{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.8673020813071388,
  "eval_steps": 3000,
  "global_step": 24000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.868724536895752,
      "epoch": 0.00038902937171756465,
      "grad_norm": 17.5,
      "learning_rate": 2e-06,
      "loss": 14.1486,
      "mean_token_accuracy": 0.0,
      "num_tokens": 8095.0,
      "step": 5
    },
    {
      "entropy": 4.891779136657715,
      "epoch": 0.0007780587434351293,
      "grad_norm": 26.125,
      "learning_rate": 4.5e-06,
      "loss": 14.148,
      "mean_token_accuracy": 0.0,
      "num_tokens": 17329.0,
      "step": 10
    },
    {
      "entropy": 4.949541425704956,
      "epoch": 0.001167088115152694,
      "grad_norm": 23.25,
      "learning_rate": 7e-06,
      "loss": 13.9134,
      "mean_token_accuracy": 0.0,
      "num_tokens": 25729.0,
      "step": 15
    },
    {
      "entropy": 5.25663423538208,
      "epoch": 0.0015561174868702586,
      "grad_norm": 35.0,
      "learning_rate": 9.5e-06,
      "loss": 13.3982,
      "mean_token_accuracy": 0.0,
      "num_tokens": 33980.0,
      "step": 20
    },
    {
      "entropy": 6.6940947532653805,
      "epoch": 0.0019451468585878235,
      "grad_norm": 26.0,
      "learning_rate": 1.2e-05,
      "loss": 12.2791,
      "mean_token_accuracy": 0.0,
      "num_tokens": 41675.0,
      "step": 25
    },
    {
      "entropy": 9.23413438796997,
      "epoch": 0.002334176230305388,
      "grad_norm": 6.84375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 11.2025,
      "mean_token_accuracy": 0.0,
      "num_tokens": 50277.0,
      "step": 30
    },
    {
      "entropy": 10.568378067016601,
      "epoch": 0.0027232056020229526,
      "grad_norm": 2.78125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.7304,
      "mean_token_accuracy": 0.0,
      "num_tokens": 58790.0,
      "step": 35
    },
    {
      "entropy": 10.704268169403075,
      "epoch": 0.0031122349737405172,
      "grad_norm": 2.90625,
      "learning_rate": 1.95e-05,
      "loss": 10.5063,
      "mean_token_accuracy": 0.024650751310400666,
      "num_tokens": 67247.0,
      "step": 40
    },
    {
      "entropy": 10.695908069610596,
      "epoch": 0.003501264345458082,
      "grad_norm": 2.703125,
      "learning_rate": 2.2e-05,
      "loss": 10.1754,
      "mean_token_accuracy": 0.04813225194811821,
      "num_tokens": 75174.0,
      "step": 45
    },
    {
      "entropy": 10.63707094192505,
      "epoch": 0.003890293717175647,
      "grad_norm": 2.15625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.8604,
      "mean_token_accuracy": 0.05086616277694702,
      "num_tokens": 83203.0,
      "step": 50
    },
    {
      "entropy": 10.510939502716065,
      "epoch": 0.004279323088893211,
      "grad_norm": 1.9140625,
      "learning_rate": 2.7e-05,
      "loss": 9.6639,
      "mean_token_accuracy": 0.04917947538197041,
      "num_tokens": 91798.0,
      "step": 55
    },
    {
      "entropy": 10.447108936309814,
      "epoch": 0.004668352460610776,
      "grad_norm": 1.7734375,
      "learning_rate": 2.95e-05,
      "loss": 9.6115,
      "mean_token_accuracy": 0.047414034977555274,
      "num_tokens": 100869.0,
      "step": 60
    },
    {
      "entropy": 10.47428493499756,
      "epoch": 0.0050573818323283405,
      "grad_norm": 1.96875,
      "learning_rate": 3.2e-05,
      "loss": 9.5064,
      "mean_token_accuracy": 0.043885057978332045,
      "num_tokens": 109697.0,
      "step": 65
    },
    {
      "entropy": 10.48178014755249,
      "epoch": 0.005446411204045905,
      "grad_norm": 1.921875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.4411,
      "mean_token_accuracy": 0.04529423713684082,
      "num_tokens": 118656.0,
      "step": 70
    },
    {
      "entropy": 10.426056289672852,
      "epoch": 0.00583544057576347,
      "grad_norm": 1.5625,
      "learning_rate": 3.7e-05,
      "loss": 9.4078,
      "mean_token_accuracy": 0.04362858422100544,
      "num_tokens": 127748.0,
      "step": 75
    },
    {
      "entropy": 10.388638496398926,
      "epoch": 0.0062244699474810344,
      "grad_norm": 1.671875,
      "learning_rate": 3.95e-05,
      "loss": 9.2589,
      "mean_token_accuracy": 0.04771117903292179,
      "num_tokens": 136621.0,
      "step": 80
    },
    {
      "entropy": 10.360412406921387,
      "epoch": 0.006613499319198599,
      "grad_norm": 1.7109375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.0947,
      "mean_token_accuracy": 0.05321100503206253,
      "num_tokens": 145793.0,
      "step": 85
    },
    {
      "entropy": 10.309724712371827,
      "epoch": 0.007002528690916164,
      "grad_norm": 1.6171875,
      "learning_rate": 4.45e-05,
      "loss": 8.9975,
      "mean_token_accuracy": 0.05680464021861553,
      "num_tokens": 154466.0,
      "step": 90
    },
    {
      "entropy": 10.248615932464599,
      "epoch": 0.007391558062633729,
      "grad_norm": 1.78125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.9498,
      "mean_token_accuracy": 0.05836099237203598,
      "num_tokens": 163297.0,
      "step": 95
    },
    {
      "entropy": 10.143774604797363,
      "epoch": 0.007780587434351294,
      "grad_norm": 1.4921875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.7819,
      "mean_token_accuracy": 0.06600868813693524,
      "num_tokens": 171747.0,
      "step": 100
    },
    {
      "entropy": 10.016874027252197,
      "epoch": 0.008169616806068859,
      "grad_norm": 1.3125,
      "learning_rate": 5.2e-05,
      "loss": 8.7019,
      "mean_token_accuracy": 0.062353368103504184,
      "num_tokens": 180499.0,
      "step": 105
    },
    {
      "entropy": 10.008924961090088,
      "epoch": 0.008558646177786422,
      "grad_norm": 1.390625,
      "learning_rate": 5.45e-05,
      "loss": 8.5723,
      "mean_token_accuracy": 0.06311165988445282,
      "num_tokens": 189227.0,
      "step": 110
    },
    {
      "entropy": 9.854091453552247,
      "epoch": 0.008947675549503988,
      "grad_norm": 1.28125,
      "learning_rate": 5.7e-05,
      "loss": 8.4385,
      "mean_token_accuracy": 0.06438967436552048,
      "num_tokens": 198388.0,
      "step": 115
    },
    {
      "entropy": 9.726921844482423,
      "epoch": 0.009336704921221552,
      "grad_norm": 1.0859375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.3286,
      "mean_token_accuracy": 0.06439770273864269,
      "num_tokens": 206854.0,
      "step": 120
    },
    {
      "entropy": 9.583395957946777,
      "epoch": 0.009725734292939117,
      "grad_norm": 1.1953125,
      "learning_rate": 6.2e-05,
      "loss": 8.2097,
      "mean_token_accuracy": 0.06303132474422454,
      "num_tokens": 215506.0,
      "step": 125
    },
    {
      "entropy": 9.42068099975586,
      "epoch": 0.010114763664656681,
      "grad_norm": 1.0390625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.1093,
      "mean_token_accuracy": 0.06823768317699433,
      "num_tokens": 223717.0,
      "step": 130
    },
    {
      "entropy": 9.187827396392823,
      "epoch": 0.010503793036374246,
      "grad_norm": 0.890625,
      "learning_rate": 6.7e-05,
      "loss": 7.9995,
      "mean_token_accuracy": 0.06632016785442829,
      "num_tokens": 232480.0,
      "step": 135
    },
    {
      "entropy": 9.082991695404052,
      "epoch": 0.01089282240809181,
      "grad_norm": 1.1015625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.9293,
      "mean_token_accuracy": 0.06358420252799987,
      "num_tokens": 241350.0,
      "step": 140
    },
    {
      "entropy": 8.822460651397705,
      "epoch": 0.011281851779809376,
      "grad_norm": 0.9765625,
      "learning_rate": 7.2e-05,
      "loss": 7.8758,
      "mean_token_accuracy": 0.06782571263611317,
      "num_tokens": 250610.0,
      "step": 145
    },
    {
      "entropy": 8.690089511871339,
      "epoch": 0.01167088115152694,
      "grad_norm": 0.8515625,
      "learning_rate": 7.45e-05,
      "loss": 7.7822,
      "mean_token_accuracy": 0.060015364363789556,
      "num_tokens": 259672.0,
      "step": 150
    },
    {
      "entropy": 8.512046813964844,
      "epoch": 0.012059910523244505,
      "grad_norm": 1.015625,
      "learning_rate": 7.7e-05,
      "loss": 7.7639,
      "mean_token_accuracy": 0.06780118122696877,
      "num_tokens": 267826.0,
      "step": 155
    },
    {
      "entropy": 8.466501140594483,
      "epoch": 0.012448939894962069,
      "grad_norm": 1.0078125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.7686,
      "mean_token_accuracy": 0.060590619966387746,
      "num_tokens": 277324.0,
      "step": 160
    },
    {
      "entropy": 8.349157810211182,
      "epoch": 0.012837969266679634,
      "grad_norm": 0.96875,
      "learning_rate": 8.2e-05,
      "loss": 7.6015,
      "mean_token_accuracy": 0.06816631369292736,
      "num_tokens": 285482.0,
      "step": 165
    },
    {
      "entropy": 8.220991134643555,
      "epoch": 0.013226998638397198,
      "grad_norm": 0.9453125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.6413,
      "mean_token_accuracy": 0.06899847462773323,
      "num_tokens": 293684.0,
      "step": 170
    },
    {
      "entropy": 8.094815015792847,
      "epoch": 0.013616028010114764,
      "grad_norm": 1.0390625,
      "learning_rate": 8.7e-05,
      "loss": 7.602,
      "mean_token_accuracy": 0.06723922677338123,
      "num_tokens": 302605.0,
      "step": 175
    },
    {
      "entropy": 8.08293595314026,
      "epoch": 0.014005057381832327,
      "grad_norm": 0.93359375,
      "learning_rate": 8.95e-05,
      "loss": 7.6409,
      "mean_token_accuracy": 0.0671521469950676,
      "num_tokens": 310849.0,
      "step": 180
    },
    {
      "entropy": 8.118443584442138,
      "epoch": 0.014394086753549893,
      "grad_norm": 0.875,
      "learning_rate": 9.2e-05,
      "loss": 7.646,
      "mean_token_accuracy": 0.06186659783124924,
      "num_tokens": 319752.0,
      "step": 185
    },
    {
      "entropy": 8.10387020111084,
      "epoch": 0.014783116125267459,
      "grad_norm": 1.1328125,
      "learning_rate": 9.45e-05,
      "loss": 7.681,
      "mean_token_accuracy": 0.06433467343449592,
      "num_tokens": 328009.0,
      "step": 190
    },
    {
      "entropy": 7.984123754501343,
      "epoch": 0.015172145496985022,
      "grad_norm": 0.83984375,
      "learning_rate": 9.7e-05,
      "loss": 7.5648,
      "mean_token_accuracy": 0.0687804438173771,
      "num_tokens": 337414.0,
      "step": 195
    },
    {
      "entropy": 7.993130111694336,
      "epoch": 0.015561174868702588,
      "grad_norm": 0.91015625,
      "learning_rate": 9.95e-05,
      "loss": 7.5609,
      "mean_token_accuracy": 0.06835784390568733,
      "num_tokens": 345419.0,
      "step": 200
    },
    {
      "entropy": 7.953881740570068,
      "epoch": 0.01595020424042015,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000102,
      "loss": 7.5415,
      "mean_token_accuracy": 0.07114643156528473,
      "num_tokens": 353435.0,
      "step": 205
    },
    {
      "entropy": 7.923952913284301,
      "epoch": 0.016339233612137717,
      "grad_norm": 1.015625,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.5578,
      "mean_token_accuracy": 0.07233858108520508,
      "num_tokens": 361840.0,
      "step": 210
    },
    {
      "entropy": 7.9594556331634525,
      "epoch": 0.016728262983855283,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000107,
      "loss": 7.6335,
      "mean_token_accuracy": 0.06823679059743881,
      "num_tokens": 370873.0,
      "step": 215
    },
    {
      "entropy": 7.93640627861023,
      "epoch": 0.017117292355572845,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001095,
      "loss": 7.5629,
      "mean_token_accuracy": 0.06850668452680111,
      "num_tokens": 379558.0,
      "step": 220
    },
    {
      "entropy": 7.873511123657226,
      "epoch": 0.01750632172729041,
      "grad_norm": 1.03125,
      "learning_rate": 0.000112,
      "loss": 7.5426,
      "mean_token_accuracy": 0.07185782194137573,
      "num_tokens": 387870.0,
      "step": 225
    },
    {
      "entropy": 7.965339994430542,
      "epoch": 0.017895351099007976,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001145,
      "loss": 7.6015,
      "mean_token_accuracy": 0.07074649557471276,
      "num_tokens": 396442.0,
      "step": 230
    },
    {
      "entropy": 7.89727029800415,
      "epoch": 0.01828438047072554,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.5481,
      "mean_token_accuracy": 0.0699117787182331,
      "num_tokens": 404764.0,
      "step": 235
    },
    {
      "entropy": 7.8868903636932375,
      "epoch": 0.018673409842443103,
      "grad_norm": 0.953125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.4854,
      "mean_token_accuracy": 0.0721463244408369,
      "num_tokens": 413355.0,
      "step": 240
    },
    {
      "entropy": 7.873054361343383,
      "epoch": 0.01906243921416067,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000122,
      "loss": 7.4,
      "mean_token_accuracy": 0.07351703196763992,
      "num_tokens": 421868.0,
      "step": 245
    },
    {
      "entropy": 7.893638324737549,
      "epoch": 0.019451468585878234,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0001245,
      "loss": 7.4402,
      "mean_token_accuracy": 0.07299891375005245,
      "num_tokens": 431106.0,
      "step": 250
    },
    {
      "entropy": 7.8072466373443605,
      "epoch": 0.0198404979575958,
      "grad_norm": 1.046875,
      "learning_rate": 0.000127,
      "loss": 7.5599,
      "mean_token_accuracy": 0.06760080493986606,
      "num_tokens": 439459.0,
      "step": 255
    },
    {
      "entropy": 7.876844310760498,
      "epoch": 0.020229527329313362,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001295,
      "loss": 7.3982,
      "mean_token_accuracy": 0.07463918179273606,
      "num_tokens": 447943.0,
      "step": 260
    },
    {
      "entropy": 7.733979797363281,
      "epoch": 0.020618556701030927,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000132,
      "loss": 7.5169,
      "mean_token_accuracy": 0.07182875908911228,
      "num_tokens": 456691.0,
      "step": 265
    },
    {
      "entropy": 7.837671566009521,
      "epoch": 0.021007586072748493,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.4317,
      "mean_token_accuracy": 0.07321453914046287,
      "num_tokens": 466388.0,
      "step": 270
    },
    {
      "entropy": 7.768101596832276,
      "epoch": 0.02139661544446606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.494,
      "mean_token_accuracy": 0.07455857545137405,
      "num_tokens": 474544.0,
      "step": 275
    },
    {
      "entropy": 7.842556571960449,
      "epoch": 0.02178564481618362,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001395,
      "loss": 7.516,
      "mean_token_accuracy": 0.074214668571949,
      "num_tokens": 483561.0,
      "step": 280
    },
    {
      "entropy": 7.771492433547974,
      "epoch": 0.022174674187901186,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.3561,
      "mean_token_accuracy": 0.07050973400473595,
      "num_tokens": 492036.0,
      "step": 285
    },
    {
      "entropy": 7.75999813079834,
      "epoch": 0.02256370355961875,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001445,
      "loss": 7.4784,
      "mean_token_accuracy": 0.06661519259214402,
      "num_tokens": 500978.0,
      "step": 290
    },
    {
      "entropy": 7.853191137313843,
      "epoch": 0.022952732931336317,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000147,
      "loss": 7.475,
      "mean_token_accuracy": 0.07539443038403988,
      "num_tokens": 509719.0,
      "step": 295
    },
    {
      "entropy": 7.738577365875244,
      "epoch": 0.02334176230305388,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0001495,
      "loss": 7.4126,
      "mean_token_accuracy": 0.07369871325790882,
      "num_tokens": 518422.0,
      "step": 300
    },
    {
      "entropy": 7.7334229946136475,
      "epoch": 0.023730791674771445,
      "grad_norm": 1.21875,
      "learning_rate": 0.000152,
      "loss": 7.4423,
      "mean_token_accuracy": 0.0712751142680645,
      "num_tokens": 527172.0,
      "step": 305
    },
    {
      "entropy": 7.780814075469971,
      "epoch": 0.02411982104648901,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.3618,
      "mean_token_accuracy": 0.07441995665431023,
      "num_tokens": 536070.0,
      "step": 310
    },
    {
      "entropy": 7.7062712669372555,
      "epoch": 0.024508850418206576,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000157,
      "loss": 7.3546,
      "mean_token_accuracy": 0.07498554512858391,
      "num_tokens": 544556.0,
      "step": 315
    },
    {
      "entropy": 7.870984935760498,
      "epoch": 0.024897879789924138,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001595,
      "loss": 7.3584,
      "mean_token_accuracy": 0.07835289165377617,
      "num_tokens": 553122.0,
      "step": 320
    },
    {
      "entropy": 7.6616284370422365,
      "epoch": 0.025286909161641703,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000162,
      "loss": 7.4064,
      "mean_token_accuracy": 0.0707610372453928,
      "num_tokens": 562265.0,
      "step": 325
    },
    {
      "entropy": 7.606665897369385,
      "epoch": 0.02567593853335927,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.359,
      "mean_token_accuracy": 0.08119581118226052,
      "num_tokens": 570938.0,
      "step": 330
    },
    {
      "entropy": 7.7274675369262695,
      "epoch": 0.026064967905076834,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.3081,
      "mean_token_accuracy": 0.07931275889277459,
      "num_tokens": 578568.0,
      "step": 335
    },
    {
      "entropy": 7.719749879837036,
      "epoch": 0.026453997276794396,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.3575,
      "mean_token_accuracy": 0.07647890225052834,
      "num_tokens": 587013.0,
      "step": 340
    },
    {
      "entropy": 7.669827508926391,
      "epoch": 0.026843026648511962,
      "grad_norm": 1.03125,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.3932,
      "mean_token_accuracy": 0.07862738743424416,
      "num_tokens": 595354.0,
      "step": 345
    },
    {
      "entropy": 7.759836196899414,
      "epoch": 0.027232056020229527,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.3681,
      "mean_token_accuracy": 0.07964789792895317,
      "num_tokens": 603971.0,
      "step": 350
    },
    {
      "entropy": 7.578008699417114,
      "epoch": 0.027621085391947093,
      "grad_norm": 1.078125,
      "learning_rate": 0.000177,
      "loss": 7.2821,
      "mean_token_accuracy": 0.08256564736366272,
      "num_tokens": 612097.0,
      "step": 355
    },
    {
      "entropy": 7.757014989852905,
      "epoch": 0.028010114763664655,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001795,
      "loss": 7.3505,
      "mean_token_accuracy": 0.07818787395954133,
      "num_tokens": 621445.0,
      "step": 360
    },
    {
      "entropy": 7.615339231491089,
      "epoch": 0.02839914413538222,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000182,
      "loss": 7.4092,
      "mean_token_accuracy": 0.0784608293324709,
      "num_tokens": 630041.0,
      "step": 365
    },
    {
      "entropy": 7.631348466873169,
      "epoch": 0.028788173507099786,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001845,
      "loss": 7.2121,
      "mean_token_accuracy": 0.08307572454214096,
      "num_tokens": 638317.0,
      "step": 370
    },
    {
      "entropy": 7.66585636138916,
      "epoch": 0.02917720287881735,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000187,
      "loss": 7.3499,
      "mean_token_accuracy": 0.07931018397212028,
      "num_tokens": 647297.0,
      "step": 375
    },
    {
      "entropy": 7.729125881195069,
      "epoch": 0.029566232250534917,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001895,
      "loss": 7.3836,
      "mean_token_accuracy": 0.08041680678725242,
      "num_tokens": 655309.0,
      "step": 380
    },
    {
      "entropy": 7.609511232376098,
      "epoch": 0.02995526162225248,
      "grad_norm": 1.140625,
      "learning_rate": 0.000192,
      "loss": 7.2463,
      "mean_token_accuracy": 0.07354939132928848,
      "num_tokens": 663879.0,
      "step": 385
    },
    {
      "entropy": 7.602015590667724,
      "epoch": 0.030344290993970045,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001945,
      "loss": 7.1702,
      "mean_token_accuracy": 0.08083694204688072,
      "num_tokens": 672913.0,
      "step": 390
    },
    {
      "entropy": 7.568642234802246,
      "epoch": 0.03073332036568761,
      "grad_norm": 1.046875,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.2686,
      "mean_token_accuracy": 0.08029174320399761,
      "num_tokens": 682282.0,
      "step": 395
    },
    {
      "entropy": 7.541975831985473,
      "epoch": 0.031122349737405176,
      "grad_norm": 1.125,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.2759,
      "mean_token_accuracy": 0.07506514936685563,
      "num_tokens": 691432.0,
      "step": 400
    },
    {
      "entropy": 7.6201324462890625,
      "epoch": 0.03151137910912274,
      "grad_norm": 1.140625,
      "learning_rate": 0.000202,
      "loss": 7.263,
      "mean_token_accuracy": 0.08165091350674629,
      "num_tokens": 700639.0,
      "step": 405
    },
    {
      "entropy": 7.666884183883667,
      "epoch": 0.0319004084808403,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.3413,
      "mean_token_accuracy": 0.07532657012343406,
      "num_tokens": 709200.0,
      "step": 410
    },
    {
      "entropy": 7.459015941619873,
      "epoch": 0.03228943785255787,
      "grad_norm": 1.15625,
      "learning_rate": 0.000207,
      "loss": 7.1665,
      "mean_token_accuracy": 0.08444926962256431,
      "num_tokens": 717266.0,
      "step": 415
    },
    {
      "entropy": 7.566874313354492,
      "epoch": 0.032678467224275434,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002095,
      "loss": 7.1734,
      "mean_token_accuracy": 0.08190047703683376,
      "num_tokens": 725799.0,
      "step": 420
    },
    {
      "entropy": 7.5747381210327145,
      "epoch": 0.033067496595993,
      "grad_norm": 1.578125,
      "learning_rate": 0.000212,
      "loss": 7.2685,
      "mean_token_accuracy": 0.08348871022462845,
      "num_tokens": 734375.0,
      "step": 425
    },
    {
      "entropy": 7.568057584762573,
      "epoch": 0.033456525967710565,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002145,
      "loss": 7.3127,
      "mean_token_accuracy": 0.07849649339914322,
      "num_tokens": 743453.0,
      "step": 430
    },
    {
      "entropy": 7.540275430679321,
      "epoch": 0.033845555339428124,
      "grad_norm": 0.9375,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.1611,
      "mean_token_accuracy": 0.0852227695286274,
      "num_tokens": 752443.0,
      "step": 435
    },
    {
      "entropy": 7.473781156539917,
      "epoch": 0.03423458471114569,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002195,
      "loss": 7.1522,
      "mean_token_accuracy": 0.07828448340296745,
      "num_tokens": 761294.0,
      "step": 440
    },
    {
      "entropy": 7.516408538818359,
      "epoch": 0.034623614082863255,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000222,
      "loss": 7.1092,
      "mean_token_accuracy": 0.08720964565873146,
      "num_tokens": 769489.0,
      "step": 445
    },
    {
      "entropy": 7.519400596618652,
      "epoch": 0.03501264345458082,
      "grad_norm": 1.375,
      "learning_rate": 0.0002245,
      "loss": 7.1954,
      "mean_token_accuracy": 0.07971547171473503,
      "num_tokens": 777836.0,
      "step": 450
    },
    {
      "entropy": 7.5304583549499515,
      "epoch": 0.035401672826298386,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.212,
      "mean_token_accuracy": 0.07866286486387253,
      "num_tokens": 786543.0,
      "step": 455
    },
    {
      "entropy": 7.456272268295288,
      "epoch": 0.03579070219801595,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.2535,
      "mean_token_accuracy": 0.08314662128686905,
      "num_tokens": 794525.0,
      "step": 460
    },
    {
      "entropy": 7.433866357803344,
      "epoch": 0.03617973156973352,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.1606,
      "mean_token_accuracy": 0.08660675287246704,
      "num_tokens": 803537.0,
      "step": 465
    },
    {
      "entropy": 7.4901923656463625,
      "epoch": 0.03656876094145108,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.1265,
      "mean_token_accuracy": 0.08419004380702973,
      "num_tokens": 811706.0,
      "step": 470
    },
    {
      "entropy": 7.434392309188842,
      "epoch": 0.03695779031316864,
      "grad_norm": 1.15625,
      "learning_rate": 0.000237,
      "loss": 7.1685,
      "mean_token_accuracy": 0.0864868015050888,
      "num_tokens": 820182.0,
      "step": 475
    },
    {
      "entropy": 7.459165287017822,
      "epoch": 0.03734681968488621,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002395,
      "loss": 7.1077,
      "mean_token_accuracy": 0.08410876803100109,
      "num_tokens": 829808.0,
      "step": 480
    },
    {
      "entropy": 7.448411321640014,
      "epoch": 0.03773584905660377,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000242,
      "loss": 7.048,
      "mean_token_accuracy": 0.08421071916818619,
      "num_tokens": 838838.0,
      "step": 485
    },
    {
      "entropy": 7.44689908027649,
      "epoch": 0.03812487842832134,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002445,
      "loss": 7.1524,
      "mean_token_accuracy": 0.08371994793415069,
      "num_tokens": 847800.0,
      "step": 490
    },
    {
      "entropy": 7.513600492477417,
      "epoch": 0.0385139078000389,
      "grad_norm": 1.5,
      "learning_rate": 0.000247,
      "loss": 7.2233,
      "mean_token_accuracy": 0.08270580545067788,
      "num_tokens": 856503.0,
      "step": 495
    },
    {
      "entropy": 7.354915380477905,
      "epoch": 0.03890293717175647,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002495,
      "loss": 7.1189,
      "mean_token_accuracy": 0.08470338881015778,
      "num_tokens": 865245.0,
      "step": 500
    },
    {
      "entropy": 7.403017473220825,
      "epoch": 0.039291966543474034,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000252,
      "loss": 7.1297,
      "mean_token_accuracy": 0.08570023030042648,
      "num_tokens": 874485.0,
      "step": 505
    },
    {
      "entropy": 7.318638038635254,
      "epoch": 0.0396809959151916,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002545,
      "loss": 6.9894,
      "mean_token_accuracy": 0.09045747816562652,
      "num_tokens": 883301.0,
      "step": 510
    },
    {
      "entropy": 7.306433153152466,
      "epoch": 0.04007002528690916,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000257,
      "loss": 7.0393,
      "mean_token_accuracy": 0.08763011917471886,
      "num_tokens": 892387.0,
      "step": 515
    },
    {
      "entropy": 7.318406486511231,
      "epoch": 0.040459054658626724,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002595,
      "loss": 7.036,
      "mean_token_accuracy": 0.09374425634741783,
      "num_tokens": 901522.0,
      "step": 520
    },
    {
      "entropy": 7.283069658279419,
      "epoch": 0.04084808403034429,
      "grad_norm": 1.171875,
      "learning_rate": 0.000262,
      "loss": 6.9894,
      "mean_token_accuracy": 0.08943136781454086,
      "num_tokens": 909848.0,
      "step": 525
    },
    {
      "entropy": 7.451201486587524,
      "epoch": 0.041237113402061855,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.2176,
      "mean_token_accuracy": 0.08226554542779922,
      "num_tokens": 919079.0,
      "step": 530
    },
    {
      "entropy": 7.3233407497406,
      "epoch": 0.04162614277377942,
      "grad_norm": 1.625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.963,
      "mean_token_accuracy": 0.09083623737096787,
      "num_tokens": 927807.0,
      "step": 535
    },
    {
      "entropy": 7.230840492248535,
      "epoch": 0.042015172145496986,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.0691,
      "mean_token_accuracy": 0.091452506929636,
      "num_tokens": 936498.0,
      "step": 540
    },
    {
      "entropy": 7.414189958572388,
      "epoch": 0.04240420151721455,
      "grad_norm": 1.046875,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.1387,
      "mean_token_accuracy": 0.09394621923565864,
      "num_tokens": 945436.0,
      "step": 545
    },
    {
      "entropy": 7.326435708999634,
      "epoch": 0.04279323088893212,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002745,
      "loss": 7.0618,
      "mean_token_accuracy": 0.08727624863386155,
      "num_tokens": 953990.0,
      "step": 550
    },
    {
      "entropy": 7.232479190826416,
      "epoch": 0.04318226026064968,
      "grad_norm": 1.140625,
      "learning_rate": 0.000277,
      "loss": 6.9777,
      "mean_token_accuracy": 0.09197546169161797,
      "num_tokens": 962008.0,
      "step": 555
    },
    {
      "entropy": 7.343172931671143,
      "epoch": 0.04357128963236724,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002795,
      "loss": 7.1391,
      "mean_token_accuracy": 0.08779035359621049,
      "num_tokens": 970935.0,
      "step": 560
    },
    {
      "entropy": 7.3697394847869875,
      "epoch": 0.04396031900408481,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.0947,
      "mean_token_accuracy": 0.08397784754633904,
      "num_tokens": 979492.0,
      "step": 565
    },
    {
      "entropy": 7.260292625427246,
      "epoch": 0.04434934837580237,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002845,
      "loss": 6.9991,
      "mean_token_accuracy": 0.08743947520852088,
      "num_tokens": 987866.0,
      "step": 570
    },
    {
      "entropy": 7.262716388702392,
      "epoch": 0.04473837774751994,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000287,
      "loss": 7.0328,
      "mean_token_accuracy": 0.09108475595712662,
      "num_tokens": 996760.0,
      "step": 575
    },
    {
      "entropy": 7.308055782318116,
      "epoch": 0.0451274071192375,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002895,
      "loss": 6.9709,
      "mean_token_accuracy": 0.09112134501338005,
      "num_tokens": 1005432.0,
      "step": 580
    },
    {
      "entropy": 7.290707445144653,
      "epoch": 0.04551643649095507,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000292,
      "loss": 7.0545,
      "mean_token_accuracy": 0.0900682970881462,
      "num_tokens": 1013729.0,
      "step": 585
    },
    {
      "entropy": 7.068312883377075,
      "epoch": 0.045905465862672634,
      "grad_norm": 1.125,
      "learning_rate": 0.0002945,
      "loss": 6.8894,
      "mean_token_accuracy": 0.09857918545603753,
      "num_tokens": 1022485.0,
      "step": 590
    },
    {
      "entropy": 7.210888338088989,
      "epoch": 0.0462944952343902,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000297,
      "loss": 6.9469,
      "mean_token_accuracy": 0.09099083542823791,
      "num_tokens": 1030523.0,
      "step": 595
    },
    {
      "entropy": 7.292349624633789,
      "epoch": 0.04668352460610776,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002995,
      "loss": 7.0544,
      "mean_token_accuracy": 0.09193994700908661,
      "num_tokens": 1039268.0,
      "step": 600
    },
    {
      "entropy": 7.170610523223877,
      "epoch": 0.047072553977825324,
      "grad_norm": 1.140625,
      "learning_rate": 0.000302,
      "loss": 6.9503,
      "mean_token_accuracy": 0.09305147007107735,
      "num_tokens": 1048590.0,
      "step": 605
    },
    {
      "entropy": 7.1161217212677,
      "epoch": 0.04746158334954289,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003045,
      "loss": 6.8817,
      "mean_token_accuracy": 0.09100724905729293,
      "num_tokens": 1057232.0,
      "step": 610
    },
    {
      "entropy": 7.188746976852417,
      "epoch": 0.047850612721260455,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000307,
      "loss": 6.9444,
      "mean_token_accuracy": 0.08997739925980568,
      "num_tokens": 1066404.0,
      "step": 615
    },
    {
      "entropy": 7.160967588424683,
      "epoch": 0.04823964209297802,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003095,
      "loss": 7.0056,
      "mean_token_accuracy": 0.08781350031495094,
      "num_tokens": 1075093.0,
      "step": 620
    },
    {
      "entropy": 7.1323071956634525,
      "epoch": 0.048628671464695586,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000312,
      "loss": 6.9276,
      "mean_token_accuracy": 0.09003649577498436,
      "num_tokens": 1084454.0,
      "step": 625
    },
    {
      "entropy": 7.157799243927002,
      "epoch": 0.04901770083641315,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003145,
      "loss": 6.8467,
      "mean_token_accuracy": 0.09403926953673362,
      "num_tokens": 1093452.0,
      "step": 630
    },
    {
      "entropy": 7.186054420471192,
      "epoch": 0.04940673020813072,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000317,
      "loss": 6.8691,
      "mean_token_accuracy": 0.09542709589004517,
      "num_tokens": 1101814.0,
      "step": 635
    },
    {
      "entropy": 7.078825283050537,
      "epoch": 0.049795759579848276,
      "grad_norm": 1.125,
      "learning_rate": 0.0003195,
      "loss": 6.9504,
      "mean_token_accuracy": 0.09376255795359612,
      "num_tokens": 1110386.0,
      "step": 640
    },
    {
      "entropy": 7.1958681583404545,
      "epoch": 0.05018478895156584,
      "grad_norm": 1.078125,
      "learning_rate": 0.000322,
      "loss": 7.0501,
      "mean_token_accuracy": 0.08292681574821473,
      "num_tokens": 1120141.0,
      "step": 645
    },
    {
      "entropy": 7.147344017028809,
      "epoch": 0.05057381832328341,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.9687,
      "mean_token_accuracy": 0.09161446392536163,
      "num_tokens": 1129920.0,
      "step": 650
    },
    {
      "entropy": 7.015273141860962,
      "epoch": 0.05096284769500097,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.8553,
      "mean_token_accuracy": 0.09384810253977775,
      "num_tokens": 1138518.0,
      "step": 655
    },
    {
      "entropy": 7.139459609985352,
      "epoch": 0.05135187706671854,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.9721,
      "mean_token_accuracy": 0.08497107923030853,
      "num_tokens": 1147113.0,
      "step": 660
    },
    {
      "entropy": 7.057343339920044,
      "epoch": 0.0517409064384361,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.8095,
      "mean_token_accuracy": 0.09642426520586014,
      "num_tokens": 1155886.0,
      "step": 665
    },
    {
      "entropy": 7.103746461868286,
      "epoch": 0.05212993581015367,
      "grad_norm": 1.25,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.9542,
      "mean_token_accuracy": 0.09668822288513183,
      "num_tokens": 1164200.0,
      "step": 670
    },
    {
      "entropy": 7.031191968917847,
      "epoch": 0.052518965181871234,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000337,
      "loss": 6.8286,
      "mean_token_accuracy": 0.09215016961097718,
      "num_tokens": 1172720.0,
      "step": 675
    },
    {
      "entropy": 7.015228414535523,
      "epoch": 0.05290799455358879,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003395,
      "loss": 6.7712,
      "mean_token_accuracy": 0.09318645074963569,
      "num_tokens": 1181225.0,
      "step": 680
    },
    {
      "entropy": 7.15850133895874,
      "epoch": 0.05329702392530636,
      "grad_norm": 1.125,
      "learning_rate": 0.000342,
      "loss": 6.9747,
      "mean_token_accuracy": 0.09126665517687797,
      "num_tokens": 1190070.0,
      "step": 685
    },
    {
      "entropy": 7.017357587814331,
      "epoch": 0.053686053297023924,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.8557,
      "mean_token_accuracy": 0.09437393769621849,
      "num_tokens": 1198995.0,
      "step": 690
    },
    {
      "entropy": 7.108525228500366,
      "epoch": 0.05407508266874149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000347,
      "loss": 6.9139,
      "mean_token_accuracy": 0.09893580600619316,
      "num_tokens": 1207664.0,
      "step": 695
    },
    {
      "entropy": 7.1310450553894045,
      "epoch": 0.054464112040459055,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003495,
      "loss": 6.8008,
      "mean_token_accuracy": 0.10891369134187698,
      "num_tokens": 1216232.0,
      "step": 700
    },
    {
      "entropy": 6.994553136825561,
      "epoch": 0.05485314141217662,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000352,
      "loss": 6.884,
      "mean_token_accuracy": 0.09593645855784416,
      "num_tokens": 1224106.0,
      "step": 705
    },
    {
      "entropy": 6.933602619171142,
      "epoch": 0.055242170783894186,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003545,
      "loss": 6.7804,
      "mean_token_accuracy": 0.10029099360108376,
      "num_tokens": 1232514.0,
      "step": 710
    },
    {
      "entropy": 7.105175018310547,
      "epoch": 0.05563120015561175,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000357,
      "loss": 6.8244,
      "mean_token_accuracy": 0.10292564630508423,
      "num_tokens": 1240178.0,
      "step": 715
    },
    {
      "entropy": 6.9011462211608885,
      "epoch": 0.05602022952732931,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003595,
      "loss": 6.75,
      "mean_token_accuracy": 0.1004354253411293,
      "num_tokens": 1248635.0,
      "step": 720
    },
    {
      "entropy": 7.058794069290161,
      "epoch": 0.056409258899046875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000362,
      "loss": 6.8454,
      "mean_token_accuracy": 0.10295553430914879,
      "num_tokens": 1256389.0,
      "step": 725
    },
    {
      "entropy": 6.923951578140259,
      "epoch": 0.05679828827076444,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003645,
      "loss": 6.8013,
      "mean_token_accuracy": 0.09758390784263611,
      "num_tokens": 1264035.0,
      "step": 730
    },
    {
      "entropy": 7.092983961105347,
      "epoch": 0.057187317642482006,
      "grad_norm": 1.234375,
      "learning_rate": 0.000367,
      "loss": 6.8054,
      "mean_token_accuracy": 0.09857210367918015,
      "num_tokens": 1272941.0,
      "step": 735
    },
    {
      "entropy": 6.776862668991089,
      "epoch": 0.05757634701419957,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003695,
      "loss": 6.7916,
      "mean_token_accuracy": 0.10023090690374374,
      "num_tokens": 1281767.0,
      "step": 740
    },
    {
      "entropy": 7.078878498077392,
      "epoch": 0.05796537638591714,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000372,
      "loss": 6.8725,
      "mean_token_accuracy": 0.1040918543934822,
      "num_tokens": 1290049.0,
      "step": 745
    },
    {
      "entropy": 6.88560905456543,
      "epoch": 0.0583544057576347,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003745,
      "loss": 6.6878,
      "mean_token_accuracy": 0.10409798175096512,
      "num_tokens": 1299032.0,
      "step": 750
    },
    {
      "entropy": 6.961802768707275,
      "epoch": 0.05874343512935227,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000377,
      "loss": 6.9603,
      "mean_token_accuracy": 0.09438887834548951,
      "num_tokens": 1308508.0,
      "step": 755
    },
    {
      "entropy": 7.064554643630982,
      "epoch": 0.059132464501069834,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003795,
      "loss": 6.7931,
      "mean_token_accuracy": 0.10270294025540352,
      "num_tokens": 1317044.0,
      "step": 760
    },
    {
      "entropy": 6.964282894134522,
      "epoch": 0.05952149387278739,
      "grad_norm": 1.015625,
      "learning_rate": 0.000382,
      "loss": 6.809,
      "mean_token_accuracy": 0.10116439014673233,
      "num_tokens": 1325554.0,
      "step": 765
    },
    {
      "entropy": 6.90094223022461,
      "epoch": 0.05991052324450496,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003845,
      "loss": 6.6974,
      "mean_token_accuracy": 0.09834365025162697,
      "num_tokens": 1334201.0,
      "step": 770
    },
    {
      "entropy": 6.887777614593506,
      "epoch": 0.060299552616222524,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.7209,
      "mean_token_accuracy": 0.10774503573775292,
      "num_tokens": 1343061.0,
      "step": 775
    },
    {
      "entropy": 6.859195232391357,
      "epoch": 0.06068858198794009,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.6786,
      "mean_token_accuracy": 0.10537986010313034,
      "num_tokens": 1352262.0,
      "step": 780
    },
    {
      "entropy": 6.866818618774414,
      "epoch": 0.061077611359657655,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.7495,
      "mean_token_accuracy": 0.09941672831773758,
      "num_tokens": 1361400.0,
      "step": 785
    },
    {
      "entropy": 7.006886291503906,
      "epoch": 0.06146664073137522,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.8369,
      "mean_token_accuracy": 0.0997631348669529,
      "num_tokens": 1369850.0,
      "step": 790
    },
    {
      "entropy": 6.788526773452759,
      "epoch": 0.061855670103092786,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.6824,
      "mean_token_accuracy": 0.10736593306064605,
      "num_tokens": 1378342.0,
      "step": 795
    },
    {
      "entropy": 6.927976512908936,
      "epoch": 0.06224469947481035,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003995,
      "loss": 6.7924,
      "mean_token_accuracy": 0.09832986444234848,
      "num_tokens": 1386981.0,
      "step": 800
    },
    {
      "entropy": 6.989340591430664,
      "epoch": 0.06263372884652792,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000402,
      "loss": 6.8623,
      "mean_token_accuracy": 0.09800197407603264,
      "num_tokens": 1395801.0,
      "step": 805
    },
    {
      "entropy": 6.974379205703736,
      "epoch": 0.06302275821824548,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004045,
      "loss": 6.7523,
      "mean_token_accuracy": 0.10530616566538811,
      "num_tokens": 1403715.0,
      "step": 810
    },
    {
      "entropy": 7.0467917919158936,
      "epoch": 0.06341178758996305,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.8484,
      "mean_token_accuracy": 0.09779960736632347,
      "num_tokens": 1413199.0,
      "step": 815
    },
    {
      "entropy": 6.8538792610168455,
      "epoch": 0.0638008169616806,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004095,
      "loss": 6.8357,
      "mean_token_accuracy": 0.10221734568476677,
      "num_tokens": 1421918.0,
      "step": 820
    },
    {
      "entropy": 6.861208772659301,
      "epoch": 0.06418984633339817,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000412,
      "loss": 6.713,
      "mean_token_accuracy": 0.099763373285532,
      "num_tokens": 1430625.0,
      "step": 825
    },
    {
      "entropy": 6.8965160846710205,
      "epoch": 0.06457887570511574,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004145,
      "loss": 6.7463,
      "mean_token_accuracy": 0.101302270591259,
      "num_tokens": 1439158.0,
      "step": 830
    },
    {
      "entropy": 6.761336946487427,
      "epoch": 0.0649679050768333,
      "grad_norm": 1.171875,
      "learning_rate": 0.000417,
      "loss": 6.7006,
      "mean_token_accuracy": 0.10375607237219811,
      "num_tokens": 1448266.0,
      "step": 835
    },
    {
      "entropy": 6.932992315292358,
      "epoch": 0.06535693444855087,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004195,
      "loss": 6.7856,
      "mean_token_accuracy": 0.10553251951932907,
      "num_tokens": 1457322.0,
      "step": 840
    },
    {
      "entropy": 6.940921783447266,
      "epoch": 0.06574596382026843,
      "grad_norm": 1.09375,
      "learning_rate": 0.000422,
      "loss": 6.905,
      "mean_token_accuracy": 0.09642096757888793,
      "num_tokens": 1466120.0,
      "step": 845
    },
    {
      "entropy": 6.877140855789184,
      "epoch": 0.066134993191986,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004245,
      "loss": 6.6655,
      "mean_token_accuracy": 0.10028111264109611,
      "num_tokens": 1474965.0,
      "step": 850
    },
    {
      "entropy": 6.782597208023072,
      "epoch": 0.06652402256370356,
      "grad_norm": 1.265625,
      "learning_rate": 0.000427,
      "loss": 6.7623,
      "mean_token_accuracy": 0.10180079340934753,
      "num_tokens": 1483483.0,
      "step": 855
    },
    {
      "entropy": 6.930035066604614,
      "epoch": 0.06691305193542113,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004295,
      "loss": 6.6985,
      "mean_token_accuracy": 0.0988282911479473,
      "num_tokens": 1492313.0,
      "step": 860
    },
    {
      "entropy": 6.802086067199707,
      "epoch": 0.06730208130713869,
      "grad_norm": 1.078125,
      "learning_rate": 0.000432,
      "loss": 6.6205,
      "mean_token_accuracy": 0.10422073453664779,
      "num_tokens": 1501081.0,
      "step": 865
    },
    {
      "entropy": 6.703395795822144,
      "epoch": 0.06769111067885625,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004345,
      "loss": 6.6475,
      "mean_token_accuracy": 0.1006127305328846,
      "num_tokens": 1510169.0,
      "step": 870
    },
    {
      "entropy": 6.842187690734863,
      "epoch": 0.06808014005057382,
      "grad_norm": 1.140625,
      "learning_rate": 0.000437,
      "loss": 6.5884,
      "mean_token_accuracy": 0.10198393240571021,
      "num_tokens": 1519546.0,
      "step": 875
    },
    {
      "entropy": 6.701787328720092,
      "epoch": 0.06846916942229138,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004395,
      "loss": 6.6955,
      "mean_token_accuracy": 0.09849147126078606,
      "num_tokens": 1529104.0,
      "step": 880
    },
    {
      "entropy": 6.727069425582886,
      "epoch": 0.06885819879400895,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000442,
      "loss": 6.653,
      "mean_token_accuracy": 0.11421328037977219,
      "num_tokens": 1537033.0,
      "step": 885
    },
    {
      "entropy": 6.807587718963623,
      "epoch": 0.06924722816572651,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004445,
      "loss": 6.6796,
      "mean_token_accuracy": 0.10756831169128418,
      "num_tokens": 1545059.0,
      "step": 890
    },
    {
      "entropy": 6.851893138885498,
      "epoch": 0.06963625753744408,
      "grad_norm": 1.125,
      "learning_rate": 0.000447,
      "loss": 6.6941,
      "mean_token_accuracy": 0.10561164915561676,
      "num_tokens": 1553784.0,
      "step": 895
    },
    {
      "entropy": 6.693432712554932,
      "epoch": 0.07002528690916164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.6167,
      "mean_token_accuracy": 0.10881831496953964,
      "num_tokens": 1562059.0,
      "step": 900
    },
    {
      "entropy": 6.777541732788086,
      "epoch": 0.0704143162808792,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.6848,
      "mean_token_accuracy": 0.10598696693778038,
      "num_tokens": 1570061.0,
      "step": 905
    },
    {
      "entropy": 6.781379699707031,
      "epoch": 0.07080334565259677,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.6588,
      "mean_token_accuracy": 0.10451298281550407,
      "num_tokens": 1578261.0,
      "step": 910
    },
    {
      "entropy": 6.708835935592651,
      "epoch": 0.07119237502431433,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.5728,
      "mean_token_accuracy": 0.108784069865942,
      "num_tokens": 1586260.0,
      "step": 915
    },
    {
      "entropy": 6.839280271530152,
      "epoch": 0.0715814043960319,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.7433,
      "mean_token_accuracy": 0.10013837367296219,
      "num_tokens": 1595234.0,
      "step": 920
    },
    {
      "entropy": 6.70526909828186,
      "epoch": 0.07197043376774946,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000462,
      "loss": 6.6588,
      "mean_token_accuracy": 0.10594909787178039,
      "num_tokens": 1604212.0,
      "step": 925
    },
    {
      "entropy": 6.769180774688721,
      "epoch": 0.07235946313946703,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004645,
      "loss": 6.6854,
      "mean_token_accuracy": 0.10409966185688972,
      "num_tokens": 1612925.0,
      "step": 930
    },
    {
      "entropy": 6.6779365062713625,
      "epoch": 0.07274849251118459,
      "grad_norm": 1.15625,
      "learning_rate": 0.000467,
      "loss": 6.5829,
      "mean_token_accuracy": 0.10564278215169906,
      "num_tokens": 1621804.0,
      "step": 935
    },
    {
      "entropy": 6.673547887802124,
      "epoch": 0.07313752188290217,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004695,
      "loss": 6.5768,
      "mean_token_accuracy": 0.10946961864829063,
      "num_tokens": 1630943.0,
      "step": 940
    },
    {
      "entropy": 6.686461353302002,
      "epoch": 0.07352655125461972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000472,
      "loss": 6.479,
      "mean_token_accuracy": 0.109093077480793,
      "num_tokens": 1638899.0,
      "step": 945
    },
    {
      "entropy": 6.665160274505615,
      "epoch": 0.07391558062633728,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004745,
      "loss": 6.6837,
      "mean_token_accuracy": 0.10972491055727004,
      "num_tokens": 1647485.0,
      "step": 950
    },
    {
      "entropy": 6.800267505645752,
      "epoch": 0.07430460999805485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000477,
      "loss": 6.6331,
      "mean_token_accuracy": 0.11150100603699684,
      "num_tokens": 1656290.0,
      "step": 955
    },
    {
      "entropy": 6.686690902709961,
      "epoch": 0.07469363936977241,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004795,
      "loss": 6.673,
      "mean_token_accuracy": 0.11151936575770378,
      "num_tokens": 1665474.0,
      "step": 960
    },
    {
      "entropy": 6.701140642166138,
      "epoch": 0.07508266874148999,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000482,
      "loss": 6.5381,
      "mean_token_accuracy": 0.11096475347876548,
      "num_tokens": 1673599.0,
      "step": 965
    },
    {
      "entropy": 6.659738826751709,
      "epoch": 0.07547169811320754,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004845,
      "loss": 6.53,
      "mean_token_accuracy": 0.1062575913965702,
      "num_tokens": 1683282.0,
      "step": 970
    },
    {
      "entropy": 6.715300607681274,
      "epoch": 0.07586072748492512,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000487,
      "loss": 6.6514,
      "mean_token_accuracy": 0.11183567643165589,
      "num_tokens": 1691094.0,
      "step": 975
    },
    {
      "entropy": 6.577898406982422,
      "epoch": 0.07624975685664268,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004895,
      "loss": 6.506,
      "mean_token_accuracy": 0.11562167555093765,
      "num_tokens": 1700053.0,
      "step": 980
    },
    {
      "entropy": 6.557030916213989,
      "epoch": 0.07663878622836025,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000492,
      "loss": 6.4294,
      "mean_token_accuracy": 0.11256621927022933,
      "num_tokens": 1707732.0,
      "step": 985
    },
    {
      "entropy": 6.711409378051758,
      "epoch": 0.0770278156000778,
      "grad_norm": 1.25,
      "learning_rate": 0.0004945,
      "loss": 6.6956,
      "mean_token_accuracy": 0.106421110779047,
      "num_tokens": 1716406.0,
      "step": 990
    },
    {
      "entropy": 6.768523073196411,
      "epoch": 0.07741684497179536,
      "grad_norm": 1.078125,
      "learning_rate": 0.000497,
      "loss": 6.7278,
      "mean_token_accuracy": 0.10514088124036788,
      "num_tokens": 1725832.0,
      "step": 995
    },
    {
      "entropy": 6.696955347061158,
      "epoch": 0.07780587434351294,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995,
      "loss": 6.6202,
      "mean_token_accuracy": 0.10349752753973007,
      "num_tokens": 1734853.0,
      "step": 1000
    },
    {
      "entropy": 6.6868829250335695,
      "epoch": 0.0781949037152305,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999999989075146,
      "loss": 6.6371,
      "mean_token_accuracy": 0.11080349832773209,
      "num_tokens": 1744240.0,
      "step": 1005
    },
    {
      "entropy": 6.718117666244507,
      "epoch": 0.07858393308694807,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999999944692927,
      "loss": 6.6374,
      "mean_token_accuracy": 0.10706623718142509,
      "num_tokens": 1753334.0,
      "step": 1010
    },
    {
      "entropy": 6.530433320999146,
      "epoch": 0.07897296245866563,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499999986617054,
      "loss": 6.5341,
      "mean_token_accuracy": 0.10991563349962234,
      "num_tokens": 1762032.0,
      "step": 1015
    },
    {
      "entropy": 6.689446020126343,
      "epoch": 0.0793619918303832,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999999753507986,
      "loss": 6.544,
      "mean_token_accuracy": 0.11081757172942161,
      "num_tokens": 1770508.0,
      "step": 1020
    },
    {
      "entropy": 6.560748434066772,
      "epoch": 0.07975102120210076,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999999606705267,
      "loss": 6.426,
      "mean_token_accuracy": 0.11921553984284401,
      "num_tokens": 1780167.0,
      "step": 1025
    },
    {
      "entropy": 6.652446746826172,
      "epoch": 0.08014005057381832,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999999425762385,
      "loss": 6.6517,
      "mean_token_accuracy": 0.10580526143312455,
      "num_tokens": 1788809.0,
      "step": 1030
    },
    {
      "entropy": 6.66084303855896,
      "epoch": 0.08052907994553589,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999999210679344,
      "loss": 6.558,
      "mean_token_accuracy": 0.11268438398838043,
      "num_tokens": 1797445.0,
      "step": 1035
    },
    {
      "entropy": 6.660087299346924,
      "epoch": 0.08091810931725345,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999998961456145,
      "loss": 6.5697,
      "mean_token_accuracy": 0.11011887267231942,
      "num_tokens": 1806488.0,
      "step": 1040
    },
    {
      "entropy": 6.649512672424317,
      "epoch": 0.08130713868897102,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999998678092794,
      "loss": 6.6358,
      "mean_token_accuracy": 0.11133731156587601,
      "num_tokens": 1815692.0,
      "step": 1045
    },
    {
      "entropy": 6.6660400390625,
      "epoch": 0.08169616806068858,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999998360589293,
      "loss": 6.5694,
      "mean_token_accuracy": 0.11109899133443832,
      "num_tokens": 1824599.0,
      "step": 1050
    },
    {
      "entropy": 6.704316091537476,
      "epoch": 0.08208519743240615,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499999800894565,
      "loss": 6.7123,
      "mean_token_accuracy": 0.10897875279188156,
      "num_tokens": 1833411.0,
      "step": 1055
    },
    {
      "entropy": 6.60129189491272,
      "epoch": 0.08247422680412371,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999997623161866,
      "loss": 6.5145,
      "mean_token_accuracy": 0.10535632520914077,
      "num_tokens": 1842177.0,
      "step": 1060
    },
    {
      "entropy": 6.544263887405395,
      "epoch": 0.08286325617584128,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999997203237951,
      "loss": 6.4816,
      "mean_token_accuracy": 0.11858308389782905,
      "num_tokens": 1850176.0,
      "step": 1065
    },
    {
      "entropy": 6.632837915420533,
      "epoch": 0.08325228554755884,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499999674917391,
      "loss": 6.5469,
      "mean_token_accuracy": 0.10982977524399758,
      "num_tokens": 1859464.0,
      "step": 1070
    },
    {
      "entropy": 6.580215501785278,
      "epoch": 0.0836413149192764,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999996260969748,
      "loss": 6.4394,
      "mean_token_accuracy": 0.11119021698832512,
      "num_tokens": 1867733.0,
      "step": 1075
    },
    {
      "entropy": 6.4762852668762205,
      "epoch": 0.08403034429099397,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999995738625475,
      "loss": 6.4287,
      "mean_token_accuracy": 0.11978307366371155,
      "num_tokens": 1876010.0,
      "step": 1080
    },
    {
      "entropy": 6.604893350601197,
      "epoch": 0.08441937366271153,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999995182141099,
      "loss": 6.6016,
      "mean_token_accuracy": 0.11037462204694748,
      "num_tokens": 1884906.0,
      "step": 1085
    },
    {
      "entropy": 6.618666505813598,
      "epoch": 0.0848084030344291,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999994591516626,
      "loss": 6.4995,
      "mean_token_accuracy": 0.11649383753538131,
      "num_tokens": 1893464.0,
      "step": 1090
    },
    {
      "entropy": 6.576593685150146,
      "epoch": 0.08519743240614666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999993966752066,
      "loss": 6.5487,
      "mean_token_accuracy": 0.11343064531683922,
      "num_tokens": 1902769.0,
      "step": 1095
    },
    {
      "entropy": 6.619402933120727,
      "epoch": 0.08558646177786423,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499999330784743,
      "loss": 6.4742,
      "mean_token_accuracy": 0.11150655373930932,
      "num_tokens": 1911968.0,
      "step": 1100
    },
    {
      "entropy": 6.530769681930542,
      "epoch": 0.08597549114958179,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004999992614802725,
      "loss": 6.463,
      "mean_token_accuracy": 0.11413491815328598,
      "num_tokens": 1920063.0,
      "step": 1105
    },
    {
      "entropy": 6.437048149108887,
      "epoch": 0.08636452052129936,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999991887617966,
      "loss": 6.4509,
      "mean_token_accuracy": 0.11341893821954727,
      "num_tokens": 1929192.0,
      "step": 1110
    },
    {
      "entropy": 6.676905298233033,
      "epoch": 0.08675354989301692,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999991126293158,
      "loss": 6.644,
      "mean_token_accuracy": 0.11029372960329056,
      "num_tokens": 1939082.0,
      "step": 1115
    },
    {
      "entropy": 6.419079971313477,
      "epoch": 0.08714257926473448,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999990330828318,
      "loss": 6.504,
      "mean_token_accuracy": 0.11239467188715935,
      "num_tokens": 1948313.0,
      "step": 1120
    },
    {
      "entropy": 6.685780143737793,
      "epoch": 0.08753160863645205,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999989501223456,
      "loss": 6.4964,
      "mean_token_accuracy": 0.11267818436026573,
      "num_tokens": 1957035.0,
      "step": 1125
    },
    {
      "entropy": 6.490918064117432,
      "epoch": 0.08792063800816961,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999988637478584,
      "loss": 6.484,
      "mean_token_accuracy": 0.1103479541838169,
      "num_tokens": 1966041.0,
      "step": 1130
    },
    {
      "entropy": 6.564761972427368,
      "epoch": 0.08830966737988719,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999987739593716,
      "loss": 6.5039,
      "mean_token_accuracy": 0.1088343806564808,
      "num_tokens": 1975019.0,
      "step": 1135
    },
    {
      "entropy": 6.590786552429199,
      "epoch": 0.08869869675160474,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999986807568865,
      "loss": 6.4347,
      "mean_token_accuracy": 0.12605319619178773,
      "num_tokens": 1982930.0,
      "step": 1140
    },
    {
      "entropy": 6.5897198677062985,
      "epoch": 0.08908772612332232,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999985841404045,
      "loss": 6.5418,
      "mean_token_accuracy": 0.11301586031913757,
      "num_tokens": 1991888.0,
      "step": 1145
    },
    {
      "entropy": 6.486571311950684,
      "epoch": 0.08947675549503988,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999984841099271,
      "loss": 6.4041,
      "mean_token_accuracy": 0.11347272992134094,
      "num_tokens": 2000202.0,
      "step": 1150
    },
    {
      "entropy": 6.517371654510498,
      "epoch": 0.08986578486675743,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499998380665456,
      "loss": 6.4403,
      "mean_token_accuracy": 0.11478241458535195,
      "num_tokens": 2009281.0,
      "step": 1155
    },
    {
      "entropy": 6.524582719802856,
      "epoch": 0.090254814238475,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999982738069924,
      "loss": 6.4493,
      "mean_token_accuracy": 0.11818324699997902,
      "num_tokens": 2017806.0,
      "step": 1160
    },
    {
      "entropy": 6.5087179183959964,
      "epoch": 0.09064384361019256,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999981635345382,
      "loss": 6.5419,
      "mean_token_accuracy": 0.11332567036151886,
      "num_tokens": 2027582.0,
      "step": 1165
    },
    {
      "entropy": 6.556246519088745,
      "epoch": 0.09103287298191014,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999980498480949,
      "loss": 6.389,
      "mean_token_accuracy": 0.11757862493395806,
      "num_tokens": 2035427.0,
      "step": 1170
    },
    {
      "entropy": 6.627065515518188,
      "epoch": 0.0914219023536277,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999979327476644,
      "loss": 6.5817,
      "mean_token_accuracy": 0.11226092055439948,
      "num_tokens": 2044603.0,
      "step": 1175
    },
    {
      "entropy": 6.501073837280273,
      "epoch": 0.09181093172534527,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999978122332485,
      "loss": 6.4133,
      "mean_token_accuracy": 0.11716802939772605,
      "num_tokens": 2053272.0,
      "step": 1180
    },
    {
      "entropy": 6.524116849899292,
      "epoch": 0.09219996109706283,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999976883048487,
      "loss": 6.4342,
      "mean_token_accuracy": 0.11659721657633781,
      "num_tokens": 2061677.0,
      "step": 1185
    },
    {
      "entropy": 6.492729616165161,
      "epoch": 0.0925889904687804,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999975609624673,
      "loss": 6.4621,
      "mean_token_accuracy": 0.11502088606357574,
      "num_tokens": 2070282.0,
      "step": 1190
    },
    {
      "entropy": 6.6285412311553955,
      "epoch": 0.09297801984049796,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999974302061059,
      "loss": 6.5401,
      "mean_token_accuracy": 0.11611524373292922,
      "num_tokens": 2078812.0,
      "step": 1195
    },
    {
      "entropy": 6.4900905132293705,
      "epoch": 0.09336704921221552,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999972960357666,
      "loss": 6.432,
      "mean_token_accuracy": 0.10967228636145591,
      "num_tokens": 2087020.0,
      "step": 1200
    },
    {
      "entropy": 6.501610469818115,
      "epoch": 0.09375607858393309,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999971584514516,
      "loss": 6.3804,
      "mean_token_accuracy": 0.11955688148736954,
      "num_tokens": 2095933.0,
      "step": 1205
    },
    {
      "entropy": 6.39524621963501,
      "epoch": 0.09414510795565065,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999970174531627,
      "loss": 6.404,
      "mean_token_accuracy": 0.11746799275279045,
      "num_tokens": 2105218.0,
      "step": 1210
    },
    {
      "entropy": 6.402376556396485,
      "epoch": 0.09453413732736822,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999968730409022,
      "loss": 6.3215,
      "mean_token_accuracy": 0.12029565200209617,
      "num_tokens": 2113808.0,
      "step": 1215
    },
    {
      "entropy": 6.5325843334198,
      "epoch": 0.09492316669908578,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999967252146723,
      "loss": 6.4858,
      "mean_token_accuracy": 0.11363726928830147,
      "num_tokens": 2123014.0,
      "step": 1220
    },
    {
      "entropy": 6.529097700119019,
      "epoch": 0.09531219607080335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999965739744752,
      "loss": 6.4078,
      "mean_token_accuracy": 0.12257340103387833,
      "num_tokens": 2131840.0,
      "step": 1225
    },
    {
      "entropy": 6.576442623138428,
      "epoch": 0.09570122544252091,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999964193203132,
      "loss": 6.5001,
      "mean_token_accuracy": 0.11559791788458824,
      "num_tokens": 2140467.0,
      "step": 1230
    },
    {
      "entropy": 6.446888017654419,
      "epoch": 0.09609025481423847,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999962612521888,
      "loss": 6.4351,
      "mean_token_accuracy": 0.12105798646807671,
      "num_tokens": 2149282.0,
      "step": 1235
    },
    {
      "entropy": 6.533386898040772,
      "epoch": 0.09647928418595604,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999960997701041,
      "loss": 6.4277,
      "mean_token_accuracy": 0.11944963932037353,
      "num_tokens": 2157781.0,
      "step": 1240
    },
    {
      "entropy": 6.454381942749023,
      "epoch": 0.0968683135576736,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999959348740618,
      "loss": 6.4188,
      "mean_token_accuracy": 0.11854116097092629,
      "num_tokens": 2166482.0,
      "step": 1245
    },
    {
      "entropy": 6.428623056411743,
      "epoch": 0.09725734292939117,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999957665640641,
      "loss": 6.4369,
      "mean_token_accuracy": 0.11538697481155395,
      "num_tokens": 2174559.0,
      "step": 1250
    },
    {
      "entropy": 6.477071905136109,
      "epoch": 0.09764637230110873,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499995594840114,
      "loss": 6.3211,
      "mean_token_accuracy": 0.12279326990246772,
      "num_tokens": 2182374.0,
      "step": 1255
    },
    {
      "entropy": 6.449340200424194,
      "epoch": 0.0980354016728263,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999954197022138,
      "loss": 6.4127,
      "mean_token_accuracy": 0.12096217796206474,
      "num_tokens": 2191427.0,
      "step": 1260
    },
    {
      "entropy": 6.572444152832031,
      "epoch": 0.09842443104454386,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999952411503661,
      "loss": 6.4932,
      "mean_token_accuracy": 0.11495474576950074,
      "num_tokens": 2199698.0,
      "step": 1265
    },
    {
      "entropy": 6.452260112762451,
      "epoch": 0.09881346041626143,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999950591845738,
      "loss": 6.3453,
      "mean_token_accuracy": 0.12533242255449295,
      "num_tokens": 2208519.0,
      "step": 1270
    },
    {
      "entropy": 6.402873182296753,
      "epoch": 0.09920248978797899,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999948738048397,
      "loss": 6.3615,
      "mean_token_accuracy": 0.12151595205068588,
      "num_tokens": 2217453.0,
      "step": 1275
    },
    {
      "entropy": 6.4074461460113525,
      "epoch": 0.09959151915969655,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999946850111663,
      "loss": 6.4644,
      "mean_token_accuracy": 0.11735093221068382,
      "num_tokens": 2225623.0,
      "step": 1280
    },
    {
      "entropy": 6.501673412322998,
      "epoch": 0.09998054853141412,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999944928035569,
      "loss": 6.4157,
      "mean_token_accuracy": 0.11887598559260368,
      "num_tokens": 2234519.0,
      "step": 1285
    },
    {
      "entropy": 6.473206901550293,
      "epoch": 0.10036957790313168,
      "grad_norm": 1.328125,
      "learning_rate": 0.000499994297182014,
      "loss": 6.4207,
      "mean_token_accuracy": 0.11706000119447708,
      "num_tokens": 2242994.0,
      "step": 1290
    },
    {
      "entropy": 6.490306758880616,
      "epoch": 0.10075860727484925,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999940981465408,
      "loss": 6.4227,
      "mean_token_accuracy": 0.12089471891522408,
      "num_tokens": 2251687.0,
      "step": 1295
    },
    {
      "entropy": 6.406135559082031,
      "epoch": 0.10114763664656681,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999938956971404,
      "loss": 6.3499,
      "mean_token_accuracy": 0.1211350217461586,
      "num_tokens": 2260751.0,
      "step": 1300
    },
    {
      "entropy": 6.49813494682312,
      "epoch": 0.10153666601828439,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999936898338156,
      "loss": 6.2935,
      "mean_token_accuracy": 0.1254826083779335,
      "num_tokens": 2269125.0,
      "step": 1305
    },
    {
      "entropy": 6.301623916625976,
      "epoch": 0.10192569539000194,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999934805565698,
      "loss": 6.2265,
      "mean_token_accuracy": 0.12722386717796325,
      "num_tokens": 2277098.0,
      "step": 1310
    },
    {
      "entropy": 6.510185956954956,
      "epoch": 0.10231472476171952,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499993267865406,
      "loss": 6.5162,
      "mean_token_accuracy": 0.117816873639822,
      "num_tokens": 2285953.0,
      "step": 1315
    },
    {
      "entropy": 6.437057304382324,
      "epoch": 0.10270375413343708,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999930517603275,
      "loss": 6.4151,
      "mean_token_accuracy": 0.12368793934583663,
      "num_tokens": 2294487.0,
      "step": 1320
    },
    {
      "entropy": 6.528111314773559,
      "epoch": 0.10309278350515463,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999928322413374,
      "loss": 6.4525,
      "mean_token_accuracy": 0.11401506140828133,
      "num_tokens": 2302915.0,
      "step": 1325
    },
    {
      "entropy": 6.398245716094971,
      "epoch": 0.1034818128768722,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999926093084394,
      "loss": 6.4152,
      "mean_token_accuracy": 0.11596530675888062,
      "num_tokens": 2310769.0,
      "step": 1330
    },
    {
      "entropy": 6.374190855026245,
      "epoch": 0.10387084224858976,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999923829616365,
      "loss": 6.2874,
      "mean_token_accuracy": 0.11599803119897842,
      "num_tokens": 2319987.0,
      "step": 1335
    },
    {
      "entropy": 6.437367486953735,
      "epoch": 0.10425987162030734,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999921532009323,
      "loss": 6.4054,
      "mean_token_accuracy": 0.1133360095322132,
      "num_tokens": 2328964.0,
      "step": 1340
    },
    {
      "entropy": 6.47161374092102,
      "epoch": 0.1046489009920249,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999919200263304,
      "loss": 6.3552,
      "mean_token_accuracy": 0.11501636877655982,
      "num_tokens": 2337945.0,
      "step": 1345
    },
    {
      "entropy": 6.456685209274292,
      "epoch": 0.10503793036374247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999916834378342,
      "loss": 6.2612,
      "mean_token_accuracy": 0.11833763271570205,
      "num_tokens": 2345989.0,
      "step": 1350
    },
    {
      "entropy": 6.304389333724975,
      "epoch": 0.10542695973546003,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999914434354472,
      "loss": 6.3725,
      "mean_token_accuracy": 0.12045516222715377,
      "num_tokens": 2353752.0,
      "step": 1355
    },
    {
      "entropy": 6.450414657592773,
      "epoch": 0.10581598910717759,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999912000191733,
      "loss": 6.3546,
      "mean_token_accuracy": 0.11809542030096054,
      "num_tokens": 2361961.0,
      "step": 1360
    },
    {
      "entropy": 6.352401399612427,
      "epoch": 0.10620501847889516,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499990953189016,
      "loss": 6.2651,
      "mean_token_accuracy": 0.12437428683042526,
      "num_tokens": 2370148.0,
      "step": 1365
    },
    {
      "entropy": 6.429590368270874,
      "epoch": 0.10659404785061272,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999907029449791,
      "loss": 6.3158,
      "mean_token_accuracy": 0.125085686147213,
      "num_tokens": 2379105.0,
      "step": 1370
    },
    {
      "entropy": 6.433341217041016,
      "epoch": 0.10698307722233029,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999904492870663,
      "loss": 6.39,
      "mean_token_accuracy": 0.12164573967456818,
      "num_tokens": 2387417.0,
      "step": 1375
    },
    {
      "entropy": 6.49081072807312,
      "epoch": 0.10737210659404785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999901922152816,
      "loss": 6.4107,
      "mean_token_accuracy": 0.12515631392598153,
      "num_tokens": 2396163.0,
      "step": 1380
    },
    {
      "entropy": 6.248341655731201,
      "epoch": 0.10776113596576542,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999899317296288,
      "loss": 6.2564,
      "mean_token_accuracy": 0.12287922278046608,
      "num_tokens": 2404661.0,
      "step": 1385
    },
    {
      "entropy": 6.356365108489991,
      "epoch": 0.10815016533748298,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999896678301121,
      "loss": 6.3524,
      "mean_token_accuracy": 0.11748461276292801,
      "num_tokens": 2413459.0,
      "step": 1390
    },
    {
      "entropy": 6.453977441787719,
      "epoch": 0.10853919470920055,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999894005167352,
      "loss": 6.3129,
      "mean_token_accuracy": 0.12602921575307846,
      "num_tokens": 2421401.0,
      "step": 1395
    },
    {
      "entropy": 6.392508220672608,
      "epoch": 0.10892822408091811,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499989129789502,
      "loss": 6.3072,
      "mean_token_accuracy": 0.12456106916069984,
      "num_tokens": 2429983.0,
      "step": 1400
    },
    {
      "entropy": 6.474859046936035,
      "epoch": 0.10931725345263567,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999888556484172,
      "loss": 6.4411,
      "mean_token_accuracy": 0.11389126777648925,
      "num_tokens": 2439206.0,
      "step": 1405
    },
    {
      "entropy": 6.379395532608032,
      "epoch": 0.10970628282435324,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999885780934845,
      "loss": 6.3657,
      "mean_token_accuracy": 0.12810833156108856,
      "num_tokens": 2448801.0,
      "step": 1410
    },
    {
      "entropy": 6.400861072540283,
      "epoch": 0.1100953121960708,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999882971247081,
      "loss": 6.4086,
      "mean_token_accuracy": 0.12141750380396843,
      "num_tokens": 2458007.0,
      "step": 1415
    },
    {
      "entropy": 6.39118766784668,
      "epoch": 0.11048434156778837,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999880127420926,
      "loss": 6.3,
      "mean_token_accuracy": 0.13179884999990463,
      "num_tokens": 2466574.0,
      "step": 1420
    },
    {
      "entropy": 6.468687915802002,
      "epoch": 0.11087337093950593,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999877249456421,
      "loss": 6.3711,
      "mean_token_accuracy": 0.11950016170740127,
      "num_tokens": 2475263.0,
      "step": 1425
    },
    {
      "entropy": 6.185061025619507,
      "epoch": 0.1112624003112235,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999874337353609,
      "loss": 6.1305,
      "mean_token_accuracy": 0.12724639102816582,
      "num_tokens": 2483738.0,
      "step": 1430
    },
    {
      "entropy": 6.326748466491699,
      "epoch": 0.11165142968294106,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999871391112535,
      "loss": 6.232,
      "mean_token_accuracy": 0.11938041299581528,
      "num_tokens": 2492937.0,
      "step": 1435
    },
    {
      "entropy": 6.304116439819336,
      "epoch": 0.11204045905465862,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999868410733244,
      "loss": 6.3029,
      "mean_token_accuracy": 0.11935702487826347,
      "num_tokens": 2500716.0,
      "step": 1440
    },
    {
      "entropy": 6.360095405578614,
      "epoch": 0.11242948842637619,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499986539621578,
      "loss": 6.3182,
      "mean_token_accuracy": 0.11632756739854813,
      "num_tokens": 2509385.0,
      "step": 1445
    },
    {
      "entropy": 6.516259574890137,
      "epoch": 0.11281851779809375,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999862347560191,
      "loss": 6.3697,
      "mean_token_accuracy": 0.11724613457918168,
      "num_tokens": 2518179.0,
      "step": 1450
    },
    {
      "entropy": 6.297982025146484,
      "epoch": 0.11320754716981132,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999859264766521,
      "loss": 6.33,
      "mean_token_accuracy": 0.12113318666815757,
      "num_tokens": 2526644.0,
      "step": 1455
    },
    {
      "entropy": 6.390083694458008,
      "epoch": 0.11359657654152888,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999856147834817,
      "loss": 6.2294,
      "mean_token_accuracy": 0.12677537202835082,
      "num_tokens": 2534998.0,
      "step": 1460
    },
    {
      "entropy": 6.299189186096191,
      "epoch": 0.11398560591324645,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999852996765129,
      "loss": 6.3009,
      "mean_token_accuracy": 0.12109575122594833,
      "num_tokens": 2543562.0,
      "step": 1465
    },
    {
      "entropy": 6.413508653640747,
      "epoch": 0.11437463528496401,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999849811557502,
      "loss": 6.3233,
      "mean_token_accuracy": 0.11768989339470863,
      "num_tokens": 2551688.0,
      "step": 1470
    },
    {
      "entropy": 6.353024053573608,
      "epoch": 0.11476366465668159,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999846592211984,
      "loss": 6.3672,
      "mean_token_accuracy": 0.12517470344901085,
      "num_tokens": 2560227.0,
      "step": 1475
    },
    {
      "entropy": 6.358402109146118,
      "epoch": 0.11515269402839914,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999843338728626,
      "loss": 6.2573,
      "mean_token_accuracy": 0.12913090363144875,
      "num_tokens": 2569245.0,
      "step": 1480
    },
    {
      "entropy": 6.319776582717895,
      "epoch": 0.1155417234001167,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999840051107476,
      "loss": 6.2771,
      "mean_token_accuracy": 0.12657519578933715,
      "num_tokens": 2577706.0,
      "step": 1485
    },
    {
      "entropy": 6.3895636081695555,
      "epoch": 0.11593075277183428,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999836729348584,
      "loss": 6.3161,
      "mean_token_accuracy": 0.12418374568223953,
      "num_tokens": 2586075.0,
      "step": 1490
    },
    {
      "entropy": 6.328500509262085,
      "epoch": 0.11631978214355183,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999833373452,
      "loss": 6.3166,
      "mean_token_accuracy": 0.11702506840229035,
      "num_tokens": 2594965.0,
      "step": 1495
    },
    {
      "entropy": 6.343495893478393,
      "epoch": 0.1167088115152694,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999829983417778,
      "loss": 6.2299,
      "mean_token_accuracy": 0.12473512664437295,
      "num_tokens": 2603355.0,
      "step": 1500
    },
    {
      "entropy": 6.428485584259033,
      "epoch": 0.11709784088698696,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999826559245965,
      "loss": 6.3809,
      "mean_token_accuracy": 0.11869822889566421,
      "num_tokens": 2611697.0,
      "step": 1505
    },
    {
      "entropy": 6.2671410083770756,
      "epoch": 0.11748687025870454,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999823100936615,
      "loss": 6.2445,
      "mean_token_accuracy": 0.12823037803173065,
      "num_tokens": 2620821.0,
      "step": 1510
    },
    {
      "entropy": 6.389608240127563,
      "epoch": 0.1178758996304221,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999819608489781,
      "loss": 6.2504,
      "mean_token_accuracy": 0.1255895458161831,
      "num_tokens": 2628891.0,
      "step": 1515
    },
    {
      "entropy": 6.372634601593018,
      "epoch": 0.11826492900213967,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999816081905515,
      "loss": 6.371,
      "mean_token_accuracy": 0.12320602610707283,
      "num_tokens": 2638641.0,
      "step": 1520
    },
    {
      "entropy": 6.3987085819244385,
      "epoch": 0.11865395837385723,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999812521183872,
      "loss": 6.264,
      "mean_token_accuracy": 0.12482255175709725,
      "num_tokens": 2646898.0,
      "step": 1525
    },
    {
      "entropy": 6.377302169799805,
      "epoch": 0.11904298774557479,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999808926324903,
      "loss": 6.3393,
      "mean_token_accuracy": 0.11689784750342369,
      "num_tokens": 2655404.0,
      "step": 1530
    },
    {
      "entropy": 6.296369886398315,
      "epoch": 0.11943201711729236,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999805297328665,
      "loss": 6.2017,
      "mean_token_accuracy": 0.1269844152033329,
      "num_tokens": 2663851.0,
      "step": 1535
    },
    {
      "entropy": 6.27218918800354,
      "epoch": 0.11982104648900992,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999801634195214,
      "loss": 6.2599,
      "mean_token_accuracy": 0.124675901979208,
      "num_tokens": 2672231.0,
      "step": 1540
    },
    {
      "entropy": 6.31765775680542,
      "epoch": 0.12021007586072749,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999797936924603,
      "loss": 6.2452,
      "mean_token_accuracy": 0.1260676108300686,
      "num_tokens": 2680198.0,
      "step": 1545
    },
    {
      "entropy": 6.392157125473022,
      "epoch": 0.12059910523244505,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499979420551689,
      "loss": 6.2862,
      "mean_token_accuracy": 0.11980032101273537,
      "num_tokens": 2689208.0,
      "step": 1550
    },
    {
      "entropy": 6.420441484451294,
      "epoch": 0.12098813460416262,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499979043997213,
      "loss": 6.3474,
      "mean_token_accuracy": 0.11860086098313331,
      "num_tokens": 2698116.0,
      "step": 1555
    },
    {
      "entropy": 6.286598205566406,
      "epoch": 0.12137716397588018,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999786640290381,
      "loss": 6.2261,
      "mean_token_accuracy": 0.11801384165883064,
      "num_tokens": 2707046.0,
      "step": 1560
    },
    {
      "entropy": 6.2947837829589846,
      "epoch": 0.12176619334759774,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999782806471701,
      "loss": 6.238,
      "mean_token_accuracy": 0.12088324874639511,
      "num_tokens": 2714929.0,
      "step": 1565
    },
    {
      "entropy": 6.297727441787719,
      "epoch": 0.12215522271931531,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999778938516147,
      "loss": 6.2441,
      "mean_token_accuracy": 0.12293168380856515,
      "num_tokens": 2723436.0,
      "step": 1570
    },
    {
      "entropy": 6.305335235595703,
      "epoch": 0.12254425209103287,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499977503642378,
      "loss": 6.3148,
      "mean_token_accuracy": 0.12060637921094894,
      "num_tokens": 2731935.0,
      "step": 1575
    },
    {
      "entropy": 6.29901385307312,
      "epoch": 0.12293328146275044,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999771100194656,
      "loss": 6.142,
      "mean_token_accuracy": 0.13157675340771674,
      "num_tokens": 2740307.0,
      "step": 1580
    },
    {
      "entropy": 6.158562135696411,
      "epoch": 0.123322310834468,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999767129828837,
      "loss": 6.1256,
      "mean_token_accuracy": 0.1322464384138584,
      "num_tokens": 2749077.0,
      "step": 1585
    },
    {
      "entropy": 6.349336957931518,
      "epoch": 0.12371134020618557,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999763125326383,
      "loss": 6.2515,
      "mean_token_accuracy": 0.12292205169796944,
      "num_tokens": 2758133.0,
      "step": 1590
    },
    {
      "entropy": 6.304586410522461,
      "epoch": 0.12410036957790313,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999759086687355,
      "loss": 6.183,
      "mean_token_accuracy": 0.12989940419793128,
      "num_tokens": 2766509.0,
      "step": 1595
    },
    {
      "entropy": 6.194839763641357,
      "epoch": 0.1244893989496207,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999755013911813,
      "loss": 6.3616,
      "mean_token_accuracy": 0.12518828511238098,
      "num_tokens": 2775908.0,
      "step": 1600
    },
    {
      "entropy": 6.344752550125122,
      "epoch": 0.12487842832133826,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499975090699982,
      "loss": 6.121,
      "mean_token_accuracy": 0.13455261215567588,
      "num_tokens": 2784803.0,
      "step": 1605
    },
    {
      "entropy": 6.332213401794434,
      "epoch": 0.12526745769305583,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999746765951438,
      "loss": 6.2889,
      "mean_token_accuracy": 0.12206416502594948,
      "num_tokens": 2793897.0,
      "step": 1610
    },
    {
      "entropy": 6.274225473403931,
      "epoch": 0.1256564870647734,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999742590766729,
      "loss": 6.122,
      "mean_token_accuracy": 0.12973754331469536,
      "num_tokens": 2802652.0,
      "step": 1615
    },
    {
      "entropy": 6.226436805725098,
      "epoch": 0.12604551643649095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999738381445757,
      "loss": 6.2668,
      "mean_token_accuracy": 0.12090609893202782,
      "num_tokens": 2811527.0,
      "step": 1620
    },
    {
      "entropy": 6.277534484863281,
      "epoch": 0.1264345458082085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999734137988586,
      "loss": 6.1126,
      "mean_token_accuracy": 0.1338588200509548,
      "num_tokens": 2819410.0,
      "step": 1625
    },
    {
      "entropy": 6.044222688674926,
      "epoch": 0.1268235751799261,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499972986039528,
      "loss": 6.12,
      "mean_token_accuracy": 0.13493081852793692,
      "num_tokens": 2828079.0,
      "step": 1630
    },
    {
      "entropy": 6.299578428268433,
      "epoch": 0.12721260455164365,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999725548665904,
      "loss": 6.2427,
      "mean_token_accuracy": 0.1260551355779171,
      "num_tokens": 2836326.0,
      "step": 1635
    },
    {
      "entropy": 6.397226285934448,
      "epoch": 0.1276016339233612,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999721202800524,
      "loss": 6.28,
      "mean_token_accuracy": 0.12439439222216606,
      "num_tokens": 2845271.0,
      "step": 1640
    },
    {
      "entropy": 6.166529750823974,
      "epoch": 0.12799066329507877,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999716822799206,
      "loss": 6.1479,
      "mean_token_accuracy": 0.13345591425895692,
      "num_tokens": 2853688.0,
      "step": 1645
    },
    {
      "entropy": 6.2988574504852295,
      "epoch": 0.12837969266679633,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999712408662016,
      "loss": 6.2329,
      "mean_token_accuracy": 0.13125427886843682,
      "num_tokens": 2862261.0,
      "step": 1650
    },
    {
      "entropy": 6.290777921676636,
      "epoch": 0.12876872203851392,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999707960389021,
      "loss": 6.2441,
      "mean_token_accuracy": 0.12467502355575562,
      "num_tokens": 2871324.0,
      "step": 1655
    },
    {
      "entropy": 6.363094997406006,
      "epoch": 0.12915775141023147,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999703477980288,
      "loss": 6.3015,
      "mean_token_accuracy": 0.11913043558597565,
      "num_tokens": 2880268.0,
      "step": 1660
    },
    {
      "entropy": 6.2725090980529785,
      "epoch": 0.12954678078194903,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999698961435885,
      "loss": 6.2154,
      "mean_token_accuracy": 0.12513224706053733,
      "num_tokens": 2888510.0,
      "step": 1665
    },
    {
      "entropy": 6.255806159973145,
      "epoch": 0.1299358101536666,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999694410755882,
      "loss": 6.1155,
      "mean_token_accuracy": 0.134560976177454,
      "num_tokens": 2897616.0,
      "step": 1670
    },
    {
      "entropy": 6.2078857421875,
      "epoch": 0.13032483952538418,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999689825940347,
      "loss": 6.2119,
      "mean_token_accuracy": 0.13180608600378035,
      "num_tokens": 2905703.0,
      "step": 1675
    },
    {
      "entropy": 6.257840871810913,
      "epoch": 0.13071386889710174,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999685206989349,
      "loss": 6.0924,
      "mean_token_accuracy": 0.13349219858646394,
      "num_tokens": 2914881.0,
      "step": 1680
    },
    {
      "entropy": 6.152332448959351,
      "epoch": 0.1311028982688193,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999680553902959,
      "loss": 6.1496,
      "mean_token_accuracy": 0.12726475298404694,
      "num_tokens": 2923563.0,
      "step": 1685
    },
    {
      "entropy": 6.314036417007446,
      "epoch": 0.13149192764053685,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999675866681247,
      "loss": 6.3568,
      "mean_token_accuracy": 0.12228120416402817,
      "num_tokens": 2932821.0,
      "step": 1690
    },
    {
      "entropy": 6.361423540115356,
      "epoch": 0.1318809570122544,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999671145324286,
      "loss": 6.2519,
      "mean_token_accuracy": 0.12628840357065202,
      "num_tokens": 2940926.0,
      "step": 1695
    },
    {
      "entropy": 6.251637029647827,
      "epoch": 0.132269986383972,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999666389832144,
      "loss": 6.175,
      "mean_token_accuracy": 0.12774975672364236,
      "num_tokens": 2950028.0,
      "step": 1700
    },
    {
      "entropy": 6.2219006538391115,
      "epoch": 0.13265901575568956,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999661600204897,
      "loss": 6.1962,
      "mean_token_accuracy": 0.1321553736925125,
      "num_tokens": 2957636.0,
      "step": 1705
    },
    {
      "entropy": 6.136737251281739,
      "epoch": 0.13304804512740712,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999656776442615,
      "loss": 5.9691,
      "mean_token_accuracy": 0.1330525793135166,
      "num_tokens": 2966544.0,
      "step": 1710
    },
    {
      "entropy": 6.201963996887207,
      "epoch": 0.13343707449912467,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999651918545372,
      "loss": 6.2068,
      "mean_token_accuracy": 0.12606080621480942,
      "num_tokens": 2976236.0,
      "step": 1715
    },
    {
      "entropy": 6.247957944869995,
      "epoch": 0.13382610387084226,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999647026513243,
      "loss": 6.1593,
      "mean_token_accuracy": 0.13033938258886338,
      "num_tokens": 2984261.0,
      "step": 1720
    },
    {
      "entropy": 6.17896146774292,
      "epoch": 0.13421513324255982,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049996421003463,
      "loss": 6.1346,
      "mean_token_accuracy": 0.12553048729896546,
      "num_tokens": 2992970.0,
      "step": 1725
    },
    {
      "entropy": 6.274186754226685,
      "epoch": 0.13460416261427738,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999637140044619,
      "loss": 6.1571,
      "mean_token_accuracy": 0.13733121678233146,
      "num_tokens": 3001356.0,
      "step": 1730
    },
    {
      "entropy": 6.217675352096558,
      "epoch": 0.13499319198599494,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999632145608275,
      "loss": 6.1756,
      "mean_token_accuracy": 0.13170652091503143,
      "num_tokens": 3009509.0,
      "step": 1735
    },
    {
      "entropy": 6.264265441894532,
      "epoch": 0.1353822213577125,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999627117037343,
      "loss": 6.1947,
      "mean_token_accuracy": 0.12841238230466842,
      "num_tokens": 3017675.0,
      "step": 1740
    },
    {
      "entropy": 6.189158773422241,
      "epoch": 0.13577125072943008,
      "grad_norm": 1.203125,
      "learning_rate": 0.00049996220543319,
      "loss": 6.1458,
      "mean_token_accuracy": 0.13517939075827598,
      "num_tokens": 3025527.0,
      "step": 1745
    },
    {
      "entropy": 6.226872253417969,
      "epoch": 0.13616028010114764,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999616957492025,
      "loss": 6.1199,
      "mean_token_accuracy": 0.12841131687164306,
      "num_tokens": 3034251.0,
      "step": 1750
    },
    {
      "entropy": 6.166691827774048,
      "epoch": 0.1365493094728652,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999611826517793,
      "loss": 6.3415,
      "mean_token_accuracy": 0.12151764705777168,
      "num_tokens": 3043272.0,
      "step": 1755
    },
    {
      "entropy": 6.374958848953247,
      "epoch": 0.13693833884458276,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999606661409281,
      "loss": 6.1882,
      "mean_token_accuracy": 0.1262716494500637,
      "num_tokens": 3051092.0,
      "step": 1760
    },
    {
      "entropy": 6.144443225860596,
      "epoch": 0.13732736821630034,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999601462166569,
      "loss": 6.1727,
      "mean_token_accuracy": 0.12414734214544296,
      "num_tokens": 3060584.0,
      "step": 1765
    },
    {
      "entropy": 6.250931119918823,
      "epoch": 0.1377163975880179,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999596228789736,
      "loss": 6.1511,
      "mean_token_accuracy": 0.13213173300027847,
      "num_tokens": 3069990.0,
      "step": 1770
    },
    {
      "entropy": 6.308782005310059,
      "epoch": 0.13810542695973546,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999590961278859,
      "loss": 6.1321,
      "mean_token_accuracy": 0.12601641267538072,
      "num_tokens": 3078837.0,
      "step": 1775
    },
    {
      "entropy": 6.052882099151612,
      "epoch": 0.13849445633145302,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999585659634021,
      "loss": 6.1353,
      "mean_token_accuracy": 0.13299693092703818,
      "num_tokens": 3087557.0,
      "step": 1780
    },
    {
      "entropy": 6.232021903991699,
      "epoch": 0.13888348570317058,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999580323855301,
      "loss": 6.1211,
      "mean_token_accuracy": 0.12982656732201575,
      "num_tokens": 3096524.0,
      "step": 1785
    },
    {
      "entropy": 6.301290035247803,
      "epoch": 0.13927251507488816,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999574953942781,
      "loss": 6.2168,
      "mean_token_accuracy": 0.13102202489972115,
      "num_tokens": 3105271.0,
      "step": 1790
    },
    {
      "entropy": 6.1726874828338625,
      "epoch": 0.13966154444660572,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999569549896541,
      "loss": 6.1126,
      "mean_token_accuracy": 0.12439190596342087,
      "num_tokens": 3114479.0,
      "step": 1795
    },
    {
      "entropy": 6.19274673461914,
      "epoch": 0.14005057381832328,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999564111716665,
      "loss": 6.1028,
      "mean_token_accuracy": 0.13765334486961364,
      "num_tokens": 3123319.0,
      "step": 1800
    },
    {
      "entropy": 6.169351577758789,
      "epoch": 0.14043960319004084,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999558639403232,
      "loss": 6.1431,
      "mean_token_accuracy": 0.1253248445689678,
      "num_tokens": 3132532.0,
      "step": 1805
    },
    {
      "entropy": 6.320477867126465,
      "epoch": 0.1408286325617584,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999553132956328,
      "loss": 6.2178,
      "mean_token_accuracy": 0.12777107432484627,
      "num_tokens": 3140798.0,
      "step": 1810
    },
    {
      "entropy": 6.137387990951538,
      "epoch": 0.14121766193347599,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999547592376035,
      "loss": 6.1878,
      "mean_token_accuracy": 0.13269276320934295,
      "num_tokens": 3149607.0,
      "step": 1815
    },
    {
      "entropy": 6.1910168647766115,
      "epoch": 0.14160669130519354,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999542017662438,
      "loss": 6.0139,
      "mean_token_accuracy": 0.1368260033428669,
      "num_tokens": 3157537.0,
      "step": 1820
    },
    {
      "entropy": 6.140592384338379,
      "epoch": 0.1419957206769111,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999536408815623,
      "loss": 6.182,
      "mean_token_accuracy": 0.12532128095626832,
      "num_tokens": 3166351.0,
      "step": 1825
    },
    {
      "entropy": 6.161713981628418,
      "epoch": 0.14238475004862866,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999530765835672,
      "loss": 6.0296,
      "mean_token_accuracy": 0.13651844039559363,
      "num_tokens": 3174690.0,
      "step": 1830
    },
    {
      "entropy": 6.136564493179321,
      "epoch": 0.14277377942034625,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999525088722673,
      "loss": 6.1637,
      "mean_token_accuracy": 0.1232847586274147,
      "num_tokens": 3183460.0,
      "step": 1835
    },
    {
      "entropy": 6.285676670074463,
      "epoch": 0.1431628087920638,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999519377476711,
      "loss": 6.1164,
      "mean_token_accuracy": 0.12833876982331277,
      "num_tokens": 3192183.0,
      "step": 1840
    },
    {
      "entropy": 6.160711622238159,
      "epoch": 0.14355183816378136,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999513632097872,
      "loss": 6.2056,
      "mean_token_accuracy": 0.12972862496972085,
      "num_tokens": 3200875.0,
      "step": 1845
    },
    {
      "entropy": 6.196098470687867,
      "epoch": 0.14394086753549892,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999507852586245,
      "loss": 6.0473,
      "mean_token_accuracy": 0.13408101424574853,
      "num_tokens": 3209797.0,
      "step": 1850
    },
    {
      "entropy": 6.224760866165161,
      "epoch": 0.14432989690721648,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999502038941917,
      "loss": 6.1883,
      "mean_token_accuracy": 0.12606970742344856,
      "num_tokens": 3218612.0,
      "step": 1855
    },
    {
      "entropy": 6.182671165466308,
      "epoch": 0.14471892627893407,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999496191164975,
      "loss": 6.1049,
      "mean_token_accuracy": 0.12787139862775804,
      "num_tokens": 3227364.0,
      "step": 1860
    },
    {
      "entropy": 6.14884614944458,
      "epoch": 0.14510795565065163,
      "grad_norm": 1.125,
      "learning_rate": 0.000499949030925551,
      "loss": 6.1123,
      "mean_token_accuracy": 0.12965136021375656,
      "num_tokens": 3236448.0,
      "step": 1865
    },
    {
      "entropy": 6.170213508605957,
      "epoch": 0.14549698502236919,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999484393213609,
      "loss": 6.0578,
      "mean_token_accuracy": 0.13230266496539117,
      "num_tokens": 3244957.0,
      "step": 1870
    },
    {
      "entropy": 6.124844217300415,
      "epoch": 0.14588601439408674,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999478443039363,
      "loss": 6.1312,
      "mean_token_accuracy": 0.12952700182795523,
      "num_tokens": 3253250.0,
      "step": 1875
    },
    {
      "entropy": 6.199846649169922,
      "epoch": 0.14627504376580433,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999472458732862,
      "loss": 6.0482,
      "mean_token_accuracy": 0.13997920900583266,
      "num_tokens": 3261455.0,
      "step": 1880
    },
    {
      "entropy": 6.061277914047241,
      "epoch": 0.1466640731375219,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999466440294198,
      "loss": 6.096,
      "mean_token_accuracy": 0.13030619099736213,
      "num_tokens": 3270108.0,
      "step": 1885
    },
    {
      "entropy": 6.309679841995239,
      "epoch": 0.14705310250923945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499946038772346,
      "loss": 6.2226,
      "mean_token_accuracy": 0.13158586323261262,
      "num_tokens": 3278425.0,
      "step": 1890
    },
    {
      "entropy": 6.191829681396484,
      "epoch": 0.147442131880957,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999454301020741,
      "loss": 6.1462,
      "mean_token_accuracy": 0.12819165959954262,
      "num_tokens": 3288358.0,
      "step": 1895
    },
    {
      "entropy": 6.18556809425354,
      "epoch": 0.14783116125267456,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999448180186133,
      "loss": 6.1303,
      "mean_token_accuracy": 0.12729835957288743,
      "num_tokens": 3297199.0,
      "step": 1900
    },
    {
      "entropy": 6.178153467178345,
      "epoch": 0.14822019062439215,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499944202521973,
      "loss": 6.1506,
      "mean_token_accuracy": 0.12357342168688774,
      "num_tokens": 3306350.0,
      "step": 1905
    },
    {
      "entropy": 6.1408665657043455,
      "epoch": 0.1486092199961097,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999435836121624,
      "loss": 5.9895,
      "mean_token_accuracy": 0.13374814838171006,
      "num_tokens": 3314921.0,
      "step": 1910
    },
    {
      "entropy": 6.011826419830323,
      "epoch": 0.14899824936782727,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499942961289191,
      "loss": 6.1307,
      "mean_token_accuracy": 0.13305978029966353,
      "num_tokens": 3323324.0,
      "step": 1915
    },
    {
      "entropy": 6.364898824691773,
      "epoch": 0.14938727873954483,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999423355530683,
      "loss": 6.2503,
      "mean_token_accuracy": 0.12814974039793015,
      "num_tokens": 3331614.0,
      "step": 1920
    },
    {
      "entropy": 6.1918212890625,
      "epoch": 0.1497763081112624,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999417064038035,
      "loss": 6.0785,
      "mean_token_accuracy": 0.13158948644995688,
      "num_tokens": 3339573.0,
      "step": 1925
    },
    {
      "entropy": 6.158113813400268,
      "epoch": 0.15016533748297997,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999410738414065,
      "loss": 6.0591,
      "mean_token_accuracy": 0.13647124394774438,
      "num_tokens": 3347775.0,
      "step": 1930
    },
    {
      "entropy": 6.0896648406982425,
      "epoch": 0.15055436685469753,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999404378658866,
      "loss": 6.0149,
      "mean_token_accuracy": 0.13345661535859107,
      "num_tokens": 3356947.0,
      "step": 1935
    },
    {
      "entropy": 6.188624572753906,
      "epoch": 0.1509433962264151,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999397984772536,
      "loss": 6.116,
      "mean_token_accuracy": 0.13309542015194892,
      "num_tokens": 3365420.0,
      "step": 1940
    },
    {
      "entropy": 6.1624736309051515,
      "epoch": 0.15133242559813265,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999391556755172,
      "loss": 6.2005,
      "mean_token_accuracy": 0.128976671397686,
      "num_tokens": 3373483.0,
      "step": 1945
    },
    {
      "entropy": 6.264059114456177,
      "epoch": 0.15172145496985023,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999385094606872,
      "loss": 6.1881,
      "mean_token_accuracy": 0.12472899630665779,
      "num_tokens": 3382557.0,
      "step": 1950
    },
    {
      "entropy": 6.211884689331055,
      "epoch": 0.1521104843415678,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999378598327733,
      "loss": 6.1242,
      "mean_token_accuracy": 0.12660180181264877,
      "num_tokens": 3392079.0,
      "step": 1955
    },
    {
      "entropy": 6.159574842453003,
      "epoch": 0.15249951371328535,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999372067917854,
      "loss": 6.1958,
      "mean_token_accuracy": 0.1291268937289715,
      "num_tokens": 3401070.0,
      "step": 1960
    },
    {
      "entropy": 6.1600908756256105,
      "epoch": 0.1528885430850029,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999365503377335,
      "loss": 6.0782,
      "mean_token_accuracy": 0.1298006922006607,
      "num_tokens": 3409921.0,
      "step": 1965
    },
    {
      "entropy": 6.146826696395874,
      "epoch": 0.1532775724567205,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999358904706273,
      "loss": 6.1052,
      "mean_token_accuracy": 0.12948946729302407,
      "num_tokens": 3419149.0,
      "step": 1970
    },
    {
      "entropy": 6.094645309448242,
      "epoch": 0.15366660182843805,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999352271904771,
      "loss": 5.9952,
      "mean_token_accuracy": 0.13366427645087242,
      "num_tokens": 3427435.0,
      "step": 1975
    },
    {
      "entropy": 6.211178398132324,
      "epoch": 0.1540556312001556,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999345604972928,
      "loss": 6.1954,
      "mean_token_accuracy": 0.12369204685091972,
      "num_tokens": 3436377.0,
      "step": 1980
    },
    {
      "entropy": 6.176980257034302,
      "epoch": 0.15444466057187317,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999338903910846,
      "loss": 6.0885,
      "mean_token_accuracy": 0.13029382526874542,
      "num_tokens": 3444997.0,
      "step": 1985
    },
    {
      "entropy": 6.191714239120484,
      "epoch": 0.15483368994359073,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999332168718626,
      "loss": 6.1348,
      "mean_token_accuracy": 0.13196599557995797,
      "num_tokens": 3454254.0,
      "step": 1990
    },
    {
      "entropy": 6.185032701492309,
      "epoch": 0.15522271931530832,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999325399396371,
      "loss": 6.0932,
      "mean_token_accuracy": 0.13326400965452195,
      "num_tokens": 3462853.0,
      "step": 1995
    },
    {
      "entropy": 6.061761045455933,
      "epoch": 0.15561174868702587,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999318595944183,
      "loss": 6.0596,
      "mean_token_accuracy": 0.12793660834431647,
      "num_tokens": 3471288.0,
      "step": 2000
    },
    {
      "entropy": 6.056248712539673,
      "epoch": 0.15600077805874343,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999311758362166,
      "loss": 5.9365,
      "mean_token_accuracy": 0.13281023353338242,
      "num_tokens": 3479479.0,
      "step": 2005
    },
    {
      "entropy": 6.097038793563843,
      "epoch": 0.156389807430461,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999304886650422,
      "loss": 5.9666,
      "mean_token_accuracy": 0.13581353425979614,
      "num_tokens": 3488621.0,
      "step": 2010
    },
    {
      "entropy": 6.058021783828735,
      "epoch": 0.15677883680217858,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999297980809058,
      "loss": 6.0255,
      "mean_token_accuracy": 0.13270611315965652,
      "num_tokens": 3497792.0,
      "step": 2015
    },
    {
      "entropy": 6.070380449295044,
      "epoch": 0.15716786617389614,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999291040838176,
      "loss": 6.04,
      "mean_token_accuracy": 0.1334388241171837,
      "num_tokens": 3506548.0,
      "step": 2020
    },
    {
      "entropy": 6.062567090988159,
      "epoch": 0.1575568955456137,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999284066737884,
      "loss": 6.0488,
      "mean_token_accuracy": 0.1377915136516094,
      "num_tokens": 3515086.0,
      "step": 2025
    },
    {
      "entropy": 6.19374680519104,
      "epoch": 0.15794592491733125,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999277058508285,
      "loss": 6.0715,
      "mean_token_accuracy": 0.12722084745764733,
      "num_tokens": 3523985.0,
      "step": 2030
    },
    {
      "entropy": 6.161553239822387,
      "epoch": 0.1583349542890488,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999270016149486,
      "loss": 6.1462,
      "mean_token_accuracy": 0.13290010020136833,
      "num_tokens": 3532575.0,
      "step": 2035
    },
    {
      "entropy": 6.14690637588501,
      "epoch": 0.1587239836607664,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999262939661596,
      "loss": 6.1174,
      "mean_token_accuracy": 0.13246945217251777,
      "num_tokens": 3541558.0,
      "step": 2040
    },
    {
      "entropy": 6.174297761917114,
      "epoch": 0.15911301303248396,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999255829044721,
      "loss": 6.0417,
      "mean_token_accuracy": 0.1349615588784218,
      "num_tokens": 3549697.0,
      "step": 2045
    },
    {
      "entropy": 6.183842468261719,
      "epoch": 0.15950204240420152,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999248684298968,
      "loss": 6.2193,
      "mean_token_accuracy": 0.12634108141064643,
      "num_tokens": 3558673.0,
      "step": 2050
    },
    {
      "entropy": 6.201715040206909,
      "epoch": 0.15989107177591907,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999241505424447,
      "loss": 6.0597,
      "mean_token_accuracy": 0.1356561675667763,
      "num_tokens": 3567176.0,
      "step": 2055
    },
    {
      "entropy": 6.166228723526001,
      "epoch": 0.16028010114763663,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999234292421265,
      "loss": 6.1923,
      "mean_token_accuracy": 0.1288422577083111,
      "num_tokens": 3576020.0,
      "step": 2060
    },
    {
      "entropy": 6.133847093582153,
      "epoch": 0.16066913051935422,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999227045289535,
      "loss": 6.0154,
      "mean_token_accuracy": 0.13646728843450545,
      "num_tokens": 3584705.0,
      "step": 2065
    },
    {
      "entropy": 6.148935461044312,
      "epoch": 0.16105815989107178,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999219764029363,
      "loss": 6.2193,
      "mean_token_accuracy": 0.12454503774642944,
      "num_tokens": 3593339.0,
      "step": 2070
    },
    {
      "entropy": 6.174338722229004,
      "epoch": 0.16144718926278934,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999212448640861,
      "loss": 5.9522,
      "mean_token_accuracy": 0.14272483959794044,
      "num_tokens": 3601206.0,
      "step": 2075
    },
    {
      "entropy": 5.9944499969482425,
      "epoch": 0.1618362186345069,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999205099124141,
      "loss": 6.049,
      "mean_token_accuracy": 0.13079817742109298,
      "num_tokens": 3610371.0,
      "step": 2080
    },
    {
      "entropy": 6.26822681427002,
      "epoch": 0.16222524800622448,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999197715479313,
      "loss": 6.0661,
      "mean_token_accuracy": 0.12999794706702233,
      "num_tokens": 3618834.0,
      "step": 2085
    },
    {
      "entropy": 5.992048168182373,
      "epoch": 0.16261427737794204,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999190297706489,
      "loss": 6.0897,
      "mean_token_accuracy": 0.13502170965075494,
      "num_tokens": 3627426.0,
      "step": 2090
    },
    {
      "entropy": 6.1902689933776855,
      "epoch": 0.1630033067496596,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999182845805783,
      "loss": 6.11,
      "mean_token_accuracy": 0.13337679505348204,
      "num_tokens": 3636020.0,
      "step": 2095
    },
    {
      "entropy": 6.184276056289673,
      "epoch": 0.16339233612137716,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999175359777306,
      "loss": 6.0313,
      "mean_token_accuracy": 0.13348037749528885,
      "num_tokens": 3644419.0,
      "step": 2100
    },
    {
      "entropy": 6.074879598617554,
      "epoch": 0.16378136549309472,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999167839621174,
      "loss": 6.1447,
      "mean_token_accuracy": 0.12506258115172386,
      "num_tokens": 3653223.0,
      "step": 2105
    },
    {
      "entropy": 6.210933208465576,
      "epoch": 0.1641703948648123,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999160285337501,
      "loss": 6.1709,
      "mean_token_accuracy": 0.13025981783866883,
      "num_tokens": 3662409.0,
      "step": 2110
    },
    {
      "entropy": 6.18992509841919,
      "epoch": 0.16455942423652986,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999152696926399,
      "loss": 6.1217,
      "mean_token_accuracy": 0.1288077488541603,
      "num_tokens": 3670635.0,
      "step": 2115
    },
    {
      "entropy": 6.055357551574707,
      "epoch": 0.16494845360824742,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999145074387985,
      "loss": 6.0277,
      "mean_token_accuracy": 0.1416458807885647,
      "num_tokens": 3679303.0,
      "step": 2120
    },
    {
      "entropy": 6.251597690582275,
      "epoch": 0.16533748297996498,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999137417722374,
      "loss": 6.2109,
      "mean_token_accuracy": 0.12960903346538544,
      "num_tokens": 3689299.0,
      "step": 2125
    },
    {
      "entropy": 6.15298285484314,
      "epoch": 0.16572651235168256,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999129726929684,
      "loss": 6.1215,
      "mean_token_accuracy": 0.13158850744366646,
      "num_tokens": 3697843.0,
      "step": 2130
    },
    {
      "entropy": 6.209942436218261,
      "epoch": 0.16611554172340012,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999122002010029,
      "loss": 6.076,
      "mean_token_accuracy": 0.1307775042951107,
      "num_tokens": 3705876.0,
      "step": 2135
    },
    {
      "entropy": 6.12683482170105,
      "epoch": 0.16650457109511768,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999114242963527,
      "loss": 6.16,
      "mean_token_accuracy": 0.12918348461389542,
      "num_tokens": 3714835.0,
      "step": 2140
    },
    {
      "entropy": 6.12851095199585,
      "epoch": 0.16689360046683524,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999106449790298,
      "loss": 6.0364,
      "mean_token_accuracy": 0.13037344813346863,
      "num_tokens": 3723421.0,
      "step": 2145
    },
    {
      "entropy": 6.1327372074127195,
      "epoch": 0.1672826298385528,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999098622490458,
      "loss": 6.0615,
      "mean_token_accuracy": 0.13242582008242607,
      "num_tokens": 3731590.0,
      "step": 2150
    },
    {
      "entropy": 6.019791984558106,
      "epoch": 0.16767165921027039,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999090761064125,
      "loss": 5.9754,
      "mean_token_accuracy": 0.13794221505522727,
      "num_tokens": 3740700.0,
      "step": 2155
    },
    {
      "entropy": 6.113742160797119,
      "epoch": 0.16806068858198794,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499908286551142,
      "loss": 6.0592,
      "mean_token_accuracy": 0.13774652481079103,
      "num_tokens": 3749516.0,
      "step": 2160
    },
    {
      "entropy": 6.097338151931763,
      "epoch": 0.1684497179537055,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999074935832463,
      "loss": 6.0011,
      "mean_token_accuracy": 0.138454669713974,
      "num_tokens": 3757696.0,
      "step": 2165
    },
    {
      "entropy": 6.080916881561279,
      "epoch": 0.16883874732542306,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999066972027373,
      "loss": 6.0766,
      "mean_token_accuracy": 0.1338729217648506,
      "num_tokens": 3765786.0,
      "step": 2170
    },
    {
      "entropy": 6.08872880935669,
      "epoch": 0.16922777669714065,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999058974096271,
      "loss": 5.9604,
      "mean_token_accuracy": 0.13815502300858498,
      "num_tokens": 3773265.0,
      "step": 2175
    },
    {
      "entropy": 6.091851711273193,
      "epoch": 0.1696168060688582,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499905094203928,
      "loss": 6.0857,
      "mean_token_accuracy": 0.12940960824489595,
      "num_tokens": 3782402.0,
      "step": 2180
    },
    {
      "entropy": 6.111265754699707,
      "epoch": 0.17000583544057576,
      "grad_norm": 1.125,
      "learning_rate": 0.000499904287585652,
      "loss": 6.1536,
      "mean_token_accuracy": 0.12703027427196503,
      "num_tokens": 3791901.0,
      "step": 2185
    },
    {
      "entropy": 6.167774677276611,
      "epoch": 0.17039486481229332,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999034775548113,
      "loss": 5.9948,
      "mean_token_accuracy": 0.1401320591568947,
      "num_tokens": 3799872.0,
      "step": 2190
    },
    {
      "entropy": 6.025020885467529,
      "epoch": 0.17078389418401088,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999026641114185,
      "loss": 5.9304,
      "mean_token_accuracy": 0.14193431213498114,
      "num_tokens": 3808949.0,
      "step": 2195
    },
    {
      "entropy": 6.096840572357178,
      "epoch": 0.17117292355572847,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999018472554857,
      "loss": 6.046,
      "mean_token_accuracy": 0.13580235317349434,
      "num_tokens": 3817429.0,
      "step": 2200
    },
    {
      "entropy": 6.2811767101287845,
      "epoch": 0.17156195292744603,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999010269870253,
      "loss": 6.3187,
      "mean_token_accuracy": 0.12173992693424225,
      "num_tokens": 3827101.0,
      "step": 2205
    },
    {
      "entropy": 6.186968183517456,
      "epoch": 0.17195098229916359,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999002033060498,
      "loss": 6.064,
      "mean_token_accuracy": 0.12950773164629936,
      "num_tokens": 3836147.0,
      "step": 2210
    },
    {
      "entropy": 6.04772801399231,
      "epoch": 0.17234001167088114,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998993762125716,
      "loss": 6.0717,
      "mean_token_accuracy": 0.13448109477758408,
      "num_tokens": 3845300.0,
      "step": 2215
    },
    {
      "entropy": 6.171705484390259,
      "epoch": 0.17272904104259873,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998985457066034,
      "loss": 6.0614,
      "mean_token_accuracy": 0.13789453059434892,
      "num_tokens": 3853620.0,
      "step": 2220
    },
    {
      "entropy": 6.069169330596924,
      "epoch": 0.1731180704143163,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998977117881576,
      "loss": 6.0595,
      "mean_token_accuracy": 0.1349787399172783,
      "num_tokens": 3861734.0,
      "step": 2225
    },
    {
      "entropy": 6.176798486709595,
      "epoch": 0.17350709978603385,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998968744572472,
      "loss": 6.0786,
      "mean_token_accuracy": 0.13317364156246186,
      "num_tokens": 3870753.0,
      "step": 2230
    },
    {
      "entropy": 6.061793375015259,
      "epoch": 0.1738961291577514,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998960337138845,
      "loss": 5.9548,
      "mean_token_accuracy": 0.13736256137490271,
      "num_tokens": 3879382.0,
      "step": 2235
    },
    {
      "entropy": 6.092678499221802,
      "epoch": 0.17428515852946896,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998951895580826,
      "loss": 6.1108,
      "mean_token_accuracy": 0.1346526972949505,
      "num_tokens": 3888417.0,
      "step": 2240
    },
    {
      "entropy": 6.073836374282837,
      "epoch": 0.17467418790118655,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998943419898542,
      "loss": 5.9869,
      "mean_token_accuracy": 0.13438915610313415,
      "num_tokens": 3897083.0,
      "step": 2245
    },
    {
      "entropy": 6.004721164703369,
      "epoch": 0.1750632172729041,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998934910092119,
      "loss": 5.9588,
      "mean_token_accuracy": 0.13743010833859443,
      "num_tokens": 3905564.0,
      "step": 2250
    },
    {
      "entropy": 6.127100706100464,
      "epoch": 0.17545224664462167,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998926366161689,
      "loss": 6.1048,
      "mean_token_accuracy": 0.12591039165854453,
      "num_tokens": 3914896.0,
      "step": 2255
    },
    {
      "entropy": 6.121838665008545,
      "epoch": 0.17584127601633923,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998917788107383,
      "loss": 6.0954,
      "mean_token_accuracy": 0.1338363006711006,
      "num_tokens": 3924151.0,
      "step": 2260
    },
    {
      "entropy": 6.209240961074829,
      "epoch": 0.17623030538805678,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998909175929326,
      "loss": 6.0515,
      "mean_token_accuracy": 0.13761386051774024,
      "num_tokens": 3933163.0,
      "step": 2265
    },
    {
      "entropy": 6.0434469223022464,
      "epoch": 0.17661933475977437,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998900529627653,
      "loss": 5.97,
      "mean_token_accuracy": 0.139447121322155,
      "num_tokens": 3941120.0,
      "step": 2270
    },
    {
      "entropy": 5.9832147598266605,
      "epoch": 0.17700836413149193,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998891849202495,
      "loss": 5.8716,
      "mean_token_accuracy": 0.13407945334911348,
      "num_tokens": 3949896.0,
      "step": 2275
    },
    {
      "entropy": 6.099995136260986,
      "epoch": 0.1773973935032095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499888313465398,
      "loss": 5.9871,
      "mean_token_accuracy": 0.13756617158651352,
      "num_tokens": 3958794.0,
      "step": 2280
    },
    {
      "entropy": 6.031581878662109,
      "epoch": 0.17778642287492705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998874385982247,
      "loss": 6.0827,
      "mean_token_accuracy": 0.13277505114674568,
      "num_tokens": 3967267.0,
      "step": 2285
    },
    {
      "entropy": 6.035991859436035,
      "epoch": 0.17817545224664463,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998865603187421,
      "loss": 5.9947,
      "mean_token_accuracy": 0.13988595828413963,
      "num_tokens": 3976536.0,
      "step": 2290
    },
    {
      "entropy": 6.2207560539245605,
      "epoch": 0.1785644816183622,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499885678626964,
      "loss": 6.0652,
      "mean_token_accuracy": 0.13461703062057495,
      "num_tokens": 3986006.0,
      "step": 2295
    },
    {
      "entropy": 5.9804434299469,
      "epoch": 0.17895351099007975,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998847935229038,
      "loss": 6.0132,
      "mean_token_accuracy": 0.129325582832098,
      "num_tokens": 3994478.0,
      "step": 2300
    },
    {
      "entropy": 6.138697957992553,
      "epoch": 0.1793425403617973,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998839050065746,
      "loss": 6.0353,
      "mean_token_accuracy": 0.13380764052271843,
      "num_tokens": 4003190.0,
      "step": 2305
    },
    {
      "entropy": 6.132336902618408,
      "epoch": 0.17973156973351487,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998830130779902,
      "loss": 5.9841,
      "mean_token_accuracy": 0.13556784838438035,
      "num_tokens": 4011886.0,
      "step": 2310
    },
    {
      "entropy": 5.926994562149048,
      "epoch": 0.18012059910523245,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499882117737164,
      "loss": 5.8736,
      "mean_token_accuracy": 0.13585432842373849,
      "num_tokens": 4020792.0,
      "step": 2315
    },
    {
      "entropy": 5.999753761291504,
      "epoch": 0.18050962847695,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998812189841096,
      "loss": 5.8524,
      "mean_token_accuracy": 0.13412006422877312,
      "num_tokens": 4029235.0,
      "step": 2320
    },
    {
      "entropy": 6.0219508647918705,
      "epoch": 0.18089865784866757,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998803168188407,
      "loss": 6.0882,
      "mean_token_accuracy": 0.13498658463358879,
      "num_tokens": 4037341.0,
      "step": 2325
    },
    {
      "entropy": 6.037039661407471,
      "epoch": 0.18128768722038513,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998794112413708,
      "loss": 5.9816,
      "mean_token_accuracy": 0.1305742658674717,
      "num_tokens": 4046324.0,
      "step": 2330
    },
    {
      "entropy": 6.062641859054565,
      "epoch": 0.18167671659210272,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998785022517137,
      "loss": 6.0358,
      "mean_token_accuracy": 0.13607652485370636,
      "num_tokens": 4054579.0,
      "step": 2335
    },
    {
      "entropy": 6.14380373954773,
      "epoch": 0.18206574596382027,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998775898498835,
      "loss": 5.9767,
      "mean_token_accuracy": 0.13501785844564437,
      "num_tokens": 4063087.0,
      "step": 2340
    },
    {
      "entropy": 6.033124876022339,
      "epoch": 0.18245477533553783,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998766740358936,
      "loss": 6.0707,
      "mean_token_accuracy": 0.1377527207136154,
      "num_tokens": 4072309.0,
      "step": 2345
    },
    {
      "entropy": 6.117186975479126,
      "epoch": 0.1828438047072554,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998757548097582,
      "loss": 6.1177,
      "mean_token_accuracy": 0.13360145166516305,
      "num_tokens": 4081017.0,
      "step": 2350
    },
    {
      "entropy": 6.075238609313965,
      "epoch": 0.18323283407897295,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998748321714911,
      "loss": 6.1195,
      "mean_token_accuracy": 0.12870020866394044,
      "num_tokens": 4089799.0,
      "step": 2355
    },
    {
      "entropy": 6.079360723495483,
      "epoch": 0.18362186345069054,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998739061211065,
      "loss": 6.0327,
      "mean_token_accuracy": 0.1406281881034374,
      "num_tokens": 4097948.0,
      "step": 2360
    },
    {
      "entropy": 6.1009117603302006,
      "epoch": 0.1840108928224081,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998729766586181,
      "loss": 5.9017,
      "mean_token_accuracy": 0.14306575134396554,
      "num_tokens": 4107050.0,
      "step": 2365
    },
    {
      "entropy": 6.032208967208862,
      "epoch": 0.18439992219412565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998720437840403,
      "loss": 5.9622,
      "mean_token_accuracy": 0.13016203418374062,
      "num_tokens": 4115547.0,
      "step": 2370
    },
    {
      "entropy": 6.0398108005523685,
      "epoch": 0.1847889515658432,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998711074973871,
      "loss": 5.9436,
      "mean_token_accuracy": 0.13297787085175514,
      "num_tokens": 4124047.0,
      "step": 2375
    },
    {
      "entropy": 5.943724346160889,
      "epoch": 0.1851779809375608,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998701677986728,
      "loss": 5.8968,
      "mean_token_accuracy": 0.145902356505394,
      "num_tokens": 4132005.0,
      "step": 2380
    },
    {
      "entropy": 6.1200315952301025,
      "epoch": 0.18556701030927836,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998692246879116,
      "loss": 6.062,
      "mean_token_accuracy": 0.1263026289641857,
      "num_tokens": 4140199.0,
      "step": 2385
    },
    {
      "entropy": 6.068143510818482,
      "epoch": 0.18595603968099592,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998682781651178,
      "loss": 5.9889,
      "mean_token_accuracy": 0.13453146889805795,
      "num_tokens": 4149353.0,
      "step": 2390
    },
    {
      "entropy": 5.977908182144165,
      "epoch": 0.18634506905271347,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998673282303059,
      "loss": 5.9865,
      "mean_token_accuracy": 0.13498907014727593,
      "num_tokens": 4158531.0,
      "step": 2395
    },
    {
      "entropy": 5.940221500396729,
      "epoch": 0.18673409842443103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998663748834901,
      "loss": 5.7874,
      "mean_token_accuracy": 0.14433993101119996,
      "num_tokens": 4166743.0,
      "step": 2400
    },
    {
      "entropy": 6.098818731307984,
      "epoch": 0.18712312779614862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998654181246851,
      "loss": 6.1332,
      "mean_token_accuracy": 0.13165030181407927,
      "num_tokens": 4175416.0,
      "step": 2405
    },
    {
      "entropy": 6.116787719726562,
      "epoch": 0.18751215716786618,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998644579539052,
      "loss": 6.0591,
      "mean_token_accuracy": 0.13178250938653946,
      "num_tokens": 4185324.0,
      "step": 2410
    },
    {
      "entropy": 6.0655097484588625,
      "epoch": 0.18790118653958374,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499863494371165,
      "loss": 5.9928,
      "mean_token_accuracy": 0.13624793738126756,
      "num_tokens": 4193361.0,
      "step": 2415
    },
    {
      "entropy": 6.1048424243927,
      "epoch": 0.1882902159113013,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998625273764793,
      "loss": 6.0361,
      "mean_token_accuracy": 0.12900563478469848,
      "num_tokens": 4202553.0,
      "step": 2420
    },
    {
      "entropy": 5.989557075500488,
      "epoch": 0.18867924528301888,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998615569698626,
      "loss": 5.9538,
      "mean_token_accuracy": 0.1428642563521862,
      "num_tokens": 4211585.0,
      "step": 2425
    },
    {
      "entropy": 6.038365507125855,
      "epoch": 0.18906827465473644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998605831513296,
      "loss": 5.9981,
      "mean_token_accuracy": 0.12849293798208236,
      "num_tokens": 4220829.0,
      "step": 2430
    },
    {
      "entropy": 5.955444860458374,
      "epoch": 0.189457304026454,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998596059208953,
      "loss": 5.9165,
      "mean_token_accuracy": 0.14009664207696915,
      "num_tokens": 4228658.0,
      "step": 2435
    },
    {
      "entropy": 6.089463567733764,
      "epoch": 0.18984633339817156,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998586252785743,
      "loss": 6.0069,
      "mean_token_accuracy": 0.13533291295170785,
      "num_tokens": 4237650.0,
      "step": 2440
    },
    {
      "entropy": 6.035800504684448,
      "epoch": 0.19023536276988912,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998576412243816,
      "loss": 5.9774,
      "mean_token_accuracy": 0.13687002435326576,
      "num_tokens": 4246701.0,
      "step": 2445
    },
    {
      "entropy": 6.017635250091553,
      "epoch": 0.1906243921416067,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998566537583321,
      "loss": 5.9504,
      "mean_token_accuracy": 0.1353508435189724,
      "num_tokens": 4254541.0,
      "step": 2450
    },
    {
      "entropy": 6.067923736572266,
      "epoch": 0.19101342151332426,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998556628804408,
      "loss": 5.9673,
      "mean_token_accuracy": 0.1318634532392025,
      "num_tokens": 4263111.0,
      "step": 2455
    },
    {
      "entropy": 5.993190145492553,
      "epoch": 0.19140245088504182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998546685907225,
      "loss": 5.9944,
      "mean_token_accuracy": 0.13662618547677993,
      "num_tokens": 4272796.0,
      "step": 2460
    },
    {
      "entropy": 6.009687709808349,
      "epoch": 0.19179148025675938,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998536708891927,
      "loss": 5.9369,
      "mean_token_accuracy": 0.1403819091618061,
      "num_tokens": 4281361.0,
      "step": 2465
    },
    {
      "entropy": 6.065948390960694,
      "epoch": 0.19218050962847694,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998526697758663,
      "loss": 5.9727,
      "mean_token_accuracy": 0.13534120246767997,
      "num_tokens": 4290470.0,
      "step": 2470
    },
    {
      "entropy": 6.068803882598877,
      "epoch": 0.19256953900019452,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998516652507585,
      "loss": 6.0308,
      "mean_token_accuracy": 0.13365093022584915,
      "num_tokens": 4299087.0,
      "step": 2475
    },
    {
      "entropy": 5.988065528869629,
      "epoch": 0.19295856837191208,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998506573138846,
      "loss": 5.8941,
      "mean_token_accuracy": 0.1456553541123867,
      "num_tokens": 4307572.0,
      "step": 2480
    },
    {
      "entropy": 6.013512992858887,
      "epoch": 0.19334759774362964,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998496459652598,
      "loss": 5.9525,
      "mean_token_accuracy": 0.13854917660355567,
      "num_tokens": 4316337.0,
      "step": 2485
    },
    {
      "entropy": 6.133624935150147,
      "epoch": 0.1937366271153472,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998486312048995,
      "loss": 6.0553,
      "mean_token_accuracy": 0.13632312491536142,
      "num_tokens": 4326258.0,
      "step": 2490
    },
    {
      "entropy": 5.984883785247803,
      "epoch": 0.19412565648706478,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998476130328193,
      "loss": 5.8958,
      "mean_token_accuracy": 0.14192752689123153,
      "num_tokens": 4334711.0,
      "step": 2495
    },
    {
      "entropy": 6.024339723587036,
      "epoch": 0.19451468585878234,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998465914490343,
      "loss": 6.0093,
      "mean_token_accuracy": 0.13177428618073464,
      "num_tokens": 4343739.0,
      "step": 2500
    },
    {
      "entropy": 5.996756887435913,
      "epoch": 0.1949037152304999,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00049984556645356,
      "loss": 5.9476,
      "mean_token_accuracy": 0.13979061022400857,
      "num_tokens": 4352905.0,
      "step": 2505
    },
    {
      "entropy": 6.098079061508178,
      "epoch": 0.19529274460221746,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998445380464124,
      "loss": 5.9867,
      "mean_token_accuracy": 0.14360451996326445,
      "num_tokens": 4361333.0,
      "step": 2510
    },
    {
      "entropy": 5.932000255584716,
      "epoch": 0.19568177397393502,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998435062276066,
      "loss": 5.9141,
      "mean_token_accuracy": 0.13802293092012405,
      "num_tokens": 4369872.0,
      "step": 2515
    },
    {
      "entropy": 6.033679389953614,
      "epoch": 0.1960708033456526,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998424709971586,
      "loss": 6.001,
      "mean_token_accuracy": 0.14147183522582055,
      "num_tokens": 4378097.0,
      "step": 2520
    },
    {
      "entropy": 5.968611574172973,
      "epoch": 0.19645983271737016,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998414323550839,
      "loss": 5.9189,
      "mean_token_accuracy": 0.14388937279582023,
      "num_tokens": 4387414.0,
      "step": 2525
    },
    {
      "entropy": 6.1040074825286865,
      "epoch": 0.19684886208908772,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004998403903013984,
      "loss": 6.0318,
      "mean_token_accuracy": 0.14406417012214662,
      "num_tokens": 4395922.0,
      "step": 2530
    },
    {
      "entropy": 5.995774126052856,
      "epoch": 0.19723789146080528,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998393448361179,
      "loss": 5.9889,
      "mean_token_accuracy": 0.13444236814975738,
      "num_tokens": 4404500.0,
      "step": 2535
    },
    {
      "entropy": 5.907389211654663,
      "epoch": 0.19762692083252287,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499838295959258,
      "loss": 5.9706,
      "mean_token_accuracy": 0.1347421184182167,
      "num_tokens": 4413433.0,
      "step": 2540
    },
    {
      "entropy": 6.097556447982788,
      "epoch": 0.19801595020424043,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499837243670835,
      "loss": 5.9971,
      "mean_token_accuracy": 0.14098930060863496,
      "num_tokens": 4421789.0,
      "step": 2545
    },
    {
      "entropy": 5.943789005279541,
      "epoch": 0.19840497957595798,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998361879708646,
      "loss": 5.8737,
      "mean_token_accuracy": 0.14472436904907227,
      "num_tokens": 4430256.0,
      "step": 2550
    },
    {
      "entropy": 5.98599910736084,
      "epoch": 0.19879400894767554,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499835128859363,
      "loss": 5.9308,
      "mean_token_accuracy": 0.1392303802073002,
      "num_tokens": 4438431.0,
      "step": 2555
    },
    {
      "entropy": 6.054162168502808,
      "epoch": 0.1991830383193931,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998340663363461,
      "loss": 5.9844,
      "mean_token_accuracy": 0.13681029602885247,
      "num_tokens": 4447150.0,
      "step": 2560
    },
    {
      "entropy": 5.967094993591308,
      "epoch": 0.1995720676911107,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998330004018301,
      "loss": 5.8782,
      "mean_token_accuracy": 0.14383712857961656,
      "num_tokens": 4455181.0,
      "step": 2565
    },
    {
      "entropy": 5.973532247543335,
      "epoch": 0.19996109706282825,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998319310558312,
      "loss": 5.8949,
      "mean_token_accuracy": 0.13384012654423713,
      "num_tokens": 4463361.0,
      "step": 2570
    },
    {
      "entropy": 5.9947929859161375,
      "epoch": 0.2003501264345458,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998308582983657,
      "loss": 5.9501,
      "mean_token_accuracy": 0.14294253960251807,
      "num_tokens": 4471853.0,
      "step": 2575
    },
    {
      "entropy": 5.944393157958984,
      "epoch": 0.20073915580626336,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998297821294497,
      "loss": 5.8836,
      "mean_token_accuracy": 0.1416514605283737,
      "num_tokens": 4480812.0,
      "step": 2580
    },
    {
      "entropy": 6.001008462905884,
      "epoch": 0.20112818517798095,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998287025490995,
      "loss": 5.9292,
      "mean_token_accuracy": 0.14392636716365814,
      "num_tokens": 4489180.0,
      "step": 2585
    },
    {
      "entropy": 5.991946077346801,
      "epoch": 0.2015172145496985,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998276195573317,
      "loss": 5.9977,
      "mean_token_accuracy": 0.1375030979514122,
      "num_tokens": 4498013.0,
      "step": 2590
    },
    {
      "entropy": 6.1099851608276365,
      "epoch": 0.20190624392141607,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998265331541627,
      "loss": 6.0428,
      "mean_token_accuracy": 0.12695643529295922,
      "num_tokens": 4506767.0,
      "step": 2595
    },
    {
      "entropy": 6.122916030883789,
      "epoch": 0.20229527329313363,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998254433396088,
      "loss": 6.0417,
      "mean_token_accuracy": 0.13464597910642623,
      "num_tokens": 4516657.0,
      "step": 2600
    },
    {
      "entropy": 5.863552236557007,
      "epoch": 0.20268430266485118,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998243501136867,
      "loss": 5.8851,
      "mean_token_accuracy": 0.14413899555802345,
      "num_tokens": 4524980.0,
      "step": 2605
    },
    {
      "entropy": 5.983160829544067,
      "epoch": 0.20307333203656877,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998232534764129,
      "loss": 5.9378,
      "mean_token_accuracy": 0.140546715259552,
      "num_tokens": 4533236.0,
      "step": 2610
    },
    {
      "entropy": 5.957806062698364,
      "epoch": 0.20346236140828633,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998221534278041,
      "loss": 5.8374,
      "mean_token_accuracy": 0.14236027672886847,
      "num_tokens": 4541778.0,
      "step": 2615
    },
    {
      "entropy": 6.048499488830567,
      "epoch": 0.2038513907800039,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998210499678769,
      "loss": 5.9587,
      "mean_token_accuracy": 0.13639849424362183,
      "num_tokens": 4551131.0,
      "step": 2620
    },
    {
      "entropy": 6.044900178909302,
      "epoch": 0.20424042015172145,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998199430966481,
      "loss": 5.9342,
      "mean_token_accuracy": 0.13491834998130797,
      "num_tokens": 4559530.0,
      "step": 2625
    },
    {
      "entropy": 5.8130594253540036,
      "epoch": 0.20462944952343903,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998188328141346,
      "loss": 5.8277,
      "mean_token_accuracy": 0.14135022163391114,
      "num_tokens": 4567533.0,
      "step": 2630
    },
    {
      "entropy": 6.064992189407349,
      "epoch": 0.2050184788951566,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998177191203531,
      "loss": 5.978,
      "mean_token_accuracy": 0.13598859161138535,
      "num_tokens": 4576414.0,
      "step": 2635
    },
    {
      "entropy": 6.019899988174439,
      "epoch": 0.20540750826687415,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998166020153204,
      "loss": 5.9119,
      "mean_token_accuracy": 0.1416245199739933,
      "num_tokens": 4585331.0,
      "step": 2640
    },
    {
      "entropy": 5.9772216796875,
      "epoch": 0.2057965376385917,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998154814990538,
      "loss": 5.9417,
      "mean_token_accuracy": 0.144110806286335,
      "num_tokens": 4593526.0,
      "step": 2645
    },
    {
      "entropy": 5.955821752548218,
      "epoch": 0.20618556701030927,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00049981435757157,
      "loss": 5.9234,
      "mean_token_accuracy": 0.1401088260114193,
      "num_tokens": 4602400.0,
      "step": 2650
    },
    {
      "entropy": 5.917225313186646,
      "epoch": 0.20657459638202685,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998132302328862,
      "loss": 5.8825,
      "mean_token_accuracy": 0.14242400079965592,
      "num_tokens": 4611037.0,
      "step": 2655
    },
    {
      "entropy": 6.003968095779419,
      "epoch": 0.2069636257537444,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998120994830195,
      "loss": 5.8919,
      "mean_token_accuracy": 0.13950355872511863,
      "num_tokens": 4619030.0,
      "step": 2660
    },
    {
      "entropy": 5.95154185295105,
      "epoch": 0.20735265512546197,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499810965321987,
      "loss": 5.8944,
      "mean_token_accuracy": 0.13457175344228745,
      "num_tokens": 4628649.0,
      "step": 2665
    },
    {
      "entropy": 5.90372486114502,
      "epoch": 0.20774168449717953,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998098277498059,
      "loss": 5.7891,
      "mean_token_accuracy": 0.14544083476066588,
      "num_tokens": 4636935.0,
      "step": 2670
    },
    {
      "entropy": 5.975887823104858,
      "epoch": 0.2081307138688971,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998086867664934,
      "loss": 5.9827,
      "mean_token_accuracy": 0.13934769332408906,
      "num_tokens": 4645532.0,
      "step": 2675
    },
    {
      "entropy": 6.0481304168701175,
      "epoch": 0.20851974324061467,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499807542372067,
      "loss": 6.0577,
      "mean_token_accuracy": 0.13306401297450066,
      "num_tokens": 4654433.0,
      "step": 2680
    },
    {
      "entropy": 6.116953277587891,
      "epoch": 0.20890877261233223,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998063945665439,
      "loss": 6.0665,
      "mean_token_accuracy": 0.13022227212786674,
      "num_tokens": 4662913.0,
      "step": 2685
    },
    {
      "entropy": 6.076945877075195,
      "epoch": 0.2092978019840498,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998052433499416,
      "loss": 6.0346,
      "mean_token_accuracy": 0.13053162321448325,
      "num_tokens": 4672270.0,
      "step": 2690
    },
    {
      "entropy": 6.091213703155518,
      "epoch": 0.20968683135576735,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998040887222776,
      "loss": 6.0588,
      "mean_token_accuracy": 0.1330445796251297,
      "num_tokens": 4681428.0,
      "step": 2695
    },
    {
      "entropy": 6.059315538406372,
      "epoch": 0.21007586072748494,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998029306835693,
      "loss": 5.889,
      "mean_token_accuracy": 0.14305261075496672,
      "num_tokens": 4690492.0,
      "step": 2700
    },
    {
      "entropy": 5.9609434604644775,
      "epoch": 0.2104648900992025,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998017692338344,
      "loss": 6.0116,
      "mean_token_accuracy": 0.13601114600896835,
      "num_tokens": 4699667.0,
      "step": 2705
    },
    {
      "entropy": 5.870917129516601,
      "epoch": 0.21085391947092005,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998006043730905,
      "loss": 5.6888,
      "mean_token_accuracy": 0.1551996573805809,
      "num_tokens": 4708529.0,
      "step": 2710
    },
    {
      "entropy": 5.874881172180176,
      "epoch": 0.2112429488426376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997994361013551,
      "loss": 5.9372,
      "mean_token_accuracy": 0.13897888511419296,
      "num_tokens": 4717684.0,
      "step": 2715
    },
    {
      "entropy": 6.014762020111084,
      "epoch": 0.21163197821435517,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997982644186461,
      "loss": 5.8632,
      "mean_token_accuracy": 0.14865887761116028,
      "num_tokens": 4727067.0,
      "step": 2720
    },
    {
      "entropy": 5.8886195659637455,
      "epoch": 0.21202100758607276,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997970893249813,
      "loss": 5.8669,
      "mean_token_accuracy": 0.14542917907238007,
      "num_tokens": 4735434.0,
      "step": 2725
    },
    {
      "entropy": 6.017463541030883,
      "epoch": 0.21241003695779032,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997959108203785,
      "loss": 5.8839,
      "mean_token_accuracy": 0.14075807183980943,
      "num_tokens": 4743843.0,
      "step": 2730
    },
    {
      "entropy": 6.023525381088257,
      "epoch": 0.21279906632950787,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997947289048554,
      "loss": 5.9474,
      "mean_token_accuracy": 0.13808054327964783,
      "num_tokens": 4753228.0,
      "step": 2735
    },
    {
      "entropy": 5.975101375579834,
      "epoch": 0.21318809570122543,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997935435784302,
      "loss": 6.0146,
      "mean_token_accuracy": 0.13901024162769318,
      "num_tokens": 4763401.0,
      "step": 2740
    },
    {
      "entropy": 5.985175466537475,
      "epoch": 0.21357712507294302,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997923548411208,
      "loss": 5.8696,
      "mean_token_accuracy": 0.13853829503059387,
      "num_tokens": 4771812.0,
      "step": 2745
    },
    {
      "entropy": 5.947464084625244,
      "epoch": 0.21396615444466058,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499791162692945,
      "loss": 5.924,
      "mean_token_accuracy": 0.14195793271064758,
      "num_tokens": 4780616.0,
      "step": 2750
    },
    {
      "entropy": 5.950645589828492,
      "epoch": 0.21435518381637814,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997899671339214,
      "loss": 5.9087,
      "mean_token_accuracy": 0.13782943710684775,
      "num_tokens": 4789886.0,
      "step": 2755
    },
    {
      "entropy": 5.94676365852356,
      "epoch": 0.2147442131880957,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997887681640675,
      "loss": 5.8879,
      "mean_token_accuracy": 0.14307310357689856,
      "num_tokens": 4799189.0,
      "step": 2760
    },
    {
      "entropy": 5.99482946395874,
      "epoch": 0.21513324255981325,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997875657834021,
      "loss": 5.9339,
      "mean_token_accuracy": 0.14143783524632453,
      "num_tokens": 4807850.0,
      "step": 2765
    },
    {
      "entropy": 5.996881151199341,
      "epoch": 0.21552227193153084,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997863599919432,
      "loss": 5.8407,
      "mean_token_accuracy": 0.14409077540040016,
      "num_tokens": 4815742.0,
      "step": 2770
    },
    {
      "entropy": 5.928023815155029,
      "epoch": 0.2159113013032484,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997851507897089,
      "loss": 5.9279,
      "mean_token_accuracy": 0.14160390421748162,
      "num_tokens": 4824761.0,
      "step": 2775
    },
    {
      "entropy": 6.08947377204895,
      "epoch": 0.21630033067496596,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997839381767178,
      "loss": 6.0725,
      "mean_token_accuracy": 0.1348046600818634,
      "num_tokens": 4832805.0,
      "step": 2780
    },
    {
      "entropy": 5.996554565429688,
      "epoch": 0.21668936004668352,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997827221529882,
      "loss": 5.9128,
      "mean_token_accuracy": 0.14150556772947312,
      "num_tokens": 4841938.0,
      "step": 2785
    },
    {
      "entropy": 5.940557479858398,
      "epoch": 0.2170783894184011,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997815027185386,
      "loss": 5.9411,
      "mean_token_accuracy": 0.1395154818892479,
      "num_tokens": 4849932.0,
      "step": 2790
    },
    {
      "entropy": 5.892970848083496,
      "epoch": 0.21746741879011866,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997802798733874,
      "loss": 5.8184,
      "mean_token_accuracy": 0.1439179942011833,
      "num_tokens": 4859054.0,
      "step": 2795
    },
    {
      "entropy": 6.062068367004395,
      "epoch": 0.21785644816183622,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997790536175534,
      "loss": 5.9724,
      "mean_token_accuracy": 0.1457817457616329,
      "num_tokens": 4867415.0,
      "step": 2800
    },
    {
      "entropy": 6.05829119682312,
      "epoch": 0.21824547753355378,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997778239510548,
      "loss": 5.9968,
      "mean_token_accuracy": 0.1384667731821537,
      "num_tokens": 4875965.0,
      "step": 2805
    },
    {
      "entropy": 5.974689674377442,
      "epoch": 0.21863450690527134,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997765908739106,
      "loss": 5.9021,
      "mean_token_accuracy": 0.1377502866089344,
      "num_tokens": 4884629.0,
      "step": 2810
    },
    {
      "entropy": 5.95506272315979,
      "epoch": 0.21902353627698892,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997753543861395,
      "loss": 5.9429,
      "mean_token_accuracy": 0.1401269853115082,
      "num_tokens": 4893573.0,
      "step": 2815
    },
    {
      "entropy": 5.953045749664307,
      "epoch": 0.21941256564870648,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049977411448776,
      "loss": 5.8819,
      "mean_token_accuracy": 0.1408964700996876,
      "num_tokens": 4902831.0,
      "step": 2820
    },
    {
      "entropy": 6.035543966293335,
      "epoch": 0.21980159502042404,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997728711787912,
      "loss": 5.9353,
      "mean_token_accuracy": 0.1345332682132721,
      "num_tokens": 4911360.0,
      "step": 2825
    },
    {
      "entropy": 5.9824096202850345,
      "epoch": 0.2201906243921416,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997716244592518,
      "loss": 5.8991,
      "mean_token_accuracy": 0.14080488830804824,
      "num_tokens": 4919758.0,
      "step": 2830
    },
    {
      "entropy": 5.828854990005493,
      "epoch": 0.22057965376385918,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997703743291607,
      "loss": 5.7909,
      "mean_token_accuracy": 0.14767343401908875,
      "num_tokens": 4928934.0,
      "step": 2835
    },
    {
      "entropy": 5.861857986450195,
      "epoch": 0.22096868313557674,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499769120788537,
      "loss": 5.8351,
      "mean_token_accuracy": 0.14466070234775544,
      "num_tokens": 4937780.0,
      "step": 2840
    },
    {
      "entropy": 6.097094631195068,
      "epoch": 0.2213577125072943,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997678638373995,
      "loss": 5.9356,
      "mean_token_accuracy": 0.13491297960281373,
      "num_tokens": 4946862.0,
      "step": 2845
    },
    {
      "entropy": 5.935989952087402,
      "epoch": 0.22174674187901186,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997666034757676,
      "loss": 5.9189,
      "mean_token_accuracy": 0.1431422159075737,
      "num_tokens": 4954883.0,
      "step": 2850
    },
    {
      "entropy": 5.854018354415894,
      "epoch": 0.22213577125072942,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997653397036603,
      "loss": 5.7182,
      "mean_token_accuracy": 0.1529537945985794,
      "num_tokens": 4962840.0,
      "step": 2855
    },
    {
      "entropy": 5.892796182632447,
      "epoch": 0.222524800622447,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997640725210965,
      "loss": 5.9556,
      "mean_token_accuracy": 0.14130814969539643,
      "num_tokens": 4972334.0,
      "step": 2860
    },
    {
      "entropy": 6.061372137069702,
      "epoch": 0.22291382999416456,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997628019280958,
      "loss": 5.916,
      "mean_token_accuracy": 0.14689642414450646,
      "num_tokens": 4981032.0,
      "step": 2865
    },
    {
      "entropy": 5.977332258224488,
      "epoch": 0.22330285936588212,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997615279246773,
      "loss": 5.9364,
      "mean_token_accuracy": 0.136233139783144,
      "num_tokens": 4990502.0,
      "step": 2870
    },
    {
      "entropy": 6.0572460174560545,
      "epoch": 0.22369188873759968,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997602505108603,
      "loss": 6.0998,
      "mean_token_accuracy": 0.13143445029854775,
      "num_tokens": 4999121.0,
      "step": 2875
    },
    {
      "entropy": 6.014243221282959,
      "epoch": 0.22408091810931724,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997589696866644,
      "loss": 5.8681,
      "mean_token_accuracy": 0.14159790500998498,
      "num_tokens": 5007271.0,
      "step": 2880
    },
    {
      "entropy": 5.872592258453369,
      "epoch": 0.22446994748103483,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997576854521088,
      "loss": 5.9069,
      "mean_token_accuracy": 0.1393379971385002,
      "num_tokens": 5015940.0,
      "step": 2885
    },
    {
      "entropy": 5.897215223312378,
      "epoch": 0.22485897685275238,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499756397807213,
      "loss": 5.8309,
      "mean_token_accuracy": 0.14309723377227784,
      "num_tokens": 5024294.0,
      "step": 2890
    },
    {
      "entropy": 5.909600305557251,
      "epoch": 0.22524800622446994,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997551067519966,
      "loss": 5.7782,
      "mean_token_accuracy": 0.1465318411588669,
      "num_tokens": 5033401.0,
      "step": 2895
    },
    {
      "entropy": 5.946053552627563,
      "epoch": 0.2256370355961875,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997538122864792,
      "loss": 5.8857,
      "mean_token_accuracy": 0.14171253219246865,
      "num_tokens": 5041955.0,
      "step": 2900
    },
    {
      "entropy": 6.011700201034546,
      "epoch": 0.2260260649679051,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997525144106804,
      "loss": 5.9156,
      "mean_token_accuracy": 0.13371872827410697,
      "num_tokens": 5051132.0,
      "step": 2905
    },
    {
      "entropy": 5.881177234649658,
      "epoch": 0.22641509433962265,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00049975121312462,
      "loss": 5.8391,
      "mean_token_accuracy": 0.15032089352607728,
      "num_tokens": 5059492.0,
      "step": 2910
    },
    {
      "entropy": 5.902948617935181,
      "epoch": 0.2268041237113402,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997499084283177,
      "loss": 5.8906,
      "mean_token_accuracy": 0.14851512461900712,
      "num_tokens": 5069015.0,
      "step": 2915
    },
    {
      "entropy": 5.968337202072144,
      "epoch": 0.22719315308305776,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997486003217932,
      "loss": 5.8976,
      "mean_token_accuracy": 0.1448035165667534,
      "num_tokens": 5077614.0,
      "step": 2920
    },
    {
      "entropy": 5.878965139389038,
      "epoch": 0.22758218245477532,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997472888050664,
      "loss": 5.8335,
      "mean_token_accuracy": 0.14750696420669557,
      "num_tokens": 5086300.0,
      "step": 2925
    },
    {
      "entropy": 5.818942022323609,
      "epoch": 0.2279712118264929,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997459738781573,
      "loss": 5.8771,
      "mean_token_accuracy": 0.14237508922815323,
      "num_tokens": 5095967.0,
      "step": 2930
    },
    {
      "entropy": 6.032419204711914,
      "epoch": 0.22836024119821047,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997446555410857,
      "loss": 5.827,
      "mean_token_accuracy": 0.14276775270700454,
      "num_tokens": 5104611.0,
      "step": 2935
    },
    {
      "entropy": 5.827592563629151,
      "epoch": 0.22874927056992803,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997433337938716,
      "loss": 5.7944,
      "mean_token_accuracy": 0.1462033912539482,
      "num_tokens": 5113452.0,
      "step": 2940
    },
    {
      "entropy": 5.833715486526489,
      "epoch": 0.22913829994164558,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997420086365351,
      "loss": 5.775,
      "mean_token_accuracy": 0.1478416822850704,
      "num_tokens": 5121558.0,
      "step": 2945
    },
    {
      "entropy": 5.861429166793823,
      "epoch": 0.22952732931336317,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997406800690966,
      "loss": 5.7527,
      "mean_token_accuracy": 0.1565452352166176,
      "num_tokens": 5129942.0,
      "step": 2950
    },
    {
      "entropy": 5.958620166778564,
      "epoch": 0.22991635868508073,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997393480915758,
      "loss": 5.8909,
      "mean_token_accuracy": 0.13326751589775085,
      "num_tokens": 5139012.0,
      "step": 2955
    },
    {
      "entropy": 5.977751779556274,
      "epoch": 0.2303053880567983,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997380127039931,
      "loss": 5.891,
      "mean_token_accuracy": 0.1418101616203785,
      "num_tokens": 5148185.0,
      "step": 2960
    },
    {
      "entropy": 5.962037992477417,
      "epoch": 0.23069441742851585,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997366739063688,
      "loss": 5.9079,
      "mean_token_accuracy": 0.13784476593136788,
      "num_tokens": 5157086.0,
      "step": 2965
    },
    {
      "entropy": 6.051717805862427,
      "epoch": 0.2310834468002334,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997353316987231,
      "loss": 5.9769,
      "mean_token_accuracy": 0.13798867762088776,
      "num_tokens": 5165842.0,
      "step": 2970
    },
    {
      "entropy": 5.972775411605835,
      "epoch": 0.231472476171951,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997339860810765,
      "loss": 5.986,
      "mean_token_accuracy": 0.13841650187969207,
      "num_tokens": 5174161.0,
      "step": 2975
    },
    {
      "entropy": 6.012486743927002,
      "epoch": 0.23186150554366855,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997326370534495,
      "loss": 5.9191,
      "mean_token_accuracy": 0.13878133594989778,
      "num_tokens": 5183830.0,
      "step": 2980
    },
    {
      "entropy": 5.924656009674072,
      "epoch": 0.2322505349153861,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997312846158622,
      "loss": 5.8498,
      "mean_token_accuracy": 0.13782228529453278,
      "num_tokens": 5192248.0,
      "step": 2985
    },
    {
      "entropy": 5.876343250274658,
      "epoch": 0.23263956428710367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997299287683355,
      "loss": 5.8061,
      "mean_token_accuracy": 0.1474793091416359,
      "num_tokens": 5201212.0,
      "step": 2990
    },
    {
      "entropy": 5.893640851974487,
      "epoch": 0.23302859365882125,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997285695108898,
      "loss": 5.8544,
      "mean_token_accuracy": 0.14440894722938538,
      "num_tokens": 5209338.0,
      "step": 2995
    },
    {
      "entropy": 6.071593999862671,
      "epoch": 0.2334176230305388,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997272068435458,
      "loss": 5.8863,
      "mean_token_accuracy": 0.13977692127227784,
      "num_tokens": 5217590.0,
      "step": 3000
    },
    {
      "epoch": 0.2334176230305388,
      "eval_entropy": 5.784550465518632,
      "eval_loss": 5.93321418762207,
      "eval_mean_token_accuracy": 0.1467212013470025,
      "eval_num_tokens": 5217590.0,
      "eval_runtime": 28.6214,
      "eval_samples_per_second": 1451.991,
      "eval_steps_per_second": 181.508,
      "step": 3000
    },
    {
      "entropy": 5.815381383895874,
      "epoch": 0.23380665240225637,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499725840766324,
      "loss": 5.8118,
      "mean_token_accuracy": 0.14703051149845123,
      "num_tokens": 5226383.0,
      "step": 3005
    },
    {
      "entropy": 6.005730104446411,
      "epoch": 0.23419568177397393,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997244712792453,
      "loss": 5.8461,
      "mean_token_accuracy": 0.13974943310022353,
      "num_tokens": 5235517.0,
      "step": 3010
    },
    {
      "entropy": 5.916208028793335,
      "epoch": 0.2345847111456915,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997230983823305,
      "loss": 5.8215,
      "mean_token_accuracy": 0.14983327984809874,
      "num_tokens": 5243812.0,
      "step": 3015
    },
    {
      "entropy": 5.8480511665344235,
      "epoch": 0.23497374051740907,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997217220756003,
      "loss": 5.8457,
      "mean_token_accuracy": 0.14438704922795295,
      "num_tokens": 5252399.0,
      "step": 3020
    },
    {
      "entropy": 5.957148313522339,
      "epoch": 0.23536276988912663,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997203423590757,
      "loss": 5.8616,
      "mean_token_accuracy": 0.14147318676114082,
      "num_tokens": 5260847.0,
      "step": 3025
    },
    {
      "entropy": 5.929324531555176,
      "epoch": 0.2357517992608442,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997189592327775,
      "loss": 5.9349,
      "mean_token_accuracy": 0.13945313170552254,
      "num_tokens": 5268847.0,
      "step": 3030
    },
    {
      "entropy": 5.9836653709411625,
      "epoch": 0.23614082863256175,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997175726967268,
      "loss": 5.8816,
      "mean_token_accuracy": 0.13880289793014527,
      "num_tokens": 5277989.0,
      "step": 3035
    },
    {
      "entropy": 5.839021635055542,
      "epoch": 0.23652985800427934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997161827509447,
      "loss": 5.8717,
      "mean_token_accuracy": 0.14272701367735863,
      "num_tokens": 5286272.0,
      "step": 3040
    },
    {
      "entropy": 5.89699158668518,
      "epoch": 0.2369188873759969,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997147893954521,
      "loss": 5.8847,
      "mean_token_accuracy": 0.145920929312706,
      "num_tokens": 5295281.0,
      "step": 3045
    },
    {
      "entropy": 5.979667329788208,
      "epoch": 0.23730791674771445,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997133926302702,
      "loss": 5.8187,
      "mean_token_accuracy": 0.14393712133169173,
      "num_tokens": 5303905.0,
      "step": 3050
    },
    {
      "entropy": 5.896984624862671,
      "epoch": 0.237696946119432,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997119924554204,
      "loss": 5.7868,
      "mean_token_accuracy": 0.14225174486637115,
      "num_tokens": 5313146.0,
      "step": 3055
    },
    {
      "entropy": 5.891476774215699,
      "epoch": 0.23808597549114957,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997105888709236,
      "loss": 5.8231,
      "mean_token_accuracy": 0.1491152212023735,
      "num_tokens": 5320838.0,
      "step": 3060
    },
    {
      "entropy": 5.914984178543091,
      "epoch": 0.23847500486286716,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997091818768015,
      "loss": 5.8439,
      "mean_token_accuracy": 0.14126314371824264,
      "num_tokens": 5329312.0,
      "step": 3065
    },
    {
      "entropy": 5.878947925567627,
      "epoch": 0.23886403423458472,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499707771473075,
      "loss": 5.8463,
      "mean_token_accuracy": 0.1426366925239563,
      "num_tokens": 5337747.0,
      "step": 3070
    },
    {
      "entropy": 5.908112478256226,
      "epoch": 0.23925306360630227,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997063576597659,
      "loss": 5.8391,
      "mean_token_accuracy": 0.14027275368571282,
      "num_tokens": 5346928.0,
      "step": 3075
    },
    {
      "entropy": 5.93986496925354,
      "epoch": 0.23964209297801983,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997049404368954,
      "loss": 5.8691,
      "mean_token_accuracy": 0.13592054694890976,
      "num_tokens": 5355827.0,
      "step": 3080
    },
    {
      "entropy": 6.056811714172364,
      "epoch": 0.2400311223497374,
      "grad_norm": 1.125,
      "learning_rate": 0.000499703519804485,
      "loss": 5.9944,
      "mean_token_accuracy": 0.13909246176481246,
      "num_tokens": 5364309.0,
      "step": 3085
    },
    {
      "entropy": 5.93008246421814,
      "epoch": 0.24042015172145498,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997020957625564,
      "loss": 5.8922,
      "mean_token_accuracy": 0.1420133113861084,
      "num_tokens": 5372979.0,
      "step": 3090
    },
    {
      "entropy": 5.913248920440674,
      "epoch": 0.24080918109317254,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997006683111312,
      "loss": 5.8543,
      "mean_token_accuracy": 0.1386646196246147,
      "num_tokens": 5381809.0,
      "step": 3095
    },
    {
      "entropy": 5.999415254592895,
      "epoch": 0.2411982104648901,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499699237450231,
      "loss": 5.9173,
      "mean_token_accuracy": 0.14021561443805694,
      "num_tokens": 5390036.0,
      "step": 3100
    },
    {
      "entropy": 6.05153455734253,
      "epoch": 0.24158723983660765,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996978031798774,
      "loss": 5.9727,
      "mean_token_accuracy": 0.13660571947693825,
      "num_tokens": 5398739.0,
      "step": 3105
    },
    {
      "entropy": 5.906278991699219,
      "epoch": 0.24197626920832524,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996963655000924,
      "loss": 5.8769,
      "mean_token_accuracy": 0.14358554780483246,
      "num_tokens": 5407441.0,
      "step": 3110
    },
    {
      "entropy": 5.92988772392273,
      "epoch": 0.2423652985800428,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996949244108977,
      "loss": 5.8332,
      "mean_token_accuracy": 0.13889154195785522,
      "num_tokens": 5416534.0,
      "step": 3115
    },
    {
      "entropy": 5.9200439453125,
      "epoch": 0.24275432795176036,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499693479912315,
      "loss": 5.8778,
      "mean_token_accuracy": 0.14260219261050225,
      "num_tokens": 5425170.0,
      "step": 3120
    },
    {
      "entropy": 5.876896524429322,
      "epoch": 0.24314335732347792,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996920320043666,
      "loss": 5.8391,
      "mean_token_accuracy": 0.13886853009462358,
      "num_tokens": 5433721.0,
      "step": 3125
    },
    {
      "entropy": 5.979009056091309,
      "epoch": 0.24353238669519547,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996905806870742,
      "loss": 5.9017,
      "mean_token_accuracy": 0.13908121958374978,
      "num_tokens": 5441708.0,
      "step": 3130
    },
    {
      "entropy": 5.855253744125366,
      "epoch": 0.24392141606691306,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996891259604598,
      "loss": 5.8035,
      "mean_token_accuracy": 0.14685206562280656,
      "num_tokens": 5449942.0,
      "step": 3135
    },
    {
      "entropy": 5.908247184753418,
      "epoch": 0.24431044543863062,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996876678245455,
      "loss": 5.8303,
      "mean_token_accuracy": 0.14304618909955025,
      "num_tokens": 5458004.0,
      "step": 3140
    },
    {
      "entropy": 5.999393939971924,
      "epoch": 0.24469947481034818,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996862062793536,
      "loss": 5.917,
      "mean_token_accuracy": 0.14605897068977355,
      "num_tokens": 5467071.0,
      "step": 3145
    },
    {
      "entropy": 5.911773157119751,
      "epoch": 0.24508850418206574,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996847413249061,
      "loss": 5.8867,
      "mean_token_accuracy": 0.1440001830458641,
      "num_tokens": 5476221.0,
      "step": 3150
    },
    {
      "entropy": 5.921053266525268,
      "epoch": 0.24547753355378332,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996832729612252,
      "loss": 5.8058,
      "mean_token_accuracy": 0.14126212745904923,
      "num_tokens": 5484549.0,
      "step": 3155
    },
    {
      "entropy": 5.818511867523194,
      "epoch": 0.24586656292550088,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996818011883333,
      "loss": 5.7911,
      "mean_token_accuracy": 0.14696895629167556,
      "num_tokens": 5492909.0,
      "step": 3160
    },
    {
      "entropy": 5.890760278701782,
      "epoch": 0.24625559229721844,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996803260062527,
      "loss": 5.857,
      "mean_token_accuracy": 0.14617310762405394,
      "num_tokens": 5501418.0,
      "step": 3165
    },
    {
      "entropy": 5.960519218444825,
      "epoch": 0.246644621668936,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996788474150057,
      "loss": 5.8585,
      "mean_token_accuracy": 0.14115139245986938,
      "num_tokens": 5510609.0,
      "step": 3170
    },
    {
      "entropy": 5.837988567352295,
      "epoch": 0.24703365104065356,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996773654146149,
      "loss": 5.7035,
      "mean_token_accuracy": 0.15274056941270828,
      "num_tokens": 5518349.0,
      "step": 3175
    },
    {
      "entropy": 5.748640775680542,
      "epoch": 0.24742268041237114,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996758800051026,
      "loss": 5.6529,
      "mean_token_accuracy": 0.15209673792123796,
      "num_tokens": 5526972.0,
      "step": 3180
    },
    {
      "entropy": 5.78744649887085,
      "epoch": 0.2478117097840887,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996743911864916,
      "loss": 5.8302,
      "mean_token_accuracy": 0.1438259944319725,
      "num_tokens": 5535683.0,
      "step": 3185
    },
    {
      "entropy": 5.8871269702911375,
      "epoch": 0.24820073915580626,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996728989588041,
      "loss": 5.7923,
      "mean_token_accuracy": 0.15241506546735764,
      "num_tokens": 5544258.0,
      "step": 3190
    },
    {
      "entropy": 5.778812265396118,
      "epoch": 0.24858976852752382,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996714033220632,
      "loss": 5.8127,
      "mean_token_accuracy": 0.14140888005495073,
      "num_tokens": 5552608.0,
      "step": 3195
    },
    {
      "entropy": 5.977884817123413,
      "epoch": 0.2489787978992414,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996699042762911,
      "loss": 5.9323,
      "mean_token_accuracy": 0.13892421647906303,
      "num_tokens": 5561357.0,
      "step": 3200
    },
    {
      "entropy": 5.935313653945923,
      "epoch": 0.24936782727095896,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996684018215111,
      "loss": 5.7652,
      "mean_token_accuracy": 0.15381811261177064,
      "num_tokens": 5569862.0,
      "step": 3205
    },
    {
      "entropy": 5.790357780456543,
      "epoch": 0.24975685664267652,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996668959577453,
      "loss": 5.8232,
      "mean_token_accuracy": 0.14730967432260514,
      "num_tokens": 5578485.0,
      "step": 3210
    },
    {
      "entropy": 5.8659741401672365,
      "epoch": 0.2501458860143941,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996653866850173,
      "loss": 5.8009,
      "mean_token_accuracy": 0.13983895629644394,
      "num_tokens": 5587713.0,
      "step": 3215
    },
    {
      "entropy": 5.916757345199585,
      "epoch": 0.25053491538611167,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996638740033495,
      "loss": 5.8827,
      "mean_token_accuracy": 0.14453418254852296,
      "num_tokens": 5596541.0,
      "step": 3220
    },
    {
      "entropy": 5.8514844417572025,
      "epoch": 0.2509239447578292,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996623579127651,
      "loss": 5.7605,
      "mean_token_accuracy": 0.14728045091032982,
      "num_tokens": 5605573.0,
      "step": 3225
    },
    {
      "entropy": 5.8693195343017575,
      "epoch": 0.2513129741295468,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996608384132868,
      "loss": 5.7652,
      "mean_token_accuracy": 0.14660658463835716,
      "num_tokens": 5614016.0,
      "step": 3230
    },
    {
      "entropy": 5.9747411727905275,
      "epoch": 0.25170200350126437,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499659315504938,
      "loss": 5.9937,
      "mean_token_accuracy": 0.13807957917451857,
      "num_tokens": 5622451.0,
      "step": 3235
    },
    {
      "entropy": 5.908698892593383,
      "epoch": 0.2520910328729819,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004996577891877417,
      "loss": 5.8501,
      "mean_token_accuracy": 0.14421772137284278,
      "num_tokens": 5631384.0,
      "step": 3240
    },
    {
      "entropy": 5.900033235549927,
      "epoch": 0.2524800622446995,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996562594617209,
      "loss": 5.8192,
      "mean_token_accuracy": 0.14661822468042374,
      "num_tokens": 5640726.0,
      "step": 3245
    },
    {
      "entropy": 5.896815347671509,
      "epoch": 0.252869091616417,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004996547263268991,
      "loss": 5.8132,
      "mean_token_accuracy": 0.14127296954393387,
      "num_tokens": 5648948.0,
      "step": 3250
    },
    {
      "entropy": 5.814871454238892,
      "epoch": 0.2532581209881346,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996531897832993,
      "loss": 5.7584,
      "mean_token_accuracy": 0.14894756972789763,
      "num_tokens": 5657951.0,
      "step": 3255
    },
    {
      "entropy": 5.990035724639893,
      "epoch": 0.2536471503598522,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499651649830945,
      "loss": 5.8802,
      "mean_token_accuracy": 0.14824058637022972,
      "num_tokens": 5667476.0,
      "step": 3260
    },
    {
      "entropy": 5.921661520004273,
      "epoch": 0.2540361797315697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996501064698593,
      "loss": 5.893,
      "mean_token_accuracy": 0.14198158904910088,
      "num_tokens": 5676627.0,
      "step": 3265
    },
    {
      "entropy": 5.8409010887146,
      "epoch": 0.2544252091032873,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499648559700066,
      "loss": 5.8149,
      "mean_token_accuracy": 0.14856610149145127,
      "num_tokens": 5684376.0,
      "step": 3270
    },
    {
      "entropy": 5.981685924530029,
      "epoch": 0.25481423847500484,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996470095215884,
      "loss": 5.9791,
      "mean_token_accuracy": 0.13242869153618814,
      "num_tokens": 5693629.0,
      "step": 3275
    },
    {
      "entropy": 5.9755712985992435,
      "epoch": 0.2552032678467224,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996454559344498,
      "loss": 5.8587,
      "mean_token_accuracy": 0.14427479952573777,
      "num_tokens": 5702010.0,
      "step": 3280
    },
    {
      "entropy": 5.843969249725342,
      "epoch": 0.25559229721844,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996438989386741,
      "loss": 5.7615,
      "mean_token_accuracy": 0.1483264982700348,
      "num_tokens": 5710823.0,
      "step": 3285
    },
    {
      "entropy": 5.92636137008667,
      "epoch": 0.25598132659015754,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996423385342847,
      "loss": 5.814,
      "mean_token_accuracy": 0.15001679211854935,
      "num_tokens": 5719389.0,
      "step": 3290
    },
    {
      "entropy": 5.908800363540649,
      "epoch": 0.25637035596187513,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996407747213055,
      "loss": 5.8199,
      "mean_token_accuracy": 0.13999297171831132,
      "num_tokens": 5727327.0,
      "step": 3295
    },
    {
      "entropy": 5.854852104187012,
      "epoch": 0.25675938533359266,
      "grad_norm": 1.078125,
      "learning_rate": 0.00049963920749976,
      "loss": 5.708,
      "mean_token_accuracy": 0.1529916137456894,
      "num_tokens": 5735886.0,
      "step": 3300
    },
    {
      "entropy": 5.847649335861206,
      "epoch": 0.25714841470531025,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996376368696721,
      "loss": 5.7647,
      "mean_token_accuracy": 0.14711963534355163,
      "num_tokens": 5744656.0,
      "step": 3305
    },
    {
      "entropy": 5.93619327545166,
      "epoch": 0.25753744407702783,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996360628310656,
      "loss": 5.9129,
      "mean_token_accuracy": 0.14037762954831123,
      "num_tokens": 5753026.0,
      "step": 3310
    },
    {
      "entropy": 5.907909679412842,
      "epoch": 0.25792647344874536,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996344853839644,
      "loss": 5.8486,
      "mean_token_accuracy": 0.14408993124961852,
      "num_tokens": 5762771.0,
      "step": 3315
    },
    {
      "entropy": 5.8601157665252686,
      "epoch": 0.25831550282046295,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996329045283924,
      "loss": 5.8555,
      "mean_token_accuracy": 0.14458582401275635,
      "num_tokens": 5771151.0,
      "step": 3320
    },
    {
      "entropy": 5.906837892532349,
      "epoch": 0.25870453219218054,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996313202643737,
      "loss": 5.7554,
      "mean_token_accuracy": 0.14421480894088745,
      "num_tokens": 5780648.0,
      "step": 3325
    },
    {
      "entropy": 5.84664740562439,
      "epoch": 0.25909356156389807,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996297325919321,
      "loss": 5.7304,
      "mean_token_accuracy": 0.14251815006136895,
      "num_tokens": 5789698.0,
      "step": 3330
    },
    {
      "entropy": 5.873911905288696,
      "epoch": 0.25948259093561565,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996281415110919,
      "loss": 5.9253,
      "mean_token_accuracy": 0.13928515315055848,
      "num_tokens": 5799218.0,
      "step": 3335
    },
    {
      "entropy": 5.826435613632202,
      "epoch": 0.2598716203073332,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996265470218773,
      "loss": 5.7899,
      "mean_token_accuracy": 0.14524126648902894,
      "num_tokens": 5808455.0,
      "step": 3340
    },
    {
      "entropy": 5.909103012084961,
      "epoch": 0.26026064967905077,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996249491243122,
      "loss": 5.826,
      "mean_token_accuracy": 0.144374630600214,
      "num_tokens": 5816464.0,
      "step": 3345
    },
    {
      "entropy": 5.935818672180176,
      "epoch": 0.26064967905076836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996233478184211,
      "loss": 5.7894,
      "mean_token_accuracy": 0.1422273501753807,
      "num_tokens": 5825188.0,
      "step": 3350
    },
    {
      "entropy": 5.828934192657471,
      "epoch": 0.2610387084224859,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004996217431042282,
      "loss": 5.8351,
      "mean_token_accuracy": 0.14534164518117904,
      "num_tokens": 5834150.0,
      "step": 3355
    },
    {
      "entropy": 5.849087810516357,
      "epoch": 0.2614277377942035,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499620134981758,
      "loss": 5.7045,
      "mean_token_accuracy": 0.15329001247882842,
      "num_tokens": 5842801.0,
      "step": 3360
    },
    {
      "entropy": 5.8749371528625485,
      "epoch": 0.261816767165921,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996185234510346,
      "loss": 5.7265,
      "mean_token_accuracy": 0.1523670181632042,
      "num_tokens": 5851547.0,
      "step": 3365
    },
    {
      "entropy": 5.8476887226104735,
      "epoch": 0.2622057965376386,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996169085120828,
      "loss": 5.8813,
      "mean_token_accuracy": 0.1440037742257118,
      "num_tokens": 5860430.0,
      "step": 3370
    },
    {
      "entropy": 5.850446796417236,
      "epoch": 0.2625948259093562,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996152901649268,
      "loss": 5.7868,
      "mean_token_accuracy": 0.1435166209936142,
      "num_tokens": 5869648.0,
      "step": 3375
    },
    {
      "entropy": 5.975506258010864,
      "epoch": 0.2629838552810737,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996136684095913,
      "loss": 5.7771,
      "mean_token_accuracy": 0.14979472607374192,
      "num_tokens": 5878606.0,
      "step": 3380
    },
    {
      "entropy": 5.846467924118042,
      "epoch": 0.2633728846527913,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996120432461009,
      "loss": 5.8541,
      "mean_token_accuracy": 0.14691687375307083,
      "num_tokens": 5886772.0,
      "step": 3385
    },
    {
      "entropy": 5.876380681991577,
      "epoch": 0.2637619140245088,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996104146744804,
      "loss": 5.7607,
      "mean_token_accuracy": 0.14366796389222145,
      "num_tokens": 5895007.0,
      "step": 3390
    },
    {
      "entropy": 5.941224193572998,
      "epoch": 0.2641509433962264,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996087826947541,
      "loss": 5.8807,
      "mean_token_accuracy": 0.13762849643826486,
      "num_tokens": 5904249.0,
      "step": 3395
    },
    {
      "entropy": 5.789490461349487,
      "epoch": 0.264539972767944,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996071473069471,
      "loss": 5.8617,
      "mean_token_accuracy": 0.14139924868941306,
      "num_tokens": 5911416.0,
      "step": 3400
    },
    {
      "entropy": 5.958128499984741,
      "epoch": 0.26492900213966153,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996055085110842,
      "loss": 5.8788,
      "mean_token_accuracy": 0.1421840764582157,
      "num_tokens": 5920370.0,
      "step": 3405
    },
    {
      "entropy": 5.946047925949097,
      "epoch": 0.2653180315113791,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996038663071902,
      "loss": 5.8511,
      "mean_token_accuracy": 0.14540338441729544,
      "num_tokens": 5928884.0,
      "step": 3410
    },
    {
      "entropy": 5.878864431381226,
      "epoch": 0.26570706088309665,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996022206952901,
      "loss": 5.8722,
      "mean_token_accuracy": 0.1360812932252884,
      "num_tokens": 5938456.0,
      "step": 3415
    },
    {
      "entropy": 5.998233032226563,
      "epoch": 0.26609609025481423,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996005716754086,
      "loss": 5.8857,
      "mean_token_accuracy": 0.1437491923570633,
      "num_tokens": 5947157.0,
      "step": 3420
    },
    {
      "entropy": 5.898933410644531,
      "epoch": 0.2664851196265318,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499598919247571,
      "loss": 5.761,
      "mean_token_accuracy": 0.15353262722492217,
      "num_tokens": 5956084.0,
      "step": 3425
    },
    {
      "entropy": 5.8378888130187985,
      "epoch": 0.26687414899824935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995972634118023,
      "loss": 5.799,
      "mean_token_accuracy": 0.14803935661911966,
      "num_tokens": 5964782.0,
      "step": 3430
    },
    {
      "entropy": 5.9612573146820065,
      "epoch": 0.26726317836996694,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995956041681277,
      "loss": 5.9992,
      "mean_token_accuracy": 0.13573580160737037,
      "num_tokens": 5974013.0,
      "step": 3435
    },
    {
      "entropy": 5.947599411010742,
      "epoch": 0.2676522077416845,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995939415165722,
      "loss": 5.8567,
      "mean_token_accuracy": 0.14528272598981856,
      "num_tokens": 5983351.0,
      "step": 3440
    },
    {
      "entropy": 5.836666822433472,
      "epoch": 0.26804123711340205,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004995922754571611,
      "loss": 5.6991,
      "mean_token_accuracy": 0.1511104792356491,
      "num_tokens": 5992084.0,
      "step": 3445
    },
    {
      "entropy": 5.743826961517334,
      "epoch": 0.26843026648511964,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995906059899197,
      "loss": 5.7129,
      "mean_token_accuracy": 0.14986460357904435,
      "num_tokens": 6001043.0,
      "step": 3450
    },
    {
      "entropy": 5.8271872997283936,
      "epoch": 0.26881929585683717,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004995889331148733,
      "loss": 5.7388,
      "mean_token_accuracy": 0.1417159467935562,
      "num_tokens": 6009620.0,
      "step": 3455
    },
    {
      "entropy": 5.798034811019898,
      "epoch": 0.26920832522855476,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004995872568320474,
      "loss": 5.7481,
      "mean_token_accuracy": 0.14688530936837196,
      "num_tokens": 6018278.0,
      "step": 3460
    },
    {
      "entropy": 5.8678796768188475,
      "epoch": 0.26959735460027234,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995855771414673,
      "loss": 5.7929,
      "mean_token_accuracy": 0.1462475687265396,
      "num_tokens": 6026741.0,
      "step": 3465
    },
    {
      "entropy": 5.977440547943115,
      "epoch": 0.2699863839719899,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995838940431585,
      "loss": 5.8137,
      "mean_token_accuracy": 0.1422023855149746,
      "num_tokens": 6035370.0,
      "step": 3470
    },
    {
      "entropy": 5.771992111206055,
      "epoch": 0.27037541334370746,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995822075371466,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1449583925306797,
      "num_tokens": 6044372.0,
      "step": 3475
    },
    {
      "entropy": 5.86431999206543,
      "epoch": 0.270764442715425,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995805176234571,
      "loss": 5.7905,
      "mean_token_accuracy": 0.14850411862134932,
      "num_tokens": 6052940.0,
      "step": 3480
    },
    {
      "entropy": 5.942733812332153,
      "epoch": 0.2711534720871426,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995788243021158,
      "loss": 5.7417,
      "mean_token_accuracy": 0.14992502331733704,
      "num_tokens": 6062048.0,
      "step": 3485
    },
    {
      "entropy": 5.786232757568359,
      "epoch": 0.27154250145886016,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995771275731483,
      "loss": 5.7535,
      "mean_token_accuracy": 0.14474001601338388,
      "num_tokens": 6071478.0,
      "step": 3490
    },
    {
      "entropy": 5.786611127853393,
      "epoch": 0.2719315308305777,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004995754274365802,
      "loss": 5.6948,
      "mean_token_accuracy": 0.15232311934232712,
      "num_tokens": 6080495.0,
      "step": 3495
    },
    {
      "entropy": 5.916463899612427,
      "epoch": 0.2723205602022953,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995737238924376,
      "loss": 5.8545,
      "mean_token_accuracy": 0.14128367006778716,
      "num_tokens": 6090141.0,
      "step": 3500
    },
    {
      "entropy": 5.9596137523651125,
      "epoch": 0.2727095895740128,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995720169407461,
      "loss": 5.8154,
      "mean_token_accuracy": 0.14249587133526803,
      "num_tokens": 6099215.0,
      "step": 3505
    },
    {
      "entropy": 5.877740001678466,
      "epoch": 0.2730986189457304,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995703065815317,
      "loss": 5.8982,
      "mean_token_accuracy": 0.13783512711524964,
      "num_tokens": 6108802.0,
      "step": 3510
    },
    {
      "entropy": 5.976074266433716,
      "epoch": 0.273487648317448,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995685928148203,
      "loss": 5.8916,
      "mean_token_accuracy": 0.13925518542528154,
      "num_tokens": 6117486.0,
      "step": 3515
    },
    {
      "entropy": 5.791416358947754,
      "epoch": 0.2738766776891655,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000499566875640638,
      "loss": 5.7068,
      "mean_token_accuracy": 0.1526058256626129,
      "num_tokens": 6125248.0,
      "step": 3520
    },
    {
      "entropy": 5.814730548858643,
      "epoch": 0.2742657070608831,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995651550590108,
      "loss": 5.7757,
      "mean_token_accuracy": 0.14225783199071884,
      "num_tokens": 6134831.0,
      "step": 3525
    },
    {
      "entropy": 5.885268449783325,
      "epoch": 0.2746547364326007,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995634310699647,
      "loss": 5.8383,
      "mean_token_accuracy": 0.1422427237033844,
      "num_tokens": 6143248.0,
      "step": 3530
    },
    {
      "entropy": 5.911109161376953,
      "epoch": 0.2750437658043182,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995617036735261,
      "loss": 5.8578,
      "mean_token_accuracy": 0.1427016444504261,
      "num_tokens": 6151768.0,
      "step": 3535
    },
    {
      "entropy": 5.817369890213013,
      "epoch": 0.2754327951760358,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995599728697211,
      "loss": 5.7362,
      "mean_token_accuracy": 0.15657260417938232,
      "num_tokens": 6160091.0,
      "step": 3540
    },
    {
      "entropy": 5.880684900283813,
      "epoch": 0.27582182454775334,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995582386585759,
      "loss": 5.8468,
      "mean_token_accuracy": 0.13603560924530028,
      "num_tokens": 6169440.0,
      "step": 3545
    },
    {
      "entropy": 5.879655599594116,
      "epoch": 0.2762108539194709,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995565010401167,
      "loss": 5.8055,
      "mean_token_accuracy": 0.14742040187120437,
      "num_tokens": 6177604.0,
      "step": 3550
    },
    {
      "entropy": 5.903544235229492,
      "epoch": 0.2765998832911885,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995547600143702,
      "loss": 5.8289,
      "mean_token_accuracy": 0.1397724151611328,
      "num_tokens": 6186252.0,
      "step": 3555
    },
    {
      "entropy": 5.854501056671142,
      "epoch": 0.27698891266290604,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995530155813625,
      "loss": 5.7552,
      "mean_token_accuracy": 0.14930889904499053,
      "num_tokens": 6194641.0,
      "step": 3560
    },
    {
      "entropy": 5.897522640228272,
      "epoch": 0.2773779420346236,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995512677411203,
      "loss": 5.899,
      "mean_token_accuracy": 0.14283184707164764,
      "num_tokens": 6203925.0,
      "step": 3565
    },
    {
      "entropy": 5.9327545166015625,
      "epoch": 0.27776697140634116,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995495164936698,
      "loss": 5.7918,
      "mean_token_accuracy": 0.1447850249707699,
      "num_tokens": 6213001.0,
      "step": 3570
    },
    {
      "entropy": 5.86150426864624,
      "epoch": 0.27815600077805874,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995477618390381,
      "loss": 5.7884,
      "mean_token_accuracy": 0.14019341319799422,
      "num_tokens": 6221754.0,
      "step": 3575
    },
    {
      "entropy": 5.8007618427276615,
      "epoch": 0.27854503014977633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995460037772513,
      "loss": 5.7046,
      "mean_token_accuracy": 0.1479455590248108,
      "num_tokens": 6229792.0,
      "step": 3580
    },
    {
      "entropy": 5.81386923789978,
      "epoch": 0.27893405952149386,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995442423083365,
      "loss": 5.7926,
      "mean_token_accuracy": 0.14790826737880708,
      "num_tokens": 6238839.0,
      "step": 3585
    },
    {
      "entropy": 5.889003133773803,
      "epoch": 0.27932308889321145,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004995424774323201,
      "loss": 5.6965,
      "mean_token_accuracy": 0.14725257232785224,
      "num_tokens": 6246939.0,
      "step": 3590
    },
    {
      "entropy": 5.8420634269714355,
      "epoch": 0.279712118264929,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499540709149229,
      "loss": 5.8416,
      "mean_token_accuracy": 0.14081727266311644,
      "num_tokens": 6255858.0,
      "step": 3595
    },
    {
      "entropy": 5.801895809173584,
      "epoch": 0.28010114763664656,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004995389374590901,
      "loss": 5.7468,
      "mean_token_accuracy": 0.14736847430467606,
      "num_tokens": 6265137.0,
      "step": 3600
    },
    {
      "entropy": 5.837459373474121,
      "epoch": 0.28049017700836415,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995371623619301,
      "loss": 5.762,
      "mean_token_accuracy": 0.14515087902545928,
      "num_tokens": 6273286.0,
      "step": 3605
    },
    {
      "entropy": 5.747409057617188,
      "epoch": 0.2808792063800817,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499535383857776,
      "loss": 5.7624,
      "mean_token_accuracy": 0.1442732572555542,
      "num_tokens": 6282477.0,
      "step": 3610
    },
    {
      "entropy": 5.910497999191284,
      "epoch": 0.28126823575179927,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499533601946655,
      "loss": 5.815,
      "mean_token_accuracy": 0.14379963874816895,
      "num_tokens": 6291756.0,
      "step": 3615
    },
    {
      "entropy": 5.8868742942810055,
      "epoch": 0.2816572651235168,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995318166285938,
      "loss": 5.8726,
      "mean_token_accuracy": 0.14048220962285995,
      "num_tokens": 6301093.0,
      "step": 3620
    },
    {
      "entropy": 5.864219665527344,
      "epoch": 0.2820462944952344,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004995300279036199,
      "loss": 5.6925,
      "mean_token_accuracy": 0.14989550560712814,
      "num_tokens": 6310379.0,
      "step": 3625
    },
    {
      "entropy": 5.8413715839385985,
      "epoch": 0.28243532386695197,
      "grad_norm": 1.140625,
      "learning_rate": 0.00049952823577176,
      "loss": 5.7247,
      "mean_token_accuracy": 0.14906262308359147,
      "num_tokens": 6318983.0,
      "step": 3630
    },
    {
      "entropy": 5.778509902954101,
      "epoch": 0.2828243532386695,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995264402330416,
      "loss": 5.7247,
      "mean_token_accuracy": 0.15141511708498,
      "num_tokens": 6327816.0,
      "step": 3635
    },
    {
      "entropy": 5.824458980560303,
      "epoch": 0.2832133826103871,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995246412874919,
      "loss": 5.8156,
      "mean_token_accuracy": 0.14643410369753837,
      "num_tokens": 6336693.0,
      "step": 3640
    },
    {
      "entropy": 5.865640687942505,
      "epoch": 0.2836024119821047,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499522838935138,
      "loss": 5.7545,
      "mean_token_accuracy": 0.14744553864002227,
      "num_tokens": 6345456.0,
      "step": 3645
    },
    {
      "entropy": 5.831955146789551,
      "epoch": 0.2839914413538222,
      "grad_norm": 1.25,
      "learning_rate": 0.0004995210331760074,
      "loss": 5.7624,
      "mean_token_accuracy": 0.15036555677652358,
      "num_tokens": 6353584.0,
      "step": 3650
    },
    {
      "entropy": 5.813273191452026,
      "epoch": 0.2843804707255398,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995192240101275,
      "loss": 5.7579,
      "mean_token_accuracy": 0.14931190088391305,
      "num_tokens": 6362258.0,
      "step": 3655
    },
    {
      "entropy": 5.788597297668457,
      "epoch": 0.2847695000972573,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995174114375258,
      "loss": 5.6938,
      "mean_token_accuracy": 0.1504710003733635,
      "num_tokens": 6371373.0,
      "step": 3660
    },
    {
      "entropy": 5.824434232711792,
      "epoch": 0.2851585294689749,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995155954582297,
      "loss": 5.7507,
      "mean_token_accuracy": 0.15634201616048812,
      "num_tokens": 6380431.0,
      "step": 3665
    },
    {
      "entropy": 5.913159561157227,
      "epoch": 0.2855475588406925,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004995137760722668,
      "loss": 5.9162,
      "mean_token_accuracy": 0.14795071631669998,
      "num_tokens": 6389305.0,
      "step": 3670
    },
    {
      "entropy": 5.898922729492187,
      "epoch": 0.28593658821241,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995119532796646,
      "loss": 5.7303,
      "mean_token_accuracy": 0.15350835621356965,
      "num_tokens": 6397065.0,
      "step": 3675
    },
    {
      "entropy": 5.886412763595581,
      "epoch": 0.2863256175841276,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499510127080451,
      "loss": 5.865,
      "mean_token_accuracy": 0.1383943647146225,
      "num_tokens": 6407053.0,
      "step": 3680
    },
    {
      "entropy": 5.9049224853515625,
      "epoch": 0.28671464695584514,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995082974746535,
      "loss": 5.8448,
      "mean_token_accuracy": 0.1437173031270504,
      "num_tokens": 6415165.0,
      "step": 3685
    },
    {
      "entropy": 5.916789102554321,
      "epoch": 0.28710367632756273,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995064644622999,
      "loss": 5.7961,
      "mean_token_accuracy": 0.148004749417305,
      "num_tokens": 6423646.0,
      "step": 3690
    },
    {
      "entropy": 5.917964601516724,
      "epoch": 0.2874927056992803,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995046280434181,
      "loss": 5.8388,
      "mean_token_accuracy": 0.14206836223602295,
      "num_tokens": 6432134.0,
      "step": 3695
    },
    {
      "entropy": 5.805697059631347,
      "epoch": 0.28788173507099785,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995027882180358,
      "loss": 5.7236,
      "mean_token_accuracy": 0.14343566447496414,
      "num_tokens": 6440456.0,
      "step": 3700
    },
    {
      "entropy": 5.831825113296508,
      "epoch": 0.28827076444271543,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995009449861812,
      "loss": 5.8005,
      "mean_token_accuracy": 0.14310564696788788,
      "num_tokens": 6448952.0,
      "step": 3705
    },
    {
      "entropy": 5.837108039855957,
      "epoch": 0.28865979381443296,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499499098347882,
      "loss": 5.6722,
      "mean_token_accuracy": 0.1598283752799034,
      "num_tokens": 6457225.0,
      "step": 3710
    },
    {
      "entropy": 5.825877094268799,
      "epoch": 0.28904882318615055,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994972483031662,
      "loss": 5.8025,
      "mean_token_accuracy": 0.14489346742630005,
      "num_tokens": 6465450.0,
      "step": 3715
    },
    {
      "entropy": 5.846252202987671,
      "epoch": 0.28943785255786814,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499495394852062,
      "loss": 5.8481,
      "mean_token_accuracy": 0.1436641126871109,
      "num_tokens": 6473736.0,
      "step": 3720
    },
    {
      "entropy": 5.85666675567627,
      "epoch": 0.28982688192958567,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994935379945977,
      "loss": 5.7896,
      "mean_token_accuracy": 0.14748275727033616,
      "num_tokens": 6482092.0,
      "step": 3725
    },
    {
      "entropy": 5.890761995315552,
      "epoch": 0.29021591130130325,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994916777308012,
      "loss": 5.7853,
      "mean_token_accuracy": 0.15057507902383804,
      "num_tokens": 6490704.0,
      "step": 3730
    },
    {
      "entropy": 5.926531505584717,
      "epoch": 0.29060494067302084,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994898140607007,
      "loss": 5.7452,
      "mean_token_accuracy": 0.15140895470976828,
      "num_tokens": 6499634.0,
      "step": 3735
    },
    {
      "entropy": 5.763195896148682,
      "epoch": 0.29099397004473837,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994879469843247,
      "loss": 5.8767,
      "mean_token_accuracy": 0.14157707393169403,
      "num_tokens": 6509261.0,
      "step": 3740
    },
    {
      "entropy": 5.91788911819458,
      "epoch": 0.29138299941645596,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994860765017013,
      "loss": 5.7395,
      "mean_token_accuracy": 0.14967054277658462,
      "num_tokens": 6517542.0,
      "step": 3745
    },
    {
      "entropy": 5.790781021118164,
      "epoch": 0.2917720287881735,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499484202612859,
      "loss": 5.7313,
      "mean_token_accuracy": 0.1393519215285778,
      "num_tokens": 6525945.0,
      "step": 3750
    },
    {
      "entropy": 5.803427267074585,
      "epoch": 0.2921610581598911,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994823253178264,
      "loss": 5.7338,
      "mean_token_accuracy": 0.1455947108566761,
      "num_tokens": 6534744.0,
      "step": 3755
    },
    {
      "entropy": 5.874011659622193,
      "epoch": 0.29255008753160866,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994804446166317,
      "loss": 5.7104,
      "mean_token_accuracy": 0.15804395973682403,
      "num_tokens": 6542640.0,
      "step": 3760
    },
    {
      "entropy": 5.830103826522827,
      "epoch": 0.2929391169033262,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994785605093035,
      "loss": 5.7759,
      "mean_token_accuracy": 0.15563864558935164,
      "num_tokens": 6551292.0,
      "step": 3765
    },
    {
      "entropy": 5.876126098632812,
      "epoch": 0.2933281462750438,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994766729958705,
      "loss": 5.8585,
      "mean_token_accuracy": 0.14294932782649994,
      "num_tokens": 6560760.0,
      "step": 3770
    },
    {
      "entropy": 5.939873170852661,
      "epoch": 0.2937171756467613,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994747820763612,
      "loss": 5.8672,
      "mean_token_accuracy": 0.14304318577051162,
      "num_tokens": 6570618.0,
      "step": 3775
    },
    {
      "entropy": 5.847003698348999,
      "epoch": 0.2941062050184789,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004994728877508046,
      "loss": 5.7315,
      "mean_token_accuracy": 0.14863356873393058,
      "num_tokens": 6579484.0,
      "step": 3780
    },
    {
      "entropy": 5.852235412597656,
      "epoch": 0.2944952343901965,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994709900192289,
      "loss": 5.7796,
      "mean_token_accuracy": 0.14796289503574372,
      "num_tokens": 6588531.0,
      "step": 3785
    },
    {
      "entropy": 5.8618687152862545,
      "epoch": 0.294884263761914,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994690888816635,
      "loss": 5.8273,
      "mean_token_accuracy": 0.15040946900844573,
      "num_tokens": 6596862.0,
      "step": 3790
    },
    {
      "entropy": 5.803598690032959,
      "epoch": 0.2952732931336316,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994671843381367,
      "loss": 5.7383,
      "mean_token_accuracy": 0.14878206551074982,
      "num_tokens": 6605594.0,
      "step": 3795
    },
    {
      "entropy": 5.934339141845703,
      "epoch": 0.29566232250534913,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004994652763886778,
      "loss": 5.7818,
      "mean_token_accuracy": 0.14747540801763534,
      "num_tokens": 6613971.0,
      "step": 3800
    },
    {
      "entropy": 5.881596517562866,
      "epoch": 0.2960513518770667,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994633650333155,
      "loss": 5.8601,
      "mean_token_accuracy": 0.13963369131088257,
      "num_tokens": 6623116.0,
      "step": 3805
    },
    {
      "entropy": 5.858915519714356,
      "epoch": 0.2964403812487843,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994614502720792,
      "loss": 5.7383,
      "mean_token_accuracy": 0.1572399616241455,
      "num_tokens": 6631910.0,
      "step": 3810
    },
    {
      "entropy": 5.847802829742432,
      "epoch": 0.29682941062050183,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994595321049974,
      "loss": 5.7597,
      "mean_token_accuracy": 0.15679761469364167,
      "num_tokens": 6640882.0,
      "step": 3815
    },
    {
      "entropy": 5.9463770389556885,
      "epoch": 0.2972184399922194,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994576105320994,
      "loss": 5.8491,
      "mean_token_accuracy": 0.1432507812976837,
      "num_tokens": 6649665.0,
      "step": 3820
    },
    {
      "entropy": 5.749441909790039,
      "epoch": 0.29760746936393695,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994556855534146,
      "loss": 5.6231,
      "mean_token_accuracy": 0.15284123718738557,
      "num_tokens": 6658725.0,
      "step": 3825
    },
    {
      "entropy": 5.689271545410156,
      "epoch": 0.29799649873565454,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994537571689719,
      "loss": 5.6734,
      "mean_token_accuracy": 0.15186043232679367,
      "num_tokens": 6667403.0,
      "step": 3830
    },
    {
      "entropy": 5.943368864059448,
      "epoch": 0.2983855281073721,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994518253788007,
      "loss": 5.7714,
      "mean_token_accuracy": 0.14333202689886093,
      "num_tokens": 6676673.0,
      "step": 3835
    },
    {
      "entropy": 5.870562648773193,
      "epoch": 0.29877455747908965,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994498901829304,
      "loss": 5.7644,
      "mean_token_accuracy": 0.1437056802213192,
      "num_tokens": 6686212.0,
      "step": 3840
    },
    {
      "entropy": 5.748106575012207,
      "epoch": 0.29916358685080724,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004994479515813903,
      "loss": 5.7866,
      "mean_token_accuracy": 0.1471409924328327,
      "num_tokens": 6694812.0,
      "step": 3845
    },
    {
      "entropy": 5.796619939804077,
      "epoch": 0.2995526162225248,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994460095742096,
      "loss": 5.7304,
      "mean_token_accuracy": 0.14804502204060555,
      "num_tokens": 6703439.0,
      "step": 3850
    },
    {
      "entropy": 5.898529577255249,
      "epoch": 0.29994164559424236,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499444064161418,
      "loss": 5.7095,
      "mean_token_accuracy": 0.14999190345406532,
      "num_tokens": 6711465.0,
      "step": 3855
    },
    {
      "entropy": 5.748019170761109,
      "epoch": 0.30033067496595994,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499442115343045,
      "loss": 5.7215,
      "mean_token_accuracy": 0.14734122455120086,
      "num_tokens": 6720526.0,
      "step": 3860
    },
    {
      "entropy": 5.8563597202301025,
      "epoch": 0.3007197043376775,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994401631191202,
      "loss": 5.921,
      "mean_token_accuracy": 0.13673450350761412,
      "num_tokens": 6728696.0,
      "step": 3865
    },
    {
      "entropy": 5.911409378051758,
      "epoch": 0.30110873370939506,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994382074896731,
      "loss": 5.7865,
      "mean_token_accuracy": 0.14312615022063255,
      "num_tokens": 6737955.0,
      "step": 3870
    },
    {
      "entropy": 5.8883514404296875,
      "epoch": 0.30149776308111265,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994362484547335,
      "loss": 5.8491,
      "mean_token_accuracy": 0.14304133355617524,
      "num_tokens": 6746413.0,
      "step": 3875
    },
    {
      "entropy": 5.817479228973388,
      "epoch": 0.3018867924528302,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004994342860143309,
      "loss": 5.7115,
      "mean_token_accuracy": 0.15062148869037628,
      "num_tokens": 6754341.0,
      "step": 3880
    },
    {
      "entropy": 5.73020977973938,
      "epoch": 0.30227582182454776,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994323201684953,
      "loss": 5.5864,
      "mean_token_accuracy": 0.1590082347393036,
      "num_tokens": 6762623.0,
      "step": 3885
    },
    {
      "entropy": 5.787652492523193,
      "epoch": 0.3026648511962653,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994303509172566,
      "loss": 5.7336,
      "mean_token_accuracy": 0.15008426159620286,
      "num_tokens": 6771270.0,
      "step": 3890
    },
    {
      "entropy": 5.886916542053223,
      "epoch": 0.3030538805679829,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994283782606444,
      "loss": 5.7834,
      "mean_token_accuracy": 0.14763306677341462,
      "num_tokens": 6780292.0,
      "step": 3895
    },
    {
      "entropy": 5.848605108261109,
      "epoch": 0.30344290993970047,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994264021986888,
      "loss": 5.6588,
      "mean_token_accuracy": 0.15541523694992065,
      "num_tokens": 6788580.0,
      "step": 3900
    },
    {
      "entropy": 5.682632970809936,
      "epoch": 0.303831939311418,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004994244227314197,
      "loss": 5.6741,
      "mean_token_accuracy": 0.1494122788310051,
      "num_tokens": 6796801.0,
      "step": 3905
    },
    {
      "entropy": 5.828753995895386,
      "epoch": 0.3042209686831356,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994224398588672,
      "loss": 5.6797,
      "mean_token_accuracy": 0.15172239989042283,
      "num_tokens": 6805630.0,
      "step": 3910
    },
    {
      "entropy": 5.685383987426758,
      "epoch": 0.3046099980548531,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994204535810615,
      "loss": 5.6544,
      "mean_token_accuracy": 0.15218328535556794,
      "num_tokens": 6814316.0,
      "step": 3915
    },
    {
      "entropy": 5.938894605636596,
      "epoch": 0.3049990274265707,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004994184638980326,
      "loss": 5.8338,
      "mean_token_accuracy": 0.1421108901500702,
      "num_tokens": 6823224.0,
      "step": 3920
    },
    {
      "entropy": 5.898503065109253,
      "epoch": 0.3053880567982883,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994164708098107,
      "loss": 5.8592,
      "mean_token_accuracy": 0.14033436849713327,
      "num_tokens": 6832104.0,
      "step": 3925
    },
    {
      "entropy": 5.834832239151001,
      "epoch": 0.3057770861700058,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499414474316426,
      "loss": 5.7494,
      "mean_token_accuracy": 0.14646502137184142,
      "num_tokens": 6841208.0,
      "step": 3930
    },
    {
      "entropy": 5.678035545349121,
      "epoch": 0.3061661155417234,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004994124744179088,
      "loss": 5.5608,
      "mean_token_accuracy": 0.15826275646686555,
      "num_tokens": 6849622.0,
      "step": 3935
    },
    {
      "entropy": 5.846903562545776,
      "epoch": 0.306555144913441,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994104711142896,
      "loss": 5.7712,
      "mean_token_accuracy": 0.1493185803294182,
      "num_tokens": 6857355.0,
      "step": 3940
    },
    {
      "entropy": 5.85138258934021,
      "epoch": 0.3069441742851585,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994084644055986,
      "loss": 5.7778,
      "mean_token_accuracy": 0.15036529302597046,
      "num_tokens": 6866173.0,
      "step": 3945
    },
    {
      "entropy": 5.803561782836914,
      "epoch": 0.3073332036568761,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004994064542918664,
      "loss": 5.775,
      "mean_token_accuracy": 0.1481854423880577,
      "num_tokens": 6874438.0,
      "step": 3950
    },
    {
      "entropy": 5.844762468338013,
      "epoch": 0.30772223302859364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994044407731235,
      "loss": 5.7668,
      "mean_token_accuracy": 0.14867377132177353,
      "num_tokens": 6882790.0,
      "step": 3955
    },
    {
      "entropy": 5.831721496582031,
      "epoch": 0.3081112624003112,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994024238494002,
      "loss": 5.7955,
      "mean_token_accuracy": 0.14073797017335893,
      "num_tokens": 6891375.0,
      "step": 3960
    },
    {
      "entropy": 5.749059724807739,
      "epoch": 0.3085002917720288,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994004035207274,
      "loss": 5.7282,
      "mean_token_accuracy": 0.14876208826899529,
      "num_tokens": 6899510.0,
      "step": 3965
    },
    {
      "entropy": 5.898362636566162,
      "epoch": 0.30888932114374634,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004993983797871356,
      "loss": 5.7866,
      "mean_token_accuracy": 0.14997804909944534,
      "num_tokens": 6908311.0,
      "step": 3970
    },
    {
      "entropy": 5.772304439544678,
      "epoch": 0.30927835051546393,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993963526486555,
      "loss": 5.7273,
      "mean_token_accuracy": 0.1477322421967983,
      "num_tokens": 6917064.0,
      "step": 3975
    },
    {
      "entropy": 5.775776720046997,
      "epoch": 0.30966737988718146,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499394322105318,
      "loss": 5.7212,
      "mean_token_accuracy": 0.15025561004877092,
      "num_tokens": 6925705.0,
      "step": 3980
    },
    {
      "entropy": 5.968403625488281,
      "epoch": 0.31005640925889905,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004993922881571538,
      "loss": 5.8721,
      "mean_token_accuracy": 0.14300914108753204,
      "num_tokens": 6934232.0,
      "step": 3985
    },
    {
      "entropy": 5.8764852523803714,
      "epoch": 0.31044543863061663,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993902508041939,
      "loss": 5.797,
      "mean_token_accuracy": 0.15004653483629227,
      "num_tokens": 6943143.0,
      "step": 3990
    },
    {
      "entropy": 5.761917543411255,
      "epoch": 0.31083446800233416,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499388210046469,
      "loss": 5.6544,
      "mean_token_accuracy": 0.1548629328608513,
      "num_tokens": 6951613.0,
      "step": 3995
    },
    {
      "entropy": 5.867331171035767,
      "epoch": 0.31122349737405175,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004993861658840102,
      "loss": 5.8942,
      "mean_token_accuracy": 0.14447247460484505,
      "num_tokens": 6960531.0,
      "step": 4000
    },
    {
      "entropy": 5.930006599426269,
      "epoch": 0.3116125267457693,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993841183168484,
      "loss": 5.7944,
      "mean_token_accuracy": 0.14980631619691848,
      "num_tokens": 6969211.0,
      "step": 4005
    },
    {
      "entropy": 5.91842622756958,
      "epoch": 0.31200155611748687,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993820673450148,
      "loss": 5.8139,
      "mean_token_accuracy": 0.14171599224209785,
      "num_tokens": 6978414.0,
      "step": 4010
    },
    {
      "entropy": 5.728317975997925,
      "epoch": 0.31239058548920445,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993800129685404,
      "loss": 5.5713,
      "mean_token_accuracy": 0.1550044909119606,
      "num_tokens": 6987350.0,
      "step": 4015
    },
    {
      "entropy": 5.772064590454102,
      "epoch": 0.312779614860922,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004993779551874565,
      "loss": 5.7179,
      "mean_token_accuracy": 0.1518314078450203,
      "num_tokens": 6995817.0,
      "step": 4020
    },
    {
      "entropy": 5.790110778808594,
      "epoch": 0.31316864423263957,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004993758940017943,
      "loss": 5.8044,
      "mean_token_accuracy": 0.14595166668295861,
      "num_tokens": 7004454.0,
      "step": 4025
    },
    {
      "entropy": 5.859895992279053,
      "epoch": 0.31355767360435716,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499373829411585,
      "loss": 5.7283,
      "mean_token_accuracy": 0.14616793543100357,
      "num_tokens": 7012126.0,
      "step": 4030
    },
    {
      "entropy": 5.816727113723755,
      "epoch": 0.3139467029760747,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049937176141686,
      "loss": 5.7947,
      "mean_token_accuracy": 0.14407731890678405,
      "num_tokens": 7021232.0,
      "step": 4035
    },
    {
      "entropy": 5.92659478187561,
      "epoch": 0.3143357323477923,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004993696900176506,
      "loss": 5.7776,
      "mean_token_accuracy": 0.1509588658809662,
      "num_tokens": 7029488.0,
      "step": 4040
    },
    {
      "entropy": 5.864295911788941,
      "epoch": 0.3147247617195098,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004993676152139883,
      "loss": 5.8162,
      "mean_token_accuracy": 0.1433218613266945,
      "num_tokens": 7037323.0,
      "step": 4045
    },
    {
      "entropy": 5.789323520660401,
      "epoch": 0.3151137910912274,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993655370059046,
      "loss": 5.7601,
      "mean_token_accuracy": 0.1462397038936615,
      "num_tokens": 7045754.0,
      "step": 4050
    },
    {
      "entropy": 5.808916187286377,
      "epoch": 0.315502820462945,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004993634553934309,
      "loss": 5.7059,
      "mean_token_accuracy": 0.14995122104883193,
      "num_tokens": 7054026.0,
      "step": 4055
    },
    {
      "entropy": 5.781992387771607,
      "epoch": 0.3158918498346625,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993613703765989,
      "loss": 5.7539,
      "mean_token_accuracy": 0.14648166447877883,
      "num_tokens": 7063383.0,
      "step": 4060
    },
    {
      "entropy": 5.775054311752319,
      "epoch": 0.3162808792063801,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004993592819554402,
      "loss": 5.7129,
      "mean_token_accuracy": 0.15067371279001235,
      "num_tokens": 7071952.0,
      "step": 4065
    },
    {
      "entropy": 5.856827211380005,
      "epoch": 0.3166699085780976,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004993571901299866,
      "loss": 5.7414,
      "mean_token_accuracy": 0.14603551775217055,
      "num_tokens": 7080283.0,
      "step": 4070
    },
    {
      "entropy": 5.743404960632324,
      "epoch": 0.3170589379498152,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993550949002697,
      "loss": 5.6868,
      "mean_token_accuracy": 0.1483638510107994,
      "num_tokens": 7088669.0,
      "step": 4075
    },
    {
      "entropy": 5.72527003288269,
      "epoch": 0.3174479673215328,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004993529962663213,
      "loss": 5.5895,
      "mean_token_accuracy": 0.1608485460281372,
      "num_tokens": 7097201.0,
      "step": 4080
    },
    {
      "entropy": 5.783625173568725,
      "epoch": 0.31783699669325033,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993508942281732,
      "loss": 5.7629,
      "mean_token_accuracy": 0.14638482481241227,
      "num_tokens": 7106509.0,
      "step": 4085
    },
    {
      "entropy": 5.800755357742309,
      "epoch": 0.3182260260649679,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004993487887858575,
      "loss": 5.6818,
      "mean_token_accuracy": 0.1535182848572731,
      "num_tokens": 7114550.0,
      "step": 4090
    },
    {
      "entropy": 5.863153505325317,
      "epoch": 0.31861505543668545,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499346679939406,
      "loss": 5.7892,
      "mean_token_accuracy": 0.14850987270474433,
      "num_tokens": 7122875.0,
      "step": 4095
    },
    {
      "entropy": 5.782676029205322,
      "epoch": 0.31900408480840303,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004993445676888507,
      "loss": 5.6042,
      "mean_token_accuracy": 0.15045301467180253,
      "num_tokens": 7132263.0,
      "step": 4100
    },
    {
      "entropy": 5.7514721870422365,
      "epoch": 0.3193931141801206,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993424520342236,
      "loss": 5.8134,
      "mean_token_accuracy": 0.14847530126571656,
      "num_tokens": 7141448.0,
      "step": 4105
    },
    {
      "entropy": 5.849138593673706,
      "epoch": 0.31978214355183815,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993403329755569,
      "loss": 5.8368,
      "mean_token_accuracy": 0.14440834596753122,
      "num_tokens": 7151021.0,
      "step": 4110
    },
    {
      "entropy": 5.936191082000732,
      "epoch": 0.32017117292355574,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993382105128828,
      "loss": 5.778,
      "mean_token_accuracy": 0.14660669937729837,
      "num_tokens": 7160103.0,
      "step": 4115
    },
    {
      "entropy": 5.807584238052368,
      "epoch": 0.32056020229527327,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993360846462333,
      "loss": 5.7746,
      "mean_token_accuracy": 0.14117759913206102,
      "num_tokens": 7169003.0,
      "step": 4120
    },
    {
      "entropy": 5.920692825317383,
      "epoch": 0.32094923166699085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993339553756408,
      "loss": 5.8161,
      "mean_token_accuracy": 0.15328474417328836,
      "num_tokens": 7177086.0,
      "step": 4125
    },
    {
      "entropy": 5.776644992828369,
      "epoch": 0.32133826103870844,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993318227011378,
      "loss": 5.7028,
      "mean_token_accuracy": 0.1498599775135517,
      "num_tokens": 7185772.0,
      "step": 4130
    },
    {
      "entropy": 5.870523262023926,
      "epoch": 0.32172729041042597,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004993296866227562,
      "loss": 5.8108,
      "mean_token_accuracy": 0.15271646976470948,
      "num_tokens": 7194394.0,
      "step": 4135
    },
    {
      "entropy": 5.794646072387695,
      "epoch": 0.32211631978214356,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004993275471405288,
      "loss": 5.645,
      "mean_token_accuracy": 0.1590360477566719,
      "num_tokens": 7202143.0,
      "step": 4140
    },
    {
      "entropy": 5.719142007827759,
      "epoch": 0.32250534915386114,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004993254042544879,
      "loss": 5.749,
      "mean_token_accuracy": 0.14643114656209946,
      "num_tokens": 7210645.0,
      "step": 4145
    },
    {
      "entropy": 5.813151884078979,
      "epoch": 0.3228943785255787,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499323257964666,
      "loss": 5.6965,
      "mean_token_accuracy": 0.1498257949948311,
      "num_tokens": 7218872.0,
      "step": 4150
    },
    {
      "entropy": 5.904293441772461,
      "epoch": 0.32328340789729626,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993211082710957,
      "loss": 5.8664,
      "mean_token_accuracy": 0.14466212093830108,
      "num_tokens": 7226977.0,
      "step": 4155
    },
    {
      "entropy": 5.869268417358398,
      "epoch": 0.3236724372690138,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993189551738097,
      "loss": 5.7094,
      "mean_token_accuracy": 0.14590476900339128,
      "num_tokens": 7235635.0,
      "step": 4160
    },
    {
      "entropy": 5.749515628814697,
      "epoch": 0.3240614666407314,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993167986728405,
      "loss": 5.7111,
      "mean_token_accuracy": 0.15503788292407988,
      "num_tokens": 7244506.0,
      "step": 4165
    },
    {
      "entropy": 5.847475337982178,
      "epoch": 0.32445049601244896,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499314638768221,
      "loss": 5.7244,
      "mean_token_accuracy": 0.14863126277923583,
      "num_tokens": 7252883.0,
      "step": 4170
    },
    {
      "entropy": 5.806128120422363,
      "epoch": 0.3248395253841665,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004993124754599839,
      "loss": 5.6889,
      "mean_token_accuracy": 0.14966899529099464,
      "num_tokens": 7261580.0,
      "step": 4175
    },
    {
      "entropy": 5.806477499008179,
      "epoch": 0.3252285547558841,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993103087481619,
      "loss": 5.7876,
      "mean_token_accuracy": 0.14564215391874313,
      "num_tokens": 7270346.0,
      "step": 4180
    },
    {
      "entropy": 5.881603717803955,
      "epoch": 0.3256175841276016,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004993081386327881,
      "loss": 5.8475,
      "mean_token_accuracy": 0.141534461081028,
      "num_tokens": 7279393.0,
      "step": 4185
    },
    {
      "entropy": 5.794528961181641,
      "epoch": 0.3260066134993192,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004993059651138953,
      "loss": 5.6858,
      "mean_token_accuracy": 0.14800167679786683,
      "num_tokens": 7287929.0,
      "step": 4190
    },
    {
      "entropy": 5.861797761917114,
      "epoch": 0.3263956428710368,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993037881915165,
      "loss": 5.736,
      "mean_token_accuracy": 0.15219995230436326,
      "num_tokens": 7296336.0,
      "step": 4195
    },
    {
      "entropy": 5.664977073669434,
      "epoch": 0.3267846722427543,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004993016078656847,
      "loss": 5.6228,
      "mean_token_accuracy": 0.15042226910591125,
      "num_tokens": 7304624.0,
      "step": 4200
    },
    {
      "entropy": 5.7365929126739506,
      "epoch": 0.3271737016144719,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499299424136433,
      "loss": 5.7584,
      "mean_token_accuracy": 0.15014015957713128,
      "num_tokens": 7312945.0,
      "step": 4205
    },
    {
      "entropy": 5.869129610061646,
      "epoch": 0.32756273098618943,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992972370037946,
      "loss": 5.6754,
      "mean_token_accuracy": 0.1548615500330925,
      "num_tokens": 7321273.0,
      "step": 4210
    },
    {
      "entropy": 5.86765604019165,
      "epoch": 0.327951760357907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992950464678026,
      "loss": 5.7643,
      "mean_token_accuracy": 0.14993700981140137,
      "num_tokens": 7330446.0,
      "step": 4215
    },
    {
      "entropy": 5.71111569404602,
      "epoch": 0.3283407897296246,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992928525284903,
      "loss": 5.7174,
      "mean_token_accuracy": 0.1519498571753502,
      "num_tokens": 7339282.0,
      "step": 4220
    },
    {
      "entropy": 5.820359325408935,
      "epoch": 0.32872981910134214,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499290655185891,
      "loss": 5.6791,
      "mean_token_accuracy": 0.14428175389766693,
      "num_tokens": 7347848.0,
      "step": 4225
    },
    {
      "entropy": 5.692390966415405,
      "epoch": 0.3291188484730597,
      "grad_norm": 1.125,
      "learning_rate": 0.000499288454440038,
      "loss": 5.6808,
      "mean_token_accuracy": 0.15128294825553895,
      "num_tokens": 7356678.0,
      "step": 4230
    },
    {
      "entropy": 5.804254579544067,
      "epoch": 0.3295078778447773,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992862502909645,
      "loss": 5.7397,
      "mean_token_accuracy": 0.15315497517585755,
      "num_tokens": 7365411.0,
      "step": 4235
    },
    {
      "entropy": 5.816961431503296,
      "epoch": 0.32989690721649484,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992840427387043,
      "loss": 5.6647,
      "mean_token_accuracy": 0.1589634746313095,
      "num_tokens": 7373635.0,
      "step": 4240
    },
    {
      "entropy": 5.800672721862793,
      "epoch": 0.3302859365882124,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992818317832908,
      "loss": 5.7654,
      "mean_token_accuracy": 0.14858496338129043,
      "num_tokens": 7382469.0,
      "step": 4245
    },
    {
      "entropy": 5.830825090408325,
      "epoch": 0.33067496595992996,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992796174247574,
      "loss": 5.7582,
      "mean_token_accuracy": 0.14710215330123902,
      "num_tokens": 7390712.0,
      "step": 4250
    },
    {
      "entropy": 5.846486043930054,
      "epoch": 0.33106399533164754,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992773996631378,
      "loss": 5.6819,
      "mean_token_accuracy": 0.15404242426156997,
      "num_tokens": 7398905.0,
      "step": 4255
    },
    {
      "entropy": 5.747831010818482,
      "epoch": 0.33145302470336513,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004992751784984656,
      "loss": 5.7602,
      "mean_token_accuracy": 0.1488395541906357,
      "num_tokens": 7407242.0,
      "step": 4260
    },
    {
      "entropy": 5.815032434463501,
      "epoch": 0.33184205407508266,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992729539307745,
      "loss": 5.7873,
      "mean_token_accuracy": 0.14241547212004663,
      "num_tokens": 7416489.0,
      "step": 4265
    },
    {
      "entropy": 5.956868600845337,
      "epoch": 0.33223108344680025,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992707259600984,
      "loss": 5.778,
      "mean_token_accuracy": 0.15013687759637834,
      "num_tokens": 7424507.0,
      "step": 4270
    },
    {
      "entropy": 5.733019733428955,
      "epoch": 0.3326201128185178,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004992684945864709,
      "loss": 5.7405,
      "mean_token_accuracy": 0.1416952669620514,
      "num_tokens": 7433343.0,
      "step": 4275
    },
    {
      "entropy": 5.758878135681153,
      "epoch": 0.33300914219023536,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992662598099261,
      "loss": 5.6729,
      "mean_token_accuracy": 0.15693875402212143,
      "num_tokens": 7441452.0,
      "step": 4280
    },
    {
      "entropy": 5.892730617523194,
      "epoch": 0.33339817156195295,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992640216304975,
      "loss": 5.7595,
      "mean_token_accuracy": 0.1469854936003685,
      "num_tokens": 7449541.0,
      "step": 4285
    },
    {
      "entropy": 5.803173208236695,
      "epoch": 0.3337872009336705,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004992617800482196,
      "loss": 5.7624,
      "mean_token_accuracy": 0.14484892785549164,
      "num_tokens": 7458812.0,
      "step": 4290
    },
    {
      "entropy": 5.836822128295898,
      "epoch": 0.33417623030538807,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499259535063126,
      "loss": 5.6959,
      "mean_token_accuracy": 0.14742081314325334,
      "num_tokens": 7467607.0,
      "step": 4295
    },
    {
      "entropy": 5.810816287994385,
      "epoch": 0.3345652596771056,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000499257286675251,
      "loss": 5.6385,
      "mean_token_accuracy": 0.1557489439845085,
      "num_tokens": 7476047.0,
      "step": 4300
    },
    {
      "entropy": 5.787456941604614,
      "epoch": 0.3349542890488232,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992550348846285,
      "loss": 5.6798,
      "mean_token_accuracy": 0.15409563034772872,
      "num_tokens": 7484496.0,
      "step": 4305
    },
    {
      "entropy": 5.701017713546753,
      "epoch": 0.33534331842054077,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992527796912928,
      "loss": 5.6092,
      "mean_token_accuracy": 0.1507042959332466,
      "num_tokens": 7493161.0,
      "step": 4310
    },
    {
      "entropy": 5.778288555145264,
      "epoch": 0.3357323477922583,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004992505210952782,
      "loss": 5.7633,
      "mean_token_accuracy": 0.14577817618846894,
      "num_tokens": 7502122.0,
      "step": 4315
    },
    {
      "entropy": 5.853675460815429,
      "epoch": 0.3361213771639759,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004992482590966188,
      "loss": 5.74,
      "mean_token_accuracy": 0.1466208964586258,
      "num_tokens": 7510369.0,
      "step": 4320
    },
    {
      "entropy": 5.6689918518066404,
      "epoch": 0.3365104065356934,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004992459936953489,
      "loss": 5.6101,
      "mean_token_accuracy": 0.15667367428541185,
      "num_tokens": 7519321.0,
      "step": 4325
    },
    {
      "entropy": 5.760290670394897,
      "epoch": 0.336899435907411,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992437248915032,
      "loss": 5.6671,
      "mean_token_accuracy": 0.1570081114768982,
      "num_tokens": 7528196.0,
      "step": 4330
    },
    {
      "entropy": 5.780711221694946,
      "epoch": 0.3372884652791286,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004992414526851157,
      "loss": 5.7155,
      "mean_token_accuracy": 0.14818665087223054,
      "num_tokens": 7536124.0,
      "step": 4335
    },
    {
      "entropy": 5.871149301528931,
      "epoch": 0.3376774946508461,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004992391770762212,
      "loss": 5.8176,
      "mean_token_accuracy": 0.14442213475704194,
      "num_tokens": 7544409.0,
      "step": 4340
    },
    {
      "entropy": 5.871113920211792,
      "epoch": 0.3380665240225637,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499236898064854,
      "loss": 5.7157,
      "mean_token_accuracy": 0.1509476438164711,
      "num_tokens": 7552803.0,
      "step": 4345
    },
    {
      "entropy": 5.715293931961059,
      "epoch": 0.3384555533942813,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004992346156510488,
      "loss": 5.6859,
      "mean_token_accuracy": 0.14793732464313508,
      "num_tokens": 7561909.0,
      "step": 4350
    },
    {
      "entropy": 5.809684705734253,
      "epoch": 0.3388445827659988,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992323298348403,
      "loss": 5.7338,
      "mean_token_accuracy": 0.14664253443479539,
      "num_tokens": 7571759.0,
      "step": 4355
    },
    {
      "entropy": 5.9138157844543455,
      "epoch": 0.3392336121377164,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499230040616263,
      "loss": 5.8678,
      "mean_token_accuracy": 0.14966133907437323,
      "num_tokens": 7580010.0,
      "step": 4360
    },
    {
      "entropy": 5.818178129196167,
      "epoch": 0.33962264150943394,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992277479953518,
      "loss": 5.6651,
      "mean_token_accuracy": 0.14617881402373314,
      "num_tokens": 7587841.0,
      "step": 4365
    },
    {
      "entropy": 5.751266050338745,
      "epoch": 0.34001167088115153,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992254519721414,
      "loss": 5.709,
      "mean_token_accuracy": 0.14966648817062378,
      "num_tokens": 7596187.0,
      "step": 4370
    },
    {
      "entropy": 5.704373407363891,
      "epoch": 0.3404007002528691,
      "grad_norm": 1.25,
      "learning_rate": 0.0004992231525466666,
      "loss": 5.6268,
      "mean_token_accuracy": 0.1541189044713974,
      "num_tokens": 7604361.0,
      "step": 4375
    },
    {
      "entropy": 5.8143958568573,
      "epoch": 0.34078972962458665,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992208497189624,
      "loss": 5.6999,
      "mean_token_accuracy": 0.1554698660969734,
      "num_tokens": 7613173.0,
      "step": 4380
    },
    {
      "entropy": 5.7870080947875975,
      "epoch": 0.34117875899630423,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992185434890637,
      "loss": 5.7947,
      "mean_token_accuracy": 0.14364824891090394,
      "num_tokens": 7621730.0,
      "step": 4385
    },
    {
      "entropy": 5.917753887176514,
      "epoch": 0.34156778836802176,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992162338570055,
      "loss": 5.7535,
      "mean_token_accuracy": 0.15033355951309205,
      "num_tokens": 7630362.0,
      "step": 4390
    },
    {
      "entropy": 5.800952816009522,
      "epoch": 0.34195681773973935,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004992139208228227,
      "loss": 5.7399,
      "mean_token_accuracy": 0.1459221735596657,
      "num_tokens": 7639368.0,
      "step": 4395
    },
    {
      "entropy": 5.784759378433227,
      "epoch": 0.34234584711145694,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004992116043865506,
      "loss": 5.7554,
      "mean_token_accuracy": 0.14871445298194885,
      "num_tokens": 7648399.0,
      "step": 4400
    },
    {
      "entropy": 5.789456224441528,
      "epoch": 0.34273487648317447,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004992092845482244,
      "loss": 5.5841,
      "mean_token_accuracy": 0.16029348224401474,
      "num_tokens": 7657298.0,
      "step": 4405
    },
    {
      "entropy": 5.80768232345581,
      "epoch": 0.34312390585489205,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992069613078791,
      "loss": 5.7289,
      "mean_token_accuracy": 0.1515160918235779,
      "num_tokens": 7665328.0,
      "step": 4410
    },
    {
      "entropy": 5.729858636856079,
      "epoch": 0.3435129352266096,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004992046346655499,
      "loss": 5.6733,
      "mean_token_accuracy": 0.15741576105356217,
      "num_tokens": 7674610.0,
      "step": 4415
    },
    {
      "entropy": 5.812822866439819,
      "epoch": 0.34390196459832717,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004992023046212723,
      "loss": 5.7089,
      "mean_token_accuracy": 0.14995060339570046,
      "num_tokens": 7682727.0,
      "step": 4420
    },
    {
      "entropy": 5.763069295883179,
      "epoch": 0.34429099397004476,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004991999711750816,
      "loss": 5.6729,
      "mean_token_accuracy": 0.15680503323674203,
      "num_tokens": 7691547.0,
      "step": 4425
    },
    {
      "entropy": 5.750293111801147,
      "epoch": 0.3446800233417623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991976343270133,
      "loss": 5.6278,
      "mean_token_accuracy": 0.14828511476516723,
      "num_tokens": 7699987.0,
      "step": 4430
    },
    {
      "entropy": 5.738767433166504,
      "epoch": 0.3450690527134799,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991952940771026,
      "loss": 5.6893,
      "mean_token_accuracy": 0.15360093712806702,
      "num_tokens": 7708397.0,
      "step": 4435
    },
    {
      "entropy": 5.768502807617187,
      "epoch": 0.34545808208519746,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991929504253852,
      "loss": 5.6858,
      "mean_token_accuracy": 0.14866187125444413,
      "num_tokens": 7716795.0,
      "step": 4440
    },
    {
      "entropy": 5.771968984603882,
      "epoch": 0.345847111456915,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004991906033718966,
      "loss": 5.6996,
      "mean_token_accuracy": 0.15170574188232422,
      "num_tokens": 7725663.0,
      "step": 4445
    },
    {
      "entropy": 5.784630203247071,
      "epoch": 0.3462361408286326,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991882529166724,
      "loss": 5.811,
      "mean_token_accuracy": 0.15142406523227692,
      "num_tokens": 7735175.0,
      "step": 4450
    },
    {
      "entropy": 5.839637041091919,
      "epoch": 0.3466251702003501,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991858990597483,
      "loss": 5.6743,
      "mean_token_accuracy": 0.15666035413742066,
      "num_tokens": 7744041.0,
      "step": 4455
    },
    {
      "entropy": 5.816845464706421,
      "epoch": 0.3470141995720677,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991835418011601,
      "loss": 5.7386,
      "mean_token_accuracy": 0.14877310395240784,
      "num_tokens": 7752168.0,
      "step": 4460
    },
    {
      "entropy": 5.776437044143677,
      "epoch": 0.3474032289437853,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004991811811409434,
      "loss": 5.6722,
      "mean_token_accuracy": 0.14695425182580948,
      "num_tokens": 7760746.0,
      "step": 4465
    },
    {
      "entropy": 5.740421104431152,
      "epoch": 0.3477922583155028,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991788170791341,
      "loss": 5.7022,
      "mean_token_accuracy": 0.14817825704813004,
      "num_tokens": 7769960.0,
      "step": 4470
    },
    {
      "entropy": 5.838571643829345,
      "epoch": 0.3481812876872204,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991764496157682,
      "loss": 5.7023,
      "mean_token_accuracy": 0.14896901845932006,
      "num_tokens": 7778475.0,
      "step": 4475
    },
    {
      "entropy": 5.7333417415618895,
      "epoch": 0.34857031705893793,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991740787508814,
      "loss": 5.5655,
      "mean_token_accuracy": 0.15769482105970384,
      "num_tokens": 7787590.0,
      "step": 4480
    },
    {
      "entropy": 5.759796142578125,
      "epoch": 0.3489593464306555,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991717044845097,
      "loss": 5.7507,
      "mean_token_accuracy": 0.14411904588341712,
      "num_tokens": 7796841.0,
      "step": 4485
    },
    {
      "entropy": 5.873869132995606,
      "epoch": 0.3493483758023731,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004991693268166892,
      "loss": 5.772,
      "mean_token_accuracy": 0.14633407816290855,
      "num_tokens": 7804889.0,
      "step": 4490
    },
    {
      "entropy": 5.843438386917114,
      "epoch": 0.34973740517409063,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991669457474559,
      "loss": 5.7387,
      "mean_token_accuracy": 0.1510257452726364,
      "num_tokens": 7813185.0,
      "step": 4495
    },
    {
      "entropy": 5.807561445236206,
      "epoch": 0.3501264345458082,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991645612768461,
      "loss": 5.7285,
      "mean_token_accuracy": 0.14689486026763915,
      "num_tokens": 7821630.0,
      "step": 4500
    },
    {
      "entropy": 5.8154500961303714,
      "epoch": 0.35051546391752575,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004991621734048958,
      "loss": 5.7572,
      "mean_token_accuracy": 0.1546427644789219,
      "num_tokens": 7829941.0,
      "step": 4505
    },
    {
      "entropy": 5.773974895477295,
      "epoch": 0.35090449328924334,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991597821316415,
      "loss": 5.7006,
      "mean_token_accuracy": 0.15282674133777618,
      "num_tokens": 7838515.0,
      "step": 4510
    },
    {
      "entropy": 5.813479900360107,
      "epoch": 0.3512935226609609,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004991573874571192,
      "loss": 5.7072,
      "mean_token_accuracy": 0.15503425449132918,
      "num_tokens": 7846575.0,
      "step": 4515
    },
    {
      "entropy": 5.666828250885009,
      "epoch": 0.35168255203267845,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004991549893813653,
      "loss": 5.5981,
      "mean_token_accuracy": 0.15574174076318742,
      "num_tokens": 7854970.0,
      "step": 4520
    },
    {
      "entropy": 5.765780401229859,
      "epoch": 0.35207158140439604,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991525879044162,
      "loss": 5.6803,
      "mean_token_accuracy": 0.15016104951500892,
      "num_tokens": 7863293.0,
      "step": 4525
    },
    {
      "entropy": 5.802773666381836,
      "epoch": 0.35246061077611357,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004991501830263084,
      "loss": 5.6825,
      "mean_token_accuracy": 0.15348625034093857,
      "num_tokens": 7871029.0,
      "step": 4530
    },
    {
      "entropy": 5.68434247970581,
      "epoch": 0.35284964014783116,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991477747470783,
      "loss": 5.544,
      "mean_token_accuracy": 0.1587501734495163,
      "num_tokens": 7879797.0,
      "step": 4535
    },
    {
      "entropy": 5.910399723052978,
      "epoch": 0.35323866951954874,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991453630667625,
      "loss": 5.8485,
      "mean_token_accuracy": 0.14082130119204522,
      "num_tokens": 7888600.0,
      "step": 4540
    },
    {
      "entropy": 5.779889965057373,
      "epoch": 0.3536276988912663,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991429479853976,
      "loss": 5.6786,
      "mean_token_accuracy": 0.15389400273561477,
      "num_tokens": 7897794.0,
      "step": 4545
    },
    {
      "entropy": 5.764513349533081,
      "epoch": 0.35401672826298386,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991405295030202,
      "loss": 5.6588,
      "mean_token_accuracy": 0.15605125278234483,
      "num_tokens": 7905822.0,
      "step": 4550
    },
    {
      "entropy": 5.730710124969482,
      "epoch": 0.35440575763470145,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991381076196671,
      "loss": 5.7318,
      "mean_token_accuracy": 0.15439019799232484,
      "num_tokens": 7914560.0,
      "step": 4555
    },
    {
      "entropy": 5.696042776107788,
      "epoch": 0.354794787006419,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004991356823353748,
      "loss": 5.5761,
      "mean_token_accuracy": 0.1586063250899315,
      "num_tokens": 7924058.0,
      "step": 4560
    },
    {
      "entropy": 5.7807642936706545,
      "epoch": 0.35518381637813656,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004991332536501804,
      "loss": 5.7269,
      "mean_token_accuracy": 0.1479206308722496,
      "num_tokens": 7932979.0,
      "step": 4565
    },
    {
      "entropy": 5.727989530563354,
      "epoch": 0.3555728457498541,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991308215641205,
      "loss": 5.5165,
      "mean_token_accuracy": 0.16339402496814728,
      "num_tokens": 7941357.0,
      "step": 4570
    },
    {
      "entropy": 5.728809642791748,
      "epoch": 0.3559618751215717,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991283860772323,
      "loss": 5.6756,
      "mean_token_accuracy": 0.15356432944536208,
      "num_tokens": 7949997.0,
      "step": 4575
    },
    {
      "entropy": 5.78854808807373,
      "epoch": 0.35635090449328927,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991259471895525,
      "loss": 5.6452,
      "mean_token_accuracy": 0.14733605086803436,
      "num_tokens": 7958246.0,
      "step": 4580
    },
    {
      "entropy": 5.713195323944092,
      "epoch": 0.3567399338650068,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004991235049011182,
      "loss": 5.7145,
      "mean_token_accuracy": 0.15279860645532609,
      "num_tokens": 7967176.0,
      "step": 4585
    },
    {
      "entropy": 5.802193689346313,
      "epoch": 0.3571289632367244,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004991210592119663,
      "loss": 5.5751,
      "mean_token_accuracy": 0.15998913794755937,
      "num_tokens": 7974761.0,
      "step": 4590
    },
    {
      "entropy": 5.7391304016113285,
      "epoch": 0.3575179926084419,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004991186101221342,
      "loss": 5.6562,
      "mean_token_accuracy": 0.1518123209476471,
      "num_tokens": 7983155.0,
      "step": 4595
    },
    {
      "entropy": 5.737704277038574,
      "epoch": 0.3579070219801595,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499116157631659,
      "loss": 5.6652,
      "mean_token_accuracy": 0.1524111345410347,
      "num_tokens": 7991276.0,
      "step": 4600
    },
    {
      "entropy": 5.830008125305175,
      "epoch": 0.3582960513518771,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991137017405777,
      "loss": 5.8022,
      "mean_token_accuracy": 0.14828523844480515,
      "num_tokens": 7999488.0,
      "step": 4605
    },
    {
      "entropy": 5.835680961608887,
      "epoch": 0.3586850807235946,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991112424489277,
      "loss": 5.6617,
      "mean_token_accuracy": 0.15669893324375153,
      "num_tokens": 8008553.0,
      "step": 4610
    },
    {
      "entropy": 5.698417472839355,
      "epoch": 0.3590741100953122,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991087797567462,
      "loss": 5.6325,
      "mean_token_accuracy": 0.1528215676546097,
      "num_tokens": 8017890.0,
      "step": 4615
    },
    {
      "entropy": 5.793908214569091,
      "epoch": 0.35946313946702974,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991063136640709,
      "loss": 5.7936,
      "mean_token_accuracy": 0.1451731190085411,
      "num_tokens": 8025782.0,
      "step": 4620
    },
    {
      "entropy": 5.853825378417969,
      "epoch": 0.3598521688387473,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991038441709388,
      "loss": 5.6466,
      "mean_token_accuracy": 0.15467004626989364,
      "num_tokens": 8034623.0,
      "step": 4625
    },
    {
      "entropy": 5.699706745147705,
      "epoch": 0.3602411982104649,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004991013712773877,
      "loss": 5.6651,
      "mean_token_accuracy": 0.15374641865491867,
      "num_tokens": 8043789.0,
      "step": 4630
    },
    {
      "entropy": 5.709196710586548,
      "epoch": 0.36063022758218244,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990988949834549,
      "loss": 5.6233,
      "mean_token_accuracy": 0.15785256177186965,
      "num_tokens": 8053456.0,
      "step": 4635
    },
    {
      "entropy": 5.729109573364258,
      "epoch": 0.3610192569539,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000499096415289178,
      "loss": 5.6069,
      "mean_token_accuracy": 0.15437927544116975,
      "num_tokens": 8061692.0,
      "step": 4640
    },
    {
      "entropy": 5.657801961898803,
      "epoch": 0.3614082863256176,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990939321945948,
      "loss": 5.5673,
      "mean_token_accuracy": 0.1534797206521034,
      "num_tokens": 8069564.0,
      "step": 4645
    },
    {
      "entropy": 5.781489706039428,
      "epoch": 0.36179731569733514,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004990914456997427,
      "loss": 5.6394,
      "mean_token_accuracy": 0.15054785460233688,
      "num_tokens": 8078490.0,
      "step": 4650
    },
    {
      "entropy": 5.748648023605346,
      "epoch": 0.36218634506905273,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990889558046598,
      "loss": 5.7189,
      "mean_token_accuracy": 0.1515136569738388,
      "num_tokens": 8086858.0,
      "step": 4655
    },
    {
      "entropy": 5.790530204772949,
      "epoch": 0.36257537444077026,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004990864625093836,
      "loss": 5.6815,
      "mean_token_accuracy": 0.15427908897399903,
      "num_tokens": 8094542.0,
      "step": 4660
    },
    {
      "entropy": 5.817634630203247,
      "epoch": 0.36296440381248785,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004990839658139519,
      "loss": 5.7239,
      "mean_token_accuracy": 0.14588729962706565,
      "num_tokens": 8102756.0,
      "step": 4665
    },
    {
      "entropy": 5.821946144104004,
      "epoch": 0.36335343318420543,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990814657184028,
      "loss": 5.7583,
      "mean_token_accuracy": 0.15718797445297242,
      "num_tokens": 8111258.0,
      "step": 4670
    },
    {
      "entropy": 5.854680871963501,
      "epoch": 0.36374246255592296,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004990789622227741,
      "loss": 5.8706,
      "mean_token_accuracy": 0.14053035229444505,
      "num_tokens": 8119624.0,
      "step": 4675
    },
    {
      "entropy": 5.7719175815582275,
      "epoch": 0.36413149192764055,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990764553271038,
      "loss": 5.6391,
      "mean_token_accuracy": 0.15386356711387633,
      "num_tokens": 8128156.0,
      "step": 4680
    },
    {
      "entropy": 5.709950256347656,
      "epoch": 0.3645205212993581,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990739450314299,
      "loss": 5.6797,
      "mean_token_accuracy": 0.14789464101195335,
      "num_tokens": 8136080.0,
      "step": 4685
    },
    {
      "entropy": 5.774757337570191,
      "epoch": 0.36490955067107567,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990714313357906,
      "loss": 5.6699,
      "mean_token_accuracy": 0.1537407785654068,
      "num_tokens": 8144742.0,
      "step": 4690
    },
    {
      "entropy": 5.755057859420776,
      "epoch": 0.36529858004279325,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499068914240224,
      "loss": 5.7261,
      "mean_token_accuracy": 0.1528101921081543,
      "num_tokens": 8153377.0,
      "step": 4695
    },
    {
      "entropy": 5.813622236251831,
      "epoch": 0.3656876094145108,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990663937447682,
      "loss": 5.67,
      "mean_token_accuracy": 0.15677759945392608,
      "num_tokens": 8162169.0,
      "step": 4700
    },
    {
      "entropy": 5.716111516952514,
      "epoch": 0.36607663878622837,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990638698494615,
      "loss": 5.6251,
      "mean_token_accuracy": 0.1552610233426094,
      "num_tokens": 8170727.0,
      "step": 4705
    },
    {
      "entropy": 5.643989419937133,
      "epoch": 0.3664656681579459,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990613425543423,
      "loss": 5.5737,
      "mean_token_accuracy": 0.15644574016332627,
      "num_tokens": 8179013.0,
      "step": 4710
    },
    {
      "entropy": 5.666074895858765,
      "epoch": 0.3668546975296635,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990588118594487,
      "loss": 5.5482,
      "mean_token_accuracy": 0.1601375326514244,
      "num_tokens": 8187030.0,
      "step": 4715
    },
    {
      "entropy": 5.649421739578247,
      "epoch": 0.3672437269013811,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990562777648194,
      "loss": 5.5248,
      "mean_token_accuracy": 0.15833457112312316,
      "num_tokens": 8195073.0,
      "step": 4720
    },
    {
      "entropy": 5.6958873748779295,
      "epoch": 0.3676327562730986,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004990537402704928,
      "loss": 5.6852,
      "mean_token_accuracy": 0.14870353862643243,
      "num_tokens": 8203626.0,
      "step": 4725
    },
    {
      "entropy": 5.8249259948730465,
      "epoch": 0.3680217856448162,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990511993765071,
      "loss": 5.685,
      "mean_token_accuracy": 0.15397124439477922,
      "num_tokens": 8212750.0,
      "step": 4730
    },
    {
      "entropy": 5.812510967254639,
      "epoch": 0.3684108150165337,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990486550829011,
      "loss": 5.6794,
      "mean_token_accuracy": 0.14738520756363868,
      "num_tokens": 8221542.0,
      "step": 4735
    },
    {
      "entropy": 5.628241109848022,
      "epoch": 0.3687998443882513,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990461073897134,
      "loss": 5.6136,
      "mean_token_accuracy": 0.14874156713485717,
      "num_tokens": 8230574.0,
      "step": 4740
    },
    {
      "entropy": 5.766415596008301,
      "epoch": 0.3691888737599689,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990435562969827,
      "loss": 5.7076,
      "mean_token_accuracy": 0.14979110062122344,
      "num_tokens": 8239400.0,
      "step": 4745
    },
    {
      "entropy": 5.732843399047852,
      "epoch": 0.3695779031316864,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004990410018047475,
      "loss": 5.5431,
      "mean_token_accuracy": 0.15745741575956346,
      "num_tokens": 8247537.0,
      "step": 4750
    },
    {
      "entropy": 5.717076635360717,
      "epoch": 0.369966932503404,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990384439130467,
      "loss": 5.644,
      "mean_token_accuracy": 0.15211451053619385,
      "num_tokens": 8256551.0,
      "step": 4755
    },
    {
      "entropy": 5.775771570205689,
      "epoch": 0.3703559618751216,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990358826219192,
      "loss": 5.6929,
      "mean_token_accuracy": 0.15063632577657698,
      "num_tokens": 8265492.0,
      "step": 4760
    },
    {
      "entropy": 5.7251763343811035,
      "epoch": 0.37074499124683913,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004990333179314038,
      "loss": 5.6087,
      "mean_token_accuracy": 0.15533217415213585,
      "num_tokens": 8273798.0,
      "step": 4765
    },
    {
      "entropy": 5.6684823513031,
      "epoch": 0.3711340206185567,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990307498415393,
      "loss": 5.5854,
      "mean_token_accuracy": 0.1592078447341919,
      "num_tokens": 8281998.0,
      "step": 4770
    },
    {
      "entropy": 5.748588037490845,
      "epoch": 0.37152304999027425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990281783523648,
      "loss": 5.7298,
      "mean_token_accuracy": 0.1477496527135372,
      "num_tokens": 8289833.0,
      "step": 4775
    },
    {
      "entropy": 5.728920888900757,
      "epoch": 0.37191207936199183,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990256034639192,
      "loss": 5.6865,
      "mean_token_accuracy": 0.15383536368608475,
      "num_tokens": 8297562.0,
      "step": 4780
    },
    {
      "entropy": 5.813459587097168,
      "epoch": 0.3723011087337094,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004990230251762418,
      "loss": 5.6778,
      "mean_token_accuracy": 0.1526397980749607,
      "num_tokens": 8305262.0,
      "step": 4785
    },
    {
      "entropy": 5.7271405220031735,
      "epoch": 0.37269013810542695,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990204434893713,
      "loss": 5.6038,
      "mean_token_accuracy": 0.1569250702857971,
      "num_tokens": 8313949.0,
      "step": 4790
    },
    {
      "entropy": 5.729286527633667,
      "epoch": 0.37307916747714454,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990178584033474,
      "loss": 5.6292,
      "mean_token_accuracy": 0.1523518294095993,
      "num_tokens": 8322801.0,
      "step": 4795
    },
    {
      "entropy": 5.772177505493164,
      "epoch": 0.37346819684886207,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990152699182089,
      "loss": 5.6307,
      "mean_token_accuracy": 0.15281645804643632,
      "num_tokens": 8331118.0,
      "step": 4800
    },
    {
      "entropy": 5.723559093475342,
      "epoch": 0.37385722622057965,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004990126780339953,
      "loss": 5.6824,
      "mean_token_accuracy": 0.15053532719612123,
      "num_tokens": 8340498.0,
      "step": 4805
    },
    {
      "entropy": 5.848631572723389,
      "epoch": 0.37424625559229724,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990100827507459,
      "loss": 5.8126,
      "mean_token_accuracy": 0.1469060242176056,
      "num_tokens": 8349235.0,
      "step": 4810
    },
    {
      "entropy": 5.725297355651856,
      "epoch": 0.37463528496401477,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004990074840684999,
      "loss": 5.7138,
      "mean_token_accuracy": 0.1571794256567955,
      "num_tokens": 8358484.0,
      "step": 4815
    },
    {
      "entropy": 5.828975582122803,
      "epoch": 0.37502431433573236,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004990048819872969,
      "loss": 5.6218,
      "mean_token_accuracy": 0.16150318682193757,
      "num_tokens": 8366894.0,
      "step": 4820
    },
    {
      "entropy": 5.722136926651001,
      "epoch": 0.3754133437074499,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990022765071765,
      "loss": 5.6374,
      "mean_token_accuracy": 0.15095509961247444,
      "num_tokens": 8375271.0,
      "step": 4825
    },
    {
      "entropy": 5.715645885467529,
      "epoch": 0.3758023730791675,
      "grad_norm": 1.125,
      "learning_rate": 0.000498999667628178,
      "loss": 5.585,
      "mean_token_accuracy": 0.15610362142324447,
      "num_tokens": 8383714.0,
      "step": 4830
    },
    {
      "entropy": 5.730840015411377,
      "epoch": 0.37619140245088506,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004989970553503411,
      "loss": 5.6789,
      "mean_token_accuracy": 0.15095645189285278,
      "num_tokens": 8392022.0,
      "step": 4835
    },
    {
      "entropy": 5.797322225570679,
      "epoch": 0.3765804318226026,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004989944396737054,
      "loss": 5.6448,
      "mean_token_accuracy": 0.15082373172044755,
      "num_tokens": 8401282.0,
      "step": 4840
    },
    {
      "entropy": 5.706689310073853,
      "epoch": 0.3769694611943202,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004989918205983106,
      "loss": 5.7247,
      "mean_token_accuracy": 0.1473517321050167,
      "num_tokens": 8411147.0,
      "step": 4845
    },
    {
      "entropy": 5.9045275211334225,
      "epoch": 0.37735849056603776,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004989891981241964,
      "loss": 5.7155,
      "mean_token_accuracy": 0.15281478613615035,
      "num_tokens": 8420434.0,
      "step": 4850
    },
    {
      "entropy": 5.711598110198975,
      "epoch": 0.3777475199377553,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004989865722514027,
      "loss": 5.6868,
      "mean_token_accuracy": 0.15038367509841918,
      "num_tokens": 8430022.0,
      "step": 4855
    },
    {
      "entropy": 5.566115379333496,
      "epoch": 0.3781365493094729,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004989839429799692,
      "loss": 5.4688,
      "mean_token_accuracy": 0.16414596736431122,
      "num_tokens": 8438898.0,
      "step": 4860
    },
    {
      "entropy": 5.673145532608032,
      "epoch": 0.3785255786811904,
      "grad_norm": 1.3125,
      "learning_rate": 0.000498981310309936,
      "loss": 5.5519,
      "mean_token_accuracy": 0.1613315686583519,
      "num_tokens": 8446931.0,
      "step": 4865
    },
    {
      "entropy": 5.75251350402832,
      "epoch": 0.378914608052908,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004989786742413428,
      "loss": 5.6516,
      "mean_token_accuracy": 0.15293653607368468,
      "num_tokens": 8455515.0,
      "step": 4870
    },
    {
      "entropy": 5.7446619987487795,
      "epoch": 0.3793036374246256,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004989760347742298,
      "loss": 5.6446,
      "mean_token_accuracy": 0.14837552085518838,
      "num_tokens": 8464135.0,
      "step": 4875
    },
    {
      "entropy": 5.800141620635986,
      "epoch": 0.3796926667963431,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004989733919086369,
      "loss": 5.7299,
      "mean_token_accuracy": 0.1505505695939064,
      "num_tokens": 8473222.0,
      "step": 4880
    },
    {
      "entropy": 5.756816864013672,
      "epoch": 0.3800816961680607,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989707456446043,
      "loss": 5.7194,
      "mean_token_accuracy": 0.15530014112591745,
      "num_tokens": 8481966.0,
      "step": 4885
    },
    {
      "entropy": 5.760518026351929,
      "epoch": 0.38047072553977823,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989680959821721,
      "loss": 5.7246,
      "mean_token_accuracy": 0.14595063030719757,
      "num_tokens": 8491276.0,
      "step": 4890
    },
    {
      "entropy": 5.801582527160645,
      "epoch": 0.3808597549114958,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004989654429213805,
      "loss": 5.6932,
      "mean_token_accuracy": 0.15818702578544616,
      "num_tokens": 8499926.0,
      "step": 4895
    },
    {
      "entropy": 5.729330110549927,
      "epoch": 0.3812487842832134,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004989627864622699,
      "loss": 5.5613,
      "mean_token_accuracy": 0.15492340922355652,
      "num_tokens": 8508544.0,
      "step": 4900
    },
    {
      "entropy": 5.755666065216064,
      "epoch": 0.38163781365493094,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004989601266048804,
      "loss": 5.6938,
      "mean_token_accuracy": 0.14578295424580573,
      "num_tokens": 8517326.0,
      "step": 4905
    },
    {
      "entropy": 5.713610363006592,
      "epoch": 0.3820268430266485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989574633492525,
      "loss": 5.6505,
      "mean_token_accuracy": 0.15423154979944229,
      "num_tokens": 8526218.0,
      "step": 4910
    },
    {
      "entropy": 5.832731914520264,
      "epoch": 0.38241587239836605,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004989547966954266,
      "loss": 5.703,
      "mean_token_accuracy": 0.14788499027490615,
      "num_tokens": 8534708.0,
      "step": 4915
    },
    {
      "entropy": 5.639141845703125,
      "epoch": 0.38280490177008364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498952126643443,
      "loss": 5.6392,
      "mean_token_accuracy": 0.15645275712013246,
      "num_tokens": 8543302.0,
      "step": 4920
    },
    {
      "entropy": 5.634526872634888,
      "epoch": 0.3831939311418012,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989494531933424,
      "loss": 5.5629,
      "mean_token_accuracy": 0.1510437771677971,
      "num_tokens": 8552586.0,
      "step": 4925
    },
    {
      "entropy": 5.693597841262817,
      "epoch": 0.38358296051351876,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004989467763451652,
      "loss": 5.5476,
      "mean_token_accuracy": 0.15474309772253036,
      "num_tokens": 8561720.0,
      "step": 4930
    },
    {
      "entropy": 5.704194402694702,
      "epoch": 0.38397198988523634,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004989440960989523,
      "loss": 5.5386,
      "mean_token_accuracy": 0.15864376723766327,
      "num_tokens": 8570510.0,
      "step": 4935
    },
    {
      "entropy": 5.695986223220825,
      "epoch": 0.3843610192569539,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000498941412454744,
      "loss": 5.6798,
      "mean_token_accuracy": 0.154207943379879,
      "num_tokens": 8579986.0,
      "step": 4940
    },
    {
      "entropy": 5.69583683013916,
      "epoch": 0.38475004862867146,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004989387254125813,
      "loss": 5.634,
      "mean_token_accuracy": 0.1502620428800583,
      "num_tokens": 8588507.0,
      "step": 4945
    },
    {
      "entropy": 5.744803714752197,
      "epoch": 0.38513907800038905,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004989360349725049,
      "loss": 5.5551,
      "mean_token_accuracy": 0.1549387276172638,
      "num_tokens": 8596651.0,
      "step": 4950
    },
    {
      "entropy": 5.704731035232544,
      "epoch": 0.3855281073721066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989333411345555,
      "loss": 5.6051,
      "mean_token_accuracy": 0.1559740722179413,
      "num_tokens": 8605080.0,
      "step": 4955
    },
    {
      "entropy": 5.644074535369873,
      "epoch": 0.38591713674382416,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004989306438987742,
      "loss": 5.5491,
      "mean_token_accuracy": 0.15472787916660308,
      "num_tokens": 8613017.0,
      "step": 4960
    },
    {
      "entropy": 5.738157081604004,
      "epoch": 0.38630616611554175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989279432652018,
      "loss": 5.7205,
      "mean_token_accuracy": 0.15107362791895868,
      "num_tokens": 8621954.0,
      "step": 4965
    },
    {
      "entropy": 5.872067451477051,
      "epoch": 0.3866951954872593,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989252392338791,
      "loss": 5.6791,
      "mean_token_accuracy": 0.15852125138044357,
      "num_tokens": 8630213.0,
      "step": 4970
    },
    {
      "entropy": 5.827118825912476,
      "epoch": 0.38708422485897687,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989225318048475,
      "loss": 5.6902,
      "mean_token_accuracy": 0.1547978088259697,
      "num_tokens": 8638360.0,
      "step": 4975
    },
    {
      "entropy": 5.739916849136352,
      "epoch": 0.3874732542306944,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989198209781478,
      "loss": 5.7395,
      "mean_token_accuracy": 0.1504393681883812,
      "num_tokens": 8647778.0,
      "step": 4980
    },
    {
      "entropy": 5.809992027282715,
      "epoch": 0.387862283602412,
      "grad_norm": 1.25,
      "learning_rate": 0.0004989171067538211,
      "loss": 5.6474,
      "mean_token_accuracy": 0.15263953059911728,
      "num_tokens": 8655584.0,
      "step": 4985
    },
    {
      "entropy": 5.710577869415284,
      "epoch": 0.38825131297412957,
      "grad_norm": 1.109375,
      "learning_rate": 0.000498914389131909,
      "loss": 5.6859,
      "mean_token_accuracy": 0.14940168410539628,
      "num_tokens": 8664159.0,
      "step": 4990
    },
    {
      "entropy": 5.716418075561523,
      "epoch": 0.3886403423458471,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989116681124523,
      "loss": 5.5888,
      "mean_token_accuracy": 0.1551659107208252,
      "num_tokens": 8672488.0,
      "step": 4995
    },
    {
      "entropy": 5.807924032211304,
      "epoch": 0.3890293717175647,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004989089436954924,
      "loss": 5.735,
      "mean_token_accuracy": 0.1466384842991829,
      "num_tokens": 8681492.0,
      "step": 5000
    },
    {
      "entropy": 5.687121295928955,
      "epoch": 0.3894184010892822,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004989062158810706,
      "loss": 5.6124,
      "mean_token_accuracy": 0.1561321273446083,
      "num_tokens": 8690453.0,
      "step": 5005
    },
    {
      "entropy": 5.823274040222168,
      "epoch": 0.3898074304609998,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004989034846692284,
      "loss": 5.7055,
      "mean_token_accuracy": 0.15221149623394012,
      "num_tokens": 8699349.0,
      "step": 5010
    },
    {
      "entropy": 5.7402672290802,
      "epoch": 0.3901964598327174,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004989007500600073,
      "loss": 5.6976,
      "mean_token_accuracy": 0.15279116183519365,
      "num_tokens": 8707512.0,
      "step": 5015
    },
    {
      "entropy": 5.817403078079224,
      "epoch": 0.3905854892044349,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988980120534486,
      "loss": 5.7166,
      "mean_token_accuracy": 0.14896553307771682,
      "num_tokens": 8715847.0,
      "step": 5020
    },
    {
      "entropy": 5.765046644210815,
      "epoch": 0.3909745185761525,
      "grad_norm": 1.171875,
      "learning_rate": 0.000498895270649594,
      "loss": 5.63,
      "mean_token_accuracy": 0.1588594913482666,
      "num_tokens": 8725548.0,
      "step": 5025
    },
    {
      "entropy": 5.6158092498779295,
      "epoch": 0.39136354794787004,
      "grad_norm": 1.171875,
      "learning_rate": 0.000498892525848485,
      "loss": 5.5736,
      "mean_token_accuracy": 0.15467785894870759,
      "num_tokens": 8733935.0,
      "step": 5030
    },
    {
      "entropy": 5.733027791976928,
      "epoch": 0.3917525773195876,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004988897776501633,
      "loss": 5.6919,
      "mean_token_accuracy": 0.15421300679445266,
      "num_tokens": 8743110.0,
      "step": 5035
    },
    {
      "entropy": 5.652926635742188,
      "epoch": 0.3921416066913052,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988870260546705,
      "loss": 5.5154,
      "mean_token_accuracy": 0.15928199589252473,
      "num_tokens": 8751166.0,
      "step": 5040
    },
    {
      "entropy": 5.7049425601959225,
      "epoch": 0.39253063606302274,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004988842710620486,
      "loss": 5.7227,
      "mean_token_accuracy": 0.1477685108780861,
      "num_tokens": 8759929.0,
      "step": 5045
    },
    {
      "entropy": 5.853726148605347,
      "epoch": 0.39291966543474033,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004988815126723391,
      "loss": 5.7228,
      "mean_token_accuracy": 0.14860548675060273,
      "num_tokens": 8768541.0,
      "step": 5050
    },
    {
      "entropy": 5.752565002441406,
      "epoch": 0.3933086948064579,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004988787508855841,
      "loss": 5.6247,
      "mean_token_accuracy": 0.15150825530290604,
      "num_tokens": 8776803.0,
      "step": 5055
    },
    {
      "entropy": 5.681401205062866,
      "epoch": 0.39369772417817545,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004988759857018253,
      "loss": 5.6202,
      "mean_token_accuracy": 0.15988384038209916,
      "num_tokens": 8784773.0,
      "step": 5060
    },
    {
      "entropy": 5.808680438995362,
      "epoch": 0.39408675354989303,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004988732171211048,
      "loss": 5.6468,
      "mean_token_accuracy": 0.15712766498327255,
      "num_tokens": 8793797.0,
      "step": 5065
    },
    {
      "entropy": 5.668197345733643,
      "epoch": 0.39447578292161056,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004988704451434644,
      "loss": 5.5289,
      "mean_token_accuracy": 0.15743918865919113,
      "num_tokens": 8802366.0,
      "step": 5070
    },
    {
      "entropy": 5.76843843460083,
      "epoch": 0.39486481229332815,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004988676697689465,
      "loss": 5.676,
      "mean_token_accuracy": 0.15063197910785675,
      "num_tokens": 8810975.0,
      "step": 5075
    },
    {
      "entropy": 5.730029964447022,
      "epoch": 0.39525384166504574,
      "grad_norm": 1.21875,
      "learning_rate": 0.000498864890997593,
      "loss": 5.7081,
      "mean_token_accuracy": 0.14807540625333787,
      "num_tokens": 8819137.0,
      "step": 5080
    },
    {
      "entropy": 5.751862573623657,
      "epoch": 0.39564287103676327,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004988621088294461,
      "loss": 5.6896,
      "mean_token_accuracy": 0.150757896900177,
      "num_tokens": 8827676.0,
      "step": 5085
    },
    {
      "entropy": 5.795054864883423,
      "epoch": 0.39603190040848085,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000498859323264548,
      "loss": 5.6405,
      "mean_token_accuracy": 0.15386464148759843,
      "num_tokens": 8836005.0,
      "step": 5090
    },
    {
      "entropy": 5.638021612167359,
      "epoch": 0.3964209297801984,
      "grad_norm": 1.203125,
      "learning_rate": 0.000498856534302941,
      "loss": 5.5501,
      "mean_token_accuracy": 0.1646844267845154,
      "num_tokens": 8844153.0,
      "step": 5095
    },
    {
      "entropy": 5.721856880187988,
      "epoch": 0.39680995915191597,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988537419446673,
      "loss": 5.6599,
      "mean_token_accuracy": 0.15477314591407776,
      "num_tokens": 8852520.0,
      "step": 5100
    },
    {
      "entropy": 5.785870218276978,
      "epoch": 0.39719898852363356,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004988509461897694,
      "loss": 5.4978,
      "mean_token_accuracy": 0.16158649921417237,
      "num_tokens": 8860609.0,
      "step": 5105
    },
    {
      "entropy": 5.681547069549561,
      "epoch": 0.3975880178953511,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988481470382897,
      "loss": 5.6596,
      "mean_token_accuracy": 0.15374966114759445,
      "num_tokens": 8869185.0,
      "step": 5110
    },
    {
      "entropy": 5.688305187225342,
      "epoch": 0.3979770472670687,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004988453444902708,
      "loss": 5.5377,
      "mean_token_accuracy": 0.15204035639762878,
      "num_tokens": 8877758.0,
      "step": 5115
    },
    {
      "entropy": 5.712265348434448,
      "epoch": 0.3983660766387862,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004988425385457549,
      "loss": 5.6976,
      "mean_token_accuracy": 0.15432764440774918,
      "num_tokens": 8887236.0,
      "step": 5120
    },
    {
      "entropy": 5.764572954177856,
      "epoch": 0.3987551060105038,
      "grad_norm": 1.375,
      "learning_rate": 0.0004988397292047848,
      "loss": 5.6586,
      "mean_token_accuracy": 0.15237996727228165,
      "num_tokens": 8895808.0,
      "step": 5125
    },
    {
      "entropy": 5.758984088897705,
      "epoch": 0.3991441353822214,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988369164674032,
      "loss": 5.6517,
      "mean_token_accuracy": 0.15155668929219246,
      "num_tokens": 8904091.0,
      "step": 5130
    },
    {
      "entropy": 5.559191513061523,
      "epoch": 0.3995331647539389,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004988341003336526,
      "loss": 5.4547,
      "mean_token_accuracy": 0.15798882991075516,
      "num_tokens": 8912046.0,
      "step": 5135
    },
    {
      "entropy": 5.7427011013031,
      "epoch": 0.3999221941256565,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004988312808035757,
      "loss": 5.6875,
      "mean_token_accuracy": 0.1444106861948967,
      "num_tokens": 8921060.0,
      "step": 5140
    },
    {
      "entropy": 5.764761924743652,
      "epoch": 0.400311223497374,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988284578772155,
      "loss": 5.6421,
      "mean_token_accuracy": 0.15285034775733947,
      "num_tokens": 8930757.0,
      "step": 5145
    },
    {
      "entropy": 5.84005560874939,
      "epoch": 0.4007002528690916,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988256315546146,
      "loss": 5.775,
      "mean_token_accuracy": 0.15048469677567483,
      "num_tokens": 8939544.0,
      "step": 5150
    },
    {
      "entropy": 5.65234785079956,
      "epoch": 0.4010892822408092,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004988228018358161,
      "loss": 5.5922,
      "mean_token_accuracy": 0.15676501244306565,
      "num_tokens": 8949347.0,
      "step": 5155
    },
    {
      "entropy": 5.734004735946655,
      "epoch": 0.40147831161252673,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004988199687208628,
      "loss": 5.6546,
      "mean_token_accuracy": 0.15331779420375824,
      "num_tokens": 8957697.0,
      "step": 5160
    },
    {
      "entropy": 5.722145748138428,
      "epoch": 0.4018673409842443,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988171322097977,
      "loss": 5.666,
      "mean_token_accuracy": 0.1463792070746422,
      "num_tokens": 8966107.0,
      "step": 5165
    },
    {
      "entropy": 5.658127212524414,
      "epoch": 0.4022563703559619,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004988142923026638,
      "loss": 5.6061,
      "mean_token_accuracy": 0.15096110850572586,
      "num_tokens": 8974128.0,
      "step": 5170
    },
    {
      "entropy": 5.703860330581665,
      "epoch": 0.40264539972767943,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004988114489995044,
      "loss": 5.5832,
      "mean_token_accuracy": 0.15542785674333573,
      "num_tokens": 8982564.0,
      "step": 5175
    },
    {
      "entropy": 5.644107627868652,
      "epoch": 0.403034429099397,
      "grad_norm": 1.125,
      "learning_rate": 0.0004988086023003624,
      "loss": 5.6091,
      "mean_token_accuracy": 0.15618483275175093,
      "num_tokens": 8990924.0,
      "step": 5180
    },
    {
      "entropy": 5.8266722679138185,
      "epoch": 0.40342345847111455,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004988057522052811,
      "loss": 5.7214,
      "mean_token_accuracy": 0.14280055314302445,
      "num_tokens": 8999827.0,
      "step": 5185
    },
    {
      "entropy": 5.740384721755982,
      "epoch": 0.40381248784283214,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988028987143037,
      "loss": 5.5516,
      "mean_token_accuracy": 0.15863237529993057,
      "num_tokens": 9007885.0,
      "step": 5190
    },
    {
      "entropy": 5.685981798171997,
      "epoch": 0.4042015172145497,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988000418274736,
      "loss": 5.6341,
      "mean_token_accuracy": 0.1547287479043007,
      "num_tokens": 9016752.0,
      "step": 5195
    },
    {
      "entropy": 5.7628861427307125,
      "epoch": 0.40459054658626725,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987971815448341,
      "loss": 5.6479,
      "mean_token_accuracy": 0.15279937982559205,
      "num_tokens": 9025547.0,
      "step": 5200
    },
    {
      "entropy": 5.736315155029297,
      "epoch": 0.40497957595798484,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004987943178664285,
      "loss": 5.6891,
      "mean_token_accuracy": 0.14981737211346627,
      "num_tokens": 9034711.0,
      "step": 5205
    },
    {
      "entropy": 5.723097658157348,
      "epoch": 0.40536860532970237,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004987914507923004,
      "loss": 5.5966,
      "mean_token_accuracy": 0.15352468937635422,
      "num_tokens": 9042565.0,
      "step": 5210
    },
    {
      "entropy": 5.658192777633667,
      "epoch": 0.40575763470141996,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004987885803224931,
      "loss": 5.6525,
      "mean_token_accuracy": 0.14747857078909873,
      "num_tokens": 9051590.0,
      "step": 5215
    },
    {
      "entropy": 5.788003396987915,
      "epoch": 0.40614666407313754,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004987857064570505,
      "loss": 5.7266,
      "mean_token_accuracy": 0.15114164650440215,
      "num_tokens": 9060415.0,
      "step": 5220
    },
    {
      "entropy": 5.685577011108398,
      "epoch": 0.4065356934448551,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004987828291960158,
      "loss": 5.5135,
      "mean_token_accuracy": 0.16069575399160385,
      "num_tokens": 9069492.0,
      "step": 5225
    },
    {
      "entropy": 5.64251971244812,
      "epoch": 0.40692472281657266,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498779948539433,
      "loss": 5.5749,
      "mean_token_accuracy": 0.15614712238311768,
      "num_tokens": 9078183.0,
      "step": 5230
    },
    {
      "entropy": 5.841134548187256,
      "epoch": 0.4073137521882902,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004987770644873455,
      "loss": 5.786,
      "mean_token_accuracy": 0.14980955868959428,
      "num_tokens": 9086661.0,
      "step": 5235
    },
    {
      "entropy": 5.732358360290528,
      "epoch": 0.4077027815600078,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004987741770397974,
      "loss": 5.5214,
      "mean_token_accuracy": 0.15990008935332298,
      "num_tokens": 9095449.0,
      "step": 5240
    },
    {
      "entropy": 5.692601490020752,
      "epoch": 0.40809181093172536,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004987712861968323,
      "loss": 5.6529,
      "mean_token_accuracy": 0.15172710865736008,
      "num_tokens": 9103347.0,
      "step": 5245
    },
    {
      "entropy": 5.628465938568115,
      "epoch": 0.4084808403034429,
      "grad_norm": 1.078125,
      "learning_rate": 0.000498768391958494,
      "loss": 5.5676,
      "mean_token_accuracy": 0.16047179102897643,
      "num_tokens": 9112382.0,
      "step": 5250
    },
    {
      "entropy": 5.741807270050049,
      "epoch": 0.4088698696751605,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004987654943248266,
      "loss": 5.5021,
      "mean_token_accuracy": 0.1566682919859886,
      "num_tokens": 9120628.0,
      "step": 5255
    },
    {
      "entropy": 5.720187997817993,
      "epoch": 0.40925889904687807,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004987625932958739,
      "loss": 5.6794,
      "mean_token_accuracy": 0.15120512545108794,
      "num_tokens": 9130280.0,
      "step": 5260
    },
    {
      "entropy": 5.857248258590698,
      "epoch": 0.4096479284185956,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004987596888716801,
      "loss": 5.7945,
      "mean_token_accuracy": 0.14930710643529893,
      "num_tokens": 9139720.0,
      "step": 5265
    },
    {
      "entropy": 5.71965537071228,
      "epoch": 0.4100369577903132,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004987567810522892,
      "loss": 5.6969,
      "mean_token_accuracy": 0.15649670362472534,
      "num_tokens": 9148537.0,
      "step": 5270
    },
    {
      "entropy": 5.788538026809692,
      "epoch": 0.4104259871620307,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004987538698377451,
      "loss": 5.7841,
      "mean_token_accuracy": 0.15080515742301942,
      "num_tokens": 9156965.0,
      "step": 5275
    },
    {
      "entropy": 5.687677955627441,
      "epoch": 0.4108150165337483,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004987509552280924,
      "loss": 5.4927,
      "mean_token_accuracy": 0.1643675222992897,
      "num_tokens": 9164982.0,
      "step": 5280
    },
    {
      "entropy": 5.7067982196807865,
      "epoch": 0.4112040459054659,
      "grad_norm": 1.21875,
      "learning_rate": 0.000498748037223375,
      "loss": 5.7084,
      "mean_token_accuracy": 0.15210769921541215,
      "num_tokens": 9173383.0,
      "step": 5285
    },
    {
      "entropy": 5.787872171401977,
      "epoch": 0.4115930752771834,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987451158236372,
      "loss": 5.7184,
      "mean_token_accuracy": 0.15715520158410073,
      "num_tokens": 9182743.0,
      "step": 5290
    },
    {
      "entropy": 5.779425477981567,
      "epoch": 0.411982104648901,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004987421910289234,
      "loss": 5.61,
      "mean_token_accuracy": 0.15556031763553618,
      "num_tokens": 9191770.0,
      "step": 5295
    },
    {
      "entropy": 5.541979122161865,
      "epoch": 0.41237113402061853,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004987392628392781,
      "loss": 5.5729,
      "mean_token_accuracy": 0.1607051208615303,
      "num_tokens": 9200885.0,
      "step": 5300
    },
    {
      "entropy": 5.896110725402832,
      "epoch": 0.4127601633923361,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004987363312547456,
      "loss": 5.6503,
      "mean_token_accuracy": 0.15051139891147614,
      "num_tokens": 9210114.0,
      "step": 5305
    },
    {
      "entropy": 5.770536279678344,
      "epoch": 0.4131491927640537,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004987333962753703,
      "loss": 5.6247,
      "mean_token_accuracy": 0.15380308628082276,
      "num_tokens": 9218608.0,
      "step": 5310
    },
    {
      "entropy": 5.705804014205933,
      "epoch": 0.41353822213577124,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004987304579011967,
      "loss": 5.6615,
      "mean_token_accuracy": 0.15385848730802537,
      "num_tokens": 9227737.0,
      "step": 5315
    },
    {
      "entropy": 5.7318719863891605,
      "epoch": 0.4139272515074888,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987275161322697,
      "loss": 5.5689,
      "mean_token_accuracy": 0.15382118821144103,
      "num_tokens": 9236589.0,
      "step": 5320
    },
    {
      "entropy": 5.665466213226319,
      "epoch": 0.41431628087920636,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004987245709686337,
      "loss": 5.6219,
      "mean_token_accuracy": 0.15973421782255173,
      "num_tokens": 9245313.0,
      "step": 5325
    },
    {
      "entropy": 5.665119791030884,
      "epoch": 0.41470531025092394,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004987216224103334,
      "loss": 5.6386,
      "mean_token_accuracy": 0.14816500768065452,
      "num_tokens": 9254001.0,
      "step": 5330
    },
    {
      "entropy": 5.834360504150391,
      "epoch": 0.41509433962264153,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004987186704574136,
      "loss": 5.7118,
      "mean_token_accuracy": 0.1468796029686928,
      "num_tokens": 9262485.0,
      "step": 5335
    },
    {
      "entropy": 5.73894944190979,
      "epoch": 0.41548336899435906,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004987157151099189,
      "loss": 5.5897,
      "mean_token_accuracy": 0.15356662422418593,
      "num_tokens": 9271490.0,
      "step": 5340
    },
    {
      "entropy": 5.6522377014160154,
      "epoch": 0.41587239836607665,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004987127563678945,
      "loss": 5.6124,
      "mean_token_accuracy": 0.1565516859292984,
      "num_tokens": 9280595.0,
      "step": 5345
    },
    {
      "entropy": 5.730470752716064,
      "epoch": 0.4162614277377942,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004987097942313852,
      "loss": 5.5446,
      "mean_token_accuracy": 0.16396524608135224,
      "num_tokens": 9289260.0,
      "step": 5350
    },
    {
      "entropy": 5.758099985122681,
      "epoch": 0.41665045710951176,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987068287004355,
      "loss": 5.7191,
      "mean_token_accuracy": 0.15302791744470595,
      "num_tokens": 9297842.0,
      "step": 5355
    },
    {
      "entropy": 5.792512607574463,
      "epoch": 0.41703948648122935,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004987038597750909,
      "loss": 5.6943,
      "mean_token_accuracy": 0.15403787642717362,
      "num_tokens": 9306533.0,
      "step": 5360
    },
    {
      "entropy": 5.735825634002685,
      "epoch": 0.4174285158529469,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004987008874553964,
      "loss": 5.5971,
      "mean_token_accuracy": 0.15138816684484482,
      "num_tokens": 9314534.0,
      "step": 5365
    },
    {
      "entropy": 5.750650882720947,
      "epoch": 0.41781754522466447,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004986979117413969,
      "loss": 5.7511,
      "mean_token_accuracy": 0.15022221952676773,
      "num_tokens": 9323179.0,
      "step": 5370
    },
    {
      "entropy": 5.689657926559448,
      "epoch": 0.41820657459638205,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004986949326331376,
      "loss": 5.5403,
      "mean_token_accuracy": 0.1646155908703804,
      "num_tokens": 9331265.0,
      "step": 5375
    },
    {
      "entropy": 5.701057577133179,
      "epoch": 0.4185956039680996,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004986919501306638,
      "loss": 5.602,
      "mean_token_accuracy": 0.1503379762172699,
      "num_tokens": 9340856.0,
      "step": 5380
    },
    {
      "entropy": 5.6915405750274655,
      "epoch": 0.41898463333981717,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986889642340207,
      "loss": 5.63,
      "mean_token_accuracy": 0.14884707257151603,
      "num_tokens": 9350728.0,
      "step": 5385
    },
    {
      "entropy": 5.803607416152954,
      "epoch": 0.4193736627115347,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004986859749432536,
      "loss": 5.7439,
      "mean_token_accuracy": 0.1483691193163395,
      "num_tokens": 9360109.0,
      "step": 5390
    },
    {
      "entropy": 5.825321292877197,
      "epoch": 0.4197626920832523,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004986829822584078,
      "loss": 5.7271,
      "mean_token_accuracy": 0.1516139253973961,
      "num_tokens": 9369101.0,
      "step": 5395
    },
    {
      "entropy": 5.772117948532104,
      "epoch": 0.4201517214549699,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004986799861795289,
      "loss": 5.6752,
      "mean_token_accuracy": 0.15280298441648482,
      "num_tokens": 9377686.0,
      "step": 5400
    },
    {
      "entropy": 5.753392028808594,
      "epoch": 0.4205407508266874,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004986769867066622,
      "loss": 5.663,
      "mean_token_accuracy": 0.154312302172184,
      "num_tokens": 9386439.0,
      "step": 5405
    },
    {
      "entropy": 5.683294582366943,
      "epoch": 0.420929780198405,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004986739838398532,
      "loss": 5.7234,
      "mean_token_accuracy": 0.1484253242611885,
      "num_tokens": 9395086.0,
      "step": 5410
    },
    {
      "entropy": 5.696198129653931,
      "epoch": 0.4213188095701225,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986709775791475,
      "loss": 5.569,
      "mean_token_accuracy": 0.15680409371852874,
      "num_tokens": 9403762.0,
      "step": 5415
    },
    {
      "entropy": 5.699469089508057,
      "epoch": 0.4217078389418401,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004986679679245907,
      "loss": 5.6157,
      "mean_token_accuracy": 0.16077837347984314,
      "num_tokens": 9412386.0,
      "step": 5420
    },
    {
      "entropy": 5.688532304763794,
      "epoch": 0.4220968683135577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004986649548762286,
      "loss": 5.566,
      "mean_token_accuracy": 0.1658398449420929,
      "num_tokens": 9420538.0,
      "step": 5425
    },
    {
      "entropy": 5.656873273849487,
      "epoch": 0.4224858976852752,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986619384341066,
      "loss": 5.5515,
      "mean_token_accuracy": 0.16055242866277694,
      "num_tokens": 9429083.0,
      "step": 5430
    },
    {
      "entropy": 5.747834634780884,
      "epoch": 0.4228749270569928,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004986589185982708,
      "loss": 5.6482,
      "mean_token_accuracy": 0.16484657377004625,
      "num_tokens": 9436892.0,
      "step": 5435
    },
    {
      "entropy": 5.733410263061524,
      "epoch": 0.42326395642871034,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004986558953687671,
      "loss": 5.6671,
      "mean_token_accuracy": 0.15706316977739335,
      "num_tokens": 9445267.0,
      "step": 5440
    },
    {
      "entropy": 5.748835802078247,
      "epoch": 0.42365298580042793,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004986528687456409,
      "loss": 5.6746,
      "mean_token_accuracy": 0.15422938019037247,
      "num_tokens": 9454292.0,
      "step": 5445
    },
    {
      "entropy": 5.657432270050049,
      "epoch": 0.4240420151721455,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004986498387289384,
      "loss": 5.5738,
      "mean_token_accuracy": 0.15715361535549163,
      "num_tokens": 9463126.0,
      "step": 5450
    },
    {
      "entropy": 5.732117414474487,
      "epoch": 0.42443104454386305,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986468053187058,
      "loss": 5.6558,
      "mean_token_accuracy": 0.15627878606319429,
      "num_tokens": 9471703.0,
      "step": 5455
    },
    {
      "entropy": 5.7013813018798825,
      "epoch": 0.42482007391558063,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986437685149887,
      "loss": 5.5405,
      "mean_token_accuracy": 0.16320954710245134,
      "num_tokens": 9480029.0,
      "step": 5460
    },
    {
      "entropy": 5.694327020645142,
      "epoch": 0.4252091032872982,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004986407283178334,
      "loss": 5.6439,
      "mean_token_accuracy": 0.15880101025104523,
      "num_tokens": 9488277.0,
      "step": 5465
    },
    {
      "entropy": 5.68750376701355,
      "epoch": 0.42559813265901575,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004986376847272861,
      "loss": 5.6981,
      "mean_token_accuracy": 0.15781251788139344,
      "num_tokens": 9496096.0,
      "step": 5470
    },
    {
      "entropy": 5.7373902797698975,
      "epoch": 0.42598716203073334,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004986346377433929,
      "loss": 5.6206,
      "mean_token_accuracy": 0.16011762917041777,
      "num_tokens": 9504653.0,
      "step": 5475
    },
    {
      "entropy": 5.759919834136963,
      "epoch": 0.42637619140245087,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004986315873661999,
      "loss": 5.5902,
      "mean_token_accuracy": 0.1531043156981468,
      "num_tokens": 9513088.0,
      "step": 5480
    },
    {
      "entropy": 5.68178882598877,
      "epoch": 0.42676522077416845,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004986285335957536,
      "loss": 5.5005,
      "mean_token_accuracy": 0.15920389890670777,
      "num_tokens": 9521714.0,
      "step": 5485
    },
    {
      "entropy": 5.625256824493408,
      "epoch": 0.42715425014588604,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004986254764321002,
      "loss": 5.5981,
      "mean_token_accuracy": 0.1587956354022026,
      "num_tokens": 9530629.0,
      "step": 5490
    },
    {
      "entropy": 5.672340631484985,
      "epoch": 0.42754327951760357,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004986224158752861,
      "loss": 5.5931,
      "mean_token_accuracy": 0.16335221081972123,
      "num_tokens": 9539293.0,
      "step": 5495
    },
    {
      "entropy": 5.606600618362426,
      "epoch": 0.42793230888932116,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004986193519253578,
      "loss": 5.563,
      "mean_token_accuracy": 0.15887336134910585,
      "num_tokens": 9547628.0,
      "step": 5500
    },
    {
      "entropy": 5.704019069671631,
      "epoch": 0.4283213382610387,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986162845823618,
      "loss": 5.5394,
      "mean_token_accuracy": 0.1612545669078827,
      "num_tokens": 9557260.0,
      "step": 5505
    },
    {
      "entropy": 5.788333702087402,
      "epoch": 0.4287103676327563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004986132138463446,
      "loss": 5.607,
      "mean_token_accuracy": 0.1461222380399704,
      "num_tokens": 9565923.0,
      "step": 5510
    },
    {
      "entropy": 5.638090944290161,
      "epoch": 0.42909939700447386,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004986101397173528,
      "loss": 5.5961,
      "mean_token_accuracy": 0.15869821310043336,
      "num_tokens": 9574373.0,
      "step": 5515
    },
    {
      "entropy": 5.75412769317627,
      "epoch": 0.4294884263761914,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000498607062195433,
      "loss": 5.6857,
      "mean_token_accuracy": 0.1535152956843376,
      "num_tokens": 9583652.0,
      "step": 5520
    },
    {
      "entropy": 5.781528186798096,
      "epoch": 0.429877455747909,
      "grad_norm": 1.1875,
      "learning_rate": 0.000498603981280632,
      "loss": 5.6521,
      "mean_token_accuracy": 0.15396686345338823,
      "num_tokens": 9592625.0,
      "step": 5525
    },
    {
      "entropy": 5.62929105758667,
      "epoch": 0.4302664851196265,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986008969729964,
      "loss": 5.5444,
      "mean_token_accuracy": 0.15661289691925048,
      "num_tokens": 9600934.0,
      "step": 5530
    },
    {
      "entropy": 5.674818420410157,
      "epoch": 0.4306555144913441,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004985978092725731,
      "loss": 5.5517,
      "mean_token_accuracy": 0.1576480060815811,
      "num_tokens": 9609481.0,
      "step": 5535
    },
    {
      "entropy": 5.6892547607421875,
      "epoch": 0.4310445438630617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000498594718179409,
      "loss": 5.5287,
      "mean_token_accuracy": 0.16367270201444625,
      "num_tokens": 9618063.0,
      "step": 5540
    },
    {
      "entropy": 5.662596416473389,
      "epoch": 0.4314335732347792,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004985916236935508,
      "loss": 5.6069,
      "mean_token_accuracy": 0.1567869372665882,
      "num_tokens": 9626144.0,
      "step": 5545
    },
    {
      "entropy": 5.608234024047851,
      "epoch": 0.4318226026064968,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004985885258150458,
      "loss": 5.557,
      "mean_token_accuracy": 0.1605278044939041,
      "num_tokens": 9635328.0,
      "step": 5550
    },
    {
      "entropy": 5.767141342163086,
      "epoch": 0.43221163197821433,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004985854245439408,
      "loss": 5.6859,
      "mean_token_accuracy": 0.14385951608419417,
      "num_tokens": 9644529.0,
      "step": 5555
    },
    {
      "entropy": 5.691818284988403,
      "epoch": 0.4326006613499319,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004985823198802827,
      "loss": 5.6672,
      "mean_token_accuracy": 0.15139155611395835,
      "num_tokens": 9653763.0,
      "step": 5560
    },
    {
      "entropy": 5.684495735168457,
      "epoch": 0.4329896907216495,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004985792118241188,
      "loss": 5.5098,
      "mean_token_accuracy": 0.16058704257011414,
      "num_tokens": 9661433.0,
      "step": 5565
    },
    {
      "entropy": 5.6012636661529545,
      "epoch": 0.43337872009336703,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004985761003754961,
      "loss": 5.5009,
      "mean_token_accuracy": 0.16053632646799088,
      "num_tokens": 9669688.0,
      "step": 5570
    },
    {
      "entropy": 5.727578449249267,
      "epoch": 0.4337677494650846,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004985729855344622,
      "loss": 5.6224,
      "mean_token_accuracy": 0.1552560567855835,
      "num_tokens": 9678403.0,
      "step": 5575
    },
    {
      "entropy": 5.799956750869751,
      "epoch": 0.4341567788368022,
      "grad_norm": 1.265625,
      "learning_rate": 0.000498569867301064,
      "loss": 5.7153,
      "mean_token_accuracy": 0.1504821889102459,
      "num_tokens": 9687504.0,
      "step": 5580
    },
    {
      "entropy": 5.787498140335083,
      "epoch": 0.43454580820851973,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004985667456753489,
      "loss": 5.5277,
      "mean_token_accuracy": 0.1625373348593712,
      "num_tokens": 9695772.0,
      "step": 5585
    },
    {
      "entropy": 5.621261739730835,
      "epoch": 0.4349348375802373,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004985636206573642,
      "loss": 5.5859,
      "mean_token_accuracy": 0.16200022250413895,
      "num_tokens": 9704150.0,
      "step": 5590
    },
    {
      "entropy": 5.5493614196777346,
      "epoch": 0.43532386695195485,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004985604922471575,
      "loss": 5.5114,
      "mean_token_accuracy": 0.1663881093263626,
      "num_tokens": 9712245.0,
      "step": 5595
    },
    {
      "entropy": 5.6750976085662845,
      "epoch": 0.43571289632367244,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004985573604447761,
      "loss": 5.6388,
      "mean_token_accuracy": 0.15445700511336327,
      "num_tokens": 9721027.0,
      "step": 5600
    },
    {
      "entropy": 5.743808555603027,
      "epoch": 0.43610192569539,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004985542252502676,
      "loss": 5.5712,
      "mean_token_accuracy": 0.1565669909119606,
      "num_tokens": 9728939.0,
      "step": 5605
    },
    {
      "entropy": 5.759398603439331,
      "epoch": 0.43649095506710756,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004985510866636796,
      "loss": 5.7077,
      "mean_token_accuracy": 0.1541939929127693,
      "num_tokens": 9737863.0,
      "step": 5610
    },
    {
      "entropy": 5.684520864486695,
      "epoch": 0.43687998443882514,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004985479446850596,
      "loss": 5.6936,
      "mean_token_accuracy": 0.14551933109760284,
      "num_tokens": 9747666.0,
      "step": 5615
    },
    {
      "entropy": 5.7133753299713135,
      "epoch": 0.4372690138105427,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004985447993144554,
      "loss": 5.5663,
      "mean_token_accuracy": 0.1538056954741478,
      "num_tokens": 9756595.0,
      "step": 5620
    },
    {
      "entropy": 5.72659740447998,
      "epoch": 0.43765804318226026,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004985416505519147,
      "loss": 5.5909,
      "mean_token_accuracy": 0.15512703582644463,
      "num_tokens": 9764747.0,
      "step": 5625
    },
    {
      "entropy": 5.634548234939575,
      "epoch": 0.43804707255397785,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004985384983974853,
      "loss": 5.5512,
      "mean_token_accuracy": 0.1591742143034935,
      "num_tokens": 9772992.0,
      "step": 5630
    },
    {
      "entropy": 5.7269854068756105,
      "epoch": 0.4384361019256954,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004985353428512148,
      "loss": 5.5239,
      "mean_token_accuracy": 0.16119461506605148,
      "num_tokens": 9781107.0,
      "step": 5635
    },
    {
      "entropy": 5.685976028442383,
      "epoch": 0.43882513129741296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004985321839131514,
      "loss": 5.6379,
      "mean_token_accuracy": 0.15349040925502777,
      "num_tokens": 9790618.0,
      "step": 5640
    },
    {
      "entropy": 5.727147150039673,
      "epoch": 0.4392141606691305,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004985290215833428,
      "loss": 5.6284,
      "mean_token_accuracy": 0.15251161977648736,
      "num_tokens": 9800147.0,
      "step": 5645
    },
    {
      "entropy": 5.7755354881286625,
      "epoch": 0.4396031900408481,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004985258558618372,
      "loss": 5.6812,
      "mean_token_accuracy": 0.15361425280570984,
      "num_tokens": 9809056.0,
      "step": 5650
    },
    {
      "entropy": 5.6240582942962645,
      "epoch": 0.43999221941256567,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004985226867486825,
      "loss": 5.5162,
      "mean_token_accuracy": 0.15899730771780013,
      "num_tokens": 9817188.0,
      "step": 5655
    },
    {
      "entropy": 5.70063624382019,
      "epoch": 0.4403812487842832,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985195142439267,
      "loss": 5.7173,
      "mean_token_accuracy": 0.1477881796658039,
      "num_tokens": 9826904.0,
      "step": 5660
    },
    {
      "entropy": 5.752689933776855,
      "epoch": 0.4407702781560008,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004985163383476181,
      "loss": 5.678,
      "mean_token_accuracy": 0.15223178938031195,
      "num_tokens": 9835937.0,
      "step": 5665
    },
    {
      "entropy": 5.704741144180298,
      "epoch": 0.44115930752771837,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004985131590598048,
      "loss": 5.5087,
      "mean_token_accuracy": 0.16645602136850357,
      "num_tokens": 9843695.0,
      "step": 5670
    },
    {
      "entropy": 5.695380735397339,
      "epoch": 0.4415483368994359,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004985099763805352,
      "loss": 5.6107,
      "mean_token_accuracy": 0.1557529628276825,
      "num_tokens": 9852693.0,
      "step": 5675
    },
    {
      "entropy": 5.63061671257019,
      "epoch": 0.4419373662711535,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004985067903098574,
      "loss": 5.5905,
      "mean_token_accuracy": 0.15602753460407257,
      "num_tokens": 9860251.0,
      "step": 5680
    },
    {
      "entropy": 5.605293035507202,
      "epoch": 0.442326395642871,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004985036008478197,
      "loss": 5.5289,
      "mean_token_accuracy": 0.15081032663583754,
      "num_tokens": 9868319.0,
      "step": 5685
    },
    {
      "entropy": 5.658757257461548,
      "epoch": 0.4427154250145886,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004985004079944707,
      "loss": 5.5591,
      "mean_token_accuracy": 0.1520108200609684,
      "num_tokens": 9877119.0,
      "step": 5690
    },
    {
      "entropy": 5.774013996124268,
      "epoch": 0.4431044543863062,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004984972117498587,
      "loss": 5.6426,
      "mean_token_accuracy": 0.15740321576595306,
      "num_tokens": 9886015.0,
      "step": 5695
    },
    {
      "entropy": 5.692266178131104,
      "epoch": 0.4434934837580237,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984940121140323,
      "loss": 5.5908,
      "mean_token_accuracy": 0.15590065643191336,
      "num_tokens": 9894829.0,
      "step": 5700
    },
    {
      "entropy": 5.881935501098633,
      "epoch": 0.4438825131297413,
      "grad_norm": 1.34375,
      "learning_rate": 0.00049849080908704,
      "loss": 5.8566,
      "mean_token_accuracy": 0.14073646143078805,
      "num_tokens": 9904840.0,
      "step": 5705
    },
    {
      "entropy": 5.82473669052124,
      "epoch": 0.44427154250145884,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984876026689302,
      "loss": 5.6914,
      "mean_token_accuracy": 0.14966955184936523,
      "num_tokens": 9913535.0,
      "step": 5710
    },
    {
      "entropy": 5.723844337463379,
      "epoch": 0.4446605718731764,
      "grad_norm": 1.125,
      "learning_rate": 0.000498484392859752,
      "loss": 5.573,
      "mean_token_accuracy": 0.1615655690431595,
      "num_tokens": 9921676.0,
      "step": 5715
    },
    {
      "entropy": 5.689015054702759,
      "epoch": 0.445049601244894,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004984811796595538,
      "loss": 5.6601,
      "mean_token_accuracy": 0.15303496792912483,
      "num_tokens": 9930321.0,
      "step": 5720
    },
    {
      "entropy": 5.747110748291016,
      "epoch": 0.44543863061661154,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004984779630683843,
      "loss": 5.5854,
      "mean_token_accuracy": 0.1555145226418972,
      "num_tokens": 9939324.0,
      "step": 5725
    },
    {
      "entropy": 5.701914501190186,
      "epoch": 0.44582765998832913,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004984747430862924,
      "loss": 5.6075,
      "mean_token_accuracy": 0.16185917109251022,
      "num_tokens": 9948603.0,
      "step": 5730
    },
    {
      "entropy": 5.572562646865845,
      "epoch": 0.44621668936004666,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984715197133271,
      "loss": 5.501,
      "mean_token_accuracy": 0.1617662340402603,
      "num_tokens": 9957165.0,
      "step": 5735
    },
    {
      "entropy": 5.7674284934997555,
      "epoch": 0.44660571873176425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004984682929495371,
      "loss": 5.6806,
      "mean_token_accuracy": 0.15358664467930794,
      "num_tokens": 9966064.0,
      "step": 5740
    },
    {
      "entropy": 5.634642744064331,
      "epoch": 0.44699474810348183,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004984650627949715,
      "loss": 5.5392,
      "mean_token_accuracy": 0.15414338409900666,
      "num_tokens": 9975146.0,
      "step": 5745
    },
    {
      "entropy": 5.681212997436523,
      "epoch": 0.44738377747519936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004984618292496792,
      "loss": 5.627,
      "mean_token_accuracy": 0.15547358840703965,
      "num_tokens": 9983725.0,
      "step": 5750
    },
    {
      "entropy": 5.781945705413818,
      "epoch": 0.44777280684691695,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004984585923137093,
      "loss": 5.6544,
      "mean_token_accuracy": 0.15781686902046205,
      "num_tokens": 9992627.0,
      "step": 5755
    },
    {
      "entropy": 5.605837392807007,
      "epoch": 0.4481618362186345,
      "grad_norm": 1.1875,
      "learning_rate": 0.000498455351987111,
      "loss": 5.4847,
      "mean_token_accuracy": 0.1617937222123146,
      "num_tokens": 10000801.0,
      "step": 5760
    },
    {
      "entropy": 5.753721618652344,
      "epoch": 0.44855086559035207,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004984521082699333,
      "loss": 5.6383,
      "mean_token_accuracy": 0.1581135392189026,
      "num_tokens": 10009421.0,
      "step": 5765
    },
    {
      "entropy": 5.682322311401367,
      "epoch": 0.44893989496206965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004984488611622256,
      "loss": 5.6355,
      "mean_token_accuracy": 0.1580435149371624,
      "num_tokens": 10018527.0,
      "step": 5770
    },
    {
      "entropy": 5.668961048126221,
      "epoch": 0.4493289243337872,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000498445610664037,
      "loss": 5.575,
      "mean_token_accuracy": 0.15585154592990874,
      "num_tokens": 10026906.0,
      "step": 5775
    },
    {
      "entropy": 5.645201253890991,
      "epoch": 0.44971795370550477,
      "grad_norm": 1.15625,
      "learning_rate": 0.000498442356775417,
      "loss": 5.5431,
      "mean_token_accuracy": 0.15861643850803375,
      "num_tokens": 10035972.0,
      "step": 5780
    },
    {
      "entropy": 5.690830707550049,
      "epoch": 0.45010698307722236,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004984390994964148,
      "loss": 5.6321,
      "mean_token_accuracy": 0.1506764143705368,
      "num_tokens": 10044544.0,
      "step": 5785
    },
    {
      "entropy": 5.784975051879883,
      "epoch": 0.4504960124489399,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00049843583882708,
      "loss": 5.6329,
      "mean_token_accuracy": 0.15364252626895905,
      "num_tokens": 10052880.0,
      "step": 5790
    },
    {
      "entropy": 5.696855068206787,
      "epoch": 0.4508850418206575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000498432574767462,
      "loss": 5.6906,
      "mean_token_accuracy": 0.1488797575235367,
      "num_tokens": 10062373.0,
      "step": 5795
    },
    {
      "entropy": 5.670724487304687,
      "epoch": 0.451274071192375,
      "grad_norm": 1.25,
      "learning_rate": 0.0004984293073176103,
      "loss": 5.5204,
      "mean_token_accuracy": 0.16071152985095977,
      "num_tokens": 10070473.0,
      "step": 5800
    },
    {
      "entropy": 5.721497535705566,
      "epoch": 0.4516631005640926,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004984260364775744,
      "loss": 5.6391,
      "mean_token_accuracy": 0.15294189602136612,
      "num_tokens": 10079857.0,
      "step": 5805
    },
    {
      "entropy": 5.674841642379761,
      "epoch": 0.4520521299358102,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000498422762247404,
      "loss": 5.5894,
      "mean_token_accuracy": 0.15466029495000838,
      "num_tokens": 10088326.0,
      "step": 5810
    },
    {
      "entropy": 5.699414777755737,
      "epoch": 0.4524411593075277,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004984194846271489,
      "loss": 5.5229,
      "mean_token_accuracy": 0.16031175404787062,
      "num_tokens": 10097208.0,
      "step": 5815
    },
    {
      "entropy": 5.710466909408569,
      "epoch": 0.4528301886792453,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004984162036168586,
      "loss": 5.6429,
      "mean_token_accuracy": 0.15294066667556763,
      "num_tokens": 10106057.0,
      "step": 5820
    },
    {
      "entropy": 5.6713909149169925,
      "epoch": 0.4532192180509628,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004984129192165831,
      "loss": 5.5578,
      "mean_token_accuracy": 0.15226832255721093,
      "num_tokens": 10115384.0,
      "step": 5825
    },
    {
      "entropy": 5.732411336898804,
      "epoch": 0.4536082474226804,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004984096314263722,
      "loss": 5.6372,
      "mean_token_accuracy": 0.15212475061416625,
      "num_tokens": 10123958.0,
      "step": 5830
    },
    {
      "entropy": 5.700859642028808,
      "epoch": 0.453997276794398,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004984063402462757,
      "loss": 5.5811,
      "mean_token_accuracy": 0.16246630400419235,
      "num_tokens": 10133199.0,
      "step": 5835
    },
    {
      "entropy": 5.719208860397339,
      "epoch": 0.45438630616611553,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004984030456763435,
      "loss": 5.5256,
      "mean_token_accuracy": 0.16822175681591034,
      "num_tokens": 10141796.0,
      "step": 5840
    },
    {
      "entropy": 5.655806589126587,
      "epoch": 0.4547753355378331,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004983997477166257,
      "loss": 5.5497,
      "mean_token_accuracy": 0.16270433813333512,
      "num_tokens": 10149941.0,
      "step": 5845
    },
    {
      "entropy": 5.678034448623658,
      "epoch": 0.45516436490955064,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983964463671723,
      "loss": 5.6349,
      "mean_token_accuracy": 0.15235331207513808,
      "num_tokens": 10158796.0,
      "step": 5850
    },
    {
      "entropy": 5.734345149993897,
      "epoch": 0.45555339428126823,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983931416280334,
      "loss": 5.5958,
      "mean_token_accuracy": 0.15618554651737213,
      "num_tokens": 10167328.0,
      "step": 5855
    },
    {
      "entropy": 5.717952537536621,
      "epoch": 0.4559424236529858,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983898334992591,
      "loss": 5.6297,
      "mean_token_accuracy": 0.15573803931474686,
      "num_tokens": 10175937.0,
      "step": 5860
    },
    {
      "entropy": 5.667964267730713,
      "epoch": 0.45633145302470335,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983865219808998,
      "loss": 5.5841,
      "mean_token_accuracy": 0.15104752033948898,
      "num_tokens": 10184828.0,
      "step": 5865
    },
    {
      "entropy": 5.666615152359009,
      "epoch": 0.45672048239642093,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004983832070730055,
      "loss": 5.5921,
      "mean_token_accuracy": 0.1605905458331108,
      "num_tokens": 10193871.0,
      "step": 5870
    },
    {
      "entropy": 5.618437767028809,
      "epoch": 0.4571095117681385,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004983798887756265,
      "loss": 5.581,
      "mean_token_accuracy": 0.15717791318893432,
      "num_tokens": 10202740.0,
      "step": 5875
    },
    {
      "entropy": 5.70032639503479,
      "epoch": 0.45749854113985605,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004983765670888133,
      "loss": 5.5301,
      "mean_token_accuracy": 0.1584470435976982,
      "num_tokens": 10210903.0,
      "step": 5880
    },
    {
      "entropy": 5.6195252418518065,
      "epoch": 0.45788757051157364,
      "grad_norm": 1.125,
      "learning_rate": 0.0004983732420126163,
      "loss": 5.486,
      "mean_token_accuracy": 0.16567425280809403,
      "num_tokens": 10219305.0,
      "step": 5885
    },
    {
      "entropy": 5.700950860977173,
      "epoch": 0.45827659988329117,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004983699135470858,
      "loss": 5.6403,
      "mean_token_accuracy": 0.14735830649733545,
      "num_tokens": 10228575.0,
      "step": 5890
    },
    {
      "entropy": 5.727937173843384,
      "epoch": 0.45866562925500876,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004983665816922723,
      "loss": 5.6365,
      "mean_token_accuracy": 0.15583132654428483,
      "num_tokens": 10237813.0,
      "step": 5895
    },
    {
      "entropy": 5.7645368576049805,
      "epoch": 0.45905465862672634,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983632464482267,
      "loss": 5.6425,
      "mean_token_accuracy": 0.15131317377090453,
      "num_tokens": 10246606.0,
      "step": 5900
    },
    {
      "entropy": 5.614892148971558,
      "epoch": 0.4594436879984439,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004983599078149991,
      "loss": 5.5914,
      "mean_token_accuracy": 0.15920354574918746,
      "num_tokens": 10254996.0,
      "step": 5905
    },
    {
      "entropy": 5.679306602478027,
      "epoch": 0.45983271737016146,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983565657926405,
      "loss": 5.632,
      "mean_token_accuracy": 0.1570995807647705,
      "num_tokens": 10263570.0,
      "step": 5910
    },
    {
      "entropy": 5.67201075553894,
      "epoch": 0.460221746741879,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004983532203812017,
      "loss": 5.4869,
      "mean_token_accuracy": 0.16051596105098725,
      "num_tokens": 10272237.0,
      "step": 5915
    },
    {
      "entropy": 5.643408918380738,
      "epoch": 0.4606107761135966,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004983498715807331,
      "loss": 5.595,
      "mean_token_accuracy": 0.15796029716730117,
      "num_tokens": 10281215.0,
      "step": 5920
    },
    {
      "entropy": 5.7696967124938965,
      "epoch": 0.46099980548531416,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004983465193912857,
      "loss": 5.6494,
      "mean_token_accuracy": 0.15233540534973145,
      "num_tokens": 10291054.0,
      "step": 5925
    },
    {
      "entropy": 5.777407598495484,
      "epoch": 0.4613888348570317,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983431638129104,
      "loss": 5.642,
      "mean_token_accuracy": 0.1585405334830284,
      "num_tokens": 10299847.0,
      "step": 5930
    },
    {
      "entropy": 5.612028884887695,
      "epoch": 0.4617778642287493,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004983398048456581,
      "loss": 5.5675,
      "mean_token_accuracy": 0.15711080208420752,
      "num_tokens": 10309195.0,
      "step": 5935
    },
    {
      "entropy": 5.657839107513428,
      "epoch": 0.4621668936004668,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983364424895796,
      "loss": 5.5919,
      "mean_token_accuracy": 0.15403160750865935,
      "num_tokens": 10318386.0,
      "step": 5940
    },
    {
      "entropy": 5.6664416790008545,
      "epoch": 0.4625559229721844,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004983330767447262,
      "loss": 5.5172,
      "mean_token_accuracy": 0.15980684906244277,
      "num_tokens": 10327230.0,
      "step": 5945
    },
    {
      "entropy": 5.629609680175781,
      "epoch": 0.462944952343902,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004983297076111489,
      "loss": 5.5206,
      "mean_token_accuracy": 0.16089206337928771,
      "num_tokens": 10335617.0,
      "step": 5950
    },
    {
      "entropy": 5.687315607070923,
      "epoch": 0.4633339817156195,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004983263350888987,
      "loss": 5.5807,
      "mean_token_accuracy": 0.15722011923789977,
      "num_tokens": 10344302.0,
      "step": 5955
    },
    {
      "entropy": 5.755109214782715,
      "epoch": 0.4637230110873371,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004983229591780268,
      "loss": 5.6401,
      "mean_token_accuracy": 0.15257431343197822,
      "num_tokens": 10352490.0,
      "step": 5960
    },
    {
      "entropy": 5.702300357818603,
      "epoch": 0.46411204045905463,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004983195798785844,
      "loss": 5.5558,
      "mean_token_accuracy": 0.1626654177904129,
      "num_tokens": 10361249.0,
      "step": 5965
    },
    {
      "entropy": 5.7142884731292725,
      "epoch": 0.4645010698307722,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983161971906228,
      "loss": 5.6692,
      "mean_token_accuracy": 0.14899954646825792,
      "num_tokens": 10370367.0,
      "step": 5970
    },
    {
      "entropy": 5.695809030532837,
      "epoch": 0.4648900992024898,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004983128111141935,
      "loss": 5.619,
      "mean_token_accuracy": 0.1533697783946991,
      "num_tokens": 10378429.0,
      "step": 5975
    },
    {
      "entropy": 5.6939444065094,
      "epoch": 0.46527912857420733,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004983094216493475,
      "loss": 5.5481,
      "mean_token_accuracy": 0.1594567194581032,
      "num_tokens": 10386762.0,
      "step": 5980
    },
    {
      "entropy": 5.665449094772339,
      "epoch": 0.4656681579459249,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004983060287961367,
      "loss": 5.5588,
      "mean_token_accuracy": 0.16088817715644838,
      "num_tokens": 10395150.0,
      "step": 5985
    },
    {
      "entropy": 5.7089026927947994,
      "epoch": 0.4660571873176425,
      "grad_norm": 1.25,
      "learning_rate": 0.0004983026325546123,
      "loss": 5.5751,
      "mean_token_accuracy": 0.16142064034938813,
      "num_tokens": 10403941.0,
      "step": 5990
    },
    {
      "entropy": 5.770590782165527,
      "epoch": 0.46644621668936004,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004982992329248257,
      "loss": 5.6574,
      "mean_token_accuracy": 0.14635860323905944,
      "num_tokens": 10413955.0,
      "step": 5995
    },
    {
      "entropy": 5.6717547416687015,
      "epoch": 0.4668352460610776,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004982958299068289,
      "loss": 5.5881,
      "mean_token_accuracy": 0.15662500709295274,
      "num_tokens": 10421676.0,
      "step": 6000
    },
    {
      "epoch": 0.4668352460610776,
      "eval_entropy": 5.492106970969706,
      "eval_loss": 5.644237518310547,
      "eval_mean_token_accuracy": 0.16206505519191844,
      "eval_num_tokens": 10421676.0,
      "eval_runtime": 28.8327,
      "eval_samples_per_second": 1441.351,
      "eval_steps_per_second": 180.178,
      "step": 6000
    },
    {
      "entropy": 5.616994762420655,
      "epoch": 0.46722427543279516,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000498292423500673,
      "loss": 5.5473,
      "mean_token_accuracy": 0.15582346469163894,
      "num_tokens": 10430714.0,
      "step": 6005
    },
    {
      "entropy": 5.760208368301392,
      "epoch": 0.46761330480451274,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004982890137064102,
      "loss": 5.6559,
      "mean_token_accuracy": 0.15567684173583984,
      "num_tokens": 10439089.0,
      "step": 6010
    },
    {
      "entropy": 5.70116400718689,
      "epoch": 0.46800233417623033,
      "grad_norm": 1.25,
      "learning_rate": 0.000498285600524092,
      "loss": 5.5757,
      "mean_token_accuracy": 0.15551938712596894,
      "num_tokens": 10447455.0,
      "step": 6015
    },
    {
      "entropy": 5.64963436126709,
      "epoch": 0.46839136354794786,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004982821839537701,
      "loss": 5.5763,
      "mean_token_accuracy": 0.15392711013555527,
      "num_tokens": 10455936.0,
      "step": 6020
    },
    {
      "entropy": 5.730429172515869,
      "epoch": 0.46878039291966545,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004982787639954966,
      "loss": 5.6874,
      "mean_token_accuracy": 0.1510494001209736,
      "num_tokens": 10465389.0,
      "step": 6025
    },
    {
      "entropy": 5.8077552318573,
      "epoch": 0.469169422291383,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004982753406493232,
      "loss": 5.5858,
      "mean_token_accuracy": 0.1518092706799507,
      "num_tokens": 10474709.0,
      "step": 6030
    },
    {
      "entropy": 5.667435598373413,
      "epoch": 0.46955845166310056,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004982719139153018,
      "loss": 5.5762,
      "mean_token_accuracy": 0.1561327449977398,
      "num_tokens": 10482972.0,
      "step": 6035
    },
    {
      "entropy": 5.668552541732788,
      "epoch": 0.46994748103481815,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004982684837934845,
      "loss": 5.5817,
      "mean_token_accuracy": 0.15674189403653144,
      "num_tokens": 10490897.0,
      "step": 6040
    },
    {
      "entropy": 5.658299112319947,
      "epoch": 0.4703365104065357,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004982650502839234,
      "loss": 5.5731,
      "mean_token_accuracy": 0.16014641672372817,
      "num_tokens": 10499679.0,
      "step": 6045
    },
    {
      "entropy": 5.7520054340362545,
      "epoch": 0.47072553977825327,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004982616133866705,
      "loss": 5.6189,
      "mean_token_accuracy": 0.15321286842226983,
      "num_tokens": 10508775.0,
      "step": 6050
    },
    {
      "entropy": 5.613485336303711,
      "epoch": 0.4711145691499708,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004982581731017779,
      "loss": 5.5167,
      "mean_token_accuracy": 0.16455159187316895,
      "num_tokens": 10517600.0,
      "step": 6055
    },
    {
      "entropy": 5.549383354187012,
      "epoch": 0.4715035985216884,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004982547294292979,
      "loss": 5.5001,
      "mean_token_accuracy": 0.16116441786289215,
      "num_tokens": 10525674.0,
      "step": 6060
    },
    {
      "entropy": 5.758707714080811,
      "epoch": 0.47189262789340597,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004982512823692828,
      "loss": 5.6816,
      "mean_token_accuracy": 0.1527387097477913,
      "num_tokens": 10533819.0,
      "step": 6065
    },
    {
      "entropy": 5.637930965423584,
      "epoch": 0.4722816572651235,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982478319217848,
      "loss": 5.5032,
      "mean_token_accuracy": 0.16184482872486114,
      "num_tokens": 10542901.0,
      "step": 6070
    },
    {
      "entropy": 5.636090660095215,
      "epoch": 0.4726706866368411,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982443780868562,
      "loss": 5.5798,
      "mean_token_accuracy": 0.15711832493543626,
      "num_tokens": 10552388.0,
      "step": 6075
    },
    {
      "entropy": 5.8238075256347654,
      "epoch": 0.4730597160085587,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004982409208645496,
      "loss": 5.6594,
      "mean_token_accuracy": 0.1539443776011467,
      "num_tokens": 10561676.0,
      "step": 6080
    },
    {
      "entropy": 5.799462556838989,
      "epoch": 0.4734487453802762,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004982374602549173,
      "loss": 5.5745,
      "mean_token_accuracy": 0.1607535168528557,
      "num_tokens": 10570037.0,
      "step": 6085
    },
    {
      "entropy": 5.6345953941345215,
      "epoch": 0.4738377747519938,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004982339962580119,
      "loss": 5.5749,
      "mean_token_accuracy": 0.15662821233272553,
      "num_tokens": 10579140.0,
      "step": 6090
    },
    {
      "entropy": 5.665529203414917,
      "epoch": 0.4742268041237113,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004982305288738859,
      "loss": 5.6131,
      "mean_token_accuracy": 0.1526251867413521,
      "num_tokens": 10587689.0,
      "step": 6095
    },
    {
      "entropy": 5.662497282028198,
      "epoch": 0.4746158334954289,
      "grad_norm": 1.125,
      "learning_rate": 0.0004982270581025919,
      "loss": 5.4874,
      "mean_token_accuracy": 0.16862452626228333,
      "num_tokens": 10596002.0,
      "step": 6100
    },
    {
      "entropy": 5.687047243118286,
      "epoch": 0.4750048628671465,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004982235839441826,
      "loss": 5.701,
      "mean_token_accuracy": 0.15273806527256967,
      "num_tokens": 10605895.0,
      "step": 6105
    },
    {
      "entropy": 5.798728561401367,
      "epoch": 0.475393892238864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004982201063987108,
      "loss": 5.6667,
      "mean_token_accuracy": 0.15098253041505813,
      "num_tokens": 10614641.0,
      "step": 6110
    },
    {
      "entropy": 5.686376237869263,
      "epoch": 0.4757829216105816,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004982166254662292,
      "loss": 5.6286,
      "mean_token_accuracy": 0.15341434478759766,
      "num_tokens": 10624381.0,
      "step": 6115
    },
    {
      "entropy": 5.665107345581054,
      "epoch": 0.47617195098229914,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004982131411467904,
      "loss": 5.6001,
      "mean_token_accuracy": 0.15612624436616898,
      "num_tokens": 10632493.0,
      "step": 6120
    },
    {
      "entropy": 5.6808939456939695,
      "epoch": 0.4765609803540167,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004982096534404476,
      "loss": 5.6045,
      "mean_token_accuracy": 0.14975400120019913,
      "num_tokens": 10641556.0,
      "step": 6125
    },
    {
      "entropy": 5.678710985183716,
      "epoch": 0.4769500097257343,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004982061623472535,
      "loss": 5.68,
      "mean_token_accuracy": 0.15386440381407737,
      "num_tokens": 10649736.0,
      "step": 6130
    },
    {
      "entropy": 5.690550470352173,
      "epoch": 0.47733903909745184,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004982026678672612,
      "loss": 5.5078,
      "mean_token_accuracy": 0.16158291548490525,
      "num_tokens": 10657991.0,
      "step": 6135
    },
    {
      "entropy": 5.5681188106536865,
      "epoch": 0.47772806846916943,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004981991700005238,
      "loss": 5.5754,
      "mean_token_accuracy": 0.1516540415585041,
      "num_tokens": 10667310.0,
      "step": 6140
    },
    {
      "entropy": 5.8310223579406735,
      "epoch": 0.47811709784088696,
      "grad_norm": 1.25,
      "learning_rate": 0.000498195668747094,
      "loss": 5.6227,
      "mean_token_accuracy": 0.1568164587020874,
      "num_tokens": 10675658.0,
      "step": 6145
    },
    {
      "entropy": 5.681654834747315,
      "epoch": 0.47850612721260455,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004981921641070254,
      "loss": 5.5003,
      "mean_token_accuracy": 0.16338133364915847,
      "num_tokens": 10684016.0,
      "step": 6150
    },
    {
      "entropy": 5.587088584899902,
      "epoch": 0.47889515658432213,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004981886560803708,
      "loss": 5.6247,
      "mean_token_accuracy": 0.15906248539686202,
      "num_tokens": 10692814.0,
      "step": 6155
    },
    {
      "entropy": 5.685698223114014,
      "epoch": 0.47928418595603967,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004981851446671838,
      "loss": 5.6274,
      "mean_token_accuracy": 0.15845713913440704,
      "num_tokens": 10701571.0,
      "step": 6160
    },
    {
      "entropy": 5.7994170665740965,
      "epoch": 0.47967321532775725,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004981816298675173,
      "loss": 5.6173,
      "mean_token_accuracy": 0.15209137052297592,
      "num_tokens": 10710569.0,
      "step": 6165
    },
    {
      "entropy": 5.715701580047607,
      "epoch": 0.4800622446994748,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004981781116814248,
      "loss": 5.6623,
      "mean_token_accuracy": 0.15391862541437148,
      "num_tokens": 10719277.0,
      "step": 6170
    },
    {
      "entropy": 5.631781816482544,
      "epoch": 0.48045127407119237,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004981745901089596,
      "loss": 5.5152,
      "mean_token_accuracy": 0.15900989323854448,
      "num_tokens": 10727174.0,
      "step": 6175
    },
    {
      "entropy": 5.6079856872558596,
      "epoch": 0.48084030344290996,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004981710651501753,
      "loss": 5.6179,
      "mean_token_accuracy": 0.1522950202226639,
      "num_tokens": 10736072.0,
      "step": 6180
    },
    {
      "entropy": 5.7692827701568605,
      "epoch": 0.4812293328146275,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004981675368051254,
      "loss": 5.6227,
      "mean_token_accuracy": 0.15922482833266258,
      "num_tokens": 10745375.0,
      "step": 6185
    },
    {
      "entropy": 5.68420033454895,
      "epoch": 0.4816183621863451,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004981640050738633,
      "loss": 5.6112,
      "mean_token_accuracy": 0.15697287023067474,
      "num_tokens": 10754049.0,
      "step": 6190
    },
    {
      "entropy": 5.599038887023926,
      "epoch": 0.48200739155806266,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004981604699564426,
      "loss": 5.5453,
      "mean_token_accuracy": 0.15898633897304534,
      "num_tokens": 10763475.0,
      "step": 6195
    },
    {
      "entropy": 5.664451837539673,
      "epoch": 0.4823964209297802,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004981569314529169,
      "loss": 5.6117,
      "mean_token_accuracy": 0.15641794353723526,
      "num_tokens": 10772100.0,
      "step": 6200
    },
    {
      "entropy": 5.654427576065063,
      "epoch": 0.4827854503014978,
      "grad_norm": 1.3125,
      "learning_rate": 0.00049815338956334,
      "loss": 5.5727,
      "mean_token_accuracy": 0.15927761197090148,
      "num_tokens": 10780605.0,
      "step": 6205
    },
    {
      "entropy": 5.730949020385742,
      "epoch": 0.4831744796732153,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004981498442877656,
      "loss": 5.5499,
      "mean_token_accuracy": 0.15757131054997445,
      "num_tokens": 10789235.0,
      "step": 6210
    },
    {
      "entropy": 5.6602763652801515,
      "epoch": 0.4835635090449329,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004981462956262474,
      "loss": 5.5668,
      "mean_token_accuracy": 0.15322166234254836,
      "num_tokens": 10798078.0,
      "step": 6215
    },
    {
      "entropy": 5.7057524681091305,
      "epoch": 0.4839525384166505,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004981427435788396,
      "loss": 5.6098,
      "mean_token_accuracy": 0.1502933382987976,
      "num_tokens": 10807667.0,
      "step": 6220
    },
    {
      "entropy": 5.705715608596802,
      "epoch": 0.484341567788368,
      "grad_norm": 1.25,
      "learning_rate": 0.0004981391881455957,
      "loss": 5.5558,
      "mean_token_accuracy": 0.15280514359474182,
      "num_tokens": 10816574.0,
      "step": 6225
    },
    {
      "entropy": 5.624035263061524,
      "epoch": 0.4847305971600856,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004981356293265696,
      "loss": 5.5286,
      "mean_token_accuracy": 0.15623718947172166,
      "num_tokens": 10824675.0,
      "step": 6230
    },
    {
      "entropy": 5.645666599273682,
      "epoch": 0.4851196265318031,
      "grad_norm": 1.25,
      "learning_rate": 0.0004981320671218157,
      "loss": 5.495,
      "mean_token_accuracy": 0.16385319828987122,
      "num_tokens": 10832855.0,
      "step": 6235
    },
    {
      "entropy": 5.596946859359742,
      "epoch": 0.4855086559035207,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004981285015313877,
      "loss": 5.3928,
      "mean_token_accuracy": 0.1628694102168083,
      "num_tokens": 10841603.0,
      "step": 6240
    },
    {
      "entropy": 5.7441243648529055,
      "epoch": 0.4858976852752383,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004981249325553399,
      "loss": 5.7359,
      "mean_token_accuracy": 0.15029820799827576,
      "num_tokens": 10850348.0,
      "step": 6245
    },
    {
      "entropy": 5.7170782566070555,
      "epoch": 0.48628671464695583,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004981213601937264,
      "loss": 5.4914,
      "mean_token_accuracy": 0.16128704100847244,
      "num_tokens": 10858955.0,
      "step": 6250
    },
    {
      "entropy": 5.631417274475098,
      "epoch": 0.4866757440186734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004981177844466013,
      "loss": 5.5844,
      "mean_token_accuracy": 0.1497468426823616,
      "num_tokens": 10867336.0,
      "step": 6255
    },
    {
      "entropy": 5.673184585571289,
      "epoch": 0.48706477339039095,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004981142053140192,
      "loss": 5.6523,
      "mean_token_accuracy": 0.15188782960176467,
      "num_tokens": 10877043.0,
      "step": 6260
    },
    {
      "entropy": 5.782298374176025,
      "epoch": 0.48745380276210853,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004981106227960339,
      "loss": 5.618,
      "mean_token_accuracy": 0.157500159740448,
      "num_tokens": 10885580.0,
      "step": 6265
    },
    {
      "entropy": 5.696793413162231,
      "epoch": 0.4878428321338261,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004981070368927001,
      "loss": 5.7061,
      "mean_token_accuracy": 0.1527958855032921,
      "num_tokens": 10894289.0,
      "step": 6270
    },
    {
      "entropy": 5.738697242736817,
      "epoch": 0.48823186150554365,
      "grad_norm": 1.296875,
      "learning_rate": 0.000498103447604072,
      "loss": 5.6488,
      "mean_token_accuracy": 0.14524348676204682,
      "num_tokens": 10902384.0,
      "step": 6275
    },
    {
      "entropy": 5.708152341842651,
      "epoch": 0.48862089087726124,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004980998549302044,
      "loss": 5.5375,
      "mean_token_accuracy": 0.1535746544599533,
      "num_tokens": 10911160.0,
      "step": 6280
    },
    {
      "entropy": 5.5684733390808105,
      "epoch": 0.4890099202489788,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004980962588711516,
      "loss": 5.4487,
      "mean_token_accuracy": 0.17255508452653884,
      "num_tokens": 10920066.0,
      "step": 6285
    },
    {
      "entropy": 5.688349103927612,
      "epoch": 0.48939894962069636,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000498092659426968,
      "loss": 5.6211,
      "mean_token_accuracy": 0.15368115305900573,
      "num_tokens": 10928969.0,
      "step": 6290
    },
    {
      "entropy": 5.7650329113006595,
      "epoch": 0.48978797899241394,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004980890565977085,
      "loss": 5.6194,
      "mean_token_accuracy": 0.15469925999641418,
      "num_tokens": 10936939.0,
      "step": 6295
    },
    {
      "entropy": 5.656690454483032,
      "epoch": 0.4901770083641315,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004980854503834276,
      "loss": 5.5723,
      "mean_token_accuracy": 0.1604826644062996,
      "num_tokens": 10945929.0,
      "step": 6300
    },
    {
      "entropy": 5.696713161468506,
      "epoch": 0.49056603773584906,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004980818407841802,
      "loss": 5.5637,
      "mean_token_accuracy": 0.15613728612661362,
      "num_tokens": 10954849.0,
      "step": 6305
    },
    {
      "entropy": 5.637706327438354,
      "epoch": 0.49095506710756665,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004980782278000207,
      "loss": 5.6192,
      "mean_token_accuracy": 0.1526903212070465,
      "num_tokens": 10963800.0,
      "step": 6310
    },
    {
      "entropy": 5.7678993225097654,
      "epoch": 0.4913440964792842,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004980746114310043,
      "loss": 5.6903,
      "mean_token_accuracy": 0.15478771477937697,
      "num_tokens": 10972477.0,
      "step": 6315
    },
    {
      "entropy": 5.787669038772583,
      "epoch": 0.49173312585100176,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004980709916771858,
      "loss": 5.6837,
      "mean_token_accuracy": 0.15601867586374282,
      "num_tokens": 10981437.0,
      "step": 6320
    },
    {
      "entropy": 5.698675918579101,
      "epoch": 0.4921221552227193,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004980673685386198,
      "loss": 5.5155,
      "mean_token_accuracy": 0.15962526202201843,
      "num_tokens": 10989973.0,
      "step": 6325
    },
    {
      "entropy": 5.665204000473023,
      "epoch": 0.4925111845944369,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004980637420153618,
      "loss": 5.5657,
      "mean_token_accuracy": 0.15236510932445527,
      "num_tokens": 10998160.0,
      "step": 6330
    },
    {
      "entropy": 5.703488445281982,
      "epoch": 0.49290021396615447,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004980601121074664,
      "loss": 5.6985,
      "mean_token_accuracy": 0.15162273794412612,
      "num_tokens": 11007127.0,
      "step": 6335
    },
    {
      "entropy": 5.749834060668945,
      "epoch": 0.493289243337872,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004980564788149889,
      "loss": 5.5436,
      "mean_token_accuracy": 0.15907759219408035,
      "num_tokens": 11015518.0,
      "step": 6340
    },
    {
      "entropy": 5.612507963180542,
      "epoch": 0.4936782727095896,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004980528421379844,
      "loss": 5.4549,
      "mean_token_accuracy": 0.16853654533624648,
      "num_tokens": 11023739.0,
      "step": 6345
    },
    {
      "entropy": 5.618532943725586,
      "epoch": 0.4940673020813071,
      "grad_norm": 1.125,
      "learning_rate": 0.000498049202076508,
      "loss": 5.4891,
      "mean_token_accuracy": 0.1704382061958313,
      "num_tokens": 11032034.0,
      "step": 6350
    },
    {
      "entropy": 5.629121494293213,
      "epoch": 0.4944563314530247,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004980455586306149,
      "loss": 5.48,
      "mean_token_accuracy": 0.16243453323841095,
      "num_tokens": 11040319.0,
      "step": 6355
    },
    {
      "entropy": 5.584317779541015,
      "epoch": 0.4948453608247423,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004980419118003605,
      "loss": 5.5495,
      "mean_token_accuracy": 0.15858743488788604,
      "num_tokens": 11049459.0,
      "step": 6360
    },
    {
      "entropy": 5.708620405197143,
      "epoch": 0.4952343901964598,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004980382615858001,
      "loss": 5.5342,
      "mean_token_accuracy": 0.16316292583942413,
      "num_tokens": 11058402.0,
      "step": 6365
    },
    {
      "entropy": 5.5763710021972654,
      "epoch": 0.4956234195681774,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004980346079869892,
      "loss": 5.4708,
      "mean_token_accuracy": 0.16301835030317308,
      "num_tokens": 11067430.0,
      "step": 6370
    },
    {
      "entropy": 5.67085747718811,
      "epoch": 0.49601244893989493,
      "grad_norm": 1.296875,
      "learning_rate": 0.000498030951003983,
      "loss": 5.4774,
      "mean_token_accuracy": 0.15947469770908357,
      "num_tokens": 11075594.0,
      "step": 6375
    },
    {
      "entropy": 5.6808929443359375,
      "epoch": 0.4964014783116125,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004980272906368371,
      "loss": 5.4516,
      "mean_token_accuracy": 0.17172312438488008,
      "num_tokens": 11083509.0,
      "step": 6380
    },
    {
      "entropy": 5.676959419250489,
      "epoch": 0.4967905076833301,
      "grad_norm": 1.25,
      "learning_rate": 0.0004980236268856071,
      "loss": 5.6169,
      "mean_token_accuracy": 0.15670440942049027,
      "num_tokens": 11091938.0,
      "step": 6385
    },
    {
      "entropy": 5.688388633728027,
      "epoch": 0.49717953705504764,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004980199597503487,
      "loss": 5.6213,
      "mean_token_accuracy": 0.1573230095207691,
      "num_tokens": 11099974.0,
      "step": 6390
    },
    {
      "entropy": 5.561890602111816,
      "epoch": 0.4975685664267652,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004980162892311171,
      "loss": 5.3004,
      "mean_token_accuracy": 0.17495135217905045,
      "num_tokens": 11107944.0,
      "step": 6395
    },
    {
      "entropy": 5.594667148590088,
      "epoch": 0.4979575957984828,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004980126153279684,
      "loss": 5.6,
      "mean_token_accuracy": 0.15356175750494003,
      "num_tokens": 11116401.0,
      "step": 6400
    },
    {
      "entropy": 5.673785638809204,
      "epoch": 0.49834662517020034,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004980089380409583,
      "loss": 5.5379,
      "mean_token_accuracy": 0.1534483790397644,
      "num_tokens": 11125430.0,
      "step": 6405
    },
    {
      "entropy": 5.718994045257569,
      "epoch": 0.4987356545419179,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004980052573701424,
      "loss": 5.632,
      "mean_token_accuracy": 0.15260782688856125,
      "num_tokens": 11133949.0,
      "step": 6410
    },
    {
      "entropy": 5.760442066192627,
      "epoch": 0.49912468391363546,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004980015733155767,
      "loss": 5.5743,
      "mean_token_accuracy": 0.15651538521051406,
      "num_tokens": 11141749.0,
      "step": 6415
    },
    {
      "entropy": 5.614003133773804,
      "epoch": 0.49951371328535304,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004979978858773171,
      "loss": 5.5513,
      "mean_token_accuracy": 0.15955282002687454,
      "num_tokens": 11150618.0,
      "step": 6420
    },
    {
      "entropy": 5.592733907699585,
      "epoch": 0.49990274265707063,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004979941950554195,
      "loss": 5.591,
      "mean_token_accuracy": 0.15830211639404296,
      "num_tokens": 11158549.0,
      "step": 6425
    },
    {
      "entropy": 5.731333684921265,
      "epoch": 0.5002917720287882,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004979905008499399,
      "loss": 5.5319,
      "mean_token_accuracy": 0.16687725335359574,
      "num_tokens": 11167131.0,
      "step": 6430
    },
    {
      "entropy": 5.690819978713989,
      "epoch": 0.5006808014005057,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004979868032609344,
      "loss": 5.6091,
      "mean_token_accuracy": 0.15680969059467315,
      "num_tokens": 11175766.0,
      "step": 6435
    },
    {
      "entropy": 5.65836272239685,
      "epoch": 0.5010698307722233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004979831022884591,
      "loss": 5.5659,
      "mean_token_accuracy": 0.16207551956176758,
      "num_tokens": 11184799.0,
      "step": 6440
    },
    {
      "entropy": 5.58139820098877,
      "epoch": 0.5014588601439409,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004979793979325701,
      "loss": 5.4977,
      "mean_token_accuracy": 0.16090358197689056,
      "num_tokens": 11193083.0,
      "step": 6445
    },
    {
      "entropy": 5.595728015899658,
      "epoch": 0.5018478895156584,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004979756901933236,
      "loss": 5.4977,
      "mean_token_accuracy": 0.16303258389234543,
      "num_tokens": 11201753.0,
      "step": 6450
    },
    {
      "entropy": 5.755453395843506,
      "epoch": 0.502236918887376,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000497971979070776,
      "loss": 5.593,
      "mean_token_accuracy": 0.15749429017305375,
      "num_tokens": 11209965.0,
      "step": 6455
    },
    {
      "entropy": 5.605485534667968,
      "epoch": 0.5026259482590936,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004979682645649834,
      "loss": 5.488,
      "mean_token_accuracy": 0.16725804656744003,
      "num_tokens": 11217961.0,
      "step": 6460
    },
    {
      "entropy": 5.6764552116394045,
      "epoch": 0.5030149776308112,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004979645466760025,
      "loss": 5.6157,
      "mean_token_accuracy": 0.15153489857912064,
      "num_tokens": 11225803.0,
      "step": 6465
    },
    {
      "entropy": 5.66771297454834,
      "epoch": 0.5034040070025287,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004979608254038892,
      "loss": 5.4608,
      "mean_token_accuracy": 0.16376415491104127,
      "num_tokens": 11234584.0,
      "step": 6470
    },
    {
      "entropy": 5.704413223266601,
      "epoch": 0.5037930363742462,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004979571007487003,
      "loss": 5.6524,
      "mean_token_accuracy": 0.15542552322149278,
      "num_tokens": 11243340.0,
      "step": 6475
    },
    {
      "entropy": 5.6457360744476315,
      "epoch": 0.5041820657459638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004979533727104924,
      "loss": 5.499,
      "mean_token_accuracy": 0.1582372471690178,
      "num_tokens": 11252383.0,
      "step": 6480
    },
    {
      "entropy": 5.662484931945801,
      "epoch": 0.5045710951176814,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000497949641289322,
      "loss": 5.6037,
      "mean_token_accuracy": 0.15492093414068223,
      "num_tokens": 11261419.0,
      "step": 6485
    },
    {
      "entropy": 5.6591331481933596,
      "epoch": 0.504960124489399,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004979459064852456,
      "loss": 5.5692,
      "mean_token_accuracy": 0.15648440569639205,
      "num_tokens": 11270743.0,
      "step": 6490
    },
    {
      "entropy": 5.724182987213135,
      "epoch": 0.5053491538611166,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004979421682983197,
      "loss": 5.4759,
      "mean_token_accuracy": 0.16309827119112014,
      "num_tokens": 11279666.0,
      "step": 6495
    },
    {
      "entropy": 5.7070385932922365,
      "epoch": 0.505738183232834,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004979384267286015,
      "loss": 5.6504,
      "mean_token_accuracy": 0.15011872351169586,
      "num_tokens": 11288710.0,
      "step": 6500
    },
    {
      "entropy": 5.638075351715088,
      "epoch": 0.5061272126045516,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004979346817761475,
      "loss": 5.5084,
      "mean_token_accuracy": 0.1591273382306099,
      "num_tokens": 11297665.0,
      "step": 6505
    },
    {
      "entropy": 5.707624340057373,
      "epoch": 0.5065162419762692,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004979309334410144,
      "loss": 5.6751,
      "mean_token_accuracy": 0.15144848972558975,
      "num_tokens": 11306751.0,
      "step": 6510
    },
    {
      "entropy": 5.762083244323731,
      "epoch": 0.5069052713479868,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004979271817232594,
      "loss": 5.672,
      "mean_token_accuracy": 0.14604207128286362,
      "num_tokens": 11315659.0,
      "step": 6515
    },
    {
      "entropy": 5.727238464355469,
      "epoch": 0.5072943007197044,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004979234266229391,
      "loss": 5.566,
      "mean_token_accuracy": 0.1598181277513504,
      "num_tokens": 11323362.0,
      "step": 6520
    },
    {
      "entropy": 5.658531379699707,
      "epoch": 0.5076833300914219,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004979196681401106,
      "loss": 5.5732,
      "mean_token_accuracy": 0.1515340879559517,
      "num_tokens": 11332343.0,
      "step": 6525
    },
    {
      "entropy": 5.640958070755005,
      "epoch": 0.5080723594631394,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000497915906274831,
      "loss": 5.4883,
      "mean_token_accuracy": 0.15905367136001586,
      "num_tokens": 11340816.0,
      "step": 6530
    },
    {
      "entropy": 5.579328536987305,
      "epoch": 0.508461388834857,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004979121410271574,
      "loss": 5.4893,
      "mean_token_accuracy": 0.1620453342795372,
      "num_tokens": 11349728.0,
      "step": 6535
    },
    {
      "entropy": 5.706458806991577,
      "epoch": 0.5088504182065746,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004979083723971468,
      "loss": 5.6243,
      "mean_token_accuracy": 0.15665159076452256,
      "num_tokens": 11358477.0,
      "step": 6540
    },
    {
      "entropy": 5.576215600967407,
      "epoch": 0.5092394475782922,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004979046003848564,
      "loss": 5.4395,
      "mean_token_accuracy": 0.16753915101289749,
      "num_tokens": 11367183.0,
      "step": 6545
    },
    {
      "entropy": 5.695451068878174,
      "epoch": 0.5096284769500097,
      "grad_norm": 1.75,
      "learning_rate": 0.0004979008249903435,
      "loss": 5.6462,
      "mean_token_accuracy": 0.14823395907878875,
      "num_tokens": 11376703.0,
      "step": 6550
    },
    {
      "entropy": 5.776837158203125,
      "epoch": 0.5100175063217273,
      "grad_norm": 1.25,
      "learning_rate": 0.0004978970462136655,
      "loss": 5.6616,
      "mean_token_accuracy": 0.15206627398729325,
      "num_tokens": 11385502.0,
      "step": 6555
    },
    {
      "entropy": 5.6919615268707275,
      "epoch": 0.5104065356934449,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004978932640548794,
      "loss": 5.5242,
      "mean_token_accuracy": 0.1622034192085266,
      "num_tokens": 11393947.0,
      "step": 6560
    },
    {
      "entropy": 5.5909508228302,
      "epoch": 0.5107955650651624,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004978894785140429,
      "loss": 5.5177,
      "mean_token_accuracy": 0.15952503234148024,
      "num_tokens": 11403077.0,
      "step": 6565
    },
    {
      "entropy": 5.611241388320923,
      "epoch": 0.51118459443688,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004978856895912132,
      "loss": 5.5061,
      "mean_token_accuracy": 0.1586812451481819,
      "num_tokens": 11411863.0,
      "step": 6570
    },
    {
      "entropy": 5.677961540222168,
      "epoch": 0.5115736238085975,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004978818972864481,
      "loss": 5.4708,
      "mean_token_accuracy": 0.16038688570261,
      "num_tokens": 11420030.0,
      "step": 6575
    },
    {
      "entropy": 5.666505241394043,
      "epoch": 0.5119626531803151,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004978781015998048,
      "loss": 5.5708,
      "mean_token_accuracy": 0.15640071481466294,
      "num_tokens": 11428534.0,
      "step": 6580
    },
    {
      "entropy": 5.6593951225280765,
      "epoch": 0.5123516825520327,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004978743025313413,
      "loss": 5.5472,
      "mean_token_accuracy": 0.16346569508314132,
      "num_tokens": 11436641.0,
      "step": 6585
    },
    {
      "entropy": 5.687247705459595,
      "epoch": 0.5127407119237503,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004978705000811148,
      "loss": 5.5275,
      "mean_token_accuracy": 0.15745826065540314,
      "num_tokens": 11444595.0,
      "step": 6590
    },
    {
      "entropy": 5.582401847839355,
      "epoch": 0.5131297412954678,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004978666942491832,
      "loss": 5.3998,
      "mean_token_accuracy": 0.16435882747173308,
      "num_tokens": 11453257.0,
      "step": 6595
    },
    {
      "entropy": 5.566832828521728,
      "epoch": 0.5135187706671853,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004978628850356043,
      "loss": 5.5401,
      "mean_token_accuracy": 0.1632547751069069,
      "num_tokens": 11461762.0,
      "step": 6600
    },
    {
      "entropy": 5.645640277862549,
      "epoch": 0.5139078000389029,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004978590724404358,
      "loss": 5.4872,
      "mean_token_accuracy": 0.1535194531083107,
      "num_tokens": 11470602.0,
      "step": 6605
    },
    {
      "entropy": 5.675325059890747,
      "epoch": 0.5142968294106205,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004978552564637356,
      "loss": 5.557,
      "mean_token_accuracy": 0.15856887847185136,
      "num_tokens": 11479826.0,
      "step": 6610
    },
    {
      "entropy": 5.696310520172119,
      "epoch": 0.5146858587823381,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004978514371055616,
      "loss": 5.6251,
      "mean_token_accuracy": 0.1528984934091568,
      "num_tokens": 11488698.0,
      "step": 6615
    },
    {
      "entropy": 5.704994869232178,
      "epoch": 0.5150748881540557,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004978476143659718,
      "loss": 5.6023,
      "mean_token_accuracy": 0.15849207788705827,
      "num_tokens": 11497434.0,
      "step": 6620
    },
    {
      "entropy": 5.6330362319946286,
      "epoch": 0.5154639175257731,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004978437882450241,
      "loss": 5.6285,
      "mean_token_accuracy": 0.15273135602474214,
      "num_tokens": 11506173.0,
      "step": 6625
    },
    {
      "entropy": 5.666847038269043,
      "epoch": 0.5158529468974907,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004978399587427766,
      "loss": 5.5179,
      "mean_token_accuracy": 0.16206059753894805,
      "num_tokens": 11514824.0,
      "step": 6630
    },
    {
      "entropy": 5.664588356018067,
      "epoch": 0.5162419762692083,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004978361258592874,
      "loss": 5.5814,
      "mean_token_accuracy": 0.14966512173414231,
      "num_tokens": 11523634.0,
      "step": 6635
    },
    {
      "entropy": 5.821097087860108,
      "epoch": 0.5166310056409259,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004978322895946147,
      "loss": 5.6959,
      "mean_token_accuracy": 0.14743875116109847,
      "num_tokens": 11531916.0,
      "step": 6640
    },
    {
      "entropy": 5.709766435623169,
      "epoch": 0.5170200350126435,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004978284499488167,
      "loss": 5.5176,
      "mean_token_accuracy": 0.1616029903292656,
      "num_tokens": 11540000.0,
      "step": 6645
    },
    {
      "entropy": 5.719381999969483,
      "epoch": 0.5174090643843611,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004978246069219516,
      "loss": 5.6161,
      "mean_token_accuracy": 0.15496881157159806,
      "num_tokens": 11548760.0,
      "step": 6650
    },
    {
      "entropy": 5.572812414169311,
      "epoch": 0.5177980937560785,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004978207605140777,
      "loss": 5.4182,
      "mean_token_accuracy": 0.16692280024290085,
      "num_tokens": 11557991.0,
      "step": 6655
    },
    {
      "entropy": 5.605830192565918,
      "epoch": 0.5181871231277961,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004978169107252534,
      "loss": 5.5,
      "mean_token_accuracy": 0.15972118824720383,
      "num_tokens": 11566638.0,
      "step": 6660
    },
    {
      "entropy": 5.590783166885376,
      "epoch": 0.5185761524995137,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004978130575555373,
      "loss": 5.4129,
      "mean_token_accuracy": 0.16473785936832427,
      "num_tokens": 11575106.0,
      "step": 6665
    },
    {
      "entropy": 5.648030757904053,
      "epoch": 0.5189651818712313,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004978092010049877,
      "loss": 5.5886,
      "mean_token_accuracy": 0.14713559448719024,
      "num_tokens": 11583971.0,
      "step": 6670
    },
    {
      "entropy": 5.675084209442138,
      "epoch": 0.5193542112429489,
      "grad_norm": 1.265625,
      "learning_rate": 0.000497805341073663,
      "loss": 5.5782,
      "mean_token_accuracy": 0.15736301690340043,
      "num_tokens": 11592936.0,
      "step": 6675
    },
    {
      "entropy": 5.617664909362793,
      "epoch": 0.5197432406146664,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004978014777616219,
      "loss": 5.4655,
      "mean_token_accuracy": 0.16452026665210723,
      "num_tokens": 11602055.0,
      "step": 6680
    },
    {
      "entropy": 5.644859457015992,
      "epoch": 0.520132269986384,
      "grad_norm": 1.25,
      "learning_rate": 0.0004977976110689229,
      "loss": 5.4895,
      "mean_token_accuracy": 0.16243117898702622,
      "num_tokens": 11610211.0,
      "step": 6685
    },
    {
      "entropy": 5.6761373519897464,
      "epoch": 0.5205212993581015,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000497793740995625,
      "loss": 5.5455,
      "mean_token_accuracy": 0.1594979152083397,
      "num_tokens": 11619096.0,
      "step": 6690
    },
    {
      "entropy": 5.648769330978394,
      "epoch": 0.5209103287298191,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004977898675417866,
      "loss": 5.6317,
      "mean_token_accuracy": 0.15871741771697997,
      "num_tokens": 11628142.0,
      "step": 6695
    },
    {
      "entropy": 5.733170652389527,
      "epoch": 0.5212993581015367,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004977859907074664,
      "loss": 5.4782,
      "mean_token_accuracy": 0.16082024127244948,
      "num_tokens": 11635961.0,
      "step": 6700
    },
    {
      "entropy": 5.6948095798492435,
      "epoch": 0.5216883874732542,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004977821104927236,
      "loss": 5.6115,
      "mean_token_accuracy": 0.15693757832050323,
      "num_tokens": 11644778.0,
      "step": 6705
    },
    {
      "entropy": 5.671348333358765,
      "epoch": 0.5220774168449718,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004977782268976167,
      "loss": 5.5851,
      "mean_token_accuracy": 0.1549125224351883,
      "num_tokens": 11654043.0,
      "step": 6710
    },
    {
      "entropy": 5.752050828933716,
      "epoch": 0.5224664462166894,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004977743399222048,
      "loss": 5.566,
      "mean_token_accuracy": 0.15431302934885024,
      "num_tokens": 11662794.0,
      "step": 6715
    },
    {
      "entropy": 5.610268497467041,
      "epoch": 0.522855475588407,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004977704495665471,
      "loss": 5.5678,
      "mean_token_accuracy": 0.15663338154554368,
      "num_tokens": 11670624.0,
      "step": 6720
    },
    {
      "entropy": 5.557846593856811,
      "epoch": 0.5232445049601245,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004977665558307023,
      "loss": 5.479,
      "mean_token_accuracy": 0.16290911734104158,
      "num_tokens": 11679366.0,
      "step": 6725
    },
    {
      "entropy": 5.537747621536255,
      "epoch": 0.523633534331842,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004977626587147295,
      "loss": 5.3729,
      "mean_token_accuracy": 0.16686793118715287,
      "num_tokens": 11688137.0,
      "step": 6730
    },
    {
      "entropy": 5.605463695526123,
      "epoch": 0.5240225637035596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000497758758218688,
      "loss": 5.5226,
      "mean_token_accuracy": 0.15901294574141503,
      "num_tokens": 11697236.0,
      "step": 6735
    },
    {
      "entropy": 5.669349527359008,
      "epoch": 0.5244115930752772,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004977548543426368,
      "loss": 5.5216,
      "mean_token_accuracy": 0.16320829093456268,
      "num_tokens": 11706513.0,
      "step": 6740
    },
    {
      "entropy": 5.6284770488739015,
      "epoch": 0.5248006224469948,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004977509470866353,
      "loss": 5.4808,
      "mean_token_accuracy": 0.15824874937534333,
      "num_tokens": 11715239.0,
      "step": 6745
    },
    {
      "entropy": 5.6246928691864015,
      "epoch": 0.5251896518187124,
      "grad_norm": 1.25,
      "learning_rate": 0.0004977470364507427,
      "loss": 5.4945,
      "mean_token_accuracy": 0.16825863867998123,
      "num_tokens": 11724086.0,
      "step": 6750
    },
    {
      "entropy": 5.6200848579406735,
      "epoch": 0.5255786811904298,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004977431224350184,
      "loss": 5.5293,
      "mean_token_accuracy": 0.16375562250614167,
      "num_tokens": 11731975.0,
      "step": 6755
    },
    {
      "entropy": 5.609702110290527,
      "epoch": 0.5259677105621474,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004977392050395217,
      "loss": 5.5229,
      "mean_token_accuracy": 0.15743479132652283,
      "num_tokens": 11741172.0,
      "step": 6760
    },
    {
      "entropy": 5.593728351593017,
      "epoch": 0.526356739933865,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004977352842643121,
      "loss": 5.4842,
      "mean_token_accuracy": 0.1634407237172127,
      "num_tokens": 11750609.0,
      "step": 6765
    },
    {
      "entropy": 5.639748382568359,
      "epoch": 0.5267457693055826,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000497731360109449,
      "loss": 5.4676,
      "mean_token_accuracy": 0.16293041706085204,
      "num_tokens": 11758901.0,
      "step": 6770
    },
    {
      "entropy": 5.551731443405151,
      "epoch": 0.5271347986773002,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004977274325749922,
      "loss": 5.4378,
      "mean_token_accuracy": 0.16170607060194014,
      "num_tokens": 11767202.0,
      "step": 6775
    },
    {
      "entropy": 5.561970853805542,
      "epoch": 0.5275238280490177,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004977235016610011,
      "loss": 5.5316,
      "mean_token_accuracy": 0.15927286446094513,
      "num_tokens": 11776164.0,
      "step": 6780
    },
    {
      "entropy": 5.672452306747436,
      "epoch": 0.5279128574207352,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004977195673675353,
      "loss": 5.4715,
      "mean_token_accuracy": 0.1680982381105423,
      "num_tokens": 11784548.0,
      "step": 6785
    },
    {
      "entropy": 5.594417142868042,
      "epoch": 0.5283018867924528,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004977156296946545,
      "loss": 5.5532,
      "mean_token_accuracy": 0.16064697206020356,
      "num_tokens": 11793249.0,
      "step": 6790
    },
    {
      "entropy": 5.630795812606811,
      "epoch": 0.5286909161641704,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004977116886424187,
      "loss": 5.5867,
      "mean_token_accuracy": 0.1527618646621704,
      "num_tokens": 11801512.0,
      "step": 6795
    },
    {
      "entropy": 5.645130920410156,
      "epoch": 0.529079945535888,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004977077442108873,
      "loss": 5.4718,
      "mean_token_accuracy": 0.1598686084151268,
      "num_tokens": 11810248.0,
      "step": 6800
    },
    {
      "entropy": 5.653914546966552,
      "epoch": 0.5294689749076055,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004977037964001204,
      "loss": 5.4582,
      "mean_token_accuracy": 0.1599445804953575,
      "num_tokens": 11819190.0,
      "step": 6805
    },
    {
      "entropy": 5.592351961135864,
      "epoch": 0.5298580042793231,
      "grad_norm": 1.28125,
      "learning_rate": 0.000497699845210178,
      "loss": 5.5339,
      "mean_token_accuracy": 0.1549182876944542,
      "num_tokens": 11827585.0,
      "step": 6810
    },
    {
      "entropy": 5.701840353012085,
      "epoch": 0.5302470336510406,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004976958906411197,
      "loss": 5.535,
      "mean_token_accuracy": 0.15848991721868516,
      "num_tokens": 11836256.0,
      "step": 6815
    },
    {
      "entropy": 5.74124116897583,
      "epoch": 0.5306360630227582,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004976919326930059,
      "loss": 5.6451,
      "mean_token_accuracy": 0.15875996947288512,
      "num_tokens": 11845022.0,
      "step": 6820
    },
    {
      "entropy": 5.613566589355469,
      "epoch": 0.5310250923944758,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004976879713658964,
      "loss": 5.5112,
      "mean_token_accuracy": 0.15980680286884308,
      "num_tokens": 11853289.0,
      "step": 6825
    },
    {
      "entropy": 5.631381845474243,
      "epoch": 0.5314141217661933,
      "grad_norm": 1.375,
      "learning_rate": 0.0004976840066598514,
      "loss": 5.4645,
      "mean_token_accuracy": 0.16436376720666884,
      "num_tokens": 11861589.0,
      "step": 6830
    },
    {
      "entropy": 5.711447811126709,
      "epoch": 0.5318031511379109,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004976800385749309,
      "loss": 5.5679,
      "mean_token_accuracy": 0.16017903834581376,
      "num_tokens": 11870313.0,
      "step": 6835
    },
    {
      "entropy": 5.680699110031128,
      "epoch": 0.5321921805096285,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004976760671111954,
      "loss": 5.5475,
      "mean_token_accuracy": 0.15322517231106758,
      "num_tokens": 11879558.0,
      "step": 6840
    },
    {
      "entropy": 5.639979362487793,
      "epoch": 0.532581209881346,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000497672092268705,
      "loss": 5.5479,
      "mean_token_accuracy": 0.1604737088084221,
      "num_tokens": 11887951.0,
      "step": 6845
    },
    {
      "entropy": 5.754343461990357,
      "epoch": 0.5329702392530636,
      "grad_norm": 1.234375,
      "learning_rate": 0.00049766811404752,
      "loss": 5.7416,
      "mean_token_accuracy": 0.14898372814059258,
      "num_tokens": 11896900.0,
      "step": 6850
    },
    {
      "entropy": 5.75668249130249,
      "epoch": 0.5333592686247812,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004976641324477008,
      "loss": 5.5669,
      "mean_token_accuracy": 0.1591862201690674,
      "num_tokens": 11905681.0,
      "step": 6855
    },
    {
      "entropy": 5.640393161773682,
      "epoch": 0.5337482979964987,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004976601474693077,
      "loss": 5.5093,
      "mean_token_accuracy": 0.16359012424945832,
      "num_tokens": 11914662.0,
      "step": 6860
    },
    {
      "entropy": 5.577741432189941,
      "epoch": 0.5341373273682163,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004976561591124014,
      "loss": 5.5305,
      "mean_token_accuracy": 0.15922741144895552,
      "num_tokens": 11923298.0,
      "step": 6865
    },
    {
      "entropy": 5.673051500320435,
      "epoch": 0.5345263567399339,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004976521673770421,
      "loss": 5.4482,
      "mean_token_accuracy": 0.1592395558953285,
      "num_tokens": 11932206.0,
      "step": 6870
    },
    {
      "entropy": 5.583999109268189,
      "epoch": 0.5349153861116515,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004976481722632907,
      "loss": 5.5095,
      "mean_token_accuracy": 0.16529154330492019,
      "num_tokens": 11940786.0,
      "step": 6875
    },
    {
      "entropy": 5.620921277999878,
      "epoch": 0.535304415483369,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004976441737712076,
      "loss": 5.5096,
      "mean_token_accuracy": 0.159516279399395,
      "num_tokens": 11949619.0,
      "step": 6880
    },
    {
      "entropy": 5.656333589553833,
      "epoch": 0.5356934448550865,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004976401719008536,
      "loss": 5.4506,
      "mean_token_accuracy": 0.16912465244531633,
      "num_tokens": 11957696.0,
      "step": 6885
    },
    {
      "entropy": 5.602833461761475,
      "epoch": 0.5360824742268041,
      "grad_norm": 1.625,
      "learning_rate": 0.0004976361666522893,
      "loss": 5.4651,
      "mean_token_accuracy": 0.16244376599788665,
      "num_tokens": 11965622.0,
      "step": 6890
    },
    {
      "entropy": 5.53003716468811,
      "epoch": 0.5364715035985217,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004976321580255755,
      "loss": 5.5423,
      "mean_token_accuracy": 0.15842145979404448,
      "num_tokens": 11973969.0,
      "step": 6895
    },
    {
      "entropy": 5.724082803726196,
      "epoch": 0.5368605329702393,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004976281460207731,
      "loss": 5.5329,
      "mean_token_accuracy": 0.16311838626861572,
      "num_tokens": 11982469.0,
      "step": 6900
    },
    {
      "entropy": 5.762820434570313,
      "epoch": 0.5372495623419569,
      "grad_norm": 1.25,
      "learning_rate": 0.000497624130637943,
      "loss": 5.6336,
      "mean_token_accuracy": 0.15604159682989122,
      "num_tokens": 11990656.0,
      "step": 6905
    },
    {
      "entropy": 5.658495044708252,
      "epoch": 0.5376385917136743,
      "grad_norm": 1.359375,
      "learning_rate": 0.000497620111877146,
      "loss": 5.6198,
      "mean_token_accuracy": 0.15872760415077208,
      "num_tokens": 11999394.0,
      "step": 6910
    },
    {
      "entropy": 5.72278323173523,
      "epoch": 0.5380276210853919,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004976160897384431,
      "loss": 5.5837,
      "mean_token_accuracy": 0.15553686544299125,
      "num_tokens": 12007692.0,
      "step": 6915
    },
    {
      "entropy": 5.683911848068237,
      "epoch": 0.5384166504571095,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004976120642218955,
      "loss": 5.5551,
      "mean_token_accuracy": 0.16157228350639344,
      "num_tokens": 12016038.0,
      "step": 6920
    },
    {
      "entropy": 5.625691270828247,
      "epoch": 0.5388056798288271,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000497608035327564,
      "loss": 5.6452,
      "mean_token_accuracy": 0.1577180117368698,
      "num_tokens": 12024504.0,
      "step": 6925
    },
    {
      "entropy": 5.609702634811401,
      "epoch": 0.5391947092005447,
      "grad_norm": 1.265625,
      "learning_rate": 0.00049760400305551,
      "loss": 5.5613,
      "mean_token_accuracy": 0.16581791788339614,
      "num_tokens": 12032935.0,
      "step": 6930
    },
    {
      "entropy": 5.701969766616822,
      "epoch": 0.5395837385722622,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004975999674057944,
      "loss": 5.5736,
      "mean_token_accuracy": 0.15595436990261077,
      "num_tokens": 12041907.0,
      "step": 6935
    },
    {
      "entropy": 5.656140375137329,
      "epoch": 0.5399727679439797,
      "grad_norm": 1.25,
      "learning_rate": 0.0004975959283784787,
      "loss": 5.4126,
      "mean_token_accuracy": 0.1687588095664978,
      "num_tokens": 12050528.0,
      "step": 6940
    },
    {
      "entropy": 5.577883768081665,
      "epoch": 0.5403617973156973,
      "grad_norm": 1.25,
      "learning_rate": 0.0004975918859736241,
      "loss": 5.4199,
      "mean_token_accuracy": 0.1674933299422264,
      "num_tokens": 12059221.0,
      "step": 6945
    },
    {
      "entropy": 5.64797010421753,
      "epoch": 0.5407508266874149,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004975878401912919,
      "loss": 5.6179,
      "mean_token_accuracy": 0.15194371491670608,
      "num_tokens": 12068131.0,
      "step": 6950
    },
    {
      "entropy": 5.703164386749267,
      "epoch": 0.5411398560591325,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004975837910315436,
      "loss": 5.4895,
      "mean_token_accuracy": 0.17003941386938096,
      "num_tokens": 12076521.0,
      "step": 6955
    },
    {
      "entropy": 5.658981227874756,
      "epoch": 0.54152888543085,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004975797384944405,
      "loss": 5.6906,
      "mean_token_accuracy": 0.15284735932946206,
      "num_tokens": 12085306.0,
      "step": 6960
    },
    {
      "entropy": 5.653558874130249,
      "epoch": 0.5419179148025676,
      "grad_norm": 1.203125,
      "learning_rate": 0.000497575682580044,
      "loss": 5.4765,
      "mean_token_accuracy": 0.1596546545624733,
      "num_tokens": 12094409.0,
      "step": 6965
    },
    {
      "entropy": 5.679685735702515,
      "epoch": 0.5423069441742852,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000497571623288416,
      "loss": 5.5063,
      "mean_token_accuracy": 0.1585929088294506,
      "num_tokens": 12103385.0,
      "step": 6970
    },
    {
      "entropy": 5.690260887145996,
      "epoch": 0.5426959735460027,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004975675606196177,
      "loss": 5.6363,
      "mean_token_accuracy": 0.15730044543743132,
      "num_tokens": 12112452.0,
      "step": 6975
    },
    {
      "entropy": 5.751817798614502,
      "epoch": 0.5430850029177203,
      "grad_norm": 1.234375,
      "learning_rate": 0.000497563494573711,
      "loss": 5.6334,
      "mean_token_accuracy": 0.15560218095779418,
      "num_tokens": 12120918.0,
      "step": 6980
    },
    {
      "entropy": 5.713543224334717,
      "epoch": 0.5434740322894378,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004975594251507575,
      "loss": 5.5804,
      "mean_token_accuracy": 0.15339445322752,
      "num_tokens": 12129797.0,
      "step": 6985
    },
    {
      "entropy": 5.713142776489258,
      "epoch": 0.5438630616611554,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000497555352350819,
      "loss": 5.5838,
      "mean_token_accuracy": 0.16098048090934752,
      "num_tokens": 12139245.0,
      "step": 6990
    },
    {
      "entropy": 5.686711215972901,
      "epoch": 0.544252091032873,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004975512761739572,
      "loss": 5.7437,
      "mean_token_accuracy": 0.1474977493286133,
      "num_tokens": 12147409.0,
      "step": 6995
    },
    {
      "entropy": 5.7700110912323,
      "epoch": 0.5446411204045906,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000497547196620234,
      "loss": 5.5477,
      "mean_token_accuracy": 0.1582572191953659,
      "num_tokens": 12156121.0,
      "step": 7000
    },
    {
      "entropy": 5.651832485198975,
      "epoch": 0.5450301497763081,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004975431136897114,
      "loss": 5.4365,
      "mean_token_accuracy": 0.1660777062177658,
      "num_tokens": 12165307.0,
      "step": 7005
    },
    {
      "entropy": 5.651866006851196,
      "epoch": 0.5454191791480256,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004975390273824512,
      "loss": 5.5238,
      "mean_token_accuracy": 0.15766593292355538,
      "num_tokens": 12173899.0,
      "step": 7010
    },
    {
      "entropy": 5.544921827316284,
      "epoch": 0.5458082085197432,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004975349376985155,
      "loss": 5.5202,
      "mean_token_accuracy": 0.15985482633113862,
      "num_tokens": 12182898.0,
      "step": 7015
    },
    {
      "entropy": 5.549176788330078,
      "epoch": 0.5461972378914608,
      "grad_norm": 1.25,
      "learning_rate": 0.0004975308446379663,
      "loss": 5.4084,
      "mean_token_accuracy": 0.16334766149520874,
      "num_tokens": 12191500.0,
      "step": 7020
    },
    {
      "entropy": 5.6007850646972654,
      "epoch": 0.5465862672631784,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004975267482008657,
      "loss": 5.5766,
      "mean_token_accuracy": 0.16620246022939683,
      "num_tokens": 12199420.0,
      "step": 7025
    },
    {
      "entropy": 5.691896724700928,
      "epoch": 0.546975296634896,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000497522648387276,
      "loss": 5.4757,
      "mean_token_accuracy": 0.16692695915699005,
      "num_tokens": 12207676.0,
      "step": 7030
    },
    {
      "entropy": 5.628739213943481,
      "epoch": 0.5473643260066134,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004975185451972592,
      "loss": 5.5355,
      "mean_token_accuracy": 0.16166269332170485,
      "num_tokens": 12216248.0,
      "step": 7035
    },
    {
      "entropy": 5.643997573852539,
      "epoch": 0.547753355378331,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004975144386308776,
      "loss": 5.5807,
      "mean_token_accuracy": 0.15384578108787536,
      "num_tokens": 12224588.0,
      "step": 7040
    },
    {
      "entropy": 5.72103214263916,
      "epoch": 0.5481423847500486,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004975103286881936,
      "loss": 5.583,
      "mean_token_accuracy": 0.1500740334391594,
      "num_tokens": 12234093.0,
      "step": 7045
    },
    {
      "entropy": 5.734965991973877,
      "epoch": 0.5485314141217662,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004975062153692696,
      "loss": 5.5026,
      "mean_token_accuracy": 0.16867420822381973,
      "num_tokens": 12242164.0,
      "step": 7050
    },
    {
      "entropy": 5.635144138336182,
      "epoch": 0.5489204434934838,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004975020986741678,
      "loss": 5.6154,
      "mean_token_accuracy": 0.15639858692884445,
      "num_tokens": 12251204.0,
      "step": 7055
    },
    {
      "entropy": 5.710094594955445,
      "epoch": 0.5493094728652014,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004974979786029509,
      "loss": 5.5651,
      "mean_token_accuracy": 0.16040020287036896,
      "num_tokens": 12259809.0,
      "step": 7060
    },
    {
      "entropy": 5.669298887252808,
      "epoch": 0.5496985022369189,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004974938551556814,
      "loss": 5.535,
      "mean_token_accuracy": 0.16596632599830627,
      "num_tokens": 12268668.0,
      "step": 7065
    },
    {
      "entropy": 5.619579315185547,
      "epoch": 0.5500875316086364,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004974897283324217,
      "loss": 5.4882,
      "mean_token_accuracy": 0.16426026821136475,
      "num_tokens": 12276823.0,
      "step": 7070
    },
    {
      "entropy": 5.528484678268432,
      "epoch": 0.550476560980354,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004974855981332343,
      "loss": 5.4385,
      "mean_token_accuracy": 0.16089818775653839,
      "num_tokens": 12285950.0,
      "step": 7075
    },
    {
      "entropy": 5.664534378051758,
      "epoch": 0.5508655903520716,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004974814645581823,
      "loss": 5.5627,
      "mean_token_accuracy": 0.1547747939825058,
      "num_tokens": 12295265.0,
      "step": 7080
    },
    {
      "entropy": 5.710403490066528,
      "epoch": 0.5512546197237892,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004974773276073282,
      "loss": 5.566,
      "mean_token_accuracy": 0.16329505890607834,
      "num_tokens": 12303649.0,
      "step": 7085
    },
    {
      "entropy": 5.591714763641358,
      "epoch": 0.5516436490955067,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004974731872807347,
      "loss": 5.6224,
      "mean_token_accuracy": 0.14819930344820023,
      "num_tokens": 12313303.0,
      "step": 7090
    },
    {
      "entropy": 5.649332284927368,
      "epoch": 0.5520326784672243,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004974690435784647,
      "loss": 5.5558,
      "mean_token_accuracy": 0.1625340536236763,
      "num_tokens": 12321692.0,
      "step": 7095
    },
    {
      "entropy": 5.619818687438965,
      "epoch": 0.5524217078389418,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004974648965005811,
      "loss": 5.3576,
      "mean_token_accuracy": 0.16542813181877136,
      "num_tokens": 12329908.0,
      "step": 7100
    },
    {
      "entropy": 5.621172046661377,
      "epoch": 0.5528107372106594,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004974607460471466,
      "loss": 5.5546,
      "mean_token_accuracy": 0.15522611290216445,
      "num_tokens": 12339270.0,
      "step": 7105
    },
    {
      "entropy": 5.630266952514648,
      "epoch": 0.553199766582377,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004974565922182246,
      "loss": 5.4416,
      "mean_token_accuracy": 0.15794442892074584,
      "num_tokens": 12347586.0,
      "step": 7110
    },
    {
      "entropy": 5.558411169052124,
      "epoch": 0.5535887959540945,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004974524350138777,
      "loss": 5.4119,
      "mean_token_accuracy": 0.16017946004867553,
      "num_tokens": 12356291.0,
      "step": 7115
    },
    {
      "entropy": 5.626056337356568,
      "epoch": 0.5539778253258121,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004974482744341693,
      "loss": 5.5933,
      "mean_token_accuracy": 0.15081025660037994,
      "num_tokens": 12364948.0,
      "step": 7120
    },
    {
      "entropy": 5.722656488418579,
      "epoch": 0.5543668546975297,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004974441104791624,
      "loss": 5.5678,
      "mean_token_accuracy": 0.1566952273249626,
      "num_tokens": 12373497.0,
      "step": 7125
    },
    {
      "entropy": 5.654025077819824,
      "epoch": 0.5547558840692473,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004974399431489201,
      "loss": 5.497,
      "mean_token_accuracy": 0.15983488410711288,
      "num_tokens": 12382198.0,
      "step": 7130
    },
    {
      "entropy": 5.5745806217193605,
      "epoch": 0.5551449134409648,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004974357724435057,
      "loss": 5.488,
      "mean_token_accuracy": 0.16202984526753425,
      "num_tokens": 12391122.0,
      "step": 7135
    },
    {
      "entropy": 5.56509256362915,
      "epoch": 0.5555339428126823,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004974315983629825,
      "loss": 5.4056,
      "mean_token_accuracy": 0.16576439291238784,
      "num_tokens": 12399338.0,
      "step": 7140
    },
    {
      "entropy": 5.603260612487793,
      "epoch": 0.5559229721843999,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004974274209074139,
      "loss": 5.527,
      "mean_token_accuracy": 0.1578882098197937,
      "num_tokens": 12408561.0,
      "step": 7145
    },
    {
      "entropy": 5.649334955215454,
      "epoch": 0.5563120015561175,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004974232400768632,
      "loss": 5.3973,
      "mean_token_accuracy": 0.1637552946805954,
      "num_tokens": 12417727.0,
      "step": 7150
    },
    {
      "entropy": 5.629194450378418,
      "epoch": 0.5567010309278351,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004974190558713939,
      "loss": 5.4356,
      "mean_token_accuracy": 0.1650727704167366,
      "num_tokens": 12426446.0,
      "step": 7155
    },
    {
      "entropy": 5.541499328613281,
      "epoch": 0.5570900602995527,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004974148682910694,
      "loss": 5.5106,
      "mean_token_accuracy": 0.16280363202095033,
      "num_tokens": 12435124.0,
      "step": 7160
    },
    {
      "entropy": 5.63429274559021,
      "epoch": 0.5574790896712701,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004974106773359533,
      "loss": 5.5994,
      "mean_token_accuracy": 0.15252877920866012,
      "num_tokens": 12444073.0,
      "step": 7165
    },
    {
      "entropy": 5.762870645523071,
      "epoch": 0.5578681190429877,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004974064830061091,
      "loss": 5.5822,
      "mean_token_accuracy": 0.16031479835510254,
      "num_tokens": 12452419.0,
      "step": 7170
    },
    {
      "entropy": 5.650390005111694,
      "epoch": 0.5582571484147053,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004974022853016006,
      "loss": 5.5335,
      "mean_token_accuracy": 0.16044811010360718,
      "num_tokens": 12461488.0,
      "step": 7175
    },
    {
      "entropy": 5.5888818264007565,
      "epoch": 0.5586461777864229,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004973980842224913,
      "loss": 5.4542,
      "mean_token_accuracy": 0.16546100229024888,
      "num_tokens": 12470344.0,
      "step": 7180
    },
    {
      "entropy": 5.641972303390503,
      "epoch": 0.5590352071581405,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004973938797688452,
      "loss": 5.4801,
      "mean_token_accuracy": 0.16285377442836763,
      "num_tokens": 12479189.0,
      "step": 7185
    },
    {
      "entropy": 5.642180395126343,
      "epoch": 0.559424236529858,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004973896719407259,
      "loss": 5.5089,
      "mean_token_accuracy": 0.16333740055561066,
      "num_tokens": 12488199.0,
      "step": 7190
    },
    {
      "entropy": 5.548586702346801,
      "epoch": 0.5598132659015755,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004973854607381972,
      "loss": 5.3915,
      "mean_token_accuracy": 0.168252657353878,
      "num_tokens": 12496704.0,
      "step": 7195
    },
    {
      "entropy": 5.485424327850342,
      "epoch": 0.5602022952732931,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004973812461613232,
      "loss": 5.3843,
      "mean_token_accuracy": 0.1694253608584404,
      "num_tokens": 12504724.0,
      "step": 7200
    },
    {
      "entropy": 5.638743734359741,
      "epoch": 0.5605913246450107,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004973770282101677,
      "loss": 5.584,
      "mean_token_accuracy": 0.1591121956706047,
      "num_tokens": 12514614.0,
      "step": 7205
    },
    {
      "entropy": 5.757522487640381,
      "epoch": 0.5609803540167283,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004973728068847947,
      "loss": 5.6017,
      "mean_token_accuracy": 0.15344584509730338,
      "num_tokens": 12522744.0,
      "step": 7210
    },
    {
      "entropy": 5.682862281799316,
      "epoch": 0.5613693833884458,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004973685821852684,
      "loss": 5.515,
      "mean_token_accuracy": 0.16239218264818192,
      "num_tokens": 12531783.0,
      "step": 7215
    },
    {
      "entropy": 5.666183757781982,
      "epoch": 0.5617584127601634,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004973643541116528,
      "loss": 5.5632,
      "mean_token_accuracy": 0.15286958813667298,
      "num_tokens": 12541768.0,
      "step": 7220
    },
    {
      "entropy": 5.710547304153442,
      "epoch": 0.562147442131881,
      "grad_norm": 1.296875,
      "learning_rate": 0.000497360122664012,
      "loss": 5.6068,
      "mean_token_accuracy": 0.15153634250164033,
      "num_tokens": 12550684.0,
      "step": 7225
    },
    {
      "entropy": 5.66706395149231,
      "epoch": 0.5625364715035985,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004973558878424104,
      "loss": 5.5335,
      "mean_token_accuracy": 0.1636952355504036,
      "num_tokens": 12559948.0,
      "step": 7230
    },
    {
      "entropy": 5.615605688095092,
      "epoch": 0.5629255008753161,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004973516496469119,
      "loss": 5.455,
      "mean_token_accuracy": 0.16776744723320008,
      "num_tokens": 12568681.0,
      "step": 7235
    },
    {
      "entropy": 5.675220489501953,
      "epoch": 0.5633145302470336,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004973474080775811,
      "loss": 5.6115,
      "mean_token_accuracy": 0.15134334862232207,
      "num_tokens": 12577724.0,
      "step": 7240
    },
    {
      "entropy": 5.755704021453857,
      "epoch": 0.5637035596187512,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004973431631344824,
      "loss": 5.5211,
      "mean_token_accuracy": 0.15602961778640748,
      "num_tokens": 12585994.0,
      "step": 7245
    },
    {
      "entropy": 5.627366352081299,
      "epoch": 0.5640925889904688,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00049733891481768,
      "loss": 5.5873,
      "mean_token_accuracy": 0.1569364219903946,
      "num_tokens": 12595525.0,
      "step": 7250
    },
    {
      "entropy": 5.607244253158569,
      "epoch": 0.5644816183621864,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004973346631272384,
      "loss": 5.4104,
      "mean_token_accuracy": 0.16649825274944305,
      "num_tokens": 12603645.0,
      "step": 7255
    },
    {
      "entropy": 5.602600908279419,
      "epoch": 0.5648706477339039,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004973304080632222,
      "loss": 5.4839,
      "mean_token_accuracy": 0.16336790174245835,
      "num_tokens": 12612165.0,
      "step": 7260
    },
    {
      "entropy": 5.60393762588501,
      "epoch": 0.5652596771056215,
      "grad_norm": 1.234375,
      "learning_rate": 0.000497326149625696,
      "loss": 5.4002,
      "mean_token_accuracy": 0.17391229122877122,
      "num_tokens": 12620678.0,
      "step": 7265
    },
    {
      "entropy": 5.708744812011719,
      "epoch": 0.565648706477339,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004973218878147244,
      "loss": 5.6067,
      "mean_token_accuracy": 0.15800948888063432,
      "num_tokens": 12628745.0,
      "step": 7270
    },
    {
      "entropy": 5.695216131210327,
      "epoch": 0.5660377358490566,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004973176226303719,
      "loss": 5.6495,
      "mean_token_accuracy": 0.1455956742167473,
      "num_tokens": 12637448.0,
      "step": 7275
    },
    {
      "entropy": 5.683227634429931,
      "epoch": 0.5664267652207742,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004973133540727033,
      "loss": 5.505,
      "mean_token_accuracy": 0.16027553081512452,
      "num_tokens": 12645804.0,
      "step": 7280
    },
    {
      "entropy": 5.610063362121582,
      "epoch": 0.5668157945924918,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004973090821417835,
      "loss": 5.4646,
      "mean_token_accuracy": 0.15778533220291138,
      "num_tokens": 12654156.0,
      "step": 7285
    },
    {
      "entropy": 5.551807832717896,
      "epoch": 0.5672048239642093,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004973048068376772,
      "loss": 5.45,
      "mean_token_accuracy": 0.1671900287270546,
      "num_tokens": 12662271.0,
      "step": 7290
    },
    {
      "entropy": 5.770065355300903,
      "epoch": 0.5675938533359268,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004973005281604492,
      "loss": 5.6205,
      "mean_token_accuracy": 0.1535290725529194,
      "num_tokens": 12671016.0,
      "step": 7295
    },
    {
      "entropy": 5.612900733947754,
      "epoch": 0.5679828827076444,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004972962461101646,
      "loss": 5.5666,
      "mean_token_accuracy": 0.1581275761127472,
      "num_tokens": 12679593.0,
      "step": 7300
    },
    {
      "entropy": 5.585796546936035,
      "epoch": 0.568371912079362,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004972919606868883,
      "loss": 5.4983,
      "mean_token_accuracy": 0.1630539119243622,
      "num_tokens": 12688715.0,
      "step": 7305
    },
    {
      "entropy": 5.647766590118408,
      "epoch": 0.5687609414510796,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004972876718906852,
      "loss": 5.5068,
      "mean_token_accuracy": 0.15621377825736998,
      "num_tokens": 12698078.0,
      "step": 7310
    },
    {
      "entropy": 5.594813871383667,
      "epoch": 0.5691499708227972,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004972833797216206,
      "loss": 5.4949,
      "mean_token_accuracy": 0.16226601600646973,
      "num_tokens": 12706375.0,
      "step": 7315
    },
    {
      "entropy": 5.714466571807861,
      "epoch": 0.5695390001945146,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004972790841797595,
      "loss": 5.681,
      "mean_token_accuracy": 0.15268146097660065,
      "num_tokens": 12715727.0,
      "step": 7320
    },
    {
      "entropy": 5.683112764358521,
      "epoch": 0.5699280295662322,
      "grad_norm": 1.171875,
      "learning_rate": 0.000497274785265167,
      "loss": 5.5742,
      "mean_token_accuracy": 0.1501793384552002,
      "num_tokens": 12725565.0,
      "step": 7325
    },
    {
      "entropy": 5.732530355453491,
      "epoch": 0.5703170589379498,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004972704829779086,
      "loss": 5.6136,
      "mean_token_accuracy": 0.1612092673778534,
      "num_tokens": 12733498.0,
      "step": 7330
    },
    {
      "entropy": 5.56683497428894,
      "epoch": 0.5707060883096674,
      "grad_norm": 1.25,
      "learning_rate": 0.0004972661773180492,
      "loss": 5.4302,
      "mean_token_accuracy": 0.16293097138404847,
      "num_tokens": 12741635.0,
      "step": 7335
    },
    {
      "entropy": 5.557634925842285,
      "epoch": 0.571095117681385,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004972618682856545,
      "loss": 5.4622,
      "mean_token_accuracy": 0.16606659442186356,
      "num_tokens": 12750973.0,
      "step": 7340
    },
    {
      "entropy": 5.6188578605651855,
      "epoch": 0.5714841470531025,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004972575558807896,
      "loss": 5.515,
      "mean_token_accuracy": 0.16870894879102707,
      "num_tokens": 12759075.0,
      "step": 7345
    },
    {
      "entropy": 5.668835544586182,
      "epoch": 0.57187317642482,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004972532401035201,
      "loss": 5.5036,
      "mean_token_accuracy": 0.15811090618371965,
      "num_tokens": 12767485.0,
      "step": 7350
    },
    {
      "entropy": 5.686478853225708,
      "epoch": 0.5722622057965376,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004972489209539114,
      "loss": 5.466,
      "mean_token_accuracy": 0.16516558080911636,
      "num_tokens": 12775128.0,
      "step": 7355
    },
    {
      "entropy": 5.515619373321533,
      "epoch": 0.5726512351682552,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004972445984320292,
      "loss": 5.3917,
      "mean_token_accuracy": 0.16754466593265532,
      "num_tokens": 12783243.0,
      "step": 7360
    },
    {
      "entropy": 5.626593446731567,
      "epoch": 0.5730402645399728,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004972402725379388,
      "loss": 5.6293,
      "mean_token_accuracy": 0.15784782022237778,
      "num_tokens": 12791613.0,
      "step": 7365
    },
    {
      "entropy": 5.677688503265381,
      "epoch": 0.5734292939116903,
      "grad_norm": 1.34375,
      "learning_rate": 0.000497235943271706,
      "loss": 5.5382,
      "mean_token_accuracy": 0.15849522203207017,
      "num_tokens": 12799817.0,
      "step": 7370
    },
    {
      "entropy": 5.699692678451538,
      "epoch": 0.5738183232834079,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004972316106333966,
      "loss": 5.5455,
      "mean_token_accuracy": 0.15867469906806947,
      "num_tokens": 12808994.0,
      "step": 7375
    },
    {
      "entropy": 5.6180887699127195,
      "epoch": 0.5742073526551255,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004972272746230761,
      "loss": 5.5102,
      "mean_token_accuracy": 0.15738145560026168,
      "num_tokens": 12817191.0,
      "step": 7380
    },
    {
      "entropy": 5.579700040817261,
      "epoch": 0.574596382026843,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004972229352408104,
      "loss": 5.4763,
      "mean_token_accuracy": 0.15756721645593644,
      "num_tokens": 12825773.0,
      "step": 7385
    },
    {
      "entropy": 5.730859375,
      "epoch": 0.5749854113985606,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004972185924866655,
      "loss": 5.5423,
      "mean_token_accuracy": 0.15951717048883438,
      "num_tokens": 12834489.0,
      "step": 7390
    },
    {
      "entropy": 5.688281154632568,
      "epoch": 0.5753744407702781,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004972142463607071,
      "loss": 5.5438,
      "mean_token_accuracy": 0.1614192619919777,
      "num_tokens": 12843179.0,
      "step": 7395
    },
    {
      "entropy": 5.756889772415161,
      "epoch": 0.5757634701419957,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004972098968630012,
      "loss": 5.5635,
      "mean_token_accuracy": 0.15483097806572915,
      "num_tokens": 12851243.0,
      "step": 7400
    },
    {
      "entropy": 5.586697053909302,
      "epoch": 0.5761524995137133,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004972055439936137,
      "loss": 5.6136,
      "mean_token_accuracy": 0.15806748121976852,
      "num_tokens": 12860267.0,
      "step": 7405
    },
    {
      "entropy": 5.7580828189849855,
      "epoch": 0.5765415288854309,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004972011877526109,
      "loss": 5.6198,
      "mean_token_accuracy": 0.15912649929523467,
      "num_tokens": 12870022.0,
      "step": 7410
    },
    {
      "entropy": 5.685928153991699,
      "epoch": 0.5769305582571485,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004971968281400587,
      "loss": 5.5692,
      "mean_token_accuracy": 0.15594390779733658,
      "num_tokens": 12879227.0,
      "step": 7415
    },
    {
      "entropy": 5.606605052947998,
      "epoch": 0.5773195876288659,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004971924651560233,
      "loss": 5.5368,
      "mean_token_accuracy": 0.15589874535799025,
      "num_tokens": 12888180.0,
      "step": 7420
    },
    {
      "entropy": 5.65610785484314,
      "epoch": 0.5777086170005835,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004971880988005709,
      "loss": 5.5442,
      "mean_token_accuracy": 0.15969115048646926,
      "num_tokens": 12896767.0,
      "step": 7425
    },
    {
      "entropy": 5.669794178009033,
      "epoch": 0.5780976463723011,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004971837290737677,
      "loss": 5.4226,
      "mean_token_accuracy": 0.16824220716953278,
      "num_tokens": 12904766.0,
      "step": 7430
    },
    {
      "entropy": 5.532117033004761,
      "epoch": 0.5784866757440187,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004971793559756801,
      "loss": 5.4926,
      "mean_token_accuracy": 0.16067837029695511,
      "num_tokens": 12913808.0,
      "step": 7435
    },
    {
      "entropy": 5.594570016860962,
      "epoch": 0.5788757051157363,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004971749795063744,
      "loss": 5.5089,
      "mean_token_accuracy": 0.15930237844586373,
      "num_tokens": 12923103.0,
      "step": 7440
    },
    {
      "entropy": 5.669405317306518,
      "epoch": 0.5792647344874537,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000497170599665917,
      "loss": 5.4628,
      "mean_token_accuracy": 0.16535604894161224,
      "num_tokens": 12931168.0,
      "step": 7445
    },
    {
      "entropy": 5.599840116500855,
      "epoch": 0.5796537638591713,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004971662164543743,
      "loss": 5.5417,
      "mean_token_accuracy": 0.16020704358816146,
      "num_tokens": 12939875.0,
      "step": 7450
    },
    {
      "entropy": 5.635864973068237,
      "epoch": 0.5800427932308889,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004971618298718131,
      "loss": 5.465,
      "mean_token_accuracy": 0.16293471604585646,
      "num_tokens": 12947883.0,
      "step": 7455
    },
    {
      "entropy": 5.806446123123169,
      "epoch": 0.5804318226026065,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004971574399182995,
      "loss": 5.7043,
      "mean_token_accuracy": 0.14791340231895447,
      "num_tokens": 12956940.0,
      "step": 7460
    },
    {
      "entropy": 5.6557777404785154,
      "epoch": 0.5808208519743241,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004971530465939005,
      "loss": 5.3983,
      "mean_token_accuracy": 0.16397302299737931,
      "num_tokens": 12965003.0,
      "step": 7465
    },
    {
      "entropy": 5.5010154247283936,
      "epoch": 0.5812098813460417,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004971486498986825,
      "loss": 5.4505,
      "mean_token_accuracy": 0.16312854290008544,
      "num_tokens": 12973323.0,
      "step": 7470
    },
    {
      "entropy": 5.580051279067993,
      "epoch": 0.5815989107177592,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004971442498327124,
      "loss": 5.4901,
      "mean_token_accuracy": 0.16068859547376632,
      "num_tokens": 12982654.0,
      "step": 7475
    },
    {
      "entropy": 5.690056228637696,
      "epoch": 0.5819879400894767,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004971398463960568,
      "loss": 5.5354,
      "mean_token_accuracy": 0.1595580130815506,
      "num_tokens": 12991781.0,
      "step": 7480
    },
    {
      "entropy": 5.718348455429077,
      "epoch": 0.5823769694611943,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004971354395887827,
      "loss": 5.6212,
      "mean_token_accuracy": 0.15529465973377227,
      "num_tokens": 13000539.0,
      "step": 7485
    },
    {
      "entropy": 5.69674596786499,
      "epoch": 0.5827659988329119,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004971310294109569,
      "loss": 5.4839,
      "mean_token_accuracy": 0.16353426575660707,
      "num_tokens": 13008638.0,
      "step": 7490
    },
    {
      "entropy": 5.573290920257568,
      "epoch": 0.5831550282046295,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004971266158626461,
      "loss": 5.4683,
      "mean_token_accuracy": 0.16310660541057587,
      "num_tokens": 13017217.0,
      "step": 7495
    },
    {
      "entropy": 5.58029465675354,
      "epoch": 0.583544057576347,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004971221989439177,
      "loss": 5.5276,
      "mean_token_accuracy": 0.1623905122280121,
      "num_tokens": 13025278.0,
      "step": 7500
    },
    {
      "entropy": 5.65991792678833,
      "epoch": 0.5839330869480646,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004971177786548384,
      "loss": 5.5386,
      "mean_token_accuracy": 0.16195032447576524,
      "num_tokens": 13034484.0,
      "step": 7505
    },
    {
      "entropy": 5.706937837600708,
      "epoch": 0.5843221163197821,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004971133549954753,
      "loss": 5.6141,
      "mean_token_accuracy": 0.1537620708346367,
      "num_tokens": 13042831.0,
      "step": 7510
    },
    {
      "entropy": 5.674683380126953,
      "epoch": 0.5847111456914997,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004971089279658954,
      "loss": 5.5786,
      "mean_token_accuracy": 0.155495323240757,
      "num_tokens": 13051614.0,
      "step": 7515
    },
    {
      "entropy": 5.6314798355102536,
      "epoch": 0.5851001750632173,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004971044975661663,
      "loss": 5.5423,
      "mean_token_accuracy": 0.1582693189382553,
      "num_tokens": 13060179.0,
      "step": 7520
    },
    {
      "entropy": 5.594365358352661,
      "epoch": 0.5854892044349348,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004971000637963549,
      "loss": 5.4412,
      "mean_token_accuracy": 0.17228898257017136,
      "num_tokens": 13068328.0,
      "step": 7525
    },
    {
      "entropy": 5.546153354644775,
      "epoch": 0.5858782338066524,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004970956266565285,
      "loss": 5.5121,
      "mean_token_accuracy": 0.1631189078092575,
      "num_tokens": 13076879.0,
      "step": 7530
    },
    {
      "entropy": 5.7159558773040775,
      "epoch": 0.58626726317837,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004970911861467544,
      "loss": 5.545,
      "mean_token_accuracy": 0.15528045892715453,
      "num_tokens": 13085123.0,
      "step": 7535
    },
    {
      "entropy": 5.565436458587646,
      "epoch": 0.5866562925500876,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004970867422671002,
      "loss": 5.4991,
      "mean_token_accuracy": 0.1610252410173416,
      "num_tokens": 13093884.0,
      "step": 7540
    },
    {
      "entropy": 5.722029304504394,
      "epoch": 0.5870453219218051,
      "grad_norm": 1.234375,
      "learning_rate": 0.000497082295017633,
      "loss": 5.5672,
      "mean_token_accuracy": 0.15952380746603012,
      "num_tokens": 13102581.0,
      "step": 7545
    },
    {
      "entropy": 5.674138116836548,
      "epoch": 0.5874343512935226,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004970778443984206,
      "loss": 5.5392,
      "mean_token_accuracy": 0.15848752111196518,
      "num_tokens": 13111307.0,
      "step": 7550
    },
    {
      "entropy": 5.6157981872558596,
      "epoch": 0.5878233806652402,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004970733904095303,
      "loss": 5.5581,
      "mean_token_accuracy": 0.1646633267402649,
      "num_tokens": 13119279.0,
      "step": 7555
    },
    {
      "entropy": 5.6678361892700195,
      "epoch": 0.5882124100369578,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004970689330510298,
      "loss": 5.5698,
      "mean_token_accuracy": 0.15818312615156174,
      "num_tokens": 13127753.0,
      "step": 7560
    },
    {
      "entropy": 5.65282506942749,
      "epoch": 0.5886014394086754,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004970644723229868,
      "loss": 5.6063,
      "mean_token_accuracy": 0.15884243100881576,
      "num_tokens": 13136881.0,
      "step": 7565
    },
    {
      "entropy": 5.603133583068848,
      "epoch": 0.588990468780393,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004970600082254687,
      "loss": 5.4849,
      "mean_token_accuracy": 0.1598943993449211,
      "num_tokens": 13145291.0,
      "step": 7570
    },
    {
      "entropy": 5.585998964309693,
      "epoch": 0.5893794981521104,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004970555407585436,
      "loss": 5.4385,
      "mean_token_accuracy": 0.16188057288527488,
      "num_tokens": 13153560.0,
      "step": 7575
    },
    {
      "entropy": 5.6350850582122805,
      "epoch": 0.589768527523828,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000497051069922279,
      "loss": 5.4911,
      "mean_token_accuracy": 0.1563738167285919,
      "num_tokens": 13162304.0,
      "step": 7580
    },
    {
      "entropy": 5.636582612991333,
      "epoch": 0.5901575568955456,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004970465957167429,
      "loss": 5.5215,
      "mean_token_accuracy": 0.16496615558862687,
      "num_tokens": 13171404.0,
      "step": 7585
    },
    {
      "entropy": 5.73821382522583,
      "epoch": 0.5905465862672632,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000497042118142003,
      "loss": 5.5863,
      "mean_token_accuracy": 0.16058476269245148,
      "num_tokens": 13180001.0,
      "step": 7590
    },
    {
      "entropy": 5.653850984573364,
      "epoch": 0.5909356156389808,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004970376371981275,
      "loss": 5.5314,
      "mean_token_accuracy": 0.16310483813285828,
      "num_tokens": 13189749.0,
      "step": 7595
    },
    {
      "entropy": 5.541713905334473,
      "epoch": 0.5913246450106983,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004970331528851842,
      "loss": 5.5307,
      "mean_token_accuracy": 0.1635299563407898,
      "num_tokens": 13199169.0,
      "step": 7600
    },
    {
      "entropy": 5.629769039154053,
      "epoch": 0.5917136743824158,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004970286652032412,
      "loss": 5.4597,
      "mean_token_accuracy": 0.17126554250717163,
      "num_tokens": 13207014.0,
      "step": 7605
    },
    {
      "entropy": 5.6623927593231205,
      "epoch": 0.5921027037541334,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004970241741523667,
      "loss": 5.4475,
      "mean_token_accuracy": 0.16537835448980331,
      "num_tokens": 13215333.0,
      "step": 7610
    },
    {
      "entropy": 5.630066680908203,
      "epoch": 0.592491733125851,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004970196797326286,
      "loss": 5.5617,
      "mean_token_accuracy": 0.15445425510406494,
      "num_tokens": 13223740.0,
      "step": 7615
    },
    {
      "entropy": 5.591768646240235,
      "epoch": 0.5928807624975686,
      "grad_norm": 1.375,
      "learning_rate": 0.0004970151819440955,
      "loss": 5.4998,
      "mean_token_accuracy": 0.16681206226348877,
      "num_tokens": 13232065.0,
      "step": 7620
    },
    {
      "entropy": 5.631966590881348,
      "epoch": 0.5932697918692861,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004970106807868351,
      "loss": 5.4803,
      "mean_token_accuracy": 0.16643381863832474,
      "num_tokens": 13241171.0,
      "step": 7625
    },
    {
      "entropy": 5.500733757019043,
      "epoch": 0.5936588212410037,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000497006176260916,
      "loss": 5.408,
      "mean_token_accuracy": 0.1683339461684227,
      "num_tokens": 13249553.0,
      "step": 7630
    },
    {
      "entropy": 5.618449926376343,
      "epoch": 0.5940478506127213,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004970016683664068,
      "loss": 5.4993,
      "mean_token_accuracy": 0.16241494119167327,
      "num_tokens": 13258585.0,
      "step": 7635
    },
    {
      "entropy": 5.63908166885376,
      "epoch": 0.5944368799844388,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004969971571033754,
      "loss": 5.5406,
      "mean_token_accuracy": 0.16231602877378465,
      "num_tokens": 13267317.0,
      "step": 7640
    },
    {
      "entropy": 5.583299398422241,
      "epoch": 0.5948259093561564,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004969926424718906,
      "loss": 5.3646,
      "mean_token_accuracy": 0.16274376809597016,
      "num_tokens": 13275705.0,
      "step": 7645
    },
    {
      "entropy": 5.621941947937012,
      "epoch": 0.5952149387278739,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004969881244720207,
      "loss": 5.5039,
      "mean_token_accuracy": 0.16544682830572127,
      "num_tokens": 13284046.0,
      "step": 7650
    },
    {
      "entropy": 5.607423925399781,
      "epoch": 0.5956039680995915,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004969836031038343,
      "loss": 5.491,
      "mean_token_accuracy": 0.16659966111183167,
      "num_tokens": 13292451.0,
      "step": 7655
    },
    {
      "entropy": 5.726130676269531,
      "epoch": 0.5959929974713091,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004969790783674002,
      "loss": 5.6048,
      "mean_token_accuracy": 0.16234291940927506,
      "num_tokens": 13300788.0,
      "step": 7660
    },
    {
      "entropy": 5.633881759643555,
      "epoch": 0.5963820268430267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004969745502627868,
      "loss": 5.5436,
      "mean_token_accuracy": 0.16315071731805803,
      "num_tokens": 13309235.0,
      "step": 7665
    },
    {
      "entropy": 5.579603385925293,
      "epoch": 0.5967710562147442,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004969700187900629,
      "loss": 5.4488,
      "mean_token_accuracy": 0.15795199573040009,
      "num_tokens": 13317331.0,
      "step": 7670
    },
    {
      "entropy": 5.571296358108521,
      "epoch": 0.5971600855864618,
      "grad_norm": 1.25,
      "learning_rate": 0.0004969654839492973,
      "loss": 5.4657,
      "mean_token_accuracy": 0.1541156552731991,
      "num_tokens": 13327296.0,
      "step": 7675
    },
    {
      "entropy": 5.680892848968506,
      "epoch": 0.5975491149581793,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004969609457405588,
      "loss": 5.5333,
      "mean_token_accuracy": 0.15591974854469298,
      "num_tokens": 13336650.0,
      "step": 7680
    },
    {
      "entropy": 5.65597710609436,
      "epoch": 0.5979381443298969,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004969564041639163,
      "loss": 5.5016,
      "mean_token_accuracy": 0.15822356045246125,
      "num_tokens": 13345353.0,
      "step": 7685
    },
    {
      "entropy": 5.650791454315185,
      "epoch": 0.5983271737016145,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004969518592194386,
      "loss": 5.493,
      "mean_token_accuracy": 0.1636109843850136,
      "num_tokens": 13354302.0,
      "step": 7690
    },
    {
      "entropy": 5.556919431686401,
      "epoch": 0.5987162030733321,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004969473109071946,
      "loss": 5.429,
      "mean_token_accuracy": 0.16622557789087294,
      "num_tokens": 13362950.0,
      "step": 7695
    },
    {
      "entropy": 5.605585622787475,
      "epoch": 0.5991052324450497,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004969427592272535,
      "loss": 5.5442,
      "mean_token_accuracy": 0.16100695878267288,
      "num_tokens": 13371383.0,
      "step": 7700
    },
    {
      "entropy": 5.676916074752808,
      "epoch": 0.5994942618167671,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004969382041796843,
      "loss": 5.5322,
      "mean_token_accuracy": 0.1646827206015587,
      "num_tokens": 13379767.0,
      "step": 7705
    },
    {
      "entropy": 5.646383142471313,
      "epoch": 0.5998832911884847,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000496933645764556,
      "loss": 5.5101,
      "mean_token_accuracy": 0.16011914163827895,
      "num_tokens": 13387970.0,
      "step": 7710
    },
    {
      "entropy": 5.502738094329834,
      "epoch": 0.6002723205602023,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004969290839819381,
      "loss": 5.4716,
      "mean_token_accuracy": 0.16803790628910065,
      "num_tokens": 13396605.0,
      "step": 7715
    },
    {
      "entropy": 5.576162385940552,
      "epoch": 0.6006613499319199,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004969245188318994,
      "loss": 5.4073,
      "mean_token_accuracy": 0.16737466305494308,
      "num_tokens": 13406028.0,
      "step": 7720
    },
    {
      "entropy": 5.620474529266358,
      "epoch": 0.6010503793036375,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004969199503145093,
      "loss": 5.4631,
      "mean_token_accuracy": 0.1679922953248024,
      "num_tokens": 13414927.0,
      "step": 7725
    },
    {
      "entropy": 5.543609142303467,
      "epoch": 0.601439408675355,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004969153784298374,
      "loss": 5.3961,
      "mean_token_accuracy": 0.1650015026330948,
      "num_tokens": 13423020.0,
      "step": 7730
    },
    {
      "entropy": 5.509767532348633,
      "epoch": 0.6018284380470725,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004969108031779527,
      "loss": 5.4256,
      "mean_token_accuracy": 0.16520768851041795,
      "num_tokens": 13431514.0,
      "step": 7735
    },
    {
      "entropy": 5.660559034347534,
      "epoch": 0.6022174674187901,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004969062245589247,
      "loss": 5.5793,
      "mean_token_accuracy": 0.15292226746678353,
      "num_tokens": 13440333.0,
      "step": 7740
    },
    {
      "entropy": 5.587594604492187,
      "epoch": 0.6026064967905077,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004969016425728231,
      "loss": 5.5293,
      "mean_token_accuracy": 0.16127558797597885,
      "num_tokens": 13448841.0,
      "step": 7745
    },
    {
      "entropy": 5.64748740196228,
      "epoch": 0.6029955261622253,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004968970572197172,
      "loss": 5.5967,
      "mean_token_accuracy": 0.1554821714758873,
      "num_tokens": 13457846.0,
      "step": 7750
    },
    {
      "entropy": 5.720643472671509,
      "epoch": 0.6033845555339428,
      "grad_norm": 1.5,
      "learning_rate": 0.0004968924684996765,
      "loss": 5.6013,
      "mean_token_accuracy": 0.15000174343585967,
      "num_tokens": 13466242.0,
      "step": 7755
    },
    {
      "entropy": 5.667198324203492,
      "epoch": 0.6037735849056604,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000496887876412771,
      "loss": 5.5521,
      "mean_token_accuracy": 0.15576667636632918,
      "num_tokens": 13474639.0,
      "step": 7760
    },
    {
      "entropy": 5.549106216430664,
      "epoch": 0.6041626142773779,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004968832809590701,
      "loss": 5.422,
      "mean_token_accuracy": 0.16609416007995606,
      "num_tokens": 13483673.0,
      "step": 7765
    },
    {
      "entropy": 5.6652998447418215,
      "epoch": 0.6045516436490955,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004968786821386435,
      "loss": 5.548,
      "mean_token_accuracy": 0.15655329674482346,
      "num_tokens": 13493001.0,
      "step": 7770
    },
    {
      "entropy": 5.741840982437134,
      "epoch": 0.6049406730208131,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004968740799515611,
      "loss": 5.5567,
      "mean_token_accuracy": 0.16150247752666474,
      "num_tokens": 13500878.0,
      "step": 7775
    },
    {
      "entropy": 5.671801662445068,
      "epoch": 0.6053297023925306,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004968694743978926,
      "loss": 5.5946,
      "mean_token_accuracy": 0.15671397149562835,
      "num_tokens": 13509945.0,
      "step": 7780
    },
    {
      "entropy": 5.571211910247802,
      "epoch": 0.6057187317642482,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004968648654777081,
      "loss": 5.5376,
      "mean_token_accuracy": 0.16182100623846055,
      "num_tokens": 13520024.0,
      "step": 7785
    },
    {
      "entropy": 5.65290937423706,
      "epoch": 0.6061077611359658,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004968602531910773,
      "loss": 5.6371,
      "mean_token_accuracy": 0.16058965027332306,
      "num_tokens": 13529136.0,
      "step": 7790
    },
    {
      "entropy": 5.689796590805054,
      "epoch": 0.6064967905076833,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004968556375380704,
      "loss": 5.4396,
      "mean_token_accuracy": 0.1580638200044632,
      "num_tokens": 13537001.0,
      "step": 7795
    },
    {
      "entropy": 5.599765300750732,
      "epoch": 0.6068858198794009,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004968510185187571,
      "loss": 5.4237,
      "mean_token_accuracy": 0.16344021111726761,
      "num_tokens": 13545520.0,
      "step": 7800
    },
    {
      "entropy": 5.5511781692504885,
      "epoch": 0.6072748492511184,
      "grad_norm": 1.265625,
      "learning_rate": 0.000496846396133208,
      "loss": 5.5149,
      "mean_token_accuracy": 0.16309064030647277,
      "num_tokens": 13554770.0,
      "step": 7805
    },
    {
      "entropy": 5.723441743850708,
      "epoch": 0.607663878622836,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004968417703814928,
      "loss": 5.5484,
      "mean_token_accuracy": 0.1591284915804863,
      "num_tokens": 13562797.0,
      "step": 7810
    },
    {
      "entropy": 5.640573644638062,
      "epoch": 0.6080529079945536,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004968371412636818,
      "loss": 5.4822,
      "mean_token_accuracy": 0.16195359975099563,
      "num_tokens": 13572017.0,
      "step": 7815
    },
    {
      "entropy": 5.6671302318573,
      "epoch": 0.6084419373662712,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004968325087798453,
      "loss": 5.5032,
      "mean_token_accuracy": 0.16161488592624665,
      "num_tokens": 13580245.0,
      "step": 7820
    },
    {
      "entropy": 5.566740989685059,
      "epoch": 0.6088309667379888,
      "grad_norm": 1.25,
      "learning_rate": 0.0004968278729300536,
      "loss": 5.4775,
      "mean_token_accuracy": 0.15461314618587493,
      "num_tokens": 13588719.0,
      "step": 7825
    },
    {
      "entropy": 5.6440144062042235,
      "epoch": 0.6092199961097062,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004968232337143769,
      "loss": 5.5267,
      "mean_token_accuracy": 0.16302523910999298,
      "num_tokens": 13597345.0,
      "step": 7830
    },
    {
      "entropy": 5.708196401596069,
      "epoch": 0.6096090254814238,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004968185911328858,
      "loss": 5.5449,
      "mean_token_accuracy": 0.15957221239805222,
      "num_tokens": 13605468.0,
      "step": 7835
    },
    {
      "entropy": 5.68671293258667,
      "epoch": 0.6099980548531414,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004968139451856506,
      "loss": 5.5597,
      "mean_token_accuracy": 0.15538589060306549,
      "num_tokens": 13614085.0,
      "step": 7840
    },
    {
      "entropy": 5.584887838363647,
      "epoch": 0.610387084224859,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004968092958727419,
      "loss": 5.3919,
      "mean_token_accuracy": 0.1681410327553749,
      "num_tokens": 13622370.0,
      "step": 7845
    },
    {
      "entropy": 5.604237079620361,
      "epoch": 0.6107761135965766,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00049680464319423,
      "loss": 5.504,
      "mean_token_accuracy": 0.16654858738183975,
      "num_tokens": 13630615.0,
      "step": 7850
    },
    {
      "entropy": 5.626179981231689,
      "epoch": 0.6111651429682942,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004967999871501858,
      "loss": 5.4418,
      "mean_token_accuracy": 0.16906313896179198,
      "num_tokens": 13638461.0,
      "step": 7855
    },
    {
      "entropy": 5.548437309265137,
      "epoch": 0.6115541723400116,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004967953277406798,
      "loss": 5.5482,
      "mean_token_accuracy": 0.15815697908401488,
      "num_tokens": 13647266.0,
      "step": 7860
    },
    {
      "entropy": 5.548516845703125,
      "epoch": 0.6119432017117292,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004967906649657828,
      "loss": 5.5184,
      "mean_token_accuracy": 0.1585494562983513,
      "num_tokens": 13655930.0,
      "step": 7865
    },
    {
      "entropy": 5.688598775863648,
      "epoch": 0.6123322310834468,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004967859988255655,
      "loss": 5.5089,
      "mean_token_accuracy": 0.16099477857351302,
      "num_tokens": 13664427.0,
      "step": 7870
    },
    {
      "entropy": 5.638605737686158,
      "epoch": 0.6127212604551644,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004967813293200985,
      "loss": 5.5267,
      "mean_token_accuracy": 0.15878960192203523,
      "num_tokens": 13673120.0,
      "step": 7875
    },
    {
      "entropy": 5.516750860214233,
      "epoch": 0.613110289826882,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004967766564494529,
      "loss": 5.3643,
      "mean_token_accuracy": 0.1739079013466835,
      "num_tokens": 13681918.0,
      "step": 7880
    },
    {
      "entropy": 5.490477180480957,
      "epoch": 0.6134993191985995,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004967719802136996,
      "loss": 5.4542,
      "mean_token_accuracy": 0.1686467006802559,
      "num_tokens": 13689904.0,
      "step": 7885
    },
    {
      "entropy": 5.61046290397644,
      "epoch": 0.613888348570317,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004967673006129094,
      "loss": 5.4858,
      "mean_token_accuracy": 0.16289936900138854,
      "num_tokens": 13697861.0,
      "step": 7890
    },
    {
      "entropy": 5.6796611785888675,
      "epoch": 0.6142773779420346,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004967626176471536,
      "loss": 5.6335,
      "mean_token_accuracy": 0.1635378211736679,
      "num_tokens": 13706058.0,
      "step": 7895
    },
    {
      "entropy": 5.670239639282227,
      "epoch": 0.6146664073137522,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004967579313165028,
      "loss": 5.5263,
      "mean_token_accuracy": 0.16258027404546738,
      "num_tokens": 13715081.0,
      "step": 7900
    },
    {
      "entropy": 5.633441066741943,
      "epoch": 0.6150554366854698,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004967532416210284,
      "loss": 5.4327,
      "mean_token_accuracy": 0.168162602186203,
      "num_tokens": 13723311.0,
      "step": 7905
    },
    {
      "entropy": 5.636973237991333,
      "epoch": 0.6154444660571873,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004967485485608016,
      "loss": 5.4742,
      "mean_token_accuracy": 0.16119187474250793,
      "num_tokens": 13731737.0,
      "step": 7910
    },
    {
      "entropy": 5.554210138320923,
      "epoch": 0.6158334954289049,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004967438521358934,
      "loss": 5.5403,
      "mean_token_accuracy": 0.16576073318719864,
      "num_tokens": 13740255.0,
      "step": 7915
    },
    {
      "entropy": 5.634121465682983,
      "epoch": 0.6162225248006225,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004967391523463754,
      "loss": 5.4753,
      "mean_token_accuracy": 0.1679953768849373,
      "num_tokens": 13748656.0,
      "step": 7920
    },
    {
      "entropy": 5.673994874954223,
      "epoch": 0.61661155417234,
      "grad_norm": 1.25,
      "learning_rate": 0.0004967344491923185,
      "loss": 5.478,
      "mean_token_accuracy": 0.16169667840003968,
      "num_tokens": 13757539.0,
      "step": 7925
    },
    {
      "entropy": 5.655036735534668,
      "epoch": 0.6170005835440576,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004967297426737943,
      "loss": 5.4913,
      "mean_token_accuracy": 0.1639968767762184,
      "num_tokens": 13766569.0,
      "step": 7930
    },
    {
      "entropy": 5.657715559005737,
      "epoch": 0.6173896129157751,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004967250327908742,
      "loss": 5.5197,
      "mean_token_accuracy": 0.1606484279036522,
      "num_tokens": 13775071.0,
      "step": 7935
    },
    {
      "entropy": 5.5887847423553465,
      "epoch": 0.6177786422874927,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004967203195436297,
      "loss": 5.5136,
      "mean_token_accuracy": 0.15838043615221978,
      "num_tokens": 13783319.0,
      "step": 7940
    },
    {
      "entropy": 5.4905599594116214,
      "epoch": 0.6181676716592103,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004967156029321322,
      "loss": 5.4393,
      "mean_token_accuracy": 0.1690957471728325,
      "num_tokens": 13792746.0,
      "step": 7945
    },
    {
      "entropy": 5.566557741165161,
      "epoch": 0.6185567010309279,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004967108829564532,
      "loss": 5.3971,
      "mean_token_accuracy": 0.16903592348098756,
      "num_tokens": 13801632.0,
      "step": 7950
    },
    {
      "entropy": 5.617258453369141,
      "epoch": 0.6189457304026454,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004967061596166646,
      "loss": 5.5283,
      "mean_token_accuracy": 0.16719207912683487,
      "num_tokens": 13810527.0,
      "step": 7955
    },
    {
      "entropy": 5.63814344406128,
      "epoch": 0.6193347597743629,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004967014329128378,
      "loss": 5.4722,
      "mean_token_accuracy": 0.1592711701989174,
      "num_tokens": 13819885.0,
      "step": 7960
    },
    {
      "entropy": 5.560896205902099,
      "epoch": 0.6197237891460805,
      "grad_norm": 1.25,
      "learning_rate": 0.0004966967028450448,
      "loss": 5.4691,
      "mean_token_accuracy": 0.16611457765102386,
      "num_tokens": 13828509.0,
      "step": 7965
    },
    {
      "entropy": 5.623311471939087,
      "epoch": 0.6201128185177981,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000496691969413357,
      "loss": 5.477,
      "mean_token_accuracy": 0.16131770461797715,
      "num_tokens": 13837196.0,
      "step": 7970
    },
    {
      "entropy": 5.655043363571167,
      "epoch": 0.6205018478895157,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004966872326178466,
      "loss": 5.5393,
      "mean_token_accuracy": 0.1589474231004715,
      "num_tokens": 13845849.0,
      "step": 7975
    },
    {
      "entropy": 5.602649211883545,
      "epoch": 0.6208908772612333,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004966824924585851,
      "loss": 5.4455,
      "mean_token_accuracy": 0.16728405505418778,
      "num_tokens": 13853430.0,
      "step": 7980
    },
    {
      "entropy": 5.655062198638916,
      "epoch": 0.6212799066329507,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004966777489356448,
      "loss": 5.6301,
      "mean_token_accuracy": 0.15381192564964294,
      "num_tokens": 13862520.0,
      "step": 7985
    },
    {
      "entropy": 5.719164943695068,
      "epoch": 0.6216689360046683,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004966730020490975,
      "loss": 5.5485,
      "mean_token_accuracy": 0.15611168891191482,
      "num_tokens": 13871064.0,
      "step": 7990
    },
    {
      "entropy": 5.582553148269653,
      "epoch": 0.6220579653763859,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000496668251799015,
      "loss": 5.4504,
      "mean_token_accuracy": 0.16296833306550979,
      "num_tokens": 13879952.0,
      "step": 7995
    },
    {
      "entropy": 5.659041118621826,
      "epoch": 0.6224469947481035,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004966634981854699,
      "loss": 5.607,
      "mean_token_accuracy": 0.15741588026285172,
      "num_tokens": 13888054.0,
      "step": 8000
    },
    {
      "entropy": 5.7299822807312015,
      "epoch": 0.6228360241198211,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004966587412085339,
      "loss": 5.5741,
      "mean_token_accuracy": 0.15745998919010162,
      "num_tokens": 13897051.0,
      "step": 8005
    },
    {
      "entropy": 5.606602478027344,
      "epoch": 0.6232250534915386,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004966539808682793,
      "loss": 5.5407,
      "mean_token_accuracy": 0.15712806433439255,
      "num_tokens": 13906175.0,
      "step": 8010
    },
    {
      "entropy": 5.629466104507446,
      "epoch": 0.6236140828632561,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004966492171647783,
      "loss": 5.5188,
      "mean_token_accuracy": 0.16155388802289963,
      "num_tokens": 13915283.0,
      "step": 8015
    },
    {
      "entropy": 5.634824514389038,
      "epoch": 0.6240031122349737,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004966444500981032,
      "loss": 5.5424,
      "mean_token_accuracy": 0.15690438449382782,
      "num_tokens": 13924007.0,
      "step": 8020
    },
    {
      "entropy": 5.610903978347778,
      "epoch": 0.6243921416066913,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004966396796683265,
      "loss": 5.4475,
      "mean_token_accuracy": 0.16363016664981841,
      "num_tokens": 13932694.0,
      "step": 8025
    },
    {
      "entropy": 5.70683536529541,
      "epoch": 0.6247811709784089,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004966349058755204,
      "loss": 5.5605,
      "mean_token_accuracy": 0.1526412174105644,
      "num_tokens": 13941817.0,
      "step": 8030
    },
    {
      "entropy": 5.672229528427124,
      "epoch": 0.6251702003501264,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004966301287197573,
      "loss": 5.4059,
      "mean_token_accuracy": 0.1685480296611786,
      "num_tokens": 13950413.0,
      "step": 8035
    },
    {
      "entropy": 5.646150207519531,
      "epoch": 0.625559229721844,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004966253482011098,
      "loss": 5.572,
      "mean_token_accuracy": 0.15196625590324403,
      "num_tokens": 13959415.0,
      "step": 8040
    },
    {
      "entropy": 5.563976001739502,
      "epoch": 0.6259482590935616,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004966205643196503,
      "loss": 5.4214,
      "mean_token_accuracy": 0.17012683898210526,
      "num_tokens": 13967889.0,
      "step": 8045
    },
    {
      "entropy": 5.507785272598267,
      "epoch": 0.6263372884652791,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004966157770754516,
      "loss": 5.4019,
      "mean_token_accuracy": 0.16366679668426515,
      "num_tokens": 13976190.0,
      "step": 8050
    },
    {
      "entropy": 5.710351610183716,
      "epoch": 0.6267263178369967,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000496610986468586,
      "loss": 5.5437,
      "mean_token_accuracy": 0.15392907708883286,
      "num_tokens": 13984940.0,
      "step": 8055
    },
    {
      "entropy": 5.647889471054077,
      "epoch": 0.6271153472087143,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004966061924991267,
      "loss": 5.5413,
      "mean_token_accuracy": 0.15577153116464615,
      "num_tokens": 13993740.0,
      "step": 8060
    },
    {
      "entropy": 5.650064134597779,
      "epoch": 0.6275043765804318,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000496601395167146,
      "loss": 5.5813,
      "mean_token_accuracy": 0.15795315206050872,
      "num_tokens": 14002401.0,
      "step": 8065
    },
    {
      "entropy": 5.644233179092407,
      "epoch": 0.6278934059521494,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004965965944727168,
      "loss": 5.5667,
      "mean_token_accuracy": 0.16592306047677993,
      "num_tokens": 14011197.0,
      "step": 8070
    },
    {
      "entropy": 5.642372131347656,
      "epoch": 0.628282435323867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004965917904159121,
      "loss": 5.4392,
      "mean_token_accuracy": 0.16610783487558364,
      "num_tokens": 14020344.0,
      "step": 8075
    },
    {
      "entropy": 5.604736757278443,
      "epoch": 0.6286714646955845,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004965869829968046,
      "loss": 5.454,
      "mean_token_accuracy": 0.16645651310682297,
      "num_tokens": 14029117.0,
      "step": 8080
    },
    {
      "entropy": 5.514641666412354,
      "epoch": 0.6290604940673021,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004965821722154674,
      "loss": 5.3465,
      "mean_token_accuracy": 0.16917043179273605,
      "num_tokens": 14038395.0,
      "step": 8085
    },
    {
      "entropy": 5.59578046798706,
      "epoch": 0.6294495234390196,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004965773580719734,
      "loss": 5.4528,
      "mean_token_accuracy": 0.16234350353479385,
      "num_tokens": 14046983.0,
      "step": 8090
    },
    {
      "entropy": 5.473298215866089,
      "epoch": 0.6298385528107372,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004965725405663957,
      "loss": 5.3554,
      "mean_token_accuracy": 0.17009160816669464,
      "num_tokens": 14056261.0,
      "step": 8095
    },
    {
      "entropy": 5.537156677246093,
      "epoch": 0.6302275821824548,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004965677196988073,
      "loss": 5.5108,
      "mean_token_accuracy": 0.16207924634218215,
      "num_tokens": 14065371.0,
      "step": 8100
    },
    {
      "entropy": 5.685234451293946,
      "epoch": 0.6306166115541724,
      "grad_norm": 1.25,
      "learning_rate": 0.0004965628954692815,
      "loss": 5.586,
      "mean_token_accuracy": 0.15611766949295997,
      "num_tokens": 14074007.0,
      "step": 8105
    },
    {
      "entropy": 5.606713151931762,
      "epoch": 0.63100564092589,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004965580678778913,
      "loss": 5.4316,
      "mean_token_accuracy": 0.16509680300951005,
      "num_tokens": 14083259.0,
      "step": 8110
    },
    {
      "entropy": 5.529731750488281,
      "epoch": 0.6313946702976074,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004965532369247102,
      "loss": 5.4341,
      "mean_token_accuracy": 0.17011858373880387,
      "num_tokens": 14090993.0,
      "step": 8115
    },
    {
      "entropy": 5.719294881820678,
      "epoch": 0.631783699669325,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004965484026098111,
      "loss": 5.5588,
      "mean_token_accuracy": 0.1669984355568886,
      "num_tokens": 14099722.0,
      "step": 8120
    },
    {
      "entropy": 5.737652587890625,
      "epoch": 0.6321727290410426,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004965435649332679,
      "loss": 5.5464,
      "mean_token_accuracy": 0.15991235971450807,
      "num_tokens": 14108483.0,
      "step": 8125
    },
    {
      "entropy": 5.670386028289795,
      "epoch": 0.6325617584127602,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004965387238951536,
      "loss": 5.5557,
      "mean_token_accuracy": 0.1556137830018997,
      "num_tokens": 14117187.0,
      "step": 8130
    },
    {
      "entropy": 5.592771053314209,
      "epoch": 0.6329507877844778,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004965338794955418,
      "loss": 5.3816,
      "mean_token_accuracy": 0.17273490279912948,
      "num_tokens": 14125424.0,
      "step": 8135
    },
    {
      "entropy": 5.595821237564087,
      "epoch": 0.6333398171561953,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004965290317345061,
      "loss": 5.4692,
      "mean_token_accuracy": 0.16249141097068787,
      "num_tokens": 14133038.0,
      "step": 8140
    },
    {
      "entropy": 5.641076898574829,
      "epoch": 0.6337288465279128,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004965241806121198,
      "loss": 5.6105,
      "mean_token_accuracy": 0.15508505403995515,
      "num_tokens": 14141114.0,
      "step": 8145
    },
    {
      "entropy": 5.65889196395874,
      "epoch": 0.6341178758996304,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004965193261284567,
      "loss": 5.4617,
      "mean_token_accuracy": 0.16860485225915908,
      "num_tokens": 14149681.0,
      "step": 8150
    },
    {
      "entropy": 5.613321352005005,
      "epoch": 0.634506905271348,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004965144682835904,
      "loss": 5.5022,
      "mean_token_accuracy": 0.15759261399507524,
      "num_tokens": 14158510.0,
      "step": 8155
    },
    {
      "entropy": 5.607683420181274,
      "epoch": 0.6348959346430656,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004965096070775946,
      "loss": 5.5648,
      "mean_token_accuracy": 0.1657479539513588,
      "num_tokens": 14166897.0,
      "step": 8160
    },
    {
      "entropy": 5.6775308609008786,
      "epoch": 0.6352849640147831,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004965047425105431,
      "loss": 5.4951,
      "mean_token_accuracy": 0.16363131403923034,
      "num_tokens": 14174692.0,
      "step": 8165
    },
    {
      "entropy": 5.577294635772705,
      "epoch": 0.6356739933865007,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004964998745825097,
      "loss": 5.4785,
      "mean_token_accuracy": 0.16154927909374237,
      "num_tokens": 14183779.0,
      "step": 8170
    },
    {
      "entropy": 5.5535430908203125,
      "epoch": 0.6360630227582182,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004964950032935682,
      "loss": 5.5367,
      "mean_token_accuracy": 0.16089922338724136,
      "num_tokens": 14191809.0,
      "step": 8175
    },
    {
      "entropy": 5.635007095336914,
      "epoch": 0.6364520521299358,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004964901286437927,
      "loss": 5.4979,
      "mean_token_accuracy": 0.1615813195705414,
      "num_tokens": 14200652.0,
      "step": 8180
    },
    {
      "entropy": 5.677788734436035,
      "epoch": 0.6368410815016534,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004964852506332568,
      "loss": 5.492,
      "mean_token_accuracy": 0.16762958317995072,
      "num_tokens": 14209694.0,
      "step": 8185
    },
    {
      "entropy": 5.596168851852417,
      "epoch": 0.6372301108733709,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000496480369262035,
      "loss": 5.5348,
      "mean_token_accuracy": 0.15523997992277144,
      "num_tokens": 14218040.0,
      "step": 8190
    },
    {
      "entropy": 5.5753233432769775,
      "epoch": 0.6376191402450885,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000496475484530201,
      "loss": 5.4708,
      "mean_token_accuracy": 0.16816412657499313,
      "num_tokens": 14225908.0,
      "step": 8195
    },
    {
      "entropy": 5.602357196807861,
      "epoch": 0.6380081696168061,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004964705964378292,
      "loss": 5.489,
      "mean_token_accuracy": 0.15687236040830613,
      "num_tokens": 14234140.0,
      "step": 8200
    },
    {
      "entropy": 5.659995603561401,
      "epoch": 0.6383971989885237,
      "grad_norm": 1.25,
      "learning_rate": 0.0004964657049849934,
      "loss": 5.5405,
      "mean_token_accuracy": 0.1630015939474106,
      "num_tokens": 14243590.0,
      "step": 8205
    },
    {
      "entropy": 5.696007823944091,
      "epoch": 0.6387862283602412,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004964608101717681,
      "loss": 5.5051,
      "mean_token_accuracy": 0.16301901191473006,
      "num_tokens": 14252249.0,
      "step": 8210
    },
    {
      "entropy": 5.566278123855591,
      "epoch": 0.6391752577319587,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004964559119982275,
      "loss": 5.4273,
      "mean_token_accuracy": 0.16251214891672133,
      "num_tokens": 14261147.0,
      "step": 8215
    },
    {
      "entropy": 5.603458547592163,
      "epoch": 0.6395642871036763,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000496451010464446,
      "loss": 5.4843,
      "mean_token_accuracy": 0.1621839627623558,
      "num_tokens": 14270309.0,
      "step": 8220
    },
    {
      "entropy": 5.588337135314942,
      "epoch": 0.6399533164753939,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004964461055704978,
      "loss": 5.4718,
      "mean_token_accuracy": 0.16583188623189926,
      "num_tokens": 14279379.0,
      "step": 8225
    },
    {
      "entropy": 5.589170265197754,
      "epoch": 0.6403423458471115,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004964411973164574,
      "loss": 5.44,
      "mean_token_accuracy": 0.157003553211689,
      "num_tokens": 14288136.0,
      "step": 8230
    },
    {
      "entropy": 5.582326507568359,
      "epoch": 0.6407313752188291,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004964362857023993,
      "loss": 5.3695,
      "mean_token_accuracy": 0.1649335041642189,
      "num_tokens": 14296437.0,
      "step": 8235
    },
    {
      "entropy": 5.573055362701416,
      "epoch": 0.6411204045905465,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000496431370728398,
      "loss": 5.4149,
      "mean_token_accuracy": 0.17282897979021072,
      "num_tokens": 14304861.0,
      "step": 8240
    },
    {
      "entropy": 5.516162967681884,
      "epoch": 0.6415094339622641,
      "grad_norm": 1.25,
      "learning_rate": 0.0004964264523945281,
      "loss": 5.4703,
      "mean_token_accuracy": 0.16430750489234924,
      "num_tokens": 14312861.0,
      "step": 8245
    },
    {
      "entropy": 5.523329734802246,
      "epoch": 0.6418984633339817,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004964215307008643,
      "loss": 5.3051,
      "mean_token_accuracy": 0.1748490273952484,
      "num_tokens": 14321651.0,
      "step": 8250
    },
    {
      "entropy": 5.620496082305908,
      "epoch": 0.6422874927056993,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004964166056474811,
      "loss": 5.4906,
      "mean_token_accuracy": 0.1738407775759697,
      "num_tokens": 14329234.0,
      "step": 8255
    },
    {
      "entropy": 5.601748752593994,
      "epoch": 0.6426765220774169,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004964116772344534,
      "loss": 5.6329,
      "mean_token_accuracy": 0.15985049307346344,
      "num_tokens": 14337603.0,
      "step": 8260
    },
    {
      "entropy": 5.657437133789062,
      "epoch": 0.6430655514491345,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004964067454618559,
      "loss": 5.4872,
      "mean_token_accuracy": 0.15789955705404282,
      "num_tokens": 14346634.0,
      "step": 8265
    },
    {
      "entropy": 5.635518598556518,
      "epoch": 0.6434545808208519,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004964018103297634,
      "loss": 5.5209,
      "mean_token_accuracy": 0.15715200453996658,
      "num_tokens": 14355336.0,
      "step": 8270
    },
    {
      "entropy": 5.623025035858154,
      "epoch": 0.6438436101925695,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004963968718382509,
      "loss": 5.4738,
      "mean_token_accuracy": 0.16136810630559922,
      "num_tokens": 14364434.0,
      "step": 8275
    },
    {
      "entropy": 5.6118512630462645,
      "epoch": 0.6442326395642871,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004963919299873931,
      "loss": 5.4195,
      "mean_token_accuracy": 0.17004749327898025,
      "num_tokens": 14372454.0,
      "step": 8280
    },
    {
      "entropy": 5.617539024353027,
      "epoch": 0.6446216689360047,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004963869847772653,
      "loss": 5.4552,
      "mean_token_accuracy": 0.1625766173005104,
      "num_tokens": 14381381.0,
      "step": 8285
    },
    {
      "entropy": 5.607727670669556,
      "epoch": 0.6450106983077223,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004963820362079423,
      "loss": 5.5705,
      "mean_token_accuracy": 0.16276483684778215,
      "num_tokens": 14390853.0,
      "step": 8290
    },
    {
      "entropy": 5.580901575088501,
      "epoch": 0.6453997276794398,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004963770842794994,
      "loss": 5.4192,
      "mean_token_accuracy": 0.16428908556699753,
      "num_tokens": 14399897.0,
      "step": 8295
    },
    {
      "entropy": 5.5460669040679935,
      "epoch": 0.6457887570511573,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004963721289920115,
      "loss": 5.4796,
      "mean_token_accuracy": 0.16398324817419052,
      "num_tokens": 14408192.0,
      "step": 8300
    },
    {
      "entropy": 5.564640665054322,
      "epoch": 0.6461777864228749,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004963671703455538,
      "loss": 5.36,
      "mean_token_accuracy": 0.17036048471927642,
      "num_tokens": 14416004.0,
      "step": 8305
    },
    {
      "entropy": 5.598507833480835,
      "epoch": 0.6465668157945925,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004963622083402018,
      "loss": 5.4028,
      "mean_token_accuracy": 0.16379630863666533,
      "num_tokens": 14424573.0,
      "step": 8310
    },
    {
      "entropy": 5.544586086273194,
      "epoch": 0.6469558451663101,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004963572429760305,
      "loss": 5.3823,
      "mean_token_accuracy": 0.1730259910225868,
      "num_tokens": 14432756.0,
      "step": 8315
    },
    {
      "entropy": 5.575740766525269,
      "epoch": 0.6473448745380276,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004963522742531155,
      "loss": 5.4877,
      "mean_token_accuracy": 0.16610362082719804,
      "num_tokens": 14441695.0,
      "step": 8320
    },
    {
      "entropy": 5.60623254776001,
      "epoch": 0.6477339039097452,
      "grad_norm": 1.328125,
      "learning_rate": 0.000496347302171532,
      "loss": 5.5827,
      "mean_token_accuracy": 0.1528534933924675,
      "num_tokens": 14451233.0,
      "step": 8325
    },
    {
      "entropy": 5.641898536682129,
      "epoch": 0.6481229332814628,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004963423267313554,
      "loss": 5.4688,
      "mean_token_accuracy": 0.16386925280094147,
      "num_tokens": 14459757.0,
      "step": 8330
    },
    {
      "entropy": 5.61696081161499,
      "epoch": 0.6485119626531803,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004963373479326614,
      "loss": 5.5488,
      "mean_token_accuracy": 0.15485638454556466,
      "num_tokens": 14468308.0,
      "step": 8335
    },
    {
      "entropy": 5.627341651916504,
      "epoch": 0.6489009920248979,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004963323657755255,
      "loss": 5.4983,
      "mean_token_accuracy": 0.1617353767156601,
      "num_tokens": 14476654.0,
      "step": 8340
    },
    {
      "entropy": 5.626075267791748,
      "epoch": 0.6492900213966154,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004963273802600231,
      "loss": 5.5665,
      "mean_token_accuracy": 0.16124435812234877,
      "num_tokens": 14484960.0,
      "step": 8345
    },
    {
      "entropy": 5.667200803756714,
      "epoch": 0.649679050768333,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00049632239138623,
      "loss": 5.5591,
      "mean_token_accuracy": 0.1546272099018097,
      "num_tokens": 14493708.0,
      "step": 8350
    },
    {
      "entropy": 5.686009931564331,
      "epoch": 0.6500680801400506,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004963173991542219,
      "loss": 5.5062,
      "mean_token_accuracy": 0.16160760223865508,
      "num_tokens": 14502123.0,
      "step": 8355
    },
    {
      "entropy": 5.639478588104248,
      "epoch": 0.6504571095117682,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004963124035640746,
      "loss": 5.4837,
      "mean_token_accuracy": 0.1642873018980026,
      "num_tokens": 14510782.0,
      "step": 8360
    },
    {
      "entropy": 5.5654679298400875,
      "epoch": 0.6508461388834857,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004963074046158637,
      "loss": 5.4363,
      "mean_token_accuracy": 0.165517221391201,
      "num_tokens": 14518560.0,
      "step": 8365
    },
    {
      "entropy": 5.578605318069458,
      "epoch": 0.6512351682552032,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004963024023096652,
      "loss": 5.4652,
      "mean_token_accuracy": 0.15973718613386154,
      "num_tokens": 14527541.0,
      "step": 8370
    },
    {
      "entropy": 5.647238302230835,
      "epoch": 0.6516241976269208,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004962973966455551,
      "loss": 5.5072,
      "mean_token_accuracy": 0.16062749028205872,
      "num_tokens": 14535590.0,
      "step": 8375
    },
    {
      "entropy": 5.529531717300415,
      "epoch": 0.6520132269986384,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004962923876236092,
      "loss": 5.3948,
      "mean_token_accuracy": 0.16741624921560289,
      "num_tokens": 14544929.0,
      "step": 8380
    },
    {
      "entropy": 5.643860578536987,
      "epoch": 0.652402256370356,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004962873752439035,
      "loss": 5.5668,
      "mean_token_accuracy": 0.15424012243747712,
      "num_tokens": 14553883.0,
      "step": 8385
    },
    {
      "entropy": 5.6864824295043945,
      "epoch": 0.6527912857420736,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004962823595065141,
      "loss": 5.5595,
      "mean_token_accuracy": 0.15886417478322984,
      "num_tokens": 14563085.0,
      "step": 8390
    },
    {
      "entropy": 5.621539354324341,
      "epoch": 0.653180315113791,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004962773404115172,
      "loss": 5.4502,
      "mean_token_accuracy": 0.16572141647338867,
      "num_tokens": 14571444.0,
      "step": 8395
    },
    {
      "entropy": 5.584204578399659,
      "epoch": 0.6535693444855086,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004962723179589886,
      "loss": 5.5285,
      "mean_token_accuracy": 0.16222902685403823,
      "num_tokens": 14580039.0,
      "step": 8400
    },
    {
      "entropy": 5.616039657592774,
      "epoch": 0.6539583738572262,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000496267292149005,
      "loss": 5.3982,
      "mean_token_accuracy": 0.17028344422578812,
      "num_tokens": 14588322.0,
      "step": 8405
    },
    {
      "entropy": 5.586518049240112,
      "epoch": 0.6543474032289438,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004962622629816423,
      "loss": 5.3553,
      "mean_token_accuracy": 0.16853828281164168,
      "num_tokens": 14597503.0,
      "step": 8410
    },
    {
      "entropy": 5.568887376785279,
      "epoch": 0.6547364326006614,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000496257230456977,
      "loss": 5.5,
      "mean_token_accuracy": 0.15949926078319548,
      "num_tokens": 14605804.0,
      "step": 8415
    },
    {
      "entropy": 5.479747915267945,
      "epoch": 0.6551254619723789,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004962521945750855,
      "loss": 5.4117,
      "mean_token_accuracy": 0.1690138041973114,
      "num_tokens": 14614064.0,
      "step": 8420
    },
    {
      "entropy": 5.656314086914063,
      "epoch": 0.6555144913440965,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000496247155336044,
      "loss": 5.4042,
      "mean_token_accuracy": 0.1671632260084152,
      "num_tokens": 14622231.0,
      "step": 8425
    },
    {
      "entropy": 5.591139268875122,
      "epoch": 0.655903520715814,
      "grad_norm": 1.375,
      "learning_rate": 0.0004962421127399291,
      "loss": 5.5285,
      "mean_token_accuracy": 0.15751022547483445,
      "num_tokens": 14631244.0,
      "step": 8430
    },
    {
      "entropy": 5.562242650985718,
      "epoch": 0.6562925500875316,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004962370667868172,
      "loss": 5.3998,
      "mean_token_accuracy": 0.1660147950053215,
      "num_tokens": 14639575.0,
      "step": 8435
    },
    {
      "entropy": 5.5867517471313475,
      "epoch": 0.6566815794592492,
      "grad_norm": 1.25,
      "learning_rate": 0.0004962320174767851,
      "loss": 5.4424,
      "mean_token_accuracy": 0.16145324409008027,
      "num_tokens": 14648810.0,
      "step": 8440
    },
    {
      "entropy": 5.55507116317749,
      "epoch": 0.6570706088309667,
      "grad_norm": 1.34375,
      "learning_rate": 0.000496226964809909,
      "loss": 5.3041,
      "mean_token_accuracy": 0.17762577682733535,
      "num_tokens": 14657783.0,
      "step": 8445
    },
    {
      "entropy": 5.605140972137451,
      "epoch": 0.6574596382026843,
      "grad_norm": 1.3125,
      "learning_rate": 0.000496221908786266,
      "loss": 5.5022,
      "mean_token_accuracy": 0.16277231127023697,
      "num_tokens": 14666872.0,
      "step": 8450
    },
    {
      "entropy": 5.670653057098389,
      "epoch": 0.6578486675744019,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004962168494059327,
      "loss": 5.6411,
      "mean_token_accuracy": 0.15669171661138534,
      "num_tokens": 14675762.0,
      "step": 8455
    },
    {
      "entropy": 5.56172194480896,
      "epoch": 0.6582376969461194,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004962117866689857,
      "loss": 5.2859,
      "mean_token_accuracy": 0.17307938039302825,
      "num_tokens": 14683928.0,
      "step": 8460
    },
    {
      "entropy": 5.565886545181274,
      "epoch": 0.658626726317837,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000496206720575502,
      "loss": 5.5287,
      "mean_token_accuracy": 0.16260690838098527,
      "num_tokens": 14692411.0,
      "step": 8465
    },
    {
      "entropy": 5.6268086433410645,
      "epoch": 0.6590157556895546,
      "grad_norm": 1.375,
      "learning_rate": 0.0004962016511255584,
      "loss": 5.529,
      "mean_token_accuracy": 0.15660884082317353,
      "num_tokens": 14701002.0,
      "step": 8470
    },
    {
      "entropy": 5.736037635803223,
      "epoch": 0.6594047850612721,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004961965783192318,
      "loss": 5.5798,
      "mean_token_accuracy": 0.1550491198897362,
      "num_tokens": 14710172.0,
      "step": 8475
    },
    {
      "entropy": 5.604672145843506,
      "epoch": 0.6597938144329897,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004961915021565992,
      "loss": 5.506,
      "mean_token_accuracy": 0.1578040361404419,
      "num_tokens": 14719446.0,
      "step": 8480
    },
    {
      "entropy": 5.669936275482177,
      "epoch": 0.6601828438047073,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004961864226377377,
      "loss": 5.5241,
      "mean_token_accuracy": 0.1672406256198883,
      "num_tokens": 14728193.0,
      "step": 8485
    },
    {
      "entropy": 5.612123584747314,
      "epoch": 0.6605718731764249,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004961813397627241,
      "loss": 5.4015,
      "mean_token_accuracy": 0.15923748910427094,
      "num_tokens": 14736930.0,
      "step": 8490
    },
    {
      "entropy": 5.519996547698975,
      "epoch": 0.6609609025481424,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004961762535316358,
      "loss": 5.4006,
      "mean_token_accuracy": 0.16979765444993972,
      "num_tokens": 14745995.0,
      "step": 8495
    },
    {
      "entropy": 5.539706087112426,
      "epoch": 0.6613499319198599,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004961711639445499,
      "loss": 5.4552,
      "mean_token_accuracy": 0.16044401675462722,
      "num_tokens": 14754827.0,
      "step": 8500
    },
    {
      "entropy": 5.6871413230896,
      "epoch": 0.6617389612915775,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004961660710015435,
      "loss": 5.5372,
      "mean_token_accuracy": 0.1606838136911392,
      "num_tokens": 14763594.0,
      "step": 8505
    },
    {
      "entropy": 5.571682167053223,
      "epoch": 0.6621279906632951,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004961609747026943,
      "loss": 5.3809,
      "mean_token_accuracy": 0.16946823447942733,
      "num_tokens": 14772366.0,
      "step": 8510
    },
    {
      "entropy": 5.559627532958984,
      "epoch": 0.6625170200350127,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004961558750480791,
      "loss": 5.4419,
      "mean_token_accuracy": 0.16206909716129303,
      "num_tokens": 14780738.0,
      "step": 8515
    },
    {
      "entropy": 5.575452375411987,
      "epoch": 0.6629060494067303,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004961507720377756,
      "loss": 5.4279,
      "mean_token_accuracy": 0.15737494230270385,
      "num_tokens": 14788982.0,
      "step": 8520
    },
    {
      "entropy": 5.612622594833374,
      "epoch": 0.6632950787784477,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004961456656718611,
      "loss": 5.4331,
      "mean_token_accuracy": 0.16051857620477678,
      "num_tokens": 14797832.0,
      "step": 8525
    },
    {
      "entropy": 5.628104496002197,
      "epoch": 0.6636841081501653,
      "grad_norm": 1.34375,
      "learning_rate": 0.000496140555950413,
      "loss": 5.4644,
      "mean_token_accuracy": 0.16462608724832534,
      "num_tokens": 14806391.0,
      "step": 8530
    },
    {
      "entropy": 5.624013757705688,
      "epoch": 0.6640731375218829,
      "grad_norm": 1.25,
      "learning_rate": 0.0004961354428735091,
      "loss": 5.4863,
      "mean_token_accuracy": 0.16068560034036636,
      "num_tokens": 14814920.0,
      "step": 8535
    },
    {
      "entropy": 5.5480677604675295,
      "epoch": 0.6644621668936005,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004961303264412267,
      "loss": 5.4286,
      "mean_token_accuracy": 0.16541460156440735,
      "num_tokens": 14823541.0,
      "step": 8540
    },
    {
      "entropy": 5.561609935760498,
      "epoch": 0.6648511962653181,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004961252066536437,
      "loss": 5.3836,
      "mean_token_accuracy": 0.165013188123703,
      "num_tokens": 14832128.0,
      "step": 8545
    },
    {
      "entropy": 5.57411208152771,
      "epoch": 0.6652402256370356,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004961200835108375,
      "loss": 5.5613,
      "mean_token_accuracy": 0.151299324631691,
      "num_tokens": 14841434.0,
      "step": 8550
    },
    {
      "entropy": 5.5968406200408936,
      "epoch": 0.6656292550087531,
      "grad_norm": 1.296875,
      "learning_rate": 0.000496114957012886,
      "loss": 5.478,
      "mean_token_accuracy": 0.1647263750433922,
      "num_tokens": 14850255.0,
      "step": 8555
    },
    {
      "entropy": 5.634866571426391,
      "epoch": 0.6660182843804707,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004961098271598671,
      "loss": 5.5512,
      "mean_token_accuracy": 0.15822786688804627,
      "num_tokens": 14858924.0,
      "step": 8560
    },
    {
      "entropy": 5.633492422103882,
      "epoch": 0.6664073137521883,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004961046939518585,
      "loss": 5.4966,
      "mean_token_accuracy": 0.16522532403469087,
      "num_tokens": 14867426.0,
      "step": 8565
    },
    {
      "entropy": 5.615775680541992,
      "epoch": 0.6667963431239059,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004960995573889379,
      "loss": 5.4596,
      "mean_token_accuracy": 0.16334087252616883,
      "num_tokens": 14876218.0,
      "step": 8570
    },
    {
      "entropy": 5.6453392028808596,
      "epoch": 0.6671853724956234,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004960944174711836,
      "loss": 5.4846,
      "mean_token_accuracy": 0.16281638890504838,
      "num_tokens": 14883995.0,
      "step": 8575
    },
    {
      "entropy": 5.5647047996521,
      "epoch": 0.667574401867341,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004960892741986734,
      "loss": 5.4822,
      "mean_token_accuracy": 0.16294094771146775,
      "num_tokens": 14892659.0,
      "step": 8580
    },
    {
      "entropy": 5.65052547454834,
      "epoch": 0.6679634312390585,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004960841275714853,
      "loss": 5.4955,
      "mean_token_accuracy": 0.1627752289175987,
      "num_tokens": 14901518.0,
      "step": 8585
    },
    {
      "entropy": 5.57407398223877,
      "epoch": 0.6683524606107761,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004960789775896975,
      "loss": 5.4921,
      "mean_token_accuracy": 0.16375407576560974,
      "num_tokens": 14910171.0,
      "step": 8590
    },
    {
      "entropy": 5.621286678314209,
      "epoch": 0.6687414899824937,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004960738242533881,
      "loss": 5.4645,
      "mean_token_accuracy": 0.16164526343345642,
      "num_tokens": 14918651.0,
      "step": 8595
    },
    {
      "entropy": 5.578404378890991,
      "epoch": 0.6691305193542112,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004960686675626353,
      "loss": 5.4767,
      "mean_token_accuracy": 0.1597375601530075,
      "num_tokens": 14926951.0,
      "step": 8600
    },
    {
      "entropy": 5.6784745216369625,
      "epoch": 0.6695195487259288,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004960635075175173,
      "loss": 5.5685,
      "mean_token_accuracy": 0.16335559040307998,
      "num_tokens": 14935528.0,
      "step": 8605
    },
    {
      "entropy": 5.6433753490448,
      "epoch": 0.6699085780976464,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004960583441181124,
      "loss": 5.416,
      "mean_token_accuracy": 0.1689871668815613,
      "num_tokens": 14944117.0,
      "step": 8610
    },
    {
      "entropy": 5.5415881156921385,
      "epoch": 0.670297607469364,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004960531773644992,
      "loss": 5.4159,
      "mean_token_accuracy": 0.16331241875886918,
      "num_tokens": 14952957.0,
      "step": 8615
    },
    {
      "entropy": 5.616297960281372,
      "epoch": 0.6706866368410815,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004960480072567557,
      "loss": 5.4821,
      "mean_token_accuracy": 0.1618007093667984,
      "num_tokens": 14960942.0,
      "step": 8620
    },
    {
      "entropy": 5.692008066177368,
      "epoch": 0.671075666212799,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004960428337949604,
      "loss": 5.5597,
      "mean_token_accuracy": 0.15529187321662902,
      "num_tokens": 14970579.0,
      "step": 8625
    },
    {
      "entropy": 5.56552357673645,
      "epoch": 0.6714646955845166,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004960376569791921,
      "loss": 5.4021,
      "mean_token_accuracy": 0.16066819429397583,
      "num_tokens": 14979353.0,
      "step": 8630
    },
    {
      "entropy": 5.633704805374146,
      "epoch": 0.6718537249562342,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004960324768095291,
      "loss": 5.5675,
      "mean_token_accuracy": 0.16641758233308793,
      "num_tokens": 14988435.0,
      "step": 8635
    },
    {
      "entropy": 5.593199825286865,
      "epoch": 0.6722427543279518,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004960272932860502,
      "loss": 5.4895,
      "mean_token_accuracy": 0.15605200678110123,
      "num_tokens": 14996834.0,
      "step": 8640
    },
    {
      "entropy": 5.654151725769043,
      "epoch": 0.6726317836996694,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004960221064088338,
      "loss": 5.5603,
      "mean_token_accuracy": 0.1535288691520691,
      "num_tokens": 15004988.0,
      "step": 8645
    },
    {
      "entropy": 5.715420198440552,
      "epoch": 0.6730208130713868,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004960169161779588,
      "loss": 5.4939,
      "mean_token_accuracy": 0.16684936881065368,
      "num_tokens": 15013482.0,
      "step": 8650
    },
    {
      "entropy": 5.570951509475708,
      "epoch": 0.6734098424431044,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004960117225935038,
      "loss": 5.4084,
      "mean_token_accuracy": 0.16789936274290085,
      "num_tokens": 15021506.0,
      "step": 8655
    },
    {
      "entropy": 5.603825616836548,
      "epoch": 0.673798871814822,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004960065256555477,
      "loss": 5.4528,
      "mean_token_accuracy": 0.168421433866024,
      "num_tokens": 15029880.0,
      "step": 8660
    },
    {
      "entropy": 5.609187889099121,
      "epoch": 0.6741879011865396,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004960013253641695,
      "loss": 5.4354,
      "mean_token_accuracy": 0.16533171981573105,
      "num_tokens": 15038080.0,
      "step": 8665
    },
    {
      "entropy": 5.571533727645874,
      "epoch": 0.6745769305582572,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004959961217194477,
      "loss": 5.4554,
      "mean_token_accuracy": 0.16327146589756011,
      "num_tokens": 15046779.0,
      "step": 8670
    },
    {
      "entropy": 5.638508653640747,
      "epoch": 0.6749659599299748,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004959909147214615,
      "loss": 5.4751,
      "mean_token_accuracy": 0.1650352582335472,
      "num_tokens": 15055557.0,
      "step": 8675
    },
    {
      "entropy": 5.57130126953125,
      "epoch": 0.6753549893016922,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004959857043702902,
      "loss": 5.3538,
      "mean_token_accuracy": 0.16675955057144165,
      "num_tokens": 15063054.0,
      "step": 8680
    },
    {
      "entropy": 5.584498167037964,
      "epoch": 0.6757440186734098,
      "grad_norm": 1.25,
      "learning_rate": 0.0004959804906660124,
      "loss": 5.4378,
      "mean_token_accuracy": 0.1649323508143425,
      "num_tokens": 15071510.0,
      "step": 8685
    },
    {
      "entropy": 5.575815725326538,
      "epoch": 0.6761330480451274,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004959752736087073,
      "loss": 5.4584,
      "mean_token_accuracy": 0.15143937170505523,
      "num_tokens": 15080524.0,
      "step": 8690
    },
    {
      "entropy": 5.618631601333618,
      "epoch": 0.676522077416845,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004959700531984543,
      "loss": 5.4776,
      "mean_token_accuracy": 0.1599847637116909,
      "num_tokens": 15089689.0,
      "step": 8695
    },
    {
      "entropy": 5.600354814529419,
      "epoch": 0.6769111067885626,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004959648294353324,
      "loss": 5.3826,
      "mean_token_accuracy": 0.16723806858062745,
      "num_tokens": 15097970.0,
      "step": 8700
    },
    {
      "entropy": 5.517818355560303,
      "epoch": 0.6773001361602801,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004959596023194208,
      "loss": 5.477,
      "mean_token_accuracy": 0.16314728260040284,
      "num_tokens": 15107786.0,
      "step": 8705
    },
    {
      "entropy": 5.614211654663086,
      "epoch": 0.6776891655319977,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000495954371850799,
      "loss": 5.4648,
      "mean_token_accuracy": 0.1633009821176529,
      "num_tokens": 15116370.0,
      "step": 8710
    },
    {
      "entropy": 5.651032733917236,
      "epoch": 0.6780781949037152,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004959491380295463,
      "loss": 5.4745,
      "mean_token_accuracy": 0.16634214967489241,
      "num_tokens": 15124961.0,
      "step": 8715
    },
    {
      "entropy": 5.592713689804077,
      "epoch": 0.6784672242754328,
      "grad_norm": 1.25,
      "learning_rate": 0.0004959439008557422,
      "loss": 5.4991,
      "mean_token_accuracy": 0.15993877798318862,
      "num_tokens": 15134230.0,
      "step": 8720
    },
    {
      "entropy": 5.596553564071655,
      "epoch": 0.6788562536471504,
      "grad_norm": 1.15625,
      "learning_rate": 0.000495938660329466,
      "loss": 5.4378,
      "mean_token_accuracy": 0.15456221699714662,
      "num_tokens": 15143585.0,
      "step": 8725
    },
    {
      "entropy": 5.575191020965576,
      "epoch": 0.6792452830188679,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004959334164507973,
      "loss": 5.3635,
      "mean_token_accuracy": 0.15851821079850198,
      "num_tokens": 15151834.0,
      "step": 8730
    },
    {
      "entropy": 5.460168266296387,
      "epoch": 0.6796343123905855,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004959281692198155,
      "loss": 5.4311,
      "mean_token_accuracy": 0.1607486918568611,
      "num_tokens": 15160630.0,
      "step": 8735
    },
    {
      "entropy": 5.635277843475341,
      "epoch": 0.6800233417623031,
      "grad_norm": 1.25,
      "learning_rate": 0.0004959229186366007,
      "loss": 5.4064,
      "mean_token_accuracy": 0.16892074495553971,
      "num_tokens": 15168999.0,
      "step": 8740
    },
    {
      "entropy": 5.635550451278687,
      "epoch": 0.6804123711340206,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000495917664701232,
      "loss": 5.4715,
      "mean_token_accuracy": 0.169201098382473,
      "num_tokens": 15177507.0,
      "step": 8745
    },
    {
      "entropy": 5.591236257553101,
      "epoch": 0.6808014005057382,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004959124074137894,
      "loss": 5.5199,
      "mean_token_accuracy": 0.15848170071840287,
      "num_tokens": 15186843.0,
      "step": 8750
    },
    {
      "entropy": 5.58202338218689,
      "epoch": 0.6811904298774557,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004959071467743526,
      "loss": 5.4138,
      "mean_token_accuracy": 0.16391640305519103,
      "num_tokens": 15195081.0,
      "step": 8755
    },
    {
      "entropy": 5.627256250381469,
      "epoch": 0.6815794592491733,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004959018827830016,
      "loss": 5.5119,
      "mean_token_accuracy": 0.16483797878026962,
      "num_tokens": 15203538.0,
      "step": 8760
    },
    {
      "entropy": 5.6962555885314945,
      "epoch": 0.6819684886208909,
      "grad_norm": 1.3125,
      "learning_rate": 0.000495896615439816,
      "loss": 5.5421,
      "mean_token_accuracy": 0.15884492993354798,
      "num_tokens": 15211845.0,
      "step": 8765
    },
    {
      "entropy": 5.626863718032837,
      "epoch": 0.6823575179926085,
      "grad_norm": 1.25,
      "learning_rate": 0.0004958913447448759,
      "loss": 5.53,
      "mean_token_accuracy": 0.157900071144104,
      "num_tokens": 15220128.0,
      "step": 8770
    },
    {
      "entropy": 5.642973899841309,
      "epoch": 0.682746547364326,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004958860706982613,
      "loss": 5.5269,
      "mean_token_accuracy": 0.15867448598146439,
      "num_tokens": 15228741.0,
      "step": 8775
    },
    {
      "entropy": 5.698912525177002,
      "epoch": 0.6831355767360435,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000495880793300052,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1600990653038025,
      "num_tokens": 15237674.0,
      "step": 8780
    },
    {
      "entropy": 5.580166912078857,
      "epoch": 0.6835246061077611,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004958755125503284,
      "loss": 5.3718,
      "mean_token_accuracy": 0.17182213217020034,
      "num_tokens": 15245727.0,
      "step": 8785
    },
    {
      "entropy": 5.5644673824310305,
      "epoch": 0.6839136354794787,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004958702284491703,
      "loss": 5.4895,
      "mean_token_accuracy": 0.16259710639715194,
      "num_tokens": 15254511.0,
      "step": 8790
    },
    {
      "entropy": 5.64516282081604,
      "epoch": 0.6843026648511963,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004958649409966581,
      "loss": 5.4614,
      "mean_token_accuracy": 0.16489705294370652,
      "num_tokens": 15263500.0,
      "step": 8795
    },
    {
      "entropy": 5.667985534667968,
      "epoch": 0.6846916942229139,
      "grad_norm": 1.46875,
      "learning_rate": 0.000495859650192872,
      "loss": 5.4711,
      "mean_token_accuracy": 0.16129500716924666,
      "num_tokens": 15271960.0,
      "step": 8800
    },
    {
      "entropy": 5.651174736022949,
      "epoch": 0.6850807235946313,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004958543560378922,
      "loss": 5.6499,
      "mean_token_accuracy": 0.15663300901651384,
      "num_tokens": 15280576.0,
      "step": 8805
    },
    {
      "entropy": 5.607110977172852,
      "epoch": 0.6854697529663489,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004958490585317991,
      "loss": 5.5239,
      "mean_token_accuracy": 0.16108635514974595,
      "num_tokens": 15288674.0,
      "step": 8810
    },
    {
      "entropy": 5.601500415802002,
      "epoch": 0.6858587823380665,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000495843757674673,
      "loss": 5.4098,
      "mean_token_accuracy": 0.16587206423282624,
      "num_tokens": 15296881.0,
      "step": 8815
    },
    {
      "entropy": 5.649654579162598,
      "epoch": 0.6862478117097841,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004958384534665945,
      "loss": 5.4535,
      "mean_token_accuracy": 0.16490767002105713,
      "num_tokens": 15305379.0,
      "step": 8820
    },
    {
      "entropy": 5.619242906570435,
      "epoch": 0.6866368410815017,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004958331459076438,
      "loss": 5.5538,
      "mean_token_accuracy": 0.15394526571035386,
      "num_tokens": 15314572.0,
      "step": 8825
    },
    {
      "entropy": 5.672193336486816,
      "epoch": 0.6870258704532192,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004958278349979018,
      "loss": 5.5112,
      "mean_token_accuracy": 0.16274864226579666,
      "num_tokens": 15323407.0,
      "step": 8830
    },
    {
      "entropy": 5.62689995765686,
      "epoch": 0.6874148998249368,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004958225207374488,
      "loss": 5.3833,
      "mean_token_accuracy": 0.16141065061092377,
      "num_tokens": 15332208.0,
      "step": 8835
    },
    {
      "entropy": 5.573613595962525,
      "epoch": 0.6878039291966543,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004958172031263655,
      "loss": 5.4373,
      "mean_token_accuracy": 0.1640392690896988,
      "num_tokens": 15340972.0,
      "step": 8840
    },
    {
      "entropy": 5.577694511413574,
      "epoch": 0.6881929585683719,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004958118821647326,
      "loss": 5.4133,
      "mean_token_accuracy": 0.16720088571310043,
      "num_tokens": 15349991.0,
      "step": 8845
    },
    {
      "entropy": 5.543353700637818,
      "epoch": 0.6885819879400895,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004958065578526308,
      "loss": 5.3598,
      "mean_token_accuracy": 0.1780972182750702,
      "num_tokens": 15358194.0,
      "step": 8850
    },
    {
      "entropy": 5.606345891952515,
      "epoch": 0.688971017311807,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000495801230190141,
      "loss": 5.3983,
      "mean_token_accuracy": 0.16778187900781633,
      "num_tokens": 15366954.0,
      "step": 8855
    },
    {
      "entropy": 5.647829484939575,
      "epoch": 0.6893600466835246,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004957958991773441,
      "loss": 5.5775,
      "mean_token_accuracy": 0.15903889238834382,
      "num_tokens": 15376107.0,
      "step": 8860
    },
    {
      "entropy": 5.557740306854248,
      "epoch": 0.6897490760552422,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004957905648143206,
      "loss": 5.4612,
      "mean_token_accuracy": 0.1630828395485878,
      "num_tokens": 15384517.0,
      "step": 8865
    },
    {
      "entropy": 5.712189245223999,
      "epoch": 0.6901381054269597,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004957852271011519,
      "loss": 5.6197,
      "mean_token_accuracy": 0.1566236935555935,
      "num_tokens": 15393544.0,
      "step": 8870
    },
    {
      "entropy": 5.634437084197998,
      "epoch": 0.6905271347986773,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004957798860379187,
      "loss": 5.4424,
      "mean_token_accuracy": 0.16527574509382248,
      "num_tokens": 15402165.0,
      "step": 8875
    },
    {
      "entropy": 5.573216724395752,
      "epoch": 0.6909161641703949,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004957745416247022,
      "loss": 5.4805,
      "mean_token_accuracy": 0.16135098785161972,
      "num_tokens": 15411430.0,
      "step": 8880
    },
    {
      "entropy": 5.56478705406189,
      "epoch": 0.6913051935421124,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004957691938615833,
      "loss": 5.3799,
      "mean_token_accuracy": 0.16975612193346024,
      "num_tokens": 15420346.0,
      "step": 8885
    },
    {
      "entropy": 5.472126531600952,
      "epoch": 0.69169422291383,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004957638427486434,
      "loss": 5.3097,
      "mean_token_accuracy": 0.1705828309059143,
      "num_tokens": 15428747.0,
      "step": 8890
    },
    {
      "entropy": 5.444356060028076,
      "epoch": 0.6920832522855476,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004957584882859636,
      "loss": 5.3225,
      "mean_token_accuracy": 0.17348022162914276,
      "num_tokens": 15437110.0,
      "step": 8895
    },
    {
      "entropy": 5.613779497146607,
      "epoch": 0.6924722816572652,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004957531304736251,
      "loss": 5.4421,
      "mean_token_accuracy": 0.16858215779066085,
      "num_tokens": 15446524.0,
      "step": 8900
    },
    {
      "entropy": 5.54238862991333,
      "epoch": 0.6928613110289827,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004957477693117091,
      "loss": 5.4762,
      "mean_token_accuracy": 0.1619431734085083,
      "num_tokens": 15455372.0,
      "step": 8905
    },
    {
      "entropy": 5.581226015090943,
      "epoch": 0.6932503404007002,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004957424048002971,
      "loss": 5.3752,
      "mean_token_accuracy": 0.16786291897296907,
      "num_tokens": 15463060.0,
      "step": 8910
    },
    {
      "entropy": 5.557521724700928,
      "epoch": 0.6936393697724178,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004957370369394706,
      "loss": 5.4993,
      "mean_token_accuracy": 0.16294360011816025,
      "num_tokens": 15471836.0,
      "step": 8915
    },
    {
      "entropy": 5.6872063159942625,
      "epoch": 0.6940283991441354,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004957316657293107,
      "loss": 5.4943,
      "mean_token_accuracy": 0.16140419095754624,
      "num_tokens": 15479819.0,
      "step": 8920
    },
    {
      "entropy": 5.613636875152588,
      "epoch": 0.694417428515853,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004957262911698992,
      "loss": 5.4975,
      "mean_token_accuracy": 0.1646125003695488,
      "num_tokens": 15488562.0,
      "step": 8925
    },
    {
      "entropy": 5.629144191741943,
      "epoch": 0.6948064578875706,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004957209132613175,
      "loss": 5.5586,
      "mean_token_accuracy": 0.1573345459997654,
      "num_tokens": 15497822.0,
      "step": 8930
    },
    {
      "entropy": 5.6370995998382565,
      "epoch": 0.695195487259288,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004957155320036473,
      "loss": 5.5316,
      "mean_token_accuracy": 0.15563015937805175,
      "num_tokens": 15506913.0,
      "step": 8935
    },
    {
      "entropy": 5.677511978149414,
      "epoch": 0.6955845166310056,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004957101473969702,
      "loss": 5.4452,
      "mean_token_accuracy": 0.1658278450369835,
      "num_tokens": 15516053.0,
      "step": 8940
    },
    {
      "entropy": 5.663633251190186,
      "epoch": 0.6959735460027232,
      "grad_norm": 1.328125,
      "learning_rate": 0.000495704759441368,
      "loss": 5.4805,
      "mean_token_accuracy": 0.15481254011392592,
      "num_tokens": 15524174.0,
      "step": 8945
    },
    {
      "entropy": 5.602047300338745,
      "epoch": 0.6963625753744408,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004956993681369221,
      "loss": 5.4151,
      "mean_token_accuracy": 0.1624005079269409,
      "num_tokens": 15533652.0,
      "step": 8950
    },
    {
      "entropy": 5.628164386749267,
      "epoch": 0.6967516047461584,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004956939734837148,
      "loss": 5.429,
      "mean_token_accuracy": 0.16301255375146867,
      "num_tokens": 15541644.0,
      "step": 8955
    },
    {
      "entropy": 5.50003662109375,
      "epoch": 0.6971406341178759,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004956885754818277,
      "loss": 5.4077,
      "mean_token_accuracy": 0.17427603900432587,
      "num_tokens": 15550298.0,
      "step": 8960
    },
    {
      "entropy": 5.595301818847656,
      "epoch": 0.6975296634895934,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004956831741313427,
      "loss": 5.4742,
      "mean_token_accuracy": 0.16779751107096672,
      "num_tokens": 15559388.0,
      "step": 8965
    },
    {
      "entropy": 5.485478496551513,
      "epoch": 0.697918692861311,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004956777694323418,
      "loss": 5.296,
      "mean_token_accuracy": 0.16711134016513823,
      "num_tokens": 15567543.0,
      "step": 8970
    },
    {
      "entropy": 5.525498628616333,
      "epoch": 0.6983077222330286,
      "grad_norm": 1.375,
      "learning_rate": 0.000495672361384907,
      "loss": 5.4018,
      "mean_token_accuracy": 0.1763388156890869,
      "num_tokens": 15575365.0,
      "step": 8975
    },
    {
      "entropy": 5.635193872451782,
      "epoch": 0.6986967516047462,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004956669499891202,
      "loss": 5.4896,
      "mean_token_accuracy": 0.16018304973840714,
      "num_tokens": 15584295.0,
      "step": 8980
    },
    {
      "entropy": 5.623423337936401,
      "epoch": 0.6990857809764637,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004956615352450639,
      "loss": 5.4715,
      "mean_token_accuracy": 0.17425471991300584,
      "num_tokens": 15592906.0,
      "step": 8985
    },
    {
      "entropy": 5.521390438079834,
      "epoch": 0.6994748103481813,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004956561171528198,
      "loss": 5.4362,
      "mean_token_accuracy": 0.16043994277715684,
      "num_tokens": 15601938.0,
      "step": 8990
    },
    {
      "entropy": 5.547690296173096,
      "epoch": 0.6998638397198989,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004956506957124704,
      "loss": 5.425,
      "mean_token_accuracy": 0.16874610781669616,
      "num_tokens": 15610173.0,
      "step": 8995
    },
    {
      "entropy": 5.641872930526733,
      "epoch": 0.7002528690916164,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000495645270924098,
      "loss": 5.511,
      "mean_token_accuracy": 0.16251857429742814,
      "num_tokens": 15618896.0,
      "step": 9000
    },
    {
      "epoch": 0.7002528690916164,
      "eval_entropy": 5.534160187292604,
      "eval_loss": 5.511755466461182,
      "eval_mean_token_accuracy": 0.16929327884368076,
      "eval_num_tokens": 15618896.0,
      "eval_runtime": 28.9098,
      "eval_samples_per_second": 1437.505,
      "eval_steps_per_second": 179.697,
      "step": 9000
    },
    {
      "entropy": 5.608629512786865,
      "epoch": 0.700641898463334,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004956398427877847,
      "loss": 5.4293,
      "mean_token_accuracy": 0.1648696929216385,
      "num_tokens": 15627890.0,
      "step": 9005
    },
    {
      "entropy": 5.5562005043029785,
      "epoch": 0.7010309278350515,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000495634411303613,
      "loss": 5.3593,
      "mean_token_accuracy": 0.1650518074631691,
      "num_tokens": 15636290.0,
      "step": 9010
    },
    {
      "entropy": 5.5294958591461185,
      "epoch": 0.7014199572067691,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004956289764716654,
      "loss": 5.4192,
      "mean_token_accuracy": 0.15961053222417831,
      "num_tokens": 15644777.0,
      "step": 9015
    },
    {
      "entropy": 5.5333515167236325,
      "epoch": 0.7018089865784867,
      "grad_norm": 1.375,
      "learning_rate": 0.0004956235382920241,
      "loss": 5.4505,
      "mean_token_accuracy": 0.16453654691576958,
      "num_tokens": 15653333.0,
      "step": 9020
    },
    {
      "entropy": 5.648467493057251,
      "epoch": 0.7021980159502043,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004956180967647717,
      "loss": 5.438,
      "mean_token_accuracy": 0.16483430713415145,
      "num_tokens": 15662140.0,
      "step": 9025
    },
    {
      "entropy": 5.603916835784912,
      "epoch": 0.7025870453219218,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004956126518899911,
      "loss": 5.53,
      "mean_token_accuracy": 0.16260873079299926,
      "num_tokens": 15671387.0,
      "step": 9030
    },
    {
      "entropy": 5.622740077972412,
      "epoch": 0.7029760746936393,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004956072036677644,
      "loss": 5.4651,
      "mean_token_accuracy": 0.16124595403671266,
      "num_tokens": 15680884.0,
      "step": 9035
    },
    {
      "entropy": 5.646053075790405,
      "epoch": 0.7033651040653569,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004956017520981746,
      "loss": 5.426,
      "mean_token_accuracy": 0.1608300492167473,
      "num_tokens": 15690030.0,
      "step": 9040
    },
    {
      "entropy": 5.554988050460816,
      "epoch": 0.7037541334370745,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004955962971813044,
      "loss": 5.4392,
      "mean_token_accuracy": 0.16510536670684814,
      "num_tokens": 15698260.0,
      "step": 9045
    },
    {
      "entropy": 5.4959269046783445,
      "epoch": 0.7041431628087921,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004955908389172364,
      "loss": 5.3346,
      "mean_token_accuracy": 0.17143793851137162,
      "num_tokens": 15706275.0,
      "step": 9050
    },
    {
      "entropy": 5.630095195770264,
      "epoch": 0.7045321921805097,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004955853773060536,
      "loss": 5.5586,
      "mean_token_accuracy": 0.1675524741411209,
      "num_tokens": 15715086.0,
      "step": 9055
    },
    {
      "entropy": 5.570626163482666,
      "epoch": 0.7049212215522271,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004955799123478388,
      "loss": 5.3808,
      "mean_token_accuracy": 0.17316911965608597,
      "num_tokens": 15723931.0,
      "step": 9060
    },
    {
      "entropy": 5.554439878463745,
      "epoch": 0.7053102509239447,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004955744440426748,
      "loss": 5.515,
      "mean_token_accuracy": 0.1656896576285362,
      "num_tokens": 15732905.0,
      "step": 9065
    },
    {
      "entropy": 5.6108321189880375,
      "epoch": 0.7056992802956623,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004955689723906448,
      "loss": 5.415,
      "mean_token_accuracy": 0.1599321410059929,
      "num_tokens": 15742355.0,
      "step": 9070
    },
    {
      "entropy": 5.723128795623779,
      "epoch": 0.7060883096673799,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004955634973918318,
      "loss": 5.4938,
      "mean_token_accuracy": 0.15792502462863922,
      "num_tokens": 15750924.0,
      "step": 9075
    },
    {
      "entropy": 5.578666114807129,
      "epoch": 0.7064773390390975,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004955580190463186,
      "loss": 5.4256,
      "mean_token_accuracy": 0.16827226132154466,
      "num_tokens": 15759677.0,
      "step": 9080
    },
    {
      "entropy": 5.566485786437989,
      "epoch": 0.7068663684108151,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004955525373541886,
      "loss": 5.4433,
      "mean_token_accuracy": 0.1610451325774193,
      "num_tokens": 15767875.0,
      "step": 9085
    },
    {
      "entropy": 5.535344076156616,
      "epoch": 0.7072553977825325,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004955470523155249,
      "loss": 5.3967,
      "mean_token_accuracy": 0.16854775696992874,
      "num_tokens": 15776194.0,
      "step": 9090
    },
    {
      "entropy": 5.586416625976563,
      "epoch": 0.7076444271542501,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004955415639304107,
      "loss": 5.4711,
      "mean_token_accuracy": 0.16787676513195038,
      "num_tokens": 15784774.0,
      "step": 9095
    },
    {
      "entropy": 5.603423166275024,
      "epoch": 0.7080334565259677,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004955360721989292,
      "loss": 5.3629,
      "mean_token_accuracy": 0.17353993505239487,
      "num_tokens": 15793121.0,
      "step": 9100
    },
    {
      "entropy": 5.706853008270263,
      "epoch": 0.7084224858976853,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004955305771211641,
      "loss": 5.577,
      "mean_token_accuracy": 0.15844244360923768,
      "num_tokens": 15801518.0,
      "step": 9105
    },
    {
      "entropy": 5.571177816390991,
      "epoch": 0.7088115152694029,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004955250786971982,
      "loss": 5.3426,
      "mean_token_accuracy": 0.1712365299463272,
      "num_tokens": 15809602.0,
      "step": 9110
    },
    {
      "entropy": 5.55650634765625,
      "epoch": 0.7092005446411204,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004955195769271154,
      "loss": 5.5286,
      "mean_token_accuracy": 0.16069710552692412,
      "num_tokens": 15818529.0,
      "step": 9115
    },
    {
      "entropy": 5.669844913482666,
      "epoch": 0.709589574012838,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004955140718109991,
      "loss": 5.415,
      "mean_token_accuracy": 0.16865902692079543,
      "num_tokens": 15827176.0,
      "step": 9120
    },
    {
      "entropy": 5.631535291671753,
      "epoch": 0.7099786033845555,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004955085633489326,
      "loss": 5.4788,
      "mean_token_accuracy": 0.16147402077913284,
      "num_tokens": 15837095.0,
      "step": 9125
    },
    {
      "entropy": 5.5327576160430905,
      "epoch": 0.7103676327562731,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004955030515409997,
      "loss": 5.462,
      "mean_token_accuracy": 0.16384090781211852,
      "num_tokens": 15845285.0,
      "step": 9130
    },
    {
      "entropy": 5.558310651779175,
      "epoch": 0.7107566621279907,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004954975363872838,
      "loss": 5.4702,
      "mean_token_accuracy": 0.172089821100235,
      "num_tokens": 15853468.0,
      "step": 9135
    },
    {
      "entropy": 5.588555479049683,
      "epoch": 0.7111456914997082,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004954920178878689,
      "loss": 5.4139,
      "mean_token_accuracy": 0.1720640629529953,
      "num_tokens": 15861699.0,
      "step": 9140
    },
    {
      "entropy": 5.6258656024932865,
      "epoch": 0.7115347208714258,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004954864960428385,
      "loss": 5.4998,
      "mean_token_accuracy": 0.16113060861825942,
      "num_tokens": 15870401.0,
      "step": 9145
    },
    {
      "entropy": 5.5766744136810305,
      "epoch": 0.7119237502431434,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004954809708522765,
      "loss": 5.2856,
      "mean_token_accuracy": 0.1793888255953789,
      "num_tokens": 15878214.0,
      "step": 9150
    },
    {
      "entropy": 5.540178632736206,
      "epoch": 0.712312779614861,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004954754423162667,
      "loss": 5.4843,
      "mean_token_accuracy": 0.1615191400051117,
      "num_tokens": 15886581.0,
      "step": 9155
    },
    {
      "entropy": 5.634097528457642,
      "epoch": 0.7127018089865785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000495469910434893,
      "loss": 5.5868,
      "mean_token_accuracy": 0.16165244430303574,
      "num_tokens": 15895748.0,
      "step": 9160
    },
    {
      "entropy": 5.677596998214722,
      "epoch": 0.713090838358296,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004954643752082392,
      "loss": 5.4521,
      "mean_token_accuracy": 0.1625576674938202,
      "num_tokens": 15905030.0,
      "step": 9165
    },
    {
      "entropy": 5.531970453262329,
      "epoch": 0.7134798677300136,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004954588366363896,
      "loss": 5.4289,
      "mean_token_accuracy": 0.16435423344373704,
      "num_tokens": 15913430.0,
      "step": 9170
    },
    {
      "entropy": 5.5162606716156,
      "epoch": 0.7138688971017312,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004954532947194279,
      "loss": 5.3503,
      "mean_token_accuracy": 0.1692097082734108,
      "num_tokens": 15921432.0,
      "step": 9175
    },
    {
      "entropy": 5.644609880447388,
      "epoch": 0.7142579264734488,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004954477494574384,
      "loss": 5.4099,
      "mean_token_accuracy": 0.16645418405532836,
      "num_tokens": 15930159.0,
      "step": 9180
    },
    {
      "entropy": 5.5702197551727295,
      "epoch": 0.7146469558451664,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004954422008505052,
      "loss": 5.4831,
      "mean_token_accuracy": 0.15969833582639695,
      "num_tokens": 15939911.0,
      "step": 9185
    },
    {
      "entropy": 5.5767840385437015,
      "epoch": 0.7150359852168838,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004954366488987125,
      "loss": 5.3969,
      "mean_token_accuracy": 0.16617162227630616,
      "num_tokens": 15948586.0,
      "step": 9190
    },
    {
      "entropy": 5.451484060287475,
      "epoch": 0.7154250145886014,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004954310936021444,
      "loss": 5.2854,
      "mean_token_accuracy": 0.1679601788520813,
      "num_tokens": 15956818.0,
      "step": 9195
    },
    {
      "entropy": 5.558848333358765,
      "epoch": 0.715814043960319,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004954255349608853,
      "loss": 5.4897,
      "mean_token_accuracy": 0.15599195212125777,
      "num_tokens": 15966027.0,
      "step": 9200
    },
    {
      "entropy": 5.673420381546021,
      "epoch": 0.7162030733320366,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004954199729750198,
      "loss": 5.5386,
      "mean_token_accuracy": 0.16199355125427245,
      "num_tokens": 15975186.0,
      "step": 9205
    },
    {
      "entropy": 5.568282270431519,
      "epoch": 0.7165921027037542,
      "grad_norm": 1.375,
      "learning_rate": 0.0004954144076446318,
      "loss": 5.3876,
      "mean_token_accuracy": 0.1700726181268692,
      "num_tokens": 15983924.0,
      "step": 9210
    },
    {
      "entropy": 5.585992002487183,
      "epoch": 0.7169811320754716,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004954088389698061,
      "loss": 5.4224,
      "mean_token_accuracy": 0.17278845161199569,
      "num_tokens": 15992906.0,
      "step": 9215
    },
    {
      "entropy": 5.543997955322266,
      "epoch": 0.7173701614471892,
      "grad_norm": 1.234375,
      "learning_rate": 0.000495403266950627,
      "loss": 5.4401,
      "mean_token_accuracy": 0.1640487238764763,
      "num_tokens": 16001378.0,
      "step": 9220
    },
    {
      "entropy": 5.569652462005616,
      "epoch": 0.7177591908189068,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004953976915871792,
      "loss": 5.4366,
      "mean_token_accuracy": 0.16413379907608033,
      "num_tokens": 16010704.0,
      "step": 9225
    },
    {
      "entropy": 5.492253684997559,
      "epoch": 0.7181482201906244,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004953921128795472,
      "loss": 5.4169,
      "mean_token_accuracy": 0.16802991628646852,
      "num_tokens": 16018936.0,
      "step": 9230
    },
    {
      "entropy": 5.583294677734375,
      "epoch": 0.718537249562342,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004953865308278156,
      "loss": 5.3302,
      "mean_token_accuracy": 0.169882196187973,
      "num_tokens": 16027295.0,
      "step": 9235
    },
    {
      "entropy": 5.529065084457398,
      "epoch": 0.7189262789340595,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004953809454320693,
      "loss": 5.3577,
      "mean_token_accuracy": 0.17003903687000274,
      "num_tokens": 16036493.0,
      "step": 9240
    },
    {
      "entropy": 5.538720512390137,
      "epoch": 0.7193153083057771,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004953753566923929,
      "loss": 5.471,
      "mean_token_accuracy": 0.16651601940393448,
      "num_tokens": 16045406.0,
      "step": 9245
    },
    {
      "entropy": 5.590838146209717,
      "epoch": 0.7197043376774946,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004953697646088712,
      "loss": 5.4458,
      "mean_token_accuracy": 0.16936541199684144,
      "num_tokens": 16054126.0,
      "step": 9250
    },
    {
      "entropy": 5.5696985721588135,
      "epoch": 0.7200933670492122,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004953641691815891,
      "loss": 5.4696,
      "mean_token_accuracy": 0.16234371364116668,
      "num_tokens": 16063921.0,
      "step": 9255
    },
    {
      "entropy": 5.569355773925781,
      "epoch": 0.7204823964209298,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004953585704106315,
      "loss": 5.4766,
      "mean_token_accuracy": 0.16223076432943345,
      "num_tokens": 16072160.0,
      "step": 9260
    },
    {
      "entropy": 5.5977160930633545,
      "epoch": 0.7208714257926473,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004953529682960832,
      "loss": 5.4117,
      "mean_token_accuracy": 0.16515904515981675,
      "num_tokens": 16080794.0,
      "step": 9265
    },
    {
      "entropy": 5.562137603759766,
      "epoch": 0.7212604551643649,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004953473628380295,
      "loss": 5.4326,
      "mean_token_accuracy": 0.1696798771619797,
      "num_tokens": 16089615.0,
      "step": 9270
    },
    {
      "entropy": 5.515734767913818,
      "epoch": 0.7216494845360825,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004953417540365553,
      "loss": 5.3497,
      "mean_token_accuracy": 0.16612076610326768,
      "num_tokens": 16098363.0,
      "step": 9275
    },
    {
      "entropy": 5.58066029548645,
      "epoch": 0.7220385139078,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004953361418917455,
      "loss": 5.4472,
      "mean_token_accuracy": 0.1622335880994797,
      "num_tokens": 16106837.0,
      "step": 9280
    },
    {
      "entropy": 5.591270780563354,
      "epoch": 0.7224275432795176,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004953305264036856,
      "loss": 5.3925,
      "mean_token_accuracy": 0.1681852363049984,
      "num_tokens": 16114923.0,
      "step": 9285
    },
    {
      "entropy": 5.568781089782715,
      "epoch": 0.7228165726512352,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004953249075724607,
      "loss": 5.4375,
      "mean_token_accuracy": 0.16472795456647873,
      "num_tokens": 16123698.0,
      "step": 9290
    },
    {
      "entropy": 5.610735368728638,
      "epoch": 0.7232056020229527,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004953192853981559,
      "loss": 5.4231,
      "mean_token_accuracy": 0.16371380388736725,
      "num_tokens": 16132361.0,
      "step": 9295
    },
    {
      "entropy": 5.654996013641357,
      "epoch": 0.7235946313946703,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004953136598808566,
      "loss": 5.4718,
      "mean_token_accuracy": 0.1594822496175766,
      "num_tokens": 16141554.0,
      "step": 9300
    },
    {
      "entropy": 5.595190000534058,
      "epoch": 0.7239836607663879,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004953080310206481,
      "loss": 5.4015,
      "mean_token_accuracy": 0.167132468521595,
      "num_tokens": 16150206.0,
      "step": 9305
    },
    {
      "entropy": 5.581953763961792,
      "epoch": 0.7243726901381055,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004953023988176162,
      "loss": 5.5345,
      "mean_token_accuracy": 0.15518438220024108,
      "num_tokens": 16158995.0,
      "step": 9310
    },
    {
      "entropy": 5.595383644104004,
      "epoch": 0.724761719509823,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004952967632718458,
      "loss": 5.509,
      "mean_token_accuracy": 0.1578101173043251,
      "num_tokens": 16167809.0,
      "step": 9315
    },
    {
      "entropy": 5.7937929153442385,
      "epoch": 0.7251507488815405,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004952911243834227,
      "loss": 5.5808,
      "mean_token_accuracy": 0.15702876150608064,
      "num_tokens": 16176257.0,
      "step": 9320
    },
    {
      "entropy": 5.65752420425415,
      "epoch": 0.7255397782532581,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004952854821524324,
      "loss": 5.5308,
      "mean_token_accuracy": 0.15480078756809235,
      "num_tokens": 16185597.0,
      "step": 9325
    },
    {
      "entropy": 5.618897438049316,
      "epoch": 0.7259288076249757,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004952798365789606,
      "loss": 5.5384,
      "mean_token_accuracy": 0.16022090762853622,
      "num_tokens": 16194591.0,
      "step": 9330
    },
    {
      "entropy": 5.6969071388244625,
      "epoch": 0.7263178369966933,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004952741876630928,
      "loss": 5.5123,
      "mean_token_accuracy": 0.15586536675691604,
      "num_tokens": 16203073.0,
      "step": 9335
    },
    {
      "entropy": 5.59380989074707,
      "epoch": 0.7267068663684109,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004952685354049148,
      "loss": 5.479,
      "mean_token_accuracy": 0.16461253315210342,
      "num_tokens": 16211208.0,
      "step": 9340
    },
    {
      "entropy": 5.624042558670044,
      "epoch": 0.7270958957401283,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004952628798045124,
      "loss": 5.4238,
      "mean_token_accuracy": 0.17028266191482544,
      "num_tokens": 16219687.0,
      "step": 9345
    },
    {
      "entropy": 5.629638671875,
      "epoch": 0.7274849251118459,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004952572208619714,
      "loss": 5.5017,
      "mean_token_accuracy": 0.1649695172905922,
      "num_tokens": 16228214.0,
      "step": 9350
    },
    {
      "entropy": 5.641639661788941,
      "epoch": 0.7278739544835635,
      "grad_norm": 1.25,
      "learning_rate": 0.0004952515585773778,
      "loss": 5.5969,
      "mean_token_accuracy": 0.15739991515874863,
      "num_tokens": 16237765.0,
      "step": 9355
    },
    {
      "entropy": 5.625778388977051,
      "epoch": 0.7282629838552811,
      "grad_norm": 1.25,
      "learning_rate": 0.0004952458929508171,
      "loss": 5.4581,
      "mean_token_accuracy": 0.1752306789159775,
      "num_tokens": 16245339.0,
      "step": 9360
    },
    {
      "entropy": 5.559713840484619,
      "epoch": 0.7286520132269987,
      "grad_norm": 1.25,
      "learning_rate": 0.0004952402239823757,
      "loss": 5.4479,
      "mean_token_accuracy": 0.1610769271850586,
      "num_tokens": 16253966.0,
      "step": 9365
    },
    {
      "entropy": 5.56492338180542,
      "epoch": 0.7290410425987162,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004952345516721393,
      "loss": 5.4214,
      "mean_token_accuracy": 0.16634339839220047,
      "num_tokens": 16262859.0,
      "step": 9370
    },
    {
      "entropy": 5.546257066726684,
      "epoch": 0.7294300719704337,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004952288760201942,
      "loss": 5.4652,
      "mean_token_accuracy": 0.16544958502054213,
      "num_tokens": 16271646.0,
      "step": 9375
    },
    {
      "entropy": 5.5463837623596195,
      "epoch": 0.7298191013421513,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004952231970266265,
      "loss": 5.4562,
      "mean_token_accuracy": 0.16257821321487426,
      "num_tokens": 16281058.0,
      "step": 9380
    },
    {
      "entropy": 5.661608028411865,
      "epoch": 0.7302081307138689,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004952175146915223,
      "loss": 5.4108,
      "mean_token_accuracy": 0.16807551681995392,
      "num_tokens": 16289039.0,
      "step": 9385
    },
    {
      "entropy": 5.426449251174927,
      "epoch": 0.7305971600855865,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004952118290149679,
      "loss": 5.3141,
      "mean_token_accuracy": 0.17522288411855697,
      "num_tokens": 16297373.0,
      "step": 9390
    },
    {
      "entropy": 5.534648466110229,
      "epoch": 0.730986189457304,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004952061399970492,
      "loss": 5.4643,
      "mean_token_accuracy": 0.16045526564121246,
      "num_tokens": 16306066.0,
      "step": 9395
    },
    {
      "entropy": 5.635262155532837,
      "epoch": 0.7313752188290216,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000495200447637853,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1583344116806984,
      "num_tokens": 16315282.0,
      "step": 9400
    },
    {
      "entropy": 5.742931938171386,
      "epoch": 0.7317642482007392,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004951947519374655,
      "loss": 5.6039,
      "mean_token_accuracy": 0.15938021540641784,
      "num_tokens": 16324830.0,
      "step": 9405
    },
    {
      "entropy": 5.657621669769287,
      "epoch": 0.7321532775724567,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004951890528959731,
      "loss": 5.4509,
      "mean_token_accuracy": 0.15992313176393508,
      "num_tokens": 16332990.0,
      "step": 9410
    },
    {
      "entropy": 5.591779947280884,
      "epoch": 0.7325423069441743,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004951833505134623,
      "loss": 5.4395,
      "mean_token_accuracy": 0.1598487064242363,
      "num_tokens": 16342115.0,
      "step": 9415
    },
    {
      "entropy": 5.601643466949463,
      "epoch": 0.7329313363158918,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004951776447900196,
      "loss": 5.4255,
      "mean_token_accuracy": 0.16582438498735427,
      "num_tokens": 16350329.0,
      "step": 9420
    },
    {
      "entropy": 5.6025560855865475,
      "epoch": 0.7333203656876094,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004951719357257317,
      "loss": 5.5261,
      "mean_token_accuracy": 0.16438100039958953,
      "num_tokens": 16358890.0,
      "step": 9425
    },
    {
      "entropy": 5.728090858459472,
      "epoch": 0.733709395059327,
      "grad_norm": 1.328125,
      "learning_rate": 0.000495166223320685,
      "loss": 5.5709,
      "mean_token_accuracy": 0.15645813196897507,
      "num_tokens": 16367338.0,
      "step": 9430
    },
    {
      "entropy": 5.677723789215088,
      "epoch": 0.7340984244310446,
      "grad_norm": 1.375,
      "learning_rate": 0.0004951605075749662,
      "loss": 5.4721,
      "mean_token_accuracy": 0.171079383790493,
      "num_tokens": 16375472.0,
      "step": 9435
    },
    {
      "entropy": 5.614687585830689,
      "epoch": 0.7344874538027621,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004951547884886623,
      "loss": 5.4712,
      "mean_token_accuracy": 0.16885554641485215,
      "num_tokens": 16384212.0,
      "step": 9440
    },
    {
      "entropy": 5.567186164855957,
      "epoch": 0.7348764831744796,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004951490660618598,
      "loss": 5.356,
      "mean_token_accuracy": 0.16907453089952468,
      "num_tokens": 16393447.0,
      "step": 9445
    },
    {
      "entropy": 5.5535173416137695,
      "epoch": 0.7352655125461972,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004951433402946457,
      "loss": 5.3585,
      "mean_token_accuracy": 0.16890464276075362,
      "num_tokens": 16401876.0,
      "step": 9450
    },
    {
      "entropy": 5.56154236793518,
      "epoch": 0.7356545419179148,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004951376111871068,
      "loss": 5.4829,
      "mean_token_accuracy": 0.1697598546743393,
      "num_tokens": 16411020.0,
      "step": 9455
    },
    {
      "entropy": 5.617890548706055,
      "epoch": 0.7360435712896324,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004951318787393299,
      "loss": 5.5083,
      "mean_token_accuracy": 0.16018458753824233,
      "num_tokens": 16420492.0,
      "step": 9460
    },
    {
      "entropy": 5.592196750640869,
      "epoch": 0.73643260066135,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004951261429514023,
      "loss": 5.4067,
      "mean_token_accuracy": 0.17297099828720092,
      "num_tokens": 16429277.0,
      "step": 9465
    },
    {
      "entropy": 5.578293752670288,
      "epoch": 0.7368216300330674,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004951204038234106,
      "loss": 5.4449,
      "mean_token_accuracy": 0.16348470598459244,
      "num_tokens": 16438008.0,
      "step": 9470
    },
    {
      "entropy": 5.596996450424195,
      "epoch": 0.737210659404785,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004951146613554424,
      "loss": 5.4572,
      "mean_token_accuracy": 0.16455768197774887,
      "num_tokens": 16446791.0,
      "step": 9475
    },
    {
      "entropy": 5.587205171585083,
      "epoch": 0.7375996887765026,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004951089155475843,
      "loss": 5.4825,
      "mean_token_accuracy": 0.1651042640209198,
      "num_tokens": 16455277.0,
      "step": 9480
    },
    {
      "entropy": 5.527266168594361,
      "epoch": 0.7379887181482202,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000495103166399924,
      "loss": 5.3946,
      "mean_token_accuracy": 0.1652072325348854,
      "num_tokens": 16463310.0,
      "step": 9485
    },
    {
      "entropy": 5.5629857063293455,
      "epoch": 0.7383777475199378,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004950974139125484,
      "loss": 5.4617,
      "mean_token_accuracy": 0.15703424960374832,
      "num_tokens": 16472421.0,
      "step": 9490
    },
    {
      "entropy": 5.631540012359619,
      "epoch": 0.7387667768916554,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004950916580855448,
      "loss": 5.5045,
      "mean_token_accuracy": 0.15918551087379457,
      "num_tokens": 16481814.0,
      "step": 9495
    },
    {
      "entropy": 5.6208983898162845,
      "epoch": 0.7391558062633728,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004950858989190007,
      "loss": 5.4831,
      "mean_token_accuracy": 0.1588638424873352,
      "num_tokens": 16490902.0,
      "step": 9500
    },
    {
      "entropy": 5.612247133255005,
      "epoch": 0.7395448356350904,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004950801364130032,
      "loss": 5.4284,
      "mean_token_accuracy": 0.16425824612379075,
      "num_tokens": 16499418.0,
      "step": 9505
    },
    {
      "entropy": 5.474906730651855,
      "epoch": 0.739933865006808,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004950743705676401,
      "loss": 5.3707,
      "mean_token_accuracy": 0.17225120663642884,
      "num_tokens": 16508578.0,
      "step": 9510
    },
    {
      "entropy": 5.533582878112793,
      "epoch": 0.7403228943785256,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004950686013829987,
      "loss": 5.438,
      "mean_token_accuracy": 0.16768757849931717,
      "num_tokens": 16517356.0,
      "step": 9515
    },
    {
      "entropy": 5.572383165359497,
      "epoch": 0.7407119237502432,
      "grad_norm": 1.375,
      "learning_rate": 0.0004950628288591665,
      "loss": 5.4223,
      "mean_token_accuracy": 0.165138640999794,
      "num_tokens": 16525834.0,
      "step": 9520
    },
    {
      "entropy": 5.594256162643433,
      "epoch": 0.7411009531219607,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004950570529962311,
      "loss": 5.4415,
      "mean_token_accuracy": 0.16531625092029573,
      "num_tokens": 16534513.0,
      "step": 9525
    },
    {
      "entropy": 5.575924730300903,
      "epoch": 0.7414899824936783,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004950512737942803,
      "loss": 5.4145,
      "mean_token_accuracy": 0.16863541603088378,
      "num_tokens": 16543326.0,
      "step": 9530
    },
    {
      "entropy": 5.534553575515747,
      "epoch": 0.7418790118653958,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004950454912534015,
      "loss": 5.4328,
      "mean_token_accuracy": 0.1633179560303688,
      "num_tokens": 16551589.0,
      "step": 9535
    },
    {
      "entropy": 5.612389421463012,
      "epoch": 0.7422680412371134,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004950397053736827,
      "loss": 5.4104,
      "mean_token_accuracy": 0.1686764821410179,
      "num_tokens": 16560261.0,
      "step": 9540
    },
    {
      "entropy": 5.550796937942505,
      "epoch": 0.742657070608831,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004950339161552117,
      "loss": 5.4049,
      "mean_token_accuracy": 0.168380768597126,
      "num_tokens": 16568645.0,
      "step": 9545
    },
    {
      "entropy": 5.526888751983643,
      "epoch": 0.7430460999805485,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004950281235980761,
      "loss": 5.4009,
      "mean_token_accuracy": 0.17568656653165818,
      "num_tokens": 16577206.0,
      "step": 9550
    },
    {
      "entropy": 5.578280305862426,
      "epoch": 0.7434351293522661,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000495022327702364,
      "loss": 5.4028,
      "mean_token_accuracy": 0.16437870264053345,
      "num_tokens": 16585011.0,
      "step": 9555
    },
    {
      "entropy": 5.589871549606324,
      "epoch": 0.7438241587239837,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004950165284681631,
      "loss": 5.4734,
      "mean_token_accuracy": 0.16122713834047317,
      "num_tokens": 16593574.0,
      "step": 9560
    },
    {
      "entropy": 5.571707105636596,
      "epoch": 0.7442131880957013,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004950107258955618,
      "loss": 5.4025,
      "mean_token_accuracy": 0.16671413630247117,
      "num_tokens": 16603023.0,
      "step": 9565
    },
    {
      "entropy": 5.641346740722656,
      "epoch": 0.7446022174674188,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004950049199846479,
      "loss": 5.4784,
      "mean_token_accuracy": 0.16376918181777,
      "num_tokens": 16611963.0,
      "step": 9570
    },
    {
      "entropy": 5.61685585975647,
      "epoch": 0.7449912468391363,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004949991107355095,
      "loss": 5.4354,
      "mean_token_accuracy": 0.16407208889722824,
      "num_tokens": 16619942.0,
      "step": 9575
    },
    {
      "entropy": 5.444828176498413,
      "epoch": 0.7453802762108539,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004949932981482347,
      "loss": 5.3243,
      "mean_token_accuracy": 0.17423411458730698,
      "num_tokens": 16628261.0,
      "step": 9580
    },
    {
      "entropy": 5.605340480804443,
      "epoch": 0.7457693055825715,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004949874822229118,
      "loss": 5.4558,
      "mean_token_accuracy": 0.16403342336416243,
      "num_tokens": 16637290.0,
      "step": 9585
    },
    {
      "entropy": 5.620204734802246,
      "epoch": 0.7461583349542891,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004949816629596288,
      "loss": 5.4391,
      "mean_token_accuracy": 0.16449353247880935,
      "num_tokens": 16646147.0,
      "step": 9590
    },
    {
      "entropy": 5.55931830406189,
      "epoch": 0.7465473643260067,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004949758403584746,
      "loss": 5.5273,
      "mean_token_accuracy": 0.16253286600112915,
      "num_tokens": 16655299.0,
      "step": 9595
    },
    {
      "entropy": 5.614705753326416,
      "epoch": 0.7469363936977241,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004949700144195368,
      "loss": 5.4198,
      "mean_token_accuracy": 0.16902676373720169,
      "num_tokens": 16664091.0,
      "step": 9600
    },
    {
      "entropy": 5.708738088607788,
      "epoch": 0.7473254230694417,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004949641851429043,
      "loss": 5.474,
      "mean_token_accuracy": 0.16058899760246276,
      "num_tokens": 16672665.0,
      "step": 9605
    },
    {
      "entropy": 5.52224907875061,
      "epoch": 0.7477144524411593,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004949583525286654,
      "loss": 5.4039,
      "mean_token_accuracy": 0.16887062191963195,
      "num_tokens": 16680735.0,
      "step": 9610
    },
    {
      "entropy": 5.579640007019043,
      "epoch": 0.7481034818128769,
      "grad_norm": 1.25,
      "learning_rate": 0.0004949525165769085,
      "loss": 5.4975,
      "mean_token_accuracy": 0.1636406421661377,
      "num_tokens": 16689560.0,
      "step": 9615
    },
    {
      "entropy": 5.630500602722168,
      "epoch": 0.7484925111845945,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004949466772877224,
      "loss": 5.4479,
      "mean_token_accuracy": 0.16279161274433135,
      "num_tokens": 16698953.0,
      "step": 9620
    },
    {
      "entropy": 5.605937719345093,
      "epoch": 0.748881540556312,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004949408346611955,
      "loss": 5.4697,
      "mean_token_accuracy": 0.16185581237077712,
      "num_tokens": 16707601.0,
      "step": 9625
    },
    {
      "entropy": 5.553993177413941,
      "epoch": 0.7492705699280295,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004949349886974165,
      "loss": 5.4711,
      "mean_token_accuracy": 0.16091223955154418,
      "num_tokens": 16716660.0,
      "step": 9630
    },
    {
      "entropy": 5.553178167343139,
      "epoch": 0.7496595992997471,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004949291393964741,
      "loss": 5.4239,
      "mean_token_accuracy": 0.16361938118934632,
      "num_tokens": 16725274.0,
      "step": 9635
    },
    {
      "entropy": 5.593573570251465,
      "epoch": 0.7500486286714647,
      "grad_norm": 1.3125,
      "learning_rate": 0.000494923286758457,
      "loss": 5.4057,
      "mean_token_accuracy": 0.16673256903886796,
      "num_tokens": 16733301.0,
      "step": 9640
    },
    {
      "entropy": 5.627401494979859,
      "epoch": 0.7504376580431823,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004949174307834541,
      "loss": 5.4443,
      "mean_token_accuracy": 0.162208454310894,
      "num_tokens": 16742072.0,
      "step": 9645
    },
    {
      "entropy": 5.5809698581695555,
      "epoch": 0.7508266874148998,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004949115714715543,
      "loss": 5.4489,
      "mean_token_accuracy": 0.16486390084028243,
      "num_tokens": 16751497.0,
      "step": 9650
    },
    {
      "entropy": 5.516124582290649,
      "epoch": 0.7512157167866174,
      "grad_norm": 1.25,
      "learning_rate": 0.0004949057088228465,
      "loss": 5.3993,
      "mean_token_accuracy": 0.17164968252182006,
      "num_tokens": 16760518.0,
      "step": 9655
    },
    {
      "entropy": 5.645664024353027,
      "epoch": 0.751604746158335,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004948998428374194,
      "loss": 5.5067,
      "mean_token_accuracy": 0.15866608619689943,
      "num_tokens": 16768689.0,
      "step": 9660
    },
    {
      "entropy": 5.613323497772217,
      "epoch": 0.7519937755300525,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004948939735153622,
      "loss": 5.3549,
      "mean_token_accuracy": 0.17620459347963333,
      "num_tokens": 16777047.0,
      "step": 9665
    },
    {
      "entropy": 5.4811015129089355,
      "epoch": 0.7523828049017701,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004948881008567641,
      "loss": 5.3239,
      "mean_token_accuracy": 0.17152850031852723,
      "num_tokens": 16785402.0,
      "step": 9670
    },
    {
      "entropy": 5.5329336643219,
      "epoch": 0.7527718342734876,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004948822248617139,
      "loss": 5.4127,
      "mean_token_accuracy": 0.1673654332756996,
      "num_tokens": 16794687.0,
      "step": 9675
    },
    {
      "entropy": 5.669200372695923,
      "epoch": 0.7531608636452052,
      "grad_norm": 1.375,
      "learning_rate": 0.000494876345530301,
      "loss": 5.4657,
      "mean_token_accuracy": 0.16521345525979997,
      "num_tokens": 16803762.0,
      "step": 9680
    },
    {
      "entropy": 5.593966245651245,
      "epoch": 0.7535498930169228,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004948704628626145,
      "loss": 5.4812,
      "mean_token_accuracy": 0.16158633828163146,
      "num_tokens": 16812872.0,
      "step": 9685
    },
    {
      "entropy": 5.6613184928894045,
      "epoch": 0.7539389223886404,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004948645768587437,
      "loss": 5.5214,
      "mean_token_accuracy": 0.16121746301651002,
      "num_tokens": 16821270.0,
      "step": 9690
    },
    {
      "entropy": 5.5620544910430905,
      "epoch": 0.7543279517603579,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004948586875187778,
      "loss": 5.4115,
      "mean_token_accuracy": 0.16581283062696456,
      "num_tokens": 16829508.0,
      "step": 9695
    },
    {
      "entropy": 5.555881786346435,
      "epoch": 0.7547169811320755,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004948527948428064,
      "loss": 5.4011,
      "mean_token_accuracy": 0.16130486726760865,
      "num_tokens": 16838327.0,
      "step": 9700
    },
    {
      "entropy": 5.620137882232666,
      "epoch": 0.755106010503793,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004948468988309187,
      "loss": 5.3888,
      "mean_token_accuracy": 0.17296596765518188,
      "num_tokens": 16846530.0,
      "step": 9705
    },
    {
      "entropy": 5.467872095108032,
      "epoch": 0.7554950398755106,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004948409994832043,
      "loss": 5.3275,
      "mean_token_accuracy": 0.16495684236288072,
      "num_tokens": 16854825.0,
      "step": 9710
    },
    {
      "entropy": 5.5626490116119385,
      "epoch": 0.7558840692472282,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004948350967997526,
      "loss": 5.4759,
      "mean_token_accuracy": 0.1575640469789505,
      "num_tokens": 16864096.0,
      "step": 9715
    },
    {
      "entropy": 5.660004615783691,
      "epoch": 0.7562730986189458,
      "grad_norm": 1.25,
      "learning_rate": 0.0004948291907806532,
      "loss": 5.5227,
      "mean_token_accuracy": 0.15829540193080902,
      "num_tokens": 16873000.0,
      "step": 9720
    },
    {
      "entropy": 5.723625516891479,
      "epoch": 0.7566621279906633,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004948232814259957,
      "loss": 5.5415,
      "mean_token_accuracy": 0.1622430920600891,
      "num_tokens": 16881374.0,
      "step": 9725
    },
    {
      "entropy": 5.621852540969849,
      "epoch": 0.7570511573623808,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004948173687358699,
      "loss": 5.5103,
      "mean_token_accuracy": 0.16261776387691498,
      "num_tokens": 16889192.0,
      "step": 9730
    },
    {
      "entropy": 5.545602655410766,
      "epoch": 0.7574401867340984,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004948114527103652,
      "loss": 5.488,
      "mean_token_accuracy": 0.16274908781051636,
      "num_tokens": 16898538.0,
      "step": 9735
    },
    {
      "entropy": 5.6392418384552006,
      "epoch": 0.757829216105816,
      "grad_norm": 1.375,
      "learning_rate": 0.0004948055333495717,
      "loss": 5.4999,
      "mean_token_accuracy": 0.15664954632520675,
      "num_tokens": 16907543.0,
      "step": 9740
    },
    {
      "entropy": 5.635542583465576,
      "epoch": 0.7582182454775336,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004947996106535791,
      "loss": 5.4135,
      "mean_token_accuracy": 0.16656369417905809,
      "num_tokens": 16916140.0,
      "step": 9745
    },
    {
      "entropy": 5.522709655761719,
      "epoch": 0.7586072748492512,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004947936846224773,
      "loss": 5.3862,
      "mean_token_accuracy": 0.1724488079547882,
      "num_tokens": 16925433.0,
      "step": 9750
    },
    {
      "entropy": 5.516055631637573,
      "epoch": 0.7589963042209686,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004947877552563562,
      "loss": 5.4164,
      "mean_token_accuracy": 0.16631170511245727,
      "num_tokens": 16933598.0,
      "step": 9755
    },
    {
      "entropy": 5.637033700942993,
      "epoch": 0.7593853335926862,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004947818225553056,
      "loss": 5.5087,
      "mean_token_accuracy": 0.1629312515258789,
      "num_tokens": 16942270.0,
      "step": 9760
    },
    {
      "entropy": 5.592579889297485,
      "epoch": 0.7597743629644038,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004947758865194156,
      "loss": 5.3557,
      "mean_token_accuracy": 0.16437536776065825,
      "num_tokens": 16951044.0,
      "step": 9765
    },
    {
      "entropy": 5.596826457977295,
      "epoch": 0.7601633923361214,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004947699471487766,
      "loss": 5.5175,
      "mean_token_accuracy": 0.1575308084487915,
      "num_tokens": 16960215.0,
      "step": 9770
    },
    {
      "entropy": 5.667195224761963,
      "epoch": 0.760552421707839,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004947640044434782,
      "loss": 5.5626,
      "mean_token_accuracy": 0.15760817527770996,
      "num_tokens": 16968515.0,
      "step": 9775
    },
    {
      "entropy": 5.638243579864502,
      "epoch": 0.7609414510795565,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004947580584036109,
      "loss": 5.5382,
      "mean_token_accuracy": 0.1537891298532486,
      "num_tokens": 16977597.0,
      "step": 9780
    },
    {
      "entropy": 5.545149517059326,
      "epoch": 0.761330480451274,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004947521090292649,
      "loss": 5.3903,
      "mean_token_accuracy": 0.16264570951461793,
      "num_tokens": 16987376.0,
      "step": 9785
    },
    {
      "entropy": 5.571061658859253,
      "epoch": 0.7617195098229916,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004947461563205304,
      "loss": 5.3842,
      "mean_token_accuracy": 0.17010038793087007,
      "num_tokens": 16995976.0,
      "step": 9790
    },
    {
      "entropy": 5.502007865905762,
      "epoch": 0.7621085391947092,
      "grad_norm": 1.375,
      "learning_rate": 0.0004947402002774977,
      "loss": 5.4084,
      "mean_token_accuracy": 0.16867428719997407,
      "num_tokens": 17003819.0,
      "step": 9795
    },
    {
      "entropy": 5.503511142730713,
      "epoch": 0.7624975685664268,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004947342409002572,
      "loss": 5.3826,
      "mean_token_accuracy": 0.17320278584957122,
      "num_tokens": 17011632.0,
      "step": 9800
    },
    {
      "entropy": 5.493191051483154,
      "epoch": 0.7628865979381443,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004947282781888994,
      "loss": 5.2979,
      "mean_token_accuracy": 0.1772227868437767,
      "num_tokens": 17019737.0,
      "step": 9805
    },
    {
      "entropy": 5.587402057647705,
      "epoch": 0.7632756273098619,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004947223121435147,
      "loss": 5.4809,
      "mean_token_accuracy": 0.16783572733402252,
      "num_tokens": 17028803.0,
      "step": 9810
    },
    {
      "entropy": 5.591305112838745,
      "epoch": 0.7636646566815795,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004947163427641936,
      "loss": 5.3113,
      "mean_token_accuracy": 0.1684025779366493,
      "num_tokens": 17036847.0,
      "step": 9815
    },
    {
      "entropy": 5.578971099853516,
      "epoch": 0.764053686053297,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004947103700510268,
      "loss": 5.4392,
      "mean_token_accuracy": 0.15950771272182465,
      "num_tokens": 17045665.0,
      "step": 9820
    },
    {
      "entropy": 5.570036697387695,
      "epoch": 0.7644427154250146,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004947043940041047,
      "loss": 5.4636,
      "mean_token_accuracy": 0.1707987070083618,
      "num_tokens": 17053842.0,
      "step": 9825
    },
    {
      "entropy": 5.623847484588623,
      "epoch": 0.7648317447967321,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004946984146235183,
      "loss": 5.5236,
      "mean_token_accuracy": 0.16092466711997985,
      "num_tokens": 17062287.0,
      "step": 9830
    },
    {
      "entropy": 5.614527702331543,
      "epoch": 0.7652207741684497,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004946924319093579,
      "loss": 5.5118,
      "mean_token_accuracy": 0.1656426951289177,
      "num_tokens": 17071495.0,
      "step": 9835
    },
    {
      "entropy": 5.550741147994995,
      "epoch": 0.7656098035401673,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004946864458617148,
      "loss": 5.4586,
      "mean_token_accuracy": 0.16478745639324188,
      "num_tokens": 17080734.0,
      "step": 9840
    },
    {
      "entropy": 5.616221523284912,
      "epoch": 0.7659988329118849,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004946804564806793,
      "loss": 5.4865,
      "mean_token_accuracy": 0.16564612537622453,
      "num_tokens": 17089683.0,
      "step": 9845
    },
    {
      "entropy": 5.6325428009033205,
      "epoch": 0.7663878622836025,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004946744637663427,
      "loss": 5.5018,
      "mean_token_accuracy": 0.16261287778615952,
      "num_tokens": 17098203.0,
      "step": 9850
    },
    {
      "entropy": 5.5683433532714846,
      "epoch": 0.7667768916553199,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004946684677187956,
      "loss": 5.3552,
      "mean_token_accuracy": 0.16821855753660203,
      "num_tokens": 17107185.0,
      "step": 9855
    },
    {
      "entropy": 5.520161914825439,
      "epoch": 0.7671659210270375,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004946624683381292,
      "loss": 5.4143,
      "mean_token_accuracy": 0.16796838492155075,
      "num_tokens": 17115168.0,
      "step": 9860
    },
    {
      "entropy": 5.54301962852478,
      "epoch": 0.7675549503987551,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004946564656244345,
      "loss": 5.4042,
      "mean_token_accuracy": 0.1668022781610489,
      "num_tokens": 17124037.0,
      "step": 9865
    },
    {
      "entropy": 5.5985424518585205,
      "epoch": 0.7679439797704727,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004946504595778026,
      "loss": 5.3303,
      "mean_token_accuracy": 0.17336662262678146,
      "num_tokens": 17132238.0,
      "step": 9870
    },
    {
      "entropy": 5.538533639907837,
      "epoch": 0.7683330091421903,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004946444501983246,
      "loss": 5.4728,
      "mean_token_accuracy": 0.16147430539131163,
      "num_tokens": 17140405.0,
      "step": 9875
    },
    {
      "entropy": 5.490571165084839,
      "epoch": 0.7687220385139077,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004946384374860916,
      "loss": 5.3864,
      "mean_token_accuracy": 0.1733921304345131,
      "num_tokens": 17149207.0,
      "step": 9880
    },
    {
      "entropy": 5.546131992340088,
      "epoch": 0.7691110678856253,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004946324214411949,
      "loss": 5.3741,
      "mean_token_accuracy": 0.16336807906627654,
      "num_tokens": 17157680.0,
      "step": 9885
    },
    {
      "entropy": 5.6140025615692135,
      "epoch": 0.7695000972573429,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004946264020637259,
      "loss": 5.4222,
      "mean_token_accuracy": 0.16739432513713837,
      "num_tokens": 17166907.0,
      "step": 9890
    },
    {
      "entropy": 5.60009126663208,
      "epoch": 0.7698891266290605,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004946203793537757,
      "loss": 5.4452,
      "mean_token_accuracy": 0.17141153067350387,
      "num_tokens": 17176318.0,
      "step": 9895
    },
    {
      "entropy": 5.576356554031372,
      "epoch": 0.7702781560007781,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004946143533114358,
      "loss": 5.4755,
      "mean_token_accuracy": 0.15591812878847122,
      "num_tokens": 17185380.0,
      "step": 9900
    },
    {
      "entropy": 5.58992714881897,
      "epoch": 0.7706671853724957,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004946083239367976,
      "loss": 5.4842,
      "mean_token_accuracy": 0.15915482491254807,
      "num_tokens": 17193177.0,
      "step": 9905
    },
    {
      "entropy": 5.519370412826538,
      "epoch": 0.7710562147442132,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004946022912299527,
      "loss": 5.4394,
      "mean_token_accuracy": 0.16229114979505538,
      "num_tokens": 17201071.0,
      "step": 9910
    },
    {
      "entropy": 5.545923805236816,
      "epoch": 0.7714452441159307,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004945962551909926,
      "loss": 5.3649,
      "mean_token_accuracy": 0.17163790464401246,
      "num_tokens": 17209375.0,
      "step": 9915
    },
    {
      "entropy": 5.579072332382202,
      "epoch": 0.7718342734876483,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004945902158200089,
      "loss": 5.3022,
      "mean_token_accuracy": 0.17475363314151765,
      "num_tokens": 17217453.0,
      "step": 9920
    },
    {
      "entropy": 5.542507314682007,
      "epoch": 0.7722233028593659,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004945841731170931,
      "loss": 5.377,
      "mean_token_accuracy": 0.16708469092845918,
      "num_tokens": 17225671.0,
      "step": 9925
    },
    {
      "entropy": 5.472960567474365,
      "epoch": 0.7726123322310835,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004945781270823369,
      "loss": 5.3431,
      "mean_token_accuracy": 0.16826416403055192,
      "num_tokens": 17234396.0,
      "step": 9930
    },
    {
      "entropy": 5.523881959915161,
      "epoch": 0.773001361602801,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004945720777158323,
      "loss": 5.4493,
      "mean_token_accuracy": 0.16446936130523682,
      "num_tokens": 17243509.0,
      "step": 9935
    },
    {
      "entropy": 5.591121244430542,
      "epoch": 0.7733903909745186,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004945660250176708,
      "loss": 5.3746,
      "mean_token_accuracy": 0.17906951308250427,
      "num_tokens": 17251447.0,
      "step": 9940
    },
    {
      "entropy": 5.5583771705627445,
      "epoch": 0.7737794203462361,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004945599689879443,
      "loss": 5.4417,
      "mean_token_accuracy": 0.17058486491441727,
      "num_tokens": 17260240.0,
      "step": 9945
    },
    {
      "entropy": 5.55849666595459,
      "epoch": 0.7741684497179537,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004945539096267448,
      "loss": 5.4316,
      "mean_token_accuracy": 0.16558820605278016,
      "num_tokens": 17267835.0,
      "step": 9950
    },
    {
      "entropy": 5.552744770050049,
      "epoch": 0.7745574790896713,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004945478469341642,
      "loss": 5.434,
      "mean_token_accuracy": 0.1676381543278694,
      "num_tokens": 17277069.0,
      "step": 9955
    },
    {
      "entropy": 5.609550714492798,
      "epoch": 0.7749465084613888,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004945417809102944,
      "loss": 5.4118,
      "mean_token_accuracy": 0.16238961815834047,
      "num_tokens": 17285014.0,
      "step": 9960
    },
    {
      "entropy": 5.5125470638275145,
      "epoch": 0.7753355378331064,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004945357115552275,
      "loss": 5.3624,
      "mean_token_accuracy": 0.17218244075775146,
      "num_tokens": 17292649.0,
      "step": 9965
    },
    {
      "entropy": 5.595477867126465,
      "epoch": 0.775724567204824,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004945296388690557,
      "loss": 5.5176,
      "mean_token_accuracy": 0.16096659675240516,
      "num_tokens": 17301792.0,
      "step": 9970
    },
    {
      "entropy": 5.573460721969605,
      "epoch": 0.7761135965765416,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000494523562851871,
      "loss": 5.2681,
      "mean_token_accuracy": 0.17233715504407882,
      "num_tokens": 17310377.0,
      "step": 9975
    },
    {
      "entropy": 5.60152645111084,
      "epoch": 0.7765026259482591,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004945174835037655,
      "loss": 5.4956,
      "mean_token_accuracy": 0.156092868745327,
      "num_tokens": 17319779.0,
      "step": 9980
    },
    {
      "entropy": 5.708641624450683,
      "epoch": 0.7768916553199766,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004945114008248319,
      "loss": 5.6191,
      "mean_token_accuracy": 0.15715108811855316,
      "num_tokens": 17328210.0,
      "step": 9985
    },
    {
      "entropy": 5.564361572265625,
      "epoch": 0.7772806846916942,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004945053148151619,
      "loss": 5.3434,
      "mean_token_accuracy": 0.1729060485959053,
      "num_tokens": 17336287.0,
      "step": 9990
    },
    {
      "entropy": 5.5056750774383545,
      "epoch": 0.7776697140634118,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004944992254748482,
      "loss": 5.3962,
      "mean_token_accuracy": 0.16017369776964188,
      "num_tokens": 17344171.0,
      "step": 9995
    },
    {
      "entropy": 5.491131114959717,
      "epoch": 0.7780587434351294,
      "grad_norm": 1.375,
      "learning_rate": 0.0004944931328039832,
      "loss": 5.4289,
      "mean_token_accuracy": 0.16791253089904784,
      "num_tokens": 17352786.0,
      "step": 10000
    },
    {
      "entropy": 5.591256380081177,
      "epoch": 0.778447772806847,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004944870368026592,
      "loss": 5.4065,
      "mean_token_accuracy": 0.17025848478078842,
      "num_tokens": 17361824.0,
      "step": 10005
    },
    {
      "entropy": 5.635959339141846,
      "epoch": 0.7788368021785644,
      "grad_norm": 1.25,
      "learning_rate": 0.0004944809374709689,
      "loss": 5.4291,
      "mean_token_accuracy": 0.16651756316423416,
      "num_tokens": 17370249.0,
      "step": 10010
    },
    {
      "entropy": 5.5495141506195065,
      "epoch": 0.779225831550282,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004944748348090045,
      "loss": 5.397,
      "mean_token_accuracy": 0.1673339620232582,
      "num_tokens": 17379650.0,
      "step": 10015
    },
    {
      "entropy": 5.537599706649781,
      "epoch": 0.7796148609219996,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004944687288168589,
      "loss": 5.4175,
      "mean_token_accuracy": 0.17041378021240233,
      "num_tokens": 17388046.0,
      "step": 10020
    },
    {
      "entropy": 5.59211835861206,
      "epoch": 0.7800038902937172,
      "grad_norm": 1.375,
      "learning_rate": 0.0004944626194946246,
      "loss": 5.4061,
      "mean_token_accuracy": 0.17399909794330598,
      "num_tokens": 17395851.0,
      "step": 10025
    },
    {
      "entropy": 5.557498359680176,
      "epoch": 0.7803929196654348,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004944565068423945,
      "loss": 5.4783,
      "mean_token_accuracy": 0.16642236411571504,
      "num_tokens": 17404273.0,
      "step": 10030
    },
    {
      "entropy": 5.511315631866455,
      "epoch": 0.7807819490371523,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004944503908602612,
      "loss": 5.4111,
      "mean_token_accuracy": 0.1673135057091713,
      "num_tokens": 17412884.0,
      "step": 10035
    },
    {
      "entropy": 5.5740358352661135,
      "epoch": 0.7811709784088698,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004944442715483174,
      "loss": 5.4002,
      "mean_token_accuracy": 0.16286081820726395,
      "num_tokens": 17420939.0,
      "step": 10040
    },
    {
      "entropy": 5.550879192352295,
      "epoch": 0.7815600077805874,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004944381489066561,
      "loss": 5.4158,
      "mean_token_accuracy": 0.16733554303646087,
      "num_tokens": 17429838.0,
      "step": 10045
    },
    {
      "entropy": 5.580566549301148,
      "epoch": 0.781949037152305,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004944320229353702,
      "loss": 5.4064,
      "mean_token_accuracy": 0.16913883984088898,
      "num_tokens": 17439164.0,
      "step": 10050
    },
    {
      "entropy": 5.607364130020142,
      "epoch": 0.7823380665240226,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004944258936345526,
      "loss": 5.3974,
      "mean_token_accuracy": 0.16829439699649812,
      "num_tokens": 17447544.0,
      "step": 10055
    },
    {
      "entropy": 5.431110858917236,
      "epoch": 0.7827270958957401,
      "grad_norm": 5.5,
      "learning_rate": 0.0004944197610042963,
      "loss": 5.2755,
      "mean_token_accuracy": 0.1729186788201332,
      "num_tokens": 17456270.0,
      "step": 10060
    },
    {
      "entropy": 5.533744430541992,
      "epoch": 0.7831161252674577,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004944136250446944,
      "loss": 5.4564,
      "mean_token_accuracy": 0.16455423831939697,
      "num_tokens": 17465241.0,
      "step": 10065
    },
    {
      "entropy": 5.645198774337769,
      "epoch": 0.7835051546391752,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004944074857558399,
      "loss": 5.4646,
      "mean_token_accuracy": 0.15913430005311965,
      "num_tokens": 17474460.0,
      "step": 10070
    },
    {
      "entropy": 5.533535814285278,
      "epoch": 0.7838941840108928,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004944013431378261,
      "loss": 5.4337,
      "mean_token_accuracy": 0.16520754545927047,
      "num_tokens": 17483256.0,
      "step": 10075
    },
    {
      "entropy": 5.610069465637207,
      "epoch": 0.7842832133826104,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004943951971907461,
      "loss": 5.4298,
      "mean_token_accuracy": 0.16506215929985046,
      "num_tokens": 17491922.0,
      "step": 10080
    },
    {
      "entropy": 5.579332876205444,
      "epoch": 0.7846722427543279,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004943890479146933,
      "loss": 5.4285,
      "mean_token_accuracy": 0.17432969957590103,
      "num_tokens": 17499763.0,
      "step": 10085
    },
    {
      "entropy": 5.546515226364136,
      "epoch": 0.7850612721260455,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004943828953097607,
      "loss": 5.4103,
      "mean_token_accuracy": 0.16522713303565978,
      "num_tokens": 17508682.0,
      "step": 10090
    },
    {
      "entropy": 5.519997215270996,
      "epoch": 0.7854503014977631,
      "grad_norm": 1.34375,
      "learning_rate": 0.000494376739376042,
      "loss": 5.337,
      "mean_token_accuracy": 0.1708441659808159,
      "num_tokens": 17517269.0,
      "step": 10095
    },
    {
      "entropy": 5.4668412685394285,
      "epoch": 0.7858393308694807,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004943705801136304,
      "loss": 5.3693,
      "mean_token_accuracy": 0.16998788565397263,
      "num_tokens": 17525135.0,
      "step": 10100
    },
    {
      "entropy": 5.567144107818604,
      "epoch": 0.7862283602411982,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004943644175226193,
      "loss": 5.4505,
      "mean_token_accuracy": 0.16855544447898865,
      "num_tokens": 17534195.0,
      "step": 10105
    },
    {
      "entropy": 5.557143831253052,
      "epoch": 0.7866173896129158,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004943582516031025,
      "loss": 5.3634,
      "mean_token_accuracy": 0.16626687496900558,
      "num_tokens": 17543016.0,
      "step": 10110
    },
    {
      "entropy": 5.5780411720275875,
      "epoch": 0.7870064189846333,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004943520823551732,
      "loss": 5.4552,
      "mean_token_accuracy": 0.16688260585069656,
      "num_tokens": 17551854.0,
      "step": 10115
    },
    {
      "entropy": 5.55659556388855,
      "epoch": 0.7873954483563509,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004943459097789252,
      "loss": 5.3728,
      "mean_token_accuracy": 0.16753175854682922,
      "num_tokens": 17560739.0,
      "step": 10120
    },
    {
      "entropy": 5.55966010093689,
      "epoch": 0.7877844777280685,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004943397338744523,
      "loss": 5.3683,
      "mean_token_accuracy": 0.16686999499797822,
      "num_tokens": 17569784.0,
      "step": 10125
    },
    {
      "entropy": 5.633488035202026,
      "epoch": 0.7881735070997861,
      "grad_norm": 1.3125,
      "learning_rate": 0.000494333554641848,
      "loss": 5.5542,
      "mean_token_accuracy": 0.1608866572380066,
      "num_tokens": 17579262.0,
      "step": 10130
    },
    {
      "entropy": 5.576036643981934,
      "epoch": 0.7885625364715037,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000494327372081206,
      "loss": 5.4029,
      "mean_token_accuracy": 0.16805678904056548,
      "num_tokens": 17587984.0,
      "step": 10135
    },
    {
      "entropy": 5.501877021789551,
      "epoch": 0.7889515658432211,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004943211861926204,
      "loss": 5.3364,
      "mean_token_accuracy": 0.17502479553222655,
      "num_tokens": 17596598.0,
      "step": 10140
    },
    {
      "entropy": 5.508901453018188,
      "epoch": 0.7893405952149387,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004943149969761848,
      "loss": 5.4658,
      "mean_token_accuracy": 0.1675132155418396,
      "num_tokens": 17605768.0,
      "step": 10145
    },
    {
      "entropy": 5.587047004699707,
      "epoch": 0.7897296245866563,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004943088044319932,
      "loss": 5.3994,
      "mean_token_accuracy": 0.16717396527528763,
      "num_tokens": 17614412.0,
      "step": 10150
    },
    {
      "entropy": 5.590031147003174,
      "epoch": 0.7901186539583739,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004943026085601395,
      "loss": 5.398,
      "mean_token_accuracy": 0.1643920585513115,
      "num_tokens": 17622515.0,
      "step": 10155
    },
    {
      "entropy": 5.502378797531128,
      "epoch": 0.7905076833300915,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004942964093607178,
      "loss": 5.3309,
      "mean_token_accuracy": 0.1700050339102745,
      "num_tokens": 17631256.0,
      "step": 10160
    },
    {
      "entropy": 5.588741827011108,
      "epoch": 0.790896712701809,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004942902068338222,
      "loss": 5.4735,
      "mean_token_accuracy": 0.1624934881925583,
      "num_tokens": 17639648.0,
      "step": 10165
    },
    {
      "entropy": 5.552893495559692,
      "epoch": 0.7912857420735265,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004942840009795466,
      "loss": 5.3947,
      "mean_token_accuracy": 0.1652170479297638,
      "num_tokens": 17648225.0,
      "step": 10170
    },
    {
      "entropy": 5.572090578079224,
      "epoch": 0.7916747714452441,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004942777917979855,
      "loss": 5.4406,
      "mean_token_accuracy": 0.17100584059953688,
      "num_tokens": 17656764.0,
      "step": 10175
    },
    {
      "entropy": 5.5802239894866945,
      "epoch": 0.7920638008169617,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004942715792892328,
      "loss": 5.398,
      "mean_token_accuracy": 0.16319905817508698,
      "num_tokens": 17665127.0,
      "step": 10180
    },
    {
      "entropy": 5.476864814758301,
      "epoch": 0.7924528301886793,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000494265363453383,
      "loss": 5.3109,
      "mean_token_accuracy": 0.17378859370946884,
      "num_tokens": 17673297.0,
      "step": 10185
    },
    {
      "entropy": 5.55854434967041,
      "epoch": 0.7928418595603968,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004942591442905302,
      "loss": 5.4598,
      "mean_token_accuracy": 0.16654280573129654,
      "num_tokens": 17681816.0,
      "step": 10190
    },
    {
      "entropy": 5.584398698806763,
      "epoch": 0.7932308889321144,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004942529218007689,
      "loss": 5.4628,
      "mean_token_accuracy": 0.16161804646253586,
      "num_tokens": 17690588.0,
      "step": 10195
    },
    {
      "entropy": 5.540796375274658,
      "epoch": 0.7936199183038319,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004942466959841936,
      "loss": 5.3697,
      "mean_token_accuracy": 0.16721461415290834,
      "num_tokens": 17699196.0,
      "step": 10200
    },
    {
      "entropy": 5.6265634059906,
      "epoch": 0.7940089476755495,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004942404668408985,
      "loss": 5.4735,
      "mean_token_accuracy": 0.1656789228320122,
      "num_tokens": 17708928.0,
      "step": 10205
    },
    {
      "entropy": 5.624050903320312,
      "epoch": 0.7943979770472671,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004942342343709783,
      "loss": 5.3659,
      "mean_token_accuracy": 0.16987982839345933,
      "num_tokens": 17717888.0,
      "step": 10210
    },
    {
      "entropy": 5.539783382415772,
      "epoch": 0.7947870064189846,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004942279985745275,
      "loss": 5.3769,
      "mean_token_accuracy": 0.162916961312294,
      "num_tokens": 17726022.0,
      "step": 10215
    },
    {
      "entropy": 5.483287954330445,
      "epoch": 0.7951760357907022,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004942217594516409,
      "loss": 5.3717,
      "mean_token_accuracy": 0.16282805055379868,
      "num_tokens": 17733998.0,
      "step": 10220
    },
    {
      "entropy": 5.60789794921875,
      "epoch": 0.7955650651624198,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000494215517002413,
      "loss": 5.4142,
      "mean_token_accuracy": 0.17179883271455765,
      "num_tokens": 17742039.0,
      "step": 10225
    },
    {
      "entropy": 5.552068567276001,
      "epoch": 0.7959540945341373,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004942092712269384,
      "loss": 5.3215,
      "mean_token_accuracy": 0.17505754977464677,
      "num_tokens": 17751474.0,
      "step": 10230
    },
    {
      "entropy": 5.480428552627563,
      "epoch": 0.7963431239058549,
      "grad_norm": 1.375,
      "learning_rate": 0.0004942030221253122,
      "loss": 5.3376,
      "mean_token_accuracy": 0.17217674106359482,
      "num_tokens": 17759483.0,
      "step": 10235
    },
    {
      "entropy": 5.503555727005005,
      "epoch": 0.7967321532775724,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004941967696976289,
      "loss": 5.448,
      "mean_token_accuracy": 0.17107067555189132,
      "num_tokens": 17768025.0,
      "step": 10240
    },
    {
      "entropy": 5.483997249603272,
      "epoch": 0.79712118264929,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004941905139439835,
      "loss": 5.2828,
      "mean_token_accuracy": 0.17359258383512496,
      "num_tokens": 17776861.0,
      "step": 10245
    },
    {
      "entropy": 5.516780042648316,
      "epoch": 0.7975102120210076,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004941842548644712,
      "loss": 5.4323,
      "mean_token_accuracy": 0.16594091355800628,
      "num_tokens": 17786221.0,
      "step": 10250
    },
    {
      "entropy": 5.628681230545044,
      "epoch": 0.7978992413927252,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004941779924591864,
      "loss": 5.4237,
      "mean_token_accuracy": 0.16071717143058778,
      "num_tokens": 17794619.0,
      "step": 10255
    },
    {
      "entropy": 5.599137878417968,
      "epoch": 0.7982882707644428,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004941717267282244,
      "loss": 5.4129,
      "mean_token_accuracy": 0.17055138796567917,
      "num_tokens": 17803288.0,
      "step": 10260
    },
    {
      "entropy": 5.498247051239014,
      "epoch": 0.7986773001361602,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004941654576716806,
      "loss": 5.4685,
      "mean_token_accuracy": 0.1604743629693985,
      "num_tokens": 17812648.0,
      "step": 10265
    },
    {
      "entropy": 5.502545595169067,
      "epoch": 0.7990663295078778,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004941591852896495,
      "loss": 5.3848,
      "mean_token_accuracy": 0.17225033342838286,
      "num_tokens": 17821955.0,
      "step": 10270
    },
    {
      "entropy": 5.631287527084351,
      "epoch": 0.7994553588795954,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004941529095822268,
      "loss": 5.4153,
      "mean_token_accuracy": 0.16885472387075423,
      "num_tokens": 17829987.0,
      "step": 10275
    },
    {
      "entropy": 5.555313205718994,
      "epoch": 0.799844388251313,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004941466305495074,
      "loss": 5.3751,
      "mean_token_accuracy": 0.1739025294780731,
      "num_tokens": 17839298.0,
      "step": 10280
    },
    {
      "entropy": 5.5623852729797365,
      "epoch": 0.8002334176230306,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004941403481915867,
      "loss": 5.3915,
      "mean_token_accuracy": 0.16719794869422913,
      "num_tokens": 17847950.0,
      "step": 10285
    },
    {
      "entropy": 5.629810810089111,
      "epoch": 0.800622446994748,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004941340625085601,
      "loss": 5.5145,
      "mean_token_accuracy": 0.16282792538404464,
      "num_tokens": 17857416.0,
      "step": 10290
    },
    {
      "entropy": 5.605551862716675,
      "epoch": 0.8010114763664656,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004941277735005228,
      "loss": 5.379,
      "mean_token_accuracy": 0.1750679075717926,
      "num_tokens": 17866516.0,
      "step": 10295
    },
    {
      "entropy": 5.661288738250732,
      "epoch": 0.8014005057381832,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004941214811675702,
      "loss": 5.5421,
      "mean_token_accuracy": 0.15751834213733673,
      "num_tokens": 17875823.0,
      "step": 10300
    },
    {
      "entropy": 5.514067506790161,
      "epoch": 0.8017895351099008,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004941151855097982,
      "loss": 5.4293,
      "mean_token_accuracy": 0.1661258801817894,
      "num_tokens": 17883597.0,
      "step": 10305
    },
    {
      "entropy": 5.609772872924805,
      "epoch": 0.8021785644816184,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004941088865273018,
      "loss": 5.3959,
      "mean_token_accuracy": 0.1608137682080269,
      "num_tokens": 17892053.0,
      "step": 10310
    },
    {
      "entropy": 5.53999342918396,
      "epoch": 0.802567593853336,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004941025842201769,
      "loss": 5.3104,
      "mean_token_accuracy": 0.17627227008342744,
      "num_tokens": 17900187.0,
      "step": 10315
    },
    {
      "entropy": 5.521771669387817,
      "epoch": 0.8029566232250535,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004940962785885189,
      "loss": 5.4731,
      "mean_token_accuracy": 0.16623075157403946,
      "num_tokens": 17908608.0,
      "step": 10320
    },
    {
      "entropy": 5.625722503662109,
      "epoch": 0.803345652596771,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004940899696324237,
      "loss": 5.371,
      "mean_token_accuracy": 0.170322048664093,
      "num_tokens": 17917331.0,
      "step": 10325
    },
    {
      "entropy": 5.65689606666565,
      "epoch": 0.8037346819684886,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004940836573519868,
      "loss": 5.4835,
      "mean_token_accuracy": 0.1651667684316635,
      "num_tokens": 17925228.0,
      "step": 10330
    },
    {
      "entropy": 5.587020063400269,
      "epoch": 0.8041237113402062,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004940773417473043,
      "loss": 5.3758,
      "mean_token_accuracy": 0.1679133266210556,
      "num_tokens": 17934110.0,
      "step": 10335
    },
    {
      "entropy": 5.547002744674683,
      "epoch": 0.8045127407119238,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004940710228184717,
      "loss": 5.3768,
      "mean_token_accuracy": 0.17224250584840775,
      "num_tokens": 17942237.0,
      "step": 10340
    },
    {
      "entropy": 5.50104341506958,
      "epoch": 0.8049017700836413,
      "grad_norm": 1.34375,
      "learning_rate": 0.000494064700565585,
      "loss": 5.2382,
      "mean_token_accuracy": 0.178866146504879,
      "num_tokens": 17951288.0,
      "step": 10345
    },
    {
      "entropy": 5.586454629898071,
      "epoch": 0.8052907994553589,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004940583749887403,
      "loss": 5.4733,
      "mean_token_accuracy": 0.15940811038017272,
      "num_tokens": 17960300.0,
      "step": 10350
    },
    {
      "entropy": 5.5388120174407955,
      "epoch": 0.8056798288270764,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004940520460880333,
      "loss": 5.3552,
      "mean_token_accuracy": 0.1721418872475624,
      "num_tokens": 17968745.0,
      "step": 10355
    },
    {
      "entropy": 5.506756687164307,
      "epoch": 0.806068858198794,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004940457138635601,
      "loss": 5.3545,
      "mean_token_accuracy": 0.1702168330550194,
      "num_tokens": 17976614.0,
      "step": 10360
    },
    {
      "entropy": 5.576000690460205,
      "epoch": 0.8064578875705116,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004940393783154169,
      "loss": 5.5197,
      "mean_token_accuracy": 0.15855003893375397,
      "num_tokens": 17984769.0,
      "step": 10365
    },
    {
      "entropy": 5.634671592712403,
      "epoch": 0.8068469169422291,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004940330394436999,
      "loss": 5.3996,
      "mean_token_accuracy": 0.16306686401367188,
      "num_tokens": 17992885.0,
      "step": 10370
    },
    {
      "entropy": 5.597893905639649,
      "epoch": 0.8072359463139467,
      "grad_norm": 1.375,
      "learning_rate": 0.000494026697248505,
      "loss": 5.4328,
      "mean_token_accuracy": 0.1607662945985794,
      "num_tokens": 18001752.0,
      "step": 10375
    },
    {
      "entropy": 5.531260824203491,
      "epoch": 0.8076249756856643,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004940203517299287,
      "loss": 5.4169,
      "mean_token_accuracy": 0.17741715013980866,
      "num_tokens": 18009939.0,
      "step": 10380
    },
    {
      "entropy": 5.635455656051636,
      "epoch": 0.8080140050573819,
      "grad_norm": 1.453125,
      "learning_rate": 0.000494014002888067,
      "loss": 5.4023,
      "mean_token_accuracy": 0.17565066665410994,
      "num_tokens": 18017993.0,
      "step": 10385
    },
    {
      "entropy": 5.587710809707642,
      "epoch": 0.8084030344290994,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004940076507230166,
      "loss": 5.4846,
      "mean_token_accuracy": 0.16297935992479323,
      "num_tokens": 18027564.0,
      "step": 10390
    },
    {
      "entropy": 5.5873720169067385,
      "epoch": 0.8087920638008169,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004940012952348735,
      "loss": 5.4717,
      "mean_token_accuracy": 0.16622072607278823,
      "num_tokens": 18036174.0,
      "step": 10395
    },
    {
      "entropy": 5.576321458816528,
      "epoch": 0.8091810931725345,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004939949364237345,
      "loss": 5.3783,
      "mean_token_accuracy": 0.16790819317102432,
      "num_tokens": 18044994.0,
      "step": 10400
    },
    {
      "entropy": 5.561063766479492,
      "epoch": 0.8095701225442521,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004939885742896958,
      "loss": 5.392,
      "mean_token_accuracy": 0.16774575859308244,
      "num_tokens": 18053789.0,
      "step": 10405
    },
    {
      "entropy": 5.608617210388184,
      "epoch": 0.8099591519159697,
      "grad_norm": 1.359375,
      "learning_rate": 0.000493982208832854,
      "loss": 5.4963,
      "mean_token_accuracy": 0.16115783751010895,
      "num_tokens": 18063483.0,
      "step": 10410
    },
    {
      "entropy": 5.5316260814666744,
      "epoch": 0.8103481812876873,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004939758400533058,
      "loss": 5.3148,
      "mean_token_accuracy": 0.16901348084211348,
      "num_tokens": 18071952.0,
      "step": 10415
    },
    {
      "entropy": 5.576836442947387,
      "epoch": 0.8107372106594047,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004939694679511478,
      "loss": 5.356,
      "mean_token_accuracy": 0.1726394325494766,
      "num_tokens": 18080323.0,
      "step": 10420
    },
    {
      "entropy": 5.609509134292603,
      "epoch": 0.8111262400311223,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004939630925264765,
      "loss": 5.4609,
      "mean_token_accuracy": 0.16328188627958298,
      "num_tokens": 18089458.0,
      "step": 10425
    },
    {
      "entropy": 5.597567987442017,
      "epoch": 0.8115152694028399,
      "grad_norm": 1.453125,
      "learning_rate": 0.000493956713779389,
      "loss": 5.3838,
      "mean_token_accuracy": 0.16132101863622667,
      "num_tokens": 18098553.0,
      "step": 10430
    },
    {
      "entropy": 5.5492987632751465,
      "epoch": 0.8119042987745575,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004939503317099817,
      "loss": 5.4736,
      "mean_token_accuracy": 0.16533359438180922,
      "num_tokens": 18106794.0,
      "step": 10435
    },
    {
      "entropy": 5.536823797225952,
      "epoch": 0.8122933281462751,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004939439463183517,
      "loss": 5.3392,
      "mean_token_accuracy": 0.17259827703237535,
      "num_tokens": 18115509.0,
      "step": 10440
    },
    {
      "entropy": 5.594719076156617,
      "epoch": 0.8126823575179926,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004939375576045958,
      "loss": 5.4878,
      "mean_token_accuracy": 0.15990200787782669,
      "num_tokens": 18124548.0,
      "step": 10445
    },
    {
      "entropy": 5.606013441085816,
      "epoch": 0.8130713868897101,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004939311655688109,
      "loss": 5.3905,
      "mean_token_accuracy": 0.16725083142518998,
      "num_tokens": 18132295.0,
      "step": 10450
    },
    {
      "entropy": 5.613696432113647,
      "epoch": 0.8134604162614277,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004939247702110941,
      "loss": 5.5058,
      "mean_token_accuracy": 0.16281792968511583,
      "num_tokens": 18140731.0,
      "step": 10455
    },
    {
      "entropy": 5.621880006790161,
      "epoch": 0.8138494456331453,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004939183715315423,
      "loss": 5.3675,
      "mean_token_accuracy": 0.16817042976617813,
      "num_tokens": 18149834.0,
      "step": 10460
    },
    {
      "entropy": 5.580629444122314,
      "epoch": 0.8142384750048629,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004939119695302526,
      "loss": 5.3169,
      "mean_token_accuracy": 0.17500500231981278,
      "num_tokens": 18158745.0,
      "step": 10465
    },
    {
      "entropy": 5.624244308471679,
      "epoch": 0.8146275043765804,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004939055642073224,
      "loss": 5.5212,
      "mean_token_accuracy": 0.15916655585169792,
      "num_tokens": 18168292.0,
      "step": 10470
    },
    {
      "entropy": 5.62207145690918,
      "epoch": 0.815016533748298,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004938991555628484,
      "loss": 5.4151,
      "mean_token_accuracy": 0.16294815689325332,
      "num_tokens": 18177592.0,
      "step": 10475
    },
    {
      "entropy": 5.580402755737305,
      "epoch": 0.8154055631200156,
      "grad_norm": 1.25,
      "learning_rate": 0.0004938927435969283,
      "loss": 5.3852,
      "mean_token_accuracy": 0.16239623576402665,
      "num_tokens": 18186845.0,
      "step": 10480
    },
    {
      "entropy": 5.637283706665039,
      "epoch": 0.8157945924917331,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004938863283096592,
      "loss": 5.4729,
      "mean_token_accuracy": 0.16739440113306045,
      "num_tokens": 18195417.0,
      "step": 10485
    },
    {
      "entropy": 5.552554273605347,
      "epoch": 0.8161836218634507,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004938799097011384,
      "loss": 5.4096,
      "mean_token_accuracy": 0.16463044583797454,
      "num_tokens": 18204009.0,
      "step": 10490
    },
    {
      "entropy": 5.497044324874878,
      "epoch": 0.8165726512351682,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004938734877714634,
      "loss": 5.3888,
      "mean_token_accuracy": 0.17515998482704162,
      "num_tokens": 18211927.0,
      "step": 10495
    },
    {
      "entropy": 5.472809934616089,
      "epoch": 0.8169616806068858,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004938670625207317,
      "loss": 5.3055,
      "mean_token_accuracy": 0.17404091656208037,
      "num_tokens": 18220394.0,
      "step": 10500
    },
    {
      "entropy": 5.583059215545655,
      "epoch": 0.8173507099786034,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004938606339490406,
      "loss": 5.4215,
      "mean_token_accuracy": 0.16456349939107895,
      "num_tokens": 18229803.0,
      "step": 10505
    },
    {
      "entropy": 5.5503377437591555,
      "epoch": 0.817739739350321,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004938542020564877,
      "loss": 5.3837,
      "mean_token_accuracy": 0.16562126725912094,
      "num_tokens": 18238431.0,
      "step": 10510
    },
    {
      "entropy": 5.582013082504273,
      "epoch": 0.8181287687220385,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004938477668431706,
      "loss": 5.3985,
      "mean_token_accuracy": 0.16772979199886323,
      "num_tokens": 18246745.0,
      "step": 10515
    },
    {
      "entropy": 5.543431568145752,
      "epoch": 0.8185177980937561,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004938413283091871,
      "loss": 5.3778,
      "mean_token_accuracy": 0.17051088362932204,
      "num_tokens": 18255569.0,
      "step": 10520
    },
    {
      "entropy": 5.5197038650512695,
      "epoch": 0.8189068274654736,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004938348864546347,
      "loss": 5.297,
      "mean_token_accuracy": 0.17485857903957366,
      "num_tokens": 18263711.0,
      "step": 10525
    },
    {
      "entropy": 5.667397832870483,
      "epoch": 0.8192958568371912,
      "grad_norm": 1.5,
      "learning_rate": 0.0004938284412796111,
      "loss": 5.4981,
      "mean_token_accuracy": 0.16424625217914582,
      "num_tokens": 18272220.0,
      "step": 10530
    },
    {
      "entropy": 5.5894135475158695,
      "epoch": 0.8196848862089088,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004938219927842144,
      "loss": 5.3978,
      "mean_token_accuracy": 0.1659523755311966,
      "num_tokens": 18281107.0,
      "step": 10535
    },
    {
      "entropy": 5.475996685028076,
      "epoch": 0.8200739155806264,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004938155409685422,
      "loss": 5.3688,
      "mean_token_accuracy": 0.17090282291173936,
      "num_tokens": 18290196.0,
      "step": 10540
    },
    {
      "entropy": 5.592043733596801,
      "epoch": 0.820462944952344,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004938090858326923,
      "loss": 5.5008,
      "mean_token_accuracy": 0.16262528747320176,
      "num_tokens": 18299418.0,
      "step": 10545
    },
    {
      "entropy": 5.506319332122803,
      "epoch": 0.8208519743240614,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000493802627376763,
      "loss": 5.2269,
      "mean_token_accuracy": 0.17988054901361467,
      "num_tokens": 18308171.0,
      "step": 10550
    },
    {
      "entropy": 5.591129875183105,
      "epoch": 0.821241003695779,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004937961656008518,
      "loss": 5.4356,
      "mean_token_accuracy": 0.1670076832175255,
      "num_tokens": 18317144.0,
      "step": 10555
    },
    {
      "entropy": 5.574430656433106,
      "epoch": 0.8216300330674966,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004937897005050573,
      "loss": 5.4899,
      "mean_token_accuracy": 0.1618042677640915,
      "num_tokens": 18326648.0,
      "step": 10560
    },
    {
      "entropy": 5.597058820724487,
      "epoch": 0.8220190624392142,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004937832320894772,
      "loss": 5.4385,
      "mean_token_accuracy": 0.16716190576553344,
      "num_tokens": 18335901.0,
      "step": 10565
    },
    {
      "entropy": 5.528044033050537,
      "epoch": 0.8224080918109318,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004937767603542098,
      "loss": 5.3635,
      "mean_token_accuracy": 0.16943397670984267,
      "num_tokens": 18344905.0,
      "step": 10570
    },
    {
      "entropy": 5.469264793395996,
      "epoch": 0.8227971211826492,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004937702852993534,
      "loss": 5.3326,
      "mean_token_accuracy": 0.1726791799068451,
      "num_tokens": 18353422.0,
      "step": 10575
    },
    {
      "entropy": 5.696764707565308,
      "epoch": 0.8231861505543668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000493763806925006,
      "loss": 5.483,
      "mean_token_accuracy": 0.15566707998514176,
      "num_tokens": 18361632.0,
      "step": 10580
    },
    {
      "entropy": 5.538483762741089,
      "epoch": 0.8235751799260844,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004937573252312661,
      "loss": 5.3249,
      "mean_token_accuracy": 0.17498582899570464,
      "num_tokens": 18370142.0,
      "step": 10585
    },
    {
      "entropy": 5.459922552108765,
      "epoch": 0.823964209297802,
      "grad_norm": 1.3125,
      "learning_rate": 0.000493750840218232,
      "loss": 5.3635,
      "mean_token_accuracy": 0.16874400824308394,
      "num_tokens": 18378245.0,
      "step": 10590
    },
    {
      "entropy": 5.541511821746826,
      "epoch": 0.8243532386695196,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004937443518860021,
      "loss": 5.3262,
      "mean_token_accuracy": 0.173284213244915,
      "num_tokens": 18386814.0,
      "step": 10595
    },
    {
      "entropy": 5.571748304367065,
      "epoch": 0.8247422680412371,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004937378602346747,
      "loss": 5.3723,
      "mean_token_accuracy": 0.1653219237923622,
      "num_tokens": 18396276.0,
      "step": 10600
    },
    {
      "entropy": 5.5731720447540285,
      "epoch": 0.8251312974129547,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004937313652643485,
      "loss": 5.3771,
      "mean_token_accuracy": 0.17177846729755403,
      "num_tokens": 18404445.0,
      "step": 10605
    },
    {
      "entropy": 5.470026350021362,
      "epoch": 0.8255203267846722,
      "grad_norm": 1.6875,
      "learning_rate": 0.000493724866975122,
      "loss": 5.2808,
      "mean_token_accuracy": 0.17622358053922654,
      "num_tokens": 18412870.0,
      "step": 10610
    },
    {
      "entropy": 5.491532564163208,
      "epoch": 0.8259093561563898,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004937183653670937,
      "loss": 5.4074,
      "mean_token_accuracy": 0.16762313693761827,
      "num_tokens": 18421084.0,
      "step": 10615
    },
    {
      "entropy": 5.556626224517823,
      "epoch": 0.8262983855281074,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004937118604403623,
      "loss": 5.4282,
      "mean_token_accuracy": 0.17020318359136583,
      "num_tokens": 18429869.0,
      "step": 10620
    },
    {
      "entropy": 5.520753955841064,
      "epoch": 0.8266874148998249,
      "grad_norm": 1.625,
      "learning_rate": 0.0004937053521950266,
      "loss": 5.3115,
      "mean_token_accuracy": 0.16983865946531296,
      "num_tokens": 18438747.0,
      "step": 10625
    },
    {
      "entropy": 5.622309637069702,
      "epoch": 0.8270764442715425,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004936988406311854,
      "loss": 5.4942,
      "mean_token_accuracy": 0.16253411620855332,
      "num_tokens": 18447241.0,
      "step": 10630
    },
    {
      "entropy": 5.629301738739014,
      "epoch": 0.8274654736432601,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004936923257489373,
      "loss": 5.5099,
      "mean_token_accuracy": 0.15812525600194932,
      "num_tokens": 18455757.0,
      "step": 10635
    },
    {
      "entropy": 5.523294401168823,
      "epoch": 0.8278545030149776,
      "grad_norm": 1.375,
      "learning_rate": 0.0004936858075483811,
      "loss": 5.3392,
      "mean_token_accuracy": 0.17323616594076158,
      "num_tokens": 18463863.0,
      "step": 10640
    },
    {
      "entropy": 5.558644199371338,
      "epoch": 0.8282435323866952,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000493679286029616,
      "loss": 5.5389,
      "mean_token_accuracy": 0.1602210983633995,
      "num_tokens": 18472396.0,
      "step": 10645
    },
    {
      "entropy": 5.6205915927886965,
      "epoch": 0.8286325617584127,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004936727611927407,
      "loss": 5.3765,
      "mean_token_accuracy": 0.1661218747496605,
      "num_tokens": 18480625.0,
      "step": 10650
    },
    {
      "entropy": 5.576664781570434,
      "epoch": 0.8290215911301303,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004936662330378546,
      "loss": 5.4521,
      "mean_token_accuracy": 0.1626016840338707,
      "num_tokens": 18489878.0,
      "step": 10655
    },
    {
      "entropy": 5.497170114517212,
      "epoch": 0.8294106205018479,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004936597015650561,
      "loss": 5.3412,
      "mean_token_accuracy": 0.1716286301612854,
      "num_tokens": 18498873.0,
      "step": 10660
    },
    {
      "entropy": 5.641657018661499,
      "epoch": 0.8297996498735655,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004936531667744448,
      "loss": 5.4719,
      "mean_token_accuracy": 0.16033537536859513,
      "num_tokens": 18507716.0,
      "step": 10665
    },
    {
      "entropy": 5.583861684799194,
      "epoch": 0.8301886792452831,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004936466286661197,
      "loss": 5.3063,
      "mean_token_accuracy": 0.17494824826717376,
      "num_tokens": 18516547.0,
      "step": 10670
    },
    {
      "entropy": 5.526273202896118,
      "epoch": 0.8305777086170005,
      "grad_norm": 1.28125,
      "learning_rate": 0.00049364008724018,
      "loss": 5.4548,
      "mean_token_accuracy": 0.16137342900037766,
      "num_tokens": 18525714.0,
      "step": 10675
    },
    {
      "entropy": 5.611880302429199,
      "epoch": 0.8309667379887181,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000493633542496725,
      "loss": 5.481,
      "mean_token_accuracy": 0.1679605707526207,
      "num_tokens": 18534981.0,
      "step": 10680
    },
    {
      "entropy": 5.625515031814575,
      "epoch": 0.8313557673604357,
      "grad_norm": 1.375,
      "learning_rate": 0.0004936269944358539,
      "loss": 5.399,
      "mean_token_accuracy": 0.16724058091640473,
      "num_tokens": 18543599.0,
      "step": 10685
    },
    {
      "entropy": 5.572318029403687,
      "epoch": 0.8317447967321533,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004936204430576664,
      "loss": 5.372,
      "mean_token_accuracy": 0.17497016936540605,
      "num_tokens": 18552207.0,
      "step": 10690
    },
    {
      "entropy": 5.528352069854736,
      "epoch": 0.8321338261038709,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004936138883622614,
      "loss": 5.4408,
      "mean_token_accuracy": 0.16688330769538878,
      "num_tokens": 18560421.0,
      "step": 10695
    },
    {
      "entropy": 5.585869073867798,
      "epoch": 0.8325228554755884,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004936073303497387,
      "loss": 5.3424,
      "mean_token_accuracy": 0.17582375705242156,
      "num_tokens": 18568812.0,
      "step": 10700
    },
    {
      "entropy": 5.548599910736084,
      "epoch": 0.8329118848473059,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004936007690201976,
      "loss": 5.3253,
      "mean_token_accuracy": 0.17208590507507324,
      "num_tokens": 18577077.0,
      "step": 10705
    },
    {
      "entropy": 5.541345024108887,
      "epoch": 0.8333009142190235,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000493594204373738,
      "loss": 5.4269,
      "mean_token_accuracy": 0.16951297223567963,
      "num_tokens": 18585754.0,
      "step": 10710
    },
    {
      "entropy": 5.511206150054932,
      "epoch": 0.8336899435907411,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004935876364104591,
      "loss": 5.3186,
      "mean_token_accuracy": 0.17156182676553727,
      "num_tokens": 18594578.0,
      "step": 10715
    },
    {
      "entropy": 5.589276885986328,
      "epoch": 0.8340789729624587,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004935810651304608,
      "loss": 5.4151,
      "mean_token_accuracy": 0.16332758963108063,
      "num_tokens": 18603810.0,
      "step": 10720
    },
    {
      "entropy": 5.628635311126709,
      "epoch": 0.8344680023341763,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004935744905338427,
      "loss": 5.4027,
      "mean_token_accuracy": 0.16077956259250642,
      "num_tokens": 18612870.0,
      "step": 10725
    },
    {
      "entropy": 5.553385639190674,
      "epoch": 0.8348570317058938,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004935679126207046,
      "loss": 5.4731,
      "mean_token_accuracy": 0.1702171102166176,
      "num_tokens": 18621353.0,
      "step": 10730
    },
    {
      "entropy": 5.4345306873321535,
      "epoch": 0.8352460610776113,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004935613313911463,
      "loss": 5.3479,
      "mean_token_accuracy": 0.17120095640420913,
      "num_tokens": 18630152.0,
      "step": 10735
    },
    {
      "entropy": 5.567927503585816,
      "epoch": 0.8356350904493289,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004935547468452678,
      "loss": 5.3876,
      "mean_token_accuracy": 0.1725356921553612,
      "num_tokens": 18639278.0,
      "step": 10740
    },
    {
      "entropy": 5.6095947265625,
      "epoch": 0.8360241198210465,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004935481589831688,
      "loss": 5.3876,
      "mean_token_accuracy": 0.17071651369333268,
      "num_tokens": 18647436.0,
      "step": 10745
    },
    {
      "entropy": 5.587734508514404,
      "epoch": 0.8364131491927641,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004935415678049492,
      "loss": 5.3894,
      "mean_token_accuracy": 0.16289528012275695,
      "num_tokens": 18655532.0,
      "step": 10750
    },
    {
      "entropy": 5.587741994857788,
      "epoch": 0.8368021785644816,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004935349733107094,
      "loss": 5.4436,
      "mean_token_accuracy": 0.1677956759929657,
      "num_tokens": 18664827.0,
      "step": 10755
    },
    {
      "entropy": 5.563502550125122,
      "epoch": 0.8371912079361992,
      "grad_norm": 1.328125,
      "learning_rate": 0.000493528375500549,
      "loss": 5.3711,
      "mean_token_accuracy": 0.17083900272846222,
      "num_tokens": 18673535.0,
      "step": 10760
    },
    {
      "entropy": 5.554038095474243,
      "epoch": 0.8375802373079168,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004935217743745685,
      "loss": 5.5188,
      "mean_token_accuracy": 0.15876322388648986,
      "num_tokens": 18682223.0,
      "step": 10765
    },
    {
      "entropy": 5.62978515625,
      "epoch": 0.8379692666796343,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004935151699328677,
      "loss": 5.422,
      "mean_token_accuracy": 0.17131631821393967,
      "num_tokens": 18690847.0,
      "step": 10770
    },
    {
      "entropy": 5.641382503509521,
      "epoch": 0.8383582960513519,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004935085621755471,
      "loss": 5.4442,
      "mean_token_accuracy": 0.16079427897930146,
      "num_tokens": 18699658.0,
      "step": 10775
    },
    {
      "entropy": 5.578688716888427,
      "epoch": 0.8387473254230694,
      "grad_norm": 1.21875,
      "learning_rate": 0.000493501951102707,
      "loss": 5.412,
      "mean_token_accuracy": 0.16344024389982223,
      "num_tokens": 18708396.0,
      "step": 10780
    },
    {
      "entropy": 5.541438055038452,
      "epoch": 0.839136354794787,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004934953367144474,
      "loss": 5.42,
      "mean_token_accuracy": 0.16848313212394714,
      "num_tokens": 18717185.0,
      "step": 10785
    },
    {
      "entropy": 5.600802850723267,
      "epoch": 0.8395253841665046,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004934887190108688,
      "loss": 5.365,
      "mean_token_accuracy": 0.16976140439510345,
      "num_tokens": 18725493.0,
      "step": 10790
    },
    {
      "entropy": 5.554530143737793,
      "epoch": 0.8399144135382222,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004934820979920719,
      "loss": 5.3434,
      "mean_token_accuracy": 0.1717671722173691,
      "num_tokens": 18733843.0,
      "step": 10795
    },
    {
      "entropy": 5.53960862159729,
      "epoch": 0.8403034429099397,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004934754736581567,
      "loss": 5.4197,
      "mean_token_accuracy": 0.16596496850252151,
      "num_tokens": 18742226.0,
      "step": 10800
    },
    {
      "entropy": 5.63180022239685,
      "epoch": 0.8406924722816572,
      "grad_norm": 1.296875,
      "learning_rate": 0.000493468846009224,
      "loss": 5.5536,
      "mean_token_accuracy": 0.16412430852651597,
      "num_tokens": 18751117.0,
      "step": 10805
    },
    {
      "entropy": 5.557992267608642,
      "epoch": 0.8410815016533748,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004934622150453742,
      "loss": 5.3745,
      "mean_token_accuracy": 0.16800056993961335,
      "num_tokens": 18759244.0,
      "step": 10810
    },
    {
      "entropy": 5.551813411712646,
      "epoch": 0.8414705310250924,
      "grad_norm": 1.296875,
      "learning_rate": 0.000493455580766708,
      "loss": 5.4475,
      "mean_token_accuracy": 0.1644230604171753,
      "num_tokens": 18768107.0,
      "step": 10815
    },
    {
      "entropy": 5.515730476379394,
      "epoch": 0.84185956039681,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004934489431733262,
      "loss": 5.2801,
      "mean_token_accuracy": 0.17706637233495712,
      "num_tokens": 18777419.0,
      "step": 10820
    },
    {
      "entropy": 5.55545563697815,
      "epoch": 0.8422485897685276,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004934423022653293,
      "loss": 5.4116,
      "mean_token_accuracy": 0.1688419684767723,
      "num_tokens": 18786410.0,
      "step": 10825
    },
    {
      "entropy": 5.538550615310669,
      "epoch": 0.842637619140245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004934356580428182,
      "loss": 5.3669,
      "mean_token_accuracy": 0.16694443672895432,
      "num_tokens": 18794409.0,
      "step": 10830
    },
    {
      "entropy": 5.553793239593506,
      "epoch": 0.8430266485119626,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004934290105058937,
      "loss": 5.4398,
      "mean_token_accuracy": 0.16828235983848572,
      "num_tokens": 18803561.0,
      "step": 10835
    },
    {
      "entropy": 5.571388149261475,
      "epoch": 0.8434156778836802,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004934223596546565,
      "loss": 5.3971,
      "mean_token_accuracy": 0.1636173039674759,
      "num_tokens": 18812096.0,
      "step": 10840
    },
    {
      "entropy": 5.530799531936646,
      "epoch": 0.8438047072553978,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004934157054892077,
      "loss": 5.4108,
      "mean_token_accuracy": 0.16458137482404708,
      "num_tokens": 18820506.0,
      "step": 10845
    },
    {
      "entropy": 5.546833610534668,
      "epoch": 0.8441937366271154,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004934090480096482,
      "loss": 5.4045,
      "mean_token_accuracy": 0.1651388466358185,
      "num_tokens": 18829437.0,
      "step": 10850
    },
    {
      "entropy": 5.665205097198486,
      "epoch": 0.8445827659988329,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004934023872160791,
      "loss": 5.5215,
      "mean_token_accuracy": 0.16349208503961563,
      "num_tokens": 18838759.0,
      "step": 10855
    },
    {
      "entropy": 5.522897720336914,
      "epoch": 0.8449717953705504,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004933957231086014,
      "loss": 5.4199,
      "mean_token_accuracy": 0.16991969496011733,
      "num_tokens": 18847000.0,
      "step": 10860
    },
    {
      "entropy": 5.602512788772583,
      "epoch": 0.845360824742268,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004933890556873161,
      "loss": 5.4879,
      "mean_token_accuracy": 0.15818663984537124,
      "num_tokens": 18855370.0,
      "step": 10865
    },
    {
      "entropy": 5.685675048828125,
      "epoch": 0.8457498541139856,
      "grad_norm": 1.25,
      "learning_rate": 0.0004933823849523246,
      "loss": 5.5571,
      "mean_token_accuracy": 0.16334884613752365,
      "num_tokens": 18863825.0,
      "step": 10870
    },
    {
      "entropy": 5.57058367729187,
      "epoch": 0.8461388834857032,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004933757109037279,
      "loss": 5.2937,
      "mean_token_accuracy": 0.17143325805664061,
      "num_tokens": 18872088.0,
      "step": 10875
    },
    {
      "entropy": 5.523991203308105,
      "epoch": 0.8465279128574207,
      "grad_norm": 1.25,
      "learning_rate": 0.0004933690335416274,
      "loss": 5.2996,
      "mean_token_accuracy": 0.17823541462421416,
      "num_tokens": 18880472.0,
      "step": 10880
    },
    {
      "entropy": 5.619044256210327,
      "epoch": 0.8469169422291383,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004933623528661245,
      "loss": 5.5106,
      "mean_token_accuracy": 0.16327719390392303,
      "num_tokens": 18889822.0,
      "step": 10885
    },
    {
      "entropy": 5.649898481369019,
      "epoch": 0.8473059716008559,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004933556688773203,
      "loss": 5.4188,
      "mean_token_accuracy": 0.16944485604763032,
      "num_tokens": 18897542.0,
      "step": 10890
    },
    {
      "entropy": 5.496301126480103,
      "epoch": 0.8476950009725734,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004933489815753165,
      "loss": 5.4865,
      "mean_token_accuracy": 0.1656087353825569,
      "num_tokens": 18906330.0,
      "step": 10895
    },
    {
      "entropy": 5.548855400085449,
      "epoch": 0.848084030344291,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004933422909602143,
      "loss": 5.3219,
      "mean_token_accuracy": 0.17305444926023483,
      "num_tokens": 18914176.0,
      "step": 10900
    },
    {
      "entropy": 5.4994221210479735,
      "epoch": 0.8484730597160085,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004933355970321155,
      "loss": 5.3393,
      "mean_token_accuracy": 0.17536135613918305,
      "num_tokens": 18922233.0,
      "step": 10905
    },
    {
      "entropy": 5.430251693725586,
      "epoch": 0.8488620890877261,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004933288997911215,
      "loss": 5.3624,
      "mean_token_accuracy": 0.17554630041122438,
      "num_tokens": 18930742.0,
      "step": 10910
    },
    {
      "entropy": 5.57297649383545,
      "epoch": 0.8492511184594437,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004933221992373338,
      "loss": 5.3294,
      "mean_token_accuracy": 0.17314134538173676,
      "num_tokens": 18938879.0,
      "step": 10915
    },
    {
      "entropy": 5.689609527587891,
      "epoch": 0.8496401478311613,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004933154953708544,
      "loss": 5.4933,
      "mean_token_accuracy": 0.15841685086488724,
      "num_tokens": 18947753.0,
      "step": 10920
    },
    {
      "entropy": 5.560961771011352,
      "epoch": 0.8500291772028788,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004933087881917848,
      "loss": 5.3961,
      "mean_token_accuracy": 0.16858956962823868,
      "num_tokens": 18956805.0,
      "step": 10925
    },
    {
      "entropy": 5.437729835510254,
      "epoch": 0.8504182065745964,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004933020777002268,
      "loss": 5.3605,
      "mean_token_accuracy": 0.17528738975524902,
      "num_tokens": 18965683.0,
      "step": 10930
    },
    {
      "entropy": 5.5141282081604,
      "epoch": 0.8508072359463139,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004932953638962823,
      "loss": 5.3449,
      "mean_token_accuracy": 0.1721363067626953,
      "num_tokens": 18973873.0,
      "step": 10935
    },
    {
      "entropy": 5.495585775375366,
      "epoch": 0.8511962653180315,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000493288646780053,
      "loss": 5.2946,
      "mean_token_accuracy": 0.1793428361415863,
      "num_tokens": 18982097.0,
      "step": 10940
    },
    {
      "entropy": 5.529202127456665,
      "epoch": 0.8515852946897491,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004932819263516409,
      "loss": 5.3599,
      "mean_token_accuracy": 0.17138735949993134,
      "num_tokens": 18991019.0,
      "step": 10945
    },
    {
      "entropy": 5.5642682075500485,
      "epoch": 0.8519743240614667,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004932752026111481,
      "loss": 5.3445,
      "mean_token_accuracy": 0.16889835894107819,
      "num_tokens": 18999231.0,
      "step": 10950
    },
    {
      "entropy": 5.596057987213134,
      "epoch": 0.8523633534331843,
      "grad_norm": 1.25,
      "learning_rate": 0.0004932684755586765,
      "loss": 5.4485,
      "mean_token_accuracy": 0.16654932051897048,
      "num_tokens": 19007807.0,
      "step": 10955
    },
    {
      "entropy": 5.543960905075073,
      "epoch": 0.8527523828049017,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004932617451943282,
      "loss": 5.4108,
      "mean_token_accuracy": 0.16886223256587982,
      "num_tokens": 19017908.0,
      "step": 10960
    },
    {
      "entropy": 5.482617568969727,
      "epoch": 0.8531414121766193,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004932550115182053,
      "loss": 5.4292,
      "mean_token_accuracy": 0.164827199280262,
      "num_tokens": 19026999.0,
      "step": 10965
    },
    {
      "entropy": 5.436986827850342,
      "epoch": 0.8535304415483369,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00049324827453041,
      "loss": 5.3229,
      "mean_token_accuracy": 0.17810767143964767,
      "num_tokens": 19035738.0,
      "step": 10970
    },
    {
      "entropy": 5.577497577667236,
      "epoch": 0.8539194709200545,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004932415342310446,
      "loss": 5.4117,
      "mean_token_accuracy": 0.16914543360471726,
      "num_tokens": 19044530.0,
      "step": 10975
    },
    {
      "entropy": 5.5313318252563475,
      "epoch": 0.8543085002917721,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004932347906202111,
      "loss": 5.4076,
      "mean_token_accuracy": 0.1639050602912903,
      "num_tokens": 19053306.0,
      "step": 10980
    },
    {
      "entropy": 5.5989118099212645,
      "epoch": 0.8546975296634896,
      "grad_norm": 1.375,
      "learning_rate": 0.000493228043698012,
      "loss": 5.401,
      "mean_token_accuracy": 0.16864163875579835,
      "num_tokens": 19062172.0,
      "step": 10985
    },
    {
      "entropy": 5.405817937850952,
      "epoch": 0.8550865590352071,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004932212934645498,
      "loss": 5.192,
      "mean_token_accuracy": 0.18007719665765762,
      "num_tokens": 19070804.0,
      "step": 10990
    },
    {
      "entropy": 5.495610094070434,
      "epoch": 0.8554755884069247,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004932145399199268,
      "loss": 5.3567,
      "mean_token_accuracy": 0.17602065652608873,
      "num_tokens": 19079055.0,
      "step": 10995
    },
    {
      "entropy": 5.50682315826416,
      "epoch": 0.8558646177786423,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004932077830642455,
      "loss": 5.3274,
      "mean_token_accuracy": 0.17099796384572982,
      "num_tokens": 19087459.0,
      "step": 11000
    },
    {
      "entropy": 5.424697017669677,
      "epoch": 0.8562536471503599,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004932010228976084,
      "loss": 5.2328,
      "mean_token_accuracy": 0.17927884683012962,
      "num_tokens": 19096271.0,
      "step": 11005
    },
    {
      "entropy": 5.582862997055054,
      "epoch": 0.8566426765220774,
      "grad_norm": 1.390625,
      "learning_rate": 0.000493194259420118,
      "loss": 5.4288,
      "mean_token_accuracy": 0.16449785083532334,
      "num_tokens": 19105257.0,
      "step": 11010
    },
    {
      "entropy": 5.569941663742066,
      "epoch": 0.857031705893795,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000493187492631877,
      "loss": 5.3768,
      "mean_token_accuracy": 0.17165541350841523,
      "num_tokens": 19113673.0,
      "step": 11015
    },
    {
      "entropy": 5.510107517242432,
      "epoch": 0.8574207352655125,
      "grad_norm": 1.25,
      "learning_rate": 0.0004931807225329882,
      "loss": 5.3308,
      "mean_token_accuracy": 0.1744338020682335,
      "num_tokens": 19121635.0,
      "step": 11020
    },
    {
      "entropy": 5.548259592056274,
      "epoch": 0.8578097646372301,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004931739491235541,
      "loss": 5.3932,
      "mean_token_accuracy": 0.17344855666160583,
      "num_tokens": 19130097.0,
      "step": 11025
    },
    {
      "entropy": 5.436193227767944,
      "epoch": 0.8581987940089477,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004931671724036777,
      "loss": 5.3109,
      "mean_token_accuracy": 0.17121241092681885,
      "num_tokens": 19138349.0,
      "step": 11030
    },
    {
      "entropy": 5.456762456893921,
      "epoch": 0.8585878233806652,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004931603923734616,
      "loss": 5.4029,
      "mean_token_accuracy": 0.17093536406755447,
      "num_tokens": 19146948.0,
      "step": 11035
    },
    {
      "entropy": 5.509184789657593,
      "epoch": 0.8589768527523828,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004931536090330088,
      "loss": 5.3581,
      "mean_token_accuracy": 0.1710792824625969,
      "num_tokens": 19155372.0,
      "step": 11040
    },
    {
      "entropy": 5.641581678390503,
      "epoch": 0.8593658821241004,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004931468223824222,
      "loss": 5.4555,
      "mean_token_accuracy": 0.16285751909017562,
      "num_tokens": 19163634.0,
      "step": 11045
    },
    {
      "entropy": 5.6391383647918705,
      "epoch": 0.859754911495818,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004931400324218048,
      "loss": 5.4481,
      "mean_token_accuracy": 0.170964215695858,
      "num_tokens": 19171582.0,
      "step": 11050
    },
    {
      "entropy": 5.528740978240966,
      "epoch": 0.8601439408675355,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004931332391512597,
      "loss": 5.4434,
      "mean_token_accuracy": 0.16314370483160018,
      "num_tokens": 19181136.0,
      "step": 11055
    },
    {
      "entropy": 5.555797100067139,
      "epoch": 0.860532970239253,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004931264425708897,
      "loss": 5.4425,
      "mean_token_accuracy": 0.16152996569871902,
      "num_tokens": 19189389.0,
      "step": 11060
    },
    {
      "entropy": 5.663418102264404,
      "epoch": 0.8609219996109706,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004931196426807983,
      "loss": 5.4271,
      "mean_token_accuracy": 0.17165378034114837,
      "num_tokens": 19198315.0,
      "step": 11065
    },
    {
      "entropy": 5.694033479690551,
      "epoch": 0.8613110289826882,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004931128394810884,
      "loss": 5.4608,
      "mean_token_accuracy": 0.16384464651346206,
      "num_tokens": 19207234.0,
      "step": 11070
    },
    {
      "entropy": 5.503334283828735,
      "epoch": 0.8617000583544058,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004931060329718634,
      "loss": 5.4651,
      "mean_token_accuracy": 0.16109324246644974,
      "num_tokens": 19216552.0,
      "step": 11075
    },
    {
      "entropy": 5.5668315410614015,
      "epoch": 0.8620890877261234,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004930992231532265,
      "loss": 5.3919,
      "mean_token_accuracy": 0.1658367931842804,
      "num_tokens": 19225237.0,
      "step": 11080
    },
    {
      "entropy": 5.529390621185303,
      "epoch": 0.8624781170978408,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000493092410025281,
      "loss": 5.3149,
      "mean_token_accuracy": 0.1733350247144699,
      "num_tokens": 19233246.0,
      "step": 11085
    },
    {
      "entropy": 5.478016662597656,
      "epoch": 0.8628671464695584,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004930855935881302,
      "loss": 5.3349,
      "mean_token_accuracy": 0.17239552438259126,
      "num_tokens": 19241748.0,
      "step": 11090
    },
    {
      "entropy": 5.601720857620239,
      "epoch": 0.863256175841276,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004930787738418777,
      "loss": 5.4945,
      "mean_token_accuracy": 0.16450771689414978,
      "num_tokens": 19250259.0,
      "step": 11095
    },
    {
      "entropy": 5.65542631149292,
      "epoch": 0.8636452052129936,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004930719507866269,
      "loss": 5.5583,
      "mean_token_accuracy": 0.16429050713777543,
      "num_tokens": 19259266.0,
      "step": 11100
    },
    {
      "entropy": 5.6297859191894535,
      "epoch": 0.8640342345847112,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004930651244224814,
      "loss": 5.4075,
      "mean_token_accuracy": 0.16003363728523254,
      "num_tokens": 19268099.0,
      "step": 11105
    },
    {
      "entropy": 5.534098100662232,
      "epoch": 0.8644232639564287,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004930582947495448,
      "loss": 5.2704,
      "mean_token_accuracy": 0.1776770383119583,
      "num_tokens": 19276894.0,
      "step": 11110
    },
    {
      "entropy": 5.452915239334106,
      "epoch": 0.8648122933281462,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004930514617679206,
      "loss": 5.363,
      "mean_token_accuracy": 0.1748938277363777,
      "num_tokens": 19285724.0,
      "step": 11115
    },
    {
      "entropy": 5.484914970397949,
      "epoch": 0.8652013226998638,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004930446254777125,
      "loss": 5.2952,
      "mean_token_accuracy": 0.17718063294887543,
      "num_tokens": 19294468.0,
      "step": 11120
    },
    {
      "entropy": 5.577092695236206,
      "epoch": 0.8655903520715814,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004930377858790242,
      "loss": 5.3608,
      "mean_token_accuracy": 0.17149281352758408,
      "num_tokens": 19303575.0,
      "step": 11125
    },
    {
      "entropy": 5.617782783508301,
      "epoch": 0.865979381443299,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004930309429719595,
      "loss": 5.5136,
      "mean_token_accuracy": 0.17120336145162582,
      "num_tokens": 19312778.0,
      "step": 11130
    },
    {
      "entropy": 5.601925373077393,
      "epoch": 0.8663684108150166,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004930240967566224,
      "loss": 5.3625,
      "mean_token_accuracy": 0.16819167584180833,
      "num_tokens": 19320725.0,
      "step": 11135
    },
    {
      "entropy": 5.562879848480224,
      "epoch": 0.8667574401867341,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004930172472331167,
      "loss": 5.3996,
      "mean_token_accuracy": 0.17437642961740493,
      "num_tokens": 19329842.0,
      "step": 11140
    },
    {
      "entropy": 5.5315632820129395,
      "epoch": 0.8671464695584516,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004930103944015463,
      "loss": 5.3952,
      "mean_token_accuracy": 0.16794670075178147,
      "num_tokens": 19338136.0,
      "step": 11145
    },
    {
      "entropy": 5.549357748031616,
      "epoch": 0.8675354989301692,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004930035382620149,
      "loss": 5.349,
      "mean_token_accuracy": 0.17086559534072876,
      "num_tokens": 19346597.0,
      "step": 11150
    },
    {
      "entropy": 5.547843408584595,
      "epoch": 0.8679245283018868,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000492996678814627,
      "loss": 5.3651,
      "mean_token_accuracy": 0.17244310677051544,
      "num_tokens": 19355595.0,
      "step": 11155
    },
    {
      "entropy": 5.536035585403442,
      "epoch": 0.8683135576736044,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004929898160594865,
      "loss": 5.4214,
      "mean_token_accuracy": 0.1709297627210617,
      "num_tokens": 19364133.0,
      "step": 11160
    },
    {
      "entropy": 5.614302682876587,
      "epoch": 0.8687025870453219,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004929829499966974,
      "loss": 5.4543,
      "mean_token_accuracy": 0.16739195585250854,
      "num_tokens": 19373628.0,
      "step": 11165
    },
    {
      "entropy": 5.555817890167236,
      "epoch": 0.8690916164170395,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004929760806263641,
      "loss": 5.3998,
      "mean_token_accuracy": 0.16885512918233872,
      "num_tokens": 19381901.0,
      "step": 11170
    },
    {
      "entropy": 5.551951885223389,
      "epoch": 0.8694806457887571,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004929692079485906,
      "loss": 5.4799,
      "mean_token_accuracy": 0.1672930136322975,
      "num_tokens": 19390390.0,
      "step": 11175
    },
    {
      "entropy": 5.576341533660889,
      "epoch": 0.8698696751604746,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004929623319634814,
      "loss": 5.439,
      "mean_token_accuracy": 0.1618825003504753,
      "num_tokens": 19399858.0,
      "step": 11180
    },
    {
      "entropy": 5.553104639053345,
      "epoch": 0.8702587045321922,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004929554526711407,
      "loss": 5.3939,
      "mean_token_accuracy": 0.16306375116109847,
      "num_tokens": 19409062.0,
      "step": 11185
    },
    {
      "entropy": 5.508029890060425,
      "epoch": 0.8706477339039097,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004929485700716729,
      "loss": 5.3775,
      "mean_token_accuracy": 0.17496417462825775,
      "num_tokens": 19417228.0,
      "step": 11190
    },
    {
      "entropy": 5.614364433288574,
      "epoch": 0.8710367632756273,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004929416841651824,
      "loss": 5.4779,
      "mean_token_accuracy": 0.16553594321012496,
      "num_tokens": 19426086.0,
      "step": 11195
    },
    {
      "entropy": 5.723991727828979,
      "epoch": 0.8714257926473449,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004929347949517739,
      "loss": 5.4621,
      "mean_token_accuracy": 0.16419120579957963,
      "num_tokens": 19435363.0,
      "step": 11200
    },
    {
      "entropy": 5.492606258392334,
      "epoch": 0.8718148220190625,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004929279024315516,
      "loss": 5.2708,
      "mean_token_accuracy": 0.17490680515766144,
      "num_tokens": 19443355.0,
      "step": 11205
    },
    {
      "entropy": 5.451952409744263,
      "epoch": 0.87220385139078,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004929210066046204,
      "loss": 5.3101,
      "mean_token_accuracy": 0.17080140113830566,
      "num_tokens": 19452217.0,
      "step": 11210
    },
    {
      "entropy": 5.5880396366119385,
      "epoch": 0.8725928807624975,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004929141074710847,
      "loss": 5.5206,
      "mean_token_accuracy": 0.16174181401729584,
      "num_tokens": 19460034.0,
      "step": 11215
    },
    {
      "entropy": 5.603846073150635,
      "epoch": 0.8729819101342151,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004929072050310491,
      "loss": 5.445,
      "mean_token_accuracy": 0.1657393604516983,
      "num_tokens": 19468403.0,
      "step": 11220
    },
    {
      "entropy": 5.577253723144532,
      "epoch": 0.8733709395059327,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004929002992846188,
      "loss": 5.4629,
      "mean_token_accuracy": 0.16801663786172866,
      "num_tokens": 19477551.0,
      "step": 11225
    },
    {
      "entropy": 5.6302289962768555,
      "epoch": 0.8737599688776503,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004928933902318981,
      "loss": 5.3521,
      "mean_token_accuracy": 0.1736658364534378,
      "num_tokens": 19486131.0,
      "step": 11230
    },
    {
      "entropy": 5.618899345397949,
      "epoch": 0.8741489982493679,
      "grad_norm": 1.21875,
      "learning_rate": 0.000492886477872992,
      "loss": 5.494,
      "mean_token_accuracy": 0.1574467435479164,
      "num_tokens": 19496212.0,
      "step": 11235
    },
    {
      "entropy": 5.564089441299439,
      "epoch": 0.8745380276210853,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004928795622080054,
      "loss": 5.3361,
      "mean_token_accuracy": 0.17063281536102295,
      "num_tokens": 19504556.0,
      "step": 11240
    },
    {
      "entropy": 5.569669103622436,
      "epoch": 0.8749270569928029,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004928726432370434,
      "loss": 5.3943,
      "mean_token_accuracy": 0.16652307659387589,
      "num_tokens": 19512832.0,
      "step": 11245
    },
    {
      "entropy": 5.585665845870972,
      "epoch": 0.8753160863645205,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004928657209602107,
      "loss": 5.4231,
      "mean_token_accuracy": 0.1680866375565529,
      "num_tokens": 19521410.0,
      "step": 11250
    },
    {
      "entropy": 5.537788772583008,
      "epoch": 0.8757051157362381,
      "grad_norm": 1.375,
      "learning_rate": 0.0004928587953776125,
      "loss": 5.3791,
      "mean_token_accuracy": 0.16848996579647063,
      "num_tokens": 19529935.0,
      "step": 11255
    },
    {
      "entropy": 5.483977890014648,
      "epoch": 0.8760941451079557,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004928518664893538,
      "loss": 5.2913,
      "mean_token_accuracy": 0.17647139877080917,
      "num_tokens": 19538330.0,
      "step": 11260
    },
    {
      "entropy": 5.49932370185852,
      "epoch": 0.8764831744796732,
      "grad_norm": 1.375,
      "learning_rate": 0.0004928449342955397,
      "loss": 5.3019,
      "mean_token_accuracy": 0.16853711307048796,
      "num_tokens": 19546860.0,
      "step": 11265
    },
    {
      "entropy": 5.57195553779602,
      "epoch": 0.8768722038513908,
      "grad_norm": 1.25,
      "learning_rate": 0.0004928379987962756,
      "loss": 5.3856,
      "mean_token_accuracy": 0.16860053837299346,
      "num_tokens": 19554945.0,
      "step": 11270
    },
    {
      "entropy": 5.582933473587036,
      "epoch": 0.8772612332231083,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004928310599916666,
      "loss": 5.4105,
      "mean_token_accuracy": 0.16590486466884613,
      "num_tokens": 19563182.0,
      "step": 11275
    },
    {
      "entropy": 5.511086559295654,
      "epoch": 0.8776502625948259,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004928241178818178,
      "loss": 5.3946,
      "mean_token_accuracy": 0.17098600119352342,
      "num_tokens": 19571405.0,
      "step": 11280
    },
    {
      "entropy": 5.641317462921142,
      "epoch": 0.8780392919665435,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004928171724668349,
      "loss": 5.4747,
      "mean_token_accuracy": 0.1652954787015915,
      "num_tokens": 19580053.0,
      "step": 11285
    },
    {
      "entropy": 5.625116157531738,
      "epoch": 0.878428321338261,
      "grad_norm": 1.375,
      "learning_rate": 0.0004928102237468229,
      "loss": 5.5013,
      "mean_token_accuracy": 0.16244151592254638,
      "num_tokens": 19588457.0,
      "step": 11290
    },
    {
      "entropy": 5.499196481704712,
      "epoch": 0.8788173507099786,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004928032717218876,
      "loss": 5.3317,
      "mean_token_accuracy": 0.16901074349880219,
      "num_tokens": 19597608.0,
      "step": 11295
    },
    {
      "entropy": 5.5090619087219235,
      "epoch": 0.8792063800816962,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004927963163921343,
      "loss": 5.3614,
      "mean_token_accuracy": 0.17164513617753982,
      "num_tokens": 19605904.0,
      "step": 11300
    },
    {
      "entropy": 5.5465319633483885,
      "epoch": 0.8795954094534137,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004927893577576685,
      "loss": 5.3499,
      "mean_token_accuracy": 0.1644853964447975,
      "num_tokens": 19614660.0,
      "step": 11305
    },
    {
      "entropy": 5.485580539703369,
      "epoch": 0.8799844388251313,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004927823958185959,
      "loss": 5.328,
      "mean_token_accuracy": 0.17479092329740525,
      "num_tokens": 19623210.0,
      "step": 11310
    },
    {
      "entropy": 5.541181421279907,
      "epoch": 0.8803734681968488,
      "grad_norm": 1.28125,
      "learning_rate": 0.000492775430575022,
      "loss": 5.4031,
      "mean_token_accuracy": 0.16457745730876921,
      "num_tokens": 19631747.0,
      "step": 11315
    },
    {
      "entropy": 5.602260255813599,
      "epoch": 0.8807624975685664,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004927684620270527,
      "loss": 5.4884,
      "mean_token_accuracy": 0.1574602872133255,
      "num_tokens": 19639902.0,
      "step": 11320
    },
    {
      "entropy": 5.613801002502441,
      "epoch": 0.881151526940284,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004927614901747935,
      "loss": 5.5253,
      "mean_token_accuracy": 0.16070694625377654,
      "num_tokens": 19648478.0,
      "step": 11325
    },
    {
      "entropy": 5.686625623703003,
      "epoch": 0.8815405563120016,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004927545150183503,
      "loss": 5.5166,
      "mean_token_accuracy": 0.16270480453968048,
      "num_tokens": 19657467.0,
      "step": 11330
    },
    {
      "entropy": 5.659714698791504,
      "epoch": 0.8819295856837192,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000492747536557829,
      "loss": 5.4751,
      "mean_token_accuracy": 0.16192390844225885,
      "num_tokens": 19665382.0,
      "step": 11335
    },
    {
      "entropy": 5.512677383422852,
      "epoch": 0.8823186150554367,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004927405547933353,
      "loss": 5.2979,
      "mean_token_accuracy": 0.17023497819900513,
      "num_tokens": 19674486.0,
      "step": 11340
    },
    {
      "entropy": 5.555436468124389,
      "epoch": 0.8827076444271542,
      "grad_norm": 1.25,
      "learning_rate": 0.0004927335697249753,
      "loss": 5.3065,
      "mean_token_accuracy": 0.16837829798460008,
      "num_tokens": 19683060.0,
      "step": 11345
    },
    {
      "entropy": 5.40867509841919,
      "epoch": 0.8830966737988718,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004927265813528549,
      "loss": 5.3004,
      "mean_token_accuracy": 0.1766051784157753,
      "num_tokens": 19691634.0,
      "step": 11350
    },
    {
      "entropy": 5.524748086929321,
      "epoch": 0.8834857031705894,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004927195896770804,
      "loss": 5.3916,
      "mean_token_accuracy": 0.16219370812177658,
      "num_tokens": 19699975.0,
      "step": 11355
    },
    {
      "entropy": 5.548884296417237,
      "epoch": 0.883874732542307,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004927125946977574,
      "loss": 5.3389,
      "mean_token_accuracy": 0.16684298664331437,
      "num_tokens": 19708431.0,
      "step": 11360
    },
    {
      "entropy": 5.52249436378479,
      "epoch": 0.8842637619140246,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004927055964149923,
      "loss": 5.3701,
      "mean_token_accuracy": 0.16658488810062408,
      "num_tokens": 19717189.0,
      "step": 11365
    },
    {
      "entropy": 5.568217086791992,
      "epoch": 0.884652791285742,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004926985948288914,
      "loss": 5.4253,
      "mean_token_accuracy": 0.17112406641244887,
      "num_tokens": 19725855.0,
      "step": 11370
    },
    {
      "entropy": 5.523195886611939,
      "epoch": 0.8850418206574596,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004926915899395608,
      "loss": 5.3462,
      "mean_token_accuracy": 0.17354673147201538,
      "num_tokens": 19734617.0,
      "step": 11375
    },
    {
      "entropy": 5.573224973678589,
      "epoch": 0.8854308500291772,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004926845817471068,
      "loss": 5.4678,
      "mean_token_accuracy": 0.16301082074642181,
      "num_tokens": 19743731.0,
      "step": 11380
    },
    {
      "entropy": 5.61586184501648,
      "epoch": 0.8858198794008948,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004926775702516358,
      "loss": 5.3967,
      "mean_token_accuracy": 0.16337064355611802,
      "num_tokens": 19752336.0,
      "step": 11385
    },
    {
      "entropy": 5.557158994674682,
      "epoch": 0.8862089087726124,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004926705554532541,
      "loss": 5.402,
      "mean_token_accuracy": 0.16458975076675414,
      "num_tokens": 19761488.0,
      "step": 11390
    },
    {
      "entropy": 5.464515256881714,
      "epoch": 0.8865979381443299,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004926635373520682,
      "loss": 5.3193,
      "mean_token_accuracy": 0.1764751359820366,
      "num_tokens": 19770330.0,
      "step": 11395
    },
    {
      "entropy": 5.530841588973999,
      "epoch": 0.8869869675160474,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004926565159481845,
      "loss": 5.3782,
      "mean_token_accuracy": 0.16833931654691697,
      "num_tokens": 19779037.0,
      "step": 11400
    },
    {
      "entropy": 5.586087989807129,
      "epoch": 0.887375996887765,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004926494912417097,
      "loss": 5.4436,
      "mean_token_accuracy": 0.1619071051478386,
      "num_tokens": 19787661.0,
      "step": 11405
    },
    {
      "entropy": 5.513906812667846,
      "epoch": 0.8877650262594826,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004926424632327503,
      "loss": 5.3556,
      "mean_token_accuracy": 0.1696711912751198,
      "num_tokens": 19796421.0,
      "step": 11410
    },
    {
      "entropy": 5.519581031799317,
      "epoch": 0.8881540556312002,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004926354319214129,
      "loss": 5.3454,
      "mean_token_accuracy": 0.17187682837247847,
      "num_tokens": 19805842.0,
      "step": 11415
    },
    {
      "entropy": 5.618935060501099,
      "epoch": 0.8885430850029177,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004926283973078041,
      "loss": 5.3802,
      "mean_token_accuracy": 0.1696264401078224,
      "num_tokens": 19814179.0,
      "step": 11420
    },
    {
      "entropy": 5.5856598854064945,
      "epoch": 0.8889321143746353,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004926213593920309,
      "loss": 5.458,
      "mean_token_accuracy": 0.1685087651014328,
      "num_tokens": 19822435.0,
      "step": 11425
    },
    {
      "entropy": 5.538632392883301,
      "epoch": 0.8893211437463528,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004926143181742,
      "loss": 5.3591,
      "mean_token_accuracy": 0.16798400282859802,
      "num_tokens": 19830300.0,
      "step": 11430
    },
    {
      "entropy": 5.531036853790283,
      "epoch": 0.8897101731180704,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004926072736544181,
      "loss": 5.4745,
      "mean_token_accuracy": 0.1687427669763565,
      "num_tokens": 19839264.0,
      "step": 11435
    },
    {
      "entropy": 5.559288549423218,
      "epoch": 0.890099202489788,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004926002258327922,
      "loss": 5.4191,
      "mean_token_accuracy": 0.16954224556684494,
      "num_tokens": 19848086.0,
      "step": 11440
    },
    {
      "entropy": 5.573931074142456,
      "epoch": 0.8904882318615055,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004925931747094292,
      "loss": 5.3998,
      "mean_token_accuracy": 0.1604166254401207,
      "num_tokens": 19856571.0,
      "step": 11445
    },
    {
      "entropy": 5.5644079685211185,
      "epoch": 0.8908772612332231,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004925861202844361,
      "loss": 5.4169,
      "mean_token_accuracy": 0.1646263226866722,
      "num_tokens": 19865272.0,
      "step": 11450
    },
    {
      "entropy": 5.536616611480713,
      "epoch": 0.8912662906049407,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00049257906255792,
      "loss": 5.3489,
      "mean_token_accuracy": 0.16775120347738265,
      "num_tokens": 19874307.0,
      "step": 11455
    },
    {
      "entropy": 5.5801863193511965,
      "epoch": 0.8916553199766583,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000492572001529988,
      "loss": 5.4075,
      "mean_token_accuracy": 0.17191922515630723,
      "num_tokens": 19882851.0,
      "step": 11460
    },
    {
      "entropy": 5.56919493675232,
      "epoch": 0.8920443493483758,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004925649372007469,
      "loss": 5.3125,
      "mean_token_accuracy": 0.17952443957328795,
      "num_tokens": 19891276.0,
      "step": 11465
    },
    {
      "entropy": 5.5681274890899655,
      "epoch": 0.8924333787200933,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004925578695703045,
      "loss": 5.47,
      "mean_token_accuracy": 0.1574510157108307,
      "num_tokens": 19900551.0,
      "step": 11470
    },
    {
      "entropy": 5.608745765686035,
      "epoch": 0.8928224080918109,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004925507986387676,
      "loss": 5.4803,
      "mean_token_accuracy": 0.1625505656003952,
      "num_tokens": 19909412.0,
      "step": 11475
    },
    {
      "entropy": 5.597529935836792,
      "epoch": 0.8932114374635285,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004925437244062436,
      "loss": 5.3124,
      "mean_token_accuracy": 0.1728732854127884,
      "num_tokens": 19918073.0,
      "step": 11480
    },
    {
      "entropy": 5.475112581253052,
      "epoch": 0.8936004668352461,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004925366468728397,
      "loss": 5.2666,
      "mean_token_accuracy": 0.18415023535490035,
      "num_tokens": 19926040.0,
      "step": 11485
    },
    {
      "entropy": 5.50698184967041,
      "epoch": 0.8939894962069637,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004925295660386635,
      "loss": 5.4267,
      "mean_token_accuracy": 0.17073747366666794,
      "num_tokens": 19934543.0,
      "step": 11490
    },
    {
      "entropy": 5.562829923629761,
      "epoch": 0.8943785255786811,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004925224819038224,
      "loss": 5.356,
      "mean_token_accuracy": 0.17162674367427827,
      "num_tokens": 19943198.0,
      "step": 11495
    },
    {
      "entropy": 5.535092496871949,
      "epoch": 0.8947675549503987,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004925153944684239,
      "loss": 5.3022,
      "mean_token_accuracy": 0.1760891482234001,
      "num_tokens": 19952319.0,
      "step": 11500
    },
    {
      "entropy": 5.526117277145386,
      "epoch": 0.8951565843221163,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004925083037325754,
      "loss": 5.2957,
      "mean_token_accuracy": 0.17167117744684218,
      "num_tokens": 19961982.0,
      "step": 11505
    },
    {
      "entropy": 5.534358739852905,
      "epoch": 0.8955456136938339,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004925012096963847,
      "loss": 5.3189,
      "mean_token_accuracy": 0.1667298972606659,
      "num_tokens": 19971055.0,
      "step": 11510
    },
    {
      "entropy": 5.439435625076294,
      "epoch": 0.8959346430655515,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004924941123599593,
      "loss": 5.2064,
      "mean_token_accuracy": 0.18462741672992705,
      "num_tokens": 19979850.0,
      "step": 11515
    },
    {
      "entropy": 5.56122579574585,
      "epoch": 0.896323672437269,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004924870117234069,
      "loss": 5.4176,
      "mean_token_accuracy": 0.166418719291687,
      "num_tokens": 19989261.0,
      "step": 11520
    },
    {
      "entropy": 5.522904396057129,
      "epoch": 0.8967127018089865,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004924799077868353,
      "loss": 5.3266,
      "mean_token_accuracy": 0.16648097783327104,
      "num_tokens": 19998321.0,
      "step": 11525
    },
    {
      "entropy": 5.445232963562011,
      "epoch": 0.8971017311807041,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004924728005503522,
      "loss": 5.395,
      "mean_token_accuracy": 0.16556968688964843,
      "num_tokens": 20007191.0,
      "step": 11530
    },
    {
      "entropy": 5.602244853973389,
      "epoch": 0.8974907605524217,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004924656900140655,
      "loss": 5.4426,
      "mean_token_accuracy": 0.16830357313156127,
      "num_tokens": 20015805.0,
      "step": 11535
    },
    {
      "entropy": 5.591095876693726,
      "epoch": 0.8978797899241393,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004924585761780831,
      "loss": 5.4042,
      "mean_token_accuracy": 0.1739838257431984,
      "num_tokens": 20024861.0,
      "step": 11540
    },
    {
      "entropy": 5.653625917434693,
      "epoch": 0.8982688192958569,
      "grad_norm": 1.28125,
      "learning_rate": 0.000492451459042513,
      "loss": 5.4575,
      "mean_token_accuracy": 0.16722671389579774,
      "num_tokens": 20034208.0,
      "step": 11545
    },
    {
      "entropy": 5.556991291046143,
      "epoch": 0.8986578486675744,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000492444338607463,
      "loss": 5.4,
      "mean_token_accuracy": 0.1693068578839302,
      "num_tokens": 20042786.0,
      "step": 11550
    },
    {
      "entropy": 5.60860595703125,
      "epoch": 0.899046878039292,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004924372148730413,
      "loss": 5.4551,
      "mean_token_accuracy": 0.16533754020929337,
      "num_tokens": 20051697.0,
      "step": 11555
    },
    {
      "entropy": 5.589045476913452,
      "epoch": 0.8994359074110095,
      "grad_norm": 1.265625,
      "learning_rate": 0.000492430087839356,
      "loss": 5.4155,
      "mean_token_accuracy": 0.17480046600103377,
      "num_tokens": 20059814.0,
      "step": 11560
    },
    {
      "entropy": 5.56453046798706,
      "epoch": 0.8998249367827271,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004924229575065152,
      "loss": 5.4341,
      "mean_token_accuracy": 0.16726895570755004,
      "num_tokens": 20068329.0,
      "step": 11565
    },
    {
      "entropy": 5.614006471633911,
      "epoch": 0.9002139661544447,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000492415823874627,
      "loss": 5.4193,
      "mean_token_accuracy": 0.17226494401693343,
      "num_tokens": 20076925.0,
      "step": 11570
    },
    {
      "entropy": 5.538260889053345,
      "epoch": 0.9006029955261622,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004924086869437998,
      "loss": 5.4277,
      "mean_token_accuracy": 0.15983839482069015,
      "num_tokens": 20085248.0,
      "step": 11575
    },
    {
      "entropy": 5.618026542663574,
      "epoch": 0.9009920248978798,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004924015467141417,
      "loss": 5.4161,
      "mean_token_accuracy": 0.16668502241373062,
      "num_tokens": 20093929.0,
      "step": 11580
    },
    {
      "entropy": 5.588715696334839,
      "epoch": 0.9013810542695974,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004923944031857613,
      "loss": 5.494,
      "mean_token_accuracy": 0.16041961461305618,
      "num_tokens": 20103141.0,
      "step": 11585
    },
    {
      "entropy": 5.571083450317383,
      "epoch": 0.901770083641315,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004923872563587668,
      "loss": 5.3159,
      "mean_token_accuracy": 0.1764917105436325,
      "num_tokens": 20111938.0,
      "step": 11590
    },
    {
      "entropy": 5.546873760223389,
      "epoch": 0.9021591130130325,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004923801062332666,
      "loss": 5.3219,
      "mean_token_accuracy": 0.17802399694919585,
      "num_tokens": 20120795.0,
      "step": 11595
    },
    {
      "entropy": 5.517357110977173,
      "epoch": 0.90254814238475,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004923729528093694,
      "loss": 5.3938,
      "mean_token_accuracy": 0.17049430608749389,
      "num_tokens": 20129816.0,
      "step": 11600
    },
    {
      "entropy": 5.606090068817139,
      "epoch": 0.9029371717564676,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004923657960871836,
      "loss": 5.5066,
      "mean_token_accuracy": 0.16241107434034346,
      "num_tokens": 20138380.0,
      "step": 11605
    },
    {
      "entropy": 5.644651412963867,
      "epoch": 0.9033262011281852,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004923586360668178,
      "loss": 5.5022,
      "mean_token_accuracy": 0.16276281028985978,
      "num_tokens": 20147201.0,
      "step": 11610
    },
    {
      "entropy": 5.5776042461395265,
      "epoch": 0.9037152304999028,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004923514727483807,
      "loss": 5.3436,
      "mean_token_accuracy": 0.17245255708694457,
      "num_tokens": 20156744.0,
      "step": 11615
    },
    {
      "entropy": 5.590355110168457,
      "epoch": 0.9041042598716204,
      "grad_norm": 1.625,
      "learning_rate": 0.000492344306131981,
      "loss": 5.3773,
      "mean_token_accuracy": 0.16877984553575515,
      "num_tokens": 20165081.0,
      "step": 11620
    },
    {
      "entropy": 5.568700361251831,
      "epoch": 0.9044932892433378,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004923371362177272,
      "loss": 5.3909,
      "mean_token_accuracy": 0.16697261333465577,
      "num_tokens": 20172893.0,
      "step": 11625
    },
    {
      "entropy": 5.555344915390014,
      "epoch": 0.9048823186150554,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004923299630057284,
      "loss": 5.44,
      "mean_token_accuracy": 0.1675187438726425,
      "num_tokens": 20181351.0,
      "step": 11630
    },
    {
      "entropy": 5.569200420379639,
      "epoch": 0.905271347986773,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004923227864960934,
      "loss": 5.442,
      "mean_token_accuracy": 0.16097819805145264,
      "num_tokens": 20189650.0,
      "step": 11635
    },
    {
      "entropy": 5.598432350158691,
      "epoch": 0.9056603773584906,
      "grad_norm": 1.453125,
      "learning_rate": 0.000492315606688931,
      "loss": 5.402,
      "mean_token_accuracy": 0.1602231442928314,
      "num_tokens": 20198184.0,
      "step": 11640
    },
    {
      "entropy": 5.600244474411011,
      "epoch": 0.9060494067302082,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004923084235843501,
      "loss": 5.3442,
      "mean_token_accuracy": 0.17479297518730164,
      "num_tokens": 20206357.0,
      "step": 11645
    },
    {
      "entropy": 5.523150157928467,
      "epoch": 0.9064384361019256,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004923012371824598,
      "loss": 5.4419,
      "mean_token_accuracy": 0.17150543183088302,
      "num_tokens": 20214384.0,
      "step": 11650
    },
    {
      "entropy": 5.582962465286255,
      "epoch": 0.9068274654736432,
      "grad_norm": 1.265625,
      "learning_rate": 0.000492294047483369,
      "loss": 5.4238,
      "mean_token_accuracy": 0.16499166190624237,
      "num_tokens": 20224103.0,
      "step": 11655
    },
    {
      "entropy": 5.56129937171936,
      "epoch": 0.9072164948453608,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004922868544871869,
      "loss": 5.4383,
      "mean_token_accuracy": 0.1643027663230896,
      "num_tokens": 20233414.0,
      "step": 11660
    },
    {
      "entropy": 5.56975302696228,
      "epoch": 0.9076055242170784,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004922796581940227,
      "loss": 5.3092,
      "mean_token_accuracy": 0.17716476172208787,
      "num_tokens": 20240988.0,
      "step": 11665
    },
    {
      "entropy": 5.466279458999634,
      "epoch": 0.907994553588796,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004922724586039855,
      "loss": 5.319,
      "mean_token_accuracy": 0.1697101265192032,
      "num_tokens": 20250066.0,
      "step": 11670
    },
    {
      "entropy": 5.472824430465698,
      "epoch": 0.9083835829605135,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004922652557171846,
      "loss": 5.269,
      "mean_token_accuracy": 0.16988119035959243,
      "num_tokens": 20258435.0,
      "step": 11675
    },
    {
      "entropy": 5.592214155197143,
      "epoch": 0.9087726123322311,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004922580495337292,
      "loss": 5.4425,
      "mean_token_accuracy": 0.16795462518930435,
      "num_tokens": 20267345.0,
      "step": 11680
    },
    {
      "entropy": 5.608999443054199,
      "epoch": 0.9091616417039486,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004922508400537288,
      "loss": 5.4584,
      "mean_token_accuracy": 0.1666748747229576,
      "num_tokens": 20275644.0,
      "step": 11685
    },
    {
      "entropy": 5.549394178390503,
      "epoch": 0.9095506710756662,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004922436272772926,
      "loss": 5.4336,
      "mean_token_accuracy": 0.16977478563785553,
      "num_tokens": 20284459.0,
      "step": 11690
    },
    {
      "entropy": 5.5712381362915036,
      "epoch": 0.9099397004473838,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004922364112045301,
      "loss": 5.3371,
      "mean_token_accuracy": 0.16603994220495225,
      "num_tokens": 20293088.0,
      "step": 11695
    },
    {
      "entropy": 5.47578763961792,
      "epoch": 0.9103287298191013,
      "grad_norm": 1.53125,
      "learning_rate": 0.000492229191835551,
      "loss": 5.3484,
      "mean_token_accuracy": 0.1726537674665451,
      "num_tokens": 20301811.0,
      "step": 11700
    },
    {
      "entropy": 5.5942809104919435,
      "epoch": 0.9107177591908189,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004922219691704645,
      "loss": 5.4915,
      "mean_token_accuracy": 0.16389033794403077,
      "num_tokens": 20310365.0,
      "step": 11705
    },
    {
      "entropy": 5.642590856552124,
      "epoch": 0.9111067885625365,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004922147432093805,
      "loss": 5.4112,
      "mean_token_accuracy": 0.16921351701021195,
      "num_tokens": 20318891.0,
      "step": 11710
    },
    {
      "entropy": 5.540099239349365,
      "epoch": 0.911495817934254,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004922075139524083,
      "loss": 5.315,
      "mean_token_accuracy": 0.1759725794196129,
      "num_tokens": 20327313.0,
      "step": 11715
    },
    {
      "entropy": 5.4736607551574705,
      "epoch": 0.9118848473059716,
      "grad_norm": 1.46875,
      "learning_rate": 0.000492200281399658,
      "loss": 5.2862,
      "mean_token_accuracy": 0.17611408978700638,
      "num_tokens": 20335511.0,
      "step": 11720
    },
    {
      "entropy": 5.528197002410889,
      "epoch": 0.9122738766776891,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000492193045551239,
      "loss": 5.3036,
      "mean_token_accuracy": 0.18054099082946778,
      "num_tokens": 20343588.0,
      "step": 11725
    },
    {
      "entropy": 5.54834623336792,
      "epoch": 0.9126629060494067,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004921858064072612,
      "loss": 5.3845,
      "mean_token_accuracy": 0.1737608551979065,
      "num_tokens": 20352793.0,
      "step": 11730
    },
    {
      "entropy": 5.508636045455932,
      "epoch": 0.9130519354211243,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004921785639678347,
      "loss": 5.4504,
      "mean_token_accuracy": 0.1690346419811249,
      "num_tokens": 20362480.0,
      "step": 11735
    },
    {
      "entropy": 5.542393541336059,
      "epoch": 0.9134409647928419,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000492171318233069,
      "loss": 5.3154,
      "mean_token_accuracy": 0.17283430695533752,
      "num_tokens": 20371192.0,
      "step": 11740
    },
    {
      "entropy": 5.632233953475952,
      "epoch": 0.9138299941645595,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004921640692030742,
      "loss": 5.4537,
      "mean_token_accuracy": 0.16680894047021866,
      "num_tokens": 20379322.0,
      "step": 11745
    },
    {
      "entropy": 5.593679428100586,
      "epoch": 0.914219023536277,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004921568168779603,
      "loss": 5.3992,
      "mean_token_accuracy": 0.15893051996827126,
      "num_tokens": 20387976.0,
      "step": 11750
    },
    {
      "entropy": 5.579742097854615,
      "epoch": 0.9146080529079945,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004921495612578374,
      "loss": 5.4603,
      "mean_token_accuracy": 0.1649359032511711,
      "num_tokens": 20397398.0,
      "step": 11755
    },
    {
      "entropy": 5.602052497863769,
      "epoch": 0.9149970822797121,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004921423023428156,
      "loss": 5.373,
      "mean_token_accuracy": 0.16892519742250442,
      "num_tokens": 20406503.0,
      "step": 11760
    },
    {
      "entropy": 5.520591402053833,
      "epoch": 0.9153861116514297,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004921350401330049,
      "loss": 5.3145,
      "mean_token_accuracy": 0.17201488614082336,
      "num_tokens": 20415117.0,
      "step": 11765
    },
    {
      "entropy": 5.502015829086304,
      "epoch": 0.9157751410231473,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004921277746285155,
      "loss": 5.3811,
      "mean_token_accuracy": 0.17021616995334626,
      "num_tokens": 20423519.0,
      "step": 11770
    },
    {
      "entropy": 5.560381603240967,
      "epoch": 0.9161641703948649,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004921205058294579,
      "loss": 5.3174,
      "mean_token_accuracy": 0.17579615712165833,
      "num_tokens": 20431831.0,
      "step": 11775
    },
    {
      "entropy": 5.6258467674255375,
      "epoch": 0.9165531997665823,
      "grad_norm": 1.3125,
      "learning_rate": 0.000492113233735942,
      "loss": 5.4698,
      "mean_token_accuracy": 0.16392302215099336,
      "num_tokens": 20440927.0,
      "step": 11780
    },
    {
      "entropy": 5.530792999267578,
      "epoch": 0.9169422291382999,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004921059583480785,
      "loss": 5.3489,
      "mean_token_accuracy": 0.17423975765705108,
      "num_tokens": 20448974.0,
      "step": 11785
    },
    {
      "entropy": 5.475788021087647,
      "epoch": 0.9173312585100175,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004920986796659775,
      "loss": 5.2639,
      "mean_token_accuracy": 0.17941995859146118,
      "num_tokens": 20457955.0,
      "step": 11790
    },
    {
      "entropy": 5.505053901672364,
      "epoch": 0.9177202878817351,
      "grad_norm": 1.375,
      "learning_rate": 0.0004920913976897498,
      "loss": 5.384,
      "mean_token_accuracy": 0.16897741407155992,
      "num_tokens": 20465942.0,
      "step": 11795
    },
    {
      "entropy": 5.535892963409424,
      "epoch": 0.9181093172534527,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004920841124195055,
      "loss": 5.444,
      "mean_token_accuracy": 0.17190477550029754,
      "num_tokens": 20474726.0,
      "step": 11800
    },
    {
      "entropy": 5.537477588653564,
      "epoch": 0.9184983466251702,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004920768238553554,
      "loss": 5.3828,
      "mean_token_accuracy": 0.17053221315145492,
      "num_tokens": 20483442.0,
      "step": 11805
    },
    {
      "entropy": 5.4853599071502686,
      "epoch": 0.9188873759968877,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004920695319974099,
      "loss": 5.3195,
      "mean_token_accuracy": 0.1739925116300583,
      "num_tokens": 20492950.0,
      "step": 11810
    },
    {
      "entropy": 5.560052919387817,
      "epoch": 0.9192764053686053,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004920622368457798,
      "loss": 5.431,
      "mean_token_accuracy": 0.16575712710618973,
      "num_tokens": 20502035.0,
      "step": 11815
    },
    {
      "entropy": 5.573536443710327,
      "epoch": 0.9196654347403229,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004920549384005759,
      "loss": 5.3046,
      "mean_token_accuracy": 0.1737214669585228,
      "num_tokens": 20510140.0,
      "step": 11820
    },
    {
      "entropy": 5.544577598571777,
      "epoch": 0.9200544641120405,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004920476366619086,
      "loss": 5.4113,
      "mean_token_accuracy": 0.16313107013702394,
      "num_tokens": 20519391.0,
      "step": 11825
    },
    {
      "entropy": 5.583065843582153,
      "epoch": 0.920443493483758,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000492040331629889,
      "loss": 5.4024,
      "mean_token_accuracy": 0.1676782801747322,
      "num_tokens": 20527635.0,
      "step": 11830
    },
    {
      "entropy": 5.568051481246949,
      "epoch": 0.9208325228554756,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004920330233046277,
      "loss": 5.3691,
      "mean_token_accuracy": 0.1756194144487381,
      "num_tokens": 20536455.0,
      "step": 11835
    },
    {
      "entropy": 5.603593826293945,
      "epoch": 0.9212215522271932,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004920257116862359,
      "loss": 5.5,
      "mean_token_accuracy": 0.1592232897877693,
      "num_tokens": 20545887.0,
      "step": 11840
    },
    {
      "entropy": 5.5125456809997555,
      "epoch": 0.9216105815989107,
      "grad_norm": 1.375,
      "learning_rate": 0.0004920183967748241,
      "loss": 5.3626,
      "mean_token_accuracy": 0.17073224037885665,
      "num_tokens": 20554415.0,
      "step": 11845
    },
    {
      "entropy": 5.566658973693848,
      "epoch": 0.9219996109706283,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004920110785705037,
      "loss": 5.4575,
      "mean_token_accuracy": 0.16421971023082732,
      "num_tokens": 20563081.0,
      "step": 11850
    },
    {
      "entropy": 5.489716815948486,
      "epoch": 0.9223886403423458,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004920037570733857,
      "loss": 5.3108,
      "mean_token_accuracy": 0.17567056119441987,
      "num_tokens": 20571800.0,
      "step": 11855
    },
    {
      "entropy": 5.592560482025147,
      "epoch": 0.9227776697140634,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004919964322835809,
      "loss": 5.4366,
      "mean_token_accuracy": 0.16670750379562377,
      "num_tokens": 20580481.0,
      "step": 11860
    },
    {
      "entropy": 5.585255765914917,
      "epoch": 0.923166699085781,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004919891042012006,
      "loss": 5.3088,
      "mean_token_accuracy": 0.1732628896832466,
      "num_tokens": 20589575.0,
      "step": 11865
    },
    {
      "entropy": 5.516901159286499,
      "epoch": 0.9235557284574986,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004919817728263562,
      "loss": 5.4008,
      "mean_token_accuracy": 0.16324003487825395,
      "num_tokens": 20598277.0,
      "step": 11870
    },
    {
      "entropy": 5.596324253082275,
      "epoch": 0.9239447578292161,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004919744381591586,
      "loss": 5.5047,
      "mean_token_accuracy": 0.16900823265314102,
      "num_tokens": 20607013.0,
      "step": 11875
    },
    {
      "entropy": 5.608691072463989,
      "epoch": 0.9243337872009336,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004919671001997193,
      "loss": 5.4384,
      "mean_token_accuracy": 0.16529636532068254,
      "num_tokens": 20615605.0,
      "step": 11880
    },
    {
      "entropy": 5.582414054870606,
      "epoch": 0.9247228165726512,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004919597589481497,
      "loss": 5.406,
      "mean_token_accuracy": 0.16373192816972731,
      "num_tokens": 20624275.0,
      "step": 11885
    },
    {
      "entropy": 5.555140590667724,
      "epoch": 0.9251118459443688,
      "grad_norm": 1.28125,
      "learning_rate": 0.000491952414404561,
      "loss": 5.3491,
      "mean_token_accuracy": 0.17381212264299392,
      "num_tokens": 20632681.0,
      "step": 11890
    },
    {
      "entropy": 5.66051869392395,
      "epoch": 0.9255008753160864,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004919450665690646,
      "loss": 5.4217,
      "mean_token_accuracy": 0.16786015331745147,
      "num_tokens": 20641432.0,
      "step": 11895
    },
    {
      "entropy": 5.568078517913818,
      "epoch": 0.925889904687804,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004919377154417724,
      "loss": 5.4194,
      "mean_token_accuracy": 0.1611281454563141,
      "num_tokens": 20650120.0,
      "step": 11900
    },
    {
      "entropy": 5.499671697616577,
      "epoch": 0.9262789340595214,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004919303610227954,
      "loss": 5.3199,
      "mean_token_accuracy": 0.16958568543195723,
      "num_tokens": 20658188.0,
      "step": 11905
    },
    {
      "entropy": 5.479451417922974,
      "epoch": 0.926667963431239,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004919230033122457,
      "loss": 5.2538,
      "mean_token_accuracy": 0.1806713670492172,
      "num_tokens": 20666157.0,
      "step": 11910
    },
    {
      "entropy": 5.538461494445801,
      "epoch": 0.9270569928029566,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004919156423102345,
      "loss": 5.4447,
      "mean_token_accuracy": 0.1682381436228752,
      "num_tokens": 20674995.0,
      "step": 11915
    },
    {
      "entropy": 5.617767477035523,
      "epoch": 0.9274460221746742,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004919082780168736,
      "loss": 5.3866,
      "mean_token_accuracy": 0.16756703555583954,
      "num_tokens": 20683786.0,
      "step": 11920
    },
    {
      "entropy": 5.570526742935181,
      "epoch": 0.9278350515463918,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004919009104322751,
      "loss": 5.4284,
      "mean_token_accuracy": 0.1617375820875168,
      "num_tokens": 20692582.0,
      "step": 11925
    },
    {
      "entropy": 5.584376668930053,
      "epoch": 0.9282240809181093,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004918935395565503,
      "loss": 5.4404,
      "mean_token_accuracy": 0.16492366492748262,
      "num_tokens": 20700892.0,
      "step": 11930
    },
    {
      "entropy": 5.5462226390838625,
      "epoch": 0.9286131102898268,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004918861653898113,
      "loss": 5.2909,
      "mean_token_accuracy": 0.18054843097925186,
      "num_tokens": 20709521.0,
      "step": 11935
    },
    {
      "entropy": 5.518969392776489,
      "epoch": 0.9290021396615444,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004918787879321701,
      "loss": 5.4044,
      "mean_token_accuracy": 0.1651698648929596,
      "num_tokens": 20717614.0,
      "step": 11940
    },
    {
      "entropy": 5.5014808177948,
      "epoch": 0.929391169033262,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004918714071837384,
      "loss": 5.3352,
      "mean_token_accuracy": 0.1689456060528755,
      "num_tokens": 20726624.0,
      "step": 11945
    },
    {
      "entropy": 5.624595785140992,
      "epoch": 0.9297801984049796,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004918640231446282,
      "loss": 5.4374,
      "mean_token_accuracy": 0.16891327053308486,
      "num_tokens": 20734738.0,
      "step": 11950
    },
    {
      "entropy": 5.625055646896362,
      "epoch": 0.9301692277766972,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004918566358149517,
      "loss": 5.3599,
      "mean_token_accuracy": 0.16594317257404329,
      "num_tokens": 20742825.0,
      "step": 11955
    },
    {
      "entropy": 5.551378202438355,
      "epoch": 0.9305582571484147,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000491849245194821,
      "loss": 5.3455,
      "mean_token_accuracy": 0.16925047039985658,
      "num_tokens": 20752132.0,
      "step": 11960
    },
    {
      "entropy": 5.594038820266723,
      "epoch": 0.9309472865201323,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004918418512843479,
      "loss": 5.4598,
      "mean_token_accuracy": 0.16059156358242035,
      "num_tokens": 20761838.0,
      "step": 11965
    },
    {
      "entropy": 5.5459983348846436,
      "epoch": 0.9313363158918498,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004918344540836451,
      "loss": 5.3593,
      "mean_token_accuracy": 0.17362731993198394,
      "num_tokens": 20770405.0,
      "step": 11970
    },
    {
      "entropy": 5.6911742210388185,
      "epoch": 0.9317253452635674,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004918270535928244,
      "loss": 5.5397,
      "mean_token_accuracy": 0.15774936378002166,
      "num_tokens": 20780130.0,
      "step": 11975
    },
    {
      "entropy": 5.567420148849488,
      "epoch": 0.932114374635285,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004918196498119984,
      "loss": 5.3971,
      "mean_token_accuracy": 0.1754030779004097,
      "num_tokens": 20788852.0,
      "step": 11980
    },
    {
      "entropy": 5.562246704101563,
      "epoch": 0.9325034040070025,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004918122427412793,
      "loss": 5.3363,
      "mean_token_accuracy": 0.16773886531591414,
      "num_tokens": 20796944.0,
      "step": 11985
    },
    {
      "entropy": 5.4880071640014645,
      "epoch": 0.9328924333787201,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004918048323807795,
      "loss": 5.4519,
      "mean_token_accuracy": 0.16756176501512526,
      "num_tokens": 20806118.0,
      "step": 11990
    },
    {
      "entropy": 5.5744466304779055,
      "epoch": 0.9332814627504377,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004917974187306114,
      "loss": 5.2584,
      "mean_token_accuracy": 0.17390068769454955,
      "num_tokens": 20814643.0,
      "step": 11995
    },
    {
      "entropy": 5.569103717803955,
      "epoch": 0.9336704921221552,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004917900017908875,
      "loss": 5.3418,
      "mean_token_accuracy": 0.16698082238435746,
      "num_tokens": 20823315.0,
      "step": 12000
    },
    {
      "epoch": 0.9336704921221552,
      "eval_entropy": 5.387608994877717,
      "eval_loss": 5.422966003417969,
      "eval_mean_token_accuracy": 0.17522683174594075,
      "eval_num_tokens": 20823315.0,
      "eval_runtime": 28.5708,
      "eval_samples_per_second": 1454.561,
      "eval_steps_per_second": 181.829,
      "step": 12000
    },
    {
      "entropy": 5.524903011322022,
      "epoch": 0.9340595214938728,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004917825815617205,
      "loss": 5.3465,
      "mean_token_accuracy": 0.1774814635515213,
      "num_tokens": 20832811.0,
      "step": 12005
    },
    {
      "entropy": 5.5779187202453615,
      "epoch": 0.9344485508655903,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004917751580432229,
      "loss": 5.3971,
      "mean_token_accuracy": 0.16667075604200363,
      "num_tokens": 20841824.0,
      "step": 12010
    },
    {
      "entropy": 5.5357122898101805,
      "epoch": 0.9348375802373079,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004917677312355072,
      "loss": 5.3416,
      "mean_token_accuracy": 0.17953069657087325,
      "num_tokens": 20850220.0,
      "step": 12015
    },
    {
      "entropy": 5.440711498260498,
      "epoch": 0.9352266096090255,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004917603011386863,
      "loss": 5.2964,
      "mean_token_accuracy": 0.17552932798862458,
      "num_tokens": 20859193.0,
      "step": 12020
    },
    {
      "entropy": 5.461758279800415,
      "epoch": 0.9356156389807431,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004917528677528727,
      "loss": 5.3226,
      "mean_token_accuracy": 0.1681724488735199,
      "num_tokens": 20867610.0,
      "step": 12025
    },
    {
      "entropy": 5.5498861312866214,
      "epoch": 0.9360046683524607,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004917454310781795,
      "loss": 5.4147,
      "mean_token_accuracy": 0.16310275644063948,
      "num_tokens": 20876991.0,
      "step": 12030
    },
    {
      "entropy": 5.523698902130127,
      "epoch": 0.9363936977241781,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004917379911147193,
      "loss": 5.2549,
      "mean_token_accuracy": 0.17352638989686966,
      "num_tokens": 20885794.0,
      "step": 12035
    },
    {
      "entropy": 5.578832674026489,
      "epoch": 0.9367827270958957,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004917305478626049,
      "loss": 5.4643,
      "mean_token_accuracy": 0.16653108596801758,
      "num_tokens": 20893487.0,
      "step": 12040
    },
    {
      "entropy": 5.538563871383667,
      "epoch": 0.9371717564676133,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004917231013219495,
      "loss": 5.4062,
      "mean_token_accuracy": 0.16739182323217391,
      "num_tokens": 20902587.0,
      "step": 12045
    },
    {
      "entropy": 5.543763732910156,
      "epoch": 0.9375607858393309,
      "grad_norm": 1.484375,
      "learning_rate": 0.000491715651492866,
      "loss": 5.3521,
      "mean_token_accuracy": 0.17395893335342408,
      "num_tokens": 20911062.0,
      "step": 12050
    },
    {
      "entropy": 5.593991708755493,
      "epoch": 0.9379498152110485,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004917081983754675,
      "loss": 5.3835,
      "mean_token_accuracy": 0.16870767027139663,
      "num_tokens": 20919369.0,
      "step": 12055
    },
    {
      "entropy": 5.5714129447937015,
      "epoch": 0.938338844582766,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004917007419698669,
      "loss": 5.4345,
      "mean_token_accuracy": 0.16696349680423736,
      "num_tokens": 20927514.0,
      "step": 12060
    },
    {
      "entropy": 5.64288592338562,
      "epoch": 0.9387278739544835,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004916932822761776,
      "loss": 5.4782,
      "mean_token_accuracy": 0.16645288318395615,
      "num_tokens": 20935750.0,
      "step": 12065
    },
    {
      "entropy": 5.560992670059204,
      "epoch": 0.9391169033262011,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004916858192945126,
      "loss": 5.3684,
      "mean_token_accuracy": 0.16627855896949767,
      "num_tokens": 20944089.0,
      "step": 12070
    },
    {
      "entropy": 5.5578186988830565,
      "epoch": 0.9395059326979187,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004916783530249852,
      "loss": 5.4618,
      "mean_token_accuracy": 0.16043560951948166,
      "num_tokens": 20952704.0,
      "step": 12075
    },
    {
      "entropy": 5.596148920059204,
      "epoch": 0.9398949620696363,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004916708834677086,
      "loss": 5.3364,
      "mean_token_accuracy": 0.16760361194610596,
      "num_tokens": 20960741.0,
      "step": 12080
    },
    {
      "entropy": 5.565456533432007,
      "epoch": 0.9402839914413538,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004916634106227962,
      "loss": 5.3356,
      "mean_token_accuracy": 0.177216799557209,
      "num_tokens": 20968778.0,
      "step": 12085
    },
    {
      "entropy": 5.571436882019043,
      "epoch": 0.9406730208130714,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004916559344903614,
      "loss": 5.4902,
      "mean_token_accuracy": 0.16506722271442414,
      "num_tokens": 20977993.0,
      "step": 12090
    },
    {
      "entropy": 5.600058841705322,
      "epoch": 0.941062050184789,
      "grad_norm": 1.5,
      "learning_rate": 0.0004916484550705176,
      "loss": 5.3368,
      "mean_token_accuracy": 0.17609670907258987,
      "num_tokens": 20986974.0,
      "step": 12095
    },
    {
      "entropy": 5.60751895904541,
      "epoch": 0.9414510795565065,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004916409723633785,
      "loss": 5.446,
      "mean_token_accuracy": 0.16376518458127975,
      "num_tokens": 20995224.0,
      "step": 12100
    },
    {
      "entropy": 5.554337644577027,
      "epoch": 0.9418401089282241,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004916334863690573,
      "loss": 5.4512,
      "mean_token_accuracy": 0.16852151602506638,
      "num_tokens": 21003270.0,
      "step": 12105
    },
    {
      "entropy": 5.560426378250122,
      "epoch": 0.9422291382999416,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004916259970876678,
      "loss": 5.4009,
      "mean_token_accuracy": 0.17196376919746398,
      "num_tokens": 21012665.0,
      "step": 12110
    },
    {
      "entropy": 5.602419424057007,
      "epoch": 0.9426181676716592,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004916185045193237,
      "loss": 5.4315,
      "mean_token_accuracy": 0.16751835793256759,
      "num_tokens": 21021043.0,
      "step": 12115
    },
    {
      "entropy": 5.564066219329834,
      "epoch": 0.9430071970433768,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004916110086641384,
      "loss": 5.3654,
      "mean_token_accuracy": 0.1720741719007492,
      "num_tokens": 21029525.0,
      "step": 12120
    },
    {
      "entropy": 5.532789134979248,
      "epoch": 0.9433962264150944,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004916035095222259,
      "loss": 5.3445,
      "mean_token_accuracy": 0.17317475825548173,
      "num_tokens": 21037687.0,
      "step": 12125
    },
    {
      "entropy": 5.581482458114624,
      "epoch": 0.9437852557868119,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004915960070936999,
      "loss": 5.4262,
      "mean_token_accuracy": 0.16739489883184433,
      "num_tokens": 21046095.0,
      "step": 12130
    },
    {
      "entropy": 5.5479920387268065,
      "epoch": 0.9441742851585294,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004915885013786742,
      "loss": 5.395,
      "mean_token_accuracy": 0.1694901019334793,
      "num_tokens": 21054565.0,
      "step": 12135
    },
    {
      "entropy": 5.462245941162109,
      "epoch": 0.944563314530247,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004915809923772628,
      "loss": 5.3071,
      "mean_token_accuracy": 0.1779477044939995,
      "num_tokens": 21062942.0,
      "step": 12140
    },
    {
      "entropy": 5.58484296798706,
      "epoch": 0.9449523439019646,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004915734800895796,
      "loss": 5.3514,
      "mean_token_accuracy": 0.17015228122472764,
      "num_tokens": 21070642.0,
      "step": 12145
    },
    {
      "entropy": 5.588027906417847,
      "epoch": 0.9453413732736822,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004915659645157383,
      "loss": 5.3402,
      "mean_token_accuracy": 0.1680173844099045,
      "num_tokens": 21079007.0,
      "step": 12150
    },
    {
      "entropy": 5.559345531463623,
      "epoch": 0.9457304026453998,
      "grad_norm": 1.25,
      "learning_rate": 0.0004915584456558535,
      "loss": 5.4077,
      "mean_token_accuracy": 0.1643287569284439,
      "num_tokens": 21087672.0,
      "step": 12155
    },
    {
      "entropy": 5.656653642654419,
      "epoch": 0.9461194320171173,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004915509235100388,
      "loss": 5.3492,
      "mean_token_accuracy": 0.173394775390625,
      "num_tokens": 21095954.0,
      "step": 12160
    },
    {
      "entropy": 5.5760596752166744,
      "epoch": 0.9465084613888348,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004915433980784086,
      "loss": 5.3303,
      "mean_token_accuracy": 0.16818120181560517,
      "num_tokens": 21104439.0,
      "step": 12165
    },
    {
      "entropy": 5.500692892074585,
      "epoch": 0.9468974907605524,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004915358693610769,
      "loss": 5.347,
      "mean_token_accuracy": 0.16612892150878905,
      "num_tokens": 21112933.0,
      "step": 12170
    },
    {
      "entropy": 5.520151805877686,
      "epoch": 0.94728652013227,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004915283373581581,
      "loss": 5.3646,
      "mean_token_accuracy": 0.17748910188674927,
      "num_tokens": 21122085.0,
      "step": 12175
    },
    {
      "entropy": 5.5588806629180905,
      "epoch": 0.9476755495039876,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004915208020697664,
      "loss": 5.4124,
      "mean_token_accuracy": 0.1652074858546257,
      "num_tokens": 21130994.0,
      "step": 12180
    },
    {
      "entropy": 5.558933115005493,
      "epoch": 0.9480645788757052,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004915132634960162,
      "loss": 5.3438,
      "mean_token_accuracy": 0.1738900989294052,
      "num_tokens": 21139744.0,
      "step": 12185
    },
    {
      "entropy": 5.579050827026367,
      "epoch": 0.9484536082474226,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004915057216370218,
      "loss": 5.3579,
      "mean_token_accuracy": 0.17661173939704894,
      "num_tokens": 21148333.0,
      "step": 12190
    },
    {
      "entropy": 5.4838090419769285,
      "epoch": 0.9488426376191402,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004914981764928977,
      "loss": 5.2827,
      "mean_token_accuracy": 0.17709265500307084,
      "num_tokens": 21157450.0,
      "step": 12195
    },
    {
      "entropy": 5.558898305892944,
      "epoch": 0.9492316669908578,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004914906280637584,
      "loss": 5.4505,
      "mean_token_accuracy": 0.16984022557735443,
      "num_tokens": 21166410.0,
      "step": 12200
    },
    {
      "entropy": 5.556618070602417,
      "epoch": 0.9496206963625754,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004914830763497183,
      "loss": 5.3826,
      "mean_token_accuracy": 0.17338816225528716,
      "num_tokens": 21175104.0,
      "step": 12205
    },
    {
      "entropy": 5.5752503871917725,
      "epoch": 0.950009725734293,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004914755213508922,
      "loss": 5.3205,
      "mean_token_accuracy": 0.17283353358507156,
      "num_tokens": 21182597.0,
      "step": 12210
    },
    {
      "entropy": 5.535672855377197,
      "epoch": 0.9503987551060105,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004914679630673945,
      "loss": 5.4469,
      "mean_token_accuracy": 0.1719596952199936,
      "num_tokens": 21192399.0,
      "step": 12215
    },
    {
      "entropy": 5.51509108543396,
      "epoch": 0.950787784477728,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004914604014993401,
      "loss": 5.3096,
      "mean_token_accuracy": 0.17505268454551698,
      "num_tokens": 21201275.0,
      "step": 12220
    },
    {
      "entropy": 5.546757364273072,
      "epoch": 0.9511768138494456,
      "grad_norm": 1.375,
      "learning_rate": 0.0004914528366468436,
      "loss": 5.4068,
      "mean_token_accuracy": 0.1677502378821373,
      "num_tokens": 21210491.0,
      "step": 12225
    },
    {
      "entropy": 5.553897714614868,
      "epoch": 0.9515658432211632,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004914452685100199,
      "loss": 5.3833,
      "mean_token_accuracy": 0.1705961272120476,
      "num_tokens": 21218899.0,
      "step": 12230
    },
    {
      "entropy": 5.443309259414673,
      "epoch": 0.9519548725928808,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004914376970889836,
      "loss": 5.3049,
      "mean_token_accuracy": 0.1672735631465912,
      "num_tokens": 21227640.0,
      "step": 12235
    },
    {
      "entropy": 5.52520112991333,
      "epoch": 0.9523439019645983,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004914301223838497,
      "loss": 5.286,
      "mean_token_accuracy": 0.16754906475543976,
      "num_tokens": 21236422.0,
      "step": 12240
    },
    {
      "entropy": 5.623691463470459,
      "epoch": 0.9527329313363159,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004914225443947334,
      "loss": 5.4694,
      "mean_token_accuracy": 0.1670359268784523,
      "num_tokens": 21245820.0,
      "step": 12245
    },
    {
      "entropy": 5.657903051376342,
      "epoch": 0.9531219607080335,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004914149631217494,
      "loss": 5.3921,
      "mean_token_accuracy": 0.1695787288248539,
      "num_tokens": 21255570.0,
      "step": 12250
    },
    {
      "entropy": 5.587461233139038,
      "epoch": 0.953510990079751,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004914073785650127,
      "loss": 5.5234,
      "mean_token_accuracy": 0.16402170360088347,
      "num_tokens": 21263712.0,
      "step": 12255
    },
    {
      "entropy": 5.606937837600708,
      "epoch": 0.9539000194514686,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004913997907246385,
      "loss": 5.4016,
      "mean_token_accuracy": 0.17134646475315093,
      "num_tokens": 21272309.0,
      "step": 12260
    },
    {
      "entropy": 5.49057903289795,
      "epoch": 0.9542890488231861,
      "grad_norm": 1.375,
      "learning_rate": 0.000491392199600742,
      "loss": 5.3238,
      "mean_token_accuracy": 0.17048763632774352,
      "num_tokens": 21281711.0,
      "step": 12265
    },
    {
      "entropy": 5.525494194030761,
      "epoch": 0.9546780781949037,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004913846051934382,
      "loss": 5.2949,
      "mean_token_accuracy": 0.17485241293907167,
      "num_tokens": 21289524.0,
      "step": 12270
    },
    {
      "entropy": 5.457579898834228,
      "epoch": 0.9550671075666213,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004913770075028425,
      "loss": 5.2266,
      "mean_token_accuracy": 0.17732639461755753,
      "num_tokens": 21298225.0,
      "step": 12275
    },
    {
      "entropy": 5.546008729934693,
      "epoch": 0.9554561369383389,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004913694065290701,
      "loss": 5.2444,
      "mean_token_accuracy": 0.17801036685705185,
      "num_tokens": 21306819.0,
      "step": 12280
    },
    {
      "entropy": 5.557561731338501,
      "epoch": 0.9558451663100564,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004913618022722363,
      "loss": 5.4061,
      "mean_token_accuracy": 0.17224710136651994,
      "num_tokens": 21315517.0,
      "step": 12285
    },
    {
      "entropy": 5.537788677215576,
      "epoch": 0.9562341956817739,
      "grad_norm": 1.375,
      "learning_rate": 0.0004913541947324566,
      "loss": 5.3003,
      "mean_token_accuracy": 0.1735188901424408,
      "num_tokens": 21324144.0,
      "step": 12290
    },
    {
      "entropy": 5.563275861740112,
      "epoch": 0.9566232250534915,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004913465839098463,
      "loss": 5.3881,
      "mean_token_accuracy": 0.1700058825314045,
      "num_tokens": 21333329.0,
      "step": 12295
    },
    {
      "entropy": 5.464137029647827,
      "epoch": 0.9570122544252091,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004913389698045209,
      "loss": 5.3323,
      "mean_token_accuracy": 0.16273706108331681,
      "num_tokens": 21341955.0,
      "step": 12300
    },
    {
      "entropy": 5.5015605926513675,
      "epoch": 0.9574012837969267,
      "grad_norm": 1.34375,
      "learning_rate": 0.000491331352416596,
      "loss": 5.3122,
      "mean_token_accuracy": 0.17130431085824965,
      "num_tokens": 21350593.0,
      "step": 12305
    },
    {
      "entropy": 5.523934459686279,
      "epoch": 0.9577903131686443,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004913237317461872,
      "loss": 5.3875,
      "mean_token_accuracy": 0.1707832008600235,
      "num_tokens": 21359007.0,
      "step": 12310
    },
    {
      "entropy": 5.580479621887207,
      "epoch": 0.9581793425403617,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00049131610779341,
      "loss": 5.4218,
      "mean_token_accuracy": 0.16613728553056717,
      "num_tokens": 21367645.0,
      "step": 12315
    },
    {
      "entropy": 5.521607542037964,
      "epoch": 0.9585683719120793,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004913084805583803,
      "loss": 5.2306,
      "mean_token_accuracy": 0.17340102046728134,
      "num_tokens": 21376040.0,
      "step": 12320
    },
    {
      "entropy": 5.54140157699585,
      "epoch": 0.9589574012837969,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004913008500412136,
      "loss": 5.3671,
      "mean_token_accuracy": 0.17132264226675034,
      "num_tokens": 21385305.0,
      "step": 12325
    },
    {
      "entropy": 5.514420652389527,
      "epoch": 0.9593464306555145,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004912932162420259,
      "loss": 5.2663,
      "mean_token_accuracy": 0.1815379187464714,
      "num_tokens": 21393705.0,
      "step": 12330
    },
    {
      "entropy": 5.547263860702515,
      "epoch": 0.9597354600272321,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004912855791609327,
      "loss": 5.387,
      "mean_token_accuracy": 0.1692749187350273,
      "num_tokens": 21401830.0,
      "step": 12335
    },
    {
      "entropy": 5.552059459686279,
      "epoch": 0.9601244893989496,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004912779387980502,
      "loss": 5.4317,
      "mean_token_accuracy": 0.16849422603845596,
      "num_tokens": 21410398.0,
      "step": 12340
    },
    {
      "entropy": 5.631909608840942,
      "epoch": 0.9605135187706672,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004912702951534942,
      "loss": 5.3822,
      "mean_token_accuracy": 0.17147708535194398,
      "num_tokens": 21419082.0,
      "step": 12345
    },
    {
      "entropy": 5.595753288269043,
      "epoch": 0.9609025481423847,
      "grad_norm": 1.25,
      "learning_rate": 0.0004912626482273809,
      "loss": 5.3895,
      "mean_token_accuracy": 0.1671099007129669,
      "num_tokens": 21427992.0,
      "step": 12350
    },
    {
      "entropy": 5.481524419784546,
      "epoch": 0.9612915775141023,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004912549980198258,
      "loss": 5.3459,
      "mean_token_accuracy": 0.1666832983493805,
      "num_tokens": 21436933.0,
      "step": 12355
    },
    {
      "entropy": 5.504237365722656,
      "epoch": 0.9616806068858199,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004912473445309455,
      "loss": 5.3543,
      "mean_token_accuracy": 0.17218171656131745,
      "num_tokens": 21445640.0,
      "step": 12360
    },
    {
      "entropy": 5.6821917533874515,
      "epoch": 0.9620696362575375,
      "grad_norm": 1.296875,
      "learning_rate": 0.000491239687760856,
      "loss": 5.5099,
      "mean_token_accuracy": 0.16260401010513306,
      "num_tokens": 21454511.0,
      "step": 12365
    },
    {
      "entropy": 5.578676223754883,
      "epoch": 0.962458665629255,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004912320277096732,
      "loss": 5.4892,
      "mean_token_accuracy": 0.16189088821411132,
      "num_tokens": 21463851.0,
      "step": 12370
    },
    {
      "entropy": 5.5483697891235355,
      "epoch": 0.9628476950009726,
      "grad_norm": 1.375,
      "learning_rate": 0.0004912243643775137,
      "loss": 5.2794,
      "mean_token_accuracy": 0.17747919857501984,
      "num_tokens": 21471620.0,
      "step": 12375
    },
    {
      "entropy": 5.531028413772583,
      "epoch": 0.9632367243726901,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004912166977644936,
      "loss": 5.3061,
      "mean_token_accuracy": 0.1717200204730034,
      "num_tokens": 21479944.0,
      "step": 12380
    },
    {
      "entropy": 5.59373631477356,
      "epoch": 0.9636257537444077,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004912090278707293,
      "loss": 5.4865,
      "mean_token_accuracy": 0.16159820631146432,
      "num_tokens": 21489175.0,
      "step": 12385
    },
    {
      "entropy": 5.5857062339782715,
      "epoch": 0.9640147831161253,
      "grad_norm": 1.359375,
      "learning_rate": 0.000491201354696337,
      "loss": 5.3936,
      "mean_token_accuracy": 0.17435883432626725,
      "num_tokens": 21498533.0,
      "step": 12390
    },
    {
      "entropy": 5.69445915222168,
      "epoch": 0.9644038124878428,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004911936782414334,
      "loss": 5.4715,
      "mean_token_accuracy": 0.16166434586048126,
      "num_tokens": 21506771.0,
      "step": 12395
    },
    {
      "entropy": 5.459056186676025,
      "epoch": 0.9647928418595604,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004911859985061348,
      "loss": 5.2755,
      "mean_token_accuracy": 0.18061659038066863,
      "num_tokens": 21515359.0,
      "step": 12400
    },
    {
      "entropy": 5.519433212280274,
      "epoch": 0.965181871231278,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004911783154905577,
      "loss": 5.3294,
      "mean_token_accuracy": 0.16903999596834182,
      "num_tokens": 21524313.0,
      "step": 12405
    },
    {
      "entropy": 5.518152189254761,
      "epoch": 0.9655709006029956,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004911706291948188,
      "loss": 5.3517,
      "mean_token_accuracy": 0.1772301584482193,
      "num_tokens": 21533211.0,
      "step": 12410
    },
    {
      "entropy": 5.527912187576294,
      "epoch": 0.9659599299747131,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004911629396190348,
      "loss": 5.2999,
      "mean_token_accuracy": 0.1734490469098091,
      "num_tokens": 21541575.0,
      "step": 12415
    },
    {
      "entropy": 5.558382415771485,
      "epoch": 0.9663489593464306,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004911552467633221,
      "loss": 5.3867,
      "mean_token_accuracy": 0.16843879073858262,
      "num_tokens": 21549569.0,
      "step": 12420
    },
    {
      "entropy": 5.530548763275147,
      "epoch": 0.9667379887181482,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004911475506277976,
      "loss": 5.337,
      "mean_token_accuracy": 0.17053512632846832,
      "num_tokens": 21557277.0,
      "step": 12425
    },
    {
      "entropy": 5.519514799118042,
      "epoch": 0.9671270180898658,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000491139851212578,
      "loss": 5.3627,
      "mean_token_accuracy": 0.16825293004512787,
      "num_tokens": 21566590.0,
      "step": 12430
    },
    {
      "entropy": 5.587489366531372,
      "epoch": 0.9675160474615834,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004911321485177802,
      "loss": 5.4348,
      "mean_token_accuracy": 0.1667655289173126,
      "num_tokens": 21575140.0,
      "step": 12435
    },
    {
      "entropy": 5.5105226039886475,
      "epoch": 0.967905076833301,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004911244425435212,
      "loss": 5.3762,
      "mean_token_accuracy": 0.16279788911342621,
      "num_tokens": 21584647.0,
      "step": 12440
    },
    {
      "entropy": 5.546506690979004,
      "epoch": 0.9682941062050184,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004911167332899176,
      "loss": 5.3398,
      "mean_token_accuracy": 0.16958120614290237,
      "num_tokens": 21593609.0,
      "step": 12445
    },
    {
      "entropy": 5.644188690185547,
      "epoch": 0.968683135576736,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004911090207570867,
      "loss": 5.4499,
      "mean_token_accuracy": 0.16757878810167312,
      "num_tokens": 21602584.0,
      "step": 12450
    },
    {
      "entropy": 5.49824333190918,
      "epoch": 0.9690721649484536,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004911013049451453,
      "loss": 5.414,
      "mean_token_accuracy": 0.17698777616024017,
      "num_tokens": 21611109.0,
      "step": 12455
    },
    {
      "entropy": 5.644853734970093,
      "epoch": 0.9694611943201712,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004910935858542106,
      "loss": 5.4057,
      "mean_token_accuracy": 0.16859019547700882,
      "num_tokens": 21620173.0,
      "step": 12460
    },
    {
      "entropy": 5.511510848999023,
      "epoch": 0.9698502236918888,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004910858634843997,
      "loss": 5.2672,
      "mean_token_accuracy": 0.1754579558968544,
      "num_tokens": 21628318.0,
      "step": 12465
    },
    {
      "entropy": 5.582004690170288,
      "epoch": 0.9702392530636063,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004910781378358297,
      "loss": 5.4718,
      "mean_token_accuracy": 0.16632996499538422,
      "num_tokens": 21637102.0,
      "step": 12470
    },
    {
      "entropy": 5.5808652400970455,
      "epoch": 0.9706282824353238,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004910704089086178,
      "loss": 5.3162,
      "mean_token_accuracy": 0.17573461830615997,
      "num_tokens": 21645055.0,
      "step": 12475
    },
    {
      "entropy": 5.55949912071228,
      "epoch": 0.9710173118070414,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004910626767028815,
      "loss": 5.3942,
      "mean_token_accuracy": 0.15920934230089187,
      "num_tokens": 21654106.0,
      "step": 12480
    },
    {
      "entropy": 5.52292160987854,
      "epoch": 0.971406341178759,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004910549412187379,
      "loss": 5.3753,
      "mean_token_accuracy": 0.17179536670446396,
      "num_tokens": 21663081.0,
      "step": 12485
    },
    {
      "entropy": 5.57421383857727,
      "epoch": 0.9717953705504766,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004910472024563045,
      "loss": 5.3563,
      "mean_token_accuracy": 0.17092758566141128,
      "num_tokens": 21671033.0,
      "step": 12490
    },
    {
      "entropy": 5.538195371627808,
      "epoch": 0.9721843999221941,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004910394604156987,
      "loss": 5.352,
      "mean_token_accuracy": 0.16927904337644578,
      "num_tokens": 21679167.0,
      "step": 12495
    },
    {
      "entropy": 5.564725351333618,
      "epoch": 0.9725734292939117,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004910317150970379,
      "loss": 5.4736,
      "mean_token_accuracy": 0.16747646331787108,
      "num_tokens": 21688365.0,
      "step": 12500
    },
    {
      "entropy": 5.519552612304688,
      "epoch": 0.9729624586656292,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004910239665004397,
      "loss": 5.3733,
      "mean_token_accuracy": 0.16787540912628174,
      "num_tokens": 21697003.0,
      "step": 12505
    },
    {
      "entropy": 5.5843085765838625,
      "epoch": 0.9733514880373468,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004910162146260216,
      "loss": 5.347,
      "mean_token_accuracy": 0.17447817772626878,
      "num_tokens": 21705947.0,
      "step": 12510
    },
    {
      "entropy": 5.573052501678466,
      "epoch": 0.9737405174090644,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004910084594739013,
      "loss": 5.3835,
      "mean_token_accuracy": 0.1666942611336708,
      "num_tokens": 21714153.0,
      "step": 12515
    },
    {
      "entropy": 5.56226692199707,
      "epoch": 0.9741295467807819,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004910007010441964,
      "loss": 5.4689,
      "mean_token_accuracy": 0.16191142797470093,
      "num_tokens": 21723480.0,
      "step": 12520
    },
    {
      "entropy": 5.634710502624512,
      "epoch": 0.9745185761524995,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004909929393370248,
      "loss": 5.3308,
      "mean_token_accuracy": 0.17485801875591278,
      "num_tokens": 21731297.0,
      "step": 12525
    },
    {
      "entropy": 5.458512926101685,
      "epoch": 0.9749076055242171,
      "grad_norm": 1.25,
      "learning_rate": 0.0004909851743525041,
      "loss": 5.332,
      "mean_token_accuracy": 0.17015228420495987,
      "num_tokens": 21740975.0,
      "step": 12530
    },
    {
      "entropy": 5.456052350997925,
      "epoch": 0.9752966348959347,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000490977406090752,
      "loss": 5.3304,
      "mean_token_accuracy": 0.17154321819543839,
      "num_tokens": 21749642.0,
      "step": 12535
    },
    {
      "entropy": 5.5707025051116945,
      "epoch": 0.9756856642676522,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004909696345518867,
      "loss": 5.3629,
      "mean_token_accuracy": 0.1709204748272896,
      "num_tokens": 21757962.0,
      "step": 12540
    },
    {
      "entropy": 5.590342092514038,
      "epoch": 0.9760746936393697,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004909618597360258,
      "loss": 5.3644,
      "mean_token_accuracy": 0.17000051885843276,
      "num_tokens": 21766053.0,
      "step": 12545
    },
    {
      "entropy": 5.486387872695923,
      "epoch": 0.9764637230110873,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004909540816432876,
      "loss": 5.3034,
      "mean_token_accuracy": 0.17007056772708892,
      "num_tokens": 21774280.0,
      "step": 12550
    },
    {
      "entropy": 5.523618841171265,
      "epoch": 0.9768527523828049,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004909463002737897,
      "loss": 5.3415,
      "mean_token_accuracy": 0.1724835976958275,
      "num_tokens": 21782879.0,
      "step": 12555
    },
    {
      "entropy": 5.597775173187256,
      "epoch": 0.9772417817545225,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004909385156276506,
      "loss": 5.4279,
      "mean_token_accuracy": 0.16808952242136002,
      "num_tokens": 21791294.0,
      "step": 12560
    },
    {
      "entropy": 5.605129337310791,
      "epoch": 0.9776308111262401,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004909307277049881,
      "loss": 5.2643,
      "mean_token_accuracy": 0.1793579190969467,
      "num_tokens": 21799304.0,
      "step": 12565
    },
    {
      "entropy": 5.58639235496521,
      "epoch": 0.9780198404979576,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004909229365059205,
      "loss": 5.4552,
      "mean_token_accuracy": 0.16180186122655868,
      "num_tokens": 21809024.0,
      "step": 12570
    },
    {
      "entropy": 5.515496301651001,
      "epoch": 0.9784088698696751,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004909151420305661,
      "loss": 5.2608,
      "mean_token_accuracy": 0.180128613114357,
      "num_tokens": 21817412.0,
      "step": 12575
    },
    {
      "entropy": 5.443965244293213,
      "epoch": 0.9787978992413927,
      "grad_norm": 1.28125,
      "learning_rate": 0.000490907344279043,
      "loss": 5.285,
      "mean_token_accuracy": 0.1706534057855606,
      "num_tokens": 21825261.0,
      "step": 12580
    },
    {
      "entropy": 5.598343181610107,
      "epoch": 0.9791869286131103,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004908995432514698,
      "loss": 5.4011,
      "mean_token_accuracy": 0.1770653709769249,
      "num_tokens": 21833318.0,
      "step": 12585
    },
    {
      "entropy": 5.543214321136475,
      "epoch": 0.9795759579848279,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004908917389479645,
      "loss": 5.368,
      "mean_token_accuracy": 0.16777470111846923,
      "num_tokens": 21841377.0,
      "step": 12590
    },
    {
      "entropy": 5.47628231048584,
      "epoch": 0.9799649873565455,
      "grad_norm": 1.25,
      "learning_rate": 0.0004908839313686456,
      "loss": 5.3015,
      "mean_token_accuracy": 0.17293741703033447,
      "num_tokens": 21850183.0,
      "step": 12595
    },
    {
      "entropy": 5.583069467544556,
      "epoch": 0.980354016728263,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004908761205136319,
      "loss": 5.4142,
      "mean_token_accuracy": 0.1798610806465149,
      "num_tokens": 21858108.0,
      "step": 12600
    },
    {
      "entropy": 5.492361688613892,
      "epoch": 0.9807430460999805,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004908683063830414,
      "loss": 5.2702,
      "mean_token_accuracy": 0.17607586979866027,
      "num_tokens": 21866330.0,
      "step": 12605
    },
    {
      "entropy": 5.557174110412598,
      "epoch": 0.9811320754716981,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004908604889769932,
      "loss": 5.3644,
      "mean_token_accuracy": 0.17308066487312318,
      "num_tokens": 21875254.0,
      "step": 12610
    },
    {
      "entropy": 5.593617582321167,
      "epoch": 0.9815211048434157,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004908526682956054,
      "loss": 5.435,
      "mean_token_accuracy": 0.16785266995429993,
      "num_tokens": 21884009.0,
      "step": 12615
    },
    {
      "entropy": 5.565900230407715,
      "epoch": 0.9819101342151333,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004908448443389972,
      "loss": 5.3584,
      "mean_token_accuracy": 0.16843471825122833,
      "num_tokens": 21892428.0,
      "step": 12620
    },
    {
      "entropy": 5.578279113769531,
      "epoch": 0.9822991635868508,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004908370171072869,
      "loss": 5.4003,
      "mean_token_accuracy": 0.17159607410430908,
      "num_tokens": 21901140.0,
      "step": 12625
    },
    {
      "entropy": 5.515268325805664,
      "epoch": 0.9826881929585684,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004908291866005933,
      "loss": 5.291,
      "mean_token_accuracy": 0.1810177117586136,
      "num_tokens": 21909072.0,
      "step": 12630
    },
    {
      "entropy": 5.488519144058228,
      "epoch": 0.9830772223302859,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004908213528190355,
      "loss": 5.3573,
      "mean_token_accuracy": 0.16976865082979203,
      "num_tokens": 21917873.0,
      "step": 12635
    },
    {
      "entropy": 5.631306171417236,
      "epoch": 0.9834662517020035,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004908135157627321,
      "loss": 5.4831,
      "mean_token_accuracy": 0.1660613462328911,
      "num_tokens": 21927302.0,
      "step": 12640
    },
    {
      "entropy": 5.614320468902588,
      "epoch": 0.9838552810737211,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004908056754318022,
      "loss": 5.391,
      "mean_token_accuracy": 0.17224741876125335,
      "num_tokens": 21936440.0,
      "step": 12645
    },
    {
      "entropy": 5.510326194763183,
      "epoch": 0.9842443104454386,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004907978318263646,
      "loss": 5.2611,
      "mean_token_accuracy": 0.1719265729188919,
      "num_tokens": 21944655.0,
      "step": 12650
    },
    {
      "entropy": 5.626398706436158,
      "epoch": 0.9846333398171562,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004907899849465383,
      "loss": 5.4974,
      "mean_token_accuracy": 0.16908804923295975,
      "num_tokens": 21953202.0,
      "step": 12655
    },
    {
      "entropy": 5.557944822311401,
      "epoch": 0.9850223691888738,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004907821347924424,
      "loss": 5.3551,
      "mean_token_accuracy": 0.17220570743083954,
      "num_tokens": 21961889.0,
      "step": 12660
    },
    {
      "entropy": 5.513428831100464,
      "epoch": 0.9854113985605913,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004907742813641963,
      "loss": 5.2421,
      "mean_token_accuracy": 0.17460116744041443,
      "num_tokens": 21970864.0,
      "step": 12665
    },
    {
      "entropy": 5.507844638824463,
      "epoch": 0.9858004279323089,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004907664246619187,
      "loss": 5.342,
      "mean_token_accuracy": 0.17051352858543395,
      "num_tokens": 21979271.0,
      "step": 12670
    },
    {
      "entropy": 5.517146348953247,
      "epoch": 0.9861894573040264,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004907585646857293,
      "loss": 5.1946,
      "mean_token_accuracy": 0.18001116067171097,
      "num_tokens": 21987321.0,
      "step": 12675
    },
    {
      "entropy": 5.557333612442017,
      "epoch": 0.986578486675744,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004907507014357467,
      "loss": 5.4685,
      "mean_token_accuracy": 0.16238831579685212,
      "num_tokens": 21996094.0,
      "step": 12680
    },
    {
      "entropy": 5.530393123626709,
      "epoch": 0.9869675160474616,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004907428349120909,
      "loss": 5.3241,
      "mean_token_accuracy": 0.16548317670822144,
      "num_tokens": 22004095.0,
      "step": 12685
    },
    {
      "entropy": 5.619708108901977,
      "epoch": 0.9873565454191792,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004907349651148809,
      "loss": 5.4298,
      "mean_token_accuracy": 0.16731757223606109,
      "num_tokens": 22013371.0,
      "step": 12690
    },
    {
      "entropy": 5.528348302841186,
      "epoch": 0.9877455747908968,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004907270920442361,
      "loss": 5.2894,
      "mean_token_accuracy": 0.17162059843540192,
      "num_tokens": 22022313.0,
      "step": 12695
    },
    {
      "entropy": 5.56501259803772,
      "epoch": 0.9881346041626142,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004907192157002762,
      "loss": 5.4691,
      "mean_token_accuracy": 0.1678902328014374,
      "num_tokens": 22031690.0,
      "step": 12700
    },
    {
      "entropy": 5.696156454086304,
      "epoch": 0.9885236335343318,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004907113360831204,
      "loss": 5.5141,
      "mean_token_accuracy": 0.15464612394571303,
      "num_tokens": 22040957.0,
      "step": 12705
    },
    {
      "entropy": 5.589044189453125,
      "epoch": 0.9889126629060494,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004907034531928886,
      "loss": 5.335,
      "mean_token_accuracy": 0.17279461920261383,
      "num_tokens": 22049623.0,
      "step": 12710
    },
    {
      "entropy": 5.493846035003662,
      "epoch": 0.989301692277767,
      "grad_norm": 1.375,
      "learning_rate": 0.0004906955670297001,
      "loss": 5.3536,
      "mean_token_accuracy": 0.1677471026778221,
      "num_tokens": 22058205.0,
      "step": 12715
    },
    {
      "entropy": 5.476700496673584,
      "epoch": 0.9896907216494846,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004906876775936746,
      "loss": 5.2271,
      "mean_token_accuracy": 0.17466342747211455,
      "num_tokens": 22067375.0,
      "step": 12720
    },
    {
      "entropy": 5.475682115554809,
      "epoch": 0.990079751021202,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004906797848849321,
      "loss": 5.2723,
      "mean_token_accuracy": 0.1794630318880081,
      "num_tokens": 22075670.0,
      "step": 12725
    },
    {
      "entropy": 5.4448870658874515,
      "epoch": 0.9904687803929196,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000490671888903592,
      "loss": 5.3626,
      "mean_token_accuracy": 0.17033643424510955,
      "num_tokens": 22084360.0,
      "step": 12730
    },
    {
      "entropy": 5.547029829025268,
      "epoch": 0.9908578097646372,
      "grad_norm": 1.375,
      "learning_rate": 0.0004906639896497744,
      "loss": 5.3582,
      "mean_token_accuracy": 0.17353748232126237,
      "num_tokens": 22092666.0,
      "step": 12735
    },
    {
      "entropy": 5.6224525451660154,
      "epoch": 0.9912468391363548,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004906560871235988,
      "loss": 5.5007,
      "mean_token_accuracy": 0.16956808120012284,
      "num_tokens": 22101311.0,
      "step": 12740
    },
    {
      "entropy": 5.575134420394898,
      "epoch": 0.9916358685080724,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004906481813251854,
      "loss": 5.3143,
      "mean_token_accuracy": 0.1791418418288231,
      "num_tokens": 22109640.0,
      "step": 12745
    },
    {
      "entropy": 5.471666574478149,
      "epoch": 0.9920248978797899,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004906402722546542,
      "loss": 5.3109,
      "mean_token_accuracy": 0.16892679482698442,
      "num_tokens": 22118721.0,
      "step": 12750
    },
    {
      "entropy": 5.469309616088867,
      "epoch": 0.9924139272515075,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000490632359912125,
      "loss": 5.3184,
      "mean_token_accuracy": 0.17011154145002366,
      "num_tokens": 22126745.0,
      "step": 12755
    },
    {
      "entropy": 5.601627016067505,
      "epoch": 0.992802956623225,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000490624444297718,
      "loss": 5.3962,
      "mean_token_accuracy": 0.16726614087820052,
      "num_tokens": 22135010.0,
      "step": 12760
    },
    {
      "entropy": 5.505506992340088,
      "epoch": 0.9931919859949426,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004906165254115533,
      "loss": 5.3337,
      "mean_token_accuracy": 0.17282298654317857,
      "num_tokens": 22144339.0,
      "step": 12765
    },
    {
      "entropy": 5.397467803955078,
      "epoch": 0.9935810153666602,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000490608603253751,
      "loss": 5.2787,
      "mean_token_accuracy": 0.18157540112733841,
      "num_tokens": 22153405.0,
      "step": 12770
    },
    {
      "entropy": 5.592889022827149,
      "epoch": 0.9939700447383778,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004906006778244312,
      "loss": 5.3677,
      "mean_token_accuracy": 0.16852252036333085,
      "num_tokens": 22161392.0,
      "step": 12775
    },
    {
      "entropy": 5.588080787658692,
      "epoch": 0.9943590741100953,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004905927491237145,
      "loss": 5.3227,
      "mean_token_accuracy": 0.1759607598185539,
      "num_tokens": 22170635.0,
      "step": 12780
    },
    {
      "entropy": 5.416839694976806,
      "epoch": 0.9947481034818129,
      "grad_norm": 1.25,
      "learning_rate": 0.0004905848171517208,
      "loss": 5.2137,
      "mean_token_accuracy": 0.17934241890907288,
      "num_tokens": 22178629.0,
      "step": 12785
    },
    {
      "entropy": 5.435929584503174,
      "epoch": 0.9951371328535304,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004905768819085706,
      "loss": 5.2841,
      "mean_token_accuracy": 0.17653844505548477,
      "num_tokens": 22187142.0,
      "step": 12790
    },
    {
      "entropy": 5.578234100341797,
      "epoch": 0.995526162225248,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004905689433943846,
      "loss": 5.3572,
      "mean_token_accuracy": 0.1716219186782837,
      "num_tokens": 22195249.0,
      "step": 12795
    },
    {
      "entropy": 5.583882570266724,
      "epoch": 0.9959151915969656,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004905610016092828,
      "loss": 5.3933,
      "mean_token_accuracy": 0.17545613944530486,
      "num_tokens": 22203621.0,
      "step": 12800
    },
    {
      "entropy": 5.500404405593872,
      "epoch": 0.9963042209686831,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004905530565533859,
      "loss": 5.3363,
      "mean_token_accuracy": 0.1793372169137001,
      "num_tokens": 22212547.0,
      "step": 12805
    },
    {
      "entropy": 5.5117346286773685,
      "epoch": 0.9966932503404007,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004905451082268146,
      "loss": 5.3314,
      "mean_token_accuracy": 0.17046946585178374,
      "num_tokens": 22221642.0,
      "step": 12810
    },
    {
      "entropy": 5.537295722961426,
      "epoch": 0.9970822797121183,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004905371566296891,
      "loss": 5.3587,
      "mean_token_accuracy": 0.1707915872335434,
      "num_tokens": 22230227.0,
      "step": 12815
    },
    {
      "entropy": 5.535537910461426,
      "epoch": 0.9974713090838359,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004905292017621305,
      "loss": 5.3452,
      "mean_token_accuracy": 0.17594442516565323,
      "num_tokens": 22238973.0,
      "step": 12820
    },
    {
      "entropy": 5.495472240447998,
      "epoch": 0.9978603384555534,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004905212436242593,
      "loss": 5.3295,
      "mean_token_accuracy": 0.17380284518003464,
      "num_tokens": 22247613.0,
      "step": 12825
    },
    {
      "entropy": 5.481318044662475,
      "epoch": 0.9982493678272709,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004905132822161962,
      "loss": 5.3526,
      "mean_token_accuracy": 0.16845973283052446,
      "num_tokens": 22255707.0,
      "step": 12830
    },
    {
      "entropy": 5.541123962402343,
      "epoch": 0.9986383971989885,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004905053175380621,
      "loss": 5.2302,
      "mean_token_accuracy": 0.1792900815606117,
      "num_tokens": 22264524.0,
      "step": 12835
    },
    {
      "entropy": 5.595702219009399,
      "epoch": 0.9990274265707061,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004904973495899778,
      "loss": 5.3511,
      "mean_token_accuracy": 0.1747155100107193,
      "num_tokens": 22272973.0,
      "step": 12840
    },
    {
      "entropy": 5.566828441619873,
      "epoch": 0.9994164559424237,
      "grad_norm": 1.25,
      "learning_rate": 0.0004904893783720642,
      "loss": 5.3866,
      "mean_token_accuracy": 0.16957555562257767,
      "num_tokens": 22281545.0,
      "step": 12845
    },
    {
      "entropy": 5.525720453262329,
      "epoch": 0.9998054853141413,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004904814038844424,
      "loss": 5.3761,
      "mean_token_accuracy": 0.17147639244794846,
      "num_tokens": 22289835.0,
      "step": 12850
    },
    {
      "entropy": 5.510589440663655,
      "epoch": 1.000155611748687,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004904734261272332,
      "loss": 5.3341,
      "mean_token_accuracy": 0.17134587797853681,
      "num_tokens": 22297794.0,
      "step": 12855
    },
    {
      "entropy": 5.476920938491821,
      "epoch": 1.0005446411204046,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004904654451005576,
      "loss": 5.2384,
      "mean_token_accuracy": 0.1745680019259453,
      "num_tokens": 22305637.0,
      "step": 12860
    },
    {
      "entropy": 5.534882068634033,
      "epoch": 1.0009336704921221,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004904574608045369,
      "loss": 5.3275,
      "mean_token_accuracy": 0.16650792956352234,
      "num_tokens": 22314796.0,
      "step": 12865
    },
    {
      "entropy": 5.559276580810547,
      "epoch": 1.0013226998638398,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004904494732392923,
      "loss": 5.2662,
      "mean_token_accuracy": 0.1689400315284729,
      "num_tokens": 22324522.0,
      "step": 12870
    },
    {
      "entropy": 5.538963031768799,
      "epoch": 1.0017117292355573,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004904414824049448,
      "loss": 5.2608,
      "mean_token_accuracy": 0.17630613446235657,
      "num_tokens": 22333404.0,
      "step": 12875
    },
    {
      "entropy": 5.482955694198608,
      "epoch": 1.0021007586072748,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004904334883016156,
      "loss": 5.3615,
      "mean_token_accuracy": 0.1660408392548561,
      "num_tokens": 22343364.0,
      "step": 12880
    },
    {
      "entropy": 5.565887594223023,
      "epoch": 1.0024897879789925,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004904254909294261,
      "loss": 5.297,
      "mean_token_accuracy": 0.1664889261126518,
      "num_tokens": 22352854.0,
      "step": 12885
    },
    {
      "entropy": 5.530776166915894,
      "epoch": 1.00287881735071,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004904174902884978,
      "loss": 5.1976,
      "mean_token_accuracy": 0.18283711969852448,
      "num_tokens": 22361916.0,
      "step": 12890
    },
    {
      "entropy": 5.479391622543335,
      "epoch": 1.0032678467224276,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004904094863789519,
      "loss": 5.2984,
      "mean_token_accuracy": 0.17210589200258256,
      "num_tokens": 22370848.0,
      "step": 12895
    },
    {
      "entropy": 5.461601734161377,
      "epoch": 1.0036568760941451,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00049040147920091,
      "loss": 5.1803,
      "mean_token_accuracy": 0.18275183886289598,
      "num_tokens": 22379945.0,
      "step": 12900
    },
    {
      "entropy": 5.4947206497192385,
      "epoch": 1.0040459054658626,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004903934687544933,
      "loss": 5.2458,
      "mean_token_accuracy": 0.17642272859811783,
      "num_tokens": 22388795.0,
      "step": 12905
    },
    {
      "entropy": 5.56704273223877,
      "epoch": 1.0044349348375803,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004903854550398237,
      "loss": 5.3014,
      "mean_token_accuracy": 0.17787829041481018,
      "num_tokens": 22396960.0,
      "step": 12910
    },
    {
      "entropy": 5.530378913879394,
      "epoch": 1.0048239642092978,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004903774380570226,
      "loss": 5.168,
      "mean_token_accuracy": 0.1767799064517021,
      "num_tokens": 22404698.0,
      "step": 12915
    },
    {
      "entropy": 5.482751893997192,
      "epoch": 1.0052129935810155,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004903694178062117,
      "loss": 5.2137,
      "mean_token_accuracy": 0.17133289724588394,
      "num_tokens": 22413391.0,
      "step": 12920
    },
    {
      "entropy": 5.561685180664062,
      "epoch": 1.005602022952733,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004903613942875126,
      "loss": 5.3245,
      "mean_token_accuracy": 0.1758733555674553,
      "num_tokens": 22421357.0,
      "step": 12925
    },
    {
      "entropy": 5.522103023529053,
      "epoch": 1.0059910523244504,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004903533675010472,
      "loss": 5.1909,
      "mean_token_accuracy": 0.17573100328445435,
      "num_tokens": 22430286.0,
      "step": 12930
    },
    {
      "entropy": 5.4525957107543945,
      "epoch": 1.006380081696168,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004903453374469373,
      "loss": 5.2638,
      "mean_token_accuracy": 0.17179507464170457,
      "num_tokens": 22439128.0,
      "step": 12935
    },
    {
      "entropy": 5.445306825637817,
      "epoch": 1.0067691110678856,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004903373041253045,
      "loss": 5.211,
      "mean_token_accuracy": 0.1774335116147995,
      "num_tokens": 22447724.0,
      "step": 12940
    },
    {
      "entropy": 5.45561900138855,
      "epoch": 1.0071581404396033,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004903292675362709,
      "loss": 5.2494,
      "mean_token_accuracy": 0.18000908493995665,
      "num_tokens": 22456792.0,
      "step": 12945
    },
    {
      "entropy": 5.442387437820434,
      "epoch": 1.0075471698113208,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004903212276799584,
      "loss": 5.1887,
      "mean_token_accuracy": 0.1780812993645668,
      "num_tokens": 22465465.0,
      "step": 12950
    },
    {
      "entropy": 5.484514951705933,
      "epoch": 1.0079361991830382,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004903131845564889,
      "loss": 5.2283,
      "mean_token_accuracy": 0.18253717869520186,
      "num_tokens": 22474498.0,
      "step": 12955
    },
    {
      "entropy": 5.450583362579346,
      "epoch": 1.008325228554756,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004903051381659847,
      "loss": 5.173,
      "mean_token_accuracy": 0.1826133444905281,
      "num_tokens": 22482926.0,
      "step": 12960
    },
    {
      "entropy": 5.448676443099975,
      "epoch": 1.0087142579264734,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004902970885085677,
      "loss": 5.2054,
      "mean_token_accuracy": 0.18125245422124864,
      "num_tokens": 22491984.0,
      "step": 12965
    },
    {
      "entropy": 5.505223941802979,
      "epoch": 1.009103287298191,
      "grad_norm": 1.296875,
      "learning_rate": 0.00049028903558436,
      "loss": 5.2335,
      "mean_token_accuracy": 0.1683299347758293,
      "num_tokens": 22501411.0,
      "step": 12970
    },
    {
      "entropy": 5.536653661727906,
      "epoch": 1.0094923166699086,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004902809793934838,
      "loss": 5.355,
      "mean_token_accuracy": 0.17801813781261444,
      "num_tokens": 22510289.0,
      "step": 12975
    },
    {
      "entropy": 5.5372443199157715,
      "epoch": 1.009881346041626,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004902729199360615,
      "loss": 5.3752,
      "mean_token_accuracy": 0.16908740103244782,
      "num_tokens": 22519061.0,
      "step": 12980
    },
    {
      "entropy": 5.5717541694641115,
      "epoch": 1.0102703754133437,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004902648572122153,
      "loss": 5.3608,
      "mean_token_accuracy": 0.16445601284503936,
      "num_tokens": 22528207.0,
      "step": 12985
    },
    {
      "entropy": 5.545996761322021,
      "epoch": 1.0106594047850612,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004902567912220674,
      "loss": 5.302,
      "mean_token_accuracy": 0.17262889742851256,
      "num_tokens": 22537055.0,
      "step": 12990
    },
    {
      "entropy": 5.41119155883789,
      "epoch": 1.011048434156779,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004902487219657404,
      "loss": 5.2733,
      "mean_token_accuracy": 0.17389650344848634,
      "num_tokens": 22545239.0,
      "step": 12995
    },
    {
      "entropy": 5.539286422729492,
      "epoch": 1.0114374635284964,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004902406494433566,
      "loss": 5.4102,
      "mean_token_accuracy": 0.1639057844877243,
      "num_tokens": 22554850.0,
      "step": 13000
    },
    {
      "entropy": 5.58484902381897,
      "epoch": 1.0118264929002139,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004902325736550386,
      "loss": 5.2621,
      "mean_token_accuracy": 0.1735207036137581,
      "num_tokens": 22563799.0,
      "step": 13005
    },
    {
      "entropy": 5.565141725540161,
      "epoch": 1.0122155222719316,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004902244946009088,
      "loss": 5.4555,
      "mean_token_accuracy": 0.16532764360308647,
      "num_tokens": 22572699.0,
      "step": 13010
    },
    {
      "entropy": 5.645598077774048,
      "epoch": 1.012604551643649,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004902164122810899,
      "loss": 5.36,
      "mean_token_accuracy": 0.16890981197357177,
      "num_tokens": 22580916.0,
      "step": 13015
    },
    {
      "entropy": 5.543091106414795,
      "epoch": 1.0129935810153667,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004902083266957045,
      "loss": 5.2406,
      "mean_token_accuracy": 0.18386773765087128,
      "num_tokens": 22589525.0,
      "step": 13020
    },
    {
      "entropy": 5.463796091079712,
      "epoch": 1.0133826103870842,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004902002378448753,
      "loss": 5.2667,
      "mean_token_accuracy": 0.17648083567619324,
      "num_tokens": 22597929.0,
      "step": 13025
    },
    {
      "entropy": 5.518269729614258,
      "epoch": 1.0137716397588017,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000490192145728725,
      "loss": 5.2844,
      "mean_token_accuracy": 0.17419883608818054,
      "num_tokens": 22606500.0,
      "step": 13030
    },
    {
      "entropy": 5.549993324279785,
      "epoch": 1.0141606691305194,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004901840503473764,
      "loss": 5.2381,
      "mean_token_accuracy": 0.1711733043193817,
      "num_tokens": 22614626.0,
      "step": 13035
    },
    {
      "entropy": 5.519760465621948,
      "epoch": 1.0145496985022369,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004901759517009522,
      "loss": 5.3013,
      "mean_token_accuracy": 0.16727600991725922,
      "num_tokens": 22623488.0,
      "step": 13040
    },
    {
      "entropy": 5.505440187454224,
      "epoch": 1.0149387278739546,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004901678497895755,
      "loss": 5.2026,
      "mean_token_accuracy": 0.17804351896047593,
      "num_tokens": 22632413.0,
      "step": 13045
    },
    {
      "entropy": 5.547317266464233,
      "epoch": 1.015327757245672,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004901597446133692,
      "loss": 5.358,
      "mean_token_accuracy": 0.16635855585336684,
      "num_tokens": 22640982.0,
      "step": 13050
    },
    {
      "entropy": 5.562605667114258,
      "epoch": 1.0157167866173895,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004901516361724564,
      "loss": 5.33,
      "mean_token_accuracy": 0.17078778743743897,
      "num_tokens": 22649809.0,
      "step": 13055
    },
    {
      "entropy": 5.5064808368682865,
      "epoch": 1.0161058159891072,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004901435244669597,
      "loss": 5.244,
      "mean_token_accuracy": 0.17461332231760024,
      "num_tokens": 22658292.0,
      "step": 13060
    },
    {
      "entropy": 5.580621147155762,
      "epoch": 1.0164948453608247,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004901354094970025,
      "loss": 5.3153,
      "mean_token_accuracy": 0.17145186066627502,
      "num_tokens": 22667063.0,
      "step": 13065
    },
    {
      "entropy": 5.498051023483276,
      "epoch": 1.0168838747325424,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004901272912627081,
      "loss": 5.2391,
      "mean_token_accuracy": 0.17214105725288392,
      "num_tokens": 22676213.0,
      "step": 13070
    },
    {
      "entropy": 5.401425981521607,
      "epoch": 1.0172729041042599,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004901191697641992,
      "loss": 5.1762,
      "mean_token_accuracy": 0.17646822184324265,
      "num_tokens": 22684948.0,
      "step": 13075
    },
    {
      "entropy": 5.571120929718018,
      "epoch": 1.0176619334759773,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004901110450015994,
      "loss": 5.283,
      "mean_token_accuracy": 0.17036380767822265,
      "num_tokens": 22693218.0,
      "step": 13080
    },
    {
      "entropy": 5.486243724822998,
      "epoch": 1.018050962847695,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004901029169750319,
      "loss": 5.144,
      "mean_token_accuracy": 0.18303845971822738,
      "num_tokens": 22701572.0,
      "step": 13085
    },
    {
      "entropy": 5.417777347564697,
      "epoch": 1.0184399922194125,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004900947856846201,
      "loss": 5.2281,
      "mean_token_accuracy": 0.17624456584453582,
      "num_tokens": 22709498.0,
      "step": 13090
    },
    {
      "entropy": 5.523627710342407,
      "epoch": 1.0188290215911302,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000490086651130487,
      "loss": 5.2401,
      "mean_token_accuracy": 0.17883387058973313,
      "num_tokens": 22717922.0,
      "step": 13095
    },
    {
      "entropy": 5.55873703956604,
      "epoch": 1.0192180509628477,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004900785133127566,
      "loss": 5.2683,
      "mean_token_accuracy": 0.18007400631904602,
      "num_tokens": 22725933.0,
      "step": 13100
    },
    {
      "entropy": 5.4631119728088375,
      "epoch": 1.0196070803345652,
      "grad_norm": 1.34375,
      "learning_rate": 0.000490070372231552,
      "loss": 5.2353,
      "mean_token_accuracy": 0.17319102138280867,
      "num_tokens": 22734198.0,
      "step": 13105
    },
    {
      "entropy": 5.497663450241089,
      "epoch": 1.0199961097062828,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004900622278869968,
      "loss": 5.2805,
      "mean_token_accuracy": 0.1735595539212227,
      "num_tokens": 22742419.0,
      "step": 13110
    },
    {
      "entropy": 5.543889570236206,
      "epoch": 1.0203851390780003,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004900540802792146,
      "loss": 5.2288,
      "mean_token_accuracy": 0.17765220254659653,
      "num_tokens": 22750576.0,
      "step": 13115
    },
    {
      "entropy": 5.593791818618774,
      "epoch": 1.020774168449718,
      "grad_norm": 1.359375,
      "learning_rate": 0.000490045929408329,
      "loss": 5.4007,
      "mean_token_accuracy": 0.1730293795466423,
      "num_tokens": 22759604.0,
      "step": 13120
    },
    {
      "entropy": 5.481210041046142,
      "epoch": 1.0211631978214355,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004900377752744637,
      "loss": 5.1526,
      "mean_token_accuracy": 0.17913611233234406,
      "num_tokens": 22768305.0,
      "step": 13125
    },
    {
      "entropy": 5.515837955474853,
      "epoch": 1.021552227193153,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004900296178777426,
      "loss": 5.3151,
      "mean_token_accuracy": 0.16377966552972795,
      "num_tokens": 22777364.0,
      "step": 13130
    },
    {
      "entropy": 5.48155779838562,
      "epoch": 1.0219412565648707,
      "grad_norm": 1.234375,
      "learning_rate": 0.000490021457218289,
      "loss": 5.2306,
      "mean_token_accuracy": 0.17472662031650543,
      "num_tokens": 22786192.0,
      "step": 13135
    },
    {
      "entropy": 5.565016174316407,
      "epoch": 1.0223302859365881,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004900132932962272,
      "loss": 5.2865,
      "mean_token_accuracy": 0.1778672769665718,
      "num_tokens": 22794244.0,
      "step": 13140
    },
    {
      "entropy": 5.505892419815064,
      "epoch": 1.0227193153083058,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000490005126111681,
      "loss": 5.2708,
      "mean_token_accuracy": 0.18059523701667785,
      "num_tokens": 22802543.0,
      "step": 13145
    },
    {
      "entropy": 5.463393592834473,
      "epoch": 1.0231083446800233,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000489996955664774,
      "loss": 5.281,
      "mean_token_accuracy": 0.1724498152732849,
      "num_tokens": 22811267.0,
      "step": 13150
    },
    {
      "entropy": 5.462934446334839,
      "epoch": 1.023497374051741,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004899887819556306,
      "loss": 5.2036,
      "mean_token_accuracy": 0.17702029049396514,
      "num_tokens": 22820196.0,
      "step": 13155
    },
    {
      "entropy": 5.514190244674682,
      "epoch": 1.0238864034234585,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004899806049843745,
      "loss": 5.3506,
      "mean_token_accuracy": 0.17238128632307054,
      "num_tokens": 22828498.0,
      "step": 13160
    },
    {
      "entropy": 5.560862827301025,
      "epoch": 1.024275432795176,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004899724247511299,
      "loss": 5.2691,
      "mean_token_accuracy": 0.17590803354978563,
      "num_tokens": 22837411.0,
      "step": 13165
    },
    {
      "entropy": 5.436128950119018,
      "epoch": 1.0246644621668937,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000489964241256021,
      "loss": 5.1931,
      "mean_token_accuracy": 0.17749083787202835,
      "num_tokens": 22846265.0,
      "step": 13170
    },
    {
      "entropy": 5.383639669418335,
      "epoch": 1.0250534915386111,
      "grad_norm": 1.625,
      "learning_rate": 0.0004899560544991718,
      "loss": 5.1662,
      "mean_token_accuracy": 0.1792393296957016,
      "num_tokens": 22854489.0,
      "step": 13175
    },
    {
      "entropy": 5.501243448257446,
      "epoch": 1.0254425209103288,
      "grad_norm": 1.375,
      "learning_rate": 0.0004899478644807068,
      "loss": 5.3568,
      "mean_token_accuracy": 0.16961019486188889,
      "num_tokens": 22863269.0,
      "step": 13180
    },
    {
      "entropy": 5.5464643955230715,
      "epoch": 1.0258315502820463,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004899396712007498,
      "loss": 5.3266,
      "mean_token_accuracy": 0.17069934457540512,
      "num_tokens": 22872304.0,
      "step": 13185
    },
    {
      "entropy": 5.457087469100952,
      "epoch": 1.0262205796537638,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004899314746594256,
      "loss": 5.1377,
      "mean_token_accuracy": 0.17494407296180725,
      "num_tokens": 22881409.0,
      "step": 13190
    },
    {
      "entropy": 5.468878316879272,
      "epoch": 1.0266096090254815,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004899232748568584,
      "loss": 5.238,
      "mean_token_accuracy": 0.1789081260561943,
      "num_tokens": 22889560.0,
      "step": 13195
    },
    {
      "entropy": 5.444890642166138,
      "epoch": 1.026998638397199,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004899150717931724,
      "loss": 5.3023,
      "mean_token_accuracy": 0.1670473337173462,
      "num_tokens": 22897434.0,
      "step": 13200
    },
    {
      "entropy": 5.522672700881958,
      "epoch": 1.0273876677689167,
      "grad_norm": 1.375,
      "learning_rate": 0.0004899068654684924,
      "loss": 5.2821,
      "mean_token_accuracy": 0.16946140974760054,
      "num_tokens": 22906050.0,
      "step": 13205
    },
    {
      "entropy": 5.512297344207764,
      "epoch": 1.0277766971406341,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004898986558829428,
      "loss": 5.2388,
      "mean_token_accuracy": 0.17587502151727677,
      "num_tokens": 22914118.0,
      "step": 13210
    },
    {
      "entropy": 5.534115409851074,
      "epoch": 1.0281657265123516,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004898904430366479,
      "loss": 5.248,
      "mean_token_accuracy": 0.1714159816503525,
      "num_tokens": 22922539.0,
      "step": 13215
    },
    {
      "entropy": 5.449799108505249,
      "epoch": 1.0285547558840693,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004898822269297328,
      "loss": 5.2477,
      "mean_token_accuracy": 0.17054136246442794,
      "num_tokens": 22932371.0,
      "step": 13220
    },
    {
      "entropy": 5.513401746749878,
      "epoch": 1.0289437852557868,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004898740075623218,
      "loss": 5.285,
      "mean_token_accuracy": 0.17473663985729218,
      "num_tokens": 22940608.0,
      "step": 13225
    },
    {
      "entropy": 5.567580366134644,
      "epoch": 1.0293328146275045,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004898657849345399,
      "loss": 5.2776,
      "mean_token_accuracy": 0.17156692147254943,
      "num_tokens": 22948905.0,
      "step": 13230
    },
    {
      "entropy": 5.455373287200928,
      "epoch": 1.029721843999222,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004898575590465116,
      "loss": 5.2375,
      "mean_token_accuracy": 0.1757283017039299,
      "num_tokens": 22957218.0,
      "step": 13235
    },
    {
      "entropy": 5.470862150192261,
      "epoch": 1.0301108733709394,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004898493298983619,
      "loss": 5.3774,
      "mean_token_accuracy": 0.17307137846946716,
      "num_tokens": 22965700.0,
      "step": 13240
    },
    {
      "entropy": 5.592911100387573,
      "epoch": 1.0304999027426571,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004898410974902155,
      "loss": 5.3185,
      "mean_token_accuracy": 0.17168628424406052,
      "num_tokens": 22973996.0,
      "step": 13245
    },
    {
      "entropy": 5.596234226226807,
      "epoch": 1.0308889321143746,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004898328618221974,
      "loss": 5.3152,
      "mean_token_accuracy": 0.1673495277762413,
      "num_tokens": 22983142.0,
      "step": 13250
    },
    {
      "entropy": 5.489285326004028,
      "epoch": 1.0312779614860923,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004898246228944327,
      "loss": 5.2474,
      "mean_token_accuracy": 0.17086656391620636,
      "num_tokens": 22991263.0,
      "step": 13255
    },
    {
      "entropy": 5.544483184814453,
      "epoch": 1.0316669908578098,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004898163807070461,
      "loss": 5.3539,
      "mean_token_accuracy": 0.1660310670733452,
      "num_tokens": 23000068.0,
      "step": 13260
    },
    {
      "entropy": 5.557637405395508,
      "epoch": 1.0320560202295272,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004898081352601629,
      "loss": 5.2436,
      "mean_token_accuracy": 0.17484168112277984,
      "num_tokens": 23007935.0,
      "step": 13265
    },
    {
      "entropy": 5.491690015792846,
      "epoch": 1.032445049601245,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004897998865539082,
      "loss": 5.3207,
      "mean_token_accuracy": 0.1596148669719696,
      "num_tokens": 23017462.0,
      "step": 13270
    },
    {
      "entropy": 5.54883918762207,
      "epoch": 1.0328340789729624,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004897916345884069,
      "loss": 5.3166,
      "mean_token_accuracy": 0.16747146099805832,
      "num_tokens": 23026681.0,
      "step": 13275
    },
    {
      "entropy": 5.498556852340698,
      "epoch": 1.0332231083446801,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004897833793637847,
      "loss": 5.2779,
      "mean_token_accuracy": 0.16919007003307343,
      "num_tokens": 23035720.0,
      "step": 13280
    },
    {
      "entropy": 5.564071178436279,
      "epoch": 1.0336121377163976,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004897751208801665,
      "loss": 5.2981,
      "mean_token_accuracy": 0.17060740143060685,
      "num_tokens": 23044491.0,
      "step": 13285
    },
    {
      "entropy": 5.524331426620483,
      "epoch": 1.034001167088115,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004897668591376777,
      "loss": 5.29,
      "mean_token_accuracy": 0.17487076222896575,
      "num_tokens": 23054034.0,
      "step": 13290
    },
    {
      "entropy": 5.509965801239014,
      "epoch": 1.0343901964598328,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004897585941364436,
      "loss": 5.2721,
      "mean_token_accuracy": 0.16823101937770843,
      "num_tokens": 23062478.0,
      "step": 13295
    },
    {
      "entropy": 5.489333009719848,
      "epoch": 1.0347792258315502,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004897503258765896,
      "loss": 5.2828,
      "mean_token_accuracy": 0.16752099394798278,
      "num_tokens": 23070776.0,
      "step": 13300
    },
    {
      "entropy": 5.5575519561767575,
      "epoch": 1.035168255203268,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004897420543582413,
      "loss": 5.327,
      "mean_token_accuracy": 0.16969966441392897,
      "num_tokens": 23078459.0,
      "step": 13305
    },
    {
      "entropy": 5.555881357192993,
      "epoch": 1.0355572845749854,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000489733779581524,
      "loss": 5.2934,
      "mean_token_accuracy": 0.17586043775081633,
      "num_tokens": 23087113.0,
      "step": 13310
    },
    {
      "entropy": 5.455705165863037,
      "epoch": 1.035946313946703,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004897255015465635,
      "loss": 5.2149,
      "mean_token_accuracy": 0.17389488816261292,
      "num_tokens": 23095989.0,
      "step": 13315
    },
    {
      "entropy": 5.467445755004883,
      "epoch": 1.0363353433184206,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004897172202534853,
      "loss": 5.2574,
      "mean_token_accuracy": 0.17328002601861953,
      "num_tokens": 23104578.0,
      "step": 13320
    },
    {
      "entropy": 5.556106424331665,
      "epoch": 1.036724372690138,
      "grad_norm": 1.28125,
      "learning_rate": 0.000489708935702415,
      "loss": 5.3105,
      "mean_token_accuracy": 0.17145204842090606,
      "num_tokens": 23114390.0,
      "step": 13325
    },
    {
      "entropy": 5.574456214904785,
      "epoch": 1.0371134020618558,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004897006478934784,
      "loss": 5.1953,
      "mean_token_accuracy": 0.1771734520792961,
      "num_tokens": 23122411.0,
      "step": 13330
    },
    {
      "entropy": 5.386401987075805,
      "epoch": 1.0375024314335732,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004896923568268011,
      "loss": 5.1519,
      "mean_token_accuracy": 0.17943936586380005,
      "num_tokens": 23130801.0,
      "step": 13335
    },
    {
      "entropy": 5.464883852005005,
      "epoch": 1.0378914608052907,
      "grad_norm": 1.34375,
      "learning_rate": 0.000489684062502509,
      "loss": 5.2025,
      "mean_token_accuracy": 0.1755169689655304,
      "num_tokens": 23138578.0,
      "step": 13340
    },
    {
      "entropy": 5.497285652160644,
      "epoch": 1.0382804901770084,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004896757649207281,
      "loss": 5.28,
      "mean_token_accuracy": 0.17274308800697327,
      "num_tokens": 23147135.0,
      "step": 13345
    },
    {
      "entropy": 5.54693341255188,
      "epoch": 1.0386695195487259,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000489667464081584,
      "loss": 5.3686,
      "mean_token_accuracy": 0.17182284146547316,
      "num_tokens": 23155791.0,
      "step": 13350
    },
    {
      "entropy": 5.534262180328369,
      "epoch": 1.0390585489204436,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004896591599852029,
      "loss": 5.2084,
      "mean_token_accuracy": 0.17880017161369324,
      "num_tokens": 23163674.0,
      "step": 13355
    },
    {
      "entropy": 5.49973578453064,
      "epoch": 1.039447578292161,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004896508526317107,
      "loss": 5.2789,
      "mean_token_accuracy": 0.16380516886711122,
      "num_tokens": 23171931.0,
      "step": 13360
    },
    {
      "entropy": 5.558356094360351,
      "epoch": 1.0398366076638785,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004896425420212334,
      "loss": 5.2894,
      "mean_token_accuracy": 0.1703491285443306,
      "num_tokens": 23180195.0,
      "step": 13365
    },
    {
      "entropy": 5.470646047592163,
      "epoch": 1.0402256370355962,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004896342281538973,
      "loss": 5.2178,
      "mean_token_accuracy": 0.17389285266399385,
      "num_tokens": 23189486.0,
      "step": 13370
    },
    {
      "entropy": 5.4824117660522464,
      "epoch": 1.0406146664073137,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004896259110298283,
      "loss": 5.3438,
      "mean_token_accuracy": 0.17094234228134156,
      "num_tokens": 23198055.0,
      "step": 13375
    },
    {
      "entropy": 5.51085844039917,
      "epoch": 1.0410036957790314,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004896175906491528,
      "loss": 5.2216,
      "mean_token_accuracy": 0.17861904948949814,
      "num_tokens": 23206092.0,
      "step": 13380
    },
    {
      "entropy": 5.510445785522461,
      "epoch": 1.0413927251507489,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004896092670119968,
      "loss": 5.3057,
      "mean_token_accuracy": 0.17596572488546372,
      "num_tokens": 23214820.0,
      "step": 13385
    },
    {
      "entropy": 5.43936767578125,
      "epoch": 1.0417817545224664,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004896009401184869,
      "loss": 5.2415,
      "mean_token_accuracy": 0.17504541724920272,
      "num_tokens": 23224033.0,
      "step": 13390
    },
    {
      "entropy": 5.5084155082702635,
      "epoch": 1.042170783894184,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004895926099687493,
      "loss": 5.2362,
      "mean_token_accuracy": 0.17828564941883088,
      "num_tokens": 23233074.0,
      "step": 13395
    },
    {
      "entropy": 5.433311700820923,
      "epoch": 1.0425598132659015,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004895842765629104,
      "loss": 5.2096,
      "mean_token_accuracy": 0.17888783067464828,
      "num_tokens": 23241167.0,
      "step": 13400
    },
    {
      "entropy": 5.509353160858154,
      "epoch": 1.0429488426376192,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004895759399010966,
      "loss": 5.2159,
      "mean_token_accuracy": 0.1758016437292099,
      "num_tokens": 23249601.0,
      "step": 13405
    },
    {
      "entropy": 5.524061107635498,
      "epoch": 1.0433378720093367,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004895675999834345,
      "loss": 5.3189,
      "mean_token_accuracy": 0.16395875662565232,
      "num_tokens": 23257868.0,
      "step": 13410
    },
    {
      "entropy": 5.532989740371704,
      "epoch": 1.0437269013810542,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004895592568100505,
      "loss": 5.3145,
      "mean_token_accuracy": 0.1718521773815155,
      "num_tokens": 23266586.0,
      "step": 13415
    },
    {
      "entropy": 5.5586975574493405,
      "epoch": 1.0441159307527719,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004895509103810714,
      "loss": 5.285,
      "mean_token_accuracy": 0.17386317551136016,
      "num_tokens": 23275114.0,
      "step": 13420
    },
    {
      "entropy": 5.411413860321045,
      "epoch": 1.0445049601244893,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004895425606966237,
      "loss": 5.1788,
      "mean_token_accuracy": 0.18079530149698259,
      "num_tokens": 23283200.0,
      "step": 13425
    },
    {
      "entropy": 5.500542402267456,
      "epoch": 1.044893989496207,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000489534207756834,
      "loss": 5.3766,
      "mean_token_accuracy": 0.1665601462125778,
      "num_tokens": 23291999.0,
      "step": 13430
    },
    {
      "entropy": 5.543798732757568,
      "epoch": 1.0452830188679245,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004895258515618293,
      "loss": 5.2593,
      "mean_token_accuracy": 0.18120337873697281,
      "num_tokens": 23300514.0,
      "step": 13435
    },
    {
      "entropy": 5.517561197280884,
      "epoch": 1.045672048239642,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004895174921117362,
      "loss": 5.2055,
      "mean_token_accuracy": 0.1731306105852127,
      "num_tokens": 23309543.0,
      "step": 13440
    },
    {
      "entropy": 5.401105213165283,
      "epoch": 1.0460610776113597,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004895091294066816,
      "loss": 5.2431,
      "mean_token_accuracy": 0.17738472521305085,
      "num_tokens": 23319002.0,
      "step": 13445
    },
    {
      "entropy": 5.536071681976319,
      "epoch": 1.0464501069830772,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004895007634467924,
      "loss": 5.2974,
      "mean_token_accuracy": 0.16986297965049743,
      "num_tokens": 23326604.0,
      "step": 13450
    },
    {
      "entropy": 5.463508796691895,
      "epoch": 1.0468391363547949,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004894923942321955,
      "loss": 5.2353,
      "mean_token_accuracy": 0.18364446014165878,
      "num_tokens": 23335492.0,
      "step": 13455
    },
    {
      "entropy": 5.467422437667847,
      "epoch": 1.0472281657265123,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004894840217630179,
      "loss": 5.2389,
      "mean_token_accuracy": 0.17657190561294556,
      "num_tokens": 23343693.0,
      "step": 13460
    },
    {
      "entropy": 5.435745191574097,
      "epoch": 1.0476171950982298,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004894756460393868,
      "loss": 5.2227,
      "mean_token_accuracy": 0.17957899123430252,
      "num_tokens": 23352587.0,
      "step": 13465
    },
    {
      "entropy": 5.504619073867798,
      "epoch": 1.0480062244699475,
      "grad_norm": 1.25,
      "learning_rate": 0.000489467267061429,
      "loss": 5.283,
      "mean_token_accuracy": 0.18057815879583358,
      "num_tokens": 23361293.0,
      "step": 13470
    },
    {
      "entropy": 5.5328404903411865,
      "epoch": 1.048395253841665,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004894588848292718,
      "loss": 5.2866,
      "mean_token_accuracy": 0.17249845266342162,
      "num_tokens": 23369663.0,
      "step": 13475
    },
    {
      "entropy": 5.4785009860992435,
      "epoch": 1.0487842832133827,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004894504993430425,
      "loss": 5.2032,
      "mean_token_accuracy": 0.17163490653038024,
      "num_tokens": 23378224.0,
      "step": 13480
    },
    {
      "entropy": 5.5824049949646,
      "epoch": 1.0491733125851002,
      "grad_norm": 1.5,
      "learning_rate": 0.000489442110602868,
      "loss": 5.3606,
      "mean_token_accuracy": 0.16755372434854507,
      "num_tokens": 23386693.0,
      "step": 13485
    },
    {
      "entropy": 5.531474494934082,
      "epoch": 1.0495623419568176,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004894337186088759,
      "loss": 5.236,
      "mean_token_accuracy": 0.1770475134253502,
      "num_tokens": 23395226.0,
      "step": 13490
    },
    {
      "entropy": 5.454203271865845,
      "epoch": 1.0499513713285353,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004894253233611934,
      "loss": 5.2495,
      "mean_token_accuracy": 0.17192675024271012,
      "num_tokens": 23403779.0,
      "step": 13495
    },
    {
      "entropy": 5.531600475311279,
      "epoch": 1.0503404007002528,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000489416924859948,
      "loss": 5.3623,
      "mean_token_accuracy": 0.17131223231554032,
      "num_tokens": 23411807.0,
      "step": 13500
    },
    {
      "entropy": 5.510690307617187,
      "epoch": 1.0507294300719705,
      "grad_norm": 1.34375,
      "learning_rate": 0.000489408523105267,
      "loss": 5.2438,
      "mean_token_accuracy": 0.1758344739675522,
      "num_tokens": 23420323.0,
      "step": 13505
    },
    {
      "entropy": 5.463551664352417,
      "epoch": 1.051118459443688,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004894001180972779,
      "loss": 5.2193,
      "mean_token_accuracy": 0.17467987835407256,
      "num_tokens": 23429220.0,
      "step": 13510
    },
    {
      "entropy": 5.460594129562378,
      "epoch": 1.0515074888154055,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004893917098361082,
      "loss": 5.2543,
      "mean_token_accuracy": 0.18091317117214203,
      "num_tokens": 23438471.0,
      "step": 13515
    },
    {
      "entropy": 5.5224944114685055,
      "epoch": 1.0518965181871232,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004893832983218856,
      "loss": 5.3227,
      "mean_token_accuracy": 0.17846597135066986,
      "num_tokens": 23447504.0,
      "step": 13520
    },
    {
      "entropy": 5.394408750534057,
      "epoch": 1.0522855475588406,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004893748835547377,
      "loss": 5.0663,
      "mean_token_accuracy": 0.18594157248735427,
      "num_tokens": 23456065.0,
      "step": 13525
    },
    {
      "entropy": 5.447011280059814,
      "epoch": 1.0526745769305583,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004893664655347921,
      "loss": 5.2544,
      "mean_token_accuracy": 0.1811427116394043,
      "num_tokens": 23463826.0,
      "step": 13530
    },
    {
      "entropy": 5.421354866027832,
      "epoch": 1.0530636063022758,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004893580442621767,
      "loss": 5.211,
      "mean_token_accuracy": 0.1802362933754921,
      "num_tokens": 23472092.0,
      "step": 13535
    },
    {
      "entropy": 5.540503644943238,
      "epoch": 1.0534526356739935,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000489349619737019,
      "loss": 5.3135,
      "mean_token_accuracy": 0.16885752528905867,
      "num_tokens": 23480934.0,
      "step": 13540
    },
    {
      "entropy": 5.53750433921814,
      "epoch": 1.053841665045711,
      "grad_norm": 1.21875,
      "learning_rate": 0.000489341191959447,
      "loss": 5.2663,
      "mean_token_accuracy": 0.17121487110853195,
      "num_tokens": 23490187.0,
      "step": 13545
    },
    {
      "entropy": 5.4766397953033445,
      "epoch": 1.0542306944174284,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004893327609295887,
      "loss": 5.2169,
      "mean_token_accuracy": 0.17307864427566527,
      "num_tokens": 23498330.0,
      "step": 13550
    },
    {
      "entropy": 5.404174518585205,
      "epoch": 1.0546197237891461,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004893243266475719,
      "loss": 5.1474,
      "mean_token_accuracy": 0.17973985970020295,
      "num_tokens": 23506219.0,
      "step": 13555
    },
    {
      "entropy": 5.466575384140015,
      "epoch": 1.0550087531608636,
      "grad_norm": 1.25,
      "learning_rate": 0.0004893158891135245,
      "loss": 5.2322,
      "mean_token_accuracy": 0.17707367539405822,
      "num_tokens": 23514585.0,
      "step": 13560
    },
    {
      "entropy": 5.51822190284729,
      "epoch": 1.055397782532581,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004893074483275746,
      "loss": 5.1952,
      "mean_token_accuracy": 0.17982145845890046,
      "num_tokens": 23522919.0,
      "step": 13565
    },
    {
      "entropy": 5.526120615005493,
      "epoch": 1.0557868119042988,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004892990042898503,
      "loss": 5.3968,
      "mean_token_accuracy": 0.16582975089550017,
      "num_tokens": 23531613.0,
      "step": 13570
    },
    {
      "entropy": 5.5679426193237305,
      "epoch": 1.0561758412760163,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004892905570004798,
      "loss": 5.3437,
      "mean_token_accuracy": 0.1721174895763397,
      "num_tokens": 23539865.0,
      "step": 13575
    },
    {
      "entropy": 5.458710861206055,
      "epoch": 1.056564870647734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000489282106459591,
      "loss": 5.1761,
      "mean_token_accuracy": 0.18362881541252135,
      "num_tokens": 23548424.0,
      "step": 13580
    },
    {
      "entropy": 5.52537841796875,
      "epoch": 1.0569539000194514,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004892736526673124,
      "loss": 5.2714,
      "mean_token_accuracy": 0.17386070936918258,
      "num_tokens": 23557199.0,
      "step": 13585
    },
    {
      "entropy": 5.530455684661865,
      "epoch": 1.0573429293911691,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004892651956237721,
      "loss": 5.384,
      "mean_token_accuracy": 0.16503250002861022,
      "num_tokens": 23565979.0,
      "step": 13590
    },
    {
      "entropy": 5.496713256835937,
      "epoch": 1.0577319587628866,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004892567353290986,
      "loss": 5.2516,
      "mean_token_accuracy": 0.17182186096906663,
      "num_tokens": 23575405.0,
      "step": 13595
    },
    {
      "entropy": 5.605249977111816,
      "epoch": 1.058120988134604,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004892482717834201,
      "loss": 5.3893,
      "mean_token_accuracy": 0.1602438807487488,
      "num_tokens": 23584589.0,
      "step": 13600
    },
    {
      "entropy": 5.519269514083862,
      "epoch": 1.0585100175063218,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004892398049868651,
      "loss": 5.2302,
      "mean_token_accuracy": 0.17371235191822051,
      "num_tokens": 23593079.0,
      "step": 13605
    },
    {
      "entropy": 5.532129001617432,
      "epoch": 1.0588990468780393,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000489231334939562,
      "loss": 5.3264,
      "mean_token_accuracy": 0.1730417564511299,
      "num_tokens": 23602010.0,
      "step": 13610
    },
    {
      "entropy": 5.543614387512207,
      "epoch": 1.059288076249757,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004892228616416395,
      "loss": 5.2735,
      "mean_token_accuracy": 0.17420105785131454,
      "num_tokens": 23610707.0,
      "step": 13615
    },
    {
      "entropy": 5.446663951873779,
      "epoch": 1.0596771056214744,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004892143850932259,
      "loss": 5.253,
      "mean_token_accuracy": 0.1703906774520874,
      "num_tokens": 23619579.0,
      "step": 13620
    },
    {
      "entropy": 5.487177038192749,
      "epoch": 1.060066134993192,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048920590529445,
      "loss": 5.2262,
      "mean_token_accuracy": 0.1814274311065674,
      "num_tokens": 23628407.0,
      "step": 13625
    },
    {
      "entropy": 5.460596513748169,
      "epoch": 1.0604551643649096,
      "grad_norm": 1.5,
      "learning_rate": 0.0004891974222454406,
      "loss": 5.1566,
      "mean_token_accuracy": 0.18144096434116364,
      "num_tokens": 23637094.0,
      "step": 13630
    },
    {
      "entropy": 5.435451555252075,
      "epoch": 1.060844193736627,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004891889359463261,
      "loss": 5.2304,
      "mean_token_accuracy": 0.17880595177412034,
      "num_tokens": 23645823.0,
      "step": 13635
    },
    {
      "entropy": 5.532078266143799,
      "epoch": 1.0612332231083448,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004891804463972354,
      "loss": 5.3348,
      "mean_token_accuracy": 0.16802839189767838,
      "num_tokens": 23655027.0,
      "step": 13640
    },
    {
      "entropy": 5.604687070846557,
      "epoch": 1.0616222524800623,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004891719535982974,
      "loss": 5.3961,
      "mean_token_accuracy": 0.16401204764842986,
      "num_tokens": 23663840.0,
      "step": 13645
    },
    {
      "entropy": 5.570868253707886,
      "epoch": 1.0620112818517797,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004891634575496408,
      "loss": 5.3475,
      "mean_token_accuracy": 0.16477967649698258,
      "num_tokens": 23672795.0,
      "step": 13650
    },
    {
      "entropy": 5.501248788833618,
      "epoch": 1.0624003112234974,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004891549582513946,
      "loss": 5.2616,
      "mean_token_accuracy": 0.17394069582223892,
      "num_tokens": 23681833.0,
      "step": 13655
    },
    {
      "entropy": 5.517497205734253,
      "epoch": 1.062789340595215,
      "grad_norm": 1.34375,
      "learning_rate": 0.000489146455703688,
      "loss": 5.2373,
      "mean_token_accuracy": 0.1761193208396435,
      "num_tokens": 23689939.0,
      "step": 13660
    },
    {
      "entropy": 5.494831895828247,
      "epoch": 1.0631783699669326,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004891379499066495,
      "loss": 5.2719,
      "mean_token_accuracy": 0.17255582362413407,
      "num_tokens": 23698535.0,
      "step": 13665
    },
    {
      "entropy": 5.493082523345947,
      "epoch": 1.06356739933865,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004891294408604087,
      "loss": 5.2121,
      "mean_token_accuracy": 0.17645900845527648,
      "num_tokens": 23707749.0,
      "step": 13670
    },
    {
      "entropy": 5.5798718452453615,
      "epoch": 1.0639564287103676,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004891209285650942,
      "loss": 5.3679,
      "mean_token_accuracy": 0.1711876004934311,
      "num_tokens": 23716503.0,
      "step": 13675
    },
    {
      "entropy": 5.571073198318482,
      "epoch": 1.0643454580820852,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004891124130208355,
      "loss": 5.3129,
      "mean_token_accuracy": 0.16990765184164047,
      "num_tokens": 23724882.0,
      "step": 13680
    },
    {
      "entropy": 5.461633253097534,
      "epoch": 1.0647344874538027,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004891038942277618,
      "loss": 5.2942,
      "mean_token_accuracy": 0.18063681572675705,
      "num_tokens": 23733680.0,
      "step": 13685
    },
    {
      "entropy": 5.505451297760009,
      "epoch": 1.0651235168255204,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004890953721860022,
      "loss": 5.225,
      "mean_token_accuracy": 0.1805526450276375,
      "num_tokens": 23742066.0,
      "step": 13690
    },
    {
      "entropy": 5.473612546920776,
      "epoch": 1.065512546197238,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004890868468956862,
      "loss": 5.1434,
      "mean_token_accuracy": 0.18179723918437957,
      "num_tokens": 23750558.0,
      "step": 13695
    },
    {
      "entropy": 5.435284614562988,
      "epoch": 1.0659015755689554,
      "grad_norm": 1.234375,
      "learning_rate": 0.000489078318356943,
      "loss": 5.2118,
      "mean_token_accuracy": 0.17735994160175322,
      "num_tokens": 23759120.0,
      "step": 13700
    },
    {
      "entropy": 5.446619606018066,
      "epoch": 1.066290604940673,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004890697865699021,
      "loss": 5.2515,
      "mean_token_accuracy": 0.17786474227905275,
      "num_tokens": 23768301.0,
      "step": 13705
    },
    {
      "entropy": 5.496279764175415,
      "epoch": 1.0666796343123905,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004890612515346929,
      "loss": 5.1959,
      "mean_token_accuracy": 0.1822264760732651,
      "num_tokens": 23776155.0,
      "step": 13710
    },
    {
      "entropy": 5.519058179855347,
      "epoch": 1.0670686636841082,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004890527132514448,
      "loss": 5.3117,
      "mean_token_accuracy": 0.17042991667985916,
      "num_tokens": 23785363.0,
      "step": 13715
    },
    {
      "entropy": 5.4911857604980465,
      "epoch": 1.0674576930558257,
      "grad_norm": 1.375,
      "learning_rate": 0.0004890441717202876,
      "loss": 5.2546,
      "mean_token_accuracy": 0.1783105954527855,
      "num_tokens": 23793817.0,
      "step": 13720
    },
    {
      "entropy": 5.430178880691528,
      "epoch": 1.0678467224275432,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004890356269413507,
      "loss": 5.297,
      "mean_token_accuracy": 0.17501659393310548,
      "num_tokens": 23802629.0,
      "step": 13725
    },
    {
      "entropy": 5.39081392288208,
      "epoch": 1.068235751799261,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000489027078914764,
      "loss": 5.0866,
      "mean_token_accuracy": 0.18784652054309844,
      "num_tokens": 23811178.0,
      "step": 13730
    },
    {
      "entropy": 5.500747299194336,
      "epoch": 1.0686247811709784,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004890185276406569,
      "loss": 5.3188,
      "mean_token_accuracy": 0.1760838136076927,
      "num_tokens": 23819685.0,
      "step": 13735
    },
    {
      "entropy": 5.4168542385101315,
      "epoch": 1.069013810542696,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004890099731191592,
      "loss": 5.1948,
      "mean_token_accuracy": 0.1728578805923462,
      "num_tokens": 23828733.0,
      "step": 13740
    },
    {
      "entropy": 5.605859375,
      "epoch": 1.0694028399144135,
      "grad_norm": 1.28125,
      "learning_rate": 0.000489001415350401,
      "loss": 5.3642,
      "mean_token_accuracy": 0.17208168059587478,
      "num_tokens": 23837095.0,
      "step": 13745
    },
    {
      "entropy": 5.543181371688843,
      "epoch": 1.069791869286131,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004889928543345118,
      "loss": 5.3255,
      "mean_token_accuracy": 0.1667383372783661,
      "num_tokens": 23846637.0,
      "step": 13750
    },
    {
      "entropy": 5.509400796890259,
      "epoch": 1.0701808986578487,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004889842900716217,
      "loss": 5.3869,
      "mean_token_accuracy": 0.16680870205163956,
      "num_tokens": 23856042.0,
      "step": 13755
    },
    {
      "entropy": 5.574008941650391,
      "epoch": 1.0705699280295662,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004889757225618606,
      "loss": 5.3408,
      "mean_token_accuracy": 0.16003026068210602,
      "num_tokens": 23865367.0,
      "step": 13760
    },
    {
      "entropy": 5.522847366333008,
      "epoch": 1.0709589574012839,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004889671518053584,
      "loss": 5.3187,
      "mean_token_accuracy": 0.17168945223093032,
      "num_tokens": 23874740.0,
      "step": 13765
    },
    {
      "entropy": 5.524417972564697,
      "epoch": 1.0713479867730014,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004889585778022453,
      "loss": 5.2836,
      "mean_token_accuracy": 0.17669813632965087,
      "num_tokens": 23884229.0,
      "step": 13770
    },
    {
      "entropy": 5.4255897998809814,
      "epoch": 1.0717370161447188,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004889500005526514,
      "loss": 5.2147,
      "mean_token_accuracy": 0.1769952341914177,
      "num_tokens": 23892149.0,
      "step": 13775
    },
    {
      "entropy": 5.56971845626831,
      "epoch": 1.0721260455164365,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004889414200567067,
      "loss": 5.4324,
      "mean_token_accuracy": 0.16193339228630066,
      "num_tokens": 23901802.0,
      "step": 13780
    },
    {
      "entropy": 5.5532365322113035,
      "epoch": 1.072515074888154,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004889328363145415,
      "loss": 5.2801,
      "mean_token_accuracy": 0.1727521687746048,
      "num_tokens": 23910268.0,
      "step": 13785
    },
    {
      "entropy": 5.407101678848266,
      "epoch": 1.0729041042598717,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004889242493262859,
      "loss": 5.1152,
      "mean_token_accuracy": 0.17757045924663545,
      "num_tokens": 23918351.0,
      "step": 13790
    },
    {
      "entropy": 5.479969167709351,
      "epoch": 1.0732931336315892,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004889156590920704,
      "loss": 5.1716,
      "mean_token_accuracy": 0.17735549956560134,
      "num_tokens": 23927367.0,
      "step": 13795
    },
    {
      "entropy": 5.48079628944397,
      "epoch": 1.0736821630033067,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004889070656120253,
      "loss": 5.2769,
      "mean_token_accuracy": 0.17862653881311416,
      "num_tokens": 23936216.0,
      "step": 13800
    },
    {
      "entropy": 5.576433610916138,
      "epoch": 1.0740711923750244,
      "grad_norm": 1.25,
      "learning_rate": 0.0004888984688862809,
      "loss": 5.3312,
      "mean_token_accuracy": 0.16522572189569473,
      "num_tokens": 23944344.0,
      "step": 13805
    },
    {
      "entropy": 5.560266971588135,
      "epoch": 1.0744602217467418,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004888898689149677,
      "loss": 5.3648,
      "mean_token_accuracy": 0.16149822473526002,
      "num_tokens": 23953390.0,
      "step": 13810
    },
    {
      "entropy": 5.636888551712036,
      "epoch": 1.0748492511184595,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004888812656982162,
      "loss": 5.3475,
      "mean_token_accuracy": 0.17412169277668,
      "num_tokens": 23962230.0,
      "step": 13815
    },
    {
      "entropy": 5.4780816555023195,
      "epoch": 1.075238280490177,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004888726592361569,
      "loss": 5.2392,
      "mean_token_accuracy": 0.17245784550905227,
      "num_tokens": 23971627.0,
      "step": 13820
    },
    {
      "entropy": 5.403182125091552,
      "epoch": 1.0756273098618945,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004888640495289204,
      "loss": 5.2306,
      "mean_token_accuracy": 0.18453577756881714,
      "num_tokens": 23980222.0,
      "step": 13825
    },
    {
      "entropy": 5.468798685073852,
      "epoch": 1.0760163392336122,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004888554365766374,
      "loss": 5.2997,
      "mean_token_accuracy": 0.16826647371053696,
      "num_tokens": 23988803.0,
      "step": 13830
    },
    {
      "entropy": 5.553125333786011,
      "epoch": 1.0764053686053296,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004888468203794385,
      "loss": 5.2835,
      "mean_token_accuracy": 0.17038446962833403,
      "num_tokens": 23997664.0,
      "step": 13835
    },
    {
      "entropy": 5.508232927322387,
      "epoch": 1.0767943979770473,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004888382009374545,
      "loss": 5.2687,
      "mean_token_accuracy": 0.1785341501235962,
      "num_tokens": 24006641.0,
      "step": 13840
    },
    {
      "entropy": 5.473012018203735,
      "epoch": 1.0771834273487648,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004888295782508161,
      "loss": 5.2178,
      "mean_token_accuracy": 0.1712936520576477,
      "num_tokens": 24014852.0,
      "step": 13845
    },
    {
      "entropy": 5.544522380828857,
      "epoch": 1.0775724567204823,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004888209523196542,
      "loss": 5.2434,
      "mean_token_accuracy": 0.18076094835996628,
      "num_tokens": 24023441.0,
      "step": 13850
    },
    {
      "entropy": 5.538372802734375,
      "epoch": 1.0779614860922,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004888123231440997,
      "loss": 5.2559,
      "mean_token_accuracy": 0.17686181962490083,
      "num_tokens": 24032120.0,
      "step": 13855
    },
    {
      "entropy": 5.506207084655761,
      "epoch": 1.0783505154639175,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004888036907242834,
      "loss": 5.2997,
      "mean_token_accuracy": 0.17518354058265687,
      "num_tokens": 24040264.0,
      "step": 13860
    },
    {
      "entropy": 5.594097757339478,
      "epoch": 1.0787395448356352,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004887950550603366,
      "loss": 5.3469,
      "mean_token_accuracy": 0.1704188108444214,
      "num_tokens": 24049194.0,
      "step": 13865
    },
    {
      "entropy": 5.505089044570923,
      "epoch": 1.0791285742073526,
      "grad_norm": 1.25,
      "learning_rate": 0.0004887864161523901,
      "loss": 5.2346,
      "mean_token_accuracy": 0.17036609798669816,
      "num_tokens": 24058004.0,
      "step": 13870
    },
    {
      "entropy": 5.371783590316772,
      "epoch": 1.0795176035790701,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004887777740005749,
      "loss": 5.1799,
      "mean_token_accuracy": 0.17829613834619523,
      "num_tokens": 24066871.0,
      "step": 13875
    },
    {
      "entropy": 5.47926959991455,
      "epoch": 1.0799066329507878,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004887691286050224,
      "loss": 5.2837,
      "mean_token_accuracy": 0.17447290420532227,
      "num_tokens": 24075409.0,
      "step": 13880
    },
    {
      "entropy": 5.541545486450195,
      "epoch": 1.0802956623225053,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004887604799658635,
      "loss": 5.2212,
      "mean_token_accuracy": 0.1748093843460083,
      "num_tokens": 24083983.0,
      "step": 13885
    },
    {
      "entropy": 5.423152208328247,
      "epoch": 1.080684691694223,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004887518280832295,
      "loss": 5.2536,
      "mean_token_accuracy": 0.17496563196182252,
      "num_tokens": 24092225.0,
      "step": 13890
    },
    {
      "entropy": 5.437602233886719,
      "epoch": 1.0810737210659405,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004887431729572519,
      "loss": 5.1638,
      "mean_token_accuracy": 0.18181942403316498,
      "num_tokens": 24100045.0,
      "step": 13895
    },
    {
      "entropy": 5.46027250289917,
      "epoch": 1.081462750437658,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004887345145880617,
      "loss": 5.2338,
      "mean_token_accuracy": 0.18062204122543335,
      "num_tokens": 24108540.0,
      "step": 13900
    },
    {
      "entropy": 5.469130277633667,
      "epoch": 1.0818517798093756,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004887258529757904,
      "loss": 5.3244,
      "mean_token_accuracy": 0.17561854869127275,
      "num_tokens": 24117819.0,
      "step": 13905
    },
    {
      "entropy": 5.432076835632325,
      "epoch": 1.082240809181093,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004887171881205696,
      "loss": 5.1201,
      "mean_token_accuracy": 0.17849269658327102,
      "num_tokens": 24127000.0,
      "step": 13910
    },
    {
      "entropy": 5.494923210144043,
      "epoch": 1.0826298385528108,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004887085200225306,
      "loss": 5.2711,
      "mean_token_accuracy": 0.1713639721274376,
      "num_tokens": 24135404.0,
      "step": 13915
    },
    {
      "entropy": 5.396844100952149,
      "epoch": 1.0830188679245283,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004886998486818049,
      "loss": 5.0886,
      "mean_token_accuracy": 0.18566545248031616,
      "num_tokens": 24143944.0,
      "step": 13920
    },
    {
      "entropy": 5.495157480239868,
      "epoch": 1.083407897296246,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004886911740985242,
      "loss": 5.2216,
      "mean_token_accuracy": 0.17162090092897414,
      "num_tokens": 24153097.0,
      "step": 13925
    },
    {
      "entropy": 5.5088883399963375,
      "epoch": 1.0837969266679635,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048868249627282,
      "loss": 5.3595,
      "mean_token_accuracy": 0.1714622676372528,
      "num_tokens": 24162499.0,
      "step": 13930
    },
    {
      "entropy": 5.4504992961883545,
      "epoch": 1.084185956039681,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000488673815204824,
      "loss": 5.1796,
      "mean_token_accuracy": 0.18298446536064147,
      "num_tokens": 24170543.0,
      "step": 13935
    },
    {
      "entropy": 5.4536065578460695,
      "epoch": 1.0845749854113986,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004886651308946681,
      "loss": 5.2652,
      "mean_token_accuracy": 0.16802462935447693,
      "num_tokens": 24180871.0,
      "step": 13940
    },
    {
      "entropy": 5.432541036605835,
      "epoch": 1.084964014783116,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004886564433424839,
      "loss": 5.2052,
      "mean_token_accuracy": 0.17641855478286744,
      "num_tokens": 24189075.0,
      "step": 13945
    },
    {
      "entropy": 5.552456283569336,
      "epoch": 1.0853530441548336,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004886477525484032,
      "loss": 5.2933,
      "mean_token_accuracy": 0.17755922079086303,
      "num_tokens": 24197686.0,
      "step": 13950
    },
    {
      "entropy": 5.368551874160767,
      "epoch": 1.0857420735265513,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004886390585125579,
      "loss": 5.1781,
      "mean_token_accuracy": 0.17651353478431703,
      "num_tokens": 24207093.0,
      "step": 13955
    },
    {
      "entropy": 5.5616272449493405,
      "epoch": 1.0861311028982688,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004886303612350799,
      "loss": 5.3113,
      "mean_token_accuracy": 0.17161279767751694,
      "num_tokens": 24216447.0,
      "step": 13960
    },
    {
      "entropy": 5.5278544425964355,
      "epoch": 1.0865201322699864,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004886216607161013,
      "loss": 5.3273,
      "mean_token_accuracy": 0.16857061982154847,
      "num_tokens": 24225041.0,
      "step": 13965
    },
    {
      "entropy": 5.495896196365356,
      "epoch": 1.086909161641704,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004886129569557538,
      "loss": 5.2548,
      "mean_token_accuracy": 0.17146216183900834,
      "num_tokens": 24233686.0,
      "step": 13970
    },
    {
      "entropy": 5.498942422866821,
      "epoch": 1.0872981910134216,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004886042499541699,
      "loss": 5.2766,
      "mean_token_accuracy": 0.17957906126976014,
      "num_tokens": 24242553.0,
      "step": 13975
    },
    {
      "entropy": 5.480491638183594,
      "epoch": 1.087687220385139,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004885955397114813,
      "loss": 5.2066,
      "mean_token_accuracy": 0.17801113277673722,
      "num_tokens": 24251658.0,
      "step": 13980
    },
    {
      "entropy": 5.5748389720916744,
      "epoch": 1.0880762497568566,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004885868262278205,
      "loss": 5.3,
      "mean_token_accuracy": 0.1718540906906128,
      "num_tokens": 24260397.0,
      "step": 13985
    },
    {
      "entropy": 5.457262134552002,
      "epoch": 1.0884652791285743,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004885781095033195,
      "loss": 5.2203,
      "mean_token_accuracy": 0.17580918222665787,
      "num_tokens": 24268789.0,
      "step": 13990
    },
    {
      "entropy": 5.5068543434143065,
      "epoch": 1.0888543085002917,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004885693895381108,
      "loss": 5.3243,
      "mean_token_accuracy": 0.168991294503212,
      "num_tokens": 24278050.0,
      "step": 13995
    },
    {
      "entropy": 5.589233016967773,
      "epoch": 1.0892433378720092,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004885606663323263,
      "loss": 5.2837,
      "mean_token_accuracy": 0.1727209821343422,
      "num_tokens": 24286477.0,
      "step": 14000
    },
    {
      "entropy": 5.403928804397583,
      "epoch": 1.089632367243727,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004885519398860987,
      "loss": 5.163,
      "mean_token_accuracy": 0.17940218150615692,
      "num_tokens": 24295704.0,
      "step": 14005
    },
    {
      "entropy": 5.386621046066284,
      "epoch": 1.0900213966154444,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004885432101995604,
      "loss": 5.2026,
      "mean_token_accuracy": 0.1841410607099533,
      "num_tokens": 24304132.0,
      "step": 14010
    },
    {
      "entropy": 5.539301776885987,
      "epoch": 1.090410425987162,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004885344772728436,
      "loss": 5.2723,
      "mean_token_accuracy": 0.17559004276990892,
      "num_tokens": 24312864.0,
      "step": 14015
    },
    {
      "entropy": 5.523755168914795,
      "epoch": 1.0907994553588796,
      "grad_norm": 1.25,
      "learning_rate": 0.0004885257411060812,
      "loss": 5.2047,
      "mean_token_accuracy": 0.1783328980207443,
      "num_tokens": 24321745.0,
      "step": 14020
    },
    {
      "entropy": 5.401197624206543,
      "epoch": 1.0911884847305973,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004885170016994053,
      "loss": 5.2244,
      "mean_token_accuracy": 0.17089144587516786,
      "num_tokens": 24331061.0,
      "step": 14025
    },
    {
      "entropy": 5.4208807945251465,
      "epoch": 1.0915775141023147,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004885082590529489,
      "loss": 5.2333,
      "mean_token_accuracy": 0.1745702862739563,
      "num_tokens": 24339935.0,
      "step": 14030
    },
    {
      "entropy": 5.532813215255738,
      "epoch": 1.0919665434740322,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004884995131668445,
      "loss": 5.2713,
      "mean_token_accuracy": 0.17376965433359146,
      "num_tokens": 24348876.0,
      "step": 14035
    },
    {
      "entropy": 5.456120920181275,
      "epoch": 1.09235557284575,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004884907640412248,
      "loss": 5.2782,
      "mean_token_accuracy": 0.17300991863012313,
      "num_tokens": 24357094.0,
      "step": 14040
    },
    {
      "entropy": 5.453493261337281,
      "epoch": 1.0927446022174674,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004884820116762225,
      "loss": 5.2397,
      "mean_token_accuracy": 0.17267200499773025,
      "num_tokens": 24366320.0,
      "step": 14045
    },
    {
      "entropy": 5.476856327056884,
      "epoch": 1.093133631589185,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004884732560719706,
      "loss": 5.1289,
      "mean_token_accuracy": 0.17867153882980347,
      "num_tokens": 24374390.0,
      "step": 14050
    },
    {
      "entropy": 5.427449750900268,
      "epoch": 1.0935226609609026,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004884644972286017,
      "loss": 5.2349,
      "mean_token_accuracy": 0.187056964635849,
      "num_tokens": 24382804.0,
      "step": 14055
    },
    {
      "entropy": 5.495086765289306,
      "epoch": 1.09391169033262,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004884557351462488,
      "loss": 5.2465,
      "mean_token_accuracy": 0.1738008365035057,
      "num_tokens": 24391352.0,
      "step": 14060
    },
    {
      "entropy": 5.539323472976685,
      "epoch": 1.0943007197043377,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004884469698250449,
      "loss": 5.3085,
      "mean_token_accuracy": 0.17266384810209273,
      "num_tokens": 24400348.0,
      "step": 14065
    },
    {
      "entropy": 5.405953788757325,
      "epoch": 1.0946897490760552,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000488438201265123,
      "loss": 5.151,
      "mean_token_accuracy": 0.18451739922165872,
      "num_tokens": 24409628.0,
      "step": 14070
    },
    {
      "entropy": 5.42804651260376,
      "epoch": 1.095078778447773,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004884294294666161,
      "loss": 5.2509,
      "mean_token_accuracy": 0.17527421563863754,
      "num_tokens": 24418379.0,
      "step": 14075
    },
    {
      "entropy": 5.546537256240844,
      "epoch": 1.0954678078194904,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004884206544296573,
      "loss": 5.383,
      "mean_token_accuracy": 0.166053669154644,
      "num_tokens": 24427719.0,
      "step": 14080
    },
    {
      "entropy": 5.5603461265563965,
      "epoch": 1.0958568371912079,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004884118761543797,
      "loss": 5.3317,
      "mean_token_accuracy": 0.17193595618009566,
      "num_tokens": 24435938.0,
      "step": 14085
    },
    {
      "entropy": 5.473726081848144,
      "epoch": 1.0962458665629256,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004884030946409167,
      "loss": 5.2218,
      "mean_token_accuracy": 0.177756629884243,
      "num_tokens": 24444286.0,
      "step": 14090
    },
    {
      "entropy": 5.427606105804443,
      "epoch": 1.096634895934643,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004883943098894013,
      "loss": 5.1696,
      "mean_token_accuracy": 0.17818832099437715,
      "num_tokens": 24452546.0,
      "step": 14095
    },
    {
      "entropy": 5.506666564941407,
      "epoch": 1.0970239253063607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000488385521899967,
      "loss": 5.3047,
      "mean_token_accuracy": 0.17128586620092393,
      "num_tokens": 24460765.0,
      "step": 14100
    },
    {
      "entropy": 5.440995216369629,
      "epoch": 1.0974129546780782,
      "grad_norm": 1.3125,
      "learning_rate": 0.000488376730672747,
      "loss": 5.1863,
      "mean_token_accuracy": 0.17720701694488525,
      "num_tokens": 24470018.0,
      "step": 14105
    },
    {
      "entropy": 5.415807247161865,
      "epoch": 1.0978019840497957,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004883679362078746,
      "loss": 5.1252,
      "mean_token_accuracy": 0.17770083397626876,
      "num_tokens": 24478882.0,
      "step": 14110
    },
    {
      "entropy": 5.474627256393433,
      "epoch": 1.0981910134215134,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004883591385054836,
      "loss": 5.3054,
      "mean_token_accuracy": 0.16687375605106353,
      "num_tokens": 24487801.0,
      "step": 14115
    },
    {
      "entropy": 5.4835120677948,
      "epoch": 1.0985800427932308,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004883503375657072,
      "loss": 5.2313,
      "mean_token_accuracy": 0.17657593488693238,
      "num_tokens": 24496082.0,
      "step": 14120
    },
    {
      "entropy": 5.436771631240845,
      "epoch": 1.0989690721649485,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004883415333886789,
      "loss": 5.1799,
      "mean_token_accuracy": 0.18368006199598313,
      "num_tokens": 24504664.0,
      "step": 14125
    },
    {
      "entropy": 5.451993036270141,
      "epoch": 1.099358101536666,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004883327259745325,
      "loss": 5.1993,
      "mean_token_accuracy": 0.18186167478561402,
      "num_tokens": 24512797.0,
      "step": 14130
    },
    {
      "entropy": 5.4107273578643795,
      "epoch": 1.0997471309083835,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004883239153234015,
      "loss": 5.1845,
      "mean_token_accuracy": 0.18125838041305542,
      "num_tokens": 24520941.0,
      "step": 14135
    },
    {
      "entropy": 5.511631345748901,
      "epoch": 1.1001361602801012,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004883151014354197,
      "loss": 5.2941,
      "mean_token_accuracy": 0.17334001362323762,
      "num_tokens": 24530087.0,
      "step": 14140
    },
    {
      "entropy": 5.499214696884155,
      "epoch": 1.1005251896518187,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004883062843107207,
      "loss": 5.2185,
      "mean_token_accuracy": 0.1741839587688446,
      "num_tokens": 24538593.0,
      "step": 14145
    },
    {
      "entropy": 5.614100122451783,
      "epoch": 1.1009142190235364,
      "grad_norm": 1.25,
      "learning_rate": 0.0004882974639494383,
      "loss": 5.3224,
      "mean_token_accuracy": 0.1650022506713867,
      "num_tokens": 24547158.0,
      "step": 14150
    },
    {
      "entropy": 5.481727647781372,
      "epoch": 1.1013032483952538,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004882886403517065,
      "loss": 5.3038,
      "mean_token_accuracy": 0.17123375684022904,
      "num_tokens": 24556266.0,
      "step": 14155
    },
    {
      "entropy": 5.5516016483306885,
      "epoch": 1.1016922777669713,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004882798135176589,
      "loss": 5.358,
      "mean_token_accuracy": 0.17056136280298234,
      "num_tokens": 24565122.0,
      "step": 14160
    },
    {
      "entropy": 5.516513872146606,
      "epoch": 1.102081307138689,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048827098344742973,
      "loss": 5.1916,
      "mean_token_accuracy": 0.18432184606790541,
      "num_tokens": 24573944.0,
      "step": 14165
    },
    {
      "entropy": 5.516745138168335,
      "epoch": 1.1024703365104065,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004882621501411528,
      "loss": 5.369,
      "mean_token_accuracy": 0.172733835875988,
      "num_tokens": 24582233.0,
      "step": 14170
    },
    {
      "entropy": 5.524791097640991,
      "epoch": 1.1028593658821242,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004882533135989622,
      "loss": 5.239,
      "mean_token_accuracy": 0.17305948287248613,
      "num_tokens": 24590855.0,
      "step": 14175
    },
    {
      "entropy": 5.593385553359985,
      "epoch": 1.1032483952538417,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000488244473820992,
      "loss": 5.2925,
      "mean_token_accuracy": 0.16824696213006973,
      "num_tokens": 24599877.0,
      "step": 14180
    },
    {
      "entropy": 5.403017568588257,
      "epoch": 1.1036374246255591,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048823563080737634,
      "loss": 5.1639,
      "mean_token_accuracy": 0.18214736580848695,
      "num_tokens": 24608450.0,
      "step": 14185
    },
    {
      "entropy": 5.380068111419678,
      "epoch": 1.1040264539972768,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004882267845582493,
      "loss": 5.2537,
      "mean_token_accuracy": 0.1745898097753525,
      "num_tokens": 24618018.0,
      "step": 14190
    },
    {
      "entropy": 5.523136854171753,
      "epoch": 1.1044154833689943,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048821793507374526,
      "loss": 5.2353,
      "mean_token_accuracy": 0.1769924983382225,
      "num_tokens": 24626388.0,
      "step": 14195
    },
    {
      "entropy": 5.58181209564209,
      "epoch": 1.104804512740712,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004882090823539984,
      "loss": 5.354,
      "mean_token_accuracy": 0.16567516326904297,
      "num_tokens": 24635284.0,
      "step": 14200
    },
    {
      "entropy": 5.484607172012329,
      "epoch": 1.1051935421124295,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004882002263991431,
      "loss": 5.187,
      "mean_token_accuracy": 0.1835028424859047,
      "num_tokens": 24644297.0,
      "step": 14205
    },
    {
      "entropy": 5.366165351867676,
      "epoch": 1.105582571484147,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048819136720931364,
      "loss": 5.0637,
      "mean_token_accuracy": 0.18631493002176286,
      "num_tokens": 24652965.0,
      "step": 14210
    },
    {
      "entropy": 5.3803304672241214,
      "epoch": 1.1059716008558647,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048818250478464457,
      "loss": 5.1329,
      "mean_token_accuracy": 0.1830570787191391,
      "num_tokens": 24661034.0,
      "step": 14215
    },
    {
      "entropy": 5.489173173904419,
      "epoch": 1.1063606302275821,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004881736391252703,
      "loss": 5.2332,
      "mean_token_accuracy": 0.1787367932498455,
      "num_tokens": 24669831.0,
      "step": 14220
    },
    {
      "entropy": 5.494463109970093,
      "epoch": 1.1067496595992998,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004881647702313253,
      "loss": 5.2953,
      "mean_token_accuracy": 0.1678923636674881,
      "num_tokens": 24678491.0,
      "step": 14225
    },
    {
      "entropy": 5.483972597122192,
      "epoch": 1.1071386889710173,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048815589810294427,
      "loss": 5.2457,
      "mean_token_accuracy": 0.17994679361581803,
      "num_tokens": 24687308.0,
      "step": 14230
    },
    {
      "entropy": 5.484911298751831,
      "epoch": 1.1075277183427348,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048814702274026174,
      "loss": 5.1955,
      "mean_token_accuracy": 0.17658255696296693,
      "num_tokens": 24696080.0,
      "step": 14235
    },
    {
      "entropy": 5.427283763885498,
      "epoch": 1.1079167477144525,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004881381441434124,
      "loss": 5.2815,
      "mean_token_accuracy": 0.17419894337654113,
      "num_tokens": 24704561.0,
      "step": 14240
    },
    {
      "entropy": 5.541844224929809,
      "epoch": 1.10830577708617,
      "grad_norm": 1.21875,
      "learning_rate": 0.000488129262312531,
      "loss": 5.3053,
      "mean_token_accuracy": 0.17608402222394942,
      "num_tokens": 24713403.0,
      "step": 14245
    },
    {
      "entropy": 5.562458515167236,
      "epoch": 1.1086948064578876,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048812037724775217,
      "loss": 5.2929,
      "mean_token_accuracy": 0.1693214550614357,
      "num_tokens": 24722031.0,
      "step": 14250
    },
    {
      "entropy": 5.495748472213745,
      "epoch": 1.1090838358296051,
      "grad_norm": 1.25,
      "learning_rate": 0.0004881114889492109,
      "loss": 5.2373,
      "mean_token_accuracy": 0.17336097359657288,
      "num_tokens": 24731176.0,
      "step": 14255
    },
    {
      "entropy": 5.543942546844482,
      "epoch": 1.1094728652013226,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048810259741704197,
      "loss": 5.2959,
      "mean_token_accuracy": 0.1724896401166916,
      "num_tokens": 24740037.0,
      "step": 14260
    },
    {
      "entropy": 5.566663837432861,
      "epoch": 1.1098618945730403,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004880937026513803,
      "loss": 5.3478,
      "mean_token_accuracy": 0.1688096582889557,
      "num_tokens": 24748778.0,
      "step": 14265
    },
    {
      "entropy": 5.502234745025635,
      "epoch": 1.1102509239447578,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004880848046523609,
      "loss": 5.1821,
      "mean_token_accuracy": 0.17502260953187943,
      "num_tokens": 24757226.0,
      "step": 14270
    },
    {
      "entropy": 5.421371603012085,
      "epoch": 1.1106399533164755,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048807590342011875,
      "loss": 5.2451,
      "mean_token_accuracy": 0.17716450542211531,
      "num_tokens": 24765734.0,
      "step": 14275
    },
    {
      "entropy": 5.417635107040406,
      "epoch": 1.111028982688193,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048806699895478895,
      "loss": 5.2269,
      "mean_token_accuracy": 0.17687494456768035,
      "num_tokens": 24773773.0,
      "step": 14280
    },
    {
      "entropy": 5.470719814300537,
      "epoch": 1.1114180120599104,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004880580912565065,
      "loss": 5.257,
      "mean_token_accuracy": 0.17438815534114838,
      "num_tokens": 24782563.0,
      "step": 14285
    },
    {
      "entropy": 5.535797023773194,
      "epoch": 1.1118070414316281,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004880491803254067,
      "loss": 5.3305,
      "mean_token_accuracy": 0.17216645032167435,
      "num_tokens": 24791492.0,
      "step": 14290
    },
    {
      "entropy": 5.476056480407715,
      "epoch": 1.1121960708033456,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004880402661616246,
      "loss": 5.1489,
      "mean_token_accuracy": 0.18349012732505798,
      "num_tokens": 24800229.0,
      "step": 14295
    },
    {
      "entropy": 5.416717147827148,
      "epoch": 1.1125851001750633,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004880313487652956,
      "loss": 5.2182,
      "mean_token_accuracy": 0.17329297810792924,
      "num_tokens": 24809454.0,
      "step": 14300
    },
    {
      "entropy": 5.439956188201904,
      "epoch": 1.1129741295467808,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048802242813655503,
      "loss": 5.2314,
      "mean_token_accuracy": 0.1731076255440712,
      "num_tokens": 24817956.0,
      "step": 14305
    },
    {
      "entropy": 5.430538177490234,
      "epoch": 1.1133631589184985,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000488013504275538,
      "loss": 5.1346,
      "mean_token_accuracy": 0.181461001932621,
      "num_tokens": 24826288.0,
      "step": 14310
    },
    {
      "entropy": 5.566921186447144,
      "epoch": 1.113752188290216,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004880045771823802,
      "loss": 5.2847,
      "mean_token_accuracy": 0.17647963464260102,
      "num_tokens": 24834720.0,
      "step": 14315
    },
    {
      "entropy": 5.457182550430298,
      "epoch": 1.1141412176619334,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048799564685721695,
      "loss": 5.254,
      "mean_token_accuracy": 0.1734289661049843,
      "num_tokens": 24843000.0,
      "step": 14320
    },
    {
      "entropy": 5.496828985214234,
      "epoch": 1.1145302470336511,
      "grad_norm": 1.25,
      "learning_rate": 0.0004879867133001837,
      "loss": 5.2301,
      "mean_token_accuracy": 0.17242724299430848,
      "num_tokens": 24851060.0,
      "step": 14325
    },
    {
      "entropy": 5.378280210494995,
      "epoch": 1.1149192764053686,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004879777765114162,
      "loss": 5.1484,
      "mean_token_accuracy": 0.18061680495738983,
      "num_tokens": 24859913.0,
      "step": 14330
    },
    {
      "entropy": 5.552402448654175,
      "epoch": 1.115308305777086,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048796883649104996,
      "loss": 5.3729,
      "mean_token_accuracy": 0.16949469298124314,
      "num_tokens": 24869648.0,
      "step": 14335
    },
    {
      "entropy": 5.52118148803711,
      "epoch": 1.1156973351488038,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004879598932392206,
      "loss": 5.2741,
      "mean_token_accuracy": 0.17817801237106323,
      "num_tokens": 24878813.0,
      "step": 14340
    },
    {
      "entropy": 5.50523943901062,
      "epoch": 1.1160863645205212,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048795094675606364,
      "loss": 5.2476,
      "mean_token_accuracy": 0.1730068951845169,
      "num_tokens": 24887714.0,
      "step": 14345
    },
    {
      "entropy": 5.485349559783936,
      "epoch": 1.116475393892239,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004879419970417152,
      "loss": 5.2342,
      "mean_token_accuracy": 0.1741853415966034,
      "num_tokens": 24895961.0,
      "step": 14350
    },
    {
      "entropy": 5.460059022903442,
      "epoch": 1.1168644232639564,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048793304409631077,
      "loss": 5.2448,
      "mean_token_accuracy": 0.17294235825538634,
      "num_tokens": 24904873.0,
      "step": 14355
    },
    {
      "entropy": 5.408315944671631,
      "epoch": 1.117253452635674,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004879240879199863,
      "loss": 5.1507,
      "mean_token_accuracy": 0.1850723698735237,
      "num_tokens": 24913486.0,
      "step": 14360
    },
    {
      "entropy": 5.478535175323486,
      "epoch": 1.1176424820073916,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048791512851287786,
      "loss": 5.2386,
      "mean_token_accuracy": 0.17304750233888627,
      "num_tokens": 24921943.0,
      "step": 14365
    },
    {
      "entropy": 5.59430513381958,
      "epoch": 1.118031511379109,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048790616587512107,
      "loss": 5.3258,
      "mean_token_accuracy": 0.1695338174700737,
      "num_tokens": 24930081.0,
      "step": 14370
    },
    {
      "entropy": 5.568580484390258,
      "epoch": 1.1184205407508268,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004878972000068521,
      "loss": 5.2573,
      "mean_token_accuracy": 0.17729070484638215,
      "num_tokens": 24938985.0,
      "step": 14375
    },
    {
      "entropy": 5.350616788864135,
      "epoch": 1.1188095701225442,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048788823090820707,
      "loss": 5.1107,
      "mean_token_accuracy": 0.18178079128265381,
      "num_tokens": 24947497.0,
      "step": 14380
    },
    {
      "entropy": 5.488047552108765,
      "epoch": 1.1191985994942617,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048787925857932194,
      "loss": 5.3198,
      "mean_token_accuracy": 0.16944557875394822,
      "num_tokens": 24956263.0,
      "step": 14385
    },
    {
      "entropy": 5.5972580909729,
      "epoch": 1.1195876288659794,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048787028302033287,
      "loss": 5.4069,
      "mean_token_accuracy": 0.16532928496599197,
      "num_tokens": 24965646.0,
      "step": 14390
    },
    {
      "entropy": 5.556589841842651,
      "epoch": 1.1199766582376969,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048786130423137606,
      "loss": 5.2873,
      "mean_token_accuracy": 0.17365862876176835,
      "num_tokens": 24973318.0,
      "step": 14395
    },
    {
      "entropy": 5.50542688369751,
      "epoch": 1.1203656876094146,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004878523222125879,
      "loss": 5.3571,
      "mean_token_accuracy": 0.17373592555522918,
      "num_tokens": 24982600.0,
      "step": 14400
    },
    {
      "entropy": 5.6085539817810055,
      "epoch": 1.120754716981132,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004878433369641044,
      "loss": 5.3429,
      "mean_token_accuracy": 0.16792995035648345,
      "num_tokens": 24990386.0,
      "step": 14405
    },
    {
      "entropy": 5.421349239349365,
      "epoch": 1.1211437463528497,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004878343484860621,
      "loss": 5.1235,
      "mean_token_accuracy": 0.1827808529138565,
      "num_tokens": 24999064.0,
      "step": 14410
    },
    {
      "entropy": 5.410102081298828,
      "epoch": 1.1215327757245672,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004878253567785972,
      "loss": 5.2988,
      "mean_token_accuracy": 0.17512336522340774,
      "num_tokens": 25008448.0,
      "step": 14415
    },
    {
      "entropy": 5.328612327575684,
      "epoch": 1.1219218050962847,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048781636184184644,
      "loss": 5.1226,
      "mean_token_accuracy": 0.18734526336193086,
      "num_tokens": 25016722.0,
      "step": 14420
    },
    {
      "entropy": 5.442304801940918,
      "epoch": 1.1223108344680024,
      "grad_norm": 1.25,
      "learning_rate": 0.000487807363675946,
      "loss": 5.2016,
      "mean_token_accuracy": 0.18224625438451766,
      "num_tokens": 25026263.0,
      "step": 14425
    },
    {
      "entropy": 5.551255321502685,
      "epoch": 1.1226998638397199,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004877983622810326,
      "loss": 5.2447,
      "mean_token_accuracy": 0.17484278976917267,
      "num_tokens": 25035202.0,
      "step": 14430
    },
    {
      "entropy": 5.532693195343017,
      "epoch": 1.1230888932114376,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004877893576572427,
      "loss": 5.3199,
      "mean_token_accuracy": 0.17524950802326203,
      "num_tokens": 25044309.0,
      "step": 14435
    },
    {
      "entropy": 5.498838233947754,
      "epoch": 1.123477922583155,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000487780349804713,
      "loss": 5.3048,
      "mean_token_accuracy": 0.17444520592689514,
      "num_tokens": 25053044.0,
      "step": 14440
    },
    {
      "entropy": 5.339668655395508,
      "epoch": 1.1238669519548725,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004877713387235802,
      "loss": 5.2019,
      "mean_token_accuracy": 0.17973190695047378,
      "num_tokens": 25061869.0,
      "step": 14445
    },
    {
      "entropy": 5.535868215560913,
      "epoch": 1.1242559813265902,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000487762324413981,
      "loss": 5.3857,
      "mean_token_accuracy": 0.16621869504451753,
      "num_tokens": 25070022.0,
      "step": 14450
    },
    {
      "entropy": 5.541877794265747,
      "epoch": 1.1246450106983077,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048775330687605213,
      "loss": 5.2763,
      "mean_token_accuracy": 0.17078239768743514,
      "num_tokens": 25078604.0,
      "step": 14455
    },
    {
      "entropy": 5.490096807479858,
      "epoch": 1.1250340400700254,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004877442861099305,
      "loss": 5.1322,
      "mean_token_accuracy": 0.18350782841444016,
      "num_tokens": 25087256.0,
      "step": 14460
    },
    {
      "entropy": 5.471586084365844,
      "epoch": 1.1254230694417429,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004877352621157529,
      "loss": 5.2143,
      "mean_token_accuracy": 0.17688586115837096,
      "num_tokens": 25095593.0,
      "step": 14465
    },
    {
      "entropy": 5.498533630371094,
      "epoch": 1.1258120988134603,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048772623489365635,
      "loss": 5.281,
      "mean_token_accuracy": 0.17353989332914352,
      "num_tokens": 25104852.0,
      "step": 14470
    },
    {
      "entropy": 5.493834066390991,
      "epoch": 1.126201128185178,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004877172044437777,
      "loss": 5.2659,
      "mean_token_accuracy": 0.17459852248430252,
      "num_tokens": 25114453.0,
      "step": 14475
    },
    {
      "entropy": 5.493443965911865,
      "epoch": 1.1265901575568955,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048770817076625416,
      "loss": 5.2513,
      "mean_token_accuracy": 0.178925584256649,
      "num_tokens": 25123733.0,
      "step": 14480
    },
    {
      "entropy": 5.4589715003967285,
      "epoch": 1.126979186928613,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048769913386122253,
      "loss": 5.1778,
      "mean_token_accuracy": 0.18383704572916032,
      "num_tokens": 25132572.0,
      "step": 14485
    },
    {
      "entropy": 5.3761186599731445,
      "epoch": 1.1273682163003307,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004876900937288202,
      "loss": 5.2174,
      "mean_token_accuracy": 0.17639420479536055,
      "num_tokens": 25141494.0,
      "step": 14490
    },
    {
      "entropy": 5.44729323387146,
      "epoch": 1.1277572456720482,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048768105036918426,
      "loss": 5.1771,
      "mean_token_accuracy": 0.17887090295553207,
      "num_tokens": 25150186.0,
      "step": 14495
    },
    {
      "entropy": 5.4999809741973875,
      "epoch": 1.1281462750437659,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048767200378245187,
      "loss": 5.2496,
      "mean_token_accuracy": 0.17497891187667847,
      "num_tokens": 25159251.0,
      "step": 14500
    },
    {
      "entropy": 5.458379793167114,
      "epoch": 1.1285353044154833,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048766295396876025,
      "loss": 5.2671,
      "mean_token_accuracy": 0.17795070111751557,
      "num_tokens": 25167780.0,
      "step": 14505
    },
    {
      "entropy": 5.445296001434326,
      "epoch": 1.128924333787201,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048765390092824683,
      "loss": 5.1676,
      "mean_token_accuracy": 0.17945482581853867,
      "num_tokens": 25175358.0,
      "step": 14510
    },
    {
      "entropy": 5.444134473800659,
      "epoch": 1.1293133631589185,
      "grad_norm": 1.21875,
      "learning_rate": 0.000487644844661049,
      "loss": 5.2067,
      "mean_token_accuracy": 0.1730176791548729,
      "num_tokens": 25184247.0,
      "step": 14515
    },
    {
      "entropy": 5.420819902420044,
      "epoch": 1.129702392530636,
      "grad_norm": 1.21875,
      "learning_rate": 0.000487635785167304,
      "loss": 5.1217,
      "mean_token_accuracy": 0.18564544767141342,
      "num_tokens": 25192311.0,
      "step": 14520
    },
    {
      "entropy": 5.378132104873657,
      "epoch": 1.1300914219023537,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048762672244714953,
      "loss": 5.1788,
      "mean_token_accuracy": 0.1821864664554596,
      "num_tokens": 25200861.0,
      "step": 14525
    },
    {
      "entropy": 5.523636388778686,
      "epoch": 1.1304804512740712,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004876176565007229,
      "loss": 5.3186,
      "mean_token_accuracy": 0.17549225986003875,
      "num_tokens": 25208938.0,
      "step": 14530
    },
    {
      "entropy": 5.552270889282227,
      "epoch": 1.1308694806457888,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048760858732816174,
      "loss": 5.2228,
      "mean_token_accuracy": 0.17584580034017563,
      "num_tokens": 25216997.0,
      "step": 14535
    },
    {
      "entropy": 5.440706348419189,
      "epoch": 1.1312585100175063,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004875995149296037,
      "loss": 5.1326,
      "mean_token_accuracy": 0.1811284303665161,
      "num_tokens": 25225205.0,
      "step": 14540
    },
    {
      "entropy": 5.432539415359497,
      "epoch": 1.1316475393892238,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004875904393051864,
      "loss": 5.2042,
      "mean_token_accuracy": 0.18095376640558242,
      "num_tokens": 25233807.0,
      "step": 14545
    },
    {
      "entropy": 5.404697704315185,
      "epoch": 1.1320365687609415,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004875813604550476,
      "loss": 5.2205,
      "mean_token_accuracy": 0.17411712408065796,
      "num_tokens": 25241715.0,
      "step": 14550
    },
    {
      "entropy": 5.543915557861328,
      "epoch": 1.132425598132659,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048757227837932494,
      "loss": 5.3464,
      "mean_token_accuracy": 0.17370496392250062,
      "num_tokens": 25250418.0,
      "step": 14555
    },
    {
      "entropy": 5.556774520874024,
      "epoch": 1.1328146275043767,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004875631930781563,
      "loss": 5.2473,
      "mean_token_accuracy": 0.166562420129776,
      "num_tokens": 25259164.0,
      "step": 14560
    },
    {
      "entropy": 5.49828872680664,
      "epoch": 1.1332036568760941,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048755410455167967,
      "loss": 5.294,
      "mean_token_accuracy": 0.17247709035873413,
      "num_tokens": 25267902.0,
      "step": 14565
    },
    {
      "entropy": 5.476278972625733,
      "epoch": 1.1335926862478116,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004875450128000326,
      "loss": 5.235,
      "mean_token_accuracy": 0.16847948133945465,
      "num_tokens": 25275991.0,
      "step": 14570
    },
    {
      "entropy": 5.483393573760987,
      "epoch": 1.1339817156195293,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004875359178233534,
      "loss": 5.2349,
      "mean_token_accuracy": 0.17800334841012955,
      "num_tokens": 25284454.0,
      "step": 14575
    },
    {
      "entropy": 5.565164709091187,
      "epoch": 1.1343707449912468,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048752681962177985,
      "loss": 5.3362,
      "mean_token_accuracy": 0.17282153964042662,
      "num_tokens": 25292867.0,
      "step": 14580
    },
    {
      "entropy": 5.463955545425415,
      "epoch": 1.1347597743629645,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048751771819545013,
      "loss": 5.2369,
      "mean_token_accuracy": 0.18253506869077682,
      "num_tokens": 25301486.0,
      "step": 14585
    },
    {
      "entropy": 5.439399242401123,
      "epoch": 1.135148803734682,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048750861354450236,
      "loss": 5.2063,
      "mean_token_accuracy": 0.17949570268392562,
      "num_tokens": 25310800.0,
      "step": 14590
    },
    {
      "entropy": 5.506656646728516,
      "epoch": 1.1355378331063994,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004874995056690746,
      "loss": 5.2692,
      "mean_token_accuracy": 0.1775213062763214,
      "num_tokens": 25319063.0,
      "step": 14595
    },
    {
      "entropy": 5.623602294921875,
      "epoch": 1.1359268624781171,
      "grad_norm": 1.1875,
      "learning_rate": 0.000487490394569305,
      "loss": 5.3247,
      "mean_token_accuracy": 0.17306364923715592,
      "num_tokens": 25327817.0,
      "step": 14600
    },
    {
      "entropy": 5.466697931289673,
      "epoch": 1.1363158918498346,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004874812802453319,
      "loss": 5.1479,
      "mean_token_accuracy": 0.17923812717199325,
      "num_tokens": 25335753.0,
      "step": 14605
    },
    {
      "entropy": 5.472228860855102,
      "epoch": 1.1367049212215523,
      "grad_norm": 1.375,
      "learning_rate": 0.0004874721626972936,
      "loss": 5.3371,
      "mean_token_accuracy": 0.17058107107877732,
      "num_tokens": 25344514.0,
      "step": 14610
    },
    {
      "entropy": 5.508408069610596,
      "epoch": 1.1370939505932698,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004874630419253284,
      "loss": 5.2414,
      "mean_token_accuracy": 0.1780880644917488,
      "num_tokens": 25353003.0,
      "step": 14615
    },
    {
      "entropy": 5.466843700408935,
      "epoch": 1.1374829799649873,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048745391792957474,
      "loss": 5.2475,
      "mean_token_accuracy": 0.17237841635942458,
      "num_tokens": 25361599.0,
      "step": 14620
    },
    {
      "entropy": 5.517878437042237,
      "epoch": 1.137872009336705,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048744479071017097,
      "loss": 5.2904,
      "mean_token_accuracy": 0.1712319955229759,
      "num_tokens": 25370566.0,
      "step": 14625
    },
    {
      "entropy": 5.527145147323608,
      "epoch": 1.1382610387084224,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004874356602672556,
      "loss": 5.2795,
      "mean_token_accuracy": 0.17387592941522598,
      "num_tokens": 25379589.0,
      "step": 14630
    },
    {
      "entropy": 5.450634145736695,
      "epoch": 1.1386500680801401,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004874265266009673,
      "loss": 5.1989,
      "mean_token_accuracy": 0.1847446233034134,
      "num_tokens": 25387481.0,
      "step": 14635
    },
    {
      "entropy": 5.432240056991577,
      "epoch": 1.1390390974518576,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004874173897114445,
      "loss": 5.2299,
      "mean_token_accuracy": 0.17731301486492157,
      "num_tokens": 25395551.0,
      "step": 14640
    },
    {
      "entropy": 5.519715690612793,
      "epoch": 1.1394281268235753,
      "grad_norm": 1.375,
      "learning_rate": 0.00048740824959882603,
      "loss": 5.2818,
      "mean_token_accuracy": 0.16925130784511566,
      "num_tokens": 25404303.0,
      "step": 14645
    },
    {
      "entropy": 5.507825469970703,
      "epoch": 1.1398171561952928,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004873991062632504,
      "loss": 5.2806,
      "mean_token_accuracy": 0.16797916367650031,
      "num_tokens": 25412749.0,
      "step": 14650
    },
    {
      "entropy": 5.473748350143433,
      "epoch": 1.1402061855670103,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048738995970485635,
      "loss": 5.2196,
      "mean_token_accuracy": 0.1748868480324745,
      "num_tokens": 25421075.0,
      "step": 14655
    },
    {
      "entropy": 5.507064580917358,
      "epoch": 1.140595214938728,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004873808099237827,
      "loss": 5.3542,
      "mean_token_accuracy": 0.17185562252998351,
      "num_tokens": 25429750.0,
      "step": 14660
    },
    {
      "entropy": 5.518696641921997,
      "epoch": 1.1409842443104454,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004873716569201684,
      "loss": 5.2278,
      "mean_token_accuracy": 0.17812446057796477,
      "num_tokens": 25437686.0,
      "step": 14665
    },
    {
      "entropy": 5.51885838508606,
      "epoch": 1.141373273682163,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048736250069415213,
      "loss": 5.2437,
      "mean_token_accuracy": 0.17689005583524703,
      "num_tokens": 25446050.0,
      "step": 14670
    },
    {
      "entropy": 5.429541206359863,
      "epoch": 1.1417623030538806,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004873533412458729,
      "loss": 5.2646,
      "mean_token_accuracy": 0.17253877073526383,
      "num_tokens": 25455134.0,
      "step": 14675
    },
    {
      "entropy": 5.42077751159668,
      "epoch": 1.142151332425598,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048734417857546986,
      "loss": 5.1366,
      "mean_token_accuracy": 0.18095289766788483,
      "num_tokens": 25463258.0,
      "step": 14680
    },
    {
      "entropy": 5.479968452453614,
      "epoch": 1.1425403617973158,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004873350126830818,
      "loss": 5.2137,
      "mean_token_accuracy": 0.1819719523191452,
      "num_tokens": 25472085.0,
      "step": 14685
    },
    {
      "entropy": 5.537111043930054,
      "epoch": 1.1429293911690332,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004873258435688479,
      "loss": 5.239,
      "mean_token_accuracy": 0.1763238161802292,
      "num_tokens": 25481233.0,
      "step": 14690
    },
    {
      "entropy": 5.439796829223633,
      "epoch": 1.143318420540751,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004873166712329073,
      "loss": 5.1992,
      "mean_token_accuracy": 0.1822013184428215,
      "num_tokens": 25490354.0,
      "step": 14695
    },
    {
      "entropy": 5.433959436416626,
      "epoch": 1.1437074499124684,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048730749567539914,
      "loss": 5.2687,
      "mean_token_accuracy": 0.1710852712392807,
      "num_tokens": 25498467.0,
      "step": 14700
    },
    {
      "entropy": 5.497142219543457,
      "epoch": 1.144096479284186,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048729831689646257,
      "loss": 5.265,
      "mean_token_accuracy": 0.17779612243175508,
      "num_tokens": 25507543.0,
      "step": 14705
    },
    {
      "entropy": 5.480814456939697,
      "epoch": 1.1444855086559036,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048728913489623705,
      "loss": 5.1541,
      "mean_token_accuracy": 0.1835753545165062,
      "num_tokens": 25516070.0,
      "step": 14710
    },
    {
      "entropy": 5.499295806884765,
      "epoch": 1.144874538027621,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004872799496748618,
      "loss": 5.2701,
      "mean_token_accuracy": 0.17517900019884108,
      "num_tokens": 25525216.0,
      "step": 14715
    },
    {
      "entropy": 5.544117975234985,
      "epoch": 1.1452635673993385,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004872707612324761,
      "loss": 5.3066,
      "mean_token_accuracy": 0.16923885345458983,
      "num_tokens": 25533822.0,
      "step": 14720
    },
    {
      "entropy": 5.581564950942993,
      "epoch": 1.1456525967710562,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004872615695692196,
      "loss": 5.3269,
      "mean_token_accuracy": 0.16834421306848527,
      "num_tokens": 25541580.0,
      "step": 14725
    },
    {
      "entropy": 5.416932010650635,
      "epoch": 1.1460416261427737,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004872523746852315,
      "loss": 5.1201,
      "mean_token_accuracy": 0.1860901027917862,
      "num_tokens": 25549860.0,
      "step": 14730
    },
    {
      "entropy": 5.460150527954101,
      "epoch": 1.1464306555144914,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048724317658065146,
      "loss": 5.2862,
      "mean_token_accuracy": 0.17425459176301955,
      "num_tokens": 25559060.0,
      "step": 14735
    },
    {
      "entropy": 5.589854145050049,
      "epoch": 1.146819684886209,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048723397525561916,
      "loss": 5.3336,
      "mean_token_accuracy": 0.16649319976568222,
      "num_tokens": 25567769.0,
      "step": 14740
    },
    {
      "entropy": 5.534193229675293,
      "epoch": 1.1472087142579266,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048722477071027394,
      "loss": 5.3166,
      "mean_token_accuracy": 0.17398206293582916,
      "num_tokens": 25577443.0,
      "step": 14745
    },
    {
      "entropy": 5.541543006896973,
      "epoch": 1.147597743629644,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048721556294475574,
      "loss": 5.2214,
      "mean_token_accuracy": 0.1864846244454384,
      "num_tokens": 25586283.0,
      "step": 14750
    },
    {
      "entropy": 5.471278333663941,
      "epoch": 1.1479867730013615,
      "grad_norm": 1.25,
      "learning_rate": 0.00048720635195920403,
      "loss": 5.3355,
      "mean_token_accuracy": 0.17660784274339675,
      "num_tokens": 25594881.0,
      "step": 14755
    },
    {
      "entropy": 5.405520915985107,
      "epoch": 1.1483758023730792,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004871971377537588,
      "loss": 5.18,
      "mean_token_accuracy": 0.17867934554815293,
      "num_tokens": 25603738.0,
      "step": 14760
    },
    {
      "entropy": 5.471324396133423,
      "epoch": 1.1487648317447967,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004871879203285597,
      "loss": 5.2526,
      "mean_token_accuracy": 0.17837337553501129,
      "num_tokens": 25611780.0,
      "step": 14765
    },
    {
      "entropy": 5.489037418365479,
      "epoch": 1.1491538611165142,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048717869968374655,
      "loss": 5.3265,
      "mean_token_accuracy": 0.17406396120786666,
      "num_tokens": 25620489.0,
      "step": 14770
    },
    {
      "entropy": 5.468789291381836,
      "epoch": 1.1495428904882319,
      "grad_norm": 1.375,
      "learning_rate": 0.00048716947581945954,
      "loss": 5.2737,
      "mean_token_accuracy": 0.17839858531951905,
      "num_tokens": 25629113.0,
      "step": 14775
    },
    {
      "entropy": 5.46669635772705,
      "epoch": 1.1499319198599494,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004871602487358383,
      "loss": 5.2205,
      "mean_token_accuracy": 0.1767689496278763,
      "num_tokens": 25637313.0,
      "step": 14780
    },
    {
      "entropy": 5.524136066436768,
      "epoch": 1.150320949231667,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000487151018433023,
      "loss": 5.3076,
      "mean_token_accuracy": 0.175931416451931,
      "num_tokens": 25646522.0,
      "step": 14785
    },
    {
      "entropy": 5.502716302871704,
      "epoch": 1.1507099786033845,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048714178491115367,
      "loss": 5.204,
      "mean_token_accuracy": 0.17807692289352417,
      "num_tokens": 25656058.0,
      "step": 14790
    },
    {
      "entropy": 5.552208280563354,
      "epoch": 1.1510990079751022,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048713254817037046,
      "loss": 5.3194,
      "mean_token_accuracy": 0.17262864708900452,
      "num_tokens": 25665072.0,
      "step": 14795
    },
    {
      "entropy": 5.525521373748779,
      "epoch": 1.1514880373468197,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004871233082108135,
      "loss": 5.3553,
      "mean_token_accuracy": 0.1660602405667305,
      "num_tokens": 25673877.0,
      "step": 14800
    },
    {
      "entropy": 5.536645364761353,
      "epoch": 1.1518770667185372,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004871140650326228,
      "loss": 5.2168,
      "mean_token_accuracy": 0.17195206135511398,
      "num_tokens": 25683238.0,
      "step": 14805
    },
    {
      "entropy": 5.523863363265991,
      "epoch": 1.1522660960902549,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004871048186359389,
      "loss": 5.3169,
      "mean_token_accuracy": 0.16545094251632692,
      "num_tokens": 25692654.0,
      "step": 14810
    },
    {
      "entropy": 5.554662799835205,
      "epoch": 1.1526551254619724,
      "grad_norm": 1.546875,
      "learning_rate": 0.000487095569020902,
      "loss": 5.2911,
      "mean_token_accuracy": 0.17407094240188598,
      "num_tokens": 25701681.0,
      "step": 14815
    },
    {
      "entropy": 5.5317220211029055,
      "epoch": 1.1530441548336898,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004870863161876524,
      "loss": 5.2007,
      "mean_token_accuracy": 0.17962604016065598,
      "num_tokens": 25709848.0,
      "step": 14820
    },
    {
      "entropy": 5.581866836547851,
      "epoch": 1.1534331842054075,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004870770601363305,
      "loss": 5.4436,
      "mean_token_accuracy": 0.17273078709840775,
      "num_tokens": 25717993.0,
      "step": 14825
    },
    {
      "entropy": 5.551402950286866,
      "epoch": 1.153822213577125,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048706780086707675,
      "loss": 5.3365,
      "mean_token_accuracy": 0.16836561262607574,
      "num_tokens": 25726379.0,
      "step": 14830
    },
    {
      "entropy": 5.487813282012939,
      "epoch": 1.1542112429488427,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004870585383800317,
      "loss": 5.2399,
      "mean_token_accuracy": 0.17993202358484267,
      "num_tokens": 25734968.0,
      "step": 14835
    },
    {
      "entropy": 5.536644840240479,
      "epoch": 1.1546002723205602,
      "grad_norm": 1.375,
      "learning_rate": 0.00048704927267533585,
      "loss": 5.2758,
      "mean_token_accuracy": 0.17906226515769957,
      "num_tokens": 25743315.0,
      "step": 14840
    },
    {
      "entropy": 5.513940095901489,
      "epoch": 1.1549893016922779,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048704000375312975,
      "loss": 5.1979,
      "mean_token_accuracy": 0.178344988822937,
      "num_tokens": 25752396.0,
      "step": 14845
    },
    {
      "entropy": 5.526822137832641,
      "epoch": 1.1553783310639953,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048703073161355414,
      "loss": 5.3136,
      "mean_token_accuracy": 0.17459192276000976,
      "num_tokens": 25761142.0,
      "step": 14850
    },
    {
      "entropy": 5.440198183059692,
      "epoch": 1.1557673604357128,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004870214562567497,
      "loss": 5.1946,
      "mean_token_accuracy": 0.17491271942853928,
      "num_tokens": 25769514.0,
      "step": 14855
    },
    {
      "entropy": 5.482697868347168,
      "epoch": 1.1561563898074305,
      "grad_norm": 1.390625,
      "learning_rate": 0.000487012177682857,
      "loss": 5.1951,
      "mean_token_accuracy": 0.17698332220315932,
      "num_tokens": 25777408.0,
      "step": 14860
    },
    {
      "entropy": 5.420403099060058,
      "epoch": 1.156545419179148,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004870028958920171,
      "loss": 5.2026,
      "mean_token_accuracy": 0.17721671164035796,
      "num_tokens": 25785646.0,
      "step": 14865
    },
    {
      "entropy": 5.519942331314087,
      "epoch": 1.1569344485508655,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004869936108843706,
      "loss": 5.2614,
      "mean_token_accuracy": 0.17413395792245864,
      "num_tokens": 25794007.0,
      "step": 14870
    },
    {
      "entropy": 5.480394506454468,
      "epoch": 1.1573234779225832,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048698432266005854,
      "loss": 5.1957,
      "mean_token_accuracy": 0.17586324959993363,
      "num_tokens": 25802765.0,
      "step": 14875
    },
    {
      "entropy": 5.530696439743042,
      "epoch": 1.1577125072943006,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004869750312192217,
      "loss": 5.3138,
      "mean_token_accuracy": 0.17150983065366746,
      "num_tokens": 25811214.0,
      "step": 14880
    },
    {
      "entropy": 5.516277170181274,
      "epoch": 1.1581015366660183,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048696573656200123,
      "loss": 5.3145,
      "mean_token_accuracy": 0.17341137528419495,
      "num_tokens": 25820350.0,
      "step": 14885
    },
    {
      "entropy": 5.505978298187256,
      "epoch": 1.1584905660377358,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048695643868853806,
      "loss": 5.2414,
      "mean_token_accuracy": 0.18219964504241942,
      "num_tokens": 25828088.0,
      "step": 14890
    },
    {
      "entropy": 5.434723520278931,
      "epoch": 1.1588795954094535,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004869471375989733,
      "loss": 5.1227,
      "mean_token_accuracy": 0.1857884258031845,
      "num_tokens": 25835909.0,
      "step": 14895
    },
    {
      "entropy": 5.381528663635254,
      "epoch": 1.159268624781171,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004869378332934481,
      "loss": 5.1922,
      "mean_token_accuracy": 0.17694181501865386,
      "num_tokens": 25844360.0,
      "step": 14900
    },
    {
      "entropy": 5.59519944190979,
      "epoch": 1.1596576541528885,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048692852577210356,
      "loss": 5.3284,
      "mean_token_accuracy": 0.16598222106695176,
      "num_tokens": 25853877.0,
      "step": 14905
    },
    {
      "entropy": 5.511133241653442,
      "epoch": 1.1600466835246062,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000486919215035081,
      "loss": 5.1959,
      "mean_token_accuracy": 0.18456930071115493,
      "num_tokens": 25862200.0,
      "step": 14910
    },
    {
      "entropy": 5.413064575195312,
      "epoch": 1.1604357128963236,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004869099010825217,
      "loss": 5.1764,
      "mean_token_accuracy": 0.17527795135974883,
      "num_tokens": 25871112.0,
      "step": 14915
    },
    {
      "entropy": 5.442687892913819,
      "epoch": 1.1608247422680413,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048690058391456684,
      "loss": 5.3402,
      "mean_token_accuracy": 0.17317889034748077,
      "num_tokens": 25880135.0,
      "step": 14920
    },
    {
      "entropy": 5.5458659648895265,
      "epoch": 1.1612137716397588,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004868912635313579,
      "loss": 5.2346,
      "mean_token_accuracy": 0.17990500777959822,
      "num_tokens": 25888609.0,
      "step": 14925
    },
    {
      "entropy": 5.481937456130981,
      "epoch": 1.1616028010114763,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004868819399330364,
      "loss": 5.2865,
      "mean_token_accuracy": 0.17359334379434585,
      "num_tokens": 25897254.0,
      "step": 14930
    },
    {
      "entropy": 5.4619469165802,
      "epoch": 1.161991830383194,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004868726131197436,
      "loss": 5.1936,
      "mean_token_accuracy": 0.1828017368912697,
      "num_tokens": 25905705.0,
      "step": 14935
    },
    {
      "entropy": 5.484757900238037,
      "epoch": 1.1623808597549115,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004868632830916212,
      "loss": 5.2544,
      "mean_token_accuracy": 0.1740899220108986,
      "num_tokens": 25914066.0,
      "step": 14940
    },
    {
      "entropy": 5.450885963439942,
      "epoch": 1.1627698891266292,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004868539498488106,
      "loss": 5.2541,
      "mean_token_accuracy": 0.1756904825568199,
      "num_tokens": 25922949.0,
      "step": 14945
    },
    {
      "entropy": 5.376556348800659,
      "epoch": 1.1631589184983466,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004868446133914536,
      "loss": 5.1144,
      "mean_token_accuracy": 0.185198412835598,
      "num_tokens": 25931465.0,
      "step": 14950
    },
    {
      "entropy": 5.49980320930481,
      "epoch": 1.163547947870064,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048683527371969185,
      "loss": 5.2935,
      "mean_token_accuracy": 0.173323193192482,
      "num_tokens": 25941014.0,
      "step": 14955
    },
    {
      "entropy": 5.412701034545899,
      "epoch": 1.1639369772417818,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048682593083366685,
      "loss": 5.153,
      "mean_token_accuracy": 0.17737683802843093,
      "num_tokens": 25950275.0,
      "step": 14960
    },
    {
      "entropy": 5.430574989318847,
      "epoch": 1.1643260066134993,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048681658473352055,
      "loss": 5.2787,
      "mean_token_accuracy": 0.17585472017526627,
      "num_tokens": 25959138.0,
      "step": 14965
    },
    {
      "entropy": 5.522162342071534,
      "epoch": 1.164715035985217,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004868072354193947,
      "loss": 5.1588,
      "mean_token_accuracy": 0.17691579163074495,
      "num_tokens": 25967221.0,
      "step": 14970
    },
    {
      "entropy": 5.476398229598999,
      "epoch": 1.1651040653569344,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004867978828914312,
      "loss": 5.2752,
      "mean_token_accuracy": 0.17386409789323806,
      "num_tokens": 25976194.0,
      "step": 14975
    },
    {
      "entropy": 5.442127990722656,
      "epoch": 1.165493094728652,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004867885271497719,
      "loss": 5.2444,
      "mean_token_accuracy": 0.17414329797029496,
      "num_tokens": 25985493.0,
      "step": 14980
    },
    {
      "entropy": 5.537582635879517,
      "epoch": 1.1658821241003696,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004867791681945588,
      "loss": 5.2021,
      "mean_token_accuracy": 0.176902873814106,
      "num_tokens": 25994213.0,
      "step": 14985
    },
    {
      "entropy": 5.3891276836395265,
      "epoch": 1.166271153472087,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048676980602593395,
      "loss": 5.157,
      "mean_token_accuracy": 0.18316584080457687,
      "num_tokens": 26003164.0,
      "step": 14990
    },
    {
      "entropy": 5.484225416183472,
      "epoch": 1.1666601828438048,
      "grad_norm": 1.375,
      "learning_rate": 0.00048676044064403936,
      "loss": 5.2852,
      "mean_token_accuracy": 0.17339709401130676,
      "num_tokens": 26012049.0,
      "step": 14995
    },
    {
      "entropy": 5.579562759399414,
      "epoch": 1.1670492122155223,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004867510720490171,
      "loss": 5.3667,
      "mean_token_accuracy": 0.1652686595916748,
      "num_tokens": 26020466.0,
      "step": 15000
    },
    {
      "epoch": 1.1670492122155223,
      "eval_entropy": 5.363034827716082,
      "eval_loss": 5.3619585037231445,
      "eval_mean_token_accuracy": 0.18032150644023792,
      "eval_num_tokens": 26020466.0,
      "eval_runtime": 29.0748,
      "eval_samples_per_second": 1429.347,
      "eval_steps_per_second": 178.677,
      "step": 15000
    },
    {
      "entropy": 5.429365777969361,
      "epoch": 1.1674382415872397,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004867417002410094,
      "loss": 5.1678,
      "mean_token_accuracy": 0.17899539917707444,
      "num_tokens": 26028815.0,
      "step": 15005
    },
    {
      "entropy": 5.56017165184021,
      "epoch": 1.1678272709589574,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048673232522015845,
      "loss": 5.3553,
      "mean_token_accuracy": 0.1708623871207237,
      "num_tokens": 26037759.0,
      "step": 15010
    },
    {
      "entropy": 5.539667701721191,
      "epoch": 1.168216300330675,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004867229469866064,
      "loss": 5.2735,
      "mean_token_accuracy": 0.16781237125396728,
      "num_tokens": 26046349.0,
      "step": 15015
    },
    {
      "entropy": 5.4890283107757565,
      "epoch": 1.1686053297023926,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004867135655404956,
      "loss": 5.2709,
      "mean_token_accuracy": 0.18057565838098527,
      "num_tokens": 26055265.0,
      "step": 15020
    },
    {
      "entropy": 5.486582279205322,
      "epoch": 1.16899435907411,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048670418088196844,
      "loss": 5.2534,
      "mean_token_accuracy": 0.17626446038484572,
      "num_tokens": 26064335.0,
      "step": 15025
    },
    {
      "entropy": 5.495803356170654,
      "epoch": 1.1693833884458276,
      "grad_norm": 1.25,
      "learning_rate": 0.0004866947930111674,
      "loss": 5.187,
      "mean_token_accuracy": 0.17860865592956543,
      "num_tokens": 26073044.0,
      "step": 15030
    },
    {
      "entropy": 5.43163685798645,
      "epoch": 1.1697724178175453,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048668540192823467,
      "loss": 5.2098,
      "mean_token_accuracy": 0.18019946962594985,
      "num_tokens": 26081361.0,
      "step": 15035
    },
    {
      "entropy": 5.441382026672363,
      "epoch": 1.1701614471892627,
      "grad_norm": 1.546875,
      "learning_rate": 0.000486676007633313,
      "loss": 5.217,
      "mean_token_accuracy": 0.17552125900983812,
      "num_tokens": 26089740.0,
      "step": 15040
    },
    {
      "entropy": 5.572740173339843,
      "epoch": 1.1705504765609804,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004866666101265448,
      "loss": 5.3051,
      "mean_token_accuracy": 0.17491288036108016,
      "num_tokens": 26098906.0,
      "step": 15045
    },
    {
      "entropy": 5.503249788284302,
      "epoch": 1.170939505932698,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048665720940807274,
      "loss": 5.286,
      "mean_token_accuracy": 0.17210738509893417,
      "num_tokens": 26107978.0,
      "step": 15050
    },
    {
      "entropy": 5.481639385223389,
      "epoch": 1.1713285353044154,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048664780547803935,
      "loss": 5.2272,
      "mean_token_accuracy": 0.17455514073371886,
      "num_tokens": 26116756.0,
      "step": 15055
    },
    {
      "entropy": 5.6110070705413815,
      "epoch": 1.171717564676133,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004866383983365873,
      "loss": 5.309,
      "mean_token_accuracy": 0.16427583545446395,
      "num_tokens": 26125078.0,
      "step": 15060
    },
    {
      "entropy": 5.558560228347778,
      "epoch": 1.1721065940478506,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048662898798385956,
      "loss": 5.3185,
      "mean_token_accuracy": 0.1728869527578354,
      "num_tokens": 26134149.0,
      "step": 15065
    },
    {
      "entropy": 5.439807510375976,
      "epoch": 1.1724956234195683,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048661957441999875,
      "loss": 5.186,
      "mean_token_accuracy": 0.18209881335496902,
      "num_tokens": 26142789.0,
      "step": 15070
    },
    {
      "entropy": 5.512610387802124,
      "epoch": 1.1728846527912857,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048661015764514765,
      "loss": 5.3502,
      "mean_token_accuracy": 0.16991154849529266,
      "num_tokens": 26151189.0,
      "step": 15075
    },
    {
      "entropy": 5.478395938873291,
      "epoch": 1.1732736821630034,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048660073765944927,
      "loss": 5.2346,
      "mean_token_accuracy": 0.17700796127319335,
      "num_tokens": 26159495.0,
      "step": 15080
    },
    {
      "entropy": 5.4695127487182615,
      "epoch": 1.173662711534721,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004865913144630465,
      "loss": 5.1739,
      "mean_token_accuracy": 0.18347302973270416,
      "num_tokens": 26167804.0,
      "step": 15085
    },
    {
      "entropy": 5.501602697372436,
      "epoch": 1.1740517409064384,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048658188805608227,
      "loss": 5.2159,
      "mean_token_accuracy": 0.17901125103235244,
      "num_tokens": 26176240.0,
      "step": 15090
    },
    {
      "entropy": 5.491896677017212,
      "epoch": 1.174440770278156,
      "grad_norm": 1.375,
      "learning_rate": 0.0004865724584386998,
      "loss": 5.2317,
      "mean_token_accuracy": 0.1771928310394287,
      "num_tokens": 26184580.0,
      "step": 15095
    },
    {
      "entropy": 5.5088879585266115,
      "epoch": 1.1748297996498736,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004865630256110418,
      "loss": 5.2881,
      "mean_token_accuracy": 0.1742125079035759,
      "num_tokens": 26193756.0,
      "step": 15100
    },
    {
      "entropy": 5.570204448699951,
      "epoch": 1.175218829021591,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048655358957325174,
      "loss": 5.261,
      "mean_token_accuracy": 0.18112905472517013,
      "num_tokens": 26201818.0,
      "step": 15105
    },
    {
      "entropy": 5.500771331787109,
      "epoch": 1.1756078583933087,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048654415032547277,
      "loss": 5.3483,
      "mean_token_accuracy": 0.16415801793336868,
      "num_tokens": 26211817.0,
      "step": 15110
    },
    {
      "entropy": 5.491998720169067,
      "epoch": 1.1759968877650262,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048653470786784794,
      "loss": 5.2023,
      "mean_token_accuracy": 0.1740380972623825,
      "num_tokens": 26220280.0,
      "step": 15115
    },
    {
      "entropy": 5.4378245830535885,
      "epoch": 1.176385917136744,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004865252622005206,
      "loss": 5.1643,
      "mean_token_accuracy": 0.18348196297883987,
      "num_tokens": 26228719.0,
      "step": 15120
    },
    {
      "entropy": 5.533568286895752,
      "epoch": 1.1767749465084614,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048651581332363405,
      "loss": 5.2636,
      "mean_token_accuracy": 0.17689931094646455,
      "num_tokens": 26237462.0,
      "step": 15125
    },
    {
      "entropy": 5.573260116577148,
      "epoch": 1.177163975880179,
      "grad_norm": 1.921875,
      "learning_rate": 0.00048650636123733176,
      "loss": 5.3615,
      "mean_token_accuracy": 0.1745436668395996,
      "num_tokens": 26245747.0,
      "step": 15130
    },
    {
      "entropy": 5.56461181640625,
      "epoch": 1.1775530052518965,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004864969059417571,
      "loss": 5.232,
      "mean_token_accuracy": 0.1735546410083771,
      "num_tokens": 26254977.0,
      "step": 15135
    },
    {
      "entropy": 5.546353769302368,
      "epoch": 1.177942034623614,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004864874474370534,
      "loss": 5.3661,
      "mean_token_accuracy": 0.16523050665855407,
      "num_tokens": 26264298.0,
      "step": 15140
    },
    {
      "entropy": 5.501612377166748,
      "epoch": 1.1783310639953317,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048647798572336445,
      "loss": 5.1925,
      "mean_token_accuracy": 0.16971518099308014,
      "num_tokens": 26273454.0,
      "step": 15145
    },
    {
      "entropy": 5.549590682983398,
      "epoch": 1.1787200933670492,
      "grad_norm": 1.375,
      "learning_rate": 0.00048646852080083355,
      "loss": 5.2297,
      "mean_token_accuracy": 0.17417798191308975,
      "num_tokens": 26282584.0,
      "step": 15150
    },
    {
      "entropy": 5.489213132858277,
      "epoch": 1.1791091227387667,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004864590526696045,
      "loss": 5.2759,
      "mean_token_accuracy": 0.1778562471270561,
      "num_tokens": 26291633.0,
      "step": 15155
    },
    {
      "entropy": 5.520056247711182,
      "epoch": 1.1794981521104844,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048644958132982094,
      "loss": 5.284,
      "mean_token_accuracy": 0.1749679461121559,
      "num_tokens": 26300808.0,
      "step": 15160
    },
    {
      "entropy": 5.477617883682251,
      "epoch": 1.1798871814822018,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048644010678162644,
      "loss": 5.2118,
      "mean_token_accuracy": 0.18650766909122468,
      "num_tokens": 26308976.0,
      "step": 15165
    },
    {
      "entropy": 5.5196428298950195,
      "epoch": 1.1802762108539195,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004864306290251649,
      "loss": 5.2656,
      "mean_token_accuracy": 0.17552779763936996,
      "num_tokens": 26316980.0,
      "step": 15170
    },
    {
      "entropy": 5.441870212554932,
      "epoch": 1.180665240225637,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048642114806058014,
      "loss": 5.1578,
      "mean_token_accuracy": 0.1854674518108368,
      "num_tokens": 26325723.0,
      "step": 15175
    },
    {
      "entropy": 5.433674240112305,
      "epoch": 1.1810542695973547,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004864116638880159,
      "loss": 5.2721,
      "mean_token_accuracy": 0.1736752763390541,
      "num_tokens": 26334851.0,
      "step": 15180
    },
    {
      "entropy": 5.571105480194092,
      "epoch": 1.1814432989690722,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004864021765076162,
      "loss": 5.2562,
      "mean_token_accuracy": 0.17699959874153137,
      "num_tokens": 26343696.0,
      "step": 15185
    },
    {
      "entropy": 5.592726135253907,
      "epoch": 1.1818323283407897,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048639268591952494,
      "loss": 5.3919,
      "mean_token_accuracy": 0.1697914034128189,
      "num_tokens": 26353054.0,
      "step": 15190
    },
    {
      "entropy": 5.50649790763855,
      "epoch": 1.1822213577125074,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048638319212388614,
      "loss": 5.2262,
      "mean_token_accuracy": 0.17555565685033797,
      "num_tokens": 26361484.0,
      "step": 15195
    },
    {
      "entropy": 5.52585277557373,
      "epoch": 1.1826103870842248,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004863736951208438,
      "loss": 5.3032,
      "mean_token_accuracy": 0.16765618026256562,
      "num_tokens": 26370402.0,
      "step": 15200
    },
    {
      "entropy": 5.5697746753692625,
      "epoch": 1.1829994164559423,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048636419491054217,
      "loss": 5.3576,
      "mean_token_accuracy": 0.16881369799375534,
      "num_tokens": 26378746.0,
      "step": 15205
    },
    {
      "entropy": 5.517532300949097,
      "epoch": 1.18338844582766,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004863546914931252,
      "loss": 5.2247,
      "mean_token_accuracy": 0.1864602491259575,
      "num_tokens": 26386984.0,
      "step": 15210
    },
    {
      "entropy": 5.61485915184021,
      "epoch": 1.1837774751993775,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004863451848687373,
      "loss": 5.3521,
      "mean_token_accuracy": 0.16712313741445542,
      "num_tokens": 26395712.0,
      "step": 15215
    },
    {
      "entropy": 5.503615188598633,
      "epoch": 1.1841665045710952,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048633567503752253,
      "loss": 5.287,
      "mean_token_accuracy": 0.17407525479793548,
      "num_tokens": 26406421.0,
      "step": 15220
    },
    {
      "entropy": 5.567384481430054,
      "epoch": 1.1845555339428127,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004863261619996253,
      "loss": 5.2609,
      "mean_token_accuracy": 0.1733366996049881,
      "num_tokens": 26416032.0,
      "step": 15225
    },
    {
      "entropy": 5.487317132949829,
      "epoch": 1.1849445633145304,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004863166457551899,
      "loss": 5.202,
      "mean_token_accuracy": 0.17901441156864167,
      "num_tokens": 26423903.0,
      "step": 15230
    },
    {
      "entropy": 5.508638477325439,
      "epoch": 1.1853335926862478,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004863071263043608,
      "loss": 5.279,
      "mean_token_accuracy": 0.1764646962285042,
      "num_tokens": 26432631.0,
      "step": 15235
    },
    {
      "entropy": 5.498473644256592,
      "epoch": 1.1857226220579653,
      "grad_norm": 1.375,
      "learning_rate": 0.0004862976036472823,
      "loss": 5.2876,
      "mean_token_accuracy": 0.17676588296890258,
      "num_tokens": 26441529.0,
      "step": 15240
    },
    {
      "entropy": 5.517298841476441,
      "epoch": 1.186111651429683,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048628807778409907,
      "loss": 5.1851,
      "mean_token_accuracy": 0.18358005285263063,
      "num_tokens": 26449928.0,
      "step": 15245
    },
    {
      "entropy": 5.493913364410401,
      "epoch": 1.1865006808014005,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004862785487149555,
      "loss": 5.1759,
      "mean_token_accuracy": 0.18420424610376357,
      "num_tokens": 26458065.0,
      "step": 15250
    },
    {
      "entropy": 5.442678070068359,
      "epoch": 1.186889710173118,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004862690164399963,
      "loss": 5.2206,
      "mean_token_accuracy": 0.1823731243610382,
      "num_tokens": 26466690.0,
      "step": 15255
    },
    {
      "entropy": 5.411856412887573,
      "epoch": 1.1872787395448356,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048625948095936593,
      "loss": 5.2752,
      "mean_token_accuracy": 0.17093705832958223,
      "num_tokens": 26476133.0,
      "step": 15260
    },
    {
      "entropy": 5.541415882110596,
      "epoch": 1.1876677689165531,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048624994227320923,
      "loss": 5.2253,
      "mean_token_accuracy": 0.174331833422184,
      "num_tokens": 26485151.0,
      "step": 15265
    },
    {
      "entropy": 5.485921955108642,
      "epoch": 1.1880567982882708,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048624040038167094,
      "loss": 5.2913,
      "mean_token_accuracy": 0.1717325419187546,
      "num_tokens": 26494178.0,
      "step": 15270
    },
    {
      "entropy": 5.541491079330444,
      "epoch": 1.1884458276599883,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048623085528489584,
      "loss": 5.278,
      "mean_token_accuracy": 0.18224940299987794,
      "num_tokens": 26503233.0,
      "step": 15275
    },
    {
      "entropy": 5.470219945907592,
      "epoch": 1.188834857031706,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048622130698302863,
      "loss": 5.1504,
      "mean_token_accuracy": 0.18698617815971375,
      "num_tokens": 26512127.0,
      "step": 15280
    },
    {
      "entropy": 5.428116750717163,
      "epoch": 1.1892238864034235,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048621175547621426,
      "loss": 5.1203,
      "mean_token_accuracy": 0.18726104497909546,
      "num_tokens": 26520239.0,
      "step": 15285
    },
    {
      "entropy": 5.367747688293457,
      "epoch": 1.189612915775141,
      "grad_norm": 1.25,
      "learning_rate": 0.0004862022007645978,
      "loss": 5.2756,
      "mean_token_accuracy": 0.17212725132703782,
      "num_tokens": 26528946.0,
      "step": 15290
    },
    {
      "entropy": 5.533008098602295,
      "epoch": 1.1900019451468586,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048619264284832403,
      "loss": 5.2551,
      "mean_token_accuracy": 0.1718164160847664,
      "num_tokens": 26536825.0,
      "step": 15295
    },
    {
      "entropy": 5.484944820404053,
      "epoch": 1.1903909745185761,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048618308172753804,
      "loss": 5.199,
      "mean_token_accuracy": 0.18474840074777604,
      "num_tokens": 26545208.0,
      "step": 15300
    },
    {
      "entropy": 5.480442190170288,
      "epoch": 1.1907800038902936,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004861735174023849,
      "loss": 5.206,
      "mean_token_accuracy": 0.17705218493938446,
      "num_tokens": 26553768.0,
      "step": 15305
    },
    {
      "entropy": 5.5430488109588625,
      "epoch": 1.1911690332620113,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048616394987300985,
      "loss": 5.3134,
      "mean_token_accuracy": 0.17014084607362748,
      "num_tokens": 26561582.0,
      "step": 15310
    },
    {
      "entropy": 5.416816759109497,
      "epoch": 1.1915580626337288,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004861543791395579,
      "loss": 5.2145,
      "mean_token_accuracy": 0.17560068964958192,
      "num_tokens": 26570632.0,
      "step": 15315
    },
    {
      "entropy": 5.442972564697266,
      "epoch": 1.1919470920054465,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004861448052021743,
      "loss": 5.1789,
      "mean_token_accuracy": 0.1821298435330391,
      "num_tokens": 26579507.0,
      "step": 15320
    },
    {
      "entropy": 5.4764632225036625,
      "epoch": 1.192336121377164,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004861352280610044,
      "loss": 5.2098,
      "mean_token_accuracy": 0.1804475247859955,
      "num_tokens": 26588765.0,
      "step": 15325
    },
    {
      "entropy": 5.531530094146729,
      "epoch": 1.1927251507488816,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048612564771619346,
      "loss": 5.2634,
      "mean_token_accuracy": 0.17586708664894105,
      "num_tokens": 26597529.0,
      "step": 15330
    },
    {
      "entropy": 5.511665201187133,
      "epoch": 1.193114180120599,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048611606416788686,
      "loss": 5.2208,
      "mean_token_accuracy": 0.17744175791740419,
      "num_tokens": 26605960.0,
      "step": 15335
    },
    {
      "entropy": 5.468547582626343,
      "epoch": 1.1935032094923166,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048610647741622996,
      "loss": 5.1382,
      "mean_token_accuracy": 0.18745063990354538,
      "num_tokens": 26614435.0,
      "step": 15340
    },
    {
      "entropy": 5.487285137176514,
      "epoch": 1.1938922388640343,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004860968874613683,
      "loss": 5.2539,
      "mean_token_accuracy": 0.1761014848947525,
      "num_tokens": 26622873.0,
      "step": 15345
    },
    {
      "entropy": 5.472904205322266,
      "epoch": 1.1942812682357518,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004860872943034474,
      "loss": 5.1813,
      "mean_token_accuracy": 0.18527811616659165,
      "num_tokens": 26631529.0,
      "step": 15350
    },
    {
      "entropy": 5.521970319747925,
      "epoch": 1.1946702976074695,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004860776979426128,
      "loss": 5.2893,
      "mean_token_accuracy": 0.16512622386217118,
      "num_tokens": 26639466.0,
      "step": 15355
    },
    {
      "entropy": 5.3834123611450195,
      "epoch": 1.195059326979187,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004860680983790101,
      "loss": 5.2185,
      "mean_token_accuracy": 0.17507579922676086,
      "num_tokens": 26648329.0,
      "step": 15360
    },
    {
      "entropy": 5.505174350738526,
      "epoch": 1.1954483563509044,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004860584956127849,
      "loss": 5.1562,
      "mean_token_accuracy": 0.17604049146175385,
      "num_tokens": 26656899.0,
      "step": 15365
    },
    {
      "entropy": 5.5684130668640135,
      "epoch": 1.195837385722622,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048604888964408306,
      "loss": 5.393,
      "mean_token_accuracy": 0.17005542516708375,
      "num_tokens": 26666155.0,
      "step": 15370
    },
    {
      "entropy": 5.426250553131103,
      "epoch": 1.1962264150943396,
      "grad_norm": 1.5,
      "learning_rate": 0.0004860392804730502,
      "loss": 5.1771,
      "mean_token_accuracy": 0.18689159005880357,
      "num_tokens": 26675567.0,
      "step": 15375
    },
    {
      "entropy": 5.480653524398804,
      "epoch": 1.1966154444660573,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048602966809983204,
      "loss": 5.2339,
      "mean_token_accuracy": 0.17842643707990646,
      "num_tokens": 26684133.0,
      "step": 15380
    },
    {
      "entropy": 5.5077722549438475,
      "epoch": 1.1970044738377748,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004860200525245746,
      "loss": 5.3589,
      "mean_token_accuracy": 0.16625821888446807,
      "num_tokens": 26692746.0,
      "step": 15385
    },
    {
      "entropy": 5.5773721694946286,
      "epoch": 1.1973935032094922,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004860104337474239,
      "loss": 5.2429,
      "mean_token_accuracy": 0.1809867650270462,
      "num_tokens": 26701887.0,
      "step": 15390
    },
    {
      "entropy": 5.48978443145752,
      "epoch": 1.19778253258121,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048600081176852555,
      "loss": 5.2221,
      "mean_token_accuracy": 0.1759009376168251,
      "num_tokens": 26710346.0,
      "step": 15395
    },
    {
      "entropy": 5.387714052200318,
      "epoch": 1.1981715619529274,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048599118658802575,
      "loss": 5.1303,
      "mean_token_accuracy": 0.18023394346237182,
      "num_tokens": 26718775.0,
      "step": 15400
    },
    {
      "entropy": 5.442866277694702,
      "epoch": 1.198560591324645,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004859815582060706,
      "loss": 5.2104,
      "mean_token_accuracy": 0.17905230522155763,
      "num_tokens": 26726683.0,
      "step": 15405
    },
    {
      "entropy": 5.456772518157959,
      "epoch": 1.1989496206963626,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004859719266228061,
      "loss": 5.2188,
      "mean_token_accuracy": 0.1807861492037773,
      "num_tokens": 26735450.0,
      "step": 15410
    },
    {
      "entropy": 5.470722723007202,
      "epoch": 1.19933865006808,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004859622918383784,
      "loss": 5.1844,
      "mean_token_accuracy": 0.17846325039863586,
      "num_tokens": 26744006.0,
      "step": 15415
    },
    {
      "entropy": 5.395093488693237,
      "epoch": 1.1997276794397977,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004859526538529337,
      "loss": 5.0791,
      "mean_token_accuracy": 0.19092137962579728,
      "num_tokens": 26751964.0,
      "step": 15420
    },
    {
      "entropy": 5.373332214355469,
      "epoch": 1.2001167088115152,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004859430126666182,
      "loss": 5.158,
      "mean_token_accuracy": 0.17934513241052627,
      "num_tokens": 26760766.0,
      "step": 15425
    },
    {
      "entropy": 5.467801952362061,
      "epoch": 1.200505738183233,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048593336827957824,
      "loss": 5.1765,
      "mean_token_accuracy": 0.17549433708190917,
      "num_tokens": 26769663.0,
      "step": 15430
    },
    {
      "entropy": 5.542212343215942,
      "epoch": 1.2008947675549504,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048592372069196024,
      "loss": 5.3952,
      "mean_token_accuracy": 0.1635242611169815,
      "num_tokens": 26778693.0,
      "step": 15435
    },
    {
      "entropy": 5.450088119506836,
      "epoch": 1.2012837969266679,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004859140699039104,
      "loss": 5.1711,
      "mean_token_accuracy": 0.18602428883314132,
      "num_tokens": 26787368.0,
      "step": 15440
    },
    {
      "entropy": 5.410736179351806,
      "epoch": 1.2016728262983856,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048590441591557526,
      "loss": 5.12,
      "mean_token_accuracy": 0.18972058296203614,
      "num_tokens": 26795559.0,
      "step": 15445
    },
    {
      "entropy": 5.416481351852417,
      "epoch": 1.202061855670103,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048589475872710134,
      "loss": 5.1157,
      "mean_token_accuracy": 0.1836517184972763,
      "num_tokens": 26803657.0,
      "step": 15450
    },
    {
      "entropy": 5.561429738998413,
      "epoch": 1.2024508850418207,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004858850983386351,
      "loss": 5.3033,
      "mean_token_accuracy": 0.17373560220003129,
      "num_tokens": 26812214.0,
      "step": 15455
    },
    {
      "entropy": 5.476353693008423,
      "epoch": 1.2028399144135382,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004858754347503231,
      "loss": 5.1494,
      "mean_token_accuracy": 0.19088565707206726,
      "num_tokens": 26820591.0,
      "step": 15460
    },
    {
      "entropy": 5.410498046875,
      "epoch": 1.203228943785256,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048586576796231216,
      "loss": 5.2493,
      "mean_token_accuracy": 0.17892842441797258,
      "num_tokens": 26829228.0,
      "step": 15465
    },
    {
      "entropy": 5.5283650875091555,
      "epoch": 1.2036179731569734,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004858560979747487,
      "loss": 5.3975,
      "mean_token_accuracy": 0.16485095471143724,
      "num_tokens": 26838083.0,
      "step": 15470
    },
    {
      "entropy": 5.475933933258057,
      "epoch": 1.2040070025286909,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048584642478777967,
      "loss": 5.0762,
      "mean_token_accuracy": 0.18039197027683257,
      "num_tokens": 26846796.0,
      "step": 15475
    },
    {
      "entropy": 5.44716854095459,
      "epoch": 1.2043960319004086,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004858367484015517,
      "loss": 5.2052,
      "mean_token_accuracy": 0.18125954568386077,
      "num_tokens": 26854874.0,
      "step": 15480
    },
    {
      "entropy": 5.431047964096069,
      "epoch": 1.204785061272126,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048582706881621166,
      "loss": 5.2305,
      "mean_token_accuracy": 0.1830142393708229,
      "num_tokens": 26863040.0,
      "step": 15485
    },
    {
      "entropy": 5.489552640914917,
      "epoch": 1.2051740906438435,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004858173860319063,
      "loss": 5.2215,
      "mean_token_accuracy": 0.17290144711732863,
      "num_tokens": 26872183.0,
      "step": 15490
    },
    {
      "entropy": 5.526428508758545,
      "epoch": 1.2055631200155612,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048580770004878277,
      "loss": 5.1964,
      "mean_token_accuracy": 0.17603176683187485,
      "num_tokens": 26880480.0,
      "step": 15495
    },
    {
      "entropy": 5.463943195343018,
      "epoch": 1.2059521493872787,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004857980108669879,
      "loss": 5.2231,
      "mean_token_accuracy": 0.18189868032932283,
      "num_tokens": 26888855.0,
      "step": 15500
    },
    {
      "entropy": 5.449655342102051,
      "epoch": 1.2063411787589964,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048578831848666875,
      "loss": 5.2273,
      "mean_token_accuracy": 0.17638375014066696,
      "num_tokens": 26897692.0,
      "step": 15505
    },
    {
      "entropy": 5.453392124176025,
      "epoch": 1.2067302081307139,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048577862290797235,
      "loss": 5.1995,
      "mean_token_accuracy": 0.1740122839808464,
      "num_tokens": 26906502.0,
      "step": 15510
    },
    {
      "entropy": 5.534039497375488,
      "epoch": 1.2071192375024316,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048576892413104586,
      "loss": 5.2325,
      "mean_token_accuracy": 0.17494663894176482,
      "num_tokens": 26915246.0,
      "step": 15515
    },
    {
      "entropy": 5.580176496505738,
      "epoch": 1.207508266874149,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004857592221560364,
      "loss": 5.3566,
      "mean_token_accuracy": 0.17636353969573976,
      "num_tokens": 26923901.0,
      "step": 15520
    },
    {
      "entropy": 5.539089488983154,
      "epoch": 1.2078972962458665,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004857495169830912,
      "loss": 5.4099,
      "mean_token_accuracy": 0.17498260587453843,
      "num_tokens": 26933246.0,
      "step": 15525
    },
    {
      "entropy": 5.490329933166504,
      "epoch": 1.2082863256175842,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004857398086123575,
      "loss": 5.2654,
      "mean_token_accuracy": 0.17963187843561174,
      "num_tokens": 26941938.0,
      "step": 15530
    },
    {
      "entropy": 5.446584558486938,
      "epoch": 1.2086753549893017,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004857300970439827,
      "loss": 5.2213,
      "mean_token_accuracy": 0.17509732097387315,
      "num_tokens": 26949967.0,
      "step": 15535
    },
    {
      "entropy": 5.465288496017456,
      "epoch": 1.2090643843610192,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048572038227811404,
      "loss": 5.2898,
      "mean_token_accuracy": 0.17573822140693665,
      "num_tokens": 26958743.0,
      "step": 15540
    },
    {
      "entropy": 5.587131595611572,
      "epoch": 1.2094534137327368,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000485710664314899,
      "loss": 5.2848,
      "mean_token_accuracy": 0.18044032901525497,
      "num_tokens": 26967338.0,
      "step": 15545
    },
    {
      "entropy": 5.43191499710083,
      "epoch": 1.2098424431044543,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000485700943154485,
      "loss": 5.1611,
      "mean_token_accuracy": 0.18039312660694123,
      "num_tokens": 26976182.0,
      "step": 15550
    },
    {
      "entropy": 5.480651760101319,
      "epoch": 1.210231472476172,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004856912187970195,
      "loss": 5.2985,
      "mean_token_accuracy": 0.1713150694966316,
      "num_tokens": 26984275.0,
      "step": 15555
    },
    {
      "entropy": 5.61712794303894,
      "epoch": 1.2106205018478895,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004856814912426502,
      "loss": 5.253,
      "mean_token_accuracy": 0.17858054488897324,
      "num_tokens": 26991891.0,
      "step": 15560
    },
    {
      "entropy": 5.474617338180542,
      "epoch": 1.2110095312196072,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004856717604915245,
      "loss": 5.1886,
      "mean_token_accuracy": 0.1808333471417427,
      "num_tokens": 27000390.0,
      "step": 15565
    },
    {
      "entropy": 5.301046848297119,
      "epoch": 1.2113985605913247,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004856620265437902,
      "loss": 5.1572,
      "mean_token_accuracy": 0.18961051106452942,
      "num_tokens": 27009348.0,
      "step": 15570
    },
    {
      "entropy": 5.474587869644165,
      "epoch": 1.2117875899630421,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000485652289399595,
      "loss": 5.2519,
      "mean_token_accuracy": 0.17714242041110992,
      "num_tokens": 27017529.0,
      "step": 15575
    },
    {
      "entropy": 5.59005765914917,
      "epoch": 1.2121766193347598,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048564254905908655,
      "loss": 5.2465,
      "mean_token_accuracy": 0.180522583425045,
      "num_tokens": 27026238.0,
      "step": 15580
    },
    {
      "entropy": 5.421262407302857,
      "epoch": 1.2125656487064773,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048563280552241266,
      "loss": 5.1722,
      "mean_token_accuracy": 0.183051635324955,
      "num_tokens": 27034995.0,
      "step": 15585
    },
    {
      "entropy": 5.573421669006348,
      "epoch": 1.2129546780781948,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004856230587897212,
      "loss": 5.3534,
      "mean_token_accuracy": 0.17322721034288407,
      "num_tokens": 27043631.0,
      "step": 15590
    },
    {
      "entropy": 5.529970502853393,
      "epoch": 1.2133437074499125,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004856133088611602,
      "loss": 5.2205,
      "mean_token_accuracy": 0.18375702500343322,
      "num_tokens": 27052491.0,
      "step": 15595
    },
    {
      "entropy": 5.48557858467102,
      "epoch": 1.21373273682163,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004856035557368773,
      "loss": 5.2146,
      "mean_token_accuracy": 0.1764917105436325,
      "num_tokens": 27061056.0,
      "step": 15600
    },
    {
      "entropy": 5.410582590103149,
      "epoch": 1.2141217661933477,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048559379941702074,
      "loss": 5.2113,
      "mean_token_accuracy": 0.18179532289505004,
      "num_tokens": 27069562.0,
      "step": 15605
    },
    {
      "entropy": 5.482656955718994,
      "epoch": 1.2145107955650651,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048558403990173844,
      "loss": 5.2115,
      "mean_token_accuracy": 0.18246349394321443,
      "num_tokens": 27077814.0,
      "step": 15610
    },
    {
      "entropy": 5.467282152175903,
      "epoch": 1.2148998249367828,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048557427719117855,
      "loss": 5.1312,
      "mean_token_accuracy": 0.18518226593732834,
      "num_tokens": 27086515.0,
      "step": 15615
    },
    {
      "entropy": 5.455596876144409,
      "epoch": 1.2152888543085003,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004855645112854891,
      "loss": 5.2456,
      "mean_token_accuracy": 0.1810719385743141,
      "num_tokens": 27095759.0,
      "step": 15620
    },
    {
      "entropy": 5.428546905517578,
      "epoch": 1.2156778836802178,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004855547421848184,
      "loss": 5.2675,
      "mean_token_accuracy": 0.17682098150253295,
      "num_tokens": 27104921.0,
      "step": 15625
    },
    {
      "entropy": 5.529678916931152,
      "epoch": 1.2160669130519355,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048554496988931456,
      "loss": 5.2638,
      "mean_token_accuracy": 0.17585633397102357,
      "num_tokens": 27113876.0,
      "step": 15630
    },
    {
      "entropy": 5.5367835521697994,
      "epoch": 1.216455942423653,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048553519439912597,
      "loss": 5.1316,
      "mean_token_accuracy": 0.18312423974275588,
      "num_tokens": 27121654.0,
      "step": 15635
    },
    {
      "entropy": 5.35203742980957,
      "epoch": 1.2168449717953704,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004855254157144009,
      "loss": 5.1731,
      "mean_token_accuracy": 0.18546899408102036,
      "num_tokens": 27130156.0,
      "step": 15640
    },
    {
      "entropy": 5.537712526321411,
      "epoch": 1.2172340011670881,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004855156338352877,
      "loss": 5.267,
      "mean_token_accuracy": 0.1744837388396263,
      "num_tokens": 27138646.0,
      "step": 15645
    },
    {
      "entropy": 5.542705965042114,
      "epoch": 1.2176230305388056,
      "grad_norm": 1.375,
      "learning_rate": 0.00048550584876193493,
      "loss": 5.2536,
      "mean_token_accuracy": 0.1840051755309105,
      "num_tokens": 27147282.0,
      "step": 15650
    },
    {
      "entropy": 5.424997234344483,
      "epoch": 1.2180120599105233,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048549606049449085,
      "loss": 5.1533,
      "mean_token_accuracy": 0.1758889749646187,
      "num_tokens": 27156238.0,
      "step": 15655
    },
    {
      "entropy": 5.462225914001465,
      "epoch": 1.2184010892822408,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004854862690331041,
      "loss": 5.2587,
      "mean_token_accuracy": 0.17398144155740738,
      "num_tokens": 27164936.0,
      "step": 15660
    },
    {
      "entropy": 5.47140531539917,
      "epoch": 1.2187901186539585,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048547647437792327,
      "loss": 5.2243,
      "mean_token_accuracy": 0.18227086067199708,
      "num_tokens": 27173362.0,
      "step": 15665
    },
    {
      "entropy": 5.416584491729736,
      "epoch": 1.219179148025676,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000485466676529097,
      "loss": 5.141,
      "mean_token_accuracy": 0.19056322425603867,
      "num_tokens": 27181611.0,
      "step": 15670
    },
    {
      "entropy": 5.51442666053772,
      "epoch": 1.2195681773973934,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048545687548677373,
      "loss": 5.2312,
      "mean_token_accuracy": 0.17836839854717254,
      "num_tokens": 27189593.0,
      "step": 15675
    },
    {
      "entropy": 5.493145847320557,
      "epoch": 1.2199572067691111,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004854470712511025,
      "loss": 5.1403,
      "mean_token_accuracy": 0.1835565060377121,
      "num_tokens": 27198006.0,
      "step": 15680
    },
    {
      "entropy": 5.459655380249023,
      "epoch": 1.2203462361408286,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048543726382223193,
      "loss": 5.2686,
      "mean_token_accuracy": 0.1758343905210495,
      "num_tokens": 27206846.0,
      "step": 15685
    },
    {
      "entropy": 5.43719253540039,
      "epoch": 1.220735265512546,
      "grad_norm": 1.25,
      "learning_rate": 0.00048542745320031075,
      "loss": 5.2166,
      "mean_token_accuracy": 0.17917377203702928,
      "num_tokens": 27214838.0,
      "step": 15690
    },
    {
      "entropy": 5.485200119018555,
      "epoch": 1.2211242948842638,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048541763938548795,
      "loss": 5.2844,
      "mean_token_accuracy": 0.17426016330718994,
      "num_tokens": 27224161.0,
      "step": 15695
    },
    {
      "entropy": 5.46514663696289,
      "epoch": 1.2215133242559812,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048540782237791244,
      "loss": 5.1556,
      "mean_token_accuracy": 0.17481624558568,
      "num_tokens": 27233084.0,
      "step": 15700
    },
    {
      "entropy": 5.609204959869385,
      "epoch": 1.221902353627699,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048539800217773293,
      "loss": 5.3495,
      "mean_token_accuracy": 0.16798435300588607,
      "num_tokens": 27242219.0,
      "step": 15705
    },
    {
      "entropy": 5.519218444824219,
      "epoch": 1.2222913829994164,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004853881787850988,
      "loss": 5.2913,
      "mean_token_accuracy": 0.17324173599481582,
      "num_tokens": 27250411.0,
      "step": 15710
    },
    {
      "entropy": 5.557796144485474,
      "epoch": 1.2226804123711341,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048537835220015886,
      "loss": 5.3118,
      "mean_token_accuracy": 0.17200906723737716,
      "num_tokens": 27259472.0,
      "step": 15715
    },
    {
      "entropy": 5.493574142456055,
      "epoch": 1.2230694417428516,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004853685224230623,
      "loss": 5.1974,
      "mean_token_accuracy": 0.17979127913713455,
      "num_tokens": 27267922.0,
      "step": 15720
    },
    {
      "entropy": 5.41370038986206,
      "epoch": 1.223458471114569,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048535868945395825,
      "loss": 5.294,
      "mean_token_accuracy": 0.18221947252750398,
      "num_tokens": 27277129.0,
      "step": 15725
    },
    {
      "entropy": 5.536149883270264,
      "epoch": 1.2238475004862868,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048534885329299586,
      "loss": 5.2771,
      "mean_token_accuracy": 0.17098883241415025,
      "num_tokens": 27285583.0,
      "step": 15730
    },
    {
      "entropy": 5.501801681518555,
      "epoch": 1.2242365298580042,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004853390139403245,
      "loss": 5.1451,
      "mean_token_accuracy": 0.1779624566435814,
      "num_tokens": 27294292.0,
      "step": 15735
    },
    {
      "entropy": 5.526371049880981,
      "epoch": 1.224625559229722,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048532917139609334,
      "loss": 5.3439,
      "mean_token_accuracy": 0.17053160518407823,
      "num_tokens": 27302606.0,
      "step": 15740
    },
    {
      "entropy": 5.544359016418457,
      "epoch": 1.2250145886014394,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004853193256604518,
      "loss": 5.2169,
      "mean_token_accuracy": 0.17790654450654983,
      "num_tokens": 27311103.0,
      "step": 15745
    },
    {
      "entropy": 5.552533388137817,
      "epoch": 1.2254036179731569,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048530947673354924,
      "loss": 5.228,
      "mean_token_accuracy": 0.1793415680527687,
      "num_tokens": 27319497.0,
      "step": 15750
    },
    {
      "entropy": 5.427660655975342,
      "epoch": 1.2257926473448746,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004852996246155351,
      "loss": 5.2639,
      "mean_token_accuracy": 0.17681593745946883,
      "num_tokens": 27328367.0,
      "step": 15755
    },
    {
      "entropy": 5.435044527053833,
      "epoch": 1.226181676716592,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048528976930655896,
      "loss": 5.1626,
      "mean_token_accuracy": 0.18119857162237168,
      "num_tokens": 27336797.0,
      "step": 15760
    },
    {
      "entropy": 5.410197496414185,
      "epoch": 1.2265707060883098,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048527991080677015,
      "loss": 5.1326,
      "mean_token_accuracy": 0.18363547176122666,
      "num_tokens": 27345052.0,
      "step": 15765
    },
    {
      "entropy": 5.470131921768188,
      "epoch": 1.2269597354600272,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048527004911631843,
      "loss": 5.1737,
      "mean_token_accuracy": 0.18862563967704774,
      "num_tokens": 27353062.0,
      "step": 15770
    },
    {
      "entropy": 5.450510454177857,
      "epoch": 1.2273487648317447,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004852601842353534,
      "loss": 5.2223,
      "mean_token_accuracy": 0.17984014749526978,
      "num_tokens": 27361471.0,
      "step": 15775
    },
    {
      "entropy": 5.527235794067383,
      "epoch": 1.2277377942034624,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048525031616402476,
      "loss": 5.2886,
      "mean_token_accuracy": 0.17228037416934966,
      "num_tokens": 27370655.0,
      "step": 15780
    },
    {
      "entropy": 5.486161518096924,
      "epoch": 1.2281268235751799,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048524044490248227,
      "loss": 5.1174,
      "mean_token_accuracy": 0.1858793541789055,
      "num_tokens": 27379329.0,
      "step": 15785
    },
    {
      "entropy": 5.465364742279053,
      "epoch": 1.2285158529468976,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048523057045087557,
      "loss": 5.2606,
      "mean_token_accuracy": 0.1761045277118683,
      "num_tokens": 27388864.0,
      "step": 15790
    },
    {
      "entropy": 5.56963095664978,
      "epoch": 1.228904882318615,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048522069280935466,
      "loss": 5.3453,
      "mean_token_accuracy": 0.1645652949810028,
      "num_tokens": 27397804.0,
      "step": 15795
    },
    {
      "entropy": 5.451493453979492,
      "epoch": 1.2292939116903325,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004852108119780693,
      "loss": 5.2186,
      "mean_token_accuracy": 0.17626747637987136,
      "num_tokens": 27406578.0,
      "step": 15800
    },
    {
      "entropy": 5.455910730361938,
      "epoch": 1.2296829410620502,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004852009279571694,
      "loss": 5.3239,
      "mean_token_accuracy": 0.16848056763410568,
      "num_tokens": 27415207.0,
      "step": 15805
    },
    {
      "entropy": 5.518950891494751,
      "epoch": 1.2300719704337677,
      "grad_norm": 1.25,
      "learning_rate": 0.000485191040746805,
      "loss": 5.242,
      "mean_token_accuracy": 0.17220761477947236,
      "num_tokens": 27424201.0,
      "step": 15810
    },
    {
      "entropy": 5.476751279830933,
      "epoch": 1.2304609998054854,
      "grad_norm": 1.25,
      "learning_rate": 0.0004851811503471262,
      "loss": 5.2439,
      "mean_token_accuracy": 0.17757963240146638,
      "num_tokens": 27433187.0,
      "step": 15815
    },
    {
      "entropy": 5.450969171524048,
      "epoch": 1.2308500291772029,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048517125675828294,
      "loss": 5.2227,
      "mean_token_accuracy": 0.1767941102385521,
      "num_tokens": 27441976.0,
      "step": 15820
    },
    {
      "entropy": 5.464339971542358,
      "epoch": 1.2312390585489204,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048516135998042536,
      "loss": 5.1558,
      "mean_token_accuracy": 0.18300871849060057,
      "num_tokens": 27450235.0,
      "step": 15825
    },
    {
      "entropy": 5.510146236419677,
      "epoch": 1.231628087920638,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004851514600137037,
      "loss": 5.3565,
      "mean_token_accuracy": 0.1700697660446167,
      "num_tokens": 27458842.0,
      "step": 15830
    },
    {
      "entropy": 5.57522873878479,
      "epoch": 1.2320171172923555,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048514155685826807,
      "loss": 5.2876,
      "mean_token_accuracy": 0.17434059083461761,
      "num_tokens": 27467383.0,
      "step": 15835
    },
    {
      "entropy": 5.466125106811523,
      "epoch": 1.2324061466640732,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004851316505142688,
      "loss": 5.2635,
      "mean_token_accuracy": 0.17546983510255815,
      "num_tokens": 27475811.0,
      "step": 15840
    },
    {
      "entropy": 5.633650255203247,
      "epoch": 1.2327951760357907,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048512174098185615,
      "loss": 5.3927,
      "mean_token_accuracy": 0.16412212550640107,
      "num_tokens": 27484032.0,
      "step": 15845
    },
    {
      "entropy": 5.490025663375855,
      "epoch": 1.2331842054075082,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048511182826118055,
      "loss": 5.0993,
      "mean_token_accuracy": 0.1899169608950615,
      "num_tokens": 27492330.0,
      "step": 15850
    },
    {
      "entropy": 5.3502439022064205,
      "epoch": 1.2335732347792259,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048510191235239246,
      "loss": 5.1979,
      "mean_token_accuracy": 0.17784707844257355,
      "num_tokens": 27501042.0,
      "step": 15855
    },
    {
      "entropy": 5.409828805923462,
      "epoch": 1.2339622641509433,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048509199325564217,
      "loss": 5.1481,
      "mean_token_accuracy": 0.18505718261003495,
      "num_tokens": 27509939.0,
      "step": 15860
    },
    {
      "entropy": 5.524032020568848,
      "epoch": 1.234351293522661,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004850820709710803,
      "loss": 5.3597,
      "mean_token_accuracy": 0.16386570036411285,
      "num_tokens": 27518653.0,
      "step": 15865
    },
    {
      "entropy": 5.533376264572143,
      "epoch": 1.2347403228943785,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004850721454988574,
      "loss": 5.205,
      "mean_token_accuracy": 0.18562995344400407,
      "num_tokens": 27526856.0,
      "step": 15870
    },
    {
      "entropy": 5.426485395431518,
      "epoch": 1.235129352266096,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048506221683912405,
      "loss": 5.164,
      "mean_token_accuracy": 0.18758039474487304,
      "num_tokens": 27535205.0,
      "step": 15875
    },
    {
      "entropy": 5.532666873931885,
      "epoch": 1.2355183816378137,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004850522849920309,
      "loss": 5.2398,
      "mean_token_accuracy": 0.18003426045179366,
      "num_tokens": 27543516.0,
      "step": 15880
    },
    {
      "entropy": 5.530383110046387,
      "epoch": 1.2359074110095312,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048504234995772863,
      "loss": 5.3029,
      "mean_token_accuracy": 0.17515851706266403,
      "num_tokens": 27552136.0,
      "step": 15885
    },
    {
      "entropy": 5.4363627433776855,
      "epoch": 1.2362964403812489,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004850324117363681,
      "loss": 5.2681,
      "mean_token_accuracy": 0.18483385741710662,
      "num_tokens": 27560805.0,
      "step": 15890
    },
    {
      "entropy": 5.426168918609619,
      "epoch": 1.2366854697529663,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048502247032809997,
      "loss": 5.1835,
      "mean_token_accuracy": 0.17700906991958618,
      "num_tokens": 27569643.0,
      "step": 15895
    },
    {
      "entropy": 5.4428849697113035,
      "epoch": 1.237074499124684,
      "grad_norm": 1.375,
      "learning_rate": 0.0004850125257330751,
      "loss": 5.1344,
      "mean_token_accuracy": 0.17863352298736573,
      "num_tokens": 27578506.0,
      "step": 15900
    },
    {
      "entropy": 5.406799268722534,
      "epoch": 1.2374635284964015,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048500257795144446,
      "loss": 5.2277,
      "mean_token_accuracy": 0.17592196613550187,
      "num_tokens": 27586911.0,
      "step": 15905
    },
    {
      "entropy": 5.506816053390503,
      "epoch": 1.237852557868119,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000484992626983359,
      "loss": 5.2781,
      "mean_token_accuracy": 0.17242311835289,
      "num_tokens": 27595112.0,
      "step": 15910
    },
    {
      "entropy": 5.540608501434326,
      "epoch": 1.2382415872398367,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004849826728289696,
      "loss": 5.2764,
      "mean_token_accuracy": 0.17014632225036622,
      "num_tokens": 27603923.0,
      "step": 15915
    },
    {
      "entropy": 5.3950207233428955,
      "epoch": 1.2386306166115542,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048497271548842737,
      "loss": 5.1009,
      "mean_token_accuracy": 0.18551628142595292,
      "num_tokens": 27612705.0,
      "step": 15920
    },
    {
      "entropy": 5.449764156341553,
      "epoch": 1.2390196459832716,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004849627549618834,
      "loss": 5.2104,
      "mean_token_accuracy": 0.1828647717833519,
      "num_tokens": 27621858.0,
      "step": 15925
    },
    {
      "entropy": 5.506116771697998,
      "epoch": 1.2394086753549893,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048495279124948886,
      "loss": 5.3158,
      "mean_token_accuracy": 0.17837489992380143,
      "num_tokens": 27629977.0,
      "step": 15930
    },
    {
      "entropy": 5.552283716201782,
      "epoch": 1.2397977047267068,
      "grad_norm": 1.375,
      "learning_rate": 0.0004849428243513948,
      "loss": 5.2405,
      "mean_token_accuracy": 0.17784868478775023,
      "num_tokens": 27638433.0,
      "step": 15935
    },
    {
      "entropy": 5.419126653671265,
      "epoch": 1.2401867340984245,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048493285426775255,
      "loss": 5.1654,
      "mean_token_accuracy": 0.18199482411146164,
      "num_tokens": 27647116.0,
      "step": 15940
    },
    {
      "entropy": 5.4253324508667,
      "epoch": 1.240575763470142,
      "grad_norm": 1.375,
      "learning_rate": 0.0004849228809987135,
      "loss": 5.2461,
      "mean_token_accuracy": 0.17561040222644805,
      "num_tokens": 27655598.0,
      "step": 15945
    },
    {
      "entropy": 5.6100245952606205,
      "epoch": 1.2409647928418597,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004849129045444288,
      "loss": 5.3394,
      "mean_token_accuracy": 0.169119693338871,
      "num_tokens": 27664308.0,
      "step": 15950
    },
    {
      "entropy": 5.409825038909912,
      "epoch": 1.2413538222135772,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004849029249050498,
      "loss": 5.0736,
      "mean_token_accuracy": 0.1861928477883339,
      "num_tokens": 27672808.0,
      "step": 15955
    },
    {
      "entropy": 5.462823152542114,
      "epoch": 1.2417428515852946,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048489294208072805,
      "loss": 5.3166,
      "mean_token_accuracy": 0.1683171272277832,
      "num_tokens": 27681091.0,
      "step": 15960
    },
    {
      "entropy": 5.489803123474121,
      "epoch": 1.2421318809570123,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048488295607161495,
      "loss": 5.2352,
      "mean_token_accuracy": 0.17938618808984758,
      "num_tokens": 27689582.0,
      "step": 15965
    },
    {
      "entropy": 5.474964809417725,
      "epoch": 1.2425209103287298,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004848729668778621,
      "loss": 5.1702,
      "mean_token_accuracy": 0.18082460314035415,
      "num_tokens": 27698509.0,
      "step": 15970
    },
    {
      "entropy": 5.450139713287354,
      "epoch": 1.2429099397004473,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004848629744996211,
      "loss": 5.2697,
      "mean_token_accuracy": 0.17326084524393082,
      "num_tokens": 27708216.0,
      "step": 15975
    },
    {
      "entropy": 5.5066461086273195,
      "epoch": 1.243298969072165,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004848529789370434,
      "loss": 5.2815,
      "mean_token_accuracy": 0.17182123064994811,
      "num_tokens": 27716801.0,
      "step": 15980
    },
    {
      "entropy": 5.485807704925537,
      "epoch": 1.2436879984438824,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004848429801902808,
      "loss": 5.2451,
      "mean_token_accuracy": 0.1763685464859009,
      "num_tokens": 27725186.0,
      "step": 15985
    },
    {
      "entropy": 5.57281756401062,
      "epoch": 1.2440770278156001,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048483297825948485,
      "loss": 5.2062,
      "mean_token_accuracy": 0.1761604592204094,
      "num_tokens": 27733386.0,
      "step": 15990
    },
    {
      "entropy": 5.4849255084991455,
      "epoch": 1.2444660571873176,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048482297314480756,
      "loss": 5.2704,
      "mean_token_accuracy": 0.17417690604925157,
      "num_tokens": 27742253.0,
      "step": 15995
    },
    {
      "entropy": 5.454342412948608,
      "epoch": 1.2448550865590353,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004848129648464006,
      "loss": 5.1955,
      "mean_token_accuracy": 0.1838082566857338,
      "num_tokens": 27751555.0,
      "step": 16000
    },
    {
      "entropy": 5.520375156402588,
      "epoch": 1.2452441159307528,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004848029533644159,
      "loss": 5.3402,
      "mean_token_accuracy": 0.1729814037680626,
      "num_tokens": 27760151.0,
      "step": 16005
    },
    {
      "entropy": 5.523137187957763,
      "epoch": 1.2456331453024703,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004847929386990052,
      "loss": 5.2526,
      "mean_token_accuracy": 0.18057575821876526,
      "num_tokens": 27770017.0,
      "step": 16010
    },
    {
      "entropy": 5.43131422996521,
      "epoch": 1.246022174674188,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048478292085032065,
      "loss": 5.105,
      "mean_token_accuracy": 0.1829829692840576,
      "num_tokens": 27777796.0,
      "step": 16015
    },
    {
      "entropy": 5.416547966003418,
      "epoch": 1.2464112040459054,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004847728998185142,
      "loss": 5.2083,
      "mean_token_accuracy": 0.1831973284482956,
      "num_tokens": 27785784.0,
      "step": 16020
    },
    {
      "entropy": 5.508535385131836,
      "epoch": 1.246800233417623,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048476287560373786,
      "loss": 5.1991,
      "mean_token_accuracy": 0.17453235238790513,
      "num_tokens": 27793911.0,
      "step": 16025
    },
    {
      "entropy": 5.441770219802857,
      "epoch": 1.2471892627893406,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048475284820614375,
      "loss": 5.215,
      "mean_token_accuracy": 0.18563712239265442,
      "num_tokens": 27803125.0,
      "step": 16030
    },
    {
      "entropy": 5.3928751945495605,
      "epoch": 1.247578292161058,
      "grad_norm": 1.25,
      "learning_rate": 0.00048474281762588407,
      "loss": 5.1765,
      "mean_token_accuracy": 0.18201385885477067,
      "num_tokens": 27812008.0,
      "step": 16035
    },
    {
      "entropy": 5.460753297805786,
      "epoch": 1.2479673215327758,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004847327838631109,
      "loss": 5.3157,
      "mean_token_accuracy": 0.1737951308488846,
      "num_tokens": 27820490.0,
      "step": 16040
    },
    {
      "entropy": 5.557893943786621,
      "epoch": 1.2483563509044933,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004847227469179766,
      "loss": 5.266,
      "mean_token_accuracy": 0.176366825401783,
      "num_tokens": 27829513.0,
      "step": 16045
    },
    {
      "entropy": 5.472574710845947,
      "epoch": 1.248745380276211,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004847127067906334,
      "loss": 5.1943,
      "mean_token_accuracy": 0.18095572143793107,
      "num_tokens": 27837811.0,
      "step": 16050
    },
    {
      "entropy": 5.548652982711792,
      "epoch": 1.2491344096479284,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048470266348123365,
      "loss": 5.3296,
      "mean_token_accuracy": 0.17246274054050445,
      "num_tokens": 27846449.0,
      "step": 16055
    },
    {
      "entropy": 5.462945652008057,
      "epoch": 1.249523439019646,
      "grad_norm": 1.375,
      "learning_rate": 0.0004846926169899298,
      "loss": 5.1343,
      "mean_token_accuracy": 0.18638257682323456,
      "num_tokens": 27854834.0,
      "step": 16060
    },
    {
      "entropy": 5.536026668548584,
      "epoch": 1.2499124683913636,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048468256731687426,
      "loss": 5.277,
      "mean_token_accuracy": 0.17591823041439056,
      "num_tokens": 27862984.0,
      "step": 16065
    },
    {
      "entropy": 5.518423175811767,
      "epoch": 1.250301497763081,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004846725144622194,
      "loss": 5.2987,
      "mean_token_accuracy": 0.1647412970662117,
      "num_tokens": 27872198.0,
      "step": 16070
    },
    {
      "entropy": 5.51891713142395,
      "epoch": 1.2506905271347986,
      "grad_norm": 1.375,
      "learning_rate": 0.000484662458426118,
      "loss": 5.2008,
      "mean_token_accuracy": 0.17768915593624116,
      "num_tokens": 27880856.0,
      "step": 16075
    },
    {
      "entropy": 5.449474620819092,
      "epoch": 1.2510795565065163,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048465239920872247,
      "loss": 5.1609,
      "mean_token_accuracy": 0.18106906861066818,
      "num_tokens": 27888677.0,
      "step": 16080
    },
    {
      "entropy": 5.409119415283203,
      "epoch": 1.2514685858782337,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048464233681018545,
      "loss": 5.1997,
      "mean_token_accuracy": 0.17691970020532607,
      "num_tokens": 27896940.0,
      "step": 16085
    },
    {
      "entropy": 5.483449029922485,
      "epoch": 1.2518576152499514,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004846322712306596,
      "loss": 5.2919,
      "mean_token_accuracy": 0.17264828085899353,
      "num_tokens": 27905739.0,
      "step": 16090
    },
    {
      "entropy": 5.578553104400635,
      "epoch": 1.252246644621669,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048462220247029783,
      "loss": 5.2507,
      "mean_token_accuracy": 0.1817007526755333,
      "num_tokens": 27914029.0,
      "step": 16095
    },
    {
      "entropy": 5.588305807113647,
      "epoch": 1.2526356739933866,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048461213052925265,
      "loss": 5.3628,
      "mean_token_accuracy": 0.17041990160942078,
      "num_tokens": 27923287.0,
      "step": 16100
    },
    {
      "entropy": 5.461779594421387,
      "epoch": 1.253024703365104,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004846020554076771,
      "loss": 5.213,
      "mean_token_accuracy": 0.18012726455926895,
      "num_tokens": 27931727.0,
      "step": 16105
    },
    {
      "entropy": 5.452313566207886,
      "epoch": 1.2534137327368216,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004845919771057239,
      "loss": 5.1394,
      "mean_token_accuracy": 0.1843203142285347,
      "num_tokens": 27939524.0,
      "step": 16110
    },
    {
      "entropy": 5.48021125793457,
      "epoch": 1.2538027621085392,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004845818956235462,
      "loss": 5.2248,
      "mean_token_accuracy": 0.17517240047454835,
      "num_tokens": 27947827.0,
      "step": 16115
    },
    {
      "entropy": 5.413819313049316,
      "epoch": 1.2541917914802567,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048457181096129664,
      "loss": 5.229,
      "mean_token_accuracy": 0.17882770746946336,
      "num_tokens": 27956447.0,
      "step": 16120
    },
    {
      "entropy": 5.513627433776856,
      "epoch": 1.2545808208519742,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004845617231191285,
      "loss": 5.2769,
      "mean_token_accuracy": 0.17298759073019027,
      "num_tokens": 27965570.0,
      "step": 16125
    },
    {
      "entropy": 5.48878698348999,
      "epoch": 1.254969850223692,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004845516320971948,
      "loss": 5.2295,
      "mean_token_accuracy": 0.17800179421901702,
      "num_tokens": 27974527.0,
      "step": 16130
    },
    {
      "entropy": 5.461105012893677,
      "epoch": 1.2553588795954094,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004845415378956486,
      "loss": 5.17,
      "mean_token_accuracy": 0.18018230944871902,
      "num_tokens": 27983190.0,
      "step": 16135
    },
    {
      "entropy": 5.480553817749024,
      "epoch": 1.255747908967127,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000484531440514643,
      "loss": 5.2707,
      "mean_token_accuracy": 0.17969854027032853,
      "num_tokens": 27992203.0,
      "step": 16140
    },
    {
      "entropy": 5.550424098968506,
      "epoch": 1.2561369383388445,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048452133995433136,
      "loss": 5.2414,
      "mean_token_accuracy": 0.17608337104320526,
      "num_tokens": 28001154.0,
      "step": 16145
    },
    {
      "entropy": 5.414363050460816,
      "epoch": 1.2565259677105622,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004845112362148668,
      "loss": 5.146,
      "mean_token_accuracy": 0.17836720794439315,
      "num_tokens": 28009841.0,
      "step": 16150
    },
    {
      "entropy": 5.409993457794189,
      "epoch": 1.2569149970822797,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004845011292964028,
      "loss": 5.2427,
      "mean_token_accuracy": 0.1806267410516739,
      "num_tokens": 28019245.0,
      "step": 16155
    },
    {
      "entropy": 5.548595142364502,
      "epoch": 1.2573040264539972,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048449101919909265,
      "loss": 5.3648,
      "mean_token_accuracy": 0.16800736784934997,
      "num_tokens": 28028342.0,
      "step": 16160
    },
    {
      "entropy": 5.560082960128784,
      "epoch": 1.257693055825715,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048448090592308955,
      "loss": 5.2954,
      "mean_token_accuracy": 0.16870589107275008,
      "num_tokens": 28037308.0,
      "step": 16165
    },
    {
      "entropy": 5.524203157424926,
      "epoch": 1.2580820851974324,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004844707894685473,
      "loss": 5.2785,
      "mean_token_accuracy": 0.17870236486196517,
      "num_tokens": 28046214.0,
      "step": 16170
    },
    {
      "entropy": 5.451855230331421,
      "epoch": 1.2584711145691498,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004844606698356192,
      "loss": 5.0894,
      "mean_token_accuracy": 0.19016786217689513,
      "num_tokens": 28054722.0,
      "step": 16175
    },
    {
      "entropy": 5.432681941986084,
      "epoch": 1.2588601439408675,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004844505470244588,
      "loss": 5.1061,
      "mean_token_accuracy": 0.17895887792110443,
      "num_tokens": 28062596.0,
      "step": 16180
    },
    {
      "entropy": 5.380811500549316,
      "epoch": 1.2592491733125852,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004844404210352197,
      "loss": 5.1584,
      "mean_token_accuracy": 0.18074051290750504,
      "num_tokens": 28071183.0,
      "step": 16185
    },
    {
      "entropy": 5.334208726882935,
      "epoch": 1.2596382026843027,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048443029186805554,
      "loss": 5.1554,
      "mean_token_accuracy": 0.18262638449668883,
      "num_tokens": 28079194.0,
      "step": 16190
    },
    {
      "entropy": 5.403448820114136,
      "epoch": 1.2600272320560202,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004844201595231201,
      "loss": 5.1686,
      "mean_token_accuracy": 0.18047152310609818,
      "num_tokens": 28087874.0,
      "step": 16195
    },
    {
      "entropy": 5.560309839248657,
      "epoch": 1.2604162614277379,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048441002400056706,
      "loss": 5.2458,
      "mean_token_accuracy": 0.1779054269194603,
      "num_tokens": 28097390.0,
      "step": 16200
    },
    {
      "entropy": 5.408809232711792,
      "epoch": 1.2608052907994554,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004843998853005502,
      "loss": 5.1573,
      "mean_token_accuracy": 0.18068110942840576,
      "num_tokens": 28105445.0,
      "step": 16205
    },
    {
      "entropy": 5.534795904159546,
      "epoch": 1.2611943201711728,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004843897434232233,
      "loss": 5.1957,
      "mean_token_accuracy": 0.1836117222905159,
      "num_tokens": 28113855.0,
      "step": 16210
    },
    {
      "entropy": 5.472966194152832,
      "epoch": 1.2615833495428905,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048437959836874047,
      "loss": 5.1488,
      "mean_token_accuracy": 0.18562332540750504,
      "num_tokens": 28123300.0,
      "step": 16215
    },
    {
      "entropy": 5.387235546112061,
      "epoch": 1.261972378914608,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048436945013725544,
      "loss": 5.1465,
      "mean_token_accuracy": 0.18177770227193832,
      "num_tokens": 28131829.0,
      "step": 16220
    },
    {
      "entropy": 5.481217002868652,
      "epoch": 1.2623614082863255,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048435929872892226,
      "loss": 5.3119,
      "mean_token_accuracy": 0.1722740799188614,
      "num_tokens": 28141060.0,
      "step": 16225
    },
    {
      "entropy": 5.527338552474975,
      "epoch": 1.2627504376580432,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004843491441438949,
      "loss": 5.3062,
      "mean_token_accuracy": 0.16698299944400788,
      "num_tokens": 28149866.0,
      "step": 16230
    },
    {
      "entropy": 5.447233581542969,
      "epoch": 1.2631394670297609,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048433898638232755,
      "loss": 5.236,
      "mean_token_accuracy": 0.18000967055559158,
      "num_tokens": 28158593.0,
      "step": 16235
    },
    {
      "entropy": 5.494152688980103,
      "epoch": 1.2635284964014784,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004843288254443742,
      "loss": 5.1695,
      "mean_token_accuracy": 0.18486100137233735,
      "num_tokens": 28167301.0,
      "step": 16240
    },
    {
      "entropy": 5.478094530105591,
      "epoch": 1.2639175257731958,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004843186613301892,
      "loss": 5.2284,
      "mean_token_accuracy": 0.18294586688280107,
      "num_tokens": 28176092.0,
      "step": 16245
    },
    {
      "entropy": 5.504894876480103,
      "epoch": 1.2643065551449135,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004843084940399266,
      "loss": 5.2907,
      "mean_token_accuracy": 0.18051428347826004,
      "num_tokens": 28184570.0,
      "step": 16250
    },
    {
      "entropy": 5.462686967849732,
      "epoch": 1.264695584516631,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004842983235737408,
      "loss": 5.2239,
      "mean_token_accuracy": 0.18603262752294542,
      "num_tokens": 28192874.0,
      "step": 16255
    },
    {
      "entropy": 5.462531423568725,
      "epoch": 1.2650846138883485,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004842881499317861,
      "loss": 5.2356,
      "mean_token_accuracy": 0.18035957515239714,
      "num_tokens": 28201546.0,
      "step": 16260
    },
    {
      "entropy": 5.416395473480224,
      "epoch": 1.2654736432600662,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004842779731142168,
      "loss": 5.2004,
      "mean_token_accuracy": 0.17683276534080505,
      "num_tokens": 28210061.0,
      "step": 16265
    },
    {
      "entropy": 5.3979864597320555,
      "epoch": 1.2658626726317836,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048426779312118735,
      "loss": 5.1262,
      "mean_token_accuracy": 0.1790337786078453,
      "num_tokens": 28219334.0,
      "step": 16270
    },
    {
      "entropy": 5.542618656158448,
      "epoch": 1.2662517020035013,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004842576099528522,
      "loss": 5.2605,
      "mean_token_accuracy": 0.17416693568229674,
      "num_tokens": 28227963.0,
      "step": 16275
    },
    {
      "entropy": 5.464198732376099,
      "epoch": 1.2666407313752188,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004842474236093659,
      "loss": 5.2236,
      "mean_token_accuracy": 0.1782044619321823,
      "num_tokens": 28236648.0,
      "step": 16280
    },
    {
      "entropy": 5.449898052215576,
      "epoch": 1.2670297607469365,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048423723409088306,
      "loss": 5.23,
      "mean_token_accuracy": 0.1760978117585182,
      "num_tokens": 28245221.0,
      "step": 16285
    },
    {
      "entropy": 5.450131464004516,
      "epoch": 1.267418790118654,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004842270413975582,
      "loss": 5.2041,
      "mean_token_accuracy": 0.1761073037981987,
      "num_tokens": 28254481.0,
      "step": 16290
    },
    {
      "entropy": 5.503116273880005,
      "epoch": 1.2678078194903715,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000484216845529546,
      "loss": 5.1793,
      "mean_token_accuracy": 0.1791584834456444,
      "num_tokens": 28263844.0,
      "step": 16295
    },
    {
      "entropy": 5.426848411560059,
      "epoch": 1.2681968488620892,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048420664648700113,
      "loss": 5.1609,
      "mean_token_accuracy": 0.17788806408643723,
      "num_tokens": 28272500.0,
      "step": 16300
    },
    {
      "entropy": 5.487860536575317,
      "epoch": 1.2685858782338066,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004841964442700784,
      "loss": 5.2427,
      "mean_token_accuracy": 0.1698075130581856,
      "num_tokens": 28281553.0,
      "step": 16305
    },
    {
      "entropy": 5.4953038692474365,
      "epoch": 1.2689749076055241,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048418623887893264,
      "loss": 5.2088,
      "mean_token_accuracy": 0.18202257305383682,
      "num_tokens": 28290078.0,
      "step": 16310
    },
    {
      "entropy": 5.426489639282226,
      "epoch": 1.2693639369772418,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048417603031371867,
      "loss": 5.0917,
      "mean_token_accuracy": 0.1774670347571373,
      "num_tokens": 28298879.0,
      "step": 16315
    },
    {
      "entropy": 5.3851676940917965,
      "epoch": 1.2697529663489593,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004841658185745913,
      "loss": 5.1976,
      "mean_token_accuracy": 0.18547892570495605,
      "num_tokens": 28307188.0,
      "step": 16320
    },
    {
      "entropy": 5.423924493789673,
      "epoch": 1.270141995720677,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048415560366170564,
      "loss": 5.1196,
      "mean_token_accuracy": 0.18174345493316652,
      "num_tokens": 28316012.0,
      "step": 16325
    },
    {
      "entropy": 5.491856098175049,
      "epoch": 1.2705310250923945,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004841453855752165,
      "loss": 5.3047,
      "mean_token_accuracy": 0.1725137159228325,
      "num_tokens": 28325750.0,
      "step": 16330
    },
    {
      "entropy": 5.45008282661438,
      "epoch": 1.2709200544641122,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004841351643152791,
      "loss": 5.2445,
      "mean_token_accuracy": 0.17500410079956055,
      "num_tokens": 28335217.0,
      "step": 16335
    },
    {
      "entropy": 5.514533948898316,
      "epoch": 1.2713090838358296,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004841249398820485,
      "loss": 5.2113,
      "mean_token_accuracy": 0.17611064016819,
      "num_tokens": 28344123.0,
      "step": 16340
    },
    {
      "entropy": 5.507355165481568,
      "epoch": 1.271698113207547,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004841147122756797,
      "loss": 5.2202,
      "mean_token_accuracy": 0.1785136118531227,
      "num_tokens": 28353230.0,
      "step": 16345
    },
    {
      "entropy": 5.464165210723877,
      "epoch": 1.2720871425792648,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000484104481496328,
      "loss": 5.2292,
      "mean_token_accuracy": 0.17608998715877533,
      "num_tokens": 28361636.0,
      "step": 16350
    },
    {
      "entropy": 5.455095863342285,
      "epoch": 1.2724761719509823,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004840942475441486,
      "loss": 5.1787,
      "mean_token_accuracy": 0.1751441016793251,
      "num_tokens": 28370066.0,
      "step": 16355
    },
    {
      "entropy": 5.519035196304321,
      "epoch": 1.2728652013226998,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004840840104192969,
      "loss": 5.2815,
      "mean_token_accuracy": 0.17137643396854402,
      "num_tokens": 28379017.0,
      "step": 16360
    },
    {
      "entropy": 5.5544782161712645,
      "epoch": 1.2732542306944175,
      "grad_norm": 1.265625,
      "learning_rate": 0.000484073770121928,
      "loss": 5.2329,
      "mean_token_accuracy": 0.1790240228176117,
      "num_tokens": 28387939.0,
      "step": 16365
    },
    {
      "entropy": 5.419571113586426,
      "epoch": 1.273643260066135,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004840635266521975,
      "loss": 5.1404,
      "mean_token_accuracy": 0.1869226261973381,
      "num_tokens": 28396341.0,
      "step": 16370
    },
    {
      "entropy": 5.388726758956909,
      "epoch": 1.2740322894378526,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004840532800102607,
      "loss": 5.1572,
      "mean_token_accuracy": 0.18105708956718444,
      "num_tokens": 28405128.0,
      "step": 16375
    },
    {
      "entropy": 5.4394447803497314,
      "epoch": 1.27442131880957,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048404303019627314,
      "loss": 5.141,
      "mean_token_accuracy": 0.18251504451036454,
      "num_tokens": 28412841.0,
      "step": 16380
    },
    {
      "entropy": 5.40770959854126,
      "epoch": 1.2748103481812878,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048403277721039036,
      "loss": 5.0922,
      "mean_token_accuracy": 0.18830534368753432,
      "num_tokens": 28421526.0,
      "step": 16385
    },
    {
      "entropy": 5.479176187515259,
      "epoch": 1.2751993775530053,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004840225210527679,
      "loss": 5.2633,
      "mean_token_accuracy": 0.17417340278625487,
      "num_tokens": 28430070.0,
      "step": 16390
    },
    {
      "entropy": 5.429286861419678,
      "epoch": 1.2755884069247228,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004840122617235613,
      "loss": 5.1459,
      "mean_token_accuracy": 0.17906307131052018,
      "num_tokens": 28438920.0,
      "step": 16395
    },
    {
      "entropy": 5.470898294448853,
      "epoch": 1.2759774362964404,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004840019992229263,
      "loss": 5.182,
      "mean_token_accuracy": 0.1817480131983757,
      "num_tokens": 28447666.0,
      "step": 16400
    },
    {
      "entropy": 5.507677459716797,
      "epoch": 1.276366465668158,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048399173355101867,
      "loss": 5.2915,
      "mean_token_accuracy": 0.1724479839205742,
      "num_tokens": 28457341.0,
      "step": 16405
    },
    {
      "entropy": 5.451514148712159,
      "epoch": 1.2767554950398754,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048398146470799417,
      "loss": 5.1946,
      "mean_token_accuracy": 0.1861010044813156,
      "num_tokens": 28466202.0,
      "step": 16410
    },
    {
      "entropy": 5.547113227844238,
      "epoch": 1.277144524411593,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048397119269400846,
      "loss": 5.2847,
      "mean_token_accuracy": 0.1746869444847107,
      "num_tokens": 28474494.0,
      "step": 16415
    },
    {
      "entropy": 5.49758129119873,
      "epoch": 1.2775335537833106,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048396091750921766,
      "loss": 5.2433,
      "mean_token_accuracy": 0.17987698167562485,
      "num_tokens": 28482931.0,
      "step": 16420
    },
    {
      "entropy": 5.440438747406006,
      "epoch": 1.2779225831550283,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004839506391537774,
      "loss": 5.1979,
      "mean_token_accuracy": 0.17847239673137666,
      "num_tokens": 28491372.0,
      "step": 16425
    },
    {
      "entropy": 5.384391355514526,
      "epoch": 1.2783116125267457,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004839403576278438,
      "loss": 5.1149,
      "mean_token_accuracy": 0.1803361564874649,
      "num_tokens": 28500201.0,
      "step": 16430
    },
    {
      "entropy": 5.425058078765869,
      "epoch": 1.2787006418984634,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004839300729315729,
      "loss": 5.1469,
      "mean_token_accuracy": 0.18302377462387084,
      "num_tokens": 28508899.0,
      "step": 16435
    },
    {
      "entropy": 5.482381105422974,
      "epoch": 1.279089671270181,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004839197850651206,
      "loss": 5.2266,
      "mean_token_accuracy": 0.17388372272253036,
      "num_tokens": 28518280.0,
      "step": 16440
    },
    {
      "entropy": 5.532951354980469,
      "epoch": 1.2794787006418984,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048390949402864317,
      "loss": 5.279,
      "mean_token_accuracy": 0.17831270545721054,
      "num_tokens": 28527113.0,
      "step": 16445
    },
    {
      "entropy": 5.436628961563111,
      "epoch": 1.279867730013616,
      "grad_norm": 1.25,
      "learning_rate": 0.00048389919982229666,
      "loss": 5.1834,
      "mean_token_accuracy": 0.18332831114530562,
      "num_tokens": 28535253.0,
      "step": 16450
    },
    {
      "entropy": 5.440378189086914,
      "epoch": 1.2802567593853336,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048388890244623723,
      "loss": 5.2262,
      "mean_token_accuracy": 0.17614058256149293,
      "num_tokens": 28543475.0,
      "step": 16455
    },
    {
      "entropy": 5.509215927124023,
      "epoch": 1.280645788757051,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004838786019006213,
      "loss": 5.2717,
      "mean_token_accuracy": 0.17863212525844574,
      "num_tokens": 28551921.0,
      "step": 16460
    },
    {
      "entropy": 5.4979002475738525,
      "epoch": 1.2810348181287687,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048386829818560493,
      "loss": 5.2447,
      "mean_token_accuracy": 0.17675622403621674,
      "num_tokens": 28560413.0,
      "step": 16465
    },
    {
      "entropy": 5.535602235794068,
      "epoch": 1.2814238475004862,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048385799130134464,
      "loss": 5.2744,
      "mean_token_accuracy": 0.17638418227434158,
      "num_tokens": 28568802.0,
      "step": 16470
    },
    {
      "entropy": 5.466368293762207,
      "epoch": 1.281812876872204,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004838476812479968,
      "loss": 5.3576,
      "mean_token_accuracy": 0.1711149886250496,
      "num_tokens": 28578164.0,
      "step": 16475
    },
    {
      "entropy": 5.548346662521363,
      "epoch": 1.2822019062439214,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048383736802571774,
      "loss": 5.146,
      "mean_token_accuracy": 0.19468737691640853,
      "num_tokens": 28586061.0,
      "step": 16480
    },
    {
      "entropy": 5.413267040252686,
      "epoch": 1.282590935615639,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000483827051634664,
      "loss": 5.1489,
      "mean_token_accuracy": 0.1780667245388031,
      "num_tokens": 28594185.0,
      "step": 16485
    },
    {
      "entropy": 5.403642034530639,
      "epoch": 1.2829799649873566,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048381673207499224,
      "loss": 5.1967,
      "mean_token_accuracy": 0.18521921336650848,
      "num_tokens": 28602427.0,
      "step": 16490
    },
    {
      "entropy": 5.4675883769989015,
      "epoch": 1.283368994359074,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004838064093468588,
      "loss": 5.1533,
      "mean_token_accuracy": 0.18010566830635072,
      "num_tokens": 28611608.0,
      "step": 16495
    },
    {
      "entropy": 5.489548921585083,
      "epoch": 1.2837580237307917,
      "grad_norm": 1.25,
      "learning_rate": 0.0004837960834504206,
      "loss": 5.1519,
      "mean_token_accuracy": 0.18218579441308974,
      "num_tokens": 28620142.0,
      "step": 16500
    },
    {
      "entropy": 5.446900796890259,
      "epoch": 1.2841470531025092,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004837857543858341,
      "loss": 5.2454,
      "mean_token_accuracy": 0.1796965166926384,
      "num_tokens": 28629401.0,
      "step": 16505
    },
    {
      "entropy": 5.46379714012146,
      "epoch": 1.2845360824742267,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004837754221532561,
      "loss": 5.1994,
      "mean_token_accuracy": 0.17722307741642,
      "num_tokens": 28637934.0,
      "step": 16510
    },
    {
      "entropy": 5.466189908981323,
      "epoch": 1.2849251118459444,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048376508675284334,
      "loss": 5.2134,
      "mean_token_accuracy": 0.17799234688282012,
      "num_tokens": 28647145.0,
      "step": 16515
    },
    {
      "entropy": 5.474808645248413,
      "epoch": 1.2853141412176619,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048375474818475274,
      "loss": 5.312,
      "mean_token_accuracy": 0.17233075201511383,
      "num_tokens": 28656146.0,
      "step": 16520
    },
    {
      "entropy": 5.475784730911255,
      "epoch": 1.2857031705893796,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048374440644914104,
      "loss": 5.1205,
      "mean_token_accuracy": 0.18288243561983109,
      "num_tokens": 28664476.0,
      "step": 16525
    },
    {
      "entropy": 5.470512390136719,
      "epoch": 1.286092199961097,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048373406154616523,
      "loss": 5.1353,
      "mean_token_accuracy": 0.18811102211475372,
      "num_tokens": 28672188.0,
      "step": 16530
    },
    {
      "entropy": 5.368266677856445,
      "epoch": 1.2864812293328147,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048372371347598226,
      "loss": 5.2316,
      "mean_token_accuracy": 0.18081342577934265,
      "num_tokens": 28680758.0,
      "step": 16535
    },
    {
      "entropy": 5.5965025424957275,
      "epoch": 1.2868702587045322,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048371336223874923,
      "loss": 5.367,
      "mean_token_accuracy": 0.1681418873369694,
      "num_tokens": 28690392.0,
      "step": 16540
    },
    {
      "entropy": 5.5204548835754395,
      "epoch": 1.2872592880762497,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000483703007834623,
      "loss": 5.215,
      "mean_token_accuracy": 0.17682894170284272,
      "num_tokens": 28699121.0,
      "step": 16545
    },
    {
      "entropy": 5.518025875091553,
      "epoch": 1.2876483174479674,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004836926502637609,
      "loss": 5.3122,
      "mean_token_accuracy": 0.16872557252645493,
      "num_tokens": 28707692.0,
      "step": 16550
    },
    {
      "entropy": 5.442875862121582,
      "epoch": 1.2880373468196848,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048368228952632005,
      "loss": 5.2366,
      "mean_token_accuracy": 0.1787099227309227,
      "num_tokens": 28716630.0,
      "step": 16555
    },
    {
      "entropy": 5.438476514816284,
      "epoch": 1.2884263761914023,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048367192562245756,
      "loss": 5.1306,
      "mean_token_accuracy": 0.18950250297784804,
      "num_tokens": 28725503.0,
      "step": 16560
    },
    {
      "entropy": 5.4652636528015135,
      "epoch": 1.28881540556312,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048366155855233067,
      "loss": 5.2349,
      "mean_token_accuracy": 0.17741916626691817,
      "num_tokens": 28734578.0,
      "step": 16565
    },
    {
      "entropy": 5.508862638473511,
      "epoch": 1.2892044349348377,
      "grad_norm": 1.25,
      "learning_rate": 0.0004836511883160968,
      "loss": 5.2908,
      "mean_token_accuracy": 0.1766727715730667,
      "num_tokens": 28743380.0,
      "step": 16570
    },
    {
      "entropy": 5.497365188598633,
      "epoch": 1.2895934643065552,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004836408149139133,
      "loss": 5.2512,
      "mean_token_accuracy": 0.17797475010156633,
      "num_tokens": 28752968.0,
      "step": 16575
    },
    {
      "entropy": 5.4591264724731445,
      "epoch": 1.2899824936782727,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048363043834593743,
      "loss": 5.2454,
      "mean_token_accuracy": 0.1733506664633751,
      "num_tokens": 28761756.0,
      "step": 16580
    },
    {
      "entropy": 5.501807689666748,
      "epoch": 1.2903715230499904,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004836200586123268,
      "loss": 5.2065,
      "mean_token_accuracy": 0.18051162660121917,
      "num_tokens": 28771110.0,
      "step": 16585
    },
    {
      "entropy": 5.534831094741821,
      "epoch": 1.2907605524217078,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048360967571323875,
      "loss": 5.2915,
      "mean_token_accuracy": 0.17516311705112458,
      "num_tokens": 28780016.0,
      "step": 16590
    },
    {
      "entropy": 5.434015274047852,
      "epoch": 1.2911495817934253,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048359928964883094,
      "loss": 5.1833,
      "mean_token_accuracy": 0.18689361810684205,
      "num_tokens": 28788150.0,
      "step": 16595
    },
    {
      "entropy": 5.461406278610229,
      "epoch": 1.291538611165143,
      "grad_norm": 1.296875,
      "learning_rate": 0.000483588900419261,
      "loss": 5.1711,
      "mean_token_accuracy": 0.1793718382716179,
      "num_tokens": 28796564.0,
      "step": 16600
    },
    {
      "entropy": 5.624229383468628,
      "epoch": 1.2919276405368605,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004835785080246864,
      "loss": 5.4027,
      "mean_token_accuracy": 0.16326048523187636,
      "num_tokens": 28804987.0,
      "step": 16605
    },
    {
      "entropy": 5.542754745483398,
      "epoch": 1.292316669908578,
      "grad_norm": 1.359375,
      "learning_rate": 0.000483568112465265,
      "loss": 5.1936,
      "mean_token_accuracy": 0.18009138554334642,
      "num_tokens": 28813752.0,
      "step": 16610
    },
    {
      "entropy": 5.426624488830567,
      "epoch": 1.2927056992802957,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048355771374115436,
      "loss": 5.1101,
      "mean_token_accuracy": 0.18327810615301132,
      "num_tokens": 28822464.0,
      "step": 16615
    },
    {
      "entropy": 5.423651218414307,
      "epoch": 1.2930947286520134,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004835473118525125,
      "loss": 5.1506,
      "mean_token_accuracy": 0.1797259271144867,
      "num_tokens": 28830997.0,
      "step": 16620
    },
    {
      "entropy": 5.399851274490357,
      "epoch": 1.2934837580237308,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004835369067994971,
      "loss": 5.1424,
      "mean_token_accuracy": 0.18947167545557023,
      "num_tokens": 28839979.0,
      "step": 16625
    },
    {
      "entropy": 5.465669822692871,
      "epoch": 1.2938727873954483,
      "grad_norm": 1.25,
      "learning_rate": 0.000483526498582266,
      "loss": 5.1805,
      "mean_token_accuracy": 0.17419132590293884,
      "num_tokens": 28848394.0,
      "step": 16630
    },
    {
      "entropy": 5.60013222694397,
      "epoch": 1.294261816767166,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048351608720097715,
      "loss": 5.483,
      "mean_token_accuracy": 0.16452204138040544,
      "num_tokens": 28857804.0,
      "step": 16635
    },
    {
      "entropy": 5.493505907058716,
      "epoch": 1.2946508461388835,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048350567265578867,
      "loss": 5.1259,
      "mean_token_accuracy": 0.17957513481378556,
      "num_tokens": 28866405.0,
      "step": 16640
    },
    {
      "entropy": 5.541100263595581,
      "epoch": 1.295039875510601,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004834952549468584,
      "loss": 5.2247,
      "mean_token_accuracy": 0.18246638625860215,
      "num_tokens": 28874751.0,
      "step": 16645
    },
    {
      "entropy": 5.505598783493042,
      "epoch": 1.2954289048823187,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004834848340743446,
      "loss": 5.3452,
      "mean_token_accuracy": 0.16946036443114282,
      "num_tokens": 28884013.0,
      "step": 16650
    },
    {
      "entropy": 5.428146743774414,
      "epoch": 1.2958179342540361,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004834744100384052,
      "loss": 5.1461,
      "mean_token_accuracy": 0.17946367263793944,
      "num_tokens": 28892175.0,
      "step": 16655
    },
    {
      "entropy": 5.544998073577881,
      "epoch": 1.2962069636257538,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004834639828391984,
      "loss": 5.208,
      "mean_token_accuracy": 0.17580684125423432,
      "num_tokens": 28901238.0,
      "step": 16660
    },
    {
      "entropy": 5.505697107315063,
      "epoch": 1.2965959929974713,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048345355247688256,
      "loss": 5.2234,
      "mean_token_accuracy": 0.17321915030479432,
      "num_tokens": 28910368.0,
      "step": 16665
    },
    {
      "entropy": 5.4913724899292,
      "epoch": 1.296985022369189,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004834431189516158,
      "loss": 5.3076,
      "mean_token_accuracy": 0.17062025964260102,
      "num_tokens": 28918854.0,
      "step": 16670
    },
    {
      "entropy": 5.498944568634033,
      "epoch": 1.2973740517409065,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004834326822635565,
      "loss": 5.2301,
      "mean_token_accuracy": 0.18061346113681792,
      "num_tokens": 28927103.0,
      "step": 16675
    },
    {
      "entropy": 5.4339886665344235,
      "epoch": 1.297763081112624,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048342224241286296,
      "loss": 5.1679,
      "mean_token_accuracy": 0.17780641466379166,
      "num_tokens": 28936257.0,
      "step": 16680
    },
    {
      "entropy": 5.582760810852051,
      "epoch": 1.2981521104843416,
      "grad_norm": 1.25,
      "learning_rate": 0.0004834117993996937,
      "loss": 5.2878,
      "mean_token_accuracy": 0.1745957612991333,
      "num_tokens": 28944991.0,
      "step": 16685
    },
    {
      "entropy": 5.497891664505005,
      "epoch": 1.2985411398560591,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004834013532242071,
      "loss": 5.2582,
      "mean_token_accuracy": 0.17548902928829194,
      "num_tokens": 28953781.0,
      "step": 16690
    },
    {
      "entropy": 5.470164442062378,
      "epoch": 1.2989301692277766,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004833909038865616,
      "loss": 5.1471,
      "mean_token_accuracy": 0.1868959367275238,
      "num_tokens": 28961916.0,
      "step": 16695
    },
    {
      "entropy": 5.47140154838562,
      "epoch": 1.2993191985994943,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004833804513869159,
      "loss": 5.1881,
      "mean_token_accuracy": 0.1751638501882553,
      "num_tokens": 28970223.0,
      "step": 16700
    },
    {
      "entropy": 5.456587314605713,
      "epoch": 1.2997082279712118,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004833699957254284,
      "loss": 5.2251,
      "mean_token_accuracy": 0.17924579083919526,
      "num_tokens": 28979063.0,
      "step": 16705
    },
    {
      "entropy": 5.551636171340943,
      "epoch": 1.3000972573429295,
      "grad_norm": 1.328125,
      "learning_rate": 0.000483359536902258,
      "loss": 5.342,
      "mean_token_accuracy": 0.1761761412024498,
      "num_tokens": 28987229.0,
      "step": 16710
    },
    {
      "entropy": 5.542795705795288,
      "epoch": 1.300486286714647,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004833490749175632,
      "loss": 5.2952,
      "mean_token_accuracy": 0.17223856896162032,
      "num_tokens": 28995889.0,
      "step": 16715
    },
    {
      "entropy": 5.51875114440918,
      "epoch": 1.3008753160863646,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048333860977150286,
      "loss": 5.2838,
      "mean_token_accuracy": 0.1746693581342697,
      "num_tokens": 29003962.0,
      "step": 16720
    },
    {
      "entropy": 5.438073396682739,
      "epoch": 1.3012643454580821,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048332814146423566,
      "loss": 5.1571,
      "mean_token_accuracy": 0.17704245448112488,
      "num_tokens": 29012652.0,
      "step": 16725
    },
    {
      "entropy": 5.576805925369262,
      "epoch": 1.3016533748297996,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004833176699959206,
      "loss": 5.3131,
      "mean_token_accuracy": 0.16424298286437988,
      "num_tokens": 29022445.0,
      "step": 16730
    },
    {
      "entropy": 5.4788501262664795,
      "epoch": 1.3020424042015173,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048330719536671633,
      "loss": 5.1979,
      "mean_token_accuracy": 0.17414999902248382,
      "num_tokens": 29030840.0,
      "step": 16735
    },
    {
      "entropy": 5.545869445800781,
      "epoch": 1.3024314335732348,
      "grad_norm": 1.40625,
      "learning_rate": 0.000483296717576782,
      "loss": 5.2592,
      "mean_token_accuracy": 0.1776003435254097,
      "num_tokens": 29039640.0,
      "step": 16740
    },
    {
      "entropy": 5.46205997467041,
      "epoch": 1.3028204629449522,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004832862366262765,
      "loss": 5.2005,
      "mean_token_accuracy": 0.18249295949935912,
      "num_tokens": 29047793.0,
      "step": 16745
    },
    {
      "entropy": 5.433813285827637,
      "epoch": 1.30320949231667,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048327575251535886,
      "loss": 5.1769,
      "mean_token_accuracy": 0.18062884509563445,
      "num_tokens": 29056880.0,
      "step": 16750
    },
    {
      "entropy": 5.46841516494751,
      "epoch": 1.3035985216883874,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004832652652441883,
      "loss": 5.2072,
      "mean_token_accuracy": 0.17528669238090516,
      "num_tokens": 29065093.0,
      "step": 16755
    },
    {
      "entropy": 5.428246641159058,
      "epoch": 1.303987551060105,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048325477481292375,
      "loss": 5.2776,
      "mean_token_accuracy": 0.1744294837117195,
      "num_tokens": 29074389.0,
      "step": 16760
    },
    {
      "entropy": 5.36859655380249,
      "epoch": 1.3043765804318226,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004832442812217244,
      "loss": 5.1093,
      "mean_token_accuracy": 0.1844582512974739,
      "num_tokens": 29082789.0,
      "step": 16765
    },
    {
      "entropy": 5.543893766403198,
      "epoch": 1.3047656098035403,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004832337844707496,
      "loss": 5.3757,
      "mean_token_accuracy": 0.17193863540887833,
      "num_tokens": 29092831.0,
      "step": 16770
    },
    {
      "entropy": 5.416094732284546,
      "epoch": 1.3051546391752578,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048322328456015855,
      "loss": 5.066,
      "mean_token_accuracy": 0.18399367481470108,
      "num_tokens": 29101067.0,
      "step": 16775
    },
    {
      "entropy": 5.500089979171753,
      "epoch": 1.3055436685469752,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048321278149011053,
      "loss": 5.227,
      "mean_token_accuracy": 0.174577197432518,
      "num_tokens": 29110423.0,
      "step": 16780
    },
    {
      "entropy": 5.483604097366333,
      "epoch": 1.305932697918693,
      "grad_norm": 1.25,
      "learning_rate": 0.00048320227526076504,
      "loss": 5.1441,
      "mean_token_accuracy": 0.17388620972633362,
      "num_tokens": 29119508.0,
      "step": 16785
    },
    {
      "entropy": 5.421654558181762,
      "epoch": 1.3063217272904104,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004831917658722814,
      "loss": 5.2299,
      "mean_token_accuracy": 0.17941801995038986,
      "num_tokens": 29127899.0,
      "step": 16790
    },
    {
      "entropy": 5.481058168411255,
      "epoch": 1.3067107566621279,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048318125332481903,
      "loss": 5.2656,
      "mean_token_accuracy": 0.18058269768953322,
      "num_tokens": 29137261.0,
      "step": 16795
    },
    {
      "entropy": 5.489014196395874,
      "epoch": 1.3070997860338456,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048317073761853764,
      "loss": 5.2868,
      "mean_token_accuracy": 0.17971061319112777,
      "num_tokens": 29145439.0,
      "step": 16800
    },
    {
      "entropy": 5.491727685928344,
      "epoch": 1.307488815405563,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004831602187535965,
      "loss": 5.3259,
      "mean_token_accuracy": 0.1692336767911911,
      "num_tokens": 29153871.0,
      "step": 16805
    },
    {
      "entropy": 5.457500648498535,
      "epoch": 1.3078778447772808,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004831496967301554,
      "loss": 5.1707,
      "mean_token_accuracy": 0.18118292689323426,
      "num_tokens": 29162343.0,
      "step": 16810
    },
    {
      "entropy": 5.448733282089234,
      "epoch": 1.3082668741489982,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048313917154837386,
      "loss": 5.2051,
      "mean_token_accuracy": 0.1833004727959633,
      "num_tokens": 29170533.0,
      "step": 16815
    },
    {
      "entropy": 5.469147729873657,
      "epoch": 1.308655903520716,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004831286432084118,
      "loss": 5.2341,
      "mean_token_accuracy": 0.17693759500980377,
      "num_tokens": 29179353.0,
      "step": 16820
    },
    {
      "entropy": 5.430820512771606,
      "epoch": 1.3090449328924334,
      "grad_norm": 1.25,
      "learning_rate": 0.0004831181117104289,
      "loss": 5.1974,
      "mean_token_accuracy": 0.1753191351890564,
      "num_tokens": 29188747.0,
      "step": 16825
    },
    {
      "entropy": 5.444255399703979,
      "epoch": 1.3094339622641509,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048310757705458476,
      "loss": 5.1985,
      "mean_token_accuracy": 0.1824263870716095,
      "num_tokens": 29197302.0,
      "step": 16830
    },
    {
      "entropy": 5.497534704208374,
      "epoch": 1.3098229916358686,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048309703924103944,
      "loss": 5.2056,
      "mean_token_accuracy": 0.17513555586338042,
      "num_tokens": 29205667.0,
      "step": 16835
    },
    {
      "entropy": 5.54513897895813,
      "epoch": 1.310212021007586,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048308649826995283,
      "loss": 5.2693,
      "mean_token_accuracy": 0.17692117691040038,
      "num_tokens": 29214475.0,
      "step": 16840
    },
    {
      "entropy": 5.508228349685669,
      "epoch": 1.3106010503793035,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048307595414148475,
      "loss": 5.1828,
      "mean_token_accuracy": 0.17780641913414003,
      "num_tokens": 29222990.0,
      "step": 16845
    },
    {
      "entropy": 5.46151909828186,
      "epoch": 1.3109900797510212,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004830654068557952,
      "loss": 5.2836,
      "mean_token_accuracy": 0.17842273265123368,
      "num_tokens": 29232351.0,
      "step": 16850
    },
    {
      "entropy": 5.4863687515258786,
      "epoch": 1.3113791091227387,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004830548564130444,
      "loss": 5.1584,
      "mean_token_accuracy": 0.1802321657538414,
      "num_tokens": 29240418.0,
      "step": 16855
    },
    {
      "entropy": 5.517590761184692,
      "epoch": 1.3117681384944564,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048304430281339216,
      "loss": 5.283,
      "mean_token_accuracy": 0.1731072261929512,
      "num_tokens": 29248975.0,
      "step": 16860
    },
    {
      "entropy": 5.425411605834961,
      "epoch": 1.3121571678661739,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004830337460569989,
      "loss": 5.2472,
      "mean_token_accuracy": 0.17618439346551895,
      "num_tokens": 29257866.0,
      "step": 16865
    },
    {
      "entropy": 5.4017597198486325,
      "epoch": 1.3125461972378916,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004830231861440246,
      "loss": 5.0254,
      "mean_token_accuracy": 0.19110275357961654,
      "num_tokens": 29266057.0,
      "step": 16870
    },
    {
      "entropy": 5.408912515640258,
      "epoch": 1.312935226609609,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004830126230746295,
      "loss": 5.1796,
      "mean_token_accuracy": 0.17994275838136672,
      "num_tokens": 29274752.0,
      "step": 16875
    },
    {
      "entropy": 5.537418508529663,
      "epoch": 1.3133242559813265,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048300205684897405,
      "loss": 5.2497,
      "mean_token_accuracy": 0.17239557206630707,
      "num_tokens": 29282781.0,
      "step": 16880
    },
    {
      "entropy": 5.393339681625366,
      "epoch": 1.3137132853530442,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004829914874672184,
      "loss": 5.2126,
      "mean_token_accuracy": 0.17632610499858856,
      "num_tokens": 29291196.0,
      "step": 16885
    },
    {
      "entropy": 5.5158288955688475,
      "epoch": 1.3141023147247617,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048298091492952297,
      "loss": 5.3026,
      "mean_token_accuracy": 0.17467147409915923,
      "num_tokens": 29300168.0,
      "step": 16890
    },
    {
      "entropy": 5.4659029006958,
      "epoch": 1.3144913440964792,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004829703392360482,
      "loss": 5.1301,
      "mean_token_accuracy": 0.17820778638124465,
      "num_tokens": 29308468.0,
      "step": 16895
    },
    {
      "entropy": 5.382673978805542,
      "epoch": 1.3148803734681969,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048295976038695455,
      "loss": 5.1511,
      "mean_token_accuracy": 0.17897773236036302,
      "num_tokens": 29317096.0,
      "step": 16900
    },
    {
      "entropy": 5.484233856201172,
      "epoch": 1.3152694028399143,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004829491783824025,
      "loss": 5.2672,
      "mean_token_accuracy": 0.16711802929639816,
      "num_tokens": 29325840.0,
      "step": 16905
    },
    {
      "entropy": 5.46508641242981,
      "epoch": 1.315658432211632,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048293859322255276,
      "loss": 5.2033,
      "mean_token_accuracy": 0.1824135959148407,
      "num_tokens": 29333874.0,
      "step": 16910
    },
    {
      "entropy": 5.467566967010498,
      "epoch": 1.3160474615833495,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004829280049075657,
      "loss": 5.1567,
      "mean_token_accuracy": 0.1817405641078949,
      "num_tokens": 29341937.0,
      "step": 16915
    },
    {
      "entropy": 5.521892023086548,
      "epoch": 1.3164364909550672,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048291741343760216,
      "loss": 5.3964,
      "mean_token_accuracy": 0.16771324574947358,
      "num_tokens": 29350231.0,
      "step": 16920
    },
    {
      "entropy": 5.48892912864685,
      "epoch": 1.3168255203267847,
      "grad_norm": 1.25,
      "learning_rate": 0.0004829068188128229,
      "loss": 5.2543,
      "mean_token_accuracy": 0.17544516921043396,
      "num_tokens": 29359036.0,
      "step": 16925
    },
    {
      "entropy": 5.542684650421142,
      "epoch": 1.3172145496985022,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004828962210333885,
      "loss": 5.1423,
      "mean_token_accuracy": 0.1826049953699112,
      "num_tokens": 29368063.0,
      "step": 16930
    },
    {
      "entropy": 5.4489278316497805,
      "epoch": 1.3176035790702199,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004828856200994598,
      "loss": 5.1149,
      "mean_token_accuracy": 0.1790579751133919,
      "num_tokens": 29376653.0,
      "step": 16935
    },
    {
      "entropy": 5.373625612258911,
      "epoch": 1.3179926084419373,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004828750160111978,
      "loss": 5.1168,
      "mean_token_accuracy": 0.1853714406490326,
      "num_tokens": 29384978.0,
      "step": 16940
    },
    {
      "entropy": 5.48098464012146,
      "epoch": 1.3183816378136548,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048286440876876325,
      "loss": 5.3055,
      "mean_token_accuracy": 0.17115148305892944,
      "num_tokens": 29393927.0,
      "step": 16945
    },
    {
      "entropy": 5.550119924545288,
      "epoch": 1.3187706671853725,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048285379837231714,
      "loss": 5.3035,
      "mean_token_accuracy": 0.17337229549884797,
      "num_tokens": 29402393.0,
      "step": 16950
    },
    {
      "entropy": 5.475613307952881,
      "epoch": 1.31915969655709,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004828431848220205,
      "loss": 5.2311,
      "mean_token_accuracy": 0.17744748890399933,
      "num_tokens": 29411011.0,
      "step": 16955
    },
    {
      "entropy": 5.496192884445191,
      "epoch": 1.3195487259288077,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004828325681180343,
      "loss": 5.2522,
      "mean_token_accuracy": 0.17789504081010818,
      "num_tokens": 29420095.0,
      "step": 16960
    },
    {
      "entropy": 5.453487682342529,
      "epoch": 1.3199377553005252,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004828219482605197,
      "loss": 5.253,
      "mean_token_accuracy": 0.17630696892738343,
      "num_tokens": 29429274.0,
      "step": 16965
    },
    {
      "entropy": 5.516122627258301,
      "epoch": 1.3203267846722428,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048281132524963786,
      "loss": 5.2082,
      "mean_token_accuracy": 0.17345507442951202,
      "num_tokens": 29437572.0,
      "step": 16970
    },
    {
      "entropy": 5.499616098403931,
      "epoch": 1.3207158140439603,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004828006990855499,
      "loss": 5.2477,
      "mean_token_accuracy": 0.18053370267152785,
      "num_tokens": 29446294.0,
      "step": 16975
    },
    {
      "entropy": 5.432140302658081,
      "epoch": 1.3211048434156778,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048279006976841704,
      "loss": 5.2067,
      "mean_token_accuracy": 0.17865952402353286,
      "num_tokens": 29454710.0,
      "step": 16980
    },
    {
      "entropy": 5.48644232749939,
      "epoch": 1.3214938727873955,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004827794372984007,
      "loss": 5.2193,
      "mean_token_accuracy": 0.17921172380447387,
      "num_tokens": 29463067.0,
      "step": 16985
    },
    {
      "entropy": 5.501447010040283,
      "epoch": 1.321882902159113,
      "grad_norm": 1.265625,
      "learning_rate": 0.000482768801675662,
      "loss": 5.2372,
      "mean_token_accuracy": 0.1698783040046692,
      "num_tokens": 29471636.0,
      "step": 16990
    },
    {
      "entropy": 5.511439561843872,
      "epoch": 1.3222719315308304,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004827581629003625,
      "loss": 5.2505,
      "mean_token_accuracy": 0.1751886785030365,
      "num_tokens": 29480772.0,
      "step": 16995
    },
    {
      "entropy": 5.450866270065307,
      "epoch": 1.3226609609025481,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048274752097266356,
      "loss": 5.1596,
      "mean_token_accuracy": 0.1820823907852173,
      "num_tokens": 29489113.0,
      "step": 17000
    },
    {
      "entropy": 5.525137948989868,
      "epoch": 1.3230499902742658,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004827368758927266,
      "loss": 5.2449,
      "mean_token_accuracy": 0.17053622007369995,
      "num_tokens": 29497721.0,
      "step": 17005
    },
    {
      "entropy": 5.47624888420105,
      "epoch": 1.3234390196459833,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004827262276607132,
      "loss": 5.216,
      "mean_token_accuracy": 0.18076564073562623,
      "num_tokens": 29506452.0,
      "step": 17010
    },
    {
      "entropy": 5.46647834777832,
      "epoch": 1.3238280490177008,
      "grad_norm": 1.34375,
      "learning_rate": 0.000482715576276785,
      "loss": 5.1864,
      "mean_token_accuracy": 0.1806783676147461,
      "num_tokens": 29514635.0,
      "step": 17015
    },
    {
      "entropy": 5.479906320571899,
      "epoch": 1.3242170783894185,
      "grad_norm": 1.25,
      "learning_rate": 0.0004827049217411035,
      "loss": 5.2565,
      "mean_token_accuracy": 0.17529368698596953,
      "num_tokens": 29523929.0,
      "step": 17020
    },
    {
      "entropy": 5.443309116363525,
      "epoch": 1.324606107761136,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048269426405383043,
      "loss": 5.162,
      "mean_token_accuracy": 0.1775990307331085,
      "num_tokens": 29533125.0,
      "step": 17025
    },
    {
      "entropy": 5.446059894561768,
      "epoch": 1.3249951371328534,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048268360321512745,
      "loss": 5.1814,
      "mean_token_accuracy": 0.18526019155979156,
      "num_tokens": 29541604.0,
      "step": 17030
    },
    {
      "entropy": 5.484731435775757,
      "epoch": 1.3253841665045711,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004826729392251564,
      "loss": 5.2034,
      "mean_token_accuracy": 0.17930059283971786,
      "num_tokens": 29550100.0,
      "step": 17035
    },
    {
      "entropy": 5.464026641845703,
      "epoch": 1.3257731958762886,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000482662272084079,
      "loss": 5.2411,
      "mean_token_accuracy": 0.17819985300302504,
      "num_tokens": 29559490.0,
      "step": 17040
    },
    {
      "entropy": 5.519045114517212,
      "epoch": 1.326162225248006,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004826516017920571,
      "loss": 5.2562,
      "mean_token_accuracy": 0.17001968175172805,
      "num_tokens": 29568734.0,
      "step": 17045
    },
    {
      "entropy": 5.507044982910156,
      "epoch": 1.3265512546197238,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004826409283492528,
      "loss": 5.243,
      "mean_token_accuracy": 0.17576885372400283,
      "num_tokens": 29577339.0,
      "step": 17050
    },
    {
      "entropy": 5.522839736938477,
      "epoch": 1.3269402839914415,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048263025175582783,
      "loss": 5.3491,
      "mean_token_accuracy": 0.17169837057590484,
      "num_tokens": 29585561.0,
      "step": 17055
    },
    {
      "entropy": 5.51704969406128,
      "epoch": 1.327329313363159,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004826195720119442,
      "loss": 5.2823,
      "mean_token_accuracy": 0.17219843715429306,
      "num_tokens": 29594663.0,
      "step": 17060
    },
    {
      "entropy": 5.561356258392334,
      "epoch": 1.3277183427348764,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004826088891177641,
      "loss": 5.2329,
      "mean_token_accuracy": 0.1816175475716591,
      "num_tokens": 29603537.0,
      "step": 17065
    },
    {
      "entropy": 5.515897226333618,
      "epoch": 1.3281073721065941,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048259820307344954,
      "loss": 5.2588,
      "mean_token_accuracy": 0.17831403762102127,
      "num_tokens": 29612234.0,
      "step": 17070
    },
    {
      "entropy": 5.388201093673706,
      "epoch": 1.3284964014783116,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048258751387916267,
      "loss": 5.0836,
      "mean_token_accuracy": 0.17967843115329743,
      "num_tokens": 29621435.0,
      "step": 17075
    },
    {
      "entropy": 5.561673927307129,
      "epoch": 1.328885430850029,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048257682153506564,
      "loss": 5.3641,
      "mean_token_accuracy": 0.1663633793592453,
      "num_tokens": 29630266.0,
      "step": 17080
    },
    {
      "entropy": 5.374151372909546,
      "epoch": 1.3292744602217468,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004825661260413208,
      "loss": 5.1483,
      "mean_token_accuracy": 0.17775199860334395,
      "num_tokens": 29639118.0,
      "step": 17085
    },
    {
      "entropy": 5.425501585006714,
      "epoch": 1.3296634895934643,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048255542739809035,
      "loss": 5.2421,
      "mean_token_accuracy": 0.1852846309542656,
      "num_tokens": 29647616.0,
      "step": 17090
    },
    {
      "entropy": 5.543737077713013,
      "epoch": 1.330052518965182,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048254472560553665,
      "loss": 5.1914,
      "mean_token_accuracy": 0.17968874275684357,
      "num_tokens": 29656264.0,
      "step": 17095
    },
    {
      "entropy": 5.519982814788818,
      "epoch": 1.3304415483368994,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048253402066382207,
      "loss": 5.2039,
      "mean_token_accuracy": 0.1792625069618225,
      "num_tokens": 29664452.0,
      "step": 17100
    },
    {
      "entropy": 5.433029651641846,
      "epoch": 1.3308305777086171,
      "grad_norm": 1.375,
      "learning_rate": 0.0004825233125731091,
      "loss": 5.1378,
      "mean_token_accuracy": 0.1870576724410057,
      "num_tokens": 29673456.0,
      "step": 17105
    },
    {
      "entropy": 5.486018991470337,
      "epoch": 1.3312196070803346,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048251260133356014,
      "loss": 5.2682,
      "mean_token_accuracy": 0.17460379749536514,
      "num_tokens": 29682232.0,
      "step": 17110
    },
    {
      "entropy": 5.460006380081177,
      "epoch": 1.331608636452052,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048250188694533774,
      "loss": 5.1788,
      "mean_token_accuracy": 0.1786024197936058,
      "num_tokens": 29690421.0,
      "step": 17115
    },
    {
      "entropy": 5.395324182510376,
      "epoch": 1.3319976658237698,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004824911694086044,
      "loss": 5.1635,
      "mean_token_accuracy": 0.18143800646066666,
      "num_tokens": 29698844.0,
      "step": 17120
    },
    {
      "entropy": 5.4610881328582765,
      "epoch": 1.3323866951954872,
      "grad_norm": 1.390625,
      "learning_rate": 0.000482480448723523,
      "loss": 5.1784,
      "mean_token_accuracy": 0.1782989591360092,
      "num_tokens": 29706738.0,
      "step": 17125
    },
    {
      "entropy": 5.501462984085083,
      "epoch": 1.3327757245672047,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048246972489025586,
      "loss": 5.2592,
      "mean_token_accuracy": 0.17563404589891435,
      "num_tokens": 29715812.0,
      "step": 17130
    },
    {
      "entropy": 5.419402408599853,
      "epoch": 1.3331647539389224,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048245899790896597,
      "loss": 5.1575,
      "mean_token_accuracy": 0.1819878950715065,
      "num_tokens": 29724488.0,
      "step": 17135
    },
    {
      "entropy": 5.420894145965576,
      "epoch": 1.33355378331064,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004824482677798159,
      "loss": 5.1467,
      "mean_token_accuracy": 0.1811024561524391,
      "num_tokens": 29733565.0,
      "step": 17140
    },
    {
      "entropy": 5.422998380661011,
      "epoch": 1.3339428126823576,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048243753450296863,
      "loss": 5.2252,
      "mean_token_accuracy": 0.18027852475643158,
      "num_tokens": 29742437.0,
      "step": 17145
    },
    {
      "entropy": 5.450380516052246,
      "epoch": 1.334331842054075,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048242679807858693,
      "loss": 5.2529,
      "mean_token_accuracy": 0.17903263717889786,
      "num_tokens": 29751636.0,
      "step": 17150
    },
    {
      "entropy": 5.498220634460449,
      "epoch": 1.3347208714257928,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048241605850683365,
      "loss": 5.2026,
      "mean_token_accuracy": 0.17844272255897523,
      "num_tokens": 29760199.0,
      "step": 17155
    },
    {
      "entropy": 5.454590940475464,
      "epoch": 1.3351099007975102,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048240531578787197,
      "loss": 5.1554,
      "mean_token_accuracy": 0.18274222761392594,
      "num_tokens": 29769169.0,
      "step": 17160
    },
    {
      "entropy": 5.435620546340942,
      "epoch": 1.3354989301692277,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004823945699218647,
      "loss": 5.1497,
      "mean_token_accuracy": 0.1851333811879158,
      "num_tokens": 29777218.0,
      "step": 17165
    },
    {
      "entropy": 5.440847492218017,
      "epoch": 1.3358879595409454,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004823838209089749,
      "loss": 5.2004,
      "mean_token_accuracy": 0.1769283175468445,
      "num_tokens": 29786221.0,
      "step": 17170
    },
    {
      "entropy": 5.471513938903809,
      "epoch": 1.3362769889126629,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048237306874936565,
      "loss": 5.235,
      "mean_token_accuracy": 0.17463277131319047,
      "num_tokens": 29794659.0,
      "step": 17175
    },
    {
      "entropy": 5.5716637134552,
      "epoch": 1.3366660182843804,
      "grad_norm": 6.65625,
      "learning_rate": 0.0004823623134432001,
      "loss": 5.2052,
      "mean_token_accuracy": 0.17814639508724212,
      "num_tokens": 29804237.0,
      "step": 17180
    },
    {
      "entropy": 5.563660335540772,
      "epoch": 1.337055047656098,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048235155499064166,
      "loss": 5.3099,
      "mean_token_accuracy": 0.17338835448026657,
      "num_tokens": 29813521.0,
      "step": 17185
    },
    {
      "entropy": 5.528487777709961,
      "epoch": 1.3374440770278155,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048234079339185335,
      "loss": 5.3651,
      "mean_token_accuracy": 0.16922378093004226,
      "num_tokens": 29822123.0,
      "step": 17190
    },
    {
      "entropy": 5.453748035430908,
      "epoch": 1.3378331063995332,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048233002864699856,
      "loss": 5.2713,
      "mean_token_accuracy": 0.17731060236692428,
      "num_tokens": 29830557.0,
      "step": 17195
    },
    {
      "entropy": 5.457331371307373,
      "epoch": 1.3382221357712507,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004823192607562405,
      "loss": 5.1854,
      "mean_token_accuracy": 0.18171877264976502,
      "num_tokens": 29839566.0,
      "step": 17200
    },
    {
      "entropy": 5.478645992279053,
      "epoch": 1.3386111651429684,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048230848971974265,
      "loss": 5.1683,
      "mean_token_accuracy": 0.18375956118106843,
      "num_tokens": 29847808.0,
      "step": 17205
    },
    {
      "entropy": 5.427330732345581,
      "epoch": 1.3390001945146859,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004822977155376684,
      "loss": 5.127,
      "mean_token_accuracy": 0.18357300460338594,
      "num_tokens": 29855974.0,
      "step": 17210
    },
    {
      "entropy": 5.520886850357056,
      "epoch": 1.3393892238864034,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004822869382101814,
      "loss": 5.2807,
      "mean_token_accuracy": 0.1766713872551918,
      "num_tokens": 29864910.0,
      "step": 17215
    },
    {
      "entropy": 5.550689029693603,
      "epoch": 1.339778253258121,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004822761577374449,
      "loss": 5.3471,
      "mean_token_accuracy": 0.17259298861026764,
      "num_tokens": 29873668.0,
      "step": 17220
    },
    {
      "entropy": 5.477415704727173,
      "epoch": 1.3401672826298385,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004822653741196227,
      "loss": 5.2585,
      "mean_token_accuracy": 0.17247983664274216,
      "num_tokens": 29882099.0,
      "step": 17225
    },
    {
      "entropy": 5.45361852645874,
      "epoch": 1.340556312001556,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004822545873568783,
      "loss": 5.1784,
      "mean_token_accuracy": 0.18330150544643403,
      "num_tokens": 29891022.0,
      "step": 17230
    },
    {
      "entropy": 5.4447942733764645,
      "epoch": 1.3409453413732737,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048224379744937545,
      "loss": 5.1835,
      "mean_token_accuracy": 0.1798565000295639,
      "num_tokens": 29899270.0,
      "step": 17235
    },
    {
      "entropy": 5.515349578857422,
      "epoch": 1.3413343707449912,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048223300439727783,
      "loss": 5.3131,
      "mean_token_accuracy": 0.16852786839008332,
      "num_tokens": 29908183.0,
      "step": 17240
    },
    {
      "entropy": 5.450714254379273,
      "epoch": 1.3417234001167089,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048222220820074926,
      "loss": 5.2287,
      "mean_token_accuracy": 0.1799772396683693,
      "num_tokens": 29917148.0,
      "step": 17245
    },
    {
      "entropy": 5.4187837600708,
      "epoch": 1.3421124294884264,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048221140885995346,
      "loss": 5.1554,
      "mean_token_accuracy": 0.1820338860154152,
      "num_tokens": 29926251.0,
      "step": 17250
    },
    {
      "entropy": 5.434029197692871,
      "epoch": 1.342501458860144,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004822006063750543,
      "loss": 5.2358,
      "mean_token_accuracy": 0.17981840223073958,
      "num_tokens": 29935000.0,
      "step": 17255
    },
    {
      "entropy": 5.40452561378479,
      "epoch": 1.3428904882318615,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048218980074621575,
      "loss": 5.1108,
      "mean_token_accuracy": 0.1944351002573967,
      "num_tokens": 29943046.0,
      "step": 17260
    },
    {
      "entropy": 5.487700319290161,
      "epoch": 1.343279517603579,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048217899197360183,
      "loss": 5.2372,
      "mean_token_accuracy": 0.17940936535596846,
      "num_tokens": 29952254.0,
      "step": 17265
    },
    {
      "entropy": 5.4296266555786135,
      "epoch": 1.3436685469752967,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004821681800573764,
      "loss": 5.1783,
      "mean_token_accuracy": 0.17732789516448974,
      "num_tokens": 29961458.0,
      "step": 17270
    },
    {
      "entropy": 5.513119268417358,
      "epoch": 1.3440575763470142,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004821573649977036,
      "loss": 5.1615,
      "mean_token_accuracy": 0.18459303230047225,
      "num_tokens": 29969865.0,
      "step": 17275
    },
    {
      "entropy": 5.474719667434693,
      "epoch": 1.3444466057187316,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048214654679474753,
      "loss": 5.2267,
      "mean_token_accuracy": 0.16858811676502228,
      "num_tokens": 29978260.0,
      "step": 17280
    },
    {
      "entropy": 5.556903934478759,
      "epoch": 1.3448356350904493,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048213572544867224,
      "loss": 5.3478,
      "mean_token_accuracy": 0.1710471212863922,
      "num_tokens": 29987528.0,
      "step": 17285
    },
    {
      "entropy": 5.469323396682739,
      "epoch": 1.3452246644621668,
      "grad_norm": 1.25,
      "learning_rate": 0.00048212490095964213,
      "loss": 5.1757,
      "mean_token_accuracy": 0.1815681740641594,
      "num_tokens": 29995991.0,
      "step": 17290
    },
    {
      "entropy": 5.488766670227051,
      "epoch": 1.3456136938338845,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004821140733278212,
      "loss": 5.1605,
      "mean_token_accuracy": 0.1843622460961342,
      "num_tokens": 30004569.0,
      "step": 17295
    },
    {
      "entropy": 5.473983860015869,
      "epoch": 1.346002723205602,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004821032425533739,
      "loss": 5.2538,
      "mean_token_accuracy": 0.17172167003154754,
      "num_tokens": 30014021.0,
      "step": 17300
    },
    {
      "entropy": 5.39105544090271,
      "epoch": 1.3463917525773197,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004820924086364646,
      "loss": 5.0613,
      "mean_token_accuracy": 0.189045749604702,
      "num_tokens": 30021803.0,
      "step": 17305
    },
    {
      "entropy": 5.41913480758667,
      "epoch": 1.3467807819490372,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048208157157725756,
      "loss": 5.2222,
      "mean_token_accuracy": 0.17783802449703218,
      "num_tokens": 30030707.0,
      "step": 17310
    },
    {
      "entropy": 5.475420761108398,
      "epoch": 1.3471698113207546,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048207073137591726,
      "loss": 5.2424,
      "mean_token_accuracy": 0.17088247388601302,
      "num_tokens": 30040235.0,
      "step": 17315
    },
    {
      "entropy": 5.566338443756104,
      "epoch": 1.3475588406924723,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048205988803260824,
      "loss": 5.2686,
      "mean_token_accuracy": 0.17804193496704102,
      "num_tokens": 30049361.0,
      "step": 17320
    },
    {
      "entropy": 5.467367315292359,
      "epoch": 1.3479478700641898,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048204904154749496,
      "loss": 5.2342,
      "mean_token_accuracy": 0.1735579863190651,
      "num_tokens": 30058896.0,
      "step": 17325
    },
    {
      "entropy": 5.524796009063721,
      "epoch": 1.3483368994359073,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048203819192074206,
      "loss": 5.2722,
      "mean_token_accuracy": 0.16515540480613708,
      "num_tokens": 30067206.0,
      "step": 17330
    },
    {
      "entropy": 5.468643045425415,
      "epoch": 1.348725928807625,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048202733915251407,
      "loss": 5.1675,
      "mean_token_accuracy": 0.18090327680110932,
      "num_tokens": 30076286.0,
      "step": 17335
    },
    {
      "entropy": 5.34609956741333,
      "epoch": 1.3491149581793425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048201648324297573,
      "loss": 5.0929,
      "mean_token_accuracy": 0.18755825161933898,
      "num_tokens": 30084681.0,
      "step": 17340
    },
    {
      "entropy": 5.476779794692993,
      "epoch": 1.3495039875510602,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048200562419229185,
      "loss": 5.1768,
      "mean_token_accuracy": 0.1760203167796135,
      "num_tokens": 30093293.0,
      "step": 17345
    },
    {
      "entropy": 5.410783100128174,
      "epoch": 1.3498930169227776,
      "grad_norm": 1.328125,
      "learning_rate": 0.000481994762000627,
      "loss": 5.0453,
      "mean_token_accuracy": 0.19152640998363496,
      "num_tokens": 30101593.0,
      "step": 17350
    },
    {
      "entropy": 5.444497394561767,
      "epoch": 1.3502820462944953,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048198389666814616,
      "loss": 5.2249,
      "mean_token_accuracy": 0.1762275367975235,
      "num_tokens": 30110328.0,
      "step": 17355
    },
    {
      "entropy": 5.458833980560303,
      "epoch": 1.3506710756662128,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048197302819501416,
      "loss": 5.1774,
      "mean_token_accuracy": 0.18262867033481597,
      "num_tokens": 30118721.0,
      "step": 17360
    },
    {
      "entropy": 5.345045137405395,
      "epoch": 1.3510601050379303,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048196215658139586,
      "loss": 5.1387,
      "mean_token_accuracy": 0.17909569889307023,
      "num_tokens": 30127238.0,
      "step": 17365
    },
    {
      "entropy": 5.407507085800171,
      "epoch": 1.351449134409648,
      "grad_norm": 1.375,
      "learning_rate": 0.0004819512818274562,
      "loss": 5.1529,
      "mean_token_accuracy": 0.18806327134370804,
      "num_tokens": 30135049.0,
      "step": 17370
    },
    {
      "entropy": 5.346249294281006,
      "epoch": 1.3518381637813655,
      "grad_norm": 1.375,
      "learning_rate": 0.0004819404039333603,
      "loss": 5.1611,
      "mean_token_accuracy": 0.17851939648389817,
      "num_tokens": 30143991.0,
      "step": 17375
    },
    {
      "entropy": 5.4415325164794925,
      "epoch": 1.352227193153083,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004819295228992731,
      "loss": 5.3438,
      "mean_token_accuracy": 0.17305629402399064,
      "num_tokens": 30153968.0,
      "step": 17380
    },
    {
      "entropy": 5.558028984069824,
      "epoch": 1.3526162225248006,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048191863872535984,
      "loss": 5.2176,
      "mean_token_accuracy": 0.1768216833472252,
      "num_tokens": 30162927.0,
      "step": 17385
    },
    {
      "entropy": 5.4435567378997805,
      "epoch": 1.3530052518965183,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004819077514117855,
      "loss": 5.1661,
      "mean_token_accuracy": 0.18392012864351273,
      "num_tokens": 30170986.0,
      "step": 17390
    },
    {
      "entropy": 5.4310197830200195,
      "epoch": 1.3533942812682358,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048189686095871545,
      "loss": 5.1422,
      "mean_token_accuracy": 0.18400197327136994,
      "num_tokens": 30179167.0,
      "step": 17395
    },
    {
      "entropy": 5.395035266876221,
      "epoch": 1.3537833106399533,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048188596736631475,
      "loss": 5.1545,
      "mean_token_accuracy": 0.18702241033315659,
      "num_tokens": 30187710.0,
      "step": 17400
    },
    {
      "entropy": 5.45907039642334,
      "epoch": 1.354172340011671,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048187507063474885,
      "loss": 5.2605,
      "mean_token_accuracy": 0.172671340405941,
      "num_tokens": 30196502.0,
      "step": 17405
    },
    {
      "entropy": 5.364462089538574,
      "epoch": 1.3545613693833884,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048186417076418294,
      "loss": 5.1091,
      "mean_token_accuracy": 0.18310269862413406,
      "num_tokens": 30205070.0,
      "step": 17410
    },
    {
      "entropy": 5.452150678634643,
      "epoch": 1.354950398755106,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048185326775478256,
      "loss": 5.1738,
      "mean_token_accuracy": 0.17719151377677916,
      "num_tokens": 30213554.0,
      "step": 17415
    },
    {
      "entropy": 5.422304391860962,
      "epoch": 1.3553394281268236,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048184236160671306,
      "loss": 5.1223,
      "mean_token_accuracy": 0.18125565946102143,
      "num_tokens": 30222687.0,
      "step": 17420
    },
    {
      "entropy": 5.409008359909057,
      "epoch": 1.355728457498541,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048183145232013995,
      "loss": 5.191,
      "mean_token_accuracy": 0.17694730162620545,
      "num_tokens": 30231086.0,
      "step": 17425
    },
    {
      "entropy": 5.405587244033813,
      "epoch": 1.3561174868702586,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048182053989522883,
      "loss": 5.0858,
      "mean_token_accuracy": 0.18508124649524688,
      "num_tokens": 30240355.0,
      "step": 17430
    },
    {
      "entropy": 5.531016492843628,
      "epoch": 1.3565065162419763,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048180962433214515,
      "loss": 5.3092,
      "mean_token_accuracy": 0.17210336029529572,
      "num_tokens": 30249586.0,
      "step": 17435
    },
    {
      "entropy": 5.50889310836792,
      "epoch": 1.356895545613694,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048179870563105453,
      "loss": 5.2417,
      "mean_token_accuracy": 0.17363675981760024,
      "num_tokens": 30257666.0,
      "step": 17440
    },
    {
      "entropy": 5.449280548095703,
      "epoch": 1.3572845749854114,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004817877837921228,
      "loss": 5.3531,
      "mean_token_accuracy": 0.1777123048901558,
      "num_tokens": 30266846.0,
      "step": 17445
    },
    {
      "entropy": 5.37691559791565,
      "epoch": 1.357673604357129,
      "grad_norm": 1.25,
      "learning_rate": 0.0004817768588155155,
      "loss": 5.127,
      "mean_token_accuracy": 0.18548429757356644,
      "num_tokens": 30275587.0,
      "step": 17450
    },
    {
      "entropy": 5.504610681533814,
      "epoch": 1.3580626337288466,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048176593070139865,
      "loss": 5.181,
      "mean_token_accuracy": 0.1836211621761322,
      "num_tokens": 30283948.0,
      "step": 17455
    },
    {
      "entropy": 5.557436275482178,
      "epoch": 1.358451663100564,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048175499944993767,
      "loss": 5.3441,
      "mean_token_accuracy": 0.17612174451351165,
      "num_tokens": 30292302.0,
      "step": 17460
    },
    {
      "entropy": 5.519558000564575,
      "epoch": 1.3588406924722816,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048174406506129884,
      "loss": 5.327,
      "mean_token_accuracy": 0.17316247373819352,
      "num_tokens": 30301414.0,
      "step": 17465
    },
    {
      "entropy": 5.520103979110718,
      "epoch": 1.3592297218439993,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048173312753564787,
      "loss": 5.2202,
      "mean_token_accuracy": 0.1834974467754364,
      "num_tokens": 30311045.0,
      "step": 17470
    },
    {
      "entropy": 5.464147710800171,
      "epoch": 1.3596187512157167,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004817221868731506,
      "loss": 5.12,
      "mean_token_accuracy": 0.18723722249269487,
      "num_tokens": 30318552.0,
      "step": 17475
    },
    {
      "entropy": 5.359785270690918,
      "epoch": 1.3600077805874344,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048171124307397334,
      "loss": 5.2386,
      "mean_token_accuracy": 0.17614844888448716,
      "num_tokens": 30327583.0,
      "step": 17480
    },
    {
      "entropy": 5.415590572357178,
      "epoch": 1.360396809959152,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004817002961382819,
      "loss": 5.1894,
      "mean_token_accuracy": 0.17462695389986038,
      "num_tokens": 30336339.0,
      "step": 17485
    },
    {
      "entropy": 5.515897512435913,
      "epoch": 1.3607858393308696,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048168934606624255,
      "loss": 5.2559,
      "mean_token_accuracy": 0.17807946503162383,
      "num_tokens": 30344878.0,
      "step": 17490
    },
    {
      "entropy": 5.472713994979858,
      "epoch": 1.361174868702587,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048167839285802116,
      "loss": 5.2193,
      "mean_token_accuracy": 0.17641102522611618,
      "num_tokens": 30353071.0,
      "step": 17495
    },
    {
      "entropy": 5.484301948547364,
      "epoch": 1.3615638980743046,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004816674365137843,
      "loss": 5.1823,
      "mean_token_accuracy": 0.17798113226890563,
      "num_tokens": 30361356.0,
      "step": 17500
    },
    {
      "entropy": 5.515079593658447,
      "epoch": 1.3619529274460223,
      "grad_norm": 1.4375,
      "learning_rate": 0.000481656477033698,
      "loss": 5.2219,
      "mean_token_accuracy": 0.17381927371025085,
      "num_tokens": 30370917.0,
      "step": 17505
    },
    {
      "entropy": 5.414577913284302,
      "epoch": 1.3623419568177397,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004816455144179286,
      "loss": 5.241,
      "mean_token_accuracy": 0.1747538834810257,
      "num_tokens": 30380304.0,
      "step": 17510
    },
    {
      "entropy": 5.473174047470093,
      "epoch": 1.3627309861894572,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004816345486666424,
      "loss": 5.1877,
      "mean_token_accuracy": 0.18343633264303208,
      "num_tokens": 30388565.0,
      "step": 17515
    },
    {
      "entropy": 5.479825687408447,
      "epoch": 1.363120015561175,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004816235797800058,
      "loss": 5.3058,
      "mean_token_accuracy": 0.1758017957210541,
      "num_tokens": 30397220.0,
      "step": 17520
    },
    {
      "entropy": 5.463488483428955,
      "epoch": 1.3635090449328924,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004816126077581852,
      "loss": 5.1816,
      "mean_token_accuracy": 0.18080277591943741,
      "num_tokens": 30405235.0,
      "step": 17525
    },
    {
      "entropy": 5.476525735855103,
      "epoch": 1.36389807430461,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048160163260134724,
      "loss": 5.2141,
      "mean_token_accuracy": 0.18124206662178038,
      "num_tokens": 30414218.0,
      "step": 17530
    },
    {
      "entropy": 5.456611013412475,
      "epoch": 1.3642871036763276,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004815906543096583,
      "loss": 5.1867,
      "mean_token_accuracy": 0.1778945431113243,
      "num_tokens": 30422665.0,
      "step": 17535
    },
    {
      "entropy": 5.467821216583252,
      "epoch": 1.3646761330480452,
      "grad_norm": 1.34375,
      "learning_rate": 0.000481579672883285,
      "loss": 5.1962,
      "mean_token_accuracy": 0.17974680364131929,
      "num_tokens": 30431868.0,
      "step": 17540
    },
    {
      "entropy": 5.483196973800659,
      "epoch": 1.3650651624197627,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048156868832239396,
      "loss": 5.22,
      "mean_token_accuracy": 0.17803621739149095,
      "num_tokens": 30440548.0,
      "step": 17545
    },
    {
      "entropy": 5.552192687988281,
      "epoch": 1.3654541917914802,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004815577006271519,
      "loss": 5.2515,
      "mean_token_accuracy": 0.1719839632511139,
      "num_tokens": 30448959.0,
      "step": 17550
    },
    {
      "entropy": 5.46440863609314,
      "epoch": 1.365843221163198,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048154670979772555,
      "loss": 5.1503,
      "mean_token_accuracy": 0.17891211807727814,
      "num_tokens": 30457193.0,
      "step": 17555
    },
    {
      "entropy": 5.524586343765259,
      "epoch": 1.3662322505349154,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004815357158342815,
      "loss": 5.2207,
      "mean_token_accuracy": 0.16915565729141235,
      "num_tokens": 30466253.0,
      "step": 17560
    },
    {
      "entropy": 5.41445746421814,
      "epoch": 1.3666212799066328,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004815247187369868,
      "loss": 5.142,
      "mean_token_accuracy": 0.18338638842105864,
      "num_tokens": 30474675.0,
      "step": 17565
    },
    {
      "entropy": 5.460443115234375,
      "epoch": 1.3670103092783505,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048151371850600814,
      "loss": 5.2015,
      "mean_token_accuracy": 0.18252433389425277,
      "num_tokens": 30483031.0,
      "step": 17570
    },
    {
      "entropy": 5.335499382019043,
      "epoch": 1.367399338650068,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048150271514151255,
      "loss": 5.1746,
      "mean_token_accuracy": 0.18291907161474227,
      "num_tokens": 30491331.0,
      "step": 17575
    },
    {
      "entropy": 5.446661472320557,
      "epoch": 1.3677883680217857,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048149170864366693,
      "loss": 5.202,
      "mean_token_accuracy": 0.17804948687553407,
      "num_tokens": 30499649.0,
      "step": 17580
    },
    {
      "entropy": 5.474894905090332,
      "epoch": 1.3681773973935032,
      "grad_norm": 1.375,
      "learning_rate": 0.00048148069901263836,
      "loss": 5.0611,
      "mean_token_accuracy": 0.1868538647890091,
      "num_tokens": 30508922.0,
      "step": 17585
    },
    {
      "entropy": 5.433991718292236,
      "epoch": 1.368566426765221,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048146968624859375,
      "loss": 5.16,
      "mean_token_accuracy": 0.17930388003587722,
      "num_tokens": 30517745.0,
      "step": 17590
    },
    {
      "entropy": 5.423880386352539,
      "epoch": 1.3689554561369384,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004814586703517003,
      "loss": 5.3126,
      "mean_token_accuracy": 0.171878582239151,
      "num_tokens": 30526597.0,
      "step": 17595
    },
    {
      "entropy": 5.556510734558105,
      "epoch": 1.3693444855086558,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004814476513221251,
      "loss": 5.1754,
      "mean_token_accuracy": 0.18382183015346526,
      "num_tokens": 30533969.0,
      "step": 17600
    },
    {
      "entropy": 5.492372608184814,
      "epoch": 1.3697335148803735,
      "grad_norm": 1.375,
      "learning_rate": 0.00048143662916003545,
      "loss": 5.2695,
      "mean_token_accuracy": 0.17043701261281968,
      "num_tokens": 30543277.0,
      "step": 17605
    },
    {
      "entropy": 5.479914712905884,
      "epoch": 1.370122544252091,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004814256038655985,
      "loss": 5.2343,
      "mean_token_accuracy": 0.16911228001117706,
      "num_tokens": 30551889.0,
      "step": 17610
    },
    {
      "entropy": 5.446575403213501,
      "epoch": 1.3705115736238085,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004814145754389816,
      "loss": 5.2,
      "mean_token_accuracy": 0.18106127679347991,
      "num_tokens": 30560926.0,
      "step": 17615
    },
    {
      "entropy": 5.567795181274414,
      "epoch": 1.3709006029955262,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048140354388035204,
      "loss": 5.2819,
      "mean_token_accuracy": 0.172747839987278,
      "num_tokens": 30570102.0,
      "step": 17620
    },
    {
      "entropy": 5.4808916568756105,
      "epoch": 1.3712896323672437,
      "grad_norm": 1.25,
      "learning_rate": 0.0004813925091898772,
      "loss": 5.241,
      "mean_token_accuracy": 0.1751165583729744,
      "num_tokens": 30578585.0,
      "step": 17625
    },
    {
      "entropy": 5.460775423049927,
      "epoch": 1.3716786617389614,
      "grad_norm": 1.25,
      "learning_rate": 0.0004813814713677246,
      "loss": 5.2241,
      "mean_token_accuracy": 0.1803095296025276,
      "num_tokens": 30587594.0,
      "step": 17630
    },
    {
      "entropy": 5.519592952728272,
      "epoch": 1.3720676911106788,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004813704304140616,
      "loss": 5.2213,
      "mean_token_accuracy": 0.1825554206967354,
      "num_tokens": 30596234.0,
      "step": 17635
    },
    {
      "entropy": 5.491425704956055,
      "epoch": 1.3724567204823965,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004813593863290559,
      "loss": 5.2931,
      "mean_token_accuracy": 0.17476817965507507,
      "num_tokens": 30605822.0,
      "step": 17640
    },
    {
      "entropy": 5.490394067764282,
      "epoch": 1.372845749854114,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004813483391128748,
      "loss": 5.1786,
      "mean_token_accuracy": 0.18474684804677963,
      "num_tokens": 30614154.0,
      "step": 17645
    },
    {
      "entropy": 5.4929540157318115,
      "epoch": 1.3732347792258315,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004813372887656862,
      "loss": 5.2168,
      "mean_token_accuracy": 0.18172314167022705,
      "num_tokens": 30622393.0,
      "step": 17650
    },
    {
      "entropy": 5.38900990486145,
      "epoch": 1.3736238085975492,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004813262352876576,
      "loss": 5.2464,
      "mean_token_accuracy": 0.17344316840171814,
      "num_tokens": 30632173.0,
      "step": 17655
    },
    {
      "entropy": 5.535798072814941,
      "epoch": 1.3740128379692667,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048131517867895684,
      "loss": 5.3273,
      "mean_token_accuracy": 0.16535733193159102,
      "num_tokens": 30640313.0,
      "step": 17660
    },
    {
      "entropy": 5.49175009727478,
      "epoch": 1.3744018673409841,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048130411893975155,
      "loss": 5.1338,
      "mean_token_accuracy": 0.1838139921426773,
      "num_tokens": 30648651.0,
      "step": 17665
    },
    {
      "entropy": 5.464856863021851,
      "epoch": 1.3747908967127018,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004812930560702097,
      "loss": 5.2053,
      "mean_token_accuracy": 0.1786033496260643,
      "num_tokens": 30657782.0,
      "step": 17670
    },
    {
      "entropy": 5.440604257583618,
      "epoch": 1.3751799260844193,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048128199007049907,
      "loss": 5.1995,
      "mean_token_accuracy": 0.18296842128038407,
      "num_tokens": 30666493.0,
      "step": 17675
    },
    {
      "entropy": 5.385986185073852,
      "epoch": 1.375568955456137,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048127092094078756,
      "loss": 5.1122,
      "mean_token_accuracy": 0.18998788595199584,
      "num_tokens": 30675054.0,
      "step": 17680
    },
    {
      "entropy": 5.4945848941802975,
      "epoch": 1.3759579848278545,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048125984868124317,
      "loss": 5.1482,
      "mean_token_accuracy": 0.19134803414344786,
      "num_tokens": 30683927.0,
      "step": 17685
    },
    {
      "entropy": 5.341000127792358,
      "epoch": 1.3763470141995722,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004812487732920338,
      "loss": 5.0636,
      "mean_token_accuracy": 0.18582557290792465,
      "num_tokens": 30691670.0,
      "step": 17690
    },
    {
      "entropy": 5.3209082126617435,
      "epoch": 1.3767360435712896,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004812376947733277,
      "loss": 5.0918,
      "mean_token_accuracy": 0.1884150743484497,
      "num_tokens": 30700312.0,
      "step": 17695
    },
    {
      "entropy": 5.427642917633056,
      "epoch": 1.3771250729430071,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048122661312529263,
      "loss": 5.1483,
      "mean_token_accuracy": 0.18579026460647582,
      "num_tokens": 30708975.0,
      "step": 17700
    },
    {
      "entropy": 5.4120324611663815,
      "epoch": 1.3775141023147248,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004812155283480971,
      "loss": 5.1685,
      "mean_token_accuracy": 0.1819477289915085,
      "num_tokens": 30718314.0,
      "step": 17705
    },
    {
      "entropy": 5.438683605194091,
      "epoch": 1.3779031316864423,
      "grad_norm": 1.375,
      "learning_rate": 0.0004812044404419091,
      "loss": 5.2098,
      "mean_token_accuracy": 0.18508134186267852,
      "num_tokens": 30726801.0,
      "step": 17710
    },
    {
      "entropy": 5.4407487392425535,
      "epoch": 1.3782921610581598,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000481193349406897,
      "loss": 5.0651,
      "mean_token_accuracy": 0.1819971904158592,
      "num_tokens": 30734884.0,
      "step": 17715
    },
    {
      "entropy": 5.430898666381836,
      "epoch": 1.3786811904298775,
      "grad_norm": 1.21875,
      "learning_rate": 0.000481182255243229,
      "loss": 5.2139,
      "mean_token_accuracy": 0.1776076763868332,
      "num_tokens": 30744986.0,
      "step": 17720
    },
    {
      "entropy": 5.377299737930298,
      "epoch": 1.379070219801595,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048117115795107335,
      "loss": 5.1798,
      "mean_token_accuracy": 0.18276818245649337,
      "num_tokens": 30753476.0,
      "step": 17725
    },
    {
      "entropy": 5.555529451370239,
      "epoch": 1.3794592491733126,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004811600575305987,
      "loss": 5.298,
      "mean_token_accuracy": 0.16841984987258912,
      "num_tokens": 30762642.0,
      "step": 17730
    },
    {
      "entropy": 5.5357105255126955,
      "epoch": 1.3798482785450301,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004811489539819731,
      "loss": 5.2546,
      "mean_token_accuracy": 0.17150311768054963,
      "num_tokens": 30771530.0,
      "step": 17735
    },
    {
      "entropy": 5.478751468658447,
      "epoch": 1.3802373079167478,
      "grad_norm": 1.25,
      "learning_rate": 0.00048113784730536544,
      "loss": 5.2516,
      "mean_token_accuracy": 0.17661058604717256,
      "num_tokens": 30780001.0,
      "step": 17740
    },
    {
      "entropy": 5.421697902679443,
      "epoch": 1.3806263372884653,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048112673750094396,
      "loss": 5.1628,
      "mean_token_accuracy": 0.18394747078418733,
      "num_tokens": 30788696.0,
      "step": 17745
    },
    {
      "entropy": 5.434032964706421,
      "epoch": 1.3810153666601828,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004811156245688773,
      "loss": 5.2307,
      "mean_token_accuracy": 0.18660632222890855,
      "num_tokens": 30797457.0,
      "step": 17750
    },
    {
      "entropy": 5.497101306915283,
      "epoch": 1.3814043960319005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004811045085093342,
      "loss": 5.2261,
      "mean_token_accuracy": 0.17334395200014113,
      "num_tokens": 30807048.0,
      "step": 17755
    },
    {
      "entropy": 5.464280128479004,
      "epoch": 1.381793425403618,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004810933893224831,
      "loss": 5.2333,
      "mean_token_accuracy": 0.17793329805135727,
      "num_tokens": 30815315.0,
      "step": 17760
    },
    {
      "entropy": 5.396453237533569,
      "epoch": 1.3821824547753354,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048108226700849287,
      "loss": 5.1526,
      "mean_token_accuracy": 0.17954117357730864,
      "num_tokens": 30823221.0,
      "step": 17765
    },
    {
      "entropy": 5.521250677108765,
      "epoch": 1.382571484147053,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004810711415675323,
      "loss": 5.3294,
      "mean_token_accuracy": 0.17431871443986893,
      "num_tokens": 30832112.0,
      "step": 17770
    },
    {
      "entropy": 5.46862473487854,
      "epoch": 1.3829605135187706,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004810600129997702,
      "loss": 5.2146,
      "mean_token_accuracy": 0.18213479071855546,
      "num_tokens": 30840597.0,
      "step": 17775
    },
    {
      "entropy": 5.407245683670044,
      "epoch": 1.3833495428904883,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048104888130537534,
      "loss": 5.0331,
      "mean_token_accuracy": 0.18684405088424683,
      "num_tokens": 30848947.0,
      "step": 17780
    },
    {
      "entropy": 5.417137145996094,
      "epoch": 1.3837385722622058,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048103774648451664,
      "loss": 5.2356,
      "mean_token_accuracy": 0.17412994503974916,
      "num_tokens": 30856630.0,
      "step": 17785
    },
    {
      "entropy": 5.46316704750061,
      "epoch": 1.3841276016339235,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048102660853736314,
      "loss": 5.2944,
      "mean_token_accuracy": 0.17976730912923813,
      "num_tokens": 30866032.0,
      "step": 17790
    },
    {
      "entropy": 5.499716520309448,
      "epoch": 1.384516631005641,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004810154674640837,
      "loss": 5.1425,
      "mean_token_accuracy": 0.1841568484902382,
      "num_tokens": 30874104.0,
      "step": 17795
    },
    {
      "entropy": 5.392666959762574,
      "epoch": 1.3849056603773584,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048100432326484744,
      "loss": 5.1445,
      "mean_token_accuracy": 0.1890779986977577,
      "num_tokens": 30882792.0,
      "step": 17800
    },
    {
      "entropy": 5.400527095794677,
      "epoch": 1.385294689749076,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004809931759398235,
      "loss": 5.1464,
      "mean_token_accuracy": 0.181832355260849,
      "num_tokens": 30891337.0,
      "step": 17805
    },
    {
      "entropy": 5.458163022994995,
      "epoch": 1.3856837191207936,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048098202548918094,
      "loss": 5.1926,
      "mean_token_accuracy": 0.1809339478611946,
      "num_tokens": 30899589.0,
      "step": 17810
    },
    {
      "entropy": 5.490861940383911,
      "epoch": 1.386072748492511,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048097087191308914,
      "loss": 5.1797,
      "mean_token_accuracy": 0.18101212084293367,
      "num_tokens": 30907893.0,
      "step": 17815
    },
    {
      "entropy": 5.48102068901062,
      "epoch": 1.3864617778642288,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004809597152117171,
      "loss": 5.3031,
      "mean_token_accuracy": 0.17819927781820297,
      "num_tokens": 30916797.0,
      "step": 17820
    },
    {
      "entropy": 5.4765736103057865,
      "epoch": 1.3868508072359464,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004809485553852342,
      "loss": 5.1739,
      "mean_token_accuracy": 0.18122555911540986,
      "num_tokens": 30925451.0,
      "step": 17825
    },
    {
      "entropy": 5.370659446716308,
      "epoch": 1.387239836607664,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004809373924338098,
      "loss": 5.1719,
      "mean_token_accuracy": 0.17870952934026718,
      "num_tokens": 30934595.0,
      "step": 17830
    },
    {
      "entropy": 5.445623731613159,
      "epoch": 1.3876288659793814,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048092622635761316,
      "loss": 5.3087,
      "mean_token_accuracy": 0.18041228353977204,
      "num_tokens": 30943896.0,
      "step": 17835
    },
    {
      "entropy": 5.549954557418824,
      "epoch": 1.388017895351099,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048091505715681395,
      "loss": 5.152,
      "mean_token_accuracy": 0.18019594252109528,
      "num_tokens": 30952255.0,
      "step": 17840
    },
    {
      "entropy": 5.5174291133880615,
      "epoch": 1.3884069247228166,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004809038848315814,
      "loss": 5.2367,
      "mean_token_accuracy": 0.1792559117078781,
      "num_tokens": 30960972.0,
      "step": 17845
    },
    {
      "entropy": 5.4613782405853275,
      "epoch": 1.388795954094534,
      "grad_norm": 1.25,
      "learning_rate": 0.0004808927093820851,
      "loss": 5.2571,
      "mean_token_accuracy": 0.18045680671930314,
      "num_tokens": 30970298.0,
      "step": 17850
    },
    {
      "entropy": 5.524462795257568,
      "epoch": 1.3891849834662517,
      "grad_norm": 1.25,
      "learning_rate": 0.0004808815308084947,
      "loss": 5.1859,
      "mean_token_accuracy": 0.1824512168765068,
      "num_tokens": 30979355.0,
      "step": 17855
    },
    {
      "entropy": 5.39074296951294,
      "epoch": 1.3895740128379692,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004808703491109798,
      "loss": 5.1455,
      "mean_token_accuracy": 0.1863921657204628,
      "num_tokens": 30988085.0,
      "step": 17860
    },
    {
      "entropy": 5.374648761749268,
      "epoch": 1.3899630422096867,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004808591642897099,
      "loss": 5.1849,
      "mean_token_accuracy": 0.18439920395612716,
      "num_tokens": 30996039.0,
      "step": 17865
    },
    {
      "entropy": 5.573904323577881,
      "epoch": 1.3903520715814044,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000480847976344855,
      "loss": 5.3296,
      "mean_token_accuracy": 0.16815685778856276,
      "num_tokens": 31004664.0,
      "step": 17870
    },
    {
      "entropy": 5.465896511077881,
      "epoch": 1.390741100953122,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048083678527658465,
      "loss": 5.1764,
      "mean_token_accuracy": 0.18277042359113693,
      "num_tokens": 31012571.0,
      "step": 17875
    },
    {
      "entropy": 5.400308752059937,
      "epoch": 1.3911301303248396,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004808255910850687,
      "loss": 5.2179,
      "mean_token_accuracy": 0.18377925008535384,
      "num_tokens": 31020901.0,
      "step": 17880
    },
    {
      "entropy": 5.477540159225464,
      "epoch": 1.391519159696557,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004808143937704769,
      "loss": 5.2231,
      "mean_token_accuracy": 0.18380358219146728,
      "num_tokens": 31029707.0,
      "step": 17885
    },
    {
      "entropy": 5.442879486083984,
      "epoch": 1.3919081890682747,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048080319333297937,
      "loss": 5.2011,
      "mean_token_accuracy": 0.18365217000246048,
      "num_tokens": 31037833.0,
      "step": 17890
    },
    {
      "entropy": 5.423038911819458,
      "epoch": 1.3922972184399922,
      "grad_norm": 1.5,
      "learning_rate": 0.00048079198977274597,
      "loss": 5.1446,
      "mean_token_accuracy": 0.1914268985390663,
      "num_tokens": 31046494.0,
      "step": 17895
    },
    {
      "entropy": 5.593123149871826,
      "epoch": 1.3926862478117097,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048078078308994666,
      "loss": 5.3345,
      "mean_token_accuracy": 0.17007485032081604,
      "num_tokens": 31054957.0,
      "step": 17900
    },
    {
      "entropy": 5.522422933578492,
      "epoch": 1.3930752771834274,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004807695732847515,
      "loss": 5.1687,
      "mean_token_accuracy": 0.1780516654253006,
      "num_tokens": 31063068.0,
      "step": 17905
    },
    {
      "entropy": 5.441025829315185,
      "epoch": 1.3934643065551449,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048075836035733053,
      "loss": 5.1862,
      "mean_token_accuracy": 0.18215351849794387,
      "num_tokens": 31072081.0,
      "step": 17910
    },
    {
      "entropy": 5.399029684066773,
      "epoch": 1.3938533359268626,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000480747144307854,
      "loss": 5.0767,
      "mean_token_accuracy": 0.1904923215508461,
      "num_tokens": 31080541.0,
      "step": 17915
    },
    {
      "entropy": 5.498310136795044,
      "epoch": 1.39424236529858,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048073592513649203,
      "loss": 5.241,
      "mean_token_accuracy": 0.17784370183944703,
      "num_tokens": 31089035.0,
      "step": 17920
    },
    {
      "entropy": 5.443628597259521,
      "epoch": 1.3946313946702977,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004807247028434148,
      "loss": 5.3007,
      "mean_token_accuracy": 0.16943613141775132,
      "num_tokens": 31096973.0,
      "step": 17925
    },
    {
      "entropy": 5.445528936386109,
      "epoch": 1.3950204240420152,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004807134774287927,
      "loss": 5.2233,
      "mean_token_accuracy": 0.1736281543970108,
      "num_tokens": 31104526.0,
      "step": 17930
    },
    {
      "entropy": 5.472922611236572,
      "epoch": 1.3954094534137327,
      "grad_norm": 2.8125,
      "learning_rate": 0.00048070224889279603,
      "loss": 5.051,
      "mean_token_accuracy": 0.1874218314886093,
      "num_tokens": 31112350.0,
      "step": 17935
    },
    {
      "entropy": 5.472572708129883,
      "epoch": 1.3957984827854504,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048069101723559505,
      "loss": 5.248,
      "mean_token_accuracy": 0.17868749648332596,
      "num_tokens": 31120333.0,
      "step": 17940
    },
    {
      "entropy": 5.493313550949097,
      "epoch": 1.3961875121571679,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004806797824573603,
      "loss": 5.2392,
      "mean_token_accuracy": 0.17073285281658174,
      "num_tokens": 31129652.0,
      "step": 17945
    },
    {
      "entropy": 5.573933172225952,
      "epoch": 1.3965765415288853,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004806685445582624,
      "loss": 5.2216,
      "mean_token_accuracy": 0.1746518686413765,
      "num_tokens": 31137542.0,
      "step": 17950
    },
    {
      "entropy": 5.415256118774414,
      "epoch": 1.396965570900603,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048065730353847143,
      "loss": 5.2073,
      "mean_token_accuracy": 0.17955394387245177,
      "num_tokens": 31145958.0,
      "step": 17955
    },
    {
      "entropy": 5.419348192214966,
      "epoch": 1.3973546002723205,
      "grad_norm": 1.25,
      "learning_rate": 0.00048064605939815837,
      "loss": 5.1729,
      "mean_token_accuracy": 0.18029398322105408,
      "num_tokens": 31154267.0,
      "step": 17960
    },
    {
      "entropy": 5.541728496551514,
      "epoch": 1.3977436296440382,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004806348121374936,
      "loss": 5.24,
      "mean_token_accuracy": 0.1802792102098465,
      "num_tokens": 31162742.0,
      "step": 17965
    },
    {
      "entropy": 5.570001125335693,
      "epoch": 1.3981326590157557,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004806235617566479,
      "loss": 5.2312,
      "mean_token_accuracy": 0.18512373566627502,
      "num_tokens": 31171275.0,
      "step": 17970
    },
    {
      "entropy": 5.361103773117065,
      "epoch": 1.3985216883874734,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000480612308255792,
      "loss": 4.9543,
      "mean_token_accuracy": 0.18870961517095566,
      "num_tokens": 31179318.0,
      "step": 17975
    },
    {
      "entropy": 5.387025976181031,
      "epoch": 1.3989107177591908,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048060105163509647,
      "loss": 5.1448,
      "mean_token_accuracy": 0.1784622699022293,
      "num_tokens": 31187129.0,
      "step": 17980
    },
    {
      "entropy": 5.516384696960449,
      "epoch": 1.3992997471309083,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004805897918947323,
      "loss": 5.3352,
      "mean_token_accuracy": 0.17232319414615632,
      "num_tokens": 31196381.0,
      "step": 17985
    },
    {
      "entropy": 5.514331865310669,
      "epoch": 1.399688776502626,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004805785290348702,
      "loss": 5.2541,
      "mean_token_accuracy": 0.17938355058431626,
      "num_tokens": 31205659.0,
      "step": 17990
    },
    {
      "entropy": 5.468711471557617,
      "epoch": 1.4000778058743435,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048056726305568124,
      "loss": 5.2176,
      "mean_token_accuracy": 0.18499796092510223,
      "num_tokens": 31214996.0,
      "step": 17995
    },
    {
      "entropy": 5.529380989074707,
      "epoch": 1.400466835246061,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048055599395733617,
      "loss": 5.1839,
      "mean_token_accuracy": 0.17765311151742935,
      "num_tokens": 31222964.0,
      "step": 18000
    },
    {
      "epoch": 1.400466835246061,
      "eval_entropy": 5.272493297624634,
      "eval_loss": 5.300766944885254,
      "eval_mean_token_accuracy": 0.184081994976644,
      "eval_num_tokens": 31222964.0,
      "eval_runtime": 28.7472,
      "eval_samples_per_second": 1445.634,
      "eval_steps_per_second": 180.713,
      "step": 18000
    },
    {
      "entropy": 5.4736542224884035,
      "epoch": 1.4008558646177787,
      "grad_norm": 1.375,
      "learning_rate": 0.00048054472174000615,
      "loss": 5.2977,
      "mean_token_accuracy": 0.17344066500663757,
      "num_tokens": 31231653.0,
      "step": 18005
    },
    {
      "entropy": 5.451776742935181,
      "epoch": 1.4012448939894961,
      "grad_norm": 1.25,
      "learning_rate": 0.00048053344640386207,
      "loss": 5.1621,
      "mean_token_accuracy": 0.17775968611240386,
      "num_tokens": 31240613.0,
      "step": 18010
    },
    {
      "entropy": 5.411409521102906,
      "epoch": 1.4016339233612138,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048052216794907505,
      "loss": 5.0973,
      "mean_token_accuracy": 0.1889883443713188,
      "num_tokens": 31249724.0,
      "step": 18015
    },
    {
      "entropy": 5.46095027923584,
      "epoch": 1.4020229527329313,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048051088637581624,
      "loss": 5.3297,
      "mean_token_accuracy": 0.17485296130180358,
      "num_tokens": 31258717.0,
      "step": 18020
    },
    {
      "entropy": 5.480286693572998,
      "epoch": 1.402411982104649,
      "grad_norm": 1.28125,
      "learning_rate": 0.000480499601684257,
      "loss": 5.1844,
      "mean_token_accuracy": 0.18745956867933272,
      "num_tokens": 31267170.0,
      "step": 18025
    },
    {
      "entropy": 5.45912275314331,
      "epoch": 1.4028010114763665,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004804883138745682,
      "loss": 5.2108,
      "mean_token_accuracy": 0.1792296662926674,
      "num_tokens": 31277787.0,
      "step": 18030
    },
    {
      "entropy": 5.463729572296143,
      "epoch": 1.403190040848084,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004804770229469214,
      "loss": 5.2181,
      "mean_token_accuracy": 0.1785406768321991,
      "num_tokens": 31286185.0,
      "step": 18035
    },
    {
      "entropy": 5.483088970184326,
      "epoch": 1.4035790702198017,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004804657289014878,
      "loss": 5.1412,
      "mean_token_accuracy": 0.18320417255163193,
      "num_tokens": 31295361.0,
      "step": 18040
    },
    {
      "entropy": 5.49319167137146,
      "epoch": 1.4039680995915191,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048045443173843884,
      "loss": 5.2737,
      "mean_token_accuracy": 0.1763720393180847,
      "num_tokens": 31304036.0,
      "step": 18045
    },
    {
      "entropy": 5.407097291946411,
      "epoch": 1.4043571289632366,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048044313145794587,
      "loss": 5.1305,
      "mean_token_accuracy": 0.1841235414147377,
      "num_tokens": 31311919.0,
      "step": 18050
    },
    {
      "entropy": 5.545353746414184,
      "epoch": 1.4047461583349543,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048043182806018034,
      "loss": 5.2544,
      "mean_token_accuracy": 0.1759094476699829,
      "num_tokens": 31320805.0,
      "step": 18055
    },
    {
      "entropy": 5.438980627059936,
      "epoch": 1.4051351877066718,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048042052154531384,
      "loss": 5.2499,
      "mean_token_accuracy": 0.1783716782927513,
      "num_tokens": 31329688.0,
      "step": 18060
    },
    {
      "entropy": 5.363278198242187,
      "epoch": 1.4055242170783895,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004804092119135179,
      "loss": 5.0917,
      "mean_token_accuracy": 0.181120166182518,
      "num_tokens": 31338301.0,
      "step": 18065
    },
    {
      "entropy": 5.557471990585327,
      "epoch": 1.405913246450107,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004803978991649641,
      "loss": 5.2992,
      "mean_token_accuracy": 0.1732984408736229,
      "num_tokens": 31347371.0,
      "step": 18070
    },
    {
      "entropy": 5.505506706237793,
      "epoch": 1.4063022758218247,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048038658329982404,
      "loss": 5.2598,
      "mean_token_accuracy": 0.1794675827026367,
      "num_tokens": 31355740.0,
      "step": 18075
    },
    {
      "entropy": 5.374305057525635,
      "epoch": 1.4066913051935421,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004803752643182695,
      "loss": 5.0764,
      "mean_token_accuracy": 0.19378335475921632,
      "num_tokens": 31364718.0,
      "step": 18080
    },
    {
      "entropy": 5.535042095184326,
      "epoch": 1.4070803345652596,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004803639422204723,
      "loss": 5.2317,
      "mean_token_accuracy": 0.17666717171669005,
      "num_tokens": 31372845.0,
      "step": 18085
    },
    {
      "entropy": 5.456090307235717,
      "epoch": 1.4074693639369773,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004803526170066041,
      "loss": 5.2188,
      "mean_token_accuracy": 0.17643413841724395,
      "num_tokens": 31380614.0,
      "step": 18090
    },
    {
      "entropy": 5.50577940940857,
      "epoch": 1.4078583933086948,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048034128867683674,
      "loss": 5.281,
      "mean_token_accuracy": 0.1745205208659172,
      "num_tokens": 31389609.0,
      "step": 18095
    },
    {
      "entropy": 5.530187749862671,
      "epoch": 1.4082474226804123,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004803299572313422,
      "loss": 5.2417,
      "mean_token_accuracy": 0.17645535618066788,
      "num_tokens": 31399046.0,
      "step": 18100
    },
    {
      "entropy": 5.4240968227386475,
      "epoch": 1.40863645205213,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004803186226702924,
      "loss": 5.1425,
      "mean_token_accuracy": 0.18600098490715028,
      "num_tokens": 31407832.0,
      "step": 18105
    },
    {
      "entropy": 5.477601766586304,
      "epoch": 1.4090254814238474,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004803072849938592,
      "loss": 5.2309,
      "mean_token_accuracy": 0.1754569962620735,
      "num_tokens": 31417097.0,
      "step": 18110
    },
    {
      "entropy": 5.522157907485962,
      "epoch": 1.4094145107955651,
      "grad_norm": 1.25,
      "learning_rate": 0.0004802959442022149,
      "loss": 5.2575,
      "mean_token_accuracy": 0.17442698031663895,
      "num_tokens": 31425448.0,
      "step": 18115
    },
    {
      "entropy": 5.407347583770752,
      "epoch": 1.4098035401672826,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048028460029553126,
      "loss": 5.0822,
      "mean_token_accuracy": 0.19378546625375748,
      "num_tokens": 31434116.0,
      "step": 18120
    },
    {
      "entropy": 5.347379112243653,
      "epoch": 1.4101925695390003,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048027325327398065,
      "loss": 5.1588,
      "mean_token_accuracy": 0.1845672219991684,
      "num_tokens": 31443456.0,
      "step": 18125
    },
    {
      "entropy": 5.471483039855957,
      "epoch": 1.4105815989107178,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004802619031377351,
      "loss": 5.1765,
      "mean_token_accuracy": 0.17862315028905867,
      "num_tokens": 31452372.0,
      "step": 18130
    },
    {
      "entropy": 5.528848648071289,
      "epoch": 1.4109706282824352,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048025054988696684,
      "loss": 5.1974,
      "mean_token_accuracy": 0.18416267931461333,
      "num_tokens": 31460446.0,
      "step": 18135
    },
    {
      "entropy": 5.402406549453735,
      "epoch": 1.411359657654153,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048023919352184827,
      "loss": 5.1541,
      "mean_token_accuracy": 0.1792267844080925,
      "num_tokens": 31468814.0,
      "step": 18140
    },
    {
      "entropy": 5.398874568939209,
      "epoch": 1.4117486870258704,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048022783404255156,
      "loss": 5.1246,
      "mean_token_accuracy": 0.1826523393392563,
      "num_tokens": 31477617.0,
      "step": 18145
    },
    {
      "entropy": 5.499728107452393,
      "epoch": 1.412137716397588,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004802164714492491,
      "loss": 5.1147,
      "mean_token_accuracy": 0.1891283929347992,
      "num_tokens": 31485226.0,
      "step": 18150
    },
    {
      "entropy": 5.399042367935181,
      "epoch": 1.4125267457693056,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048020510574211335,
      "loss": 5.1589,
      "mean_token_accuracy": 0.18539402037858962,
      "num_tokens": 31493425.0,
      "step": 18155
    },
    {
      "entropy": 5.411426258087158,
      "epoch": 1.412915775141023,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048019373692131674,
      "loss": 5.132,
      "mean_token_accuracy": 0.18215943425893782,
      "num_tokens": 31502499.0,
      "step": 18160
    },
    {
      "entropy": 5.442806959152222,
      "epoch": 1.4133048045127408,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048018236498703176,
      "loss": 5.1541,
      "mean_token_accuracy": 0.18791037797927856,
      "num_tokens": 31510376.0,
      "step": 18165
    },
    {
      "entropy": 5.463871240615845,
      "epoch": 1.4136938338844582,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048017098993943097,
      "loss": 5.3297,
      "mean_token_accuracy": 0.17313849478960036,
      "num_tokens": 31519101.0,
      "step": 18170
    },
    {
      "entropy": 5.426382350921631,
      "epoch": 1.414082863256176,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000480159611778687,
      "loss": 5.1558,
      "mean_token_accuracy": 0.18520258218050004,
      "num_tokens": 31527221.0,
      "step": 18175
    },
    {
      "entropy": 5.404942369461059,
      "epoch": 1.4144718926278934,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048014823050497243,
      "loss": 5.1731,
      "mean_token_accuracy": 0.17832941561937332,
      "num_tokens": 31535711.0,
      "step": 18180
    },
    {
      "entropy": 5.4997752666473385,
      "epoch": 1.4148609219996109,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048013684611846,
      "loss": 5.3189,
      "mean_token_accuracy": 0.1758361354470253,
      "num_tokens": 31545363.0,
      "step": 18185
    },
    {
      "entropy": 5.55587477684021,
      "epoch": 1.4152499513713286,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048012545861932245,
      "loss": 5.1589,
      "mean_token_accuracy": 0.18458433598279952,
      "num_tokens": 31553760.0,
      "step": 18190
    },
    {
      "entropy": 5.463839530944824,
      "epoch": 1.415638980743046,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004801140680077325,
      "loss": 5.1216,
      "mean_token_accuracy": 0.18480269908905028,
      "num_tokens": 31561625.0,
      "step": 18195
    },
    {
      "entropy": 5.383902549743652,
      "epoch": 1.4160280101147635,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004801026742838631,
      "loss": 5.1949,
      "mean_token_accuracy": 0.18475404232740403,
      "num_tokens": 31569345.0,
      "step": 18200
    },
    {
      "entropy": 5.472035932540893,
      "epoch": 1.4164170394864812,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004800912774478871,
      "loss": 5.1681,
      "mean_token_accuracy": 0.1754018098115921,
      "num_tokens": 31577990.0,
      "step": 18205
    },
    {
      "entropy": 5.424478244781494,
      "epoch": 1.416806068858199,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004800798774999773,
      "loss": 5.123,
      "mean_token_accuracy": 0.18509931415319442,
      "num_tokens": 31586556.0,
      "step": 18210
    },
    {
      "entropy": 5.451363658905029,
      "epoch": 1.4171950982299164,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048006847444030686,
      "loss": 5.2498,
      "mean_token_accuracy": 0.1771626204252243,
      "num_tokens": 31595697.0,
      "step": 18215
    },
    {
      "entropy": 5.437131261825561,
      "epoch": 1.4175841276016339,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004800570682690487,
      "loss": 5.1408,
      "mean_token_accuracy": 0.18580842167139053,
      "num_tokens": 31604760.0,
      "step": 18220
    },
    {
      "entropy": 5.506507921218872,
      "epoch": 1.4179731569733516,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004800456589863759,
      "loss": 5.2826,
      "mean_token_accuracy": 0.17110908180475234,
      "num_tokens": 31614115.0,
      "step": 18225
    },
    {
      "entropy": 5.48956265449524,
      "epoch": 1.418362186345069,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004800342465924616,
      "loss": 5.2547,
      "mean_token_accuracy": 0.18479448258876802,
      "num_tokens": 31622492.0,
      "step": 18230
    },
    {
      "entropy": 5.522823667526245,
      "epoch": 1.4187512157167865,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000480022831087479,
      "loss": 5.2703,
      "mean_token_accuracy": 0.1762626886367798,
      "num_tokens": 31632269.0,
      "step": 18235
    },
    {
      "entropy": 5.488721895217895,
      "epoch": 1.4191402450885042,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004800114124716012,
      "loss": 5.2524,
      "mean_token_accuracy": 0.18024343401193618,
      "num_tokens": 31640374.0,
      "step": 18240
    },
    {
      "entropy": 5.400678825378418,
      "epoch": 1.4195292744602217,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004799999907450015,
      "loss": 5.1681,
      "mean_token_accuracy": 0.1865179494023323,
      "num_tokens": 31648406.0,
      "step": 18245
    },
    {
      "entropy": 5.421573638916016,
      "epoch": 1.4199183038319392,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004799885659078533,
      "loss": 5.112,
      "mean_token_accuracy": 0.18376338928937913,
      "num_tokens": 31656478.0,
      "step": 18250
    },
    {
      "entropy": 5.403015947341919,
      "epoch": 1.4203073332036569,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004799771379603299,
      "loss": 5.1401,
      "mean_token_accuracy": 0.18152912706136703,
      "num_tokens": 31665106.0,
      "step": 18255
    },
    {
      "entropy": 5.485751628875732,
      "epoch": 1.4206963625753746,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004799657069026046,
      "loss": 5.2527,
      "mean_token_accuracy": 0.1769503176212311,
      "num_tokens": 31673557.0,
      "step": 18260
    },
    {
      "entropy": 5.411700439453125,
      "epoch": 1.421085391947092,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047995427273485097,
      "loss": 5.0729,
      "mean_token_accuracy": 0.19002403020858766,
      "num_tokens": 31682123.0,
      "step": 18265
    },
    {
      "entropy": 5.459918260574341,
      "epoch": 1.4214744213188095,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047994283545724247,
      "loss": 5.3352,
      "mean_token_accuracy": 0.17166938930749892,
      "num_tokens": 31690387.0,
      "step": 18270
    },
    {
      "entropy": 5.45930118560791,
      "epoch": 1.4218634506905272,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047993139506995273,
      "loss": 5.1459,
      "mean_token_accuracy": 0.18426873832941054,
      "num_tokens": 31699814.0,
      "step": 18275
    },
    {
      "entropy": 5.42512149810791,
      "epoch": 1.4222524800622447,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004799199515731551,
      "loss": 5.1194,
      "mean_token_accuracy": 0.18865997195243836,
      "num_tokens": 31708437.0,
      "step": 18280
    },
    {
      "entropy": 5.439579010009766,
      "epoch": 1.4226415094339622,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047990850496702346,
      "loss": 5.0998,
      "mean_token_accuracy": 0.19117797315120696,
      "num_tokens": 31716935.0,
      "step": 18285
    },
    {
      "entropy": 5.5585535049438475,
      "epoch": 1.4230305388056799,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004798970552517314,
      "loss": 5.3338,
      "mean_token_accuracy": 0.1727210611104965,
      "num_tokens": 31725997.0,
      "step": 18290
    },
    {
      "entropy": 5.46288104057312,
      "epoch": 1.4234195681773973,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047988560242745264,
      "loss": 5.1366,
      "mean_token_accuracy": 0.18529358953237535,
      "num_tokens": 31733651.0,
      "step": 18295
    },
    {
      "entropy": 5.404537296295166,
      "epoch": 1.423808597549115,
      "grad_norm": 1.265625,
      "learning_rate": 0.000479874146494361,
      "loss": 5.1398,
      "mean_token_accuracy": 0.18074147403240204,
      "num_tokens": 31742465.0,
      "step": 18300
    },
    {
      "entropy": 5.483858203887939,
      "epoch": 1.4241976269208325,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004798626874526302,
      "loss": 5.2133,
      "mean_token_accuracy": 0.1797075778245926,
      "num_tokens": 31751547.0,
      "step": 18305
    },
    {
      "entropy": 5.501356506347657,
      "epoch": 1.4245866562925502,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047985122530243426,
      "loss": 5.3235,
      "mean_token_accuracy": 0.17224632501602172,
      "num_tokens": 31761027.0,
      "step": 18310
    },
    {
      "entropy": 5.47981104850769,
      "epoch": 1.4249756856642677,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000479839760043947,
      "loss": 5.2358,
      "mean_token_accuracy": 0.17579241394996642,
      "num_tokens": 31770166.0,
      "step": 18315
    },
    {
      "entropy": 5.4511512279510494,
      "epoch": 1.4253647150359852,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047982829167734245,
      "loss": 5.0721,
      "mean_token_accuracy": 0.19054419845342635,
      "num_tokens": 31778606.0,
      "step": 18320
    },
    {
      "entropy": 5.4705792427062985,
      "epoch": 1.4257537444077029,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047981682020279456,
      "loss": 5.3073,
      "mean_token_accuracy": 0.17375074774026872,
      "num_tokens": 31787637.0,
      "step": 18325
    },
    {
      "entropy": 5.39207820892334,
      "epoch": 1.4261427737794203,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004798053456204775,
      "loss": 5.1349,
      "mean_token_accuracy": 0.19069287776947022,
      "num_tokens": 31796401.0,
      "step": 18330
    },
    {
      "entropy": 5.516585397720337,
      "epoch": 1.4265318031511378,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004797938679305651,
      "loss": 5.2624,
      "mean_token_accuracy": 0.16957140117883682,
      "num_tokens": 31805917.0,
      "step": 18335
    },
    {
      "entropy": 5.543400764465332,
      "epoch": 1.4269208325228555,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047978238713323193,
      "loss": 5.2739,
      "mean_token_accuracy": 0.17382530570030214,
      "num_tokens": 31814890.0,
      "step": 18340
    },
    {
      "entropy": 5.447697353363037,
      "epoch": 1.427309861894573,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047977090322865183,
      "loss": 5.1334,
      "mean_token_accuracy": 0.1870914250612259,
      "num_tokens": 31822469.0,
      "step": 18345
    },
    {
      "entropy": 5.457222938537598,
      "epoch": 1.4276988912662907,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004797594162169993,
      "loss": 5.2635,
      "mean_token_accuracy": 0.17286391407251359,
      "num_tokens": 31831030.0,
      "step": 18350
    },
    {
      "entropy": 5.476812696456909,
      "epoch": 1.4280879206380082,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004797479260984485,
      "loss": 5.1665,
      "mean_token_accuracy": 0.17754374593496322,
      "num_tokens": 31840137.0,
      "step": 18355
    },
    {
      "entropy": 5.349786043167114,
      "epoch": 1.4284769500097259,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047973643287317386,
      "loss": 5.062,
      "mean_token_accuracy": 0.1877047076821327,
      "num_tokens": 31849258.0,
      "step": 18360
    },
    {
      "entropy": 5.371751356124878,
      "epoch": 1.4288659793814433,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047972493654134963,
      "loss": 5.0763,
      "mean_token_accuracy": 0.18509842157363893,
      "num_tokens": 31857506.0,
      "step": 18365
    },
    {
      "entropy": 5.476739168167114,
      "epoch": 1.4292550087531608,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047971343710315043,
      "loss": 5.1959,
      "mean_token_accuracy": 0.17901685386896132,
      "num_tokens": 31866685.0,
      "step": 18370
    },
    {
      "entropy": 5.534469652175903,
      "epoch": 1.4296440381248785,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004797019345587506,
      "loss": 5.2794,
      "mean_token_accuracy": 0.16997788101434708,
      "num_tokens": 31874404.0,
      "step": 18375
    },
    {
      "entropy": 5.5420740127563475,
      "epoch": 1.430033067496596,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004796904289083248,
      "loss": 5.2685,
      "mean_token_accuracy": 0.1790207251906395,
      "num_tokens": 31883075.0,
      "step": 18380
    },
    {
      "entropy": 5.4539999008178714,
      "epoch": 1.4304220968683135,
      "grad_norm": 1.25,
      "learning_rate": 0.0004796789201520474,
      "loss": 5.17,
      "mean_token_accuracy": 0.18373835384845733,
      "num_tokens": 31891539.0,
      "step": 18385
    },
    {
      "entropy": 5.4749456405639645,
      "epoch": 1.4308111262400312,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047966740829009333,
      "loss": 5.2142,
      "mean_token_accuracy": 0.1836016908288002,
      "num_tokens": 31900575.0,
      "step": 18390
    },
    {
      "entropy": 5.483533620834351,
      "epoch": 1.4312001556117486,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000479655893322637,
      "loss": 5.1913,
      "mean_token_accuracy": 0.17648310512304305,
      "num_tokens": 31909256.0,
      "step": 18395
    },
    {
      "entropy": 5.4319721221923825,
      "epoch": 1.4315891849834663,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004796443752498532,
      "loss": 5.1434,
      "mean_token_accuracy": 0.1856548845767975,
      "num_tokens": 31917098.0,
      "step": 18400
    },
    {
      "entropy": 5.549627876281738,
      "epoch": 1.4319782143551838,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004796328540719169,
      "loss": 5.29,
      "mean_token_accuracy": 0.17716718316078187,
      "num_tokens": 31925331.0,
      "step": 18405
    },
    {
      "entropy": 5.482576370239258,
      "epoch": 1.4323672437269015,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004796213297890026,
      "loss": 5.1468,
      "mean_token_accuracy": 0.17827855050563812,
      "num_tokens": 31933769.0,
      "step": 18410
    },
    {
      "entropy": 5.496219873428345,
      "epoch": 1.432756273098619,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004796098024012853,
      "loss": 5.2389,
      "mean_token_accuracy": 0.17835263758897782,
      "num_tokens": 31943089.0,
      "step": 18415
    },
    {
      "entropy": 5.514734601974487,
      "epoch": 1.4331453024703364,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047959827190894,
      "loss": 5.2009,
      "mean_token_accuracy": 0.17688598185777665,
      "num_tokens": 31951452.0,
      "step": 18420
    },
    {
      "entropy": 5.5234983444213865,
      "epoch": 1.4335343318420541,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047958673831214157,
      "loss": 5.2363,
      "mean_token_accuracy": 0.17741104066371918,
      "num_tokens": 31959625.0,
      "step": 18425
    },
    {
      "entropy": 5.445345211029053,
      "epoch": 1.4339233612137716,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047957520161106496,
      "loss": 5.1895,
      "mean_token_accuracy": 0.1802542343735695,
      "num_tokens": 31967906.0,
      "step": 18430
    },
    {
      "entropy": 5.400961828231812,
      "epoch": 1.434312390585489,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004795636618058853,
      "loss": 5.1477,
      "mean_token_accuracy": 0.1850086122751236,
      "num_tokens": 31976183.0,
      "step": 18435
    },
    {
      "entropy": 5.435861492156983,
      "epoch": 1.4347014199572068,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004795521188967777,
      "loss": 5.1463,
      "mean_token_accuracy": 0.18644186854362488,
      "num_tokens": 31984513.0,
      "step": 18440
    },
    {
      "entropy": 5.573759984970093,
      "epoch": 1.4350904493289243,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004795405728839172,
      "loss": 5.3187,
      "mean_token_accuracy": 0.17402064353227614,
      "num_tokens": 31992994.0,
      "step": 18445
    },
    {
      "entropy": 5.59650673866272,
      "epoch": 1.435479478700642,
      "grad_norm": 1.25,
      "learning_rate": 0.0004795290237674792,
      "loss": 5.2034,
      "mean_token_accuracy": 0.1843166545033455,
      "num_tokens": 32002043.0,
      "step": 18450
    },
    {
      "entropy": 5.480944633483887,
      "epoch": 1.4358685080723594,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047951747154763866,
      "loss": 5.0912,
      "mean_token_accuracy": 0.18237205147743224,
      "num_tokens": 32009901.0,
      "step": 18455
    },
    {
      "entropy": 5.3438026905059814,
      "epoch": 1.4362575374440771,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047950591622457106,
      "loss": 5.1714,
      "mean_token_accuracy": 0.17852087765932084,
      "num_tokens": 32017947.0,
      "step": 18460
    },
    {
      "entropy": 5.465938568115234,
      "epoch": 1.4366465668157946,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047949435779845174,
      "loss": 5.1746,
      "mean_token_accuracy": 0.1812824234366417,
      "num_tokens": 32026537.0,
      "step": 18465
    },
    {
      "entropy": 5.461041831970215,
      "epoch": 1.437035596187512,
      "grad_norm": 1.375,
      "learning_rate": 0.00047948279626945596,
      "loss": 5.1411,
      "mean_token_accuracy": 0.18737553507089616,
      "num_tokens": 32034874.0,
      "step": 18470
    },
    {
      "entropy": 5.430530834197998,
      "epoch": 1.4374246255592298,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047947123163775924,
      "loss": 5.2402,
      "mean_token_accuracy": 0.17809039056301118,
      "num_tokens": 32043829.0,
      "step": 18475
    },
    {
      "entropy": 5.483400630950928,
      "epoch": 1.4378136549309473,
      "grad_norm": 1.171875,
      "learning_rate": 0.000479459663903537,
      "loss": 5.2777,
      "mean_token_accuracy": 0.1736520677804947,
      "num_tokens": 32052883.0,
      "step": 18480
    },
    {
      "entropy": 5.544664669036865,
      "epoch": 1.4382026843026647,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004794480930669649,
      "loss": 5.2895,
      "mean_token_accuracy": 0.18204477578401565,
      "num_tokens": 32061421.0,
      "step": 18485
    },
    {
      "entropy": 5.497006273269653,
      "epoch": 1.4385917136743824,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004794365191282183,
      "loss": 5.1987,
      "mean_token_accuracy": 0.18338433057069778,
      "num_tokens": 32069957.0,
      "step": 18490
    },
    {
      "entropy": 5.401572036743164,
      "epoch": 1.4389807430461,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047942494208747297,
      "loss": 5.0825,
      "mean_token_accuracy": 0.18388008922338486,
      "num_tokens": 32079375.0,
      "step": 18495
    },
    {
      "entropy": 5.381296825408936,
      "epoch": 1.4393697724178176,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004794133619449045,
      "loss": 5.1194,
      "mean_token_accuracy": 0.18615116477012633,
      "num_tokens": 32088331.0,
      "step": 18500
    },
    {
      "entropy": 5.431476593017578,
      "epoch": 1.439758801789535,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004794017787006886,
      "loss": 5.1051,
      "mean_token_accuracy": 0.1823434442281723,
      "num_tokens": 32096897.0,
      "step": 18505
    },
    {
      "entropy": 5.421221303939819,
      "epoch": 1.4401478311612528,
      "grad_norm": 1.296875,
      "learning_rate": 0.000479390192355001,
      "loss": 5.1278,
      "mean_token_accuracy": 0.19198459684848784,
      "num_tokens": 32105942.0,
      "step": 18510
    },
    {
      "entropy": 5.554152154922486,
      "epoch": 1.4405368605329703,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004793786029080176,
      "loss": 5.3206,
      "mean_token_accuracy": 0.1747378796339035,
      "num_tokens": 32114540.0,
      "step": 18515
    },
    {
      "entropy": 5.422180271148681,
      "epoch": 1.4409258899046877,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004793670103599143,
      "loss": 5.1427,
      "mean_token_accuracy": 0.17922163009643555,
      "num_tokens": 32123214.0,
      "step": 18520
    },
    {
      "entropy": 5.467208814620972,
      "epoch": 1.4413149192764054,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047935541471086677,
      "loss": 5.1902,
      "mean_token_accuracy": 0.1838235765695572,
      "num_tokens": 32132369.0,
      "step": 18525
    },
    {
      "entropy": 5.4175615310668945,
      "epoch": 1.441703948648123,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004793438159610511,
      "loss": 5.1205,
      "mean_token_accuracy": 0.1836361438035965,
      "num_tokens": 32141037.0,
      "step": 18530
    },
    {
      "entropy": 5.446578788757324,
      "epoch": 1.4420929780198404,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047933221411064334,
      "loss": 5.1378,
      "mean_token_accuracy": 0.1856568157672882,
      "num_tokens": 32149515.0,
      "step": 18535
    },
    {
      "entropy": 5.453254127502442,
      "epoch": 1.442482007391558,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004793206091598194,
      "loss": 5.2152,
      "mean_token_accuracy": 0.1853768527507782,
      "num_tokens": 32158099.0,
      "step": 18540
    },
    {
      "entropy": 5.488775539398193,
      "epoch": 1.4428710367632755,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004793090011087554,
      "loss": 5.1835,
      "mean_token_accuracy": 0.18350646793842315,
      "num_tokens": 32167059.0,
      "step": 18545
    },
    {
      "entropy": 5.497556829452515,
      "epoch": 1.4432600661349932,
      "grad_norm": 1.375,
      "learning_rate": 0.00047929738995762755,
      "loss": 5.3464,
      "mean_token_accuracy": 0.17596405297517775,
      "num_tokens": 32175988.0,
      "step": 18550
    },
    {
      "entropy": 5.378277921676636,
      "epoch": 1.4436490955067107,
      "grad_norm": 1.5,
      "learning_rate": 0.000479285775706612,
      "loss": 5.0903,
      "mean_token_accuracy": 0.1829860419034958,
      "num_tokens": 32185362.0,
      "step": 18555
    },
    {
      "entropy": 5.443124389648437,
      "epoch": 1.4440381248784284,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047927415835588496,
      "loss": 5.1676,
      "mean_token_accuracy": 0.18192839175462722,
      "num_tokens": 32194599.0,
      "step": 18560
    },
    {
      "entropy": 5.489502763748169,
      "epoch": 1.444427154250146,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047926253790562265,
      "loss": 5.1775,
      "mean_token_accuracy": 0.18466929495334625,
      "num_tokens": 32203106.0,
      "step": 18565
    },
    {
      "entropy": 5.443100547790527,
      "epoch": 1.4448161836218634,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047925091435600147,
      "loss": 5.2314,
      "mean_token_accuracy": 0.1805298298597336,
      "num_tokens": 32211652.0,
      "step": 18570
    },
    {
      "entropy": 5.533976268768311,
      "epoch": 1.445205212993581,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047923928770719776,
      "loss": 5.2165,
      "mean_token_accuracy": 0.1746003434062004,
      "num_tokens": 32220435.0,
      "step": 18575
    },
    {
      "entropy": 5.486438274383545,
      "epoch": 1.4455942423652985,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047922765795938797,
      "loss": 5.1587,
      "mean_token_accuracy": 0.1795663818717003,
      "num_tokens": 32228368.0,
      "step": 18580
    },
    {
      "entropy": 5.420218086242675,
      "epoch": 1.445983271737016,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004792160251127485,
      "loss": 5.2194,
      "mean_token_accuracy": 0.1768001914024353,
      "num_tokens": 32237367.0,
      "step": 18585
    },
    {
      "entropy": 5.473250102996826,
      "epoch": 1.4463723011087337,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047920438916745586,
      "loss": 5.2216,
      "mean_token_accuracy": 0.17834050059318543,
      "num_tokens": 32245503.0,
      "step": 18590
    },
    {
      "entropy": 5.500887155532837,
      "epoch": 1.4467613304804512,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004791927501236866,
      "loss": 5.213,
      "mean_token_accuracy": 0.17574112266302108,
      "num_tokens": 32254565.0,
      "step": 18595
    },
    {
      "entropy": 5.495170307159424,
      "epoch": 1.4471503598521689,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047918110798161754,
      "loss": 5.2383,
      "mean_token_accuracy": 0.1742665708065033,
      "num_tokens": 32263701.0,
      "step": 18600
    },
    {
      "entropy": 5.417531108856201,
      "epoch": 1.4475393892238864,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047916946274142503,
      "loss": 5.187,
      "mean_token_accuracy": 0.18311120569705963,
      "num_tokens": 32272205.0,
      "step": 18605
    },
    {
      "entropy": 5.508271169662476,
      "epoch": 1.447928418595604,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047915781440328593,
      "loss": 5.2202,
      "mean_token_accuracy": 0.17715062946081161,
      "num_tokens": 32279918.0,
      "step": 18610
    },
    {
      "entropy": 5.500790691375732,
      "epoch": 1.4483174479673215,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004791461629673769,
      "loss": 5.2148,
      "mean_token_accuracy": 0.18780752867460251,
      "num_tokens": 32288600.0,
      "step": 18615
    },
    {
      "entropy": 5.550813913345337,
      "epoch": 1.448706477339039,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004791345084338748,
      "loss": 5.3296,
      "mean_token_accuracy": 0.1790144056081772,
      "num_tokens": 32298334.0,
      "step": 18620
    },
    {
      "entropy": 5.456232023239136,
      "epoch": 1.4490955067107567,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004791228508029565,
      "loss": 5.1084,
      "mean_token_accuracy": 0.18531265407800673,
      "num_tokens": 32306785.0,
      "step": 18625
    },
    {
      "entropy": 5.411680698394775,
      "epoch": 1.4494845360824742,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047911119007479873,
      "loss": 5.2463,
      "mean_token_accuracy": 0.17908914834260942,
      "num_tokens": 32315419.0,
      "step": 18630
    },
    {
      "entropy": 5.473725891113281,
      "epoch": 1.4498735654541917,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047909952624957866,
      "loss": 5.1102,
      "mean_token_accuracy": 0.18894828259944915,
      "num_tokens": 32323655.0,
      "step": 18635
    },
    {
      "entropy": 5.431823492050171,
      "epoch": 1.4502625948259094,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000479087859327473,
      "loss": 5.1153,
      "mean_token_accuracy": 0.18009142428636551,
      "num_tokens": 32331465.0,
      "step": 18640
    },
    {
      "entropy": 5.43390998840332,
      "epoch": 1.450651624197627,
      "grad_norm": 1.375,
      "learning_rate": 0.000479076189308659,
      "loss": 5.3008,
      "mean_token_accuracy": 0.17412006556987764,
      "num_tokens": 32341211.0,
      "step": 18645
    },
    {
      "entropy": 5.543957757949829,
      "epoch": 1.4510406535693445,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047906451619331364,
      "loss": 5.3411,
      "mean_token_accuracy": 0.17637332379817963,
      "num_tokens": 32349344.0,
      "step": 18650
    },
    {
      "entropy": 5.560733938217163,
      "epoch": 1.451429682941062,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047905283998161404,
      "loss": 5.187,
      "mean_token_accuracy": 0.18302991688251496,
      "num_tokens": 32357797.0,
      "step": 18655
    },
    {
      "entropy": 5.342849540710449,
      "epoch": 1.4518187123127797,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047904116067373743,
      "loss": 5.1231,
      "mean_token_accuracy": 0.1856083557009697,
      "num_tokens": 32366131.0,
      "step": 18660
    },
    {
      "entropy": 5.4081315994262695,
      "epoch": 1.4522077416844972,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004790294782698609,
      "loss": 5.2674,
      "mean_token_accuracy": 0.17749192118644713,
      "num_tokens": 32374738.0,
      "step": 18665
    },
    {
      "entropy": 5.523671245574951,
      "epoch": 1.4525967710562147,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004790177927701618,
      "loss": 5.2087,
      "mean_token_accuracy": 0.18124039620161056,
      "num_tokens": 32383638.0,
      "step": 18670
    },
    {
      "entropy": 5.5095795631408695,
      "epoch": 1.4529858004279324,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004790061041748174,
      "loss": 5.2668,
      "mean_token_accuracy": 0.17633054554462432,
      "num_tokens": 32393269.0,
      "step": 18675
    },
    {
      "entropy": 5.521847915649414,
      "epoch": 1.4533748297996498,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004789944124840051,
      "loss": 5.22,
      "mean_token_accuracy": 0.1749240592122078,
      "num_tokens": 32401568.0,
      "step": 18680
    },
    {
      "entropy": 5.437391138076782,
      "epoch": 1.4537638591713673,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004789827176979022,
      "loss": 5.1255,
      "mean_token_accuracy": 0.18432992696762085,
      "num_tokens": 32409996.0,
      "step": 18685
    },
    {
      "entropy": 5.395550775527954,
      "epoch": 1.454152888543085,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004789710198166864,
      "loss": 5.1909,
      "mean_token_accuracy": 0.17648541033267975,
      "num_tokens": 32418886.0,
      "step": 18690
    },
    {
      "entropy": 5.390384817123413,
      "epoch": 1.4545419179148027,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047895931884053497,
      "loss": 5.0353,
      "mean_token_accuracy": 0.18794193416833876,
      "num_tokens": 32427999.0,
      "step": 18695
    },
    {
      "entropy": 5.505612754821778,
      "epoch": 1.4549309472865202,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047894761476962547,
      "loss": 5.3113,
      "mean_token_accuracy": 0.16816525608301164,
      "num_tokens": 32437115.0,
      "step": 18700
    },
    {
      "entropy": 5.471020364761353,
      "epoch": 1.4553199766582376,
      "grad_norm": 1.25,
      "learning_rate": 0.00047893590760413545,
      "loss": 5.1229,
      "mean_token_accuracy": 0.18578119128942489,
      "num_tokens": 32445668.0,
      "step": 18705
    },
    {
      "entropy": 5.3939516067504885,
      "epoch": 1.4557090060299553,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047892419734424276,
      "loss": 5.1532,
      "mean_token_accuracy": 0.18389686048030854,
      "num_tokens": 32454022.0,
      "step": 18710
    },
    {
      "entropy": 5.538009548187256,
      "epoch": 1.4560980354016728,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047891248399012495,
      "loss": 5.2969,
      "mean_token_accuracy": 0.1747129425406456,
      "num_tokens": 32462997.0,
      "step": 18715
    },
    {
      "entropy": 5.562886142730713,
      "epoch": 1.4564870647733903,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004789007675419597,
      "loss": 5.1503,
      "mean_token_accuracy": 0.18320504426956177,
      "num_tokens": 32471184.0,
      "step": 18720
    },
    {
      "entropy": 5.373207902908325,
      "epoch": 1.456876094145108,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047888904799992486,
      "loss": 5.1711,
      "mean_token_accuracy": 0.18299435079097748,
      "num_tokens": 32480051.0,
      "step": 18725
    },
    {
      "entropy": 5.45562515258789,
      "epoch": 1.4572651235168255,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047887732536419826,
      "loss": 5.2033,
      "mean_token_accuracy": 0.17768926471471785,
      "num_tokens": 32489269.0,
      "step": 18730
    },
    {
      "entropy": 5.488498640060425,
      "epoch": 1.4576541528885432,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004788655996349577,
      "loss": 5.139,
      "mean_token_accuracy": 0.18547287583351135,
      "num_tokens": 32497941.0,
      "step": 18735
    },
    {
      "entropy": 5.489590215682983,
      "epoch": 1.4580431822602606,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047885387081238125,
      "loss": 5.1925,
      "mean_token_accuracy": 0.1797911137342453,
      "num_tokens": 32507679.0,
      "step": 18740
    },
    {
      "entropy": 5.437524032592774,
      "epoch": 1.4584322116319783,
      "grad_norm": 1.375,
      "learning_rate": 0.0004788421388966467,
      "loss": 5.2188,
      "mean_token_accuracy": 0.17793875932693481,
      "num_tokens": 32516964.0,
      "step": 18745
    },
    {
      "entropy": 5.431038904190063,
      "epoch": 1.4588212410036958,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047883040388793207,
      "loss": 5.042,
      "mean_token_accuracy": 0.1924495056271553,
      "num_tokens": 32525178.0,
      "step": 18750
    },
    {
      "entropy": 5.551911449432373,
      "epoch": 1.4592102703754133,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047881866578641563,
      "loss": 5.2174,
      "mean_token_accuracy": 0.17587279230356218,
      "num_tokens": 32533694.0,
      "step": 18755
    },
    {
      "entropy": 5.427580070495606,
      "epoch": 1.459599299747131,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004788069245922753,
      "loss": 5.2121,
      "mean_token_accuracy": 0.1809308186173439,
      "num_tokens": 32542529.0,
      "step": 18760
    },
    {
      "entropy": 5.458859539031982,
      "epoch": 1.4599883291188485,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004787951803056893,
      "loss": 5.2072,
      "mean_token_accuracy": 0.17117880135774613,
      "num_tokens": 32550987.0,
      "step": 18765
    },
    {
      "entropy": 5.390578031539917,
      "epoch": 1.460377358490566,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004787834329268358,
      "loss": 5.1553,
      "mean_token_accuracy": 0.18641261309385299,
      "num_tokens": 32559812.0,
      "step": 18770
    },
    {
      "entropy": 5.462271308898925,
      "epoch": 1.4607663878622836,
      "grad_norm": 1.359375,
      "learning_rate": 0.000478771682455893,
      "loss": 5.1903,
      "mean_token_accuracy": 0.1863370031118393,
      "num_tokens": 32568477.0,
      "step": 18775
    },
    {
      "entropy": 5.589956665039063,
      "epoch": 1.461155417234001,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004787599288930393,
      "loss": 5.2388,
      "mean_token_accuracy": 0.17859549969434738,
      "num_tokens": 32577921.0,
      "step": 18780
    },
    {
      "entropy": 5.46247410774231,
      "epoch": 1.4615444466057188,
      "grad_norm": 1.34375,
      "learning_rate": 0.000478748172238453,
      "loss": 5.1314,
      "mean_token_accuracy": 0.1846252888441086,
      "num_tokens": 32587028.0,
      "step": 18785
    },
    {
      "entropy": 5.370759773254394,
      "epoch": 1.4619334759774363,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004787364124923125,
      "loss": 5.1575,
      "mean_token_accuracy": 0.18063351660966873,
      "num_tokens": 32596092.0,
      "step": 18790
    },
    {
      "entropy": 5.42425045967102,
      "epoch": 1.462322505349154,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047872464965479625,
      "loss": 5.1264,
      "mean_token_accuracy": 0.17888969480991362,
      "num_tokens": 32605271.0,
      "step": 18795
    },
    {
      "entropy": 5.517597579956055,
      "epoch": 1.4627115347208715,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004787128837260826,
      "loss": 5.307,
      "mean_token_accuracy": 0.17559807896614074,
      "num_tokens": 32614240.0,
      "step": 18800
    },
    {
      "entropy": 5.524364614486695,
      "epoch": 1.463100564092589,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004787011147063503,
      "loss": 5.2198,
      "mean_token_accuracy": 0.1774027928709984,
      "num_tokens": 32623390.0,
      "step": 18805
    },
    {
      "entropy": 5.453616142272949,
      "epoch": 1.4634895934643066,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004786893425957777,
      "loss": 5.1869,
      "mean_token_accuracy": 0.18418461829423904,
      "num_tokens": 32631649.0,
      "step": 18810
    },
    {
      "entropy": 5.379134178161621,
      "epoch": 1.463878622836024,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004786775673945436,
      "loss": 5.1172,
      "mean_token_accuracy": 0.18605041205883027,
      "num_tokens": 32640844.0,
      "step": 18815
    },
    {
      "entropy": 5.475657749176025,
      "epoch": 1.4642676522077416,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047866578910282656,
      "loss": 5.1936,
      "mean_token_accuracy": 0.18568813055753708,
      "num_tokens": 32649458.0,
      "step": 18820
    },
    {
      "entropy": 5.419367980957031,
      "epoch": 1.4646566815794593,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047865400772080535,
      "loss": 5.1737,
      "mean_token_accuracy": 0.18384962379932404,
      "num_tokens": 32657649.0,
      "step": 18825
    },
    {
      "entropy": 5.515106058120727,
      "epoch": 1.4650457109511767,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047864222324865875,
      "loss": 5.3076,
      "mean_token_accuracy": 0.17915474474430085,
      "num_tokens": 32665407.0,
      "step": 18830
    },
    {
      "entropy": 5.479231262207032,
      "epoch": 1.4654347403228944,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004786304356865655,
      "loss": 5.1291,
      "mean_token_accuracy": 0.18186154812574387,
      "num_tokens": 32673760.0,
      "step": 18835
    },
    {
      "entropy": 5.412498044967651,
      "epoch": 1.465823769694612,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047861864503470457,
      "loss": 5.159,
      "mean_token_accuracy": 0.179190731048584,
      "num_tokens": 32682118.0,
      "step": 18840
    },
    {
      "entropy": 5.476997184753418,
      "epoch": 1.4662127990663296,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004786068512932547,
      "loss": 5.1874,
      "mean_token_accuracy": 0.18081807792186738,
      "num_tokens": 32690430.0,
      "step": 18845
    },
    {
      "entropy": 5.479546594619751,
      "epoch": 1.466601828438047,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000478595054462395,
      "loss": 5.0338,
      "mean_token_accuracy": 0.18940188735723495,
      "num_tokens": 32698776.0,
      "step": 18850
    },
    {
      "entropy": 5.534380865097046,
      "epoch": 1.4669908578097646,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047858325454230437,
      "loss": 5.2923,
      "mean_token_accuracy": 0.17772379964590074,
      "num_tokens": 32707299.0,
      "step": 18855
    },
    {
      "entropy": 5.395196199417114,
      "epoch": 1.4673798871814823,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004785714515331619,
      "loss": 5.1139,
      "mean_token_accuracy": 0.19416551291942596,
      "num_tokens": 32716041.0,
      "step": 18860
    },
    {
      "entropy": 5.458668422698975,
      "epoch": 1.4677689165531997,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047855964543514666,
      "loss": 5.1505,
      "mean_token_accuracy": 0.18225469440221786,
      "num_tokens": 32724182.0,
      "step": 18865
    },
    {
      "entropy": 5.41873254776001,
      "epoch": 1.4681579459249172,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047854783624843786,
      "loss": 5.0916,
      "mean_token_accuracy": 0.18801273703575133,
      "num_tokens": 32732858.0,
      "step": 18870
    },
    {
      "entropy": 5.531584644317627,
      "epoch": 1.468546975296635,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047853602397321453,
      "loss": 5.243,
      "mean_token_accuracy": 0.17630415856838227,
      "num_tokens": 32741422.0,
      "step": 18875
    },
    {
      "entropy": 5.546292543411255,
      "epoch": 1.4689360046683524,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047852420860965605,
      "loss": 5.2894,
      "mean_token_accuracy": 0.17414565235376359,
      "num_tokens": 32750720.0,
      "step": 18880
    },
    {
      "entropy": 5.398320960998535,
      "epoch": 1.46932503404007,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047851239015794166,
      "loss": 5.1143,
      "mean_token_accuracy": 0.18058869540691375,
      "num_tokens": 32759418.0,
      "step": 18885
    },
    {
      "entropy": 5.424722909927368,
      "epoch": 1.4697140634117876,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047850056861825066,
      "loss": 5.1973,
      "mean_token_accuracy": 0.18210604786872864,
      "num_tokens": 32767604.0,
      "step": 18890
    },
    {
      "entropy": 5.4482746601104735,
      "epoch": 1.4701030927835053,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047848874399076245,
      "loss": 5.2897,
      "mean_token_accuracy": 0.17169092446565629,
      "num_tokens": 32776535.0,
      "step": 18895
    },
    {
      "entropy": 5.553547096252442,
      "epoch": 1.4704921221552227,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004784769162756563,
      "loss": 5.191,
      "mean_token_accuracy": 0.18324317336082457,
      "num_tokens": 32785633.0,
      "step": 18900
    },
    {
      "entropy": 5.507122278213501,
      "epoch": 1.4708811515269402,
      "grad_norm": 1.421875,
      "learning_rate": 0.000478465085473112,
      "loss": 5.1548,
      "mean_token_accuracy": 0.17950839549303055,
      "num_tokens": 32794281.0,
      "step": 18905
    },
    {
      "entropy": 5.486583280563354,
      "epoch": 1.471270180898658,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004784532515833088,
      "loss": 5.3328,
      "mean_token_accuracy": 0.1767190083861351,
      "num_tokens": 32802312.0,
      "step": 18910
    },
    {
      "entropy": 5.4171489715576175,
      "epoch": 1.4716592102703754,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047844141460642636,
      "loss": 5.0919,
      "mean_token_accuracy": 0.18468550145626067,
      "num_tokens": 32810983.0,
      "step": 18915
    },
    {
      "entropy": 5.453736305236816,
      "epoch": 1.4720482396420929,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047842957454264425,
      "loss": 5.1452,
      "mean_token_accuracy": 0.1813129812479019,
      "num_tokens": 32820695.0,
      "step": 18920
    },
    {
      "entropy": 5.460472583770752,
      "epoch": 1.4724372690138106,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047841773139214213,
      "loss": 5.1868,
      "mean_token_accuracy": 0.18236879259347916,
      "num_tokens": 32828483.0,
      "step": 18925
    },
    {
      "entropy": 5.395666885375976,
      "epoch": 1.472826298385528,
      "grad_norm": 1.25,
      "learning_rate": 0.0004784058851550997,
      "loss": 5.1852,
      "mean_token_accuracy": 0.17760971188545227,
      "num_tokens": 32837827.0,
      "step": 18930
    },
    {
      "entropy": 5.533575820922851,
      "epoch": 1.4732153277572457,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004783940358316967,
      "loss": 5.2871,
      "mean_token_accuracy": 0.1733487978577614,
      "num_tokens": 32846872.0,
      "step": 18935
    },
    {
      "entropy": 5.48499870300293,
      "epoch": 1.4736043571289632,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047838218342211296,
      "loss": 5.0974,
      "mean_token_accuracy": 0.19082875400781632,
      "num_tokens": 32855418.0,
      "step": 18940
    },
    {
      "entropy": 5.363893032073975,
      "epoch": 1.473993386500681,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047837032792652837,
      "loss": 5.1696,
      "mean_token_accuracy": 0.1762622758746147,
      "num_tokens": 32864075.0,
      "step": 18945
    },
    {
      "entropy": 5.473411798477173,
      "epoch": 1.4743824158723984,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004783584693451226,
      "loss": 5.2691,
      "mean_token_accuracy": 0.1723635584115982,
      "num_tokens": 32872615.0,
      "step": 18950
    },
    {
      "entropy": 5.560315370559692,
      "epoch": 1.4747714452441159,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004783466076780759,
      "loss": 5.1347,
      "mean_token_accuracy": 0.18082498759031296,
      "num_tokens": 32881313.0,
      "step": 18955
    },
    {
      "entropy": 5.514806890487671,
      "epoch": 1.4751604746158336,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004783347429255679,
      "loss": 5.2078,
      "mean_token_accuracy": 0.17956217676401137,
      "num_tokens": 32889323.0,
      "step": 18960
    },
    {
      "entropy": 5.328566694259644,
      "epoch": 1.475549503987551,
      "grad_norm": 1.21875,
      "learning_rate": 0.000478322875087779,
      "loss": 5.1478,
      "mean_token_accuracy": 0.179301455616951,
      "num_tokens": 32898045.0,
      "step": 18965
    },
    {
      "entropy": 5.512156200408936,
      "epoch": 1.4759385333592685,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047831100416488906,
      "loss": 5.2463,
      "mean_token_accuracy": 0.17141567915678024,
      "num_tokens": 32906646.0,
      "step": 18970
    },
    {
      "entropy": 5.525881338119507,
      "epoch": 1.4763275627309862,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047829913015707816,
      "loss": 5.2329,
      "mean_token_accuracy": 0.17813896387815475,
      "num_tokens": 32915714.0,
      "step": 18975
    },
    {
      "entropy": 5.484026193618774,
      "epoch": 1.4767165921027037,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047828725306452657,
      "loss": 5.2933,
      "mean_token_accuracy": 0.17183134704828262,
      "num_tokens": 32924996.0,
      "step": 18980
    },
    {
      "entropy": 5.473680734634399,
      "epoch": 1.4771056214744214,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047827537288741453,
      "loss": 5.1214,
      "mean_token_accuracy": 0.18252688646316528,
      "num_tokens": 32933550.0,
      "step": 18985
    },
    {
      "entropy": 5.513328599929809,
      "epoch": 1.4774946508461388,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004782634896259222,
      "loss": 5.244,
      "mean_token_accuracy": 0.17530196458101271,
      "num_tokens": 32941800.0,
      "step": 18990
    },
    {
      "entropy": 5.426942634582519,
      "epoch": 1.4778836802178565,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047825160328023,
      "loss": 5.2044,
      "mean_token_accuracy": 0.1841472178697586,
      "num_tokens": 32951674.0,
      "step": 18995
    },
    {
      "entropy": 5.446544122695923,
      "epoch": 1.478272709589574,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004782397138505181,
      "loss": 5.144,
      "mean_token_accuracy": 0.1814299240708351,
      "num_tokens": 32959880.0,
      "step": 19000
    },
    {
      "entropy": 5.4063066959381105,
      "epoch": 1.4786617389612915,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047822782133696715,
      "loss": 5.1423,
      "mean_token_accuracy": 0.18353772163391113,
      "num_tokens": 32968557.0,
      "step": 19005
    },
    {
      "entropy": 5.53695592880249,
      "epoch": 1.4790507683330092,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004782159257397574,
      "loss": 5.2596,
      "mean_token_accuracy": 0.17830706387758255,
      "num_tokens": 32977519.0,
      "step": 19010
    },
    {
      "entropy": 5.558176136016845,
      "epoch": 1.4794397977047267,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047820402705906953,
      "loss": 5.2479,
      "mean_token_accuracy": 0.18025315999984742,
      "num_tokens": 32985917.0,
      "step": 19015
    },
    {
      "entropy": 5.453706359863281,
      "epoch": 1.4798288270764441,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047819212529508394,
      "loss": 5.1557,
      "mean_token_accuracy": 0.18149400651454925,
      "num_tokens": 32994612.0,
      "step": 19020
    },
    {
      "entropy": 5.453710079193115,
      "epoch": 1.4802178564481618,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004781802204479812,
      "loss": 5.2531,
      "mean_token_accuracy": 0.17939510717988014,
      "num_tokens": 33003528.0,
      "step": 19025
    },
    {
      "entropy": 5.400583982467651,
      "epoch": 1.4806068858198795,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004781683125179421,
      "loss": 5.1373,
      "mean_token_accuracy": 0.18666049391031264,
      "num_tokens": 33011685.0,
      "step": 19030
    },
    {
      "entropy": 5.477034711837769,
      "epoch": 1.480995915191597,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004781564015051472,
      "loss": 5.1765,
      "mean_token_accuracy": 0.1856398642063141,
      "num_tokens": 33020313.0,
      "step": 19035
    },
    {
      "entropy": 5.419596862792969,
      "epoch": 1.4813849445633145,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004781444874097772,
      "loss": 5.2078,
      "mean_token_accuracy": 0.1799188733100891,
      "num_tokens": 33029143.0,
      "step": 19040
    },
    {
      "entropy": 5.430913972854614,
      "epoch": 1.4817739739350322,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047813257023201307,
      "loss": 5.1413,
      "mean_token_accuracy": 0.18184129446744918,
      "num_tokens": 33038359.0,
      "step": 19045
    },
    {
      "entropy": 5.393340158462524,
      "epoch": 1.4821630033067497,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004781206499720354,
      "loss": 5.1865,
      "mean_token_accuracy": 0.17605089098215104,
      "num_tokens": 33046914.0,
      "step": 19050
    },
    {
      "entropy": 5.520650196075439,
      "epoch": 1.4825520326784671,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047810872663002523,
      "loss": 5.2769,
      "mean_token_accuracy": 0.17467247396707536,
      "num_tokens": 33054490.0,
      "step": 19055
    },
    {
      "entropy": 5.512266731262207,
      "epoch": 1.4829410620501848,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047809680020616333,
      "loss": 5.325,
      "mean_token_accuracy": 0.1759907066822052,
      "num_tokens": 33064624.0,
      "step": 19060
    },
    {
      "entropy": 5.44154577255249,
      "epoch": 1.4833300914219023,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047808487070063083,
      "loss": 5.0458,
      "mean_token_accuracy": 0.18376300632953643,
      "num_tokens": 33072581.0,
      "step": 19065
    },
    {
      "entropy": 5.43805456161499,
      "epoch": 1.4837191207936198,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004780729381136086,
      "loss": 5.2112,
      "mean_token_accuracy": 0.18010231107473373,
      "num_tokens": 33081696.0,
      "step": 19070
    },
    {
      "entropy": 5.471293306350708,
      "epoch": 1.4841081501653375,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004780610024452778,
      "loss": 5.2007,
      "mean_token_accuracy": 0.17797369211912156,
      "num_tokens": 33090470.0,
      "step": 19075
    },
    {
      "entropy": 5.5131958484649655,
      "epoch": 1.4844971795370552,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047804906369581954,
      "loss": 5.2395,
      "mean_token_accuracy": 0.18039721101522446,
      "num_tokens": 33099006.0,
      "step": 19080
    },
    {
      "entropy": 5.413989305496216,
      "epoch": 1.4848862089087727,
      "grad_norm": 1.25,
      "learning_rate": 0.0004780371218654148,
      "loss": 5.0978,
      "mean_token_accuracy": 0.18693019300699235,
      "num_tokens": 33107715.0,
      "step": 19085
    },
    {
      "entropy": 5.427972745895386,
      "epoch": 1.4852752382804901,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047802517695424496,
      "loss": 5.1309,
      "mean_token_accuracy": 0.18932367414236068,
      "num_tokens": 33116895.0,
      "step": 19090
    },
    {
      "entropy": 5.442896032333374,
      "epoch": 1.4856642676522078,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004780132289624913,
      "loss": 5.1827,
      "mean_token_accuracy": 0.17944522053003312,
      "num_tokens": 33125676.0,
      "step": 19095
    },
    {
      "entropy": 5.450321960449219,
      "epoch": 1.4860532970239253,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004780012778903349,
      "loss": 5.1863,
      "mean_token_accuracy": 0.18306608945131303,
      "num_tokens": 33134870.0,
      "step": 19100
    },
    {
      "entropy": 5.477551651000977,
      "epoch": 1.4864423263956428,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047798932373795724,
      "loss": 5.2545,
      "mean_token_accuracy": 0.17817706018686294,
      "num_tokens": 33144258.0,
      "step": 19105
    },
    {
      "entropy": 5.492812299728394,
      "epoch": 1.4868313557673605,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047797736650553977,
      "loss": 5.1957,
      "mean_token_accuracy": 0.18045669943094253,
      "num_tokens": 33153703.0,
      "step": 19110
    },
    {
      "entropy": 5.51462574005127,
      "epoch": 1.487220385139078,
      "grad_norm": 1.375,
      "learning_rate": 0.00047796540619326366,
      "loss": 5.1943,
      "mean_token_accuracy": 0.17473009675741197,
      "num_tokens": 33163092.0,
      "step": 19115
    },
    {
      "entropy": 5.436188840866089,
      "epoch": 1.4876094145107956,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004779534428013107,
      "loss": 5.1894,
      "mean_token_accuracy": 0.18153058290481566,
      "num_tokens": 33172133.0,
      "step": 19120
    },
    {
      "entropy": 5.412543296813965,
      "epoch": 1.4879984438825131,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047794147632986223,
      "loss": 5.2025,
      "mean_token_accuracy": 0.1763584703207016,
      "num_tokens": 33180629.0,
      "step": 19125
    },
    {
      "entropy": 5.486550521850586,
      "epoch": 1.4883874732542308,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004779295067790999,
      "loss": 5.1547,
      "mean_token_accuracy": 0.18087733536958694,
      "num_tokens": 33189602.0,
      "step": 19130
    },
    {
      "entropy": 5.459705781936646,
      "epoch": 1.4887765026259483,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004779175341492053,
      "loss": 5.1832,
      "mean_token_accuracy": 0.18326559513807297,
      "num_tokens": 33198403.0,
      "step": 19135
    },
    {
      "entropy": 5.3958992004394535,
      "epoch": 1.4891655319976658,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004779055584403601,
      "loss": 5.128,
      "mean_token_accuracy": 0.18013326227664947,
      "num_tokens": 33207131.0,
      "step": 19140
    },
    {
      "entropy": 5.456138324737549,
      "epoch": 1.4895545613693835,
      "grad_norm": 1.328125,
      "learning_rate": 0.000477893579652746,
      "loss": 5.1872,
      "mean_token_accuracy": 0.1814579799771309,
      "num_tokens": 33216143.0,
      "step": 19145
    },
    {
      "entropy": 5.458274841308594,
      "epoch": 1.489943590741101,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047788159778654475,
      "loss": 5.1689,
      "mean_token_accuracy": 0.1896282911300659,
      "num_tokens": 33224558.0,
      "step": 19150
    },
    {
      "entropy": 5.369024419784546,
      "epoch": 1.4903326201128184,
      "grad_norm": 1.25,
      "learning_rate": 0.00047786961284193813,
      "loss": 5.1532,
      "mean_token_accuracy": 0.18241242021322251,
      "num_tokens": 33232737.0,
      "step": 19155
    },
    {
      "entropy": 5.507910490036011,
      "epoch": 1.4907216494845361,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004778576248191081,
      "loss": 5.1792,
      "mean_token_accuracy": 0.17595004737377168,
      "num_tokens": 33240845.0,
      "step": 19160
    },
    {
      "entropy": 5.4341333389282225,
      "epoch": 1.4911106788562536,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004778456337182365,
      "loss": 5.0925,
      "mean_token_accuracy": 0.18541962802410125,
      "num_tokens": 33249101.0,
      "step": 19165
    },
    {
      "entropy": 5.500579833984375,
      "epoch": 1.4914997082279713,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004778336395395052,
      "loss": 5.1879,
      "mean_token_accuracy": 0.18232096880674362,
      "num_tokens": 33258403.0,
      "step": 19170
    },
    {
      "entropy": 5.375124073028564,
      "epoch": 1.4918887375996888,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047782164228309636,
      "loss": 5.1035,
      "mean_token_accuracy": 0.18395027071237563,
      "num_tokens": 33266679.0,
      "step": 19175
    },
    {
      "entropy": 5.409463739395141,
      "epoch": 1.4922777669714065,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047780964194919193,
      "loss": 5.134,
      "mean_token_accuracy": 0.18880752772092818,
      "num_tokens": 33274996.0,
      "step": 19180
    },
    {
      "entropy": 5.5391641616821286,
      "epoch": 1.492666796343124,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00047779763853797387,
      "loss": 5.2379,
      "mean_token_accuracy": 0.18079362362623214,
      "num_tokens": 33282558.0,
      "step": 19185
    },
    {
      "entropy": 5.497524404525757,
      "epoch": 1.4930558257148414,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004777856320496245,
      "loss": 5.2227,
      "mean_token_accuracy": 0.17295868545770646,
      "num_tokens": 33291233.0,
      "step": 19190
    },
    {
      "entropy": 5.438090848922729,
      "epoch": 1.493444855086559,
      "grad_norm": 1.203125,
      "learning_rate": 0.000477773622484326,
      "loss": 5.098,
      "mean_token_accuracy": 0.18422650396823884,
      "num_tokens": 33299821.0,
      "step": 19195
    },
    {
      "entropy": 5.404525709152222,
      "epoch": 1.4938338844582766,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004777616098422604,
      "loss": 5.1795,
      "mean_token_accuracy": 0.18257853090763093,
      "num_tokens": 33309624.0,
      "step": 19200
    },
    {
      "entropy": 5.4740571022033695,
      "epoch": 1.494222913829994,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047774959412361014,
      "loss": 5.2002,
      "mean_token_accuracy": 0.18061966747045516,
      "num_tokens": 33318727.0,
      "step": 19205
    },
    {
      "entropy": 5.54583773612976,
      "epoch": 1.4946119432017118,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004777375753285575,
      "loss": 5.1927,
      "mean_token_accuracy": 0.1786590725183487,
      "num_tokens": 33327219.0,
      "step": 19210
    },
    {
      "entropy": 5.4156818866729735,
      "epoch": 1.4950009725734292,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047772555345728486,
      "loss": 5.1955,
      "mean_token_accuracy": 0.1830497995018959,
      "num_tokens": 33336466.0,
      "step": 19215
    },
    {
      "entropy": 5.54693603515625,
      "epoch": 1.495390001945147,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004777135285099746,
      "loss": 5.2526,
      "mean_token_accuracy": 0.17527057081460953,
      "num_tokens": 33346040.0,
      "step": 19220
    },
    {
      "entropy": 5.553634214401245,
      "epoch": 1.4957790313168644,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004777015004868092,
      "loss": 5.1988,
      "mean_token_accuracy": 0.17920879274606705,
      "num_tokens": 33355476.0,
      "step": 19225
    },
    {
      "entropy": 5.40829062461853,
      "epoch": 1.496168060688582,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004776894693879712,
      "loss": 5.1524,
      "mean_token_accuracy": 0.18685372620821,
      "num_tokens": 33364455.0,
      "step": 19230
    },
    {
      "entropy": 5.409610223770142,
      "epoch": 1.4965570900602996,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004776774352136431,
      "loss": 5.1482,
      "mean_token_accuracy": 0.18166572153568267,
      "num_tokens": 33373129.0,
      "step": 19235
    },
    {
      "entropy": 5.420319700241089,
      "epoch": 1.496946119432017,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047766539796400756,
      "loss": 5.1652,
      "mean_token_accuracy": 0.18454618006944656,
      "num_tokens": 33382775.0,
      "step": 19240
    },
    {
      "entropy": 5.43588433265686,
      "epoch": 1.4973351488037348,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004776533576392471,
      "loss": 5.1656,
      "mean_token_accuracy": 0.1824536368250847,
      "num_tokens": 33391370.0,
      "step": 19245
    },
    {
      "entropy": 5.539198684692383,
      "epoch": 1.4977241781754522,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004776413142395445,
      "loss": 5.1707,
      "mean_token_accuracy": 0.18307846039533615,
      "num_tokens": 33399832.0,
      "step": 19250
    },
    {
      "entropy": 5.464285612106323,
      "epoch": 1.4981132075471697,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047762926776508244,
      "loss": 5.201,
      "mean_token_accuracy": 0.18457026928663253,
      "num_tokens": 33408709.0,
      "step": 19255
    },
    {
      "entropy": 5.391808080673218,
      "epoch": 1.4985022369188874,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047761721821604387,
      "loss": 5.1984,
      "mean_token_accuracy": 0.1779489278793335,
      "num_tokens": 33416983.0,
      "step": 19260
    },
    {
      "entropy": 5.449897718429566,
      "epoch": 1.4988912662906049,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004776051655926115,
      "loss": 5.223,
      "mean_token_accuracy": 0.18205749988555908,
      "num_tokens": 33426692.0,
      "step": 19265
    },
    {
      "entropy": 5.504791641235352,
      "epoch": 1.4992802956623226,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047759310989496813,
      "loss": 5.2082,
      "mean_token_accuracy": 0.1874220997095108,
      "num_tokens": 33434895.0,
      "step": 19270
    },
    {
      "entropy": 5.426663589477539,
      "epoch": 1.49966932503404,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004775810511232969,
      "loss": 5.1128,
      "mean_token_accuracy": 0.18144633769989013,
      "num_tokens": 33443378.0,
      "step": 19275
    },
    {
      "entropy": 5.527096891403199,
      "epoch": 1.5000583544057577,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047756898927778056,
      "loss": 5.2556,
      "mean_token_accuracy": 0.1791997507214546,
      "num_tokens": 33452753.0,
      "step": 19280
    },
    {
      "entropy": 5.563938999176026,
      "epoch": 1.5004473837774752,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047755692435860227,
      "loss": 5.2358,
      "mean_token_accuracy": 0.17396415621042252,
      "num_tokens": 33461625.0,
      "step": 19285
    },
    {
      "entropy": 5.435961532592773,
      "epoch": 1.5008364131491927,
      "grad_norm": 1.296875,
      "learning_rate": 0.000477544856365945,
      "loss": 5.1602,
      "mean_token_accuracy": 0.18510159403085708,
      "num_tokens": 33471228.0,
      "step": 19290
    },
    {
      "entropy": 5.383442306518555,
      "epoch": 1.5012254425209104,
      "grad_norm": 1.375,
      "learning_rate": 0.00047753278529999205,
      "loss": 5.1708,
      "mean_token_accuracy": 0.18318298012018203,
      "num_tokens": 33480137.0,
      "step": 19295
    },
    {
      "entropy": 5.445921945571899,
      "epoch": 1.5016144718926279,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047752071116092637,
      "loss": 5.1714,
      "mean_token_accuracy": 0.18468228429555894,
      "num_tokens": 33488860.0,
      "step": 19300
    },
    {
      "entropy": 5.4348413944244385,
      "epoch": 1.5020035012643453,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004775086339489312,
      "loss": 5.1373,
      "mean_token_accuracy": 0.17986771166324617,
      "num_tokens": 33497625.0,
      "step": 19305
    },
    {
      "entropy": 5.47045865058899,
      "epoch": 1.502392530636063,
      "grad_norm": 1.375,
      "learning_rate": 0.0004774965536641899,
      "loss": 5.3012,
      "mean_token_accuracy": 0.16962795704603195,
      "num_tokens": 33506422.0,
      "step": 19310
    },
    {
      "entropy": 5.417447757720947,
      "epoch": 1.5027815600077807,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047748447030688575,
      "loss": 5.0787,
      "mean_token_accuracy": 0.18490252494812012,
      "num_tokens": 33515638.0,
      "step": 19315
    },
    {
      "entropy": 5.4492974281311035,
      "epoch": 1.503170589379498,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047747238387720194,
      "loss": 5.0926,
      "mean_token_accuracy": 0.18031395375728607,
      "num_tokens": 33524287.0,
      "step": 19320
    },
    {
      "entropy": 5.410359287261963,
      "epoch": 1.5035596187512157,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000477460294375322,
      "loss": 5.1151,
      "mean_token_accuracy": 0.17429952174425126,
      "num_tokens": 33532579.0,
      "step": 19325
    },
    {
      "entropy": 5.396300315856934,
      "epoch": 1.5039486481229334,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047744820180142935,
      "loss": 5.1861,
      "mean_token_accuracy": 0.17657063901424408,
      "num_tokens": 33540816.0,
      "step": 19330
    },
    {
      "entropy": 5.33559365272522,
      "epoch": 1.5043376774946509,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047743610615570746,
      "loss": 5.0516,
      "mean_token_accuracy": 0.19526432305574418,
      "num_tokens": 33549383.0,
      "step": 19335
    },
    {
      "entropy": 5.362539386749267,
      "epoch": 1.5047267068663683,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047742400743833993,
      "loss": 5.1149,
      "mean_token_accuracy": 0.18991010636091232,
      "num_tokens": 33557457.0,
      "step": 19340
    },
    {
      "entropy": 5.483785581588745,
      "epoch": 1.505115736238086,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004774119056495102,
      "loss": 5.219,
      "mean_token_accuracy": 0.1809105783700943,
      "num_tokens": 33565479.0,
      "step": 19345
    },
    {
      "entropy": 5.441138839721679,
      "epoch": 1.5055047656098035,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047739980078940206,
      "loss": 5.1926,
      "mean_token_accuracy": 0.17683461606502532,
      "num_tokens": 33574691.0,
      "step": 19350
    },
    {
      "entropy": 5.3751363277435305,
      "epoch": 1.505893794981521,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047738769285819894,
      "loss": 5.0432,
      "mean_token_accuracy": 0.19726763367652894,
      "num_tokens": 33582360.0,
      "step": 19355
    },
    {
      "entropy": 5.381725931167603,
      "epoch": 1.5062828243532387,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004773755818560849,
      "loss": 5.1495,
      "mean_token_accuracy": 0.190781632065773,
      "num_tokens": 33590977.0,
      "step": 19360
    },
    {
      "entropy": 5.394251489639283,
      "epoch": 1.5066718537249564,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004773634677832434,
      "loss": 5.1314,
      "mean_token_accuracy": 0.18340077847242356,
      "num_tokens": 33600227.0,
      "step": 19365
    },
    {
      "entropy": 5.519790363311768,
      "epoch": 1.5070608830966739,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004773513506398584,
      "loss": 5.1771,
      "mean_token_accuracy": 0.18774376511573793,
      "num_tokens": 33608408.0,
      "step": 19370
    },
    {
      "entropy": 5.4828001976013185,
      "epoch": 1.5074499124683913,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004773392304261137,
      "loss": 5.1068,
      "mean_token_accuracy": 0.18125121295452118,
      "num_tokens": 33616892.0,
      "step": 19375
    },
    {
      "entropy": 5.425474214553833,
      "epoch": 1.507838941840109,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004773271071421933,
      "loss": 5.2209,
      "mean_token_accuracy": 0.183016075193882,
      "num_tokens": 33626090.0,
      "step": 19380
    },
    {
      "entropy": 5.321277713775634,
      "epoch": 1.5082279712118265,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047731498078828105,
      "loss": 5.0533,
      "mean_token_accuracy": 0.19009934961795807,
      "num_tokens": 33634855.0,
      "step": 19385
    },
    {
      "entropy": 5.447555732727051,
      "epoch": 1.508617000583544,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000477302851364561,
      "loss": 5.2121,
      "mean_token_accuracy": 0.1811250776052475,
      "num_tokens": 33643425.0,
      "step": 19390
    },
    {
      "entropy": 5.559801435470581,
      "epoch": 1.5090060299552617,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047729071887121717,
      "loss": 5.3185,
      "mean_token_accuracy": 0.17334826290607452,
      "num_tokens": 33652926.0,
      "step": 19395
    },
    {
      "entropy": 5.5478527545928955,
      "epoch": 1.5093950593269791,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004772785833084337,
      "loss": 5.2185,
      "mean_token_accuracy": 0.1802562028169632,
      "num_tokens": 33661279.0,
      "step": 19400
    },
    {
      "entropy": 5.440386056900024,
      "epoch": 1.5097840886986966,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004772664446763946,
      "loss": 5.0925,
      "mean_token_accuracy": 0.18016327023506165,
      "num_tokens": 33669717.0,
      "step": 19405
    },
    {
      "entropy": 5.461114072799683,
      "epoch": 1.5101731180704143,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004772543029752842,
      "loss": 5.226,
      "mean_token_accuracy": 0.17862799316644667,
      "num_tokens": 33679141.0,
      "step": 19410
    },
    {
      "entropy": 5.397663211822509,
      "epoch": 1.510562147442132,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047724215820528666,
      "loss": 5.0457,
      "mean_token_accuracy": 0.18993809521198274,
      "num_tokens": 33686867.0,
      "step": 19415
    },
    {
      "entropy": 5.443642902374267,
      "epoch": 1.5109511768138495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004772300103665863,
      "loss": 5.1624,
      "mean_token_accuracy": 0.18685533255338668,
      "num_tokens": 33695073.0,
      "step": 19420
    },
    {
      "entropy": 5.469837427139282,
      "epoch": 1.511340206185567,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047721785945936733,
      "loss": 5.2347,
      "mean_token_accuracy": 0.17580808699131012,
      "num_tokens": 33703212.0,
      "step": 19425
    },
    {
      "entropy": 5.475648832321167,
      "epoch": 1.5117292355572847,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004772057054838143,
      "loss": 5.1733,
      "mean_token_accuracy": 0.1816648557782173,
      "num_tokens": 33711487.0,
      "step": 19430
    },
    {
      "entropy": 5.452206707000732,
      "epoch": 1.5121182649290021,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047719354844011146,
      "loss": 5.2552,
      "mean_token_accuracy": 0.1790287971496582,
      "num_tokens": 33719642.0,
      "step": 19435
    },
    {
      "entropy": 5.405749034881592,
      "epoch": 1.5125072943007196,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004771813883284434,
      "loss": 5.1556,
      "mean_token_accuracy": 0.18315218836069108,
      "num_tokens": 33728167.0,
      "step": 19440
    },
    {
      "entropy": 5.399593496322632,
      "epoch": 1.5128963236724373,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047716922514899455,
      "loss": 5.1622,
      "mean_token_accuracy": 0.18719623982906342,
      "num_tokens": 33736094.0,
      "step": 19445
    },
    {
      "entropy": 5.404392099380493,
      "epoch": 1.5132853530441548,
      "grad_norm": 1.25,
      "learning_rate": 0.00047715705890194953,
      "loss": 5.0369,
      "mean_token_accuracy": 0.1970661148428917,
      "num_tokens": 33744416.0,
      "step": 19450
    },
    {
      "entropy": 5.48582911491394,
      "epoch": 1.5136743824158723,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047714488958749285,
      "loss": 5.1575,
      "mean_token_accuracy": 0.18625634163618088,
      "num_tokens": 33752873.0,
      "step": 19455
    },
    {
      "entropy": 5.399388790130615,
      "epoch": 1.51406341178759,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047713271720580924,
      "loss": 5.1483,
      "mean_token_accuracy": 0.1816815882921219,
      "num_tokens": 33761604.0,
      "step": 19460
    },
    {
      "entropy": 5.442336988449097,
      "epoch": 1.5144524411593077,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004771205417570834,
      "loss": 5.1857,
      "mean_token_accuracy": 0.1827282965183258,
      "num_tokens": 33770562.0,
      "step": 19465
    },
    {
      "entropy": 5.453472042083741,
      "epoch": 1.5148414705310251,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047710836324150004,
      "loss": 5.1307,
      "mean_token_accuracy": 0.18890078365802765,
      "num_tokens": 33779037.0,
      "step": 19470
    },
    {
      "entropy": 5.463383579254151,
      "epoch": 1.5152304999027426,
      "grad_norm": 1.21875,
      "learning_rate": 0.000477096181659244,
      "loss": 5.1802,
      "mean_token_accuracy": 0.18278009295463563,
      "num_tokens": 33787949.0,
      "step": 19475
    },
    {
      "entropy": 5.399554014205933,
      "epoch": 1.5156195292744603,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004770839970105,
      "loss": 5.1132,
      "mean_token_accuracy": 0.1884308248758316,
      "num_tokens": 33796783.0,
      "step": 19480
    },
    {
      "entropy": 5.425920581817627,
      "epoch": 1.5160085586461778,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047707180929545295,
      "loss": 5.1513,
      "mean_token_accuracy": 0.1831379532814026,
      "num_tokens": 33804975.0,
      "step": 19485
    },
    {
      "entropy": 5.382816171646118,
      "epoch": 1.5163975880178953,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047705961851428786,
      "loss": 5.1643,
      "mean_token_accuracy": 0.18120018541812896,
      "num_tokens": 33813355.0,
      "step": 19490
    },
    {
      "entropy": 5.425685691833496,
      "epoch": 1.516786617389613,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047704742466718973,
      "loss": 5.1325,
      "mean_token_accuracy": 0.19100488424301149,
      "num_tokens": 33821710.0,
      "step": 19495
    },
    {
      "entropy": 5.509184217453003,
      "epoch": 1.5171756467613304,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047703522775434354,
      "loss": 5.2833,
      "mean_token_accuracy": 0.17496250718832015,
      "num_tokens": 33830674.0,
      "step": 19500
    },
    {
      "entropy": 5.373120069503784,
      "epoch": 1.517564676133048,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047702302777593425,
      "loss": 5.0991,
      "mean_token_accuracy": 0.18148758709430696,
      "num_tokens": 33838963.0,
      "step": 19505
    },
    {
      "entropy": 5.513600254058838,
      "epoch": 1.5179537055047656,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047701082473214715,
      "loss": 5.2376,
      "mean_token_accuracy": 0.18190186768770217,
      "num_tokens": 33847728.0,
      "step": 19510
    },
    {
      "entropy": 5.469587755203247,
      "epoch": 1.5183427348764833,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047699861862316725,
      "loss": 5.2198,
      "mean_token_accuracy": 0.18141587376594542,
      "num_tokens": 33856170.0,
      "step": 19515
    },
    {
      "entropy": 5.441420984268189,
      "epoch": 1.5187317642482008,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004769864094491798,
      "loss": 5.2012,
      "mean_token_accuracy": 0.18403111398220062,
      "num_tokens": 33865531.0,
      "step": 19520
    },
    {
      "entropy": 5.471396160125733,
      "epoch": 1.5191207936199183,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004769741972103702,
      "loss": 5.1167,
      "mean_token_accuracy": 0.17649407535791398,
      "num_tokens": 33873976.0,
      "step": 19525
    },
    {
      "entropy": 5.397882699966431,
      "epoch": 1.519509822991636,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047696198190692353,
      "loss": 5.0092,
      "mean_token_accuracy": 0.1983886942267418,
      "num_tokens": 33881884.0,
      "step": 19530
    },
    {
      "entropy": 5.37314453125,
      "epoch": 1.5198988523633534,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004769497635390253,
      "loss": 5.2311,
      "mean_token_accuracy": 0.1836245357990265,
      "num_tokens": 33890932.0,
      "step": 19535
    },
    {
      "entropy": 5.453801870346069,
      "epoch": 1.520287881735071,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004769375421068608,
      "loss": 5.1796,
      "mean_token_accuracy": 0.1802298069000244,
      "num_tokens": 33899271.0,
      "step": 19540
    },
    {
      "entropy": 5.491240930557251,
      "epoch": 1.5206769111067886,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047692531761061555,
      "loss": 5.2109,
      "mean_token_accuracy": 0.1750801056623459,
      "num_tokens": 33908776.0,
      "step": 19545
    },
    {
      "entropy": 5.45037260055542,
      "epoch": 1.521065940478506,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000476913090050475,
      "loss": 5.1042,
      "mean_token_accuracy": 0.19254542738199235,
      "num_tokens": 33917424.0,
      "step": 19550
    },
    {
      "entropy": 5.495787191390991,
      "epoch": 1.5214549698502235,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047690085942662464,
      "loss": 5.1851,
      "mean_token_accuracy": 0.1823791444301605,
      "num_tokens": 33925948.0,
      "step": 19555
    },
    {
      "entropy": 5.45155701637268,
      "epoch": 1.5218439992219412,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047688862573925015,
      "loss": 5.2078,
      "mean_token_accuracy": 0.18373111486434937,
      "num_tokens": 33934623.0,
      "step": 19560
    },
    {
      "entropy": 5.438210391998291,
      "epoch": 1.522233028593659,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047687638898853707,
      "loss": 5.2098,
      "mean_token_accuracy": 0.17873280346393586,
      "num_tokens": 33943871.0,
      "step": 19565
    },
    {
      "entropy": 5.529087877273559,
      "epoch": 1.5226220579653764,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004768641491746711,
      "loss": 5.1751,
      "mean_token_accuracy": 0.18006608337163926,
      "num_tokens": 33953032.0,
      "step": 19570
    },
    {
      "entropy": 5.458274030685425,
      "epoch": 1.523011087337094,
      "grad_norm": 1.296875,
      "learning_rate": 0.000476851906297838,
      "loss": 5.222,
      "mean_token_accuracy": 0.1831106424331665,
      "num_tokens": 33961702.0,
      "step": 19575
    },
    {
      "entropy": 5.461389684677124,
      "epoch": 1.5234001167088116,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047683966035822346,
      "loss": 5.2361,
      "mean_token_accuracy": 0.17599067091941833,
      "num_tokens": 33970996.0,
      "step": 19580
    },
    {
      "entropy": 5.499028825759888,
      "epoch": 1.523789146080529,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047682741135601336,
      "loss": 5.2013,
      "mean_token_accuracy": 0.18630655109882355,
      "num_tokens": 33980445.0,
      "step": 19585
    },
    {
      "entropy": 5.5005824089050295,
      "epoch": 1.5241781754522465,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047681515929139356,
      "loss": 5.1412,
      "mean_token_accuracy": 0.18286774307489395,
      "num_tokens": 33989042.0,
      "step": 19590
    },
    {
      "entropy": 5.328094720840454,
      "epoch": 1.5245672048239642,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047680290416454995,
      "loss": 5.0114,
      "mean_token_accuracy": 0.19995274096727372,
      "num_tokens": 33998162.0,
      "step": 19595
    },
    {
      "entropy": 5.400707817077636,
      "epoch": 1.524956234195682,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004767906459756684,
      "loss": 5.1598,
      "mean_token_accuracy": 0.1828813686966896,
      "num_tokens": 34007480.0,
      "step": 19600
    },
    {
      "entropy": 5.46451210975647,
      "epoch": 1.5253452635673992,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047677838472493504,
      "loss": 5.1313,
      "mean_token_accuracy": 0.1834305003285408,
      "num_tokens": 34015605.0,
      "step": 19605
    },
    {
      "entropy": 5.527031946182251,
      "epoch": 1.5257342929391169,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004767661204125358,
      "loss": 5.2896,
      "mean_token_accuracy": 0.17471335381269454,
      "num_tokens": 34024847.0,
      "step": 19610
    },
    {
      "entropy": 5.451772022247314,
      "epoch": 1.5261233223108346,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004767538530386569,
      "loss": 5.1602,
      "mean_token_accuracy": 0.17560100704431533,
      "num_tokens": 34034205.0,
      "step": 19615
    },
    {
      "entropy": 5.4462199211120605,
      "epoch": 1.526512351682552,
      "grad_norm": 1.25,
      "learning_rate": 0.00047674158260348437,
      "loss": 5.0656,
      "mean_token_accuracy": 0.18876087069511413,
      "num_tokens": 34042578.0,
      "step": 19620
    },
    {
      "entropy": 5.3750834465026855,
      "epoch": 1.5269013810542695,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047672930910720436,
      "loss": 5.1568,
      "mean_token_accuracy": 0.18691648840904235,
      "num_tokens": 34051699.0,
      "step": 19625
    },
    {
      "entropy": 5.342703533172608,
      "epoch": 1.5272904104259872,
      "grad_norm": 1.375,
      "learning_rate": 0.00047671703255000326,
      "loss": 5.1695,
      "mean_token_accuracy": 0.17989095598459243,
      "num_tokens": 34060872.0,
      "step": 19630
    },
    {
      "entropy": 5.452771139144898,
      "epoch": 1.5276794397977047,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004767047529320673,
      "loss": 5.2144,
      "mean_token_accuracy": 0.17854063212871552,
      "num_tokens": 34069501.0,
      "step": 19635
    },
    {
      "entropy": 5.485051584243775,
      "epoch": 1.5280684691694222,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047669247025358257,
      "loss": 5.241,
      "mean_token_accuracy": 0.17649343162775039,
      "num_tokens": 34078573.0,
      "step": 19640
    },
    {
      "entropy": 5.4484395503997805,
      "epoch": 1.5284574985411399,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047668018451473584,
      "loss": 5.1907,
      "mean_token_accuracy": 0.18057778775691985,
      "num_tokens": 34087270.0,
      "step": 19645
    },
    {
      "entropy": 5.461319065093994,
      "epoch": 1.5288465279128576,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004766678957157132,
      "loss": 5.2298,
      "mean_token_accuracy": 0.18080998659133912,
      "num_tokens": 34096205.0,
      "step": 19650
    },
    {
      "entropy": 5.434656763076783,
      "epoch": 1.5292355572845748,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004766556038567013,
      "loss": 5.0366,
      "mean_token_accuracy": 0.18941207081079484,
      "num_tokens": 34104736.0,
      "step": 19655
    },
    {
      "entropy": 5.455489540100098,
      "epoch": 1.5296245866562925,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004766433089378865,
      "loss": 5.153,
      "mean_token_accuracy": 0.18687899112701417,
      "num_tokens": 34113420.0,
      "step": 19660
    },
    {
      "entropy": 5.4722803115844725,
      "epoch": 1.5300136160280102,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047663101095945544,
      "loss": 5.2563,
      "mean_token_accuracy": 0.17361120730638505,
      "num_tokens": 34121702.0,
      "step": 19665
    },
    {
      "entropy": 5.4608762741088865,
      "epoch": 1.5304026453997277,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047661870992159476,
      "loss": 5.1425,
      "mean_token_accuracy": 0.18361746221780778,
      "num_tokens": 34129657.0,
      "step": 19670
    },
    {
      "entropy": 5.487381315231323,
      "epoch": 1.5307916747714452,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047660640582449106,
      "loss": 5.2817,
      "mean_token_accuracy": 0.1790304586291313,
      "num_tokens": 34138078.0,
      "step": 19675
    },
    {
      "entropy": 5.501763153076172,
      "epoch": 1.5311807041431629,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047659409866833104,
      "loss": 5.1481,
      "mean_token_accuracy": 0.18649304956197738,
      "num_tokens": 34146402.0,
      "step": 19680
    },
    {
      "entropy": 5.505959177017212,
      "epoch": 1.5315697335148803,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004765817884533014,
      "loss": 5.2802,
      "mean_token_accuracy": 0.18149096220731736,
      "num_tokens": 34155909.0,
      "step": 19685
    },
    {
      "entropy": 5.36584587097168,
      "epoch": 1.5319587628865978,
      "grad_norm": 1.34375,
      "learning_rate": 0.000476569475179589,
      "loss": 5.1325,
      "mean_token_accuracy": 0.18313600718975068,
      "num_tokens": 34164598.0,
      "step": 19690
    },
    {
      "entropy": 5.441551733016968,
      "epoch": 1.5323477922583155,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047655715884738074,
      "loss": 5.1855,
      "mean_token_accuracy": 0.17998702377080916,
      "num_tokens": 34173476.0,
      "step": 19695
    },
    {
      "entropy": 5.476489162445068,
      "epoch": 1.5327368216300332,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004765448394568632,
      "loss": 5.203,
      "mean_token_accuracy": 0.17815272957086564,
      "num_tokens": 34181986.0,
      "step": 19700
    },
    {
      "entropy": 5.4478555679321286,
      "epoch": 1.5331258510017505,
      "grad_norm": 1.25,
      "learning_rate": 0.0004765325170082237,
      "loss": 5.1691,
      "mean_token_accuracy": 0.18304586559534072,
      "num_tokens": 34190477.0,
      "step": 19705
    },
    {
      "entropy": 5.3794825077056885,
      "epoch": 1.5335148803734682,
      "grad_norm": 1.1875,
      "learning_rate": 0.000476520191501649,
      "loss": 5.0784,
      "mean_token_accuracy": 0.18856527805328369,
      "num_tokens": 34198955.0,
      "step": 19710
    },
    {
      "entropy": 5.509455633163452,
      "epoch": 1.5339039097451859,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047650786293732607,
      "loss": 5.276,
      "mean_token_accuracy": 0.17426880300045014,
      "num_tokens": 34207744.0,
      "step": 19715
    },
    {
      "entropy": 5.448606967926025,
      "epoch": 1.5342929391169033,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004764955313154421,
      "loss": 5.1697,
      "mean_token_accuracy": 0.18940844982862473,
      "num_tokens": 34215935.0,
      "step": 19720
    },
    {
      "entropy": 5.43552598953247,
      "epoch": 1.5346819684886208,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047648319663618415,
      "loss": 5.1999,
      "mean_token_accuracy": 0.1778809979557991,
      "num_tokens": 34224990.0,
      "step": 19725
    },
    {
      "entropy": 5.414816236495971,
      "epoch": 1.5350709978603385,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047647085889973936,
      "loss": 5.0955,
      "mean_token_accuracy": 0.18890078812837602,
      "num_tokens": 34232928.0,
      "step": 19730
    },
    {
      "entropy": 5.399786472320557,
      "epoch": 1.535460027232056,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047645851810629503,
      "loss": 5.0619,
      "mean_token_accuracy": 0.1924973323941231,
      "num_tokens": 34241434.0,
      "step": 19735
    },
    {
      "entropy": 5.37187647819519,
      "epoch": 1.5358490566037735,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047644617425603825,
      "loss": 5.1143,
      "mean_token_accuracy": 0.19306736886501313,
      "num_tokens": 34250067.0,
      "step": 19740
    },
    {
      "entropy": 5.4581928730010985,
      "epoch": 1.5362380859754912,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004764338273491565,
      "loss": 5.1849,
      "mean_token_accuracy": 0.18465080708265305,
      "num_tokens": 34258656.0,
      "step": 19745
    },
    {
      "entropy": 5.490884256362915,
      "epoch": 1.5366271153472089,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047642147738583695,
      "loss": 5.2826,
      "mean_token_accuracy": 0.16672633737325668,
      "num_tokens": 34267417.0,
      "step": 19750
    },
    {
      "entropy": 5.442201137542725,
      "epoch": 1.5370161447189261,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004764091243662671,
      "loss": 5.0453,
      "mean_token_accuracy": 0.1949620544910431,
      "num_tokens": 34275334.0,
      "step": 19755
    },
    {
      "entropy": 5.398856449127197,
      "epoch": 1.5374051740906438,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047639676829063437,
      "loss": 5.2005,
      "mean_token_accuracy": 0.17791488021612167,
      "num_tokens": 34284016.0,
      "step": 19760
    },
    {
      "entropy": 5.328921747207642,
      "epoch": 1.5377942034623615,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047638440915912623,
      "loss": 5.1139,
      "mean_token_accuracy": 0.1966216668486595,
      "num_tokens": 34292747.0,
      "step": 19765
    },
    {
      "entropy": 5.40814995765686,
      "epoch": 1.538183232834079,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004763720469719303,
      "loss": 5.0821,
      "mean_token_accuracy": 0.1847347602248192,
      "num_tokens": 34301764.0,
      "step": 19770
    },
    {
      "entropy": 5.425303554534912,
      "epoch": 1.5385722622057965,
      "grad_norm": 1.203125,
      "learning_rate": 0.000476359681729234,
      "loss": 5.1929,
      "mean_token_accuracy": 0.1850317418575287,
      "num_tokens": 34310522.0,
      "step": 19775
    },
    {
      "entropy": 5.503285694122314,
      "epoch": 1.5389612915775142,
      "grad_norm": 1.25,
      "learning_rate": 0.000476347313431225,
      "loss": 5.1842,
      "mean_token_accuracy": 0.18381678760051728,
      "num_tokens": 34319342.0,
      "step": 19780
    },
    {
      "entropy": 5.523685169219971,
      "epoch": 1.5393503209492316,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047633494207809096,
      "loss": 5.2927,
      "mean_token_accuracy": 0.17465928494930266,
      "num_tokens": 34327468.0,
      "step": 19785
    },
    {
      "entropy": 5.4529660701751705,
      "epoch": 1.539739350320949,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047632256767001977,
      "loss": 5.2357,
      "mean_token_accuracy": 0.1800572782754898,
      "num_tokens": 34336484.0,
      "step": 19790
    },
    {
      "entropy": 5.388904094696045,
      "epoch": 1.5401283796926668,
      "grad_norm": 1.265625,
      "learning_rate": 0.000476310190207199,
      "loss": 5.0724,
      "mean_token_accuracy": 0.19283018112182618,
      "num_tokens": 34344708.0,
      "step": 19795
    },
    {
      "entropy": 5.4474772930145265,
      "epoch": 1.5405174090643845,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047629780968981653,
      "loss": 5.1197,
      "mean_token_accuracy": 0.18426511138677598,
      "num_tokens": 34352757.0,
      "step": 19800
    },
    {
      "entropy": 5.435916900634766,
      "epoch": 1.540906438436102,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047628542611806007,
      "loss": 5.126,
      "mean_token_accuracy": 0.18959590047597885,
      "num_tokens": 34362070.0,
      "step": 19805
    },
    {
      "entropy": 5.416981363296509,
      "epoch": 1.5412954678078195,
      "grad_norm": 1.375,
      "learning_rate": 0.00047627303949211773,
      "loss": 5.1412,
      "mean_token_accuracy": 0.180282923579216,
      "num_tokens": 34370136.0,
      "step": 19810
    },
    {
      "entropy": 5.3737874031066895,
      "epoch": 1.5416844971795371,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004762606498121774,
      "loss": 5.1099,
      "mean_token_accuracy": 0.18142778277397156,
      "num_tokens": 34378534.0,
      "step": 19815
    },
    {
      "entropy": 5.362152528762818,
      "epoch": 1.5420735265512546,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000476248257078427,
      "loss": 5.1066,
      "mean_token_accuracy": 0.18241876512765884,
      "num_tokens": 34386313.0,
      "step": 19820
    },
    {
      "entropy": 5.409736442565918,
      "epoch": 1.542462555922972,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004762358612910547,
      "loss": 5.1521,
      "mean_token_accuracy": 0.18856621384620667,
      "num_tokens": 34394566.0,
      "step": 19825
    },
    {
      "entropy": 5.411107873916626,
      "epoch": 1.5428515852946898,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004762234624502484,
      "loss": 5.0341,
      "mean_token_accuracy": 0.18770351260900497,
      "num_tokens": 34403177.0,
      "step": 19830
    },
    {
      "entropy": 5.483054685592651,
      "epoch": 1.5432406146664073,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047621106055619644,
      "loss": 5.184,
      "mean_token_accuracy": 0.18565118610858916,
      "num_tokens": 34411577.0,
      "step": 19835
    },
    {
      "entropy": 5.401013374328613,
      "epoch": 1.5436296440381247,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047619865560908687,
      "loss": 5.1923,
      "mean_token_accuracy": 0.17836122810840607,
      "num_tokens": 34420649.0,
      "step": 19840
    },
    {
      "entropy": 5.428129577636719,
      "epoch": 1.5440186734098424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004761862476091079,
      "loss": 5.0931,
      "mean_token_accuracy": 0.189253968000412,
      "num_tokens": 34429778.0,
      "step": 19845
    },
    {
      "entropy": 5.47112135887146,
      "epoch": 1.5444077027815601,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047617383655644785,
      "loss": 5.1899,
      "mean_token_accuracy": 0.1845293790102005,
      "num_tokens": 34438412.0,
      "step": 19850
    },
    {
      "entropy": 5.460814428329468,
      "epoch": 1.5447967321532776,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004761614224512951,
      "loss": 5.1738,
      "mean_token_accuracy": 0.1763644114136696,
      "num_tokens": 34447022.0,
      "step": 19855
    },
    {
      "entropy": 5.484309959411621,
      "epoch": 1.545185761524995,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004761490052938379,
      "loss": 5.207,
      "mean_token_accuracy": 0.17770828157663346,
      "num_tokens": 34455502.0,
      "step": 19860
    },
    {
      "entropy": 5.433563613891602,
      "epoch": 1.5455747908967128,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004761365850842646,
      "loss": 5.1467,
      "mean_token_accuracy": 0.18777014166116715,
      "num_tokens": 34463682.0,
      "step": 19865
    },
    {
      "entropy": 5.417294025421143,
      "epoch": 1.5459638202684303,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004761241618227639,
      "loss": 5.0848,
      "mean_token_accuracy": 0.1871474951505661,
      "num_tokens": 34471678.0,
      "step": 19870
    },
    {
      "entropy": 5.368842029571534,
      "epoch": 1.5463528496401477,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047611173550952414,
      "loss": 5.1018,
      "mean_token_accuracy": 0.18713496774435043,
      "num_tokens": 34479830.0,
      "step": 19875
    },
    {
      "entropy": 5.428650999069214,
      "epoch": 1.5467418790118654,
      "grad_norm": 1.25,
      "learning_rate": 0.00047609930614473387,
      "loss": 5.1739,
      "mean_token_accuracy": 0.17468727827072145,
      "num_tokens": 34487445.0,
      "step": 19880
    },
    {
      "entropy": 5.405205774307251,
      "epoch": 1.547130908383583,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047608687372858175,
      "loss": 5.0588,
      "mean_token_accuracy": 0.18379866927862168,
      "num_tokens": 34496306.0,
      "step": 19885
    },
    {
      "entropy": 5.404586219787598,
      "epoch": 1.5475199377553004,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047607443826125633,
      "loss": 5.1784,
      "mean_token_accuracy": 0.18396103680133818,
      "num_tokens": 34505067.0,
      "step": 19890
    },
    {
      "entropy": 5.537490177154541,
      "epoch": 1.547908967127018,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047606199974294637,
      "loss": 5.2712,
      "mean_token_accuracy": 0.17477076947689058,
      "num_tokens": 34513947.0,
      "step": 19895
    },
    {
      "entropy": 5.5512245178222654,
      "epoch": 1.5482979964987358,
      "grad_norm": 1.375,
      "learning_rate": 0.0004760495581738406,
      "loss": 5.2911,
      "mean_token_accuracy": 0.17187391370534896,
      "num_tokens": 34523383.0,
      "step": 19900
    },
    {
      "entropy": 5.463081884384155,
      "epoch": 1.5486870258704533,
      "grad_norm": 1.25,
      "learning_rate": 0.0004760371135541278,
      "loss": 5.0937,
      "mean_token_accuracy": 0.18152161091566085,
      "num_tokens": 34532376.0,
      "step": 19905
    },
    {
      "entropy": 5.5154829025268555,
      "epoch": 1.5490760552421707,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004760246658839967,
      "loss": 5.2597,
      "mean_token_accuracy": 0.17626759111881257,
      "num_tokens": 34541075.0,
      "step": 19910
    },
    {
      "entropy": 5.36005277633667,
      "epoch": 1.5494650846138884,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004760122151636364,
      "loss": 5.0748,
      "mean_token_accuracy": 0.19100635647773742,
      "num_tokens": 34550296.0,
      "step": 19915
    },
    {
      "entropy": 5.41638069152832,
      "epoch": 1.549854113985606,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004759997613932356,
      "loss": 5.2195,
      "mean_token_accuracy": 0.18080566078424454,
      "num_tokens": 34559185.0,
      "step": 19920
    },
    {
      "entropy": 5.521035480499267,
      "epoch": 1.5502431433573234,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047598730457298335,
      "loss": 5.2034,
      "mean_token_accuracy": 0.18026041239500046,
      "num_tokens": 34568018.0,
      "step": 19925
    },
    {
      "entropy": 5.529689884185791,
      "epoch": 1.550632172729041,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047597484470306866,
      "loss": 5.2061,
      "mean_token_accuracy": 0.18496567904949188,
      "num_tokens": 34576835.0,
      "step": 19930
    },
    {
      "entropy": 5.47203688621521,
      "epoch": 1.5510212021007586,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004759623817836806,
      "loss": 5.1372,
      "mean_token_accuracy": 0.18821623027324677,
      "num_tokens": 34585426.0,
      "step": 19935
    },
    {
      "entropy": 5.4083505153656,
      "epoch": 1.551410231472476,
      "grad_norm": 1.25,
      "learning_rate": 0.0004759499158150082,
      "loss": 5.2078,
      "mean_token_accuracy": 0.1717604950070381,
      "num_tokens": 34594428.0,
      "step": 19940
    },
    {
      "entropy": 5.465669727325439,
      "epoch": 1.5517992608441937,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047593744679724076,
      "loss": 5.2205,
      "mean_token_accuracy": 0.18423704653978348,
      "num_tokens": 34603736.0,
      "step": 19945
    },
    {
      "entropy": 5.537930917739868,
      "epoch": 1.5521882902159114,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047592497473056733,
      "loss": 5.2093,
      "mean_token_accuracy": 0.17881692349910736,
      "num_tokens": 34613295.0,
      "step": 19950
    },
    {
      "entropy": 5.410644912719727,
      "epoch": 1.552577319587629,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047591249961517724,
      "loss": 5.146,
      "mean_token_accuracy": 0.18195830434560775,
      "num_tokens": 34622955.0,
      "step": 19955
    },
    {
      "entropy": 5.3897435665130615,
      "epoch": 1.5529663489593464,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004759000214512598,
      "loss": 5.1504,
      "mean_token_accuracy": 0.18565933406352997,
      "num_tokens": 34631866.0,
      "step": 19960
    },
    {
      "entropy": 5.477232027053833,
      "epoch": 1.553355378331064,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004758875402390042,
      "loss": 5.2286,
      "mean_token_accuracy": 0.1820620894432068,
      "num_tokens": 34640281.0,
      "step": 19965
    },
    {
      "entropy": 5.492070055007934,
      "epoch": 1.5537444077027815,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047587505597859996,
      "loss": 5.1993,
      "mean_token_accuracy": 0.17850820273160933,
      "num_tokens": 34649080.0,
      "step": 19970
    },
    {
      "entropy": 5.42692461013794,
      "epoch": 1.554133437074499,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047586256867023646,
      "loss": 5.1576,
      "mean_token_accuracy": 0.18843939006328583,
      "num_tokens": 34657969.0,
      "step": 19975
    },
    {
      "entropy": 5.509852027893066,
      "epoch": 1.5545224664462167,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004758500783141032,
      "loss": 5.2886,
      "mean_token_accuracy": 0.1782894939184189,
      "num_tokens": 34667187.0,
      "step": 19980
    },
    {
      "entropy": 5.457784795761109,
      "epoch": 1.5549114958179344,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004758375849103897,
      "loss": 5.1427,
      "mean_token_accuracy": 0.1821802243590355,
      "num_tokens": 34676292.0,
      "step": 19985
    },
    {
      "entropy": 5.499884366989136,
      "epoch": 1.5553005251896517,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004758250884592855,
      "loss": 5.2265,
      "mean_token_accuracy": 0.18101574331521988,
      "num_tokens": 34685198.0,
      "step": 19990
    },
    {
      "entropy": 5.398608064651489,
      "epoch": 1.5556895545613694,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047581258896098024,
      "loss": 5.0414,
      "mean_token_accuracy": 0.1933036372065544,
      "num_tokens": 34694019.0,
      "step": 19995
    },
    {
      "entropy": 5.469468879699707,
      "epoch": 1.556078583933087,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004758000864156636,
      "loss": 5.1959,
      "mean_token_accuracy": 0.17521129548549652,
      "num_tokens": 34703183.0,
      "step": 20000
    },
    {
      "entropy": 5.446364068984986,
      "epoch": 1.5564676133048045,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047578758082352527,
      "loss": 5.1763,
      "mean_token_accuracy": 0.1795782119035721,
      "num_tokens": 34711049.0,
      "step": 20005
    },
    {
      "entropy": 5.43844165802002,
      "epoch": 1.556856642676522,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047577507218475487,
      "loss": 5.259,
      "mean_token_accuracy": 0.17457854449748994,
      "num_tokens": 34720130.0,
      "step": 20010
    },
    {
      "entropy": 5.494384002685547,
      "epoch": 1.5572456720482397,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047576256049954236,
      "loss": 5.2128,
      "mean_token_accuracy": 0.18601434230804442,
      "num_tokens": 34728791.0,
      "step": 20015
    },
    {
      "entropy": 5.355014228820801,
      "epoch": 1.5576347014199572,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004757500457680776,
      "loss": 4.9855,
      "mean_token_accuracy": 0.1934950828552246,
      "num_tokens": 34737470.0,
      "step": 20020
    },
    {
      "entropy": 5.43953127861023,
      "epoch": 1.5580237307916747,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004757375279905504,
      "loss": 5.1721,
      "mean_token_accuracy": 0.1818319782614708,
      "num_tokens": 34745774.0,
      "step": 20025
    },
    {
      "entropy": 5.364687538146972,
      "epoch": 1.5584127601633924,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047572500716715075,
      "loss": 5.0751,
      "mean_token_accuracy": 0.1950848624110222,
      "num_tokens": 34753649.0,
      "step": 20030
    },
    {
      "entropy": 5.422722864151001,
      "epoch": 1.55880178953511,
      "grad_norm": 1.25,
      "learning_rate": 0.00047571248329806855,
      "loss": 5.1602,
      "mean_token_accuracy": 0.1822699561715126,
      "num_tokens": 34761833.0,
      "step": 20035
    },
    {
      "entropy": 5.373681163787841,
      "epoch": 1.5591908189068273,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047569995638349383,
      "loss": 5.1437,
      "mean_token_accuracy": 0.18113805502653121,
      "num_tokens": 34770969.0,
      "step": 20040
    },
    {
      "entropy": 5.412510347366333,
      "epoch": 1.559579848278545,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004756874264236168,
      "loss": 5.1935,
      "mean_token_accuracy": 0.17828971296548843,
      "num_tokens": 34779860.0,
      "step": 20045
    },
    {
      "entropy": 5.465016222000122,
      "epoch": 1.5599688776502627,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047567489341862753,
      "loss": 5.1225,
      "mean_token_accuracy": 0.18685101270675658,
      "num_tokens": 34788816.0,
      "step": 20050
    },
    {
      "entropy": 5.386428260803223,
      "epoch": 1.5603579070219802,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047566235736871607,
      "loss": 5.2118,
      "mean_token_accuracy": 0.1802176684141159,
      "num_tokens": 34797784.0,
      "step": 20055
    },
    {
      "entropy": 5.38468918800354,
      "epoch": 1.5607469363936977,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047564981827407277,
      "loss": 5.0405,
      "mean_token_accuracy": 0.19266285747289658,
      "num_tokens": 34806553.0,
      "step": 20060
    },
    {
      "entropy": 5.386757755279541,
      "epoch": 1.5611359657654154,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047563727613488785,
      "loss": 5.0752,
      "mean_token_accuracy": 0.1877358615398407,
      "num_tokens": 34814854.0,
      "step": 20065
    },
    {
      "entropy": 5.417456912994385,
      "epoch": 1.5615249951371328,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047562473095135154,
      "loss": 5.2507,
      "mean_token_accuracy": 0.18363254517316818,
      "num_tokens": 34823147.0,
      "step": 20070
    },
    {
      "entropy": 5.463807058334351,
      "epoch": 1.5619140245088503,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004756121827236544,
      "loss": 5.1426,
      "mean_token_accuracy": 0.18998710960149764,
      "num_tokens": 34831335.0,
      "step": 20075
    },
    {
      "entropy": 5.4331310272216795,
      "epoch": 1.562303053880568,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004755996314519866,
      "loss": 5.2046,
      "mean_token_accuracy": 0.17962484657764435,
      "num_tokens": 34840118.0,
      "step": 20080
    },
    {
      "entropy": 5.4149970531463625,
      "epoch": 1.5626920832522857,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004755870771365387,
      "loss": 5.1894,
      "mean_token_accuracy": 0.1785154551267624,
      "num_tokens": 34848851.0,
      "step": 20085
    },
    {
      "entropy": 5.4209503650665285,
      "epoch": 1.563081112624003,
      "grad_norm": 1.25,
      "learning_rate": 0.00047557451977750113,
      "loss": 5.0561,
      "mean_token_accuracy": 0.18795352429151535,
      "num_tokens": 34857660.0,
      "step": 20090
    },
    {
      "entropy": 5.420278644561767,
      "epoch": 1.5634701419957207,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004755619593750645,
      "loss": 5.1274,
      "mean_token_accuracy": 0.1873987063765526,
      "num_tokens": 34866828.0,
      "step": 20095
    },
    {
      "entropy": 5.448787784576416,
      "epoch": 1.5638591713674383,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004755493959294194,
      "loss": 5.2704,
      "mean_token_accuracy": 0.17430488467216493,
      "num_tokens": 34875461.0,
      "step": 20100
    },
    {
      "entropy": 5.348835849761963,
      "epoch": 1.5642482007391558,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004755368294407564,
      "loss": 5.1077,
      "mean_token_accuracy": 0.1797124281525612,
      "num_tokens": 34883914.0,
      "step": 20105
    },
    {
      "entropy": 5.4778890132904055,
      "epoch": 1.5646372301108733,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004755242599092662,
      "loss": 5.2404,
      "mean_token_accuracy": 0.1802492469549179,
      "num_tokens": 34892191.0,
      "step": 20110
    },
    {
      "entropy": 5.5175557136535645,
      "epoch": 1.565026259482591,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047551168733513956,
      "loss": 5.198,
      "mean_token_accuracy": 0.18306575268507003,
      "num_tokens": 34900450.0,
      "step": 20115
    },
    {
      "entropy": 5.432807493209839,
      "epoch": 1.5654152888543085,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047549911171856717,
      "loss": 5.0941,
      "mean_token_accuracy": 0.19144162088632583,
      "num_tokens": 34908706.0,
      "step": 20120
    },
    {
      "entropy": 5.423538732528686,
      "epoch": 1.565804318226026,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004754865330597398,
      "loss": 5.1323,
      "mean_token_accuracy": 0.1927052319049835,
      "num_tokens": 34916598.0,
      "step": 20125
    },
    {
      "entropy": 5.418301200866699,
      "epoch": 1.5661933475977436,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047547395135884854,
      "loss": 5.146,
      "mean_token_accuracy": 0.18573373556137085,
      "num_tokens": 34925850.0,
      "step": 20130
    },
    {
      "entropy": 5.497565221786499,
      "epoch": 1.5665823769694613,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004754613666160841,
      "loss": 5.2336,
      "mean_token_accuracy": 0.18274517953395844,
      "num_tokens": 34934919.0,
      "step": 20135
    },
    {
      "entropy": 5.560720634460449,
      "epoch": 1.5669714063411786,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047544877883163753,
      "loss": 5.3388,
      "mean_token_accuracy": 0.16754353195428848,
      "num_tokens": 34943875.0,
      "step": 20140
    },
    {
      "entropy": 5.456881093978882,
      "epoch": 1.5673604357128963,
      "grad_norm": 1.25,
      "learning_rate": 0.00047543618800569975,
      "loss": 5.1312,
      "mean_token_accuracy": 0.1823203057050705,
      "num_tokens": 34952458.0,
      "step": 20145
    },
    {
      "entropy": 5.423003435134888,
      "epoch": 1.567749465084614,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047542359413846184,
      "loss": 5.2183,
      "mean_token_accuracy": 0.1793881446123123,
      "num_tokens": 34961550.0,
      "step": 20150
    },
    {
      "entropy": 5.444183254241944,
      "epoch": 1.5681384944563315,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004754109972301149,
      "loss": 5.1151,
      "mean_token_accuracy": 0.17867797017097473,
      "num_tokens": 34969561.0,
      "step": 20155
    },
    {
      "entropy": 5.463370323181152,
      "epoch": 1.568527523828049,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047539839728085007,
      "loss": 5.1999,
      "mean_token_accuracy": 0.18843604028224945,
      "num_tokens": 34977904.0,
      "step": 20160
    },
    {
      "entropy": 5.4362757205963135,
      "epoch": 1.5689165531997666,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004753857942908585,
      "loss": 5.2198,
      "mean_token_accuracy": 0.1807292491197586,
      "num_tokens": 34986646.0,
      "step": 20165
    },
    {
      "entropy": 5.495504570007324,
      "epoch": 1.5693055825714841,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004753731882603315,
      "loss": 5.2281,
      "mean_token_accuracy": 0.18250027298927307,
      "num_tokens": 34995387.0,
      "step": 20170
    },
    {
      "entropy": 5.509100151062012,
      "epoch": 1.5696946119432016,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047536057918946026,
      "loss": 5.1568,
      "mean_token_accuracy": 0.1850757271051407,
      "num_tokens": 35004668.0,
      "step": 20175
    },
    {
      "entropy": 5.347309827804565,
      "epoch": 1.5700836413149193,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004753479670784361,
      "loss": 5.0403,
      "mean_token_accuracy": 0.19209699630737304,
      "num_tokens": 35012881.0,
      "step": 20180
    },
    {
      "entropy": 5.457665824890137,
      "epoch": 1.570472670686637,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004753353519274505,
      "loss": 5.1837,
      "mean_token_accuracy": 0.17480671256780625,
      "num_tokens": 35022127.0,
      "step": 20185
    },
    {
      "entropy": 5.4176140308380125,
      "epoch": 1.5708617000583545,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004753227337366948,
      "loss": 5.0815,
      "mean_token_accuracy": 0.1908511623740196,
      "num_tokens": 35030066.0,
      "step": 20190
    },
    {
      "entropy": 5.363081693649292,
      "epoch": 1.571250729430072,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047531011250636046,
      "loss": 5.1066,
      "mean_token_accuracy": 0.19017705768346788,
      "num_tokens": 35038560.0,
      "step": 20195
    },
    {
      "entropy": 5.469505453109742,
      "epoch": 1.5716397588017896,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047529748823663896,
      "loss": 5.2841,
      "mean_token_accuracy": 0.17911580353975295,
      "num_tokens": 35048011.0,
      "step": 20200
    },
    {
      "entropy": 5.541485071182251,
      "epoch": 1.572028788173507,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004752848609277219,
      "loss": 5.2301,
      "mean_token_accuracy": 0.1801591157913208,
      "num_tokens": 35056247.0,
      "step": 20205
    },
    {
      "entropy": 5.449982070922852,
      "epoch": 1.5724178175452246,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047527223057980086,
      "loss": 5.1599,
      "mean_token_accuracy": 0.17973894327878953,
      "num_tokens": 35064885.0,
      "step": 20210
    },
    {
      "entropy": 5.3860053539276125,
      "epoch": 1.5728068469169423,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047525959719306736,
      "loss": 5.092,
      "mean_token_accuracy": 0.17897505611181258,
      "num_tokens": 35073452.0,
      "step": 20215
    },
    {
      "entropy": 5.4172852516174315,
      "epoch": 1.5731958762886598,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004752469607677134,
      "loss": 5.2416,
      "mean_token_accuracy": 0.17938838005065919,
      "num_tokens": 35082459.0,
      "step": 20220
    },
    {
      "entropy": 5.453011846542358,
      "epoch": 1.5735849056603772,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004752343213039305,
      "loss": 5.1914,
      "mean_token_accuracy": 0.1760311096906662,
      "num_tokens": 35091336.0,
      "step": 20225
    },
    {
      "entropy": 5.470207166671753,
      "epoch": 1.573973935032095,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004752216788019104,
      "loss": 5.1423,
      "mean_token_accuracy": 0.18535609990358354,
      "num_tokens": 35099802.0,
      "step": 20230
    },
    {
      "entropy": 5.431559467315674,
      "epoch": 1.5743629644038126,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004752090332618451,
      "loss": 5.1761,
      "mean_token_accuracy": 0.17834040969610215,
      "num_tokens": 35108520.0,
      "step": 20235
    },
    {
      "entropy": 5.463664436340332,
      "epoch": 1.57475199377553,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004751963846839263,
      "loss": 5.2081,
      "mean_token_accuracy": 0.1903899610042572,
      "num_tokens": 35117481.0,
      "step": 20240
    },
    {
      "entropy": 5.413418197631836,
      "epoch": 1.5751410231472476,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047518373306834605,
      "loss": 5.1848,
      "mean_token_accuracy": 0.1765264391899109,
      "num_tokens": 35126605.0,
      "step": 20245
    },
    {
      "entropy": 5.433182334899902,
      "epoch": 1.5755300525189653,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047517107841529626,
      "loss": 5.0869,
      "mean_token_accuracy": 0.1884714737534523,
      "num_tokens": 35135312.0,
      "step": 20250
    },
    {
      "entropy": 5.484035348892212,
      "epoch": 1.5759190818906827,
      "grad_norm": 1.25,
      "learning_rate": 0.000475158420724969,
      "loss": 5.1487,
      "mean_token_accuracy": 0.18331928551197052,
      "num_tokens": 35144759.0,
      "step": 20255
    },
    {
      "entropy": 5.442784214019776,
      "epoch": 1.5763081112624002,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047514575999755627,
      "loss": 5.2482,
      "mean_token_accuracy": 0.18137912154197694,
      "num_tokens": 35153701.0,
      "step": 20260
    },
    {
      "entropy": 5.48476390838623,
      "epoch": 1.576697140634118,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047513309623325024,
      "loss": 5.3192,
      "mean_token_accuracy": 0.1754847288131714,
      "num_tokens": 35162503.0,
      "step": 20265
    },
    {
      "entropy": 5.432541465759277,
      "epoch": 1.5770861700058354,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004751204294322429,
      "loss": 5.1462,
      "mean_token_accuracy": 0.18605234026908873,
      "num_tokens": 35171209.0,
      "step": 20270
    },
    {
      "entropy": 5.4098015308380125,
      "epoch": 1.5774751993775529,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047510775959472675,
      "loss": 5.1444,
      "mean_token_accuracy": 0.18065100759267808,
      "num_tokens": 35179623.0,
      "step": 20275
    },
    {
      "entropy": 5.41809196472168,
      "epoch": 1.5778642287492706,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004750950867208937,
      "loss": 5.1326,
      "mean_token_accuracy": 0.18475190103054046,
      "num_tokens": 35188679.0,
      "step": 20280
    },
    {
      "entropy": 5.402343988418579,
      "epoch": 1.5782532581209883,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004750824108109362,
      "loss": 5.1177,
      "mean_token_accuracy": 0.18779564499855042,
      "num_tokens": 35197325.0,
      "step": 20285
    },
    {
      "entropy": 5.520298480987549,
      "epoch": 1.5786422874927057,
      "grad_norm": 1.125,
      "learning_rate": 0.0004750697318650466,
      "loss": 5.2563,
      "mean_token_accuracy": 0.1752472549676895,
      "num_tokens": 35206741.0,
      "step": 20290
    },
    {
      "entropy": 5.48109655380249,
      "epoch": 1.5790313168644232,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004750570498834173,
      "loss": 5.2024,
      "mean_token_accuracy": 0.17644497007131577,
      "num_tokens": 35215346.0,
      "step": 20295
    },
    {
      "entropy": 5.445080089569092,
      "epoch": 1.579420346236141,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004750443648662407,
      "loss": 5.1183,
      "mean_token_accuracy": 0.18400296568870544,
      "num_tokens": 35223811.0,
      "step": 20300
    },
    {
      "entropy": 5.468844938278198,
      "epoch": 1.5798093756078584,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047503167681370924,
      "loss": 5.2159,
      "mean_token_accuracy": 0.17576158791780472,
      "num_tokens": 35232438.0,
      "step": 20305
    },
    {
      "entropy": 5.446767807006836,
      "epoch": 1.5801984049795759,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004750189857260154,
      "loss": 5.1734,
      "mean_token_accuracy": 0.17409084290266036,
      "num_tokens": 35242529.0,
      "step": 20310
    },
    {
      "entropy": 5.34234471321106,
      "epoch": 1.5805874343512936,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004750062916033519,
      "loss": 4.9969,
      "mean_token_accuracy": 0.1972213312983513,
      "num_tokens": 35250695.0,
      "step": 20315
    },
    {
      "entropy": 5.430249118804932,
      "epoch": 1.580976463723011,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004749935944459113,
      "loss": 5.135,
      "mean_token_accuracy": 0.18306117206811906,
      "num_tokens": 35259212.0,
      "step": 20320
    },
    {
      "entropy": 5.392402029037475,
      "epoch": 1.5813654930947285,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004749808942538862,
      "loss": 5.126,
      "mean_token_accuracy": 0.18606866002082825,
      "num_tokens": 35267667.0,
      "step": 20325
    },
    {
      "entropy": 5.447392511367798,
      "epoch": 1.5817545224664462,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004749681910274693,
      "loss": 5.1623,
      "mean_token_accuracy": 0.1818079337477684,
      "num_tokens": 35275850.0,
      "step": 20330
    },
    {
      "entropy": 5.539931440353394,
      "epoch": 1.582143551838164,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047495548476685334,
      "loss": 5.1582,
      "mean_token_accuracy": 0.18419691026210785,
      "num_tokens": 35284635.0,
      "step": 20335
    },
    {
      "entropy": 5.459666204452515,
      "epoch": 1.5825325812098814,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047494277547223125,
      "loss": 5.1639,
      "mean_token_accuracy": 0.18826763331890106,
      "num_tokens": 35292335.0,
      "step": 20340
    },
    {
      "entropy": 5.37278413772583,
      "epoch": 1.5829216105815989,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047493006314379573,
      "loss": 5.1572,
      "mean_token_accuracy": 0.1870950400829315,
      "num_tokens": 35300709.0,
      "step": 20345
    },
    {
      "entropy": 5.444381856918335,
      "epoch": 1.5833106399533166,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047491734778173976,
      "loss": 5.138,
      "mean_token_accuracy": 0.18770199567079543,
      "num_tokens": 35308956.0,
      "step": 20350
    },
    {
      "entropy": 5.367072629928589,
      "epoch": 1.583699669325034,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047490462938625617,
      "loss": 5.1492,
      "mean_token_accuracy": 0.18154312372207643,
      "num_tokens": 35317663.0,
      "step": 20355
    },
    {
      "entropy": 5.430141019821167,
      "epoch": 1.5840886986967515,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047489190795753806,
      "loss": 5.231,
      "mean_token_accuracy": 0.1757897511124611,
      "num_tokens": 35327412.0,
      "step": 20360
    },
    {
      "entropy": 5.554189920425415,
      "epoch": 1.5844777280684692,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004748791834957784,
      "loss": 5.2545,
      "mean_token_accuracy": 0.17551352083683014,
      "num_tokens": 35335888.0,
      "step": 20365
    },
    {
      "entropy": 5.439113903045654,
      "epoch": 1.5848667574401867,
      "grad_norm": 1.96875,
      "learning_rate": 0.00047486645600117037,
      "loss": 5.1884,
      "mean_token_accuracy": 0.17896692156791688,
      "num_tokens": 35345122.0,
      "step": 20370
    },
    {
      "entropy": 5.442428064346314,
      "epoch": 1.5852557868119042,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004748537254739068,
      "loss": 5.2474,
      "mean_token_accuracy": 0.18319374024868013,
      "num_tokens": 35353751.0,
      "step": 20375
    },
    {
      "entropy": 5.451111221313477,
      "epoch": 1.5856448161836219,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004748409919141812,
      "loss": 5.1743,
      "mean_token_accuracy": 0.18260872811079026,
      "num_tokens": 35363083.0,
      "step": 20380
    },
    {
      "entropy": 5.34975528717041,
      "epoch": 1.5860338455553395,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004748282553221865,
      "loss": 5.1213,
      "mean_token_accuracy": 0.1911944255232811,
      "num_tokens": 35371198.0,
      "step": 20385
    },
    {
      "entropy": 5.445730590820313,
      "epoch": 1.586422874927057,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004748155156981162,
      "loss": 5.1983,
      "mean_token_accuracy": 0.1885846361517906,
      "num_tokens": 35379633.0,
      "step": 20390
    },
    {
      "entropy": 5.5097638130187985,
      "epoch": 1.5868119042987745,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047480277304216346,
      "loss": 5.1489,
      "mean_token_accuracy": 0.184230937063694,
      "num_tokens": 35388172.0,
      "step": 20395
    },
    {
      "entropy": 5.411647319793701,
      "epoch": 1.5872009336704922,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047479002735452164,
      "loss": 5.1333,
      "mean_token_accuracy": 0.17945477217435837,
      "num_tokens": 35397126.0,
      "step": 20400
    },
    {
      "entropy": 5.451480865478516,
      "epoch": 1.5875899630422097,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047477727863538415,
      "loss": 5.1266,
      "mean_token_accuracy": 0.18773161619901657,
      "num_tokens": 35405364.0,
      "step": 20405
    },
    {
      "entropy": 5.437002468109131,
      "epoch": 1.5879789924139271,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047476452688494444,
      "loss": 5.1258,
      "mean_token_accuracy": 0.18287321627140046,
      "num_tokens": 35414187.0,
      "step": 20410
    },
    {
      "entropy": 5.428719425201416,
      "epoch": 1.5883680217856448,
      "grad_norm": 1.265625,
      "learning_rate": 0.000474751772103396,
      "loss": 5.1312,
      "mean_token_accuracy": 0.18583427369594574,
      "num_tokens": 35422694.0,
      "step": 20415
    },
    {
      "entropy": 5.425508069992065,
      "epoch": 1.5887570511573625,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004747390142909323,
      "loss": 5.2392,
      "mean_token_accuracy": 0.1849235028028488,
      "num_tokens": 35431176.0,
      "step": 20420
    },
    {
      "entropy": 5.500851202011108,
      "epoch": 1.5891460805290798,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047472625344774713,
      "loss": 5.1953,
      "mean_token_accuracy": 0.17497323602437972,
      "num_tokens": 35439938.0,
      "step": 20425
    },
    {
      "entropy": 5.489838743209839,
      "epoch": 1.5895351099007975,
      "grad_norm": 1.25,
      "learning_rate": 0.00047471348957403374,
      "loss": 5.1128,
      "mean_token_accuracy": 0.18068841546773912,
      "num_tokens": 35447864.0,
      "step": 20430
    },
    {
      "entropy": 5.485489463806152,
      "epoch": 1.5899241392725152,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004747007226699862,
      "loss": 5.1701,
      "mean_token_accuracy": 0.18190595805644988,
      "num_tokens": 35456047.0,
      "step": 20435
    },
    {
      "entropy": 5.381927490234375,
      "epoch": 1.5903131686442327,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047468795273579803,
      "loss": 5.092,
      "mean_token_accuracy": 0.19141326397657393,
      "num_tokens": 35464510.0,
      "step": 20440
    },
    {
      "entropy": 5.442649221420288,
      "epoch": 1.5907021980159501,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004746751797716629,
      "loss": 5.2099,
      "mean_token_accuracy": 0.17924860566854478,
      "num_tokens": 35472715.0,
      "step": 20445
    },
    {
      "entropy": 5.425706911087036,
      "epoch": 1.5910912273876678,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004746624037777748,
      "loss": 5.0953,
      "mean_token_accuracy": 0.18678919225931168,
      "num_tokens": 35482094.0,
      "step": 20450
    },
    {
      "entropy": 5.434487771987915,
      "epoch": 1.5914802567593853,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047464962475432754,
      "loss": 5.1203,
      "mean_token_accuracy": 0.18367568254470826,
      "num_tokens": 35489864.0,
      "step": 20455
    },
    {
      "entropy": 5.448923349380493,
      "epoch": 1.5918692861311028,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000474636842701515,
      "loss": 5.2266,
      "mean_token_accuracy": 0.1773795261979103,
      "num_tokens": 35498840.0,
      "step": 20460
    },
    {
      "entropy": 5.367321443557739,
      "epoch": 1.5922583155028205,
      "grad_norm": 1.25,
      "learning_rate": 0.0004746240576195311,
      "loss": 5.0779,
      "mean_token_accuracy": 0.18357114046812056,
      "num_tokens": 35506954.0,
      "step": 20465
    },
    {
      "entropy": 5.412641334533691,
      "epoch": 1.5926473448745382,
      "grad_norm": 1.25,
      "learning_rate": 0.00047461126950856987,
      "loss": 5.223,
      "mean_token_accuracy": 0.1811554878950119,
      "num_tokens": 35515796.0,
      "step": 20470
    },
    {
      "entropy": 5.4620177268981935,
      "epoch": 1.5930363742462554,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004745984783688253,
      "loss": 5.1241,
      "mean_token_accuracy": 0.19150683134794236,
      "num_tokens": 35523713.0,
      "step": 20475
    },
    {
      "entropy": 5.44192590713501,
      "epoch": 1.5934254036179731,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047458568420049153,
      "loss": 5.262,
      "mean_token_accuracy": 0.1779765546321869,
      "num_tokens": 35532800.0,
      "step": 20480
    },
    {
      "entropy": 5.499974393844605,
      "epoch": 1.5938144329896908,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047457288700376274,
      "loss": 5.1735,
      "mean_token_accuracy": 0.187871952354908,
      "num_tokens": 35540865.0,
      "step": 20485
    },
    {
      "entropy": 5.49722113609314,
      "epoch": 1.5942034623614083,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047456008677883304,
      "loss": 5.2719,
      "mean_token_accuracy": 0.18069226294755936,
      "num_tokens": 35549498.0,
      "step": 20490
    },
    {
      "entropy": 5.469437122344971,
      "epoch": 1.5945924917331258,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004745472835258966,
      "loss": 5.1669,
      "mean_token_accuracy": 0.18711107969284058,
      "num_tokens": 35557756.0,
      "step": 20495
    },
    {
      "entropy": 5.526432991027832,
      "epoch": 1.5949815211048435,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047453447724514773,
      "loss": 5.2504,
      "mean_token_accuracy": 0.17852569222450257,
      "num_tokens": 35566062.0,
      "step": 20500
    },
    {
      "entropy": 5.423308563232422,
      "epoch": 1.595370550476561,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004745216679367808,
      "loss": 5.075,
      "mean_token_accuracy": 0.18053892850875855,
      "num_tokens": 35574377.0,
      "step": 20505
    },
    {
      "entropy": 5.440949869155884,
      "epoch": 1.5957595798482784,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004745088556009901,
      "loss": 5.1582,
      "mean_token_accuracy": 0.18533771485090256,
      "num_tokens": 35582254.0,
      "step": 20510
    },
    {
      "entropy": 5.449945735931396,
      "epoch": 1.5961486092199961,
      "grad_norm": 1.25,
      "learning_rate": 0.0004744960402379701,
      "loss": 5.218,
      "mean_token_accuracy": 0.17885905504226685,
      "num_tokens": 35590804.0,
      "step": 20515
    },
    {
      "entropy": 5.422127389907837,
      "epoch": 1.5965376385917138,
      "grad_norm": 1.25,
      "learning_rate": 0.00047448322184791525,
      "loss": 5.1521,
      "mean_token_accuracy": 0.18596051782369613,
      "num_tokens": 35599547.0,
      "step": 20520
    },
    {
      "entropy": 5.52576003074646,
      "epoch": 1.596926667963431,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047447040043101994,
      "loss": 5.2831,
      "mean_token_accuracy": 0.17599363178014754,
      "num_tokens": 35608948.0,
      "step": 20525
    },
    {
      "entropy": 5.582568168640137,
      "epoch": 1.5973156973351488,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047445757598747886,
      "loss": 5.3554,
      "mean_token_accuracy": 0.16876927241683007,
      "num_tokens": 35618243.0,
      "step": 20530
    },
    {
      "entropy": 5.436730861663818,
      "epoch": 1.5977047267068665,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004744447485174864,
      "loss": 5.1448,
      "mean_token_accuracy": 0.1886093258857727,
      "num_tokens": 35626938.0,
      "step": 20535
    },
    {
      "entropy": 5.4523481845855715,
      "epoch": 1.598093756078584,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047443191802123746,
      "loss": 5.1819,
      "mean_token_accuracy": 0.19275401383638383,
      "num_tokens": 35635612.0,
      "step": 20540
    },
    {
      "entropy": 5.443666601181031,
      "epoch": 1.5984827854503014,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047441908449892664,
      "loss": 5.182,
      "mean_token_accuracy": 0.18847143054008483,
      "num_tokens": 35644320.0,
      "step": 20545
    },
    {
      "entropy": 5.414632892608642,
      "epoch": 1.5988718148220191,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047440624795074855,
      "loss": 5.1006,
      "mean_token_accuracy": 0.18560975193977355,
      "num_tokens": 35652606.0,
      "step": 20550
    },
    {
      "entropy": 5.381332445144653,
      "epoch": 1.5992608441937366,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047439340837689804,
      "loss": 5.095,
      "mean_token_accuracy": 0.19038425832986833,
      "num_tokens": 35661082.0,
      "step": 20555
    },
    {
      "entropy": 5.3651916027069095,
      "epoch": 1.599649873565454,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047438056577756986,
      "loss": 5.1967,
      "mean_token_accuracy": 0.1791358470916748,
      "num_tokens": 35669926.0,
      "step": 20560
    },
    {
      "entropy": 5.4771589756011965,
      "epoch": 1.6000389029371718,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047436772015295903,
      "loss": 5.1853,
      "mean_token_accuracy": 0.1797611430287361,
      "num_tokens": 35678964.0,
      "step": 20565
    },
    {
      "entropy": 5.4834822654724125,
      "epoch": 1.6004279323088895,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047435487150326036,
      "loss": 5.1364,
      "mean_token_accuracy": 0.19150885790586472,
      "num_tokens": 35687943.0,
      "step": 20570
    },
    {
      "entropy": 5.370457410812378,
      "epoch": 1.6008169616806067,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004743420198286688,
      "loss": 5.1863,
      "mean_token_accuracy": 0.18146563470363616,
      "num_tokens": 35696684.0,
      "step": 20575
    },
    {
      "entropy": 5.427177762985229,
      "epoch": 1.6012059910523244,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047432916512937936,
      "loss": 5.1877,
      "mean_token_accuracy": 0.18546193689107895,
      "num_tokens": 35705038.0,
      "step": 20580
    },
    {
      "entropy": 5.500246381759643,
      "epoch": 1.6015950204240421,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004743163074055871,
      "loss": 5.1682,
      "mean_token_accuracy": 0.1812863826751709,
      "num_tokens": 35713638.0,
      "step": 20585
    },
    {
      "entropy": 5.477024126052856,
      "epoch": 1.6019840497957596,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004743034466574871,
      "loss": 5.1507,
      "mean_token_accuracy": 0.18801537603139878,
      "num_tokens": 35721935.0,
      "step": 20590
    },
    {
      "entropy": 5.463110828399659,
      "epoch": 1.602373079167477,
      "grad_norm": 1.25,
      "learning_rate": 0.0004742905828852746,
      "loss": 5.2114,
      "mean_token_accuracy": 0.17838673293590546,
      "num_tokens": 35731610.0,
      "step": 20595
    },
    {
      "entropy": 5.496250772476197,
      "epoch": 1.6027621085391948,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004742777160891447,
      "loss": 5.2141,
      "mean_token_accuracy": 0.18232737183570863,
      "num_tokens": 35740956.0,
      "step": 20600
    },
    {
      "entropy": 5.409165334701538,
      "epoch": 1.6031511379109122,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004742648462692926,
      "loss": 5.0717,
      "mean_token_accuracy": 0.19261856079101564,
      "num_tokens": 35749316.0,
      "step": 20605
    },
    {
      "entropy": 5.46047534942627,
      "epoch": 1.6035401672826297,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047425197342591363,
      "loss": 5.2532,
      "mean_token_accuracy": 0.1810690939426422,
      "num_tokens": 35758875.0,
      "step": 20610
    },
    {
      "entropy": 5.41661581993103,
      "epoch": 1.6039291966543474,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004742390975592031,
      "loss": 5.2071,
      "mean_token_accuracy": 0.18556952029466628,
      "num_tokens": 35767851.0,
      "step": 20615
    },
    {
      "entropy": 5.490317916870117,
      "epoch": 1.604318226026065,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047422621866935645,
      "loss": 5.0972,
      "mean_token_accuracy": 0.189169342815876,
      "num_tokens": 35776393.0,
      "step": 20620
    },
    {
      "entropy": 5.365667295455933,
      "epoch": 1.6047072553977826,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000474213336756569,
      "loss": 5.0909,
      "mean_token_accuracy": 0.19008722603321077,
      "num_tokens": 35784868.0,
      "step": 20625
    },
    {
      "entropy": 5.42010498046875,
      "epoch": 1.6050962847695,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004742004518210362,
      "loss": 5.1695,
      "mean_token_accuracy": 0.17457361966371537,
      "num_tokens": 35793812.0,
      "step": 20630
    },
    {
      "entropy": 5.471294689178467,
      "epoch": 1.6054853141412178,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004741875638629536,
      "loss": 5.1585,
      "mean_token_accuracy": 0.1743136912584305,
      "num_tokens": 35802206.0,
      "step": 20635
    },
    {
      "entropy": 5.440286636352539,
      "epoch": 1.6058743435129352,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004741746728825168,
      "loss": 5.1444,
      "mean_token_accuracy": 0.18190768957138062,
      "num_tokens": 35811095.0,
      "step": 20640
    },
    {
      "entropy": 5.5096173763275145,
      "epoch": 1.6062633728846527,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004741617788799213,
      "loss": 5.2527,
      "mean_token_accuracy": 0.17459849417209625,
      "num_tokens": 35819799.0,
      "step": 20645
    },
    {
      "entropy": 5.432175779342652,
      "epoch": 1.6066524022563704,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047414888185536285,
      "loss": 5.1828,
      "mean_token_accuracy": 0.18460266441106796,
      "num_tokens": 35829130.0,
      "step": 20650
    },
    {
      "entropy": 5.434774446487427,
      "epoch": 1.6070414316280879,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004741359818090371,
      "loss": 5.1194,
      "mean_token_accuracy": 0.18427557796239852,
      "num_tokens": 35838020.0,
      "step": 20655
    },
    {
      "entropy": 5.355521965026855,
      "epoch": 1.6074304609998054,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047412307874113976,
      "loss": 5.0443,
      "mean_token_accuracy": 0.19493936449289323,
      "num_tokens": 35845843.0,
      "step": 20660
    },
    {
      "entropy": 5.350476980209351,
      "epoch": 1.607819490371523,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004741101726518667,
      "loss": 5.0776,
      "mean_token_accuracy": 0.1854074627161026,
      "num_tokens": 35854505.0,
      "step": 20665
    },
    {
      "entropy": 5.409836149215698,
      "epoch": 1.6082085197432407,
      "grad_norm": 1.25,
      "learning_rate": 0.0004740972635414136,
      "loss": 5.1609,
      "mean_token_accuracy": 0.18538784682750703,
      "num_tokens": 35862847.0,
      "step": 20670
    },
    {
      "entropy": 5.402918815612793,
      "epoch": 1.6085975491149582,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004740843514099765,
      "loss": 5.1182,
      "mean_token_accuracy": 0.1794407397508621,
      "num_tokens": 35871471.0,
      "step": 20675
    },
    {
      "entropy": 5.488820648193359,
      "epoch": 1.6089865784866757,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004740714362577512,
      "loss": 5.1417,
      "mean_token_accuracy": 0.17694109976291655,
      "num_tokens": 35880266.0,
      "step": 20680
    },
    {
      "entropy": 5.431238174438477,
      "epoch": 1.6093756078583934,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047405851808493364,
      "loss": 5.0557,
      "mean_token_accuracy": 0.19038631170988082,
      "num_tokens": 35888727.0,
      "step": 20685
    },
    {
      "entropy": 5.3681519508361815,
      "epoch": 1.6097646372301109,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047404559689172006,
      "loss": 5.1532,
      "mean_token_accuracy": 0.18077253103256224,
      "num_tokens": 35897750.0,
      "step": 20690
    },
    {
      "entropy": 5.412582921981811,
      "epoch": 1.6101536666018283,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047403267267830617,
      "loss": 5.1024,
      "mean_token_accuracy": 0.18918199539184571,
      "num_tokens": 35906254.0,
      "step": 20695
    },
    {
      "entropy": 5.420455265045166,
      "epoch": 1.610542695973546,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047401974544488844,
      "loss": 5.2109,
      "mean_token_accuracy": 0.1716628849506378,
      "num_tokens": 35915180.0,
      "step": 20700
    },
    {
      "entropy": 5.483662843704224,
      "epoch": 1.6109317253452635,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004740068151916628,
      "loss": 5.1797,
      "mean_token_accuracy": 0.17427147030830384,
      "num_tokens": 35924336.0,
      "step": 20705
    },
    {
      "entropy": 5.470591878890991,
      "epoch": 1.611320754716981,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004739938819188255,
      "loss": 5.1719,
      "mean_token_accuracy": 0.18257935494184493,
      "num_tokens": 35933031.0,
      "step": 20710
    },
    {
      "entropy": 5.450223445892334,
      "epoch": 1.6117097840886987,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004739809456265727,
      "loss": 5.0416,
      "mean_token_accuracy": 0.19644464701414108,
      "num_tokens": 35941794.0,
      "step": 20715
    },
    {
      "entropy": 5.519880533218384,
      "epoch": 1.6120988134604164,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004739680063151008,
      "loss": 5.2414,
      "mean_token_accuracy": 0.18030249923467637,
      "num_tokens": 35949636.0,
      "step": 20720
    },
    {
      "entropy": 5.465391159057617,
      "epoch": 1.6124878428321339,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004739550639846061,
      "loss": 5.2702,
      "mean_token_accuracy": 0.1798594266176224,
      "num_tokens": 35958842.0,
      "step": 20725
    },
    {
      "entropy": 5.408455896377563,
      "epoch": 1.6128768722038513,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047394211863528496,
      "loss": 5.1087,
      "mean_token_accuracy": 0.1803465738892555,
      "num_tokens": 35967897.0,
      "step": 20730
    },
    {
      "entropy": 5.347626447677612,
      "epoch": 1.613265901575569,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004739291702673338,
      "loss": 5.0231,
      "mean_token_accuracy": 0.1928437888622284,
      "num_tokens": 35976309.0,
      "step": 20735
    },
    {
      "entropy": 5.544116020202637,
      "epoch": 1.6136549309472865,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004739162188809492,
      "loss": 5.2511,
      "mean_token_accuracy": 0.18134445697069168,
      "num_tokens": 35984401.0,
      "step": 20740
    },
    {
      "entropy": 5.445677089691162,
      "epoch": 1.614043960319004,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047390326447632747,
      "loss": 5.1498,
      "mean_token_accuracy": 0.17752983421087265,
      "num_tokens": 35993068.0,
      "step": 20745
    },
    {
      "entropy": 5.539250612258911,
      "epoch": 1.6144329896907217,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004738903070536654,
      "loss": 5.3153,
      "mean_token_accuracy": 0.18138758689165116,
      "num_tokens": 36001559.0,
      "step": 20750
    },
    {
      "entropy": 5.500018358230591,
      "epoch": 1.6148220190624392,
      "grad_norm": 1.375,
      "learning_rate": 0.0004738773466131594,
      "loss": 5.2313,
      "mean_token_accuracy": 0.18270974457263947,
      "num_tokens": 36010675.0,
      "step": 20755
    },
    {
      "entropy": 5.37749662399292,
      "epoch": 1.6152110484341566,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004738643831550063,
      "loss": 4.9685,
      "mean_token_accuracy": 0.19576405584812165,
      "num_tokens": 36019498.0,
      "step": 20760
    },
    {
      "entropy": 5.330859804153443,
      "epoch": 1.6156000778058743,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004738514166794026,
      "loss": 5.0387,
      "mean_token_accuracy": 0.19797454327344893,
      "num_tokens": 36028206.0,
      "step": 20765
    },
    {
      "entropy": 5.492358016967773,
      "epoch": 1.615989107177592,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047383844718654525,
      "loss": 5.2561,
      "mean_token_accuracy": 0.18545452058315276,
      "num_tokens": 36037341.0,
      "step": 20770
    },
    {
      "entropy": 5.447669172286988,
      "epoch": 1.6163781365493095,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004738254746766309,
      "loss": 5.1395,
      "mean_token_accuracy": 0.19092552065849305,
      "num_tokens": 36046596.0,
      "step": 20775
    },
    {
      "entropy": 5.407862043380737,
      "epoch": 1.616767165921027,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004738124991498565,
      "loss": 5.1,
      "mean_token_accuracy": 0.18787268102169036,
      "num_tokens": 36055031.0,
      "step": 20780
    },
    {
      "entropy": 5.495108413696289,
      "epoch": 1.6171561952927447,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047379952060641866,
      "loss": 5.2504,
      "mean_token_accuracy": 0.1894080713391304,
      "num_tokens": 36064049.0,
      "step": 20785
    },
    {
      "entropy": 5.506619834899903,
      "epoch": 1.6175452246644622,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047378653904651475,
      "loss": 5.2307,
      "mean_token_accuracy": 0.17482070922851561,
      "num_tokens": 36072886.0,
      "step": 20790
    },
    {
      "entropy": 5.501468658447266,
      "epoch": 1.6179342540361796,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047377355447034136,
      "loss": 5.2065,
      "mean_token_accuracy": 0.1820884108543396,
      "num_tokens": 36082133.0,
      "step": 20795
    },
    {
      "entropy": 5.464742851257324,
      "epoch": 1.6183232834078973,
      "grad_norm": 1.375,
      "learning_rate": 0.0004737605668780958,
      "loss": 5.1124,
      "mean_token_accuracy": 0.18634761720895768,
      "num_tokens": 36090980.0,
      "step": 20800
    },
    {
      "entropy": 5.428704690933228,
      "epoch": 1.618712312779615,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047374757626997494,
      "loss": 5.2079,
      "mean_token_accuracy": 0.18200210928916932,
      "num_tokens": 36099584.0,
      "step": 20805
    },
    {
      "entropy": 5.385202121734619,
      "epoch": 1.6191013421513323,
      "grad_norm": 1.375,
      "learning_rate": 0.0004737345826461759,
      "loss": 5.1537,
      "mean_token_accuracy": 0.1783108800649643,
      "num_tokens": 36108052.0,
      "step": 20810
    },
    {
      "entropy": 5.4375892162323,
      "epoch": 1.61949037152305,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004737215860068959,
      "loss": 5.0322,
      "mean_token_accuracy": 0.19564128369092942,
      "num_tokens": 36116623.0,
      "step": 20815
    },
    {
      "entropy": 5.481699228286743,
      "epoch": 1.6198794008947677,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047370858635233223,
      "loss": 5.171,
      "mean_token_accuracy": 0.18492969274520873,
      "num_tokens": 36124953.0,
      "step": 20820
    },
    {
      "entropy": 5.398912382125855,
      "epoch": 1.6202684302664851,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047369558368268194,
      "loss": 5.158,
      "mean_token_accuracy": 0.1828717440366745,
      "num_tokens": 36133500.0,
      "step": 20825
    },
    {
      "entropy": 5.3912111759185795,
      "epoch": 1.6206574596382026,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004736825779981424,
      "loss": 5.1125,
      "mean_token_accuracy": 0.18695752769708635,
      "num_tokens": 36141934.0,
      "step": 20830
    },
    {
      "entropy": 5.402982997894287,
      "epoch": 1.6210464890099203,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004736695692989111,
      "loss": 5.1223,
      "mean_token_accuracy": 0.18809404969215393,
      "num_tokens": 36150592.0,
      "step": 20835
    },
    {
      "entropy": 5.3729877948760985,
      "epoch": 1.6214355183816378,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004736565575851852,
      "loss": 4.9945,
      "mean_token_accuracy": 0.19637948870658875,
      "num_tokens": 36159377.0,
      "step": 20840
    },
    {
      "entropy": 5.425131320953369,
      "epoch": 1.6218245477533553,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004736435428571622,
      "loss": 5.118,
      "mean_token_accuracy": 0.18691076785326005,
      "num_tokens": 36168159.0,
      "step": 20845
    },
    {
      "entropy": 5.3226207256317135,
      "epoch": 1.622213577125073,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004736305251150397,
      "loss": 5.0777,
      "mean_token_accuracy": 0.18796844482421876,
      "num_tokens": 36176108.0,
      "step": 20850
    },
    {
      "entropy": 5.4310914993286135,
      "epoch": 1.6226026064967907,
      "grad_norm": 1.375,
      "learning_rate": 0.0004736175043590151,
      "loss": 5.2595,
      "mean_token_accuracy": 0.1762857273221016,
      "num_tokens": 36184930.0,
      "step": 20855
    },
    {
      "entropy": 5.513455104827881,
      "epoch": 1.622991635868508,
      "grad_norm": 1.25,
      "learning_rate": 0.000473604480589286,
      "loss": 5.1856,
      "mean_token_accuracy": 0.18750521540641785,
      "num_tokens": 36193308.0,
      "step": 20860
    },
    {
      "entropy": 5.388754940032959,
      "epoch": 1.6233806652402256,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047359145380605007,
      "loss": 5.133,
      "mean_token_accuracy": 0.1900997206568718,
      "num_tokens": 36202016.0,
      "step": 20865
    },
    {
      "entropy": 5.459857654571533,
      "epoch": 1.6237696946119433,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004735784240095049,
      "loss": 5.1196,
      "mean_token_accuracy": 0.18242949843406678,
      "num_tokens": 36210757.0,
      "step": 20870
    },
    {
      "entropy": 5.389839553833008,
      "epoch": 1.6241587239836608,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047356539119984813,
      "loss": 5.066,
      "mean_token_accuracy": 0.18458036929368973,
      "num_tokens": 36219052.0,
      "step": 20875
    },
    {
      "entropy": 5.308129405975341,
      "epoch": 1.6245477533553783,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004735523553772777,
      "loss": 5.0533,
      "mean_token_accuracy": 0.1916509047150612,
      "num_tokens": 36227890.0,
      "step": 20880
    },
    {
      "entropy": 5.351578140258789,
      "epoch": 1.624936782727096,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004735393165419912,
      "loss": 5.1744,
      "mean_token_accuracy": 0.1826167434453964,
      "num_tokens": 36236725.0,
      "step": 20885
    },
    {
      "entropy": 5.529488277435303,
      "epoch": 1.6253258120988134,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047352627469418666,
      "loss": 5.1855,
      "mean_token_accuracy": 0.18249355256557465,
      "num_tokens": 36246220.0,
      "step": 20890
    },
    {
      "entropy": 5.4847249507904055,
      "epoch": 1.625714841470531,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004735132298340619,
      "loss": 5.2039,
      "mean_token_accuracy": 0.17801870852708818,
      "num_tokens": 36254662.0,
      "step": 20895
    },
    {
      "entropy": 5.383641242980957,
      "epoch": 1.6261038708422486,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004735001819618148,
      "loss": 5.167,
      "mean_token_accuracy": 0.18517387360334397,
      "num_tokens": 36262446.0,
      "step": 20900
    },
    {
      "entropy": 5.468044185638428,
      "epoch": 1.6264929002139663,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004734871310776434,
      "loss": 5.1691,
      "mean_token_accuracy": 0.17988636940717698,
      "num_tokens": 36271284.0,
      "step": 20905
    },
    {
      "entropy": 5.400888586044312,
      "epoch": 1.6268819295856836,
      "grad_norm": 1.375,
      "learning_rate": 0.0004734740771817457,
      "loss": 5.0499,
      "mean_token_accuracy": 0.19043746143579482,
      "num_tokens": 36279180.0,
      "step": 20910
    },
    {
      "entropy": 5.387185907363891,
      "epoch": 1.6272709589574013,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004734610202743198,
      "loss": 5.178,
      "mean_token_accuracy": 0.1855251282453537,
      "num_tokens": 36287714.0,
      "step": 20915
    },
    {
      "entropy": 5.465124654769897,
      "epoch": 1.627659988329119,
      "grad_norm": 1.25,
      "learning_rate": 0.0004734479603555638,
      "loss": 5.1885,
      "mean_token_accuracy": 0.1842973753809929,
      "num_tokens": 36297120.0,
      "step": 20920
    },
    {
      "entropy": 5.494491910934448,
      "epoch": 1.6280490177008364,
      "grad_norm": 1.25,
      "learning_rate": 0.00047343489742567593,
      "loss": 5.1845,
      "mean_token_accuracy": 0.18406689763069153,
      "num_tokens": 36305232.0,
      "step": 20925
    },
    {
      "entropy": 5.493837356567383,
      "epoch": 1.628438047072554,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047342183148485424,
      "loss": 5.2181,
      "mean_token_accuracy": 0.18834635764360427,
      "num_tokens": 36313587.0,
      "step": 20930
    },
    {
      "entropy": 5.4487566471099855,
      "epoch": 1.6288270764442716,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047340876253329706,
      "loss": 5.1621,
      "mean_token_accuracy": 0.19119502902030944,
      "num_tokens": 36322054.0,
      "step": 20935
    },
    {
      "entropy": 5.426901388168335,
      "epoch": 1.629216105815989,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047339569057120275,
      "loss": 5.0777,
      "mean_token_accuracy": 0.19229902774095536,
      "num_tokens": 36330226.0,
      "step": 20940
    },
    {
      "entropy": 5.382965421676635,
      "epoch": 1.6296051351877066,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047338261559876966,
      "loss": 5.1499,
      "mean_token_accuracy": 0.17930882573127746,
      "num_tokens": 36338436.0,
      "step": 20945
    },
    {
      "entropy": 5.369858121871948,
      "epoch": 1.6299941645594243,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047336953761619604,
      "loss": 5.1205,
      "mean_token_accuracy": 0.18394071012735366,
      "num_tokens": 36347204.0,
      "step": 20950
    },
    {
      "entropy": 5.416620302200317,
      "epoch": 1.630383193931142,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047335645662368046,
      "loss": 5.117,
      "mean_token_accuracy": 0.19004431664943694,
      "num_tokens": 36356225.0,
      "step": 20955
    },
    {
      "entropy": 5.416371822357178,
      "epoch": 1.6307722233028592,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004733433726214214,
      "loss": 5.1828,
      "mean_token_accuracy": 0.17974185645580293,
      "num_tokens": 36364526.0,
      "step": 20960
    },
    {
      "entropy": 5.521980094909668,
      "epoch": 1.631161252674577,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047333028560961733,
      "loss": 5.2603,
      "mean_token_accuracy": 0.1762666717171669,
      "num_tokens": 36373162.0,
      "step": 20965
    },
    {
      "entropy": 5.484997797012329,
      "epoch": 1.6315502820462946,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047331719558846687,
      "loss": 5.1528,
      "mean_token_accuracy": 0.17944801449775696,
      "num_tokens": 36381545.0,
      "step": 20970
    },
    {
      "entropy": 5.400477170944214,
      "epoch": 1.631939311418012,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047330410255816863,
      "loss": 5.1243,
      "mean_token_accuracy": 0.18846172243356704,
      "num_tokens": 36390267.0,
      "step": 20975
    },
    {
      "entropy": 5.450103664398194,
      "epoch": 1.6323283407897295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047329100651892123,
      "loss": 5.1545,
      "mean_token_accuracy": 0.1901376724243164,
      "num_tokens": 36398910.0,
      "step": 20980
    },
    {
      "entropy": 5.449186229705811,
      "epoch": 1.6327173701614472,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004732779074709234,
      "loss": 5.1287,
      "mean_token_accuracy": 0.186529041826725,
      "num_tokens": 36407382.0,
      "step": 20985
    },
    {
      "entropy": 5.3768504619598385,
      "epoch": 1.6331063995331647,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000473264805414374,
      "loss": 5.1368,
      "mean_token_accuracy": 0.18458705842494966,
      "num_tokens": 36415734.0,
      "step": 20990
    },
    {
      "entropy": 5.433033418655396,
      "epoch": 1.6334954289048822,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047325170034947167,
      "loss": 5.2093,
      "mean_token_accuracy": 0.17567485421895981,
      "num_tokens": 36424537.0,
      "step": 20995
    },
    {
      "entropy": 5.471461057662964,
      "epoch": 1.6338844582766,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047323859227641537,
      "loss": 5.1021,
      "mean_token_accuracy": 0.18617338687181473,
      "num_tokens": 36433276.0,
      "step": 21000
    },
    {
      "epoch": 1.6338844582766,
      "eval_entropy": 5.3532049030381055,
      "eval_loss": 5.248416900634766,
      "eval_mean_token_accuracy": 0.1883855755687391,
      "eval_num_tokens": 36433276.0,
      "eval_runtime": 28.7023,
      "eval_samples_per_second": 1447.898,
      "eval_steps_per_second": 180.996,
      "step": 21000
    },
    {
      "entropy": 5.4729327201843265,
      "epoch": 1.6342734876483176,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047322548119540397,
      "loss": 5.1478,
      "mean_token_accuracy": 0.1813689276576042,
      "num_tokens": 36442154.0,
      "step": 21005
    },
    {
      "entropy": 5.3455397605896,
      "epoch": 1.634662517020035,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047321236710663636,
      "loss": 5.001,
      "mean_token_accuracy": 0.1939090222120285,
      "num_tokens": 36450625.0,
      "step": 21010
    },
    {
      "entropy": 5.434037733078003,
      "epoch": 1.6350515463917525,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047319925001031165,
      "loss": 5.1604,
      "mean_token_accuracy": 0.18046659976243973,
      "num_tokens": 36460082.0,
      "step": 21015
    },
    {
      "entropy": 5.487746334075927,
      "epoch": 1.6354405757634702,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004731861299066287,
      "loss": 5.1928,
      "mean_token_accuracy": 0.18331139236688615,
      "num_tokens": 36469372.0,
      "step": 21020
    },
    {
      "entropy": 5.381699323654175,
      "epoch": 1.6358296051351877,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004731730067957867,
      "loss": 4.9904,
      "mean_token_accuracy": 0.1906184270977974,
      "num_tokens": 36478190.0,
      "step": 21025
    },
    {
      "entropy": 5.347916460037231,
      "epoch": 1.6362186345069052,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047315988067798476,
      "loss": 5.1225,
      "mean_token_accuracy": 0.19068235605955125,
      "num_tokens": 36486245.0,
      "step": 21030
    },
    {
      "entropy": 5.401310634613037,
      "epoch": 1.6366076638786229,
      "grad_norm": 1.234375,
      "learning_rate": 0.000473146751553422,
      "loss": 5.1813,
      "mean_token_accuracy": 0.18376985043287278,
      "num_tokens": 36495858.0,
      "step": 21035
    },
    {
      "entropy": 5.5129273414611815,
      "epoch": 1.6369966932503404,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004731336194222978,
      "loss": 5.185,
      "mean_token_accuracy": 0.17840277701616286,
      "num_tokens": 36504564.0,
      "step": 21040
    },
    {
      "entropy": 5.321560668945312,
      "epoch": 1.6373857226220578,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047312048428481114,
      "loss": 5.064,
      "mean_token_accuracy": 0.18959666043519974,
      "num_tokens": 36512758.0,
      "step": 21045
    },
    {
      "entropy": 5.362840700149536,
      "epoch": 1.6377747519937755,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047310734614116153,
      "loss": 5.0353,
      "mean_token_accuracy": 0.1852007269859314,
      "num_tokens": 36521429.0,
      "step": 21050
    },
    {
      "entropy": 5.42672381401062,
      "epoch": 1.6381637813654932,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004730942049915483,
      "loss": 5.1089,
      "mean_token_accuracy": 0.19230006486177445,
      "num_tokens": 36530161.0,
      "step": 21055
    },
    {
      "entropy": 5.423188066482544,
      "epoch": 1.6385528107372107,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004730810608361707,
      "loss": 5.108,
      "mean_token_accuracy": 0.18639254122972487,
      "num_tokens": 36538184.0,
      "step": 21060
    },
    {
      "entropy": 5.448139429092407,
      "epoch": 1.6389418401089282,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047306791367522833,
      "loss": 5.128,
      "mean_token_accuracy": 0.19195273220539094,
      "num_tokens": 36547418.0,
      "step": 21065
    },
    {
      "entropy": 5.573483514785766,
      "epoch": 1.6393308694806459,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004730547635089206,
      "loss": 5.306,
      "mean_token_accuracy": 0.17890384197235107,
      "num_tokens": 36556542.0,
      "step": 21070
    },
    {
      "entropy": 5.448286008834839,
      "epoch": 1.6397198988523634,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047304161033744714,
      "loss": 5.0955,
      "mean_token_accuracy": 0.1809767410159111,
      "num_tokens": 36565903.0,
      "step": 21075
    },
    {
      "entropy": 5.382310152053833,
      "epoch": 1.6401089282240808,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004730284541610075,
      "loss": 5.1151,
      "mean_token_accuracy": 0.18813490867614746,
      "num_tokens": 36574010.0,
      "step": 21080
    },
    {
      "entropy": 5.400834751129151,
      "epoch": 1.6404979575957985,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004730152949798012,
      "loss": 5.1447,
      "mean_token_accuracy": 0.19069821685552596,
      "num_tokens": 36582142.0,
      "step": 21085
    },
    {
      "entropy": 5.441775941848755,
      "epoch": 1.640886986967516,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047300213279402787,
      "loss": 5.1451,
      "mean_token_accuracy": 0.18750930726528167,
      "num_tokens": 36590676.0,
      "step": 21090
    },
    {
      "entropy": 5.397728109359742,
      "epoch": 1.6412760163392335,
      "grad_norm": 1.25,
      "learning_rate": 0.00047298896760388745,
      "loss": 5.0334,
      "mean_token_accuracy": 0.20079335570335388,
      "num_tokens": 36599489.0,
      "step": 21095
    },
    {
      "entropy": 5.435423040390015,
      "epoch": 1.6416650457109512,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004729757994095796,
      "loss": 5.1181,
      "mean_token_accuracy": 0.1788809269666672,
      "num_tokens": 36608231.0,
      "step": 21100
    },
    {
      "entropy": 5.37393627166748,
      "epoch": 1.6420540750826689,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047296262821130405,
      "loss": 5.1725,
      "mean_token_accuracy": 0.1817234843969345,
      "num_tokens": 36617335.0,
      "step": 21105
    },
    {
      "entropy": 5.481614398956299,
      "epoch": 1.6424431044543863,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047294945400926065,
      "loss": 5.1787,
      "mean_token_accuracy": 0.1827680453658104,
      "num_tokens": 36626249.0,
      "step": 21110
    },
    {
      "entropy": 5.484762811660767,
      "epoch": 1.6428321338261038,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004729362768036494,
      "loss": 5.2268,
      "mean_token_accuracy": 0.1786796733736992,
      "num_tokens": 36634602.0,
      "step": 21115
    },
    {
      "entropy": 5.485878133773804,
      "epoch": 1.6432211631978215,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047292309659467024,
      "loss": 5.1811,
      "mean_token_accuracy": 0.18429100662469863,
      "num_tokens": 36643091.0,
      "step": 21120
    },
    {
      "entropy": 5.486299133300781,
      "epoch": 1.643610192569539,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004729099133825231,
      "loss": 5.1857,
      "mean_token_accuracy": 0.18498112857341767,
      "num_tokens": 36652530.0,
      "step": 21125
    },
    {
      "entropy": 5.472276210784912,
      "epoch": 1.6439992219412565,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000472896727167408,
      "loss": 5.2067,
      "mean_token_accuracy": 0.18065759986639024,
      "num_tokens": 36660858.0,
      "step": 21130
    },
    {
      "entropy": 5.54839596748352,
      "epoch": 1.6443882513129742,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047288353794952505,
      "loss": 5.267,
      "mean_token_accuracy": 0.16871827840805054,
      "num_tokens": 36670211.0,
      "step": 21135
    },
    {
      "entropy": 5.423988723754883,
      "epoch": 1.6447772806846916,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004728703457290744,
      "loss": 5.1797,
      "mean_token_accuracy": 0.17730396389961242,
      "num_tokens": 36679604.0,
      "step": 21140
    },
    {
      "entropy": 5.478514575958252,
      "epoch": 1.6451663100564091,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004728571505062562,
      "loss": 5.2613,
      "mean_token_accuracy": 0.17477357089519502,
      "num_tokens": 36687908.0,
      "step": 21145
    },
    {
      "entropy": 5.435767650604248,
      "epoch": 1.6455553394281268,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004728439522812707,
      "loss": 5.0412,
      "mean_token_accuracy": 0.18947336822748184,
      "num_tokens": 36696675.0,
      "step": 21150
    },
    {
      "entropy": 5.414568614959717,
      "epoch": 1.6459443687998445,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047283075105431806,
      "loss": 5.1536,
      "mean_token_accuracy": 0.1790348395705223,
      "num_tokens": 36705594.0,
      "step": 21155
    },
    {
      "entropy": 5.497033596038818,
      "epoch": 1.646333398171562,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004728175468255987,
      "loss": 5.2032,
      "mean_token_accuracy": 0.18610911071300507,
      "num_tokens": 36714008.0,
      "step": 21160
    },
    {
      "entropy": 5.373608064651489,
      "epoch": 1.6467224275432795,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047280433959531287,
      "loss": 5.0406,
      "mean_token_accuracy": 0.1946001723408699,
      "num_tokens": 36723093.0,
      "step": 21165
    },
    {
      "entropy": 5.375360202789307,
      "epoch": 1.6471114569149972,
      "grad_norm": 1.1875,
      "learning_rate": 0.000472791129363661,
      "loss": 5.0544,
      "mean_token_accuracy": 0.18963581770658494,
      "num_tokens": 36732119.0,
      "step": 21170
    },
    {
      "entropy": 5.354014730453491,
      "epoch": 1.6475004862867146,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004727779161308436,
      "loss": 5.1331,
      "mean_token_accuracy": 0.18068700134754181,
      "num_tokens": 36740593.0,
      "step": 21175
    },
    {
      "entropy": 5.518045902252197,
      "epoch": 1.6478895156584321,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004727646998970612,
      "loss": 5.3118,
      "mean_token_accuracy": 0.17549846619367598,
      "num_tokens": 36749083.0,
      "step": 21180
    },
    {
      "entropy": 5.482885646820068,
      "epoch": 1.6482785450301498,
      "grad_norm": 1.25,
      "learning_rate": 0.00047275148066251417,
      "loss": 5.1191,
      "mean_token_accuracy": 0.1857609435915947,
      "num_tokens": 36757469.0,
      "step": 21185
    },
    {
      "entropy": 5.508031797409058,
      "epoch": 1.6486675744018673,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004727382584274032,
      "loss": 5.2954,
      "mean_token_accuracy": 0.1673070102930069,
      "num_tokens": 36766197.0,
      "step": 21190
    },
    {
      "entropy": 5.4474118709564205,
      "epoch": 1.6490566037735848,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047272503319192887,
      "loss": 5.1113,
      "mean_token_accuracy": 0.1860267177224159,
      "num_tokens": 36774172.0,
      "step": 21195
    },
    {
      "entropy": 5.4273491382598875,
      "epoch": 1.6494456331453025,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004727118049562919,
      "loss": 5.163,
      "mean_token_accuracy": 0.18042041659355162,
      "num_tokens": 36783549.0,
      "step": 21200
    },
    {
      "entropy": 5.520964574813843,
      "epoch": 1.6498346625170202,
      "grad_norm": 1.453125,
      "learning_rate": 0.000472698573720693,
      "loss": 5.2205,
      "mean_token_accuracy": 0.18044029772281647,
      "num_tokens": 36791755.0,
      "step": 21205
    },
    {
      "entropy": 5.492210960388183,
      "epoch": 1.6502236918887376,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004726853394853329,
      "loss": 5.2627,
      "mean_token_accuracy": 0.1747572898864746,
      "num_tokens": 36801253.0,
      "step": 21210
    },
    {
      "entropy": 5.376606035232544,
      "epoch": 1.650612721260455,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047267210225041234,
      "loss": 5.1089,
      "mean_token_accuracy": 0.19334858655929565,
      "num_tokens": 36810497.0,
      "step": 21215
    },
    {
      "entropy": 5.371975564956665,
      "epoch": 1.6510017506321728,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047265886201613234,
      "loss": 5.0899,
      "mean_token_accuracy": 0.19561389237642288,
      "num_tokens": 36819823.0,
      "step": 21220
    },
    {
      "entropy": 5.422666931152344,
      "epoch": 1.6513907800038903,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047264561878269365,
      "loss": 5.0695,
      "mean_token_accuracy": 0.18724533468484877,
      "num_tokens": 36828935.0,
      "step": 21225
    },
    {
      "entropy": 5.390351009368897,
      "epoch": 1.6517798093756078,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004726323725502973,
      "loss": 5.0185,
      "mean_token_accuracy": 0.18440023064613342,
      "num_tokens": 36837906.0,
      "step": 21230
    },
    {
      "entropy": 5.426978254318238,
      "epoch": 1.6521688387473255,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004726191233191443,
      "loss": 5.1145,
      "mean_token_accuracy": 0.18981438577175141,
      "num_tokens": 36846041.0,
      "step": 21235
    },
    {
      "entropy": 5.498830318450928,
      "epoch": 1.6525578681190431,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004726058710894357,
      "loss": 5.2227,
      "mean_token_accuracy": 0.17717188447713852,
      "num_tokens": 36854493.0,
      "step": 21240
    },
    {
      "entropy": 5.378484582901001,
      "epoch": 1.6529468974907604,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004725926158613724,
      "loss": 5.0481,
      "mean_token_accuracy": 0.1898688107728958,
      "num_tokens": 36863042.0,
      "step": 21245
    },
    {
      "entropy": 5.466516065597534,
      "epoch": 1.653335926862478,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004725793576351557,
      "loss": 5.1789,
      "mean_token_accuracy": 0.17678294628858565,
      "num_tokens": 36872513.0,
      "step": 21250
    },
    {
      "entropy": 5.480537414550781,
      "epoch": 1.6537249562341958,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004725660964109867,
      "loss": 5.1629,
      "mean_token_accuracy": 0.18508963733911515,
      "num_tokens": 36881149.0,
      "step": 21255
    },
    {
      "entropy": 5.433595132827759,
      "epoch": 1.6541139856059133,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047255283218906673,
      "loss": 5.1545,
      "mean_token_accuracy": 0.1884353280067444,
      "num_tokens": 36889372.0,
      "step": 21260
    },
    {
      "entropy": 5.415424442291259,
      "epoch": 1.6545030149776307,
      "grad_norm": 1.484375,
      "learning_rate": 0.000472539564969597,
      "loss": 5.1184,
      "mean_token_accuracy": 0.18358090668916702,
      "num_tokens": 36897727.0,
      "step": 21265
    },
    {
      "entropy": 5.3983986377716064,
      "epoch": 1.6548920443493484,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047252629475277863,
      "loss": 5.1248,
      "mean_token_accuracy": 0.1870673790574074,
      "num_tokens": 36906288.0,
      "step": 21270
    },
    {
      "entropy": 5.419629430770874,
      "epoch": 1.655281073721066,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004725130215388132,
      "loss": 5.1868,
      "mean_token_accuracy": 0.1779185950756073,
      "num_tokens": 36915614.0,
      "step": 21275
    },
    {
      "entropy": 5.427306985855102,
      "epoch": 1.6556701030927834,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004724997453279021,
      "loss": 5.0514,
      "mean_token_accuracy": 0.19173242002725602,
      "num_tokens": 36923631.0,
      "step": 21280
    },
    {
      "entropy": 5.462836599349975,
      "epoch": 1.656059132464501,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004724864661202467,
      "loss": 5.2205,
      "mean_token_accuracy": 0.17995547205209733,
      "num_tokens": 36932296.0,
      "step": 21285
    },
    {
      "entropy": 5.345575952529908,
      "epoch": 1.6564481618362188,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047247318391604846,
      "loss": 5.0264,
      "mean_token_accuracy": 0.1976824313402176,
      "num_tokens": 36940785.0,
      "step": 21290
    },
    {
      "entropy": 5.453766393661499,
      "epoch": 1.656837191207936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047245989871550897,
      "loss": 5.1466,
      "mean_token_accuracy": 0.18947572410106658,
      "num_tokens": 36948797.0,
      "step": 21295
    },
    {
      "entropy": 5.478200006484985,
      "epoch": 1.6572262205796537,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004724466105188299,
      "loss": 5.2357,
      "mean_token_accuracy": 0.1832130178809166,
      "num_tokens": 36957808.0,
      "step": 21300
    },
    {
      "entropy": 5.440634393692017,
      "epoch": 1.6576152499513714,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047243331932621263,
      "loss": 5.1391,
      "mean_token_accuracy": 0.189691162109375,
      "num_tokens": 36966260.0,
      "step": 21305
    },
    {
      "entropy": 5.456794357299804,
      "epoch": 1.658004279323089,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004724200251378591,
      "loss": 5.1888,
      "mean_token_accuracy": 0.18019301146268846,
      "num_tokens": 36975436.0,
      "step": 21310
    },
    {
      "entropy": 5.544637441635132,
      "epoch": 1.6583933086948064,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004724067279539709,
      "loss": 5.2971,
      "mean_token_accuracy": 0.17637204080820085,
      "num_tokens": 36983485.0,
      "step": 21315
    },
    {
      "entropy": 5.567764568328857,
      "epoch": 1.658782338066524,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047239342777474975,
      "loss": 5.2716,
      "mean_token_accuracy": 0.17408886104822158,
      "num_tokens": 36992632.0,
      "step": 21320
    },
    {
      "entropy": 5.4704078197479244,
      "epoch": 1.6591713674382416,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047238012460039765,
      "loss": 5.0715,
      "mean_token_accuracy": 0.18698741793632506,
      "num_tokens": 37000694.0,
      "step": 21325
    },
    {
      "entropy": 5.414481830596924,
      "epoch": 1.659560396809959,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004723668184311162,
      "loss": 5.1517,
      "mean_token_accuracy": 0.18763884007930756,
      "num_tokens": 37009514.0,
      "step": 21330
    },
    {
      "entropy": 5.521216821670532,
      "epoch": 1.6599494261816767,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004723535092671074,
      "loss": 5.2767,
      "mean_token_accuracy": 0.18242395520210267,
      "num_tokens": 37019434.0,
      "step": 21335
    },
    {
      "entropy": 5.3953691005706785,
      "epoch": 1.6603384555533944,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047234019710857335,
      "loss": 5.0802,
      "mean_token_accuracy": 0.18350424021482467,
      "num_tokens": 37028673.0,
      "step": 21340
    },
    {
      "entropy": 5.381939125061035,
      "epoch": 1.6607274849251117,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004723268819557158,
      "loss": 5.1229,
      "mean_token_accuracy": 0.18639254420995713,
      "num_tokens": 37038204.0,
      "step": 21345
    },
    {
      "entropy": 5.384921026229859,
      "epoch": 1.6611165142968294,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047231356380873687,
      "loss": 5.0391,
      "mean_token_accuracy": 0.18720462024211884,
      "num_tokens": 37047301.0,
      "step": 21350
    },
    {
      "entropy": 5.447371959686279,
      "epoch": 1.661505543668547,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004723002426678388,
      "loss": 5.1375,
      "mean_token_accuracy": 0.18476439118385315,
      "num_tokens": 37054892.0,
      "step": 21355
    },
    {
      "entropy": 5.422904396057129,
      "epoch": 1.6618945730402646,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004722869185332235,
      "loss": 5.0497,
      "mean_token_accuracy": 0.18697649985551834,
      "num_tokens": 37063139.0,
      "step": 21360
    },
    {
      "entropy": 5.475736331939697,
      "epoch": 1.662283602411982,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047227359140509324,
      "loss": 5.2266,
      "mean_token_accuracy": 0.17226823717355727,
      "num_tokens": 37072019.0,
      "step": 21365
    },
    {
      "entropy": 5.483352327346802,
      "epoch": 1.6626726317836997,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004722602612836501,
      "loss": 5.2973,
      "mean_token_accuracy": 0.18010493516921997,
      "num_tokens": 37081209.0,
      "step": 21370
    },
    {
      "entropy": 5.371938562393188,
      "epoch": 1.6630616611554172,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047224692816909665,
      "loss": 5.0447,
      "mean_token_accuracy": 0.18997300565242767,
      "num_tokens": 37089464.0,
      "step": 21375
    },
    {
      "entropy": 5.338663578033447,
      "epoch": 1.6634506905271347,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004722335920616349,
      "loss": 5.0413,
      "mean_token_accuracy": 0.18887903392314911,
      "num_tokens": 37098224.0,
      "step": 21380
    },
    {
      "entropy": 5.380928039550781,
      "epoch": 1.6638397198988524,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047222025296146745,
      "loss": 5.2006,
      "mean_token_accuracy": 0.18501993417739868,
      "num_tokens": 37107039.0,
      "step": 21385
    },
    {
      "entropy": 5.469821310043335,
      "epoch": 1.66422874927057,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047220691086879643,
      "loss": 5.1969,
      "mean_token_accuracy": 0.18015396893024443,
      "num_tokens": 37116029.0,
      "step": 21390
    },
    {
      "entropy": 5.424716949462891,
      "epoch": 1.6646177786422873,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004721935657838245,
      "loss": 5.0103,
      "mean_token_accuracy": 0.1982838749885559,
      "num_tokens": 37124553.0,
      "step": 21395
    },
    {
      "entropy": 5.431406545639038,
      "epoch": 1.665006808014005,
      "grad_norm": 1.203125,
      "learning_rate": 0.000472180217706754,
      "loss": 5.1796,
      "mean_token_accuracy": 0.18586851805448532,
      "num_tokens": 37132670.0,
      "step": 21400
    },
    {
      "entropy": 5.458803606033325,
      "epoch": 1.6653958373857227,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004721668666377876,
      "loss": 5.2606,
      "mean_token_accuracy": 0.18487101644277573,
      "num_tokens": 37141534.0,
      "step": 21405
    },
    {
      "entropy": 5.463641023635864,
      "epoch": 1.6657848667574402,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047215351257712783,
      "loss": 5.1856,
      "mean_token_accuracy": 0.17626679986715316,
      "num_tokens": 37150895.0,
      "step": 21410
    },
    {
      "entropy": 5.4487415790557865,
      "epoch": 1.6661738961291577,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047214015552497735,
      "loss": 5.1156,
      "mean_token_accuracy": 0.18438707143068314,
      "num_tokens": 37159244.0,
      "step": 21415
    },
    {
      "entropy": 5.428887939453125,
      "epoch": 1.6665629255008754,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047212679548153867,
      "loss": 5.198,
      "mean_token_accuracy": 0.18428165316581727,
      "num_tokens": 37167762.0,
      "step": 21420
    },
    {
      "entropy": 5.364361810684204,
      "epoch": 1.6669519548725928,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047211343244701475,
      "loss": 5.1021,
      "mean_token_accuracy": 0.18570891618728638,
      "num_tokens": 37176470.0,
      "step": 21425
    },
    {
      "entropy": 5.480764436721802,
      "epoch": 1.6673409842443103,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004721000664216081,
      "loss": 5.2309,
      "mean_token_accuracy": 0.17923157215118407,
      "num_tokens": 37185923.0,
      "step": 21430
    },
    {
      "entropy": 5.489883184432983,
      "epoch": 1.667730013616028,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004720866974055219,
      "loss": 5.2146,
      "mean_token_accuracy": 0.17731408178806304,
      "num_tokens": 37195400.0,
      "step": 21435
    },
    {
      "entropy": 5.510724401473999,
      "epoch": 1.6681190429877457,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004720733253989585,
      "loss": 5.2531,
      "mean_token_accuracy": 0.17790626138448715,
      "num_tokens": 37203862.0,
      "step": 21440
    },
    {
      "entropy": 5.37012939453125,
      "epoch": 1.6685080723594632,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047205995040212114,
      "loss": 5.02,
      "mean_token_accuracy": 0.19433459937572478,
      "num_tokens": 37212518.0,
      "step": 21445
    },
    {
      "entropy": 5.544953870773315,
      "epoch": 1.6688971017311807,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004720465724152127,
      "loss": 5.2487,
      "mean_token_accuracy": 0.17534533590078355,
      "num_tokens": 37221342.0,
      "step": 21450
    },
    {
      "entropy": 5.4481607437133786,
      "epoch": 1.6692861311028984,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047203319143843613,
      "loss": 5.174,
      "mean_token_accuracy": 0.1844106510281563,
      "num_tokens": 37230477.0,
      "step": 21455
    },
    {
      "entropy": 5.430639839172363,
      "epoch": 1.6696751604746158,
      "grad_norm": 1.25,
      "learning_rate": 0.0004720198074719946,
      "loss": 5.1289,
      "mean_token_accuracy": 0.18741045743227006,
      "num_tokens": 37238590.0,
      "step": 21460
    },
    {
      "entropy": 5.423798704147339,
      "epoch": 1.6700641898463333,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047200642051609094,
      "loss": 5.164,
      "mean_token_accuracy": 0.18442320376634597,
      "num_tokens": 37248065.0,
      "step": 21465
    },
    {
      "entropy": 5.461537170410156,
      "epoch": 1.670453219218051,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004719930305709285,
      "loss": 5.1544,
      "mean_token_accuracy": 0.188926000893116,
      "num_tokens": 37256467.0,
      "step": 21470
    },
    {
      "entropy": 5.460478687286377,
      "epoch": 1.6708422485897685,
      "grad_norm": 1.25,
      "learning_rate": 0.00047197963763671033,
      "loss": 5.1476,
      "mean_token_accuracy": 0.18160920441150666,
      "num_tokens": 37264255.0,
      "step": 21475
    },
    {
      "entropy": 5.43180627822876,
      "epoch": 1.671231277961486,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004719662417136397,
      "loss": 5.0613,
      "mean_token_accuracy": 0.18735998272895812,
      "num_tokens": 37272671.0,
      "step": 21480
    },
    {
      "entropy": 5.394109773635864,
      "epoch": 1.6716203073332037,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047195284280191987,
      "loss": 5.1737,
      "mean_token_accuracy": 0.18495950400829314,
      "num_tokens": 37281895.0,
      "step": 21485
    },
    {
      "entropy": 5.35521764755249,
      "epoch": 1.6720093367049214,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004719394409017541,
      "loss": 5.0715,
      "mean_token_accuracy": 0.1959013670682907,
      "num_tokens": 37290098.0,
      "step": 21490
    },
    {
      "entropy": 5.444878816604614,
      "epoch": 1.6723983660766388,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047192603601334585,
      "loss": 5.1084,
      "mean_token_accuracy": 0.18685898184776306,
      "num_tokens": 37299494.0,
      "step": 21495
    },
    {
      "entropy": 5.465498638153076,
      "epoch": 1.6727873954483563,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047191262813689835,
      "loss": 5.1708,
      "mean_token_accuracy": 0.18296178430318832,
      "num_tokens": 37308366.0,
      "step": 21500
    },
    {
      "entropy": 5.413819646835327,
      "epoch": 1.673176424820074,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004718992172726152,
      "loss": 5.1612,
      "mean_token_accuracy": 0.18549337089061738,
      "num_tokens": 37316774.0,
      "step": 21505
    },
    {
      "entropy": 5.458289003372192,
      "epoch": 1.6735654541917915,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047188580342069983,
      "loss": 5.1863,
      "mean_token_accuracy": 0.181383615732193,
      "num_tokens": 37325388.0,
      "step": 21510
    },
    {
      "entropy": 5.522270917892456,
      "epoch": 1.673954483563509,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004718723865813557,
      "loss": 5.2136,
      "mean_token_accuracy": 0.1795085296034813,
      "num_tokens": 37333476.0,
      "step": 21515
    },
    {
      "entropy": 5.53579044342041,
      "epoch": 1.6743435129352267,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004718589667547865,
      "loss": 5.2709,
      "mean_token_accuracy": 0.18239307254552842,
      "num_tokens": 37342556.0,
      "step": 21520
    },
    {
      "entropy": 5.495479393005371,
      "epoch": 1.6747325423069441,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004718455439411958,
      "loss": 5.1591,
      "mean_token_accuracy": 0.19303099066019058,
      "num_tokens": 37351082.0,
      "step": 21525
    },
    {
      "entropy": 5.491023969650269,
      "epoch": 1.6751215716786616,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004718321181407873,
      "loss": 5.138,
      "mean_token_accuracy": 0.18363266736268996,
      "num_tokens": 37359083.0,
      "step": 21530
    },
    {
      "entropy": 5.39815731048584,
      "epoch": 1.6755106010503793,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004718186893537647,
      "loss": 5.1273,
      "mean_token_accuracy": 0.18681024014949799,
      "num_tokens": 37367982.0,
      "step": 21535
    },
    {
      "entropy": 5.477026891708374,
      "epoch": 1.675899630422097,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004718052575803318,
      "loss": 5.207,
      "mean_token_accuracy": 0.1800020933151245,
      "num_tokens": 37376932.0,
      "step": 21540
    },
    {
      "entropy": 5.469567775726318,
      "epoch": 1.6762886597938145,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004717918228206923,
      "loss": 5.215,
      "mean_token_accuracy": 0.18191076219081878,
      "num_tokens": 37386248.0,
      "step": 21545
    },
    {
      "entropy": 5.473011779785156,
      "epoch": 1.676677689165532,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004717783850750501,
      "loss": 5.1328,
      "mean_token_accuracy": 0.18218936771154404,
      "num_tokens": 37395412.0,
      "step": 21550
    },
    {
      "entropy": 5.436119794845581,
      "epoch": 1.6770667185372496,
      "grad_norm": 1.375,
      "learning_rate": 0.0004717649443436091,
      "loss": 5.0999,
      "mean_token_accuracy": 0.1876216173171997,
      "num_tokens": 37404332.0,
      "step": 21555
    },
    {
      "entropy": 5.474289417266846,
      "epoch": 1.6774557479089671,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047175150062657336,
      "loss": 5.2014,
      "mean_token_accuracy": 0.18636367321014405,
      "num_tokens": 37413365.0,
      "step": 21560
    },
    {
      "entropy": 5.47145414352417,
      "epoch": 1.6778447772806846,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047173805392414664,
      "loss": 5.0937,
      "mean_token_accuracy": 0.184091779589653,
      "num_tokens": 37422164.0,
      "step": 21565
    },
    {
      "entropy": 5.448068571090698,
      "epoch": 1.6782338066524023,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004717246042365332,
      "loss": 5.1833,
      "mean_token_accuracy": 0.1809279814362526,
      "num_tokens": 37431092.0,
      "step": 21570
    },
    {
      "entropy": 5.44748330116272,
      "epoch": 1.6786228360241198,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047171115156393695,
      "loss": 5.2675,
      "mean_token_accuracy": 0.17721026688814162,
      "num_tokens": 37440832.0,
      "step": 21575
    },
    {
      "entropy": 5.456909799575806,
      "epoch": 1.6790118653958372,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047169769590656207,
      "loss": 5.1799,
      "mean_token_accuracy": 0.18156476616859435,
      "num_tokens": 37449317.0,
      "step": 21580
    },
    {
      "entropy": 5.512399101257325,
      "epoch": 1.679400894767555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047168423726461273,
      "loss": 5.2147,
      "mean_token_accuracy": 0.17965819388628007,
      "num_tokens": 37457820.0,
      "step": 21585
    },
    {
      "entropy": 5.440832233428955,
      "epoch": 1.6797899241392726,
      "grad_norm": 1.375,
      "learning_rate": 0.0004716707756382931,
      "loss": 5.0568,
      "mean_token_accuracy": 0.1934329405426979,
      "num_tokens": 37466378.0,
      "step": 21590
    },
    {
      "entropy": 5.438096761703491,
      "epoch": 1.6801789535109901,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047165731102780757,
      "loss": 5.1962,
      "mean_token_accuracy": 0.17562576085329057,
      "num_tokens": 37475114.0,
      "step": 21595
    },
    {
      "entropy": 5.443257570266724,
      "epoch": 1.6805679828827076,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047164384343336037,
      "loss": 5.2217,
      "mean_token_accuracy": 0.17551061511039734,
      "num_tokens": 37483898.0,
      "step": 21600
    },
    {
      "entropy": 5.47622127532959,
      "epoch": 1.6809570122544253,
      "grad_norm": 1.25,
      "learning_rate": 0.00047163037285515583,
      "loss": 5.2077,
      "mean_token_accuracy": 0.18221314996480942,
      "num_tokens": 37492420.0,
      "step": 21605
    },
    {
      "entropy": 5.415828895568848,
      "epoch": 1.6813460416261428,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004716168992933982,
      "loss": 5.1011,
      "mean_token_accuracy": 0.191717991232872,
      "num_tokens": 37501090.0,
      "step": 21610
    },
    {
      "entropy": 5.4700616836547855,
      "epoch": 1.6817350709978602,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047160342274829234,
      "loss": 5.2373,
      "mean_token_accuracy": 0.1767852634191513,
      "num_tokens": 37510465.0,
      "step": 21615
    },
    {
      "entropy": 5.435352087020874,
      "epoch": 1.682124100369578,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047158994322004223,
      "loss": 5.1407,
      "mean_token_accuracy": 0.18671649992465972,
      "num_tokens": 37519155.0,
      "step": 21620
    },
    {
      "entropy": 5.446352815628051,
      "epoch": 1.6825131297412956,
      "grad_norm": 1.25,
      "learning_rate": 0.0004715764607088527,
      "loss": 5.1363,
      "mean_token_accuracy": 0.18951668888330458,
      "num_tokens": 37527134.0,
      "step": 21625
    },
    {
      "entropy": 5.452503108978272,
      "epoch": 1.6829021591130129,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004715629752149283,
      "loss": 5.1719,
      "mean_token_accuracy": 0.18442397862672805,
      "num_tokens": 37536087.0,
      "step": 21630
    },
    {
      "entropy": 5.410755586624146,
      "epoch": 1.6832911884847306,
      "grad_norm": 1.375,
      "learning_rate": 0.00047154948673847356,
      "loss": 5.107,
      "mean_token_accuracy": 0.18967499434947968,
      "num_tokens": 37543896.0,
      "step": 21635
    },
    {
      "entropy": 5.456842756271362,
      "epoch": 1.6836802178564483,
      "grad_norm": 1.375,
      "learning_rate": 0.0004715359952796933,
      "loss": 5.1811,
      "mean_token_accuracy": 0.18201919496059418,
      "num_tokens": 37552571.0,
      "step": 21640
    },
    {
      "entropy": 5.50685338973999,
      "epoch": 1.6840692472281658,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047152250083879203,
      "loss": 5.2068,
      "mean_token_accuracy": 0.18273255079984665,
      "num_tokens": 37560823.0,
      "step": 21645
    },
    {
      "entropy": 5.403844976425171,
      "epoch": 1.6844582765998832,
      "grad_norm": 1.375,
      "learning_rate": 0.00047150900341597464,
      "loss": 5.0781,
      "mean_token_accuracy": 0.18405091166496276,
      "num_tokens": 37568312.0,
      "step": 21650
    },
    {
      "entropy": 5.411937522888183,
      "epoch": 1.684847305971601,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004714955030114459,
      "loss": 5.1463,
      "mean_token_accuracy": 0.19270649254322053,
      "num_tokens": 37577008.0,
      "step": 21655
    },
    {
      "entropy": 5.5054105758667,
      "epoch": 1.6852363353433184,
      "grad_norm": 1.25,
      "learning_rate": 0.00047148199962541065,
      "loss": 5.1738,
      "mean_token_accuracy": 0.1781419962644577,
      "num_tokens": 37586252.0,
      "step": 21660
    },
    {
      "entropy": 5.460412120819091,
      "epoch": 1.6856253647150359,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004714684932580738,
      "loss": 5.1512,
      "mean_token_accuracy": 0.18525373041629792,
      "num_tokens": 37594699.0,
      "step": 21665
    },
    {
      "entropy": 5.4064796447753904,
      "epoch": 1.6860143940867536,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004714549839096403,
      "loss": 5.1288,
      "mean_token_accuracy": 0.18649631291627883,
      "num_tokens": 37603891.0,
      "step": 21670
    },
    {
      "entropy": 5.451591634750367,
      "epoch": 1.6864034234584713,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047144147158031507,
      "loss": 5.1514,
      "mean_token_accuracy": 0.1827014446258545,
      "num_tokens": 37612809.0,
      "step": 21675
    },
    {
      "entropy": 5.488323068618774,
      "epoch": 1.6867924528301885,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004714279562703032,
      "loss": 5.0923,
      "mean_token_accuracy": 0.18619071394205094,
      "num_tokens": 37621787.0,
      "step": 21680
    },
    {
      "entropy": 5.38452181816101,
      "epoch": 1.6871814822019062,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004714144379798098,
      "loss": 5.045,
      "mean_token_accuracy": 0.19893213361501694,
      "num_tokens": 37629796.0,
      "step": 21685
    },
    {
      "entropy": 5.44609751701355,
      "epoch": 1.687570511573624,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047140091670903986,
      "loss": 5.1042,
      "mean_token_accuracy": 0.1861880525946617,
      "num_tokens": 37638467.0,
      "step": 21690
    },
    {
      "entropy": 5.418607711791992,
      "epoch": 1.6879595409453414,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047138739245819857,
      "loss": 5.1638,
      "mean_token_accuracy": 0.183639857172966,
      "num_tokens": 37647159.0,
      "step": 21695
    },
    {
      "entropy": 5.427438449859619,
      "epoch": 1.6883485703170589,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047137386522749124,
      "loss": 5.131,
      "mean_token_accuracy": 0.17802622318267822,
      "num_tokens": 37655986.0,
      "step": 21700
    },
    {
      "entropy": 5.387999153137207,
      "epoch": 1.6887375996887766,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004713603350171231,
      "loss": 5.0458,
      "mean_token_accuracy": 0.19104453325271606,
      "num_tokens": 37663932.0,
      "step": 21705
    },
    {
      "entropy": 5.488088130950928,
      "epoch": 1.689126629060494,
      "grad_norm": 1.25,
      "learning_rate": 0.00047134680182729926,
      "loss": 5.1902,
      "mean_token_accuracy": 0.1780229166150093,
      "num_tokens": 37672142.0,
      "step": 21710
    },
    {
      "entropy": 5.470323944091797,
      "epoch": 1.6895156584322115,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004713332656582254,
      "loss": 5.0824,
      "mean_token_accuracy": 0.18836161941289903,
      "num_tokens": 37680569.0,
      "step": 21715
    },
    {
      "entropy": 5.394380807876587,
      "epoch": 1.6899046878039292,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004713197265101066,
      "loss": 5.1781,
      "mean_token_accuracy": 0.17837965041399,
      "num_tokens": 37688852.0,
      "step": 21720
    },
    {
      "entropy": 5.455995225906372,
      "epoch": 1.690293717175647,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047130618438314856,
      "loss": 5.1417,
      "mean_token_accuracy": 0.18777900636196138,
      "num_tokens": 37696900.0,
      "step": 21725
    },
    {
      "entropy": 5.520359039306641,
      "epoch": 1.6906827465473642,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004712926392775565,
      "loss": 5.2853,
      "mean_token_accuracy": 0.1797661527991295,
      "num_tokens": 37706292.0,
      "step": 21730
    },
    {
      "entropy": 5.4113188743591305,
      "epoch": 1.6910717759190819,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004712790911935362,
      "loss": 5.0791,
      "mean_token_accuracy": 0.18715739846229554,
      "num_tokens": 37715025.0,
      "step": 21735
    },
    {
      "entropy": 5.4513444900512695,
      "epoch": 1.6914608052907996,
      "grad_norm": 1.21875,
      "learning_rate": 0.000471265540131293,
      "loss": 5.2119,
      "mean_token_accuracy": 0.1773126021027565,
      "num_tokens": 37724124.0,
      "step": 21740
    },
    {
      "entropy": 5.42948055267334,
      "epoch": 1.691849834662517,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047125198609103267,
      "loss": 5.1591,
      "mean_token_accuracy": 0.18089642077684404,
      "num_tokens": 37732689.0,
      "step": 21745
    },
    {
      "entropy": 5.455805778503418,
      "epoch": 1.6922388640342345,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004712384290729607,
      "loss": 5.1152,
      "mean_token_accuracy": 0.18241401463747026,
      "num_tokens": 37741206.0,
      "step": 21750
    },
    {
      "entropy": 5.434431505203247,
      "epoch": 1.6926278934059522,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000471224869077283,
      "loss": 5.1169,
      "mean_token_accuracy": 0.18820763975381852,
      "num_tokens": 37749549.0,
      "step": 21755
    },
    {
      "entropy": 5.318625640869141,
      "epoch": 1.6930169227776697,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047121130610420527,
      "loss": 4.9356,
      "mean_token_accuracy": 0.19977205246686935,
      "num_tokens": 37758332.0,
      "step": 21760
    },
    {
      "entropy": 5.402372264862061,
      "epoch": 1.6934059521493872,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047119774015393323,
      "loss": 5.1414,
      "mean_token_accuracy": 0.18190839141607285,
      "num_tokens": 37767368.0,
      "step": 21765
    },
    {
      "entropy": 5.439834785461426,
      "epoch": 1.6937949815211049,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047118417122667285,
      "loss": 5.1151,
      "mean_token_accuracy": 0.18551575839519502,
      "num_tokens": 37776426.0,
      "step": 21770
    },
    {
      "entropy": 5.533604383468628,
      "epoch": 1.6941840108928226,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004711705993226298,
      "loss": 5.2741,
      "mean_token_accuracy": 0.17538267076015474,
      "num_tokens": 37784464.0,
      "step": 21775
    },
    {
      "entropy": 5.378828668594361,
      "epoch": 1.6945730402645398,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004711570244420102,
      "loss": 5.1061,
      "mean_token_accuracy": 0.17925872057676315,
      "num_tokens": 37794307.0,
      "step": 21780
    },
    {
      "entropy": 5.456859493255616,
      "epoch": 1.6949620696362575,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004711434465850199,
      "loss": 5.1918,
      "mean_token_accuracy": 0.17484943717718124,
      "num_tokens": 37803151.0,
      "step": 21785
    },
    {
      "entropy": 5.408019256591797,
      "epoch": 1.6953510990079752,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004711298657518651,
      "loss": 5.1662,
      "mean_token_accuracy": 0.18202330470085143,
      "num_tokens": 37812433.0,
      "step": 21790
    },
    {
      "entropy": 5.46373438835144,
      "epoch": 1.6957401283796927,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004711162819427518,
      "loss": 5.0961,
      "mean_token_accuracy": 0.18412057906389237,
      "num_tokens": 37820627.0,
      "step": 21795
    },
    {
      "entropy": 5.461150360107422,
      "epoch": 1.6961291577514102,
      "grad_norm": 1.28125,
      "learning_rate": 0.000471102695157886,
      "loss": 5.1466,
      "mean_token_accuracy": 0.18467850536108016,
      "num_tokens": 37829015.0,
      "step": 21800
    },
    {
      "entropy": 5.470170164108277,
      "epoch": 1.6965181871231279,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004710891053974739,
      "loss": 5.209,
      "mean_token_accuracy": 0.18138865679502486,
      "num_tokens": 37837097.0,
      "step": 21805
    },
    {
      "entropy": 5.46840181350708,
      "epoch": 1.6969072164948453,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004710755126617217,
      "loss": 5.1742,
      "mean_token_accuracy": 0.17659877985715866,
      "num_tokens": 37845414.0,
      "step": 21810
    },
    {
      "entropy": 5.521837282180786,
      "epoch": 1.6972962458665628,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004710619169508358,
      "loss": 5.1684,
      "mean_token_accuracy": 0.1857956036925316,
      "num_tokens": 37855006.0,
      "step": 21815
    },
    {
      "entropy": 5.554569578170776,
      "epoch": 1.6976852752382805,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004710483182650223,
      "loss": 5.3044,
      "mean_token_accuracy": 0.17741592824459076,
      "num_tokens": 37863611.0,
      "step": 21820
    },
    {
      "entropy": 5.437148761749268,
      "epoch": 1.6980743046099982,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047103471660448767,
      "loss": 5.1849,
      "mean_token_accuracy": 0.18329339027404784,
      "num_tokens": 37871935.0,
      "step": 21825
    },
    {
      "entropy": 5.417216777801514,
      "epoch": 1.6984633339817157,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047102111196943813,
      "loss": 5.1475,
      "mean_token_accuracy": 0.1842946633696556,
      "num_tokens": 37880760.0,
      "step": 21830
    },
    {
      "entropy": 5.473556089401245,
      "epoch": 1.6988523633534331,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047100750436008035,
      "loss": 5.0983,
      "mean_token_accuracy": 0.18626521527767181,
      "num_tokens": 37889457.0,
      "step": 21835
    },
    {
      "entropy": 5.47004280090332,
      "epoch": 1.6992413927251508,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047099389377662054,
      "loss": 5.1443,
      "mean_token_accuracy": 0.18277855068445206,
      "num_tokens": 37897805.0,
      "step": 21840
    },
    {
      "entropy": 5.429114866256714,
      "epoch": 1.6996304220968683,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004709802802192654,
      "loss": 5.2114,
      "mean_token_accuracy": 0.17950051426887512,
      "num_tokens": 37906292.0,
      "step": 21845
    },
    {
      "entropy": 5.427227735519409,
      "epoch": 1.7000194514685858,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047096666368822153,
      "loss": 5.1139,
      "mean_token_accuracy": 0.18379775881767274,
      "num_tokens": 37914513.0,
      "step": 21850
    },
    {
      "entropy": 5.466753578186035,
      "epoch": 1.7004084808403035,
      "grad_norm": 1.25,
      "learning_rate": 0.00047095304418369536,
      "loss": 5.1231,
      "mean_token_accuracy": 0.18563194572925568,
      "num_tokens": 37923097.0,
      "step": 21855
    },
    {
      "entropy": 5.481782674789429,
      "epoch": 1.700797510212021,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004709394217058936,
      "loss": 5.1754,
      "mean_token_accuracy": 0.18829378932714463,
      "num_tokens": 37931617.0,
      "step": 21860
    },
    {
      "entropy": 5.474587917327881,
      "epoch": 1.7011865395837384,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004709257962550231,
      "loss": 5.125,
      "mean_token_accuracy": 0.189526928961277,
      "num_tokens": 37939777.0,
      "step": 21865
    },
    {
      "entropy": 5.408878231048584,
      "epoch": 1.7015755689554561,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047091216783129035,
      "loss": 5.1649,
      "mean_token_accuracy": 0.18576932996511458,
      "num_tokens": 37949375.0,
      "step": 21870
    },
    {
      "entropy": 5.437985038757324,
      "epoch": 1.7019645983271738,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004708985364349024,
      "loss": 5.1181,
      "mean_token_accuracy": 0.19108730107545852,
      "num_tokens": 37958724.0,
      "step": 21875
    },
    {
      "entropy": 5.426161575317383,
      "epoch": 1.7023536276988913,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047088490206606586,
      "loss": 5.1131,
      "mean_token_accuracy": 0.1854175001382828,
      "num_tokens": 37967646.0,
      "step": 21880
    },
    {
      "entropy": 5.409465360641479,
      "epoch": 1.7027426570706088,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047087126472498785,
      "loss": 5.1838,
      "mean_token_accuracy": 0.18243782222270966,
      "num_tokens": 37976058.0,
      "step": 21885
    },
    {
      "entropy": 5.370332098007202,
      "epoch": 1.7031316864423265,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000470857624411875,
      "loss": 5.0915,
      "mean_token_accuracy": 0.18778082728385925,
      "num_tokens": 37984895.0,
      "step": 21890
    },
    {
      "entropy": 5.433343887329102,
      "epoch": 1.703520715814044,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047084398112693456,
      "loss": 5.0589,
      "mean_token_accuracy": 0.18817091286182402,
      "num_tokens": 37993514.0,
      "step": 21895
    },
    {
      "entropy": 5.466991710662842,
      "epoch": 1.7039097451857614,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004708303348703734,
      "loss": 5.1258,
      "mean_token_accuracy": 0.18037586510181428,
      "num_tokens": 38002274.0,
      "step": 21900
    },
    {
      "entropy": 5.419135665893554,
      "epoch": 1.7042987745574791,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004708166856423986,
      "loss": 5.0272,
      "mean_token_accuracy": 0.18908049464225768,
      "num_tokens": 38010336.0,
      "step": 21905
    },
    {
      "entropy": 5.426383113861084,
      "epoch": 1.7046878039291966,
      "grad_norm": 1.25,
      "learning_rate": 0.00047080303344321727,
      "loss": 5.1034,
      "mean_token_accuracy": 0.18969349712133407,
      "num_tokens": 38019166.0,
      "step": 21910
    },
    {
      "entropy": 5.357455539703369,
      "epoch": 1.705076833300914,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047078937827303653,
      "loss": 5.1533,
      "mean_token_accuracy": 0.18501580506563187,
      "num_tokens": 38027634.0,
      "step": 21915
    },
    {
      "entropy": 5.472870969772339,
      "epoch": 1.7054658626726318,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004707757201320636,
      "loss": 5.1494,
      "mean_token_accuracy": 0.18131695836782455,
      "num_tokens": 38036611.0,
      "step": 21920
    },
    {
      "entropy": 5.394597673416138,
      "epoch": 1.7058548920443495,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047076205902050577,
      "loss": 5.0512,
      "mean_token_accuracy": 0.18909388035535812,
      "num_tokens": 38045815.0,
      "step": 21925
    },
    {
      "entropy": 5.490310287475586,
      "epoch": 1.706243921416067,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047074839493857024,
      "loss": 5.2238,
      "mean_token_accuracy": 0.17757856696844102,
      "num_tokens": 38054973.0,
      "step": 21930
    },
    {
      "entropy": 5.471278190612793,
      "epoch": 1.7066329507877844,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004707347278864644,
      "loss": 5.0979,
      "mean_token_accuracy": 0.1889703869819641,
      "num_tokens": 38064084.0,
      "step": 21935
    },
    {
      "entropy": 5.410388326644897,
      "epoch": 1.7070219801595021,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047072105786439563,
      "loss": 5.1262,
      "mean_token_accuracy": 0.1869691103696823,
      "num_tokens": 38073838.0,
      "step": 21940
    },
    {
      "entropy": 5.508942031860352,
      "epoch": 1.7074110095312196,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047070738487257137,
      "loss": 5.2171,
      "mean_token_accuracy": 0.17555944472551346,
      "num_tokens": 38082531.0,
      "step": 21945
    },
    {
      "entropy": 5.495730876922607,
      "epoch": 1.707800038902937,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047069370891119895,
      "loss": 5.222,
      "mean_token_accuracy": 0.1795056328177452,
      "num_tokens": 38091499.0,
      "step": 21950
    },
    {
      "entropy": 5.449326086044311,
      "epoch": 1.7081890682746548,
      "grad_norm": 1.171875,
      "learning_rate": 0.000470680029980486,
      "loss": 5.0479,
      "mean_token_accuracy": 0.20004933327436447,
      "num_tokens": 38099488.0,
      "step": 21955
    },
    {
      "entropy": 5.481084489822388,
      "epoch": 1.7085780976463723,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047066634808064006,
      "loss": 5.1757,
      "mean_token_accuracy": 0.1800842046737671,
      "num_tokens": 38107535.0,
      "step": 21960
    },
    {
      "entropy": 5.46864948272705,
      "epoch": 1.7089671270180897,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047065266321186873,
      "loss": 5.1349,
      "mean_token_accuracy": 0.18380848169326783,
      "num_tokens": 38116457.0,
      "step": 21965
    },
    {
      "entropy": 5.4725213050842285,
      "epoch": 1.7093561563898074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004706389753743797,
      "loss": 5.0609,
      "mean_token_accuracy": 0.19020908027887345,
      "num_tokens": 38124571.0,
      "step": 21970
    },
    {
      "entropy": 5.357316017150879,
      "epoch": 1.7097451857615251,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004706252845683805,
      "loss": 5.0997,
      "mean_token_accuracy": 0.18343816697597504,
      "num_tokens": 38132789.0,
      "step": 21975
    },
    {
      "entropy": 5.303881597518921,
      "epoch": 1.7101342151332426,
      "grad_norm": 1.25,
      "learning_rate": 0.00047061159079407903,
      "loss": 5.0069,
      "mean_token_accuracy": 0.1901726931333542,
      "num_tokens": 38141413.0,
      "step": 21980
    },
    {
      "entropy": 5.393996286392212,
      "epoch": 1.71052324450496,
      "grad_norm": 1.171875,
      "learning_rate": 0.000470597894051683,
      "loss": 5.0477,
      "mean_token_accuracy": 0.1885794773697853,
      "num_tokens": 38149869.0,
      "step": 21985
    },
    {
      "entropy": 5.361879491806031,
      "epoch": 1.7109122738766778,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004705841943414003,
      "loss": 4.9809,
      "mean_token_accuracy": 0.20070360600948334,
      "num_tokens": 38157356.0,
      "step": 21990
    },
    {
      "entropy": 5.4140547752380375,
      "epoch": 1.7113013032483952,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047057049166343876,
      "loss": 5.1517,
      "mean_token_accuracy": 0.18470899611711503,
      "num_tokens": 38166125.0,
      "step": 21995
    },
    {
      "entropy": 5.423274898529053,
      "epoch": 1.7116903326201127,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047055678601800623,
      "loss": 5.0924,
      "mean_token_accuracy": 0.18635322898626328,
      "num_tokens": 38174171.0,
      "step": 22000
    },
    {
      "entropy": 5.502792072296143,
      "epoch": 1.7120793619918304,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047054307740531076,
      "loss": 5.194,
      "mean_token_accuracy": 0.18177500367164612,
      "num_tokens": 38182755.0,
      "step": 22005
    },
    {
      "entropy": 5.4396730899810795,
      "epoch": 1.712468391363548,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047052936582556035,
      "loss": 5.1303,
      "mean_token_accuracy": 0.18818821161985397,
      "num_tokens": 38191289.0,
      "step": 22010
    },
    {
      "entropy": 5.4758796215057375,
      "epoch": 1.7128574207352654,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047051565127896307,
      "loss": 5.2686,
      "mean_token_accuracy": 0.1749943882226944,
      "num_tokens": 38200624.0,
      "step": 22015
    },
    {
      "entropy": 5.49514594078064,
      "epoch": 1.713246450106983,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047050193376572686,
      "loss": 5.0681,
      "mean_token_accuracy": 0.1872066229581833,
      "num_tokens": 38209272.0,
      "step": 22020
    },
    {
      "entropy": 5.492700862884521,
      "epoch": 1.7136354794787008,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047048821328606,
      "loss": 5.236,
      "mean_token_accuracy": 0.1797465831041336,
      "num_tokens": 38218851.0,
      "step": 22025
    },
    {
      "entropy": 5.425061511993408,
      "epoch": 1.7140245088504182,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004704744898401708,
      "loss": 5.1354,
      "mean_token_accuracy": 0.18684813529253005,
      "num_tokens": 38227367.0,
      "step": 22030
    },
    {
      "entropy": 5.481336402893066,
      "epoch": 1.7144135382221357,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004704607634282672,
      "loss": 5.1309,
      "mean_token_accuracy": 0.18998834788799285,
      "num_tokens": 38235721.0,
      "step": 22035
    },
    {
      "entropy": 5.497761058807373,
      "epoch": 1.7148025675938534,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047044703405055765,
      "loss": 5.2155,
      "mean_token_accuracy": 0.17843061834573745,
      "num_tokens": 38244686.0,
      "step": 22040
    },
    {
      "entropy": 5.472004461288452,
      "epoch": 1.7151915969655709,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047043330170725046,
      "loss": 5.2024,
      "mean_token_accuracy": 0.18386606723070145,
      "num_tokens": 38253609.0,
      "step": 22045
    },
    {
      "entropy": 5.438859176635742,
      "epoch": 1.7155806263372884,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047041956639855406,
      "loss": 5.1293,
      "mean_token_accuracy": 0.18245308995246887,
      "num_tokens": 38261746.0,
      "step": 22050
    },
    {
      "entropy": 5.446011352539062,
      "epoch": 1.715969655709006,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004704058281246766,
      "loss": 5.1383,
      "mean_token_accuracy": 0.18560929000377654,
      "num_tokens": 38270573.0,
      "step": 22055
    },
    {
      "entropy": 5.431704902648926,
      "epoch": 1.7163586850807238,
      "grad_norm": 1.25,
      "learning_rate": 0.0004703920868858268,
      "loss": 5.133,
      "mean_token_accuracy": 0.18807685673236846,
      "num_tokens": 38279031.0,
      "step": 22060
    },
    {
      "entropy": 5.408733367919922,
      "epoch": 1.716747714452441,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047037834268221315,
      "loss": 5.1736,
      "mean_token_accuracy": 0.18206825703382493,
      "num_tokens": 38287060.0,
      "step": 22065
    },
    {
      "entropy": 5.471729135513305,
      "epoch": 1.7171367438241587,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004703645955140441,
      "loss": 5.1455,
      "mean_token_accuracy": 0.1903379127383232,
      "num_tokens": 38295192.0,
      "step": 22070
    },
    {
      "entropy": 5.425918292999268,
      "epoch": 1.7175257731958764,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047035084538152815,
      "loss": 5.0662,
      "mean_token_accuracy": 0.18836621046066285,
      "num_tokens": 38303932.0,
      "step": 22075
    },
    {
      "entropy": 5.435115051269531,
      "epoch": 1.7179148025675939,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004703370922848742,
      "loss": 5.2182,
      "mean_token_accuracy": 0.1749226674437523,
      "num_tokens": 38312906.0,
      "step": 22080
    },
    {
      "entropy": 5.440869665145874,
      "epoch": 1.7183038319393114,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047032333622429074,
      "loss": 5.2034,
      "mean_token_accuracy": 0.17831271141767502,
      "num_tokens": 38322359.0,
      "step": 22085
    },
    {
      "entropy": 5.580415821075439,
      "epoch": 1.718692861311029,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047030957719998656,
      "loss": 5.245,
      "mean_token_accuracy": 0.17385976612567902,
      "num_tokens": 38331880.0,
      "step": 22090
    },
    {
      "entropy": 5.505927610397339,
      "epoch": 1.7190818906827465,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004702958152121704,
      "loss": 5.323,
      "mean_token_accuracy": 0.1714931845664978,
      "num_tokens": 38340763.0,
      "step": 22095
    },
    {
      "entropy": 5.463445329666138,
      "epoch": 1.719470920054464,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004702820502610511,
      "loss": 5.1265,
      "mean_token_accuracy": 0.18389199376106263,
      "num_tokens": 38349427.0,
      "step": 22100
    },
    {
      "entropy": 5.4333680152893065,
      "epoch": 1.7198599494261817,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004702682823468375,
      "loss": 5.1,
      "mean_token_accuracy": 0.19248313456773758,
      "num_tokens": 38358025.0,
      "step": 22105
    },
    {
      "entropy": 5.454819297790527,
      "epoch": 1.7202489787978994,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047025451146973844,
      "loss": 5.091,
      "mean_token_accuracy": 0.18219396024942397,
      "num_tokens": 38366923.0,
      "step": 22110
    },
    {
      "entropy": 5.479520177841186,
      "epoch": 1.7206380081696167,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047024073762996305,
      "loss": 5.176,
      "mean_token_accuracy": 0.187580443918705,
      "num_tokens": 38375397.0,
      "step": 22115
    },
    {
      "entropy": 5.420974636077881,
      "epoch": 1.7210270375413343,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004702269608277202,
      "loss": 5.0844,
      "mean_token_accuracy": 0.19060172885656357,
      "num_tokens": 38384279.0,
      "step": 22120
    },
    {
      "entropy": 5.466910266876221,
      "epoch": 1.721416066913052,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004702131810632189,
      "loss": 5.1235,
      "mean_token_accuracy": 0.18644751757383346,
      "num_tokens": 38392087.0,
      "step": 22125
    },
    {
      "entropy": 5.442948341369629,
      "epoch": 1.7218050962847695,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047019939833666837,
      "loss": 5.2512,
      "mean_token_accuracy": 0.1766302242875099,
      "num_tokens": 38400616.0,
      "step": 22130
    },
    {
      "entropy": 5.489051723480225,
      "epoch": 1.722194125656487,
      "grad_norm": 1.25,
      "learning_rate": 0.0004701856126482776,
      "loss": 5.1335,
      "mean_token_accuracy": 0.18316368758678436,
      "num_tokens": 38408975.0,
      "step": 22135
    },
    {
      "entropy": 5.464526557922364,
      "epoch": 1.7225831550282047,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004701718239982559,
      "loss": 5.2475,
      "mean_token_accuracy": 0.1856542006134987,
      "num_tokens": 38417161.0,
      "step": 22140
    },
    {
      "entropy": 5.500979328155518,
      "epoch": 1.7229721843999222,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047015803238681234,
      "loss": 5.2012,
      "mean_token_accuracy": 0.18200944364070892,
      "num_tokens": 38425078.0,
      "step": 22145
    },
    {
      "entropy": 5.445772933959961,
      "epoch": 1.7233612137716396,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004701442378141563,
      "loss": 5.086,
      "mean_token_accuracy": 0.18872628957033158,
      "num_tokens": 38433347.0,
      "step": 22150
    },
    {
      "entropy": 5.528460931777954,
      "epoch": 1.7237502431433573,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000470130440280497,
      "loss": 5.1985,
      "mean_token_accuracy": 0.17880051583051682,
      "num_tokens": 38442373.0,
      "step": 22155
    },
    {
      "entropy": 5.50367488861084,
      "epoch": 1.724139272515075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004701166397860439,
      "loss": 5.2713,
      "mean_token_accuracy": 0.17187037020921708,
      "num_tokens": 38452325.0,
      "step": 22160
    },
    {
      "entropy": 5.508620977401733,
      "epoch": 1.7245283018867923,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004701028363310064,
      "loss": 5.257,
      "mean_token_accuracy": 0.1799311012029648,
      "num_tokens": 38462317.0,
      "step": 22165
    },
    {
      "entropy": 5.458264398574829,
      "epoch": 1.72491733125851,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047008902991559385,
      "loss": 5.0771,
      "mean_token_accuracy": 0.18792858421802522,
      "num_tokens": 38471149.0,
      "step": 22170
    },
    {
      "entropy": 5.509291505813598,
      "epoch": 1.7253063606302277,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004700752205400158,
      "loss": 5.2018,
      "mean_token_accuracy": 0.17911841422319413,
      "num_tokens": 38480427.0,
      "step": 22175
    },
    {
      "entropy": 5.352899980545044,
      "epoch": 1.7256953900019452,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047006140820448174,
      "loss": 5.1146,
      "mean_token_accuracy": 0.18891166597604753,
      "num_tokens": 38489283.0,
      "step": 22180
    },
    {
      "entropy": 5.442649698257446,
      "epoch": 1.7260844193736626,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047004759290920136,
      "loss": 5.1527,
      "mean_token_accuracy": 0.1839055374264717,
      "num_tokens": 38497557.0,
      "step": 22185
    },
    {
      "entropy": 5.436860799789429,
      "epoch": 1.7264734487453803,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004700337746543841,
      "loss": 5.1811,
      "mean_token_accuracy": 0.18733938187360763,
      "num_tokens": 38506589.0,
      "step": 22190
    },
    {
      "entropy": 5.417081594467163,
      "epoch": 1.7268624781170978,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004700199534402398,
      "loss": 5.1472,
      "mean_token_accuracy": 0.1824677288532257,
      "num_tokens": 38515016.0,
      "step": 22195
    },
    {
      "entropy": 5.507541799545288,
      "epoch": 1.7272515074888153,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004700061292669781,
      "loss": 5.1824,
      "mean_token_accuracy": 0.18618014752864837,
      "num_tokens": 38523792.0,
      "step": 22200
    },
    {
      "entropy": 5.353345680236816,
      "epoch": 1.727640536860533,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004699923021348088,
      "loss": 5.0047,
      "mean_token_accuracy": 0.1917987048625946,
      "num_tokens": 38532007.0,
      "step": 22205
    },
    {
      "entropy": 5.374248886108399,
      "epoch": 1.7280295662322507,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046997847204394166,
      "loss": 5.124,
      "mean_token_accuracy": 0.18407183438539504,
      "num_tokens": 38541044.0,
      "step": 22210
    },
    {
      "entropy": 5.422600650787354,
      "epoch": 1.728418595603968,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046996463899458653,
      "loss": 5.0661,
      "mean_token_accuracy": 0.1963285982608795,
      "num_tokens": 38549365.0,
      "step": 22215
    },
    {
      "entropy": 5.4137725830078125,
      "epoch": 1.7288076249756856,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004699508029869533,
      "loss": 5.0876,
      "mean_token_accuracy": 0.1927267551422119,
      "num_tokens": 38558289.0,
      "step": 22220
    },
    {
      "entropy": 5.45611982345581,
      "epoch": 1.7291966543474033,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046993696402125205,
      "loss": 5.1932,
      "mean_token_accuracy": 0.17374160289764404,
      "num_tokens": 38568422.0,
      "step": 22225
    },
    {
      "entropy": 5.488032817840576,
      "epoch": 1.7295856837191208,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004699231220976925,
      "loss": 5.1645,
      "mean_token_accuracy": 0.18438005447387695,
      "num_tokens": 38577009.0,
      "step": 22230
    },
    {
      "entropy": 5.402946281433105,
      "epoch": 1.7299747130908383,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004699092772164849,
      "loss": 4.9827,
      "mean_token_accuracy": 0.19345647990703582,
      "num_tokens": 38584920.0,
      "step": 22235
    },
    {
      "entropy": 5.276201772689819,
      "epoch": 1.730363742462556,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004698954293778392,
      "loss": 5.0638,
      "mean_token_accuracy": 0.19083783477544786,
      "num_tokens": 38593533.0,
      "step": 22240
    },
    {
      "entropy": 5.355232524871826,
      "epoch": 1.7307527718342735,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046988157858196555,
      "loss": 5.0762,
      "mean_token_accuracy": 0.18871882557868958,
      "num_tokens": 38602300.0,
      "step": 22245
    },
    {
      "entropy": 5.408234739303589,
      "epoch": 1.731141801205991,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046986772482907426,
      "loss": 5.1552,
      "mean_token_accuracy": 0.18473425060510634,
      "num_tokens": 38610729.0,
      "step": 22250
    },
    {
      "entropy": 5.396394109725952,
      "epoch": 1.7315308305777086,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004698538681193754,
      "loss": 5.0296,
      "mean_token_accuracy": 0.19235869348049164,
      "num_tokens": 38618192.0,
      "step": 22255
    },
    {
      "entropy": 5.4040412425994875,
      "epoch": 1.7319198599494263,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046984000845307907,
      "loss": 5.1385,
      "mean_token_accuracy": 0.18972633630037308,
      "num_tokens": 38627030.0,
      "step": 22260
    },
    {
      "entropy": 5.421328401565551,
      "epoch": 1.7323088893211438,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046982614583039584,
      "loss": 5.2184,
      "mean_token_accuracy": 0.1786823108792305,
      "num_tokens": 38635761.0,
      "step": 22265
    },
    {
      "entropy": 5.38336410522461,
      "epoch": 1.7326979186928613,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004698122802515359,
      "loss": 5.1234,
      "mean_token_accuracy": 0.18818623423576356,
      "num_tokens": 38643952.0,
      "step": 22270
    },
    {
      "entropy": 5.461874151229859,
      "epoch": 1.733086948064579,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004697984117167097,
      "loss": 5.0876,
      "mean_token_accuracy": 0.19542839825153352,
      "num_tokens": 38651888.0,
      "step": 22275
    },
    {
      "entropy": 5.455183362960815,
      "epoch": 1.7334759774362964,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046978454022612767,
      "loss": 5.2163,
      "mean_token_accuracy": 0.18202774226665497,
      "num_tokens": 38660466.0,
      "step": 22280
    },
    {
      "entropy": 5.460018014907837,
      "epoch": 1.733865006808014,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046977066578000025,
      "loss": 5.1641,
      "mean_token_accuracy": 0.17939212769269944,
      "num_tokens": 38668429.0,
      "step": 22285
    },
    {
      "entropy": 5.419499731063842,
      "epoch": 1.7342540361797316,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046975678837853806,
      "loss": 5.1741,
      "mean_token_accuracy": 0.18542421013116836,
      "num_tokens": 38676391.0,
      "step": 22290
    },
    {
      "entropy": 5.458867454528809,
      "epoch": 1.734643065551449,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046974290802195156,
      "loss": 5.1905,
      "mean_token_accuracy": 0.19043862521648408,
      "num_tokens": 38684662.0,
      "step": 22295
    },
    {
      "entropy": 5.463813352584839,
      "epoch": 1.7350320949231666,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004697290247104513,
      "loss": 5.1174,
      "mean_token_accuracy": 0.18328982591629028,
      "num_tokens": 38693497.0,
      "step": 22300
    },
    {
      "entropy": 5.375125980377197,
      "epoch": 1.7354211242948843,
      "grad_norm": 1.25,
      "learning_rate": 0.00046971513844424814,
      "loss": 5.036,
      "mean_token_accuracy": 0.18585280776023866,
      "num_tokens": 38701898.0,
      "step": 22305
    },
    {
      "entropy": 5.4715338230133055,
      "epoch": 1.735810153666602,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046970124922355265,
      "loss": 5.1602,
      "mean_token_accuracy": 0.1767078310251236,
      "num_tokens": 38710910.0,
      "step": 22310
    },
    {
      "entropy": 5.519257402420044,
      "epoch": 1.7361991830383194,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004696873570485756,
      "loss": 5.1741,
      "mean_token_accuracy": 0.18417732417583466,
      "num_tokens": 38719612.0,
      "step": 22315
    },
    {
      "entropy": 5.337896394729614,
      "epoch": 1.736588212410037,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004696734619195278,
      "loss": 5.1706,
      "mean_token_accuracy": 0.18064161241054535,
      "num_tokens": 38728332.0,
      "step": 22320
    },
    {
      "entropy": 5.3912616729736325,
      "epoch": 1.7369772417817546,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046965956383662,
      "loss": 5.1734,
      "mean_token_accuracy": 0.18458604216575622,
      "num_tokens": 38736830.0,
      "step": 22325
    },
    {
      "entropy": 5.446973657608032,
      "epoch": 1.737366271153472,
      "grad_norm": 1.25,
      "learning_rate": 0.0004696456628000632,
      "loss": 5.1377,
      "mean_token_accuracy": 0.1844444453716278,
      "num_tokens": 38745972.0,
      "step": 22330
    },
    {
      "entropy": 5.3612282276153564,
      "epoch": 1.7377553005251896,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046963175881006825,
      "loss": 5.0353,
      "mean_token_accuracy": 0.1815722569823265,
      "num_tokens": 38754287.0,
      "step": 22335
    },
    {
      "entropy": 5.3481049060821535,
      "epoch": 1.7381443298969073,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004696178518668462,
      "loss": 5.1246,
      "mean_token_accuracy": 0.18977776914834976,
      "num_tokens": 38763212.0,
      "step": 22340
    },
    {
      "entropy": 5.353709125518799,
      "epoch": 1.7385333592686247,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046960394197060804,
      "loss": 4.977,
      "mean_token_accuracy": 0.1947174370288849,
      "num_tokens": 38771668.0,
      "step": 22345
    },
    {
      "entropy": 5.449677801132202,
      "epoch": 1.7389223886403422,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046959002912156473,
      "loss": 5.2029,
      "mean_token_accuracy": 0.18209309428930281,
      "num_tokens": 38781184.0,
      "step": 22350
    },
    {
      "entropy": 5.394895744323731,
      "epoch": 1.73931141801206,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004695761133199275,
      "loss": 5.0378,
      "mean_token_accuracy": 0.1968079537153244,
      "num_tokens": 38789622.0,
      "step": 22355
    },
    {
      "entropy": 5.427113437652588,
      "epoch": 1.7397004473837776,
      "grad_norm": 1.125,
      "learning_rate": 0.0004695621945659074,
      "loss": 5.1192,
      "mean_token_accuracy": 0.19229973405599593,
      "num_tokens": 38798250.0,
      "step": 22360
    },
    {
      "entropy": 5.4141088962554935,
      "epoch": 1.740089476755495,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004695482728597157,
      "loss": 5.1254,
      "mean_token_accuracy": 0.185532408952713,
      "num_tokens": 38807724.0,
      "step": 22365
    },
    {
      "entropy": 5.4058685302734375,
      "epoch": 1.7404785061272126,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046953434820156363,
      "loss": 5.1732,
      "mean_token_accuracy": 0.18877536207437515,
      "num_tokens": 38816017.0,
      "step": 22370
    },
    {
      "entropy": 5.451808500289917,
      "epoch": 1.7408675354989303,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004695204205916624,
      "loss": 5.0885,
      "mean_token_accuracy": 0.19340557605028152,
      "num_tokens": 38824976.0,
      "step": 22375
    },
    {
      "entropy": 5.334464883804321,
      "epoch": 1.7412565648706477,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004695064900302235,
      "loss": 5.0944,
      "mean_token_accuracy": 0.18830258697271346,
      "num_tokens": 38834358.0,
      "step": 22380
    },
    {
      "entropy": 5.403951978683471,
      "epoch": 1.7416455942423652,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004694925565174581,
      "loss": 5.1765,
      "mean_token_accuracy": 0.17911430448293686,
      "num_tokens": 38842801.0,
      "step": 22385
    },
    {
      "entropy": 5.436860132217407,
      "epoch": 1.742034623614083,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046947862005357777,
      "loss": 5.2018,
      "mean_token_accuracy": 0.17850478887557983,
      "num_tokens": 38851335.0,
      "step": 22390
    },
    {
      "entropy": 5.488598585128784,
      "epoch": 1.7424236529858004,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004694646806387939,
      "loss": 5.1867,
      "mean_token_accuracy": 0.17903123795986176,
      "num_tokens": 38860074.0,
      "step": 22395
    },
    {
      "entropy": 5.439096736907959,
      "epoch": 1.7428126823575179,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004694507382733181,
      "loss": 5.1042,
      "mean_token_accuracy": 0.18824402987957,
      "num_tokens": 38868319.0,
      "step": 22400
    },
    {
      "entropy": 5.382623767852783,
      "epoch": 1.7432017117292355,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004694367929573618,
      "loss": 5.1618,
      "mean_token_accuracy": 0.19231386482715607,
      "num_tokens": 38876971.0,
      "step": 22405
    },
    {
      "entropy": 5.4630286693573,
      "epoch": 1.7435907411009532,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004694228446911367,
      "loss": 5.2755,
      "mean_token_accuracy": 0.17482255697250365,
      "num_tokens": 38886032.0,
      "step": 22410
    },
    {
      "entropy": 5.484811210632325,
      "epoch": 1.7439797704726707,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004694088934748542,
      "loss": 5.152,
      "mean_token_accuracy": 0.18637422025203704,
      "num_tokens": 38894505.0,
      "step": 22415
    },
    {
      "entropy": 5.320993709564209,
      "epoch": 1.7443687998443882,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004693949393087263,
      "loss": 5.0449,
      "mean_token_accuracy": 0.1930392265319824,
      "num_tokens": 38903450.0,
      "step": 22420
    },
    {
      "entropy": 5.409434461593628,
      "epoch": 1.744757829216106,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004693809821929647,
      "loss": 5.115,
      "mean_token_accuracy": 0.1896115943789482,
      "num_tokens": 38911326.0,
      "step": 22425
    },
    {
      "entropy": 5.39502182006836,
      "epoch": 1.7451468585878234,
      "grad_norm": 1.28125,
      "learning_rate": 0.000469367022127781,
      "loss": 5.1012,
      "mean_token_accuracy": 0.19049326479434966,
      "num_tokens": 38920035.0,
      "step": 22430
    },
    {
      "entropy": 5.456334972381592,
      "epoch": 1.7455358879595408,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046935305911338703,
      "loss": 5.1463,
      "mean_token_accuracy": 0.18700133562088012,
      "num_tokens": 38928542.0,
      "step": 22435
    },
    {
      "entropy": 5.4703648567199705,
      "epoch": 1.7459249173312585,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004693390931499948,
      "loss": 5.13,
      "mean_token_accuracy": 0.19097038805484773,
      "num_tokens": 38937486.0,
      "step": 22440
    },
    {
      "entropy": 5.413763904571534,
      "epoch": 1.7463139467029762,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004693251242378162,
      "loss": 5.0738,
      "mean_token_accuracy": 0.19065851271152495,
      "num_tokens": 38946171.0,
      "step": 22445
    },
    {
      "entropy": 5.367866468429566,
      "epoch": 1.7467029760746935,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046931115237706304,
      "loss": 5.0568,
      "mean_token_accuracy": 0.1883680358529091,
      "num_tokens": 38954214.0,
      "step": 22450
    },
    {
      "entropy": 5.488908672332764,
      "epoch": 1.7470920054464112,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004692971775679475,
      "loss": 5.2087,
      "mean_token_accuracy": 0.17844777405261994,
      "num_tokens": 38962102.0,
      "step": 22455
    },
    {
      "entropy": 5.447727251052856,
      "epoch": 1.7474810348181289,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046928319981068154,
      "loss": 5.0963,
      "mean_token_accuracy": 0.1914510726928711,
      "num_tokens": 38970472.0,
      "step": 22460
    },
    {
      "entropy": 5.39034481048584,
      "epoch": 1.7478700641898464,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004692692191054773,
      "loss": 5.1026,
      "mean_token_accuracy": 0.1826579064130783,
      "num_tokens": 38979151.0,
      "step": 22465
    },
    {
      "entropy": 5.391091966629029,
      "epoch": 1.7482590935615638,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004692552354525468,
      "loss": 5.1187,
      "mean_token_accuracy": 0.1896912485361099,
      "num_tokens": 38987044.0,
      "step": 22470
    },
    {
      "entropy": 5.4516983985900875,
      "epoch": 1.7486481229332815,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004692412488521023,
      "loss": 5.2187,
      "mean_token_accuracy": 0.1817756101489067,
      "num_tokens": 38995882.0,
      "step": 22475
    },
    {
      "entropy": 5.474606084823608,
      "epoch": 1.749037152304999,
      "grad_norm": 1.234375,
      "learning_rate": 0.000469227259304356,
      "loss": 5.1967,
      "mean_token_accuracy": 0.18416704535484313,
      "num_tokens": 39004456.0,
      "step": 22480
    },
    {
      "entropy": 5.500409317016602,
      "epoch": 1.7494261816767165,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046921326680952023,
      "loss": 5.1555,
      "mean_token_accuracy": 0.18533004075288773,
      "num_tokens": 39013363.0,
      "step": 22485
    },
    {
      "entropy": 5.448150014877319,
      "epoch": 1.7498152110484342,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004691992713678072,
      "loss": 5.1072,
      "mean_token_accuracy": 0.19047252237796783,
      "num_tokens": 39022116.0,
      "step": 22490
    },
    {
      "entropy": 5.443274927139282,
      "epoch": 1.7502042404201519,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004691852729794293,
      "loss": 5.1415,
      "mean_token_accuracy": 0.1873203694820404,
      "num_tokens": 39030737.0,
      "step": 22495
    },
    {
      "entropy": 5.342491579055786,
      "epoch": 1.7505932697918691,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004691712716445991,
      "loss": 5.0753,
      "mean_token_accuracy": 0.19463072568178177,
      "num_tokens": 39039511.0,
      "step": 22500
    },
    {
      "entropy": 5.355319452285767,
      "epoch": 1.7509822991635868,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004691572673635288,
      "loss": 5.066,
      "mean_token_accuracy": 0.19230345338582994,
      "num_tokens": 39048745.0,
      "step": 22505
    },
    {
      "entropy": 5.417779588699341,
      "epoch": 1.7513713285353045,
      "grad_norm": 1.21875,
      "learning_rate": 0.000469143260136431,
      "loss": 5.1416,
      "mean_token_accuracy": 0.18639375269412994,
      "num_tokens": 39057311.0,
      "step": 22510
    },
    {
      "entropy": 5.508921813964844,
      "epoch": 1.751760357907022,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004691292499635183,
      "loss": 5.1502,
      "mean_token_accuracy": 0.1874221533536911,
      "num_tokens": 39065630.0,
      "step": 22515
    },
    {
      "entropy": 5.357827472686767,
      "epoch": 1.7521493872787395,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004691152368450032,
      "loss": 5.1125,
      "mean_token_accuracy": 0.18569216281175613,
      "num_tokens": 39074284.0,
      "step": 22520
    },
    {
      "entropy": 5.377844619750976,
      "epoch": 1.7525384166504572,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046910122078109835,
      "loss": 5.1434,
      "mean_token_accuracy": 0.1895149528980255,
      "num_tokens": 39083284.0,
      "step": 22525
    },
    {
      "entropy": 5.45479679107666,
      "epoch": 1.7529274460221747,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004690872017720163,
      "loss": 5.0955,
      "mean_token_accuracy": 0.19052730947732927,
      "num_tokens": 39091106.0,
      "step": 22530
    },
    {
      "entropy": 5.381279706954956,
      "epoch": 1.7533164753938921,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046907317981797006,
      "loss": 5.0526,
      "mean_token_accuracy": 0.19442886859178543,
      "num_tokens": 39100499.0,
      "step": 22535
    },
    {
      "entropy": 5.399704647064209,
      "epoch": 1.7537055047656098,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046905915491917213,
      "loss": 5.1615,
      "mean_token_accuracy": 0.18966828137636185,
      "num_tokens": 39108895.0,
      "step": 22540
    },
    {
      "entropy": 5.303766250610352,
      "epoch": 1.7540945341373275,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004690451270758354,
      "loss": 5.0018,
      "mean_token_accuracy": 0.20048508942127227,
      "num_tokens": 39117053.0,
      "step": 22545
    },
    {
      "entropy": 5.398172044754029,
      "epoch": 1.7544835635090448,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046903109628817276,
      "loss": 5.1364,
      "mean_token_accuracy": 0.18973526060581208,
      "num_tokens": 39125750.0,
      "step": 22550
    },
    {
      "entropy": 5.392284154891968,
      "epoch": 1.7548725928807625,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046901706255639703,
      "loss": 5.0213,
      "mean_token_accuracy": 0.19316612482070922,
      "num_tokens": 39134105.0,
      "step": 22555
    },
    {
      "entropy": 5.529622316360474,
      "epoch": 1.7552616222524802,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004690030258807212,
      "loss": 5.2942,
      "mean_token_accuracy": 0.18487648516893387,
      "num_tokens": 39143543.0,
      "step": 22560
    },
    {
      "entropy": 5.442528295516968,
      "epoch": 1.7556506516241976,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046898898626135827,
      "loss": 5.1655,
      "mean_token_accuracy": 0.18303528279066086,
      "num_tokens": 39152826.0,
      "step": 22565
    },
    {
      "entropy": 5.4463824272155765,
      "epoch": 1.7560396809959151,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004689749436985211,
      "loss": 5.0879,
      "mean_token_accuracy": 0.18753847032785415,
      "num_tokens": 39160999.0,
      "step": 22570
    },
    {
      "entropy": 5.41607666015625,
      "epoch": 1.7564287103676328,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046896089819242304,
      "loss": 5.1257,
      "mean_token_accuracy": 0.1821245953440666,
      "num_tokens": 39169134.0,
      "step": 22575
    },
    {
      "entropy": 5.373354721069336,
      "epoch": 1.7568177397393503,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004689468497432771,
      "loss": 5.0666,
      "mean_token_accuracy": 0.19241163432598113,
      "num_tokens": 39177545.0,
      "step": 22580
    },
    {
      "entropy": 5.317069435119629,
      "epoch": 1.7572067691110678,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004689327983512963,
      "loss": 5.0121,
      "mean_token_accuracy": 0.19345415532588958,
      "num_tokens": 39185671.0,
      "step": 22585
    },
    {
      "entropy": 5.459634780883789,
      "epoch": 1.7575957984827855,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046891874401669404,
      "loss": 5.1596,
      "mean_token_accuracy": 0.18063502609729767,
      "num_tokens": 39194420.0,
      "step": 22590
    },
    {
      "entropy": 5.395160055160522,
      "epoch": 1.7579848278545032,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004689046867396834,
      "loss": 5.1201,
      "mean_token_accuracy": 0.18944093137979506,
      "num_tokens": 39203295.0,
      "step": 22595
    },
    {
      "entropy": 5.358353757858277,
      "epoch": 1.7583738572262204,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004688906265204779,
      "loss": 5.107,
      "mean_token_accuracy": 0.18821542263031005,
      "num_tokens": 39211838.0,
      "step": 22600
    },
    {
      "entropy": 5.404463005065918,
      "epoch": 1.7587628865979381,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004688765633592907,
      "loss": 5.0664,
      "mean_token_accuracy": 0.18429576903581618,
      "num_tokens": 39219842.0,
      "step": 22605
    },
    {
      "entropy": 5.4071526527404785,
      "epoch": 1.7591519159696558,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004688624972563352,
      "loss": 5.1255,
      "mean_token_accuracy": 0.18849987238645555,
      "num_tokens": 39228167.0,
      "step": 22610
    },
    {
      "entropy": 5.427981233596801,
      "epoch": 1.7595409453413733,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004688484282118249,
      "loss": 5.1667,
      "mean_token_accuracy": 0.1800422415137291,
      "num_tokens": 39236546.0,
      "step": 22615
    },
    {
      "entropy": 5.441767883300781,
      "epoch": 1.7599299747130908,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004688343562259732,
      "loss": 5.1005,
      "mean_token_accuracy": 0.19728597700595857,
      "num_tokens": 39244212.0,
      "step": 22620
    },
    {
      "entropy": 5.346495246887207,
      "epoch": 1.7603190040848085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004688202812989937,
      "loss": 5.0536,
      "mean_token_accuracy": 0.19228919595479965,
      "num_tokens": 39252892.0,
      "step": 22625
    },
    {
      "entropy": 5.327357435226441,
      "epoch": 1.760708033456526,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004688062034311,
      "loss": 5.0591,
      "mean_token_accuracy": 0.19770820289850236,
      "num_tokens": 39262261.0,
      "step": 22630
    },
    {
      "entropy": 5.433740758895874,
      "epoch": 1.7610970628282434,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046879212262250546,
      "loss": 5.0901,
      "mean_token_accuracy": 0.1918582245707512,
      "num_tokens": 39270790.0,
      "step": 22635
    },
    {
      "entropy": 5.375422477722168,
      "epoch": 1.761486092199961,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046877803887342406,
      "loss": 5.1054,
      "mean_token_accuracy": 0.18144760131835938,
      "num_tokens": 39278825.0,
      "step": 22640
    },
    {
      "entropy": 5.411475324630738,
      "epoch": 1.7618751215716788,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004687639521840693,
      "loss": 5.1485,
      "mean_token_accuracy": 0.18198432624340058,
      "num_tokens": 39288343.0,
      "step": 22645
    },
    {
      "entropy": 5.474473142623902,
      "epoch": 1.7622641509433963,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046874986255465495,
      "loss": 5.1711,
      "mean_token_accuracy": 0.18388526439666747,
      "num_tokens": 39296562.0,
      "step": 22650
    },
    {
      "entropy": 5.466700077056885,
      "epoch": 1.7626531803151138,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046873576998539485,
      "loss": 5.1057,
      "mean_token_accuracy": 0.1921567738056183,
      "num_tokens": 39305433.0,
      "step": 22655
    },
    {
      "entropy": 5.364537334442138,
      "epoch": 1.7630422096868315,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004687216744765028,
      "loss": 5.1662,
      "mean_token_accuracy": 0.18480238765478135,
      "num_tokens": 39313244.0,
      "step": 22660
    },
    {
      "entropy": 5.411498737335205,
      "epoch": 1.763431239058549,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004687075760281927,
      "loss": 5.2047,
      "mean_token_accuracy": 0.18049863278865813,
      "num_tokens": 39321933.0,
      "step": 22665
    },
    {
      "entropy": 5.486029767990113,
      "epoch": 1.7638202684302664,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004686934746406784,
      "loss": 5.1839,
      "mean_token_accuracy": 0.18438055813312532,
      "num_tokens": 39331194.0,
      "step": 22670
    },
    {
      "entropy": 5.435326051712036,
      "epoch": 1.764209297801984,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046867937031417397,
      "loss": 5.0938,
      "mean_token_accuracy": 0.18311641365289688,
      "num_tokens": 39339762.0,
      "step": 22675
    },
    {
      "entropy": 5.379807329177856,
      "epoch": 1.7645983271737016,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004686652630488933,
      "loss": 5.1744,
      "mean_token_accuracy": 0.18560923933982848,
      "num_tokens": 39349020.0,
      "step": 22680
    },
    {
      "entropy": 5.4603485584259035,
      "epoch": 1.764987356545419,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046865115284505063,
      "loss": 5.2778,
      "mean_token_accuracy": 0.17935481816530227,
      "num_tokens": 39357889.0,
      "step": 22685
    },
    {
      "entropy": 5.464339399337769,
      "epoch": 1.7653763859171367,
      "grad_norm": 1.25,
      "learning_rate": 0.00046863703970285986,
      "loss": 5.1266,
      "mean_token_accuracy": 0.1911150962114334,
      "num_tokens": 39366120.0,
      "step": 22690
    },
    {
      "entropy": 5.394922924041748,
      "epoch": 1.7657654152888544,
      "grad_norm": 1.25,
      "learning_rate": 0.0004686229236225352,
      "loss": 4.9798,
      "mean_token_accuracy": 0.19347111880779266,
      "num_tokens": 39374056.0,
      "step": 22695
    },
    {
      "entropy": 5.4045867919921875,
      "epoch": 1.766154444660572,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004686088046042909,
      "loss": 5.185,
      "mean_token_accuracy": 0.18702104985713958,
      "num_tokens": 39383170.0,
      "step": 22700
    },
    {
      "entropy": 5.432613897323608,
      "epoch": 1.7665434740322894,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046859468264834114,
      "loss": 5.1369,
      "mean_token_accuracy": 0.18840789347887038,
      "num_tokens": 39391599.0,
      "step": 22705
    },
    {
      "entropy": 5.46447262763977,
      "epoch": 1.766932503404007,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046858055775490017,
      "loss": 5.1931,
      "mean_token_accuracy": 0.17587807774543762,
      "num_tokens": 39400217.0,
      "step": 22710
    },
    {
      "entropy": 5.440601873397827,
      "epoch": 1.7673215327757246,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004685664299241825,
      "loss": 5.1631,
      "mean_token_accuracy": 0.1882198855280876,
      "num_tokens": 39408943.0,
      "step": 22715
    },
    {
      "entropy": 5.430352878570557,
      "epoch": 1.767710562147442,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004685522991564022,
      "loss": 5.1303,
      "mean_token_accuracy": 0.18365857005119324,
      "num_tokens": 39417361.0,
      "step": 22720
    },
    {
      "entropy": 5.254755067825317,
      "epoch": 1.7680995915191597,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004685381654517738,
      "loss": 4.8857,
      "mean_token_accuracy": 0.20069003850221634,
      "num_tokens": 39425896.0,
      "step": 22725
    },
    {
      "entropy": 5.32765212059021,
      "epoch": 1.7684886208908772,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004685240288105119,
      "loss": 5.1279,
      "mean_token_accuracy": 0.19241410940885545,
      "num_tokens": 39434719.0,
      "step": 22730
    },
    {
      "entropy": 5.49894642829895,
      "epoch": 1.7688776502625947,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046850988923283085,
      "loss": 5.1122,
      "mean_token_accuracy": 0.18351818174123763,
      "num_tokens": 39443731.0,
      "step": 22735
    },
    {
      "entropy": 5.483583927154541,
      "epoch": 1.7692666796343124,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046849574671894523,
      "loss": 5.1933,
      "mean_token_accuracy": 0.1874889001250267,
      "num_tokens": 39452430.0,
      "step": 22740
    },
    {
      "entropy": 5.409409284591675,
      "epoch": 1.76965570900603,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046848160126906956,
      "loss": 5.0171,
      "mean_token_accuracy": 0.19297200590372085,
      "num_tokens": 39460560.0,
      "step": 22745
    },
    {
      "entropy": 5.386553525924683,
      "epoch": 1.7700447383777476,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004684674528834186,
      "loss": 5.1529,
      "mean_token_accuracy": 0.17860287576913833,
      "num_tokens": 39469575.0,
      "step": 22750
    },
    {
      "entropy": 5.418479061126709,
      "epoch": 1.770433767749465,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046845330156220703,
      "loss": 5.1049,
      "mean_token_accuracy": 0.18631057143211366,
      "num_tokens": 39478297.0,
      "step": 22755
    },
    {
      "entropy": 5.438140726089477,
      "epoch": 1.7708227971211827,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004684391473056495,
      "loss": 5.1494,
      "mean_token_accuracy": 0.1765715166926384,
      "num_tokens": 39486947.0,
      "step": 22760
    },
    {
      "entropy": 5.444226884841919,
      "epoch": 1.7712118264929002,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046842499011396076,
      "loss": 5.1418,
      "mean_token_accuracy": 0.1859448328614235,
      "num_tokens": 39495649.0,
      "step": 22765
    },
    {
      "entropy": 5.360508918762207,
      "epoch": 1.7716008558646177,
      "grad_norm": 1.25,
      "learning_rate": 0.00046841082998735575,
      "loss": 5.1125,
      "mean_token_accuracy": 0.18740831613540648,
      "num_tokens": 39504733.0,
      "step": 22770
    },
    {
      "entropy": 5.340368938446045,
      "epoch": 1.7719898852363354,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046839666692604916,
      "loss": 5.1416,
      "mean_token_accuracy": 0.19150551706552504,
      "num_tokens": 39513007.0,
      "step": 22775
    },
    {
      "entropy": 5.467089605331421,
      "epoch": 1.7723789146080529,
      "grad_norm": 1.3125,
      "learning_rate": 0.000468382500930256,
      "loss": 5.2046,
      "mean_token_accuracy": 0.1933479428291321,
      "num_tokens": 39521734.0,
      "step": 22780
    },
    {
      "entropy": 5.519225120544434,
      "epoch": 1.7727679439797703,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046836833200019116,
      "loss": 5.1908,
      "mean_token_accuracy": 0.18199897557497025,
      "num_tokens": 39530525.0,
      "step": 22785
    },
    {
      "entropy": 5.462545013427734,
      "epoch": 1.773156973351488,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004683541601360697,
      "loss": 5.1252,
      "mean_token_accuracy": 0.19222628623247145,
      "num_tokens": 39538938.0,
      "step": 22790
    },
    {
      "entropy": 5.309070682525634,
      "epoch": 1.7735460027232057,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046833998533810653,
      "loss": 5.0104,
      "mean_token_accuracy": 0.20251239836215973,
      "num_tokens": 39547342.0,
      "step": 22795
    },
    {
      "entropy": 5.31598219871521,
      "epoch": 1.7739350320949232,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004683258076065169,
      "loss": 5.0932,
      "mean_token_accuracy": 0.18373397886753082,
      "num_tokens": 39555107.0,
      "step": 22800
    },
    {
      "entropy": 5.478404951095581,
      "epoch": 1.7743240614666407,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046831162694151586,
      "loss": 5.1299,
      "mean_token_accuracy": 0.18252855241298677,
      "num_tokens": 39563226.0,
      "step": 22805
    },
    {
      "entropy": 5.5176427364349365,
      "epoch": 1.7747130908383584,
      "grad_norm": 1.25,
      "learning_rate": 0.00046829744334331855,
      "loss": 5.1797,
      "mean_token_accuracy": 0.17923756837844848,
      "num_tokens": 39571689.0,
      "step": 22810
    },
    {
      "entropy": 5.365311861038208,
      "epoch": 1.7751021202100759,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046828325681214013,
      "loss": 5.0657,
      "mean_token_accuracy": 0.19048167616128922,
      "num_tokens": 39579985.0,
      "step": 22815
    },
    {
      "entropy": 5.413469552993774,
      "epoch": 1.7754911495817933,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000468269067348196,
      "loss": 5.1645,
      "mean_token_accuracy": 0.18184514045715333,
      "num_tokens": 39587892.0,
      "step": 22820
    },
    {
      "entropy": 5.419785213470459,
      "epoch": 1.775880178953511,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004682548749517014,
      "loss": 5.2183,
      "mean_token_accuracy": 0.18386710286140442,
      "num_tokens": 39597516.0,
      "step": 22825
    },
    {
      "entropy": 5.4607429027557375,
      "epoch": 1.7762692083252287,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046824067962287163,
      "loss": 5.1313,
      "mean_token_accuracy": 0.18487349152565002,
      "num_tokens": 39605860.0,
      "step": 22830
    },
    {
      "entropy": 5.4320276260375975,
      "epoch": 1.776658237696946,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004682264813619221,
      "loss": 5.1991,
      "mean_token_accuracy": 0.1852819502353668,
      "num_tokens": 39614915.0,
      "step": 22835
    },
    {
      "entropy": 5.5204099178314205,
      "epoch": 1.7770472670686637,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046821228016906836,
      "loss": 5.1678,
      "mean_token_accuracy": 0.18448693603277205,
      "num_tokens": 39623588.0,
      "step": 22840
    },
    {
      "entropy": 5.4841631889343265,
      "epoch": 1.7774362964403814,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004681980760445257,
      "loss": 5.2049,
      "mean_token_accuracy": 0.1817408949136734,
      "num_tokens": 39632351.0,
      "step": 22845
    },
    {
      "entropy": 5.403576898574829,
      "epoch": 1.7778253258120988,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004681838689885098,
      "loss": 5.1457,
      "mean_token_accuracy": 0.18567555248737336,
      "num_tokens": 39641053.0,
      "step": 22850
    },
    {
      "entropy": 5.327241992950439,
      "epoch": 1.7782143551838163,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004681696590012362,
      "loss": 4.981,
      "mean_token_accuracy": 0.19379662424325944,
      "num_tokens": 39649838.0,
      "step": 22855
    },
    {
      "entropy": 5.364695882797241,
      "epoch": 1.778603384555534,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046815544608292043,
      "loss": 5.0427,
      "mean_token_accuracy": 0.1902295768260956,
      "num_tokens": 39658886.0,
      "step": 22860
    },
    {
      "entropy": 5.447053623199463,
      "epoch": 1.7789924139272515,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004681412302337782,
      "loss": 5.1463,
      "mean_token_accuracy": 0.18964761197566987,
      "num_tokens": 39667279.0,
      "step": 22865
    },
    {
      "entropy": 5.37525520324707,
      "epoch": 1.779381443298969,
      "grad_norm": 1.125,
      "learning_rate": 0.0004681270114540252,
      "loss": 5.0978,
      "mean_token_accuracy": 0.1837271049618721,
      "num_tokens": 39676213.0,
      "step": 22870
    },
    {
      "entropy": 5.3852849960327145,
      "epoch": 1.7797704726706867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004681127897438772,
      "loss": 5.1074,
      "mean_token_accuracy": 0.18815388083457946,
      "num_tokens": 39685859.0,
      "step": 22875
    },
    {
      "entropy": 5.370520305633545,
      "epoch": 1.7801595020424044,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046809856510355007,
      "loss": 5.0513,
      "mean_token_accuracy": 0.18763694912195206,
      "num_tokens": 39695240.0,
      "step": 22880
    },
    {
      "entropy": 5.344392538070679,
      "epoch": 1.7805485314141216,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004680843375332595,
      "loss": 5.0962,
      "mean_token_accuracy": 0.1879494830965996,
      "num_tokens": 39704441.0,
      "step": 22885
    },
    {
      "entropy": 5.458462810516357,
      "epoch": 1.7809375607858393,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046807010703322143,
      "loss": 5.1819,
      "mean_token_accuracy": 0.182386115193367,
      "num_tokens": 39714074.0,
      "step": 22890
    },
    {
      "entropy": 5.477794361114502,
      "epoch": 1.781326590157557,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004680558736036518,
      "loss": 5.173,
      "mean_token_accuracy": 0.17655222564935685,
      "num_tokens": 39722527.0,
      "step": 22895
    },
    {
      "entropy": 5.402607345581055,
      "epoch": 1.7817156195292745,
      "grad_norm": 1.125,
      "learning_rate": 0.0004680416372447666,
      "loss": 5.1357,
      "mean_token_accuracy": 0.1886889085173607,
      "num_tokens": 39730983.0,
      "step": 22900
    },
    {
      "entropy": 5.428813314437866,
      "epoch": 1.782104648900992,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046802739795678177,
      "loss": 5.1871,
      "mean_token_accuracy": 0.17829882353544235,
      "num_tokens": 39741114.0,
      "step": 22905
    },
    {
      "entropy": 5.483025693893433,
      "epoch": 1.7824936782727097,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046801315573991346,
      "loss": 5.1279,
      "mean_token_accuracy": 0.18192223459482193,
      "num_tokens": 39749281.0,
      "step": 22910
    },
    {
      "entropy": 5.393791341781617,
      "epoch": 1.7828827076444271,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046799891059437764,
      "loss": 4.9667,
      "mean_token_accuracy": 0.20066469460725783,
      "num_tokens": 39758136.0,
      "step": 22915
    },
    {
      "entropy": 5.381140279769897,
      "epoch": 1.7832717370161446,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046798466252039056,
      "loss": 5.1723,
      "mean_token_accuracy": 0.1910592719912529,
      "num_tokens": 39767104.0,
      "step": 22920
    },
    {
      "entropy": 5.44423394203186,
      "epoch": 1.7836607663878623,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046797041151816845,
      "loss": 5.1499,
      "mean_token_accuracy": 0.1908529967069626,
      "num_tokens": 39775979.0,
      "step": 22925
    },
    {
      "entropy": 5.349611091613769,
      "epoch": 1.78404979575958,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046795615758792747,
      "loss": 5.0101,
      "mean_token_accuracy": 0.19025155156850815,
      "num_tokens": 39783740.0,
      "step": 22930
    },
    {
      "entropy": 5.40112829208374,
      "epoch": 1.7844388251312973,
      "grad_norm": 1.125,
      "learning_rate": 0.0004679419007298839,
      "loss": 5.1368,
      "mean_token_accuracy": 0.19482991546392442,
      "num_tokens": 39791836.0,
      "step": 22935
    },
    {
      "entropy": 5.438749122619629,
      "epoch": 1.784827854503015,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004679276409442541,
      "loss": 5.0375,
      "mean_token_accuracy": 0.18928542882204055,
      "num_tokens": 39800694.0,
      "step": 22940
    },
    {
      "entropy": 5.376036071777344,
      "epoch": 1.7852168838747327,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004679133782312544,
      "loss": 5.1847,
      "mean_token_accuracy": 0.18330350071191787,
      "num_tokens": 39809108.0,
      "step": 22945
    },
    {
      "entropy": 5.409468364715576,
      "epoch": 1.7856059132464501,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004678991125911013,
      "loss": 5.0925,
      "mean_token_accuracy": 0.18675657212734223,
      "num_tokens": 39817608.0,
      "step": 22950
    },
    {
      "entropy": 5.37347412109375,
      "epoch": 1.7859949426181676,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004678848440240111,
      "loss": 5.072,
      "mean_token_accuracy": 0.1887265056371689,
      "num_tokens": 39826596.0,
      "step": 22955
    },
    {
      "entropy": 5.383779907226563,
      "epoch": 1.7863839719898853,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004678705725302005,
      "loss": 5.0969,
      "mean_token_accuracy": 0.1867112025618553,
      "num_tokens": 39834440.0,
      "step": 22960
    },
    {
      "entropy": 5.373521423339843,
      "epoch": 1.7867730013616028,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004678562981098859,
      "loss": 5.1224,
      "mean_token_accuracy": 0.1824068084359169,
      "num_tokens": 39842617.0,
      "step": 22965
    },
    {
      "entropy": 5.355612754821777,
      "epoch": 1.7871620307333203,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046784202076328394,
      "loss": 4.9978,
      "mean_token_accuracy": 0.20619504898786545,
      "num_tokens": 39850688.0,
      "step": 22970
    },
    {
      "entropy": 5.399077939987182,
      "epoch": 1.787551060105038,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046782774049061124,
      "loss": 5.107,
      "mean_token_accuracy": 0.18841083645820617,
      "num_tokens": 39859347.0,
      "step": 22975
    },
    {
      "entropy": 5.4002612113952635,
      "epoch": 1.7879400894767556,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004678134572920845,
      "loss": 5.1262,
      "mean_token_accuracy": 0.19255193322896957,
      "num_tokens": 39867090.0,
      "step": 22980
    },
    {
      "entropy": 5.451544666290284,
      "epoch": 1.788329118848473,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004677991711679204,
      "loss": 5.0526,
      "mean_token_accuracy": 0.1871644750237465,
      "num_tokens": 39875653.0,
      "step": 22985
    },
    {
      "entropy": 5.422796773910522,
      "epoch": 1.7887181482201906,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046778488211833585,
      "loss": 5.154,
      "mean_token_accuracy": 0.1802382305264473,
      "num_tokens": 39884781.0,
      "step": 22990
    },
    {
      "entropy": 5.422377443313598,
      "epoch": 1.7891071775919083,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004677705901435475,
      "loss": 5.0687,
      "mean_token_accuracy": 0.18729550987482071,
      "num_tokens": 39892619.0,
      "step": 22995
    },
    {
      "entropy": 5.407084131240845,
      "epoch": 1.7894962069636258,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004677562952437723,
      "loss": 5.1224,
      "mean_token_accuracy": 0.1878099784255028,
      "num_tokens": 39901518.0,
      "step": 23000
    },
    {
      "entropy": 5.357982540130616,
      "epoch": 1.7898852363353432,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046774199741922705,
      "loss": 5.087,
      "mean_token_accuracy": 0.19374388903379441,
      "num_tokens": 39909550.0,
      "step": 23005
    },
    {
      "entropy": 5.440994644165039,
      "epoch": 1.790274265707061,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046772769667012884,
      "loss": 5.1411,
      "mean_token_accuracy": 0.1880270779132843,
      "num_tokens": 39918177.0,
      "step": 23010
    },
    {
      "entropy": 5.376241254806518,
      "epoch": 1.7906632950787784,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004677133929966946,
      "loss": 5.107,
      "mean_token_accuracy": 0.18376608937978745,
      "num_tokens": 39926812.0,
      "step": 23015
    },
    {
      "entropy": 5.400599384307862,
      "epoch": 1.791052324450496,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046769908639914134,
      "loss": 5.119,
      "mean_token_accuracy": 0.18576649576425552,
      "num_tokens": 39934506.0,
      "step": 23020
    },
    {
      "entropy": 5.363254833221435,
      "epoch": 1.7914413538222136,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004676847768776861,
      "loss": 5.1113,
      "mean_token_accuracy": 0.1854304775595665,
      "num_tokens": 39943252.0,
      "step": 23025
    },
    {
      "entropy": 5.416454792022705,
      "epoch": 1.7918303831939313,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004676704644325462,
      "loss": 5.1504,
      "mean_token_accuracy": 0.18110511898994447,
      "num_tokens": 39952585.0,
      "step": 23030
    },
    {
      "entropy": 5.464705562591552,
      "epoch": 1.7922194125656488,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004676561490639385,
      "loss": 5.1086,
      "mean_token_accuracy": 0.18881328850984574,
      "num_tokens": 39961805.0,
      "step": 23035
    },
    {
      "entropy": 5.422229242324829,
      "epoch": 1.7926084419373662,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004676418307720805,
      "loss": 5.1247,
      "mean_token_accuracy": 0.18650590628385544,
      "num_tokens": 39970045.0,
      "step": 23040
    },
    {
      "entropy": 5.375871801376343,
      "epoch": 1.792997471309084,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004676275095571892,
      "loss": 5.0738,
      "mean_token_accuracy": 0.18585753440856934,
      "num_tokens": 39979073.0,
      "step": 23045
    },
    {
      "entropy": 5.383953857421875,
      "epoch": 1.7933865006808014,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046761318541948215,
      "loss": 5.1003,
      "mean_token_accuracy": 0.18986997604370118,
      "num_tokens": 39987936.0,
      "step": 23050
    },
    {
      "entropy": 5.350383996963501,
      "epoch": 1.7937755300525189,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004675988583591766,
      "loss": 5.1514,
      "mean_token_accuracy": 0.18082786053419114,
      "num_tokens": 39996547.0,
      "step": 23055
    },
    {
      "entropy": 5.4402717590332035,
      "epoch": 1.7941645594242366,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046758452837648997,
      "loss": 5.1871,
      "mean_token_accuracy": 0.1778293415904045,
      "num_tokens": 40005717.0,
      "step": 23060
    },
    {
      "entropy": 5.401093435287476,
      "epoch": 1.794553588795954,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004675701954716395,
      "loss": 5.0097,
      "mean_token_accuracy": 0.1930085003376007,
      "num_tokens": 40014799.0,
      "step": 23065
    },
    {
      "entropy": 5.368622350692749,
      "epoch": 1.7949426181676715,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046755585964484286,
      "loss": 5.1177,
      "mean_token_accuracy": 0.18343613743782045,
      "num_tokens": 40023899.0,
      "step": 23070
    },
    {
      "entropy": 5.3760278701782225,
      "epoch": 1.7953316475393892,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004675415208963177,
      "loss": 5.0765,
      "mean_token_accuracy": 0.1906256139278412,
      "num_tokens": 40032629.0,
      "step": 23075
    },
    {
      "entropy": 5.418430137634277,
      "epoch": 1.795720676911107,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046752717922628125,
      "loss": 5.1502,
      "mean_token_accuracy": 0.1917804852128029,
      "num_tokens": 40040575.0,
      "step": 23080
    },
    {
      "entropy": 5.372248697280884,
      "epoch": 1.7961097062828244,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004675128346349514,
      "loss": 5.1479,
      "mean_token_accuracy": 0.17960894256830215,
      "num_tokens": 40049221.0,
      "step": 23085
    },
    {
      "entropy": 5.524859619140625,
      "epoch": 1.7964987356545419,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046749848712254554,
      "loss": 5.1666,
      "mean_token_accuracy": 0.17856574207544326,
      "num_tokens": 40057737.0,
      "step": 23090
    },
    {
      "entropy": 5.377202749252319,
      "epoch": 1.7968877650262596,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046748413668928164,
      "loss": 5.1003,
      "mean_token_accuracy": 0.19216603338718413,
      "num_tokens": 40066447.0,
      "step": 23095
    },
    {
      "entropy": 5.3553324222564695,
      "epoch": 1.797276794397977,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004674697833353774,
      "loss": 5.1464,
      "mean_token_accuracy": 0.18779952228069305,
      "num_tokens": 40075825.0,
      "step": 23100
    },
    {
      "entropy": 5.454073190689087,
      "epoch": 1.7976658237696945,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046745542706105045,
      "loss": 5.1684,
      "mean_token_accuracy": 0.1887335181236267,
      "num_tokens": 40084342.0,
      "step": 23105
    },
    {
      "entropy": 5.544976711273193,
      "epoch": 1.7980548531414122,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046744106786651875,
      "loss": 5.1878,
      "mean_token_accuracy": 0.17412700355052949,
      "num_tokens": 40092650.0,
      "step": 23110
    },
    {
      "entropy": 5.3703960418701175,
      "epoch": 1.7984438825131297,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004674267057520001,
      "loss": 5.0838,
      "mean_token_accuracy": 0.1860807552933693,
      "num_tokens": 40100738.0,
      "step": 23115
    },
    {
      "entropy": 5.350365829467774,
      "epoch": 1.7988329118848472,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004674123407177125,
      "loss": 4.9884,
      "mean_token_accuracy": 0.19439310878515242,
      "num_tokens": 40109213.0,
      "step": 23120
    },
    {
      "entropy": 5.330119371414185,
      "epoch": 1.7992219412565649,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046739797276387394,
      "loss": 4.9491,
      "mean_token_accuracy": 0.1969822734594345,
      "num_tokens": 40117622.0,
      "step": 23125
    },
    {
      "entropy": 5.354555320739746,
      "epoch": 1.7996109706282826,
      "grad_norm": 1.125,
      "learning_rate": 0.00046738360189070235,
      "loss": 5.1208,
      "mean_token_accuracy": 0.18660174459218978,
      "num_tokens": 40126347.0,
      "step": 23130
    },
    {
      "entropy": 5.530633211135864,
      "epoch": 1.8,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004673692280984157,
      "loss": 5.2001,
      "mean_token_accuracy": 0.17241753488779069,
      "num_tokens": 40134245.0,
      "step": 23135
    },
    {
      "entropy": 5.416616296768188,
      "epoch": 1.8003890293717175,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004673548513872324,
      "loss": 5.056,
      "mean_token_accuracy": 0.1839473158121109,
      "num_tokens": 40142304.0,
      "step": 23140
    },
    {
      "entropy": 5.309227561950683,
      "epoch": 1.8007780587434352,
      "grad_norm": 1.25,
      "learning_rate": 0.00046734047175737026,
      "loss": 5.0659,
      "mean_token_accuracy": 0.18389488458633424,
      "num_tokens": 40150999.0,
      "step": 23145
    },
    {
      "entropy": 5.402010536193847,
      "epoch": 1.8011670881151527,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046732608920904754,
      "loss": 5.0934,
      "mean_token_accuracy": 0.19258983731269835,
      "num_tokens": 40158659.0,
      "step": 23150
    },
    {
      "entropy": 5.363977098464966,
      "epoch": 1.8015561174868702,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004673117037424827,
      "loss": 5.0378,
      "mean_token_accuracy": 0.19345849454402925,
      "num_tokens": 40167343.0,
      "step": 23155
    },
    {
      "entropy": 5.359388732910157,
      "epoch": 1.8019451468585879,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046729731535789375,
      "loss": 5.0879,
      "mean_token_accuracy": 0.19060981124639512,
      "num_tokens": 40176433.0,
      "step": 23160
    },
    {
      "entropy": 5.44537091255188,
      "epoch": 1.8023341762303053,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046728292405549913,
      "loss": 5.1852,
      "mean_token_accuracy": 0.18350686132907867,
      "num_tokens": 40185826.0,
      "step": 23165
    },
    {
      "entropy": 5.404057598114013,
      "epoch": 1.8027232056020228,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004672685298355172,
      "loss": 5.1755,
      "mean_token_accuracy": 0.18615381717681884,
      "num_tokens": 40194359.0,
      "step": 23170
    },
    {
      "entropy": 5.310923099517822,
      "epoch": 1.8031122349737405,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004672541326981664,
      "loss": 4.9662,
      "mean_token_accuracy": 0.1955039009451866,
      "num_tokens": 40202812.0,
      "step": 23175
    },
    {
      "entropy": 5.37564172744751,
      "epoch": 1.8035012643454582,
      "grad_norm": 1.125,
      "learning_rate": 0.000467239732643665,
      "loss": 5.1384,
      "mean_token_accuracy": 0.1807870849967003,
      "num_tokens": 40212344.0,
      "step": 23180
    },
    {
      "entropy": 5.430229043960571,
      "epoch": 1.8038902937171757,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046722532967223183,
      "loss": 5.1953,
      "mean_token_accuracy": 0.18510631322860718,
      "num_tokens": 40220974.0,
      "step": 23185
    },
    {
      "entropy": 5.4251398086547855,
      "epoch": 1.8042793230888932,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004672109237840851,
      "loss": 5.1528,
      "mean_token_accuracy": 0.18590725660324098,
      "num_tokens": 40229616.0,
      "step": 23190
    },
    {
      "entropy": 5.39336256980896,
      "epoch": 1.8046683524606109,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046719651497944355,
      "loss": 5.1356,
      "mean_token_accuracy": 0.18449864834547042,
      "num_tokens": 40238505.0,
      "step": 23195
    },
    {
      "entropy": 5.539396572113037,
      "epoch": 1.8050573818323283,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004671821032585259,
      "loss": 5.2561,
      "mean_token_accuracy": 0.18170075714588166,
      "num_tokens": 40246716.0,
      "step": 23200
    },
    {
      "entropy": 5.497548294067383,
      "epoch": 1.8054464112040458,
      "grad_norm": 1.125,
      "learning_rate": 0.0004671676886215506,
      "loss": 5.226,
      "mean_token_accuracy": 0.16857113242149352,
      "num_tokens": 40256146.0,
      "step": 23205
    },
    {
      "entropy": 5.475355815887451,
      "epoch": 1.8058354405757635,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004671532710687365,
      "loss": 5.1812,
      "mean_token_accuracy": 0.18431262522935868,
      "num_tokens": 40265304.0,
      "step": 23210
    },
    {
      "entropy": 5.330702781677246,
      "epoch": 1.806224469947481,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004671388506003024,
      "loss": 4.9874,
      "mean_token_accuracy": 0.18526601046323776,
      "num_tokens": 40274055.0,
      "step": 23215
    },
    {
      "entropy": 5.402827262878418,
      "epoch": 1.8066134993191985,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004671244272164671,
      "loss": 5.0881,
      "mean_token_accuracy": 0.18802326321601867,
      "num_tokens": 40283184.0,
      "step": 23220
    },
    {
      "entropy": 5.503146648406982,
      "epoch": 1.8070025286909162,
      "grad_norm": 1.25,
      "learning_rate": 0.00046711000091744935,
      "loss": 5.176,
      "mean_token_accuracy": 0.18659207969903946,
      "num_tokens": 40292604.0,
      "step": 23225
    },
    {
      "entropy": 5.291101503372192,
      "epoch": 1.8073915580626339,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004670955717034681,
      "loss": 4.9051,
      "mean_token_accuracy": 0.20310263633728026,
      "num_tokens": 40301096.0,
      "step": 23230
    },
    {
      "entropy": 5.446319961547852,
      "epoch": 1.8077805874343513,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046708113957474233,
      "loss": 5.3098,
      "mean_token_accuracy": 0.17133473455905915,
      "num_tokens": 40309976.0,
      "step": 23235
    },
    {
      "entropy": 5.553260469436646,
      "epoch": 1.8081696168060688,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000467066704531491,
      "loss": 5.2543,
      "mean_token_accuracy": 0.18162506371736525,
      "num_tokens": 40318594.0,
      "step": 23240
    },
    {
      "entropy": 5.401522636413574,
      "epoch": 1.8085586461777865,
      "grad_norm": 1.25,
      "learning_rate": 0.0004670522665739332,
      "loss": 5.0641,
      "mean_token_accuracy": 0.19017927944660187,
      "num_tokens": 40326480.0,
      "step": 23245
    },
    {
      "entropy": 5.353560972213745,
      "epoch": 1.808947675549504,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004670378257022878,
      "loss": 5.1954,
      "mean_token_accuracy": 0.18831604421138765,
      "num_tokens": 40334646.0,
      "step": 23250
    },
    {
      "entropy": 5.467828273773193,
      "epoch": 1.8093367049212215,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046702338191677414,
      "loss": 5.1263,
      "mean_token_accuracy": 0.18383677452802658,
      "num_tokens": 40344007.0,
      "step": 23255
    },
    {
      "entropy": 5.542666387557984,
      "epoch": 1.8097257342929391,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004670089352176113,
      "loss": 5.1791,
      "mean_token_accuracy": 0.18141606599092483,
      "num_tokens": 40352075.0,
      "step": 23260
    },
    {
      "entropy": 5.381451654434204,
      "epoch": 1.8101147636646568,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046699448560501847,
      "loss": 5.1067,
      "mean_token_accuracy": 0.18415171653032303,
      "num_tokens": 40360725.0,
      "step": 23265
    },
    {
      "entropy": 5.399619770050049,
      "epoch": 1.810503793036374,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004669800330792149,
      "loss": 5.0929,
      "mean_token_accuracy": 0.19199528843164443,
      "num_tokens": 40369022.0,
      "step": 23270
    },
    {
      "entropy": 5.411592435836792,
      "epoch": 1.8108928224080918,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046696557764041994,
      "loss": 5.0834,
      "mean_token_accuracy": 0.18862043023109437,
      "num_tokens": 40377624.0,
      "step": 23275
    },
    {
      "entropy": 5.37959656715393,
      "epoch": 1.8112818517798095,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004669511192888528,
      "loss": 5.1413,
      "mean_token_accuracy": 0.18544863909482956,
      "num_tokens": 40385678.0,
      "step": 23280
    },
    {
      "entropy": 5.3820795059204105,
      "epoch": 1.811670881151527,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046693665802473294,
      "loss": 5.1453,
      "mean_token_accuracy": 0.19133729636669158,
      "num_tokens": 40394565.0,
      "step": 23285
    },
    {
      "entropy": 5.471668720245361,
      "epoch": 1.8120599105232444,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046692219384827986,
      "loss": 5.1532,
      "mean_token_accuracy": 0.18566900491714478,
      "num_tokens": 40402910.0,
      "step": 23290
    },
    {
      "entropy": 5.414111995697022,
      "epoch": 1.8124489398949621,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004669077267597129,
      "loss": 5.025,
      "mean_token_accuracy": 0.19056606590747832,
      "num_tokens": 40411352.0,
      "step": 23295
    },
    {
      "entropy": 5.414443778991699,
      "epoch": 1.8128379692666796,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046689325675925174,
      "loss": 5.1646,
      "mean_token_accuracy": 0.18766914308071136,
      "num_tokens": 40419925.0,
      "step": 23300
    },
    {
      "entropy": 5.393743228912354,
      "epoch": 1.813226998638397,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046687878384711574,
      "loss": 5.108,
      "mean_token_accuracy": 0.18476015627384185,
      "num_tokens": 40428948.0,
      "step": 23305
    },
    {
      "entropy": 5.422899675369263,
      "epoch": 1.8136160280101148,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046686430802352476,
      "loss": 5.1494,
      "mean_token_accuracy": 0.1831866905093193,
      "num_tokens": 40437625.0,
      "step": 23310
    },
    {
      "entropy": 5.38987979888916,
      "epoch": 1.8140050573818325,
      "grad_norm": 1.25,
      "learning_rate": 0.0004668498292886982,
      "loss": 5.089,
      "mean_token_accuracy": 0.18844496607780456,
      "num_tokens": 40446435.0,
      "step": 23315
    },
    {
      "entropy": 5.491609287261963,
      "epoch": 1.8143940867535497,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046683534764285577,
      "loss": 5.1296,
      "mean_token_accuracy": 0.1774972528219223,
      "num_tokens": 40455368.0,
      "step": 23320
    },
    {
      "entropy": 5.432666063308716,
      "epoch": 1.8147831161252674,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046682086308621725,
      "loss": 5.128,
      "mean_token_accuracy": 0.1860368564724922,
      "num_tokens": 40464461.0,
      "step": 23325
    },
    {
      "entropy": 5.492969751358032,
      "epoch": 1.8151721454969851,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004668063756190026,
      "loss": 5.2349,
      "mean_token_accuracy": 0.17239613234996795,
      "num_tokens": 40473143.0,
      "step": 23330
    },
    {
      "entropy": 5.42249698638916,
      "epoch": 1.8155611748687026,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046679188524143136,
      "loss": 5.0826,
      "mean_token_accuracy": 0.19292087405920028,
      "num_tokens": 40481592.0,
      "step": 23335
    },
    {
      "entropy": 5.308642339706421,
      "epoch": 1.81595020424042,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004667773919537235,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20094687342643738,
      "num_tokens": 40489445.0,
      "step": 23340
    },
    {
      "entropy": 5.320721387863159,
      "epoch": 1.8163392336121378,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046676289575609914,
      "loss": 5.0471,
      "mean_token_accuracy": 0.19274564534425737,
      "num_tokens": 40498279.0,
      "step": 23345
    },
    {
      "entropy": 5.432632207870483,
      "epoch": 1.8167282629838553,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046674839664877796,
      "loss": 5.0442,
      "mean_token_accuracy": 0.19146710336208345,
      "num_tokens": 40507051.0,
      "step": 23350
    },
    {
      "entropy": 5.428834915161133,
      "epoch": 1.8171172923555727,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004667338946319801,
      "loss": 4.9959,
      "mean_token_accuracy": 0.19711204022169113,
      "num_tokens": 40515213.0,
      "step": 23355
    },
    {
      "entropy": 5.384689617156982,
      "epoch": 1.8175063217272904,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004667193897059255,
      "loss": 5.1683,
      "mean_token_accuracy": 0.1817286342382431,
      "num_tokens": 40523859.0,
      "step": 23360
    },
    {
      "entropy": 5.46763916015625,
      "epoch": 1.8178953510990081,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046670488187083436,
      "loss": 5.2286,
      "mean_token_accuracy": 0.18000826388597488,
      "num_tokens": 40533236.0,
      "step": 23365
    },
    {
      "entropy": 5.471030187606812,
      "epoch": 1.8182843804707254,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004666903711269267,
      "loss": 5.0315,
      "mean_token_accuracy": 0.19386110454797745,
      "num_tokens": 40541670.0,
      "step": 23370
    },
    {
      "entropy": 5.384460306167602,
      "epoch": 1.818673409842443,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004666758574744228,
      "loss": 5.1428,
      "mean_token_accuracy": 0.19036951065063476,
      "num_tokens": 40551411.0,
      "step": 23375
    },
    {
      "entropy": 5.400392246246338,
      "epoch": 1.8190624392141608,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004666613409135429,
      "loss": 5.1112,
      "mean_token_accuracy": 0.18403472155332565,
      "num_tokens": 40560351.0,
      "step": 23380
    },
    {
      "entropy": 5.356915903091431,
      "epoch": 1.8194514685858783,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004666468214445072,
      "loss": 5.0506,
      "mean_token_accuracy": 0.1960597515106201,
      "num_tokens": 40569363.0,
      "step": 23385
    },
    {
      "entropy": 5.431414365768433,
      "epoch": 1.8198404979575957,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046663229906753595,
      "loss": 5.0954,
      "mean_token_accuracy": 0.184905444085598,
      "num_tokens": 40577425.0,
      "step": 23390
    },
    {
      "entropy": 5.402434539794922,
      "epoch": 1.8202295273293134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046661777378284965,
      "loss": 5.1326,
      "mean_token_accuracy": 0.18556924611330033,
      "num_tokens": 40585718.0,
      "step": 23395
    },
    {
      "entropy": 5.486553812026978,
      "epoch": 1.820618556701031,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004666032455906685,
      "loss": 5.2117,
      "mean_token_accuracy": 0.18363087475299836,
      "num_tokens": 40594886.0,
      "step": 23400
    },
    {
      "entropy": 5.515270566940307,
      "epoch": 1.8210075860727484,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046658871449121325,
      "loss": 5.2446,
      "mean_token_accuracy": 0.1745639219880104,
      "num_tokens": 40603737.0,
      "step": 23405
    },
    {
      "entropy": 5.4223428726196286,
      "epoch": 1.821396615444466,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004665741804847041,
      "loss": 5.105,
      "mean_token_accuracy": 0.18091865330934526,
      "num_tokens": 40612473.0,
      "step": 23410
    },
    {
      "entropy": 5.4242401123046875,
      "epoch": 1.8217856448161838,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046655964357136164,
      "loss": 5.0666,
      "mean_token_accuracy": 0.18724408596754075,
      "num_tokens": 40620517.0,
      "step": 23415
    },
    {
      "entropy": 5.446871948242188,
      "epoch": 1.822174674187901,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046654510375140657,
      "loss": 5.1647,
      "mean_token_accuracy": 0.1764190226793289,
      "num_tokens": 40629530.0,
      "step": 23420
    },
    {
      "entropy": 5.399376296997071,
      "epoch": 1.8225637035596187,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004665305610250594,
      "loss": 5.0748,
      "mean_token_accuracy": 0.19316655695438384,
      "num_tokens": 40638616.0,
      "step": 23425
    },
    {
      "entropy": 5.4894993782043455,
      "epoch": 1.8229527329313364,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004665160153925408,
      "loss": 5.13,
      "mean_token_accuracy": 0.1794159308075905,
      "num_tokens": 40646945.0,
      "step": 23430
    },
    {
      "entropy": 5.3997547149658205,
      "epoch": 1.823341762303054,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046650146685407154,
      "loss": 5.1073,
      "mean_token_accuracy": 0.19166603535413743,
      "num_tokens": 40654845.0,
      "step": 23435
    },
    {
      "entropy": 5.300659418106079,
      "epoch": 1.8237307916747714,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046648691540987226,
      "loss": 4.9706,
      "mean_token_accuracy": 0.19795334786176683,
      "num_tokens": 40663776.0,
      "step": 23440
    },
    {
      "entropy": 5.405929899215698,
      "epoch": 1.824119821046489,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046647236106016387,
      "loss": 5.0301,
      "mean_token_accuracy": 0.1922718182206154,
      "num_tokens": 40672071.0,
      "step": 23445
    },
    {
      "entropy": 5.426625680923462,
      "epoch": 1.8245088504182065,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004664578038051672,
      "loss": 5.2053,
      "mean_token_accuracy": 0.1843576043844223,
      "num_tokens": 40681264.0,
      "step": 23450
    },
    {
      "entropy": 5.384003305435181,
      "epoch": 1.824897879789924,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000466443243645103,
      "loss": 5.1082,
      "mean_token_accuracy": 0.18642598092556,
      "num_tokens": 40690091.0,
      "step": 23455
    },
    {
      "entropy": 5.511582851409912,
      "epoch": 1.8252869091616417,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004664286805801923,
      "loss": 5.2193,
      "mean_token_accuracy": 0.18088100403547286,
      "num_tokens": 40699623.0,
      "step": 23460
    },
    {
      "entropy": 5.412361192703247,
      "epoch": 1.8256759385333594,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004664141146106562,
      "loss": 5.0604,
      "mean_token_accuracy": 0.18808915466070175,
      "num_tokens": 40708288.0,
      "step": 23465
    },
    {
      "entropy": 5.473953628540039,
      "epoch": 1.8260649679050769,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046639954573671547,
      "loss": 5.178,
      "mean_token_accuracy": 0.18310302644968032,
      "num_tokens": 40716357.0,
      "step": 23470
    },
    {
      "entropy": 5.455427551269532,
      "epoch": 1.8264539972767944,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046638497395859127,
      "loss": 5.1838,
      "mean_token_accuracy": 0.18217079043388368,
      "num_tokens": 40724775.0,
      "step": 23475
    },
    {
      "entropy": 5.387223625183106,
      "epoch": 1.826843026648512,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004663703992765047,
      "loss": 5.1358,
      "mean_token_accuracy": 0.19118365198373793,
      "num_tokens": 40733642.0,
      "step": 23480
    },
    {
      "entropy": 5.379017496109009,
      "epoch": 1.8272320560202295,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046635582169067693,
      "loss": 5.0609,
      "mean_token_accuracy": 0.19094339311122893,
      "num_tokens": 40742201.0,
      "step": 23485
    },
    {
      "entropy": 5.377694892883301,
      "epoch": 1.827621085391947,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046634124120132915,
      "loss": 5.0442,
      "mean_token_accuracy": 0.1951903745532036,
      "num_tokens": 40750379.0,
      "step": 23490
    },
    {
      "entropy": 5.402575826644897,
      "epoch": 1.8280101147636647,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004663266578086826,
      "loss": 5.1531,
      "mean_token_accuracy": 0.17934572845697402,
      "num_tokens": 40759232.0,
      "step": 23495
    },
    {
      "entropy": 5.412865114212036,
      "epoch": 1.8283991441353822,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004663120715129585,
      "loss": 5.1193,
      "mean_token_accuracy": 0.1840099111199379,
      "num_tokens": 40768247.0,
      "step": 23500
    },
    {
      "entropy": 5.462263059616089,
      "epoch": 1.8287881735070997,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004662974823143783,
      "loss": 5.1842,
      "mean_token_accuracy": 0.18503874838352202,
      "num_tokens": 40777085.0,
      "step": 23505
    },
    {
      "entropy": 5.454936695098877,
      "epoch": 1.8291772028788174,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004662828902131632,
      "loss": 5.2057,
      "mean_token_accuracy": 0.18238736987113952,
      "num_tokens": 40785466.0,
      "step": 23510
    },
    {
      "entropy": 5.427293062210083,
      "epoch": 1.829566232250535,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004662682952095348,
      "loss": 5.095,
      "mean_token_accuracy": 0.1877877652645111,
      "num_tokens": 40794165.0,
      "step": 23515
    },
    {
      "entropy": 5.407046985626221,
      "epoch": 1.8299552616222525,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046625369730371436,
      "loss": 5.0729,
      "mean_token_accuracy": 0.1905936688184738,
      "num_tokens": 40802497.0,
      "step": 23520
    },
    {
      "entropy": 5.387262058258057,
      "epoch": 1.83034429099397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004662390964959235,
      "loss": 5.0592,
      "mean_token_accuracy": 0.1915487140417099,
      "num_tokens": 40811022.0,
      "step": 23525
    },
    {
      "entropy": 5.463723850250244,
      "epoch": 1.8307333203656877,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046622449278638375,
      "loss": 5.1521,
      "mean_token_accuracy": 0.1861449509859085,
      "num_tokens": 40820430.0,
      "step": 23530
    },
    {
      "entropy": 5.4452940940856935,
      "epoch": 1.8311223497374052,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004662098861753167,
      "loss": 5.1649,
      "mean_token_accuracy": 0.1854791134595871,
      "num_tokens": 40829245.0,
      "step": 23535
    },
    {
      "entropy": 5.331739044189453,
      "epoch": 1.8315113791091227,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046619527666294394,
      "loss": 5.0686,
      "mean_token_accuracy": 0.1933434784412384,
      "num_tokens": 40838576.0,
      "step": 23540
    },
    {
      "entropy": 5.474249696731567,
      "epoch": 1.8319004084808403,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004661806642494872,
      "loss": 5.2017,
      "mean_token_accuracy": 0.17841590493917464,
      "num_tokens": 40847097.0,
      "step": 23545
    },
    {
      "entropy": 5.486355590820312,
      "epoch": 1.8322894378525578,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004661660489351681,
      "loss": 5.0801,
      "mean_token_accuracy": 0.19482356756925584,
      "num_tokens": 40855135.0,
      "step": 23550
    },
    {
      "entropy": 5.4612339496612545,
      "epoch": 1.8326784672242753,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004661514307202086,
      "loss": 5.1816,
      "mean_token_accuracy": 0.1843657225370407,
      "num_tokens": 40864218.0,
      "step": 23555
    },
    {
      "entropy": 5.436075830459595,
      "epoch": 1.833067496595993,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046613680960483036,
      "loss": 5.1346,
      "mean_token_accuracy": 0.18343111425638198,
      "num_tokens": 40873147.0,
      "step": 23560
    },
    {
      "entropy": 5.477634572982788,
      "epoch": 1.8334565259677107,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004661221855892552,
      "loss": 5.2082,
      "mean_token_accuracy": 0.18481239676475525,
      "num_tokens": 40881340.0,
      "step": 23565
    },
    {
      "entropy": 5.408070802688599,
      "epoch": 1.8338455553394282,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046610755867370506,
      "loss": 5.048,
      "mean_token_accuracy": 0.19115407168865203,
      "num_tokens": 40889246.0,
      "step": 23570
    },
    {
      "entropy": 5.425782680511475,
      "epoch": 1.8342345847111456,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000466092928858402,
      "loss": 5.1347,
      "mean_token_accuracy": 0.1860576316714287,
      "num_tokens": 40898285.0,
      "step": 23575
    },
    {
      "entropy": 5.40475378036499,
      "epoch": 1.8346236140828633,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046607829614356787,
      "loss": 5.1705,
      "mean_token_accuracy": 0.18485250622034072,
      "num_tokens": 40907012.0,
      "step": 23580
    },
    {
      "entropy": 5.390592288970947,
      "epoch": 1.8350126434545808,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004660636605294247,
      "loss": 5.037,
      "mean_token_accuracy": 0.19614625126123428,
      "num_tokens": 40915857.0,
      "step": 23585
    },
    {
      "entropy": 5.463746452331543,
      "epoch": 1.8354016728262983,
      "grad_norm": 1.125,
      "learning_rate": 0.0004660490220161946,
      "loss": 5.0885,
      "mean_token_accuracy": 0.18818917572498323,
      "num_tokens": 40924364.0,
      "step": 23590
    },
    {
      "entropy": 5.362799072265625,
      "epoch": 1.835790702198016,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046603438060409966,
      "loss": 5.0835,
      "mean_token_accuracy": 0.18742408752441406,
      "num_tokens": 40934092.0,
      "step": 23595
    },
    {
      "entropy": 5.352373361587524,
      "epoch": 1.8361797315697335,
      "grad_norm": 1.265625,
      "learning_rate": 0.000466019736293362,
      "loss": 4.9465,
      "mean_token_accuracy": 0.19245639741420745,
      "num_tokens": 40942066.0,
      "step": 23600
    },
    {
      "entropy": 5.445371150970459,
      "epoch": 1.836568760941451,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046600508908420396,
      "loss": 5.1386,
      "mean_token_accuracy": 0.18403899669647217,
      "num_tokens": 40951445.0,
      "step": 23605
    },
    {
      "entropy": 5.382031869888306,
      "epoch": 1.8369577903131686,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046599043897684766,
      "loss": 5.0896,
      "mean_token_accuracy": 0.18989260643720626,
      "num_tokens": 40960111.0,
      "step": 23610
    },
    {
      "entropy": 5.383875894546509,
      "epoch": 1.8373468196848863,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004659757859715155,
      "loss": 5.0562,
      "mean_token_accuracy": 0.19691762030124665,
      "num_tokens": 40968036.0,
      "step": 23615
    },
    {
      "entropy": 5.3615926742553714,
      "epoch": 1.8377358490566038,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046596113006842975,
      "loss": 5.1107,
      "mean_token_accuracy": 0.18735164999961854,
      "num_tokens": 40977032.0,
      "step": 23620
    },
    {
      "entropy": 5.397118473052979,
      "epoch": 1.8381248784283213,
      "grad_norm": 1.25,
      "learning_rate": 0.0004659464712678128,
      "loss": 5.0948,
      "mean_token_accuracy": 0.18611981719732285,
      "num_tokens": 40985523.0,
      "step": 23625
    },
    {
      "entropy": 5.3890852451324465,
      "epoch": 1.838513907800039,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004659318095698871,
      "loss": 4.9855,
      "mean_token_accuracy": 0.19653391391038894,
      "num_tokens": 40993617.0,
      "step": 23630
    },
    {
      "entropy": 5.277385568618774,
      "epoch": 1.8389029371717565,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000465917144974875,
      "loss": 5.0755,
      "mean_token_accuracy": 0.18780739307403566,
      "num_tokens": 41002424.0,
      "step": 23635
    },
    {
      "entropy": 5.424960374832153,
      "epoch": 1.839291966543474,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004659024774829992,
      "loss": 5.1868,
      "mean_token_accuracy": 0.17984769940376283,
      "num_tokens": 41011150.0,
      "step": 23640
    },
    {
      "entropy": 5.465500926971435,
      "epoch": 1.8396809959151916,
      "grad_norm": 1.203125,
      "learning_rate": 0.000465887807094482,
      "loss": 5.078,
      "mean_token_accuracy": 0.19085448682308198,
      "num_tokens": 41020112.0,
      "step": 23645
    },
    {
      "entropy": 5.469873428344727,
      "epoch": 1.8400700252869093,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046587313380954635,
      "loss": 5.1335,
      "mean_token_accuracy": 0.18261379599571229,
      "num_tokens": 41029470.0,
      "step": 23650
    },
    {
      "entropy": 5.42513861656189,
      "epoch": 1.8404590546586266,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046585845762841453,
      "loss": 5.1271,
      "mean_token_accuracy": 0.18214615434408188,
      "num_tokens": 41038622.0,
      "step": 23655
    },
    {
      "entropy": 5.5608875274658205,
      "epoch": 1.8408480840303443,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004658437785513095,
      "loss": 5.2541,
      "mean_token_accuracy": 0.1795591726899147,
      "num_tokens": 41047573.0,
      "step": 23660
    },
    {
      "entropy": 5.428491878509521,
      "epoch": 1.841237113402062,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004658290965784539,
      "loss": 5.0536,
      "mean_token_accuracy": 0.19276835024356842,
      "num_tokens": 41057084.0,
      "step": 23665
    },
    {
      "entropy": 5.2946899890899655,
      "epoch": 1.8416261427737795,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004658144117100704,
      "loss": 4.9846,
      "mean_token_accuracy": 0.19335988610982896,
      "num_tokens": 41065297.0,
      "step": 23670
    },
    {
      "entropy": 5.353234052658081,
      "epoch": 1.842015172145497,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046579972394638204,
      "loss": 5.0893,
      "mean_token_accuracy": 0.19417805075645447,
      "num_tokens": 41073748.0,
      "step": 23675
    },
    {
      "entropy": 5.439117908477783,
      "epoch": 1.8424042015172146,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046578503328761146,
      "loss": 5.0456,
      "mean_token_accuracy": 0.1927301436662674,
      "num_tokens": 41082304.0,
      "step": 23680
    },
    {
      "entropy": 5.408972120285034,
      "epoch": 1.842793230888932,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046577033973398173,
      "loss": 5.1865,
      "mean_token_accuracy": 0.1802212566137314,
      "num_tokens": 41091130.0,
      "step": 23685
    },
    {
      "entropy": 5.386506414413452,
      "epoch": 1.8431822602606496,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004657556432857157,
      "loss": 5.1174,
      "mean_token_accuracy": 0.18833518475294114,
      "num_tokens": 41100413.0,
      "step": 23690
    },
    {
      "entropy": 5.2698663711547855,
      "epoch": 1.8435712896323673,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004657409439430364,
      "loss": 4.9552,
      "mean_token_accuracy": 0.20194483250379563,
      "num_tokens": 41108582.0,
      "step": 23695
    },
    {
      "entropy": 5.3586596012115475,
      "epoch": 1.843960319004085,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004657262417061669,
      "loss": 5.0443,
      "mean_token_accuracy": 0.19317821264266968,
      "num_tokens": 41117067.0,
      "step": 23700
    },
    {
      "entropy": 5.3617500305175785,
      "epoch": 1.8443493483758022,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004657115365753302,
      "loss": 5.0631,
      "mean_token_accuracy": 0.18658435195684434,
      "num_tokens": 41125559.0,
      "step": 23705
    },
    {
      "entropy": 5.366607332229615,
      "epoch": 1.84473837774752,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046569682855074953,
      "loss": 5.092,
      "mean_token_accuracy": 0.1863200530409813,
      "num_tokens": 41134503.0,
      "step": 23710
    },
    {
      "entropy": 5.508480501174927,
      "epoch": 1.8451274071192376,
      "grad_norm": 2.09375,
      "learning_rate": 0.00046568211763264796,
      "loss": 5.2119,
      "mean_token_accuracy": 0.1762565165758133,
      "num_tokens": 41143903.0,
      "step": 23715
    },
    {
      "entropy": 5.4346095561981205,
      "epoch": 1.845516436490955,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004656674038212488,
      "loss": 5.1058,
      "mean_token_accuracy": 0.18354797810316087,
      "num_tokens": 41153082.0,
      "step": 23720
    },
    {
      "entropy": 5.453666925430298,
      "epoch": 1.8459054658626726,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046565268711677525,
      "loss": 5.073,
      "mean_token_accuracy": 0.1941823586821556,
      "num_tokens": 41161749.0,
      "step": 23725
    },
    {
      "entropy": 5.379222249984741,
      "epoch": 1.8462944952343903,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046563796751945065,
      "loss": 5.1141,
      "mean_token_accuracy": 0.1840197041630745,
      "num_tokens": 41170969.0,
      "step": 23730
    },
    {
      "entropy": 5.307048988342285,
      "epoch": 1.8466835246061077,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046562324502949834,
      "loss": 4.9756,
      "mean_token_accuracy": 0.1964436173439026,
      "num_tokens": 41179512.0,
      "step": 23735
    },
    {
      "entropy": 5.356303977966308,
      "epoch": 1.8470725539778252,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004656085196471416,
      "loss": 5.0126,
      "mean_token_accuracy": 0.1920081853866577,
      "num_tokens": 41188109.0,
      "step": 23740
    },
    {
      "entropy": 5.393094205856324,
      "epoch": 1.847461583349543,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046559379137260404,
      "loss": 5.1316,
      "mean_token_accuracy": 0.18232296854257585,
      "num_tokens": 41196708.0,
      "step": 23745
    },
    {
      "entropy": 5.515661334991455,
      "epoch": 1.8478506127212606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000465579060206109,
      "loss": 5.2563,
      "mean_token_accuracy": 0.1825646549463272,
      "num_tokens": 41205013.0,
      "step": 23750
    },
    {
      "entropy": 5.342746925354004,
      "epoch": 1.8482396420929779,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046556432614788015,
      "loss": 4.9786,
      "mean_token_accuracy": 0.19502511471509934,
      "num_tokens": 41213701.0,
      "step": 23755
    },
    {
      "entropy": 5.333768653869629,
      "epoch": 1.8486286714646956,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004655495891981409,
      "loss": 5.0136,
      "mean_token_accuracy": 0.19132963120937346,
      "num_tokens": 41222450.0,
      "step": 23760
    },
    {
      "entropy": 5.458305406570434,
      "epoch": 1.8490177008364133,
      "grad_norm": 1.1875,
      "learning_rate": 0.000465534849357115,
      "loss": 5.1253,
      "mean_token_accuracy": 0.18563984632492064,
      "num_tokens": 41230259.0,
      "step": 23765
    },
    {
      "entropy": 5.368938255310058,
      "epoch": 1.8494067302081307,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046552010662502595,
      "loss": 5.0743,
      "mean_token_accuracy": 0.1948613241314888,
      "num_tokens": 41238863.0,
      "step": 23770
    },
    {
      "entropy": 5.327775526046753,
      "epoch": 1.8497957595798482,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004655053610020976,
      "loss": 5.0865,
      "mean_token_accuracy": 0.18872832059860228,
      "num_tokens": 41247260.0,
      "step": 23775
    },
    {
      "entropy": 5.352063131332398,
      "epoch": 1.850184788951566,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004654906124885535,
      "loss": 5.0104,
      "mean_token_accuracy": 0.19859888702630996,
      "num_tokens": 41255475.0,
      "step": 23780
    },
    {
      "entropy": 5.327923536300659,
      "epoch": 1.8505738183232834,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004654758610846176,
      "loss": 4.8986,
      "mean_token_accuracy": 0.20200287848711013,
      "num_tokens": 41263858.0,
      "step": 23785
    },
    {
      "entropy": 5.368563032150268,
      "epoch": 1.8509628476950009,
      "grad_norm": 1.375,
      "learning_rate": 0.0004654611067905137,
      "loss": 5.1471,
      "mean_token_accuracy": 0.18483048677444458,
      "num_tokens": 41272527.0,
      "step": 23790
    },
    {
      "entropy": 5.418375444412232,
      "epoch": 1.8513518770667186,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046544634960646563,
      "loss": 5.0998,
      "mean_token_accuracy": 0.19133739918470383,
      "num_tokens": 41280647.0,
      "step": 23795
    },
    {
      "entropy": 5.537930631637574,
      "epoch": 1.8517409064384363,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004654315895326974,
      "loss": 5.1419,
      "mean_token_accuracy": 0.19020063430070877,
      "num_tokens": 41289462.0,
      "step": 23800
    },
    {
      "entropy": 5.4853416919708256,
      "epoch": 1.8521299358101535,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004654168265694328,
      "loss": 5.0443,
      "mean_token_accuracy": 0.19345538765192033,
      "num_tokens": 41297178.0,
      "step": 23805
    },
    {
      "entropy": 5.384818649291992,
      "epoch": 1.8525189651818712,
      "grad_norm": 1.125,
      "learning_rate": 0.000465402060716896,
      "loss": 5.0303,
      "mean_token_accuracy": 0.19304134249687194,
      "num_tokens": 41305405.0,
      "step": 23810
    },
    {
      "entropy": 5.373839807510376,
      "epoch": 1.852907994553589,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000465387291975311,
      "loss": 5.0599,
      "mean_token_accuracy": 0.19269389510154725,
      "num_tokens": 41313330.0,
      "step": 23815
    },
    {
      "entropy": 5.429249095916748,
      "epoch": 1.8532970239253064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046537252034490185,
      "loss": 5.0769,
      "mean_token_accuracy": 0.1960735037922859,
      "num_tokens": 41321245.0,
      "step": 23820
    },
    {
      "entropy": 5.410325002670288,
      "epoch": 1.8536860532970239,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046535774582589275,
      "loss": 5.1395,
      "mean_token_accuracy": 0.18483452647924423,
      "num_tokens": 41330013.0,
      "step": 23825
    },
    {
      "entropy": 5.412614250183106,
      "epoch": 1.8540750826687415,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046534296841850776,
      "loss": 5.1021,
      "mean_token_accuracy": 0.19254927486181259,
      "num_tokens": 41338134.0,
      "step": 23830
    },
    {
      "entropy": 5.375621509552002,
      "epoch": 1.854464112040459,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046532818812297124,
      "loss": 5.014,
      "mean_token_accuracy": 0.18926284462213516,
      "num_tokens": 41346693.0,
      "step": 23835
    },
    {
      "entropy": 5.434927034378052,
      "epoch": 1.8548531414121765,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004653134049395074,
      "loss": 5.191,
      "mean_token_accuracy": 0.18164955973625183,
      "num_tokens": 41354987.0,
      "step": 23840
    },
    {
      "entropy": 5.431974029541015,
      "epoch": 1.8552421707838942,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004652986188683406,
      "loss": 5.1634,
      "mean_token_accuracy": 0.183976186811924,
      "num_tokens": 41363703.0,
      "step": 23845
    },
    {
      "entropy": 5.414255905151367,
      "epoch": 1.855631200155612,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046528382990969504,
      "loss": 5.118,
      "mean_token_accuracy": 0.18959563076496125,
      "num_tokens": 41371983.0,
      "step": 23850
    },
    {
      "entropy": 5.396398401260376,
      "epoch": 1.8560202295273294,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004652690380637953,
      "loss": 5.1396,
      "mean_token_accuracy": 0.1860457792878151,
      "num_tokens": 41380215.0,
      "step": 23855
    },
    {
      "entropy": 5.401802015304566,
      "epoch": 1.8564092588990468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004652542433308657,
      "loss": 5.0555,
      "mean_token_accuracy": 0.18831026554107666,
      "num_tokens": 41388927.0,
      "step": 23860
    },
    {
      "entropy": 5.4216851711273195,
      "epoch": 1.8567982882707645,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004652394457111309,
      "loss": 5.1542,
      "mean_token_accuracy": 0.1792440265417099,
      "num_tokens": 41397823.0,
      "step": 23865
    },
    {
      "entropy": 5.41874361038208,
      "epoch": 1.857187317642482,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046522464520481517,
      "loss": 5.1313,
      "mean_token_accuracy": 0.18840500712394714,
      "num_tokens": 41406322.0,
      "step": 23870
    },
    {
      "entropy": 5.404124927520752,
      "epoch": 1.8575763470141995,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046520984181214333,
      "loss": 5.0803,
      "mean_token_accuracy": 0.18615312725305558,
      "num_tokens": 41415118.0,
      "step": 23875
    },
    {
      "entropy": 5.387262153625488,
      "epoch": 1.8579653763859172,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004651950355333398,
      "loss": 5.1327,
      "mean_token_accuracy": 0.1892439231276512,
      "num_tokens": 41424139.0,
      "step": 23880
    },
    {
      "entropy": 5.4476062774658205,
      "epoch": 1.8583544057576347,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004651802263686294,
      "loss": 5.1589,
      "mean_token_accuracy": 0.1785170093178749,
      "num_tokens": 41432017.0,
      "step": 23885
    },
    {
      "entropy": 5.434660148620606,
      "epoch": 1.8587434351293521,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004651654143182367,
      "loss": 5.0652,
      "mean_token_accuracy": 0.18635716736316682,
      "num_tokens": 41440279.0,
      "step": 23890
    },
    {
      "entropy": 5.340988111495972,
      "epoch": 1.8591324645010698,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004651505993823866,
      "loss": 5.0034,
      "mean_token_accuracy": 0.19016454815864564,
      "num_tokens": 41449325.0,
      "step": 23895
    },
    {
      "entropy": 5.345670127868653,
      "epoch": 1.8595214938727875,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046513578156130383,
      "loss": 5.0066,
      "mean_token_accuracy": 0.18737512677907944,
      "num_tokens": 41457898.0,
      "step": 23900
    },
    {
      "entropy": 5.365746021270752,
      "epoch": 1.859910523244505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004651209608552131,
      "loss": 5.041,
      "mean_token_accuracy": 0.19412644803524018,
      "num_tokens": 41466410.0,
      "step": 23905
    },
    {
      "entropy": 5.467255640029907,
      "epoch": 1.8602995526162225,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046510613726433945,
      "loss": 5.1488,
      "mean_token_accuracy": 0.18901192992925644,
      "num_tokens": 41475061.0,
      "step": 23910
    },
    {
      "entropy": 5.285003566741944,
      "epoch": 1.8606885819879402,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004650913107889078,
      "loss": 5.1142,
      "mean_token_accuracy": 0.1854396179318428,
      "num_tokens": 41483569.0,
      "step": 23915
    },
    {
      "entropy": 5.363281536102295,
      "epoch": 1.8610776113596577,
      "grad_norm": 1.1875,
      "learning_rate": 0.000465076481429143,
      "loss": 5.1252,
      "mean_token_accuracy": 0.18486839681863784,
      "num_tokens": 41492113.0,
      "step": 23920
    },
    {
      "entropy": 5.457696628570557,
      "epoch": 1.8614666407313751,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004650616491852701,
      "loss": 5.1846,
      "mean_token_accuracy": 0.18312276005744935,
      "num_tokens": 41500722.0,
      "step": 23925
    },
    {
      "entropy": 5.368642902374267,
      "epoch": 1.8618556701030928,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046504681405751426,
      "loss": 5.0199,
      "mean_token_accuracy": 0.19694875031709672,
      "num_tokens": 41509272.0,
      "step": 23930
    },
    {
      "entropy": 5.414296817779541,
      "epoch": 1.8622446994748103,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046503197604610047,
      "loss": 5.1946,
      "mean_token_accuracy": 0.1829557701945305,
      "num_tokens": 41517379.0,
      "step": 23935
    },
    {
      "entropy": 5.385214471817017,
      "epoch": 1.8626337288465278,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046501713515125393,
      "loss": 5.0581,
      "mean_token_accuracy": 0.18976382166147232,
      "num_tokens": 41525737.0,
      "step": 23940
    },
    {
      "entropy": 5.391162872314453,
      "epoch": 1.8630227582182455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004650022913731998,
      "loss": 5.0964,
      "mean_token_accuracy": 0.18889861702919006,
      "num_tokens": 41535016.0,
      "step": 23945
    },
    {
      "entropy": 5.405890941619873,
      "epoch": 1.8634117875899632,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046498744471216335,
      "loss": 5.1803,
      "mean_token_accuracy": 0.18907341808080674,
      "num_tokens": 41544253.0,
      "step": 23950
    },
    {
      "entropy": 5.43790636062622,
      "epoch": 1.8638008169616807,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046497259516836974,
      "loss": 5.1431,
      "mean_token_accuracy": 0.18406940698623658,
      "num_tokens": 41551907.0,
      "step": 23955
    },
    {
      "entropy": 5.519653606414795,
      "epoch": 1.8641898463333981,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046495774274204446,
      "loss": 5.1601,
      "mean_token_accuracy": 0.18881761878728867,
      "num_tokens": 41560675.0,
      "step": 23960
    },
    {
      "entropy": 5.449908924102783,
      "epoch": 1.8645788757051158,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004649428874334127,
      "loss": 5.1521,
      "mean_token_accuracy": 0.18223918378353118,
      "num_tokens": 41570256.0,
      "step": 23965
    },
    {
      "entropy": 5.435288381576538,
      "epoch": 1.8649679050768333,
      "grad_norm": 1.125,
      "learning_rate": 0.0004649280292427,
      "loss": 5.115,
      "mean_token_accuracy": 0.18666388243436813,
      "num_tokens": 41578990.0,
      "step": 23970
    },
    {
      "entropy": 5.375592947006226,
      "epoch": 1.8653569344485508,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004649131681701318,
      "loss": 5.1178,
      "mean_token_accuracy": 0.18590096086263658,
      "num_tokens": 41587896.0,
      "step": 23975
    },
    {
      "entropy": 5.402475357055664,
      "epoch": 1.8657459638202685,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046489830421593347,
      "loss": 5.1118,
      "mean_token_accuracy": 0.1851385548710823,
      "num_tokens": 41596160.0,
      "step": 23980
    },
    {
      "entropy": 5.433856630325318,
      "epoch": 1.866134993191986,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004648834373803307,
      "loss": 5.0959,
      "mean_token_accuracy": 0.18642021268606185,
      "num_tokens": 41605025.0,
      "step": 23985
    },
    {
      "entropy": 5.428277635574341,
      "epoch": 1.8665240225637034,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046486856766354893,
      "loss": 5.0699,
      "mean_token_accuracy": 0.19110781699419022,
      "num_tokens": 41612768.0,
      "step": 23990
    },
    {
      "entropy": 5.3803609848022464,
      "epoch": 1.8669130519354211,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046485369506581387,
      "loss": 5.1278,
      "mean_token_accuracy": 0.1848846361041069,
      "num_tokens": 41620892.0,
      "step": 23995
    },
    {
      "entropy": 5.438239717483521,
      "epoch": 1.8673020813071388,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046483881958735124,
      "loss": 5.2143,
      "mean_token_accuracy": 0.18417198956012726,
      "num_tokens": 41629708.0,
      "step": 24000
    },
    {
      "epoch": 1.8673020813071388,
      "eval_entropy": 5.300835802282016,
      "eval_loss": 5.197207927703857,
      "eval_mean_token_accuracy": 0.1921778305518845,
      "eval_num_tokens": 41629708.0,
      "eval_runtime": 29.3576,
      "eval_samples_per_second": 1415.578,
      "eval_steps_per_second": 176.956,
      "step": 24000
    }
  ],
  "logging_steps": 5,
  "max_steps": 128520,
  "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": 5.6550519963648e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}