accuracy": 0.18231951594352722,
      "num_tokens": 39930128.0,
      "step": 21415
    },
    {
      "entropy": 5.926708507537842,
      "epoch": 1.8915577534440127,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004642359074006389,
      "loss": 5.5628,
      "mean_token_accuracy": 0.18981165289878846,
      "num_tokens": 39939177.0,
      "step": 21420
    },
    {
      "entropy": 5.983507871627808,
      "epoch": 1.8919992935358532,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004642188699096656,
      "loss": 5.659,
      "mean_token_accuracy": 0.18031961023807525,
      "num_tokens": 39948537.0,
      "step": 21425
    },
    {
      "entropy": 5.840392446517944,
      "epoch": 1.8924408336276934,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004642018287120022,
      "loss": 5.555,
      "mean_token_accuracy": 0.18544627875089645,
      "num_tokens": 39958623.0,
      "step": 21430
    },
    {
      "entropy": 5.904071855545044,
      "epoch": 1.8928823737195337,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00046418478380798263,
      "loss": 5.6658,
      "mean_token_accuracy": 0.18190988004207612,
      "num_tokens": 39968736.0,
      "step": 21435
    },
    {
      "entropy": 5.779025936126709,
      "epoch": 1.8933239138113742,
      "grad_norm": 1.25,
      "learning_rate": 0.0004641677351979407,
      "loss": 5.5063,
      "mean_token_accuracy": 0.18626141995191575,
      "num_tokens": 39977600.0,
      "step": 21440
    },
    {
      "entropy": 5.854047441482544,
      "epoch": 1.8937654539032143,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004641506828822104,
      "loss": 5.6101,
      "mean_token_accuracy": 0.1787608191370964,
      "num_tokens": 39987808.0,
      "step": 21445
    },
    {
      "entropy": 5.95890622138977,
      "epoch": 1.8942069939950548,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004641336268611258,
      "loss": 5.7461,
      "mean_token_accuracy": 0.18296887576580048,
      "num_tokens": 39998553.0,
      "step": 21450
    },
    {
      "entropy": 6.0352030277252195,
      "epoch": 1.894648534086895,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046411656713502106,
      "loss": 5.6456,
      "mean_token_accuracy": 0.18325041979551315,
      "num_tokens": 40007529.0,
      "step": 21455
    },
    {
      "entropy": 5.919585466384888,
      "epoch": 1.8950900741787353,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004640995037042303,
      "loss": 5.6343,
      "mean_token_accuracy": 0.18108050972223283,
      "num_tokens": 40017116.0,
      "step": 21460
    },
    {
      "entropy": 5.896755075454712,
      "epoch": 1.8955316142705758,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004640824365690878,
      "loss": 5.6453,
      "mean_token_accuracy": 0.18372589945793152,
      "num_tokens": 40026601.0,
      "step": 21465
    },
    {
      "entropy": 5.888518142700195,
      "epoch": 1.895973154362416,
      "grad_norm": 1.25,
      "learning_rate": 0.00046406536572992795,
      "loss": 5.544,
      "mean_token_accuracy": 0.19155555963516235,
      "num_tokens": 40036029.0,
      "step": 21470
    },
    {
      "entropy": 5.927377939224243,
      "epoch": 1.8964146944542564,
      "grad_norm": 1.8125,
      "learning_rate": 0.000464048291187085,
      "loss": 5.5933,
      "mean_token_accuracy": 0.178773233294487,
      "num_tokens": 40045892.0,
      "step": 21475
    },
    {
      "entropy": 5.927532815933228,
      "epoch": 1.8968562345460969,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004640312129408936,
      "loss": 5.5463,
      "mean_token_accuracy": 0.19168782979249954,
      "num_tokens": 40055411.0,
      "step": 21480
    },
    {
      "entropy": 5.914432001113892,
      "epoch": 1.8972977746379371,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004640141309916883,
      "loss": 5.6188,
      "mean_token_accuracy": 0.1852860540151596,
      "num_tokens": 40063988.0,
      "step": 21485
    },
    {
      "entropy": 5.851445484161377,
      "epoch": 1.8977393147297774,
      "grad_norm": 1.5,
      "learning_rate": 0.0004639970453398037,
      "loss": 5.57,
      "mean_token_accuracy": 0.18855149298906326,
      "num_tokens": 40072671.0,
      "step": 21490
    },
    {
      "entropy": 5.906409835815429,
      "epoch": 1.898180854821618,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046397995598557434,
      "loss": 5.6624,
      "mean_token_accuracy": 0.1834797203540802,
      "num_tokens": 40082191.0,
      "step": 21495
    },
    {
      "entropy": 5.889163637161255,
      "epoch": 1.8986223949134582,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004639628629293352,
      "loss": 5.6064,
      "mean_token_accuracy": 0.18027965277433394,
      "num_tokens": 40090823.0,
      "step": 21500
    },
    {
      "entropy": 5.922574758529663,
      "epoch": 1.8990639350052985,
      "grad_norm": 1.5625,
      "learning_rate": 0.000463945766171421,
      "loss": 5.5808,
      "mean_token_accuracy": 0.181588277220726,
      "num_tokens": 40099825.0,
      "step": 21505
    },
    {
      "entropy": 5.928248929977417,
      "epoch": 1.899505475097139,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004639286657121668,
      "loss": 5.6084,
      "mean_token_accuracy": 0.186471089720726,
      "num_tokens": 40109589.0,
      "step": 21510
    },
    {
      "entropy": 5.8467834949493405,
      "epoch": 1.899947015188979,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004639115615519074,
      "loss": 5.5523,
      "mean_token_accuracy": 0.18722670078277587,
      "num_tokens": 40119194.0,
      "step": 21515
    },
    {
      "entropy": 5.873023653030396,
      "epoch": 1.9003885552808195,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000463894453690978,
      "loss": 5.6127,
      "mean_token_accuracy": 0.1814122125506401,
      "num_tokens": 40128184.0,
      "step": 21520
    },
    {
      "entropy": 5.872552537918091,
      "epoch": 1.9008300953726598,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046387734212971363,
      "loss": 5.6361,
      "mean_token_accuracy": 0.17812780141830445,
      "num_tokens": 40137842.0,
      "step": 21525
    },
    {
      "entropy": 5.9212267875671385,
      "epoch": 1.9012716354645,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004638602268684496,
      "loss": 5.6015,
      "mean_token_accuracy": 0.18230361938476564,
      "num_tokens": 40147151.0,
      "step": 21530
    },
    {
      "entropy": 5.89022479057312,
      "epoch": 1.9017131755563406,
      "grad_norm": 1.875,
      "learning_rate": 0.00046384310790752105,
      "loss": 5.5543,
      "mean_token_accuracy": 0.18481654077768325,
      "num_tokens": 40157170.0,
      "step": 21535
    },
    {
      "entropy": 5.8558649063110355,
      "epoch": 1.9021547156481808,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004638259852472635,
      "loss": 5.5561,
      "mean_token_accuracy": 0.185704043507576,
      "num_tokens": 40166787.0,
      "step": 21540
    },
    {
      "entropy": 5.865142774581909,
      "epoch": 1.9025962557400211,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004638088588880122,
      "loss": 5.5377,
      "mean_token_accuracy": 0.18989952355623246,
      "num_tokens": 40176640.0,
      "step": 21545
    },
    {
      "entropy": 5.8601646900177,
      "epoch": 1.9030377958318616,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004637917288301029,
      "loss": 5.5924,
      "mean_token_accuracy": 0.18633182644844054,
      "num_tokens": 40185732.0,
      "step": 21550
    },
    {
      "entropy": 5.897035312652588,
      "epoch": 1.9034793359237019,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004637745950738709,
      "loss": 5.5083,
      "mean_token_accuracy": 0.1918432295322418,
      "num_tokens": 40194798.0,
      "step": 21555
    },
    {
      "entropy": 5.893478631973267,
      "epoch": 1.9039208760155422,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004637574576196519,
      "loss": 5.6883,
      "mean_token_accuracy": 0.1795911431312561,
      "num_tokens": 40204507.0,
      "step": 21560
    },
    {
      "entropy": 5.949166965484619,
      "epoch": 1.9043624161073827,
      "grad_norm": 1.625,
      "learning_rate": 0.0004637403164677817,
      "loss": 5.5985,
      "mean_token_accuracy": 0.17833209931850433,
      "num_tokens": 40213546.0,
      "step": 21565
    },
    {
      "entropy": 5.9098906993865965,
      "epoch": 1.904803956199223,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046372317161859605,
      "loss": 5.5335,
      "mean_token_accuracy": 0.1850186586380005,
      "num_tokens": 40222571.0,
      "step": 21570
    },
    {
      "entropy": 5.8939453125,
      "epoch": 1.9052454962910632,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046370602307243083,
      "loss": 5.6266,
      "mean_token_accuracy": 0.18180839866399764,
      "num_tokens": 40231803.0,
      "step": 21575
    },
    {
      "entropy": 5.874032211303711,
      "epoch": 1.9056870363829037,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046368887082962187,
      "loss": 5.6685,
      "mean_token_accuracy": 0.17801109552383423,
      "num_tokens": 40241755.0,
      "step": 21580
    },
    {
      "entropy": 5.87813982963562,
      "epoch": 1.9061285764747438,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004636717148905053,
      "loss": 5.5582,
      "mean_token_accuracy": 0.18466489911079406,
      "num_tokens": 40250963.0,
      "step": 21585
    },
    {
      "entropy": 5.929403400421142,
      "epoch": 1.9065701165665843,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004636545552554171,
      "loss": 5.5858,
      "mean_token_accuracy": 0.1837208777666092,
      "num_tokens": 40260626.0,
      "step": 21590
    },
    {
      "entropy": 5.861831378936768,
      "epoch": 1.9070116566584245,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004636373919246935,
      "loss": 5.5537,
      "mean_token_accuracy": 0.18895742148160935,
      "num_tokens": 40269852.0,
      "step": 21595
    },
    {
      "entropy": 5.857721948623658,
      "epoch": 1.9074531967502648,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004636202248986707,
      "loss": 5.5702,
      "mean_token_accuracy": 0.18645345121622087,
      "num_tokens": 40279499.0,
      "step": 21600
    },
    {
      "entropy": 5.974339246749878,
      "epoch": 1.9078947368421053,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046360305417768484,
      "loss": 5.6596,
      "mean_token_accuracy": 0.1790601894259453,
      "num_tokens": 40288131.0,
      "step": 21605
    },
    {
      "entropy": 5.910491514205932,
      "epoch": 1.9083362769339456,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004635858797620725,
      "loss": 5.614,
      "mean_token_accuracy": 0.1825230285525322,
      "num_tokens": 40296951.0,
      "step": 21610
    },
    {
      "entropy": 5.849309253692627,
      "epoch": 1.9087778170257859,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00046356870165217004,
      "loss": 5.5433,
      "mean_token_accuracy": 0.1886252522468567,
      "num_tokens": 40306250.0,
      "step": 21615
    },
    {
      "entropy": 5.764451265335083,
      "epoch": 1.9092193571176264,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046355151984831395,
      "loss": 5.4914,
      "mean_token_accuracy": 0.19467723667621611,
      "num_tokens": 40314561.0,
      "step": 21620
    },
    {
      "entropy": 5.7682184219360355,
      "epoch": 1.9096608972094666,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046353433435084083,
      "loss": 5.4605,
      "mean_token_accuracy": 0.18976385444402694,
      "num_tokens": 40323580.0,
      "step": 21625
    },
    {
      "entropy": 5.837805652618409,
      "epoch": 1.910102437301307,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004635171451600872,
      "loss": 5.5692,
      "mean_token_accuracy": 0.18621437549591063,
      "num_tokens": 40333747.0,
      "step": 21630
    },
    {
      "entropy": 5.950639629364014,
      "epoch": 1.9105439773931474,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046349995227639015,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17875823825597764,
      "num_tokens": 40343435.0,
      "step": 21635
    },
    {
      "entropy": 5.954769563674927,
      "epoch": 1.9109855174849877,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004634827557000861,
      "loss": 5.6596,
      "mean_token_accuracy": 0.18330047875642777,
      "num_tokens": 40353282.0,
      "step": 21640
    },
    {
      "entropy": 5.890699052810669,
      "epoch": 1.911427057576828,
      "grad_norm": 1.421875,
      "learning_rate": 0.000463465555431512,
      "loss": 5.6445,
      "mean_token_accuracy": 0.18176864832639694,
      "num_tokens": 40363781.0,
      "step": 21645
    },
    {
      "entropy": 5.8521575927734375,
      "epoch": 1.9118685976686685,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046344835147100495,
      "loss": 5.5724,
      "mean_token_accuracy": 0.18312713056802749,
      "num_tokens": 40373240.0,
      "step": 21650
    },
    {
      "entropy": 5.919853782653808,
      "epoch": 1.9123101377605085,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046343114381890187,
      "loss": 5.6511,
      "mean_token_accuracy": 0.1841177150607109,
      "num_tokens": 40383088.0,
      "step": 21655
    },
    {
      "entropy": 5.946359062194825,
      "epoch": 1.912751677852349,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004634139324755398,
      "loss": 5.5948,
      "mean_token_accuracy": 0.19097699373960494,
      "num_tokens": 40391654.0,
      "step": 21660
    },
    {
      "entropy": 5.864236545562744,
      "epoch": 1.9131932179441893,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000463396717441256,
      "loss": 5.498,
      "mean_token_accuracy": 0.19835550338029861,
      "num_tokens": 40400616.0,
      "step": 21665
    },
    {
      "entropy": 5.774488496780395,
      "epoch": 1.9136347580360296,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004633794987163876,
      "loss": 5.579,
      "mean_token_accuracy": 0.18006569296121597,
      "num_tokens": 40411271.0,
      "step": 21670
    },
    {
      "entropy": 5.8680394172668455,
      "epoch": 1.91407629812787,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046336227630127205,
      "loss": 5.4812,
      "mean_token_accuracy": 0.1815154880285263,
      "num_tokens": 40420019.0,
      "step": 21675
    },
    {
      "entropy": 5.930383777618408,
      "epoch": 1.9145178382197103,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046334505019624656,
      "loss": 5.5901,
      "mean_token_accuracy": 0.18592734783887863,
      "num_tokens": 40429858.0,
      "step": 21680
    },
    {
      "entropy": 5.830162286758423,
      "epoch": 1.9149593783115506,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046332782040164867,
      "loss": 5.5532,
      "mean_token_accuracy": 0.19306197315454482,
      "num_tokens": 40439566.0,
      "step": 21685
    },
    {
      "entropy": 5.80991005897522,
      "epoch": 1.915400918403391,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004633105869178159,
      "loss": 5.5058,
      "mean_token_accuracy": 0.19334145337343217,
      "num_tokens": 40448527.0,
      "step": 21690
    },
    {
      "entropy": 5.854040050506592,
      "epoch": 1.9158424584952314,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004632933497450858,
      "loss": 5.5235,
      "mean_token_accuracy": 0.18921538740396499,
      "num_tokens": 40458076.0,
      "step": 21695
    },
    {
      "entropy": 5.830065155029297,
      "epoch": 1.9162839985870717,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004632761088837961,
      "loss": 5.4466,
      "mean_token_accuracy": 0.20447590351104736,
      "num_tokens": 40467639.0,
      "step": 21700
    },
    {
      "entropy": 5.885995483398437,
      "epoch": 1.9167255386789122,
      "grad_norm": 1.375,
      "learning_rate": 0.00046325886433428446,
      "loss": 5.5402,
      "mean_token_accuracy": 0.19170498102903366,
      "num_tokens": 40476243.0,
      "step": 21705
    },
    {
      "entropy": 5.85709285736084,
      "epoch": 1.9171670787707524,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046324161609688876,
      "loss": 5.5793,
      "mean_token_accuracy": 0.19165739566087722,
      "num_tokens": 40485789.0,
      "step": 21710
    },
    {
      "entropy": 5.811823225021362,
      "epoch": 1.9176086188625927,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004632243641719469,
      "loss": 5.5157,
      "mean_token_accuracy": 0.18816226124763488,
      "num_tokens": 40494139.0,
      "step": 21715
    },
    {
      "entropy": 5.819171333312989,
      "epoch": 1.9180501589544332,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046320710855979677,
      "loss": 5.5323,
      "mean_token_accuracy": 0.1887034222483635,
      "num_tokens": 40503965.0,
      "step": 21720
    },
    {
      "entropy": 5.884352111816407,
      "epoch": 1.9184916990462733,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046318984926077637,
      "loss": 5.5498,
      "mean_token_accuracy": 0.19060949683189393,
      "num_tokens": 40513140.0,
      "step": 21725
    },
    {
      "entropy": 5.832448196411133,
      "epoch": 1.9189332391381138,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00046317258627522393,
      "loss": 5.5531,
      "mean_token_accuracy": 0.1951460599899292,
      "num_tokens": 40522858.0,
      "step": 21730
    },
    {
      "entropy": 5.869863557815552,
      "epoch": 1.919374779229954,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004631553196034774,
      "loss": 5.5391,
      "mean_token_accuracy": 0.18948689550161363,
      "num_tokens": 40532050.0,
      "step": 21735
    },
    {
      "entropy": 5.89995493888855,
      "epoch": 1.9198163193217943,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004631380492458753,
      "loss": 5.5326,
      "mean_token_accuracy": 0.18957667350769042,
      "num_tokens": 40540942.0,
      "step": 21740
    },
    {
      "entropy": 5.853212881088257,
      "epoch": 1.9202578594136348,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004631207752027558,
      "loss": 5.5669,
      "mean_token_accuracy": 0.1890183061361313,
      "num_tokens": 40550158.0,
      "step": 21745
    },
    {
      "entropy": 5.8892449855804445,
      "epoch": 1.920699399505475,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004631034974744573,
      "loss": 5.6818,
      "mean_token_accuracy": 0.17951171696186066,
      "num_tokens": 40559267.0,
      "step": 21750
    },
    {
      "entropy": 5.911639595031739,
      "epoch": 1.9211409395973154,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004630862160613183,
      "loss": 5.5682,
      "mean_token_accuracy": 0.17996702939271927,
      "num_tokens": 40568503.0,
      "step": 21755
    },
    {
      "entropy": 5.906809663772583,
      "epoch": 1.9215824796891559,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004630689309636772,
      "loss": 5.5284,
      "mean_token_accuracy": 0.184407639503479,
      "num_tokens": 40578464.0,
      "step": 21760
    },
    {
      "entropy": 5.93601541519165,
      "epoch": 1.9220240197809961,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004630516421818728,
      "loss": 5.6883,
      "mean_token_accuracy": 0.1760425478219986,
      "num_tokens": 40588338.0,
      "step": 21765
    },
    {
      "entropy": 5.848486137390137,
      "epoch": 1.9224655598728364,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046303434971624357,
      "loss": 5.556,
      "mean_token_accuracy": 0.18509652763605117,
      "num_tokens": 40598912.0,
      "step": 21770
    },
    {
      "entropy": 5.784008502960205,
      "epoch": 1.922907099964677,
      "grad_norm": 1.375,
      "learning_rate": 0.00046301705356712846,
      "loss": 5.4532,
      "mean_token_accuracy": 0.1894407942891121,
      "num_tokens": 40607084.0,
      "step": 21775
    },
    {
      "entropy": 5.8884907245635985,
      "epoch": 1.9233486400565172,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046299975373486626,
      "loss": 5.5805,
      "mean_token_accuracy": 0.18067300617694854,
      "num_tokens": 40616422.0,
      "step": 21780
    },
    {
      "entropy": 5.834081840515137,
      "epoch": 1.9237901801483575,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004629824502197956,
      "loss": 5.5081,
      "mean_token_accuracy": 0.18654191493988037,
      "num_tokens": 40626189.0,
      "step": 21785
    },
    {
      "entropy": 5.801152610778809,
      "epoch": 1.924231720240198,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00046296514302225585,
      "loss": 5.4884,
      "mean_token_accuracy": 0.19019856601953505,
      "num_tokens": 40636053.0,
      "step": 21790
    },
    {
      "entropy": 5.869901847839356,
      "epoch": 1.924673260332038,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004629478321425858,
      "loss": 5.5986,
      "mean_token_accuracy": 0.17946066558361054,
      "num_tokens": 40645159.0,
      "step": 21795
    },
    {
      "entropy": 5.824137401580811,
      "epoch": 1.9251148004238785,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046293051758112457,
      "loss": 5.4565,
      "mean_token_accuracy": 0.19672543853521346,
      "num_tokens": 40653890.0,
      "step": 21800
    },
    {
      "entropy": 5.900637149810791,
      "epoch": 1.9255563405157188,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046291319933821143,
      "loss": 5.6279,
      "mean_token_accuracy": 0.17608252614736558,
      "num_tokens": 40663041.0,
      "step": 21805
    },
    {
      "entropy": 5.842786169052124,
      "epoch": 1.925997880607559,
      "grad_norm": 1.546875,
      "learning_rate": 0.00046289587741418553,
      "loss": 5.4938,
      "mean_token_accuracy": 0.19391832649707794,
      "num_tokens": 40670972.0,
      "step": 21810
    },
    {
      "entropy": 5.842030572891235,
      "epoch": 1.9264394206993996,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046287855180938633,
      "loss": 5.6169,
      "mean_token_accuracy": 0.19054969698190688,
      "num_tokens": 40680503.0,
      "step": 21815
    },
    {
      "entropy": 5.876213884353637,
      "epoch": 1.9268809607912398,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046286122252415303,
      "loss": 5.6088,
      "mean_token_accuracy": 0.1837650254368782,
      "num_tokens": 40691339.0,
      "step": 21820
    },
    {
      "entropy": 5.932537794113159,
      "epoch": 1.92732250088308,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046284388955882526,
      "loss": 5.5745,
      "mean_token_accuracy": 0.18529122471809387,
      "num_tokens": 40699744.0,
      "step": 21825
    },
    {
      "entropy": 5.91429648399353,
      "epoch": 1.9277640409749206,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046282655291374257,
      "loss": 5.6121,
      "mean_token_accuracy": 0.18367340564727783,
      "num_tokens": 40709144.0,
      "step": 21830
    },
    {
      "entropy": 5.803210353851318,
      "epoch": 1.9282055810667609,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00046280921258924455,
      "loss": 5.6361,
      "mean_token_accuracy": 0.18360882103443146,
      "num_tokens": 40719513.0,
      "step": 21835
    },
    {
      "entropy": 5.877527475357056,
      "epoch": 1.9286471211586012,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046279186858567083,
      "loss": 5.5764,
      "mean_token_accuracy": 0.18332892060279846,
      "num_tokens": 40728846.0,
      "step": 21840
    },
    {
      "entropy": 5.907170104980469,
      "epoch": 1.9290886612504416,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004627745209033613,
      "loss": 5.4907,
      "mean_token_accuracy": 0.19370485991239547,
      "num_tokens": 40737904.0,
      "step": 21845
    },
    {
      "entropy": 5.775260877609253,
      "epoch": 1.929530201342282,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004627571695426556,
      "loss": 5.5043,
      "mean_token_accuracy": 0.1977008745074272,
      "num_tokens": 40747328.0,
      "step": 21850
    },
    {
      "entropy": 5.8171745300292965,
      "epoch": 1.9299717414341222,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046273981450389374,
      "loss": 5.5392,
      "mean_token_accuracy": 0.18335571140050888,
      "num_tokens": 40755567.0,
      "step": 21855
    },
    {
      "entropy": 5.873519515991211,
      "epoch": 1.9304132815259627,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046272245578741567,
      "loss": 5.578,
      "mean_token_accuracy": 0.1852467879652977,
      "num_tokens": 40764684.0,
      "step": 21860
    },
    {
      "entropy": 5.907274293899536,
      "epoch": 1.9308548216178028,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004627050933935615,
      "loss": 5.5731,
      "mean_token_accuracy": 0.1929454281926155,
      "num_tokens": 40774050.0,
      "step": 21865
    },
    {
      "entropy": 5.80935697555542,
      "epoch": 1.9312963617096433,
      "grad_norm": 1.859375,
      "learning_rate": 0.00046268772732267126,
      "loss": 5.5251,
      "mean_token_accuracy": 0.18524304479360582,
      "num_tokens": 40782722.0,
      "step": 21870
    },
    {
      "entropy": 5.847461080551147,
      "epoch": 1.9317379018014835,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004626703575750852,
      "loss": 5.5969,
      "mean_token_accuracy": 0.18245240449905395,
      "num_tokens": 40791978.0,
      "step": 21875
    },
    {
      "entropy": 5.910322427749634,
      "epoch": 1.9321794418933238,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004626529841511436,
      "loss": 5.603,
      "mean_token_accuracy": 0.18323240727186202,
      "num_tokens": 40801855.0,
      "step": 21880
    },
    {
      "entropy": 5.890281009674072,
      "epoch": 1.9326209819851643,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046263560705118675,
      "loss": 5.6642,
      "mean_token_accuracy": 0.18072142004966735,
      "num_tokens": 40812025.0,
      "step": 21885
    },
    {
      "entropy": 5.866515302658081,
      "epoch": 1.9330625220770046,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004626182262755551,
      "loss": 5.4804,
      "mean_token_accuracy": 0.1882502481341362,
      "num_tokens": 40820758.0,
      "step": 21890
    },
    {
      "entropy": 5.886205339431763,
      "epoch": 1.9335040621688449,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046260084182458906,
      "loss": 5.6071,
      "mean_token_accuracy": 0.1817648857831955,
      "num_tokens": 40830653.0,
      "step": 21895
    },
    {
      "entropy": 5.87415337562561,
      "epoch": 1.9339456022606853,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00046258345369862915,
      "loss": 5.6857,
      "mean_token_accuracy": 0.17563393712043762,
      "num_tokens": 40840370.0,
      "step": 21900
    },
    {
      "entropy": 5.8749549865722654,
      "epoch": 1.9343871423525256,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004625660618980162,
      "loss": 5.5745,
      "mean_token_accuracy": 0.18597546368837356,
      "num_tokens": 40849590.0,
      "step": 21905
    },
    {
      "entropy": 5.916858720779419,
      "epoch": 1.934828682444366,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046254866642309067,
      "loss": 5.5723,
      "mean_token_accuracy": 0.18577001094818116,
      "num_tokens": 40859271.0,
      "step": 21910
    },
    {
      "entropy": 5.848455381393433,
      "epoch": 1.9352702225362064,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004625312672741935,
      "loss": 5.5833,
      "mean_token_accuracy": 0.18564672023057938,
      "num_tokens": 40869272.0,
      "step": 21915
    },
    {
      "entropy": 5.95191593170166,
      "epoch": 1.9357117626280467,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004625138644516655,
      "loss": 5.6518,
      "mean_token_accuracy": 0.1782410427927971,
      "num_tokens": 40880327.0,
      "step": 21920
    },
    {
      "entropy": 5.8515904426574705,
      "epoch": 1.936153302719887,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004624964579558474,
      "loss": 5.577,
      "mean_token_accuracy": 0.1865433484315872,
      "num_tokens": 40890407.0,
      "step": 21925
    },
    {
      "entropy": 5.9165160179138185,
      "epoch": 1.9365948428117274,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004624790477870804,
      "loss": 5.5479,
      "mean_token_accuracy": 0.18294069916009903,
      "num_tokens": 40899306.0,
      "step": 21930
    },
    {
      "entropy": 5.883017587661743,
      "epoch": 1.9370363829035675,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004624616339457055,
      "loss": 5.6288,
      "mean_token_accuracy": 0.18869634568691254,
      "num_tokens": 40908908.0,
      "step": 21935
    },
    {
      "entropy": 5.908078289031982,
      "epoch": 1.937477922995408,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004624442164320638,
      "loss": 5.5826,
      "mean_token_accuracy": 0.18585500866174698,
      "num_tokens": 40917960.0,
      "step": 21940
    },
    {
      "entropy": 5.78179702758789,
      "epoch": 1.9379194630872483,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004624267952464965,
      "loss": 5.3814,
      "mean_token_accuracy": 0.19974540174007416,
      "num_tokens": 40926364.0,
      "step": 21945
    },
    {
      "entropy": 5.806497049331665,
      "epoch": 1.9383610031790885,
      "grad_norm": 1.5,
      "learning_rate": 0.00046240937038934484,
      "loss": 5.5837,
      "mean_token_accuracy": 0.18583552688360214,
      "num_tokens": 40936558.0,
      "step": 21950
    },
    {
      "entropy": 5.848411798477173,
      "epoch": 1.938802543270929,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004623919418609502,
      "loss": 5.5654,
      "mean_token_accuracy": 0.1896674305200577,
      "num_tokens": 40945501.0,
      "step": 21955
    },
    {
      "entropy": 5.81519079208374,
      "epoch": 1.9392440833627693,
      "grad_norm": 1.71875,
      "learning_rate": 0.00046237450966165407,
      "loss": 5.5463,
      "mean_token_accuracy": 0.189144004881382,
      "num_tokens": 40954740.0,
      "step": 21960
    },
    {
      "entropy": 5.913944673538208,
      "epoch": 1.9396856234546096,
      "grad_norm": 1.5,
      "learning_rate": 0.00046235707379179775,
      "loss": 5.5929,
      "mean_token_accuracy": 0.18746706694364548,
      "num_tokens": 40963748.0,
      "step": 21965
    },
    {
      "entropy": 5.919274425506591,
      "epoch": 1.94012716354645,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046233963425172306,
      "loss": 5.5845,
      "mean_token_accuracy": 0.18539750725030898,
      "num_tokens": 40973907.0,
      "step": 21970
    },
    {
      "entropy": 5.866813278198242,
      "epoch": 1.9405687036382904,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046232219104177136,
      "loss": 5.5919,
      "mean_token_accuracy": 0.18351520597934723,
      "num_tokens": 40983374.0,
      "step": 21975
    },
    {
      "entropy": 5.924910259246826,
      "epoch": 1.9410102437301306,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004623047441622845,
      "loss": 5.6143,
      "mean_token_accuracy": 0.17464358657598494,
      "num_tokens": 40993305.0,
      "step": 21980
    },
    {
      "entropy": 5.9403526306152346,
      "epoch": 1.9414517838219711,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046228729361360427,
      "loss": 5.5596,
      "mean_token_accuracy": 0.17691201865673065,
      "num_tokens": 41002438.0,
      "step": 21985
    },
    {
      "entropy": 5.756121492385864,
      "epoch": 1.9418933239138114,
      "grad_norm": 1.53125,
      "learning_rate": 0.00046226983939607236,
      "loss": 5.5138,
      "mean_token_accuracy": 0.1898392289876938,
      "num_tokens": 41011491.0,
      "step": 21990
    },
    {
      "entropy": 5.843390989303589,
      "epoch": 1.9423348640056517,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004622523815100309,
      "loss": 5.6057,
      "mean_token_accuracy": 0.181699438393116,
      "num_tokens": 41021091.0,
      "step": 21995
    },
    {
      "entropy": 5.869727945327758,
      "epoch": 1.9427764040974922,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046223491995582187,
      "loss": 5.5471,
      "mean_token_accuracy": 0.1857534170150757,
      "num_tokens": 41030719.0,
      "step": 22000
    },
    {
      "entropy": 5.875077390670777,
      "epoch": 1.9432179441893322,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004622174547337871,
      "loss": 5.6138,
      "mean_token_accuracy": 0.17680706679821015,
      "num_tokens": 41039651.0,
      "step": 22005
    },
    {
      "entropy": 5.9215034484863285,
      "epoch": 1.9436594842811727,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046219998584426885,
      "loss": 5.5142,
      "mean_token_accuracy": 0.18798035830259324,
      "num_tokens": 41047914.0,
      "step": 22010
    },
    {
      "entropy": 5.828591775894165,
      "epoch": 1.944101024373013,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004621825132876094,
      "loss": 5.6301,
      "mean_token_accuracy": 0.18659648150205613,
      "num_tokens": 41056779.0,
      "step": 22015
    },
    {
      "entropy": 5.827025890350342,
      "epoch": 1.9445425644648533,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046216503706415096,
      "loss": 5.564,
      "mean_token_accuracy": 0.18955283612012863,
      "num_tokens": 41066356.0,
      "step": 22020
    },
    {
      "entropy": 5.912852954864502,
      "epoch": 1.9449841045566938,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004621475571742359,
      "loss": 5.593,
      "mean_token_accuracy": 0.1923756405711174,
      "num_tokens": 41075766.0,
      "step": 22025
    },
    {
      "entropy": 5.929812860488892,
      "epoch": 1.945425644648534,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046213007361820663,
      "loss": 5.5767,
      "mean_token_accuracy": 0.18535844087600709,
      "num_tokens": 41085503.0,
      "step": 22030
    },
    {
      "entropy": 5.966259288787842,
      "epoch": 1.9458671847403743,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004621125863964057,
      "loss": 5.7294,
      "mean_token_accuracy": 0.1671004742383957,
      "num_tokens": 41096581.0,
      "step": 22035
    },
    {
      "entropy": 5.919198703765869,
      "epoch": 1.9463087248322148,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046209509550917553,
      "loss": 5.5431,
      "mean_token_accuracy": 0.19107319712638854,
      "num_tokens": 41105226.0,
      "step": 22040
    },
    {
      "entropy": 5.943841934204102,
      "epoch": 1.9467502649240551,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004620776009568589,
      "loss": 5.5402,
      "mean_token_accuracy": 0.18965779691934587,
      "num_tokens": 41113998.0,
      "step": 22045
    },
    {
      "entropy": 5.806170415878296,
      "epoch": 1.9471918050158954,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004620601027397984,
      "loss": 5.575,
      "mean_token_accuracy": 0.1788436159491539,
      "num_tokens": 41123147.0,
      "step": 22050
    },
    {
      "entropy": 5.89486083984375,
      "epoch": 1.947633345107736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046204260085833695,
      "loss": 5.5475,
      "mean_token_accuracy": 0.19059031456708908,
      "num_tokens": 41132317.0,
      "step": 22055
    },
    {
      "entropy": 5.92190580368042,
      "epoch": 1.9480748851995762,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004620250953128173,
      "loss": 5.6124,
      "mean_token_accuracy": 0.1858062833547592,
      "num_tokens": 41140795.0,
      "step": 22060
    },
    {
      "entropy": 5.859563159942627,
      "epoch": 1.9485164252914164,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004620075861035824,
      "loss": 5.5734,
      "mean_token_accuracy": 0.18185152262449264,
      "num_tokens": 41149959.0,
      "step": 22065
    },
    {
      "entropy": 5.892498397827149,
      "epoch": 1.948957965383257,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004619900732309753,
      "loss": 5.6352,
      "mean_token_accuracy": 0.1752442479133606,
      "num_tokens": 41159388.0,
      "step": 22070
    },
    {
      "entropy": 6.004470252990723,
      "epoch": 1.949399505475097,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046197255669533886,
      "loss": 5.807,
      "mean_token_accuracy": 0.17370293140411378,
      "num_tokens": 41168875.0,
      "step": 22075
    },
    {
      "entropy": 5.979822969436645,
      "epoch": 1.9498410455669375,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046195503649701646,
      "loss": 5.6101,
      "mean_token_accuracy": 0.18280381709337234,
      "num_tokens": 41177621.0,
      "step": 22080
    },
    {
      "entropy": 5.934388542175293,
      "epoch": 1.9502825856587778,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004619375126363513,
      "loss": 5.631,
      "mean_token_accuracy": 0.1877792403101921,
      "num_tokens": 41185911.0,
      "step": 22085
    },
    {
      "entropy": 5.856566619873047,
      "epoch": 1.950724125750618,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046191998511368646,
      "loss": 5.5415,
      "mean_token_accuracy": 0.1902650162577629,
      "num_tokens": 41194811.0,
      "step": 22090
    },
    {
      "entropy": 5.7244771957397464,
      "epoch": 1.9511656658424585,
      "grad_norm": 1.25,
      "learning_rate": 0.00046190245392936554,
      "loss": 5.4742,
      "mean_token_accuracy": 0.19547055959701537,
      "num_tokens": 41203955.0,
      "step": 22095
    },
    {
      "entropy": 5.84662880897522,
      "epoch": 1.9516072059342988,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004618849190837317,
      "loss": 5.5017,
      "mean_token_accuracy": 0.19271224737167358,
      "num_tokens": 41212635.0,
      "step": 22100
    },
    {
      "entropy": 5.886985921859742,
      "epoch": 1.952048746026139,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004618673805771287,
      "loss": 5.5674,
      "mean_token_accuracy": 0.18330346345901488,
      "num_tokens": 41221443.0,
      "step": 22105
    },
    {
      "entropy": 5.857612419128418,
      "epoch": 1.9524902861179796,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004618498384098999,
      "loss": 5.5499,
      "mean_token_accuracy": 0.18194970339536667,
      "num_tokens": 41230089.0,
      "step": 22110
    },
    {
      "entropy": 5.889239549636841,
      "epoch": 1.9529318262098199,
      "grad_norm": 1.71875,
      "learning_rate": 0.00046183229258238905,
      "loss": 5.6657,
      "mean_token_accuracy": 0.1788114294409752,
      "num_tokens": 41239569.0,
      "step": 22115
    },
    {
      "entropy": 5.966203927993774,
      "epoch": 1.9533733663016601,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004618147430949399,
      "loss": 5.5652,
      "mean_token_accuracy": 0.18671686798334122,
      "num_tokens": 41248394.0,
      "step": 22120
    },
    {
      "entropy": 5.865884590148926,
      "epoch": 1.9538149063935006,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046179718994789624,
      "loss": 5.569,
      "mean_token_accuracy": 0.18799942433834077,
      "num_tokens": 41258148.0,
      "step": 22125
    },
    {
      "entropy": 5.797039556503296,
      "epoch": 1.954256446485341,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046177963314160175,
      "loss": 5.4966,
      "mean_token_accuracy": 0.19413871169090272,
      "num_tokens": 41266984.0,
      "step": 22130
    },
    {
      "entropy": 5.869829940795898,
      "epoch": 1.9546979865771812,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046176207267640054,
      "loss": 5.6411,
      "mean_token_accuracy": 0.17991742491722107,
      "num_tokens": 41276119.0,
      "step": 22135
    },
    {
      "entropy": 5.814297533035278,
      "epoch": 1.9551395266690217,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004617445085526365,
      "loss": 5.5016,
      "mean_token_accuracy": 0.19034779369831084,
      "num_tokens": 41284945.0,
      "step": 22140
    },
    {
      "entropy": 5.794466495513916,
      "epoch": 1.9555810667608617,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004617269407706537,
      "loss": 5.4896,
      "mean_token_accuracy": 0.19688279330730438,
      "num_tokens": 41294255.0,
      "step": 22145
    },
    {
      "entropy": 5.830201053619385,
      "epoch": 1.9560226068527022,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004617093693307964,
      "loss": 5.6041,
      "mean_token_accuracy": 0.18337610214948655,
      "num_tokens": 41303056.0,
      "step": 22150
    },
    {
      "entropy": 5.908079957962036,
      "epoch": 1.9564641469445425,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046169179423340866,
      "loss": 5.6243,
      "mean_token_accuracy": 0.17839315086603164,
      "num_tokens": 41312444.0,
      "step": 22155
    },
    {
      "entropy": 6.003380966186524,
      "epoch": 1.9569056870363828,
      "grad_norm": 2.015625,
      "learning_rate": 0.000461674215478835,
      "loss": 5.716,
      "mean_token_accuracy": 0.17448977679014205,
      "num_tokens": 41322507.0,
      "step": 22160
    },
    {
      "entropy": 5.908987092971802,
      "epoch": 1.9573472271282233,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004616566330674195,
      "loss": 5.5679,
      "mean_token_accuracy": 0.1868058666586876,
      "num_tokens": 41332507.0,
      "step": 22165
    },
    {
      "entropy": 5.7980804443359375,
      "epoch": 1.9577887672200636,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004616390469995068,
      "loss": 5.5491,
      "mean_token_accuracy": 0.1838330775499344,
      "num_tokens": 41342085.0,
      "step": 22170
    },
    {
      "entropy": 5.837388896942139,
      "epoch": 1.9582303073119038,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046162145727544126,
      "loss": 5.5262,
      "mean_token_accuracy": 0.19608274847269058,
      "num_tokens": 41351422.0,
      "step": 22175
    },
    {
      "entropy": 5.830231857299805,
      "epoch": 1.9586718474037443,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004616038638955675,
      "loss": 5.5491,
      "mean_token_accuracy": 0.19123305827379228,
      "num_tokens": 41361337.0,
      "step": 22180
    },
    {
      "entropy": 5.804967546463013,
      "epoch": 1.9591133874955846,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046158626686023014,
      "loss": 5.4989,
      "mean_token_accuracy": 0.19177144318819045,
      "num_tokens": 41370734.0,
      "step": 22185
    },
    {
      "entropy": 5.859177875518799,
      "epoch": 1.9595549275874249,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046156866616977406,
      "loss": 5.5196,
      "mean_token_accuracy": 0.18846632838249205,
      "num_tokens": 41380090.0,
      "step": 22190
    },
    {
      "entropy": 5.904728507995605,
      "epoch": 1.9599964676792654,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046155106182454384,
      "loss": 5.5786,
      "mean_token_accuracy": 0.19160428792238235,
      "num_tokens": 41388967.0,
      "step": 22195
    },
    {
      "entropy": 5.875503635406494,
      "epoch": 1.9604380077711057,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004615334538248844,
      "loss": 5.5547,
      "mean_token_accuracy": 0.18320715427398682,
      "num_tokens": 41397429.0,
      "step": 22200
    },
    {
      "entropy": 5.855077123641967,
      "epoch": 1.960879547862946,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046151584217114073,
      "loss": 5.514,
      "mean_token_accuracy": 0.18516878932714462,
      "num_tokens": 41406420.0,
      "step": 22205
    },
    {
      "entropy": 5.76237301826477,
      "epoch": 1.9613210879547864,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004614982268636577,
      "loss": 5.4543,
      "mean_token_accuracy": 0.20035306364297867,
      "num_tokens": 41415994.0,
      "step": 22210
    },
    {
      "entropy": 5.809692859649658,
      "epoch": 1.9617626280466265,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004614806079027806,
      "loss": 5.4608,
      "mean_token_accuracy": 0.1982736185193062,
      "num_tokens": 41425744.0,
      "step": 22215
    },
    {
      "entropy": 5.882755994796753,
      "epoch": 1.962204168138467,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046146298528885435,
      "loss": 5.6506,
      "mean_token_accuracy": 0.18129936456680298,
      "num_tokens": 41435452.0,
      "step": 22220
    },
    {
      "entropy": 5.924654960632324,
      "epoch": 1.9626457082303073,
      "grad_norm": 2.625,
      "learning_rate": 0.0004614453590222243,
      "loss": 5.6075,
      "mean_token_accuracy": 0.1865273892879486,
      "num_tokens": 41444853.0,
      "step": 22225
    },
    {
      "entropy": 5.789183759689331,
      "epoch": 1.9630872483221475,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004614277291032357,
      "loss": 5.4731,
      "mean_token_accuracy": 0.2088585302233696,
      "num_tokens": 41454569.0,
      "step": 22230
    },
    {
      "entropy": 5.920354318618775,
      "epoch": 1.963528788413988,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004614100955322339,
      "loss": 5.6384,
      "mean_token_accuracy": 0.18249636888504028,
      "num_tokens": 41464159.0,
      "step": 22235
    },
    {
      "entropy": 5.909075498580933,
      "epoch": 1.9639703285058283,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004613924583095644,
      "loss": 5.663,
      "mean_token_accuracy": 0.17826283872127532,
      "num_tokens": 41474354.0,
      "step": 22240
    },
    {
      "entropy": 5.880492687225342,
      "epoch": 1.9644118685976686,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004613748174355726,
      "loss": 5.5272,
      "mean_token_accuracy": 0.18999718129634857,
      "num_tokens": 41484053.0,
      "step": 22245
    },
    {
      "entropy": 5.887611198425293,
      "epoch": 1.964853408689509,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00046135717291060415,
      "loss": 5.5658,
      "mean_token_accuracy": 0.18109649568796157,
      "num_tokens": 41493377.0,
      "step": 22250
    },
    {
      "entropy": 5.764782047271728,
      "epoch": 1.9652949487813494,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046133952473500464,
      "loss": 5.5185,
      "mean_token_accuracy": 0.19298349022865297,
      "num_tokens": 41502361.0,
      "step": 22255
    },
    {
      "entropy": 5.853038120269775,
      "epoch": 1.9657364888731896,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004613218729091198,
      "loss": 5.5619,
      "mean_token_accuracy": 0.18252271711826323,
      "num_tokens": 41511083.0,
      "step": 22260
    },
    {
      "entropy": 5.955618047714234,
      "epoch": 1.9661780289650301,
      "grad_norm": 1.25,
      "learning_rate": 0.0004613042174332955,
      "loss": 5.6536,
      "mean_token_accuracy": 0.1824835404753685,
      "num_tokens": 41520264.0,
      "step": 22265
    },
    {
      "entropy": 5.876659870147705,
      "epoch": 1.9666195690568704,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004612865583078775,
      "loss": 5.506,
      "mean_token_accuracy": 0.19064094722270966,
      "num_tokens": 41529805.0,
      "step": 22270
    },
    {
      "entropy": 5.765671920776367,
      "epoch": 1.9670611091487107,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004612688955332118,
      "loss": 5.4388,
      "mean_token_accuracy": 0.1888108178973198,
      "num_tokens": 41538348.0,
      "step": 22275
    },
    {
      "entropy": 5.856932973861694,
      "epoch": 1.9675026492405512,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046125122910964427,
      "loss": 5.5649,
      "mean_token_accuracy": 0.18541598170995713,
      "num_tokens": 41547433.0,
      "step": 22280
    },
    {
      "entropy": 5.875472211837769,
      "epoch": 1.9679441893323912,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004612335590375212,
      "loss": 5.5687,
      "mean_token_accuracy": 0.1829586699604988,
      "num_tokens": 41556162.0,
      "step": 22285
    },
    {
      "entropy": 5.907475471496582,
      "epoch": 1.9683857294242317,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004612158853171886,
      "loss": 5.5366,
      "mean_token_accuracy": 0.18603765815496445,
      "num_tokens": 41565415.0,
      "step": 22290
    },
    {
      "entropy": 5.790291404724121,
      "epoch": 1.968827269516072,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004611982079489928,
      "loss": 5.4883,
      "mean_token_accuracy": 0.19619899541139602,
      "num_tokens": 41574728.0,
      "step": 22295
    },
    {
      "entropy": 5.8016759872436525,
      "epoch": 1.9692688096079123,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004611805269332798,
      "loss": 5.6117,
      "mean_token_accuracy": 0.17586255073547363,
      "num_tokens": 41584749.0,
      "step": 22300
    },
    {
      "entropy": 5.856154727935791,
      "epoch": 1.9697103496997528,
      "grad_norm": 1.609375,
      "learning_rate": 0.00046116284227039626,
      "loss": 5.5196,
      "mean_token_accuracy": 0.19622166007757186,
      "num_tokens": 41594658.0,
      "step": 22305
    },
    {
      "entropy": 5.86662278175354,
      "epoch": 1.970151889791593,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004611451539606886,
      "loss": 5.6238,
      "mean_token_accuracy": 0.18310556262731553,
      "num_tokens": 41604548.0,
      "step": 22310
    },
    {
      "entropy": 5.879286241531372,
      "epoch": 1.9705934298834333,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004611274620045031,
      "loss": 5.5758,
      "mean_token_accuracy": 0.1832590565085411,
      "num_tokens": 41614002.0,
      "step": 22315
    },
    {
      "entropy": 5.9100230693817135,
      "epoch": 1.9710349699752738,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004611097664021866,
      "loss": 5.5912,
      "mean_token_accuracy": 0.18254173994064332,
      "num_tokens": 41623793.0,
      "step": 22320
    },
    {
      "entropy": 5.875743818283081,
      "epoch": 1.971476510067114,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046109206715408554,
      "loss": 5.5223,
      "mean_token_accuracy": 0.18192688822746278,
      "num_tokens": 41632608.0,
      "step": 22325
    },
    {
      "entropy": 5.8847327709198,
      "epoch": 1.9719180501589544,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046107436426054676,
      "loss": 5.6075,
      "mean_token_accuracy": 0.18179901987314223,
      "num_tokens": 41641814.0,
      "step": 22330
    },
    {
      "entropy": 5.8777852058410645,
      "epoch": 1.9723595902507949,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000461056657721917,
      "loss": 5.5777,
      "mean_token_accuracy": 0.1865222319960594,
      "num_tokens": 41650643.0,
      "step": 22335
    },
    {
      "entropy": 5.866038846969604,
      "epoch": 1.9728011303426352,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046103894753854317,
      "loss": 5.5984,
      "mean_token_accuracy": 0.18318722844123841,
      "num_tokens": 41661308.0,
      "step": 22340
    },
    {
      "entropy": 5.934350299835205,
      "epoch": 1.9732426704344754,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004610212337107721,
      "loss": 5.6334,
      "mean_token_accuracy": 0.17698234021663667,
      "num_tokens": 41671090.0,
      "step": 22345
    },
    {
      "entropy": 5.845554208755493,
      "epoch": 1.973684210526316,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046100351623895087,
      "loss": 5.5717,
      "mean_token_accuracy": 0.18084558695554734,
      "num_tokens": 41679584.0,
      "step": 22350
    },
    {
      "entropy": 5.820106840133667,
      "epoch": 1.974125750618156,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046098579512342654,
      "loss": 5.5787,
      "mean_token_accuracy": 0.1724436551332474,
      "num_tokens": 41689772.0,
      "step": 22355
    },
    {
      "entropy": 5.863456106185913,
      "epoch": 1.9745672907099965,
      "grad_norm": 7.8125,
      "learning_rate": 0.0004609680703645463,
      "loss": 5.5585,
      "mean_token_accuracy": 0.18134803473949432,
      "num_tokens": 41699196.0,
      "step": 22360
    },
    {
      "entropy": 5.83959469795227,
      "epoch": 1.9750088308018368,
      "grad_norm": 1.546875,
      "learning_rate": 0.00046095034196265726,
      "loss": 5.541,
      "mean_token_accuracy": 0.19384464025497436,
      "num_tokens": 41707899.0,
      "step": 22365
    },
    {
      "entropy": 5.814503622055054,
      "epoch": 1.975450370893677,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046093260991810685,
      "loss": 5.5884,
      "mean_token_accuracy": 0.18825089782476426,
      "num_tokens": 41716903.0,
      "step": 22370
    },
    {
      "entropy": 5.9151726245880125,
      "epoch": 1.9758919109855175,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046091487423124234,
      "loss": 5.4881,
      "mean_token_accuracy": 0.19342504143714906,
      "num_tokens": 41725676.0,
      "step": 22375
    },
    {
      "entropy": 5.867004919052124,
      "epoch": 1.9763334510773578,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046089713490241117,
      "loss": 5.4663,
      "mean_token_accuracy": 0.18849319219589233,
      "num_tokens": 41734720.0,
      "step": 22380
    },
    {
      "entropy": 5.780643272399902,
      "epoch": 1.976774991169198,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004608793919319608,
      "loss": 5.5797,
      "mean_token_accuracy": 0.1914470836520195,
      "num_tokens": 41743866.0,
      "step": 22385
    },
    {
      "entropy": 5.854308462142944,
      "epoch": 1.9772165312610386,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00046086164532023894,
      "loss": 5.569,
      "mean_token_accuracy": 0.19187165498733522,
      "num_tokens": 41754157.0,
      "step": 22390
    },
    {
      "entropy": 5.94117922782898,
      "epoch": 1.9776580713528789,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046084389506759313,
      "loss": 5.5547,
      "mean_token_accuracy": 0.18472244441509247,
      "num_tokens": 41763105.0,
      "step": 22395
    },
    {
      "entropy": 5.853835821151733,
      "epoch": 1.9780996114447191,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046082614117437103,
      "loss": 5.4645,
      "mean_token_accuracy": 0.19168703407049179,
      "num_tokens": 41772623.0,
      "step": 22400
    },
    {
      "entropy": 5.788105726242065,
      "epoch": 1.9785411515365596,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00046080838364092057,
      "loss": 5.5862,
      "mean_token_accuracy": 0.1801395907998085,
      "num_tokens": 41781331.0,
      "step": 22405
    },
    {
      "entropy": 5.908712720870971,
      "epoch": 1.9789826916284,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004607906224675896,
      "loss": 5.7011,
      "mean_token_accuracy": 0.17572573125362395,
      "num_tokens": 41789727.0,
      "step": 22410
    },
    {
      "entropy": 5.9187171459198,
      "epoch": 1.9794242317202402,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004607728576547259,
      "loss": 5.5734,
      "mean_token_accuracy": 0.18436077684164048,
      "num_tokens": 41798531.0,
      "step": 22415
    },
    {
      "entropy": 5.851548862457276,
      "epoch": 1.9798657718120807,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046075508920267766,
      "loss": 5.5733,
      "mean_token_accuracy": 0.18235182613134385,
      "num_tokens": 41808304.0,
      "step": 22420
    },
    {
      "entropy": 5.7556235790252686,
      "epoch": 1.9803073119039207,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004607373171117929,
      "loss": 5.464,
      "mean_token_accuracy": 0.19313152134418488,
      "num_tokens": 41817427.0,
      "step": 22425
    },
    {
      "entropy": 5.834622097015381,
      "epoch": 1.9807488519957612,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004607195413824197,
      "loss": 5.5433,
      "mean_token_accuracy": 0.1937289521098137,
      "num_tokens": 41827202.0,
      "step": 22430
    },
    {
      "entropy": 5.887858724594116,
      "epoch": 1.9811903920876015,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004607017620149062,
      "loss": 5.4959,
      "mean_token_accuracy": 0.1958024889230728,
      "num_tokens": 41836213.0,
      "step": 22435
    },
    {
      "entropy": 5.853147554397583,
      "epoch": 1.9816319321794418,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004606839790096009,
      "loss": 5.6249,
      "mean_token_accuracy": 0.17938733696937562,
      "num_tokens": 41845358.0,
      "step": 22440
    },
    {
      "entropy": 5.869603633880615,
      "epoch": 1.9820734722712823,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004606661923668521,
      "loss": 5.5683,
      "mean_token_accuracy": 0.1915348917245865,
      "num_tokens": 41854572.0,
      "step": 22445
    },
    {
      "entropy": 5.856293249130249,
      "epoch": 1.9825150123631226,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004606484020870081,
      "loss": 5.5322,
      "mean_token_accuracy": 0.1887270137667656,
      "num_tokens": 41863643.0,
      "step": 22450
    },
    {
      "entropy": 5.792924165725708,
      "epoch": 1.9829565524549628,
      "grad_norm": 1.53125,
      "learning_rate": 0.00046063060817041747,
      "loss": 5.4599,
      "mean_token_accuracy": 0.19091761261224746,
      "num_tokens": 41873004.0,
      "step": 22455
    },
    {
      "entropy": 5.8782883644104,
      "epoch": 1.9833980925468033,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004606128106174288,
      "loss": 5.541,
      "mean_token_accuracy": 0.19107352793216706,
      "num_tokens": 41881530.0,
      "step": 22460
    },
    {
      "entropy": 5.866301345825195,
      "epoch": 1.9838396326386436,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004605950094283908,
      "loss": 5.5796,
      "mean_token_accuracy": 0.1824715793132782,
      "num_tokens": 41890342.0,
      "step": 22465
    },
    {
      "entropy": 5.826586675643921,
      "epoch": 1.9842811727304839,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046057720460365205,
      "loss": 5.5082,
      "mean_token_accuracy": 0.18979047536849974,
      "num_tokens": 41899274.0,
      "step": 22470
    },
    {
      "entropy": 5.889193964004517,
      "epoch": 1.9847227128223244,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046055939614356146,
      "loss": 5.5579,
      "mean_token_accuracy": 0.18893688172101974,
      "num_tokens": 41908835.0,
      "step": 22475
    },
    {
      "entropy": 5.796959686279297,
      "epoch": 1.9851642529141647,
      "grad_norm": 1.375,
      "learning_rate": 0.0004605415840484678,
      "loss": 5.4761,
      "mean_token_accuracy": 0.1970122918486595,
      "num_tokens": 41917900.0,
      "step": 22480
    },
    {
      "entropy": 5.848794317245483,
      "epoch": 1.985605793006005,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046052376831872007,
      "loss": 5.5521,
      "mean_token_accuracy": 0.19222814589738846,
      "num_tokens": 41926964.0,
      "step": 22485
    },
    {
      "entropy": 5.9426634311676025,
      "epoch": 1.9860473330978454,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004605059489546672,
      "loss": 5.6342,
      "mean_token_accuracy": 0.187479405105114,
      "num_tokens": 41936401.0,
      "step": 22490
    },
    {
      "entropy": 5.815711593627929,
      "epoch": 1.9864888731896855,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046048812595665824,
      "loss": 5.4726,
      "mean_token_accuracy": 0.1895550012588501,
      "num_tokens": 41944665.0,
      "step": 22495
    },
    {
      "entropy": 5.7870808124542235,
      "epoch": 1.986930413281526,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004604702993250425,
      "loss": 5.5599,
      "mean_token_accuracy": 0.17513049244880677,
      "num_tokens": 41953868.0,
      "step": 22500
    },
    {
      "entropy": 5.7854469299316404,
      "epoch": 1.9873719533733663,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00046045246906016904,
      "loss": 5.5714,
      "mean_token_accuracy": 0.19452321231365205,
      "num_tokens": 41963013.0,
      "step": 22505
    },
    {
      "entropy": 5.856605291366577,
      "epoch": 1.9878134934652065,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004604346351623872,
      "loss": 5.5704,
      "mean_token_accuracy": 0.18064149618148803,
      "num_tokens": 41972926.0,
      "step": 22510
    },
    {
      "entropy": 5.946653175354004,
      "epoch": 1.988255033557047,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046041679763204625,
      "loss": 5.562,
      "mean_token_accuracy": 0.18679905384778978,
      "num_tokens": 41981773.0,
      "step": 22515
    },
    {
      "entropy": 5.87430214881897,
      "epoch": 1.9886965736488873,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004603989564694957,
      "loss": 5.5492,
      "mean_token_accuracy": 0.18773624449968337,
      "num_tokens": 41990893.0,
      "step": 22520
    },
    {
      "entropy": 5.777024412155152,
      "epoch": 1.9891381137407276,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004603811116750851,
      "loss": 5.572,
      "mean_token_accuracy": 0.18655288219451904,
      "num_tokens": 42000842.0,
      "step": 22525
    },
    {
      "entropy": 5.950747728347778,
      "epoch": 1.989579653832568,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004603632632491639,
      "loss": 5.5504,
      "mean_token_accuracy": 0.19242268651723862,
      "num_tokens": 42009430.0,
      "step": 22530
    },
    {
      "entropy": 5.886588096618652,
      "epoch": 1.9900211939244083,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046034541119208177,
      "loss": 5.6638,
      "mean_token_accuracy": 0.1848289281129837,
      "num_tokens": 42018985.0,
      "step": 22535
    },
    {
      "entropy": 5.83545618057251,
      "epoch": 1.9904627340162486,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046032755550418845,
      "loss": 5.5426,
      "mean_token_accuracy": 0.18397380113601686,
      "num_tokens": 42028885.0,
      "step": 22540
    },
    {
      "entropy": 5.997489500045776,
      "epoch": 1.9909042741080891,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004603096961858338,
      "loss": 5.6546,
      "mean_token_accuracy": 0.1801191136240959,
      "num_tokens": 42037838.0,
      "step": 22545
    },
    {
      "entropy": 5.849003696441651,
      "epoch": 1.9913458141999294,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046029183323736755,
      "loss": 5.5334,
      "mean_token_accuracy": 0.1854138970375061,
      "num_tokens": 42047212.0,
      "step": 22550
    },
    {
      "entropy": 5.876599836349487,
      "epoch": 1.9917873542917697,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004602739666591397,
      "loss": 5.6091,
      "mean_token_accuracy": 0.18454371094703675,
      "num_tokens": 42056680.0,
      "step": 22555
    },
    {
      "entropy": 5.848644828796386,
      "epoch": 1.9922288943836102,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046025609645150014,
      "loss": 5.6213,
      "mean_token_accuracy": 0.18110444247722626,
      "num_tokens": 42065604.0,
      "step": 22560
    },
    {
      "entropy": 5.923292446136474,
      "epoch": 1.9926704344754502,
      "grad_norm": 1.3125,
      "learning_rate": 0.000460238222614799,
      "loss": 5.5485,
      "mean_token_accuracy": 0.18929385989904404,
      "num_tokens": 42074672.0,
      "step": 22565
    },
    {
      "entropy": 5.852539300918579,
      "epoch": 1.9931119745672907,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004602203451493865,
      "loss": 5.5208,
      "mean_token_accuracy": 0.1880763754248619,
      "num_tokens": 42083586.0,
      "step": 22570
    },
    {
      "entropy": 5.808136749267578,
      "epoch": 1.993553514659131,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046020246405561283,
      "loss": 5.4984,
      "mean_token_accuracy": 0.19051387161016464,
      "num_tokens": 42092328.0,
      "step": 22575
    },
    {
      "entropy": 5.78072829246521,
      "epoch": 1.9939950547509713,
      "grad_norm": 1.546875,
      "learning_rate": 0.00046018457933382807,
      "loss": 5.5258,
      "mean_token_accuracy": 0.19395862370729447,
      "num_tokens": 42102154.0,
      "step": 22580
    },
    {
      "entropy": 5.895393991470337,
      "epoch": 1.9944365948428118,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046016669098438286,
      "loss": 5.6001,
      "mean_token_accuracy": 0.18295152634382247,
      "num_tokens": 42111762.0,
      "step": 22585
    },
    {
      "entropy": 5.88388237953186,
      "epoch": 1.994878134934652,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004601487990076274,
      "loss": 5.5831,
      "mean_token_accuracy": 0.18515295684337615,
      "num_tokens": 42120631.0,
      "step": 22590
    },
    {
      "entropy": 5.800492143630981,
      "epoch": 1.9953196750264923,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004601309034039123,
      "loss": 5.4307,
      "mean_token_accuracy": 0.20304099470376968,
      "num_tokens": 42129311.0,
      "step": 22595
    },
    {
      "entropy": 5.871335887908936,
      "epoch": 1.9957612151183328,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004601130041735881,
      "loss": 5.6781,
      "mean_token_accuracy": 0.17517089247703552,
      "num_tokens": 42138735.0,
      "step": 22600
    },
    {
      "entropy": 5.942437934875488,
      "epoch": 1.996202755210173,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004600951013170054,
      "loss": 5.6159,
      "mean_token_accuracy": 0.19071706235408784,
      "num_tokens": 42148416.0,
      "step": 22605
    },
    {
      "entropy": 5.9012127876281735,
      "epoch": 1.9966442953020134,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046007719483451493,
      "loss": 5.5342,
      "mean_token_accuracy": 0.19162551015615464,
      "num_tokens": 42157191.0,
      "step": 22610
    },
    {
      "entropy": 5.73353009223938,
      "epoch": 1.9970858353938539,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046005928472646747,
      "loss": 5.4915,
      "mean_token_accuracy": 0.20012723356485368,
      "num_tokens": 42165798.0,
      "step": 22615
    },
    {
      "entropy": 5.8414887428283695,
      "epoch": 1.9975273754856941,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046004137099321386,
      "loss": 5.5892,
      "mean_token_accuracy": 0.18004080802202224,
      "num_tokens": 42175318.0,
      "step": 22620
    },
    {
      "entropy": 5.9408117771148685,
      "epoch": 1.9979689155775344,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046002345363510506,
      "loss": 5.6175,
      "mean_token_accuracy": 0.17718316614627838,
      "num_tokens": 42184800.0,
      "step": 22625
    },
    {
      "entropy": 5.915752172470093,
      "epoch": 1.998410455669375,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004600055326524921,
      "loss": 5.4539,
      "mean_token_accuracy": 0.1973792567849159,
      "num_tokens": 42194172.0,
      "step": 22630
    },
    {
      "entropy": 5.833796882629395,
      "epoch": 1.998851995761215,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045998760804572583,
      "loss": 5.4843,
      "mean_token_accuracy": 0.19283089637756348,
      "num_tokens": 42203102.0,
      "step": 22635
    },
    {
      "entropy": 5.735662746429443,
      "epoch": 1.9992935358530555,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045996967981515763,
      "loss": 5.4583,
      "mean_token_accuracy": 0.19784962832927705,
      "num_tokens": 42212421.0,
      "step": 22640
    },
    {
      "entropy": 5.868128299713135,
      "epoch": 1.9997350759448957,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045995174796113856,
      "loss": 5.5649,
      "mean_token_accuracy": 0.18520353883504867,
      "num_tokens": 42221528.0,
      "step": 22645
    },
    {
      "entropy": 5.839421129226684,
      "epoch": 2.000176616036736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045993381248402,
      "loss": 5.5228,
      "mean_token_accuracy": 0.1844082862138748,
      "num_tokens": 42230628.0,
      "step": 22650
    },
    {
      "entropy": 5.806019258499146,
      "epoch": 2.0006181561285765,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045991587338415306,
      "loss": 5.4818,
      "mean_token_accuracy": 0.19325753301382065,
      "num_tokens": 42239571.0,
      "step": 22655
    },
    {
      "entropy": 5.84874792098999,
      "epoch": 2.001059696220417,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045989793066188953,
      "loss": 5.4629,
      "mean_token_accuracy": 0.18989743441343307,
      "num_tokens": 42248010.0,
      "step": 22660
    },
    {
      "entropy": 5.757057428359985,
      "epoch": 2.001501236312257,
      "grad_norm": 1.515625,
      "learning_rate": 0.00045987998431758063,
      "loss": 5.4495,
      "mean_token_accuracy": 0.1924544796347618,
      "num_tokens": 42257251.0,
      "step": 22665
    },
    {
      "entropy": 5.853150463104248,
      "epoch": 2.0019427764040976,
      "grad_norm": 1.375,
      "learning_rate": 0.0004598620343515779,
      "loss": 5.5734,
      "mean_token_accuracy": 0.18373352140188218,
      "num_tokens": 42266441.0,
      "step": 22670
    },
    {
      "entropy": 5.829499244689941,
      "epoch": 2.0023843164959376,
      "grad_norm": 1.8125,
      "learning_rate": 0.00045984408076423313,
      "loss": 5.4965,
      "mean_token_accuracy": 0.18809223771095276,
      "num_tokens": 42275995.0,
      "step": 22675
    },
    {
      "entropy": 5.870511198043824,
      "epoch": 2.002825856587778,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004598261235558979,
      "loss": 5.4847,
      "mean_token_accuracy": 0.19318899214267732,
      "num_tokens": 42284065.0,
      "step": 22680
    },
    {
      "entropy": 5.811155986785889,
      "epoch": 2.0032673966796186,
      "grad_norm": 1.25,
      "learning_rate": 0.00045980816272692403,
      "loss": 5.5817,
      "mean_token_accuracy": 0.18625827878713608,
      "num_tokens": 42295309.0,
      "step": 22685
    },
    {
      "entropy": 5.942724132537842,
      "epoch": 2.0037089367714587,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045979019827766345,
      "loss": 5.5041,
      "mean_token_accuracy": 0.19084416180849076,
      "num_tokens": 42304390.0,
      "step": 22690
    },
    {
      "entropy": 5.883755922317505,
      "epoch": 2.004150476863299,
      "grad_norm": 1.546875,
      "learning_rate": 0.00045977223020846785,
      "loss": 5.5098,
      "mean_token_accuracy": 0.19103406816720964,
      "num_tokens": 42313830.0,
      "step": 22695
    },
    {
      "entropy": 5.890994644165039,
      "epoch": 2.0045920169551397,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004597542585196895,
      "loss": 5.5085,
      "mean_token_accuracy": 0.18334235846996308,
      "num_tokens": 42323827.0,
      "step": 22700
    },
    {
      "entropy": 5.940227031707764,
      "epoch": 2.0050335570469797,
      "grad_norm": 1.25,
      "learning_rate": 0.00045973628321168016,
      "loss": 5.608,
      "mean_token_accuracy": 0.1798837289214134,
      "num_tokens": 42333221.0,
      "step": 22705
    },
    {
      "entropy": 5.782166004180908,
      "epoch": 2.00547509713882,
      "grad_norm": 1.453125,
      "learning_rate": 0.00045971830428479215,
      "loss": 5.4852,
      "mean_token_accuracy": 0.196746626496315,
      "num_tokens": 42343036.0,
      "step": 22710
    },
    {
      "entropy": 5.8064612865448,
      "epoch": 2.0059166372306607,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00045970032173937766,
      "loss": 5.4536,
      "mean_token_accuracy": 0.19639276713132858,
      "num_tokens": 42351916.0,
      "step": 22715
    },
    {
      "entropy": 5.903685617446899,
      "epoch": 2.0063581773225008,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00045968233557578883,
      "loss": 5.5769,
      "mean_token_accuracy": 0.18607098013162612,
      "num_tokens": 42362412.0,
      "step": 22720
    },
    {
      "entropy": 5.839944458007812,
      "epoch": 2.0067997174143413,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004596643457943781,
      "loss": 5.4502,
      "mean_token_accuracy": 0.19429419487714766,
      "num_tokens": 42372773.0,
      "step": 22725
    },
    {
      "entropy": 5.855646562576294,
      "epoch": 2.0072412575061818,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004596463523954979,
      "loss": 5.4541,
      "mean_token_accuracy": 0.19093629270792006,
      "num_tokens": 42382009.0,
      "step": 22730
    },
    {
      "entropy": 5.810858821868896,
      "epoch": 2.007682797598022,
      "grad_norm": 1.5,
      "learning_rate": 0.00045962835537950063,
      "loss": 5.4824,
      "mean_token_accuracy": 0.19049643129110336,
      "num_tokens": 42390961.0,
      "step": 22735
    },
    {
      "entropy": 5.783297443389893,
      "epoch": 2.0081243376898623,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045961035474673893,
      "loss": 5.4852,
      "mean_token_accuracy": 0.19171667844057083,
      "num_tokens": 42400483.0,
      "step": 22740
    },
    {
      "entropy": 5.715685558319092,
      "epoch": 2.0085658777817024,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00045959235049756537,
      "loss": 5.3193,
      "mean_token_accuracy": 0.21058298647403717,
      "num_tokens": 42409285.0,
      "step": 22745
    },
    {
      "entropy": 5.817371463775634,
      "epoch": 2.009007417873543,
      "grad_norm": 1.453125,
      "learning_rate": 0.00045957434263233253,
      "loss": 5.5129,
      "mean_token_accuracy": 0.1871025249361992,
      "num_tokens": 42418950.0,
      "step": 22750
    },
    {
      "entropy": 5.818934345245362,
      "epoch": 2.0094489579653834,
      "grad_norm": 1.703125,
      "learning_rate": 0.00045955633115139343,
      "loss": 5.5907,
      "mean_token_accuracy": 0.1795429915189743,
      "num_tokens": 42428285.0,
      "step": 22755
    },
    {
      "entropy": 5.857922887802124,
      "epoch": 2.0098904980572234,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004595383160551007,
      "loss": 5.4543,
      "mean_token_accuracy": 0.19346287846565247,
      "num_tokens": 42437832.0,
      "step": 22760
    },
    {
      "entropy": 5.798523569107056,
      "epoch": 2.010332038149064,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004595202973438074,
      "loss": 5.4787,
      "mean_token_accuracy": 0.1867272287607193,
      "num_tokens": 42447704.0,
      "step": 22765
    },
    {
      "entropy": 5.862973022460937,
      "epoch": 2.0107735782409044,
      "grad_norm": 1.375,
      "learning_rate": 0.0004595022750178663,
      "loss": 5.5375,
      "mean_token_accuracy": 0.19111043959856033,
      "num_tokens": 42456571.0,
      "step": 22770
    },
    {
      "entropy": 5.835361957550049,
      "epoch": 2.0112151183327445,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004594842490776306,
      "loss": 5.438,
      "mean_token_accuracy": 0.19449933767318725,
      "num_tokens": 42466522.0,
      "step": 22775
    },
    {
      "entropy": 5.8186040878295895,
      "epoch": 2.011656658424585,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045946621952345335,
      "loss": 5.487,
      "mean_token_accuracy": 0.19044015258550645,
      "num_tokens": 42475940.0,
      "step": 22780
    },
    {
      "entropy": 5.873414659500122,
      "epoch": 2.0120981985164255,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004594481863556879,
      "loss": 5.5722,
      "mean_token_accuracy": 0.18056110590696334,
      "num_tokens": 42486456.0,
      "step": 22785
    },
    {
      "entropy": 5.871607732772827,
      "epoch": 2.0125397386082655,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045943014957468726,
      "loss": 5.5507,
      "mean_token_accuracy": 0.19045219719409942,
      "num_tokens": 42495578.0,
      "step": 22790
    },
    {
      "entropy": 5.809818267822266,
      "epoch": 2.012981278700106,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004594121091808049,
      "loss": 5.3966,
      "mean_token_accuracy": 0.19554734528064727,
      "num_tokens": 42504532.0,
      "step": 22795
    },
    {
      "entropy": 5.838075637817383,
      "epoch": 2.0134228187919465,
      "grad_norm": 1.75,
      "learning_rate": 0.00045939406517439427,
      "loss": 5.4543,
      "mean_token_accuracy": 0.18982529044151306,
      "num_tokens": 42513683.0,
      "step": 22800
    },
    {
      "entropy": 5.85420069694519,
      "epoch": 2.0138643588837866,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00045937601755580864,
      "loss": 5.557,
      "mean_token_accuracy": 0.1823415368795395,
      "num_tokens": 42522717.0,
      "step": 22805
    },
    {
      "entropy": 5.871758222579956,
      "epoch": 2.014305898975627,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004593579663254018,
      "loss": 5.5632,
      "mean_token_accuracy": 0.18341727703809738,
      "num_tokens": 42531847.0,
      "step": 22810
    },
    {
      "entropy": 5.842437267303467,
      "epoch": 2.014747439067467,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004593399114835272,
      "loss": 5.4833,
      "mean_token_accuracy": 0.1937089204788208,
      "num_tokens": 42540869.0,
      "step": 22815
    },
    {
      "entropy": 5.763482666015625,
      "epoch": 2.0151889791593076,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045932185303053857,
      "loss": 5.4709,
      "mean_token_accuracy": 0.19459028095006942,
      "num_tokens": 42550875.0,
      "step": 22820
    },
    {
      "entropy": 5.783092212677002,
      "epoch": 2.015630519251148,
      "grad_norm": 1.375,
      "learning_rate": 0.0004593037909667898,
      "loss": 5.4483,
      "mean_token_accuracy": 0.19772506207227708,
      "num_tokens": 42560066.0,
      "step": 22825
    },
    {
      "entropy": 5.768995761871338,
      "epoch": 2.016072059342988,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004592857252926344,
      "loss": 5.4607,
      "mean_token_accuracy": 0.19899552762508393,
      "num_tokens": 42568733.0,
      "step": 22830
    },
    {
      "entropy": 5.8458092212677,
      "epoch": 2.0165135994348287,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045926765600842653,
      "loss": 5.4817,
      "mean_token_accuracy": 0.19108146727085112,
      "num_tokens": 42577565.0,
      "step": 22835
    },
    {
      "entropy": 5.821941661834717,
      "epoch": 2.016955139526669,
      "grad_norm": 1.59375,
      "learning_rate": 0.00045924958311452006,
      "loss": 5.5285,
      "mean_token_accuracy": 0.19317487180233,
      "num_tokens": 42586489.0,
      "step": 22840
    },
    {
      "entropy": 5.876119565963745,
      "epoch": 2.017396679618509,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045923150661126904,
      "loss": 5.5643,
      "mean_token_accuracy": 0.18604032844305038,
      "num_tokens": 42596263.0,
      "step": 22845
    },
    {
      "entropy": 5.832976770401001,
      "epoch": 2.0178382197103497,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004592134264990276,
      "loss": 5.4694,
      "mean_token_accuracy": 0.18742228746414186,
      "num_tokens": 42605787.0,
      "step": 22850
    },
    {
      "entropy": 5.917246150970459,
      "epoch": 2.01827975980219,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045919534277814996,
      "loss": 5.6428,
      "mean_token_accuracy": 0.1784665420651436,
      "num_tokens": 42616014.0,
      "step": 22855
    },
    {
      "entropy": 5.871256685256958,
      "epoch": 2.0187212998940303,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004591772554489902,
      "loss": 5.523,
      "mean_token_accuracy": 0.19144122302532196,
      "num_tokens": 42624839.0,
      "step": 22860
    },
    {
      "entropy": 5.870565700531006,
      "epoch": 2.0191628399858708,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004591591645119028,
      "loss": 5.5234,
      "mean_token_accuracy": 0.18872184306383133,
      "num_tokens": 42633811.0,
      "step": 22865
    },
    {
      "entropy": 5.735362148284912,
      "epoch": 2.0196043800777113,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045914106996724217,
      "loss": 5.4112,
      "mean_token_accuracy": 0.1999753698706627,
      "num_tokens": 42643140.0,
      "step": 22870
    },
    {
      "entropy": 5.779294633865357,
      "epoch": 2.0200459201695513,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045912297181536257,
      "loss": 5.3921,
      "mean_token_accuracy": 0.19984357357025145,
      "num_tokens": 42651123.0,
      "step": 22875
    },
    {
      "entropy": 5.8172910690307615,
      "epoch": 2.020487460261392,
      "grad_norm": 3.171875,
      "learning_rate": 0.00045910487005661877,
      "loss": 5.462,
      "mean_token_accuracy": 0.19957963973283768,
      "num_tokens": 42659880.0,
      "step": 22880
    },
    {
      "entropy": 5.813700151443482,
      "epoch": 2.020929000353232,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004590867646913653,
      "loss": 5.427,
      "mean_token_accuracy": 0.19572501331567765,
      "num_tokens": 42668689.0,
      "step": 22885
    },
    {
      "entropy": 5.81799726486206,
      "epoch": 2.0213705404450724,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00045906865571995666,
      "loss": 5.4843,
      "mean_token_accuracy": 0.19426885843276978,
      "num_tokens": 42678403.0,
      "step": 22890
    },
    {
      "entropy": 5.7387031555175785,
      "epoch": 2.021812080536913,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004590505431427479,
      "loss": 5.4862,
      "mean_token_accuracy": 0.19383792877197265,
      "num_tokens": 42688203.0,
      "step": 22895
    },
    {
      "entropy": 5.832409524917603,
      "epoch": 2.022253620628753,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004590324269600936,
      "loss": 5.5641,
      "mean_token_accuracy": 0.18232239931821823,
      "num_tokens": 42698343.0,
      "step": 22900
    },
    {
      "entropy": 5.815659713745117,
      "epoch": 2.0226951607205934,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045901430717234875,
      "loss": 5.3787,
      "mean_token_accuracy": 0.20219678282737732,
      "num_tokens": 42707483.0,
      "step": 22905
    },
    {
      "entropy": 5.83566312789917,
      "epoch": 2.023136700812434,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004589961837798682,
      "loss": 5.5576,
      "mean_token_accuracy": 0.18750419914722444,
      "num_tokens": 42717820.0,
      "step": 22910
    },
    {
      "entropy": 5.862226772308349,
      "epoch": 2.023578240904274,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045897805678300714,
      "loss": 5.5015,
      "mean_token_accuracy": 0.19326231330633165,
      "num_tokens": 42726996.0,
      "step": 22915
    },
    {
      "entropy": 5.834670066833496,
      "epoch": 2.0240197809961145,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004589599261821205,
      "loss": 5.5028,
      "mean_token_accuracy": 0.19280921220779418,
      "num_tokens": 42735779.0,
      "step": 22920
    },
    {
      "entropy": 5.830481433868409,
      "epoch": 2.024461321087955,
      "grad_norm": 1.5,
      "learning_rate": 0.00045894179197756364,
      "loss": 5.5948,
      "mean_token_accuracy": 0.18763089776039124,
      "num_tokens": 42745089.0,
      "step": 22925
    },
    {
      "entropy": 5.87225751876831,
      "epoch": 2.024902861179795,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00045892365416969164,
      "loss": 5.5474,
      "mean_token_accuracy": 0.18282757103443145,
      "num_tokens": 42754963.0,
      "step": 22930
    },
    {
      "entropy": 5.9579816341400145,
      "epoch": 2.0253444012716355,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00045890551275885985,
      "loss": 5.5739,
      "mean_token_accuracy": 0.18304225355386733,
      "num_tokens": 42764856.0,
      "step": 22935
    },
    {
      "entropy": 5.867724943161011,
      "epoch": 2.025785941363476,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045888736774542363,
      "loss": 5.5456,
      "mean_token_accuracy": 0.18808584809303283,
      "num_tokens": 42775028.0,
      "step": 22940
    },
    {
      "entropy": 5.793195915222168,
      "epoch": 2.026227481455316,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004588692191297385,
      "loss": 5.5161,
      "mean_token_accuracy": 0.19287458956241607,
      "num_tokens": 42784288.0,
      "step": 22945
    },
    {
      "entropy": 5.862449693679809,
      "epoch": 2.0266690215471566,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004588510669121599,
      "loss": 5.5743,
      "mean_token_accuracy": 0.1813485234975815,
      "num_tokens": 42793356.0,
      "step": 22950
    },
    {
      "entropy": 5.786140584945679,
      "epoch": 2.0271105616389966,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045883291109304346,
      "loss": 5.4217,
      "mean_token_accuracy": 0.19651312679052352,
      "num_tokens": 42802287.0,
      "step": 22955
    },
    {
      "entropy": 5.846931314468383,
      "epoch": 2.027552101730837,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004588147516727449,
      "loss": 5.5543,
      "mean_token_accuracy": 0.18946802765130996,
      "num_tokens": 42812206.0,
      "step": 22960
    },
    {
      "entropy": 5.869739723205567,
      "epoch": 2.0279936418226776,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004587965886516199,
      "loss": 5.5499,
      "mean_token_accuracy": 0.18370679020881653,
      "num_tokens": 42822046.0,
      "step": 22965
    },
    {
      "entropy": 5.8442333221435545,
      "epoch": 2.0284351819145177,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045877842203002417,
      "loss": 5.4803,
      "mean_token_accuracy": 0.1942022532224655,
      "num_tokens": 42831228.0,
      "step": 22970
    },
    {
      "entropy": 5.857371664047241,
      "epoch": 2.028876722006358,
      "grad_norm": 1.453125,
      "learning_rate": 0.00045876025180831373,
      "loss": 5.5619,
      "mean_token_accuracy": 0.18672799468040466,
      "num_tokens": 42840683.0,
      "step": 22975
    },
    {
      "entropy": 5.874697685241699,
      "epoch": 2.0293182620981987,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045874207798684446,
      "loss": 5.4923,
      "mean_token_accuracy": 0.18958508521318435,
      "num_tokens": 42849731.0,
      "step": 22980
    },
    {
      "entropy": 5.8211428165435795,
      "epoch": 2.0297598021900387,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045872390056597235,
      "loss": 5.54,
      "mean_token_accuracy": 0.1890716850757599,
      "num_tokens": 42858028.0,
      "step": 22985
    },
    {
      "entropy": 5.82522177696228,
      "epoch": 2.030201342281879,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004587057195460536,
      "loss": 5.5371,
      "mean_token_accuracy": 0.1830664485692978,
      "num_tokens": 42867340.0,
      "step": 22990
    },
    {
      "entropy": 5.973309469223023,
      "epoch": 2.0306428823737197,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045868753492744424,
      "loss": 5.5508,
      "mean_token_accuracy": 0.1874582961201668,
      "num_tokens": 42877006.0,
      "step": 22995
    },
    {
      "entropy": 5.817297792434692,
      "epoch": 2.0310844224655598,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004586693467105005,
      "loss": 5.4191,
      "mean_token_accuracy": 0.19256477952003478,
      "num_tokens": 42886324.0,
      "step": 23000
    },
    {
      "entropy": 5.770321226119995,
      "epoch": 2.0315259625574003,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004586511548955788,
      "loss": 5.5196,
      "mean_token_accuracy": 0.19171882420778275,
      "num_tokens": 42896411.0,
      "step": 23005
    },
    {
      "entropy": 5.705739450454712,
      "epoch": 2.0319675026492408,
      "grad_norm": 1.25,
      "learning_rate": 0.0004586329594830353,
      "loss": 5.399,
      "mean_token_accuracy": 0.19676823318004608,
      "num_tokens": 42905606.0,
      "step": 23010
    },
    {
      "entropy": 5.725629663467407,
      "epoch": 2.032409042741081,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045861476047322664,
      "loss": 5.3611,
      "mean_token_accuracy": 0.1984923869371414,
      "num_tokens": 42914226.0,
      "step": 23015
    },
    {
      "entropy": 5.849899530410767,
      "epoch": 2.0328505828329213,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045859655786650925,
      "loss": 5.5696,
      "mean_token_accuracy": 0.1839447095990181,
      "num_tokens": 42923910.0,
      "step": 23020
    },
    {
      "entropy": 5.785299015045166,
      "epoch": 2.0332921229247614,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004585783516632397,
      "loss": 5.4121,
      "mean_token_accuracy": 0.2079802080988884,
      "num_tokens": 42933203.0,
      "step": 23025
    },
    {
      "entropy": 5.885000133514405,
      "epoch": 2.033733663016602,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004585601418637747,
      "loss": 5.4877,
      "mean_token_accuracy": 0.18946128934621811,
      "num_tokens": 42942143.0,
      "step": 23030
    },
    {
      "entropy": 5.809240865707397,
      "epoch": 2.0341752031084424,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004585419284684708,
      "loss": 5.5126,
      "mean_token_accuracy": 0.18697378039360046,
      "num_tokens": 42951432.0,
      "step": 23035
    },
    {
      "entropy": 5.7970703125,
      "epoch": 2.0346167432002824,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000458523711477685,
      "loss": 5.4277,
      "mean_token_accuracy": 0.1922255724668503,
      "num_tokens": 42961033.0,
      "step": 23040
    },
    {
      "entropy": 5.817980670928955,
      "epoch": 2.035058283292123,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00045850549089177406,
      "loss": 5.3986,
      "mean_token_accuracy": 0.20276558697223662,
      "num_tokens": 42971371.0,
      "step": 23045
    },
    {
      "entropy": 5.858118104934692,
      "epoch": 2.0354998233839634,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045848726671109493,
      "loss": 5.4789,
      "mean_token_accuracy": 0.18047891408205033,
      "num_tokens": 42980445.0,
      "step": 23050
    },
    {
      "entropy": 5.821119689941407,
      "epoch": 2.0359413634758035,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045846903893600457,
      "loss": 5.5047,
      "mean_token_accuracy": 0.18881763219833375,
      "num_tokens": 42989923.0,
      "step": 23055
    },
    {
      "entropy": 5.7100324630737305,
      "epoch": 2.036382903567644,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045845080756686007,
      "loss": 5.3725,
      "mean_token_accuracy": 0.19411784112453462,
      "num_tokens": 42998370.0,
      "step": 23060
    },
    {
      "entropy": 5.823016738891601,
      "epoch": 2.0368244436594845,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045843257260401855,
      "loss": 5.4752,
      "mean_token_accuracy": 0.1927361011505127,
      "num_tokens": 43007641.0,
      "step": 23065
    },
    {
      "entropy": 5.748564052581787,
      "epoch": 2.0372659837513245,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00045841433404783733,
      "loss": 5.3989,
      "mean_token_accuracy": 0.19936287850141526,
      "num_tokens": 43016271.0,
      "step": 23070
    },
    {
      "entropy": 5.758531522750855,
      "epoch": 2.037707523843165,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004583960918986736,
      "loss": 5.3843,
      "mean_token_accuracy": 0.199419304728508,
      "num_tokens": 43024525.0,
      "step": 23075
    },
    {
      "entropy": 5.784092617034912,
      "epoch": 2.0381490639350055,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045837784615688473,
      "loss": 5.5123,
      "mean_token_accuracy": 0.1909510537981987,
      "num_tokens": 43033248.0,
      "step": 23080
    },
    {
      "entropy": 5.944856929779053,
      "epoch": 2.0385906040268456,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004583595968228281,
      "loss": 5.6602,
      "mean_token_accuracy": 0.18228560984134673,
      "num_tokens": 43042171.0,
      "step": 23085
    },
    {
      "entropy": 5.866355991363525,
      "epoch": 2.039032144118686,
      "grad_norm": 1.578125,
      "learning_rate": 0.00045834134389686137,
      "loss": 5.4453,
      "mean_token_accuracy": 0.1933089539408684,
      "num_tokens": 43051447.0,
      "step": 23090
    },
    {
      "entropy": 5.767711782455445,
      "epoch": 2.039473684210526,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004583230873793419,
      "loss": 5.3989,
      "mean_token_accuracy": 0.19309349358081818,
      "num_tokens": 43059869.0,
      "step": 23095
    },
    {
      "entropy": 5.707624816894532,
      "epoch": 2.0399152243023666,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045830482727062736,
      "loss": 5.3656,
      "mean_token_accuracy": 0.1994778260588646,
      "num_tokens": 43068412.0,
      "step": 23100
    },
    {
      "entropy": 5.847116756439209,
      "epoch": 2.040356764394207,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045828656357107557,
      "loss": 5.5258,
      "mean_token_accuracy": 0.18999373465776442,
      "num_tokens": 43077130.0,
      "step": 23105
    },
    {
      "entropy": 5.90912504196167,
      "epoch": 2.040798304486047,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004582682962810442,
      "loss": 5.5448,
      "mean_token_accuracy": 0.18716578781604767,
      "num_tokens": 43086304.0,
      "step": 23110
    },
    {
      "entropy": 5.862572097778321,
      "epoch": 2.0412398445778877,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004582500254008912,
      "loss": 5.4788,
      "mean_token_accuracy": 0.1982337400317192,
      "num_tokens": 43095734.0,
      "step": 23115
    },
    {
      "entropy": 5.801224851608277,
      "epoch": 2.041681384669728,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00045823175093097444,
      "loss": 5.4617,
      "mean_token_accuracy": 0.19180019050836564,
      "num_tokens": 43104761.0,
      "step": 23120
    },
    {
      "entropy": 5.828719997406006,
      "epoch": 2.042122924761568,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004582134728716518,
      "loss": 5.5219,
      "mean_token_accuracy": 0.18840306848287583,
      "num_tokens": 43114182.0,
      "step": 23125
    },
    {
      "entropy": 5.78659200668335,
      "epoch": 2.0425644648534087,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004581951912232814,
      "loss": 5.4803,
      "mean_token_accuracy": 0.1911533534526825,
      "num_tokens": 43124893.0,
      "step": 23130
    },
    {
      "entropy": 5.8717292785644535,
      "epoch": 2.043006004945249,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004581769059862215,
      "loss": 5.4701,
      "mean_token_accuracy": 0.19179306030273438,
      "num_tokens": 43133514.0,
      "step": 23135
    },
    {
      "entropy": 5.83880648612976,
      "epoch": 2.0434475450370893,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004581586171608301,
      "loss": 5.5698,
      "mean_token_accuracy": 0.18596016764640808,
      "num_tokens": 43142082.0,
      "step": 23140
    },
    {
      "entropy": 5.796996259689331,
      "epoch": 2.0438890851289298,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004581403247474657,
      "loss": 5.522,
      "mean_token_accuracy": 0.18253479152917862,
      "num_tokens": 43151466.0,
      "step": 23145
    },
    {
      "entropy": 5.906253004074097,
      "epoch": 2.0443306252207702,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004581220287464864,
      "loss": 5.5937,
      "mean_token_accuracy": 0.1803050234913826,
      "num_tokens": 43160631.0,
      "step": 23150
    },
    {
      "entropy": 5.867755508422851,
      "epoch": 2.0447721653126103,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004581037291582507,
      "loss": 5.5754,
      "mean_token_accuracy": 0.1890497699379921,
      "num_tokens": 43170968.0,
      "step": 23155
    },
    {
      "entropy": 5.845655155181885,
      "epoch": 2.045213705404451,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004580854259831171,
      "loss": 5.4801,
      "mean_token_accuracy": 0.19253343641757964,
      "num_tokens": 43179661.0,
      "step": 23160
    },
    {
      "entropy": 5.777335453033447,
      "epoch": 2.045655245496291,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004580671192214441,
      "loss": 5.3243,
      "mean_token_accuracy": 0.20820141285657884,
      "num_tokens": 43188994.0,
      "step": 23165
    },
    {
      "entropy": 5.860503625869751,
      "epoch": 2.0460967855881314,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045804880887359036,
      "loss": 5.5964,
      "mean_token_accuracy": 0.179094335436821,
      "num_tokens": 43199001.0,
      "step": 23170
    },
    {
      "entropy": 5.837001991271973,
      "epoch": 2.046538325679972,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004580304949399146,
      "loss": 5.4576,
      "mean_token_accuracy": 0.19199587404727936,
      "num_tokens": 43208113.0,
      "step": 23175
    },
    {
      "entropy": 5.792837381362915,
      "epoch": 2.046979865771812,
      "grad_norm": 1.53125,
      "learning_rate": 0.00045801217742077544,
      "loss": 5.4674,
      "mean_token_accuracy": 0.1965992972254753,
      "num_tokens": 43216983.0,
      "step": 23180
    },
    {
      "entropy": 5.77666449546814,
      "epoch": 2.0474214058636524,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004579938563165319,
      "loss": 5.4545,
      "mean_token_accuracy": 0.19846420735120773,
      "num_tokens": 43225965.0,
      "step": 23185
    },
    {
      "entropy": 5.826109123229981,
      "epoch": 2.047862945955493,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004579755316275427,
      "loss": 5.3913,
      "mean_token_accuracy": 0.2007564499974251,
      "num_tokens": 43235078.0,
      "step": 23190
    },
    {
      "entropy": 5.797417116165161,
      "epoch": 2.048304486047333,
      "grad_norm": 1.34375,
      "learning_rate": 0.000457957203354167,
      "loss": 5.5157,
      "mean_token_accuracy": 0.19373500496149063,
      "num_tokens": 43244398.0,
      "step": 23195
    },
    {
      "entropy": 5.899709749221802,
      "epoch": 2.0487460261391734,
      "grad_norm": 1.375,
      "learning_rate": 0.00045793887149676366,
      "loss": 5.5442,
      "mean_token_accuracy": 0.18914303481578826,
      "num_tokens": 43254638.0,
      "step": 23200
    },
    {
      "entropy": 5.801394462585449,
      "epoch": 2.049187566231014,
      "grad_norm": 1.234375,
      "learning_rate": 0.000457920536055692,
      "loss": 5.4592,
      "mean_token_accuracy": 0.18810865879058838,
      "num_tokens": 43263287.0,
      "step": 23205
    },
    {
      "entropy": 5.751893711090088,
      "epoch": 2.049629106322854,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045790219703131097,
      "loss": 5.4015,
      "mean_token_accuracy": 0.19506020694971085,
      "num_tokens": 43272921.0,
      "step": 23210
    },
    {
      "entropy": 5.816206979751587,
      "epoch": 2.0500706464146945,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045788385442397994,
      "loss": 5.4773,
      "mean_token_accuracy": 0.19404996782541276,
      "num_tokens": 43283423.0,
      "step": 23215
    },
    {
      "entropy": 5.912694215774536,
      "epoch": 2.050512186506535,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004578655082340582,
      "loss": 5.5698,
      "mean_token_accuracy": 0.18758049011230468,
      "num_tokens": 43293031.0,
      "step": 23220
    },
    {
      "entropy": 5.845752477645874,
      "epoch": 2.050953726598375,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004578471584619051,
      "loss": 5.5266,
      "mean_token_accuracy": 0.1938727468252182,
      "num_tokens": 43302775.0,
      "step": 23225
    },
    {
      "entropy": 5.7124258518219,
      "epoch": 2.0513952666902155,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004578288051078803,
      "loss": 5.4443,
      "mean_token_accuracy": 0.19571835547685623,
      "num_tokens": 43312341.0,
      "step": 23230
    },
    {
      "entropy": 5.84419617652893,
      "epoch": 2.0518368067820556,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045781044817234305,
      "loss": 5.4836,
      "mean_token_accuracy": 0.18900103271007537,
      "num_tokens": 43321358.0,
      "step": 23235
    },
    {
      "entropy": 5.79281096458435,
      "epoch": 2.052278346873896,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045779208765565324,
      "loss": 5.4657,
      "mean_token_accuracy": 0.1897926613688469,
      "num_tokens": 43331839.0,
      "step": 23240
    },
    {
      "entropy": 5.871837997436524,
      "epoch": 2.0527198869657366,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004577737235581702,
      "loss": 5.5489,
      "mean_token_accuracy": 0.18492138981819153,
      "num_tokens": 43340384.0,
      "step": 23245
    },
    {
      "entropy": 5.961942052841186,
      "epoch": 2.0531614270575766,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045775535588025404,
      "loss": 5.5243,
      "mean_token_accuracy": 0.18939695358276368,
      "num_tokens": 43349264.0,
      "step": 23250
    },
    {
      "entropy": 5.816296052932739,
      "epoch": 2.053602967149417,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004577369846222643,
      "loss": 5.5139,
      "mean_token_accuracy": 0.1890059605240822,
      "num_tokens": 43358598.0,
      "step": 23255
    },
    {
      "entropy": 5.7959442138671875,
      "epoch": 2.0540445072412576,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004577186097845609,
      "loss": 5.4624,
      "mean_token_accuracy": 0.1879880115389824,
      "num_tokens": 43368357.0,
      "step": 23260
    },
    {
      "entropy": 5.935980987548828,
      "epoch": 2.0544860473330977,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045770023136750405,
      "loss": 5.5659,
      "mean_token_accuracy": 0.18948266953229903,
      "num_tokens": 43378143.0,
      "step": 23265
    },
    {
      "entropy": 5.804943561553955,
      "epoch": 2.054927587424938,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004576818493714534,
      "loss": 5.4644,
      "mean_token_accuracy": 0.19774741381406785,
      "num_tokens": 43387010.0,
      "step": 23270
    },
    {
      "entropy": 5.699764776229858,
      "epoch": 2.0553691275167787,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004576634637967693,
      "loss": 5.3501,
      "mean_token_accuracy": 0.19878350645303727,
      "num_tokens": 43395864.0,
      "step": 23275
    },
    {
      "entropy": 5.7939551830291744,
      "epoch": 2.0558106676086187,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004576450746438118,
      "loss": 5.4278,
      "mean_token_accuracy": 0.19400671124458313,
      "num_tokens": 43404662.0,
      "step": 23280
    },
    {
      "entropy": 5.817852783203125,
      "epoch": 2.0562522077004592,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004576266819129412,
      "loss": 5.48,
      "mean_token_accuracy": 0.18671859800815582,
      "num_tokens": 43413927.0,
      "step": 23285
    },
    {
      "entropy": 5.885161828994751,
      "epoch": 2.0566937477922997,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004576082856045177,
      "loss": 5.6346,
      "mean_token_accuracy": 0.17245729118585587,
      "num_tokens": 43424197.0,
      "step": 23290
    },
    {
      "entropy": 5.924931478500366,
      "epoch": 2.05713528788414,
      "grad_norm": 1.484375,
      "learning_rate": 0.00045758988571890176,
      "loss": 5.5216,
      "mean_token_accuracy": 0.19060292094945908,
      "num_tokens": 43432630.0,
      "step": 23295
    },
    {
      "entropy": 5.805805349349976,
      "epoch": 2.0575768279759803,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045757148225645383,
      "loss": 5.5015,
      "mean_token_accuracy": 0.18768813014030455,
      "num_tokens": 43442186.0,
      "step": 23300
    },
    {
      "entropy": 5.7738290309906,
      "epoch": 2.0580183680678203,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00045755307521753443,
      "loss": 5.4528,
      "mean_token_accuracy": 0.1994573652744293,
      "num_tokens": 43451927.0,
      "step": 23305
    },
    {
      "entropy": 5.8813951969146725,
      "epoch": 2.058459908159661,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045753466460250405,
      "loss": 5.4852,
      "mean_token_accuracy": 0.1878998175263405,
      "num_tokens": 43460879.0,
      "step": 23310
    },
    {
      "entropy": 5.833441352844238,
      "epoch": 2.0589014482515013,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004575162504117234,
      "loss": 5.5775,
      "mean_token_accuracy": 0.17835736721754075,
      "num_tokens": 43469918.0,
      "step": 23315
    },
    {
      "entropy": 5.834415578842163,
      "epoch": 2.0593429883433414,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004574978326455532,
      "loss": 5.4888,
      "mean_token_accuracy": 0.19142780154943467,
      "num_tokens": 43478779.0,
      "step": 23320
    },
    {
      "entropy": 5.847833061218262,
      "epoch": 2.059784528435182,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004574794113043543,
      "loss": 5.5618,
      "mean_token_accuracy": 0.19210575371980668,
      "num_tokens": 43488128.0,
      "step": 23325
    },
    {
      "entropy": 5.810982799530029,
      "epoch": 2.0602260685270224,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004574609863884875,
      "loss": 5.4812,
      "mean_token_accuracy": 0.1880495071411133,
      "num_tokens": 43497567.0,
      "step": 23330
    },
    {
      "entropy": 5.856006336212158,
      "epoch": 2.0606676086188624,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045744255789831377,
      "loss": 5.5134,
      "mean_token_accuracy": 0.18637235462665558,
      "num_tokens": 43507360.0,
      "step": 23335
    },
    {
      "entropy": 5.833869552612304,
      "epoch": 2.061109148710703,
      "grad_norm": 1.59375,
      "learning_rate": 0.00045742412583419403,
      "loss": 5.5116,
      "mean_token_accuracy": 0.18985875695943832,
      "num_tokens": 43517601.0,
      "step": 23340
    },
    {
      "entropy": 5.79422516822815,
      "epoch": 2.0615506888025434,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045740569019648945,
      "loss": 5.4165,
      "mean_token_accuracy": 0.2012377843260765,
      "num_tokens": 43526348.0,
      "step": 23345
    },
    {
      "entropy": 5.813654851913452,
      "epoch": 2.0619922288943835,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004573872509855612,
      "loss": 5.4528,
      "mean_token_accuracy": 0.19668517112731934,
      "num_tokens": 43536106.0,
      "step": 23350
    },
    {
      "entropy": 5.902621507644653,
      "epoch": 2.062433768986224,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004573688082017703,
      "loss": 5.5873,
      "mean_token_accuracy": 0.1810501292347908,
      "num_tokens": 43545110.0,
      "step": 23355
    },
    {
      "entropy": 5.822573709487915,
      "epoch": 2.0628753090780645,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045735036184547825,
      "loss": 5.4873,
      "mean_token_accuracy": 0.18644196391105652,
      "num_tokens": 43554226.0,
      "step": 23360
    },
    {
      "entropy": 5.781835269927979,
      "epoch": 2.0633168491699045,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045733191191704633,
      "loss": 5.4984,
      "mean_token_accuracy": 0.19133814573287963,
      "num_tokens": 43563184.0,
      "step": 23365
    },
    {
      "entropy": 5.834351921081543,
      "epoch": 2.063758389261745,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004573134584168359,
      "loss": 5.4754,
      "mean_token_accuracy": 0.20317726135253905,
      "num_tokens": 43572595.0,
      "step": 23370
    },
    {
      "entropy": 5.8380004405975345,
      "epoch": 2.064199929353585,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004572950013452086,
      "loss": 5.5478,
      "mean_token_accuracy": 0.1834607407450676,
      "num_tokens": 43582211.0,
      "step": 23375
    },
    {
      "entropy": 5.778594827651977,
      "epoch": 2.0646414694454256,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004572765407025257,
      "loss": 5.4192,
      "mean_token_accuracy": 0.19411822110414506,
      "num_tokens": 43591060.0,
      "step": 23380
    },
    {
      "entropy": 5.907498502731324,
      "epoch": 2.065083009537266,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004572580764891493,
      "loss": 5.6069,
      "mean_token_accuracy": 0.1840766966342926,
      "num_tokens": 43600057.0,
      "step": 23385
    },
    {
      "entropy": 5.900477266311645,
      "epoch": 2.065524549629106,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004572396087054406,
      "loss": 5.4037,
      "mean_token_accuracy": 0.1968561142683029,
      "num_tokens": 43608609.0,
      "step": 23390
    },
    {
      "entropy": 5.774201202392578,
      "epoch": 2.0659660897209466,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004572211373517618,
      "loss": 5.5271,
      "mean_token_accuracy": 0.19169795215129853,
      "num_tokens": 43618751.0,
      "step": 23395
    },
    {
      "entropy": 5.881857776641846,
      "epoch": 2.066407629812787,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00045720266242847446,
      "loss": 5.6057,
      "mean_token_accuracy": 0.18803844153881072,
      "num_tokens": 43628934.0,
      "step": 23400
    },
    {
      "entropy": 5.939468908309936,
      "epoch": 2.066849169904627,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045718418393594055,
      "loss": 5.5144,
      "mean_token_accuracy": 0.18798298686742781,
      "num_tokens": 43638512.0,
      "step": 23405
    },
    {
      "entropy": 5.7612076759338375,
      "epoch": 2.0672907099964677,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045716570187452214,
      "loss": 5.4494,
      "mean_token_accuracy": 0.19146936982870102,
      "num_tokens": 43647708.0,
      "step": 23410
    },
    {
      "entropy": 5.821790742874145,
      "epoch": 2.067732250088308,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004571472162445813,
      "loss": 5.5276,
      "mean_token_accuracy": 0.1885530099272728,
      "num_tokens": 43657096.0,
      "step": 23415
    },
    {
      "entropy": 5.85364179611206,
      "epoch": 2.0681737901801482,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00045712872704647994,
      "loss": 5.5251,
      "mean_token_accuracy": 0.1849769815802574,
      "num_tokens": 43667410.0,
      "step": 23420
    },
    {
      "entropy": 5.805835485458374,
      "epoch": 2.0686153302719887,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004571102342805805,
      "loss": 5.4781,
      "mean_token_accuracy": 0.19379069209098815,
      "num_tokens": 43675800.0,
      "step": 23425
    },
    {
      "entropy": 5.830232191085815,
      "epoch": 2.0690568703638292,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004570917379472451,
      "loss": 5.4869,
      "mean_token_accuracy": 0.1983436241745949,
      "num_tokens": 43684802.0,
      "step": 23430
    },
    {
      "entropy": 5.831229543685913,
      "epoch": 2.0694984104556693,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00045707323804683616,
      "loss": 5.4817,
      "mean_token_accuracy": 0.19437631368637084,
      "num_tokens": 43693968.0,
      "step": 23435
    },
    {
      "entropy": 5.944815254211425,
      "epoch": 2.06993995054751,
      "grad_norm": 1.671875,
      "learning_rate": 0.000457054734579716,
      "loss": 5.5026,
      "mean_token_accuracy": 0.18232344537973405,
      "num_tokens": 43703177.0,
      "step": 23440
    },
    {
      "entropy": 5.797865724563598,
      "epoch": 2.07038149063935,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00045703622754624716,
      "loss": 5.5218,
      "mean_token_accuracy": 0.19008992463350297,
      "num_tokens": 43712894.0,
      "step": 23445
    },
    {
      "entropy": 5.802890872955322,
      "epoch": 2.0708230307311903,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00045701771694679213,
      "loss": 5.4122,
      "mean_token_accuracy": 0.1940461128950119,
      "num_tokens": 43722591.0,
      "step": 23450
    },
    {
      "entropy": 5.760197639465332,
      "epoch": 2.071264570823031,
      "grad_norm": 1.765625,
      "learning_rate": 0.00045699920278171355,
      "loss": 5.4219,
      "mean_token_accuracy": 0.20174515396356582,
      "num_tokens": 43732251.0,
      "step": 23455
    },
    {
      "entropy": 5.925807142257691,
      "epoch": 2.071706110914871,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004569806850513741,
      "loss": 5.5796,
      "mean_token_accuracy": 0.1862985149025917,
      "num_tokens": 43741061.0,
      "step": 23460
    },
    {
      "entropy": 5.909265184402466,
      "epoch": 2.0721476510067114,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045696216375613647,
      "loss": 5.5696,
      "mean_token_accuracy": 0.17900942713022233,
      "num_tokens": 43750147.0,
      "step": 23465
    },
    {
      "entropy": 5.861504936218262,
      "epoch": 2.072589191098552,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004569436388963636,
      "loss": 5.5783,
      "mean_token_accuracy": 0.17887588441371918,
      "num_tokens": 43760960.0,
      "step": 23470
    },
    {
      "entropy": 5.824364042282104,
      "epoch": 2.073030731190392,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004569251104724184,
      "loss": 5.5101,
      "mean_token_accuracy": 0.19029182493686675,
      "num_tokens": 43771306.0,
      "step": 23475
    },
    {
      "entropy": 5.777688217163086,
      "epoch": 2.0734722712822324,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004569065784846636,
      "loss": 5.4084,
      "mean_token_accuracy": 0.19898889809846879,
      "num_tokens": 43780520.0,
      "step": 23480
    },
    {
      "entropy": 5.86636061668396,
      "epoch": 2.073913811374073,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004568880429334624,
      "loss": 5.539,
      "mean_token_accuracy": 0.1897300124168396,
      "num_tokens": 43789472.0,
      "step": 23485
    },
    {
      "entropy": 5.863546419143677,
      "epoch": 2.074355351465913,
      "grad_norm": 1.421875,
      "learning_rate": 0.00045686950381917805,
      "loss": 5.5358,
      "mean_token_accuracy": 0.18233954310417175,
      "num_tokens": 43798728.0,
      "step": 23490
    },
    {
      "entropy": 5.828519105911255,
      "epoch": 2.0747968915577535,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045685096114217336,
      "loss": 5.4498,
      "mean_token_accuracy": 0.1936558037996292,
      "num_tokens": 43807959.0,
      "step": 23495
    },
    {
      "entropy": 5.840868949890137,
      "epoch": 2.075238431649594,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004568324149028119,
      "loss": 5.4647,
      "mean_token_accuracy": 0.19330984354019165,
      "num_tokens": 43816922.0,
      "step": 23500
    },
    {
      "entropy": 5.802131605148316,
      "epoch": 2.075679971741434,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004568138651014567,
      "loss": 5.472,
      "mean_token_accuracy": 0.19077959805727004,
      "num_tokens": 43827195.0,
      "step": 23505
    },
    {
      "entropy": 5.779759359359741,
      "epoch": 2.0761215118332745,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004567953117384714,
      "loss": 5.4323,
      "mean_token_accuracy": 0.18778088986873626,
      "num_tokens": 43836289.0,
      "step": 23510
    },
    {
      "entropy": 5.881022357940674,
      "epoch": 2.0765630519251146,
      "grad_norm": 1.5,
      "learning_rate": 0.0004567767548142193,
      "loss": 5.5416,
      "mean_token_accuracy": 0.18635546863079072,
      "num_tokens": 43845918.0,
      "step": 23515
    },
    {
      "entropy": 5.867420244216919,
      "epoch": 2.077004592016955,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045675819432906394,
      "loss": 5.5737,
      "mean_token_accuracy": 0.18462501466274261,
      "num_tokens": 43853907.0,
      "step": 23520
    },
    {
      "entropy": 5.843808603286743,
      "epoch": 2.0774461321087956,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045673963028336896,
      "loss": 5.4905,
      "mean_token_accuracy": 0.19451006948947908,
      "num_tokens": 43862992.0,
      "step": 23525
    },
    {
      "entropy": 5.9186375617980955,
      "epoch": 2.0778876722006356,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045672106267749795,
      "loss": 5.5189,
      "mean_token_accuracy": 0.18590945303440093,
      "num_tokens": 43871686.0,
      "step": 23530
    },
    {
      "entropy": 5.771477699279785,
      "epoch": 2.078329212292476,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004567024915118148,
      "loss": 5.4989,
      "mean_token_accuracy": 0.18971027284860612,
      "num_tokens": 43881302.0,
      "step": 23535
    },
    {
      "entropy": 5.8828541278839115,
      "epoch": 2.0787707523843166,
      "grad_norm": 1.703125,
      "learning_rate": 0.00045668391678668296,
      "loss": 5.5828,
      "mean_token_accuracy": 0.1819699823856354,
      "num_tokens": 43889982.0,
      "step": 23540
    },
    {
      "entropy": 5.886616230010986,
      "epoch": 2.0792122924761567,
      "grad_norm": 2.546875,
      "learning_rate": 0.00045666533850246665,
      "loss": 5.522,
      "mean_token_accuracy": 0.188995298743248,
      "num_tokens": 43898385.0,
      "step": 23545
    },
    {
      "entropy": 5.816378545761109,
      "epoch": 2.079653832567997,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004566467566595297,
      "loss": 5.4892,
      "mean_token_accuracy": 0.1814259946346283,
      "num_tokens": 43907882.0,
      "step": 23550
    },
    {
      "entropy": 5.933834743499756,
      "epoch": 2.0800953726598377,
      "grad_norm": 1.796875,
      "learning_rate": 0.00045662817125823605,
      "loss": 5.5271,
      "mean_token_accuracy": 0.18953393250703812,
      "num_tokens": 43916222.0,
      "step": 23555
    },
    {
      "entropy": 5.883642911911011,
      "epoch": 2.0805369127516777,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045660958229894985,
      "loss": 5.4477,
      "mean_token_accuracy": 0.19311333000659942,
      "num_tokens": 43925181.0,
      "step": 23560
    },
    {
      "entropy": 5.704629278182983,
      "epoch": 2.0809784528435182,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00045659098978203524,
      "loss": 5.503,
      "mean_token_accuracy": 0.1868574559688568,
      "num_tokens": 43933973.0,
      "step": 23565
    },
    {
      "entropy": 5.729162740707397,
      "epoch": 2.0814199929353587,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004565723937078563,
      "loss": 5.4668,
      "mean_token_accuracy": 0.19088664501905442,
      "num_tokens": 43943297.0,
      "step": 23570
    },
    {
      "entropy": 5.912590265274048,
      "epoch": 2.081861533027199,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045655379407677754,
      "loss": 5.5257,
      "mean_token_accuracy": 0.19417939931154252,
      "num_tokens": 43952679.0,
      "step": 23575
    },
    {
      "entropy": 5.851170301437378,
      "epoch": 2.0823030731190393,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004565351908891632,
      "loss": 5.4868,
      "mean_token_accuracy": 0.18757954686880113,
      "num_tokens": 43961629.0,
      "step": 23580
    },
    {
      "entropy": 5.79420690536499,
      "epoch": 2.0827446132108793,
      "grad_norm": 1.671875,
      "learning_rate": 0.00045651658414537765,
      "loss": 5.5232,
      "mean_token_accuracy": 0.19245149940252304,
      "num_tokens": 43970880.0,
      "step": 23585
    },
    {
      "entropy": 5.9122278690338135,
      "epoch": 2.08318615330272,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004564979738457855,
      "loss": 5.4518,
      "mean_token_accuracy": 0.19733261913061143,
      "num_tokens": 43980449.0,
      "step": 23590
    },
    {
      "entropy": 5.842273235321045,
      "epoch": 2.0836276933945603,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00045647935999075127,
      "loss": 5.4804,
      "mean_token_accuracy": 0.18452073633670807,
      "num_tokens": 43988716.0,
      "step": 23595
    },
    {
      "entropy": 5.824282503128051,
      "epoch": 2.0840692334864004,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00045646074258063956,
      "loss": 5.5242,
      "mean_token_accuracy": 0.19429346919059753,
      "num_tokens": 43998464.0,
      "step": 23600
    },
    {
      "entropy": 5.835815525054931,
      "epoch": 2.084510773578241,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004564421216158152,
      "loss": 5.5075,
      "mean_token_accuracy": 0.19346749782562256,
      "num_tokens": 44006993.0,
      "step": 23605
    },
    {
      "entropy": 5.885981178283691,
      "epoch": 2.0849523136700814,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004564234970966428,
      "loss": 5.5453,
      "mean_token_accuracy": 0.18467033356428147,
      "num_tokens": 44015801.0,
      "step": 23610
    },
    {
      "entropy": 5.771956205368042,
      "epoch": 2.0853938537619214,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004564048690234872,
      "loss": 5.4865,
      "mean_token_accuracy": 0.18435074239969254,
      "num_tokens": 44025069.0,
      "step": 23615
    },
    {
      "entropy": 5.927738237380981,
      "epoch": 2.085835393853762,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045638623739671343,
      "loss": 5.6262,
      "mean_token_accuracy": 0.17667030096054076,
      "num_tokens": 44034700.0,
      "step": 23620
    },
    {
      "entropy": 5.884667873382568,
      "epoch": 2.0862769339456024,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00045636760221668646,
      "loss": 5.4423,
      "mean_token_accuracy": 0.19094647318124772,
      "num_tokens": 44044417.0,
      "step": 23625
    },
    {
      "entropy": 5.831897735595703,
      "epoch": 2.0867184740374425,
      "grad_norm": 1.25,
      "learning_rate": 0.0004563489634837713,
      "loss": 5.5031,
      "mean_token_accuracy": 0.1882332295179367,
      "num_tokens": 44053617.0,
      "step": 23630
    },
    {
      "entropy": 5.797045087814331,
      "epoch": 2.087160014129283,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045633032119833314,
      "loss": 5.4617,
      "mean_token_accuracy": 0.18652137219905854,
      "num_tokens": 44062738.0,
      "step": 23635
    },
    {
      "entropy": 5.838134145736694,
      "epoch": 2.0876015542211235,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000456311675360737,
      "loss": 5.5291,
      "mean_token_accuracy": 0.18335009664297103,
      "num_tokens": 44072154.0,
      "step": 23640
    },
    {
      "entropy": 5.932612752914428,
      "epoch": 2.0880430943129635,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004562930259713483,
      "loss": 5.6033,
      "mean_token_accuracy": 0.18032784014940262,
      "num_tokens": 44081239.0,
      "step": 23645
    },
    {
      "entropy": 5.830880451202392,
      "epoch": 2.088484634404804,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045627437303053236,
      "loss": 5.4744,
      "mean_token_accuracy": 0.19371165931224824,
      "num_tokens": 44090866.0,
      "step": 23650
    },
    {
      "entropy": 5.8248772621154785,
      "epoch": 2.088926174496644,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004562557165386546,
      "loss": 5.5781,
      "mean_token_accuracy": 0.18577031195163726,
      "num_tokens": 44100225.0,
      "step": 23655
    },
    {
      "entropy": 5.8547426700592045,
      "epoch": 2.0893677145884846,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004562370564960804,
      "loss": 5.5096,
      "mean_token_accuracy": 0.19123829156160355,
      "num_tokens": 44110139.0,
      "step": 23660
    },
    {
      "entropy": 5.832539129257202,
      "epoch": 2.089809254680325,
      "grad_norm": 1.546875,
      "learning_rate": 0.00045621839290317526,
      "loss": 5.5049,
      "mean_token_accuracy": 0.1937221333384514,
      "num_tokens": 44118687.0,
      "step": 23665
    },
    {
      "entropy": 5.781908988952637,
      "epoch": 2.090250794772165,
      "grad_norm": 1.421875,
      "learning_rate": 0.00045619972576030495,
      "loss": 5.4939,
      "mean_token_accuracy": 0.18959296345710755,
      "num_tokens": 44127377.0,
      "step": 23670
    },
    {
      "entropy": 5.781782102584839,
      "epoch": 2.0906923348640056,
      "grad_norm": 1.234375,
      "learning_rate": 0.000456181055067835,
      "loss": 5.5242,
      "mean_token_accuracy": 0.19189699441194535,
      "num_tokens": 44136583.0,
      "step": 23675
    },
    {
      "entropy": 5.824638414382934,
      "epoch": 2.091133874955846,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004561623808261313,
      "loss": 5.5024,
      "mean_token_accuracy": 0.18895828872919082,
      "num_tokens": 44145467.0,
      "step": 23680
    },
    {
      "entropy": 5.849105596542358,
      "epoch": 2.091575415047686,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004561437030355595,
      "loss": 5.4736,
      "mean_token_accuracy": 0.194976282119751,
      "num_tokens": 44154932.0,
      "step": 23685
    },
    {
      "entropy": 5.806014537811279,
      "epoch": 2.0920169551395267,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004561250216964857,
      "loss": 5.4225,
      "mean_token_accuracy": 0.1914547026157379,
      "num_tokens": 44162791.0,
      "step": 23690
    },
    {
      "entropy": 5.8623411655426025,
      "epoch": 2.092458495231367,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004561063368092757,
      "loss": 5.536,
      "mean_token_accuracy": 0.19175465106964112,
      "num_tokens": 44171993.0,
      "step": 23695
    },
    {
      "entropy": 5.8406270980834964,
      "epoch": 2.0929000353232072,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045608764837429543,
      "loss": 5.5412,
      "mean_token_accuracy": 0.1846216380596161,
      "num_tokens": 44180807.0,
      "step": 23700
    },
    {
      "entropy": 5.934008169174194,
      "epoch": 2.0933415754150477,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045606895639191126,
      "loss": 5.594,
      "mean_token_accuracy": 0.18228478878736495,
      "num_tokens": 44191764.0,
      "step": 23705
    },
    {
      "entropy": 5.866354370117188,
      "epoch": 2.0937831155068882,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045605026086248916,
      "loss": 5.5495,
      "mean_token_accuracy": 0.18812790811061858,
      "num_tokens": 44201165.0,
      "step": 23710
    },
    {
      "entropy": 5.906619882583618,
      "epoch": 2.0942246555987283,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004560315617863954,
      "loss": 5.5613,
      "mean_token_accuracy": 0.185288305580616,
      "num_tokens": 44210709.0,
      "step": 23715
    },
    {
      "entropy": 5.86715145111084,
      "epoch": 2.0946661956905688,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004560128591639964,
      "loss": 5.5245,
      "mean_token_accuracy": 0.18567841351032258,
      "num_tokens": 44219827.0,
      "step": 23720
    },
    {
      "entropy": 5.865751791000366,
      "epoch": 2.095107735782409,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004559941529956584,
      "loss": 5.5671,
      "mean_token_accuracy": 0.18254009932279586,
      "num_tokens": 44229498.0,
      "step": 23725
    },
    {
      "entropy": 5.824496269226074,
      "epoch": 2.0955492758742493,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004559754432817479,
      "loss": 5.4794,
      "mean_token_accuracy": 0.1899287909269333,
      "num_tokens": 44238844.0,
      "step": 23730
    },
    {
      "entropy": 5.839661312103272,
      "epoch": 2.09599081596609,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045595673002263125,
      "loss": 5.5,
      "mean_token_accuracy": 0.18952344208955765,
      "num_tokens": 44247512.0,
      "step": 23735
    },
    {
      "entropy": 5.845702123641968,
      "epoch": 2.09643235605793,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004559380132186753,
      "loss": 5.4354,
      "mean_token_accuracy": 0.1971040740609169,
      "num_tokens": 44256485.0,
      "step": 23740
    },
    {
      "entropy": 5.719971799850464,
      "epoch": 2.0968738961497704,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045591929287024655,
      "loss": 5.3349,
      "mean_token_accuracy": 0.20143299102783202,
      "num_tokens": 44265095.0,
      "step": 23745
    },
    {
      "entropy": 5.882918167114258,
      "epoch": 2.097315436241611,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004559005689777117,
      "loss": 5.5118,
      "mean_token_accuracy": 0.18531536161899567,
      "num_tokens": 44274582.0,
      "step": 23750
    },
    {
      "entropy": 5.939994049072266,
      "epoch": 2.097756976333451,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004558818415414377,
      "loss": 5.5606,
      "mean_token_accuracy": 0.18787501007318497,
      "num_tokens": 44285752.0,
      "step": 23755
    },
    {
      "entropy": 5.909236907958984,
      "epoch": 2.0981985164252914,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004558631105617912,
      "loss": 5.5504,
      "mean_token_accuracy": 0.1837774321436882,
      "num_tokens": 44295315.0,
      "step": 23760
    },
    {
      "entropy": 5.834135484695435,
      "epoch": 2.098640056517132,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045584437603913923,
      "loss": 5.5915,
      "mean_token_accuracy": 0.18536292165517806,
      "num_tokens": 44305151.0,
      "step": 23765
    },
    {
      "entropy": 5.835593223571777,
      "epoch": 2.099081596608972,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045582563797384886,
      "loss": 5.5682,
      "mean_token_accuracy": 0.18494853973388672,
      "num_tokens": 44314462.0,
      "step": 23770
    },
    {
      "entropy": 5.821321153640747,
      "epoch": 2.0995231367008125,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000455806896366287,
      "loss": 5.3682,
      "mean_token_accuracy": 0.2100304663181305,
      "num_tokens": 44322782.0,
      "step": 23775
    },
    {
      "entropy": 5.806733226776123,
      "epoch": 2.099964676792653,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004557881512168209,
      "loss": 5.4344,
      "mean_token_accuracy": 0.19457061886787413,
      "num_tokens": 44331588.0,
      "step": 23780
    },
    {
      "entropy": 5.785691022872925,
      "epoch": 2.100406216884493,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004557694025258177,
      "loss": 5.4739,
      "mean_token_accuracy": 0.19819646030664445,
      "num_tokens": 44340997.0,
      "step": 23785
    },
    {
      "entropy": 5.813557720184326,
      "epoch": 2.1008477569763335,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004557506502936448,
      "loss": 5.5265,
      "mean_token_accuracy": 0.18558417409658431,
      "num_tokens": 44350679.0,
      "step": 23790
    },
    {
      "entropy": 5.821036767959595,
      "epoch": 2.1012892970681736,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045573189452066936,
      "loss": 5.4504,
      "mean_token_accuracy": 0.19006657898426055,
      "num_tokens": 44359934.0,
      "step": 23795
    },
    {
      "entropy": 5.828332376480103,
      "epoch": 2.101730837160014,
      "grad_norm": 2.546875,
      "learning_rate": 0.00045571313520725897,
      "loss": 5.4902,
      "mean_token_accuracy": 0.1863226369023323,
      "num_tokens": 44369104.0,
      "step": 23800
    },
    {
      "entropy": 5.774208927154541,
      "epoch": 2.1021723772518546,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00045569437235378103,
      "loss": 5.4812,
      "mean_token_accuracy": 0.18889510929584502,
      "num_tokens": 44379645.0,
      "step": 23805
    },
    {
      "entropy": 5.880788135528564,
      "epoch": 2.1026139173436946,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004556756059606031,
      "loss": 5.5395,
      "mean_token_accuracy": 0.1904294401407242,
      "num_tokens": 44389276.0,
      "step": 23810
    },
    {
      "entropy": 5.868474674224854,
      "epoch": 2.103055457435535,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004556568360280928,
      "loss": 5.5001,
      "mean_token_accuracy": 0.18494162559509278,
      "num_tokens": 44399114.0,
      "step": 23815
    },
    {
      "entropy": 5.649212455749511,
      "epoch": 2.1034969975273756,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00045563806255661784,
      "loss": 5.2371,
      "mean_token_accuracy": 0.21943083703517913,
      "num_tokens": 44406917.0,
      "step": 23820
    },
    {
      "entropy": 5.8273571014404295,
      "epoch": 2.1039385376192157,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004556192855465459,
      "loss": 5.5079,
      "mean_token_accuracy": 0.1829274833202362,
      "num_tokens": 44416678.0,
      "step": 23825
    },
    {
      "entropy": 5.748898935317993,
      "epoch": 2.104380077711056,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004556005049982449,
      "loss": 5.4112,
      "mean_token_accuracy": 0.1965800255537033,
      "num_tokens": 44425427.0,
      "step": 23830
    },
    {
      "entropy": 5.899160957336425,
      "epoch": 2.1048216178028967,
      "grad_norm": 1.421875,
      "learning_rate": 0.00045558172091208284,
      "loss": 5.4846,
      "mean_token_accuracy": 0.18587032407522203,
      "num_tokens": 44434267.0,
      "step": 23835
    },
    {
      "entropy": 5.871176528930664,
      "epoch": 2.1052631578947367,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00045556293328842746,
      "loss": 5.4996,
      "mean_token_accuracy": 0.19470616132020951,
      "num_tokens": 44444037.0,
      "step": 23840
    },
    {
      "entropy": 5.820915269851684,
      "epoch": 2.1057046979865772,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004555441421276469,
      "loss": 5.4496,
      "mean_token_accuracy": 0.19403326511383057,
      "num_tokens": 44453707.0,
      "step": 23845
    },
    {
      "entropy": 5.823236560821533,
      "epoch": 2.1061462380784177,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045552534743010934,
      "loss": 5.4708,
      "mean_token_accuracy": 0.18922748863697053,
      "num_tokens": 44464357.0,
      "step": 23850
    },
    {
      "entropy": 5.882349967956543,
      "epoch": 2.1065877781702578,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045550654919618283,
      "loss": 5.5708,
      "mean_token_accuracy": 0.18147553354501725,
      "num_tokens": 44473803.0,
      "step": 23855
    },
    {
      "entropy": 5.880772638320923,
      "epoch": 2.1070293182620983,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004554877474262357,
      "loss": 5.5123,
      "mean_token_accuracy": 0.1919246256351471,
      "num_tokens": 44482830.0,
      "step": 23860
    },
    {
      "entropy": 5.873776578903199,
      "epoch": 2.1074708583539383,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004554689421206362,
      "loss": 5.4854,
      "mean_token_accuracy": 0.1912272348999977,
      "num_tokens": 44493045.0,
      "step": 23865
    },
    {
      "entropy": 5.825336074829101,
      "epoch": 2.107912398445779,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004554501332797529,
      "loss": 5.4671,
      "mean_token_accuracy": 0.1979750797152519,
      "num_tokens": 44503304.0,
      "step": 23870
    },
    {
      "entropy": 5.8808599472045895,
      "epoch": 2.1083539385376193,
      "grad_norm": 1.765625,
      "learning_rate": 0.00045543132090395403,
      "loss": 5.5591,
      "mean_token_accuracy": 0.18794138878583908,
      "num_tokens": 44512337.0,
      "step": 23875
    },
    {
      "entropy": 5.842969512939453,
      "epoch": 2.1087954786294594,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004554125049936082,
      "loss": 5.4916,
      "mean_token_accuracy": 0.19451663196086882,
      "num_tokens": 44521881.0,
      "step": 23880
    },
    {
      "entropy": 5.807424831390381,
      "epoch": 2.1092370187213,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000455393685549084,
      "loss": 5.5223,
      "mean_token_accuracy": 0.19130836576223373,
      "num_tokens": 44531351.0,
      "step": 23885
    },
    {
      "entropy": 5.880102729797363,
      "epoch": 2.1096785588131404,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004553748625707501,
      "loss": 5.5451,
      "mean_token_accuracy": 0.1914011150598526,
      "num_tokens": 44540592.0,
      "step": 23890
    },
    {
      "entropy": 5.841157531738281,
      "epoch": 2.1101200989049804,
      "grad_norm": 1.421875,
      "learning_rate": 0.00045535603605897516,
      "loss": 5.4812,
      "mean_token_accuracy": 0.19782116413116455,
      "num_tokens": 44550019.0,
      "step": 23895
    },
    {
      "entropy": 5.770502519607544,
      "epoch": 2.110561638996821,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004553372060141282,
      "loss": 5.4717,
      "mean_token_accuracy": 0.18889374732971193,
      "num_tokens": 44559423.0,
      "step": 23900
    },
    {
      "entropy": 5.77593297958374,
      "epoch": 2.1110031790886614,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004553183724365778,
      "loss": 5.4693,
      "mean_token_accuracy": 0.19716155529022217,
      "num_tokens": 44568792.0,
      "step": 23905
    },
    {
      "entropy": 5.861880970001221,
      "epoch": 2.1114447191805015,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004552995353266931,
      "loss": 5.4974,
      "mean_token_accuracy": 0.1876992702484131,
      "num_tokens": 44578191.0,
      "step": 23910
    },
    {
      "entropy": 5.857769632339478,
      "epoch": 2.111886259272342,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004552806946848431,
      "loss": 5.4506,
      "mean_token_accuracy": 0.1942305862903595,
      "num_tokens": 44587402.0,
      "step": 23915
    },
    {
      "entropy": 5.882945013046265,
      "epoch": 2.1123277993641825,
      "grad_norm": 1.5,
      "learning_rate": 0.00045526185051139676,
      "loss": 5.595,
      "mean_token_accuracy": 0.17564960122108458,
      "num_tokens": 44596241.0,
      "step": 23920
    },
    {
      "entropy": 5.8343367099761965,
      "epoch": 2.1127693394560225,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00045524300280672326,
      "loss": 5.3876,
      "mean_token_accuracy": 0.1953383356332779,
      "num_tokens": 44604796.0,
      "step": 23925
    },
    {
      "entropy": 5.8684093952178955,
      "epoch": 2.113210879547863,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045522415157119185,
      "loss": 5.5479,
      "mean_token_accuracy": 0.18495705872774124,
      "num_tokens": 44615208.0,
      "step": 23930
    },
    {
      "entropy": 5.843853187561035,
      "epoch": 2.113652419639703,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004552052968051719,
      "loss": 5.4998,
      "mean_token_accuracy": 0.19595929682254792,
      "num_tokens": 44623937.0,
      "step": 23935
    },
    {
      "entropy": 5.824685049057007,
      "epoch": 2.1140939597315436,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004551864385090326,
      "loss": 5.4734,
      "mean_token_accuracy": 0.19011862725019454,
      "num_tokens": 44634116.0,
      "step": 23940
    },
    {
      "entropy": 5.900545501708985,
      "epoch": 2.114535499823384,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004551675766831434,
      "loss": 5.5036,
      "mean_token_accuracy": 0.18802441358566285,
      "num_tokens": 44643436.0,
      "step": 23945
    },
    {
      "entropy": 5.872172021865845,
      "epoch": 2.114977039915224,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00045514871132787394,
      "loss": 5.5466,
      "mean_token_accuracy": 0.1909226730465889,
      "num_tokens": 44654077.0,
      "step": 23950
    },
    {
      "entropy": 5.863510036468506,
      "epoch": 2.1154185800070646,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045512984244359364,
      "loss": 5.5148,
      "mean_token_accuracy": 0.19068895876407624,
      "num_tokens": 44663785.0,
      "step": 23955
    },
    {
      "entropy": 5.7422069072723385,
      "epoch": 2.115860120098905,
      "grad_norm": 4.21875,
      "learning_rate": 0.00045511097003067216,
      "loss": 5.362,
      "mean_token_accuracy": 0.20457408726215362,
      "num_tokens": 44672750.0,
      "step": 23960
    },
    {
      "entropy": 5.8131671905517575,
      "epoch": 2.116301660190745,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045509209408947914,
      "loss": 5.4471,
      "mean_token_accuracy": 0.19683307111263276,
      "num_tokens": 44681814.0,
      "step": 23965
    },
    {
      "entropy": 5.8329836368560795,
      "epoch": 2.1167432002825857,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004550732146203845,
      "loss": 5.4713,
      "mean_token_accuracy": 0.19108713716268538,
      "num_tokens": 44690926.0,
      "step": 23970
    },
    {
      "entropy": 5.91849913597107,
      "epoch": 2.117184740374426,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00045505433162375806,
      "loss": 5.6187,
      "mean_token_accuracy": 0.17896629720926285,
      "num_tokens": 44700625.0,
      "step": 23975
    },
    {
      "entropy": 5.889288902282715,
      "epoch": 2.117626280466266,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004550354450999696,
      "loss": 5.5186,
      "mean_token_accuracy": 0.1899600550532341,
      "num_tokens": 44710703.0,
      "step": 23980
    },
    {
      "entropy": 5.824130296707153,
      "epoch": 2.1180678205581067,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004550165550493891,
      "loss": 5.4433,
      "mean_token_accuracy": 0.19549671262502671,
      "num_tokens": 44719412.0,
      "step": 23985
    },
    {
      "entropy": 5.851120805740356,
      "epoch": 2.118509360649947,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045499766147238675,
      "loss": 5.563,
      "mean_token_accuracy": 0.19024336338043213,
      "num_tokens": 44729161.0,
      "step": 23990
    },
    {
      "entropy": 5.91692156791687,
      "epoch": 2.1189509007417873,
      "grad_norm": 1.25,
      "learning_rate": 0.00045497876436933257,
      "loss": 5.549,
      "mean_token_accuracy": 0.18451321572065355,
      "num_tokens": 44738985.0,
      "step": 23995
    },
    {
      "entropy": 5.8291112899780275,
      "epoch": 2.1193924408336278,
      "grad_norm": 1.890625,
      "learning_rate": 0.00045495986374059674,
      "loss": 5.4784,
      "mean_token_accuracy": 0.19504982233047485,
      "num_tokens": 44748070.0,
      "step": 24000
    },
    {
      "epoch": 2.1193924408336278,
      "eval_entropy": 5.625643400373708,
      "eval_loss": 5.593287944793701,
      "eval_mean_token_accuracy": 0.19461520391127193,
      "eval_num_tokens": 44748070.0,
      "eval_runtime": 26.1111,
      "eval_samples_per_second": 1352.489,
      "eval_steps_per_second": 169.085,
      "step": 24000
    },
    {
      "entropy": 5.796493625640869,
      "epoch": 2.119833980925468,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004549409595865496,
      "loss": 5.424,
      "mean_token_accuracy": 0.19606630206108094,
      "num_tokens": 44756607.0,
      "step": 24005
    },
    {
      "entropy": 5.816817808151245,
      "epoch": 2.1202755210173083,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045492205190756133,
      "loss": 5.5458,
      "mean_token_accuracy": 0.19332479685544968,
      "num_tokens": 44765435.0,
      "step": 24010
    },
    {
      "entropy": 5.826447582244873,
      "epoch": 2.120717061109149,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004549031407040024,
      "loss": 5.5036,
      "mean_token_accuracy": 0.19069400876760484,
      "num_tokens": 44775589.0,
      "step": 24015
    },
    {
      "entropy": 5.891171407699585,
      "epoch": 2.121158601200989,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004548842259762432,
      "loss": 5.5254,
      "mean_token_accuracy": 0.18796810656785964,
      "num_tokens": 44785692.0,
      "step": 24020
    },
    {
      "entropy": 5.776044750213623,
      "epoch": 2.1216001412928294,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004548653077246544,
      "loss": 5.4156,
      "mean_token_accuracy": 0.19309582263231279,
      "num_tokens": 44794074.0,
      "step": 24025
    },
    {
      "entropy": 5.829913997650147,
      "epoch": 2.12204168138467,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004548463859496065,
      "loss": 5.5201,
      "mean_token_accuracy": 0.18859413713216783,
      "num_tokens": 44803959.0,
      "step": 24030
    },
    {
      "entropy": 5.781261396408081,
      "epoch": 2.12248322147651,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045482746065147014,
      "loss": 5.4558,
      "mean_token_accuracy": 0.19489885717630387,
      "num_tokens": 44813294.0,
      "step": 24035
    },
    {
      "entropy": 5.8588423252105715,
      "epoch": 2.1229247615683504,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004548085318306162,
      "loss": 5.4656,
      "mean_token_accuracy": 0.19109636843204497,
      "num_tokens": 44822788.0,
      "step": 24040
    },
    {
      "entropy": 5.799039840698242,
      "epoch": 2.123366301660191,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00045478959948741526,
      "loss": 5.3899,
      "mean_token_accuracy": 0.195839124917984,
      "num_tokens": 44832236.0,
      "step": 24045
    },
    {
      "entropy": 5.873392963409424,
      "epoch": 2.123807841752031,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004547706636222384,
      "loss": 5.4916,
      "mean_token_accuracy": 0.19321658015251159,
      "num_tokens": 44841258.0,
      "step": 24050
    },
    {
      "entropy": 5.851976633071899,
      "epoch": 2.1242493818438715,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004547517242354565,
      "loss": 5.4796,
      "mean_token_accuracy": 0.1880541041493416,
      "num_tokens": 44850365.0,
      "step": 24055
    },
    {
      "entropy": 5.848421382904053,
      "epoch": 2.124690921935712,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004547327813274405,
      "loss": 5.4981,
      "mean_token_accuracy": 0.19205972999334336,
      "num_tokens": 44860451.0,
      "step": 24060
    },
    {
      "entropy": 5.841442632675171,
      "epoch": 2.125132462027552,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004547138348985615,
      "loss": 5.4321,
      "mean_token_accuracy": 0.19586741775274277,
      "num_tokens": 44868815.0,
      "step": 24065
    },
    {
      "entropy": 5.825587749481201,
      "epoch": 2.1255740021193925,
      "grad_norm": 1.609375,
      "learning_rate": 0.00045469488494919076,
      "loss": 5.5433,
      "mean_token_accuracy": 0.19162357598543167,
      "num_tokens": 44878046.0,
      "step": 24070
    },
    {
      "entropy": 5.76623764038086,
      "epoch": 2.126015542211233,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004546759314796994,
      "loss": 5.4303,
      "mean_token_accuracy": 0.197223699092865,
      "num_tokens": 44886915.0,
      "step": 24075
    },
    {
      "entropy": 5.863237953186035,
      "epoch": 2.126457082303073,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004546569744904586,
      "loss": 5.4853,
      "mean_token_accuracy": 0.1891980767250061,
      "num_tokens": 44895629.0,
      "step": 24080
    },
    {
      "entropy": 5.886371898651123,
      "epoch": 2.1268986223949136,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045463801398184004,
      "loss": 5.6252,
      "mean_token_accuracy": 0.18007916808128357,
      "num_tokens": 44905913.0,
      "step": 24085
    },
    {
      "entropy": 5.951461935043335,
      "epoch": 2.1273401624867536,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004546190499542149,
      "loss": 5.5174,
      "mean_token_accuracy": 0.18525456488132477,
      "num_tokens": 44915329.0,
      "step": 24090
    },
    {
      "entropy": 5.8908061504364015,
      "epoch": 2.127781702578594,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004546000824079547,
      "loss": 5.4817,
      "mean_token_accuracy": 0.1913362219929695,
      "num_tokens": 44923810.0,
      "step": 24095
    },
    {
      "entropy": 5.820202732086182,
      "epoch": 2.1282232426704346,
      "grad_norm": 1.890625,
      "learning_rate": 0.000454581111343431,
      "loss": 5.5208,
      "mean_token_accuracy": 0.18676782250404358,
      "num_tokens": 44932375.0,
      "step": 24100
    },
    {
      "entropy": 5.8393913269042965,
      "epoch": 2.1286647827622747,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004545621367610155,
      "loss": 5.5057,
      "mean_token_accuracy": 0.18498091399669647,
      "num_tokens": 44941341.0,
      "step": 24105
    },
    {
      "entropy": 5.91528549194336,
      "epoch": 2.129106322854115,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004545431586610799,
      "loss": 5.5255,
      "mean_token_accuracy": 0.19203294962644576,
      "num_tokens": 44951221.0,
      "step": 24110
    },
    {
      "entropy": 5.709408903121949,
      "epoch": 2.1295478629459557,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004545241770439958,
      "loss": 5.4316,
      "mean_token_accuracy": 0.20288825184106826,
      "num_tokens": 44960254.0,
      "step": 24115
    },
    {
      "entropy": 5.854597377777099,
      "epoch": 2.1299894030377957,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004545051919101353,
      "loss": 5.6114,
      "mean_token_accuracy": 0.18054451048374176,
      "num_tokens": 44969023.0,
      "step": 24120
    },
    {
      "entropy": 5.826132917404175,
      "epoch": 2.130430943129636,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045448620325987013,
      "loss": 5.475,
      "mean_token_accuracy": 0.1924513578414917,
      "num_tokens": 44978942.0,
      "step": 24125
    },
    {
      "entropy": 5.829759025573731,
      "epoch": 2.1308724832214767,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045446721109357243,
      "loss": 5.4243,
      "mean_token_accuracy": 0.18823812901973724,
      "num_tokens": 44987794.0,
      "step": 24130
    },
    {
      "entropy": 5.880821466445923,
      "epoch": 2.1313140233133168,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045444821541161404,
      "loss": 5.5635,
      "mean_token_accuracy": 0.18524927496910096,
      "num_tokens": 44995781.0,
      "step": 24135
    },
    {
      "entropy": 5.861410188674927,
      "epoch": 2.1317555634051573,
      "grad_norm": 1.25,
      "learning_rate": 0.0004544292162143671,
      "loss": 5.4912,
      "mean_token_accuracy": 0.18971319794654845,
      "num_tokens": 45004731.0,
      "step": 24140
    },
    {
      "entropy": 5.894813966751099,
      "epoch": 2.1321971034969973,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004544102135022039,
      "loss": 5.5231,
      "mean_token_accuracy": 0.1819302797317505,
      "num_tokens": 45014312.0,
      "step": 24145
    },
    {
      "entropy": 5.90670599937439,
      "epoch": 2.132638643588838,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004543912072754967,
      "loss": 5.5945,
      "mean_token_accuracy": 0.18106300234794617,
      "num_tokens": 45023903.0,
      "step": 24150
    },
    {
      "entropy": 5.849125814437866,
      "epoch": 2.1330801836806783,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004543721975346177,
      "loss": 5.5145,
      "mean_token_accuracy": 0.18510211408138275,
      "num_tokens": 45033330.0,
      "step": 24155
    },
    {
      "entropy": 5.7271417617797855,
      "epoch": 2.1335217237725184,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045435318427993946,
      "loss": 5.3821,
      "mean_token_accuracy": 0.20090160220861436,
      "num_tokens": 45042862.0,
      "step": 24160
    },
    {
      "entropy": 5.814025163650513,
      "epoch": 2.133963263864359,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004543341675118343,
      "loss": 5.4646,
      "mean_token_accuracy": 0.19433607012033463,
      "num_tokens": 45052198.0,
      "step": 24165
    },
    {
      "entropy": 5.885628986358642,
      "epoch": 2.1344048039561994,
      "grad_norm": 1.765625,
      "learning_rate": 0.00045431514723067487,
      "loss": 5.5049,
      "mean_token_accuracy": 0.1944395661354065,
      "num_tokens": 45060625.0,
      "step": 24170
    },
    {
      "entropy": 5.799396610260009,
      "epoch": 2.1348463440480394,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045429612343683354,
      "loss": 5.4285,
      "mean_token_accuracy": 0.19408624321222306,
      "num_tokens": 45069557.0,
      "step": 24175
    },
    {
      "entropy": 5.827454376220703,
      "epoch": 2.13528788413988,
      "grad_norm": 2.421875,
      "learning_rate": 0.00045427709613068317,
      "loss": 5.557,
      "mean_token_accuracy": 0.18755442649126053,
      "num_tokens": 45079281.0,
      "step": 24180
    },
    {
      "entropy": 5.939194869995117,
      "epoch": 2.1357294242317204,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004542580653125965,
      "loss": 5.5438,
      "mean_token_accuracy": 0.18911024779081345,
      "num_tokens": 45089363.0,
      "step": 24185
    },
    {
      "entropy": 5.841699552536011,
      "epoch": 2.1361709643235605,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045423903098294624,
      "loss": 5.4457,
      "mean_token_accuracy": 0.18621841222047805,
      "num_tokens": 45098735.0,
      "step": 24190
    },
    {
      "entropy": 5.896944618225097,
      "epoch": 2.136612504415401,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045421999314210535,
      "loss": 5.6043,
      "mean_token_accuracy": 0.18231717199087144,
      "num_tokens": 45109173.0,
      "step": 24195
    },
    {
      "entropy": 5.846737051010132,
      "epoch": 2.1370540445072415,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004542009517904466,
      "loss": 5.5973,
      "mean_token_accuracy": 0.18257020711898803,
      "num_tokens": 45119153.0,
      "step": 24200
    },
    {
      "entropy": 5.7546892166137695,
      "epoch": 2.1374955845990815,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004541819069283432,
      "loss": 5.457,
      "mean_token_accuracy": 0.19127636700868605,
      "num_tokens": 45128219.0,
      "step": 24205
    },
    {
      "entropy": 5.856060409545899,
      "epoch": 2.137937124690922,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004541628585561682,
      "loss": 5.521,
      "mean_token_accuracy": 0.19440967589616776,
      "num_tokens": 45137777.0,
      "step": 24210
    },
    {
      "entropy": 5.813267183303833,
      "epoch": 2.1383786647827625,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004541438066742946,
      "loss": 5.4681,
      "mean_token_accuracy": 0.19002658128738403,
      "num_tokens": 45147553.0,
      "step": 24215
    },
    {
      "entropy": 5.843930387496949,
      "epoch": 2.1388202048746026,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004541247512830958,
      "loss": 5.5236,
      "mean_token_accuracy": 0.18054716438055038,
      "num_tokens": 45157443.0,
      "step": 24220
    },
    {
      "entropy": 5.924603033065796,
      "epoch": 2.139261744966443,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000454105692382945,
      "loss": 5.5602,
      "mean_token_accuracy": 0.18949793875217438,
      "num_tokens": 45167279.0,
      "step": 24225
    },
    {
      "entropy": 5.943673706054687,
      "epoch": 2.139703285058283,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045408662997421546,
      "loss": 5.594,
      "mean_token_accuracy": 0.18798813074827195,
      "num_tokens": 45176515.0,
      "step": 24230
    },
    {
      "entropy": 5.872622489929199,
      "epoch": 2.1401448251501236,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004540675640572807,
      "loss": 5.5091,
      "mean_token_accuracy": 0.18781062960624695,
      "num_tokens": 45185501.0,
      "step": 24235
    },
    {
      "entropy": 5.867527770996094,
      "epoch": 2.140586365241964,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004540484946325143,
      "loss": 5.5943,
      "mean_token_accuracy": 0.18061817139387132,
      "num_tokens": 45194881.0,
      "step": 24240
    },
    {
      "entropy": 5.8454665184021,
      "epoch": 2.141027905333804,
      "grad_norm": 1.25,
      "learning_rate": 0.00045402942170028965,
      "loss": 5.4854,
      "mean_token_accuracy": 0.19364893585443496,
      "num_tokens": 45204463.0,
      "step": 24245
    },
    {
      "entropy": 5.853513193130493,
      "epoch": 2.1414694454256447,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00045401034526098053,
      "loss": 5.5202,
      "mean_token_accuracy": 0.18671790659427642,
      "num_tokens": 45214302.0,
      "step": 24250
    },
    {
      "entropy": 5.861611604690552,
      "epoch": 2.141910985517485,
      "grad_norm": 1.375,
      "learning_rate": 0.0004539912653149604,
      "loss": 5.5353,
      "mean_token_accuracy": 0.18561749011278153,
      "num_tokens": 45224565.0,
      "step": 24255
    },
    {
      "entropy": 5.760694742202759,
      "epoch": 2.142352525609325,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004539721818626033,
      "loss": 5.4059,
      "mean_token_accuracy": 0.19817485064268112,
      "num_tokens": 45234574.0,
      "step": 24260
    },
    {
      "entropy": 5.76415057182312,
      "epoch": 2.1427940657011657,
      "grad_norm": 1.375,
      "learning_rate": 0.0004539530949042829,
      "loss": 5.3848,
      "mean_token_accuracy": 0.20028116106986998,
      "num_tokens": 45242707.0,
      "step": 24265
    },
    {
      "entropy": 5.845958995819092,
      "epoch": 2.143235605793006,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004539340044403731,
      "loss": 5.6479,
      "mean_token_accuracy": 0.17921966463327407,
      "num_tokens": 45252873.0,
      "step": 24270
    },
    {
      "entropy": 5.891510486602783,
      "epoch": 2.1436771458848463,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000453914910471248,
      "loss": 5.4974,
      "mean_token_accuracy": 0.18940561264753342,
      "num_tokens": 45262833.0,
      "step": 24275
    },
    {
      "entropy": 5.822903108596802,
      "epoch": 2.1441186859766868,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004538958129972815,
      "loss": 5.4825,
      "mean_token_accuracy": 0.196443173289299,
      "num_tokens": 45271798.0,
      "step": 24280
    },
    {
      "entropy": 5.8312659740448,
      "epoch": 2.144560226068527,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004538767120188479,
      "loss": 5.4421,
      "mean_token_accuracy": 0.19434908628463746,
      "num_tokens": 45281470.0,
      "step": 24285
    },
    {
      "entropy": 5.874058103561401,
      "epoch": 2.1450017661603673,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004538576075363211,
      "loss": 5.4951,
      "mean_token_accuracy": 0.18975018262863158,
      "num_tokens": 45290685.0,
      "step": 24290
    },
    {
      "entropy": 5.8517309665679935,
      "epoch": 2.145443306252208,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045383849955007554,
      "loss": 5.4746,
      "mean_token_accuracy": 0.19198302030563355,
      "num_tokens": 45299631.0,
      "step": 24295
    },
    {
      "entropy": 5.903859949111938,
      "epoch": 2.145884846344048,
      "grad_norm": 1.375,
      "learning_rate": 0.00045381938806048544,
      "loss": 5.5906,
      "mean_token_accuracy": 0.1862277716398239,
      "num_tokens": 45309871.0,
      "step": 24300
    },
    {
      "entropy": 5.7893448829650875,
      "epoch": 2.1463263864358884,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004538002730679253,
      "loss": 5.3543,
      "mean_token_accuracy": 0.20121535807847976,
      "num_tokens": 45319233.0,
      "step": 24305
    },
    {
      "entropy": 5.727677631378174,
      "epoch": 2.146767926527729,
      "grad_norm": 1.375,
      "learning_rate": 0.00045378115457276946,
      "loss": 5.4517,
      "mean_token_accuracy": 0.19504198133945466,
      "num_tokens": 45329467.0,
      "step": 24310
    },
    {
      "entropy": 5.800041961669922,
      "epoch": 2.147209466619569,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045376203257539244,
      "loss": 5.4694,
      "mean_token_accuracy": 0.19157497584819794,
      "num_tokens": 45339124.0,
      "step": 24315
    },
    {
      "entropy": 5.895149040222168,
      "epoch": 2.1476510067114094,
      "grad_norm": 1.65625,
      "learning_rate": 0.00045374290707616893,
      "loss": 5.5899,
      "mean_token_accuracy": 0.18409787118434906,
      "num_tokens": 45348973.0,
      "step": 24320
    },
    {
      "entropy": 5.882757902145386,
      "epoch": 2.14809254680325,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004537237780754735,
      "loss": 5.5311,
      "mean_token_accuracy": 0.18773435950279235,
      "num_tokens": 45359161.0,
      "step": 24325
    },
    {
      "entropy": 5.855289316177368,
      "epoch": 2.14853408689509,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004537046455736809,
      "loss": 5.5231,
      "mean_token_accuracy": 0.19090772718191146,
      "num_tokens": 45369019.0,
      "step": 24330
    },
    {
      "entropy": 5.797219467163086,
      "epoch": 2.1489756269869305,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004536855095711659,
      "loss": 5.4589,
      "mean_token_accuracy": 0.1997300460934639,
      "num_tokens": 45378491.0,
      "step": 24335
    },
    {
      "entropy": 5.862304925918579,
      "epoch": 2.149417167078771,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004536663700683035,
      "loss": 5.5175,
      "mean_token_accuracy": 0.18702490031719207,
      "num_tokens": 45387651.0,
      "step": 24340
    },
    {
      "entropy": 5.925714635848999,
      "epoch": 2.149858707170611,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004536472270654685,
      "loss": 5.4976,
      "mean_token_accuracy": 0.1871267229318619,
      "num_tokens": 45396524.0,
      "step": 24345
    },
    {
      "entropy": 5.8798564910888675,
      "epoch": 2.1503002472624515,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004536280805630358,
      "loss": 5.5509,
      "mean_token_accuracy": 0.19305554330348967,
      "num_tokens": 45404849.0,
      "step": 24350
    },
    {
      "entropy": 5.872095918655395,
      "epoch": 2.150741787354292,
      "grad_norm": 1.671875,
      "learning_rate": 0.00045360893056138076,
      "loss": 5.4519,
      "mean_token_accuracy": 0.19218512624502182,
      "num_tokens": 45413628.0,
      "step": 24355
    },
    {
      "entropy": 5.840049505233765,
      "epoch": 2.151183327446132,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045358977706087827,
      "loss": 5.4548,
      "mean_token_accuracy": 0.19698928892612458,
      "num_tokens": 45423190.0,
      "step": 24360
    },
    {
      "entropy": 5.753255176544189,
      "epoch": 2.1516248675379726,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004535706200619036,
      "loss": 5.4631,
      "mean_token_accuracy": 0.195666079223156,
      "num_tokens": 45432075.0,
      "step": 24365
    },
    {
      "entropy": 5.808006763458252,
      "epoch": 2.1520664076298126,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045355145956483205,
      "loss": 5.5902,
      "mean_token_accuracy": 0.18292289972305298,
      "num_tokens": 45442627.0,
      "step": 24370
    },
    {
      "entropy": 5.842738962173462,
      "epoch": 2.152507947721653,
      "grad_norm": 1.625,
      "learning_rate": 0.00045353229557003896,
      "loss": 5.4638,
      "mean_token_accuracy": 0.20253982096910478,
      "num_tokens": 45452818.0,
      "step": 24375
    },
    {
      "entropy": 5.752610206604004,
      "epoch": 2.1529494878134936,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004535131280778997,
      "loss": 5.4547,
      "mean_token_accuracy": 0.19988421797752381,
      "num_tokens": 45461964.0,
      "step": 24380
    },
    {
      "entropy": 5.763576173782349,
      "epoch": 2.1533910279053337,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004534939570887898,
      "loss": 5.4785,
      "mean_token_accuracy": 0.19461813271045686,
      "num_tokens": 45471700.0,
      "step": 24385
    },
    {
      "entropy": 5.839812755584717,
      "epoch": 2.153832567997174,
      "grad_norm": 1.25,
      "learning_rate": 0.0004534747826030848,
      "loss": 5.4436,
      "mean_token_accuracy": 0.18726915419101714,
      "num_tokens": 45481291.0,
      "step": 24390
    },
    {
      "entropy": 5.815150499343872,
      "epoch": 2.1542741080890146,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004534556046211604,
      "loss": 5.4844,
      "mean_token_accuracy": 0.19308825582265854,
      "num_tokens": 45490317.0,
      "step": 24395
    },
    {
      "entropy": 5.8201611042022705,
      "epoch": 2.1547156481808547,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045343642314339206,
      "loss": 5.5493,
      "mean_token_accuracy": 0.18320144712924957,
      "num_tokens": 45499688.0,
      "step": 24400
    },
    {
      "entropy": 5.831044006347656,
      "epoch": 2.155157188272695,
      "grad_norm": 1.375,
      "learning_rate": 0.0004534172381701558,
      "loss": 5.4891,
      "mean_token_accuracy": 0.19496040493249894,
      "num_tokens": 45508951.0,
      "step": 24405
    },
    {
      "entropy": 5.8629741191864015,
      "epoch": 2.1555987283645357,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004533980497018272,
      "loss": 5.5519,
      "mean_token_accuracy": 0.18769272267818451,
      "num_tokens": 45518955.0,
      "step": 24410
    },
    {
      "entropy": 5.875511741638183,
      "epoch": 2.1560402684563758,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004533788577387824,
      "loss": 5.5116,
      "mean_token_accuracy": 0.18611525893211364,
      "num_tokens": 45527869.0,
      "step": 24415
    },
    {
      "entropy": 5.844658088684082,
      "epoch": 2.1564818085482163,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004533596622813971,
      "loss": 5.4351,
      "mean_token_accuracy": 0.19384365230798722,
      "num_tokens": 45537250.0,
      "step": 24420
    },
    {
      "entropy": 5.766973352432251,
      "epoch": 2.1569233486400563,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004533404633300475,
      "loss": 5.4528,
      "mean_token_accuracy": 0.19093670696020126,
      "num_tokens": 45547204.0,
      "step": 24425
    },
    {
      "entropy": 5.759471797943116,
      "epoch": 2.157364888731897,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045332126088510965,
      "loss": 5.4549,
      "mean_token_accuracy": 0.18839672803878785,
      "num_tokens": 45555464.0,
      "step": 24430
    },
    {
      "entropy": 5.853564214706421,
      "epoch": 2.1578064288237373,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004533020549469597,
      "loss": 5.4132,
      "mean_token_accuracy": 0.1992808237671852,
      "num_tokens": 45564797.0,
      "step": 24435
    },
    {
      "entropy": 5.845434665679932,
      "epoch": 2.1582479689155774,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004532828455159739,
      "loss": 5.5387,
      "mean_token_accuracy": 0.18283658921718599,
      "num_tokens": 45573317.0,
      "step": 24440
    },
    {
      "entropy": 5.710386133193969,
      "epoch": 2.158689509007418,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004532636325925286,
      "loss": 5.4608,
      "mean_token_accuracy": 0.18618085682392121,
      "num_tokens": 45582838.0,
      "step": 24445
    },
    {
      "entropy": 5.773668479919434,
      "epoch": 2.1591310490992583,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045324441617700014,
      "loss": 5.3568,
      "mean_token_accuracy": 0.19913267940282822,
      "num_tokens": 45592031.0,
      "step": 24450
    },
    {
      "entropy": 5.831067800521851,
      "epoch": 2.1595725891910984,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004532251962697649,
      "loss": 5.5705,
      "mean_token_accuracy": 0.1879842311143875,
      "num_tokens": 45601353.0,
      "step": 24455
    },
    {
      "entropy": 5.768992805480957,
      "epoch": 2.160014129282939,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045320597287119943,
      "loss": 5.4086,
      "mean_token_accuracy": 0.1947968751192093,
      "num_tokens": 45611064.0,
      "step": 24460
    },
    {
      "entropy": 5.806193733215332,
      "epoch": 2.1604556693747794,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004531867459816803,
      "loss": 5.4707,
      "mean_token_accuracy": 0.19143056869506836,
      "num_tokens": 45620704.0,
      "step": 24465
    },
    {
      "entropy": 5.918283081054687,
      "epoch": 2.1608972094666195,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045316751560158414,
      "loss": 5.549,
      "mean_token_accuracy": 0.1885583519935608,
      "num_tokens": 45629967.0,
      "step": 24470
    },
    {
      "entropy": 5.896517086029053,
      "epoch": 2.16133874955846,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004531482817312877,
      "loss": 5.5516,
      "mean_token_accuracy": 0.18481461554765702,
      "num_tokens": 45639104.0,
      "step": 24475
    },
    {
      "entropy": 5.842583799362183,
      "epoch": 2.1617802896503004,
      "grad_norm": 1.515625,
      "learning_rate": 0.00045312904437116774,
      "loss": 5.6124,
      "mean_token_accuracy": 0.17967153042554856,
      "num_tokens": 45649217.0,
      "step": 24480
    },
    {
      "entropy": 5.881931495666504,
      "epoch": 2.1622218297421405,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004531098035216011,
      "loss": 5.583,
      "mean_token_accuracy": 0.18388250172138215,
      "num_tokens": 45658969.0,
      "step": 24485
    },
    {
      "entropy": 5.776577520370483,
      "epoch": 2.162663369833981,
      "grad_norm": 1.125,
      "learning_rate": 0.00045309055918296465,
      "loss": 5.3821,
      "mean_token_accuracy": 0.20095578730106353,
      "num_tokens": 45668427.0,
      "step": 24490
    },
    {
      "entropy": 5.786289215087891,
      "epoch": 2.1631049099258215,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004530713113556355,
      "loss": 5.3371,
      "mean_token_accuracy": 0.2001229926943779,
      "num_tokens": 45676675.0,
      "step": 24495
    },
    {
      "entropy": 5.736879062652588,
      "epoch": 2.1635464500176615,
      "grad_norm": 1.453125,
      "learning_rate": 0.00045305206003999057,
      "loss": 5.378,
      "mean_token_accuracy": 0.20214212238788604,
      "num_tokens": 45685756.0,
      "step": 24500
    },
    {
      "entropy": 5.775613164901733,
      "epoch": 2.163987990109502,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004530328052364071,
      "loss": 5.5149,
      "mean_token_accuracy": 0.19777699261903764,
      "num_tokens": 45694770.0,
      "step": 24505
    },
    {
      "entropy": 5.812645053863525,
      "epoch": 2.164429530201342,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004530135469452622,
      "loss": 5.3968,
      "mean_token_accuracy": 0.20096614807844163,
      "num_tokens": 45703940.0,
      "step": 24510
    },
    {
      "entropy": 5.850827741622925,
      "epoch": 2.1648710702931826,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004529942851669333,
      "loss": 5.5435,
      "mean_token_accuracy": 0.19113972038030624,
      "num_tokens": 45713875.0,
      "step": 24515
    },
    {
      "entropy": 5.882442235946655,
      "epoch": 2.165312610385023,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004529750199017974,
      "loss": 5.5539,
      "mean_token_accuracy": 0.18415285348892213,
      "num_tokens": 45723870.0,
      "step": 24520
    },
    {
      "entropy": 5.781935739517212,
      "epoch": 2.165754150476863,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004529557511502322,
      "loss": 5.4724,
      "mean_token_accuracy": 0.19772023111581802,
      "num_tokens": 45733269.0,
      "step": 24525
    },
    {
      "entropy": 5.863057661056518,
      "epoch": 2.1661956905687036,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045293647891261494,
      "loss": 5.517,
      "mean_token_accuracy": 0.18393759280443192,
      "num_tokens": 45743208.0,
      "step": 24530
    },
    {
      "entropy": 5.867428922653199,
      "epoch": 2.166637230660544,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004529172031893233,
      "loss": 5.4615,
      "mean_token_accuracy": 0.19411846548318862,
      "num_tokens": 45752411.0,
      "step": 24535
    },
    {
      "entropy": 5.753816604614258,
      "epoch": 2.167078770752384,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045289792398073487,
      "loss": 5.4644,
      "mean_token_accuracy": 0.19603102505207062,
      "num_tokens": 45761770.0,
      "step": 24540
    },
    {
      "entropy": 5.794659996032715,
      "epoch": 2.1675203108442247,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045287864128722724,
      "loss": 5.439,
      "mean_token_accuracy": 0.18859423249959945,
      "num_tokens": 45771227.0,
      "step": 24545
    },
    {
      "entropy": 5.779013347625733,
      "epoch": 2.167961850936065,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004528593551091782,
      "loss": 5.4064,
      "mean_token_accuracy": 0.1987183764576912,
      "num_tokens": 45780558.0,
      "step": 24550
    },
    {
      "entropy": 5.87250452041626,
      "epoch": 2.1684033910279052,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045284006544696554,
      "loss": 5.5182,
      "mean_token_accuracy": 0.18928080350160598,
      "num_tokens": 45790078.0,
      "step": 24555
    },
    {
      "entropy": 5.804422426223755,
      "epoch": 2.1688449311197457,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045282077230096715,
      "loss": 5.4602,
      "mean_token_accuracy": 0.19373112171888351,
      "num_tokens": 45799962.0,
      "step": 24560
    },
    {
      "entropy": 5.7634828090667725,
      "epoch": 2.169286471211586,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045280147567156097,
      "loss": 5.3559,
      "mean_token_accuracy": 0.20571949630975722,
      "num_tokens": 45809599.0,
      "step": 24565
    },
    {
      "entropy": 5.711798429489136,
      "epoch": 2.1697280113034263,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000452782175559125,
      "loss": 5.4187,
      "mean_token_accuracy": 0.1996865302324295,
      "num_tokens": 45818318.0,
      "step": 24570
    },
    {
      "entropy": 5.870879983901977,
      "epoch": 2.170169551395267,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004527628719640374,
      "loss": 5.5485,
      "mean_token_accuracy": 0.18635669499635696,
      "num_tokens": 45828083.0,
      "step": 24575
    },
    {
      "entropy": 5.865608024597168,
      "epoch": 2.170611091487107,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045274356488667614,
      "loss": 5.4835,
      "mean_token_accuracy": 0.18908601999282837,
      "num_tokens": 45838441.0,
      "step": 24580
    },
    {
      "entropy": 5.82575421333313,
      "epoch": 2.1710526315789473,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045272425432741953,
      "loss": 5.4943,
      "mean_token_accuracy": 0.19075985997915268,
      "num_tokens": 45848047.0,
      "step": 24585
    },
    {
      "entropy": 5.808656358718872,
      "epoch": 2.171494171670788,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045270494028664586,
      "loss": 5.5026,
      "mean_token_accuracy": 0.18866759091615676,
      "num_tokens": 45857675.0,
      "step": 24590
    },
    {
      "entropy": 5.824594593048095,
      "epoch": 2.171935711762628,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004526856227647335,
      "loss": 5.4265,
      "mean_token_accuracy": 0.19348441809415817,
      "num_tokens": 45866506.0,
      "step": 24595
    },
    {
      "entropy": 5.89686017036438,
      "epoch": 2.1723772518544684,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004526663017620609,
      "loss": 5.4995,
      "mean_token_accuracy": 0.19201127141714097,
      "num_tokens": 45875955.0,
      "step": 24600
    },
    {
      "entropy": 5.877992820739746,
      "epoch": 2.172818791946309,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004526469772790063,
      "loss": 5.4712,
      "mean_token_accuracy": 0.19096562415361404,
      "num_tokens": 45884518.0,
      "step": 24605
    },
    {
      "entropy": 5.806688213348389,
      "epoch": 2.173260332038149,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004526276493159487,
      "loss": 5.4423,
      "mean_token_accuracy": 0.19852527976036072,
      "num_tokens": 45894006.0,
      "step": 24610
    },
    {
      "entropy": 5.729820966720581,
      "epoch": 2.1737018721299894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004526083178732663,
      "loss": 5.3952,
      "mean_token_accuracy": 0.19543572068214415,
      "num_tokens": 45902739.0,
      "step": 24615
    },
    {
      "entropy": 5.693874216079712,
      "epoch": 2.17414341222183,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045258898295133797,
      "loss": 5.3297,
      "mean_token_accuracy": 0.20247333645820617,
      "num_tokens": 45911418.0,
      "step": 24620
    },
    {
      "entropy": 5.808693742752075,
      "epoch": 2.17458495231367,
      "grad_norm": 1.53125,
      "learning_rate": 0.00045256964455054255,
      "loss": 5.4795,
      "mean_token_accuracy": 0.19458792060613633,
      "num_tokens": 45921135.0,
      "step": 24625
    },
    {
      "entropy": 5.848145055770874,
      "epoch": 2.1750264924055105,
      "grad_norm": 1.421875,
      "learning_rate": 0.00045255030267125874,
      "loss": 5.4195,
      "mean_token_accuracy": 0.19139019250869752,
      "num_tokens": 45930276.0,
      "step": 24630
    },
    {
      "entropy": 5.832551097869873,
      "epoch": 2.175468032497351,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045253095731386553,
      "loss": 5.4533,
      "mean_token_accuracy": 0.18873689770698548,
      "num_tokens": 45939466.0,
      "step": 24635
    },
    {
      "entropy": 5.81713638305664,
      "epoch": 2.175909572589191,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045251160847874174,
      "loss": 5.4052,
      "mean_token_accuracy": 0.1930929034948349,
      "num_tokens": 45948158.0,
      "step": 24640
    },
    {
      "entropy": 5.865111637115478,
      "epoch": 2.1763511126810315,
      "grad_norm": 2.734375,
      "learning_rate": 0.00045249225616626656,
      "loss": 5.5863,
      "mean_token_accuracy": 0.18632323741912843,
      "num_tokens": 45957759.0,
      "step": 24645
    },
    {
      "entropy": 5.8245165824890135,
      "epoch": 2.1767926527728716,
      "grad_norm": 1.453125,
      "learning_rate": 0.000452472900376819,
      "loss": 5.4885,
      "mean_token_accuracy": 0.18858775943517686,
      "num_tokens": 45966689.0,
      "step": 24650
    },
    {
      "entropy": 5.831399011611938,
      "epoch": 2.177234192864712,
      "grad_norm": 1.25,
      "learning_rate": 0.00045245354111077827,
      "loss": 5.5488,
      "mean_token_accuracy": 0.18896131515502929,
      "num_tokens": 45975618.0,
      "step": 24655
    },
    {
      "entropy": 5.862236595153808,
      "epoch": 2.1776757329565526,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045243417836852355,
      "loss": 5.4799,
      "mean_token_accuracy": 0.19411153346300125,
      "num_tokens": 45984500.0,
      "step": 24660
    },
    {
      "entropy": 5.874811315536499,
      "epoch": 2.1781172730483926,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004524148121504343,
      "loss": 5.4627,
      "mean_token_accuracy": 0.18772868812084198,
      "num_tokens": 45994471.0,
      "step": 24665
    },
    {
      "entropy": 5.962192583084106,
      "epoch": 2.178558813140233,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045239544245688967,
      "loss": 5.5577,
      "mean_token_accuracy": 0.1865691363811493,
      "num_tokens": 46003757.0,
      "step": 24670
    },
    {
      "entropy": 5.773612976074219,
      "epoch": 2.1790003532320736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004523760692882693,
      "loss": 5.3911,
      "mean_token_accuracy": 0.2032511666417122,
      "num_tokens": 46013512.0,
      "step": 24675
    },
    {
      "entropy": 5.7819726943969725,
      "epoch": 2.1794418933239137,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004523566926449526,
      "loss": 5.4749,
      "mean_token_accuracy": 0.19569631963968276,
      "num_tokens": 46022867.0,
      "step": 24680
    },
    {
      "entropy": 5.805155324935913,
      "epoch": 2.179883433415754,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045233731252731905,
      "loss": 5.5015,
      "mean_token_accuracy": 0.1861927703022957,
      "num_tokens": 46031799.0,
      "step": 24685
    },
    {
      "entropy": 5.8516394138336185,
      "epoch": 2.1803249735075947,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045231792893574846,
      "loss": 5.5284,
      "mean_token_accuracy": 0.18067081719636918,
      "num_tokens": 46041917.0,
      "step": 24690
    },
    {
      "entropy": 5.873839855194092,
      "epoch": 2.1807665135994347,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045229854187062054,
      "loss": 5.4979,
      "mean_token_accuracy": 0.19369535893201828,
      "num_tokens": 46051629.0,
      "step": 24695
    },
    {
      "entropy": 5.721113300323486,
      "epoch": 2.1812080536912752,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004522791513323149,
      "loss": 5.4072,
      "mean_token_accuracy": 0.20364447087049484,
      "num_tokens": 46060840.0,
      "step": 24700
    },
    {
      "entropy": 5.695599699020386,
      "epoch": 2.1816495937831153,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045225975732121165,
      "loss": 5.4417,
      "mean_token_accuracy": 0.19405079483985901,
      "num_tokens": 46069774.0,
      "step": 24705
    },
    {
      "entropy": 5.866010141372681,
      "epoch": 2.182091133874956,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045224035983769054,
      "loss": 5.5527,
      "mean_token_accuracy": 0.18599710315465928,
      "num_tokens": 46079377.0,
      "step": 24710
    },
    {
      "entropy": 5.851829719543457,
      "epoch": 2.1825326739667963,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045222095888213155,
      "loss": 5.452,
      "mean_token_accuracy": 0.19813531190156936,
      "num_tokens": 46088190.0,
      "step": 24715
    },
    {
      "entropy": 5.8324666023254395,
      "epoch": 2.1829742140586363,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045220155445491465,
      "loss": 5.5028,
      "mean_token_accuracy": 0.19790344536304474,
      "num_tokens": 46097487.0,
      "step": 24720
    },
    {
      "entropy": 5.873967695236206,
      "epoch": 2.183415754150477,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004521821465564203,
      "loss": 5.5726,
      "mean_token_accuracy": 0.18620011061429978,
      "num_tokens": 46106554.0,
      "step": 24725
    },
    {
      "entropy": 5.870430946350098,
      "epoch": 2.1838572942423173,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004521627351870283,
      "loss": 5.5744,
      "mean_token_accuracy": 0.1822180926799774,
      "num_tokens": 46115665.0,
      "step": 24730
    },
    {
      "entropy": 5.832641077041626,
      "epoch": 2.1842988343341574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004521433203471192,
      "loss": 5.44,
      "mean_token_accuracy": 0.19783271551132203,
      "num_tokens": 46124863.0,
      "step": 24735
    },
    {
      "entropy": 5.847654819488525,
      "epoch": 2.184740374425998,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045212390203707315,
      "loss": 5.4739,
      "mean_token_accuracy": 0.18850009143352509,
      "num_tokens": 46133986.0,
      "step": 24740
    },
    {
      "entropy": 5.789244890213013,
      "epoch": 2.1851819145178384,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004521044802572705,
      "loss": 5.3987,
      "mean_token_accuracy": 0.20067705810070038,
      "num_tokens": 46142234.0,
      "step": 24745
    },
    {
      "entropy": 5.787045669555664,
      "epoch": 2.1856234546096784,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045208505500809196,
      "loss": 5.5002,
      "mean_token_accuracy": 0.196712464094162,
      "num_tokens": 46150976.0,
      "step": 24750
    },
    {
      "entropy": 5.864871883392334,
      "epoch": 2.186064994701519,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045206562628991785,
      "loss": 5.5025,
      "mean_token_accuracy": 0.19419410526752473,
      "num_tokens": 46160541.0,
      "step": 24755
    },
    {
      "entropy": 5.881729316711426,
      "epoch": 2.1865065347933594,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045204619410312875,
      "loss": 5.5427,
      "mean_token_accuracy": 0.18436092585325242,
      "num_tokens": 46170958.0,
      "step": 24760
    },
    {
      "entropy": 5.82484803199768,
      "epoch": 2.1869480748851995,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004520267584481055,
      "loss": 5.4809,
      "mean_token_accuracy": 0.19420868307352065,
      "num_tokens": 46180293.0,
      "step": 24765
    },
    {
      "entropy": 5.791618585586548,
      "epoch": 2.18738961497704,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00045200731932522873,
      "loss": 5.4535,
      "mean_token_accuracy": 0.19792397916316987,
      "num_tokens": 46189264.0,
      "step": 24770
    },
    {
      "entropy": 5.895522785186768,
      "epoch": 2.1878311550688805,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004519878767348792,
      "loss": 5.5932,
      "mean_token_accuracy": 0.17784345597028733,
      "num_tokens": 46199417.0,
      "step": 24775
    },
    {
      "entropy": 5.788554048538208,
      "epoch": 2.1882726951607205,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004519684306774378,
      "loss": 5.4212,
      "mean_token_accuracy": 0.19652555584907533,
      "num_tokens": 46209360.0,
      "step": 24780
    },
    {
      "entropy": 5.78150954246521,
      "epoch": 2.188714235252561,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045194898115328555,
      "loss": 5.4456,
      "mean_token_accuracy": 0.1861520379781723,
      "num_tokens": 46218414.0,
      "step": 24785
    },
    {
      "entropy": 5.818973588943481,
      "epoch": 2.189155775344401,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004519295281628033,
      "loss": 5.4618,
      "mean_token_accuracy": 0.19485165178775787,
      "num_tokens": 46227526.0,
      "step": 24790
    },
    {
      "entropy": 5.848794078826904,
      "epoch": 2.1895973154362416,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004519100717063723,
      "loss": 5.3999,
      "mean_token_accuracy": 0.19451722502708435,
      "num_tokens": 46236041.0,
      "step": 24795
    },
    {
      "entropy": 5.733925819396973,
      "epoch": 2.190038855528082,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045189061178437363,
      "loss": 5.4078,
      "mean_token_accuracy": 0.19611672759056092,
      "num_tokens": 46244557.0,
      "step": 24800
    },
    {
      "entropy": 5.781933069229126,
      "epoch": 2.190480395619922,
      "grad_norm": 1.25,
      "learning_rate": 0.0004518711483971885,
      "loss": 5.4107,
      "mean_token_accuracy": 0.19884918481111527,
      "num_tokens": 46253957.0,
      "step": 24805
    },
    {
      "entropy": 5.899817276000976,
      "epoch": 2.1909219357117626,
      "grad_norm": 1.3125,
      "learning_rate": 0.000451851681545198,
      "loss": 5.5312,
      "mean_token_accuracy": 0.1804847851395607,
      "num_tokens": 46263110.0,
      "step": 24810
    },
    {
      "entropy": 5.7922547340393065,
      "epoch": 2.191363475803603,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045183221122878375,
      "loss": 5.5252,
      "mean_token_accuracy": 0.1867988333106041,
      "num_tokens": 46272163.0,
      "step": 24815
    },
    {
      "entropy": 5.7954329490661625,
      "epoch": 2.191805015895443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045181273744832707,
      "loss": 5.5128,
      "mean_token_accuracy": 0.18288943469524382,
      "num_tokens": 46281097.0,
      "step": 24820
    },
    {
      "entropy": 5.9173722743988035,
      "epoch": 2.1922465559872837,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004517932602042094,
      "loss": 5.4651,
      "mean_token_accuracy": 0.1884808823466301,
      "num_tokens": 46291552.0,
      "step": 24825
    },
    {
      "entropy": 5.831365251541138,
      "epoch": 2.192688096079124,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00045177377949681235,
      "loss": 5.5099,
      "mean_token_accuracy": 0.18850334137678146,
      "num_tokens": 46300661.0,
      "step": 24830
    },
    {
      "entropy": 5.791829061508179,
      "epoch": 2.1931296361709642,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045175429532651743,
      "loss": 5.4239,
      "mean_token_accuracy": 0.19353751987218856,
      "num_tokens": 46309705.0,
      "step": 24835
    },
    {
      "entropy": 5.788273477554322,
      "epoch": 2.1935711762628047,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004517348076937064,
      "loss": 5.4373,
      "mean_token_accuracy": 0.19672725349664688,
      "num_tokens": 46319701.0,
      "step": 24840
    },
    {
      "entropy": 5.757566022872925,
      "epoch": 2.194012716354645,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045171531659876104,
      "loss": 5.4526,
      "mean_token_accuracy": 0.19884357899427413,
      "num_tokens": 46328195.0,
      "step": 24845
    },
    {
      "entropy": 5.870284223556519,
      "epoch": 2.1944542564464853,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004516958220420631,
      "loss": 5.4934,
      "mean_token_accuracy": 0.1921232059597969,
      "num_tokens": 46337518.0,
      "step": 24850
    },
    {
      "entropy": 5.815572595596313,
      "epoch": 2.194895796538326,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004516763240239945,
      "loss": 5.4116,
      "mean_token_accuracy": 0.19979784041643142,
      "num_tokens": 46346406.0,
      "step": 24855
    },
    {
      "entropy": 5.763052988052368,
      "epoch": 2.195337336630166,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004516568225449372,
      "loss": 5.4439,
      "mean_token_accuracy": 0.19697344601154326,
      "num_tokens": 46355284.0,
      "step": 24860
    },
    {
      "entropy": 5.832168292999268,
      "epoch": 2.1957788767220063,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045163731760527326,
      "loss": 5.5265,
      "mean_token_accuracy": 0.1870936080813408,
      "num_tokens": 46364982.0,
      "step": 24865
    },
    {
      "entropy": 5.779687070846558,
      "epoch": 2.196220416813847,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004516178092053847,
      "loss": 5.3645,
      "mean_token_accuracy": 0.19884507805109025,
      "num_tokens": 46373923.0,
      "step": 24870
    },
    {
      "entropy": 5.779660749435425,
      "epoch": 2.196661956905687,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004515982973456537,
      "loss": 5.386,
      "mean_token_accuracy": 0.20329754948616027,
      "num_tokens": 46382194.0,
      "step": 24875
    },
    {
      "entropy": 5.773639869689942,
      "epoch": 2.1971034969975274,
      "grad_norm": 2.125,
      "learning_rate": 0.0004515787820264624,
      "loss": 5.4563,
      "mean_token_accuracy": 0.1914518043398857,
      "num_tokens": 46391463.0,
      "step": 24880
    },
    {
      "entropy": 5.8006994247436525,
      "epoch": 2.197545037089368,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045155926324819335,
      "loss": 5.4962,
      "mean_token_accuracy": 0.19368625730276107,
      "num_tokens": 46401235.0,
      "step": 24885
    },
    {
      "entropy": 5.798978090286255,
      "epoch": 2.197986577181208,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045153974101122866,
      "loss": 5.5295,
      "mean_token_accuracy": 0.18957090824842454,
      "num_tokens": 46410631.0,
      "step": 24890
    },
    {
      "entropy": 5.866455554962158,
      "epoch": 2.1984281172730484,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045152021531595085,
      "loss": 5.5026,
      "mean_token_accuracy": 0.17974088042974473,
      "num_tokens": 46420467.0,
      "step": 24895
    },
    {
      "entropy": 5.828551626205444,
      "epoch": 2.198869657364889,
      "grad_norm": 1.125,
      "learning_rate": 0.0004515006861627424,
      "loss": 5.4333,
      "mean_token_accuracy": 0.19829555600881577,
      "num_tokens": 46430276.0,
      "step": 24900
    },
    {
      "entropy": 5.8535057544708256,
      "epoch": 2.199311197456729,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00045148115355198596,
      "loss": 5.5458,
      "mean_token_accuracy": 0.188438281416893,
      "num_tokens": 46439227.0,
      "step": 24905
    },
    {
      "entropy": 5.78479642868042,
      "epoch": 2.1997527375485695,
      "grad_norm": 1.5625,
      "learning_rate": 0.000451461617484064,
      "loss": 5.4933,
      "mean_token_accuracy": 0.19270177632570268,
      "num_tokens": 46449043.0,
      "step": 24910
    },
    {
      "entropy": 5.78171672821045,
      "epoch": 2.20019427764041,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004514420779593594,
      "loss": 5.4787,
      "mean_token_accuracy": 0.18799485117197037,
      "num_tokens": 46458332.0,
      "step": 24915
    },
    {
      "entropy": 5.867466402053833,
      "epoch": 2.20063581773225,
      "grad_norm": 1.25,
      "learning_rate": 0.00045142253497825477,
      "loss": 5.4285,
      "mean_token_accuracy": 0.1930414542555809,
      "num_tokens": 46467208.0,
      "step": 24920
    },
    {
      "entropy": 5.806895160675049,
      "epoch": 2.2010773578240905,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045140298854113306,
      "loss": 5.4442,
      "mean_token_accuracy": 0.1992160990834236,
      "num_tokens": 46476715.0,
      "step": 24925
    },
    {
      "entropy": 5.713755702972412,
      "epoch": 2.2015188979159306,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045138343864837703,
      "loss": 5.3804,
      "mean_token_accuracy": 0.1993671327829361,
      "num_tokens": 46484783.0,
      "step": 24930
    },
    {
      "entropy": 5.766569566726685,
      "epoch": 2.201960438007771,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004513638853003699,
      "loss": 5.3909,
      "mean_token_accuracy": 0.1968960702419281,
      "num_tokens": 46493616.0,
      "step": 24935
    },
    {
      "entropy": 5.875431251525879,
      "epoch": 2.2024019780996116,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00045134432849749446,
      "loss": 5.4723,
      "mean_token_accuracy": 0.19105059504508973,
      "num_tokens": 46502779.0,
      "step": 24940
    },
    {
      "entropy": 5.817952108383179,
      "epoch": 2.2028435181914516,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045132476824013396,
      "loss": 5.4675,
      "mean_token_accuracy": 0.1919872209429741,
      "num_tokens": 46511506.0,
      "step": 24945
    },
    {
      "entropy": 5.844209766387939,
      "epoch": 2.203285058283292,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004513052045286715,
      "loss": 5.5014,
      "mean_token_accuracy": 0.18878966867923735,
      "num_tokens": 46520219.0,
      "step": 24950
    },
    {
      "entropy": 5.869622659683228,
      "epoch": 2.2037265983751326,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004512856373634904,
      "loss": 5.4724,
      "mean_token_accuracy": 0.19558568596839904,
      "num_tokens": 46528737.0,
      "step": 24955
    },
    {
      "entropy": 5.834464073181152,
      "epoch": 2.2041681384669727,
      "grad_norm": 1.25,
      "learning_rate": 0.00045126606674497394,
      "loss": 5.5413,
      "mean_token_accuracy": 0.19161373525857925,
      "num_tokens": 46538802.0,
      "step": 24960
    },
    {
      "entropy": 5.80554780960083,
      "epoch": 2.204609678558813,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004512464926735054,
      "loss": 5.4839,
      "mean_token_accuracy": 0.19260286092758178,
      "num_tokens": 46548369.0,
      "step": 24965
    },
    {
      "entropy": 5.813347721099854,
      "epoch": 2.2050512186506537,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045122691514946835,
      "loss": 5.5219,
      "mean_token_accuracy": 0.19132929742336274,
      "num_tokens": 46557876.0,
      "step": 24970
    },
    {
      "entropy": 5.751050662994385,
      "epoch": 2.2054927587424937,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004512073341732463,
      "loss": 5.4758,
      "mean_token_accuracy": 0.19926581680774688,
      "num_tokens": 46566747.0,
      "step": 24975
    },
    {
      "entropy": 5.8220850944519045,
      "epoch": 2.2059342988343342,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004511877497452227,
      "loss": 5.4633,
      "mean_token_accuracy": 0.19455233663320542,
      "num_tokens": 46576196.0,
      "step": 24980
    },
    {
      "entropy": 5.915978813171387,
      "epoch": 2.2063758389261743,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004511681618657813,
      "loss": 5.5394,
      "mean_token_accuracy": 0.18776900470256805,
      "num_tokens": 46585734.0,
      "step": 24985
    },
    {
      "entropy": 5.7694963932037355,
      "epoch": 2.206817379018015,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045114857053530586,
      "loss": 5.4042,
      "mean_token_accuracy": 0.1960630387067795,
      "num_tokens": 46595931.0,
      "step": 24990
    },
    {
      "entropy": 5.893754577636718,
      "epoch": 2.2072589191098553,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045112897575418006,
      "loss": 5.4681,
      "mean_token_accuracy": 0.18560731858015062,
      "num_tokens": 46604688.0,
      "step": 24995
    },
    {
      "entropy": 5.744379568099975,
      "epoch": 2.2077004592016953,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004511093775227878,
      "loss": 5.4687,
      "mean_token_accuracy": 0.19870156198740005,
      "num_tokens": 46613860.0,
      "step": 25000
    },
    {
      "entropy": 5.8514824390411375,
      "epoch": 2.208141999293536,
      "grad_norm": 1.40625,
      "learning_rate": 0.000451089775841513,
      "loss": 5.5314,
      "mean_token_accuracy": 0.188914255797863,
      "num_tokens": 46623041.0,
      "step": 25005
    },
    {
      "entropy": 5.844975709915161,
      "epoch": 2.2085835393853763,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045107017071073963,
      "loss": 5.4889,
      "mean_token_accuracy": 0.18771113604307174,
      "num_tokens": 46633025.0,
      "step": 25010
    },
    {
      "entropy": 5.8896478652954105,
      "epoch": 2.2090250794772164,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004510505621308517,
      "loss": 5.488,
      "mean_token_accuracy": 0.19414360523223878,
      "num_tokens": 46641826.0,
      "step": 25015
    },
    {
      "entropy": 5.863744735717773,
      "epoch": 2.209466619569057,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045103095010223354,
      "loss": 5.5808,
      "mean_token_accuracy": 0.18572961837053298,
      "num_tokens": 46652028.0,
      "step": 25020
    },
    {
      "entropy": 5.83162031173706,
      "epoch": 2.2099081596608974,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045101133462526903,
      "loss": 5.5833,
      "mean_token_accuracy": 0.18462528437376022,
      "num_tokens": 46660763.0,
      "step": 25025
    },
    {
      "entropy": 5.913536834716797,
      "epoch": 2.2103496997527374,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004509917157003427,
      "loss": 5.5213,
      "mean_token_accuracy": 0.18523661345243453,
      "num_tokens": 46670213.0,
      "step": 25030
    },
    {
      "entropy": 5.90562949180603,
      "epoch": 2.210791239844578,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004509720933278386,
      "loss": 5.4863,
      "mean_token_accuracy": 0.18758353292942048,
      "num_tokens": 46679658.0,
      "step": 25035
    },
    {
      "entropy": 5.753462076187134,
      "epoch": 2.2112327799364184,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004509524675081414,
      "loss": 5.4438,
      "mean_token_accuracy": 0.19901863783597945,
      "num_tokens": 46687920.0,
      "step": 25040
    },
    {
      "entropy": 5.815122842788696,
      "epoch": 2.2116743200282585,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004509328382416354,
      "loss": 5.5406,
      "mean_token_accuracy": 0.180429208278656,
      "num_tokens": 46696779.0,
      "step": 25045
    },
    {
      "entropy": 5.906235265731811,
      "epoch": 2.212115860120099,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004509132055287051,
      "loss": 5.4974,
      "mean_token_accuracy": 0.18752871453762054,
      "num_tokens": 46706568.0,
      "step": 25050
    },
    {
      "entropy": 5.889072322845459,
      "epoch": 2.2125574002119395,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004508935693697352,
      "loss": 5.5409,
      "mean_token_accuracy": 0.19087894707918168,
      "num_tokens": 46715369.0,
      "step": 25055
    },
    {
      "entropy": 5.750374507904053,
      "epoch": 2.2129989403037795,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004508739297651103,
      "loss": 5.4476,
      "mean_token_accuracy": 0.19119653403759002,
      "num_tokens": 46724867.0,
      "step": 25060
    },
    {
      "entropy": 5.79560923576355,
      "epoch": 2.21344048039562,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045085428671521514,
      "loss": 5.5167,
      "mean_token_accuracy": 0.192353555560112,
      "num_tokens": 46734611.0,
      "step": 25065
    },
    {
      "entropy": 5.8448686599731445,
      "epoch": 2.21388202048746,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045083464022043456,
      "loss": 5.5114,
      "mean_token_accuracy": 0.18568000048398972,
      "num_tokens": 46744031.0,
      "step": 25070
    },
    {
      "entropy": 5.79198579788208,
      "epoch": 2.2143235605793006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045081499028115337,
      "loss": 5.4279,
      "mean_token_accuracy": 0.20164393484592438,
      "num_tokens": 46753379.0,
      "step": 25075
    },
    {
      "entropy": 5.856601047515869,
      "epoch": 2.214765100671141,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004507953368977565,
      "loss": 5.5776,
      "mean_token_accuracy": 0.18002419769763947,
      "num_tokens": 46762625.0,
      "step": 25080
    },
    {
      "entropy": 5.870411777496338,
      "epoch": 2.215206640762981,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045077568007062893,
      "loss": 5.5201,
      "mean_token_accuracy": 0.1896331265568733,
      "num_tokens": 46772577.0,
      "step": 25085
    },
    {
      "entropy": 5.854209089279175,
      "epoch": 2.2156481808548216,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045075601980015573,
      "loss": 5.5069,
      "mean_token_accuracy": 0.1931350439786911,
      "num_tokens": 46782226.0,
      "step": 25090
    },
    {
      "entropy": 5.811879110336304,
      "epoch": 2.216089720946662,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045073635608672215,
      "loss": 5.3912,
      "mean_token_accuracy": 0.19804896265268326,
      "num_tokens": 46789988.0,
      "step": 25095
    },
    {
      "entropy": 5.826391792297363,
      "epoch": 2.216531261038502,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045071668893071324,
      "loss": 5.5206,
      "mean_token_accuracy": 0.1839609995484352,
      "num_tokens": 46799448.0,
      "step": 25100
    },
    {
      "entropy": 5.875655889511108,
      "epoch": 2.2169728011303427,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045069701833251434,
      "loss": 5.5467,
      "mean_token_accuracy": 0.18638881891965867,
      "num_tokens": 46809829.0,
      "step": 25105
    },
    {
      "entropy": 5.824425983428955,
      "epoch": 2.217414341222183,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045067734429251086,
      "loss": 5.4661,
      "mean_token_accuracy": 0.19296102374792098,
      "num_tokens": 46817975.0,
      "step": 25110
    },
    {
      "entropy": 5.756117248535157,
      "epoch": 2.2178558813140232,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045065766681108793,
      "loss": 5.4134,
      "mean_token_accuracy": 0.19789931774139405,
      "num_tokens": 46827350.0,
      "step": 25115
    },
    {
      "entropy": 5.8187994956970215,
      "epoch": 2.2182974214058637,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004506379858886314,
      "loss": 5.4727,
      "mean_token_accuracy": 0.19579324424266814,
      "num_tokens": 46835944.0,
      "step": 25120
    },
    {
      "entropy": 5.797871065139771,
      "epoch": 2.2187389614977038,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045061830152552646,
      "loss": 5.4287,
      "mean_token_accuracy": 0.19577191025018692,
      "num_tokens": 46844843.0,
      "step": 25125
    },
    {
      "entropy": 5.80336184501648,
      "epoch": 2.2191805015895443,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000450598613722159,
      "loss": 5.5564,
      "mean_token_accuracy": 0.18089923411607742,
      "num_tokens": 46854188.0,
      "step": 25130
    },
    {
      "entropy": 5.8582189083099365,
      "epoch": 2.2196220416813848,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004505789224789145,
      "loss": 5.5305,
      "mean_token_accuracy": 0.19339046478271485,
      "num_tokens": 46863279.0,
      "step": 25135
    },
    {
      "entropy": 5.84236512184143,
      "epoch": 2.220063581773225,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045055922779617874,
      "loss": 5.4777,
      "mean_token_accuracy": 0.1864086166024208,
      "num_tokens": 46873336.0,
      "step": 25140
    },
    {
      "entropy": 5.8729517459869385,
      "epoch": 2.2205051218650653,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004505395296743376,
      "loss": 5.4924,
      "mean_token_accuracy": 0.19338965713977813,
      "num_tokens": 46882497.0,
      "step": 25145
    },
    {
      "entropy": 5.796002340316773,
      "epoch": 2.220946661956906,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045051982811377696,
      "loss": 5.4211,
      "mean_token_accuracy": 0.19478584378957747,
      "num_tokens": 46891260.0,
      "step": 25150
    },
    {
      "entropy": 5.755101108551026,
      "epoch": 2.221388202048746,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045050012311488265,
      "loss": 5.3883,
      "mean_token_accuracy": 0.1999176785349846,
      "num_tokens": 46899969.0,
      "step": 25155
    },
    {
      "entropy": 5.747582292556762,
      "epoch": 2.2218297421405864,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045048041467804067,
      "loss": 5.4174,
      "mean_token_accuracy": 0.2002062052488327,
      "num_tokens": 46909206.0,
      "step": 25160
    },
    {
      "entropy": 5.791477394104004,
      "epoch": 2.222271282232427,
      "grad_norm": 1.421875,
      "learning_rate": 0.00045046070280363725,
      "loss": 5.4139,
      "mean_token_accuracy": 0.20424625128507615,
      "num_tokens": 46918423.0,
      "step": 25165
    },
    {
      "entropy": 5.856847429275513,
      "epoch": 2.222712822324267,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004504409874920585,
      "loss": 5.4642,
      "mean_token_accuracy": 0.18323842138051988,
      "num_tokens": 46927817.0,
      "step": 25170
    },
    {
      "entropy": 5.881532621383667,
      "epoch": 2.2231543624161074,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004504212687436905,
      "loss": 5.5588,
      "mean_token_accuracy": 0.18405516892671586,
      "num_tokens": 46936880.0,
      "step": 25175
    },
    {
      "entropy": 5.843238496780396,
      "epoch": 2.223595902507948,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004504015465589196,
      "loss": 5.5026,
      "mean_token_accuracy": 0.183936707675457,
      "num_tokens": 46946722.0,
      "step": 25180
    },
    {
      "entropy": 5.747507190704345,
      "epoch": 2.224037442599788,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004503818209381322,
      "loss": 5.4347,
      "mean_token_accuracy": 0.19356664270162582,
      "num_tokens": 46956660.0,
      "step": 25185
    },
    {
      "entropy": 5.852346467971802,
      "epoch": 2.2244789826916285,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004503620918817147,
      "loss": 5.4859,
      "mean_token_accuracy": 0.1834660068154335,
      "num_tokens": 46965823.0,
      "step": 25190
    },
    {
      "entropy": 5.873174238204956,
      "epoch": 2.224920522783469,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045034235939005353,
      "loss": 5.4837,
      "mean_token_accuracy": 0.1923976182937622,
      "num_tokens": 46974933.0,
      "step": 25195
    },
    {
      "entropy": 5.874148559570313,
      "epoch": 2.225362062875309,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045032262346353526,
      "loss": 5.5511,
      "mean_token_accuracy": 0.19035923928022386,
      "num_tokens": 46983506.0,
      "step": 25200
    },
    {
      "entropy": 5.810177659988403,
      "epoch": 2.2258036029671495,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004503028841025465,
      "loss": 5.4659,
      "mean_token_accuracy": 0.19440919756889344,
      "num_tokens": 46992234.0,
      "step": 25205
    },
    {
      "entropy": 5.758495569229126,
      "epoch": 2.2262451430589896,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045028314130747397,
      "loss": 5.4169,
      "mean_token_accuracy": 0.19392835050821305,
      "num_tokens": 47002249.0,
      "step": 25210
    },
    {
      "entropy": 5.82074007987976,
      "epoch": 2.22668668315083,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045026339507870435,
      "loss": 5.5428,
      "mean_token_accuracy": 0.18752488046884536,
      "num_tokens": 47010721.0,
      "step": 25215
    },
    {
      "entropy": 5.904180526733398,
      "epoch": 2.2271282232426706,
      "grad_norm": 1.484375,
      "learning_rate": 0.00045024364541662457,
      "loss": 5.5209,
      "mean_token_accuracy": 0.19064456969499588,
      "num_tokens": 47019675.0,
      "step": 25220
    },
    {
      "entropy": 5.846822595596313,
      "epoch": 2.2275697633345106,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004502238923216213,
      "loss": 5.4634,
      "mean_token_accuracy": 0.1949811816215515,
      "num_tokens": 47028293.0,
      "step": 25225
    },
    {
      "entropy": 5.881525707244873,
      "epoch": 2.228011303426351,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004502041357940817,
      "loss": 5.5266,
      "mean_token_accuracy": 0.18526853173971175,
      "num_tokens": 47037605.0,
      "step": 25230
    },
    {
      "entropy": 5.808894491195678,
      "epoch": 2.2284528435181916,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045018437583439275,
      "loss": 5.4499,
      "mean_token_accuracy": 0.19487264305353164,
      "num_tokens": 47045889.0,
      "step": 25235
    },
    {
      "entropy": 5.7865321159362795,
      "epoch": 2.2288943836100317,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045016461244294154,
      "loss": 5.4898,
      "mean_token_accuracy": 0.19571455121040343,
      "num_tokens": 47055896.0,
      "step": 25240
    },
    {
      "entropy": 5.795222187042237,
      "epoch": 2.229335923701872,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045014484562011514,
      "loss": 5.5251,
      "mean_token_accuracy": 0.18705832958221436,
      "num_tokens": 47064980.0,
      "step": 25245
    },
    {
      "entropy": 5.778140926361084,
      "epoch": 2.2297774637937127,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004501250753663008,
      "loss": 5.4099,
      "mean_token_accuracy": 0.19639189243316652,
      "num_tokens": 47074045.0,
      "step": 25250
    },
    {
      "entropy": 5.8183379650115965,
      "epoch": 2.2302190038855527,
      "grad_norm": 1.375,
      "learning_rate": 0.0004501053016818859,
      "loss": 5.4371,
      "mean_token_accuracy": 0.19379499554634094,
      "num_tokens": 47083439.0,
      "step": 25255
    },
    {
      "entropy": 5.782951402664184,
      "epoch": 2.230660543977393,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045008552456725766,
      "loss": 5.4513,
      "mean_token_accuracy": 0.1937849000096321,
      "num_tokens": 47091977.0,
      "step": 25260
    },
    {
      "entropy": 5.772253513336182,
      "epoch": 2.2311020840692333,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004500657440228037,
      "loss": 5.4926,
      "mean_token_accuracy": 0.19118792414665223,
      "num_tokens": 47102197.0,
      "step": 25265
    },
    {
      "entropy": 5.890456056594848,
      "epoch": 2.2315436241610738,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045004596004891124,
      "loss": 5.5118,
      "mean_token_accuracy": 0.1905220329761505,
      "num_tokens": 47111563.0,
      "step": 25270
    },
    {
      "entropy": 5.791409730911255,
      "epoch": 2.2319851642529143,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045002617264596807,
      "loss": 5.4768,
      "mean_token_accuracy": 0.20022563636302948,
      "num_tokens": 47120272.0,
      "step": 25275
    },
    {
      "entropy": 5.758007621765136,
      "epoch": 2.2324267043447543,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004500063818143617,
      "loss": 5.4666,
      "mean_token_accuracy": 0.1940328598022461,
      "num_tokens": 47131023.0,
      "step": 25280
    },
    {
      "entropy": 5.73273344039917,
      "epoch": 2.232868244436595,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004499865875544797,
      "loss": 5.4229,
      "mean_token_accuracy": 0.19780974835157394,
      "num_tokens": 47139968.0,
      "step": 25285
    },
    {
      "entropy": 5.784685468673706,
      "epoch": 2.2333097845284353,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004499667898667101,
      "loss": 5.4252,
      "mean_token_accuracy": 0.19823746979236603,
      "num_tokens": 47149843.0,
      "step": 25290
    },
    {
      "entropy": 5.837689971923828,
      "epoch": 2.2337513246202754,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044994698875144064,
      "loss": 5.4888,
      "mean_token_accuracy": 0.1953107312321663,
      "num_tokens": 47159025.0,
      "step": 25295
    },
    {
      "entropy": 5.817866802215576,
      "epoch": 2.234192864712116,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004499271842090591,
      "loss": 5.4983,
      "mean_token_accuracy": 0.18778059482574463,
      "num_tokens": 47168644.0,
      "step": 25300
    },
    {
      "entropy": 5.846137952804566,
      "epoch": 2.2346344048039564,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044990737623995357,
      "loss": 5.4169,
      "mean_token_accuracy": 0.1980884552001953,
      "num_tokens": 47178051.0,
      "step": 25305
    },
    {
      "entropy": 5.880659770965576,
      "epoch": 2.2350759448957964,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004498875648445119,
      "loss": 5.5539,
      "mean_token_accuracy": 0.18635555505752563,
      "num_tokens": 47187855.0,
      "step": 25310
    },
    {
      "entropy": 5.726792097091675,
      "epoch": 2.235517484987637,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004498677500231224,
      "loss": 5.4913,
      "mean_token_accuracy": 0.1924949899315834,
      "num_tokens": 47196700.0,
      "step": 25315
    },
    {
      "entropy": 5.875244569778443,
      "epoch": 2.2359590250794774,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004498479317761731,
      "loss": 5.5389,
      "mean_token_accuracy": 0.1857009768486023,
      "num_tokens": 47205202.0,
      "step": 25320
    },
    {
      "entropy": 5.789200305938721,
      "epoch": 2.2364005651713175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004498281101040522,
      "loss": 5.4641,
      "mean_token_accuracy": 0.20116854459047318,
      "num_tokens": 47214803.0,
      "step": 25325
    },
    {
      "entropy": 5.8758245468139645,
      "epoch": 2.236842105263158,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044980828500714814,
      "loss": 5.5679,
      "mean_token_accuracy": 0.18465539515018464,
      "num_tokens": 47224641.0,
      "step": 25330
    },
    {
      "entropy": 5.840283489227295,
      "epoch": 2.2372836453549985,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004497884564858492,
      "loss": 5.5164,
      "mean_token_accuracy": 0.1873737931251526,
      "num_tokens": 47233905.0,
      "step": 25335
    },
    {
      "entropy": 5.814819240570069,
      "epoch": 2.2377251854468385,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044976862454054376,
      "loss": 5.4409,
      "mean_token_accuracy": 0.19483238458633423,
      "num_tokens": 47242209.0,
      "step": 25340
    },
    {
      "entropy": 5.92715163230896,
      "epoch": 2.238166725538679,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004497487891716204,
      "loss": 5.5082,
      "mean_token_accuracy": 0.19499906599521638,
      "num_tokens": 47252609.0,
      "step": 25345
    },
    {
      "entropy": 5.8767046451568605,
      "epoch": 2.238608265630519,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044972895037946755,
      "loss": 5.5419,
      "mean_token_accuracy": 0.18804085701704026,
      "num_tokens": 47261868.0,
      "step": 25350
    },
    {
      "entropy": 5.8350849628448485,
      "epoch": 2.2390498057223596,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044970910816447403,
      "loss": 5.4466,
      "mean_token_accuracy": 0.19747993499040603,
      "num_tokens": 47271404.0,
      "step": 25355
    },
    {
      "entropy": 5.85529260635376,
      "epoch": 2.2394913458142,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044968926252702835,
      "loss": 5.5269,
      "mean_token_accuracy": 0.18492158204317094,
      "num_tokens": 47280669.0,
      "step": 25360
    },
    {
      "entropy": 5.79316725730896,
      "epoch": 2.23993288590604,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004496694134675195,
      "loss": 5.5307,
      "mean_token_accuracy": 0.18617749214172363,
      "num_tokens": 47290096.0,
      "step": 25365
    },
    {
      "entropy": 5.874289178848267,
      "epoch": 2.2403744259978806,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004496495609863361,
      "loss": 5.5126,
      "mean_token_accuracy": 0.19238028079271316,
      "num_tokens": 47298767.0,
      "step": 25370
    },
    {
      "entropy": 5.770969438552856,
      "epoch": 2.240815966089721,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044962970508386716,
      "loss": 5.3966,
      "mean_token_accuracy": 0.2008310854434967,
      "num_tokens": 47307950.0,
      "step": 25375
    },
    {
      "entropy": 5.77340669631958,
      "epoch": 2.241257506181561,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004496098457605016,
      "loss": 5.498,
      "mean_token_accuracy": 0.1964834973216057,
      "num_tokens": 47317963.0,
      "step": 25380
    },
    {
      "entropy": 5.832563829421997,
      "epoch": 2.2416990462734017,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004495899830166284,
      "loss": 5.4801,
      "mean_token_accuracy": 0.19169073700904846,
      "num_tokens": 47326728.0,
      "step": 25385
    },
    {
      "entropy": 5.964149141311646,
      "epoch": 2.242140586365242,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004495701168526368,
      "loss": 5.4938,
      "mean_token_accuracy": 0.18821703642606735,
      "num_tokens": 47336386.0,
      "step": 25390
    },
    {
      "entropy": 5.858840036392212,
      "epoch": 2.242582126457082,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044955024726891585,
      "loss": 5.4964,
      "mean_token_accuracy": 0.1849285289645195,
      "num_tokens": 47345958.0,
      "step": 25395
    },
    {
      "entropy": 5.8304114818573,
      "epoch": 2.2430236665489227,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004495303742658549,
      "loss": 5.5117,
      "mean_token_accuracy": 0.18737585395574569,
      "num_tokens": 47355758.0,
      "step": 25400
    },
    {
      "entropy": 5.8843621730804445,
      "epoch": 2.2434652066407628,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004495104978438431,
      "loss": 5.5241,
      "mean_token_accuracy": 0.190809927880764,
      "num_tokens": 47365186.0,
      "step": 25405
    },
    {
      "entropy": 5.794332361221313,
      "epoch": 2.2439067467326033,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00044949061800327,
      "loss": 5.4911,
      "mean_token_accuracy": 0.19598740339279175,
      "num_tokens": 47374443.0,
      "step": 25410
    },
    {
      "entropy": 5.734339046478271,
      "epoch": 2.2443482868244438,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004494707347445248,
      "loss": 5.4038,
      "mean_token_accuracy": 0.19521724283695222,
      "num_tokens": 47383369.0,
      "step": 25415
    },
    {
      "entropy": 5.770232534408569,
      "epoch": 2.244789826916284,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004494508480679972,
      "loss": 5.4511,
      "mean_token_accuracy": 0.19747331887483596,
      "num_tokens": 47393104.0,
      "step": 25420
    },
    {
      "entropy": 5.76449761390686,
      "epoch": 2.2452313670081243,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044943095797407676,
      "loss": 5.4336,
      "mean_token_accuracy": 0.1985210582613945,
      "num_tokens": 47401954.0,
      "step": 25425
    },
    {
      "entropy": 5.83755145072937,
      "epoch": 2.245672907099965,
      "grad_norm": 1.359375,
      "learning_rate": 0.000449411064463153,
      "loss": 5.4458,
      "mean_token_accuracy": 0.19645402878522872,
      "num_tokens": 47411463.0,
      "step": 25430
    },
    {
      "entropy": 5.782803344726562,
      "epoch": 2.246114447191805,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044939116753561564,
      "loss": 5.3757,
      "mean_token_accuracy": 0.1959196001291275,
      "num_tokens": 47421405.0,
      "step": 25435
    },
    {
      "entropy": 5.810201597213745,
      "epoch": 2.2465559872836454,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004493712671918546,
      "loss": 5.495,
      "mean_token_accuracy": 0.1954319953918457,
      "num_tokens": 47430268.0,
      "step": 25440
    },
    {
      "entropy": 5.850427913665771,
      "epoch": 2.246997527375486,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004493513634322595,
      "loss": 5.526,
      "mean_token_accuracy": 0.18183660507202148,
      "num_tokens": 47439571.0,
      "step": 25445
    },
    {
      "entropy": 5.815749597549439,
      "epoch": 2.247439067467326,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004493314562572204,
      "loss": 5.4728,
      "mean_token_accuracy": 0.19367465227842331,
      "num_tokens": 47449085.0,
      "step": 25450
    },
    {
      "entropy": 5.725525617599487,
      "epoch": 2.2478806075591664,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004493115456671272,
      "loss": 5.375,
      "mean_token_accuracy": 0.19903378337621688,
      "num_tokens": 47458740.0,
      "step": 25455
    },
    {
      "entropy": 5.804879522323608,
      "epoch": 2.248322147651007,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044929163166236997,
      "loss": 5.5071,
      "mean_token_accuracy": 0.19289081692695617,
      "num_tokens": 47467458.0,
      "step": 25460
    },
    {
      "entropy": 5.888482093811035,
      "epoch": 2.248763687742847,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004492717142433388,
      "loss": 5.4751,
      "mean_token_accuracy": 0.18869138062000274,
      "num_tokens": 47476845.0,
      "step": 25465
    },
    {
      "entropy": 5.810908842086792,
      "epoch": 2.2492052278346875,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044925179341042385,
      "loss": 5.466,
      "mean_token_accuracy": 0.19501332193613052,
      "num_tokens": 47485632.0,
      "step": 25470
    },
    {
      "entropy": 5.846181917190552,
      "epoch": 2.249646767926528,
      "grad_norm": 1.25,
      "learning_rate": 0.00044923186916401536,
      "loss": 5.4731,
      "mean_token_accuracy": 0.19316439032554628,
      "num_tokens": 47494681.0,
      "step": 25475
    },
    {
      "entropy": 5.73270959854126,
      "epoch": 2.250088308018368,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004492119415045036,
      "loss": 5.49,
      "mean_token_accuracy": 0.19691043645143508,
      "num_tokens": 47504482.0,
      "step": 25480
    },
    {
      "entropy": 5.750381231307983,
      "epoch": 2.2505298481102085,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004491920104322791,
      "loss": 5.4796,
      "mean_token_accuracy": 0.19634248167276383,
      "num_tokens": 47513601.0,
      "step": 25485
    },
    {
      "entropy": 5.840913343429565,
      "epoch": 2.2509713882020486,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004491720759477321,
      "loss": 5.4321,
      "mean_token_accuracy": 0.19842663109302522,
      "num_tokens": 47522811.0,
      "step": 25490
    },
    {
      "entropy": 5.833033895492553,
      "epoch": 2.251412928293889,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004491521380512533,
      "loss": 5.4579,
      "mean_token_accuracy": 0.1915070593357086,
      "num_tokens": 47532471.0,
      "step": 25495
    },
    {
      "entropy": 5.712247657775879,
      "epoch": 2.2518544683857296,
      "grad_norm": 1.3125,
      "learning_rate": 0.000449132196743233,
      "loss": 5.4675,
      "mean_token_accuracy": 0.20076923668384553,
      "num_tokens": 47541336.0,
      "step": 25500
    },
    {
      "entropy": 5.774264144897461,
      "epoch": 2.2522960084775696,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004491122520240621,
      "loss": 5.3979,
      "mean_token_accuracy": 0.19915998131036758,
      "num_tokens": 47551446.0,
      "step": 25505
    },
    {
      "entropy": 5.777588987350464,
      "epoch": 2.25273754856941,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004490923038941312,
      "loss": 5.4517,
      "mean_token_accuracy": 0.1953926518559456,
      "num_tokens": 47560596.0,
      "step": 25510
    },
    {
      "entropy": 5.90223126411438,
      "epoch": 2.2531790886612506,
      "grad_norm": 1.375,
      "learning_rate": 0.000449072352353831,
      "loss": 5.5637,
      "mean_token_accuracy": 0.1894544020295143,
      "num_tokens": 47570612.0,
      "step": 25515
    },
    {
      "entropy": 5.850888967514038,
      "epoch": 2.2536206287530907,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004490523974035525,
      "loss": 5.4953,
      "mean_token_accuracy": 0.19093021601438523,
      "num_tokens": 47580234.0,
      "step": 25520
    },
    {
      "entropy": 5.8616673946380615,
      "epoch": 2.254062168844931,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004490324390436866,
      "loss": 5.492,
      "mean_token_accuracy": 0.19487694054841995,
      "num_tokens": 47589777.0,
      "step": 25525
    },
    {
      "entropy": 5.812765598297119,
      "epoch": 2.254503708936771,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000449012477274624,
      "loss": 5.3955,
      "mean_token_accuracy": 0.20291702449321747,
      "num_tokens": 47597951.0,
      "step": 25530
    },
    {
      "entropy": 5.7980562210083,
      "epoch": 2.2549452490286117,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044899251209675606,
      "loss": 5.5257,
      "mean_token_accuracy": 0.19260721355676652,
      "num_tokens": 47606054.0,
      "step": 25535
    },
    {
      "entropy": 5.90138578414917,
      "epoch": 2.255386789120452,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004489725435104737,
      "loss": 5.4964,
      "mean_token_accuracy": 0.19121690541505815,
      "num_tokens": 47614704.0,
      "step": 25540
    },
    {
      "entropy": 5.766203737258911,
      "epoch": 2.2558283292122923,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004489525715161682,
      "loss": 5.4874,
      "mean_token_accuracy": 0.1950251504778862,
      "num_tokens": 47623189.0,
      "step": 25545
    },
    {
      "entropy": 5.794699621200562,
      "epoch": 2.2562698693041328,
      "grad_norm": 1.453125,
      "learning_rate": 0.00044893259611423076,
      "loss": 5.4621,
      "mean_token_accuracy": 0.1911979779601097,
      "num_tokens": 47633831.0,
      "step": 25550
    },
    {
      "entropy": 5.898767900466919,
      "epoch": 2.2567114093959733,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004489126173050527,
      "loss": 5.4135,
      "mean_token_accuracy": 0.19719500541687013,
      "num_tokens": 47642155.0,
      "step": 25555
    },
    {
      "entropy": 5.910791826248169,
      "epoch": 2.2571529494878133,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004488926350890253,
      "loss": 5.621,
      "mean_token_accuracy": 0.17774686366319656,
      "num_tokens": 47651042.0,
      "step": 25560
    },
    {
      "entropy": 5.872401857376099,
      "epoch": 2.257594489579654,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004488726494665402,
      "loss": 5.5511,
      "mean_token_accuracy": 0.18611159175634384,
      "num_tokens": 47661424.0,
      "step": 25565
    },
    {
      "entropy": 5.848009300231934,
      "epoch": 2.2580360296714943,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044885266043798866,
      "loss": 5.4827,
      "mean_token_accuracy": 0.19317018687725068,
      "num_tokens": 47671603.0,
      "step": 25570
    },
    {
      "entropy": 5.856680679321289,
      "epoch": 2.2584775697633344,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004488326680037624,
      "loss": 5.476,
      "mean_token_accuracy": 0.19141558408737183,
      "num_tokens": 47679861.0,
      "step": 25575
    },
    {
      "entropy": 5.696392059326172,
      "epoch": 2.258919109855175,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004488126721642531,
      "loss": 5.3846,
      "mean_token_accuracy": 0.19610664248466492,
      "num_tokens": 47689702.0,
      "step": 25580
    },
    {
      "entropy": 5.723728275299072,
      "epoch": 2.2593606499470154,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044879267291985243,
      "loss": 5.4585,
      "mean_token_accuracy": 0.18794025480747223,
      "num_tokens": 47699261.0,
      "step": 25585
    },
    {
      "entropy": 5.907329893112182,
      "epoch": 2.2598021900388554,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044877267027095203,
      "loss": 5.5435,
      "mean_token_accuracy": 0.18834418952465057,
      "num_tokens": 47709079.0,
      "step": 25590
    },
    {
      "entropy": 5.775431966781616,
      "epoch": 2.260243730130696,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004487526642179439,
      "loss": 5.4124,
      "mean_token_accuracy": 0.19673028886318206,
      "num_tokens": 47718641.0,
      "step": 25595
    },
    {
      "entropy": 5.875782203674317,
      "epoch": 2.2606852702225364,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044873265476121993,
      "loss": 5.5526,
      "mean_token_accuracy": 0.18769804239273072,
      "num_tokens": 47729156.0,
      "step": 25600
    },
    {
      "entropy": 5.811559247970581,
      "epoch": 2.2611268103143765,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044871264190117196,
      "loss": 5.4877,
      "mean_token_accuracy": 0.19046588242053986,
      "num_tokens": 47738424.0,
      "step": 25605
    },
    {
      "entropy": 5.852538108825684,
      "epoch": 2.261568350406217,
      "grad_norm": 1.375,
      "learning_rate": 0.0004486926256381922,
      "loss": 5.6185,
      "mean_token_accuracy": 0.17737132012844087,
      "num_tokens": 47749032.0,
      "step": 25610
    },
    {
      "entropy": 5.894007158279419,
      "epoch": 2.2620098904980575,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004486726059726727,
      "loss": 5.4909,
      "mean_token_accuracy": 0.19146089106798173,
      "num_tokens": 47758979.0,
      "step": 25615
    },
    {
      "entropy": 5.84003529548645,
      "epoch": 2.2624514305898975,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004486525829050056,
      "loss": 5.4733,
      "mean_token_accuracy": 0.18953819274902345,
      "num_tokens": 47768752.0,
      "step": 25620
    },
    {
      "entropy": 5.90866985321045,
      "epoch": 2.262892970681738,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004486325564355831,
      "loss": 5.5068,
      "mean_token_accuracy": 0.18795130997896195,
      "num_tokens": 47778175.0,
      "step": 25625
    },
    {
      "entropy": 5.770099830627442,
      "epoch": 2.263334510773578,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004486125265647977,
      "loss": 5.4781,
      "mean_token_accuracy": 0.19935380071401596,
      "num_tokens": 47787864.0,
      "step": 25630
    },
    {
      "entropy": 5.842896366119385,
      "epoch": 2.2637760508654186,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004485924932930416,
      "loss": 5.5113,
      "mean_token_accuracy": 0.18563829362392426,
      "num_tokens": 47796912.0,
      "step": 25635
    },
    {
      "entropy": 5.895245599746704,
      "epoch": 2.264217590957259,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004485724566207073,
      "loss": 5.5572,
      "mean_token_accuracy": 0.18363770097494125,
      "num_tokens": 47807452.0,
      "step": 25640
    },
    {
      "entropy": 5.903549003601074,
      "epoch": 2.264659131049099,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044855241654818723,
      "loss": 5.5827,
      "mean_token_accuracy": 0.18519142270088196,
      "num_tokens": 47817575.0,
      "step": 25645
    },
    {
      "entropy": 5.9211201667785645,
      "epoch": 2.2651006711409396,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004485323730758741,
      "loss": 5.4842,
      "mean_token_accuracy": 0.19600235670804977,
      "num_tokens": 47826177.0,
      "step": 25650
    },
    {
      "entropy": 5.782153511047364,
      "epoch": 2.26554221123278,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004485123262041604,
      "loss": 5.425,
      "mean_token_accuracy": 0.1882389724254608,
      "num_tokens": 47835622.0,
      "step": 25655
    },
    {
      "entropy": 5.832113790512085,
      "epoch": 2.26598375132462,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004484922759334389,
      "loss": 5.4746,
      "mean_token_accuracy": 0.19404146820306778,
      "num_tokens": 47845552.0,
      "step": 25660
    },
    {
      "entropy": 5.799503564834595,
      "epoch": 2.2664252914164607,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004484722222641024,
      "loss": 5.4353,
      "mean_token_accuracy": 0.19176384806632996,
      "num_tokens": 47853820.0,
      "step": 25665
    },
    {
      "entropy": 5.761084079742432,
      "epoch": 2.2668668315083007,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004484521651965438,
      "loss": 5.423,
      "mean_token_accuracy": 0.19777773469686508,
      "num_tokens": 47863416.0,
      "step": 25670
    },
    {
      "entropy": 5.873932933807373,
      "epoch": 2.267308371600141,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044843210473115584,
      "loss": 5.5023,
      "mean_token_accuracy": 0.19091290682554246,
      "num_tokens": 47871407.0,
      "step": 25675
    },
    {
      "entropy": 5.831028461456299,
      "epoch": 2.2677499116919817,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004484120408683316,
      "loss": 5.456,
      "mean_token_accuracy": 0.19168506264686586,
      "num_tokens": 47880742.0,
      "step": 25680
    },
    {
      "entropy": 5.794584131240844,
      "epoch": 2.2681914517838218,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044839197360846416,
      "loss": 5.5256,
      "mean_token_accuracy": 0.18471176028251649,
      "num_tokens": 47889573.0,
      "step": 25685
    },
    {
      "entropy": 5.874758195877075,
      "epoch": 2.2686329918756623,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044837190295194654,
      "loss": 5.4647,
      "mean_token_accuracy": 0.1902127519249916,
      "num_tokens": 47898855.0,
      "step": 25690
    },
    {
      "entropy": 5.877256298065186,
      "epoch": 2.2690745319675028,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044835182889917195,
      "loss": 5.5,
      "mean_token_accuracy": 0.1927628695964813,
      "num_tokens": 47908981.0,
      "step": 25695
    },
    {
      "entropy": 5.75017147064209,
      "epoch": 2.269516072059343,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004483317514505336,
      "loss": 5.4957,
      "mean_token_accuracy": 0.19619455933570862,
      "num_tokens": 47918313.0,
      "step": 25700
    },
    {
      "entropy": 5.743401622772216,
      "epoch": 2.2699576121511833,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004483116706064248,
      "loss": 5.4153,
      "mean_token_accuracy": 0.193598909676075,
      "num_tokens": 47927838.0,
      "step": 25705
    },
    {
      "entropy": 5.824086427688599,
      "epoch": 2.270399152243024,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004482915863672389,
      "loss": 5.4566,
      "mean_token_accuracy": 0.194514200091362,
      "num_tokens": 47937774.0,
      "step": 25710
    },
    {
      "entropy": 5.87535982131958,
      "epoch": 2.270840692334864,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004482714987333695,
      "loss": 5.4313,
      "mean_token_accuracy": 0.19510626196861267,
      "num_tokens": 47947091.0,
      "step": 25715
    },
    {
      "entropy": 5.783803796768188,
      "epoch": 2.2712822324267044,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044825140770520997,
      "loss": 5.3984,
      "mean_token_accuracy": 0.20082199275493623,
      "num_tokens": 47955715.0,
      "step": 25720
    },
    {
      "entropy": 5.664998006820679,
      "epoch": 2.271723772518545,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044823131328315387,
      "loss": 5.3465,
      "mean_token_accuracy": 0.2093337133526802,
      "num_tokens": 47965991.0,
      "step": 25725
    },
    {
      "entropy": 5.877130174636841,
      "epoch": 2.272165312610385,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004482112154675948,
      "loss": 5.5293,
      "mean_token_accuracy": 0.1891583561897278,
      "num_tokens": 47975610.0,
      "step": 25730
    },
    {
      "entropy": 5.844792032241822,
      "epoch": 2.2726068527022254,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044819111425892667,
      "loss": 5.4558,
      "mean_token_accuracy": 0.201219742000103,
      "num_tokens": 47984977.0,
      "step": 25735
    },
    {
      "entropy": 5.805962419509887,
      "epoch": 2.273048392794066,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044817100965754297,
      "loss": 5.3964,
      "mean_token_accuracy": 0.19502853751182556,
      "num_tokens": 47994205.0,
      "step": 25740
    },
    {
      "entropy": 5.750817394256591,
      "epoch": 2.273489932885906,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044815090166383775,
      "loss": 5.4898,
      "mean_token_accuracy": 0.19316319674253463,
      "num_tokens": 48003715.0,
      "step": 25745
    },
    {
      "entropy": 5.849080038070679,
      "epoch": 2.2739314729777464,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004481307902782048,
      "loss": 5.5856,
      "mean_token_accuracy": 0.1847197964787483,
      "num_tokens": 48013200.0,
      "step": 25750
    },
    {
      "entropy": 5.945283651351929,
      "epoch": 2.274373013069587,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004481106755010382,
      "loss": 5.4717,
      "mean_token_accuracy": 0.1915651321411133,
      "num_tokens": 48022366.0,
      "step": 25755
    },
    {
      "entropy": 5.862729454040528,
      "epoch": 2.274814553161427,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044809055733273183,
      "loss": 5.5127,
      "mean_token_accuracy": 0.1900693044066429,
      "num_tokens": 48031968.0,
      "step": 25760
    },
    {
      "entropy": 5.814993143081665,
      "epoch": 2.2752560932532675,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044807043577367996,
      "loss": 5.451,
      "mean_token_accuracy": 0.19703570157289504,
      "num_tokens": 48041069.0,
      "step": 25765
    },
    {
      "entropy": 5.66896424293518,
      "epoch": 2.2756976333451076,
      "grad_norm": 1.53125,
      "learning_rate": 0.00044805031082427666,
      "loss": 5.3752,
      "mean_token_accuracy": 0.19795995354652404,
      "num_tokens": 48050550.0,
      "step": 25770
    },
    {
      "entropy": 5.844475841522216,
      "epoch": 2.276139173436948,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044803018248491615,
      "loss": 5.4254,
      "mean_token_accuracy": 0.19506329596042632,
      "num_tokens": 48059971.0,
      "step": 25775
    },
    {
      "entropy": 5.832293701171875,
      "epoch": 2.2765807135287885,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004480100507559928,
      "loss": 5.3193,
      "mean_token_accuracy": 0.1938258245587349,
      "num_tokens": 48068566.0,
      "step": 25780
    },
    {
      "entropy": 5.704541635513306,
      "epoch": 2.2770222536206286,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000447989915637901,
      "loss": 5.3857,
      "mean_token_accuracy": 0.199770687520504,
      "num_tokens": 48078352.0,
      "step": 25785
    },
    {
      "entropy": 5.775565767288208,
      "epoch": 2.277463793712469,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004479697771310351,
      "loss": 5.477,
      "mean_token_accuracy": 0.1939407095313072,
      "num_tokens": 48087089.0,
      "step": 25790
    },
    {
      "entropy": 5.850719451904297,
      "epoch": 2.2779053338043096,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044794963523578965,
      "loss": 5.5222,
      "mean_token_accuracy": 0.19206084311008453,
      "num_tokens": 48096762.0,
      "step": 25795
    },
    {
      "entropy": 5.838743543624878,
      "epoch": 2.2783468738961496,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044792948995255917,
      "loss": 5.4512,
      "mean_token_accuracy": 0.20486154407262802,
      "num_tokens": 48105994.0,
      "step": 25800
    },
    {
      "entropy": 5.764887952804566,
      "epoch": 2.27878841398799,
      "grad_norm": 1.125,
      "learning_rate": 0.0004479093412817383,
      "loss": 5.4328,
      "mean_token_accuracy": 0.20131944566965104,
      "num_tokens": 48115148.0,
      "step": 25805
    },
    {
      "entropy": 5.792564964294433,
      "epoch": 2.27922995407983,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044788918922372196,
      "loss": 5.5149,
      "mean_token_accuracy": 0.18834081292152405,
      "num_tokens": 48124696.0,
      "step": 25810
    },
    {
      "entropy": 5.8370733737945555,
      "epoch": 2.2796714941716707,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004478690337789046,
      "loss": 5.4204,
      "mean_token_accuracy": 0.1984834522008896,
      "num_tokens": 48134165.0,
      "step": 25815
    },
    {
      "entropy": 5.878579902648926,
      "epoch": 2.280113034263511,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044784887494768113,
      "loss": 5.4594,
      "mean_token_accuracy": 0.18531199395656586,
      "num_tokens": 48144080.0,
      "step": 25820
    },
    {
      "entropy": 5.764784860610962,
      "epoch": 2.2805545743553512,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044782871273044654,
      "loss": 5.4905,
      "mean_token_accuracy": 0.20070475041866304,
      "num_tokens": 48156161.0,
      "step": 25825
    },
    {
      "entropy": 5.850653171539307,
      "epoch": 2.2809961144471917,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004478085471275958,
      "loss": 5.4876,
      "mean_token_accuracy": 0.18649315387010573,
      "num_tokens": 48165317.0,
      "step": 25830
    },
    {
      "entropy": 5.855970335006714,
      "epoch": 2.2814376545390322,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004477883781395239,
      "loss": 5.434,
      "mean_token_accuracy": 0.19774698615074157,
      "num_tokens": 48173639.0,
      "step": 25835
    },
    {
      "entropy": 5.790425157546997,
      "epoch": 2.2818791946308723,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044776820576662593,
      "loss": 5.4668,
      "mean_token_accuracy": 0.1975530594587326,
      "num_tokens": 48183596.0,
      "step": 25840
    },
    {
      "entropy": 5.896965265274048,
      "epoch": 2.282320734722713,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004477480300092971,
      "loss": 5.5213,
      "mean_token_accuracy": 0.196102474629879,
      "num_tokens": 48194269.0,
      "step": 25845
    },
    {
      "entropy": 5.865816354751587,
      "epoch": 2.2827622748145533,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044772785086793257,
      "loss": 5.4754,
      "mean_token_accuracy": 0.19177181869745255,
      "num_tokens": 48203180.0,
      "step": 25850
    },
    {
      "entropy": 5.811283826828003,
      "epoch": 2.2832038149063933,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044770766834292765,
      "loss": 5.4863,
      "mean_token_accuracy": 0.1945757433772087,
      "num_tokens": 48212768.0,
      "step": 25855
    },
    {
      "entropy": 5.813453960418701,
      "epoch": 2.283645354998234,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004476874824346777,
      "loss": 5.5,
      "mean_token_accuracy": 0.18934702277183532,
      "num_tokens": 48221620.0,
      "step": 25860
    },
    {
      "entropy": 5.85723123550415,
      "epoch": 2.2840868950900743,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044766729314357827,
      "loss": 5.5119,
      "mean_token_accuracy": 0.18555721640586853,
      "num_tokens": 48231379.0,
      "step": 25865
    },
    {
      "entropy": 5.850969219207764,
      "epoch": 2.2845284351819144,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004476471004700248,
      "loss": 5.53,
      "mean_token_accuracy": 0.19127794355154037,
      "num_tokens": 48241827.0,
      "step": 25870
    },
    {
      "entropy": 5.853958225250244,
      "epoch": 2.284969975273755,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004476269044144127,
      "loss": 5.4649,
      "mean_token_accuracy": 0.18532384037971497,
      "num_tokens": 48250747.0,
      "step": 25875
    },
    {
      "entropy": 5.842100954055786,
      "epoch": 2.2854115153655954,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004476067049771378,
      "loss": 5.4664,
      "mean_token_accuracy": 0.18382894098758698,
      "num_tokens": 48260909.0,
      "step": 25880
    },
    {
      "entropy": 5.828786945343017,
      "epoch": 2.2858530554574354,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004475865021585957,
      "loss": 5.5294,
      "mean_token_accuracy": 0.18705177456140518,
      "num_tokens": 48270974.0,
      "step": 25885
    },
    {
      "entropy": 5.74724497795105,
      "epoch": 2.286294595549276,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004475662959591822,
      "loss": 5.4015,
      "mean_token_accuracy": 0.18853551894426346,
      "num_tokens": 48279466.0,
      "step": 25890
    },
    {
      "entropy": 5.835370826721191,
      "epoch": 2.2867361356411164,
      "grad_norm": 1.1875,
      "learning_rate": 0.000447546086379293,
      "loss": 5.5087,
      "mean_token_accuracy": 0.18842430859804155,
      "num_tokens": 48289830.0,
      "step": 25895
    },
    {
      "entropy": 5.8478161811828615,
      "epoch": 2.2871776757329565,
      "grad_norm": 1.25,
      "learning_rate": 0.0004475258734193242,
      "loss": 5.4571,
      "mean_token_accuracy": 0.1980438932776451,
      "num_tokens": 48299493.0,
      "step": 25900
    },
    {
      "entropy": 5.818922662734986,
      "epoch": 2.287619215824797,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004475056570796717,
      "loss": 5.4928,
      "mean_token_accuracy": 0.1978315830230713,
      "num_tokens": 48308366.0,
      "step": 25905
    },
    {
      "entropy": 5.797279500961304,
      "epoch": 2.288060755916637,
      "grad_norm": 1.125,
      "learning_rate": 0.0004474854373607314,
      "loss": 5.4678,
      "mean_token_accuracy": 0.1920726180076599,
      "num_tokens": 48317760.0,
      "step": 25910
    },
    {
      "entropy": 5.824224233627319,
      "epoch": 2.2885022960084775,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004474652142628995,
      "loss": 5.4704,
      "mean_token_accuracy": 0.18624207675457,
      "num_tokens": 48325847.0,
      "step": 25915
    },
    {
      "entropy": 5.865761709213257,
      "epoch": 2.288943836100318,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004474449877865722,
      "loss": 5.4947,
      "mean_token_accuracy": 0.18513823449611663,
      "num_tokens": 48334938.0,
      "step": 25920
    },
    {
      "entropy": 5.794952869415283,
      "epoch": 2.289385376192158,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004474247579321456,
      "loss": 5.477,
      "mean_token_accuracy": 0.19102026522159576,
      "num_tokens": 48344414.0,
      "step": 25925
    },
    {
      "entropy": 5.761966896057129,
      "epoch": 2.2898269162839986,
      "grad_norm": 1.140625,
      "learning_rate": 0.000447404524700016,
      "loss": 5.4173,
      "mean_token_accuracy": 0.19473017007112503,
      "num_tokens": 48353711.0,
      "step": 25930
    },
    {
      "entropy": 5.756228303909301,
      "epoch": 2.290268456375839,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004473842880905799,
      "loss": 5.3804,
      "mean_token_accuracy": 0.2050639122724533,
      "num_tokens": 48362907.0,
      "step": 25935
    },
    {
      "entropy": 5.876056766510009,
      "epoch": 2.290709996467679,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044736404810423366,
      "loss": 5.4487,
      "mean_token_accuracy": 0.19491581320762635,
      "num_tokens": 48372450.0,
      "step": 25940
    },
    {
      "entropy": 5.727972602844238,
      "epoch": 2.2911515365595196,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004473438047413737,
      "loss": 5.3948,
      "mean_token_accuracy": 0.19993823021650314,
      "num_tokens": 48381521.0,
      "step": 25945
    },
    {
      "entropy": 5.761495637893677,
      "epoch": 2.2915930766513597,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004473235580023966,
      "loss": 5.4254,
      "mean_token_accuracy": 0.20014556497335434,
      "num_tokens": 48390635.0,
      "step": 25950
    },
    {
      "entropy": 5.820792388916016,
      "epoch": 2.2920346167432,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044730330788769903,
      "loss": 5.4937,
      "mean_token_accuracy": 0.18611569702625275,
      "num_tokens": 48400199.0,
      "step": 25955
    },
    {
      "entropy": 5.786763668060303,
      "epoch": 2.2924761568350407,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044728305439767764,
      "loss": 5.4425,
      "mean_token_accuracy": 0.19558358043432236,
      "num_tokens": 48409266.0,
      "step": 25960
    },
    {
      "entropy": 5.817392301559448,
      "epoch": 2.2929176969268807,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004472627975327293,
      "loss": 5.4861,
      "mean_token_accuracy": 0.19454698860645295,
      "num_tokens": 48419195.0,
      "step": 25965
    },
    {
      "entropy": 5.900789451599121,
      "epoch": 2.2933592370187212,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004472425372932506,
      "loss": 5.564,
      "mean_token_accuracy": 0.18804969936609267,
      "num_tokens": 48430637.0,
      "step": 25970
    },
    {
      "entropy": 5.735941076278687,
      "epoch": 2.2938007771105617,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004472222736796386,
      "loss": 5.335,
      "mean_token_accuracy": 0.20890549272298814,
      "num_tokens": 48439929.0,
      "step": 25975
    },
    {
      "entropy": 5.8466877937316895,
      "epoch": 2.294242317202402,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004472020066922902,
      "loss": 5.5812,
      "mean_token_accuracy": 0.18509045094251633,
      "num_tokens": 48449398.0,
      "step": 25980
    },
    {
      "entropy": 5.797989845275879,
      "epoch": 2.2946838572942423,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004471817363316024,
      "loss": 5.4665,
      "mean_token_accuracy": 0.19597396850585938,
      "num_tokens": 48458843.0,
      "step": 25985
    },
    {
      "entropy": 5.818384742736816,
      "epoch": 2.295125397386083,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004471614625979723,
      "loss": 5.3983,
      "mean_token_accuracy": 0.19862309843301773,
      "num_tokens": 48468111.0,
      "step": 25990
    },
    {
      "entropy": 5.800868892669678,
      "epoch": 2.295566937477923,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004471411854917972,
      "loss": 5.4601,
      "mean_token_accuracy": 0.199945567548275,
      "num_tokens": 48477501.0,
      "step": 25995
    },
    {
      "entropy": 5.926042270660401,
      "epoch": 2.2960084775697633,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000447120905013474,
      "loss": 5.6086,
      "mean_token_accuracy": 0.18069386780261992,
      "num_tokens": 48487576.0,
      "step": 26000
    },
    {
      "entropy": 5.751308155059815,
      "epoch": 2.296450017661604,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004471006211634002,
      "loss": 5.3812,
      "mean_token_accuracy": 0.20907239317893983,
      "num_tokens": 48496230.0,
      "step": 26005
    },
    {
      "entropy": 5.808518791198731,
      "epoch": 2.296891557753444,
      "grad_norm": 1.25,
      "learning_rate": 0.00044708033394197323,
      "loss": 5.4883,
      "mean_token_accuracy": 0.1904683381319046,
      "num_tokens": 48505871.0,
      "step": 26010
    },
    {
      "entropy": 5.846672439575196,
      "epoch": 2.2973330978452844,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044706004334959024,
      "loss": 5.4995,
      "mean_token_accuracy": 0.18570457994937897,
      "num_tokens": 48515298.0,
      "step": 26015
    },
    {
      "entropy": 5.77729902267456,
      "epoch": 2.297774637937125,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004470397493866489,
      "loss": 5.5007,
      "mean_token_accuracy": 0.18769943118095397,
      "num_tokens": 48524573.0,
      "step": 26020
    },
    {
      "entropy": 5.803730010986328,
      "epoch": 2.298216178028965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004470194520535467,
      "loss": 5.5468,
      "mean_token_accuracy": 0.18673465102910997,
      "num_tokens": 48533244.0,
      "step": 26025
    },
    {
      "entropy": 5.9086791515350345,
      "epoch": 2.2986577181208054,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004469991513506814,
      "loss": 5.4186,
      "mean_token_accuracy": 0.19607501327991486,
      "num_tokens": 48542464.0,
      "step": 26030
    },
    {
      "entropy": 5.803478479385376,
      "epoch": 2.299099258212646,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004469788472784504,
      "loss": 5.4947,
      "mean_token_accuracy": 0.19603319317102433,
      "num_tokens": 48551947.0,
      "step": 26035
    },
    {
      "entropy": 5.7796834945678714,
      "epoch": 2.299540798304486,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044695853983725163,
      "loss": 5.4127,
      "mean_token_accuracy": 0.1991957902908325,
      "num_tokens": 48561287.0,
      "step": 26040
    },
    {
      "entropy": 5.817529344558716,
      "epoch": 2.2999823383963265,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004469382290274829,
      "loss": 5.5521,
      "mean_token_accuracy": 0.18562526404857635,
      "num_tokens": 48570800.0,
      "step": 26045
    },
    {
      "entropy": 5.824733686447144,
      "epoch": 2.3004238784881665,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044691791484954195,
      "loss": 5.3945,
      "mean_token_accuracy": 0.2001741573214531,
      "num_tokens": 48579705.0,
      "step": 26050
    },
    {
      "entropy": 5.837799072265625,
      "epoch": 2.300865418580007,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000446897597303827,
      "loss": 5.5212,
      "mean_token_accuracy": 0.19584323465824127,
      "num_tokens": 48588793.0,
      "step": 26055
    },
    {
      "entropy": 5.787786960601807,
      "epoch": 2.3013069586718475,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004468772763907357,
      "loss": 5.3762,
      "mean_token_accuracy": 0.1980183780193329,
      "num_tokens": 48598361.0,
      "step": 26060
    },
    {
      "entropy": 5.783269357681275,
      "epoch": 2.3017484987636876,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004468569521106664,
      "loss": 5.4865,
      "mean_token_accuracy": 0.18588589429855346,
      "num_tokens": 48607703.0,
      "step": 26065
    },
    {
      "entropy": 5.761405372619629,
      "epoch": 2.302190038855528,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004468366244640171,
      "loss": 5.4344,
      "mean_token_accuracy": 0.19833621829748155,
      "num_tokens": 48616345.0,
      "step": 26070
    },
    {
      "entropy": 5.8821690559387205,
      "epoch": 2.3026315789473686,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044681629345118615,
      "loss": 5.5037,
      "mean_token_accuracy": 0.1889692485332489,
      "num_tokens": 48624907.0,
      "step": 26075
    },
    {
      "entropy": 5.808244466781616,
      "epoch": 2.3030731190392086,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004467959590725716,
      "loss": 5.4392,
      "mean_token_accuracy": 0.19780541956424713,
      "num_tokens": 48634766.0,
      "step": 26080
    },
    {
      "entropy": 5.808417320251465,
      "epoch": 2.303514659131049,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044677562132857204,
      "loss": 5.4682,
      "mean_token_accuracy": 0.19334697723388672,
      "num_tokens": 48644321.0,
      "step": 26085
    },
    {
      "entropy": 5.780827522277832,
      "epoch": 2.303956199222889,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044675528021958567,
      "loss": 5.4685,
      "mean_token_accuracy": 0.1972940370440483,
      "num_tokens": 48653507.0,
      "step": 26090
    },
    {
      "entropy": 5.79469051361084,
      "epoch": 2.3043977393147297,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000446734935746011,
      "loss": 5.458,
      "mean_token_accuracy": 0.19987148344516753,
      "num_tokens": 48662186.0,
      "step": 26095
    },
    {
      "entropy": 5.787240028381348,
      "epoch": 2.30483927940657,
      "grad_norm": 1.5,
      "learning_rate": 0.0004467145879082467,
      "loss": 5.4548,
      "mean_token_accuracy": 0.1982992023229599,
      "num_tokens": 48671605.0,
      "step": 26100
    },
    {
      "entropy": 5.763897848129273,
      "epoch": 2.3052808194984102,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004466942367066912,
      "loss": 5.3863,
      "mean_token_accuracy": 0.1970063790678978,
      "num_tokens": 48679873.0,
      "step": 26105
    },
    {
      "entropy": 5.796825647354126,
      "epoch": 2.3057223595902507,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044667388214174336,
      "loss": 5.3663,
      "mean_token_accuracy": 0.20377321392297745,
      "num_tokens": 48688678.0,
      "step": 26110
    },
    {
      "entropy": 5.832673263549805,
      "epoch": 2.3061638996820912,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004466535242138018,
      "loss": 5.4521,
      "mean_token_accuracy": 0.19757949113845824,
      "num_tokens": 48697794.0,
      "step": 26115
    },
    {
      "entropy": 5.857972049713135,
      "epoch": 2.3066054397739313,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044663316292326526,
      "loss": 5.6135,
      "mean_token_accuracy": 0.18361402451992034,
      "num_tokens": 48706859.0,
      "step": 26120
    },
    {
      "entropy": 5.817218208312989,
      "epoch": 2.307046979865772,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044661279827053267,
      "loss": 5.4319,
      "mean_token_accuracy": 0.19962913990020753,
      "num_tokens": 48716627.0,
      "step": 26125
    },
    {
      "entropy": 5.8496666431427,
      "epoch": 2.3074885199576123,
      "grad_norm": 2.359375,
      "learning_rate": 0.00044659243025600297,
      "loss": 5.4721,
      "mean_token_accuracy": 0.1911683574318886,
      "num_tokens": 48726514.0,
      "step": 26130
    },
    {
      "entropy": 5.793008756637573,
      "epoch": 2.3079300600494523,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044657205888007525,
      "loss": 5.4453,
      "mean_token_accuracy": 0.19732450544834138,
      "num_tokens": 48735546.0,
      "step": 26135
    },
    {
      "entropy": 5.729039239883423,
      "epoch": 2.308371600141293,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004465516841431483,
      "loss": 5.3609,
      "mean_token_accuracy": 0.2047617107629776,
      "num_tokens": 48744719.0,
      "step": 26140
    },
    {
      "entropy": 5.864201498031616,
      "epoch": 2.3088131402331333,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004465313060456216,
      "loss": 5.4643,
      "mean_token_accuracy": 0.19282019287347793,
      "num_tokens": 48754201.0,
      "step": 26145
    },
    {
      "entropy": 5.808637905120849,
      "epoch": 2.3092546803249734,
      "grad_norm": 1.84375,
      "learning_rate": 0.00044651092458789405,
      "loss": 5.4664,
      "mean_token_accuracy": 0.18865385502576829,
      "num_tokens": 48763552.0,
      "step": 26150
    },
    {
      "entropy": 5.732302856445313,
      "epoch": 2.309696220416814,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004464905397703651,
      "loss": 5.4564,
      "mean_token_accuracy": 0.19386611133813858,
      "num_tokens": 48772494.0,
      "step": 26155
    },
    {
      "entropy": 5.8114790439605715,
      "epoch": 2.3101377605086544,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044647015159343396,
      "loss": 5.4765,
      "mean_token_accuracy": 0.19221725761890412,
      "num_tokens": 48782121.0,
      "step": 26160
    },
    {
      "entropy": 5.8091401100158695,
      "epoch": 2.3105793006004944,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044644976005750003,
      "loss": 5.3779,
      "mean_token_accuracy": 0.20416889637708663,
      "num_tokens": 48792040.0,
      "step": 26165
    },
    {
      "entropy": 5.770216369628907,
      "epoch": 2.311020840692335,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004464293651629629,
      "loss": 5.446,
      "mean_token_accuracy": 0.19817576557397842,
      "num_tokens": 48801429.0,
      "step": 26170
    },
    {
      "entropy": 5.797077369689942,
      "epoch": 2.3114623807841754,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044640896691022195,
      "loss": 5.4791,
      "mean_token_accuracy": 0.19153914600610733,
      "num_tokens": 48811185.0,
      "step": 26175
    },
    {
      "entropy": 5.888717317581177,
      "epoch": 2.3119039208760155,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004463885652996768,
      "loss": 5.5328,
      "mean_token_accuracy": 0.1898137614130974,
      "num_tokens": 48820691.0,
      "step": 26180
    },
    {
      "entropy": 5.864487409591675,
      "epoch": 2.312345460967856,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044636816033172716,
      "loss": 5.5386,
      "mean_token_accuracy": 0.1875033974647522,
      "num_tokens": 48829960.0,
      "step": 26185
    },
    {
      "entropy": 5.802185153961181,
      "epoch": 2.312787001059696,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004463477520067728,
      "loss": 5.4299,
      "mean_token_accuracy": 0.18894340544939042,
      "num_tokens": 48839107.0,
      "step": 26190
    },
    {
      "entropy": 5.887163352966309,
      "epoch": 2.3132285411515365,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004463273403252133,
      "loss": 5.519,
      "mean_token_accuracy": 0.1928054139018059,
      "num_tokens": 48848430.0,
      "step": 26195
    },
    {
      "entropy": 5.7720277309417725,
      "epoch": 2.313670081243377,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004463069252874487,
      "loss": 5.4029,
      "mean_token_accuracy": 0.19546396285295486,
      "num_tokens": 48857869.0,
      "step": 26200
    },
    {
      "entropy": 5.804399204254151,
      "epoch": 2.314111621335217,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044628650689387887,
      "loss": 5.5292,
      "mean_token_accuracy": 0.18817066699266433,
      "num_tokens": 48867730.0,
      "step": 26205
    },
    {
      "entropy": 5.898250961303711,
      "epoch": 2.3145531614270576,
      "grad_norm": 2.5625,
      "learning_rate": 0.00044626608514490384,
      "loss": 5.5022,
      "mean_token_accuracy": 0.18918419927358626,
      "num_tokens": 48876615.0,
      "step": 26210
    },
    {
      "entropy": 5.836344909667969,
      "epoch": 2.314994701518898,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004462456600409235,
      "loss": 5.4942,
      "mean_token_accuracy": 0.19699017703533173,
      "num_tokens": 48885732.0,
      "step": 26215
    },
    {
      "entropy": 5.81521258354187,
      "epoch": 2.315436241610738,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004462252315823381,
      "loss": 5.4099,
      "mean_token_accuracy": 0.19746698439121246,
      "num_tokens": 48894372.0,
      "step": 26220
    },
    {
      "entropy": 5.8408623218536375,
      "epoch": 2.3158777817025786,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004462047997695478,
      "loss": 5.5493,
      "mean_token_accuracy": 0.18693243563175202,
      "num_tokens": 48904321.0,
      "step": 26225
    },
    {
      "entropy": 5.820669364929199,
      "epoch": 2.3163193217944187,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044618436460295294,
      "loss": 5.501,
      "mean_token_accuracy": 0.19121033251285552,
      "num_tokens": 48914215.0,
      "step": 26230
    },
    {
      "entropy": 5.844702863693238,
      "epoch": 2.316760861886259,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004461639260829535,
      "loss": 5.4541,
      "mean_token_accuracy": 0.18890009224414825,
      "num_tokens": 48923383.0,
      "step": 26235
    },
    {
      "entropy": 5.892622947692871,
      "epoch": 2.3172024019780997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004461434842099503,
      "loss": 5.4835,
      "mean_token_accuracy": 0.1886167749762535,
      "num_tokens": 48933560.0,
      "step": 26240
    },
    {
      "entropy": 5.700312995910645,
      "epoch": 2.3176439420699397,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004461230389843435,
      "loss": 5.4032,
      "mean_token_accuracy": 0.1998525321483612,
      "num_tokens": 48943111.0,
      "step": 26245
    },
    {
      "entropy": 5.793135118484497,
      "epoch": 2.3180854821617802,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004461025904065338,
      "loss": 5.4641,
      "mean_token_accuracy": 0.19515701234340668,
      "num_tokens": 48952470.0,
      "step": 26250
    },
    {
      "entropy": 5.889206409454346,
      "epoch": 2.3185270222536207,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044608213847692154,
      "loss": 5.5031,
      "mean_token_accuracy": 0.19048506170511245,
      "num_tokens": 48961273.0,
      "step": 26255
    },
    {
      "entropy": 5.875008201599121,
      "epoch": 2.318968562345461,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004460616831959075,
      "loss": 5.5282,
      "mean_token_accuracy": 0.19091587662696838,
      "num_tokens": 48971078.0,
      "step": 26260
    },
    {
      "entropy": 5.783533573150635,
      "epoch": 2.3194101024373013,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004460412245638924,
      "loss": 5.506,
      "mean_token_accuracy": 0.19132618308067323,
      "num_tokens": 48980711.0,
      "step": 26265
    },
    {
      "entropy": 5.786983776092529,
      "epoch": 2.3198516425291418,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000446020762581277,
      "loss": 5.4147,
      "mean_token_accuracy": 0.1935329869389534,
      "num_tokens": 48990195.0,
      "step": 26270
    },
    {
      "entropy": 5.678952550888061,
      "epoch": 2.320293182620982,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044600029724846206,
      "loss": 5.3206,
      "mean_token_accuracy": 0.2075416326522827,
      "num_tokens": 49000271.0,
      "step": 26275
    },
    {
      "entropy": 5.778700876235962,
      "epoch": 2.3207347227128223,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004459798285658487,
      "loss": 5.4481,
      "mean_token_accuracy": 0.18878479599952697,
      "num_tokens": 49009624.0,
      "step": 26280
    },
    {
      "entropy": 5.729268407821655,
      "epoch": 2.321176262804663,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044595935653383756,
      "loss": 5.4266,
      "mean_token_accuracy": 0.19408130496740342,
      "num_tokens": 49018845.0,
      "step": 26285
    },
    {
      "entropy": 5.7314056873321535,
      "epoch": 2.321617802896503,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044593888115282995,
      "loss": 5.3871,
      "mean_token_accuracy": 0.2050270602107048,
      "num_tokens": 49028352.0,
      "step": 26290
    },
    {
      "entropy": 5.782250022888183,
      "epoch": 2.3220593429883434,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044591840242322686,
      "loss": 5.489,
      "mean_token_accuracy": 0.19465503096580505,
      "num_tokens": 49038180.0,
      "step": 26295
    },
    {
      "entropy": 5.803694820404052,
      "epoch": 2.322500883080184,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004458979203454295,
      "loss": 5.3693,
      "mean_token_accuracy": 0.20835692286491395,
      "num_tokens": 49048038.0,
      "step": 26300
    },
    {
      "entropy": 5.826607322692871,
      "epoch": 2.322942423172024,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044587743491983903,
      "loss": 5.4657,
      "mean_token_accuracy": 0.19492330253124238,
      "num_tokens": 49057304.0,
      "step": 26305
    },
    {
      "entropy": 5.798757934570313,
      "epoch": 2.3233839632638644,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004458569461468568,
      "loss": 5.5649,
      "mean_token_accuracy": 0.188813716173172,
      "num_tokens": 49067071.0,
      "step": 26310
    },
    {
      "entropy": 5.8162391662597654,
      "epoch": 2.323825503355705,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044583645402688414,
      "loss": 5.4911,
      "mean_token_accuracy": 0.1947309747338295,
      "num_tokens": 49075814.0,
      "step": 26315
    },
    {
      "entropy": 5.932412767410279,
      "epoch": 2.324267043447545,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004458159585603226,
      "loss": 5.5689,
      "mean_token_accuracy": 0.18717724829912186,
      "num_tokens": 49085831.0,
      "step": 26320
    },
    {
      "entropy": 5.889236402511597,
      "epoch": 2.3247085835393855,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00044579545974757353,
      "loss": 5.476,
      "mean_token_accuracy": 0.1929934337735176,
      "num_tokens": 49095893.0,
      "step": 26325
    },
    {
      "entropy": 5.727242851257325,
      "epoch": 2.3251501236312255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004457749575890385,
      "loss": 5.4323,
      "mean_token_accuracy": 0.19429387599229814,
      "num_tokens": 49104810.0,
      "step": 26330
    },
    {
      "entropy": 5.733926773071289,
      "epoch": 2.325591663723066,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004457544520851192,
      "loss": 5.4942,
      "mean_token_accuracy": 0.1881291002035141,
      "num_tokens": 49114108.0,
      "step": 26335
    },
    {
      "entropy": 5.84914755821228,
      "epoch": 2.3260332038149065,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044573394323621735,
      "loss": 5.4418,
      "mean_token_accuracy": 0.19207162111997605,
      "num_tokens": 49123371.0,
      "step": 26340
    },
    {
      "entropy": 5.855066919326783,
      "epoch": 2.3264747439067466,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004457134310427346,
      "loss": 5.4848,
      "mean_token_accuracy": 0.193508842587471,
      "num_tokens": 49132056.0,
      "step": 26345
    },
    {
      "entropy": 5.805853843688965,
      "epoch": 2.326916283998587,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004456929155050729,
      "loss": 5.4875,
      "mean_token_accuracy": 0.18859068751335145,
      "num_tokens": 49141316.0,
      "step": 26350
    },
    {
      "entropy": 5.797541856765747,
      "epoch": 2.3273578240904276,
      "grad_norm": 1.125,
      "learning_rate": 0.00044567239662363393,
      "loss": 5.4266,
      "mean_token_accuracy": 0.20104713290929793,
      "num_tokens": 49151337.0,
      "step": 26355
    },
    {
      "entropy": 5.904226589202881,
      "epoch": 2.3277993641822676,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004456518743988199,
      "loss": 5.4481,
      "mean_token_accuracy": 0.19236117601394653,
      "num_tokens": 49160970.0,
      "step": 26360
    },
    {
      "entropy": 5.807942056655884,
      "epoch": 2.328240904274108,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004456313488310326,
      "loss": 5.4406,
      "mean_token_accuracy": 0.19995508939027787,
      "num_tokens": 49170727.0,
      "step": 26365
    },
    {
      "entropy": 5.791996669769287,
      "epoch": 2.328682444365948,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044561081992067425,
      "loss": 5.4956,
      "mean_token_accuracy": 0.18973645567893982,
      "num_tokens": 49179485.0,
      "step": 26370
    },
    {
      "entropy": 5.844925785064698,
      "epoch": 2.3291239844577887,
      "grad_norm": 1.09375,
      "learning_rate": 0.000445590287668147,
      "loss": 5.5577,
      "mean_token_accuracy": 0.18149023205041886,
      "num_tokens": 49189339.0,
      "step": 26375
    },
    {
      "entropy": 5.806834173202515,
      "epoch": 2.329565524549629,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044556975207385305,
      "loss": 5.4123,
      "mean_token_accuracy": 0.19929081052541733,
      "num_tokens": 49198387.0,
      "step": 26380
    },
    {
      "entropy": 5.789703512191773,
      "epoch": 2.3300070646414692,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044554921313819465,
      "loss": 5.4983,
      "mean_token_accuracy": 0.18572574257850646,
      "num_tokens": 49207518.0,
      "step": 26385
    },
    {
      "entropy": 5.747111988067627,
      "epoch": 2.3304486047333097,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004455286708615742,
      "loss": 5.3937,
      "mean_token_accuracy": 0.2016517251729965,
      "num_tokens": 49217031.0,
      "step": 26390
    },
    {
      "entropy": 5.867904615402222,
      "epoch": 2.3308901448251502,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000445508125244394,
      "loss": 5.4778,
      "mean_token_accuracy": 0.19199148714542388,
      "num_tokens": 49225500.0,
      "step": 26395
    },
    {
      "entropy": 5.822998809814453,
      "epoch": 2.3313316849169903,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044548757628705654,
      "loss": 5.3949,
      "mean_token_accuracy": 0.20013906806707382,
      "num_tokens": 49235088.0,
      "step": 26400
    },
    {
      "entropy": 5.749777030944824,
      "epoch": 2.3317732250088308,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004454670239899646,
      "loss": 5.5065,
      "mean_token_accuracy": 0.1874376803636551,
      "num_tokens": 49245597.0,
      "step": 26405
    },
    {
      "entropy": 5.748863172531128,
      "epoch": 2.3322147651006713,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044544646835352047,
      "loss": 5.3129,
      "mean_token_accuracy": 0.21170300841331482,
      "num_tokens": 49254284.0,
      "step": 26410
    },
    {
      "entropy": 5.867848253250122,
      "epoch": 2.3326563051925113,
      "grad_norm": 1.21875,
      "learning_rate": 0.000445425909378127,
      "loss": 5.5447,
      "mean_token_accuracy": 0.18970918655395508,
      "num_tokens": 49264226.0,
      "step": 26415
    },
    {
      "entropy": 5.794131135940551,
      "epoch": 2.333097845284352,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044540534706418697,
      "loss": 5.4811,
      "mean_token_accuracy": 0.1972442612051964,
      "num_tokens": 49274022.0,
      "step": 26420
    },
    {
      "entropy": 5.856608438491821,
      "epoch": 2.3335393853761923,
      "grad_norm": 1.359375,
      "learning_rate": 0.000445384781412103,
      "loss": 5.467,
      "mean_token_accuracy": 0.19661829024553298,
      "num_tokens": 49283376.0,
      "step": 26425
    },
    {
      "entropy": 5.842105293273926,
      "epoch": 2.3339809254680324,
      "grad_norm": 1.296875,
      "learning_rate": 0.000445364212422278,
      "loss": 5.5107,
      "mean_token_accuracy": 0.18965207934379577,
      "num_tokens": 49292994.0,
      "step": 26430
    },
    {
      "entropy": 5.905291128158569,
      "epoch": 2.334422465559873,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004453436400951152,
      "loss": 5.5642,
      "mean_token_accuracy": 0.19325237423181535,
      "num_tokens": 49303599.0,
      "step": 26435
    },
    {
      "entropy": 5.867442798614502,
      "epoch": 2.3348640056517134,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004453230644310172,
      "loss": 5.5402,
      "mean_token_accuracy": 0.19058137536048889,
      "num_tokens": 49313397.0,
      "step": 26440
    },
    {
      "entropy": 5.844014263153076,
      "epoch": 2.3353055457435534,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044530248543038725,
      "loss": 5.4701,
      "mean_token_accuracy": 0.1924525812268257,
      "num_tokens": 49323288.0,
      "step": 26445
    },
    {
      "entropy": 5.705079984664917,
      "epoch": 2.335747085835394,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044528190309362853,
      "loss": 5.3327,
      "mean_token_accuracy": 0.21225903183221817,
      "num_tokens": 49332153.0,
      "step": 26450
    },
    {
      "entropy": 5.809781169891357,
      "epoch": 2.3361886259272344,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004452613174211441,
      "loss": 5.4578,
      "mean_token_accuracy": 0.18997620344161986,
      "num_tokens": 49340773.0,
      "step": 26455
    },
    {
      "entropy": 5.7577112197875975,
      "epoch": 2.3366301660190745,
      "grad_norm": 1.125,
      "learning_rate": 0.00044524072841333736,
      "loss": 5.4852,
      "mean_token_accuracy": 0.1922910302877426,
      "num_tokens": 49349863.0,
      "step": 26460
    },
    {
      "entropy": 5.760468769073486,
      "epoch": 2.337071706110915,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004452201360706115,
      "loss": 5.3157,
      "mean_token_accuracy": 0.20647781491279601,
      "num_tokens": 49358518.0,
      "step": 26465
    },
    {
      "entropy": 5.7695225238800045,
      "epoch": 2.337513246202755,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004451995403933701,
      "loss": 5.4225,
      "mean_token_accuracy": 0.20418915450572966,
      "num_tokens": 49367509.0,
      "step": 26470
    },
    {
      "entropy": 5.796112012863159,
      "epoch": 2.3379547862945955,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004451789413820164,
      "loss": 5.475,
      "mean_token_accuracy": 0.19947231262922288,
      "num_tokens": 49377157.0,
      "step": 26475
    },
    {
      "entropy": 5.891912603378296,
      "epoch": 2.338396326386436,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004451583390369541,
      "loss": 5.5199,
      "mean_token_accuracy": 0.19015612751245498,
      "num_tokens": 49386580.0,
      "step": 26480
    },
    {
      "entropy": 5.775360488891602,
      "epoch": 2.338837866478276,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004451377333585866,
      "loss": 5.4257,
      "mean_token_accuracy": 0.1934190571308136,
      "num_tokens": 49395894.0,
      "step": 26485
    },
    {
      "entropy": 5.767916250228882,
      "epoch": 2.3392794065701166,
      "grad_norm": 1.25,
      "learning_rate": 0.00044511712434731774,
      "loss": 5.4565,
      "mean_token_accuracy": 0.20091599225997925,
      "num_tokens": 49405475.0,
      "step": 26490
    },
    {
      "entropy": 5.825179195404052,
      "epoch": 2.339720946661957,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004450965120035511,
      "loss": 5.4599,
      "mean_token_accuracy": 0.1931513249874115,
      "num_tokens": 49414982.0,
      "step": 26495
    },
    {
      "entropy": 5.769105052947998,
      "epoch": 2.340162486753797,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004450758963276907,
      "loss": 5.4004,
      "mean_token_accuracy": 0.1994806408882141,
      "num_tokens": 49425502.0,
      "step": 26500
    },
    {
      "entropy": 5.744030714035034,
      "epoch": 2.3406040268456376,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044505527732014,
      "loss": 5.4106,
      "mean_token_accuracy": 0.19488929212093353,
      "num_tokens": 49435278.0,
      "step": 26505
    },
    {
      "entropy": 5.741600275039673,
      "epoch": 2.3410455669374777,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004450346549813033,
      "loss": 5.4381,
      "mean_token_accuracy": 0.1948704943060875,
      "num_tokens": 49444505.0,
      "step": 26510
    },
    {
      "entropy": 5.7899861335754395,
      "epoch": 2.341487107029318,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044501402931158433,
      "loss": 5.4334,
      "mean_token_accuracy": 0.19716664999723435,
      "num_tokens": 49453485.0,
      "step": 26515
    },
    {
      "entropy": 5.836404943466187,
      "epoch": 2.3419286471211587,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004449934003113872,
      "loss": 5.3705,
      "mean_token_accuracy": 0.20223315060138702,
      "num_tokens": 49462319.0,
      "step": 26520
    },
    {
      "entropy": 5.778238677978516,
      "epoch": 2.3423701872129987,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004449727679811161,
      "loss": 5.4447,
      "mean_token_accuracy": 0.19691283404827117,
      "num_tokens": 49470949.0,
      "step": 26525
    },
    {
      "entropy": 5.781292390823364,
      "epoch": 2.342811727304839,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004449521323211751,
      "loss": 5.4261,
      "mean_token_accuracy": 0.1975690618157387,
      "num_tokens": 49480269.0,
      "step": 26530
    },
    {
      "entropy": 5.808577680587769,
      "epoch": 2.3432532673966797,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004449314933319685,
      "loss": 5.4536,
      "mean_token_accuracy": 0.19258012473583222,
      "num_tokens": 49489364.0,
      "step": 26535
    },
    {
      "entropy": 5.864972400665283,
      "epoch": 2.3436948074885198,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044491085101390065,
      "loss": 5.4461,
      "mean_token_accuracy": 0.19563986212015153,
      "num_tokens": 49499360.0,
      "step": 26540
    },
    {
      "entropy": 5.787024879455567,
      "epoch": 2.3441363475803603,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004448902053673758,
      "loss": 5.3722,
      "mean_token_accuracy": 0.1970494344830513,
      "num_tokens": 49507770.0,
      "step": 26545
    },
    {
      "entropy": 5.678903961181641,
      "epoch": 2.3445778876722008,
      "grad_norm": 1.25,
      "learning_rate": 0.00044486955639279845,
      "loss": 5.4431,
      "mean_token_accuracy": 0.20237322449684142,
      "num_tokens": 49516824.0,
      "step": 26550
    },
    {
      "entropy": 5.783845663070679,
      "epoch": 2.345019427764041,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004448489040905731,
      "loss": 5.4453,
      "mean_token_accuracy": 0.1943666309118271,
      "num_tokens": 49525176.0,
      "step": 26555
    },
    {
      "entropy": 5.896867752075195,
      "epoch": 2.3454609678558813,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004448282484611043,
      "loss": 5.5446,
      "mean_token_accuracy": 0.19206922203302385,
      "num_tokens": 49535701.0,
      "step": 26560
    },
    {
      "entropy": 5.8251899719238285,
      "epoch": 2.345902507947722,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044480758950479673,
      "loss": 5.4691,
      "mean_token_accuracy": 0.20910117626190186,
      "num_tokens": 49545473.0,
      "step": 26565
    },
    {
      "entropy": 5.680620718002319,
      "epoch": 2.346344048039562,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00044478692722205507,
      "loss": 5.3824,
      "mean_token_accuracy": 0.1957469940185547,
      "num_tokens": 49553955.0,
      "step": 26570
    },
    {
      "entropy": 5.793128442764282,
      "epoch": 2.3467855881314024,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000444766261613284,
      "loss": 5.502,
      "mean_token_accuracy": 0.18559064269065856,
      "num_tokens": 49563305.0,
      "step": 26575
    },
    {
      "entropy": 5.846720123291016,
      "epoch": 2.347227128223243,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044474559267888847,
      "loss": 5.4809,
      "mean_token_accuracy": 0.19634333103895188,
      "num_tokens": 49571841.0,
      "step": 26580
    },
    {
      "entropy": 5.859693622589111,
      "epoch": 2.347668668315083,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004447249204192733,
      "loss": 5.4529,
      "mean_token_accuracy": 0.19736134707927705,
      "num_tokens": 49582862.0,
      "step": 26585
    },
    {
      "entropy": 5.845638275146484,
      "epoch": 2.3481102084069234,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004447042448348435,
      "loss": 5.5012,
      "mean_token_accuracy": 0.1945428505539894,
      "num_tokens": 49592410.0,
      "step": 26590
    },
    {
      "entropy": 5.79921145439148,
      "epoch": 2.348551748498764,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044468356592600405,
      "loss": 5.435,
      "mean_token_accuracy": 0.2051417887210846,
      "num_tokens": 49600582.0,
      "step": 26595
    },
    {
      "entropy": 5.810662698745728,
      "epoch": 2.348993288590604,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044466288369316006,
      "loss": 5.447,
      "mean_token_accuracy": 0.19424409419298172,
      "num_tokens": 49610165.0,
      "step": 26600
    },
    {
      "entropy": 5.7455915927886965,
      "epoch": 2.3494348286824445,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044464219813671664,
      "loss": 5.4317,
      "mean_token_accuracy": 0.20726447850465773,
      "num_tokens": 49619231.0,
      "step": 26605
    },
    {
      "entropy": 5.782672643661499,
      "epoch": 2.3498763687742845,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000444621509257079,
      "loss": 5.3311,
      "mean_token_accuracy": 0.20634853839874268,
      "num_tokens": 49627899.0,
      "step": 26610
    },
    {
      "entropy": 5.820750141143799,
      "epoch": 2.350317908866125,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00044460081705465247,
      "loss": 5.5379,
      "mean_token_accuracy": 0.18359910398721696,
      "num_tokens": 49637866.0,
      "step": 26615
    },
    {
      "entropy": 5.864542388916016,
      "epoch": 2.3507594489579655,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044458012152984253,
      "loss": 5.5093,
      "mean_token_accuracy": 0.19459903240203857,
      "num_tokens": 49648273.0,
      "step": 26620
    },
    {
      "entropy": 5.884262609481811,
      "epoch": 2.3512009890498056,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004445594226830544,
      "loss": 5.473,
      "mean_token_accuracy": 0.18934902548789978,
      "num_tokens": 49657944.0,
      "step": 26625
    },
    {
      "entropy": 5.824350023269654,
      "epoch": 2.351642529141646,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044453872051469345,
      "loss": 5.462,
      "mean_token_accuracy": 0.1969756543636322,
      "num_tokens": 49667158.0,
      "step": 26630
    },
    {
      "entropy": 5.746573495864868,
      "epoch": 2.3520840692334866,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044451801502516555,
      "loss": 5.3787,
      "mean_token_accuracy": 0.19574964046478271,
      "num_tokens": 49675535.0,
      "step": 26635
    },
    {
      "entropy": 5.742319631576538,
      "epoch": 2.3525256093253266,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004444973062148761,
      "loss": 5.4773,
      "mean_token_accuracy": 0.19494964629411698,
      "num_tokens": 49684922.0,
      "step": 26640
    },
    {
      "entropy": 5.79100604057312,
      "epoch": 2.352967149417167,
      "grad_norm": 1.125,
      "learning_rate": 0.0004444765940842308,
      "loss": 5.4319,
      "mean_token_accuracy": 0.1954280376434326,
      "num_tokens": 49695302.0,
      "step": 26645
    },
    {
      "entropy": 5.7877397537231445,
      "epoch": 2.3534086895090076,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004444558786336355,
      "loss": 5.4674,
      "mean_token_accuracy": 0.19683394879102706,
      "num_tokens": 49704668.0,
      "step": 26650
    },
    {
      "entropy": 5.7255322456359865,
      "epoch": 2.3538502296008477,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004444351598634958,
      "loss": 5.4035,
      "mean_token_accuracy": 0.20060806274414061,
      "num_tokens": 49713704.0,
      "step": 26655
    },
    {
      "entropy": 5.818120861053467,
      "epoch": 2.354291769692688,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004444144377742178,
      "loss": 5.4806,
      "mean_token_accuracy": 0.19872302263975145,
      "num_tokens": 49723743.0,
      "step": 26660
    },
    {
      "entropy": 5.866615581512451,
      "epoch": 2.354733309784528,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004443937123662071,
      "loss": 5.5524,
      "mean_token_accuracy": 0.18852213472127916,
      "num_tokens": 49734760.0,
      "step": 26665
    },
    {
      "entropy": 5.8492485046386715,
      "epoch": 2.3551748498763687,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004443729836398702,
      "loss": 5.4296,
      "mean_token_accuracy": 0.18895757347345352,
      "num_tokens": 49744836.0,
      "step": 26670
    },
    {
      "entropy": 5.779052066802978,
      "epoch": 2.355616389968209,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044435225159561263,
      "loss": 5.4022,
      "mean_token_accuracy": 0.20673364847898484,
      "num_tokens": 49753646.0,
      "step": 26675
    },
    {
      "entropy": 5.715762281417847,
      "epoch": 2.3560579300600493,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004443315162338409,
      "loss": 5.4545,
      "mean_token_accuracy": 0.19697688072919844,
      "num_tokens": 49762576.0,
      "step": 26680
    },
    {
      "entropy": 5.809338665008545,
      "epoch": 2.3564994701518898,
      "grad_norm": 1.21875,
      "learning_rate": 0.000444310777554961,
      "loss": 5.4431,
      "mean_token_accuracy": 0.19267212301492692,
      "num_tokens": 49772614.0,
      "step": 26685
    },
    {
      "entropy": 5.7954765319824215,
      "epoch": 2.3569410102437303,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004442900355593793,
      "loss": 5.4081,
      "mean_token_accuracy": 0.19419508576393127,
      "num_tokens": 49781005.0,
      "step": 26690
    },
    {
      "entropy": 5.732917928695679,
      "epoch": 2.3573825503355703,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044426929024750204,
      "loss": 5.4349,
      "mean_token_accuracy": 0.19058915376663207,
      "num_tokens": 49790639.0,
      "step": 26695
    },
    {
      "entropy": 5.818629503250122,
      "epoch": 2.357824090427411,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004442485416197357,
      "loss": 5.5247,
      "mean_token_accuracy": 0.1899215504527092,
      "num_tokens": 49800010.0,
      "step": 26700
    },
    {
      "entropy": 5.820417022705078,
      "epoch": 2.3582656305192513,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044422778967648675,
      "loss": 5.4301,
      "mean_token_accuracy": 0.195323246717453,
      "num_tokens": 49808510.0,
      "step": 26705
    },
    {
      "entropy": 5.804453134536743,
      "epoch": 2.3587071706110914,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044420703441816153,
      "loss": 5.4572,
      "mean_token_accuracy": 0.19267254769802095,
      "num_tokens": 49818017.0,
      "step": 26710
    },
    {
      "entropy": 5.776887893676758,
      "epoch": 2.359148710702932,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004441862758451668,
      "loss": 5.4398,
      "mean_token_accuracy": 0.1931275397539139,
      "num_tokens": 49826797.0,
      "step": 26715
    },
    {
      "entropy": 5.825407886505127,
      "epoch": 2.3595902507947724,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044416551395790916,
      "loss": 5.4178,
      "mean_token_accuracy": 0.19414039850234985,
      "num_tokens": 49835632.0,
      "step": 26720
    },
    {
      "entropy": 5.819238615036011,
      "epoch": 2.3600317908866124,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004441447487567953,
      "loss": 5.4444,
      "mean_token_accuracy": 0.19963242411613463,
      "num_tokens": 49844770.0,
      "step": 26725
    },
    {
      "entropy": 5.78247971534729,
      "epoch": 2.360473330978453,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044412398024223194,
      "loss": 5.4741,
      "mean_token_accuracy": 0.19482817202806474,
      "num_tokens": 49854170.0,
      "step": 26730
    },
    {
      "entropy": 5.872864627838135,
      "epoch": 2.3609148710702934,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004441032084146261,
      "loss": 5.4964,
      "mean_token_accuracy": 0.19717595875263214,
      "num_tokens": 49863541.0,
      "step": 26735
    },
    {
      "entropy": 5.848886775970459,
      "epoch": 2.3613564111621335,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044408243327438446,
      "loss": 5.4144,
      "mean_token_accuracy": 0.1957680657505989,
      "num_tokens": 49872556.0,
      "step": 26740
    },
    {
      "entropy": 5.8052267074584964,
      "epoch": 2.361797951253974,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044406165482191424,
      "loss": 5.4945,
      "mean_token_accuracy": 0.19622038602828978,
      "num_tokens": 49882188.0,
      "step": 26745
    },
    {
      "entropy": 5.854679012298584,
      "epoch": 2.362239491345814,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004440408730576223,
      "loss": 5.5295,
      "mean_token_accuracy": 0.18286992609500885,
      "num_tokens": 49891980.0,
      "step": 26750
    },
    {
      "entropy": 5.870315790176392,
      "epoch": 2.3626810314376545,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044402008798191575,
      "loss": 5.4515,
      "mean_token_accuracy": 0.19767990410327912,
      "num_tokens": 49900783.0,
      "step": 26755
    },
    {
      "entropy": 5.754424810409546,
      "epoch": 2.363122571529495,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004439992995952019,
      "loss": 5.461,
      "mean_token_accuracy": 0.19795602262020112,
      "num_tokens": 49910525.0,
      "step": 26760
    },
    {
      "entropy": 5.769544553756714,
      "epoch": 2.363564111621335,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044397850789788774,
      "loss": 5.457,
      "mean_token_accuracy": 0.1949768304824829,
      "num_tokens": 49919286.0,
      "step": 26765
    },
    {
      "entropy": 5.815430641174316,
      "epoch": 2.3640056517131756,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004439577128903808,
      "loss": 5.4031,
      "mean_token_accuracy": 0.19950215369462967,
      "num_tokens": 49927882.0,
      "step": 26770
    },
    {
      "entropy": 5.859138298034668,
      "epoch": 2.364447191805016,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004439369145730884,
      "loss": 5.4682,
      "mean_token_accuracy": 0.19486029744148253,
      "num_tokens": 49936289.0,
      "step": 26775
    },
    {
      "entropy": 5.796139192581177,
      "epoch": 2.364888731896856,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004439161129464179,
      "loss": 5.4673,
      "mean_token_accuracy": 0.1919807806611061,
      "num_tokens": 49945356.0,
      "step": 26780
    },
    {
      "entropy": 5.807422018051147,
      "epoch": 2.3653302719886966,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044389530801077675,
      "loss": 5.3908,
      "mean_token_accuracy": 0.19784061461687089,
      "num_tokens": 49954216.0,
      "step": 26785
    },
    {
      "entropy": 5.838868284225464,
      "epoch": 2.365771812080537,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004438744997665726,
      "loss": 5.5284,
      "mean_token_accuracy": 0.1922971323132515,
      "num_tokens": 49963179.0,
      "step": 26790
    },
    {
      "entropy": 5.801823520660401,
      "epoch": 2.366213352172377,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004438536882142131,
      "loss": 5.512,
      "mean_token_accuracy": 0.1880034789443016,
      "num_tokens": 49973108.0,
      "step": 26795
    },
    {
      "entropy": 5.868147468566894,
      "epoch": 2.3666548922642177,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004438328733541059,
      "loss": 5.4192,
      "mean_token_accuracy": 0.18786296397447586,
      "num_tokens": 49982003.0,
      "step": 26800
    },
    {
      "entropy": 5.787125205993652,
      "epoch": 2.3670964323560577,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004438120551866586,
      "loss": 5.4657,
      "mean_token_accuracy": 0.1896170750260353,
      "num_tokens": 49992034.0,
      "step": 26805
    },
    {
      "entropy": 5.75523567199707,
      "epoch": 2.367537972447898,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004437912337122792,
      "loss": 5.4477,
      "mean_token_accuracy": 0.19774802923202514,
      "num_tokens": 50001379.0,
      "step": 26810
    },
    {
      "entropy": 5.8554576396942135,
      "epoch": 2.3679795125397387,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004437704089313756,
      "loss": 5.5233,
      "mean_token_accuracy": 0.18914915770292282,
      "num_tokens": 50010721.0,
      "step": 26815
    },
    {
      "entropy": 5.880755662918091,
      "epoch": 2.3684210526315788,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044374958084435564,
      "loss": 5.5231,
      "mean_token_accuracy": 0.1873500630259514,
      "num_tokens": 50019571.0,
      "step": 26820
    },
    {
      "entropy": 5.7836438655853275,
      "epoch": 2.3688625927234193,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044372874945162734,
      "loss": 5.4311,
      "mean_token_accuracy": 0.19768593162298204,
      "num_tokens": 50028045.0,
      "step": 26825
    },
    {
      "entropy": 5.820521593093872,
      "epoch": 2.3693041328152598,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044370791475359877,
      "loss": 5.4405,
      "mean_token_accuracy": 0.1935562789440155,
      "num_tokens": 50036769.0,
      "step": 26830
    },
    {
      "entropy": 5.843263244628906,
      "epoch": 2.3697456729071,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044368707675067814,
      "loss": 5.4562,
      "mean_token_accuracy": 0.19285148680210112,
      "num_tokens": 50045973.0,
      "step": 26835
    },
    {
      "entropy": 5.757670259475708,
      "epoch": 2.3701872129989403,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004436662354432736,
      "loss": 5.3757,
      "mean_token_accuracy": 0.19810771942138672,
      "num_tokens": 50055243.0,
      "step": 26840
    },
    {
      "entropy": 5.803348636627197,
      "epoch": 2.370628753090781,
      "grad_norm": 1.25,
      "learning_rate": 0.00044364539083179347,
      "loss": 5.4741,
      "mean_token_accuracy": 0.19753337651491165,
      "num_tokens": 50064573.0,
      "step": 26845
    },
    {
      "entropy": 5.661289978027344,
      "epoch": 2.371070293182621,
      "grad_norm": 1.3125,
      "learning_rate": 0.000443624542916646,
      "loss": 5.309,
      "mean_token_accuracy": 0.2058949053287506,
      "num_tokens": 50073578.0,
      "step": 26850
    },
    {
      "entropy": 5.762441825866699,
      "epoch": 2.3715118332744614,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004436036916982397,
      "loss": 5.464,
      "mean_token_accuracy": 0.1986655741930008,
      "num_tokens": 50083620.0,
      "step": 26855
    },
    {
      "entropy": 5.792378711700439,
      "epoch": 2.371953373366302,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004435828371769829,
      "loss": 5.3796,
      "mean_token_accuracy": 0.20742227286100387,
      "num_tokens": 50092050.0,
      "step": 26860
    },
    {
      "entropy": 5.856243467330932,
      "epoch": 2.372394913458142,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044356197935328426,
      "loss": 5.4687,
      "mean_token_accuracy": 0.19440920054912567,
      "num_tokens": 50102425.0,
      "step": 26865
    },
    {
      "entropy": 5.7744041919708256,
      "epoch": 2.3728364535499824,
      "grad_norm": 1.671875,
      "learning_rate": 0.00044354111822755225,
      "loss": 5.3713,
      "mean_token_accuracy": 0.19534678310155867,
      "num_tokens": 50112338.0,
      "step": 26870
    },
    {
      "entropy": 5.757501697540283,
      "epoch": 2.373277993641823,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044352025380019563,
      "loss": 5.4584,
      "mean_token_accuracy": 0.19618681967258453,
      "num_tokens": 50121302.0,
      "step": 26875
    },
    {
      "entropy": 5.745772171020508,
      "epoch": 2.373719533733663,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044349938607162305,
      "loss": 5.3874,
      "mean_token_accuracy": 0.19699536710977555,
      "num_tokens": 50129820.0,
      "step": 26880
    },
    {
      "entropy": 5.8713147163391115,
      "epoch": 2.3741610738255035,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044347851504224336,
      "loss": 5.5155,
      "mean_token_accuracy": 0.19419267326593398,
      "num_tokens": 50138860.0,
      "step": 26885
    },
    {
      "entropy": 5.760289812088013,
      "epoch": 2.3746026139173435,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044345764071246536,
      "loss": 5.4136,
      "mean_token_accuracy": 0.20277315080165864,
      "num_tokens": 50148257.0,
      "step": 26890
    },
    {
      "entropy": 5.7196681022644045,
      "epoch": 2.375044154009184,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044343676308269796,
      "loss": 5.4154,
      "mean_token_accuracy": 0.2065388187766075,
      "num_tokens": 50157380.0,
      "step": 26895
    },
    {
      "entropy": 5.89013147354126,
      "epoch": 2.3754856941010245,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044341588215335016,
      "loss": 5.4579,
      "mean_token_accuracy": 0.1960765078663826,
      "num_tokens": 50167806.0,
      "step": 26900
    },
    {
      "entropy": 5.901745414733886,
      "epoch": 2.3759272341928646,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044339499792483104,
      "loss": 5.5063,
      "mean_token_accuracy": 0.19504934549331665,
      "num_tokens": 50176763.0,
      "step": 26905
    },
    {
      "entropy": 5.693349170684814,
      "epoch": 2.376368774284705,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004433741103975498,
      "loss": 5.4497,
      "mean_token_accuracy": 0.19445212185382843,
      "num_tokens": 50186735.0,
      "step": 26910
    },
    {
      "entropy": 5.711969566345215,
      "epoch": 2.3768103143765456,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044335321957191535,
      "loss": 5.4023,
      "mean_token_accuracy": 0.19913543462753297,
      "num_tokens": 50195206.0,
      "step": 26915
    },
    {
      "entropy": 5.877222156524658,
      "epoch": 2.3772518544683856,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044333232544833707,
      "loss": 5.5055,
      "mean_token_accuracy": 0.18688288033008577,
      "num_tokens": 50203706.0,
      "step": 26920
    },
    {
      "entropy": 5.8361481666564945,
      "epoch": 2.377693394560226,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004433114280272243,
      "loss": 5.4687,
      "mean_token_accuracy": 0.1964350238442421,
      "num_tokens": 50213007.0,
      "step": 26925
    },
    {
      "entropy": 5.761434698104859,
      "epoch": 2.3781349346520666,
      "grad_norm": 1.25,
      "learning_rate": 0.0004432905273089864,
      "loss": 5.3809,
      "mean_token_accuracy": 0.20320215523242952,
      "num_tokens": 50223155.0,
      "step": 26930
    },
    {
      "entropy": 5.699112272262573,
      "epoch": 2.3785764747439067,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044326962329403275,
      "loss": 5.3337,
      "mean_token_accuracy": 0.20851715952157973,
      "num_tokens": 50231788.0,
      "step": 26935
    },
    {
      "entropy": 5.763467645645141,
      "epoch": 2.379018014835747,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044324871598277294,
      "loss": 5.4153,
      "mean_token_accuracy": 0.20152414888143538,
      "num_tokens": 50241557.0,
      "step": 26940
    },
    {
      "entropy": 5.8116774559021,
      "epoch": 2.379459554927587,
      "grad_norm": 1.859375,
      "learning_rate": 0.00044322780537561643,
      "loss": 5.4799,
      "mean_token_accuracy": 0.19544838070869447,
      "num_tokens": 50252141.0,
      "step": 26945
    },
    {
      "entropy": 5.871216297149658,
      "epoch": 2.3799010950194277,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004432068914729729,
      "loss": 5.525,
      "mean_token_accuracy": 0.19071032702922822,
      "num_tokens": 50261361.0,
      "step": 26950
    },
    {
      "entropy": 5.8036610126495365,
      "epoch": 2.380342635111268,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044318597427525197,
      "loss": 5.5043,
      "mean_token_accuracy": 0.19229744821786882,
      "num_tokens": 50270193.0,
      "step": 26955
    },
    {
      "entropy": 5.839167356491089,
      "epoch": 2.3807841752031083,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00044316505378286356,
      "loss": 5.4732,
      "mean_token_accuracy": 0.1911521315574646,
      "num_tokens": 50280217.0,
      "step": 26960
    },
    {
      "entropy": 5.732112979888916,
      "epoch": 2.3812257152949488,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004431441299962173,
      "loss": 5.392,
      "mean_token_accuracy": 0.19684236794710158,
      "num_tokens": 50289610.0,
      "step": 26965
    },
    {
      "entropy": 5.728816366195678,
      "epoch": 2.3816672553867893,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004431232029157232,
      "loss": 5.3487,
      "mean_token_accuracy": 0.19886577129364014,
      "num_tokens": 50298734.0,
      "step": 26970
    },
    {
      "entropy": 5.739948987960815,
      "epoch": 2.3821087954786293,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004431022725417912,
      "loss": 5.3784,
      "mean_token_accuracy": 0.20103285312652588,
      "num_tokens": 50307679.0,
      "step": 26975
    },
    {
      "entropy": 5.814243602752685,
      "epoch": 2.38255033557047,
      "grad_norm": 1.125,
      "learning_rate": 0.0004430813388748312,
      "loss": 5.368,
      "mean_token_accuracy": 0.20035041570663453,
      "num_tokens": 50315983.0,
      "step": 26980
    },
    {
      "entropy": 5.79416937828064,
      "epoch": 2.3829918756623103,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044306040191525347,
      "loss": 5.4807,
      "mean_token_accuracy": 0.19486231058835984,
      "num_tokens": 50325222.0,
      "step": 26985
    },
    {
      "entropy": 5.709028339385986,
      "epoch": 2.3834334157541504,
      "grad_norm": 1.359375,
      "learning_rate": 0.000443039461663468,
      "loss": 5.4046,
      "mean_token_accuracy": 0.20301312655210496,
      "num_tokens": 50333417.0,
      "step": 26990
    },
    {
      "entropy": 5.749329376220703,
      "epoch": 2.383874955845991,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044301851811988504,
      "loss": 5.4036,
      "mean_token_accuracy": 0.20817478150129318,
      "num_tokens": 50342836.0,
      "step": 26995
    },
    {
      "entropy": 5.866056871414185,
      "epoch": 2.3843164959378313,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044299757128491487,
      "loss": 5.5389,
      "mean_token_accuracy": 0.19110322296619414,
      "num_tokens": 50351988.0,
      "step": 27000
    },
    {
      "epoch": 2.3843164959378313,
      "eval_entropy": 5.581728348186575,
      "eval_loss": 5.544463634490967,
      "eval_mean_token_accuracy": 0.19847757581089825,
      "eval_num_tokens": 50351988.0,
      "eval_runtime": 26.1507,
      "eval_samples_per_second": 1350.441,
      "eval_steps_per_second": 168.829,
      "step": 27000
    },
    {
      "entropy": 5.767484188079834,
      "epoch": 2.3847580360296714,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044297662115896783,
      "loss": 5.4964,
      "mean_token_accuracy": 0.19078738391399383,
      "num_tokens": 50361087.0,
      "step": 27005
    },
    {
      "entropy": 5.789765119552612,
      "epoch": 2.385199576121512,
      "grad_norm": 1.578125,
      "learning_rate": 0.00044295566774245435,
      "loss": 5.4658,
      "mean_token_accuracy": 0.19095250070095063,
      "num_tokens": 50369842.0,
      "step": 27010
    },
    {
      "entropy": 5.90677924156189,
      "epoch": 2.3856411162133524,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044293471103578475,
      "loss": 5.538,
      "mean_token_accuracy": 0.1858898103237152,
      "num_tokens": 50379125.0,
      "step": 27015
    },
    {
      "entropy": 5.724384450912476,
      "epoch": 2.3860826563051925,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044291375103936977,
      "loss": 5.3544,
      "mean_token_accuracy": 0.2072057083249092,
      "num_tokens": 50388031.0,
      "step": 27020
    },
    {
      "entropy": 5.764397048950196,
      "epoch": 2.386524196397033,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044289278775361984,
      "loss": 5.4742,
      "mean_token_accuracy": 0.1928586483001709,
      "num_tokens": 50397348.0,
      "step": 27025
    },
    {
      "entropy": 5.809425687789917,
      "epoch": 2.386965736488873,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044287182117894576,
      "loss": 5.4806,
      "mean_token_accuracy": 0.1992364451289177,
      "num_tokens": 50406782.0,
      "step": 27030
    },
    {
      "entropy": 5.791159820556641,
      "epoch": 2.3874072765807135,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044285085131575807,
      "loss": 5.4104,
      "mean_token_accuracy": 0.19899265915155412,
      "num_tokens": 50415527.0,
      "step": 27035
    },
    {
      "entropy": 5.811756134033203,
      "epoch": 2.387848816672554,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044282987816446774,
      "loss": 5.5139,
      "mean_token_accuracy": 0.1928165763616562,
      "num_tokens": 50424852.0,
      "step": 27040
    },
    {
      "entropy": 5.783355760574341,
      "epoch": 2.388290356764394,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004428089017254856,
      "loss": 5.469,
      "mean_token_accuracy": 0.19296583533287048,
      "num_tokens": 50434371.0,
      "step": 27045
    },
    {
      "entropy": 5.891714859008789,
      "epoch": 2.3887318968562345,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044278792199922245,
      "loss": 5.5391,
      "mean_token_accuracy": 0.18711710572242737,
      "num_tokens": 50443358.0,
      "step": 27050
    },
    {
      "entropy": 5.850710725784301,
      "epoch": 2.389173436948075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044276693898608933,
      "loss": 5.4585,
      "mean_token_accuracy": 0.19316650032997132,
      "num_tokens": 50453132.0,
      "step": 27055
    },
    {
      "entropy": 5.798442554473877,
      "epoch": 2.389614977039915,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044274595268649727,
      "loss": 5.3908,
      "mean_token_accuracy": 0.20346576124429702,
      "num_tokens": 50461837.0,
      "step": 27060
    },
    {
      "entropy": 5.792890834808349,
      "epoch": 2.3900565171317556,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044272496310085745,
      "loss": 5.4182,
      "mean_token_accuracy": 0.1978772535920143,
      "num_tokens": 50470482.0,
      "step": 27065
    },
    {
      "entropy": 5.773432493209839,
      "epoch": 2.390498057223596,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044270397022958094,
      "loss": 5.4609,
      "mean_token_accuracy": 0.1913543552160263,
      "num_tokens": 50480168.0,
      "step": 27070
    },
    {
      "entropy": 5.759543228149414,
      "epoch": 2.390939597315436,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044268297407307905,
      "loss": 5.4511,
      "mean_token_accuracy": 0.19320129007101058,
      "num_tokens": 50489937.0,
      "step": 27075
    },
    {
      "entropy": 5.751318168640137,
      "epoch": 2.3913811374072766,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044266197463176305,
      "loss": 5.4577,
      "mean_token_accuracy": 0.1920393168926239,
      "num_tokens": 50499151.0,
      "step": 27080
    },
    {
      "entropy": 5.874878740310669,
      "epoch": 2.3918226774991167,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044264097190604436,
      "loss": 5.5247,
      "mean_token_accuracy": 0.19691094160079955,
      "num_tokens": 50508413.0,
      "step": 27085
    },
    {
      "entropy": 5.901967000961304,
      "epoch": 2.392264217590957,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004426199658963344,
      "loss": 5.5453,
      "mean_token_accuracy": 0.19210318177938462,
      "num_tokens": 50517513.0,
      "step": 27090
    },
    {
      "entropy": 5.810697031021118,
      "epoch": 2.3927057576827977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044259895660304455,
      "loss": 5.4025,
      "mean_token_accuracy": 0.20152654945850373,
      "num_tokens": 50526057.0,
      "step": 27095
    },
    {
      "entropy": 5.6975555419921875,
      "epoch": 2.3931472977746377,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044257794402658655,
      "loss": 5.3866,
      "mean_token_accuracy": 0.19631059765815734,
      "num_tokens": 50535388.0,
      "step": 27100
    },
    {
      "entropy": 5.838620948791504,
      "epoch": 2.3935888378664782,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004425569281673719,
      "loss": 5.4615,
      "mean_token_accuracy": 0.1940155506134033,
      "num_tokens": 50544606.0,
      "step": 27105
    },
    {
      "entropy": 5.75078272819519,
      "epoch": 2.3940303779583187,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044253590902581223,
      "loss": 5.4336,
      "mean_token_accuracy": 0.19741683155298234,
      "num_tokens": 50553822.0,
      "step": 27110
    },
    {
      "entropy": 5.750366926193237,
      "epoch": 2.394471918050159,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004425148866023195,
      "loss": 5.4117,
      "mean_token_accuracy": 0.2044105812907219,
      "num_tokens": 50562879.0,
      "step": 27115
    },
    {
      "entropy": 5.765265941619873,
      "epoch": 2.3949134581419993,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044249386089730534,
      "loss": 5.372,
      "mean_token_accuracy": 0.19866358935832978,
      "num_tokens": 50571536.0,
      "step": 27120
    },
    {
      "entropy": 5.711014080047607,
      "epoch": 2.39535499823384,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004424728319111818,
      "loss": 5.3759,
      "mean_token_accuracy": 0.2048347532749176,
      "num_tokens": 50580692.0,
      "step": 27125
    },
    {
      "entropy": 5.865678358078003,
      "epoch": 2.39579653832568,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004424517996443606,
      "loss": 5.6114,
      "mean_token_accuracy": 0.19015679955482484,
      "num_tokens": 50589456.0,
      "step": 27130
    },
    {
      "entropy": 5.797392511367798,
      "epoch": 2.3962380784175203,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044243076409725395,
      "loss": 5.4081,
      "mean_token_accuracy": 0.19937923401594163,
      "num_tokens": 50598947.0,
      "step": 27135
    },
    {
      "entropy": 5.815799045562744,
      "epoch": 2.396679618509361,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044240972527027385,
      "loss": 5.4679,
      "mean_token_accuracy": 0.20117950588464736,
      "num_tokens": 50607373.0,
      "step": 27140
    },
    {
      "entropy": 5.869526481628418,
      "epoch": 2.397121158601201,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044238868316383246,
      "loss": 5.5642,
      "mean_token_accuracy": 0.18869248926639556,
      "num_tokens": 50616930.0,
      "step": 27145
    },
    {
      "entropy": 5.762528371810913,
      "epoch": 2.3975626986930414,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000442367637778342,
      "loss": 5.4661,
      "mean_token_accuracy": 0.19484712183475494,
      "num_tokens": 50626658.0,
      "step": 27150
    },
    {
      "entropy": 5.761983442306518,
      "epoch": 2.398004238784882,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004423465891142146,
      "loss": 5.4911,
      "mean_token_accuracy": 0.19051604419946672,
      "num_tokens": 50635692.0,
      "step": 27155
    },
    {
      "entropy": 5.818031597137451,
      "epoch": 2.398445778876722,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044232553717186274,
      "loss": 5.3767,
      "mean_token_accuracy": 0.20459626913070678,
      "num_tokens": 50644036.0,
      "step": 27160
    },
    {
      "entropy": 5.788651371002198,
      "epoch": 2.3988873189685624,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004423044819516988,
      "loss": 5.5627,
      "mean_token_accuracy": 0.1864360049366951,
      "num_tokens": 50652363.0,
      "step": 27165
    },
    {
      "entropy": 5.762420892715454,
      "epoch": 2.3993288590604025,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004422834234541352,
      "loss": 5.4328,
      "mean_token_accuracy": 0.1913999781012535,
      "num_tokens": 50661985.0,
      "step": 27170
    },
    {
      "entropy": 5.770860004425049,
      "epoch": 2.399770399152243,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004422623616795845,
      "loss": 5.4424,
      "mean_token_accuracy": 0.19952959269285203,
      "num_tokens": 50671171.0,
      "step": 27175
    },
    {
      "entropy": 5.789981698989868,
      "epoch": 2.4002119392440835,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044224129662845916,
      "loss": 5.3889,
      "mean_token_accuracy": 0.19809174835681914,
      "num_tokens": 50679792.0,
      "step": 27180
    },
    {
      "entropy": 5.7889509201049805,
      "epoch": 2.4006534793359235,
      "grad_norm": 1.28125,
      "learning_rate": 0.000442220228301172,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1885645255446434,
      "num_tokens": 50690082.0,
      "step": 27185
    },
    {
      "entropy": 5.748192977905274,
      "epoch": 2.401095019427764,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004421991566981357,
      "loss": 5.396,
      "mean_token_accuracy": 0.2059655249118805,
      "num_tokens": 50698351.0,
      "step": 27190
    },
    {
      "entropy": 5.694017934799194,
      "epoch": 2.4015365595196045,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000442178081819763,
      "loss": 5.3669,
      "mean_token_accuracy": 0.19994862228631974,
      "num_tokens": 50706977.0,
      "step": 27195
    },
    {
      "entropy": 5.726548910140991,
      "epoch": 2.4019780996114446,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004421570036664667,
      "loss": 5.4397,
      "mean_token_accuracy": 0.19963235706090926,
      "num_tokens": 50716168.0,
      "step": 27200
    },
    {
      "entropy": 5.838878774642945,
      "epoch": 2.402419639703285,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044213592223865976,
      "loss": 5.4823,
      "mean_token_accuracy": 0.1933092787861824,
      "num_tokens": 50725301.0,
      "step": 27205
    },
    {
      "entropy": 5.858595895767212,
      "epoch": 2.4028611797951256,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044211483753675524,
      "loss": 5.4467,
      "mean_token_accuracy": 0.19555934518575668,
      "num_tokens": 50734494.0,
      "step": 27210
    },
    {
      "entropy": 5.769343376159668,
      "epoch": 2.4033027198869656,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004420937495611661,
      "loss": 5.4943,
      "mean_token_accuracy": 0.1883556753396988,
      "num_tokens": 50743650.0,
      "step": 27215
    },
    {
      "entropy": 5.7020594596862795,
      "epoch": 2.403744259978806,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004420726583123054,
      "loss": 5.3985,
      "mean_token_accuracy": 0.1981179744005203,
      "num_tokens": 50753538.0,
      "step": 27220
    },
    {
      "entropy": 5.840254497528076,
      "epoch": 2.404185800070646,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004420515637905863,
      "loss": 5.4996,
      "mean_token_accuracy": 0.19112931340932846,
      "num_tokens": 50763349.0,
      "step": 27225
    },
    {
      "entropy": 5.878906297683716,
      "epoch": 2.4046273401624867,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004420304659964221,
      "loss": 5.4856,
      "mean_token_accuracy": 0.1864248037338257,
      "num_tokens": 50772348.0,
      "step": 27230
    },
    {
      "entropy": 5.858193111419678,
      "epoch": 2.405068880254327,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044200936493022616,
      "loss": 5.5653,
      "mean_token_accuracy": 0.18714066296815873,
      "num_tokens": 50782054.0,
      "step": 27235
    },
    {
      "entropy": 5.852827501296997,
      "epoch": 2.4055104203461672,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004419882605924116,
      "loss": 5.4655,
      "mean_token_accuracy": 0.19169602841138839,
      "num_tokens": 50791955.0,
      "step": 27240
    },
    {
      "entropy": 5.784445762634277,
      "epoch": 2.4059519604380077,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004419671529833921,
      "loss": 5.4926,
      "mean_token_accuracy": 0.1976884976029396,
      "num_tokens": 50801907.0,
      "step": 27245
    },
    {
      "entropy": 5.8000876903533936,
      "epoch": 2.4063935005298482,
      "grad_norm": 1.125,
      "learning_rate": 0.0004419460421035809,
      "loss": 5.5651,
      "mean_token_accuracy": 0.1873959258198738,
      "num_tokens": 50811413.0,
      "step": 27250
    },
    {
      "entropy": 5.788710832595825,
      "epoch": 2.4068350406216883,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004419249279533918,
      "loss": 5.4177,
      "mean_token_accuracy": 0.1975828692317009,
      "num_tokens": 50820080.0,
      "step": 27255
    },
    {
      "entropy": 5.813731718063354,
      "epoch": 2.407276580713529,
      "grad_norm": 1.359375,
      "learning_rate": 0.00044190381053323824,
      "loss": 5.52,
      "mean_token_accuracy": 0.19082509279251098,
      "num_tokens": 50829622.0,
      "step": 27260
    },
    {
      "entropy": 5.808641576766968,
      "epoch": 2.4077181208053693,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044188268984353396,
      "loss": 5.3907,
      "mean_token_accuracy": 0.2056523159146309,
      "num_tokens": 50838112.0,
      "step": 27265
    },
    {
      "entropy": 5.777623033523559,
      "epoch": 2.4081596608972093,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004418615658846927,
      "loss": 5.4733,
      "mean_token_accuracy": 0.19321490228176116,
      "num_tokens": 50848014.0,
      "step": 27270
    },
    {
      "entropy": 5.901413202285767,
      "epoch": 2.40860120098905,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004418404386571283,
      "loss": 5.553,
      "mean_token_accuracy": 0.17960018515586854,
      "num_tokens": 50857103.0,
      "step": 27275
    },
    {
      "entropy": 5.7475231170654295,
      "epoch": 2.4090427410808903,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004418193081612545,
      "loss": 5.4197,
      "mean_token_accuracy": 0.19779509007930757,
      "num_tokens": 50865866.0,
      "step": 27280
    },
    {
      "entropy": 5.8875304698944095,
      "epoch": 2.4094842811727304,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044179817439748547,
      "loss": 5.5022,
      "mean_token_accuracy": 0.1901942789554596,
      "num_tokens": 50874694.0,
      "step": 27285
    },
    {
      "entropy": 5.751189374923706,
      "epoch": 2.409925821264571,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044177703736623497,
      "loss": 5.4046,
      "mean_token_accuracy": 0.1958903968334198,
      "num_tokens": 50884965.0,
      "step": 27290
    },
    {
      "entropy": 5.774834251403808,
      "epoch": 2.4103673613564114,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044175589706791724,
      "loss": 5.5047,
      "mean_token_accuracy": 0.19201705306768418,
      "num_tokens": 50894192.0,
      "step": 27295
    },
    {
      "entropy": 5.797828531265258,
      "epoch": 2.4108089014482514,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044173475350294623,
      "loss": 5.4132,
      "mean_token_accuracy": 0.19056604504585267,
      "num_tokens": 50902458.0,
      "step": 27300
    },
    {
      "entropy": 5.820057773590088,
      "epoch": 2.411250441540092,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044171360667173633,
      "loss": 5.4041,
      "mean_token_accuracy": 0.19907789975404738,
      "num_tokens": 50912038.0,
      "step": 27305
    },
    {
      "entropy": 5.828496074676513,
      "epoch": 2.411691981631932,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004416924565747016,
      "loss": 5.5615,
      "mean_token_accuracy": 0.18043221086263656,
      "num_tokens": 50921164.0,
      "step": 27310
    },
    {
      "entropy": 5.789948129653931,
      "epoch": 2.4121335217237725,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044167130321225644,
      "loss": 5.4307,
      "mean_token_accuracy": 0.19509036093950272,
      "num_tokens": 50930359.0,
      "step": 27315
    },
    {
      "entropy": 5.7664436340332035,
      "epoch": 2.412575061815613,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044165014658481533,
      "loss": 5.4742,
      "mean_token_accuracy": 0.19249194115400314,
      "num_tokens": 50940675.0,
      "step": 27320
    },
    {
      "entropy": 5.809629297256469,
      "epoch": 2.413016601907453,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004416289866927926,
      "loss": 5.5644,
      "mean_token_accuracy": 0.18708542883396148,
      "num_tokens": 50949402.0,
      "step": 27325
    },
    {
      "entropy": 5.763953495025635,
      "epoch": 2.4134581419992935,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004416078235366027,
      "loss": 5.4198,
      "mean_token_accuracy": 0.19288770854473114,
      "num_tokens": 50958348.0,
      "step": 27330
    },
    {
      "entropy": 5.797935342788696,
      "epoch": 2.413899682091134,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044158665711666034,
      "loss": 5.421,
      "mean_token_accuracy": 0.19387879818677903,
      "num_tokens": 50967348.0,
      "step": 27335
    },
    {
      "entropy": 5.850919532775879,
      "epoch": 2.414341222182974,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044156548743338016,
      "loss": 5.4594,
      "mean_token_accuracy": 0.19667820483446122,
      "num_tokens": 50975882.0,
      "step": 27340
    },
    {
      "entropy": 5.822178888320923,
      "epoch": 2.4147827622748146,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044154431448717674,
      "loss": 5.4174,
      "mean_token_accuracy": 0.19828397780656815,
      "num_tokens": 50985241.0,
      "step": 27345
    },
    {
      "entropy": 5.793811273574829,
      "epoch": 2.415224302366655,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044152313827846494,
      "loss": 5.4486,
      "mean_token_accuracy": 0.19668371528387069,
      "num_tokens": 50994834.0,
      "step": 27350
    },
    {
      "entropy": 5.782118034362793,
      "epoch": 2.415665842458495,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044150195880765964,
      "loss": 5.4422,
      "mean_token_accuracy": 0.19498540312051774,
      "num_tokens": 51003919.0,
      "step": 27355
    },
    {
      "entropy": 5.836624717712402,
      "epoch": 2.4161073825503356,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044148077607517554,
      "loss": 5.4143,
      "mean_token_accuracy": 0.1957525134086609,
      "num_tokens": 51012868.0,
      "step": 27360
    },
    {
      "entropy": 5.811529207229614,
      "epoch": 2.4165489226421757,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044145959008142784,
      "loss": 5.4567,
      "mean_token_accuracy": 0.19387556314468385,
      "num_tokens": 51021693.0,
      "step": 27365
    },
    {
      "entropy": 5.7912615776062015,
      "epoch": 2.416990462734016,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004414384008268314,
      "loss": 5.4541,
      "mean_token_accuracy": 0.1924322098493576,
      "num_tokens": 51031100.0,
      "step": 27370
    },
    {
      "entropy": 5.768317222595215,
      "epoch": 2.4174320028258567,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044141720831180134,
      "loss": 5.5008,
      "mean_token_accuracy": 0.19407751262187958,
      "num_tokens": 51041077.0,
      "step": 27375
    },
    {
      "entropy": 5.78004264831543,
      "epoch": 2.4178735429176967,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004413960125367527,
      "loss": 5.4287,
      "mean_token_accuracy": 0.19105933904647826,
      "num_tokens": 51050403.0,
      "step": 27380
    },
    {
      "entropy": 5.801247930526733,
      "epoch": 2.4183150830095372,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044137481350210093,
      "loss": 5.4665,
      "mean_token_accuracy": 0.18857946544885634,
      "num_tokens": 51060119.0,
      "step": 27385
    },
    {
      "entropy": 5.8120245933532715,
      "epoch": 2.4187566231013777,
      "grad_norm": 1.359375,
      "learning_rate": 0.00044135361120826116,
      "loss": 5.5316,
      "mean_token_accuracy": 0.18843471854925156,
      "num_tokens": 51070355.0,
      "step": 27390
    },
    {
      "entropy": 5.771323585510254,
      "epoch": 2.419198163193218,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004413324056556488,
      "loss": 5.3839,
      "mean_token_accuracy": 0.20373543500900268,
      "num_tokens": 51079386.0,
      "step": 27395
    },
    {
      "entropy": 5.766569375991821,
      "epoch": 2.4196397032850583,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004413111968446792,
      "loss": 5.5265,
      "mean_token_accuracy": 0.18583218306303023,
      "num_tokens": 51089234.0,
      "step": 27400
    },
    {
      "entropy": 5.862168312072754,
      "epoch": 2.420081243376899,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004412899847757678,
      "loss": 5.4571,
      "mean_token_accuracy": 0.19632271975278853,
      "num_tokens": 51099062.0,
      "step": 27405
    },
    {
      "entropy": 5.797570371627808,
      "epoch": 2.420522783468739,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004412687694493302,
      "loss": 5.4056,
      "mean_token_accuracy": 0.19142901450395583,
      "num_tokens": 51108278.0,
      "step": 27410
    },
    {
      "entropy": 5.813800621032715,
      "epoch": 2.4209643235605793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000441247550865782,
      "loss": 5.5135,
      "mean_token_accuracy": 0.1939265877008438,
      "num_tokens": 51117378.0,
      "step": 27415
    },
    {
      "entropy": 5.773219060897827,
      "epoch": 2.42140586365242,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044122632902553877,
      "loss": 5.429,
      "mean_token_accuracy": 0.2043195426464081,
      "num_tokens": 51126323.0,
      "step": 27420
    },
    {
      "entropy": 5.74927773475647,
      "epoch": 2.42184740374426,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004412051039290163,
      "loss": 5.4218,
      "mean_token_accuracy": 0.20480754226446152,
      "num_tokens": 51135734.0,
      "step": 27425
    },
    {
      "entropy": 5.63138370513916,
      "epoch": 2.4222889438361004,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00044118387557663034,
      "loss": 5.3446,
      "mean_token_accuracy": 0.20629447996616362,
      "num_tokens": 51145020.0,
      "step": 27430
    },
    {
      "entropy": 5.821722030639648,
      "epoch": 2.422730483927941,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00044116264396879684,
      "loss": 5.4647,
      "mean_token_accuracy": 0.19139105081558228,
      "num_tokens": 51153908.0,
      "step": 27435
    },
    {
      "entropy": 5.754867506027222,
      "epoch": 2.423172024019781,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004411414091059316,
      "loss": 5.3946,
      "mean_token_accuracy": 0.2011575296521187,
      "num_tokens": 51163547.0,
      "step": 27440
    },
    {
      "entropy": 5.854429769515991,
      "epoch": 2.4236135641116214,
      "grad_norm": 1.125,
      "learning_rate": 0.00044112017098845064,
      "loss": 5.535,
      "mean_token_accuracy": 0.18703172653913497,
      "num_tokens": 51172559.0,
      "step": 27445
    },
    {
      "entropy": 5.767384815216064,
      "epoch": 2.4240551042034615,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004410989296167701,
      "loss": 5.3881,
      "mean_token_accuracy": 0.20362590104341508,
      "num_tokens": 51181455.0,
      "step": 27450
    },
    {
      "entropy": 5.772586297988892,
      "epoch": 2.424496644295302,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044107768499130587,
      "loss": 5.4918,
      "mean_token_accuracy": 0.19115618914365767,
      "num_tokens": 51191149.0,
      "step": 27455
    },
    {
      "entropy": 5.723204469680786,
      "epoch": 2.4249381843871425,
      "grad_norm": 1.125,
      "learning_rate": 0.0004410564371124743,
      "loss": 5.3931,
      "mean_token_accuracy": 0.20494572520256044,
      "num_tokens": 51200702.0,
      "step": 27460
    },
    {
      "entropy": 5.798267364501953,
      "epoch": 2.4253797244789825,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044103518598069156,
      "loss": 5.4928,
      "mean_token_accuracy": 0.19259083569049834,
      "num_tokens": 51209058.0,
      "step": 27465
    },
    {
      "entropy": 5.812151193618774,
      "epoch": 2.425821264570823,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044101393159637396,
      "loss": 5.4526,
      "mean_token_accuracy": 0.2009863883256912,
      "num_tokens": 51219295.0,
      "step": 27470
    },
    {
      "entropy": 5.728121042251587,
      "epoch": 2.4262628046626635,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044099267395993785,
      "loss": 5.4005,
      "mean_token_accuracy": 0.18871483206748962,
      "num_tokens": 51228430.0,
      "step": 27475
    },
    {
      "entropy": 5.764014959335327,
      "epoch": 2.4267043447545036,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004409714130717997,
      "loss": 5.3654,
      "mean_token_accuracy": 0.1990705281496048,
      "num_tokens": 51237804.0,
      "step": 27480
    },
    {
      "entropy": 5.77218337059021,
      "epoch": 2.427145884846344,
      "grad_norm": 1.4375,
      "learning_rate": 0.000440950148932376,
      "loss": 5.4116,
      "mean_token_accuracy": 0.20348803251981734,
      "num_tokens": 51247313.0,
      "step": 27485
    },
    {
      "entropy": 5.72153639793396,
      "epoch": 2.4275874249381846,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044092888154208314,
      "loss": 5.4731,
      "mean_token_accuracy": 0.1907047986984253,
      "num_tokens": 51257310.0,
      "step": 27490
    },
    {
      "entropy": 5.783519649505616,
      "epoch": 2.4280289650300246,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004409076109013379,
      "loss": 5.4174,
      "mean_token_accuracy": 0.19640354067087173,
      "num_tokens": 51267028.0,
      "step": 27495
    },
    {
      "entropy": 5.836852741241455,
      "epoch": 2.428470505121865,
      "grad_norm": 1.671875,
      "learning_rate": 0.000440886337010557,
      "loss": 5.4137,
      "mean_token_accuracy": 0.19373875707387925,
      "num_tokens": 51276756.0,
      "step": 27500
    },
    {
      "entropy": 5.7927490234375,
      "epoch": 2.428912045213705,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000440865059870157,
      "loss": 5.4474,
      "mean_token_accuracy": 0.18949896544218064,
      "num_tokens": 51286491.0,
      "step": 27505
    },
    {
      "entropy": 5.8461638450622555,
      "epoch": 2.4293535853055457,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044084377948055487,
      "loss": 5.5191,
      "mean_token_accuracy": 0.18764615207910537,
      "num_tokens": 51295221.0,
      "step": 27510
    },
    {
      "entropy": 5.8099103450775145,
      "epoch": 2.429795125397386,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044082249584216735,
      "loss": 5.4727,
      "mean_token_accuracy": 0.1933288499712944,
      "num_tokens": 51304637.0,
      "step": 27515
    },
    {
      "entropy": 5.725382661819458,
      "epoch": 2.4302366654892262,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004408012089554115,
      "loss": 5.3714,
      "mean_token_accuracy": 0.20288594216108322,
      "num_tokens": 51314107.0,
      "step": 27520
    },
    {
      "entropy": 5.816642713546753,
      "epoch": 2.4306782055810667,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044077991882070423,
      "loss": 5.4095,
      "mean_token_accuracy": 0.19880228340625763,
      "num_tokens": 51323353.0,
      "step": 27525
    },
    {
      "entropy": 5.762641763687133,
      "epoch": 2.4311197456729072,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004407586254384627,
      "loss": 5.4109,
      "mean_token_accuracy": 0.19983568042516708,
      "num_tokens": 51333771.0,
      "step": 27530
    },
    {
      "entropy": 5.78665943145752,
      "epoch": 2.4315612857647473,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004407373288091039,
      "loss": 5.4261,
      "mean_token_accuracy": 0.1897443041205406,
      "num_tokens": 51343476.0,
      "step": 27535
    },
    {
      "entropy": 5.834537124633789,
      "epoch": 2.432002825856588,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004407160289330452,
      "loss": 5.4065,
      "mean_token_accuracy": 0.1993083342909813,
      "num_tokens": 51352903.0,
      "step": 27540
    },
    {
      "entropy": 5.79502854347229,
      "epoch": 2.4324443659484283,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044069472581070363,
      "loss": 5.4513,
      "mean_token_accuracy": 0.19830765575170517,
      "num_tokens": 51362922.0,
      "step": 27545
    },
    {
      "entropy": 5.791099739074707,
      "epoch": 2.4328859060402683,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044067341944249667,
      "loss": 5.4613,
      "mean_token_accuracy": 0.19405430257320405,
      "num_tokens": 51371789.0,
      "step": 27550
    },
    {
      "entropy": 5.92032380104065,
      "epoch": 2.433327446132109,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044065210982884164,
      "loss": 5.5241,
      "mean_token_accuracy": 0.19215291738510132,
      "num_tokens": 51380989.0,
      "step": 27555
    },
    {
      "entropy": 5.795026445388794,
      "epoch": 2.4337689862239493,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004406307969701559,
      "loss": 5.4949,
      "mean_token_accuracy": 0.19223501831293105,
      "num_tokens": 51390191.0,
      "step": 27560
    },
    {
      "entropy": 5.771899271011352,
      "epoch": 2.4342105263157894,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044060948086685714,
      "loss": 5.4859,
      "mean_token_accuracy": 0.19079049229621886,
      "num_tokens": 51399133.0,
      "step": 27565
    },
    {
      "entropy": 5.753968238830566,
      "epoch": 2.43465206640763,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044058816151936283,
      "loss": 5.3575,
      "mean_token_accuracy": 0.19717190116643907,
      "num_tokens": 51408473.0,
      "step": 27570
    },
    {
      "entropy": 5.708935070037842,
      "epoch": 2.4350936064994704,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044056683892809055,
      "loss": 5.3508,
      "mean_token_accuracy": 0.20753249675035476,
      "num_tokens": 51416944.0,
      "step": 27575
    },
    {
      "entropy": 5.758321809768677,
      "epoch": 2.4355351465913104,
      "grad_norm": 1.125,
      "learning_rate": 0.00044054551309345816,
      "loss": 5.5651,
      "mean_token_accuracy": 0.1917170912027359,
      "num_tokens": 51426193.0,
      "step": 27580
    },
    {
      "entropy": 5.857351970672608,
      "epoch": 2.435976686683151,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004405241840158833,
      "loss": 5.5216,
      "mean_token_accuracy": 0.19644786715507506,
      "num_tokens": 51435718.0,
      "step": 27585
    },
    {
      "entropy": 5.797277212142944,
      "epoch": 2.436418226774991,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004405028516957837,
      "loss": 5.4071,
      "mean_token_accuracy": 0.19782092273235322,
      "num_tokens": 51445127.0,
      "step": 27590
    },
    {
      "entropy": 5.871546936035156,
      "epoch": 2.4368597668668315,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044048151613357745,
      "loss": 5.5167,
      "mean_token_accuracy": 0.1931655749678612,
      "num_tokens": 51454813.0,
      "step": 27595
    },
    {
      "entropy": 5.818877649307251,
      "epoch": 2.437301306958672,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004404601773296824,
      "loss": 5.4511,
      "mean_token_accuracy": 0.2002747744321823,
      "num_tokens": 51464454.0,
      "step": 27600
    },
    {
      "entropy": 5.7700324058532715,
      "epoch": 2.437742847050512,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004404388352845166,
      "loss": 5.4633,
      "mean_token_accuracy": 0.1848347619175911,
      "num_tokens": 51473535.0,
      "step": 27605
    },
    {
      "entropy": 5.8193036079406735,
      "epoch": 2.4381843871423525,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00044041748999849807,
      "loss": 5.4181,
      "mean_token_accuracy": 0.19572326838970183,
      "num_tokens": 51482490.0,
      "step": 27610
    },
    {
      "entropy": 5.791394901275635,
      "epoch": 2.438625927234193,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000440396141472045,
      "loss": 5.4611,
      "mean_token_accuracy": 0.20226321667432784,
      "num_tokens": 51493604.0,
      "step": 27615
    },
    {
      "entropy": 5.738651943206787,
      "epoch": 2.439067467326033,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004403747897055756,
      "loss": 5.458,
      "mean_token_accuracy": 0.1979442983865738,
      "num_tokens": 51502980.0,
      "step": 27620
    },
    {
      "entropy": 5.769508409500122,
      "epoch": 2.4395090074178736,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004403534346995081,
      "loss": 5.43,
      "mean_token_accuracy": 0.1966283291578293,
      "num_tokens": 51512131.0,
      "step": 27625
    },
    {
      "entropy": 5.830193901062012,
      "epoch": 2.439950547509714,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044033207645426093,
      "loss": 5.5191,
      "mean_token_accuracy": 0.1845942422747612,
      "num_tokens": 51521600.0,
      "step": 27630
    },
    {
      "entropy": 5.717329835891723,
      "epoch": 2.440392087601554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004403107149702524,
      "loss": 5.3334,
      "mean_token_accuracy": 0.21512241363525392,
      "num_tokens": 51531123.0,
      "step": 27635
    },
    {
      "entropy": 5.785354709625244,
      "epoch": 2.4408336276933946,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004402893502479009,
      "loss": 5.4239,
      "mean_token_accuracy": 0.19917442351579667,
      "num_tokens": 51540395.0,
      "step": 27640
    },
    {
      "entropy": 5.771020555496216,
      "epoch": 2.4412751677852347,
      "grad_norm": 1.25,
      "learning_rate": 0.000440267982287625,
      "loss": 5.4143,
      "mean_token_accuracy": 0.19777994304895402,
      "num_tokens": 51549683.0,
      "step": 27645
    },
    {
      "entropy": 5.849586534500122,
      "epoch": 2.441716707877075,
      "grad_norm": 1.25,
      "learning_rate": 0.0004402466110898435,
      "loss": 5.6021,
      "mean_token_accuracy": 0.18185747861862184,
      "num_tokens": 51559800.0,
      "step": 27650
    },
    {
      "entropy": 5.794493675231934,
      "epoch": 2.4421582479689157,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044022523665497473,
      "loss": 5.4188,
      "mean_token_accuracy": 0.19704046547412873,
      "num_tokens": 51569788.0,
      "step": 27655
    },
    {
      "entropy": 5.806346845626831,
      "epoch": 2.4425997880607557,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004402038589834375,
      "loss": 5.4645,
      "mean_token_accuracy": 0.19441838562488556,
      "num_tokens": 51578598.0,
      "step": 27660
    },
    {
      "entropy": 5.8465879440307615,
      "epoch": 2.4430413281525962,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004401824780756507,
      "loss": 5.5514,
      "mean_token_accuracy": 0.17931957840919494,
      "num_tokens": 51589203.0,
      "step": 27665
    },
    {
      "entropy": 5.83199372291565,
      "epoch": 2.4434828682444367,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004401610939320331,
      "loss": 5.43,
      "mean_token_accuracy": 0.19927516877651213,
      "num_tokens": 51597813.0,
      "step": 27670
    },
    {
      "entropy": 5.798725271224976,
      "epoch": 2.4439244083362768,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044013970655300364,
      "loss": 5.4844,
      "mean_token_accuracy": 0.1946522042155266,
      "num_tokens": 51607386.0,
      "step": 27675
    },
    {
      "entropy": 5.8737931728363035,
      "epoch": 2.4443659484281173,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044011831593898117,
      "loss": 5.5277,
      "mean_token_accuracy": 0.18864864856004715,
      "num_tokens": 51617091.0,
      "step": 27680
    },
    {
      "entropy": 5.852108097076416,
      "epoch": 2.4448074885199578,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044009692209038485,
      "loss": 5.5324,
      "mean_token_accuracy": 0.19230537116527557,
      "num_tokens": 51625616.0,
      "step": 27685
    },
    {
      "entropy": 5.751078987121582,
      "epoch": 2.445249028611798,
      "grad_norm": 1.125,
      "learning_rate": 0.0004400755250076337,
      "loss": 5.363,
      "mean_token_accuracy": 0.2072572559118271,
      "num_tokens": 51634693.0,
      "step": 27690
    },
    {
      "entropy": 5.769868087768555,
      "epoch": 2.4456905687036383,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044005412469114684,
      "loss": 5.34,
      "mean_token_accuracy": 0.21002181470394135,
      "num_tokens": 51642549.0,
      "step": 27695
    },
    {
      "entropy": 5.800993585586548,
      "epoch": 2.446132108795479,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044003272114134367,
      "loss": 5.4223,
      "mean_token_accuracy": 0.2014555886387825,
      "num_tokens": 51651742.0,
      "step": 27700
    },
    {
      "entropy": 5.8393933296203615,
      "epoch": 2.446573648887319,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004400113143586433,
      "loss": 5.4797,
      "mean_token_accuracy": 0.19699356853961944,
      "num_tokens": 51660929.0,
      "step": 27705
    },
    {
      "entropy": 5.826835966110229,
      "epoch": 2.4470151889791594,
      "grad_norm": 1.125,
      "learning_rate": 0.00043998990434346514,
      "loss": 5.4636,
      "mean_token_accuracy": 0.18739264607429504,
      "num_tokens": 51670580.0,
      "step": 27710
    },
    {
      "entropy": 5.7449744701385494,
      "epoch": 2.447456729071,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004399684910962286,
      "loss": 5.4633,
      "mean_token_accuracy": 0.19812580794095994,
      "num_tokens": 51680184.0,
      "step": 27715
    },
    {
      "entropy": 5.708809471130371,
      "epoch": 2.44789826916284,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004399470746173531,
      "loss": 5.4244,
      "mean_token_accuracy": 0.19982363283634186,
      "num_tokens": 51690277.0,
      "step": 27720
    },
    {
      "entropy": 5.784022378921509,
      "epoch": 2.4483398092546804,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004399256549072583,
      "loss": 5.4107,
      "mean_token_accuracy": 0.2034461036324501,
      "num_tokens": 51699316.0,
      "step": 27725
    },
    {
      "entropy": 5.80828104019165,
      "epoch": 2.4487813493465205,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043990423196636357,
      "loss": 5.4391,
      "mean_token_accuracy": 0.19884395748376846,
      "num_tokens": 51708552.0,
      "step": 27730
    },
    {
      "entropy": 5.887029075622559,
      "epoch": 2.449222889438361,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004398828057950888,
      "loss": 5.4165,
      "mean_token_accuracy": 0.19312669485807418,
      "num_tokens": 51717082.0,
      "step": 27735
    },
    {
      "entropy": 5.756124496459961,
      "epoch": 2.4496644295302015,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004398613763938537,
      "loss": 5.4913,
      "mean_token_accuracy": 0.19275465607643127,
      "num_tokens": 51727059.0,
      "step": 27740
    },
    {
      "entropy": 5.673369884490967,
      "epoch": 2.4501059696220415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043983994376307797,
      "loss": 5.3514,
      "mean_token_accuracy": 0.21173029094934465,
      "num_tokens": 51735889.0,
      "step": 27745
    },
    {
      "entropy": 5.827050590515137,
      "epoch": 2.450547509713882,
      "grad_norm": 1.25,
      "learning_rate": 0.00043981850790318145,
      "loss": 5.522,
      "mean_token_accuracy": 0.1873005822300911,
      "num_tokens": 51744856.0,
      "step": 27750
    },
    {
      "entropy": 5.77336015701294,
      "epoch": 2.4509890498057225,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043979706881458415,
      "loss": 5.4108,
      "mean_token_accuracy": 0.2039515718817711,
      "num_tokens": 51753552.0,
      "step": 27755
    },
    {
      "entropy": 5.804682779312134,
      "epoch": 2.4514305898975626,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043977562649770587,
      "loss": 5.5314,
      "mean_token_accuracy": 0.18602571189403533,
      "num_tokens": 51762929.0,
      "step": 27760
    },
    {
      "entropy": 5.896048545837402,
      "epoch": 2.451872129989403,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043975418095296696,
      "loss": 5.5404,
      "mean_token_accuracy": 0.18902628272771835,
      "num_tokens": 51772732.0,
      "step": 27765
    },
    {
      "entropy": 5.8919970989227295,
      "epoch": 2.4523136700812436,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043973273218078716,
      "loss": 5.5409,
      "mean_token_accuracy": 0.19017400741577148,
      "num_tokens": 51782304.0,
      "step": 27770
    },
    {
      "entropy": 5.766681861877442,
      "epoch": 2.4527552101730836,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043971128018158693,
      "loss": 5.3803,
      "mean_token_accuracy": 0.2057773768901825,
      "num_tokens": 51791492.0,
      "step": 27775
    },
    {
      "entropy": 5.764233493804932,
      "epoch": 2.453196750264924,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004396898249557864,
      "loss": 5.4801,
      "mean_token_accuracy": 0.19061116725206376,
      "num_tokens": 51800906.0,
      "step": 27780
    },
    {
      "entropy": 5.756083917617798,
      "epoch": 2.453638290356764,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043966836650380587,
      "loss": 5.332,
      "mean_token_accuracy": 0.2068623349070549,
      "num_tokens": 51809071.0,
      "step": 27785
    },
    {
      "entropy": 5.821525573730469,
      "epoch": 2.4540798304486047,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043964690482606566,
      "loss": 5.4204,
      "mean_token_accuracy": 0.19826697707176208,
      "num_tokens": 51817450.0,
      "step": 27790
    },
    {
      "entropy": 5.825331735610962,
      "epoch": 2.454521370540445,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043962543992298624,
      "loss": 5.4575,
      "mean_token_accuracy": 0.1933839052915573,
      "num_tokens": 51826400.0,
      "step": 27795
    },
    {
      "entropy": 5.783517503738404,
      "epoch": 2.4549629106322852,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004396039717949881,
      "loss": 5.4442,
      "mean_token_accuracy": 0.19957921355962754,
      "num_tokens": 51835415.0,
      "step": 27800
    },
    {
      "entropy": 5.813741874694824,
      "epoch": 2.4554044507241257,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004395825004424917,
      "loss": 5.4702,
      "mean_token_accuracy": 0.2017652302980423,
      "num_tokens": 51844914.0,
      "step": 27805
    },
    {
      "entropy": 5.777150392532349,
      "epoch": 2.455845990815966,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004395610258659178,
      "loss": 5.4559,
      "mean_token_accuracy": 0.196439927816391,
      "num_tokens": 51854178.0,
      "step": 27810
    },
    {
      "entropy": 5.811684417724609,
      "epoch": 2.4562875309078063,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043953954806568693,
      "loss": 5.568,
      "mean_token_accuracy": 0.18368038386106492,
      "num_tokens": 51863184.0,
      "step": 27815
    },
    {
      "entropy": 5.806237506866455,
      "epoch": 2.4567290709996468,
      "grad_norm": 1.25,
      "learning_rate": 0.00043951806704221994,
      "loss": 5.4887,
      "mean_token_accuracy": 0.20097100883722305,
      "num_tokens": 51872651.0,
      "step": 27820
    },
    {
      "entropy": 5.737590169906616,
      "epoch": 2.4571706110914873,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004394965827959376,
      "loss": 5.3979,
      "mean_token_accuracy": 0.20108057409524918,
      "num_tokens": 51881527.0,
      "step": 27825
    },
    {
      "entropy": 5.854833269119263,
      "epoch": 2.4576121511833273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043947509532726067,
      "loss": 5.5259,
      "mean_token_accuracy": 0.18836090266704558,
      "num_tokens": 51892083.0,
      "step": 27830
    },
    {
      "entropy": 5.782758808135986,
      "epoch": 2.458053691275168,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004394536046366102,
      "loss": 5.4031,
      "mean_token_accuracy": 0.1980675756931305,
      "num_tokens": 51900758.0,
      "step": 27835
    },
    {
      "entropy": 5.805979251861572,
      "epoch": 2.4584952313670083,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004394321107244072,
      "loss": 5.48,
      "mean_token_accuracy": 0.19284168183803557,
      "num_tokens": 51910624.0,
      "step": 27840
    },
    {
      "entropy": 5.616035890579224,
      "epoch": 2.4589367714588484,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004394106135910726,
      "loss": 5.2375,
      "mean_token_accuracy": 0.20686671286821365,
      "num_tokens": 51919306.0,
      "step": 27845
    },
    {
      "entropy": 5.802901840209961,
      "epoch": 2.459378311550689,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043938911323702766,
      "loss": 5.4618,
      "mean_token_accuracy": 0.19242972433567046,
      "num_tokens": 51928650.0,
      "step": 27850
    },
    {
      "entropy": 5.788751554489136,
      "epoch": 2.4598198516425294,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004393676096626934,
      "loss": 5.4923,
      "mean_token_accuracy": 0.19150221794843675,
      "num_tokens": 51939198.0,
      "step": 27855
    },
    {
      "entropy": 5.63243556022644,
      "epoch": 2.4602613917343694,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043934610286849114,
      "loss": 5.1903,
      "mean_token_accuracy": 0.22791864573955536,
      "num_tokens": 51948641.0,
      "step": 27860
    },
    {
      "entropy": 5.795641803741455,
      "epoch": 2.46070293182621,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004393245928548423,
      "loss": 5.424,
      "mean_token_accuracy": 0.20119185298681258,
      "num_tokens": 51958258.0,
      "step": 27865
    },
    {
      "entropy": 5.793450593948364,
      "epoch": 2.46114447191805,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043930307962216806,
      "loss": 5.3914,
      "mean_token_accuracy": 0.19903823137283325,
      "num_tokens": 51967295.0,
      "step": 27870
    },
    {
      "entropy": 5.592277336120605,
      "epoch": 2.4615860120098905,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043928156317088994,
      "loss": 5.3167,
      "mean_token_accuracy": 0.20957217067480088,
      "num_tokens": 51976450.0,
      "step": 27875
    },
    {
      "entropy": 5.747349262237549,
      "epoch": 2.462027552101731,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004392600435014295,
      "loss": 5.4258,
      "mean_token_accuracy": 0.19364499002695085,
      "num_tokens": 51985782.0,
      "step": 27880
    },
    {
      "entropy": 5.862791919708252,
      "epoch": 2.462469092193571,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043923852061420815,
      "loss": 5.4826,
      "mean_token_accuracy": 0.1991618439555168,
      "num_tokens": 51994745.0,
      "step": 27885
    },
    {
      "entropy": 5.865579223632812,
      "epoch": 2.4629106322854115,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004392169945096476,
      "loss": 5.515,
      "mean_token_accuracy": 0.18966845721006392,
      "num_tokens": 52004369.0,
      "step": 27890
    },
    {
      "entropy": 5.790476083755493,
      "epoch": 2.463352172377252,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043919546518816953,
      "loss": 5.4705,
      "mean_token_accuracy": 0.20003703385591506,
      "num_tokens": 52012556.0,
      "step": 27895
    },
    {
      "entropy": 5.78197546005249,
      "epoch": 2.463793712469092,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004391739326501957,
      "loss": 5.5309,
      "mean_token_accuracy": 0.18818745464086534,
      "num_tokens": 52022854.0,
      "step": 27900
    },
    {
      "entropy": 5.828859663009643,
      "epoch": 2.4642352525609326,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004391523968961479,
      "loss": 5.4645,
      "mean_token_accuracy": 0.19385757744312287,
      "num_tokens": 52032924.0,
      "step": 27905
    },
    {
      "entropy": 5.792382764816284,
      "epoch": 2.464676792652773,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000439130857926448,
      "loss": 5.429,
      "mean_token_accuracy": 0.19936145842075348,
      "num_tokens": 52042625.0,
      "step": 27910
    },
    {
      "entropy": 5.736405706405639,
      "epoch": 2.465118332744613,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004391093157415179,
      "loss": 5.4195,
      "mean_token_accuracy": 0.202858367562294,
      "num_tokens": 52051964.0,
      "step": 27915
    },
    {
      "entropy": 5.756160545349121,
      "epoch": 2.4655598728364536,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004390877703417797,
      "loss": 5.4449,
      "mean_token_accuracy": 0.1960089847445488,
      "num_tokens": 52060877.0,
      "step": 27920
    },
    {
      "entropy": 5.8335737705230715,
      "epoch": 2.4660014129282937,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043906622172765533,
      "loss": 5.5366,
      "mean_token_accuracy": 0.19096793979406357,
      "num_tokens": 52069585.0,
      "step": 27925
    },
    {
      "entropy": 5.815985774993896,
      "epoch": 2.466442953020134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043904466989956706,
      "loss": 5.4994,
      "mean_token_accuracy": 0.19221670180559158,
      "num_tokens": 52078435.0,
      "step": 27930
    },
    {
      "entropy": 5.845651912689209,
      "epoch": 2.4668844931119747,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043902311485793694,
      "loss": 5.4796,
      "mean_token_accuracy": 0.1910724312067032,
      "num_tokens": 52088227.0,
      "step": 27935
    },
    {
      "entropy": 5.7881818294525145,
      "epoch": 2.4673260332038147,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004390015566031873,
      "loss": 5.4645,
      "mean_token_accuracy": 0.19076981991529465,
      "num_tokens": 52098101.0,
      "step": 27940
    },
    {
      "entropy": 5.762052202224732,
      "epoch": 2.467767573295655,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043897999513574045,
      "loss": 5.409,
      "mean_token_accuracy": 0.2057751640677452,
      "num_tokens": 52107370.0,
      "step": 27945
    },
    {
      "entropy": 5.659012031555176,
      "epoch": 2.4682091133874957,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043895843045601874,
      "loss": 5.4225,
      "mean_token_accuracy": 0.2001279413700104,
      "num_tokens": 52116808.0,
      "step": 27950
    },
    {
      "entropy": 5.835477447509765,
      "epoch": 2.4686506534793358,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043893686256444463,
      "loss": 5.3702,
      "mean_token_accuracy": 0.20513463765382767,
      "num_tokens": 52125724.0,
      "step": 27955
    },
    {
      "entropy": 5.852635765075684,
      "epoch": 2.4690921935711763,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004389152914614407,
      "loss": 5.5178,
      "mean_token_accuracy": 0.1892518639564514,
      "num_tokens": 52135530.0,
      "step": 27960
    },
    {
      "entropy": 5.7819251537323,
      "epoch": 2.4695337336630168,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004388937171474293,
      "loss": 5.4763,
      "mean_token_accuracy": 0.1932560384273529,
      "num_tokens": 52144195.0,
      "step": 27965
    },
    {
      "entropy": 5.8022668838500975,
      "epoch": 2.469975273754857,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043887213962283335,
      "loss": 5.5075,
      "mean_token_accuracy": 0.1915068283677101,
      "num_tokens": 52153675.0,
      "step": 27970
    },
    {
      "entropy": 5.757124948501587,
      "epoch": 2.4704168138466973,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043885055888807516,
      "loss": 5.4224,
      "mean_token_accuracy": 0.19865225553512572,
      "num_tokens": 52163455.0,
      "step": 27975
    },
    {
      "entropy": 5.78288311958313,
      "epoch": 2.470858353938538,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004388289749435779,
      "loss": 5.4418,
      "mean_token_accuracy": 0.19803967475891113,
      "num_tokens": 52172054.0,
      "step": 27980
    },
    {
      "entropy": 5.821660709381104,
      "epoch": 2.471299894030378,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004388073877897641,
      "loss": 5.5391,
      "mean_token_accuracy": 0.18846694082021714,
      "num_tokens": 52181667.0,
      "step": 27985
    },
    {
      "entropy": 5.8501382827758786,
      "epoch": 2.4717414341222184,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043878579742705675,
      "loss": 5.4301,
      "mean_token_accuracy": 0.1943674147129059,
      "num_tokens": 52190472.0,
      "step": 27990
    },
    {
      "entropy": 5.778086566925049,
      "epoch": 2.472182974214059,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043876420385587886,
      "loss": 5.3657,
      "mean_token_accuracy": 0.2015601634979248,
      "num_tokens": 52199905.0,
      "step": 27995
    },
    {
      "entropy": 5.759410333633423,
      "epoch": 2.472624514305899,
      "grad_norm": 1.125,
      "learning_rate": 0.00043874260707665327,
      "loss": 5.4703,
      "mean_token_accuracy": 0.18932579606771469,
      "num_tokens": 52209270.0,
      "step": 28000
    },
    {
      "entropy": 5.752660083770752,
      "epoch": 2.4730660543977394,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004387210070898032,
      "loss": 5.4474,
      "mean_token_accuracy": 0.1922023266553879,
      "num_tokens": 52218144.0,
      "step": 28005
    },
    {
      "entropy": 5.825842952728271,
      "epoch": 2.4735075944895795,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004386994038957516,
      "loss": 5.4088,
      "mean_token_accuracy": 0.19953516870737076,
      "num_tokens": 52227128.0,
      "step": 28010
    },
    {
      "entropy": 5.776297998428345,
      "epoch": 2.47394913458142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004386777974949219,
      "loss": 5.4327,
      "mean_token_accuracy": 0.19490528255701065,
      "num_tokens": 52235823.0,
      "step": 28015
    },
    {
      "entropy": 5.7777674198150635,
      "epoch": 2.4743906746732605,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004386561878877372,
      "loss": 5.5593,
      "mean_token_accuracy": 0.19079759120941162,
      "num_tokens": 52245510.0,
      "step": 28020
    },
    {
      "entropy": 5.859488010406494,
      "epoch": 2.4748322147651005,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043863457507462076,
      "loss": 5.4146,
      "mean_token_accuracy": 0.19871685057878494,
      "num_tokens": 52254583.0,
      "step": 28025
    },
    {
      "entropy": 5.847831058502197,
      "epoch": 2.475273754856941,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004386129590559962,
      "loss": 5.4598,
      "mean_token_accuracy": 0.19620157331228255,
      "num_tokens": 52264778.0,
      "step": 28030
    },
    {
      "entropy": 5.733920478820801,
      "epoch": 2.4757152949487815,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004385913398322867,
      "loss": 5.4717,
      "mean_token_accuracy": 0.19173682928085328,
      "num_tokens": 52274959.0,
      "step": 28035
    },
    {
      "entropy": 5.781056594848633,
      "epoch": 2.4761568350406216,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000438569717403916,
      "loss": 5.4102,
      "mean_token_accuracy": 0.19583964645862578,
      "num_tokens": 52284606.0,
      "step": 28040
    },
    {
      "entropy": 5.768957042694092,
      "epoch": 2.476598375132462,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004385480917713075,
      "loss": 5.483,
      "mean_token_accuracy": 0.19487336724996568,
      "num_tokens": 52294571.0,
      "step": 28045
    },
    {
      "entropy": 5.823685884475708,
      "epoch": 2.4770399152243026,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004385264629348849,
      "loss": 5.4246,
      "mean_token_accuracy": 0.1963655322790146,
      "num_tokens": 52303575.0,
      "step": 28050
    },
    {
      "entropy": 5.729952096939087,
      "epoch": 2.4774814553161426,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004385048308950719,
      "loss": 5.3919,
      "mean_token_accuracy": 0.1963716119527817,
      "num_tokens": 52312863.0,
      "step": 28055
    },
    {
      "entropy": 5.809867095947266,
      "epoch": 2.477922995407983,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004384831956522922,
      "loss": 5.4705,
      "mean_token_accuracy": 0.1987553521990776,
      "num_tokens": 52322493.0,
      "step": 28060
    },
    {
      "entropy": 5.76922721862793,
      "epoch": 2.478364535499823,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043846155720696976,
      "loss": 5.3777,
      "mean_token_accuracy": 0.20554141849279403,
      "num_tokens": 52331159.0,
      "step": 28065
    },
    {
      "entropy": 5.801751899719238,
      "epoch": 2.4788060755916637,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043843991555952847,
      "loss": 5.4322,
      "mean_token_accuracy": 0.19246553182601928,
      "num_tokens": 52340276.0,
      "step": 28070
    },
    {
      "entropy": 5.808242702484131,
      "epoch": 2.479247615683504,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043841827071039197,
      "loss": 5.4577,
      "mean_token_accuracy": 0.19149312525987625,
      "num_tokens": 52349666.0,
      "step": 28075
    },
    {
      "entropy": 5.702538251876831,
      "epoch": 2.479689155775344,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043839662265998463,
      "loss": 5.4045,
      "mean_token_accuracy": 0.20727392584085463,
      "num_tokens": 52358238.0,
      "step": 28080
    },
    {
      "entropy": 5.735316276550293,
      "epoch": 2.4801306958671847,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043837497140873037,
      "loss": 5.4074,
      "mean_token_accuracy": 0.20141619443893433,
      "num_tokens": 52367663.0,
      "step": 28085
    },
    {
      "entropy": 5.755250597000122,
      "epoch": 2.480572235959025,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043835331695705326,
      "loss": 5.441,
      "mean_token_accuracy": 0.20140092372894286,
      "num_tokens": 52377320.0,
      "step": 28090
    },
    {
      "entropy": 5.794341611862182,
      "epoch": 2.4810137760508653,
      "grad_norm": 1.125,
      "learning_rate": 0.0004383316593053777,
      "loss": 5.4912,
      "mean_token_accuracy": 0.19072606265544892,
      "num_tokens": 52386231.0,
      "step": 28095
    },
    {
      "entropy": 5.795045042037964,
      "epoch": 2.4814553161427058,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004383099984541277,
      "loss": 5.5131,
      "mean_token_accuracy": 0.18681149333715438,
      "num_tokens": 52395487.0,
      "step": 28100
    },
    {
      "entropy": 5.8046690940856935,
      "epoch": 2.4818968562345463,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043828833440372775,
      "loss": 5.4502,
      "mean_token_accuracy": 0.19617307037115098,
      "num_tokens": 52404372.0,
      "step": 28105
    },
    {
      "entropy": 5.670149850845337,
      "epoch": 2.4823383963263863,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004382666671546022,
      "loss": 5.3051,
      "mean_token_accuracy": 0.21108764559030532,
      "num_tokens": 52413156.0,
      "step": 28110
    },
    {
      "entropy": 5.775538063049316,
      "epoch": 2.482779936418227,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043824499670717547,
      "loss": 5.4502,
      "mean_token_accuracy": 0.19392977505922318,
      "num_tokens": 52422782.0,
      "step": 28115
    },
    {
      "entropy": 5.84697265625,
      "epoch": 2.4832214765100673,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043822332306187213,
      "loss": 5.462,
      "mean_token_accuracy": 0.1879718467593193,
      "num_tokens": 52432363.0,
      "step": 28120
    },
    {
      "entropy": 5.754677581787109,
      "epoch": 2.4836630166019074,
      "grad_norm": 1.125,
      "learning_rate": 0.0004382016462191167,
      "loss": 5.3987,
      "mean_token_accuracy": 0.1948802649974823,
      "num_tokens": 52440995.0,
      "step": 28125
    },
    {
      "entropy": 5.701654005050659,
      "epoch": 2.484104556693748,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043817996617933393,
      "loss": 5.4185,
      "mean_token_accuracy": 0.19335890114307402,
      "num_tokens": 52450135.0,
      "step": 28130
    },
    {
      "entropy": 5.795718765258789,
      "epoch": 2.4845460967855884,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004381582829429483,
      "loss": 5.4025,
      "mean_token_accuracy": 0.20046584010124208,
      "num_tokens": 52459065.0,
      "step": 28135
    },
    {
      "entropy": 5.77596230506897,
      "epoch": 2.4849876368774284,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004381365965103848,
      "loss": 5.4166,
      "mean_token_accuracy": 0.20168082267045975,
      "num_tokens": 52468382.0,
      "step": 28140
    },
    {
      "entropy": 5.712625026702881,
      "epoch": 2.485429176969269,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004381149068820681,
      "loss": 5.4115,
      "mean_token_accuracy": 0.19912420958280563,
      "num_tokens": 52476940.0,
      "step": 28145
    },
    {
      "entropy": 5.81462459564209,
      "epoch": 2.485870717061109,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043809321405842314,
      "loss": 5.4992,
      "mean_token_accuracy": 0.19091802835464478,
      "num_tokens": 52486496.0,
      "step": 28150
    },
    {
      "entropy": 5.864582633972168,
      "epoch": 2.4863122571529495,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004380715180398749,
      "loss": 5.4728,
      "mean_token_accuracy": 0.189738430082798,
      "num_tokens": 52495778.0,
      "step": 28155
    },
    {
      "entropy": 5.762618732452393,
      "epoch": 2.48675379724479,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004380498188268483,
      "loss": 5.3386,
      "mean_token_accuracy": 0.20340392291545867,
      "num_tokens": 52504855.0,
      "step": 28160
    },
    {
      "entropy": 5.736073589324951,
      "epoch": 2.48719533733663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043802811641976855,
      "loss": 5.3895,
      "mean_token_accuracy": 0.19577704519033431,
      "num_tokens": 52513159.0,
      "step": 28165
    },
    {
      "entropy": 5.784492540359497,
      "epoch": 2.4876368774284705,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004380064108190607,
      "loss": 5.4414,
      "mean_token_accuracy": 0.19104363024234772,
      "num_tokens": 52521869.0,
      "step": 28170
    },
    {
      "entropy": 5.783083534240722,
      "epoch": 2.488078417520311,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004379847020251499,
      "loss": 5.5277,
      "mean_token_accuracy": 0.19667840749025345,
      "num_tokens": 52531886.0,
      "step": 28175
    },
    {
      "entropy": 5.771805620193481,
      "epoch": 2.488519957612151,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043796299003846155,
      "loss": 5.3806,
      "mean_token_accuracy": 0.20314180850982666,
      "num_tokens": 52540331.0,
      "step": 28180
    },
    {
      "entropy": 5.795155763626099,
      "epoch": 2.4889614977039916,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043794127485942094,
      "loss": 5.5282,
      "mean_token_accuracy": 0.1924733802676201,
      "num_tokens": 52549686.0,
      "step": 28185
    },
    {
      "entropy": 5.715784215927124,
      "epoch": 2.489403037795832,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004379195564884534,
      "loss": 5.3768,
      "mean_token_accuracy": 0.20315005779266357,
      "num_tokens": 52558722.0,
      "step": 28190
    },
    {
      "entropy": 5.813825464248657,
      "epoch": 2.489844577887672,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004378978349259844,
      "loss": 5.4951,
      "mean_token_accuracy": 0.19003532975912094,
      "num_tokens": 52568884.0,
      "step": 28195
    },
    {
      "entropy": 5.799049949645996,
      "epoch": 2.4902861179795126,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043787611017243944,
      "loss": 5.4576,
      "mean_token_accuracy": 0.19243974685668946,
      "num_tokens": 52578519.0,
      "step": 28200
    },
    {
      "entropy": 5.7391821384429935,
      "epoch": 2.4907276580713527,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043785438222824406,
      "loss": 5.3517,
      "mean_token_accuracy": 0.20697413384914398,
      "num_tokens": 52588496.0,
      "step": 28205
    },
    {
      "entropy": 5.807396602630615,
      "epoch": 2.491169198163193,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043783265109382407,
      "loss": 5.4646,
      "mean_token_accuracy": 0.1915655717253685,
      "num_tokens": 52598463.0,
      "step": 28210
    },
    {
      "entropy": 5.704466676712036,
      "epoch": 2.4916107382550337,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043781091676960507,
      "loss": 5.3726,
      "mean_token_accuracy": 0.19998743236064911,
      "num_tokens": 52607257.0,
      "step": 28215
    },
    {
      "entropy": 5.814695596694946,
      "epoch": 2.4920522783468737,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004377891792560127,
      "loss": 5.4562,
      "mean_token_accuracy": 0.19117660224437713,
      "num_tokens": 52616434.0,
      "step": 28220
    },
    {
      "entropy": 5.831390857696533,
      "epoch": 2.492493818438714,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043776743855347287,
      "loss": 5.5598,
      "mean_token_accuracy": 0.19129034280776977,
      "num_tokens": 52625564.0,
      "step": 28225
    },
    {
      "entropy": 5.881352758407592,
      "epoch": 2.4929353585305547,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043774569466241165,
      "loss": 5.5631,
      "mean_token_accuracy": 0.1898106098175049,
      "num_tokens": 52634081.0,
      "step": 28230
    },
    {
      "entropy": 5.74567756652832,
      "epoch": 2.4933768986223948,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004377239475832547,
      "loss": 5.4312,
      "mean_token_accuracy": 0.2028225392103195,
      "num_tokens": 52643722.0,
      "step": 28235
    },
    {
      "entropy": 5.78576922416687,
      "epoch": 2.4938184387142353,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004377021973164281,
      "loss": 5.4297,
      "mean_token_accuracy": 0.19984285831451415,
      "num_tokens": 52652831.0,
      "step": 28240
    },
    {
      "entropy": 5.911652946472168,
      "epoch": 2.4942599788060758,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004376804438623582,
      "loss": 5.5929,
      "mean_token_accuracy": 0.18875700384378433,
      "num_tokens": 52663139.0,
      "step": 28245
    },
    {
      "entropy": 5.828060626983643,
      "epoch": 2.494701518897916,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043765868722147076,
      "loss": 5.4454,
      "mean_token_accuracy": 0.20011227130889891,
      "num_tokens": 52672281.0,
      "step": 28250
    },
    {
      "entropy": 5.755427312850952,
      "epoch": 2.4951430589897563,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043763692739419224,
      "loss": 5.3858,
      "mean_token_accuracy": 0.20224322080612184,
      "num_tokens": 52681644.0,
      "step": 28255
    },
    {
      "entropy": 5.879546070098877,
      "epoch": 2.495584599081597,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004376151643809487,
      "loss": 5.503,
      "mean_token_accuracy": 0.187303626537323,
      "num_tokens": 52690706.0,
      "step": 28260
    },
    {
      "entropy": 5.860677909851074,
      "epoch": 2.496026139173437,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004375933981821667,
      "loss": 5.4433,
      "mean_token_accuracy": 0.19580219238996505,
      "num_tokens": 52700391.0,
      "step": 28265
    },
    {
      "entropy": 5.7731680393219,
      "epoch": 2.4964676792652774,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043757162879827245,
      "loss": 5.44,
      "mean_token_accuracy": 0.19711066484451295,
      "num_tokens": 52709586.0,
      "step": 28270
    },
    {
      "entropy": 5.710849332809448,
      "epoch": 2.496909219357118,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004375498562296924,
      "loss": 5.4261,
      "mean_token_accuracy": 0.20396245419979095,
      "num_tokens": 52718926.0,
      "step": 28275
    },
    {
      "entropy": 5.780233764648438,
      "epoch": 2.497350759448958,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004375280804768532,
      "loss": 5.3699,
      "mean_token_accuracy": 0.20284511148929596,
      "num_tokens": 52728311.0,
      "step": 28280
    },
    {
      "entropy": 5.9010144710540775,
      "epoch": 2.4977922995407984,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004375063015401813,
      "loss": 5.4949,
      "mean_token_accuracy": 0.19400326162576675,
      "num_tokens": 52737758.0,
      "step": 28285
    },
    {
      "entropy": 5.781233596801758,
      "epoch": 2.4982338396326385,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043748451942010337,
      "loss": 5.4906,
      "mean_token_accuracy": 0.19875889420509338,
      "num_tokens": 52746959.0,
      "step": 28290
    },
    {
      "entropy": 5.789085006713867,
      "epoch": 2.498675379724479,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004374627341170461,
      "loss": 5.4874,
      "mean_token_accuracy": 0.19044910371303558,
      "num_tokens": 52757315.0,
      "step": 28295
    },
    {
      "entropy": 5.817146968841553,
      "epoch": 2.4991169198163194,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004374409456314362,
      "loss": 5.4781,
      "mean_token_accuracy": 0.19748447388410567,
      "num_tokens": 52766927.0,
      "step": 28300
    },
    {
      "entropy": 5.769997787475586,
      "epoch": 2.4995584599081595,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004374191539637007,
      "loss": 5.4826,
      "mean_token_accuracy": 0.19313886612653733,
      "num_tokens": 52775103.0,
      "step": 28305
    },
    {
      "entropy": 5.78001856803894,
      "epoch": 2.5,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004373973591142663,
      "loss": 5.496,
      "mean_token_accuracy": 0.19855393320322037,
      "num_tokens": 52785281.0,
      "step": 28310
    },
    {
      "entropy": 5.785654878616333,
      "epoch": 2.5004415400918405,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004373755610835599,
      "loss": 5.3611,
      "mean_token_accuracy": 0.19849399626255035,
      "num_tokens": 52793593.0,
      "step": 28315
    },
    {
      "entropy": 5.8562541007995605,
      "epoch": 2.5008830801836806,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043735375987200877,
      "loss": 5.4341,
      "mean_token_accuracy": 0.19048701822757722,
      "num_tokens": 52803103.0,
      "step": 28320
    },
    {
      "entropy": 5.8651299476623535,
      "epoch": 2.501324620275521,
      "grad_norm": 1.25,
      "learning_rate": 0.0004373319554800397,
      "loss": 5.4896,
      "mean_token_accuracy": 0.1937492683529854,
      "num_tokens": 52811928.0,
      "step": 28325
    },
    {
      "entropy": 5.692289400100708,
      "epoch": 2.501766160367361,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043731014790808,
      "loss": 5.3018,
      "mean_token_accuracy": 0.2115025192499161,
      "num_tokens": 52821352.0,
      "step": 28330
    },
    {
      "entropy": 5.669208574295044,
      "epoch": 2.5022077004592016,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004372883371565568,
      "loss": 5.3535,
      "mean_token_accuracy": 0.20945358723402024,
      "num_tokens": 52830688.0,
      "step": 28335
    },
    {
      "entropy": 5.743948602676392,
      "epoch": 2.502649240551042,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043726652322589735,
      "loss": 5.398,
      "mean_token_accuracy": 0.20054597556591033,
      "num_tokens": 52840804.0,
      "step": 28340
    },
    {
      "entropy": 5.83082275390625,
      "epoch": 2.503090780642882,
      "grad_norm": 1.171875,
      "learning_rate": 0.000437244706116529,
      "loss": 5.4343,
      "mean_token_accuracy": 0.1878958150744438,
      "num_tokens": 52850476.0,
      "step": 28345
    },
    {
      "entropy": 5.7955970287323,
      "epoch": 2.5035323207347226,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043722288582887913,
      "loss": 5.5288,
      "mean_token_accuracy": 0.18858749866485597,
      "num_tokens": 52859254.0,
      "step": 28350
    },
    {
      "entropy": 5.755211305618286,
      "epoch": 2.503973860826563,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004372010623633753,
      "loss": 5.4399,
      "mean_token_accuracy": 0.19506631493568422,
      "num_tokens": 52867814.0,
      "step": 28355
    },
    {
      "entropy": 5.748987102508545,
      "epoch": 2.504415400918403,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043717923572044473,
      "loss": 5.3369,
      "mean_token_accuracy": 0.19535596817731857,
      "num_tokens": 52876958.0,
      "step": 28360
    },
    {
      "entropy": 5.691254186630249,
      "epoch": 2.5048569410102437,
      "grad_norm": 1.125,
      "learning_rate": 0.00043715740590051527,
      "loss": 5.3697,
      "mean_token_accuracy": 0.19980336725711823,
      "num_tokens": 52885978.0,
      "step": 28365
    },
    {
      "entropy": 5.728985977172852,
      "epoch": 2.505298481102084,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004371355729040144,
      "loss": 5.3622,
      "mean_token_accuracy": 0.19754606783390044,
      "num_tokens": 52894081.0,
      "step": 28370
    },
    {
      "entropy": 5.7884711742401125,
      "epoch": 2.5057400211939242,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004371137367313699,
      "loss": 5.4672,
      "mean_token_accuracy": 0.19743925780057908,
      "num_tokens": 52904012.0,
      "step": 28375
    },
    {
      "entropy": 5.818908071517944,
      "epoch": 2.5061815612857647,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043709189738300956,
      "loss": 5.4214,
      "mean_token_accuracy": 0.19316205382347107,
      "num_tokens": 52913319.0,
      "step": 28380
    },
    {
      "entropy": 5.726255416870117,
      "epoch": 2.5066231013776052,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004370700548593611,
      "loss": 5.4142,
      "mean_token_accuracy": 0.19974658340215684,
      "num_tokens": 52923539.0,
      "step": 28385
    },
    {
      "entropy": 5.703447914123535,
      "epoch": 2.5070646414694453,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004370482091608523,
      "loss": 5.366,
      "mean_token_accuracy": 0.2008258208632469,
      "num_tokens": 52931856.0,
      "step": 28390
    },
    {
      "entropy": 5.772503900527954,
      "epoch": 2.507506181561286,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043702636028791135,
      "loss": 5.4854,
      "mean_token_accuracy": 0.18657277077436446,
      "num_tokens": 52942270.0,
      "step": 28395
    },
    {
      "entropy": 5.744302654266358,
      "epoch": 2.5079477216531263,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004370045082409662,
      "loss": 5.3785,
      "mean_token_accuracy": 0.20486633479595184,
      "num_tokens": 52951503.0,
      "step": 28400
    },
    {
      "entropy": 5.764918279647827,
      "epoch": 2.5083892617449663,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004369826530204449,
      "loss": 5.4403,
      "mean_token_accuracy": 0.19728654623031616,
      "num_tokens": 52960288.0,
      "step": 28405
    },
    {
      "entropy": 5.764428186416626,
      "epoch": 2.508830801836807,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043696079462677537,
      "loss": 5.4739,
      "mean_token_accuracy": 0.19717622250318528,
      "num_tokens": 52969313.0,
      "step": 28410
    },
    {
      "entropy": 5.7625306129455565,
      "epoch": 2.5092723419286473,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004369389330603861,
      "loss": 5.4604,
      "mean_token_accuracy": 0.1906989261507988,
      "num_tokens": 52978924.0,
      "step": 28415
    },
    {
      "entropy": 5.799484014511108,
      "epoch": 2.5097138820204874,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043691706832170527,
      "loss": 5.4423,
      "mean_token_accuracy": 0.1957998663187027,
      "num_tokens": 52989027.0,
      "step": 28420
    },
    {
      "entropy": 5.852815961837768,
      "epoch": 2.510155422112328,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043689520041116116,
      "loss": 5.4689,
      "mean_token_accuracy": 0.19582595378160478,
      "num_tokens": 52999436.0,
      "step": 28425
    },
    {
      "entropy": 5.7015465259552,
      "epoch": 2.5105969622041684,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043687332932918204,
      "loss": 5.4644,
      "mean_token_accuracy": 0.19745858162641525,
      "num_tokens": 53008153.0,
      "step": 28430
    },
    {
      "entropy": 5.705590677261353,
      "epoch": 2.5110385022960084,
      "grad_norm": 1.125,
      "learning_rate": 0.0004368514550761965,
      "loss": 5.4462,
      "mean_token_accuracy": 0.19962492883205413,
      "num_tokens": 53017344.0,
      "step": 28435
    },
    {
      "entropy": 5.742638301849365,
      "epoch": 2.511480042387849,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043682957765263304,
      "loss": 5.3662,
      "mean_token_accuracy": 0.20635992735624314,
      "num_tokens": 53026966.0,
      "step": 28440
    },
    {
      "entropy": 5.766981506347657,
      "epoch": 2.511921582479689,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043680769705892013,
      "loss": 5.4131,
      "mean_token_accuracy": 0.1940276190638542,
      "num_tokens": 53035879.0,
      "step": 28445
    },
    {
      "entropy": 5.7688507556915285,
      "epoch": 2.5123631225715295,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004367858132954866,
      "loss": 5.4436,
      "mean_token_accuracy": 0.19538591653108597,
      "num_tokens": 53044727.0,
      "step": 28450
    },
    {
      "entropy": 5.688084697723388,
      "epoch": 2.51280466266337,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004367639263627609,
      "loss": 5.2507,
      "mean_token_accuracy": 0.2138822168111801,
      "num_tokens": 53054152.0,
      "step": 28455
    },
    {
      "entropy": 5.786692047119141,
      "epoch": 2.51324620275521,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004367420362611718,
      "loss": 5.4499,
      "mean_token_accuracy": 0.18903737366199494,
      "num_tokens": 53063259.0,
      "step": 28460
    },
    {
      "entropy": 5.827687215805054,
      "epoch": 2.5136877428470505,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004367201429911483,
      "loss": 5.4061,
      "mean_token_accuracy": 0.2000429794192314,
      "num_tokens": 53072139.0,
      "step": 28465
    },
    {
      "entropy": 5.824159336090088,
      "epoch": 2.5141292829388906,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004366982465531192,
      "loss": 5.4549,
      "mean_token_accuracy": 0.1923385038971901,
      "num_tokens": 53081833.0,
      "step": 28470
    },
    {
      "entropy": 5.668659305572509,
      "epoch": 2.514570823030731,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004366763469475134,
      "loss": 5.4087,
      "mean_token_accuracy": 0.20471657067537308,
      "num_tokens": 53091952.0,
      "step": 28475
    },
    {
      "entropy": 5.748886775970459,
      "epoch": 2.5150123631225716,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004366544441747599,
      "loss": 5.4498,
      "mean_token_accuracy": 0.20181348472833632,
      "num_tokens": 53101619.0,
      "step": 28480
    },
    {
      "entropy": 5.857583045959473,
      "epoch": 2.5154539032144116,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004366325382352878,
      "loss": 5.4211,
      "mean_token_accuracy": 0.19582653492689134,
      "num_tokens": 53110931.0,
      "step": 28485
    },
    {
      "entropy": 5.752022647857666,
      "epoch": 2.515895443306252,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043661062912952607,
      "loss": 5.424,
      "mean_token_accuracy": 0.2030706822872162,
      "num_tokens": 53119949.0,
      "step": 28490
    },
    {
      "entropy": 5.745273399353027,
      "epoch": 2.5163369833980926,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004365887168579042,
      "loss": 5.4278,
      "mean_token_accuracy": 0.20144165307283401,
      "num_tokens": 53129048.0,
      "step": 28495
    },
    {
      "entropy": 5.882178783416748,
      "epoch": 2.5167785234899327,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004365668014208512,
      "loss": 5.4848,
      "mean_token_accuracy": 0.19157612174749375,
      "num_tokens": 53139121.0,
      "step": 28500
    },
    {
      "entropy": 5.840753984451294,
      "epoch": 2.517220063581773,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043654488281879645,
      "loss": 5.3778,
      "mean_token_accuracy": 0.19854766875505447,
      "num_tokens": 53148098.0,
      "step": 28505
    },
    {
      "entropy": 5.728205776214599,
      "epoch": 2.5176616036736137,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004365229610521693,
      "loss": 5.4338,
      "mean_token_accuracy": 0.1930241584777832,
      "num_tokens": 53156992.0,
      "step": 28510
    },
    {
      "entropy": 5.7697656631469725,
      "epoch": 2.5181031437654537,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043650103612139917,
      "loss": 5.4413,
      "mean_token_accuracy": 0.1948065534234047,
      "num_tokens": 53166985.0,
      "step": 28515
    },
    {
      "entropy": 5.862615251541138,
      "epoch": 2.5185446838572942,
      "grad_norm": 1.125,
      "learning_rate": 0.0004364791080269156,
      "loss": 5.4562,
      "mean_token_accuracy": 0.19395162910223007,
      "num_tokens": 53176924.0,
      "step": 28520
    },
    {
      "entropy": 5.774032402038574,
      "epoch": 2.5189862239491347,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004364571767691482,
      "loss": 5.4571,
      "mean_token_accuracy": 0.19397223591804505,
      "num_tokens": 53186966.0,
      "step": 28525
    },
    {
      "entropy": 5.75743727684021,
      "epoch": 2.519427764040975,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043643524234852655,
      "loss": 5.4956,
      "mean_token_accuracy": 0.19541149139404296,
      "num_tokens": 53196147.0,
      "step": 28530
    },
    {
      "entropy": 5.776669979095459,
      "epoch": 2.5198693041328153,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043641330476548024,
      "loss": 5.3459,
      "mean_token_accuracy": 0.2071385383605957,
      "num_tokens": 53204446.0,
      "step": 28535
    },
    {
      "entropy": 5.791177415847779,
      "epoch": 2.520310844224656,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004363913640204392,
      "loss": 5.4668,
      "mean_token_accuracy": 0.18681320250034333,
      "num_tokens": 53213528.0,
      "step": 28540
    },
    {
      "entropy": 5.829217433929443,
      "epoch": 2.520752384316496,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000436369420113833,
      "loss": 5.4635,
      "mean_token_accuracy": 0.19685283601284026,
      "num_tokens": 53223574.0,
      "step": 28545
    },
    {
      "entropy": 5.689559364318848,
      "epoch": 2.5211939244083363,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043634747304609173,
      "loss": 5.3318,
      "mean_token_accuracy": 0.19779066145420074,
      "num_tokens": 53233011.0,
      "step": 28550
    },
    {
      "entropy": 5.799593496322632,
      "epoch": 2.521635464500177,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004363255228176452,
      "loss": 5.5053,
      "mean_token_accuracy": 0.19593330472707748,
      "num_tokens": 53241835.0,
      "step": 28555
    },
    {
      "entropy": 5.888135671615601,
      "epoch": 2.522077004592017,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004363035694289234,
      "loss": 5.4376,
      "mean_token_accuracy": 0.2012831375002861,
      "num_tokens": 53250666.0,
      "step": 28560
    },
    {
      "entropy": 5.7357419490814205,
      "epoch": 2.5225185446838574,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043628161288035644,
      "loss": 5.375,
      "mean_token_accuracy": 0.19981009215116502,
      "num_tokens": 53259386.0,
      "step": 28565
    },
    {
      "entropy": 5.740294790267944,
      "epoch": 2.522960084775698,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043625965317237435,
      "loss": 5.4467,
      "mean_token_accuracy": 0.1937156781554222,
      "num_tokens": 53268760.0,
      "step": 28570
    },
    {
      "entropy": 5.793749141693115,
      "epoch": 2.523401624867538,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00043623769030540747,
      "loss": 5.4356,
      "mean_token_accuracy": 0.1925848141312599,
      "num_tokens": 53277817.0,
      "step": 28575
    },
    {
      "entropy": 5.770207357406616,
      "epoch": 2.5238431649593784,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043621572427988585,
      "loss": 5.3355,
      "mean_token_accuracy": 0.2127211794257164,
      "num_tokens": 53286699.0,
      "step": 28580
    },
    {
      "entropy": 5.7656858444213865,
      "epoch": 2.5242847050512185,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004361937550962399,
      "loss": 5.3802,
      "mean_token_accuracy": 0.20198010206222533,
      "num_tokens": 53295431.0,
      "step": 28585
    },
    {
      "entropy": 5.716836595535279,
      "epoch": 2.524726245143059,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043617178275489997,
      "loss": 5.3779,
      "mean_token_accuracy": 0.19430874139070511,
      "num_tokens": 53304986.0,
      "step": 28590
    },
    {
      "entropy": 5.821677350997925,
      "epoch": 2.5251677852348995,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043614980725629663,
      "loss": 5.5298,
      "mean_token_accuracy": 0.1865036740899086,
      "num_tokens": 53313908.0,
      "step": 28595
    },
    {
      "entropy": 5.834622955322265,
      "epoch": 2.5256093253267395,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043612782860086,
      "loss": 5.3885,
      "mean_token_accuracy": 0.2008374124765396,
      "num_tokens": 53323553.0,
      "step": 28600
    },
    {
      "entropy": 5.843484497070312,
      "epoch": 2.52605086541858,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043610584678902096,
      "loss": 5.4792,
      "mean_token_accuracy": 0.1921474739909172,
      "num_tokens": 53333236.0,
      "step": 28605
    },
    {
      "entropy": 5.765245532989502,
      "epoch": 2.52649240551042,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043608386182120997,
      "loss": 5.4509,
      "mean_token_accuracy": 0.1958117738366127,
      "num_tokens": 53342344.0,
      "step": 28610
    },
    {
      "entropy": 5.759790229797363,
      "epoch": 2.5269339456022606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004360618736978577,
      "loss": 5.4466,
      "mean_token_accuracy": 0.20542844533920288,
      "num_tokens": 53351764.0,
      "step": 28615
    },
    {
      "entropy": 5.645538473129273,
      "epoch": 2.527375485694101,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000436039882419395,
      "loss": 5.2655,
      "mean_token_accuracy": 0.2072388768196106,
      "num_tokens": 53360667.0,
      "step": 28620
    },
    {
      "entropy": 5.806263542175293,
      "epoch": 2.527817025785941,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043601788798625253,
      "loss": 5.4495,
      "mean_token_accuracy": 0.1919879898428917,
      "num_tokens": 53370011.0,
      "step": 28625
    },
    {
      "entropy": 5.831104898452759,
      "epoch": 2.5282585658777816,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043599589039886125,
      "loss": 5.5392,
      "mean_token_accuracy": 0.19971051216125488,
      "num_tokens": 53379035.0,
      "step": 28630
    },
    {
      "entropy": 5.810754728317261,
      "epoch": 2.528700105969622,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000435973889657652,
      "loss": 5.4107,
      "mean_token_accuracy": 0.2035320833325386,
      "num_tokens": 53387739.0,
      "step": 28635
    },
    {
      "entropy": 5.8440595149993895,
      "epoch": 2.529141646061462,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043595188576305585,
      "loss": 5.5029,
      "mean_token_accuracy": 0.18726643770933152,
      "num_tokens": 53398219.0,
      "step": 28640
    },
    {
      "entropy": 5.793629932403564,
      "epoch": 2.5295831861533027,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043592987871550373,
      "loss": 5.4802,
      "mean_token_accuracy": 0.19207824170589446,
      "num_tokens": 53407337.0,
      "step": 28645
    },
    {
      "entropy": 5.886902904510498,
      "epoch": 2.530024726245143,
      "grad_norm": 1.453125,
      "learning_rate": 0.00043590786851542687,
      "loss": 5.4903,
      "mean_token_accuracy": 0.19625934362411498,
      "num_tokens": 53416223.0,
      "step": 28650
    },
    {
      "entropy": 5.739646863937378,
      "epoch": 2.5304662663369832,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043588585516325635,
      "loss": 5.3932,
      "mean_token_accuracy": 0.19775305986404418,
      "num_tokens": 53424719.0,
      "step": 28655
    },
    {
      "entropy": 5.837287282943725,
      "epoch": 2.5309078064288237,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043586383865942347,
      "loss": 5.3984,
      "mean_token_accuracy": 0.20004155933856965,
      "num_tokens": 53433763.0,
      "step": 28660
    },
    {
      "entropy": 5.815133953094483,
      "epoch": 2.5313493465206642,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043584181900435947,
      "loss": 5.4251,
      "mean_token_accuracy": 0.1956007182598114,
      "num_tokens": 53443270.0,
      "step": 28665
    },
    {
      "entropy": 5.775971937179565,
      "epoch": 2.5317908866125043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004358197961984957,
      "loss": 5.3996,
      "mean_token_accuracy": 0.1924089789390564,
      "num_tokens": 53451555.0,
      "step": 28670
    },
    {
      "entropy": 5.693482542037964,
      "epoch": 2.532232426704345,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004357977702422635,
      "loss": 5.4752,
      "mean_token_accuracy": 0.1908907935023308,
      "num_tokens": 53460845.0,
      "step": 28675
    },
    {
      "entropy": 5.76548957824707,
      "epoch": 2.5326739667961853,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043577574113609456,
      "loss": 5.3742,
      "mean_token_accuracy": 0.2076229065656662,
      "num_tokens": 53470169.0,
      "step": 28680
    },
    {
      "entropy": 5.778090524673462,
      "epoch": 2.5331155068880253,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004357537088804203,
      "loss": 5.4185,
      "mean_token_accuracy": 0.19986058473587037,
      "num_tokens": 53478484.0,
      "step": 28685
    },
    {
      "entropy": 5.725354719161987,
      "epoch": 2.533557046979866,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004357316734756722,
      "loss": 5.4024,
      "mean_token_accuracy": 0.19645469039678573,
      "num_tokens": 53488791.0,
      "step": 28690
    },
    {
      "entropy": 5.789385557174683,
      "epoch": 2.5339985870717063,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004357096349222821,
      "loss": 5.3764,
      "mean_token_accuracy": 0.2057168573141098,
      "num_tokens": 53498310.0,
      "step": 28695
    },
    {
      "entropy": 5.769964790344238,
      "epoch": 2.5344401271635464,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043568759322068165,
      "loss": 5.4379,
      "mean_token_accuracy": 0.2003674477338791,
      "num_tokens": 53508145.0,
      "step": 28700
    },
    {
      "entropy": 5.770507764816284,
      "epoch": 2.534881667255387,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004356655483713026,
      "loss": 5.4697,
      "mean_token_accuracy": 0.19780299067497253,
      "num_tokens": 53517473.0,
      "step": 28705
    },
    {
      "entropy": 5.818856859207154,
      "epoch": 2.5353232073472274,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043564350037457683,
      "loss": 5.4667,
      "mean_token_accuracy": 0.1959809824824333,
      "num_tokens": 53526625.0,
      "step": 28710
    },
    {
      "entropy": 5.829560089111328,
      "epoch": 2.5357647474390674,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043562144923093627,
      "loss": 5.3342,
      "mean_token_accuracy": 0.20327746570110322,
      "num_tokens": 53535480.0,
      "step": 28715
    },
    {
      "entropy": 5.7290387630462645,
      "epoch": 2.536206287530908,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043559939494081286,
      "loss": 5.4819,
      "mean_token_accuracy": 0.19229964464902877,
      "num_tokens": 53544368.0,
      "step": 28720
    },
    {
      "entropy": 5.722231674194336,
      "epoch": 2.536647827622748,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004355773375046387,
      "loss": 5.4307,
      "mean_token_accuracy": 0.20221597999334334,
      "num_tokens": 53553235.0,
      "step": 28725
    },
    {
      "entropy": 5.820092678070068,
      "epoch": 2.5370893677145885,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004355552769228457,
      "loss": 5.4639,
      "mean_token_accuracy": 0.18702964931726457,
      "num_tokens": 53562635.0,
      "step": 28730
    },
    {
      "entropy": 5.8562620162963865,
      "epoch": 2.537530907806429,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004355332131958663,
      "loss": 5.3878,
      "mean_token_accuracy": 0.19481416493654252,
      "num_tokens": 53571394.0,
      "step": 28735
    },
    {
      "entropy": 5.838503217697143,
      "epoch": 2.537972447898269,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004355111463241324,
      "loss": 5.4909,
      "mean_token_accuracy": 0.1905012235045433,
      "num_tokens": 53580718.0,
      "step": 28740
    },
    {
      "entropy": 5.764612340927124,
      "epoch": 2.5384139879901095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004354890763080765,
      "loss": 5.5094,
      "mean_token_accuracy": 0.19739625751972198,
      "num_tokens": 53590120.0,
      "step": 28745
    },
    {
      "entropy": 5.760209465026856,
      "epoch": 2.5388555280819496,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004354670031481309,
      "loss": 5.454,
      "mean_token_accuracy": 0.19419420659542083,
      "num_tokens": 53599783.0,
      "step": 28750
    },
    {
      "entropy": 5.775852537155151,
      "epoch": 2.53929706817379,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004354449268447279,
      "loss": 5.4249,
      "mean_token_accuracy": 0.19272384643554688,
      "num_tokens": 53608663.0,
      "step": 28755
    },
    {
      "entropy": 5.807463884353638,
      "epoch": 2.5397386082656306,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043542284739830006,
      "loss": 5.5145,
      "mean_token_accuracy": 0.19426106065511703,
      "num_tokens": 53618239.0,
      "step": 28760
    },
    {
      "entropy": 5.869434881210327,
      "epoch": 2.5401801483574706,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043540076480927994,
      "loss": 5.4602,
      "mean_token_accuracy": 0.20004933327436447,
      "num_tokens": 53627683.0,
      "step": 28765
    },
    {
      "entropy": 5.740294885635376,
      "epoch": 2.540621688449311,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043537867907809995,
      "loss": 5.3397,
      "mean_token_accuracy": 0.20390724688768386,
      "num_tokens": 53636980.0,
      "step": 28770
    },
    {
      "entropy": 5.7058045864105225,
      "epoch": 2.5410632285411516,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004353565902051929,
      "loss": 5.4243,
      "mean_token_accuracy": 0.18868485689163209,
      "num_tokens": 53647310.0,
      "step": 28775
    },
    {
      "entropy": 5.798305082321167,
      "epoch": 2.5415047686329917,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00043533449819099145,
      "loss": 5.5034,
      "mean_token_accuracy": 0.1972437471151352,
      "num_tokens": 53657433.0,
      "step": 28780
    },
    {
      "entropy": 5.784423971176148,
      "epoch": 2.541946308724832,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004353124030359284,
      "loss": 5.4121,
      "mean_token_accuracy": 0.1977531597018242,
      "num_tokens": 53666969.0,
      "step": 28785
    },
    {
      "entropy": 5.812146854400635,
      "epoch": 2.5423878488166727,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043529030474043657,
      "loss": 5.4372,
      "mean_token_accuracy": 0.19523836821317672,
      "num_tokens": 53676332.0,
      "step": 28790
    },
    {
      "entropy": 5.75386233329773,
      "epoch": 2.5428293889085127,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043526820330494877,
      "loss": 5.4802,
      "mean_token_accuracy": 0.1927498236298561,
      "num_tokens": 53685766.0,
      "step": 28795
    },
    {
      "entropy": 5.830302476882935,
      "epoch": 2.5432709290003532,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004352460987298981,
      "loss": 5.4241,
      "mean_token_accuracy": 0.19463727474212647,
      "num_tokens": 53694744.0,
      "step": 28800
    },
    {
      "entropy": 5.704309558868408,
      "epoch": 2.5437124690921937,
      "grad_norm": 1.125,
      "learning_rate": 0.0004352239910157174,
      "loss": 5.3028,
      "mean_token_accuracy": 0.21227456331253053,
      "num_tokens": 53704266.0,
      "step": 28805
    },
    {
      "entropy": 5.689721632003784,
      "epoch": 2.544154009184034,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00043520188016283977,
      "loss": 5.3636,
      "mean_token_accuracy": 0.19870593398809433,
      "num_tokens": 53712070.0,
      "step": 28810
    },
    {
      "entropy": 5.74771032333374,
      "epoch": 2.5445955492758743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043517976617169857,
      "loss": 5.4418,
      "mean_token_accuracy": 0.19082449823617936,
      "num_tokens": 53720152.0,
      "step": 28815
    },
    {
      "entropy": 5.689497709274292,
      "epoch": 2.5450370893677148,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004351576490427268,
      "loss": 5.2725,
      "mean_token_accuracy": 0.2103740781545639,
      "num_tokens": 53729451.0,
      "step": 28820
    },
    {
      "entropy": 5.721990919113159,
      "epoch": 2.545478629459555,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004351355287763578,
      "loss": 5.4398,
      "mean_token_accuracy": 0.19491227716207504,
      "num_tokens": 53738866.0,
      "step": 28825
    },
    {
      "entropy": 5.7315572738647464,
      "epoch": 2.5459201695513953,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043511340537302485,
      "loss": 5.2833,
      "mean_token_accuracy": 0.20673781037330627,
      "num_tokens": 53748006.0,
      "step": 28830
    },
    {
      "entropy": 5.80237135887146,
      "epoch": 2.546361709643236,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004350912788331613,
      "loss": 5.3369,
      "mean_token_accuracy": 0.21034676879644393,
      "num_tokens": 53756412.0,
      "step": 28835
    },
    {
      "entropy": 5.744651126861572,
      "epoch": 2.546803249735076,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004350691491572007,
      "loss": 5.4661,
      "mean_token_accuracy": 0.1979224681854248,
      "num_tokens": 53765509.0,
      "step": 28840
    },
    {
      "entropy": 5.7155601501464846,
      "epoch": 2.5472447898269164,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004350470163455764,
      "loss": 5.4571,
      "mean_token_accuracy": 0.19682829678058625,
      "num_tokens": 53774883.0,
      "step": 28845
    },
    {
      "entropy": 5.825737380981446,
      "epoch": 2.547686329918757,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043502488039872224,
      "loss": 5.4346,
      "mean_token_accuracy": 0.18774766027927398,
      "num_tokens": 53783650.0,
      "step": 28850
    },
    {
      "entropy": 5.880618000030518,
      "epoch": 2.548127870010597,
      "grad_norm": 1.125,
      "learning_rate": 0.00043500274131707156,
      "loss": 5.5112,
      "mean_token_accuracy": 0.18918595761060714,
      "num_tokens": 53793031.0,
      "step": 28855
    },
    {
      "entropy": 5.87454137802124,
      "epoch": 2.5485694101024374,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004349805991010582,
      "loss": 5.5156,
      "mean_token_accuracy": 0.1809660866856575,
      "num_tokens": 53803967.0,
      "step": 28860
    },
    {
      "entropy": 5.789888525009156,
      "epoch": 2.5490109501942775,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043495845375111594,
      "loss": 5.4384,
      "mean_token_accuracy": 0.19400778859853746,
      "num_tokens": 53813575.0,
      "step": 28865
    },
    {
      "entropy": 5.768313884735107,
      "epoch": 2.549452490286118,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043493630526767846,
      "loss": 5.4072,
      "mean_token_accuracy": 0.1989683762192726,
      "num_tokens": 53823250.0,
      "step": 28870
    },
    {
      "entropy": 5.785819149017334,
      "epoch": 2.5498940303779585,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00043491415365117984,
      "loss": 5.3957,
      "mean_token_accuracy": 0.20140734016895295,
      "num_tokens": 53832300.0,
      "step": 28875
    },
    {
      "entropy": 5.757631254196167,
      "epoch": 2.5503355704697985,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043489199890205376,
      "loss": 5.3588,
      "mean_token_accuracy": 0.2000524416565895,
      "num_tokens": 53841385.0,
      "step": 28880
    },
    {
      "entropy": 5.808206272125244,
      "epoch": 2.550777110561639,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043486984102073445,
      "loss": 5.47,
      "mean_token_accuracy": 0.1866510331630707,
      "num_tokens": 53850349.0,
      "step": 28885
    },
    {
      "entropy": 5.712516260147095,
      "epoch": 2.551218650653479,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043484768000765585,
      "loss": 5.3115,
      "mean_token_accuracy": 0.20711386054754258,
      "num_tokens": 53859547.0,
      "step": 28890
    },
    {
      "entropy": 5.7390281677246096,
      "epoch": 2.5516601907453196,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004348255158632521,
      "loss": 5.4307,
      "mean_token_accuracy": 0.20011014193296434,
      "num_tokens": 53868497.0,
      "step": 28895
    },
    {
      "entropy": 5.707411623001098,
      "epoch": 2.55210173083716,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043480334858795737,
      "loss": 5.3634,
      "mean_token_accuracy": 0.19865658283233642,
      "num_tokens": 53877200.0,
      "step": 28900
    },
    {
      "entropy": 5.738184070587158,
      "epoch": 2.552543270929,
      "grad_norm": 1.109375,
      "learning_rate": 0.000434781178182206,
      "loss": 5.3176,
      "mean_token_accuracy": 0.20996322184801103,
      "num_tokens": 53887081.0,
      "step": 28905
    },
    {
      "entropy": 5.739811420440674,
      "epoch": 2.5529848110208406,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004347590046464322,
      "loss": 5.3783,
      "mean_token_accuracy": 0.19938085973262787,
      "num_tokens": 53896334.0,
      "step": 28910
    },
    {
      "entropy": 5.735810089111328,
      "epoch": 2.553426351112681,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043473682798107034,
      "loss": 5.4185,
      "mean_token_accuracy": 0.19817714989185334,
      "num_tokens": 53905658.0,
      "step": 28915
    },
    {
      "entropy": 5.836450433731079,
      "epoch": 2.553867891204521,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004347146481865549,
      "loss": 5.5208,
      "mean_token_accuracy": 0.19437649846076965,
      "num_tokens": 53915747.0,
      "step": 28920
    },
    {
      "entropy": 5.872328662872315,
      "epoch": 2.5543094312963617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004346924652633203,
      "loss": 5.4535,
      "mean_token_accuracy": 0.1987748146057129,
      "num_tokens": 53924636.0,
      "step": 28925
    },
    {
      "entropy": 5.7363121032714846,
      "epoch": 2.554750971388202,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004346702792118011,
      "loss": 5.368,
      "mean_token_accuracy": 0.20651305466890335,
      "num_tokens": 53934104.0,
      "step": 28930
    },
    {
      "entropy": 5.772401905059814,
      "epoch": 2.5551925114800422,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043464809003243197,
      "loss": 5.4448,
      "mean_token_accuracy": 0.20323298275470733,
      "num_tokens": 53943055.0,
      "step": 28935
    },
    {
      "entropy": 5.7606922626495365,
      "epoch": 2.5556340515718827,
      "grad_norm": 1.125,
      "learning_rate": 0.0004346258977256475,
      "loss": 5.4163,
      "mean_token_accuracy": 0.20295822471380234,
      "num_tokens": 53952489.0,
      "step": 28940
    },
    {
      "entropy": 5.8200689315795895,
      "epoch": 2.5560755916637232,
      "grad_norm": 1.125,
      "learning_rate": 0.00043460370229188257,
      "loss": 5.3658,
      "mean_token_accuracy": 0.19540989696979522,
      "num_tokens": 53962274.0,
      "step": 28945
    },
    {
      "entropy": 5.815763807296753,
      "epoch": 2.5565171317555633,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043458150373157183,
      "loss": 5.4557,
      "mean_token_accuracy": 0.19341008514165878,
      "num_tokens": 53971152.0,
      "step": 28950
    },
    {
      "entropy": 5.785996437072754,
      "epoch": 2.5569586718474038,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004345593020451503,
      "loss": 5.4103,
      "mean_token_accuracy": 0.19476097971200942,
      "num_tokens": 53979510.0,
      "step": 28955
    },
    {
      "entropy": 5.646920013427734,
      "epoch": 2.5574002119392443,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043453709723305264,
      "loss": 5.3765,
      "mean_token_accuracy": 0.20258827209472657,
      "num_tokens": 53987917.0,
      "step": 28960
    },
    {
      "entropy": 5.760303783416748,
      "epoch": 2.5578417520310843,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043451488929571405,
      "loss": 5.4604,
      "mean_token_accuracy": 0.19450544863939284,
      "num_tokens": 53997924.0,
      "step": 28965
    },
    {
      "entropy": 5.769318389892578,
      "epoch": 2.558283292122925,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004344926782335695,
      "loss": 5.4185,
      "mean_token_accuracy": 0.19725882709026338,
      "num_tokens": 54006505.0,
      "step": 28970
    },
    {
      "entropy": 5.8751014232635494,
      "epoch": 2.5587248322147653,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043447046404705404,
      "loss": 5.5228,
      "mean_token_accuracy": 0.18943820893764496,
      "num_tokens": 54016794.0,
      "step": 28975
    },
    {
      "entropy": 5.7381346225738525,
      "epoch": 2.5591663723066054,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004344482467366029,
      "loss": 5.3277,
      "mean_token_accuracy": 0.2091084748506546,
      "num_tokens": 54025995.0,
      "step": 28980
    },
    {
      "entropy": 5.7646091938018795,
      "epoch": 2.559607912398446,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004344260263026513,
      "loss": 5.3897,
      "mean_token_accuracy": 0.2003390833735466,
      "num_tokens": 54035501.0,
      "step": 28985
    },
    {
      "entropy": 5.733925867080688,
      "epoch": 2.5600494524902864,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004344038027456346,
      "loss": 5.4869,
      "mean_token_accuracy": 0.19293991476297379,
      "num_tokens": 54045283.0,
      "step": 28990
    },
    {
      "entropy": 5.7945905208587645,
      "epoch": 2.5604909925821264,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004343815760659879,
      "loss": 5.5003,
      "mean_token_accuracy": 0.19780979603528975,
      "num_tokens": 54056250.0,
      "step": 28995
    },
    {
      "entropy": 5.8140233039855955,
      "epoch": 2.560932532673967,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004343593462641469,
      "loss": 5.3802,
      "mean_token_accuracy": 0.19925764948129654,
      "num_tokens": 54065225.0,
      "step": 29000
    },
    {
      "entropy": 5.783453130722046,
      "epoch": 2.561374072765807,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004343371133405469,
      "loss": 5.5091,
      "mean_token_accuracy": 0.19582258462905883,
      "num_tokens": 54074056.0,
      "step": 29005
    },
    {
      "entropy": 5.797067260742187,
      "epoch": 2.5618156128576475,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043431487729562345,
      "loss": 5.3722,
      "mean_token_accuracy": 0.20344914942979814,
      "num_tokens": 54083202.0,
      "step": 29010
    },
    {
      "entropy": 5.731200265884399,
      "epoch": 2.562257152949488,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004342926381298122,
      "loss": 5.2988,
      "mean_token_accuracy": 0.21091606467962265,
      "num_tokens": 54092088.0,
      "step": 29015
    },
    {
      "entropy": 5.757012414932251,
      "epoch": 2.562698693041328,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004342703958435487,
      "loss": 5.5093,
      "mean_token_accuracy": 0.18892745226621627,
      "num_tokens": 54101113.0,
      "step": 29020
    },
    {
      "entropy": 5.730999565124511,
      "epoch": 2.5631402331331685,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004342481504372688,
      "loss": 5.4309,
      "mean_token_accuracy": 0.20664407312870026,
      "num_tokens": 54109832.0,
      "step": 29025
    },
    {
      "entropy": 5.774690723419189,
      "epoch": 2.5635817732250086,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043422590191140813,
      "loss": 5.3372,
      "mean_token_accuracy": 0.2030613586306572,
      "num_tokens": 54117783.0,
      "step": 29030
    },
    {
      "entropy": 5.849211597442627,
      "epoch": 2.564023313316849,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004342036502664026,
      "loss": 5.4944,
      "mean_token_accuracy": 0.1856474310159683,
      "num_tokens": 54126539.0,
      "step": 29035
    },
    {
      "entropy": 5.758130168914795,
      "epoch": 2.5644648534086896,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004341813955026882,
      "loss": 5.3314,
      "mean_token_accuracy": 0.21000847816467286,
      "num_tokens": 54135316.0,
      "step": 29040
    },
    {
      "entropy": 5.733311700820923,
      "epoch": 2.5649063935005296,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043415913762070074,
      "loss": 5.3642,
      "mean_token_accuracy": 0.2052118331193924,
      "num_tokens": 54144164.0,
      "step": 29045
    },
    {
      "entropy": 5.764106321334839,
      "epoch": 2.56534793359237,
      "grad_norm": 1.125,
      "learning_rate": 0.0004341368766208763,
      "loss": 5.4424,
      "mean_token_accuracy": 0.19778802394866943,
      "num_tokens": 54154030.0,
      "step": 29050
    },
    {
      "entropy": 5.666384601593018,
      "epoch": 2.5657894736842106,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004341146125036509,
      "loss": 5.375,
      "mean_token_accuracy": 0.21361307054758072,
      "num_tokens": 54163507.0,
      "step": 29055
    },
    {
      "entropy": 5.698010587692261,
      "epoch": 2.5662310137760507,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043409234526946077,
      "loss": 5.3403,
      "mean_token_accuracy": 0.20406097173690796,
      "num_tokens": 54173016.0,
      "step": 29060
    },
    {
      "entropy": 5.848013687133789,
      "epoch": 2.566672553867891,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043407007491874216,
      "loss": 5.4656,
      "mean_token_accuracy": 0.19684356898069383,
      "num_tokens": 54181965.0,
      "step": 29065
    },
    {
      "entropy": 5.732329082489014,
      "epoch": 2.5671140939597317,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043404780145193115,
      "loss": 5.3967,
      "mean_token_accuracy": 0.20090592205524443,
      "num_tokens": 54190574.0,
      "step": 29070
    },
    {
      "entropy": 5.8309248924255375,
      "epoch": 2.5675556340515717,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043402552486946424,
      "loss": 5.4843,
      "mean_token_accuracy": 0.19718263000249864,
      "num_tokens": 54200105.0,
      "step": 29075
    },
    {
      "entropy": 5.8412824153900145,
      "epoch": 2.567997174143412,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043400324517177765,
      "loss": 5.4401,
      "mean_token_accuracy": 0.19517897963523864,
      "num_tokens": 54209247.0,
      "step": 29080
    },
    {
      "entropy": 5.7865907669067385,
      "epoch": 2.5684387142352527,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043398096235930795,
      "loss": 5.3782,
      "mean_token_accuracy": 0.19714784473180771,
      "num_tokens": 54218122.0,
      "step": 29085
    },
    {
      "entropy": 5.6919773578643795,
      "epoch": 2.5688802543270928,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043395867643249164,
      "loss": 5.2927,
      "mean_token_accuracy": 0.21359048485755922,
      "num_tokens": 54226706.0,
      "step": 29090
    },
    {
      "entropy": 5.741680049896241,
      "epoch": 2.5693217944189333,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004339363873917652,
      "loss": 5.406,
      "mean_token_accuracy": 0.19813846796751022,
      "num_tokens": 54235927.0,
      "step": 29095
    },
    {
      "entropy": 5.758253335952759,
      "epoch": 2.5697633345107738,
      "grad_norm": 1.125,
      "learning_rate": 0.00043391409523756537,
      "loss": 5.4979,
      "mean_token_accuracy": 0.18753288239240645,
      "num_tokens": 54245939.0,
      "step": 29100
    },
    {
      "entropy": 5.769730281829834,
      "epoch": 2.570204874602614,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043389179997032885,
      "loss": 5.4768,
      "mean_token_accuracy": 0.1926875039935112,
      "num_tokens": 54255726.0,
      "step": 29105
    },
    {
      "entropy": 5.758779096603393,
      "epoch": 2.5706464146944543,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004338695015904922,
      "loss": 5.3729,
      "mean_token_accuracy": 0.20391585528850556,
      "num_tokens": 54264485.0,
      "step": 29110
    },
    {
      "entropy": 5.735147619247437,
      "epoch": 2.571087954786295,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004338472000984924,
      "loss": 5.3707,
      "mean_token_accuracy": 0.20225958079099654,
      "num_tokens": 54272717.0,
      "step": 29115
    },
    {
      "entropy": 5.727560567855835,
      "epoch": 2.571529494878135,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043382489549476636,
      "loss": 5.3201,
      "mean_token_accuracy": 0.20892775654792786,
      "num_tokens": 54281806.0,
      "step": 29120
    },
    {
      "entropy": 5.7599231719970705,
      "epoch": 2.5719710349699754,
      "grad_norm": 1.4375,
      "learning_rate": 0.000433802587779751,
      "loss": 5.4614,
      "mean_token_accuracy": 0.19367914646863937,
      "num_tokens": 54291147.0,
      "step": 29125
    },
    {
      "entropy": 5.82211742401123,
      "epoch": 2.572412575061816,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043378027695388315,
      "loss": 5.4529,
      "mean_token_accuracy": 0.1985684111714363,
      "num_tokens": 54299696.0,
      "step": 29130
    },
    {
      "entropy": 5.809426212310791,
      "epoch": 2.572854115153656,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043375796301760007,
      "loss": 5.4067,
      "mean_token_accuracy": 0.20589210093021393,
      "num_tokens": 54309320.0,
      "step": 29135
    },
    {
      "entropy": 5.809529256820679,
      "epoch": 2.5732956552454964,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004337356459713387,
      "loss": 5.5011,
      "mean_token_accuracy": 0.19486991316080093,
      "num_tokens": 54318794.0,
      "step": 29140
    },
    {
      "entropy": 5.774178838729858,
      "epoch": 2.5737371953373365,
      "grad_norm": 1.125,
      "learning_rate": 0.0004337133258155363,
      "loss": 5.4217,
      "mean_token_accuracy": 0.19729449450969697,
      "num_tokens": 54327792.0,
      "step": 29145
    },
    {
      "entropy": 5.805655479431152,
      "epoch": 2.574178735429177,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043369100255063016,
      "loss": 5.442,
      "mean_token_accuracy": 0.19064365476369857,
      "num_tokens": 54337524.0,
      "step": 29150
    },
    {
      "entropy": 5.713745021820069,
      "epoch": 2.5746202755210175,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004336686761770575,
      "loss": 5.3244,
      "mean_token_accuracy": 0.21334663927555084,
      "num_tokens": 54347409.0,
      "step": 29155
    },
    {
      "entropy": 5.676456928253174,
      "epoch": 2.5750618156128575,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043364634669525577,
      "loss": 5.3429,
      "mean_token_accuracy": 0.20273261219263078,
      "num_tokens": 54356744.0,
      "step": 29160
    },
    {
      "entropy": 5.776691579818726,
      "epoch": 2.575503355704698,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043362401410566224,
      "loss": 5.4217,
      "mean_token_accuracy": 0.19282619059085845,
      "num_tokens": 54365274.0,
      "step": 29165
    },
    {
      "entropy": 5.716569995880127,
      "epoch": 2.575944895796538,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043360167840871455,
      "loss": 5.3622,
      "mean_token_accuracy": 0.20331128239631652,
      "num_tokens": 54374448.0,
      "step": 29170
    },
    {
      "entropy": 5.6801964282989506,
      "epoch": 2.5763864358883786,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043357933960485017,
      "loss": 5.3994,
      "mean_token_accuracy": 0.19360854476690292,
      "num_tokens": 54383599.0,
      "step": 29175
    },
    {
      "entropy": 5.668877124786377,
      "epoch": 2.576827975980219,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043355699769450674,
      "loss": 5.2958,
      "mean_token_accuracy": 0.2115686282515526,
      "num_tokens": 54392294.0,
      "step": 29180
    },
    {
      "entropy": 5.791758251190186,
      "epoch": 2.577269516072059,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043353465267812185,
      "loss": 5.411,
      "mean_token_accuracy": 0.20109790712594985,
      "num_tokens": 54401854.0,
      "step": 29185
    },
    {
      "entropy": 5.825889015197754,
      "epoch": 2.5777110561638996,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004335123045561333,
      "loss": 5.5208,
      "mean_token_accuracy": 0.18984647542238237,
      "num_tokens": 54411255.0,
      "step": 29190
    },
    {
      "entropy": 5.830274057388306,
      "epoch": 2.57815259625574,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043348995332897887,
      "loss": 5.4456,
      "mean_token_accuracy": 0.18819218277931213,
      "num_tokens": 54420437.0,
      "step": 29195
    },
    {
      "entropy": 5.6708907127380375,
      "epoch": 2.57859413634758,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043346759899709643,
      "loss": 5.2744,
      "mean_token_accuracy": 0.21146924197673797,
      "num_tokens": 54429166.0,
      "step": 29200
    },
    {
      "entropy": 5.75497465133667,
      "epoch": 2.5790356764394207,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004334452415609238,
      "loss": 5.3642,
      "mean_token_accuracy": 0.19620211273431779,
      "num_tokens": 54438060.0,
      "step": 29205
    },
    {
      "entropy": 5.684785175323486,
      "epoch": 2.579477216531261,
      "grad_norm": 1.171875,
      "learning_rate": 0.000433422881020899,
      "loss": 5.3457,
      "mean_token_accuracy": 0.2079796865582466,
      "num_tokens": 54447503.0,
      "step": 29210
    },
    {
      "entropy": 5.7222391128540036,
      "epoch": 2.579918756623101,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043340051737746,
      "loss": 5.3714,
      "mean_token_accuracy": 0.20409584045410156,
      "num_tokens": 54456741.0,
      "step": 29215
    },
    {
      "entropy": 5.764773797988892,
      "epoch": 2.5803602967149417,
      "grad_norm": 1.171875,
      "learning_rate": 0.000433378150631045,
      "loss": 5.4379,
      "mean_token_accuracy": 0.19358309507369995,
      "num_tokens": 54466614.0,
      "step": 29220
    },
    {
      "entropy": 5.82527494430542,
      "epoch": 2.580801836806782,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004333557807820922,
      "loss": 5.4048,
      "mean_token_accuracy": 0.20082637816667556,
      "num_tokens": 54475191.0,
      "step": 29225
    },
    {
      "entropy": 5.700796031951905,
      "epoch": 2.5812433768986223,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004333334078310396,
      "loss": 5.353,
      "mean_token_accuracy": 0.21042647808790207,
      "num_tokens": 54483924.0,
      "step": 29230
    },
    {
      "entropy": 5.6190284252166744,
      "epoch": 2.5816849169904628,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043331103177832566,
      "loss": 5.3552,
      "mean_token_accuracy": 0.2001377284526825,
      "num_tokens": 54493739.0,
      "step": 29235
    },
    {
      "entropy": 5.7288634300231935,
      "epoch": 2.5821264570823033,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004332886526243886,
      "loss": 5.3851,
      "mean_token_accuracy": 0.19583563655614852,
      "num_tokens": 54503760.0,
      "step": 29240
    },
    {
      "entropy": 5.835544395446777,
      "epoch": 2.5825679971741433,
      "grad_norm": 1.125,
      "learning_rate": 0.0004332662703696669,
      "loss": 5.3752,
      "mean_token_accuracy": 0.20669446140527725,
      "num_tokens": 54512884.0,
      "step": 29245
    },
    {
      "entropy": 5.807183933258057,
      "epoch": 2.583009537265984,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004332438850145989,
      "loss": 5.4385,
      "mean_token_accuracy": 0.19752825498580934,
      "num_tokens": 54522747.0,
      "step": 29250
    },
    {
      "entropy": 5.734337377548218,
      "epoch": 2.5834510773578243,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004332214965596232,
      "loss": 5.33,
      "mean_token_accuracy": 0.20519591122865677,
      "num_tokens": 54532109.0,
      "step": 29255
    },
    {
      "entropy": 5.689410209655762,
      "epoch": 2.5838926174496644,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004331991050051785,
      "loss": 5.3835,
      "mean_token_accuracy": 0.2049789696931839,
      "num_tokens": 54541160.0,
      "step": 29260
    },
    {
      "entropy": 5.746656894683838,
      "epoch": 2.584334157541505,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043317671035170315,
      "loss": 5.3851,
      "mean_token_accuracy": 0.20824730396270752,
      "num_tokens": 54550162.0,
      "step": 29265
    },
    {
      "entropy": 5.681261873245239,
      "epoch": 2.5847756976333454,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043315431259963606,
      "loss": 5.2505,
      "mean_token_accuracy": 0.20989584177732468,
      "num_tokens": 54559355.0,
      "step": 29270
    },
    {
      "entropy": 5.716899776458741,
      "epoch": 2.5852172377251854,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043313191174941594,
      "loss": 5.3298,
      "mean_token_accuracy": 0.20784920454025269,
      "num_tokens": 54568628.0,
      "step": 29275
    },
    {
      "entropy": 5.708792400360108,
      "epoch": 2.585658777817026,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004331095078014816,
      "loss": 5.3909,
      "mean_token_accuracy": 0.20116203129291535,
      "num_tokens": 54578744.0,
      "step": 29280
    },
    {
      "entropy": 5.811019563674927,
      "epoch": 2.586100317908866,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043308710075627195,
      "loss": 5.4269,
      "mean_token_accuracy": 0.19751086682081223,
      "num_tokens": 54588654.0,
      "step": 29285
    },
    {
      "entropy": 5.80687575340271,
      "epoch": 2.5865418580007065,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004330646906142259,
      "loss": 5.3679,
      "mean_token_accuracy": 0.20043631345033647,
      "num_tokens": 54597085.0,
      "step": 29290
    },
    {
      "entropy": 5.733868837356567,
      "epoch": 2.586983398092547,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004330422773757824,
      "loss": 5.4754,
      "mean_token_accuracy": 0.19520022124052047,
      "num_tokens": 54606151.0,
      "step": 29295
    },
    {
      "entropy": 5.707451057434082,
      "epoch": 2.587424938184387,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004330198610413806,
      "loss": 5.3367,
      "mean_token_accuracy": 0.2096156433224678,
      "num_tokens": 54615085.0,
      "step": 29300
    },
    {
      "entropy": 5.766754388809204,
      "epoch": 2.5878664782762275,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004329974416114596,
      "loss": 5.3334,
      "mean_token_accuracy": 0.20052669942378998,
      "num_tokens": 54623895.0,
      "step": 29305
    },
    {
      "entropy": 5.726675939559937,
      "epoch": 2.5883080183680676,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004329750190864586,
      "loss": 5.3032,
      "mean_token_accuracy": 0.2080240622162819,
      "num_tokens": 54633096.0,
      "step": 29310
    },
    {
      "entropy": 5.766446876525879,
      "epoch": 2.588749558459908,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043295259346681677,
      "loss": 5.489,
      "mean_token_accuracy": 0.18700897991657256,
      "num_tokens": 54642210.0,
      "step": 29315
    },
    {
      "entropy": 5.831173038482666,
      "epoch": 2.5891910985517486,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004329301647529734,
      "loss": 5.5543,
      "mean_token_accuracy": 0.18292182683944702,
      "num_tokens": 54651960.0,
      "step": 29320
    },
    {
      "entropy": 5.843255043029785,
      "epoch": 2.5896326386435886,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043290773294536814,
      "loss": 5.4016,
      "mean_token_accuracy": 0.20197472721338272,
      "num_tokens": 54661560.0,
      "step": 29325
    },
    {
      "entropy": 5.817828941345215,
      "epoch": 2.590074178735429,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043288529804443997,
      "loss": 5.4265,
      "mean_token_accuracy": 0.19668363630771638,
      "num_tokens": 54671302.0,
      "step": 29330
    },
    {
      "entropy": 5.710333156585693,
      "epoch": 2.5905157188272696,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043286286005062875,
      "loss": 5.4438,
      "mean_token_accuracy": 0.19679137021303178,
      "num_tokens": 54681641.0,
      "step": 29335
    },
    {
      "entropy": 5.727007818222046,
      "epoch": 2.5909572589191097,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043284041896437386,
      "loss": 5.398,
      "mean_token_accuracy": 0.19494490176439286,
      "num_tokens": 54690190.0,
      "step": 29340
    },
    {
      "entropy": 5.829875802993774,
      "epoch": 2.59139879901095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004328179747861149,
      "loss": 5.4664,
      "mean_token_accuracy": 0.19485388100147247,
      "num_tokens": 54699164.0,
      "step": 29345
    },
    {
      "entropy": 5.730369663238525,
      "epoch": 2.5918403391027907,
      "grad_norm": 1.125,
      "learning_rate": 0.00043279552751629153,
      "loss": 5.288,
      "mean_token_accuracy": 0.2118021622300148,
      "num_tokens": 54708509.0,
      "step": 29350
    },
    {
      "entropy": 5.740078687667847,
      "epoch": 2.5922818791946307,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043277307715534365,
      "loss": 5.447,
      "mean_token_accuracy": 0.19706758856773376,
      "num_tokens": 54717382.0,
      "step": 29355
    },
    {
      "entropy": 5.68647985458374,
      "epoch": 2.592723419286471,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004327506237037109,
      "loss": 5.4176,
      "mean_token_accuracy": 0.20332204550504684,
      "num_tokens": 54726191.0,
      "step": 29360
    },
    {
      "entropy": 5.820627689361572,
      "epoch": 2.5931649593783117,
      "grad_norm": 1.078125,
      "learning_rate": 0.000432728167161833,
      "loss": 5.4633,
      "mean_token_accuracy": 0.1922199085354805,
      "num_tokens": 54736748.0,
      "step": 29365
    },
    {
      "entropy": 5.76673994064331,
      "epoch": 2.5936064994701518,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004327057075301501,
      "loss": 5.3706,
      "mean_token_accuracy": 0.1989672973752022,
      "num_tokens": 54746623.0,
      "step": 29370
    },
    {
      "entropy": 5.743926239013672,
      "epoch": 2.5940480395619923,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043268324480910203,
      "loss": 5.3342,
      "mean_token_accuracy": 0.20460993349552153,
      "num_tokens": 54755778.0,
      "step": 29375
    },
    {
      "entropy": 5.784788036346436,
      "epoch": 2.5944895796538328,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000432660778999129,
      "loss": 5.5205,
      "mean_token_accuracy": 0.19481238722801208,
      "num_tokens": 54765410.0,
      "step": 29380
    },
    {
      "entropy": 5.861114501953125,
      "epoch": 2.594931119745673,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004326383101006709,
      "loss": 5.5352,
      "mean_token_accuracy": 0.18480388969182968,
      "num_tokens": 54776115.0,
      "step": 29385
    },
    {
      "entropy": 5.7684986114501955,
      "epoch": 2.5953726598375133,
      "grad_norm": 1.25,
      "learning_rate": 0.000432615838114168,
      "loss": 5.3831,
      "mean_token_accuracy": 0.19928668588399887,
      "num_tokens": 54784719.0,
      "step": 29390
    },
    {
      "entropy": 5.802220296859741,
      "epoch": 2.595814199929354,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043259336304006036,
      "loss": 5.3598,
      "mean_token_accuracy": 0.2038702741265297,
      "num_tokens": 54794003.0,
      "step": 29395
    },
    {
      "entropy": 5.765779113769531,
      "epoch": 2.596255740021194,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004325708848787884,
      "loss": 5.4402,
      "mean_token_accuracy": 0.1984880670905113,
      "num_tokens": 54803634.0,
      "step": 29400
    },
    {
      "entropy": 5.74446177482605,
      "epoch": 2.5966972801130344,
      "grad_norm": 1.25,
      "learning_rate": 0.0004325484036307925,
      "loss": 5.4563,
      "mean_token_accuracy": 0.20203528106212615,
      "num_tokens": 54813466.0,
      "step": 29405
    },
    {
      "entropy": 5.866516828536987,
      "epoch": 2.597138820204875,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004325259192965129,
      "loss": 5.4062,
      "mean_token_accuracy": 0.19515435844659806,
      "num_tokens": 54823855.0,
      "step": 29410
    },
    {
      "entropy": 5.7668064594268795,
      "epoch": 2.597580360296715,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004325034318763902,
      "loss": 5.4131,
      "mean_token_accuracy": 0.2014845073223114,
      "num_tokens": 54832696.0,
      "step": 29415
    },
    {
      "entropy": 5.68567385673523,
      "epoch": 2.5980219003885554,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004324809413708648,
      "loss": 5.4794,
      "mean_token_accuracy": 0.19361677169799804,
      "num_tokens": 54841552.0,
      "step": 29420
    },
    {
      "entropy": 5.72100944519043,
      "epoch": 2.5984634404803955,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004324584477803773,
      "loss": 5.3826,
      "mean_token_accuracy": 0.19885178059339523,
      "num_tokens": 54851196.0,
      "step": 29425
    },
    {
      "entropy": 5.895408964157104,
      "epoch": 2.598904980572236,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043243595110536846,
      "loss": 5.339,
      "mean_token_accuracy": 0.19950600266456603,
      "num_tokens": 54860334.0,
      "step": 29430
    },
    {
      "entropy": 5.713431119918823,
      "epoch": 2.5993465206640765,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043241345134627885,
      "loss": 5.3927,
      "mean_token_accuracy": 0.1996796727180481,
      "num_tokens": 54870056.0,
      "step": 29435
    },
    {
      "entropy": 5.751983404159546,
      "epoch": 2.5997880607559165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043239094850354914,
      "loss": 5.4914,
      "mean_token_accuracy": 0.19016815721988678,
      "num_tokens": 54880788.0,
      "step": 29440
    },
    {
      "entropy": 5.754257869720459,
      "epoch": 2.600229600847757,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004323684425776203,
      "loss": 5.418,
      "mean_token_accuracy": 0.19768842458724975,
      "num_tokens": 54890021.0,
      "step": 29445
    },
    {
      "entropy": 5.675734233856201,
      "epoch": 2.600671140939597,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043234593356893325,
      "loss": 5.2929,
      "mean_token_accuracy": 0.21934398710727693,
      "num_tokens": 54898884.0,
      "step": 29450
    },
    {
      "entropy": 5.78735728263855,
      "epoch": 2.6011126810314376,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043232342147792887,
      "loss": 5.4696,
      "mean_token_accuracy": 0.19620538800954818,
      "num_tokens": 54907742.0,
      "step": 29455
    },
    {
      "entropy": 5.790899562835693,
      "epoch": 2.601554221123278,
      "grad_norm": 1.15625,
      "learning_rate": 0.000432300906305048,
      "loss": 5.4459,
      "mean_token_accuracy": 0.19843090772628785,
      "num_tokens": 54916838.0,
      "step": 29460
    },
    {
      "entropy": 5.799296426773071,
      "epoch": 2.601995761215118,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004322783880507319,
      "loss": 5.505,
      "mean_token_accuracy": 0.18852390050888063,
      "num_tokens": 54925942.0,
      "step": 29465
    },
    {
      "entropy": 5.709391689300537,
      "epoch": 2.6024373013069586,
      "grad_norm": 1.375,
      "learning_rate": 0.00043225586671542164,
      "loss": 5.3154,
      "mean_token_accuracy": 0.19917650073766707,
      "num_tokens": 54934611.0,
      "step": 29470
    },
    {
      "entropy": 5.837875413894653,
      "epoch": 2.602878841398799,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004322333422995584,
      "loss": 5.4252,
      "mean_token_accuracy": 0.1924605831503868,
      "num_tokens": 54944354.0,
      "step": 29475
    },
    {
      "entropy": 5.768097352981568,
      "epoch": 2.603320381490639,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043221081480358337,
      "loss": 5.4912,
      "mean_token_accuracy": 0.19013358503580094,
      "num_tokens": 54954070.0,
      "step": 29480
    },
    {
      "entropy": 5.821988821029663,
      "epoch": 2.6037619215824797,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004321882842279379,
      "loss": 5.4838,
      "mean_token_accuracy": 0.18796878755092622,
      "num_tokens": 54963922.0,
      "step": 29485
    },
    {
      "entropy": 5.789308261871338,
      "epoch": 2.60420346167432,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004321657505730634,
      "loss": 5.3494,
      "mean_token_accuracy": 0.20446356385946274,
      "num_tokens": 54973202.0,
      "step": 29490
    },
    {
      "entropy": 5.798629808425903,
      "epoch": 2.60464500176616,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043214321383940117,
      "loss": 5.4229,
      "mean_token_accuracy": 0.19408866465091706,
      "num_tokens": 54982318.0,
      "step": 29495
    },
    {
      "entropy": 5.6570611000061035,
      "epoch": 2.6050865418580007,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043212067402739274,
      "loss": 5.2631,
      "mean_token_accuracy": 0.2152974784374237,
      "num_tokens": 54991397.0,
      "step": 29500
    },
    {
      "entropy": 5.8300128936767575,
      "epoch": 2.605528081949841,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004320981311374797,
      "loss": 5.47,
      "mean_token_accuracy": 0.19794840663671492,
      "num_tokens": 55001160.0,
      "step": 29505
    },
    {
      "entropy": 5.802194976806641,
      "epoch": 2.6059696220416813,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043207558517010364,
      "loss": 5.4616,
      "mean_token_accuracy": 0.20230181366205216,
      "num_tokens": 55010311.0,
      "step": 29510
    },
    {
      "entropy": 5.681719207763672,
      "epoch": 2.6064111621335218,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004320530361257061,
      "loss": 5.414,
      "mean_token_accuracy": 0.20637890547513962,
      "num_tokens": 55020017.0,
      "step": 29515
    },
    {
      "entropy": 5.843117713928223,
      "epoch": 2.6068527022253623,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000432030484004729,
      "loss": 5.5058,
      "mean_token_accuracy": 0.18809847086668013,
      "num_tokens": 55029930.0,
      "step": 29520
    },
    {
      "entropy": 5.799219083786011,
      "epoch": 2.6072942423172023,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043200792880761403,
      "loss": 5.392,
      "mean_token_accuracy": 0.1977636322379112,
      "num_tokens": 55039065.0,
      "step": 29525
    },
    {
      "entropy": 5.721060419082642,
      "epoch": 2.607735782409043,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000431985370534803,
      "loss": 5.3096,
      "mean_token_accuracy": 0.20606800317764282,
      "num_tokens": 55047592.0,
      "step": 29530
    },
    {
      "entropy": 5.744743680953979,
      "epoch": 2.6081773225008833,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043196280918673783,
      "loss": 5.4376,
      "mean_token_accuracy": 0.19209142178297042,
      "num_tokens": 55058031.0,
      "step": 29535
    },
    {
      "entropy": 5.808378839492798,
      "epoch": 2.6086188625927234,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043194024476386045,
      "loss": 5.4647,
      "mean_token_accuracy": 0.2009682312607765,
      "num_tokens": 55068283.0,
      "step": 29540
    },
    {
      "entropy": 5.7854651927948,
      "epoch": 2.609060402684564,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000431917677266613,
      "loss": 5.2773,
      "mean_token_accuracy": 0.20518157333135606,
      "num_tokens": 55077480.0,
      "step": 29545
    },
    {
      "entropy": 5.73414363861084,
      "epoch": 2.6095019427764043,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043189510669543754,
      "loss": 5.4474,
      "mean_token_accuracy": 0.1949603736400604,
      "num_tokens": 55086783.0,
      "step": 29550
    },
    {
      "entropy": 5.745943260192871,
      "epoch": 2.6099434828682444,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004318725330507762,
      "loss": 5.4916,
      "mean_token_accuracy": 0.1874939039349556,
      "num_tokens": 55096861.0,
      "step": 29555
    },
    {
      "entropy": 5.8044881343841555,
      "epoch": 2.610385022960085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043184995633307094,
      "loss": 5.4254,
      "mean_token_accuracy": 0.20245304852724075,
      "num_tokens": 55107504.0,
      "step": 29560
    },
    {
      "entropy": 5.879988813400269,
      "epoch": 2.610826563051925,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004318273765427644,
      "loss": 5.5478,
      "mean_token_accuracy": 0.1899382710456848,
      "num_tokens": 55116620.0,
      "step": 29565
    },
    {
      "entropy": 5.79324369430542,
      "epoch": 2.6112681031437655,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004318047936802987,
      "loss": 5.4056,
      "mean_token_accuracy": 0.2016095444560051,
      "num_tokens": 55126428.0,
      "step": 29570
    },
    {
      "entropy": 5.748732042312622,
      "epoch": 2.611709643235606,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004317822077461164,
      "loss": 5.3976,
      "mean_token_accuracy": 0.19265949130058288,
      "num_tokens": 55135369.0,
      "step": 29575
    },
    {
      "entropy": 5.770861530303955,
      "epoch": 2.612151183327446,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004317596187406596,
      "loss": 5.4138,
      "mean_token_accuracy": 0.19180414825677872,
      "num_tokens": 55144742.0,
      "step": 29580
    },
    {
      "entropy": 5.816020727157593,
      "epoch": 2.6125927234192865,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043173702666437114,
      "loss": 5.4824,
      "mean_token_accuracy": 0.1972362592816353,
      "num_tokens": 55154665.0,
      "step": 29585
    },
    {
      "entropy": 5.802307653427124,
      "epoch": 2.6130342635111266,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043171443151769347,
      "loss": 5.4097,
      "mean_token_accuracy": 0.20173251032829284,
      "num_tokens": 55163288.0,
      "step": 29590
    },
    {
      "entropy": 5.765371370315552,
      "epoch": 2.613475803602967,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004316918333010692,
      "loss": 5.4686,
      "mean_token_accuracy": 0.1924753949046135,
      "num_tokens": 55172775.0,
      "step": 29595
    },
    {
      "entropy": 5.8597166538238525,
      "epoch": 2.6139173436948075,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004316692320149411,
      "loss": 5.3961,
      "mean_token_accuracy": 0.1984267443418503,
      "num_tokens": 55182227.0,
      "step": 29600
    },
    {
      "entropy": 5.846681880950928,
      "epoch": 2.6143588837866476,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004316466276597517,
      "loss": 5.4561,
      "mean_token_accuracy": 0.19555874317884445,
      "num_tokens": 55191562.0,
      "step": 29605
    },
    {
      "entropy": 5.70208773612976,
      "epoch": 2.614800423878488,
      "grad_norm": 1.953125,
      "learning_rate": 0.000431624020235944,
      "loss": 5.3095,
      "mean_token_accuracy": 0.20652801543474197,
      "num_tokens": 55200439.0,
      "step": 29610
    },
    {
      "entropy": 5.687596654891967,
      "epoch": 2.6152419639703286,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004316014097439609,
      "loss": 5.3561,
      "mean_token_accuracy": 0.21022915244102477,
      "num_tokens": 55208850.0,
      "step": 29615
    },
    {
      "entropy": 5.830326795578003,
      "epoch": 2.6156835040621687,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004315787961842452,
      "loss": 5.4618,
      "mean_token_accuracy": 0.1959400549530983,
      "num_tokens": 55218364.0,
      "step": 29620
    },
    {
      "entropy": 5.765350389480591,
      "epoch": 2.616125044154009,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043155617955723994,
      "loss": 5.3941,
      "mean_token_accuracy": 0.202898745238781,
      "num_tokens": 55227201.0,
      "step": 29625
    },
    {
      "entropy": 5.77248740196228,
      "epoch": 2.6165665842458496,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004315335598633882,
      "loss": 5.5052,
      "mean_token_accuracy": 0.19262831658124924,
      "num_tokens": 55236907.0,
      "step": 29630
    },
    {
      "entropy": 5.876423168182373,
      "epoch": 2.6170081243376897,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043151093710313296,
      "loss": 5.4508,
      "mean_token_accuracy": 0.1915866419672966,
      "num_tokens": 55246795.0,
      "step": 29635
    },
    {
      "entropy": 5.801009941101074,
      "epoch": 2.61744966442953,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043148831127691754,
      "loss": 5.4258,
      "mean_token_accuracy": 0.1939913585782051,
      "num_tokens": 55255313.0,
      "step": 29640
    },
    {
      "entropy": 5.770849704742432,
      "epoch": 2.6178912045213707,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043146568238518505,
      "loss": 5.3864,
      "mean_token_accuracy": 0.19240757524967195,
      "num_tokens": 55264698.0,
      "step": 29645
    },
    {
      "entropy": 5.805172777175903,
      "epoch": 2.6183327446132107,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004314430504283789,
      "loss": 5.4816,
      "mean_token_accuracy": 0.19193679243326187,
      "num_tokens": 55274180.0,
      "step": 29650
    },
    {
      "entropy": 5.8388509273529055,
      "epoch": 2.6187742847050512,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004314204154069423,
      "loss": 5.4669,
      "mean_token_accuracy": 0.1947869062423706,
      "num_tokens": 55283552.0,
      "step": 29655
    },
    {
      "entropy": 5.828304195404053,
      "epoch": 2.6192158247968917,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004313977773213188,
      "loss": 5.4941,
      "mean_token_accuracy": 0.19182902574539185,
      "num_tokens": 55293929.0,
      "step": 29660
    },
    {
      "entropy": 5.768464660644531,
      "epoch": 2.619657364888732,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004313751361719517,
      "loss": 5.3576,
      "mean_token_accuracy": 0.2047067850828171,
      "num_tokens": 55303586.0,
      "step": 29665
    },
    {
      "entropy": 5.844944429397583,
      "epoch": 2.6200989049805723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043135249195928464,
      "loss": 5.3815,
      "mean_token_accuracy": 0.20325529724359512,
      "num_tokens": 55313032.0,
      "step": 29670
    },
    {
      "entropy": 5.747390556335449,
      "epoch": 2.620540445072413,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043132984468376124,
      "loss": 5.459,
      "mean_token_accuracy": 0.19661171436309816,
      "num_tokens": 55323046.0,
      "step": 29675
    },
    {
      "entropy": 5.826638793945312,
      "epoch": 2.620981985164253,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043130719434582505,
      "loss": 5.3917,
      "mean_token_accuracy": 0.20414864271879196,
      "num_tokens": 55331956.0,
      "step": 29680
    },
    {
      "entropy": 5.654925966262818,
      "epoch": 2.6214235252560933,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004312845409459198,
      "loss": 5.3609,
      "mean_token_accuracy": 0.19676022380590438,
      "num_tokens": 55341573.0,
      "step": 29685
    },
    {
      "entropy": 5.74471697807312,
      "epoch": 2.621865065347934,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043126188448448934,
      "loss": 5.3662,
      "mean_token_accuracy": 0.20449464172124862,
      "num_tokens": 55350325.0,
      "step": 29690
    },
    {
      "entropy": 5.881163787841797,
      "epoch": 2.622306605439774,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004312392249619774,
      "loss": 5.5134,
      "mean_token_accuracy": 0.1936175510287285,
      "num_tokens": 55359716.0,
      "step": 29695
    },
    {
      "entropy": 5.922883129119873,
      "epoch": 2.6227481455316144,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043121656237882786,
      "loss": 5.5141,
      "mean_token_accuracy": 0.1859311729669571,
      "num_tokens": 55369443.0,
      "step": 29700
    },
    {
      "entropy": 5.793255710601807,
      "epoch": 2.6231896856234544,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043119389673548477,
      "loss": 5.4581,
      "mean_token_accuracy": 0.19019715785980223,
      "num_tokens": 55379018.0,
      "step": 29705
    },
    {
      "entropy": 5.756792306900024,
      "epoch": 2.623631225715295,
      "grad_norm": 1.171875,
      "learning_rate": 0.000431171228032392,
      "loss": 5.4442,
      "mean_token_accuracy": 0.1938459977507591,
      "num_tokens": 55388681.0,
      "step": 29710
    },
    {
      "entropy": 5.816942596435547,
      "epoch": 2.6240727658071354,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004311485562699938,
      "loss": 5.4028,
      "mean_token_accuracy": 0.19151002019643784,
      "num_tokens": 55398495.0,
      "step": 29715
    },
    {
      "entropy": 5.774779224395752,
      "epoch": 2.6245143058989755,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004311258814487342,
      "loss": 5.4172,
      "mean_token_accuracy": 0.19862231165170668,
      "num_tokens": 55408498.0,
      "step": 29720
    },
    {
      "entropy": 5.735433149337768,
      "epoch": 2.624955845990816,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004311032035690572,
      "loss": 5.4409,
      "mean_token_accuracy": 0.19164654910564421,
      "num_tokens": 55417336.0,
      "step": 29725
    },
    {
      "entropy": 5.795939826965332,
      "epoch": 2.625397386082656,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004310805226314074,
      "loss": 5.4034,
      "mean_token_accuracy": 0.19357756972312928,
      "num_tokens": 55426828.0,
      "step": 29730
    },
    {
      "entropy": 5.7857901096344,
      "epoch": 2.6258389261744965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004310578386362288,
      "loss": 5.3547,
      "mean_token_accuracy": 0.19664252698421478,
      "num_tokens": 55435325.0,
      "step": 29735
    },
    {
      "entropy": 5.774301958084107,
      "epoch": 2.626280466266337,
      "grad_norm": 1.046875,
      "learning_rate": 0.000431035151583966,
      "loss": 5.4985,
      "mean_token_accuracy": 0.19261418133974076,
      "num_tokens": 55444892.0,
      "step": 29740
    },
    {
      "entropy": 5.684619760513305,
      "epoch": 2.626722006358177,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004310124614750632,
      "loss": 5.2802,
      "mean_token_accuracy": 0.21055570542812346,
      "num_tokens": 55453816.0,
      "step": 29745
    },
    {
      "entropy": 5.806141901016235,
      "epoch": 2.6271635464500176,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004309897683099652,
      "loss": 5.4852,
      "mean_token_accuracy": 0.19311239421367646,
      "num_tokens": 55463411.0,
      "step": 29750
    },
    {
      "entropy": 5.852423524856567,
      "epoch": 2.627605086541858,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004309670720891162,
      "loss": 5.5496,
      "mean_token_accuracy": 0.18464104384183883,
      "num_tokens": 55472939.0,
      "step": 29755
    },
    {
      "entropy": 5.7048133373260494,
      "epoch": 2.628046626633698,
      "grad_norm": 1.359375,
      "learning_rate": 0.000430944372812961,
      "loss": 5.2732,
      "mean_token_accuracy": 0.20721607506275178,
      "num_tokens": 55481149.0,
      "step": 29760
    },
    {
      "entropy": 5.745597457885742,
      "epoch": 2.6284881667255386,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004309216704819442,
      "loss": 5.3275,
      "mean_token_accuracy": 0.19859185218811035,
      "num_tokens": 55489858.0,
      "step": 29765
    },
    {
      "entropy": 5.7147603034973145,
      "epoch": 2.628929706817379,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004308989650965106,
      "loss": 5.4095,
      "mean_token_accuracy": 0.19465689957141877,
      "num_tokens": 55498835.0,
      "step": 29770
    },
    {
      "entropy": 5.7022651672363285,
      "epoch": 2.629371246909219,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004308762566571049,
      "loss": 5.3058,
      "mean_token_accuracy": 0.20810796618461608,
      "num_tokens": 55507437.0,
      "step": 29775
    },
    {
      "entropy": 5.73891863822937,
      "epoch": 2.6298127870010597,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004308535451641721,
      "loss": 5.4326,
      "mean_token_accuracy": 0.19521166980266572,
      "num_tokens": 55516304.0,
      "step": 29780
    },
    {
      "entropy": 5.778234672546387,
      "epoch": 2.6302543270929,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004308308306181569,
      "loss": 5.3044,
      "mean_token_accuracy": 0.2085944876074791,
      "num_tokens": 55525720.0,
      "step": 29785
    },
    {
      "entropy": 5.800660705566406,
      "epoch": 2.6306958671847402,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004308081130195042,
      "loss": 5.4961,
      "mean_token_accuracy": 0.1884981095790863,
      "num_tokens": 55534882.0,
      "step": 29790
    },
    {
      "entropy": 5.689759540557861,
      "epoch": 2.6311374072765807,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004307853923686594,
      "loss": 5.3495,
      "mean_token_accuracy": 0.20342826545238496,
      "num_tokens": 55544337.0,
      "step": 29795
    },
    {
      "entropy": 5.752853298187256,
      "epoch": 2.6315789473684212,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004307626686660673,
      "loss": 5.3714,
      "mean_token_accuracy": 0.19840333759784698,
      "num_tokens": 55553616.0,
      "step": 29800
    },
    {
      "entropy": 5.815249061584472,
      "epoch": 2.6320204874602613,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043073994191217307,
      "loss": 5.4518,
      "mean_token_accuracy": 0.1998927652835846,
      "num_tokens": 55564066.0,
      "step": 29805
    },
    {
      "entropy": 5.711155939102173,
      "epoch": 2.632462027552102,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043071721210742196,
      "loss": 5.3269,
      "mean_token_accuracy": 0.2074647828936577,
      "num_tokens": 55573189.0,
      "step": 29810
    },
    {
      "entropy": 5.7029369354248045,
      "epoch": 2.6329035676439423,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004306944792522593,
      "loss": 5.3527,
      "mean_token_accuracy": 0.20312198549509047,
      "num_tokens": 55581421.0,
      "step": 29815
    },
    {
      "entropy": 5.7491742134094235,
      "epoch": 2.6333451077357823,
      "grad_norm": 1.125,
      "learning_rate": 0.00043067174334713025,
      "loss": 5.4399,
      "mean_token_accuracy": 0.18612775355577468,
      "num_tokens": 55591323.0,
      "step": 29820
    },
    {
      "entropy": 5.727654266357422,
      "epoch": 2.633786647827623,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004306490043924803,
      "loss": 5.3505,
      "mean_token_accuracy": 0.2101445361971855,
      "num_tokens": 55600352.0,
      "step": 29825
    },
    {
      "entropy": 5.780841398239136,
      "epoch": 2.6342281879194633,
      "grad_norm": 1.125,
      "learning_rate": 0.0004306262623887549,
      "loss": 5.4294,
      "mean_token_accuracy": 0.19946608394384385,
      "num_tokens": 55609678.0,
      "step": 29830
    },
    {
      "entropy": 5.753357172012329,
      "epoch": 2.6346697280113034,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004306035173363995,
      "loss": 5.4598,
      "mean_token_accuracy": 0.196833273768425,
      "num_tokens": 55619401.0,
      "step": 29835
    },
    {
      "entropy": 5.718058681488037,
      "epoch": 2.635111268103144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043058076923585976,
      "loss": 5.4858,
      "mean_token_accuracy": 0.1924581855535507,
      "num_tokens": 55629208.0,
      "step": 29840
    },
    {
      "entropy": 5.736233568191528,
      "epoch": 2.635552808194984,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043055801808758123,
      "loss": 5.378,
      "mean_token_accuracy": 0.19561349004507064,
      "num_tokens": 55638304.0,
      "step": 29845
    },
    {
      "entropy": 5.779783868789673,
      "epoch": 2.6359943482868244,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043053526389200955,
      "loss": 5.4333,
      "mean_token_accuracy": 0.19789528250694274,
      "num_tokens": 55647982.0,
      "step": 29850
    },
    {
      "entropy": 5.721602439880371,
      "epoch": 2.636435888378665,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043051250664959053,
      "loss": 5.3431,
      "mean_token_accuracy": 0.20186741948127745,
      "num_tokens": 55657088.0,
      "step": 29855
    },
    {
      "entropy": 5.801981210708618,
      "epoch": 2.636877428470505,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004304897463607699,
      "loss": 5.4378,
      "mean_token_accuracy": 0.19602265059947968,
      "num_tokens": 55666451.0,
      "step": 29860
    },
    {
      "entropy": 5.686094188690186,
      "epoch": 2.6373189685623455,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004304669830259936,
      "loss": 5.3437,
      "mean_token_accuracy": 0.20267942249774934,
      "num_tokens": 55675636.0,
      "step": 29865
    },
    {
      "entropy": 5.746966361999512,
      "epoch": 2.6377605086541855,
      "grad_norm": 1.125,
      "learning_rate": 0.00043044421664570755,
      "loss": 5.4202,
      "mean_token_accuracy": 0.20154542475938797,
      "num_tokens": 55686092.0,
      "step": 29870
    },
    {
      "entropy": 5.714355182647705,
      "epoch": 2.638202048746026,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043042144722035765,
      "loss": 5.3801,
      "mean_token_accuracy": 0.19892633110284805,
      "num_tokens": 55695212.0,
      "step": 29875
    },
    {
      "entropy": 5.687577962875366,
      "epoch": 2.6386435888378665,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043039867475039,
      "loss": 5.3962,
      "mean_token_accuracy": 0.2062084048986435,
      "num_tokens": 55704471.0,
      "step": 29880
    },
    {
      "entropy": 5.798611783981324,
      "epoch": 2.6390851289297066,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004303758992362507,
      "loss": 5.4264,
      "mean_token_accuracy": 0.19213131368160247,
      "num_tokens": 55712915.0,
      "step": 29885
    },
    {
      "entropy": 5.806986141204834,
      "epoch": 2.639526669021547,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004303531206783859,
      "loss": 5.4417,
      "mean_token_accuracy": 0.19197893738746644,
      "num_tokens": 55722045.0,
      "step": 29890
    },
    {
      "entropy": 5.728535938262939,
      "epoch": 2.6399682091133876,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004303303390772419,
      "loss": 5.4028,
      "mean_token_accuracy": 0.20341577827930452,
      "num_tokens": 55731548.0,
      "step": 29895
    },
    {
      "entropy": 5.837156629562378,
      "epoch": 2.6404097492052276,
      "grad_norm": 1.515625,
      "learning_rate": 0.00043030755443326476,
      "loss": 5.5039,
      "mean_token_accuracy": 0.1926216796040535,
      "num_tokens": 55741494.0,
      "step": 29900
    },
    {
      "entropy": 5.835375690460205,
      "epoch": 2.640851289297068,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043028476674690107,
      "loss": 5.4254,
      "mean_token_accuracy": 0.19677566438913346,
      "num_tokens": 55750000.0,
      "step": 29905
    },
    {
      "entropy": 5.7540384292602536,
      "epoch": 2.6412928293889086,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004302619760185971,
      "loss": 5.4492,
      "mean_token_accuracy": 0.1950097292661667,
      "num_tokens": 55760375.0,
      "step": 29910
    },
    {
      "entropy": 5.776131582260132,
      "epoch": 2.6417343694807487,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004302391822487993,
      "loss": 5.3855,
      "mean_token_accuracy": 0.20069534182548524,
      "num_tokens": 55768726.0,
      "step": 29915
    },
    {
      "entropy": 5.754797887802124,
      "epoch": 2.642175909572589,
      "grad_norm": 1.125,
      "learning_rate": 0.0004302163854379541,
      "loss": 5.4457,
      "mean_token_accuracy": 0.19527125209569932,
      "num_tokens": 55778203.0,
      "step": 29920
    },
    {
      "entropy": 5.753163528442383,
      "epoch": 2.6426174496644297,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043019358558650837,
      "loss": 5.3541,
      "mean_token_accuracy": 0.20570621341466905,
      "num_tokens": 55787719.0,
      "step": 29925
    },
    {
      "entropy": 5.68881402015686,
      "epoch": 2.6430589897562697,
      "grad_norm": 1.25,
      "learning_rate": 0.00043017078269490846,
      "loss": 5.361,
      "mean_token_accuracy": 0.20189491361379625,
      "num_tokens": 55796824.0,
      "step": 29930
    },
    {
      "entropy": 5.805902624130249,
      "epoch": 2.6435005298481102,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043014797676360124,
      "loss": 5.3673,
      "mean_token_accuracy": 0.21006380915641784,
      "num_tokens": 55805460.0,
      "step": 29935
    },
    {
      "entropy": 5.624522018432617,
      "epoch": 2.6439420699399507,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004301251677930332,
      "loss": 5.3647,
      "mean_token_accuracy": 0.20150776952505112,
      "num_tokens": 55815112.0,
      "step": 29940
    },
    {
      "entropy": 5.693876886367798,
      "epoch": 2.644383610031791,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004301023557836516,
      "loss": 5.3834,
      "mean_token_accuracy": 0.2042497605085373,
      "num_tokens": 55823815.0,
      "step": 29945
    },
    {
      "entropy": 5.758120965957642,
      "epoch": 2.6448251501236313,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004300795407359029,
      "loss": 5.3564,
      "mean_token_accuracy": 0.20449803322553634,
      "num_tokens": 55833513.0,
      "step": 29950
    },
    {
      "entropy": 5.75466103553772,
      "epoch": 2.645266690215472,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043005672265023425,
      "loss": 5.3589,
      "mean_token_accuracy": 0.20376587212085723,
      "num_tokens": 55841765.0,
      "step": 29955
    },
    {
      "entropy": 5.759252691268921,
      "epoch": 2.645708230307312,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043003390152709264,
      "loss": 5.4342,
      "mean_token_accuracy": 0.20084785670042038,
      "num_tokens": 55851125.0,
      "step": 29960
    },
    {
      "entropy": 5.792479228973389,
      "epoch": 2.6461497703991523,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043001107736692504,
      "loss": 5.4058,
      "mean_token_accuracy": 0.2044704556465149,
      "num_tokens": 55861826.0,
      "step": 29965
    },
    {
      "entropy": 5.840427732467651,
      "epoch": 2.646591310490993,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004299882501701785,
      "loss": 5.4442,
      "mean_token_accuracy": 0.19263314753770827,
      "num_tokens": 55870658.0,
      "step": 29970
    },
    {
      "entropy": 5.753293180465699,
      "epoch": 2.647032850582833,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042996541993730035,
      "loss": 5.4204,
      "mean_token_accuracy": 0.1930813178420067,
      "num_tokens": 55880228.0,
      "step": 29975
    },
    {
      "entropy": 5.741337871551513,
      "epoch": 2.6474743906746734,
      "grad_norm": 1.125,
      "learning_rate": 0.00042994258666873773,
      "loss": 5.4711,
      "mean_token_accuracy": 0.19120197892189025,
      "num_tokens": 55890101.0,
      "step": 29980
    },
    {
      "entropy": 5.776100301742554,
      "epoch": 2.6479159307665134,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000429919750364938,
      "loss": 5.3347,
      "mean_token_accuracy": 0.20646623224020005,
      "num_tokens": 55899164.0,
      "step": 29985
    },
    {
      "entropy": 5.7620923042297365,
      "epoch": 2.648357470858354,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042989691102634847,
      "loss": 5.3737,
      "mean_token_accuracy": 0.20528269708156585,
      "num_tokens": 55907735.0,
      "step": 29990
    },
    {
      "entropy": 5.787467908859253,
      "epoch": 2.6487990109501944,
      "grad_norm": 1.125,
      "learning_rate": 0.00042987406865341657,
      "loss": 5.5114,
      "mean_token_accuracy": 0.18882398605346679,
      "num_tokens": 55917374.0,
      "step": 29995
    },
    {
      "entropy": 5.8128828525543215,
      "epoch": 2.6492405510420345,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004298512232465897,
      "loss": 5.4571,
      "mean_token_accuracy": 0.18937179893255235,
      "num_tokens": 55926430.0,
      "step": 30000
    },
    {
      "epoch": 2.6492405510420345,
      "eval_entropy": 5.586035329542236,
      "eval_loss": 5.488365650177002,
      "eval_mean_token_accuracy": 0.2036818137814292,
      "eval_num_tokens": 55926430.0,
      "eval_runtime": 26.7657,
      "eval_samples_per_second": 1319.414,
      "eval_steps_per_second": 164.95,
      "step": 30000
    }
  ],
  "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": 8.2769570878464e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}