ind-latn-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed10/checkpoint-6000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4668352460610776,
  "eval_steps": 3000,
  "global_step": 6000,
  "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
    }
  ],
  "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": 1.4145558571008e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}